PRELIMINARY 12-MONTH REPORT
                      of the
          LEAD PAINT ABATEMENT AND
 REPAIR AND MAINTENANCE STUDY IN BALTIMORE


                    Conducted by
       KENNEDY KRIEGER RESEARCH INSTITUTE
                       for
    U.S. ENVIRONMENTAL PROTECTION AGENCY
               Ben Lim, EPA Task Manager
               Technical Programs Branch
              Chemical Management Division
          Office of Pollution Prevention and Toxics
July 21, 1995

-------
          PRELIMINARY 12-MONTH REPORT
                        of the

            LEAD PAINT ABATEMENT AND
 REPAIR AND MAINTENANCE STUDY IN BALTIMORE

                       Conducted by
        KENNEDY KRIEGER RESEARCH INSTITUTE

                          for

    U.S. ENVIRONMENTAL PROTECTION AGENCY
                 Ben Lim, EPA Task Manager
                 Technical Programs Branch
               Chemical Management Division
            Office of Pollution Prevention and Toxics
PREPARED BY:

Mark R. Farfel, Sc.D., Project Director
Charles Rohde, Ph.D., Project Statistician
Peter S. J. Lees, Ph.D., QC Officer
Brian Rooney, Data Analyst
Desmond I. Bannon, Laboratory Chief
Luban Alvi, Data Manager

July 21, 1995

-------
                         EXECUTIVE SUMMARY

      This  preliminary  report is  based  on data  from  60  (80%)
 intervention houses  and  29 controls that have completed  one  year  of
 follow-up   in   the   Lead-Based  Paint  Abatement  and  Repair   &
 Maintenance  (R&M) Study.  The research aim  is to characterize and
 compare  the  short  (2-6  mo.)  and  longer-term  (12-24  mo.)
 effectiveness  of three  levels of  lower cost  R&M interventions
 designed to reduce children's exposure to lead in paint and dust,
 known sources of high lead exposure in U.S. children.  The study
 population  consisted of  non-Hispanic  black households   (mostly
 renters)  residing in  Baltimore rowhouses with reported  low  monthly
 housing payments  (mean  $324). At the outset, mean ages of study
 children  ranged from  25 to 33 months across groups, and the GM PbB
 levels in the three R&M groups and previously abated (PA) controls
 ranged from 10  to 14.5 ng/dL. The GM PbB in children in the modern
 urban (MU,  post-1979) control houses was 4
      The report presents longitudinal data from five study groups
 across  five campaigns. A number of types of graphical displays are
 included to  facilitate understanding of the changes found in dust
 lead and children's blood lead. The reader who has a preferred type
 of  data display or limited time is directed to specific tables and
 figures as follows:  data  tables- see Tables 5-8;  box  plots with
 median  traces- see Figures 1-13, 17-19;  bargraphs - see Figures 14-
 16,  20-23; for plots-  see Figures 24-28  (blood data only).   The
 main findings are listed below:

 •     All three levels  of R&M intervention were associated with
      sustained reductions  in dust  lead (PbD)  loadings,  lead
      concentrations   (PbD-C) ,  and  dust loadings.  The  degree  of
      reduction in PbD and PbD-C was positively associated with the
      level of  the  R&M intervention.  Statistically  significant
      differences  were found  between R&M groups  after intervention
      and at two,  six,  and  twelve  months.

 •     Over time, the  MU  control group had significantly  lower dust
      lead levels than the other four study groups. Dust lead levels
      in the  PA control group  were  generally  intermediate between
      the middle  (R&M II) and high (R&M  III) level of intervention.

 •     Children with pre-intervention PbBs^20 jug/dL had statistically
      significant  reductions  in PbB to  levels <20 jug/dL  at twelve
      months post-intervention.  Children  with pre-intervention PbBs
      <20  /ig/dL did not  have  statistically significant PbB changes
      within or between R&M groups. An  increasingly downward,  but
      not  statistically  significant,  trend in  PbB was  found  in
      children in the  high  level intervention group  (R&M  III) .

•     Children in the  MU control group  had  significantly lower PbBs
      over time as compared with each  of the other four groups.

•    Using all  five  study groups in longitudinal data  analysis,
     children's PbB was significantly related to dust lead.

-------
                         TABLE OF CONTENTS
 1.0   INTRODUCTION 	  1

 2.0   STUDY  DESIGN AND SAMPLE COLLECTION  PROCEDURES   	  3
      2.1  Selection Criteria for  Houses  and Children  	  5
      2.2  Repair & Maintenance  Interventions  	  6
      2.3  Recruitment and  Enrollment   	  7
      2.4  Sample Collection  Procedures   	  8

 3.0   LABORATORY  ANALYSIS PROCEDURES  	  10

 4.0   DATA PROCESSING AND STATISTICAL ANALYSIS  PROCEDURES   .  .  11
      4.1  Data Processing	11
      4.2  Data Summary	12
      4.3  Quality Control	13
      4.4  Statistical Analysis  	  14
      5.1  Side-by-Side Boxplots with Median Traces   	  23
      5.2  Descriptive Statistics  - Twelve Month Campaign   .  .  25
      5.3  Longitudinal Data  Analysis	29

 6.0   DISCUSSION	34

 7.0   REFERENCES	47
TABLES  1-10

FIGURES 1-28


APPENDIX A:  Descriptive Statistics

APPENDIX B:  Longitudinal Data Analysis: Dust Models

APPENDIX C:  Longitudinal Data Analysis: Blood Lead Models

-------
                          List of Tables
Table  1:

Table  2:


Table  3:


Table  4:
Data collection plan

Types and numbers of 12-month campaign  samples
collected and analyzed for lead

Types and numbers of 12-month campaign  samples
collected by study group

Frequency of family moves and reoccupancies between the
initial and the twelve month campaigns  among the 60 R&M
houses included in this preliminary report and the
control houses.

Overall dust lead loadings for houses completing the
12 -month campaign

Overall dust lead concentrations for houses completing
the 12 -month campaign

Overall dust loadings for houses completing the 12-
month campaign

Descriptive Statistics for Blood Lead Concentrations
for Children with Initial PbB <20 //g/dL

Twelve month campaign correlations of In (lead loading),
In (lead concentration)  and In (dust loading) with
In (blood Atg/dL)  for the youngest child per household
Table 5:


Table 6:


Table 7:


Table 8:


Table 9:
Table 10: Twelve month campaign correlations of In (lead loading),
          In (lead concentration) and In (dust loading) with
          In (blood /ug/dL) for all children
                                ii

-------
                          List of Figures
 Figure l:


 Figure 2:


 Figure 3:


 Figure 4:


 Figure 5:


 Figure 6:


 Figure 7:


 Figure 8:


 Figure 9:


 Figure 10:


 Figure 11:


 Figure 12:


 Figure 13:


 Figure  14:

 Figure  15:

 Figure  16:

Figure  17:


Figure  18:
 Boxplot:  Dust Lead Loadings across Campaigns for
 Floor Surfaces

 Boxplot:  Dust Lead Loadings across Campaigns for
 Window Sill Surfaces

 Boxplot:  Dust Lead Loadings across Campaigns for
 Window Well Surfaces

 Boxplot:  Dust Lead Loadings across Campaigns for
 Interior  Entryway Surfaces

 Boxplot:  Dust Lead Concentrations  across Campaigns
 for  Floor Surfaces

 Boxplot:  Dust Lead Concentrations  across Campaigns
 for  Window Sill  Surfaces

 Boxplot:  Dust Lead Concentrations  across Campaigns
 for  Window Well  Surfaces

 Boxplot:  Dust Lead Concentrations  across Campaigns
 for  Interior  Entryway  Surfaces

 Boxplot:  Dust Loadings across Campaigns  for  Floor
 Surfaces

 Boxplot:  Dust Loadings across Campaigns  for  Window
 Sill  Surfaces

 Boxplot:  Dust Loadings across Campaigns  for  Window
 Well  Surfaces

 Boxplot:  Dust Loadings across Campaigns  for
 Interior  Entryway  Surfaces

 Boxplot:  Blood Lead  Concentrations for Children
 with  Initial  Blood Pb  <20 ug/dL

 Bar Graph: Dust Lead Loadings over Time

 Bar Graph: Dust Lead Concentrations over Time

 Bar Graph: Dust Loadings over Time

 Boxplot: Dust  Lead Loadings across Groups at the
 12 Month Campaign

Boxplot: Dust  Lead Concentrations across Groups at
the 12 Month Campaign
                               111

-------
                     List of Figures  (cont.)
Figure  19:


Figure  20:


Figure  21:

Figure  22:


Figure  23:

Figure  24:


Figure  25:


Figure  26:


Figure  27:


Figure  28:
Boxplot: Dust Loadings across Groups at the 12
Month Campaign

Bar Graph: Overall Lead Levels and Dust Loadings
by Group

Bar Graph: Overall Dust Lead Loadings by Group

Bar Graph: Overall Dust Lead Concentrations by
Group

Bar Graph: Overall Dust Loadings by Group

Plots of Children's Blood Lead Levels across Tine
- R&M I

Plots of Children's Blood Lead Levels across Tine
- R&M II

Plots of Children's Blood Lead Levels across Tine
- R&M III

Plots of Children's Blood Lead Levels across Tine
- Modern Urban

Plots of Children's Blood Lead Levels across Tine
- Previously Abated
                               IV

-------
                                                 Preliminary RtM Study Report-
                                                 Partial 12 Month Data  7/95
 1.0  INTRODUCTION
      This preliminary report is based on data from eighty percent
 of  the  intervention houses that have completed the twelve  month
 campaign of the Lead-Based Paint Abatement and Repair  &  Maintenance
 (R&M)  Study  which  is  underway at  the  Kennedy Krieger  Research
 Institute (KKRI)  in Baltimore.  The R&M study  is being conducted
 according to the Quality Assurance Project Plan  approved by EPA in
 November 1992.1
      Included in  this  report are  twelve month data from twenty
 houses in each of the three R&M intervention groups (Levels I-III,
 n=60  houses),  15  (100%) modern urban (MU)  control houses, and 14
 (93%)  previously and comprehensively abated (PA) control houses.
 The twelve month  data from these 89 houses were collected between
 March 1994 and May 1995.  The level  of participant cooperation with
 home  and  clinic  visits  has been  high.   Consequently,  we  have
 sufficient preliminary data to begin to address the following study
 objectives:

 •    Describe changes in lead levels in settled house dust for R&M
     Levels I-III  across five campaigns, i.e., pre- and immediate
     post-intervention, and  two months,  six months,  and twelve
     months post-intervention.
 •    Describe changes in environmental variables between baseline
     and the  twelve month  campaign for  the  modern  urban  and
     previously abated  control  houses.
 •    Fit the  study's  statistical   model  for   longitudinal  data
     analysis to the dust  lead  and  blood lead data.

     An  earlier report provided  descriptive statistics on baseline
 environmental,  biological,  and demographic data for the five study
 groups (total of  107 houses).:  At  baseline,  the study population
 consisted  of  non-Hispanic black households with reported  low-to-
moderate monthly  rents and  mortgages  (mean $324)  residing  in
Baltimore  City  row houses.   Most  (81%)  households were  renters.

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 The median household  size  was 4 persons, and the number  of  study
 children per household ranged from 1 to 4 children.  Mean  baseline
 ages of children ranged from 25 to 33 months across study groups.
 Accounting  for  multiple  children  per  household   (clustering),
 geometric  mean  baseline blood lead concentrations (PbB)  were 5,  13,
 10,  14.5,  and 13 ng/dL for the modern urban,  previously  abated,  and
 R&M I-III  groups, respectively.
      At baseline, the three R&M groups were similar with respect to
 demographic  characteristics   and  lead  levels   in  blood   and
 environmental samples.  However, lead levels in settled house  dust
 and children's  blood at baseline tended to be highest  for the R&M
 Level III  houses (vacant at baseline), lowest for the R&M Level  I
 houses (occupied at baseline), and intermediate for the R&M Level
 II  houses  (mix  of vacant and occupied houses at baseline).2   Some
 of  the owners of the vacant R&M III  and  II  houses accepted study
 families with children with blood lead elevations >20 /xg/dL as  per
 their policy.  Consequently, unlike R&M I  households, R&M Levels II
 and  III  had  some  children  with  initial  baseline   blood   lead
 concentrations >20 /xg/dL.  These children were analyzed separately
 in  this report.
 Purpose of R&M  Study
      The R&M study  is designed  to  characterize and  compare  the
 short  (2 to 6 month)  and longer-term (12  to  24 month)  efficacy of
 comprehensive lead-paint abatement and less costly and potentially
more cost-effective Repair and  Maintenance (R&M)  interventions  for
 reducing lead  in settled house dust  and children's blood.    The
 investigators plan to  extend  the follow-up  to five years  post-
 intervention.    This research  is important because  residential
paints  and dusts have  been identified as major sources  of  lead
exposure for  U.S. children, particularly those living in houses
with  deteriorated paint and high  dust lead  levels.3 Furthermore,
there  is   a  dearth  of  research  literature  on the outcomes of
interventions for children with blood lead  concentrations in  our

-------
                                                 Preliminary R*M Study Report-
                                                 Partial 12 Month Data  7/95
 primary study range of 10 to 20 jtg/dL.
      The R&M  approach may provide  a  practical means  of  reducing
 exposure for  future generations  of children who will  continue to
 occupy housing containing  lead-based paint.  However,  systematic
 studies of  the R&M approach are needed in order to provide a sound
 scientific  basis  for  prevention  policies.    EPA   funding  is
 anticipated for two years of follow-up testing in each  study house
 to periodically  measure  lead  in  settled dust,  soil,  water,  and
 children's  blood.  Pre- and immediately post-R&M intervention data
 provide baseline measurements  for the study of short  and  longer-
 term changes in  lead levels in children's blood and settled house
 dust.
      The overall  research goal of the R&M study is to contribute to
 the  scientific basis for a standard of care  for lead-painted houses
 via  the conduct  of  a  longitudinal  intervention  study.   Specific
 study objectives  are as follows:
1.   Measure and compare the short and longer-term changes of  lead
     in settled house dust and children's blood associated with R&M
     Levels  l-lll  and previous comprehensive abatement.
2.   Characterize  the nature  of  the  relationship between lead in
     children's blood and settled house dust.
3.   Evaluate  dust collection methodologies for the determination
     of  lead  in  residential  dusts,   including wipe  and  cyclone
     methods.  This objective was addressed in past reports.4"6
2.0  STUDY DESIGN AMD  SAMPLE COLLECTION PROCEDURES
     This prospective study has two main components and five groups
of  study  houses (Table  1) .   The  first component is  designed to
obtain serial measurements of lead in  venous  blood  of children 6
months through  4 years  of age at baseline,  settled  house dust,
soil, and drinking water  in three groups of 25 R&M houses (total of
75 houses),  each receiving  one of three  levels  of R&M intervention

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data   7/95
 (Levels I-III)  (Table  1).   The study  questionnaire, designed to
 obtain information  on  demographics and  covariates  which could
 influence  lead  exposure  in the  home  (e.g.,  hobbies  and child
 behavior) are done at six month intervals beginning at enrollment.
      Occupied study houses (target n=37) were randomly assigned to
 receive either  R&M Level I or R&M Level II interventions.  Houses
 vacant at  the  time of  intervention (target n=38)  were randomly
 assigned to receive R&M Level II or Level III  interventions.  Since
 R&M  Level  II interventions were done  in both occupied and vacant
 houses,  the randomization scheme  was  designed  to  ensure equal
 numbers of  houses (n=25) at each R&M level.  More frequent sampling
 campaigns  were planned  for  R&M houses  during the  first  year to
 allow for the estimation of  the rate of reaccumulation of lead in
 dust and  an early assessment  of the need for  further cleanup or
 repairs.
      The second  component of the study design is to obtain serial
 measurements  of  lead  in house dust,  soil,  drinking water,  and
 venous  blood  of  children  (aged   6  months  through  4  years  at
 baseline),  in  two groups of control houses  (Table  1).   The study
 questionnaire  is  administered  at  six  month  intervals  in  both
 control groups.
      One control  group  is a  group  of 16 scattered  site row houses
 in older housing neighborhoods  which received  a comprehensive type
 of lead-paint  abatement performed  by pilot abatement projects in
 Baltimore between May 1988 and February 1991.   The  two years of
 planned  follow-up  will  provide an  opportunity to measure  the
 efficacy of comprehensive abatement  practices at 4-to-6 years post-
 abatement.   Pre- and  immediate post-abatement  dust  lead data  are
 available for these previously  abated  houses  from past studies.
     Modern urban houses  built  after 1979, and  presumably free of
 lead-based  paint, constitute   the  second control  group  (n=16).
These houses are located in modern  urban subdivisions in  which  all
houses are  of this type.

-------
                                                 Preliminary RsM Study Report-
                                                 Partial 12 Month Data  7/95
 2.1   Selection Criteria for Houses and Children

      The following selection criteria were applied to all groups:

 1.    House size:  approximately 800-1200 square feet.

 2.    House Condition: Structurally sound without pre-existing
      conditions that could impede or adversely affect the R&M
      treatments and the safety of the workers and field staff
      (e.g. roof leaks  or unsafe floor structures).

 3.    Utilities (heat,  electric, and water) were available to
      facilitate intervention and field sampling.

 4.    Occupants: Household included at least one child 6 months
      through 4 years of age who was neither mentally retarded
      nor  physically  handicapped  with restricted  movement;
      family had no definite and immediate plans to move.

 5.    House was  not excessively  furnished.  This  criterion
      enabled  dust collection in  all  houses and intervention
      and cleanup  efforts in occupied R&M houses.

      The following selection criteria were applied to R&M candidate
      houses only:

 6.    House contained lead-based paint (defined in Maryland as
      *0.7  mg  Pb/cm2  or  *0.5   percent   lead  by  weight  as
      determined  by wet  chemical analysis)  on  at  least  one
      surface in a  minimum of two rooms or, in the absence of
      testing,  year of construction pre-1941.

 7.    House dust lead levels prior to intervention exceeded
      Maryland's  interim  post-abatement  clearance  levels
      (<200 jiig/ft2, <500 Atg/ft2 and <800 Mg/ft2,  respectively,
      for  floors, window sills, and window wells)  at a minimum
      of  three  locations.

8.    House had 12 or fewer windows needing  R&M work.  This was
      to  facilitate the R&M work with  limited  resources.

     Additional criterion  applied to  previously  abated houses:

9.   At  least  two pairs each of pre- and  immediately  post-
     abatement  dust-wipe  lead  measurements  from  the  same
     floor,  window  sill,  and  window  well  surfaces  were
     available  from previously  collected  data.   These  data
     provide baseline dust lead  levels.

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 2.2  Repair & Maintenance Interventions
      The Repair & Maintenance interventions under study were  all
 financed  by the  Maryland  Department of  Housing  and Community
 Development  (DHCD) via a special loan program available to property
 owners.  The initial R&M intervention costs were capped by  DHCD as
 follows:  Level  I:  $1,650;  Level  II:  $3,500;  Level  III:   $6,000-
 $7,000.   All  R&M  work was  performed by workers  trained  in lead
 paint abatement work  according to  Maryland regulations.   These
 interventions,   described   in  detail  elsewhere,1  are  briefly
 described  below.
      R&M  Level   I   interventions   included  wet   scraping   of
 deteriorating  lead-based  paint  on   interior  surfaces,  limited
 repainting of scraped surfaces, installation of an entryway mat,
 wet cleaning and vacuuming with a  high efficiency particulate  air
 (HEPA) vacuum to the extent possible  in an occupied unit, education
 of  occupants and  owners,  and stabilization of  exterior  painted
 surfaces to  the extent possible  given the budget  cap.    R&M I
 interventions were done exclusively in  occupied  units and generally
 took  a full  day to complete.
      Two key elements  added to R&M  Level  II  interventions were
 floor treatments to make floors smooth  and easily cleanable and  in-
 place window and door treatments to reduce abrasion of lead painted
 surfaces.   Half of  the R&M II interventions  were performed   in
 vacant units.  In the occupied units, floor treatments were limited
 to  rooms and areas  where the work was  feasible.   R&M  II  work was
 generally completed  within  two days.  Precautions were taken in the
 occupied R&M I  and  II  houses to protect children  and  furnishings
 from lead exposure during the work  phase.  These measures  included
having children out of the  house  until all work and  cleanup was
completed  and the  use of  containment measures  such  as  plastic
sheeting to protect belongings.
     For  R&M  Level  III,   the  highest  level  of  intervention,

-------
                                                 Preliminary RiM Study Report-
                                                 Partial 12 Month Data  7/95
 additional work included window replacement as the primary  window
 treatment, encapsulation  of exterior  window and door trim with
 aluminum coil stock, and more durable floor and stairway treatments
 (e.g.  coverings).  All R&M  households  received cleaning  kits  for
 their  own wet cleaning efforts which included a bucket,  sponge mop,
 sponges,  a replacement sponge mop head,  TSP cleaning agent, and  the
 EPA brochure entitled "Lead  Poisoning and Your Children."
 2.3 Recruitment and Enrollment
     The enrollment process entailed a three step process of pre-
 enrollment, formal  enrollment,  and ongoing  pre-enrollment  and
 enrollment activity as described below.
     Extensive home visiting activity (1100 home visits  to  over  650
 modern urban, previously abated, and older occupied dwellings)  was
 performed by field staff as part of pre-enrollment field activities
 during  the spring  and  summer  of  1992.    Over 90  percent  of
 households  identified as  potentially  eligible  for  the  study
 indicated an interest in participating.   This early pre-enrollment
 activity  yielded 100 interested and eligible households for  formal
 enrollment.
     Formal enrollment refers to the obtaining of signed  informed
 consent  statements for study participation  from parents or legal
 guardians  for both environmental and biological  sampling.   Separate
 consent  statements were obtained  for each child enrolled in  the
 study  using forms  approved  by the  Joint Committee on  Clinical
 Investigation of  the Johns Hopkins Medical Institutions.
     Between the  time of formal enrollment  and the  initiation of
 the  initial data  collection campaign  in January  of  1993, some
 enrolled households  were  lost to  the study,  primarily due  to
 children  aging and the moving of families to other dwellings.  In
 some cases, the losses  necessitated ongoing pre-enrollment activity
to identify an increased pool of potential study participants.  No
evidence  was found for selection bias when excluded  R&M-candidate
houses were compared to the  R&M study houses.2

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 2.4   Sample Collection Procedures
      Venous blood  was  collected from  children at  the  KKRI Lead
 Clinic by a pediatric phlebotomist into 3  mL Vacutainersฎ with EDTA
 added as an anticoagulant.
      Trained  field teams  collected  all  environmental  samples,
 including field QC samples (blanks and duplicates).  Settled house
 dust was collected  using a modified high  volume cyclone sampler
 originally developed  for  EPA  for  the  evaluation of  pesticide
 residues in house dust.7  The modified device,  referred to as the
 R&M  cyclone,  is described in detail and  characterized elsewhere.4
 The  device  consists of  a  cast aluminum cyclone attached to a hand
 held Dirt Devilฎ vacuum as the air mover for the system.   A 100 mL
 Teflonฎ microwave digestion  liner was used as the sample collection
 container  to eliminate a sample  transfer step  in the laboratory,
 thereby  reducing  the risk of sample  loss.
      The sampling plan for settled dust included the collection of
 three composite floor samples  per house  -  one across rooms with
 windows  on the first story,  one  across rooms with windows on the
 second story and one  from  first and  second story  rooms  without
 windows.   Two randomly selected 1-ft2  (929-cm2)  perimeter floor
 locations were  sampled  in each room  designated  for inclusion in a
 composite  sample.   Composite window sill and window well  samples
 were  collected  separately from all first  and second story windows
 available  for sampling at  each campaign,  respectively.   Settled
 dust  was  collected as  individual  samples  from horizontal  and
 accessible portions  of air ducts,  interior and exterior entryways,
 and the main  item of upholstered furnishing.
      Soil core samples were  collected as separate composites of the
top 0.5  inch (1.3  cm) of soil  from 3  randomly selected locations at
the drip  line.  Cores were collected  into  polystyrene liners using
a 6-inch (15.2 cm)  stainless  steel recovery probe.

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
      Drinking water  samples were collected  as 2-hour fixed-time
 stagnation samples from the kitchen faucet rather than first draw
 samples to avoid  disrupting families early  in the  morning.  The
 procedure was to  run the cold water for  at  least two minutes to
 flush the pipes and then to collect the first flush of water  after
 a 2-hour  interval.  A list of field sample types is provided below.
         Settled-dust  fR&M cyclone)
           Perimeter floor composite - 1st and  2nd  story
            rooms with windows and rooms without windows
           Window sill composite     - 1st and  2nd  story
           Window well composite     - 1st and  2nd  story
           Air duct/upholstery
           Interior entryway
           Exterior entryway
         Soil core  Drip-line composite
         Drinking water  Kitchen faucet
         Field QC
           Blanks (one per house) and duplicates
          (every tenth house) for all sample types.	

      Information on the  study  children and their  households was
collected using  a  structured interview questionnaire. As planned,
families were informed by letter of the results of all dust lead
and  blood  lead  tests.   Dust  test results  were  provided  on  a
qualitative   basis  with  recommendations   for  priorities  for
housekeeping.   Families with water  and soil lead  concentrations
that  exceeded EPA guidance  levels  were provided with  additional
recommendations for avoiding  lead exposure  (e.g. EPA's guidance to
run the  water to flush the pipes prior to use  whenever the water
has been in  contact with the  plumbing for two  hours or  more) .
Additionally, separate letters were sent to the parents/guardians
with  the results of the  blood  lead tests so  that they could  be
shared with  the child's  primary care provider.   All  blood lead
results  were  reported  to  the Maryland Childhood Lead Registry  as
required by Maryland law.

-------
                                                  Preliminary RSM Study Report-
                                                  Partial 12 Month Data   7/95
3.0  LABORATORY ANALYSIS  PROCEDURES
     Interior  and exterior settled dust, exterior  soil,  water, and
venous   blood   samples  were   analyzed  at  the   Kennedy  Krieger
Institute's Trace Metal Laboratory using established analytical
methods.   Microwave  digestion was used for dust, soil,  and water
samples  (modified SW  846 Methods 3015 and 3051).   Analysis of dust
and soil digestates was performed using Inductively Coupled Plasma-
Atomic  Emission  Spectrometry  (ICP)   (SW 846  Method  6010)  and/or
Graphite Furnace Atomic  Absorption  Spectrometry  (GFAA)  (SW 846
Method  7421).   Analysis  of drinking water  was  by GFAA  (SW 846
Method  7421).    Blood was analyzed  by GFAA using matrix  matched
standards.3  The  table below summarizes  the procedures.

                    Summary of Laboratory Procedures
Sample Type
Dust
Soil
Drinking Water
Blood
Pre- Preparation
Summary
Post-field drying
and gravimetrics
Sample drying and
homogenization
none
Stabilized in EDTA
Preparation Summary
Digest using 1:1 HN03: H20
with microwave heating
Digest using 1:1 HNO3: H20
with microwave heating
Digest using 1:1 HNO3: H20
with microwave heating
Addition of matrix
modifier/triton X-100
solution
Analysis
Summary
ICP/GFAA'
GFAA
GFAA
GFAA
  Samples with lead concentrations below the limit of quantitation
  of the ICP instrument were analyzed by GFAA.
                                 10

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data   7/95

 4.0   DATA PROCESSING AND STATISTICAL ANALYSIS PROCEDURES
 4.1   Data Processing

      Sources  of  data  include  the field  collection  forms,  the

 questionnaire, and the laboratory data packages.  Raw data of all

 types were transferred to the Data Manager who  uploaded the data to

 a  VAXStation   3100  computer  and  prepared  the  data for  later

 analysis.  Below is a summary of the data processing steps employed
 for  the  three  sources of data.


 •     The field data set consists of all data recorded on the field
      collection forms for settled  dust,  soil,  and drinking water
      samples as well as  room and window inventory data.   A data
      entry firm keypunched and  verified (double entry)  data from
      the field forms into ASCII data files.  These raw data files
      were transferred to the Data Management team for management,
      storage,  and later analysis.   Field data forms  were checked
      for completeness and accuracy by both the  Outreach Coordinator
      and Data  Manager prior  to data entry.   Data were re-verified
      by  laboratory staff after data entry.


 •     Laboratory data were electronically stored  for each laboratory
      instrument as  follows:  Mettler  Balance:  gravimetric  data
      (tared  and loaded weights for cyclone-dust and soil samples)
      ICE:  lead measurements  for  cyclone-dust  samples.  Graphite
      Furnace-AAS;  lead content of drinking water, soil, blood,  and
      low lead  concentration  cyclone dust samples.   Electronically
      stored  laboratory  data  from  the  Mettler,  ICP,  and  GFAA
      instruments were moved to Paradox (v.4.0)  by laboratory staff
      for tracking  of  samples.  The  data  were  re-verified  by
      laboratory staff after  data entry. Paradox  data were  then
      converted to ASCII  files  by  the  Data  Management team  for
      uploading to the  VAXStation.   A  SAS program  read in  the
      laboratory data  for  environmental and  blood  samples  and
      created SAS datasets  for data  analysis.


•    Questionnaire data forms were keypunched and verified (double
     entry)  by the data  entry firm  into  ASCII data files.   These
     raw data  files were verified in-house  and transferred to  the
     Data Manager.   A SAS program read  in the raw data and created
     SAS datasets for data analysis.
                                11

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 4.2  Data Summary
      This preliminary  report includes data through the twelve month
 campaign for 60 R&M houses and 29 control houses as  follows:
        R&M Level I:       20 (80%)  of 25 planned houses
        R&M Level II:      20 (80%)  of 25 planned houses
        R&M Level III:      20 (80%)  of 25 planned houses
        Modern Urban:      15 (100%) of 15 planned houses
        Previously Abated: 14 (93%)  of 15 planned houses.

      Environmental dust  data  from four  surface types (perimeter
 floor,  window sill, window  well, and interior entryway)  included
 in  each of the first five data collection  campaigns  (pre-R&M, post-
 R&M,  two  months,  six months,  and twelve  months post-R&M)  are
 included in  this report as well as data collected less frequently
 (i.e.,  airduct dust, upholstery dust,  soil,  and water).  Tables 2
 and 3 display the types and numbers of  12-month campaign samples
 planned, collected, and analyzed for lead  by  study group for the 89
 houses  included  in this report.
      Some  of  the original  study  families  moved or voluntarily
 withdrew from the study between the initial  and twelve month data
 collection campaigns.  Table 4  reports  the frequency of family and
 child moves, and  reoccupancies by new  study families,  by study
 group.   Approximately  21 percent (13/60)  of  the original families
 in the 60 R&M houses moved prior to the twelve month campaign.  By
 the  twelve  month  campaign,  all  of  these   study  families  were
 replaced by the next family that moved  into the house.   Despite our
 success  in gaining the participation of  these  new  families, they
 had  fewer  eligible children than  the  original families.   By the
 twelve month campaign,  the study also  gained 15 children who were
 newborns that became of age (ฃ6 months)  for  blood lead testing.
      One R&M II  house  was vacant at the  time of  the  twelve month
 sampling.  None of the  houses included  in  this report  are known to
have had any  major renovations or repairs  during the  first year of
 follow-up.   One R&M I  house had its front and back  doors replaced

                                12

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data   7/95
 due to break-ins that damaged the  original doors,  and in another
 house the wallpaper was  removed by the occupants  from the first
 floor rooms by a steam process.
      Children  in most  of  the  occupied  R&M  I   and II  houses
 experienced an extended time interval of  6  to 10 months between the
 initial PbB and the completion of the R&M work.  This interval was
 due to the  time taken  by collaborating property owners to provide
 a  completed R&M  loan  application and the  time  required  for loan
 processing,  loan approval, and  loan  closing.  The  actual  R&M work
 contributed relatively little to  the actual time intervals  between
 the initial  and subsequent  campaigns.    An effort  is  currently
 underway,  with the  permission of  study families,  to  obtain  any
 additional  PbB data  that might be available from these children's
 medical records for  the period between  the initial study PbB and
 the start of the R&M work that  could  serve as new baseline study
 values closer to the times of the interventions.   This effort was
 not needed  for children who moved  into  the vacant  R&M II and III
 houses after  the R&M work was completed.

 4.3  Quality  Control
     At  the  time of  this report,  the performance audit  which
 consists  of  a review of  the laboratory  QC  charts  has  been  an
 ongoing activity of the QC Officer.   QC samples include instrument
 QC  samples   (initial  and continuing  calibration  verification
 samples,  initial  and  continuing  calibration  blanks,   standard
 reference materials, spikes and  spike duplicates),  method blanks,
 and field QC  samples (field blanks, field  duplicates).   The data
quality  objectives  related  to  laboratory and field  performance
continue to be met.  The QC Officer has  also completed the  system
audit  of  the  laboratory and  field procedures.   No  significant
deficiencies have been  noted.   Since the  data audit  is ongoing  and
the twelve month  campaign has not been completed,  this report  is
termed preliminary and partial in nature.

                                13

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
      The ongoing  data  review has  recently  revealed occasional
 inconsistencies over  time in  the  measurement  of  various window
 components in some of the study houses.  The measurement of window
 components  can  be a  problem  due  to  the  multiple  window  types
 encountered;    selection  of   the   appropriate   dimensions  to
 characterize  the component requires considerable judgement by the
 field teams.  Shortly after field work commenced, the protocol for
 window measurement was greatly expanded and was training conducted
 to  reduce potential measurement problems.    Some problems remain
 however.   At  this time,  the  principal  investigator  and the QC
 Officer are in the process of personally inspecting  and  remeasuring
 all  windows identified with discrepancies  in measurements in any
 campaign.  As a result of these remeasurements,  it is expected that
 some lead and  dust  loadings and statistical  analyses based upon
 these measures in the  six- and twelve-month  preliminary reports
 will  be revised slightly.

 4.4   Statistical Analysis
      In order to compare the same  houses over time, we restricted
 the longitudinal data analysis  to the 60 R&M houses and 29 control
 houses for which data were  available  from the twelve month campaign
 as of May 1995.
      For  data  analysis  purposes,  lead  values   less  than  the
 instrument detection limit  (IDL) were  coded  as the IDL/v/2.9   For
 lead values less than the  limit of  quantitation (LOQ)  but greater
 than the IDL, the  observed value was used  in  the data analysis.
     One  child in a previously abated house  had a  blood  lead
 increase  to a concentration of 53 ng/dL at the 12  month campaign
 and was  provided with chelation therapy.  This child is an outlier
 in this  study  and was excluded  from  the statistical  data analysis.
The mother reported that the affected child has more frequent hand-
to-mouth activity than  the  siblings who did  not experience similar
blood lead increases.   Inspection of the house revealed exterior

                                14

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 surfaces with  deteriorated lead paint which had been  stabilized,
 but not  removed,  as part  of the original abatement  work 4  to  5
 years  ago.   Also, lead  in paint and dust was identified in the
 basement  which  had not been treated at the time of abatement  since
 the basement was not a finished living area.  The property owner
 plans to make additional repairs.
 Descriptive statistics
      SASฎ  PROC UNIVARIATE  indicated that the  environmental  and
 blood lead  distributions were skewed.  As expected, use of the log
 transformation  reduced  the  amount  of  skewness  and  produced
 histograms  and boxplots  which  were  approximately  normal   (see
 boxplot Figures 1 to  13) .   Descriptive  statistics  for blood  and
 environmental variables were produced after transforming the  data
 using the natural  logarithm (In).
      SASฎ  PROC  UNIVARIATE  was   used   to  produce  descriptive
 statistics for all  sample  types with one  observation per house.
 Since multiple observations were  available per house for settled
 dust  from window  sills,  and window  wells,  floors  in  rooms  with
 windows,  as well as  for children's blood, additional analysis  was
 performed using SASฎ PROC MIXED with house as  a random effect to
 address the issue of  clustering  (i.e. multiple  observations  per
 house).   Geometric  mean  (GM)  values, standard  errors,  and  95%
 confidence  intervals were obtained using  the  following PROC MIXED
 models  fitted separately for each  study  group  (R&M  Levels l-ni,
 modern  urban,  previously abated),  surface type  (floors  in  rooms
 with windows, window  sill, window well), and matrix (dust,  blood):

     In(PbD)    = 30 +  ln(E)   (Eq.l)
     In(PbD-C)  = 3o +  ln(E)   (Eq.2)
     ln(DL)     = 3o +  ln(E)   (Eq.3)
     In(PbB)    = 3o +  ln(E)   (Eq.4)

where,  PbD  =  dust  lead  loading  (/zg/ft2);  PbD-C  =  dust  lead

                                15

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 concentration  (jig/g); DL = dust  loading  (mg/ft2) ; PbB = blood lead
 concentration  (ng/dL); 60 =ln(a); a = a constant; ln(E) is  normally
 distributed.
 Side-by-side  boxplots
      Side-by-side boxplot figures with median traces are presented
 in  this  report  as  a  means  of  displaying  lead   levels across
 campaigns within and across  study groups.   For  this preliminary
 report,  the boxplots across  time for the  environmental  data are
 limited  to the 89 houses with data  available at the twelve  month
 campaign.  Boxplots were  generated using  S-Plusฎ software10 (see
 section   5.1   for   a  description  of  boxplot  components).  The
 descriptive statistics  presented in this  report  include  "extreme
 values"  that  are   indicated  by  the symbol  '*'  in the  boxplot
 displays.
 Statistical method  for  analysis  of longitudinal data
      Statistical methods for the  analysis of longitudinal data have
 developed rapidly over  the  last  decade.  These methods,  which are
 natural extensions of multiple regression and  analysis of variance,
 are extremely flexible.  Current  longitudinal  methods allow for the
 inclusion  of   random  and  fixed  effects,   longitudinal   (time
dependent) covariates and constant  covariates, as well as discrete
and continuous  covariates, all  in a  multiple regression context.
In  this   study,  for  example,  we  have the following  types  of
covariates:

      (1)  type of house  - fixed effect,  discrete
      (2)  house - random effect,  discrete
      (3)  dust lead  - fixed time  dependent  continuous covariate
      (4)  child - random effect,  discrete
      (5)  R&M Level  - fixed effect, discrete
      (6)  time - fixed time dependent continuous covariate
      (7)  age of child - fixed time dependent covariate
      (8)  season - fixed discrete covariate

                                16

-------
                                                  Preliminary RซM Study Report-
                                                  Partial 12 Month Data   7/95
      The response variable modeled  was dust lead reading or blood
 lead concentration (transformed  using  logarithms).   These response
 variables as  well as their  associated time  dependent covariates
 will be observed at times described in Table 1.
      For the  dust lead measurements  let Y1 denote the  vector of
 responses over time  for the ith house  i.e.  x  is an nix  vector of
 the form Y, =  (yiifyl2,... ,yllu)T where  yx]  is  the response for the ith
 house at time t:  and "T" stands for the transpose  operation. Then
 the general form of the model is:

           Y1 = XX0 + Z.b, + e1

 where Xt is an  /i xp   matrix of  covariate values  for  the  fixed
 effects,  3 is  a pxl vector of parameters for the fixed effects, Zi
 is  an nAxq matrix of covariate values for the random effects, bx is
 a qxl vector of random  effect parameters and e,.  is an r\xl vector
 representing random error.   We have N such models,  one for each
 house.
      Estimates of the  parameters in  the  overall model  are obtained
 using the methods outlined  in  published papers.11"17 The  essential
 feature  of these methods is the use  of weighted least  squares with
 a "working" estimate of the covariance matrix followed  by  iteration
 with  an  updated   estimate  of  the   covariance  matrix   until
 convergence.  The estimate of the variance-covariance matrix of the
 fixed effects is  robust,  in the   sense  that it  is   consistent,
 regardless of the form of the "working" estimate  of the covariance
 matrix.  The model for blood lead readings will be similar with the
 above model specified for each  child.
     Our primary interest in this study is in the  parameters of the
model which represent the effect of R&M interventions on  dust lead
and  blood lead.    The fact that this model  allows estimation of
these parameters  in the presence  of heterogeneity between houses

                                 17

-------
                                                 Preliminary RSM Study Reporc-
                                                 Parcial 12 Month Data  7/95
 and temporal  correlation, and produces variance estimates which are
 robust is extremely important.  Should it  be  necessary we can,  in
 this  framework,  also  consider  response variables   which  are
 discrete.
      The general nature  of  the model makes  it  ideal  for a study of
 this type where  there  is  the potential for unbalance.   Since the
 model is house or child  specific, depending on whether dust lead or
 blood lead is being modelled, we do not require that the number of
 observations  through time be equal.  Thus, should a child move or
 otherwise be eliminated  from  the  study  the  house data can  be
 analyzed while the data for that  child can be included  up to the
 point of departure.  Should another child be entered  into the  study
 at  that  house his or her blood lead readings can be included in the
 blood lead analysis for the remainder of the study,  thus  providing
 partial  information for that child.   The common residence of the
 children is  included  in  the  house  covariate which  allows  for
 correlation structure between these observations.
      Age-related effects in  the analysis of blood  lead  responses
 need  to  take  into account the fact that blood lead is not linearly
 related  to age since it tends to  increase between  six months and
 two years and decrease slowly among children over two years  of  age.
 This  is  done  by the use of  linear  and quadratic  terms for age in
 the  model.    The  presence  of several  children in a house, which
 introduces another source of correlation,  (i.e.  between children in
 the same house)  is accounted for  by using house as a random effect
 which introduces  the  required correlation.
      SAS PROC MIXED AND GEE18 software were used  for longitudinal
 data  analysis.
 Specifications of longitudinal models  for dust
      In the analysis of the preliminary twelve month dust data, we
have  fit the  statistical model proposed in the QAPjP to the data.
The results of the compositing self  study  indicated that an overall
measure  of lead  exposure could be considered  with little loss of

                                18

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95

 information.   This was true for both dust lead concentrations and

 lead  loadings.    These results  suggest  that  the  readings from

 multiple sample sites  in a  house  might be combined to produce an

 overall  measure  to  use  as  a covariate in  the  model   relating

 environmental  lead levels to blood lead levels.  Consequently, we

 have explored the use of factor analysis as a method for combining

 individual  sample  results.  The results indicate that:
     One  factor,  the  first,  accounts  for  (79%  to  83%)  of the
     variability of environmental dust  lead when all five groups
     are  analyzed together.   The findings were  similar when the
     three R&M groups  were analyzed separately.

     The  second  factor measures  the  difference between the floor
     lead readings and window sill and  window well lead readings
     and  accounts for (12% to 14%) of  the variability when all five
     groups are analyzed together.  The findings again were similar
     when the three R&M  groups were analyzed  separately.
Thus far, the percentages  of  the variability of the dust readings

accounted  for  by the  factor loadings  have remained  stable over
study groups and campaigns (see  below).
Five Study Groups:
Oust
Measure

Pb Loading
Pb Cone.
Dust Loading
Initial
Campaign
factorl factor2
.81 .14
.82 .12
.61 .22
Six Month
Campaign
factorl factor2
.83 .12
.81 .14
.67 .18
Twelve Month
Campaign
factorl factor2
.83 .12
.79 .14
.69 .17
                                19

-------
                                                Preliminary RSM Study Report-
                                                Partial 12 Month Data  7/95
 R&M Groups Only:
Dust
Measure

Pb Loading
Pb Cone.
Dust Loading
Initial
Campaign
factorl factor2
.80 .15
.73 .19
.69 .17
Six Month
Campaign
factorl factor2
.79 .15
.74 .18
.70 .16
Twelve Month
Campaign
factorl factor2
.79 .15
.75 .17
.70 .16
 Given  the  stability of the factors over time, we feel comfortable
 using  them as the variable to measure environmental lead levels.
 The  first  factor was  used  as  the  dependent  variable in  the
 longitudinal  data analysis of dust changes.

 The following models were fit to the dust data:
ln(factorl)13kl =   00 + jS^season^ +  02*levellk
                    + @3*campaign1 (level ฑk)
                    + b^house, +  eljkl

In(factorl)
                                        +  ]32*levellk
                    + 03*levelk (campaign^)
                    + bl*housel  +
where,
I refers  to house, j to  season,  k to R&M Level, 1  to  campaign,
level(campaign)= campaign nested within R&M level to compare dust
levels over time within R&M groups, and campaign(level)=  R&M level
nested within campaign to compare R&M groups at each  campaign over
                               20

-------
                                                 Preliminary RซM Study Report-
                                                 Partial 12 Month Data   7/95
 time.   Following   standard   practice  regression   coefficients
 corresponding to "fixed effects" are denoted by Greek letters while
 regression coefficients  corresponding  to "random  effects"  are
 denoted by ordinary letters (e.g.  b).
      SAS  PROC MIXED has been used  for  the analysis thus far since
 a  log  transformation  of lead  values  is indicated  as  previously
 mentioned.
 Specifications  of  longitudinal models  for blood lead
      In order to address the study  objectives  with regard to blood
 lead changes,  we  fit  two main types  of models to the  data.  The
 first model,  hereinafter referred  to  as the exposure  model,  was
 used to characterize the relationship  between blood  lead and dust
 lead.   in this  model,  the two  dust  lead factors were  included as
 dependent  variables  along   with   demographic  and   behavioral
 variables.   The  second  model, hereinafter  referred  to  as  the
 comparison model, was used to investigate blood  lead  levels across
 groups  and within groups over time.   The  two models are as follows:
 Exposure  Model
 ln(PbB)lklm = 0o  + P!*factorliklm +  ]32* factor 2 lklm
                + P3*ageiklm +  P,*age2,klm + j35*seasonlklm
                + (36*genderlk,. +  J37*mouthing,klm
                + 08*campaign1 +  f3g* factor I*campaign1
                + jB10*factor2*campaign1
               + b^house, +  bm(I)*childm(I)  +  eiklm

The  initial campaign blood and  dust  lead values  for children   who
moved into the vacant R&M  II  and III houses after intervention were
excluded from the exposure model.  Their  initial  PbB values at  the
time of move-in reflect body burdens associated  with exposures in
their past living environments and not their new  home  environments.
                                21

-------
 Comparison  Models
                                                Preliminary RSM Study Report-
                                                Partial 12 Month Data  7/95
ln(PbB)lklm  =
                +  02*age2lklm
+ J34*genderiklm + j35*mouthing,klm
+ J36*groupk  +  J37*campalgn1
+ j3g*groupk (campaignL)
+ b^house,  +  bm(I)*childmm +  e
                                                (33*seasonLklm
                                                  ''iklm
ln(PbB),klm =
                                              + j33*seasoniklm
                  04*genderlklm +  (35*mouthing,klm
                  06*groupk  + /37*campaign1
                  jBg*campaign1(groupk)
                               b2*childm(I) + eiklm
where,
I refers  to house,  k to group, 1 to  campaign,  m to child within
house,  group(campaign)= campaign  nested within  study  group  to
compare blood levels  over time within groups, and campaign(group)=
group  nested within campaign  to  compare groups at each campaign
over time.  Following  standard practice  regression coefficients
corresponding to  "fixed effects" are denoted  by Greek letters while
regression  coefficients corresponding  to  "random effects"  are
denoted by ordinary letters (e.g.  b).   The comparison models were
fit using  all blood lead data and then  separately for children with
PbB levels <20/Ltg/dL and 220/jg/dL.
                               22

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data   7/95
 5.0  RESULTS
 5.1  Side-by-Side Boxplots with Median Traces
      In  order  to  graphically  display   changes   over  time  in
 environmental  and blood  data this  report provides  a series  of
 boxplot displays  with median trace lines  connecting  the  median
 values across time.   These  descriptive displays do not take into
 account season or any other potential covariates.
      In a boxplot display, 50 percent of the data  is  contained in
 the box shown  in  the figure; the  bottom  of the box is the lower
 quartile  and  the  top of the  box  is the third  quartile,  the
 horizontal line inside the box represents  the sample median.   The
 vertical  lines extending from the box represent  the  expected lower
 and upper  range of the data based on the variability of  the  central
 portion of the data.  The fences are 1.5 interquartile  ranges from
 the upper  and lower edges of the box.  Extreme values are indicated
 by  an  asterisk.19  The widths of the boxes in any  given  side-by-side
 boxplot are  proportional  to the number of observations.
 Dust Data Boxplots
     Figures 1-12 are boxplot displays presented  by  study group for
 each of the  main surface  types  (floors  in rooms  with windows,
 window  sills,  window wells,  interior  entryways)  showing  the
 distributions of dust lead loadings, dust lead concentrations,  and
 dust loadings, respectively,  across  campaigns.   These displays
 allow  for  comparisons both within and across groups over time.
     The median traces for lead loadings across most surface types
 in R&M Level I-III houses  show a pattern of  reduced  levels at post-
 intervention, most pronounced for  R&M III,  followed by increases  at
 two  months to levels  that  remained below pre-intervention levels.
At six and twelve  months,  the median lead loadings tended to  remain
relatively stable;  R&M I and II houses had  unchanged or moderately
 increased  levels  across  surface  types and R&M  III   houses had
unchanged or moderately decreased  levels  (Figures 1-4).    Deviations
from this  pattern  include:  (a)  interior entryways in R&M I  houses

                                23

-------
                                                 Preliminary RsM Study Report-
                                                 Parcial 12 Month Data  7/95
which did not have reduced levels at post-intervention and  floors
in R&M I  houses which did  not show a pattern  of  rebound at  two
months and  (b)  floors and  entryways  in R&M  II  houses which  had
decreased levels  at  post-intervention  but  no  increases  at  two
months (Figures 1 and 4).
      For  dust  lead concentrations,  the  median traces  reveal  a
downward trend at post-intervention and at  two months across  sample
types  that is most pronounced in R&M III  and R&M  II houses compared
with   R&M I  houses.     At  six  and  twelve  months,  the  lead
concentrations tend to remain relatively  stable in R&M  I houses  and
show unchanged or moderately decreased levels across surface  types
in R&M Level II  and III houses (Figures  5-8).
     The  median traces for dust loadings  in the three R&M groups
show  patterns  similar to  those  described  above  for the lead
loadings  (Figures 9-12).
     In the  modern urban  and previously  abated control houses  the
median traces for dust show a pattern of  relatively  stable lead
loadings,  lead concentrations,  and dust  loadings over time with a
slight downward  trend at  six and  twelve months  for  lead loadings
and dust  loadings (Figures 1-12).

Blood  Data Boxplots  with Median Traces
     Figure  13  provides boxplot displays  of  the unadjusted blood
lead concentrations (jug/dL)  over time by study group for the subset
of children  with  initial  PbB <20 /ng/dL.   The outlier in  the  PA
group  at  twelve months with a PbB of 53  /ug/dL is  excluded from
Figure  13 and the  statistical data analysis for  reasons explained
in section  4.4.   The median  traces  for  all  five study  groups
indicate modest or  little  change over  time.
                                24

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data   T/95

 5.2  Descriptive Statistics  - Twelve Month Campaign
 Settled Dust

      Descriptive statistics  for settled dust at  the twelve month
 campaign are displayed as follows:


 •     Bar graphs of GM dust lead loadings (PbD,  Mg/ft2),  dust lead
      concentrations  (PbD-C,  Mg/g),  and dust loadings (DL,  mg/ft2)
      by study group and surface type  (Figures 14  to 16).

 •     Boxplot displays of PbD, PbD-C and  DL distributions by study
      group and surface type  (Figures 17-19).

 •     Descriptive statistics  (geo.  mean,  n, min, max, sd)  for the
      twelve month PbD, PbD-C, DL respectively,  for  the  five study
      groups and seven surface types  (interior entryways,  floors in
      rooms with and  without  windows,  window sills,  window wells,
      upholstery and air ducts) are presented in  Appendix  A,  Tables
      a-c) .    Sample  sizes  for  upholstery  and   air  duct data  are
      limited for reasons provided in Table  2.  The geometric means
      and 95% confidence intervals for floors in rooms with windows,
      window sills,  and window wells are  the values  produced by an
      analysis that takes into account clustering  (section  4.4).

      Geometric  mean   (GM)  lead  loadings  across  all  groups  and

 surface  types at the twelve month campaign were <603  Mg/ft2,  except

 for air  ducts in all groups (range  of GMs 856 jug/ft2 (MU) to 18073

 Mg/ft2 (R&M  I))  and window wells  in R&M  I  (19412  ng/fi  ),  R&M II

 (1,761 /ig/ft2),  and previously abated houses (1172  ng/ft2) (Appendix

 A, Table b).  For R&M Levels I-lll,  respectively, the GM PbD  values

 were  112, 79, and 36 fig/ft2 for  floors in rooms  with windows;  603,

 201,  and 23 fig/ft2 for window  sills  and 19412, 1761,  and  268 Mg/ft2
 for window  wells.

      GM  dust lead  concentrations  across all groups and surface

 types at twelve months were <4000 Mg/g  (ppm), equivalent to <0.40%,

 except for  window sills  (7754 Mg/g) and window wells (22963 Mg/g)
 in R&M I houses (Appendix  A,  Table a).

     At the twelve month campaign,  GM dust loadings  across surface

types were  <340 mg/ft2 except for window wells  in  MU, PA, and  R&M

I and R&M II groups  (range 456 to 841 mg/ft2) and air ducts  (range


                                25

-------
                                                 Preliminary FUM Study Report-
                                                 Partial 12 Month Data  7/95
 8474 to 15184 mg/ft2)(Appendix A Table c).
      Modern urban houses  continued to have the lowest lead loadings
 at the  twelve month campaign.  GM values  across  surface types were
 all <217 Mg/ft2 except  for air ducts.   For floors,  windows sills,
 and window wells at twelve months, R&M I houses had significantly
 higher  GM PbD  values  (112,  603,  and  19412  jug/ft2, respectively)
 compared to  R&M III houses (36,  23,  and  268  /ig/ft2,  respectively);
 GM values in R&M II  houses were  intermediate  (79, 201, and 1761
 Mg/ft2,  respectively).
      At twelve months, modern  urban houses  continued to have the
 lowest  GM PbD-C  levels across  surface types  (<440 /ug/g) .   The GM
 PbD-C values  for  interior entryways and interior floors  across the
 other four study groups (R&M I-III and PA) were not statistically
 different  from  each other  (Appendix A,  Table a) . For windows sills
 and  window wells, R&M  I  houses  had  significantly higher GM PbD-C
 values  (7754 and 22963 M9/9/  respectively)   compared  to R&M  III
 houses   (796  and  1130  ng/g,  respectively),  with R&M  II  houses
 intermediate (2941 and 3862 pq/q,  respectively).
      Study groups were most similar to  each other  in terms  of dust
 loadings.  However, dust  loadings  tended  to  be highest in R&M  I
 houses,  lowest  in R&M III  houses and  intermediate in R&M  II  houses.
 For windows sills and window wells,  R&M I houses had significantly
 higher  GM  dust loading values  (78  and  841  mg/ft2, respectively)
 compared to R&M III houses  (29 and 237 mg/ft2, respectively), with
 R&M  II  houses  intermediate  (68  and  456  mg/ft2,  respectively)
 (Appendix  A  Table c).
 Dust  Lead  Correlations  between  Surface Types
 Statistically significant correlations (p<.05) were found  between
 the dust lead levels from  most of the surfaces types at the twelve
month campaign.   Tables d-f in  Appendix  A display the correlation
matrices for dust  load loadings,  lead  concentrations,  and dust
 loadings, respectively.  The highest  correlation coefficients were
 for window sills and window wells (.73  for PbD and .67 for PbD-C).

                                26

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 Overall Summary Measures of Dust Data
      Summary  measures  of  the dust  data  for  each  house  were
 calculated based on the weighted average of the readings across the
 four  surface  types  common   to  all  campaigns,   i.e.,   floors,
 entryways, window sills,  and  window wells.   The weighting factor
 for lead loadings and dust loadings was the surface area  sampled.
 For lead concentrations the weighting factor was the  sample mass.
      Figure 20 displays the overall geometric mean  PbD, PBD-C and
 DL values  at the twelve month campaign by study group.   Figures 21-
 23 display the GM values for the three overall  dust measurements by
 study group for each of the five data collection campaigns  through
 the twelve month campaign. Descriptive statistics (GM,  range)  for
 the overall PbD, PbD-C, and DL readings respectively by campaign
 are displayed in  Tables 5-7.   Based on  these  summary measures,
 overall  GM lead loadings were approximately 35 times higher in R&M
 I  houses as compared with R&M III houses.  This difference is due
 to order of magnitude higher lead concentrations and  several  fold
 higher dust loadings in R&M I houses relative to R&M  III houses.
 Drip-Line  Soil
      Drip-line soil samples were not collected at the  twelve month
 campaign.  Therefore,  this  report provides preliminary six month
 data  on  lead concentrations of drip-line soil (PbS)  by study group
 (Appendix A Table g) .  These data are limited due to the fact  that
 most  study houses have no drip-line  soil.   The PbS levels in modern
 urban houses remained similar to the  initial levels (six month
 GM=73  nq/g, range  34  to 229 Mg/g versus initial GM=63 ng/g, range
 29  to  154  Mg/g)•    Across  previously   abated  and  R&M  houses
 individual PbS  values ranged from 182 to  7845  /xg/g  at six months
 compared to the  range  of  233 to  15968  Mg/g  observed  at pre-
 intervention/baseline.
 Drinking Water
      Drinking water samples also were  not collected  at the twelve
month campaign.   Water lead concentrations  (PbW) at six months were

                                27

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Daca  7/95
 unchanged from  their  geometric mean  baseline levels  of ฃ4 /ig/L
 (ppb)  across groups.  The range of values also remained the  same  -
 less  than the instrumental  limit of detection  (
-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 clustering)  (max n=79).
 5.3  Longitudinal Data Analysis
 Dust Lead
      The longitudinal models described in section  4.4 were  fit to
 the data for  lead  loadings, lead concentrations, and dust  loadings
 from the three R&M groups and all five study groups.  Fitting  the
 models  to the data  from  the   three   R&M groups  allowed  for
 investigation of changes across all five campaigns  (initial,  post-
 intervention, and  two, six, and twelve month) , two  of which  (post-
 intervention  and two month) were not common  to the  control groups.
 The three  campaigns common to all study groups were  included  in  the
 five group models.
      In these models the  dependent  variable was factorl obtained
 from the factor analysis,  a factor that accounted  for most  of  the
 variability of environmental dust lead.   For  each dust variable
 (PbD,  PbD-C,  DL), the model was fit  to the data with campaign
 nested within study groups  in order to  examine  trends  over time
 within study groups, and with study group  nested within campaign  in
 order  to compare study  groups at each campaign.
     The main findings of the longitudinal analysis of the dust
 data  after controlling for  season  are   listed  below.   In the
 presence  of  other  covariates  (campaign,   group),  season  was a
 significant fixed  effect  in  the lead loading and dust  loading
 models but not  in  the lead concentration  models.
     The  PROC  MIXED  output  is  displayed   in  Appendix  B.  The
 printouts have the  following codes for campaign  and group:  00=pre-
 intervention,  PI=post-intervention,  02=two  months,  06=six months
 and 12=twelve  month, IN=initial  campaign (5 group model), 1=R&M I,
 2=R&M II,  3=R&M III). Interpretation of the  estimates  obtained by
 SAS PROC MIXED obey the  usual  rules of interpretation of regression
coefficients,  i.e., the  coefficient of  a covariate is the expected
change in  the response  variable associated  with a  unit  change in

                                29

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data   7/95

 the covariate in the presence  of  the  other covariates.   When the

 covariate is  a  dummy variable,  a unit  change in  the  covariate

 corresponds to the expected difference between the response at the

 level of  the covariate  compared  to the  omitted  level.   If  the

 effect  is a  nested  effect,  e.g.,  level nested within campaign the

 coefficient  represents  the comparison of  that level versus  the

 omitted  level  for  a   fixed  campaign  and   this  allows  the

 determination of differences between levels of intervention.


 Dust  Lead loadings:

 R&M group comparisons across time

 •     Despite   higher   baseline   levels,    lead  loadings   were
      significantly  lower (p<.01)  in R&M III  and R&M II  houses as
      compared with R&M I houses at immediately post-intervention,
      two months  (R&M III  only),  six months,  and twelve months.   At
      each campaign,  PbD  was lowest in R&M  III,  highest  in  R&M I,
      and  intermediate in R&M II.

 Changes over  time within R&M groups

 •     For all three R&M groups,  post-intervention  lead loadings were
      significantly  lower than their pre-intervention  levels.  For
      R&M  II and  III houses,  PbD levels at  two months,  six months,
      and  twelve  months  were  significantly  higher  than   their
      corresponding post-intervention levels. In R&M  I houses  PbD
      levels over time were  not  significantly different from their
      immediate post-intervention levels.


Lead  concentrations:

R&M group comparisons

•    Pre-intervention dust  lead concentrations  (PbD-C)  were  not
     significantly different across the three R&M groups.   PbD-C
      levels were significantly lower (generally p<.01)  in  R&M II
     and  III  as  compared  with   R&M  I   at  immediately   post-
     intervention, two months,  six months  and twelve months.  At
     each campaign PbD-C was lowest in R&M  III,  highest  in  R&M I,
     and intermediate in R&M II.
                                30

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 Changes over time within R&M groups
 •     For R&M II and III (but not R&M I), lead concentrations were
      significantly   reduced    (p<.01)   at   immediately   post-
      intervention.  For both R&M I and II houses, PbD-C  levels at
      two months,  six months  and twelve months were  significantly
      lower than  their  corresponding levels  at  immediately  post-
      intervention.  In  R&M III, only the  twelve month  level  was
      significantly lower (p<.01) than the post-intervention  level.

 Dust loadings:

 R&M oroup comparisons

 •     At pre-intervention,  dust loadings  (DL)  were  significantly
      higher  in  R&M II and III houses as compared  with R&M  I houses.
      At each  point in time post-intervention the estimates for  R&M
      II and  III were negative relative to R&M I; however, none of
      differences reached statistical significance.   R&M  II houses
      had intermediate DL levels at each follow-up campaign.

 Changes over  time within R&M groups

 •     In all  three R&M groups,  dust loadings were  significantly
      reduced  (p<.01) at immediately post-intervention.  None of  the
      follow-up  levels were significantly different from their  post-
      intervention  levels although the levels tended to be  higher at
      follow-up  (positive coefficients).

 Control houses:

 •     Dust lead loadings  in  the  modern urban  (MU)  houses  were
      significantly lower (p<.01) than those  in the three R&M groups
      and the  previously  abated (PA)  houses at baseline,  six month
      and twelve months.   PbD levels tended to decrease over time in
      MU  and  PA   houses;   however,   none  of  the  changes  were
      statistically significant.

 •     Dust lead  concentrations  at  baseline,  six  month and twelve
      months were significantly lower in  MU houses as compared with
      each of the other four groups.   At six and twelve months, lead
      concentrations in MU  and PA  houses were not  significantly
      different  from their  baseline levels.

•     Dust loadings at baseline  were significantly lower in modern
      urban houses  as compared with each  of  the R&M groups but not
      the previously abated  group.   At follow-up,  none of the study
      groups had dust loadings  that were statistically different
      from those in the modern urban study houses.


                                31

-------
                                                 Preliminary RSM Study Report-
                                                 Parcial 12 Month Data  7/95
 Blood Lead

      The exposure and comparison  models  described in section 4.4

 were  fit to the blood lead data for children  in the three R&M study

 groups and for children in all five  study groups.   Fitting the

 models to data from the three R&M groups allowed for investigation

 of  changes across  all four PbB  testing campaigns  (initial,  two

 month, six month and twelve month).  The three campaigns (initial,

 six  month, twelve  month) common  to all  five  study  groups were

 included in the five  group models.

      The main  findings of the  comparison models for investigating

 PbB changes within and between groups  are  listed  below.  The PROC

 MIXED output is displayed in Appendix  C with the following codes:

 (campaigns)0=pre-intervention/initial, 2=two month, 6=six month and

 12=twelve month,  L1=R&M  Level  I, L2=R&M II,  L3=R&M III.)


 R&M group comparisons

 •     For  children with  initial PbB <20  jiig/dL,  no  statistically
      significant  blood  lead differences were  found  between  the
      three R&M groups at baseline and  two, six  and  twelve months
      post-intervention controlling  for  age and  season.  R&M  I
      children  tended  to  have  lower  PbB levels at each  campaign,
      including baseline,  relative  to R&M  III  children.


Changes over time within  R&M groups

•     For children with initial  PbB <20  M9/dL  in  each of  the three
      R&M groups, no statistically  significant changes in PbB were
      found at the two, six and  twelve month campaigns controlling
      for age  and  season.   However, a  consistently downward  but
      nonsignificant trend in blood lead was  found in  children in
      R&M III  houses.   (The  coefficients at two,  six, and  twelve
     months were negative and  increased in magnitude over time.)
                                32

-------
                                                 Preliminary R4M Study Report-
                                                 Parcial 12 Month Data  7/95

 Control Houses; Chances within groups and group comparisons

 •     For children  with initial  PbB  <20 jxg/dL  in the previously
      abated  control  houses,   no  statistically  significant  PbB
      changes were  found  at  the  six  and twelve  month  campaigns
      controlling for age and season.

 •     Controlling for  age  and  season,  children  in  modern  urban
      houses  had PbB  levels that were  lower  and statistically
      different from those  of  children in each  of the other  four
      study groups at the initial, six and twelve  month campaigns.
      Modern  urban  children   had  a   small  but  statistically
      significant  increase in blood lead concentration over baseline
      at the six month campaign only.


 PbB  changes in children with baseline PbB s?20 ug/dL

 •     For the  small numbers of children with initial PbB 120 /ig/dL,
      a  downward trend in blood lead concentration was found at  the
      two,   six  and  twelve month campaigns.  These  changes  were
      statistically significant (p<.05)  at twelve months in R&M  II,
      R&M III  and PA groups, controlling for age and season.


 Exposure Model

      The main findings of the exposure  models  for investigating  the
 relationship between blood lead  and dust  lead  are listed below  (see
 Appendix C  for  the  PROC MIXED output):

 •     Age,  age2, and  season  (summer vs  nonsummer)  were significant
      contributors to the 3 and 5 group  models; gender and hand-to-
      mouth activity  (high vs low)  were  not. Blood lead tended to be
      higher  in  males and significantly  higher in males  in  the 5
      group model  with dust  loadings only.

•     Controlling  for age and other covariates  (campaign,  group,
      dust factors)  included in  the various  blood lead models, the
      seasonal change in children's  PbB  was estimated  to be  +1.3
      /xg/dL in summer relative to  winter.

•     Using  all   five   study  groupsr   dust   lead  loadings   and
      concentrations  (factor1   and  factor2)   and   dust  loadings
      (factor2 only), were significantly  related  to children's PbB
      after adjusting for age, season, campaign and the inclusion of
     random effects  for houses  and multiple children per  house.
                                33

-------
                                                 Preliminary RsM Study Report-
                                                 Partial 12 Month Data   7/95
      Using  the  three   R&M  groups.  dust   lead  loadings  and
      concentrations  (factor1  and  factor2)   and  dust  loadings
      (factorl),  were not significantly related  to children's PbB
      after adjusting for age, season,  campaign and  the inclusion of
      random  effects for  houses and multiple  children per house.
      Factor2  was a significant factor in  the PbB  model with dust
      loading  factors.
      The  interactions of factorl and factor2 with campaign were not
      statistically  significant for lead  concentration factors and
      dust loading  factors.  For lead loading, the interaction of
      factor2  and campaign  was only marginally  significant.   For
      these reason,  the exposure  models  were  rerun  without these
      interaction terms.
 6.0  DISCUSSION
     To  date,  this  study has  met its  enrollment goals  and  has
 completed  nearly all  of the planned  data collection  across  the
 first  four sampling campaigns.  Laboratory  performance  and  data
 quality objectives continue to be met.  The study also continues to
 benefit  from  a high level  of family cooperation  with  both blood
 lead  and  environmental  sampling  well  into  the  twelve  month
 campaign.  Through  the  twelve   month  campaign  approximately  20
 percent of the  original  families in  the  60 R&M houses included in
 this report moved.   In  every case,  we successfully  enrolled  the
 next family to occupy the house, assuring that  at minimum we  can
 continue to do the environmental sampling.  Most new families also
 had eligible  children  who were  enrolled  in the blood lead testing
 component of  the  study.
 Dust Lead
     All three levels of  R&M intervention under investigation were
 found to  be  associated  with  overall reductions in  interior dust
 lead loadings,  lead concentrations, and dust loadings that were
 sustained for  twelve months  below pre-treatment levels.  Moreover,
the degree of  reduction in lead  exposure  was positively  associated
with  the  level  of   the R&M   intervention,   and  statistically
significant differences  were found between R&M  groups  over time.
                                34

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 After  intervention and  at  two,   six,  and  twelve  months,  lead
 loadings and  lead  concentrations were  lowest  in R&M  III houses,
 intermediate  in  R&M II  houses,  and  highest in R&M  I houses.  A
 similar but nonstatistically significant pattern was  observed  for
 dust loadings over time post-intervention  (Figures 21-23).
      At the outset, R&M II and III houses had significantly  higher
 lead loadings  and  dust  loadings  as  compared  with R&M  I  houses
 despite random  assignment  of  houses  to   intervention groups as
 explained in section 2.0.   Since lead concentrations at the  outset
 were similar,  the R&M group differences in lead loadings were  due
 to higher dust loadings which in turn may be attributed to the fact
 that half of the  R&M II and all of the R&M III houses  were  vacant
 at the  outset.  Lead dust  has been shown to reaccumulate in  vacant
 houses.20
      Reaccumulation of  lead-containing dust in  R&M  houses  was
 greatest during the first two months post-intervention  as compared
 to the  relatively stable period from  two  months to twelve  months
 (Figures 1-8 and  21-23).  This early reaccumulation in R&M  II  and
 III  houses may be due to a combination  of reaccumulation of lead in
 dust after   intervention  and prior  to  occupancy  as well  as  the
 potential importation of lead containing dust  into the house  during
 move-in by  study  families.
      At the twelve month campaign,  overall  weighted  averages  of  the
 dust lead loadings  (based on floor,  window sill, window well,  and
 interior entryway samples)  were  26-fold higher  in R&M  I houses as
 compared with R&M III houses, and  5-fold higher  in R&M  I houses as
 compared  with  R&M  II  houses   (Table  5).   These  substantial
differences  in lead  exposure are attributable mainly to differences
 in lead concentrations  and  secondarily to  differences  in dust
 loadings  (Figure 20) .    This was also true with  regard to the
observed 2.7-fold  difference in overall PbD levels between R&M III
and the modern urban houses  at twelve  months (Figure 20).
                                35

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
      Differences in overall  dust  lead concentrations  between  R&M
 groups  following  intervention   were  expected  since   R&M   III
 interventions addressed  lead paint  sources to a  greater degree
 (e.g.   by  window  replacement)  than  either  R&M   I   or  R&M  II
 interventions. Despite this,  it is  interesting to  note that dust
 lead concentrations  in R&M  I and II  houses at two  months,  six
 months,  and  twelve months  were  significantly lower  than their
 corresponding levels  immediately post-intervention. This  was also
 the case with R&M  III houses at the  twelve month campaign. These
 findings may be due to a reduced  rate of input of  lead into dust
 from high lead-content paint sources during the one-year period of
 follow-up.   Given a reduced rate of high lead input, housekeeping
 by   families  might then  further  remove  some  of   the remaining
 reservoir of highly concentrated  lead dust.   Such  dust may have
 accumulated in R&M houses  during the  last fifty  years or more since
 the  time they were built.   A rise in the lead concentration (Mg/g)
 of   settled  house  dust  in  future   campaigns  might   signal  the
 presence  of  new lead  paint  hazards  and  the  need  for  further
 remediation  activities.
     The  patterns  observed  in dust  loadings  and  concentrations
 across R&M groups (highest in R&M I,  lowest in R&M III) may be due
 to  the  degree to which smooth and easily cleanable surfaces were
 provided as part of the interventions. The provision of smooth and
 easily cleanable surfaces  has been  shown to be an important element
 of  effective residential  lead paint  abatement.21  In  this study,
 surface conditions  would  have influenced the effectiveness of the
post-R&M cleanup by contractors as  well as the housekeeping of the
resident  families.
     One  year following  intervention,  R&M  III  was the only R&M
group  to  have dust lead  loadings  on  a geometric mean basis  that
were  less than  the current  EPA  guidance  levels  of  100 /^g/ft2
 (floors) ,  500 M9/ft2 (window sills), and 800 nq/ft2  (window wells).
The twelve month GM PbD levels in the  R&M groups for floors, window

                                36

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 sills and  window  wells,  respectively  were 112,  603,  and  19412
 jug/ft2 for R&M  I;  79,  201,  and 1761 /ig/ft2  for  R&M II;  and 36, 23,
 and 268 jttg/ft2 for R&M  III.  It should be noted that although the
 R&M cyclone  has been shown to produce higher estimates of dust lead
 loadings compared  to  wipes across a range  of surface types and
 conditions,   the  cyclone tends to  yield lower  PbD estimates  on
 smooth surfaces with  low lead  loadings  (<~100  /xg/ft2) as  compared
 with wipe sampling.6
 Control  groups
      The modern urban and previously abated control houses  were
 characterized  by  the  relative  stability of lead loadings,  lead
 concentration  and dust loadings  over  time  (Tables  5-7) .   The
 downward trend found  in dust  loadings and lead loadings  in  these
 houses may be attributable in  some part to  (a) families  becoming
 more aware of the  importance  of lead dust control as a result of
 their study  participation and  (b)  the fact that dust was  removed
 from their homes  by  the process of  environmental sampling.  This
 applies  to the intervention groups as well. Moreover,  this study
 was  not designed to assess the specific effects  of  education on the
 study outcomes across groups  (e.g.  study has  no education  only
 control  group)  but rather to  investigate the effectiveness of the
 R&M  interventions  as  a whole  which included the  provision of
 information  to  all families as explained in section 2.0.
      The  modern urban study  houses are  all  located  in  housing
 subdivisions built after 1978  and were presumed  to  be free of lead-
 based paint.  This presumption  is supported by the consistently low
 overall  interior dust lead concentrations  (GM *300 jug/g (ppm),
 equivalent to 0.03%) and low soil lead concentrations  (GM*75 /xg/g).
At each campaign in which they  were  tested, the  modern urban houses
had  significantly  lower  lead  loadings  and  concentrations as
compared  with each  of the other study  groups.  Dust loadings in the
modern urban houses were  significantly  lower  than those  in the
other study groups  at  the initial campaign only.

                                37

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
      In the previously abated group, geometric mean  lead  loadings
 and  concentrations tended to be  bracketed by the levels  found in
 R&M  II and III houses at  the six and twelve month campaigns. The
 latter campaigns were conducted  approximately four to five  years
 after the  houses  had been abated.  As was illustrated  by  the case
 in which a child's blood  lead  rose  to 53 /ig/dL during follow-up,
 the  previously abated houses  were comprehensively  but not  fully
 abated of  lead paint.   Some interior (basement) surfaces  that had
 not  been treated originally due  to  resource limitations  and some
 exterior surfaces that had been  stabilized as part of the  original
 abatement  were found to  be in deteriorated condition and  a likely
 source of  this child's exposure  along with deteriorated  exterior
 paint  observed on neighboring houses.   This case also  points to the
 need for ongoing inspection and  maintenance of houses that receive
 any type of interim  control or  partial abatement intervention and
 the need for long-term follow-up of  study houses to assess  long-
 term effectiveness of  the  R&M interventions.
 Soil and Drinking Water
     Due to low concentrations,  drinking water was not found to be
 an  important  source  of   lead  exposure  in  study children.  The
 geometric  mean water  lead  concentrations across study groups were
 all ฃ4 M9/L (ppb) at the  initial  and six month campaigns.   Only a
 small  number of readings exceeded the EPA drinking water standard
 of 15  jug/L.. The maximum reading was  44 Aig/L.
     Soil  lead data  were  limited due to  the absence of drip-line
 soil at most study houses  except for the modern urban houses.   Soil
 lead concentrations ranged  from  30 to 16,000 Mg/g across houses at
the initial and six month campaigns.   GM levels ranged from 700 to
730 ng/g in R&M I  and II houses  at the six month campaign (Appendix
A, Table g).  The  low levels  (GMs  of 63 to 73 /xg/g) measured in the
drip-line  soil  next  to the modern urban  houses over time  suggest
that replacement sod or soil might have been used at these houses
at,  or since, the time of  construction.

                                38

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data   7/95
 Blood lead
      The majority of U.S.  children with  elevated blood lead levels
 as defined by the U.S.  CDC (*10 nq/dL), have levels in the range of
 10 to 20 ng/dL-22  Little  is known about PbB changes associated with
 lead  paint  hazard  remediation  interventions in  the  homes  of
 children with this low-to-moderate level of lead toxicity.  Most of
 the children  in  the three  R&M groups had baseline PbBs in the range
 of 5 to  20 /jg/dL.  They  were recruited from households  residing in,
 or  moving into,   eligible study  houses  identified  from  lists
 provided  by  collaborating property  owners  who manage  low-income
 rental  property  in minority  neighborhoods  in  Baltimore  City.
 Children in these  types of  neighborhoods are at  known  high risk of
 lead poisoning.
      All  study  children  are African Americans living in low-to-
 moderate income  housing.  At the outset, their geometric  mean (GM)
 ages across  the  five  study groups  ranged  from  25  months to  33
 months.   Their baseline GM  PbB values across groups ranged from 10
 to 14.5  /*g/dL. It  can be noted here that in U.S.  non-Hispanic black
 children  from  low-income  families  living  in  central  cities
 (populations   *1 million,  1988-1991),  the  GM  PbB  was  recently
 estimated at  9.7 /ng/dL.22
      The R&M  study's maximum baseline PbB was in a child  who moved
 into a vacant R&M Level III house with a PbB reading of  42  nq/dL.
As  anticipated, the largest numbers of children with baseline  PbBs
>20ng/dL were in the R&M II and III groups.  This was due the  fact
that one of the  housing organizations had a  policy of renting  its
improved properties to  families with lead poisoned children.
      Children in the three R&M groups with PbB <20  /xg/dL at the
outset tended to remain in that range throughout the twelve month
period  of  follow-up.  After controlling for  age  and  season,  an
increasingly downward trend in  PbB level was  noted across  the two,
six,  and twelve  month  campaigns  in  the R&M III group,  the group
that  also  had  the lowest house  dust  lead   levels  over  time.

                                39

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 Furthermore, no statistically significant PbB changes were found in
 any  of  the R&M groups in children with baseline PbB  <20 /xg/dL.  One
 could hypothesize that the R&M interventions prevented  an increase
 in blood  lead that study children might have experienced  otherwise
 in the  absence of the R&M interventions.  For ethical reasons,  the
 study  design  did not  include  a  non-intervention  control  group
 against which  to  test this hypothesis.
     R&M   children  with  pre-intervention  PbBs  *20  ng/dL  did
 experience statistically significant reductions in blood  lead over
 time to levels below 20 nq/dL  at  twelve months.  Swindell et.  al.
 recently  reported that blood lead  levels also  declined up to  one
 year post-abatement in children with baseline PbBs ^20 /ig/dL  among
 moderately lead poisoned children in central  Massachusetts.23   Their
 study  was based  on  a  retrospective  record review and  did  not
 include the  collection of environmental  lead data pre- and  post-
 abatement.  Our finding of  an  absence  of a blood  lead  rise in
 children with initial PbBs <20 M9/dL differ from those of Swindell
 et. al. who reported an overall rise in PbB from 16.8  ^g/dL to 19.3
 jig/dL  at  up  to  one  year  post-abatement in  children with pre-
 abatement PbB <20  jitg/dL.23
     A  noteworthy  finding in  the R&M  study is  the  absence of
 statistically significant  increases in children's PbB measured at
 two months post-intervention,  especially  in the R&M I  and II houses
 that  were occupied at  the  time   of  treatment.     This  finding,
 together with the  dust  lead  data,  provides some evidence that the
 precautions  used  in R&M  houses  were  successful  in  preventing
 significant short-term  increases in children's  PbB that have been
 attributed  to improper  abatement  practices in past studies.24'25
 Precautions  included having  children  out of the  house while R&M
work was  in  progress,  and the use  of  work practices to minimize,
contain, and cleanup lead  dust.
                                40

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 Control groups
      The findings for children with PbBs <20 /ig/dL and *20 jug/dL in
 the previously abated group were similar to those described  above
 for children in the R&M groups. In contrast to this group and  the
 three R&M groups,  the GM  PbBs  across time  (range:3-4 jiig/dL)  in
 children living in the modern urban  study houses were similar to
 the national GM of 3.6 /xg/dL reported for U.S. children aged  12 to
 60  months.22  The  GM estimated  for  all  U.S.  non-Hispanic  black
 children in  this age range was 5.6 /xg/dL.22  Unlike the other four
 study groups,  nearly all of the modern urban houses were  occupied
 by  owner occupants.  Also, as reported previously, the mean monthly
 mortgage/rent  payment was higher ($406) in this group as  compared
 with the other groups (range of means:$288 to $330).2
      Additionally,  all but one of the PbB readings in the  modern
 urban group  across time were below the CDC's level of concern  (10
 /xg/dL) . One child had a PbB increase from 3 to 10 jug/dL at the  six
 month campaign which may have contributed  to the significant  PbB
 rise  found between  the initial and  six  month campaigns  in this
 group.  The PbB levels  in children  in the  modern urban group were
 significantly lower than those in the  children  in each of the  other
 four  study groups at the initial,  six and twelve month campaigns.
 This  finding is the only statistically significant difference in
 PbB  found in  the R&M  study to  date among  children who  had an
 initial PbB <20 /xg/dL.
 Exposure Model
     Using data from all  five  study groups,  the  exposure models
 (longitudinal data analysis)  indicated that the blood lead levels
 of  study  children were significantly related  to  their  house dust
 lead  loadings  and  lead  concentrations  after  controlling  for
 significant  fixed effects,  i.e.,  age,  (age2),  and  season.  The
guadratic  term indicates a  nonlinear relationship between  blood
 lead and children's age.  The  absence of a statistically significant
relationship between dust lead  and blood  lead  in the three R&M

                                41

-------
                                                 Preliminary RsM Study Report-
                                                 Partial 12 Month Data  7/95
 group exposure model was likely due to the narrower range of dust
 lead levels  over  time  post-intervention  as  compared with  pre-
 intervention and the absence  of  the low-lead modern urban houses
 from the analysis.  Other studies,  including the recent  study  in
 Rochester,26 have found  a  relationship between lead in children's
 blood and settled dust in their homes.
      Based  on  the various longitudinal data analyses performed, the
 seasonal  change in children's blood lead levels was estimated to  be
 +1.3 /ig/dL  in summer  relative to  winter.   Others have  reported
 seasonal  trends in children's blood lead levels for different years
 and populations  which  vary  in  the  estimated magnitude  of the
 seasonal  difference (unpublished review, PL  Reagan, July  1992) .
 Considerations in interpretation of PbB findings
      A  number of  factors that can  mediate  a  child's  blood  lead
 response  to an intervention, including age, degree  of hand-to-mouth
 activity, total cumulative body lead burden,   the timing  of  the PbB
 measurements,  and neighborhood housing characteristics  need to be
 considered  in  the interpretation of blood lead findings from this
 or  any other intervention study.   For this  study, these  factors are
 of  particular relevance with  regard to the  following  blood  lead
 findings:

 •    The  lack  of  statistically significant PbB changes within and
     across  R&M groups  in children with  baseline  PbB <20  /zg/dL
     despite  significant  differences in  house dust  lead  levels
     between R&M  groups.
 •    The consistently downward, but not statistically significant,
     trend in PbB over time post-intervention  in the R&M III  group.
 •    The significantly lower PbBs  in modern urban children relative
     to all  other groups,  in  particular R&M  III  where dust  lead
     exposures were  most similar  to those in  modern urban houses.

     At enrollment,  the mean ages of  the  children in  the R&M and
previously abated  groups  were in the range of  25 to 33  months.  On
average,  these children were past the age range at which blood  lead
                                42

-------
                                                 Preliminary RtM Study Report-
                                                 Partial 12 Month Data  7/95
 levels  tend  to peak  (i.e.,  18-24  months of  age).27 Given  their
 reported  housing  histories,   it  is  likely that  the  R&M  study
 children spent  most  or all of  their young lives prior to enrollment,
 in low-income  housing  with potentially  high lead dust and  paint
 exposures. For the modern urban group, the blood lead  and  housing
 history data suggest that the  children were  at  lower risk  of high
 lead exposure  prior to enrollment  compared  to the other  groups.
 Most children in the modern urban group were  residents  of the same
 house since birth.
      Assuming that children  in  the R&M and previously abated houses
 had chronically elevated PbBs prior to enrollment, additional time
 beyond twelve  months post-intervention may  be  needed to  measure
 significant PbB changes in R&M children for  the reasons mentioned
 briefly  below.  Lead in blood reflects a mixture of recent exposure
 and body stores of  lead. Most (approximately 70%)  of the  lead in
 children is stored in bone.28 The half-life of lead in human  adult
 cortical bone  is estimated to  be 20 years.29 Skeletal lead can be
 an  ongoing internal source  of lead measured in blood even  after
 external  exposure  and  children's   lead  ingestion  are   reduced
 following lead remediation interventions.   Unfortunately, the bone
 lead concentrations of study children are unknown and the kinetics
 of  lead mobilization from bone is not well understood in children.
 This makes  it difficult  to  estimate the magnitude and duration of
 bone lead's contribution to children's blood lead measured in the
 post-intervention  phase  of  this study.
     In  an  earlier study in Baltimore,  children with GM PbBs  of 63
 Mg/dL prior to receiving inpatient chelation therapy were monitored
 after discharge to "lead free" public housing and  housing abated
 according  to  local  ordinances in  effect prior  to 1982.   These
 children had post-discharge  GM PbBs  at one month that changed very
 little  prior  to 24  to  30  months  post-discharge.30  This  earlier
 study highlights  the need  to  continue the   investigation  of PbB
changes  in  the  R&M  Study  at  the eighteen and  twenty-four month

                                43

-------
                                                 Preliminary RsM Study Report-
                                                 Partial 12 Month Data  7/95
 campaigns.
      At the twelve month campaign, most study children were 36 to
 48  months of age,  an age range in which the frequency  of  children
 engaging in mouthing  behavior is  likely  to be  less compared to
 younger children.  Since hand-to-mouth activity is recognized as  a
 a  major  route  of  entry of  lead  into pre-school  children,27  less
 frequent hand-to-mouth activity over time might account  in  part for
 the lack of statistically significant blood lead changes within and
 between R&M groups in children with baseline PbB <20  jug/dL,  despite
 the differences in  dust lead exposure  between and within  groups
 over time. To date, hand-to-mouth  activity  has not been  found to be
 a significant  covariate in the blood lead exposure models.   This
 may be  due  in part to  the  more or less truncated PbB distribution
 and relatively  small sample sizes.
     The small number of children with initial PbB  *20 /ig/dL at the
 outset  may  have  had higher blood  lead levels due to more  frequent
 hand-to-mouth activity.  Therefore, they may have had a relatively
 greater  contribution  to  their  blood  lead from   their  current
 exposure versus bone lead  compared to the children with PbB  <20
 /xg/dL. Thus, their PbB  levels may  have been more responsive  to the
 reduction in lead exposure  associated with  the R&M interventions as
 compared to children with  lower baseline PbB levels.
     Neighborhood  housing  characteristics  may have contributed to
 the observed PbB differences between the  modern urban group and the
 other four study groups.  By design, the modern  urban study  houses
were  all located  in housing subdivisions  built  after  1978  and
 presumably  free of  lead-based paint.   As mentioned above,  this
presumption  is  supported  by the  very low  dust  and  soil  lead
 concentrations  found in the modern urban  study houses.  In addition
to  having low  exposures to lead  in their  own house  dust and  soil
 (up to  more than two orders  of magnitude  lower than those  in  the
other study groups) ,  children  in the modern urban  houses also  have
fewer opportunities to  be exposed  to high lead  sources  in, or

                                44

-------
                                                  Preliminary RCM Study Reporc-
                                                  Partial 12 Month Data   7/95
 originating  from,  neighboring houses  by virtue of  the  fact that
 they are located in subdivisions  with  similar low lead housing.
      On the other hand,  children in  each  of the other study groups,
 including   the   previously   abated  group,    share   a   common
 characteristic of being  located in  neighborhoods with  older lead-
 painted  rowhousing  built  before  1940  and  in  generally  poor
 condition. These  neighborhoods,  and the blocks in which  the R&M
 houses are  located,  often have abandoned and boarded housing as
 well.  Compared to the modern urban  group, children in  the  R&M and
 previously abated groups have had more opportunities for exposure
 to high lead  sources  in paint and  dust  from surrounding  houses.
 Neighboring houses can be sources of lead particles and  paint chips
 which can be  tracked  or blown inside  study  houses. This  was the
 rationale for  including walk-off mats at  the main entryways in R&M
 houses.   Neighboring houses can also be  sites where children are
 directly exposed to lead during visits and outside play  activities.
     The  study sample  sizes of 20 to 26  children per group across
 campaigns limits the  degree to which  small  proportional  changes
 over time in PbB can be detected with high statistical power.   The
 group differences in PbB between the modern urban  group and  each of
 the  other  four  study  groups  were  found   to   be statistically
 significant  (differences on the In  scale of  approximately  0.8  or
 ratio of approximately 2:1), whereas the smaller  log differences  in
 blood  lead  levels over time in the R&M  III  group  were not  large
 enough to be statistically significant (alpha s.05) in this  study.
     In closing, it should be emphasized that by their nature, the
 three R&M interventions under investigation are interim control  or
 partial abatement approaches to reducing  lead hazards in housing.
 As such they  are not expected to  be long-lasting or permanent. A
major study objective remains the determination of the longevity  of
the R&M  interventions. To  date,  dust  lead  loadings at  specific
sites in individual study houses  (particularly R&M I houses) have
reaccumulated  to levels close to pre-intervention  levels.  However,
                                45

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
none  of the  R&M interventions  in individual  houses through the
twelve month campaign have failed on a wholesale basis,  whereby all
or most of  the interior dust lead samples have had  lead  levels at
or above pre-intervention levels. If wholesale failures  do occur,
we  will use contingency funds  to  make  additional  repairs  and
reclean houses  as needed.
     Lastly, it is recognized that the findings reported herein are
preliminary in nature and that the results can change as  more  data
from  the twelve month and subsequent campaigns become available.
We anticipate that  data from  the twelve  month  campaign in  R&M
houses will be complete and available for inclusion in the eighteen
month preliminary report scheduled for  February of 1996.   Future
campaigns will allow us  to  investigate the longer-term changes in
lead in settled  house dust and children's  blood and move  towards
the overall research goal of  contributing to the scientific basis
for a standard of care for lead-painted houses.  Also, it  is hoped
that future campaigns will provide sufficient longitudinal  data on
children born into study  houses  to allow  for a  separate analysis of
a subgroup  which has the  potential to  increase  our  understanding of
the role of R&M  intervention in  the primary prevention of  lead
poisoning in children.
                                46

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95
 7.0  REFERENCES
 1.    "Quality  Assurance  Project Plan  for  the  Kennedy  Krieger
      Institute Lead  Paint Abatement and Repair and  Maintenance
      Study in Baltimore"   November 1992. (Battelle Subcontract No.
      41950(2348)-2207;  EPA Contract No.  68-DO-0126)  Office  of
      Pollution Prevention  and  Toxics,   Design  and  Development
      Branch,  Washington DC.

 2.    "Draft Final Report - Descriptive Statistics from  the  Initial
      Sampling Campaign  of the Lead  Paint Abatement and Repair  &
      Maintenance Study in Baltimore,"  Submitted by Kennedy  Krieger
      Research Institute, December 6,   1994,  (under EPA Contract No.
      68-D4-0001) to the Office of Pollution Prevention  and  Toxics,
      Design and Development Branch,  Washington DC.

 3.    U.S.  Agency  for Toxic  Substances and Disease  Registry.  "The
      Nature and Extent of Lead Poisoning in United States Children:
      A Report to Congress,"  Washington, DC  1988.

 4.    Farfel MR,  Bannon D,  Lees  PSJ,  Lim BS and  Rohde CA.   1994.
      Comparison of Two  Cyclone-Based Collection  Devices for  the
      Evaluation of Lead-Containing Residential Dusts.  Journal of
      Industrial and Occupational Hygiene.  9:212-217.

 5.    Farfel MR, Bannon D, Chisolm JJ  Jr,  Lees PSJ,  Lim BS  and Rohde
      CA. 1994.  Comparison of a Wipe  and a  Vacuum Collection Method
      for   the  Determination   of  Lead   in   Residential  Dusts.
      Environmental Research,  65:291-301.

 6.    Farfel MR,  Lees PSJ,  Rohde CA,  Lim  BS  and Bannon D. 1994.
      Comparison of Wipe  and Cyclone  Methods  for the Determination
      of  Lead  in  Residential  Dusts.    Applied  Occupational &
      Environmental Hygienet  9:1006-1012.

7.   Research Triangle Institute and  Engineering Plus. "Development
     of a  High Volume  Small Surface Sampler  for Pesticides   and
     Toxics in House Dust  -  Final Report"   Submitted  to  US   EPA
     Exposure  Assessment Research Division. Research Triangle Park,
     NC (June 29,  1990) .
                                47

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data  7/95


 8.    Bannon DI, Murashchik C,  Zapf  CR,  Farfel MR and  Chisolm, JJ
      Jr.    1994.   A Graphite  Furnace AAS  Method  of  Blood  Lead
      Measurement  Using  Matrix   Matched  Standards.     Clinical
      Chemistry. 40:1730-1734.

 9.    Hornung RW, Reed LD. 1990. Estimation of Average Concentration
      in the Presence of Nondetectable Values. Applied Occupational
      &  Environmental Hygienef 5(1):46-51.

 10.   S-PLUSฎ Version 3.2 for Windows, March 1993, "Read Me  First"
      StatSci,  A Division of Microsoft, Inc., Seattle,  Washington.

 11.   Laird  Nan,   Ware   James.   1982.  Random-Effects  Model  for
      Longitudinal  Data.   Biometrics 38:963-974.

 12.   Zeger  SL, Liang K-Y (1986).   Longitudinal Data Analysis for
      Discrete  and  Continuous Outcomes.  Biometrics 42:121-130.

 13.   Zeger  SL,  Liang K-Y, Albert PS.  1988. Models for Longitudinal
      Data:  A  Generalized Estimating Equation Approach.  Biometrics
      44:1049-1060.

 14.   Waternaux  C, Laird N, Ware J.  1989.   Methods for the Analysis
      of Longitudinal Data: Blood Lead Concentrations and Cognitive
      Development.  J  Amer Stat Assoc 84:33-41.

 15.   Liang  K-Y,  Zeger SL. 1986.   Longitudinal Data Analysis Using
      Generalized Linear  Models.   Biometrics  73:13-22.

 16.   Moulton LH,  Zeger  SL.  1989.  Analyzing Repeated Measures on
      Generalized Linear  Models via  the Bootstrap.  Biometrics. 45:
      381-394.

 17.   Royall RM.  1986.    Model Robust   Inference  using  Maximum
      Likelihood Estimators.   Internatl Stat  Rev 54:221-226.

 18.  GEE: A SAS Macro for  Longitudinal  Data Analysis. Technical
     Report #674  Department  of  Biostatistics,  The  Johns  Hopkins
     University.

19.  Tukey  JW.  1977. Exploratory  Data Analysis.  Addison-Wesley,
     Reading,  Massachusetts.
                                48

-------
                                                 Preliminary RSM study Report-
                                                 Partial 12 Month Data  7/95


 20.   Sayre JW, Katzel MD.  1979.  Household surface lead dust:  its
      accumulation  in  vacant  homes.   Environ.  Health Perspect.
      29:179-182.

 21.   Farfel MR,  Chisolra JJ  Jr,  Rohde CA.  1994.  The  longer-term
      effectiveness of residential  lead paint abatement.  Environ.
      Research 66:217-221.

 22.   Brody DJ,  Pirkle JL, Kramer RA, et al. 1994.  Blood lead  levels
      in the US  population: Phase 1 of the third national health  and
      nutrition  examination survey (NHANES III, 1988 to  1991). JAMA
      272:277-283.

 23.   Swindell  SL,  Charney  E,  Brown  MJ,  Delaney  J.  1994. Home
      Abatement and Blood Lead  Changes  in Children with Class  III
      Lead  Poisoning.  Clin.  Pediatrics  Sept.1994: 536-541.

 24.   Rey-Alvarez  S and  Menke-Hargrave T. 1987. Deleading dilemma:
      pitfall  in   the  management   of   childhood  lead  poisoning.
      Pediatrics 79:214-217.

 25.   Farfel MR and Chisolm  JJ.  1990.  Health and  environmental
      outcomes of  traditional and modified practices for abatement
      of  residential lead-based paint.  Am.  J. Public Health  80:1240-
      1245.

 26.   Lanphear B.   et.  al. The  Relationship  of  Lead-Contaminated
      House Dust and Blood-Lead Level Among Urban  Children.  Final
      Report to  the National  Center  for Lead-Safe Housing,  1994.

 27.   U.S. Centers for Disease Control.  Preventing Lead Poisoning in
      Young Children; A Statement by  the  Centers for Disease Control
      - October  1991. USDHHS/PHS/CDC Atlanta, Georgia.

 28.   Barry,  PSI.   1981.  Concentrations  of  lead  in  tissues  of
      children.  Br.  J. Industrial Med.  316:1037-1043.

29.  Rabinowitz  MB,  Kopple  JD,  Whetherhill GW.  1976.  Kinetic
     analysis  of  lead mobilization in  healthy  humans. J.  Clin.
     Invest. 58:260.

30.  Chisolm  JJ,   Mellits  ED,  Quaskey.  1985.  The  relationship
     between the level of lead absorption in children  and the age,
     type,  and condition  of housing.  Environ.  Research 38:31-45.

                                49

-------
                                                            Preliminary RCM Study Report-
                                                            Partial 12 Month Data   7/95
Table  l:   Data collection  plan*
Study
Group

R&M I




R&M II




R&M III




Previously
Abated



Modern
Urban


Pre-R&M/
Initial
B
D
S
w
Q
B
D
S
W
Q
B
D
S
-
Q
B
D
S
W
Q
B
D
S
W
Q
Post-
R&M
-
D
S
-
"
-
D
S
-
"
-
D
S
W
"

n/a




n/a


2
Month
B
D
-
-
™
B
D
-
-
"
B
D
-
-
™
-
-
-
"
-
-
-
—

6
Month
B
D
S
w
Q
B
D
S
W
Q
B
D
S
W
Q
B
D
S
W
Q
B
D
S
W
Q
12
Month
B
D
-
-
Q
B
D
-
-
Q
B
D
-
-
Q
B
D
-
Q
B
D
-
-
Q
18
Month
B
D
S
W
Q
B
D
S
W
Q
B
D
S
W
Q
B
D
S
W
Q
B
D
S
W
Q
24
Month
B
D
_
_
Q
B
D
_
_
Q
B
D
_
_
Q
B
D
-
Q
B
D
-
-
Q
  B=Blood; D=Dust; S=Soil; W=Water;  Q=Questionnaire
                                      50

-------
                                                           Preliminary RsM Study Report-
                                                           Partial 12 Month Data   7/95

 Table  2:   Types  and numbers  of 12-month campaign samples  collected
              and analyzed for lead  (not including  field  QC)
SAMPLE TYPE
SBTTLKD DOST:
Perimeter Floor
Composite-- room*
with window.
Ferinater Floor
Composite — room
without window*
Window Sill Ccopoaita
Window Wall ConpoBite
Interior Bntryway
Exterior Bntryway
Air Duct/
Upholstery
Total Dust par
Dwelling
TOTAL DOST

Soil Cora -
drip line
Drinking Mater*
Venous Blood -
JUKI 6 TO 60 MDl.
TOTAL
Planned
per House
2a
1
2
2
1
0
I9
9


0
0
1 per
child
10+
Collected in
89 Study
Bouse*
182b
45
177
174
89
0
53
35

755

0
0
109
864
Chemically
Analyzed
for Lead
182
45
177
174
89
0
53
35

755

0
0
109
864
Not
Collected
in the 89
Bouses
0
44C
ld
4e
0
Of
lh

50

Of
Of
51
55
* One composite  sample was obtained per story.
" Includes  4  samples  collected  from basements used as living spaces  in 3  RSM I
  houses and 1 Modern Urban  house.
c 44  houses did  not have rooms without windows.
d Sills  on one story were inaccessible in 1 RSM I  house.
• Wells  on one story  were inaccessible in 2 R&M I houses and on both stories in  1
  RSM I house,  for a total  of 4 samples  not collected, primarily due to plastic
  being sealed over  the  windows  in winter.
' This sample type was not part of the 12-month campaign.
9 Upholstery sample was collected if air  duct sample  could not be obtained.
h Air duct & upholstery were inaccessible/not present  in 1 RSM II house.
1 3 Modern Urban children had moved at the time of blood collection,  and  2 Modern
  Urban  children were missed by  the Primary Care Physician.
                                       51

-------
                                                     Preliminary RSM Study Report-
                                                     Partial 12 Month Data   7/95
Table 3:  Types and numbers  of 12-month campaign samples  collected
           by  study group (not  including field QC)
SAMPLE TYPE
SETTLED DUST:
Perimeter Floor
Composite — room*
with window*
Perimeter Floor
Composite—rooms
without window*
Window Sill
Composite
Window Nell
Composite
Interior Batryway
Exterior Entryway
Air Duct
Upholstery

TOTAL DOST

Soil Cora
- drip lino
Drinking Mater
Venou* Blood -
agea 6 to 60 no*.
Collected
in IS
Modern
Urban
Bouses
31
4
30
30
15
0
11
4

125

0
0
14
Collected
in 14
Previously
Abated
Houses
28
6
28
28
14
0
5
9

118

0
0
24
Collected
in 20
Rat i
House*
43
15
39
36
20
0
8
12

173

0
0
21
Collected
in 20
IUM II
Houses
40
11
40
40
20
0
14
5

170

0
0
26
Collected
in 20
RCM III
Bouses
40
9
40
40
20
0
15
5

169

0
0
24
                                  52

-------
                                                      Preliminary RSM Study Report-
                                                      Partial 12 Month Data   7/95
Table 4:  Frequency of family moves and reoccupancies between the
           initial and the  twelve month campaigns  among the 60 R&M
           houses  included  in  this  preliminary  report  and  the
           control houses.
STUDY GROUP
R&M I
(20 houses)
R&M II
(20 houses)
R&M III
(20 houses)
Modern Urban
(15 houses )b
Previously Abated
(15 houses )b
Total
Moved
No.
Families
6
3
4
1
1
15
No.
Children3
12
5
6
3
1
27
Replaced
No.
Families
6
3
4
1
0
14
No.
Children
1
2
6
0
0
9
'  Includes children who moved although family did not.
b  One extra house was included for a  total of  16 houses.  Of  these one household
  withdrew from the study before the  6 month campaign leaving 15 houses.
                                    53

-------
                                                                 Preliminary RIM Study Report-
                                                                 Partial 12 Month Data   7/95
 Table  5:  Overall dust  lead  loadings* (f/g/ft2)  for houses completing the 12-month  campaign
Study
Group
(n)
Modern
Urban
(15)
Previously
Abated
(14)
R&M
Level I
(20)
R&M
Level II
(20)b
R&M
Level III
Pre-
Intervention/
Initial Campaign
GM
min max
90
10 540
890
120 5420
16200
1440 70700
23800
3100 124000
39300
3260 127000
Post-
Intervention/
R&M Campaign
GM
min max
N/A
N/A
1490
280 13100
300
3 4910
90
7 3760
Two Month
Campaign
GM
min max
N/A
N/A
3420
560 11200
1440
160 13560
220
40 1140
Six Month
Campaign
GM
min max
70
6 750
470
50 2970
4020
950 19100
1230
40 14000
180
10 1610
Twelve Month
Campaign
GM
min max
60
10 260
630
100 5300
4140
590 22600
840
40 24900
160
30 3050
'      Based on weighted  averages of floor, entryway,  window sill, and window well samples within  houses.
      Eighteen houses  were tested in the two month campaign.
                                                        54

-------
                                                                Preliminary RtM Study Report-
                                                                Partial 12 Month Data   7/95
Table 6:   Overall dust lead  concentrations*  C^g/g)  for houses completing the  12-month campaign
Study
Group
(n)
Modern
Urban
(15)
Previously
Abated
(14)
R&M
Level I
(20)
R&M
Level II
(20)b
R&M
Level III
(20)b
Pre-
Intervention/
Initial Campaign
GM
min max
260
90 510
2360
430 16400
20500
3300 106000
13900
1240 65000
16800
1440 64800
Post-
Intervention/
R&M Campaign
GM
min max
N/A
N/A
8300
730 66500
6000
480 38400
2800
830 47100
Two Month
Campaign
GM
min max
N/A
N/A
13400
3800 34900
6200
700 53300
1500
500 6700
Six Month
Campaign
GM
min max
300
100 830
1900
180 18000
12700
2400 50900
4700
500 45500
1400
360 12600
Twelve Month
Campaign
GM
min max
300
150 1500
2800
500 15600
12900
2300 54400
3400
300 36900
1100
200 16300
      Based on weighted averages  of  floor, entryway,  window sill,  and window well samples within houses
      Eighteen houses were tested in the  two month campaign.
                                                       55

-------
                                                             Preliminary RSM Study Report-
                                                             Partial 12 Month Data   7/95
Table 7:   Overall dust loadings*  (mg/ft2) for houses completing the  12-month campaign
Study
Group
(n)
Modern
Urban
(15)
Previously
Abated
(14)
R&M
Level I
(20)
R&M
Level II
(20)b
R&M
Level III
Pre-
Intervention/
Initial Campaign
6M
min max
330
50 1690
380
160 1340
790
300 3360
1710
350 7920
2340
970 6570
Post-
Intervention/
R&M Campaign
GM
min max
N/A
N/A
180
50 1530
50
10 260
30
10 710
Two Month
Campaign
GM
min max
N/A
N/A
260
100 490
230
100 1090
140
70 640
Six Month
Campaign
GM
min max
240
60 2950
250
100 830
320
140 1330
260
50 1230
130
30 300
Twelve Month
Campaign
GM
rain max
200
90 680
220
50 620
320
100 1340
250
40 930
140
40 320
  Based  on weighted averages of floor,  entryway,  window sill, and window well samples  within houses
  Eighteen houses tested included in the two month  campaign.
                                                        56

-------
                                                    Preliminary RcM Study Report-
                                                    Partial 12 Month Data   7/95
Table  8:
Descriptive Statistics  for Blood Lead  Concentrations (PbB //g/dL)  for
Children  with Initial PbB < 20 //g/dL

STUDY GROUP
R&M Level I
R&M Level II
R&M Level III
Previously
Abated*
Modern Urban


GM
range
(n)
GM
range
(n)
GM
range
(n)
GM
range
(n)
GM
range
(n)
CAMPAIGN
INITIAL
8
2 to 16
(24)
10
3 to 19
(20)
11
2 to 19
(19)
12
6 to 18
(18)
3
1 to 6
(18)
2 MONTH
8
1 to 16
(22)
10
4 to 20
(15)
11
3 to 25
(16)
N/A
N/A
6 MONTH
9
2 to 17
(19)
12
5 to 21
(16)
10
3 to 23
(18)
13
5 to 32
(17)
4
2 to 10
(16)
12 MONTH
8
2 to 20
(15)
10
5 to 18
(15)
10
6 to 16
(14)
12
7 to 21
(16)
3
2 to 6
(14)
   * Excludes child requiring chelation therapy during follow-up.
                                                57

-------
                                                Preliminary RซM Study Report-
                                                Partial 12 Month Data  1/95
Table 9:  Twelve month campaign correlations of In (lead loading). In (lead concentration),
           and In (dust  loading) with  In (blood ^g/dL) for the youngest child per household.
Pearson  correlation coefficients / Prob >  |R|  under Ho: Rho=0  /  Number of observations

DUST
VARIABLE

In (lead loading)
A/9/ft2

In (lead cone)
/*g/g

In (dust loading)
mg/ft2
Surface Type

r
P
n
r
P
n
r
P
n
Floor
.32
< .01
79
.45
< .01
79
.01
NS
79
Window
Sill
.22
.05
79
.22
.05
79
.13
NS
79
Window
Well
.22
.05
78
.29
.01
78
- .06
NS
78
Interior
Entryway
.15
NS
79
.31
< .01
79
- .003
NS
79
Floors - Rms
w/o Windows
.30
.05
41
.36
.02
41
.01
NS
41
Air
Duct
.31
.03
50
.37
< .01
50
.12
NS
50
Upholstery
.49
< .01
29
===^=^=
.61
< .01
29
.19
NS
29
NS =  Non-significant p-value>.05
                                            58

-------
                                                 Preliminary RSM Study Report-
                                                 Partial 12 Month Data   7/95
Table 10: Twelve month campaign  correlations of  In(lead loading),  In(lead concentration)
           and In (dust loading) with In (blood //g/dL)  for all children.


Pearson  correlation coefficients /  Prob  >  |R|  under Ho:  Rho=0 / Number  of observations

DUST
VARIABLE

In (lead loading)
MQ/ft2

In (lead cone)
M9/g

In (dust loading)
mg/ft2


r
P
n
i— i
r
P
n
r
P
n
—
Surface Type
Floor
.34
< .01
109
.45
< .01
109
< .01
NS
109
====d
Window
Sill
.22
.02
109
.23
.02
109
.10
NS
109
Window
Well
.22
.02
107
.27
.01
107
- .04
NS
107
Interior
Entryway
.11
NS
109
.34
< .01
109
- .01
NS
109
Floors -Rms
w/o Windows
.36
.01
55
.37
< .01
55
.02
NS
55
Air
Duct
.39
< .01
62
.46
< .01
62
.14
NS
62
Upholstery
.30
.04
47
.50
< .01
47
.09
NS
47
NS = Non-significant p-value>.05
                                            59

-------
                                         R&M Study: Preliminary 12 Month Report (7/95)
                    Figure 1   Dust Lead Loadings (ug/ftA2) across Campaigns for Floor Surfaces
                    RM-I
Q
.a
 o,
Q
.0
Q-
o,
               Previously-Abated
                                       RM-II
                                                                                                    RM-III
7 -
6 -
5
4
2
1 -
0 -
-1 -
r-r-i r^-i |
MUb 	 g— 	 g

IN PI 2M 6M 12M
7
6
5
4 •
3
2
1 -
0
-1 -
^-•fl — h h n
ug u 2
IN PI 2M 6M 12M
7 -
6 -
5
4
3
2
1 -
0 -
-1 •
Vn JH Pi '

IN PI 2M 6M 12M
KENNEDY KREGER RES
                                      Modern
                         7
                         6
                         5
                         4
                         3
                         2
                         1
                         0
                         -1 i
                                                 &
                             -&
                                                                                                                        O
                                                                                         H
                                                                                         H
                                                                                         C
                                                                                         H
                                                                                         m
          IN
6M
12M
IN
                                                            6M
12M
                                 data based on the 89 houses completing the twelve month campaiqn

-------
                               R&M Study: Preliminary 12 Month Report (7/95)
       Figure 2  Dust Lead Loadings (ug/ftA2) across Campaigns for Window Sill Surfaces
          RM-I
                                       RM-II
                                                                   RM-III



6
.ฃ)
CL
0
0)
o







Q
_Q
CL
O
^j>
0



7 •
6 -
5
A -
*t
3 -
2 -

1 -
0 -
-1


T
_L -T-
i i i ' y T"1 j . ^
S^sb rh n -Fi
M U U r_p
1 1 *-^J
! ' — '
^


IN PI 2M 6M 12M
Previously-Abated
7 -
6
5 -
4

3 •
2
1 -
0
-1


-r

Bn I3!
M g
"


                                   7
                                   6
                                   5
                                   4
                                   3
                                   2 ^
                                   1
                                   0
                                   -1
                                       IN  PI 2M
                                          6M
                                        12M
                        5
                        4 -
                        3
                        2
                        1
                        0
                                                Modern
IN
6M
12M
IN
                                                  -Q-
6M
                                                                7
                                                                6
                                                                5
                                                                4
                                                                3
                                                                2 -\
                                                                1
                                                                0 •
                                                                -1 -
                                                                                        m
                                                                                        m
                                                                                        o
                                                                                        rn
                                       IN  PI 2M
                                          6M
                                                                                                     12M
                                                                                                             w
                                                                                                             m
                                                                                                             O
                                                                                                             •nvซ
                                                                                                             CU.SJ
                                                                                                             c
                                                                                                             H
12M
                       data based on the 89 houses completing the twelve month campaign

-------
                               R&M Study:  Preliminary 12 Month Report (7/95)
      Figure 3   Dust Lead Loadings (ug/ftA2) across Campaigns for Window Well Surfaces
o
o
7
6
5
4
3
2
1
0
-1
          RM-I
     Previously-Abated
                     a
                                      RM-II
                                                                   RM-III
7 -
6 -
5 J
og10(PbD)
M CO -ฃ>.
1 '
0 -
-1 -

PI
Bv - ^
N—I— r1-! [-1-] i i
— a — tj LJ
IP -



7 -
6 -
5 -
4 -
3 -
2 -
1 -
0 •
-1 -

-T-
ง T- T-
\T n h n
^ " !



7 -
6
5 -
4
3 -
2
1 -
0
-1

j=
s
\ . *
\* n n R
ps/^r M y
-^J


_
m
z:
z
ru
3
3?
JS
rn
o
IN PI 2M 6M 12M IN PI 2M 6M 12M IN PI 2M 6M 12M
^J
ni
m
                                      Modern
5
4
3
2
1

-1
                                       B-
                                       -e
                                       -B
                                                                                                            o
                                                                     m
IN
6M
12M
IN
          6M
                                                                     12M
                       data based on the 89 houses completing the twelve month campaign

-------
                                         R&M Study: Preliminary 12 Month Report (7/95)
              Figure 4  Dust Lead Loadings (ug/ftA2) across Campaigns for Interior Entryway Surfaces
                    RM-I
                                                       RM-II
                                                                    RM-III
O)
o
0
.a
CD
O
 7
 6
 5
 4
 3

 1
 0
-1
7
6
5
4

2
1
0
-1
         B-e-f
         IN  PI 2M     6M
                           12M
              Previously-Abated
              7
              6
              5
              4
              3
              2 \
              1
              0
              -1 -I
              7
              e ^
              5
              4
              3
              2
              1
              0 -
              -1 -
                                   7 -
                                   6
                                   5
                                   4
                                   3
                                   2
                                   1
                                   0
                                   -1 ^
                  IN  PI 2M
             6M
           12M
                                                      Modern
9-
                                                                      -B
IN  PI 2M
6M
12M
K
                                                           rn
                                                           a
                                                           -<
rn
Q
m
po
20
m
c#
g
73
O
T
z
w
H
H
C
H
m
          IN
                6M
12M
IN
6M
                                                                       12M
                                 data based on the 89 houses completing the twelve month campaign

-------
                              R&M Study: Preliminary 12 Month Report (7/95)
        Figure 5  Dust Lead Concentrations (ug/g) across Campaigns for Floor Surfaces
          RM-I
                                      RM-II
                                                                   RM-III
6

5
9 4
Q
{ฃ 3
0
g, 2 -
1 -
0 -



~ n ^
B ^~-fฐ~i fl r~i
^ .L 2 H hd



IN PI 2M 6M 12M
6 -

5 -
4 -
3 -

2 -
1 -
0 -


m
5n -f- T ;
U tl — -pi— 	 n -"
" J^ : M ^P



IN PI 2M 6M 12M
6 -

5 -
4 J
3 •

2 •
1 -
0 •


^i-
n^T T :
: fj^CL 	 5 "^
: Sr M 	 Q



IN PI 2M 6M 12M
^
m
•z.
z
m
^
"v
ง
m
0
m
73
70
m
C0
5
5
Previously-Abated Modern O
X
6

5 •

9 4
Q
n
t 3 -
o
D) 2 -
1 -
0 -



*
i— •— i *
r-i .-i-.
— S
ซ_L_i •— *— '


6

5 -

4 -
3
2 -
1 -
0 -





-r
PI n rn
~ rn — 	 F— I


Z
Cfl
-^

H
c
H
m



IN
6M
12M
IN
6M
12M
                       data based on the 89 houses completing the twelve month campaign

-------
                              R&M Study: Preliminary 12 Month Report (7/95)


     Figure 6  Dust Lead Concentrations (ug/g)  across Campaigns for Window Sill Surfaces
          RM-I
                                      RM-II
                                                                  RM-III
o
Q
a
o
en
a

O
Q
.0
O
O)
0

6
5
4 -
3 -

2 -
1 -
0 -
6
5 -
4 -
3 •
2 -
1 -
0 -
F^ — Fh~PI n n
_J_ LJ [J ^J — LJ
•—• •-j-j j^ '-^



IN PI 2M 6M 12M
Previously-Abated

PI n 5
tpr— LJ M
^L.

                                  6



                                  5



                                  4



                                  3



                                  2 H


                                  1



                                  0 H
                                                  S
                                       IN  PI 2M
                                         6M
                                       12M
                                  6


                                  5


                                  4


                                  3


                                  2 -I


                                  1


                                  o H
                                               Modern
IN
6M
12M
IN
                                      -B-
6M
                                       -a
12M
                                                    6



                                                    5



                                                    4



                                                    3 -


                                                    2 -



                                                    1



                                                    0
                                                                             IN  PI 2M
                                                   6M
                                                                                         12M
                                                                                                 m
                                                                    m
                                                                    o
                                                                                                 m
                                                                                                 Q
                                                                                                 m
                                                                                                 5
                                                                                                 ^0
                                                                                                 O
                                                                                                 X
                                                                    c

                                                                    m
                      data based on the 89 houses completing the twelve month campaign

-------
                               R&M Study: Preliminary 12 Month Report (7/95)


     Figure 7  Dust Lead Concentrations (ug/g) across Campaigns for Window Well Surfaces
          RM-I
IN  PI  2M
  6M
12M
     Previously-Abated
IN
6M
12M
                                       RM-II
IN  PI  2M
  6M
12M
IN
                                      Modern


0
Q
Q.
O)
0


6 -
5 -
4 -
3
2
1 -
0 -


i rh ri
n n — -— Hi
u u
L-^J



6 -
5
4 -
3 •
2
1 -
0 -



ฃ—
^ j^


6M
12M
                                                                   RM-III
6 -
5 -
9 4 •
Q
t 3 -
o
| 2 -
1 -
0 -
&Pl n rh n
U U u3~~ u
6 J
5
4
3
2
1 -
0 -
M-a-S T
1
6 -
5 -
4 -
3 -
2 •
1 -
0 -
N-D D S
I I i-J LJ
KENNEDY KRIEGI
                                                          IN  PI  2M
6M
                                                                                                    1?M
                                                                                                            m
                                                                                       3)
                                                                                       o
                                                                                                             c

                                                                                                             m
                       data based on the 89 houses completing the twelve month campaign

-------
                              R&M Study: Preliminary 12 Month Report (7/95)
  Figure 8  Dust Lead Concentrations (ug/g) across Campaigns for Interior Entryway Surfaces
RM-I
                                                RM-II
                                                                             RM-III
6
5 -
• 4 -
-Q
cT
*a> 2 -
1 -
0 1
*

EHIH3 Q Q


*
*
6
5 -
4 -
3
2 -
1 -
0 -


CrQ^fu J5
. • UJ I 	 1 	 	
' 	 •-" IH. , .
J^


6
5 -
4
3 -
2
1 -
0

.
ri T "
•



rn
z
m
D
^
2
m
o
IN PI 2M 6M 12M IN PI 2M 6M 12M IN PI 2M 6M 12M ^
PI
Previously-Abated Modern ^J
o
6
5 •
9 4 -

J3
n 3 -
^> 2 -
o
1 -
0 -





i-^-i •
: Un~~ — -M
•— -• ' *


6
5
4 -

3
2
1 -
0 -





*
g^ ง ~^^~^
,

X
nra
z
H

H
m


IN
6M
           12M
IN
6M
12M
                      data based on the 89 houses completing the twelve month campaign

-------
                               R&M Study:  Preliminary 12 Month Report (7/95)
             Figure 9  Dust Loadings (mg/ftA2) across Campaigns for Floor Surfaces
RM-I
                                                 RM-II
                                                                             RM-III
o
2
1 -
Q
o 0 -
D)
-1 -
-2 -
-3 -
3 -
2
1 '
Q
o 0 -
O)
-1 -
-2


-
Eh-T| P 	 R 	 —5


IN PI 2M 6M 12M
Previously-Abated


ง— n 	 a
'-p-J I I UJ
o
2 -
1 -

0 -
-1 -
-2 -
-3 -
3 -
2 •
1 -
0 -
-1 -
-2


_
^\T " 5
. b^ง &~~ S
1
IN PI 2M 6M 12M
Modern


*
J-
o
2 •
1 -

0
-1
-2
-3 -




" Vx& S ~ 13
,_^
IN PI 2M 6M 12M
<*^
rn
z
mm
2,
m
Q
m
Q
m
71
m
m
>
0
S
z
c
m
IN
6M
           12M
IN
6M
12M
                       data based on the 89 houses completing the twelve month campaign

-------
                               R&M Study: Preliminary 12 Month Report (7/95)
         Figure 10   Dust Loadings (mg/ftA2) across Campaigns for Window Sill Surfaces
          RM-I
                                       RM-II
                                                                    RM-III
3
2 •
^ 1 -
Q
o 0
en
-1 -
-2 -
-3 -
:\rr jru -R 	 Pi
,-ij b^-f — trJ — i i
IN PI 2M 6M 12M
3 •
2
1 -
0
-1 -
-2 -
-3
^N R— rn rh
V-T J&- — td — — tp
IN PI 2M 6M 12M
3 -
2 -
1 -
0 -
-1 -
-2
-3
w n n n
Yl^L U V
IN PI 2M 6M 12M
KENNEDY KRSEGER RES
     Previously-Abated
                                      Modern


^
Q
o
o


3 -
2 -
1 -

0 •
-1
-2
-3 -



,

n n a
u U tj
1 *

3 •
2
1

0
-1 -
-2
-3 •





^~— -S— — i
* I
1 	 1
                                                                                         O
                                                                                                              f.v.fl

                                                                                                              •m-fl
                                                                                                              m
IN
6M
12M
IN
6M
12M
                       data based on the 89 houses completing the twelve month campaign

-------
Q
o
cr>
o
3
2
1
0
-1
2
-3 -
                               R&M Study:  Preliminary 12 Month Report (7/95)
        Figure 11  Dust Loadings (mg/ftA2) across Campaigns for Window Well Surfaces
          RM-I
-9-
IN  PI  2M
                      6M
                           12M
                                    3 ^
                                    2
                                    1
                                    0 i
                                    -1
                                    -2 -
                                    -3 -
                                                            RM-II
                                                              -B-
                                                  IN  PI  2M
                                                          6M
                                                 12M
                                                                                 3
                                                                                 2 ^
                                                                                 1
                                                                                 0
                                                                                 -1
                                                                                 -2
                                                                                 -3
                                                                                               RM-III
IN  PI  2M
6M
12M
                                                                                                                        m
                                                                                                                        •%.
                                                                                                                        z:
                                                                                                                        ni
                                                                                                                        m
                                                                                                                        Q
                                                                                                                        m
                                                                                                               m
Q
o
^i
o
3
2
1
0
-1
-2
-3 -
     Previously-Abated
                                             3
                                             2
                                             1
                                             0


                                             -2 -
                                             -3 -
IN
                    6M
                          12M
                                                  IN
                                                           Modern
                                                                                                                         o
                                                                                                               H
                                                                                                               m
                                                        6M
                                                 12M
                       data based on the 89 houses completing the twelve month campaign

-------
                                        R&M Study:  Preliminary 12 Month Report (7/95)


               Figure 12  Dust Loadings (mg/ftA2) across Campaigns for Interior Entryway Surfaces
                   RM-I
                                      RM-II
                                                                   RM-III
O)
o
O)
o
3 -
2 •
1 -

0 -
-1 -
-2 -
-3 -












3
2 -
1 -
0 -
-1 -
-2 -
-3 -











9
•

N






—








Ji H 	 .a 	 R
Lj-tr~" XT M
: -^ ^

3 -
2 -
1 -

0 -
-1 -
-2 -
-3 -


,_„.
: p_
rk^ h fl 	 Q
; fJ-'-V U
uL

3 •
2 -
1 -

0 -
-1 -
-2
-3


_
ri "^
Mv ^__ ^
B^i ~~^
ฑ^ ^
'

m
MBBi
m
o

Q
PI 2M 6M 12M IN PI 2M 6M 12M IN PI 2M 6M 12M HI
20
0)
ง
Previously-Abated Modern ^
O
X




— -Q — o
^_, ;
l_i_,
3 -
2 -
1 -
0 -
-1 -
-2 -
-3 -


^^-, *
: 	 	 g
y~ ~&~ ~^
•

=,

H

m


         IN
6M
12M
IN
6M
12M
                                data based on the 89 houses completing the twelve month campaign

-------
                              R&M Study:  Preliminary 12 Month Report (7/95)
       Figure 13  Blood Lead Concentrations for Children with Initial Blood Pb < 20 ug/dL
         RM-I
                                       RM-II
                                                                      RM-III
2.0
1.5-
CD
1 1.0-
O)
0
0.5-
o.o-


5—5-
iiL
*
*
*
IN 2M





-R-— i
ฑ B


* 1

2.0-
1.5-
1.0-
0.5-
0.0-
6M 12M

Previously
2.0
1.5-
m
JD
| 1.0-
O)
o
0.5-
0.0-



^~


_L



E3-B-&— e
.

IN 2M 6M 12M
2.0-
1.5
1.0-
0.5-
0.0


B—B — n n
^. V LJ LJ
ซ-!-

IN 2M 6M 12M
m
;z
ni
71
ni
Q
rn
m
w
m
-Abated



2.0
1.5
1.0-
0.5-
0.0-
Modern Urban
. 	 .
S" ง S
>
ya
o
z
d
c
m

IN
6M
12M
                                        IN
6M
12M
                      data based on the 89 houses completing the twelve month campaign

-------
            Figure  14: Dust Lead  Loadings

          R & M Study - Partial 12 Month Campaign
     100,000
             Geo. Mean 0"g/ft2)
              Ifit Eniry  Floors w/windows Floors w/o windows  Sills
                                       Wells
                                             Air Due!
                                                   Upholstery
                             Surface Type
                MU • PA • R&M I   R&M II • R&M III
                                                                m

                                                                3
                                                                7;
                                                                3D
                                                                m
                                                                m
I


2


H





I
7/95

-------
        Figure 15: Dust Lead Concentrations
         R & M Study - Partial 12 Month Campaign
     100,000
      10,000
       1,000
        100
         10
            Geo. Mean (/ug/g)
Int Entry   Floors w/windows Floors w/o windows   Sills    Wells    Air Duct    Upholstery
                            Surface Type
               MU   I PA  IR&MI   R&M II  IR&MIII
                                                I
                                                m
                                                5
                                                              m
                                                              c
                                                              MM
                                                              m
7/95

-------
                Figure 16: Dust Loadings

          R & M Study - Partial 12 Month Campaign
     100,000
      10,000
       1,000
        100
         10
             Geo. mean (mg/ft2)
              Int Entry  Floors w/windows Floors w/o windows   Sills     Wells    Air Duct    Upholstery
                             Surface Type
                MU   I PAH R&M I    R&M II  I R&M III
                                                                 m

                                                                 5
5
m
30

3D
m
I
mm


I
7/95

-------
                               R&M Study: Preliminary 12 Month Report (7/95)
       Figure 17  Dust Lead Loadings (ug/ftA2) across Groups at the 12 Month Campaign
          Floor
      Window Sill
Window Well
7
6
5 -
logld(PbD)
ro co .u
1 -
0 -
-1 -



B a ph D _
1 • 1 — *~l
— B


7 -
6
5 -
4
3
2
1 -
0
-1 -



*
Q P ^ ^
• ; ^


7
6
5
4
3
2
1 -
0
-1 -


. 	 . T^
R . _
2 Q _
M H S p



•y-
m
z
z
m
a
F

m
Q
RMI RMII RMIII PA MU RMI RMII RMIII PA MU RMI RMII RMIII PA MU ^
m
C/>
     Interior Entryway
   Floors w/o windows
7 -
6 -
5 -
4 •
3 •
2 -
1 -
0
-1 -



—— ^T^ *
i B B B s
-;-

7 -
6 -
5 -
4 -
3 -
2
H
1 '
o -
-1 -



,— -^
f=l rn ^ D
B a
. : . p

                                                                                                               5
                                                                                                               I
                                                                                                               C
                                                                                                               rn
RMI  RMII  RMIII  PA  MU
RMI   RMII  RMIII  PA  MU
                       data based on the 89 houses completing the twelve month campaign

-------
                                        R&M Study: Preliminary 12 Month Report (7/95)


               Figure 18   Dust Lead Concentrations (ug/g) across Groups at the 12 Month Campaign
                   Floor
      Window Sill
Window Well
9
Q
.0
f
o

Q
n
6
5
4 -
3 -
2 -
1 -
0 -


i i 5 i
ง


6
5
4
3
2 -
1 -
0

*
a 3
-0555
^^ r~l



6
5 -
4
3
2
1 -
0 -

T r~:~1
_^ rb Pi
d g Q




m
z
m
3
7;
^
Q
RMI RMII RMIII PA MU RMI RMII RMIII PA MU RMI RMII RMIII PA MU ITS
7ฎ
m
v*
Interior Entryway Floors w/o windows 20
6 •
5 -

4 •
3 •
2 •
1 -
0 -
.


r— ^-> .-^-,
Q i ฃ Q ^
" Q


6 -
5 -

4 •
3
2
1 -
0 -


*
m


._

m
H
ซv|
c
H
m


          RMI   RMII  RMIII  PA   MU
RMI   RMII  RMIII  PA  MU
                                data based on the 89 houses completing the twelve month campaiqn

-------
                               R&M Study:  Preliminary 12 Month Report (7/95)
         Figure 19  Dust Loadings (mg/ftA2) across Groups at the 12 Month Campaign
         Floor
      Window Sill
Window Well
3

2 •
IT 1 '
Q
o 0 •
CT)
-1 •
-2 -
-3 -



*
3 a a 9 |


3 -

2
1 -
0 ~
-1 -
-2 -
-3 -








a




RMI RMII RMIII PA MU RMI

Interior Entryway
3
2
1 .
Q
0 0
cn
-1 •
-2 •
-3

, 	 ,
i „, ^_,
.J-. : : : 	 ,
ง 5 i P a
! *
>-i-'
3
2
1 -

0
-1 •
-2
-3 -













	 ' 	 '


	 ^






ง ง B 5
l-i--

3 -

2
1 -
0
-1 •
-2
-3 -



*
- ? Q i I


5^
m
z
m
^
ffi
STJ
RMII RMIII PA MU RMI RMII RMIII PA MU 94jj
*>?#
m
CO
s
P"
Floors w/o windows Q




B g 5 I


Z
(ft

swvjj




RMI  RMII  RMIII   PA  MU
RMI  RMII  RMIII  PA  MU
                       data based on the 89 houses completinq the twelve month campaign

-------
 Figure 20:  Overall lead levels and dust loadings by group*

                R & M  Study - Partial 12-Month  Campaign
               Geo. Mean
      100,000
        10,000
         1,000
           100
                                          m

                                          5
                                          5
                                          m
                                          33

                                          X
                                          m
                 Lead Concentration
Lead Loading

   (/jg/ft" 2)
Dust Loading

   (mg/ft~2)
                 Modern Urban 11 Previously Abated • R&M I
            R&M
R&M III
7/95


* Summary measures are based on weighted averages of

floors, entryways, window sills, and window wells within houses

-------
 Figure 21:   Overall dust lead  loadings by group
           R &  M Study - Partial 12 Month Campaign
      100,000
       10,000
        1,000
             Geo. Mean (jug/ft2)
         100
                Initial
Post-IN
2 month
6 month
12 month
               Modern Urban • Previously Abated • R&M I  R&M II • R&M III
7/95
* Summary measures are based on weighted averages of
floors, entryways, window sills, and window wells within houses
                                             i
                                             m
                                             s
                                             33
                                             I
                                             m
                                             ao
                                             33
                                             m
                                                                       c
                                                                       FT

-------
 Figure 22:  Overall dust lead  concentrations  by  group*

             R & M Study - Partial 12 Month Campaign
              Geo. Mean
      100,000
       10,000
        1,000
          100
                 Initial
Post-IN
2 month
6 month
12 month
                Modern Urban • Previously Abated • R&M I  R&M II • R&M
7/95

* Summary measures are based on weighted averages of

floors, entryways, window sills, and window wells within houses
                                               i


                                               I
                                               *
                                               35
                                               m
                                               o
                                               m
                                                                         m
                                                                         (ft

-------
     Figure 23:  Overall  dust  loadings by  group

           R & M  Study - Partial 12 Month Campaign
      10,000
       1,000
        100
            Geo. Mean (mg/ft2)
               Initial
Post-IN
2 month
6 month
12 month
               Modern Urban • Previously Abated • R&M I   R&M II • R&M
7/95

* Summary measures are based on weighted averages of

floors, entryways, window sills, and window wells within houses
                                              m

                                              3
                                              *
                                              35
                                              m
                                              6
                                              m
                                              30

                                              as
c

ft

-------
                        Figure 24:
  40
                     R &M Study- Preliminary 12 Month Report (7/95)
                     Children's Blood Lead Levels Across Time - R &M I
                           for children with 12-month values
  35-
  30-
D>
g
8 20
I

^ 15-
CQ
 Dec-1992
Jun-1993
Dec-1993           Jun-1994
    Blood Collection Date
Dec-1994
Jun-1995

-------
                     Figure 25:
  40
  35-
  30-
  25-
8 20


<3

|


^ 15~
o
_o

CD



  10-
   5_
 Dec-1992
                  R & M Study - Preliminary 12 Month R eport (7/95)

                  Children's Blood Lead Levels Across Time - R&M I

                         for children with 12-month values
Jun-1993
Dec-1993          Jun-1994

    Blood Collection Date
                                                                                           m

                                                                                           1
                                                                                           s

                                                                                           i
                                                                                           s
                                                                                           m
                                                                                           o
                                                                                           z
                                                                                           C0
                                                                                                                nl
Dec-1994
Jun-1995

-------
                              Figure 26:
 R &M Study- Preliminary 12 Month Report (7/95)
Children's Blood Lead Levels Across Time - R&M II
        for children with 12-month values
          Dec-1992           Jun-1993           Dec-1993           Jun-1994
                                                     Blood Collection Date
•The bloodleadvaluelhatappearsas40isactjallya value of43ug/dL.
                                        Dec-1994
Jun-1995

-------
  40
  35_
  30-
  20-
8
1
0)
-1 15-


8
CD
  10-
   5_
 Dec-1992
                     Figure 27:           R &M Study- Preliminary 12 Month Report (7/95)

                                   Children's Blood Lead Levels Across Time - Modern Urban

                                              for children with 12-month values

                        —T~

                     Jun-1993
                                             T
                                                                 T
                                                                                                                m


                                                                                                                m
                                                                                                                m
                                                                                                                o
                                                                                                                m
                                                                                                                50
                                                                                                                o
                                                                       ^r^



                                                                       |

                                                                       m
Dec-1993            Jun-1994

     Blood Collection Date
Dec-1994
Jun-1995

-------
    40
     35-
    30-
  125
  o
  8 20
  8
  _g
  en
    10-1
      5-
   Dec-1992
                        Figure 28:           R &M Study- Preliminary 12 Month Report (7/95)
                                      Children's Blood Lead Levels Across Time - Previously Abated
                                                  for children vvith 12-month values


Jun-1993
Dec-1993

                                                                                                m

                                                                Jun-1994
                                                 Blood Collection Date
"The child whh blood lead value of 53ug/dL is excluded from analysis.
Dec-1994
Jun-1995

-------
                        Preliminary R&M Study Report
                        Partial 12 Month Data   7/95
   APPENDIX A
Descriptive Statistics

-------
Table at Descriptive Statistics for 12  Month Campaign Dust Lead Concentrations (j/g/g) by Surface Type and Group
  Surface

  Air Duct
  Floor
  Floor w/o windows
  Interior  Entrance
  Sill
 Upholstery
 Well
Study
Group
Modern Urban
Pre-Abated
RM-Z
RM-II
RM-III
Modern Urban
Pre-Abated
RM-I
RM-II
RM-III
Modern Urban
Pre-Abated
RM-I
RM-II
RM-III
Modern Urban
Pre-Abated
RM-I
RM-II
RM-III
Modern Urban
Pre-Abated
RM-I
RM-II
RM-III
Modern Urban
Pre-Abated
RM-I
RM-II
RM-III
Modern Urban
Pre-Abated
RM-I
RM-II
RM-III

N
11
5
8
14
IS
31
28
43
40
40
4
6
15
11
9
15
14
20
20
20
30
28
39
40
40
4
9
12
5
5
30
28
36
40
40
                                     Minimum
                                                   Maximum
18
167
332
52
168
13
11
205
85
67
2
56
175
96
169
21
196
262
155
37
31
193
626
192
172
89
215
91
225
202
91
274
4,012
65
3
4,464
5,841
24,150
5,594
5,226
1,085
8,894
11,977
8,714
19,604
152
6,190
9,674
58,840
1,419
777
7,741
704,065
4,251
4,887
1,742
132,312
41,979
39,009
52,598
461
770
2,195
473
973
8,734
45,214
493,006
151,924
29,576
Geometric
Mean
101
1,138
1,592
500
859
79
570
819
627
465
15
966
575
615
565
119
1,106
1,695
819
701
267
1,138
7,754
2,941
796
206
394
457
374
392
43B
3,031
22,963
3,862
1,130

std(ln)
1.585
1.632
1.371
1.360
0.876
0.796
1.332
1.000
0.977
1.084
1.758
1.636
1.012
1.647
0.727
0.961
1.108
1.692
0.814
1.078
0.968
1.388
1.072
1.470
1.209
0.679
0.485
0.808
0.299
0.592
0.909
1.650
1.020
1.618
1.347
 Lower
95% CI

    35
   150
   506
   228
   529

    SB
   319
   579
   421
   299

     1
   174
   328
   204
   323

    70
   583
   768
   559
   423

   171
   608
 5,110
 1,707
   505

    70
   271
   274
   258
   188

   297
 1,341
16,041
 1,931
   711
 Upper
95% CI

   293
 8,628
 5,009
 1,096
 1,395

   108
 1,018
 1,159
   935
   724

   252
 5,377
 1,006
 1,860
   988

   203
 2,096
 3,741
 1,198
 1,160

   416
 2,129
11,767
 5,067
 1,255

   606
   571
   764
   542
   817

   647
 6,852
32,872
 7,723
 1,797
"ฐฐ"'  ?indฐw 8i118'  and
prevlously Abated
                                             wells  account for  multiple samples from a house

                                                           Page   A-l

-------
    Table bi  Descriptive  Statistics for 12 Month Campaign Dust Lead Loadings (fjg/ft2)  by  Surface Type and Group
Surface

Air Duct
 Floor
 Floor w/o windows
 Interior Entrance
 Sill
 Upholstery
 Well
Study
Group
Modern Urban
Pro-Abated
RM-I
RM-II
RM-III
Modern Urban
Fre-Abated
RM-I
RM-II
RM-III
Modern Urban
Pro-Abated
RM-I
RM-II
RM-III
Modern Urban
Pro-Abated
RM-I
RM-II
RM-III
Modern Urban
Pro-Abated
RM-I
RM-II
RM-III
Modern Urban
Pro-Abated
RM-I
RM-II
RM-III
Modern Urban
Fre-Abated
RM-I
RM-II
RM-III

N
11
5
8
14
15
31
28
43
40
40
4
6
15
11
9
15
14
20
20
20
30
28
39
40
40
4
9
12
5
5
30
28
36
40
40
                                                 Minimum
                                                               Maximum
112
326
245
15
144
< 1
< 1
13
8
2
< 1
5
8
1
7
1
8
35
7
1
2
4
5
1
1
1
6
1
13
7
9
52
485
6
2
14,428
180,703
3,755,278
413,541
874,350
107
2,424
23,466
4,416
1,304
3
856
204
7,580
444
391
15,204
45,201
9,574
1,452
36
24,481
7,523
9,921
617
24
82
158
824
744
2,410
22,872
367,432
163,334
30,614
Geometric
Mean
856
10,020
18,073
6,675
13,040
8
77
112
79
36
1
108
55
42
34
30
119
514
141
89
9
74
603
201
23
7
20
36
70
131
217
1,172
19,412
1,761
268

std(ln)
1.760
2.394
3.357
2.876
2.077
1.138
2.006
1.488
1.342
1.466
0.901
1.724
1.240
2.300
1.425
1.383
2.260
1.840
1.931
1.847
0.809
1.770
1.770
2.339
1.510
1.318
0.982
1.453
1.552
1.897
1.470
1.887
1.320
2.375
1.618
 Lower
95% CI

   262
   513
 1,092
 1,268
 4,127

     5
    32
    69
    45
    20

     0
    18
    28
     9
    11

    14
    32
   217
    57
    38

     6
    33
   307
    78
    13

     1
    10
    14
    10
    12

   108
   504
11,947
   641
   157
  Upper
 95% CI

  2,794
195,779
299,134
 35,126
 41,199

     13
    186
    181
    139
     65

      4
    662
    109
    195
    101

     64
    440
  1,216
    349
    212

     12
    166
  1,185
    520
     41

     61
     43
     90
    483
  1,380

    439
  2,723
 31,542
  4,840
    459
Values for floors,  window sills,  and window wells  account  for multiple samples from a house
Pre-Abated=Previously Abated
                                                           Page   A-2

-------
  Table ei


Surface

Air Duct
Floor
Floor w/o windows
Interior Entrance
Sill
Upholstery
Hell
Descriptive Statistics Cor the 12 Month Campaign Dust  Loadings  (mg/ft') by Surface Type and Oroupt
Study
Group
Modern Urban
Pro-Abated
RM-I
RM-II
RM-III
Modern Urban
Pre-Abated
RM-I
RM-II
RM-III
Modern Urban
Pre-Abated
RM-I
RM-II
RM-III
Modern Urban
Pre-Abated
RM-I
RM-II
RM-III
Modern Urban
Pre-Abated
RM-I
RM-II
RM-III
Modern Urban
Pre-Abated
RM-I
RM-II
RM-III
Modern Urban
Pre-Abated
RM-I
RM-II
RM-III

N
11
5
8
14
15
31
28
43
40
40
4
6
15
11
9
15
14
20
20
20
30
28
39
40
40
4
9
12
5
5
30
28
36
40
40
                                    Minimum

                                      1,222
                                      1,951
                                        354
                                        283
                                         94

                                          6
                                         18
                                         14
                                         19
                                         17

                                         21
                                         49
                                         17
                                          3
                                         13

                                         11
                                          2
                                         19
                                         19
                                         13

                                          8
                                          8
                                          5
                                          4
                                        < 1

                                         14
                                         12
                                         11
                                         56
                                         36

                                         28
                                         33
                                         76
                                         12
                                          9
Maximum

176,990
 34,803
155,499
103,878
167,309

    675
    749
  8,924
    981
    617

    292
    156
    459
    562
    313

    760
  2,669
  8,077
  2,330
  1,941

    354
    518
  1,901
  1,742
    264

     87
    214
    407
  1,854
  2,704

  3,441
  5,496
 10,988
  7,097
  1,592
Geometric
Mean
8,474
8,807
11,353
13,349
15,184
100
135
134
126
78
62
112
96
67
60
250
108
303
172
127
32
65
78
68
29
36
52
79
188
334
497
387
841
456
237

std(ln)
1.560
1.090
2.172
2.046
1.808
1.072
1.141
1.094
0.986
0.968
1.110
0.451
1.172
1.510
1.005
1.026
1.835
1.486
1.356
1.350
0.949
1.169
1.209
1.234
1.257
0.791
0.937
1.108
1.398
1.601
1.265
1.041
0.955
1.316
1.245
 Lower
95% CI

 2,972
 2,274
 1,846
 4,096
 5,579

    62
    82
    90
    87
    56

    11
    70
    50
    24
    28

   141
    37
   151
    91
    68

    21
    40
    49
    43
    18

    10
    25
    39
    33
    46

   261
   253
   587
   272
   145
 Upper
95% CI

24,161
34,107
69,803
43,505
41,331

   161
   220
   201
   184
   107

   361
   180
   183
   186
   129

   441
   311
   608
   325
   240

    50
   106
   124
   109
    46

   128
   107
   159
 1,067
 2,439

   945
   590
 1,205
   765
   388
                    Values for floors,  window sills,  and window wells account for multiple samples from a house
           Pre-Abated=Previously Abated
                                                          Page   A-3

-------
Table di      Pearson correlation coefficients between  log transformed dust  lead loadings (*/g/ft') at  the twelve month  campaign


Floors

Window Sills

Window Wells

Interior
Entryway
Floors - Rms
w/o Windows
Air Ducts
Window Sills

.49"
(89)





"

"

Window Wells

.49"
(88)
.73"
(88)



~

*

Interior
Entryway
.38"
(89)
.39"
(89)
.40"
(88)

•

•

Floors -
Rms w/o Windows
.46"
(45)
.27
(45)
.17
(44)
.29'
(45)

-

Air
Ducts
.16
(53)
.25
(53)
.26
(52)
.15
(53)
-.11
(28)

Upholstery

.28
(35)
.04
(35)
.07
(35)
.16
(35)
.17
(17)

              p < .05;       •*      p < .01
              Air Duct or Upholstery Samples were collected, not both.
                                                       Page  A-4

-------
Table ei
             Pearson correlation coefficients between log transformed dust lead concentrations at the  twelve month campaign


Floors

Window Sills

Window Hells

Interior
Entryway
Floors - Rms
w/o Windows
Air
Ducts
Window
Sills

.51"
(89)





™
-

~
Window Wells

.54"
(88)
.67"
(88)



~
-

~
Interior
Entryway

.54"
(89)
.47"
(89)
.51"
(88)

—
-

—
Floors -
Rms w/o
Windows
.55"
(45)
.46"
(45)
.30
(44)
.48"
(45)
-

-
Air
Ducts

.47"
(53)
.42"
(53)
.33
(52)
.34
(53)
.37
(28)

-
Upholstery

.30
(35)
.30
(35)
.31
(35)
. 38
(35)
.43
(17)

^ซ
                    p < .05;    **    p <  .01
                   Air Duct or Upholstery  Samples were collected, not both.
                                                  Page   A-5

-------
Table f:     Pearson correlation coefficients between  log transformed dust  loadings  (mg/ftj) at the twelve  month campaign


Floors

Window Sills

Window Wells

Interior Entryway

Floors - Rms w/o
Windows
Air
Ducts
Window Sills

.08
(89)








-
Window
Wells
.14
(88)
.66"
(88)





"
-
Interior
Entryway
.08
(89)
.56"
(89)
29
(88)



"
-
Floors -
Rms w/o Windows
.27
(45)
.04
(45)
.18
(44)
- .01
(45)

-
-
Air
Ducts
- .10
(53)
.05
(53)
- 07
(52)
-.04
(53)
- .35
(28)
-
Upholstery

.04
(35)
- .14
(35)
- .23
(35)
- .12
(35)
- .15
(17)
_a
                    p < .05; **  p <  .01
                    Air Duct or Upholstery Samples were collected, not both
                                                     Page   A-6

-------
Table gi     Descriptive statistics on soil lead concentrations O/g/g) from the preliminary six month campaign report
               Study
               Group

               Modern Urban
               Prev'ly Abated
               RM-I
               RM-II
13
 3
10
 9

35
Minimum

     34
    304
    182
    428
Maximum

    229
  7,845
  4,530
  2,608
Geometric
     Mean

       73
    1,521
      730
      708
std(ln)

  0.489
  1.625
  1.000
  0.599
Table hi
             Descriptive statistics on water lead concentrations 
-------
                                   Preliminary R&M study Report
                                   Partial 12 Month Data   7/95
               APPENDIX  B
Longitudinal data analysis output for dust lead

-------
                 DUST  LEAD LOADINGS WITHIN COMPLETE R_M HOUSES AS AN OUTCOME OF ENVIRONMENTAL FACTORS: 07/17/95
                    12-month campaign all data model: Tactorl = level + season + eampaign(level) + randomjiouse

                                                       The MIXED Procedure
                                                     Class Level Information

                                         Class     Levels  Values	
                                         DID           60  310 316 343 347 348 353 355
                                                           357 358 359 366 372 373 381
                                                           386 403 406 409 418 419 302
                                                           312 335 338 345 350 354 365
                                                           370 374 376 377 384 385 391
                                                           397 407 412 416 420 306 309
                                                           317 320 323 325 326 328 332
                                                           336 383 389 390 402 405 411
                                                           413 415 417 404
                                         LEVEL          3321
                                         CAMPAIGN       5  00 02 06 12 PI
UST3.LIS   (source)                                        Page  B-1

-------
                 DUST LEAD LOADINGS WITHIN COMPLETE R_H HOUSES AS AN OUTCOME OF ENVIRONMENTAL FACTORS: 07/17/95
                    12-month campaign all data model: Tactor1 = level + season + campaign(level) + random house

                                                       The NIXED Procedure
                                              Covariance Parameter Estimates (REML)
                              Cov Parm
     Ratio
  Estimate
                                                                        Std Error
                 Z  Pr > !Zi
                              DID
                              Residual
0.52166498
1.00000000
0.09194687
0.17625655
0.02427637
0.01687193
 3.79
10.45
0.0002
0.0000
                                                       The MIXED Procedure
                                              Model Fitting Information for FACTOR1
                                             Description
                                     value
Observations 293.0000
Variance Estimate 0.1763
Standard Deviation Estimate 0.4198
REML Log Likelihood -215.239
Akaike's Information Criterion -217.239
Schwartz's Bayesian Criterion -220.855
-2 REML Log Likelihood 430.4771
Null Model LRT Chi -Square 44.3687
Null Model LRT OF 1.0000
Null Model LRT P-Value 0.0000
The MIXED Procedure
Solution for Fixed Effects
Parameter
INTERCEPT
SPRING
SUMMER
FALL
LEVEL 3
LEVEL 2
LEVEL 1
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAHPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAHPAIGN( LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAHPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAHPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN LEVEL)








00
02
06
12
PI
00
02
06
12
PI
00
02
06
12
PI








3
3
3
3
3
2
2
2
2
2





Estimate
0.21071261
-0.05365035
-0.15434789
-0.00393296
-1.57870478
-0.79301191
0.00000000
3.05571972
0.70179258
0.57870463
0.47188651
0.00000000
2.03467544
0.60115318
0.41829973
0.42521529
0.00000000
0.93998960
0.06323570
0.17211068
0.14439255
0.00000000
Std Error
0.
0.
0.
0.
0.
0.

0.
0.
0.
12117595
07966301
07064688
08553026
16707386
16524626
B
13688774
13823696
13526552
0.13338638

0.
0.
0.
0.

0.
0.
0.
0.

•
13359146
13780818
13331476
13336752
•
13534171
13992696
13372528
13503716
.
DDF
218
218
218
218
218
218
m
218
218
218
218
•
218
218
218
218
•
218
218
218
218
.
T
1.74
-0.67
-2.18
-0.05
-9.45
-4.80
<
22.32
5.08
4.28
3.54
•
15.23
4.36
3.14
3.19
•
6.95
0.45
1.29
1.07
.
Pr > !T!
0.0835
0.5014
0.0300
0.9634
0.0000
0.0000
.
0.0000
0.0000
0.0000
0.0005
•
0.0000
0.0000
0.0019
0.0016
•
0.0000
0.6518
0.1994
0.2861
.
Aloha
0.05
0.05
0.05
0.05
0.05
0.05
_
0.05
0.05
0.05
0.05
•
0.05
0.05
0.05
0.05
•
0.05
0.05
0.05
0.05
.
Lower
-0.0281
-0.2107
-0.2936
-0.1725
-1.9080
-1.1187

2.7859
0.4293
0.3121
0.2090
•
1.7714
0.3295
0.1555
0.1624
•
0.6732
-0.2125
-0.0914
-0.1218
m
Itooer
0.4495
0.1034
-0.0151
0.1646
-1.2494
-0.4673

3.3255
0.9742
0.8453
0.7348
•
2.2980
0.8728
0.6811
0.6881
.
1.2067
0.3390
0.4357
0.4105

                                        Source
           The MIXED Procedure
         Tests of Fixed Effects

               NDF   DDF  Type IMF  Pr > F
SPRING
SUMMER
FALL
LEVEL
CAMPAIGN(LEVEL)
1
1
1
2
12
218
218
218
218
218
0.45
4.77
0.00
30.48
78.00
0.5014
0.0300
0.9634
0.0000
0.0000
JST3.LIS   (source)
                Page   B-2

-------
                  DUST LEAD LOADINGS WITHIN COMPLETE R_M HOUSES AS AN OUTCOME OF ENVIRONMENTAL  FACTORS:  07/17/95
                   12-month campaign all data model: factor!  = campaign + season + level(campaign)  *  random house

                                                        The MIXED Procedure
                                               Covariance Parameter Estimates (REHL)
                               Cov Parm
                                                 Ratio
                Estimate
               Std Error
                                                                                         Z   Pr  >  1Z!
                               DID
                               Residual
0.52166498
1.00000000
0.09194687
0.17625655
0.02427637
0.01687193
 3.79
10.45
0.0002
0.0000
                                                        The MIXED  Procedure
                                               Model  Fitting Information for  FACTOR1
                                              Description
                                     Value
                                              Observat i ons                     293.0000
                                              Variance Estimate                 0.1763
                                              Standard Deviation Estimate       0.4198
                                              REML  Log Likelihood              -215.239
                                              Akaike's Information  Criterion   -217.239
                                              Schwartz's  Bayesian Criterion    -220.855
                                              -2  REML Log Likelihood           430.4771
                                              Null  Model  LRT Chi-Square        44.3687
                                              Null  Model  LRT DF                 1.0000
                                              Null  Model  LRT P-Value            0.0000

                                                       The MIXED Procedure
                                                    Solution  for Fixed Effects
               Parameter
                                          Estimate
                                                       Std Error   DDF
                                                                             T  Pr >  !T!  Aloha
                                                                                                    Lower
                                                                                                              Uooer
INTERCEPT
SPRING
SUMMER
FALL
CAMPAIGN 00
CAMPAIGN 02
CAMPAIGN 06
CAMPAIGN 12
CAMPAIGN PI
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL (CAMPAIGN)
LEVEL (CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAHPAIGN)
LEVEL(CAMPAIGN)
LEVEL (CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)









3
2
1
3
2
1
3
2
1
3
2
1
3
2
1









00
00
00
02
02
02
06
06
06
12
12
12
PI
PI
PI
0.21071261
-0.05365035
-0.15434789
-0.00393296
0.93998960
0.06323570
0.17211068
0.14439255
0.00000000
0.53702534
0.30167393
0.00000000
-0.94014791
-0.25509443
0.00000000
-1.17211083
-0.54682286
0.00000000
-1.25121082
-0.51218917
0.00000000
•1.57870478
•0.79301191
0.00000000
0
0
0
0
0
0
0
0

0
0

0
0

0
0

.12117595
.07966301
.07064688
.08553026
.13534171
.13992696
.13372528
.13503716
•
.16628366
.16559959
•
.17149305
.16909995
•
.16424355
.16376917
•
0.16609913
0.16548192

•
0.16707386
0.16524626

.
218
218
218
218
218
218
218
218
•
218
218
•
218
218
•
218
218
•
218
218
•
218
218
.
1.74
-0.67
-2.18
-0.05
6.95
0.45
1.29
1.07
•
3.23
1.82
•
-5.48
-1.51
•
-7.14
-3.34
•
-7.53
-3.10
•
-9.45
-4.80
.
0.0835
0.5014
0.0300
0.9634
0.0000
0.6518
0.1994
0.2861
•
0.0014
0.0699
•
0.0000
0.1329
•
0.0000
0.0010
•
0.0000
0.0022
•
0.0000
0.0000
.
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
.
-0.0281
-0.2107
-0.2936
-0.1725
0.6732
-0.2125
-0.0914
-0.1218
.
0.2093
-0.0247
•
-1.2781
-0.5884
•
-1.4958
-0.8696
•
•1.5786
-0.8383
•
-1.9080
-1.1187
.
0.4495
0.1034
-0.0151
0.1646
1.2067
0.3390
0.4357
0.4105
.
0.8648
0.6281
.
-0.6022
0.0782
-
-0.8484
-0.2240
•
-0.9238
-0.1860
•
-1.2494
-0.4673
„
                                         Source
           The MIXED Procedure
         Tests of Fixed Effects

               NDF   DDF  Type III  F   Pr >  F
SPRING
SUMMER
FALL
CAMPAIGN
LEVEL (CAMPAIGN)
1
1
1
4
10
218
218
218
218
218
0.45
4.77
0.00
199.07
20.67
0.5014
0.0300
0.9634
0.0000
0.0000
HJST3.LIS    (source)
               Page  B-3

-------
               DUST  LEAD CONCENTRATION WITHIN COMPLETE R_M HOUSES AS AN OUTCOME OF ENVIRONMENTAL FACTORS: 07/17/95
                    12-month campaign all data model: factoM = level + season + campaign(level) + randomjiouse

                                                       The MIXED Procedure
                                              Covariance Parameter Estimates (REML)
                              Cov Farm
                                                Ratio
                 Estimate
                                                                        Std Error
                                                                                        Z  Pr > !Z!
                              DID
                              Residual
  1.49789517
  1.00000000
0.30631481
0.20449683
0.06542666
0.01958334
 4.68
10.44
0.0000
0.0000
                                                       The MIXED Procedure
                                              Model Fitting Information for FACTOR1
                                             Description
                                      Value
                                             Observations                    293.0000
                                             Variance Estimate                 0.2045
                                             Standard Deviation Estimate       0.4522
                                             REML Log Likelihood             -259.876
                                             Akaike's Information Criterion  -261.876
                                             Schwartz's Bayesian Criterion   -265.493
                                             -2 REML Log Likelihood          519.7517
                                             Null Model LRT Chi-Square       132.5278
                                             Null Model LRT DF                 1.0000
                                             Null Model LRT P-Value            0.0000
                                                       The MIXED Procedure
                                                    Solution for Fixed Effects
               Parameter
Estimate
                                                       Std Error   DDF
                    T   Pr  >  IT!  Alpha
                                                                                                    Lower
                                                                                                              Upper
INTERCEPT
SPRING
SUMMER
FALL
LEVEL 3
LEVEL 2
LEVEL 1
CAHPAIGN(LEVEL)
CAHPAIGN(LEVEL)
CAMPAIGN LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN (LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)







00
02
06
12
PI
00
02
06
12
PI
00
02
06
12
PI







3
3
3
3
3
2
2
2
2
2
1
1
1
1
1
0.79154396
0.01012971
0.03712852
0.18065916
•1.67197043
-0.49669126
0.00000000
1.69044516
-0.11563154
-0.21343092
-0.43154675
0.00000000
0.38176648
-0.42809703
-0.61411477
-0.75782173
0.00000000
-0.02383807
-0.37532179
-0.50021153
-0.40423858
0.00000000
0.16450453
0.08789402
0.07678562
0.09400117
0.
0.

0.
0.
0.
0.

0.
0.
0.
0.

0.
22894525
22731269
i
14764545
14910152
14582414
14370749
•
14392992
14866878
14363061
14369051
•
14593333
0.15104645
0.
0.

14407048
14554370
.
218
218
218
218
218
218
B
218
218
218
218
•
218
218
218
218
•
218
218
218
218
.
4.81
0.12
0.48
1.92
-7.30
-2.19
m
11.45
-0.78
-1.46
-3.00
•
2.65
-2.88
-4.28
•5.27
•
-0.16
-2.48
-3.47
-2.78
.
0.0000
0.9084
0.6292
0.0559
0.0000
0.0300
t
0.0000
0.4389
0.1447
0.0030
•
0.0086
0.0044
0.0000
0.0000
•
0.8704
0.0137
0.0006
0.0060
.
0.05
0.05
0.05
0.05
0.05
0.05
t
0.05
0.05
0.05
0.05
•
0.05
0.05
0.05
0.05
•
0.05
0.05
0.05
0.05
.
0.4673
-0.1631
-0.1142
-0.0046
-2.1232
-0.9447
m
1.3994
-0.4095
-0.5008
-0.7148
•
0.0981
-0.7211
-0.8972
-1.0410
•
-0.3115
-0.6730
-0.7842
-0.6911
,
1.1158
0.1834
0.1885
0.3659
-1.2207
-0.0487

1.9814
0.1782
0.0740
-0.1483
.
0.6654
-0.1351
-0.3310
-0.4746
•
0.2638
-0.0776
-0.2163
-0.1174
„
                                        Source
            The MIXED Procedure
          Tests of Fixed Effects

                NDF   DDF  Type IMF  Pr > F
SPRING
SUMMER
FALL
LEVEL
CAMPAIGN(LEVEL)
1
1
1
2
12
218
218
218
218
218
0.01
0.23
3.69
21.67
31.03
0.9084
0.6292
0.0559
0.0000
0.0000
'UST3.LIS   (source)
                                                           Page  B-4

-------
              DUST LEAD CONCENTRATION WITHIN COMPLETE R_M HOUSES AS AN OUTCOME OF ENVIRONMENTAL  FACTORS:  07/17/95
                  12-month campaign all data model: facTorl = campaign * season * level(campaign)  +  random_house

                                                       The MIXED Procedure
                                              Covanance Parameter Estimates (REML)
                              Cov Parm
                                                Ratio
                 Estimate
               Std Error
                 Z  Pr > !Z!
                              DID
                              Residual
 1.49789517
 1.00000000
0.30631481
0.20449683
0.06542666
0.01958334
 4.68
10.44
0.0000
0.0000
                                                       The MIXED Procedure
                                              Model Fitting Information for FACTOR1
                                             Description
                                      Value
                                             Observations                    293.0000
                                             Variance Estimate                 0.2045
                                             Standard Deviation Estimate       0.4522
                                             REML Log Likelihood             -259.876
                                             Akaike's Information Criterion  -261.876
                                             Schwartz's Bayesian Criterion   -265.493
                                             -2 REML Log Likelihood          519.7517
                                             Null Model LRT Chi-Square       132.5278
                                             Null Model LRT DF                 1.0000
                                             Null Model LRT P-Value            0.0000

                                                       The MIXED Procedure
                                                    Solution for Fixed Effects
               Parameter
Estimate
                                                       Std Error    DDF
                   T  Pr >  IT!  Alpha
                                                                                                   Lower
                                                                                                             Upper
INTERCEPT
SPRING
SUMMER
FALL
CAMPAIGN 00
CAMPAIGN 02
CAMPAIGN 06
CAMPAIGN 12
CAMPAIGN PI
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL (CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL (CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)









3
2
1
3
2
1
3
2
1
3
2
1
3
2
1









00
00
00
02
02
02
06
06
06
12
12
12
PI
PI
PI
0.79154396
0.01012971
0.03712852
0.18065916
-0.02383807
-0.37532179
-0.50021153
-0.40423858
0.00000000
0.04231281
-0.09108671
0.00000000
-1.41228018
-0.54946649
0.00000000
-1.38518981
-0.61059450
0.00000000
-1.69927860
-0.85027441
0.00000000
-1.67197043
-0.49669126
0.00000000
0.16450453
0.08789402
0.07678562
0.09400117
0.14593333
0.15104645
0.14407048
0.14554370
_
0.22822452
0.22762099
•
0.23276139
0.23068327
•
0.22641807
0.22601142
•
0.22805884
0.22750959
•
0.22894525
0.22731269
.
218
218
218
218
218
218
218
218
_
218
218
•
218
218
•
218
218
•
218
218
•
218
218
.
4.81
0.12
0.48
1.92
-0.16
-2.48
-3.47
-2.78
_
0.19
-0.40
•
-6.07
-2.38
•
-6.12
-2.70
•
-7.45
-3.74
•
-7.30
-2.19
.
0
0
0
0
0
0
0
0

0
0

0
0

0
0

0
.0000
.9084
.6292
.0559
.8704
.0137
.0006
.0060

.8531
.6894
•
.0000
.0181
•
.0000
.0074
•
.0000
0.0002

•
0.0000
0.0300

.
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05

0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
.
0.4673
-0.1631
-0.1142
-0.0046
-0.3115
-0.6730
-0.7842
-0.6911

-0.4075
•0.5397
•
-1.8710
-1.0041
•
-1.8314
-1.0560
•
-2.1488
•1.2987
•
-2.1232
-0.9447
,
1.1158
0.1834
0.1885
0.3659
0.2638
-0.0776
•0.2163
-0.1174

0.4921
0.3575
•
-0.9535
-0.0948
•
-0.9389
-0.1651
•
-1.2498
-0.4019
•
-1.2207
-0.0487
.
                                        Source
            The MIXED Procedure
          Tests of  Fixed Effects

                NDF   DDF  Type III F  Pr > F
SPRING
SUMMER
FALL
CAMPAIGN
1
1
1
4
218
218
218
218
0.01
0.23
3.69
67.22
0.9084
0.6292
0.0559
0.0000
                                        LEVEL(CAMPAIGN)    10   218
                                14.63  0.0000
UST3.LIS   (source)
                 Page  8-5

-------
                    DUST  LOADINGS  WITHIN  COMPLETE R_M  HOUSES AS AN OUTCOME OF ENVIRONMENTAL FACTORS: 07/17/95
                    12-month  campaign alt  data model:  factorl = level + season + campaign(level) + random house

                                                       The MIXED Procedure
                                              Covariance Parameter Estimates (REML)
                              Cov Parm
                                                Ratio
                Estimate
                                                                        Std Error
                                             Z  Pr > !Z!
                              DID
                              Residual
0.02848508
1.00000000
0.01173763
0.41206246
0.01986539
0.03944356
 0.59
10.45
0.5546
0.0000
                                                       The MIXED Procedure
                                              Model Fitting Information for FACTOR1
                                             Description
                                     Value
Observations 293.0000
Variance Estimate 0.4121
Standard Deviation Estimate 0.6419
REML Log Likelihood -299.850
Akaike's Information Criterion -301.850
Schwartz's Bayesian Criterion -305.466
-2 REML Log Likelihood 599.6990
Null Model LRT Chi -Square 0.3884
Null Model LRT DF 1.0000
Null Model LRT P-Value 0.5331
The MIXED Procedure
Solution for Fixed Effects
Parameter
INTERCEPT
SPRING
SUMMER
FALL
LEVEL 3
LEVEL 2
LEVEL 1
CAMPAIGN (LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN (LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN (LEVEL)
CAHPAIGN(LEVEL)
CAMPAIGN (LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN LEVEL)
CAMPAIGN(LEVEL)
CAMPAIGN(LEVEL)








00
02
06
12
PI
00
02
06
12
PI
00
02
06
12
PI








3
3
3
3
3
2
2
2
2
2
1
1
1
1
1
Estimate
-0.17237997
-0.23095597
-0.24665377
-0.28858532
-0.10243114
-0.10783173
0.00000000
2.47313079
0.13786048
0.03977628
0.08857254
0.00000000
2.00774854
0.17599146
0.12030923
0.18485027
0.00000000
0.83066936
0.09064261
0.10539994
0.07875213
0.00000000
Std Error
0.15458120
0.11266415
0.10469955
0.12244765
0.21105977
0.20823255
.
0.20846672
0.21046377
0.20629882
0.20381395
•
0.20411718
0.20970352
0.20370511
0.20377294
•
0.20630818
0.21254020
0.20433152
0.20606596
.
DDF
218
218
218
218
218
218
m
218
218
218
218
•
218
218
218
218
•
218
218
218
218
.
T
-1.12
-2.05
-2.36
-2.36
-0.49
-0.52
_
11.86
0.66
0.19
0.43
•
9.84
0.84
0.59
0.91
•
4.03
0.43
0.52
0.38
.
Pr > !T!
0.2660
0.0416
0.0194
0.0193
0.6279
0.6051

0.0000
0.5131
0.8473
0.6643
•
0.0000
0.4023
0.5554
0.3653
•
0.0001
0.6702
0.6065
0.7027
.
Aloha
0.05
0.05
0.05
0.05
0.05
0.05

0.05
0.05
0.05
0.05
•
0.05
0.05
0.05
0.05
•
0.05
0.05
0.05
0.05
.

-0
-0
-0
-0
-0
-0

2
-0
-0
-0

1
-0
-0
-0

0
Lower
.4770
.4530
.4530
.5299
.5184
.5182

.0623
.2769
.3668
.3131
•
.6055
.2373
.2812
.2168
•
.4241
-0.3283
-0.2973
-0.3274

.
Upper
0.1323
-0.0089
-0.0403
-0.0473
0.3135
0.3026

2.8840
0.5527
0.4464
0.4903

2.4100
0.5893
0.5218
0.5865
•
1.2373
0.5095
0.5081
0.4849
a
                                        Source
           The MIXED Procedure
         Tests of Fixed Effects

               NDF   DDF  Type IMF  Pr > F
SPRING
SUMMER
FALL
LEVEL
1
1
1
2
218
218
218
218
4.20
5.55
5.55
2.80
0.0416
0.0194
0.0193
0.0630
                                        CAMPAIGN(LEVEL)    12   218
                               30.48  0.0000
UST3.LIS   (source)
                                                           Page  B-6

-------
                    DUST  LOADINGS WITHIN COMPLETE  R_M  HOUSES AS AN OUTCOME OF  ENVIRONMENTAL  FACTORS: 07/17/95
                   12-month campaign all data model: factorl = campaign + season +  IeveI(campaign) + random house

                                                        The MIXED Procedure
                               Cov Parm
   Covariance Parameter Estimates (REML)

     Ratio	Estimate     Std Error
                                                                                        Z  Pr >  !Z!
                              DID
                              Residual
0.02848508
1.00000000
0.01173763
0.41206246
0.01986539
0.03944356
 0.59
10.45
0.5546
0.0000
                                                       The MIXED Procedure
                                              Model Fitting Information for FACTOR1
                                             Description
                                                                                Value
Observations 293.0000
Variance Estimate 0.4121
Standard Deviation Estimate 0.6419
REML Log Likelihood -299.850
Akaike's Information Criterion -301.850
Schwartz's Bayesian Criterion -305.466
-2 REML Log Likelihood 599.6990
Null Model LRT Chi -Square 0.3884
Null Model LRT DF 1.0000
Null Model LRT P-Value 0.5331
The MIXED Procedure
Solution for Fixed Effects
Parameter
INTERCEPT
SPRING
SUMMER
FALL
CAMPAIGN 00
CAMPAIGN 02
CAMPAIGN 06
CAMPAIGN 12
CAMPAIGN PI
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL (CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL (CAMPAIGN)
LEVEL (CAMPAIGN)
LEVEL(CAMPAIGN)
LEVEL (CAMPAIGN)











3
2
1
3
2
1
3
2
1
3
2
1
3
2
1











00
00
00
02
02
02
06
06
06
12
12
12
PI
PI
PI

Estimate
-0.17237997
-0.23095597
-0.24665377
-0.28858532
0.83066936
0.09064261
0.10539994
0.07875213
0.00000000
1.54003029
1.06924744
0.00000000
-0.05521327
-0.02248289
0.00000000
-0.16805480
-0.09292245
0.00000000
-0.09261073
-0.00173360
0.00000000
-0.10243114
-0.10783173
0.00000000

Std Error
0.15458120
0.11266415
0.10469955
0.12244765
0.20630818
0.21254020
0.20433152
0.20606596
m
0.20991871
0.20881251
•
0.21872343
0.21459388
•
0.20669036
0.20586405
•
0.20962813
0.20865833
•
0.21105977
0.20823255
•
The MIXED
DDF
218
218
218
218
218
218
218
218
.
218
218
•
218
218
•
218
218
•
218
218
•
218
218
•
T
-1.12
•2.05
-2.36
-2.36
4.03
0.43
0.52
0.38
_
7.34
5.12
•
•0.25
•0.10
•
-0.81
-0.45
•
-0.44
-0.01
•
-0.49
-0.52
•
Pr > !T!
0.2660
0.0416
0.0194
0.0193
0.0001
0.6702
0.6065
0.7027

0.0000
0.0000
•
0.8009
0.9167
•
0.4171
0.6522
•
0.6591
0.9934
•
0.6279
0.6051
•
Aloha
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05

0.05
0.05
.
0.05
0.05
.
0.05
0.05
•
0.05
0.05
.
0.05
0.05
•

-0
-0
-0
-0
0
-0
-0
-0

1
0

-0
-0

Lower
.4770
.4530
.4530
.5299
.4241
.3283
.2973
.3274

.1263
.6577
.
.4863
.4454
.
-0.5754
-0.4987

.
-0.5058
-0.4130

.
-0.5184
-0.5182

.

0.1323
-0.0089
-0.0403
-0.0473
1.2373
0.5095
0.5081
0.4849

1 .9538
1 .4808

0.3759
0.4005

0.2393
0.3128

0.3205
0.4095

0.3135
0.3026

Procedure
Tests of Fixed Effects















Source
SPRING
SUMMER
FALL
CAMPAIGN
NDF
1
1
1
4
LEVEL(CAMPAIGN) 10
DDF Tvoe IMF
218
218
218
218
218
4.20
5.55
5.55
78.49
5.83
Pr > F
0.0416
0.0194
0.0193
0.0000
0.0000
























UST3.LIS   (source)
               Page  B-7

-------
                                              R&M Study   Preliminary 12-Month Report
                                                        The NIXED Procedure

                                                      Class Level Information

                                          Class     Levels  Values

                                          DID           89  102 103 104 105 106 110 111
                                                            112 120 121 127 131 135 136
                                                            306 309 317 320 323 325 326
                                                            328 332 336 383 389 390 402
                                                            404 405 411 413 415 417 302
                                                            312 335 338 345 350 354 365
                                                            370 374 376 377 384 385 391
                                                            397 407 412 416 420 310 316
                                                            343 347 348 353 355 357 358
                                                            359 366 372 373 381 386 403
                                                            406 409 418 419 204 205 207
                                                            209 210 212 214 215 217 223
                                                            224 227 233 234 235
                                          GROUP          5  Abated Level 1  Level2 Levels
                                                            Modern
                                          CAMPAIGN       3  06 12 IN
HJST5.LIS   (source)                                         Page   B-8

-------
                                             R&M Study   Preliminary 12-Month Report
                        DUST LEAD LOADINGS WITHIN HOUSE AS AN OUTCOME OF ENVIRONMENTAL FACTORS:  07/17/95
                   12-month campaign all data model: factorl = season + group + campaignCgroup)  + randomjiouse

                                                       The MIXED Procedure
                                              Covanance Parameter Estimates (REML)
                              Cov Parm
     Ratio
                                                           Estimate
                             Std Error
                               Z  Pr > !2!
                              DID
                              Residual
0.80897052
1.00000000
0.09396472
0.11615345
0.02098848
0.01284375
4.48
9.04
0.0000
0.0000
                                                       The MIXED Procedure
                                              Model Fitting Information for  FACTOR1
                                             Description
                                     Value
Observations 265.0000
Variance Estimate 0.1162
Standard Deviation Estimate 0.3408
REML Log Likelihood -162.465
Akaike's Information Criterion -164.465
Schwartz's Bayesian Criterion -167.975
-2 REML Log Likelihood 324.9307
Null Model LRT Chi -Square 45.0299
Null Model LRT OF 1.0000
Null Model LRT P-Value 0.0000
The MIXED Procedure
Solution for Fixed Effects
Parameter
INTERCEPT
SPRING
SUMMER
FALL
GROUP Abated
GROUP Level 1
GROUP Level 2
GROUP Level3
GROUP Modern
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN (GROUP)
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)










06
12
IN
06
12
IN
06
12
IN
06
12
IN
06
12
IN










Abated
Abated
Abated
L eve 11
Level 1
Levell
Level2
Level 2
Level2
Level3
LevelJ
Level3
Modern
Modern
Modern
Estimate
-1.05314514
-0.05772997
•0.10581868
-0.08705832
1.03232696
2.21960088
2.49893493
2.72239565
0.00000000
-0.19014737
-0.17750863
0.00000000
-0.71012419
-0.72935979
0.00000000
•1.49895219
-1.49883812
0.00000000
-2.29129807
-2.40480584
0.00000000
-0.07346004
-0.21028855
0.00000000
Std Error
0.12262196
0.06871415
0.06197884
0.08076030
0.17077146
0.15908373
0.15779129
0.15759853
f
0.14028701
0.12925200
•
0.11081962
0.11175151
•
0.10867341
0.10811371
•
0.10899821
0.10980518
•
0.13623480
0.12496062
.
DDF
163
163
163
163
163
163
163
163
.
163
163
•
163
163
•
163
163
•
163
163
•
163
163
.
T
-8.59
-0.84
-1.71
-1.08
6.05
13.95
15.84
17.27
.
-1.36
-1.37
•
-6.41
-6.53
•
-13.79
-13.86
•
-21.02
-21.90
•
-0.54
-1.68
.
Pr > IT!
0.0000
0.4021
0.0897
0.2826
0.0000
0.0000
0.0000
0.0000
_
0.1772
0.1715
•
0.0000
0.0000
•
0.0000
0.0000
•
0.0000
0.0000
•
0.5905
0.0943
.
Aloha
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05

0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
m

-1
-0
-0
-0
0
1
2
2

-0
•0

-0
-0

-1
•1

Lower
.2953
.1934
.2282
.2465
.6951
.9055
.1874
.4112

.4672
.4327
•
.9290
.9500
•
.7135
.7123
•
-2.5065
-2.6216

•
-0.3425
-0.4570

_
Uooer
-0.8110
0.0780
0.0166
0.0724
1 .3695
2.5337
2.8105
3.0336

0.0869
0.0777
•
-0.4913
-0.5087
•
-1.2844
-1.2854
.
-2.0761
-2.1880
.
0.1956
0.0365
m
                                        Source
           The MIXED Procedure
         Tests of Fixed Effects

               NDF   DDF  Type IMF   Pr  >  F
SPRING
SUMMER
FALL
GROUP
CAMPAIGN (GROUP)
1
1
1
4
10
163
163
163
163
163
0.71
2.91
1.16
61.47
92.23
0.4021
0.0897
0.2826
0.0000
0.0000
JST5.LIS   (source)
                Page  B-9

-------
                                             R&M Study   Preliminary 12-Month Report
                         OUST  LEAD  LOADINGS WITHIN HOUSE AS AN OUTCOME OF ENVIRONMENTAL FACTORS: 07/17/95
                    12-month campaign all data model: factorl = season + campaign group(canpaign) + random house
                              Cov Parm
                                                       The MIXED Procedure
                                              Covariance Parameter Estimates (REML)
     Ratio
  Estimate
                                                                        Std Error
                                                                                        Z  Pr > !Z!
                              DID
                              Residual
0.80897052
1.00000000
0.09396472
0.11615345
0.02098848
0.01284375
4.48
9.04
0.0000
0.0000
                                                       The MIXED Procedure
                                              Model Fitting Information for FACTOR1
                                             Description
                                                                                Value
                                             Observations                    265.0000
                                             Variance Estimate                 0.1162
                                             Standard Deviation Estimate       0.3408
                                             REML Log Likelihood             -162.465
                                             Akaike's Information Criterion  -164.465
                                             Schwartz's Bayesian Criterion   -167.975
                                             -2 REML Log Likelihood          324.9307
                                             Null Model LRT Chi-Square        45.0299
                                             Null Model LRT DF                 1.0000
                                             Null Model LRT P-Value            0.0000
                                                       The MIXED Procedure
                                                   Solution for Fixed Effects
Parameter
INTERCEPT
SPRING
SUMMER
FALL
CAMPAIGN 06
CAMPAIGN 12
CAMPAIGN IN
GROUP ( CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP (CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP (CAMPAIGN)








Abated
Level 1
Level 2
Level3
Modern
Abated
Level!
Level2
Level3
Modern
Abated
Level 1
Level2
Level3
Modern








06
06
06
06
06
12
12
12
12
12
IN
IN
IN
IN
IN
Estimate
-1.05314514
-0.05772997
-0.10581868
-0.08705832
-0.07346004
-0.21028855
0.00000000
0.91563962
1.58293672
1 .07344278
0.50455762
0.00000000
1.06510688
1.70052964
1.21038536
0.52787836
0.00000000
1.03232696
2.21960088
2.49893493
2.72239565
0.00000000
Std Error
0.12262196
0.06871415
0.06197884
0.08076030
0.13623480
0.12496062

0.
0.
0.
0.

0.
0.
0.
0.

0.
0.
0.
0.

.
17051242
15925483
15925483
15883916
•
17047779
16010886
15953155
15837732
•
17077146
15908373
15779129
15759853
.
00 F
163
163
163
163
163
163
_
163
163
163
163
•
163
163
163
163
-
163
163
163
163
.
T
-8.59
-0.84
•1.71
-1.08
-0.54
-1.68

5.37
9.94
6.74
3.18
•
6.25
10.62
7.59
3.33
•
6.05
13.95
15.84
17.27
.
Pr > IT!
0.0000
0.4021
0.0897
0.2826
0.5905
0.0943

0.0000
0.0000
0.0000
0.0018
.
0.0000
0.0000
0.0000
0.0011
.
0.0000
0.0000
0.0000
0.0000
.
Aloha
0.05
0.05
0.05
0.05
0.05
0.05

0.05
0.05
0.05
0.05
.
0.05
0.05
0.05
0.05

0.05
0.05
0.05
0.05

Lower
-1.2953
-0.1934
-0.2282
-0.2465
-0.3425
-0.4570

0.5789
1.2685
0.7590
0.1909

0.7285
1.3844
0.8954
0.2151

0.6951
1 .9055
2.1874
2.4112


-0.8110
0.0780
0.0166
0.0724
0.1956
0.0365

1.2523
1 .8974
1.3879
0.8182
•
1.4017
2.0167
1.5254
0.8406

1.3695
2.5337
2.8105
3.0336

                                        Source
           The MIXED Procedure
         Tests of Fixed Effects

               NDF   DDF  Type IMF   Pr  >  F
SPRING
SUMMER
FALL
CAMPAIGN
GROUP (CAMPAIGN)
1
1
1
2
12
163
163
163
163
163
0.71
2.91
1.16
210.76
49.20
0.4021
0.0897
0.2826
0.0000
0.0000
UST5.LIS   (source)
                                                          Page  B-10

-------
                                              R&M Study   Preliminary 12-Month  Report
                  DUST LEAD CONCENTRATIONS WITHIN AN R_M HOUSE AS AN OUTCOME OF ENVIRONMENTAL  FACTORS: 07/17/95
                      12-month campaign all model: faclorl =  season * group +  campaign(group)  +  random_house

                                                        The MIXED Procedure
                                               Covanance Parameter  Estimates (REML)
                               Cov Parm
                               DID
                               Residual
     Ratio
  Estimate
 Std Error
                                                                                         Z  Pr  >  !Z!
1.44284416
1.00000000
0.16851088
0.11679077
0.03246000
0.01293660
5.19
9.03
0.0000
0.0000
                                                        The  MIXED  Procedure
                                               Model  Fitting Information  for  FACTOR1
                                              Description
                                                                                Value
                                              Observations
                                              Variance  Estimate
                                              Standard  Deviation Estimate
                                              REML  Log  Likelihood
                                              Akaike's  Information Criterion
                                              Schwartz's Bayesian Criterion
                                              -2  REML Log Likelihood
                                              Null  Model LRT Chi-Square
                                              Null  Model LRT OF
                                              Null  Model LRT P-Value

                                                       The MIXED Procedure
                                                   Solution for Fixed Effects
            Parameter
                                            Estimate
                                                         Std Error   DDF
                                 265.0000
                                   0.1168
                                   0.3417
                                 -181.573
                                 -183.573
                                 -187.083
                                 363.1463
                                  81.9813
                                   1.0000
                                   0.0000
                                   T  Pr > IT!  Aloha
                                                                                                      Lower
INTERCEPT
SPRING
SUMMER
FALL
GROUP Abated
GROUP L eve 11
GROUP Level2
GROUP Level3
GROUP Modern
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN(GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)









06
12
IN
06
12
IN
06
12
IN
06
12
IN
06
12
IN









Abated
Abated
Abated
L eve 11
Level!
Level!
Level2
Level 2
Level 2
Level3
Level3
Level3
Modern
Modern
Modern
-1.54143049
0.02539044
0.05558741
0.07776581
1.33661492
2.53300167
2.48788625
2.60867180
0.00000000
-0.15247858
0.03286790
0.00000000
-0.37402960
-0.28932540
0.00000000
-0.77923434
-0.89675944
0.00000000
-1.45873074
-1. 65968050
0.00000000
0.02241664
0.04948173
0.00000000
0
0
0
0
0
0
0
0

0
0

0
0

0
0

.14186450
.07124566
.06271464
.08294464
.1 9888784
.18468302
.1 8352396
.1 8336602
m
.1 4092658
.12962777
•
.11125439
.11227454
•
.10899098
.10841849
•
0.10933164
0.11019498

•
0.13686564
0.12533191

.
163
163
163
163
163
163
163
163
m
163
163
•
163
163
•
163
163
•
163
163
•
163
163
.
-10.87
0.36
0.89
0.94
6.72
13.72
13.56
14.23
m
-1.08
0.25
•
-3.36
-2.58
•
-7.15
-8.27
•
-13.34
-15.06
•
0.16
0.39
.
0.0000
0.7220
0.3767
0.3499
0.0000
0.0000
0.0000
0.0000
.
0.2809
0.8002
•
0.0010
0.0109
•
0.0000
0.0000
•
0.0000
0.0000
•
0.8701
0.6935
.
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
_
0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
.
0.05
0.05
.
-1.8216
•0.1153
-0.0683
-0.0860
0.9439
2.1683
2.1255
2.2466

-0.4308
-0.2231
•
-0.5937
-0.5110
•
-0.9945
-1.1108
•
-1.6746
-1.8773
.
-0.2478
-0.1980
.
-1.2613
0.1661
0.1794
0.2416
1.7293
2.8977
2.8503
2.9708

0.1258
0.2888
•
-0.1543
-0.0676
.
-0.5640
-0.6827
•
-1.2428
-1.4421
.
0.2927
0.2970
.
                                         Source
           The MIXED  Procedure
         Tests of  Fixed  Effects

               NDF    OOF  Type IMF  Pr  >  F
SPRING
SUMMER
FALL
GROUP
CAMPAIGN (GROUP)
1
1
1
4
10
163
163
163
163
163
0.13
0.79
0.88
59.63
36.27
0.7220
0.3767
0.3499
0.0000
0.0000
5UST5.LIS    (source)
                                                           Page  B-11

-------
                                             R&H Study   Preliminary 12-Month Report
                 DUST LEAD CONCENTRATIONS WITHIN AN R_M HOUSE AS AN OUTCOME OF ENVIRONMENTAL FACTORS:  07/17/95
                  12-month campaign all data model: factor! = season + campaign + group(campaign) + randomjiouse

                                                       The MIXED Procedure
                                              Covariance Parameter Estimates (REML)
                              Cov Perm
                                                Ratio
                                                           Estimate
                                                                        Std Error
                                             Z   Pr  >  'Z!
                              DID
                              Residual
1.44284416
1.00000000
0.16851088
0.11679077
0.03246000
0.01293660
5.19
9.03
0.0000
0.0000
                                                       The MIXED Procedure
                                              Model Fitting Information for  FACTOR1
                                             Description
                                    Value
            Parameter
  Observations                    265.0000
  Variance Estimate                 0.1168
  Standard Deviation  Estimate       0.3417
  REML  Log Likelihood            -181.573
  Akaike's Information Criterion  -183.573
  Schwartz's  Bayesian Criterion   -187.083
  -2  REML  Log Likelihood          363.1463
  Null  Model  LRT Chi-Square        81.9813
  Null  Model  LRT OF                 1.0000
  Null  Model  LRT P-Value            0.0000

           The MIXED Procedure
        Solution for  Fixed Effects

Estimate     Std Error   DDF
                                                                              T  Pr >  !T!  Aloha
INTERCEPT
SPRING
SUMMER
FALL
CAMPAIGN 06
CAMPAIGN 12
CAMPAIGN IN
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP (CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)







Abated
Level 1
Level2
Levels
Modern
Abated
Levell
Level 2
Level 3
Modern
Abated
Levell
Level2
Level3
Modern







06
06
06
06
06
12
12
12
12
12
IN
IN
IN
IN
IN
-1.54143049
0.02539044
0.05558741
0.07776581
0.02241664
0.04948173
0.00000000
1.16171970
2.13655543
1.68623527
1.12752442
0.00000000
1.32000108
2.19419454
1.54164508
0.89950957
0.00000000
1.33661492
2.53300167
2.48788625
2.60867180
0.00000000
0.14186450
0.07124566
0.
0.
0.
0.

0.
0.
0.
0.

0.
06271464
08294464
13686564
12533191
m
19864512
18485584
18485584
18446740
•
19861562
0.18566652
0.
0.

0.
0.
0.
0.

18510615
18403946
•
19888784
18468302
18352396
18336602
.
163
163
163
163
163
163
^
163
163
163
163
•
163
163
163
163
•
163
163
163
163
.
-10.87
0.36
0.89
0.94
0.16
0.39
m
5.85
11.56
9.12
6.11
•
6.65
11.82
8.33
4.89
•
6.72
13.72
13.56
14.23
.
0.0000
0.7220
0.3767
0.3499
0.8701
0.6935
a
0.0000
0.0000
0.0000
0.0000
•
0.0000
0.0000
0.0000
0.0000
•
0.0000
0.0000
0.0000
0.0000
.
0.05
0.05
0.05
0.05
0.05
0.05
.
0.05
0.05
0.05
0.05
•
0.05
0.05
0.05
0.05
•
0.05
0.05
0.05
0.05
.
-1.8216
-0.1153
-0.0683
-0.0860
-0.2478
-0.1980

0.7695
1.7715
1.3212
0.7633
•
0.9278
1 .8276
1.1761
0.5361
•
0.9439
2.1683
2.1255
2.2466
.
-1.2613
0.1661
0.1794
0.2416
0.2927
0.2970

1.5540
2.5016
2.0513
1.4918
.
1.7122
2.5608
1.9072
1.2629
•
1.7293
2.8977
2.8503
2.9708
.
                                        Source
           The MIXED Procedure
         Tests of Fixed Effects

               NDF   DDF  Type IMF  Pr >  F
SPRING
SUMMER
FALL
CAMPAIGN
GROUP( CAMPAIGN)
1
1
1
2
12
163
163
163
163
163
0.13
0.79
0.88
65.30
34.72
0.7220
0.3767
0.3499
0.0000
0.0000
UST5.LIS   (source)
               Page  B-12

-------
                                             R&M Study   Preliminary 12-Month Report
                       DUST LOADINGS WITHIN AN R_H HOUSE AS AN OUTCOME OF ENVIRONMENTAL FACTORS: 07/17/95
                    12-month campaign all data model: factor! =  season + group + campaign(group) + random house

                                                       The MIXED Procedure
                                              Covariance Parameter Estimates (REHL)
                              Cov Parm
                              DID
                              Residual
                                                Ratio
                Estimate
                                                                        Std Error
                                                                                        Z  Pr > !Z!
0.07308933
1.00000000
0.03213037
0.43960407
0.03218445
0.048S221S
1.00
9.06
0.3181
0.0000
                                                       The MIXED Procedure
                                              Model Fitting Information for FACTOR1
                                             Description
                                     Value
Observations 265.0000
Variance Estimate 0.4396
Standard Deviation Estimate 0.6630
REML Log Likelihood -283.905
Alt a ike's Information Criterion -285.905
Schwartz's Bayesian Criterion -289.414
-2 REML Log Likelihood 567.8100
Null Model LRT Chi -Square 1.1004
Null Model LRT OF 1.0000
Null Model LRT P-Value 0.2942
The MIXED Procedure
Solution for Fixed Effects
Parameter
INTERCEPT
SPRING
SUMMER
FALL
GROUP Abated
GROUP Level 1
GROUP Level2
GROUP Level3
GROUP Modern
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)










06
12
IN
06
12
IN
06
12
IN
06
12
IN
06
12
IN










Abated
Abated
Abated
Level 1
Level 1
Level 1
Level2
Level 2
Level 2
Level3
Level3
Level3
Modern
Modern
Modern

-0
-0
-0
-0
0
0
1
2
0
-0
-0
0
-0
Estimate
.06361615
.26336293
.26787536
.39004848
.12970839
.62413482
.65035420
.08996275
.00000000
.05159171
.23083312
.00000000
.69699967
-0.73166898
Std Error
0.18551362
0.11670888
0.11564992
0.14246430
0.25606076
0.24012806
0.23736687
0.23681208
m
0.27075516
0.25130120
•
0.21461631
0.21584130
DDF
163
163
163
163
163
163
163
163
m
163
163
•
163
163
T
-0.34
-2.26
-2.32
-2.74
0.51
2.60
6.95
8.83
_
-0.19
-0.92
•
-3.25
-3.39
Pr > !T!
0.7321
0.0254
0.0218
0.0069
0.6132
0.0102
0.0000
0.0000

0.8491
0.3597
•
0.0014
0.0009
Aloha
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05

0.05
0.05
•
0.05
0.05
Lower
-0.4299
-0.4938
-0.4962
-0.6714
-0.3759
0.1500
1.1816
1.6223

-0.5862
-0.7271
.
-1.1208
-1.1579

0.3027
-0.0329
-0.0395
-0.1087
0.6353
1.0983
2.1191
2.5576

0.4830
0.2654
.
-0.2732
-0.3055
0.00000000 ... . .
-1.81109843
-1.75627120
0.21124969
0.21025838
163
163
-8.57
-8.35
0.0000
0.0000
0.05
0.05
-2.2282
-2.1715
-1.3940
-1.3411
0.00000000 ... .
-2.31890512
-2.27389472
0.21176967
0.21299118
163
163
-10.95
-10.68
0.0000
0.0000
0.05
0.05
-2.7371
-2.6945
-1.9007
-1.8533
0.00000000 ... . .
0.08782490
-0.18976017
0.26283695
0.24290377
163
163
0.33
-0.78
0.7387
0.4358
0.05
0.05
-0.4312
-0.6694
0.6068
0.2899
0.00000000 ... . .
                                        Source
           The MIXED Procedure
         Tests of Fixed Effects

               NDF   DDF  Type III  F   Pr >  F
SPRING
SUMMER
FALL
GROUP
CAMPAIGN (GROUP)
1
1
1
4
10
163
163
163
163
163
5.09
5.37
7.50
6.61
26.80
0.0254
0.0218
0.0069
0.0001
0.0000
JST5.LIS   (source)
               Page  B-13

-------
                                              RSM Study   Preliminary 12-Honth Report
                         DUST  LOADINGS  WITHIN AN R_H  HOUSE AS AN OUTCOME OF ENVIRONMENTAL FACTORS: 07/17/95
                   12-month  campaign  all data model:  factod =  season + campaign * group(campaign) + randomjiouse
                               Cov Parm
                                                        The MIXED Procedure
                                               Covariance Parameter Estimates (REML)
                                                 Ratio
                Estimate
                                                                         Std Error
                                             Z   Pr >  \2\
                               010
                               Residual
0.07308933
1.00000000
0.03213037
0.43960407
0.03218445
0.04852215
1.00
9.06
0.3181
0.0000
                                                        The MIXED Procedure
                                               Model Fitting Information for FACTOR1
                                              Description
                                     Value
Observations 265.0000
Variance Estimate 0.4396
Standard Deviation Estimate 0.6630
REML Log Likelihood -283.905
Akaike's Information Criterion -285.905
Schwartz's Bayesian Criterion -289.414
-2 REML Log Likelihood 567.8100
Null Model LRT Chi -Square 1.1004
Null Model LRT OF 1.0000
Null Model LRT P-Value 0.2942
The MIXED Procedure
Solution for Fixed Effects
Parameter
INTERCEPT
SPRING
SUMMER
FALL
CAMPAIGN 06
CAMPAIGN 12
CAMPAIGN IN
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP( CAMPAIGN)
GROUP(CAMPAIGN)
GROUP( CAMPAIGN)
GROUP( CAMPAIGN)
GROUP(CAHPAIGN)
GROUP( CAMPAIGN)
GROUP(CAMPAIGN)
GROUP( CAMPAIGN)








Abated
Level 1
Level 2
Levels
Modern
Abated
Level 1
Level 2
Levels
Modern
Abated
Levell
Level 2
LevelS
Modern








06
06
06
06
06
12
12
12
12
12
IN
IN
IN
IN
IN
Estimate
-0.06361615
-0.26336293
-0.26787536
-0.39004848
0.08782490
•0.18976017
0.00000000
-0.00970822
-0.16068975
-0.24856913
-0.31676727
0.00000000
0.08863544
0.08222601
0.08384317
0.00582820
0.00000000
0.12970839
0.62413482
1.65035420
2.08996275
0.00000000

0
0
0
0
0
0

0
0
0
0

0
0
0
0

Std Error
.18551362
.11670888
.11564992
.14246430
.26283695
.24290377
t
.25559433
.24034132
.24034132
.23959103
•
.25550685
.24177361
.24093726
.23872095
•
0.25606076
0.24012806
0.23736687
0.23681208

m
DDF
163
163
163
163
163
163
.
163
163
163
163
•
163
163
163
163
•
163
163
163
163
m
T
-0.34
-2.26
-2.32
-2.74
0.33
-0.78
.
-0.04
-0.67
-1.03
-1.32
•
0.35
0.34
0.35
0.02
•
0.51
2.60
6.95
8.83
f
Pr > IT!
0.7321
0.0254
0.0218
0.0069
0.7387
0.4358
m
0.9697
0.5047
0.3026
0.1880
•
0.7291
0.7342
0.7283
0.9806
•
0.6132
0.0102
0.0000
0.0000
.
Aloha
0.05
0.05
0.05
0.05
0.05
0.05

0.05
0.05
0.05
0.05
•
0.05
0.05
0.05
0.05
.
0.05
0.05
0.05
0.05
_
Lower
-0.4299
-0.4938
-0.4962
-0.6714
-0.4312
-0.6694

-0.5144
-0.6353
-0.7232
-0.7899
.
-0.4159
-0.3952
-0.3919
-0.4656
.
-0.3759
0.1500
1.1816
1.6223

UDOer
0.3027
-0.0329
•0.0395
-0.1087
0.6068
0.2899

0.4950
0.3139
0.2260
0.1563
.
0.5932
0.5596
0.5596
0.4772
.
0.6353
1.0983
2.1191
2.5576

                                         Source
           The MIXED Procedure
         Tests of Fixed Effects

               NDF   PDF  Type 111  F   Pr  >  F
SPRING
SUMMER
FALL
CAMPAIGN
GROUP (CAMPAIGN)
1
1
1
2
12
163
163
163
163
163
5.09
5.37
7.50
58.99
10.74
0.0254
0.0218
0.0069
0.0000
0.0000
DUSTS.LIS   (source)
                                                           Page  B-14

-------
                                    Preliminary R&N Study Report
                                    Partial 12  Month Data   7/95
               APPENDIX C
Longitudinal data analysis output for blood lead

-------
                                                 R&H Study - Preliminary 12-Month
                                    CHILD  BLOOD AS AN OUTCOME OF GROUP AND  CAMPAIGN:  07/21/95
                                        COMPARISON MODEL -  3 R&M Groups,  Initial  PbB  <  20 jsg/dL

                                                        The MIXED Procedure
                                               Covariance Parameter Estimates  (REML)
                            Cov Parm
                                                    Ratio
          Estimate
                                                                            Std Error
                                       Z  Pr > !Z!
DID
CHILDNUM(DID)
Residual
0.85205541 0.07286326 0.05414210 1.35
0.96599689 0.08260693 0.04623754 1.79
1.00000000 0.08551470 0.01020211 8.38
The MIXED Procedure
Model Fitting Information for LNBLOOD
Desc riot ion Value
0.1784
0.0740
0.0000

Observations 216.0000
Variance Estimate 0.0855
Standard Deviation Estimate 0.2924
REML Log Likelihood -127.523
Akaike's Information Criterion -130.523
Schwartz's Bayesian Criterion -135.471
-2 REML Log Likelihood 255.0463
Null Model LRT Chi -Square 92.5209
Null Model LRT DF 2.0000
Null Model LRT P-Value 0.0000
The MIXED Procedure
Solution for Fixed Effects
Parameter
INTERCEPT
AGE
AGESQ
SUMMER
MALE
GROUP L1
GROUP L2
GROUP L3
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN (GROUP)
CANPAIGN(GROUP)



1
Estimate
.64501827
0.03230535






12 LI
6 LI
2 LI
0 L1
12 L2
6 L2
2 L2
0 L2
12 L3
6 L3
2 L3
0 L3
-0
0
0
•0
•0
0
-0
0
0
0
.00034535
.21044428
.06101732
.22873767
.02765123
.00000000
.11238864
.09320553
.00011891
.00000000
0.08397407
0.07399752
0.05035725

0
0
0
0
0
0
0

0
Std Error
.22727941
.00901870
.00011024
.05149566
.10689932
.15750261
.16968811
•
.11601262
0.10094505
0.09336397

•
0.11341595
0.10589662
0.10518037
DDF
139
139
139
139
139
139
139
•
139
139
139
.
139
139
139
T
7.24
3.58
-3.13
4.09
0.57
-1.45
-0.16
•
-0.97
0.92
0.00
.
0.74
0.70
0.48
Pr > !T!
0.0000
0.0005
0.0021
0.0001
0.5691
0.1487
0.8708
.
0.3343
0.3574
0.9990
.
0.4603
0.4859
0.6329

0.05
0.05
0.05
0.05
0.05
0.05
0.05
•
0.05
0.05
0.05
.
0.05
0.05
0.05
Lower
1.1956
0.0145
-0.0006
0.1086
-0.1503
-0.5401
-0.3632
.
-0.3418
-0.1064
-0.1845

-0.1403
-0.1354
-0.1576
0.00000000 ... .
-0.12363470
-0.09071779
-0.03502872
0.11155345
0.09824989
0.10107311
139
139
139
-1.11
-0.92
-0.35
0.2696
0.3574
0.7294
0.05
0.05
0.05
-0.3442
-0.2850
-0.2349
0.00000000 ... .

2.0944
0.0501
-0.0001
0.3123
0.2724
0.0827
0.3079

0.1170
0.2928
0.1847

0.3082
0.2834
0.2583

0.0969
0.1035
0.1648

                                        Source
  The MIXED Procedure
Tests of Fixed Effects

      NDF   DDF  Type IMF  Pr > F
AGE
AGESQ
SUMMER
MALE
GROUP
CAMPAIGN (GROUP)
1
1
1
1
2
9
139
139
139
139
139
139
12.83
9.81
16.70
0.33
1.79
0.72
0.0005
0.0021
0.0001
0.5691
0.1712
0.6922
COM3A.LIS  (source)
                                                           Page  C-1

-------
                                                R&M Study -  Preliminary 12-Month
                                   CHILD  BLOOD AS AN OUTCOME OF GROUP AMD CAMPAIGN: 07/21/95
                                       COMPARISON MODEL - 3 R&M Groups,  Initial PbB < 20 j/g/dL

                                                      The MIXED Procedure
                                             Covanance Parameter Estimates (REML)
                           Cov Parm
                                                  Ratio
          Estimate
                                                                          Std Error
                                                                                          Z  Pr > !2!
DID
CHILDNUN(DID)
Residual
0.85205541
0.96599689
1.00000000
0.07286326
0.08260693
0.08551470
0.05414210
0.04623756
0.01020211
1.35
1.79
8.38
0.1784
0.0740
0.0000
                                                      The MIXED Procedure
                                             Model Fitting Information for LNBLOOD
                                            Description
                            Value
                                            Observations                    216.0000
                                            Variance Estimate                 0.0855
                                            Standard Deviation Estimate       0.2924
                                            REML Log Likelihood             -127.523
                                            Akaike's Information Criterion  -130.523
                                            Schwartz's Bayesian Criterion   -135.471
                                            -2 REML Log Likelihood          255.0463
                                            Null Model LRT Chi-Square        92.5209
                                            Null Model LRT DF                 2.0000
                                            Null Model LRT P-Value            0.0000

                                                      The MIXED Procedure
                                                  Solution for Fixed Effects
             Parameter
                                         Estimate
                                                      Std Error   DDF
                        TPr >  !T!  Alpha
                                                                                                   Lower
INTERCEPT
AGE
AGESQ
SUMMER
MALE
CAMPAIGN 12
CAMPAIGN 6
CAMPAIGN 2
CAMPAIGN 0
GROUPC CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP( CAMPAIGN)
GROUP (CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP (CAMPAIGN)
GROUP (CAMPAIGN)









LI
L2
L3
LI
L2
L3
LI
L2
L3
LI
L2
L3









12
12
12
6
6
6
2
2
2
0
0
0
1.64501827
0.03230535
-0.00034535
0.21044428
0.06101732
-0.12363470
-0.09071779
-0.03502872
0.00000000
-0.21749161
0.17995755
0.00000000
-0.04481435
0.13706408
0.00000000
-0.19359004
0.05773474
0.00000000
-0.22873767
-0.02765123
0.00000000
0.
22727941
0.00901870
0.
0.
0.
0.
0.
0.

0.
0.

0.
0.

0.
0.

0.
0.

00011024
05149566
10689932
11155345
09824989
10107311
•
17169710
17870556
•
16234649
17455214
•
16287630
17839586
•
15750261
16968811
.
139
139
139
139
139
139
139
139
•
139
139
•
139
139
•
139
139
•
139
139
.
7.24
3.58
-3.13
4.09
0.57
-1.11
-0.92
-0.35
•
-1.27
1.01
•
-0.28
0.79
•
-1.19
0.32
•
-1.45
-0.16
.
0.0000
0.0005
0.0021
0.0001
0.5691
0.2696
0.3574
0.7294
•
0.2074
0.3157
•
0.7829
0.4337
•
0.2366
0.7467
•
0.1487
0.8708
•
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
.
1.1956
0.0145
-0.0006
0.1086
-0.1503
-0.3442
-0.2850
-0.2349
•
-0.5570
-0.1734
•
-0.3658
-0.2081
•
-0.5156
-0.2950
•
-0.5401
-0.3632
.
2.0944
0.0501
-0.0001
0.3123
0.2724
0.0969
0.1035
0.1648
•
0.1220
0.5333
•
0.2762
0.4822
.
0.1284
0.4105
•
0.0827
0.3079
u
                                        Source
  The MIXED Procedure
Tests of Fixed Effects

      NDF   DDF  Type 111  F  Pr >  F
AGE
AGESQ
SUMMER
HALE
CAMPAIGN
GROUP (CAMPAIGN)
1
1
1
1
3
8
139
139
139
139
139
139
12.83
9.81
16.70
0.33
0.48
1.05
0.0005
0.0021
0.0001
0.5691
0.6936
0.4025
:OM3A.LIS  (source)
       Page   C-2

-------
                                             R&N Study - Preliminary 12-Month Report
                                   CHILD BLOOD AS AN OUTCOME OF CAMPAIGN AND GROUP: 07/18/95
                                    COMPARISON MODEL • 5 STUDY GROUPS,  INITIAL PbB  <20 //g/dL
                                                      Excluding Child 120-3

                                                       The MIXED Procedure
                                              Covariance Parameter Estimates (REML)
                           Cov Parm
                                                   Ratio
          Estimate
Std Error
                                                                                           Z  Pr > !Z!
DID 0.04259006 0.00389418 0.05408292 0.07 0.
CHILDNUH(DID) 1.16521862 0.10654064 0.05590865 1.91 0.
Residual 1.00000000 0.09143404 0.01056290 8.66 0.
The MIXED Procedure
Model Fitting Information for LNBLOOD
Description Value
9426
0567
0000

Observations 258.0000
Variance Estimate 0.0914
Standard Deviation Estimate 0.3024
REML Log Likelihood -156.727
Akaike's Information Criterion -159.727
Schwartz's Bayesian Criterion -164.935
-2 REML Log Likelihood 313.4532
Null Model LRT Chi -Square 63.3846
Null Model LRT DF 2.0000
Null Model LRT P-Value 0.0000
The MIXED Procedure
Solution for Fixed Effects
Parameter
INTERCEPT
AGE
AGESO
SUMMER
MOUTH
MALE
GROUP Abated
GROUP Levell
GROUP LevelZ
GROUP Level3
GROUP Modern
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN(GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)












12 Abated
6 Abated
0 Abated
12 Levell
6 Levell
0 Levell
12 Level2
6 LcvelZ
0 Level2
12 Levels
6 Level3
0 Level3
12 Modern
6 Modern
0 Modern
Estimate
0.50079743
0.02793169
-0.00030776
0.23637030
0.02567014
-0.02425782
1.37964782
1.02492547
1.23246481
1.30079046
0.00000000
-0.07624198
0.03998553
0.00000000
•0.07493976
0.12636933
0.00000000
0.09055015
0.06079373
0.00000000
-0.12328829
-0.08705515
0.00000000
0.01684491
0.29389836
0.00000000

0
0
0
0
0
0
0
Std Error
.19908714
.00823559
.00010332
.05080301
.05899198
.08174074
.15533753
0.14587305
0
0

0
0

0
.15147403
.15235511
m
.11563578
.10970510
•
.11728104
0.10309675

•
0.11521128
0.10894878

•
0.11595457
0.10211355

•
0.11635452
0.10802618

.
DDF
146
146
146
146
146
146
146
146
146
146
B
146
146
•
146
146
•
146
146
•
146
146
•
146
146
.
T
2.52
3.39
-2.98
4.65
0.44
-0.30
8.88
7.03
8.14
8.54
B
-0.66
0.36
•
-0.64
1.23
•
0.79
0.56
•
-1.06
-0.85
•
0.14
2.72
.
Pr
0
0
0
0
0
0
0
0
0
0

0
0

0
0

> !T!
.0130
.0009
.0034
.0000
.6641
.7671
.0000
.0000
.0000
.0000
.
.5107
.7160
•
.5238
.2223
•
0.4332
0.5777

•
0.2894
0.3953

•
0.8851
0.0073

.
Aloha
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
_
0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
.
Lower
0.1073
0.0117
-0.0005
0.1360
-0.0909
-0.1858
1.0726
0.7366
0.9331
0.9997

-0.3048
-0.1768

-0.3067
-0.0774
.
-0.1371
-0.1545
.
-0.3525
-0.2889
•
-0.2131
0.0804
.
Uooer
0.8943
0.0442
-0.0001
0.3368
0.1423
0.1373
1.6866
1.3132
1.5318
1.6019

0.1523
0.2568

0.1568
0.3301

0.3182
0.2761
•
0.1059
0.1148
.
0.2468
0.5074
m
                                        Source
  The MIXED Procedure
Tests of Fixed Effects

      NDF   DDF  Type III F  Pr > F
AGE
AGESO
SUMMER
MOUTH
MALE
GROUP
CAMPAIGN (GROUP)





t
1C
146
146
146
146
146
> 146
> 146
11.50
8.87
21.65
0.19
0.09
29.87
1.61
0.0009
0.0034
0.0000
0.6641
0.7671
0.0000
0.1085
LCOM5.LIS  (source)
       Page  C-3

-------
                                             R&M Study - Preliminary 12-Month Report
                                   CHILD BLOOD AS AN OUTCOME OF CAMPAIGN AND GROUP: 07/18/95
                                     COMPARISON MODEL  - 5  STUDY GROUPS,  INITIAL  PbB  <20 jyg/dL
                                                      Excluding Child 120-3

                                                       The MIXED Procedure
                                              Covariance Parameter Estimates (REML)
                           Cov Parm
                                                   Ratio
                                                              Estimate
                       Std Error
Z  Pr > !Z!
DID 0.04259006 0.00389418 0.05408292 0.07 0.
CHILDNUM(DID) 1.16521862 0.10654064 0.05590865 1.91 0.
Residual 1.00000000 0.09143404 0.01056290 8.66 0.
The MIXED Procedure
Model Fitting Information for LNBLOOO
Description Value
Parameter
INTERCEPT
AGE
AGESO
SUMMER
MOUTH
MALE
CAMPAIGN 12
CAMPAIGN 6
CAMPAIGN 0
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAHPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP(CAMPAIGN)
GROUP (CAMPAIGN)
GROUP(CAMPAIGN)
Observations 258.0000
Variance Estimate 0.0914
Standard Deviation Estimate 0.3024
REML Log Likelihood -156.727
Akaike's Information Criterion -159.727
Schwartz's Bayesian Criterion -164.935
-2 REML Log Likelihood 313.4532
Null Model LRT Chi -Square 63.3846
Null Model LRT OF 2.0000
Null Model LRT P-Value 0.0000
The MIXED Procedure
Solution for Fixed Effects
Estimate Std Error DDF T Pr > !T!







Abated
Levell
Level2
Level3
Modern
Abated
Levell
Level 2
Level3
Modern
Abated
Levell
Level2
LevelS
Modern







12
12
12
12
12
6
6
6
6
6
0
0
0
0
0
0.50079743
0.02793169
-0.00030776
0.23637030
0.02567014
-0.02425782
0.01684491
0.29389836
0.00000000
1.28656093
0.93314079
1.30617006
1.16065726
0.00000000
1.12573499
0.85739644
0.99936019
0.91983696
0.00000000
1.37964782
1 .02492547
1 .23246481
1.30079046
0.00000000
0.19908714
0.00823559
0.00010332
0.05080301
0.
0.
0.
0.

0.
0.
0.
0.

0.
0.
0.
0.

0.
0.
0.
0.

05899198
08174074
11635452
10802618
.
16572324
16217733
16271769
16588817
•
15928529
15272310
15874806
15796763
•
15533753
14587305
15147403
15235511
.
146
146
146
146
146
146
146
146
.
146
146
146
146
•
146
146
146
146
•
146
146
146
146
.
2.52
3.39
-2.98
4.65
0.44
-0.30
0.14
2.72

7.76
5.75
8.03
7.00
•
7.07
5.61
6.30
5.82
•
8.88
7.03
8.14
8.54
.
0.0130
0.0009
0.0034
0.0000
0.6641
0.7671
0.8851
0.0073

0.0000
0.0000
0.0000
0.0000
•
0.0000
0.0000
0.0000
0.0000
•
0.0000
0.0000
0.0000
0.0000
.
Aloha
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05

0.05
0.05
0.05
0.05
•
0.05
0.05
0.05
0.05
-
0.05
0.05
0.05
0.05
.
9426
0567
0000

0.1073
0.0117
-0.0005
0.1360
-0.0909
-0.1858
-0.2131
0.0804

0.9590
0.6126
0.9846
0.8328
.
0.8109
0.5556
0.6856
0.6076
•
1.0726
0.7366
0.9331
0.9997
.


0.8943
0.0442
-0.0001
0.3368
0.1423
0.1373
0.2468
0.5074

1.6141
1.2537
1.6278
1.4885

1.4405
1.1592
1.3131
1.2320
•
1.6866
1.3132
1.5318
1.6019

                                        Source
  The MIXED Procedure
Tests of Fixed Effects

      HDF   OOF  Type III  F  Pr > F
AGE
AGESO
SUMMER
MOUTH
MALE
CAMPAIGN
GROUP (CAMPAIGN)
1
1
1
1
1
2
12
146
146
146
146
146
146
146
11.50 0.0009
8.87 0.0034
21.65 0.0000
0.19 0.6641
0.09 0.7671
3.21 0.0434
10.97 0.0000
LCOM5.LIS  (source)
      Page  C-4

-------
                                              R&M Study - Preliminary 12-Month Report
                                    CHILD BLOOD AS AN OUTCOME OF CAMPAIGN AND GROUP: 07/21/95
                                     COMPARISON MODEL - 5 R&M Groups. Initial PbB Lead >= 20 j/g/dL
                                                       Excluding Child 120-3

                                                        The MIXED Procedure
                                               Covariance Parameter Estimates (REML)



















Parameter
INTERCEPT
AGE
AGESO
SUMMER
MOUTH
MALE
GROUP Abated
GROUP Level 1
GROUP LevelZ
GROUP Level3
CAMPAIGN(GROUP)
CAMPAIGN (GROUP)
CAMPAIGN(GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
CAMPAIGN (GROUP)
Cov Parm
DID
CHILDNLH(DID)
Residual


























12 Abated
6 Abated
0 Abated
12 Level 1
6 Levell
0 Levell
12 Level2
6 Level 2
0 Level 2
12 Level3
6 Level 3
0 Level 3
Ratio Estimate Std Error
0.00000000 0.00000000
1.10267685 0.02667178 0.01892920 1.
1.00000000 0.02(18821 0.0088S830 2.
The MIXED Procedure
Model Fitting Information for LNBLOOD
Description Value
Observations 42.0000
Variance Estimate 0.0242
Standard Deviation Estimate 0.1555
REML Log Likelihood -14.0979
Akaike's Information Criterion -17.0979
Schwartz's Bayesian Criterion -18.9262
-2 REML Log Likelihood 28.1958
Null Model LRT Chi -Square 4.2254
Null Model LRT DF 2.0000
Null Model LRT P-Value 0.1209
The MIXED Procedure
Solution for Fixed Effects
Estimate Std Error DDF T Pr > !T!
3.32192091 0.40476213 14 8.21 0.0000
-0.00396229 0.02344561 14 -0.17 0.8682
0.00010886 0.00029944 14 0.36 0.7216
0.17840855 0.08458966 14 2.11 0.0534
-0.16715983 0.12202463 14 -1.37 0.1923
0.12980143 0.13033147 14 1.00 0.3362
0.00657403 0.19694529 14 0.03 0.9738
-0.25847934 0.31804706 14 -0.81 0.4300
0.01256838 0.14383319 14 0.09 0.9316
0.00000000 ...
-0.42713657 0.15132765 14 -2.82 0.0136
-0.29740206 0.13639076 14 -2.18 0.0468
0.00000000 ...
-0.27732451 0.30801683 14 -0.90 0.3832
-0.19704254 0.29229920 14 -0.67 0.5112
0.00000000 ...
-0.66198479 0.11149930 14 -5.94 0.0000
-0.39811222 0.10745245 14 -3.71 0.0024
0.00000000 ...
-0.49295156 0.14726691 14 -3.35 0.0048
-0.27348742 0.13468986 14 -2.03 0.0617
0.00000000 ...
Z Pr
m
41 0
73 0















Aloha
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
.
0.05
0.05
•
0.05
0.05
•
0.05
0.05
•
0.05
0.05
.
> |7|
m
.1588
.0063















Lower
2.4538
-0.0542
-0.0005
-0.0030
-0.4289
-0.1497
-0.4158
-0.9406
-0.2959
.
-0.7517
-0.5899
•
-0.9380
-0.8240
•
-0.9011
-0.6286
•
-0.8088
-0.5624
.



















Uooer
4.1900
0.0463
0.0008
0.3598
0.0946
0.4093
0.4290
0.4237
0.3211
.
-0.1026
-0.0049
•
0.3833
0.4299
•
-0.4228
-0.1676
•
-0.1771
0.0154
B
                                         Source
  The MIXED Procedure
Tests of Fixed Effects

      NDF   OOF  Type IMF  Pr > F
AGE
AGESO
SUMMER
MOUTH
MALE
GROUP :
CAMPAIGN (GROUP) i
14
14
14
14
14
5 14
J 14
0.03 0.8682
0.13 0.7216
4.45 0.0534
1.88 0.1923
0.99 0.3362
0.36 0.7835
6.06 0.0018
iCOMSA.LIS  (source)
      Page  C-5

-------
                                             R&M  Study  -  Preliminary  12-Month Report
                            CHILD BLOOD AS AH OUTCOME OF  DUST  LEAD  LOADINGS WITHIN A HOUSE: 07/18/95
                                                    EXPOSURE MODEL  -  5  STUDY GROUPS
                                                   Excluding Initial  Campaign
                                                      Excluding Child 120-3
                           Cov Perm
                                                       The MIXED Procedure
                                              Covariance  Parameter Estimates (REML)
         Ratio
Estimate
                                                                          Std Error
Z  Pr > !Z!
DID
CHILDNUM(DID)
Residual
1 .28375240
1.42763103
1.00000000
0.12646785
0.14064194
0.09851421
0.05124132
0.04261544
0.01131555
2.47
3.30
8.71
0.0136
0.0010
0.0000
                                                      The MIXED Procedure
                                             Model Fitting  Information for LNBLOOD
                                            Description
                                                                               Value
                                            Observations                    298.0000
                                            Variance Estimate                 0.0985
                                            Standard Deviation Estimate       0.3139
                                            REML Log Likelihood             -222.940
                                            Akaike's Information Criterion  -225.940
                                            Schwartz's Bayesian Criterion   -231.445
                                            -2 REML Log Likelihood          445.8798
                                            Null Model LRT Chi-Square       123.6121
                                            Null Model LRT DF                 2.0000
                                            Null Model LRT P-Value            0.0000
              Parameter
                                                      The MIXED Procedure
                                                   Solution for Fixed Effects
Estimate
                                                      Std Error   DDF
              T  Pr >  !T!  Alpha
                                                                                                   Lower
                                                                                                             Uooer
INTERCEPT
F ACTOR 1
FACTOR2
AGE
AGESO
SUMMER
CAMPAIGN
CAMPAIGN
CAMPAIGN





12
6
0
1
0
0
0
-0
0
0
0
0
.52270520
.21944775
.09866321
.02683210
.00028313
.27630658
.11875049
.20315084
.00000000
0
0
0
0
0
0
0
0

.14555515
.05373446
.03409240
.00785970
.00010549
.05158814
.06826404
.06053807
.
290
290
290
290
290
290
290
290
.
10.46
4.08
2.89
3.41
-2.68
5.36
1.74
3.36
.
0.0000
0.0001
0.0041
0.0007
0.0077
0.0000
0.0830
0.0009
.
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
.
1.2362
0.1137
0.0316
0.0114
-0.0005
0.1748
-0.0156
0.0840
.
1 .8092
0.3252
0.1658
0.0423
-0.0001
0.3778
0.2531
0.3223
m
                                           Source
                                                      The MIXED Procedure
                                                    Tests of Fixed Effects

                                                       NOF   PDF  Type 111 F  Pr >  F
FACTOR 1
FACTOR2
AGE
AGESO
SUMMER
CAMPAIGN
1
1
1
1
1
2
290
290
290
290
290
290
16.68
8.38
11.65
7.20
28.69
6.20
0.0001
0.0041
0.0007
0.0077
0.0000
0.0023
LEXP5.LIS  (source)
                  Page C-6

-------
                                             R&M Study - Preliminary 12-Month Report
                         CHILD BLOOD AS AN OUTCOME OF DUST LEAD CONCENTRATIONS WITHIN A HOUSE: 07/18/95
                                                    EXPOSURE MODEL - 5 STUDY GROUPS
                                                   Excluding  Initial Campaign
                                                      Excluding Child  120-3
                           Cov Parm
                                                       The MIXED Procedure
                                              Covariance Parameter Estimates (REML)
                                                   Ratio
          Estimate
Std Error
                                                                                           Z  Pr > !Z!
DID
CHILDNUN(DID)
Residual
1 .03540303
1.49613331
1.00000000
0.10133702
0.14642964
0.09787205
0.04785436
0.04298802
0.01123861
2.12
3.41
8.71
0.0342
0.0007
0.0000
                                                       The MIXED Procedure
                                              Model Fitting Information for LNBLOOD
                                             Description
                                                                                Value
                                             Observations                    298.0000
                                             Variance Estimate                 0.0979
                                             Standard Deviation Estimate       0.3128
                                             REML Log Likelihood             -219.668
                                             Akaike's Information Criterion  -222.668
                                             Schwartz's Bayesian Criterion   -228.172
                                             -2 REML Log Likelihood          439.3351
                                             Null Model LRT Chi-Square       118.6309
                                             Null Model LRT DF                 2.0000
                                             Null Model LRT P-Value            0.0000
                                                       The MIXED  Procedure
                                                    Solution for  Fixed  Effects
Parameter
INTERCEPT
FACTOR 1
FACTOR2
AGE
AGESO
SUMMER
CAMPAIGN 12
CAMPAIGN 6
CAMPAIGN 0

1
0
0
0
-0
0
0
0
0
Estimate
.50731275
.23402683
.09405059
.02926403
.00031463
.24116023
.09721856
.16523515
.00000000

0
0
0
0
0
0
0
0

Std Error
. 14385447
.04599675
.03037429
.00784485
.00010522
.04979517
.06544177
.05548312
.
DDF
290
290
290
290
290
290
290
290
.
T
10.48
5.09
3.10
3.73
-2.99
4.84
1.49
2.98
.
Pr > IT!
0.0000
0.0000
0.0022
0.0002
0.0030
0.0000
0.1385
0.0031
.
Aloha
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
.

1
0
0
0
-0
0
-0
0

Lower
.2242
.1435
.0343
.0138
.0005
.1432
.0316
.0560
.
Uooer
1.7904
0.3246
0.1538
0.0447
-0.0001
0.3392
0.2260
0.2744
m
                                           Source
  The MIXED Procedure
Tests of Fixed Effects

   HDF   DDF  Type 111 F  Pr > F
FACTOR 1
FACTOR2
AGE
AGESO
SUMMER
CAMPAIGN J
290
290
290
290
290
! 290
25.89
9.59
13.92
8.94
23.46
4.76
0.0000
0.0022
0.0002
0.0030
0.0000
0.0092
LEXP5.LIS  (source)
      Page  C-7

-------
                                                R&M Study - Preliminary 12-Month  Report
                                 CHILD BLOOD AS AN OUTCOME OF DUST LOADINGS WITHIN A HOUSE:  07/18/95
                                                       EXPOSURE MODEL - 5 STUDY GROUPS
                                                      Excluding Initial Campaign
                                                         Excluding Child 120-3
                                                          The MIXED Procedure
                                                 Covariance Parameter Estimates  (REML)
                                                                                              Z   Pr  >  !Z!
DID
CHILDNUM(DID)
Residual
2.08521083
1.56930768
1.00000000
0.19587611
0.14741429
0.09393588
0.06245621
0.04482045
0.01066665
3.14
3.29
8.81
0.0017
0.0010
0.0000
                                                          The MIXED Procedure
                                                 Model  Fitting Information  for LNBLOOD
                                                Description
                                                                                  Value
                                                Observations                    298.0000
                                                Variance Estimate                 0.0939
                                                Standard Deviation Estimate       0.3065
                                                REML  Log Likelihood             -229.720
                                                Akaike's Information Criterion  -232.720
                                                Schwartz's Bayesian Criterion   -238.225
                                                -2 REML Log Likelihood          459.4406
                                                Null  Model LRT Chi-Square       170.0766
                                                Null  Model LRT OF                 2.0000
                                                Null  Model LRT P-Value            0.0000
                                                         The MIXED Procedure
                                                      Solution for Fixed Effects
Parameter
INTERCEPT
F ACTOR 1
FACTOR2
AGE
AGESO
SUMMER
CAMPAIGN 12
CAMPAIGN 6
CAMPAIGN 0

1
-0
0
0
-0
0
0
0
0
Estimate
.56679839
.02335899
.10909206
.02442296
.00026090
.20622855
.03415141
.12365619
.00000000
Std Error
0.14923971
0.07296066
0.04120194
0.00786114
0.00010527
0.05049457
0.07221552
0.06438353
m
DDF
290
290
290
290
290
290
290
290
B
T
10.50
-0.32
2.65
3.11
-2.48
4.08
0.47
1.92
.
Pr > !T!
0.0000
0.7491
0.0085
0.0021
0.0138
0.0001
0.6366
0.0558
,
Aloha
0
0
0
0
0
0
0
0

.05
.05
.05
.05
.05
.05
.05
.05

Lower
1.
-0.
0.
0.
-0.
0.
-0.
-0.

2731
1670
0280
0090
0005
1068
1080
0031

Uooer
1.8605
0.1202
0.1902
0.0399
-0.0001
0.3056
0.1763
0.2504

                                              Source
  The MIXED Procedure
Tests of Fixed Effects

   HDF   DDF  Type IMF   Pr > F
F ACTOR 1
FACTOR2
AGE
AGESO
SUMMER
CAMPAIGN
1
1
1
1
1
2
290
290
290
290
290
290
0.10
7.01
9.65
6.14
16.68
3.12
0.7491
0.0085
0.0021
0.0138
0.0001
0.0456
1    EXP5.LIS   
-------
                                             R&H Study - Preliminary 12-Month Report
                            CHILD BLOOD AS AN OUTCOME OF OUST LEAD LOADINGS WITHIN A  HOUSE:  07/18/95
                                                    EXPOSURE MODEL - 3 R&M GROUPS
                                    Excluding initial campaign observations for  vacant  houses
                                                      Excluding Child 120-3
                           Cov Parm
                                                       The MIXED Procedure
                                              Covariance Parameter Estimates (REHL)
                                                   Ratio
          Estimate
std Error
                                                                                           Z   Pr  >  !2!
DID
CHILDNUM(DID)
Residual
0.00000000
1.99151207
1.00000000
0.00000000
0.17645345
0.08860275
0.03217351
0.01018259
5.48
8.70
0.0000
0.0000
                                                       The MIXED Procedure
                                              Model  Fitting Information for  LNBLOOD
                                             Description
                                                                               Value
                                             Observations                    251.0000
                                             Variance Estimate                0.0886
                                             Standard Deviation Estimate       0.2977
                                             REML  Log Likelihood             -159.293
                                             Akaike's Information Criterion   -162.293
                                             Schwartz's Bayesian Criterion    -167.526
                                             -2 REML  Log Likelihood           318.5859
                                             Null  Model LRT  Chi-Square        104.5399
                                             Null  Model LRT  DF                2.0000
                                             Null  Model LRT  P-Value            0.0000
               Parameter
                                                       The MIXED  Procedure
                                                    Solution  for  Fixed Effects
                                         Estimate
                                                                  DDF
                                                                            T  Pr >  !T!
                                                                                                   Lower
INTERCEPT
FACTOR 1
FACTOR2
AGE
AGESO
SUMMER
CAMPAIGN
CAMPAIGN
CAMPAIGN
CAMPAIGN





0
2
6
12
1
-0
-0
0
-0
0
-0
0
0
0
.60601235
.01703229
.00001820
.03289107
.00035615
.21936204
.01882390
.06013966
.05093620
.00000000
0
0
0
0
0
0
0
0
0

.16597611
.05615890
.03461233
.00808884
.00010730
.04813242
.10149030
.06212353
.05434423
.
149
149
149
149
149
149
149
149
149
.
9.68
-0.30
-0.00
4.07
-3.32
4.56
-0.19
0.97
0.94
.
0.0000
0.7621
0.9996
0.0001
0.0011
0.0000
0.8531
0.3346
0.3501
.
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
.
1.2780
-0.1280
-0.0684
0.0169
-0.0006
0.1243
-0.2194
-0.0626
-0.0564
.
1.9340
0.0939
0.0684
0.0489
-0.0001
0.3145
0.1817
0.1829
0.1583
B
                                           Source
  The MIXED Procedure
Tests of Fixed Effects

   NDF   DDF  Type IMF   Pr  >  F
F ACTOR 1
FACTOR2
AGE
AGESO
SUMMER
CAMPAIGN
1
1
1
1
1
3
149
149
149
149
149
149
0.09
0.00
16.53
11.02
20.77
0.65
0.7621
0.9996
0.0001
0.0011
0.0000
0.5815
LEXP3.LIS (source)
      Page  C-9

-------
                                             R&N Study • Preliminary 12-Month Report
                         CHILD BLOOD AS AN OUTCOME OF DUST LEAD CONCENTRATIONS UITHIN A HOUSE: 07/18/95
                                                    EXPOSURE MODEL - 3 R&M GROUPS
                                    Excluding initial campaign observations for vacant houses
                                                      Excluding Child 120-3
                           Cov Parm
                                                       The MIXED Procedure
                                              Covariance Parameter Estimates (REML)
Ratio
                                                              Estimate
                        Std Error
                                                                                           Z  Pr > !Z!
DID
CHILDNUM(DID)
Residual
0.00000000
1.98825144
1.00000000
0.00000000
0.17627361
0.08865760
0.03227193
0.01020893
5.46
8.68
0.0000
0.0000
                                                       The MIXED Procedure
                                              Model Fitting Information for LNBLOOD
                                             Description
                                                                                Value
                                             Observat ions                    251.0000
                                             Variance Estimate                 0.0887
                                             Standard Deviation Estimate       0.2978
                                             REML Log Likelihood             -159.529
                                             Akaike's Information Criterion  -162.529
                                             Schwartz's Bayesian Criterion   -167.763
                                             -2 REHL  Log Likelihood          319.0589
                                             Null Model LRT  Chi-Square       103.6870
                                             Null Model LRT  DF                 2.0000
                                             Null Model LRT  P-Value            0.0000
                                                       The MIXED  Procedure
                                                    Solution for  Fixed  Effects
Parameter
INTERCEPT
F ACTOR 1
FACTOR2
AGE
AGESO
SUMMER
CAMPAIGN
CAMPAIGN
CAMPAIGN
CAMPAIGN





0
2
6
12
1
-0
0
0
-0
0
-0
0
0
0
Estimate
.61016168
.01022023
.00402246
.03278622
.00035444
.22052014
.03074231
.05906940
.05019800
.00000000
Std Error
0.16746488
0.04934901
0.03168466
0.00809707
0.00010755
0.04728677
0.09171529
0.06491925
0.05555614
.
DDF
149
149
149
149
149
149
149
149
149
.

9.
-0.
0.
4.
-3.
4.
-0.
0.
0.

T
61
21
13
05
30
66
34
91
90
.
Pr > IT!
0.0000
0.8362
0.8991
0.0001
0.0012
0.0000
0.7380
0.3643
0.3677
.
Aloha
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
.
Lower
1.2792
•0.1077
-0.0586
0.0168
-0.0006
0.1271
-0.2120
-0.0692
-0.0596
_
Uuoer
1.9411
0.0873
0.0666
0.0488
-0.0001
0.3140
0.1505
0.1874
0.1600
m
                                           Source
                                                      The MIXED Procedure
                                                    Tests of Fixed Effects

                                                       NDF   DDF  Type 111 F  Pr > F
F ACTOR 1
FACTOR2
AGE
AGESO
SUMMER
CAMPAIGN
1
1
1
1
1
3
149
149
149
149
149
149
0.04
0.02
16.40
10.86
21.75
0.87
0.8362
0.8991
0.0001
0.0012
0.0000
0.4577
LEXP3.LIS (source)
       Page  C-10

-------
                                             Rtt Study • Preliminary 12-Month Report
                              CHILD BLOOD AS AN OUTCOME OF DUST LOADINGS WITHIN A HOUSE:  07/18/95
                                                    EXPOSURE MODEL - 3 R&M GROUPS
                                    Excluding initial campaign observations for vacant  houses
                                                      Excluding Child 120-3
                           Cov Pann
                                                       The MIXED Procedure
                                              Covariance Parameter Estimates (REML)
                                                   Ratio
          Estimate
                                                                           Std Error
                                                                                           Z   Pr  >  !Z!
DID
CHILDNUM(DID)
Residual
0.00000000
2.15588581
1.00000000
0.00000000
0.17826129
0.08268587
0.03177693
0.00949693
5.61
8.71
0.0000
0.0000
                                                       The MIXED  Procedure
                                              Model  Fitting Information for  LNBLOOD
                                             Description
                                                                               Value
                                             Observations                     251.0000
                                             Variance Estimate                  0.0827
                                             Standard Deviation  Estimate        0.2876
                                             REML  Log Likelihood             -153.603
                                             Akaike's Information Criterion   -156.603
                                             Schwartz's Bayesian Criterion    -161.836
                                             -2 RENL  Log Likelihood           307.2051
                                             Null  Model LRT Chi-Square        113.9592
                                             Null  Model LRT OF                  2.0000
                                             Null  Model LRT P-Value             0.0000
               Parameter
                                                      The MIXED Procedure
                                                   Solution for Fixed Effects
                                          Estimate
                                                      Std Error   DDF
                                                                            T  Pr > IT!  Aloha
INTERCEPT
FACTOR 1
FACTOR2
AGE
AGESQ
SUMMER
CAMPAIGN
CAMPAIGN
CAMPAIGN
CAMPAIGN





0
2
6
12
1.65881354
-0.06292760
0.10428400
0.03033036
-0.00032941
0.20778324
-0.07107343
0.03901309
0.03189224
0.00000000
0
0
0
0
0
0
0
0
0

.16363744
.07467186
.03415829
.00791841
.00010444
.04607974
.10952151
.06015643
.05295620
B
149
149
149
149
149
149
149
149
149
m
10.14
-0.84
3.05
3.83
-3.15
4.51
-0.65
0.65
0.60
m
0.0000
0.4007
0.0027
0.0002
0.0019
0.0000
0.5174
0.5176
0.5479
_
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05

1
-0
0
0
-0
0
-0
-0
-0

.3355
.2105
.0368
.0147
.0005
.1167
.2875
.0799
.0727

1 .9822
0.0846
0.1718
0.0460
-0.0001
0.2988
0.1453
0.1579
0.1365

                                           Source
  The MIXED Procedure
Tests of Fixed Effects

   NDF   DDF  Type IMF  Pr > F
FACTOR 1
FACTOR2
ACE
AGESQ
SUMMER
CAMPAIGN ]
149
149
149
149
149
! 149
0.71
9.32
14.67
9.95
20.33
0.50
0.4007
0.0027
0.0002
0.0019
0.0000
0.6794
LEXP3.LIS (source)
                                                          Page  C-11

-------