Final Risk Evaluation for N- Methylpyrrolidone NMP Systematic Review Supplemental File Data Quality Evaluation of Human Health Studies Epidemiological Studies CASRN 872 50 4

United States
hi	Environmental Protection Agency

Office of Chemical Safety and
Pollution Prevention

Final Risk Evaluation for
n-Methylpyrrolidone

Systematic Review Supplemental File:

Data Quality Evaluation of Human Health Hazard Studies

Epidemiological Studies

CASRN: 872-50-4

December 2020

-------
Table Listing

1	Nishimura et al., 2009: Evaluation of Musculoskeletal/Motor Function Outcomes	2

2	Nishimura et al., 2009: Evaluation of Neurological/Behavior Outcomes		6

3	Haufroid et al., 2014: Evaluation of Renal Outcomes	10

4	Haufroid et al., 2014: Evaluation of Respiratory Outcomes	14

5	Haufroid et al., 2014: Evaluation of Hepatic Outcomes 	20

6	Bader et al., 2006: Evaluation of Irritation Outcomes 	22

This document presents data quality evaluation results for epidemiological studies evaluated for
the NMP Risk Evaluation.

EPA's Office of Pollution Prevention and Toxics (OPPT) developed data quality criteria for
epidemiological studies. The first version of the criteria was documented in the Application of
Systematic Review in TSCA Risk Evaluations document (EPA Document #740-Pl-8001). The
initial criteria were updated as described in the supplemental file Final Risk Evaluation for n-
Methylpyrrolidone (NMP) Systematic Review Supplemental File: Updates to the Data Quality
Criteria for Epidemiological Studies, Docket EPA-HQ-OPPT-2019-0236.

Page 1 of 25

-------
Table 1: Nishimura et al., 2009: Evaluation of Musculoskeletal/Motor Function Outcomes

Study Citation:

Data Type:
HERO ID:

Nishimura, S., Yasui, H., Miyauchi, H., Kikuchi, Y., Kondo, N., Takebayashi, T., Tanaka, S., Mikoshiba, Y., Omae, K., Nomiyama,
T. (2009). A cross-sectional observation of effect of exposure to N-methyl-2-pyrrolidone (NMP) on workers' health Industrial Health,
47(4), 355-362

Cross-sectional Occupational NMP MotorFunction MCV Mean-Musculoskeletal/Motor Function

735269

Domain

Metric

Rating^ MWF* Score

Comments^

Domain 1: Study Participation

Metric 1: Participant Selection

Low

Metric 2: Attrition
Metric 3: Comparison Group

High
High

X 0.4 1.2 Participants include 15 male workers in a factory

using NMP for cleaning instruments without protec-
tive respiratory devices or clothing (wore polyethy-
lene gloves). Not stated if these 15 encompassed the
entire exposed workforce or a select subset. No in-
formation provided on participation rate, inclusion
or exclusion criteria, or methods of participation se-
lection.

X 0.4 0.4 One exposed worked excluded from study, because

he missed work on the day of health effects exam.

X 0.2 0.2 Controls selected from workers at the same factory

with no occupational NMP exposure, matched by
age, education and work load. No significant dif-
ferences in age, physical status, education, drinking
levels or smoking habits. Controls were only sam-
pled on the last day of the 5 day study, compared to
daily sampling in exposed group.

Domain 2: Exposure Characterization

Metric 4: Measurement of Exposure

Metric 5: Exposure Levels

Metric 6: Temporality

High X 0.4 0.4 Sampling tube of 400 mg activated charcoal and air

sampling pump (flow rate 0.1 L /min) worn for 8 hr/
day for 1 week (exposed) or 1 day (controls). An-
alyzed with GC-MS. See reference (Xiafei 200) for
details.

Low X 0.2 0.6 Exposure maximum (0.80 ppm) and daily means

(0.14-0.26 ppm) were below the OEL of 1 ppm rec-
ommended by the Japan Society for Occupational
Health (JSOH). Likely to result in a bias towards
the null.

Low X 0.4 1.2 Outcomes measured directly after a 1-5 days of ex-

posure, but history of exposure not stated. Out-
comes of skin irritation/headaches expected to fall
within this window, but some neurobehavioral out-
comes (depression, response time, and nerve conduc-
tivity) may fall outside of this exposure window.

Domain 3: Outcome Assessment

Continued on next page . ..

Page 2 of 25

-------
. continued from previous page

Study Citation:

Data Type:
HERO ID:

Cross-sectional Occupational NMP MotorFunction MCV Mean-Musculoskeletal/Motor Function

735269

Domain

Metric

Rating^ MWF* Score

Comments^

Metric 7: Outcome Measurement or Characterization Medium

Metric 8: Reporting Bias

Medium

X 0.667 1.33 Motor and sensory nerve conduction velocities of
median nerve of dominant arm (Neuropack). Neu-
robehavioral tests (finger tapping, response time,
reaction time, digit span, and Benton visual re-
tention test) carried out on a personal computer.
These objective metrics would be ranked high. Sub-
jective symptoms (>50 subjective symptoms, de-
pression, and anxiety) were determined from self-
administered questionnaires, which would be ranked
as low. Therefore, the full study was ranked as
medium.

X 0.333 0.67 States that no significant differences were reported
in symptoms related to irritation, but no data pro-
vided. All other outcomes fully reported and ex-
tractable.

Domain 4: Potential Confounding/Variable Control
Metric 9: Covariate Adjustment

Metric 10: Covariate Characterization

Metric 11: Co-exposure Confounding

Medium X 0.5 1 Multiple regression, multiple logistic regression and

stratification were used to adjust for potential con-
founders including age, education, BMI and smok-
ing/drinking habits. These results were not quanti-
tatively reported, however, the exposed and control
groups do not have significant differences with re-
gards to these covariates.

Medium X 0.25 0.5 Smoking and medical histories collected from self-

administered questionnaires. Source of age, body
weight information not stated.

Medium X 0.25 0.5 Identified co-exposure to xylene (10% of cleaning so-
lution), which was measured by a NIOSH method.
Primary xylene metabolite (methylhippuric acid)
was measured in urine. Both measurements fell be-
low the limits of detection (0.1 ppm in air and 0.01
mg/dL in urine).

Domain 5: Analysis

Metric 12: Study Design and Methods

Medium X 0.4 0.8 Study design is appropriate for the outcomes mea-

sured. Means, standard deviations and number of
participants reported for outcomes. Linear regres-
sion conducted, but quantitative results not pre-
sented.

Continued on next page

Page 3 of 25

-------
.. . continued from previous page

Study Citation: Nishimura, S., Yasui, H., Miyauchi, H., Kikuchi, Y., Kondo, N., Takebayashi, T., Tanaka, S., Mikoshiba, Y., Omae, K., Nomiyama,
T. (2009). A cross-sectional observation of effect of exposure to N-methyl-2-pyrrolidone (NMP) on workers' health Industrial Health,
47(4), 355-362

Data Type: Cross-sectional Occupational NMP MotorFunction MCV Mean-Musculoskeletal/Motor Function

HERO ID: 735269

Domain

Metric

Rating^

MWF*

Score

Comments^

Metric 13:

Statistical Power

Medium

x 0.2

0.4

Number of participants (14 exposed, 15 controls) is

small and no information on the derivation of statis-

tical power is provided. The number of participants

is assumed to be adequate.

Metric 14:

Reproducibility of Analyses

Medium

x 0.2

0.4

Simple analysis is reproducible.

Metric 15:

Statistical Models

Medium

x 0.2

0.4

Means with standard deviations presented for out-

comes. Regression models not discussed in detail,

but not reported either. The presented analysis is

sufficient.

Domain 6: Other Considerations for Biomarker Selection and Measurement

Metric 16:

Use of Biomarker of Exposure

Medium

x 0.167

0.33

NMP was used as a biomarker, not its metabolites.

Previous study showed that it can be reflective of

exposure (Bader 2007), but it was not a quantitative

association in this study. All workers with inhalation

exposure had NMP in urine, while all controls had

NMP below the limit of detection.

Metric 17:

Effect Biomarker

Not Rated

Biomarker not used for effects.

Metric 18:

Method Sensitivity

Medium

x 0.167

0.33

LOD stated and sufficiently low to detect biomarker

in all exposed samples.

Metric 19:

Biomarker Stability

Medium

x 0.167

0.33

Urine samples stored at 4C, which differs from the

80C stated in the method reference (Xiaofei 2000).

Stability and time between collection and analysis

not stated in either study.

Metric 20:

Sample Contamination

Medium

x 0.167

0.33

Aside from requesting that participants washed their

hands before providing samples, no information is

provided regarding contamination.

Metric 21:

Method Requirements

Medium

x 0.167

0.33

GC-MS used for high degree of confidence in chem-

ical identification.

Metric 22:

Matrix Adjustment

High

x 0.167

0.17

Creatinine adjusted and unadjusted values provided

(Table 2).

Overall Quality Determination'

Medium

2.0

Extracted

Yes

Continued on

next page .

Page 4 of 25

-------
. continued from previous page

Study Citation: Nishimura, S., Yasui, H., Miyauchi, H., Kikuchi, Y., Kondo, N., Takebayashi, T., Tanaka, S., Mikoshiba, Y., Omae, K., Nomiyama,
T. (2009). A cross-sectional observation of effect of exposure to N-methyl-2-pyrrolidone (NMP) on workers' health Industrial Health,

47(4), 355-362

Data Type:	Cross-sectional_Occupational_NMP_MotorFunction_MCV_Mean-Musculoskeletal/Motor Function

HERO ID:	735269

Domain	Metric	Rating ^ MWF* Score	Comments^

* MWF = Metric Weighting Factor

t High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value.

+ The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High.

if any metric is unacceptable

Overall rating =

J]. (Metric Score; x MWF;) / ]T\ MWFj

(round to the nearest tenth) otherwise

where High: > 1 to < 1.7; Medium: > 1.7 to < 2.3; Low: > 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating
is crossed out and an arrow points to the new rating.

^ This metric met the criteria for high confidence as expected for this type of study.

Page 5 of25

-------
Table 2: Nishimura et al., 2009: Evaluation of Neurological/Behavior Outcomes

Study Citation:

Data Type:
HERO ID:

Cross-sectional Occupational NMP Neurological BentonVisual Mean-Neurological/Behavior

735269

Domain

Metric

Rating^ MWF* Score

Comments^

Domain 1: Study Participation

Metric 1: Participant Selection

Low

Metric 2: Attrition
Metric 3: Comparison Group

High
High

X 0.4 1.2 Participants include 15 male workers in a factory

X 0.4 0.4 One exposed worked excluded from study, because

he missed work on the day of health effects exam.

X 0.2 0.2 Controls selected from workers at the same factory

Domain 2: Exposure Characterization

Metric 4: Measurement of Exposure

Metric 5: Exposure Levels

Metric 6: Temporality

High X 0.4 0.4 Sampling tube of 400 mg activated charcoal and air

sampling pump (flow rate 0.1 L /min) worn f or 8
hr/day for 1 week (exposed) or 1 day (controls).
An-alyzed with GC-MS. See reference (Xiafei 200)
for details.

Low X 0.2 0.6 Exposure maximum (0.80 ppm) and daily means

(0.14-0.26 ppm) were below the OEL of 1 ppm rec-
ommended by the Japan Society for Occupational
Health (JSOH). Likely to result in a bias towards
the null.

Low X 0.4 1.2 Outcomes measured directly after a 1-5 days of ex-

Domain 3: Outcome Assessment

Continued on next page

Page 6 of 25

-------
. continued from previous page

Study Citation:

Data Type:
HERO ID:

Cross-sectional Occupational NMP Neurological BentonVisual Mean-Neurological/Behavior

735269

Domain

Metric

Rating^ MWF* Score

Comments^

Metric 7: Outcome Measurement or Characterization Medium

Metric 8: Reporting Bias

Medium

X 0.333 0.67 States that no significant differences were reported
in symptoms related to irritation, but no data pro-
vided. All other outcomes fully reported and ex-
tractable.

Domain 4: Potential Confounding/Variable Control
Metric 9: Covariate Adjustment

Metric 10: Covariate Characterization

Metric 11: Co-exposure Confounding

Medium X 0.5 1 Multiple regression, multiple logistic regression and

Medium X 0.25 0.5 Smoking and medical histories collected from self-

administered questionnaires. Source of age, body
weight information not stated.

Domain 5: Analysis

Metric 12: Study Design and Methods

Medium X 0.4 0.8 Study design is appropriate for the outcomes mea-

sured. Means, standard deviations and number of
participants reported for outcomes. Linear regres-
sion conducted, but quantitative results not pre-
sented.

Continued on next page

Page 7 of 25

-------
.. . continued from previous page

Data Type: Cross-sectional Occupational NMP Neurological BentonVisual Mean-Neurological/Behavior

HERO ID: 735269

Domain

Metric

Rating^

MWF*

Score

Comments^

Metric 13:

Statistical Power

Medium

x 0.2

0.4

Number of participants (14 exposed, 15 controls) is

small and no information on the derivation of statis-

tical power is provided. The number of participants

is assumed to be adequate.

Metric 14:

Reproducibility of Analyses

Medium

x 0.2

0.4

Simple analysis is reproducible.

Metric 15:

Statistical Models

Medium

x 0.2

0.4

Means with standard deviations presented for out-

comes. Regression models not discussed in detail,

but not reported either. The presented analysis is

sufficient.

Domain 6: Other Considerations for Biomarker Selection and Measurement

Metric 16:

Use of Biomarker of Exposure

Medium

x 0.167

0.33

NMP was used as a biomarker, not its metabolites.

Previous study showed that it can be reflective of

exposure (Bader 2007), but it was not a quantitative

association in this study. All workers with inhalation

exposure had NMP in urine, while all controls had

NMP below the limit of detection.

Metric 17:

Effect Biomarker

Not Rated

Biomarker not used for effects.

Metric 18:

Method Sensitivity

Medium

x 0.167

0.33

LOD stated and sufficiently low to detect biomarker

in all exposed samples.

Metric 19:

Biomarker Stability

Medium

x 0.167

0.33

Urine samples stored at 4C, which differs from the

80C stated in the method reference (Xiaofei 2000).

Stability and time between collection and analysis

not stated in either study.

Metric 20:

Sample Contamination

Medium

x 0.167

0.33

Aside from requesting that participants washed their

hands before providing samples, no information is

provided regarding contamination.

Metric 21:

Method Requirements

Medium

x 0.167

0.33

GC-MS used for high degree of confidence in chem-

ical identification.

Metric 22:

Matrix Adjustment

High

x 0.167

0.17

Creatinine adjusted and unadjusted values provided

(Table 2).

Overall Quality Determination'

Medium

2.0

Extracted

Yes

Continued on

next page .

Page 8 of 25

-------
. continued from previous page

Study Citation: Nishimura, S., Yasui, H., Miyauchi, H., Kikuchi, Y., Kondo, N., Takebayashi, T., Tanaka, S., Mikoshiba, Y., Omae, K., Nomiyama,
T. (2009). A cross-sectional observation of effect of exposure to N-methyl-2-pyrrolidone (NMP) on workers' health Industrial Health,

47(4), 355-362

Data Type:	Cross-sectional_Occupational_NMP_Neurological_BentonVisual_Mean-Neurological/Behavior

HERO ID:	735269

Domain	Metric	Rating ^ MWF* Score	Comments^

* MWF = Metric Weighting Factor

t High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value.

+ The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High.

4	if any metric is unacceptable

Overall rating =

J]. (Metric Score; x MWF;) / ]T\ MWFj

(round to the nearest tenth) otherwise

where High: > 1 to < 1.7; Medium: > 1.7 to < 2.3; Low: > 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating
is crossed out and an arrow points to the new rating.

^ This metric met the criteria for high confidence as expected for this type of study.

Page 9 of25

-------
Table 3: Haufroid et al., 2014: Evaluation of Renal Outcomes

Study Citation:

Data Type:
HERO ID:

Haufroid, V; Jaeger, VK; Jeggli, S; Eisenegger, R; Bernard, A; Friedli, D; Lison, D; Hotz, P (2014). Biological monitoring and
health effects of low-level exposure to N-methyl-2-pyrrolidone: a cross-sectional study International Archives of Occupational and
Environmental Health, 87(6), 663-674

Cross-sectional Occupational NMP Renalserumcreatinine Median-Renal

2654929

Domain

Metric

Rating^ MWF* Score

Comments^

Domain 1: Study Participation

Metric 1: Participant Selection

Metric 2: Attrition

Metric 3: Comparison Group

High X 0.4 0.4 Occupational NMP study in Switzerland from 2006-

2011. From an initial list of 61 potential compa-
nies, 21 were included in the final study. Exposures
were related to graffiti removal or by solvent expo-
sure (production or synthesis); not always daily ex-
posures. Analysis conducted on 91 exposed workers
and 114 unexposed workers (all males)

Medium X 0.4 0.8 Of the 327 eligible workers, 207 (63%) participated.

Exclusion based on desire of participants (113), or-
ganization reasons (7) and gender (1 woman). No
indication of bias from non-participation.

High X 0.2 0.2 Unexposed and exposed workers has similar distri-

butions of age. Unexposed workers had a slightly
higher education levels, were less likely to be smok-
ers and had lower alcohol consumption. However,
these differences were not large.

Domain 2: Exposure Characterization

Metric 4: Measurement of Exposure

Metric 5: Exposure Levels

Medium x 0.4

Low

x 0.2

0.8 Personal air sampling for a full day with solid sor-
bent tubes and pumps (150 ml/min); NMP deter-
mined with NIOSH method. Exposure noted to vary
greatly by days, but samples only collected from one
day. Monthly exposures estimated by occupational
history.

0.6 Range of NMP: below LOD-25.8 mg/m3 (median
0.18 mg/m). Participants categorized into 5 groups
based on current and past exposures: never ex-
posed, former solvent exposure, current NMP expo-
sure only, current solvent exposure (no NMP), cur-
rent exposure to NMP and other solvents. Relatively
low NMP exposure and use of protect equipment re-
sult in a limited ability to determine dose-response.

Continued on next page . ..

Page 10 of 25

-------
. continued from previous page

Study Citation:

Data Type:
HERO ID:

Cross-sectional Occupational NMP Renalserumcreatinine Median-Renal

2654929

Domain

Metric

Rating^ MWF* Score

Comments^

Metric 6: Temporality

Medium x 0.4

0.8 Biomarkers for health outcomes measured directly
after shift with air monitoring and again before next
shift (16 hrs off of work). Clinical symptoms, such as
skin irritation/headaches, determined within a week
of air monitoring; only 37 workers (43% of "exposed"
group) worked with NMP the day before clinical as-
sessments. Biomarkers for liver, renal and respira-
tory health also expected to fall within this exposure
window.

Domain 3: Outcome Assessment

Metric 7: Outcome Measurement or Characterization Medium x 0.667

1.33

Metric 8: Reporting Bias

Medium x 0.333 0.67

Clinical outcomes (skin irritation, headaches, neu-
rotoxic outcomes) assessed with a semi-structured
clinical examination and questionnaires. Spirome-
try results assessed by 2 physicians. Biomarkers for
haematological, renal, liver and respiratory health
also used (see metrics 16-22).

Clinical outcomes are briefly qualitatively described,
and thus cannot be extracted. The outcome
biomarkers are fully reported.

Domain 4: Potential Confounding/Variable Control
Metric 9: Covariate Adjustment

Low

x 0.5

Metric 10: Covariate Characterization

Metric 11: Co-exposure Confounding

Medium x 0.25

1.5 Adjustment for age, smoking (pack years & number
of years since smoking cessation), skin disease, glove
usage, and genetic factors considered in various anal-
yses. Differences in nationality between exposed and
controls were not provided, but most non-Swiss par-
ticipants were German or Italian. Data on education
provided, but not adjustment for this factor or SES.
However, it is unclear if these covariates were con-
sidered in the analysis between NMP exposure and
health outcomes.

0.5 Smoking status/history determined with question-
naire. Other covariates assumed to be collected from
employment records, but this is not explicitly states.

0.5 Categorized based on exposure to additional organic
solvents. Hand washing with organic solvents also
noted on the day of biomonitoring data collection.

Domain 5: Analysis

Continued on next page . ..

Page 11 of 25

-------
. continued from previous page

Study Citation:

Data Type:
HERO ID:

Cross-sectional Occupational NMP Renalserumcreatinine Median-Renal

2654929

Domain

Metric

Rating^ MWF* Score

Comments^

Metric 12: Study Design and Methods

Medium x 0.4

Metric 13: Statistical Power

Medium x 0.2

Metric 14: Reproducibility of Analyses
Metric 15: Statistical Models

Medium x 0.2
Medium x 0.2

0.8 The study design chosen was appropriate for the re-
search questions however the scarce
data on symptomatic effects limited the analysis.
Due to wide variation in daily NMP exposure for in-
dividual participants, only 43% of "exposed" workers
worked with NMP the day before clinical examina-
tion. So determination of acute health effects in this
population is somewhat compromised.

0.4 Only 8 participants had exposure to only NMP,
while 38 had current exposure a mix of organic sol-
vents (including NMP), and 30 were never exposed
to NMP or organic solvents. Although power cal-
culations were done apriori, the number of symp-
tomatic cases was low making interpretation diffi-
cult.

0.4 Description of analysis sufficient to understand and
reproduce.

0.4 Multiple linear regression models used for exposed
group and for the entire group for 2-HSMI/5-HNMP
and s-creatinine.

Domain 6: Other Considerations for Biomarker Selection and Measurement

Metric 16: Use of Biomarker of Exposure High

Metric 17: Effect Biomarker

Metric 18: Method Sensitivity

Metric 19: Biomarker Stability

Metric 20: Sample Contamination

Metric 21: Method Requirements

Metric 22: Matrix Adjustment

Medium x 0.143

Medium x 0.143
Medium x 0.143
Low

High
High

X 0.143 0.14 2-HMSI (mg/l; before next shift) covered 70% of
variance. Metabolites measured in urine have long
half-lives (6-26 hrs) and are unique to NMP.

0.29 Biomarkers for renal health (urinary RBP, uri-
nary albumin, and serum creatinine), hepatic health
(GGT expression), and respiratory health (serum
CC16) were used. Well established, but mechanisms
of action not described.

0.29 Metabolites measured with LC-MS/MS and a LOQ
of 0.2 mg/L.

0.29 Storage history not described, but do not have a high
likelihood of biomarker instability.

X 0.143 0.43 Blanks used for NMP metabolites, but no documen-
tation of steps used to ensure contamination free
from collection to measurement.

X 0.143 0.14 LC-MS/MS used for NMP metabolites

X 0.143 0.14 Creatinine adjusted and unadjusted values provided
(Table 2).

Continued on next page . ..

Page 12 of 25

-------
.. . continued from previous page

Study Citation:

Data Type:
HERO ID:

Haufroid, V; Jaeger, VK; Jeggli, S; Eisenegger, R; Bernard, A; Friedli, D; Lison, D; Hotz, P (2014). Biological monitoring and
health effects of low-level exposure to N-methyl-2-pyrrolidone: a cross-sectional study International Archives of Occupational and
Environmental Health, 87(6), 663-674

Cross-sectional	Occupational	NMP	Renalserumcreatinine	Median-Renal

2654929

Domain

Metric

Rating^ MWF* Score

Comments^

Overall Quality Determination1"



Medium —> Low§ 3r©



Extracted



Yes





MWF = Metric Weighting Factor

High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value.

The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High.

Overall rating =

]T\ (Metric Score; X MWF;) / J] . MWFj

if any metric is unacceptable
(round to the nearest tenth) otherwise

where High: > 1 to < 1.7; Medium: > 1.7 to < 2.3; Low: > 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating
is crossed out and an arrow points to the new rating.

This metric met the criteria for high confidence as expected for this type of study.

§ Evaluator's explanation for rating change: "The effect outcome being reported is based on a very small number of exposed workers and the authors do not that the
small number makes interpretation difficult."

Page 13 of 25

-------
Table 4: Haufroid et al., 2014: Evaluation of Respiratory Outcomes

Study Citation:

Data Type:
HERO ID:

Cross-sectional Occupational NMP Lung Capacity FEV1 Median-Respiratory

2654929

Domain

Metric

Rating^ MWF* Score

Comments^

Domain 1: Study Participation

Metric 1: Participant Selection

Metric 2: Attrition

Metric 3: Comparison Group

High X 0.4 0.4 Occupational NMP study in Switzerland from 2006-

Medium X 0.4 0.8 Of the 327 eligible workers, 207 (63%) participated.

Exclusion based on desire of participants (113), or-
ganization reasons (7) and gender (1 woman). No
indication of bias from non-participation.

High X 0.2 0.2 Unexposed and exposed workers has similar distri-

butions of age. Unexposed workers had a slightly
higher education levels, were less likely to be smok-
ers and had lower alcohol consumption. However,
these differences were not large.

Domain 2: Exposure Characterization

Metric 4: Measurement of Exposure

Metric 5: Exposure Levels

Medium x 0.4

Low

x 0.2

Continued on next page . ..

Page 14 of 25

-------
. continued from previous page

Study Citation:

Data Type:
HERO ID:

Cross-sectional Occupational NMP Lung Capacity FEV1 Median-Respiratory

2654929

Domain

Metric

Rating^ MWF* Score

Comments^

Metric 6: Temporality

Medium x 0.4

Domain 3: Outcome Assessment

Metric 7: Outcome Measurement or Characterization Medium x 0.667

1.33

Metric 8: Reporting Bias

Medium x 0.333 0.67

Clinical outcomes are briefly qualitatively described,
and thus cannot be extracted. The outcome
biomarkers are fully reported.

Domain 4: Potential Confounding/Variable Control
Metric 9: Covariate Adjustment

Low

x 0.5

Metric 10: Covariate Characterization

Metric 11: Co-exposure Confounding

Medium x 0.25

0.5 Smoking status/history determined with question-
naire. Other covariates assumed to be collected from
employment records, but this is not explicitly states.

0.5 Categorized based on exposure to additional organic
solvents. Hand washing with organic solvents also
noted on the day of biomonitoring data collection.

Domain 5: Analysis

Continued on next page . ..

Page 15 of 25

-------
.. . continued from previous page

Study Citation: Haufroid, V; Jaeger, VK; Jeggli, S; Eisenegger, R; Bernard, A; Friedli, D; Lison, D; Hotz, P (2014). Biological monitoring and
health effects of low-level exposure to N-methyl-2-pyrrolidone: a cross-sectional study International Archives of Occupational and
Environmental Health, 87(6), 663-674

Data Type: Cross-sectional Occupational NMP LungCapacity FEV1 Median-Respiratory

HERO ID: 2654929

Domain

Metric

Rating^

MWF*

Score

Comments^

Metric 12:

Study Design and Methods

Medium

0.4

0.8

The study design chosen was appropriate for the re-
search questions however the scarce data on symp-
tomatic effects limited the analysis. Due to wide
variation in daily NMP exposure for individual par-
ticipants, only 43% of "exposed" workers worked
with NMP the day before clinical examination, so
determination of acute health effects in this popula-
tion is somewhat compromised.

Metric 13:

Statistical Power

Medium

0.2

0.4

Only 8 participants had exposure to only NMP,
while 38 had current exposure a mix of organic sol-
vents (including NMP). For reported outcomes, and
30 were never exposed to NMP or organic solvents.
Although power calculations were done apriori, the
number of symptomatic cases was low making inter-
pretation difficult.

Metric 14:

Reproducibility of Analyses

Medium

0.2

0.4

Description of analysis sufficient to understand and
reproduce.

Metric 15:

Statistical Models

Low

0.2

0.6

Multiple linear regression models used for exposed
group and for the entire group for some outcome
but not FEV1.

Domain 6: Other Considerations for Biomarker Selection and Measurement

Metric 16:

Use of Biomarker of Exposure

High

0.143

0.14

2-HMSI (mg/1; before next shift) covered 70% of
variance. Metabolites measured in urine have long
half-lives (6-26 hrs) and are unique to NMP.

Metric 17:

Effect Biomarker

Medium

0.143

0.29

Biomarkers for renal health (urinary RBP, uri-
nary albumin, and serum creatinine), hepatic health
(GGT expression), and respiratory health (serum
CC16) were used. Well established, but mechanisms
of action not described.

Metric 18:

Method Sensitivity

Medium

0.143

0.29

Metabolites measured with LC-MS/MS and a LOQ
of 0.2 mg/L.

Metric 19:

Biomarker Stability

Medium

0.143

0.29

Storage history not described, but do not have a high
likelihood of biomarker instability.

Metric 20:

Sample Contamination

Low

0.143

0.43

Blanks used for NMP metabolites, but no documen-
tation of steps used to ensure contamination free
from collection to measurement.

Metric 21:

Method Requirements

High

0.143

0.14

LC-MS/MS used for NMP metabolites

Metric 22:

Matrix Adjustment

High

0.143

0.14

Creatinine adjusted and unadjusted values provided
(Table 2).

Continued on next page . ..

Page 16 of 25

-------
.. . continued from previous page

Study Citation:

Data Type:
HERO ID:

Haufroid, V; Jaeger, VK; Jeggli, S; Eisenegger, R; Bernard, A; Friedli, D; Lison, D; Hotz, P (2014). Biological monitoring and
health effects of low-level exposure to N-methyl-2-pyrrolidone: a cross-sectional study International Archives of Occupational and
Environmental Health, 87(6), 663-674

Cross-sectional	Occupational	NMP	Lung Capacity	FEV1	Median-Respiratory

2654929

Domain

Metric

Rating^ MWF* Score

Comments^

Overall Quality Determination1"



Medium —> Low§ 3r©



Extracted



Yes





MWF = Metric Weighting Factor

High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value.

The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High.

Overall rating =

]T\ (Metric Score; X MWF;) / J] . MWFj

if any metric is unacceptable
(round to the nearest tenth) otherwise

where High: > 1 to < 1.7; Medium: > 1.7 to < 2.3; Low: > 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating
is crossed out and an arrow points to the new rating.

This metric met the criteria for high confidence as expected for this type of study.

§ Evaluator's explanation for rating change: "Effect based on a very small number of workers."

Page 17 of 25

-------
Table 5: Haufroid et al., 2014: Evaluation of Hepatic Outcomes

Study Citation:

Data Type:
HERO ID:

Cross-sectional Occupational NMP Liver GGTlevels Median-Hepatic

2654929

Domain

Metric

Rating^ MWF* Score

Comments^

Domain 1: Study Participation

Metric 1: Participant Selection

Metric 2: Attrition

Metric 3: Comparison Group

High X 0.4 0.4 Occupational NMP study in Switzerland from 2006-

Medium X 0.4 0.8 Of the 327 eligible workers, 207 (63%) participated.

Exclusion based on desire of participants (113), or-
ganization reasons (7) and gender (1 woman). No
indication of bias from non-participation.

High X 0.2 0.2 Unexposed and exposed workers has similar distri-

butions of age. Unexposed workers had a slightly
higher education levels, were less likely to be smok-
ers and had lower alcohol consumption. However,
these differences were not large.

Domain 2: Exposure Characterization

Metric 4: Measurement of Exposure

Metric 5: Exposure Levels

Medium x 0.4

Low

x 0.2

Continued on next page . ..

Page 18 of 25

-------
. continued from previous page

Study Citation:

Data Type:
HERO ID:

Cross-sectional Occupational NMP Liver GGTlevels Median-Hepatic

2654929

Domain

Metric

Rating^ MWF* Score

Comments^

Metric 6: Temporality

Medium x 0.4

Domain 3: Outcome Assessment

Metric 7: Outcome Measurement or Characterization Medium x 0.667

1.33

Metric 8: Reporting Bias

Medium x 0.333 0.67

Clinical outcomes are briefly qualitatively described,
and thus cannot be extracted. The outcome
biomarkers are fully reported.

Domain 4: Potential Confounding/Variable Control
Metric 9: Covariate Adjustment

Low

x 0.5

Metric 10: Covariate Characterization

Metric 11: Co-exposure Confounding

Medium x 0.25

0.5 Smoking status/history determined with question-
naire. Other covariates assumed to be collected from
employment records, but this is not explicitly states.

0.5 Categorized based on exposure to additional organic
solvents. Hand washing with organic solvents also
noted on the day of biomonitoring data collection.

Domain 5: Analysis

Continued on next page . ..

Page 19 of 25

-------
.. . continued from previous page

Data Type: Cross-sectional Occupational NMP Liver GGTlevels Median-Hepatic

HERO ID: 2654929

Domain

Metric

Rating^

MWF*

Score

Comments^

Metric 12:

Study Design and Methods

Medium

0.4

0.8

Metric 13:

Statistical Power

Medium

0.2

0.4

Metric 14:

Reproducibility of Analyses

Medium

0.2

0.4

Description of analysis sufficient to understand and
reproduce.

Metric 15:

Statistical Models

Medium

0.2

0.4

Multiple linear regression models used for exposed
group and for the entire group for 2-HSMI/5-HNMP
and GGT.

Domain 6: Other Considerations for Biomarker Selection and Measurement

Metric 16:

Use of Biomarker of Exposure

High

0.143

0.14

2-HMSI (mg/1; before next shift) covered 70% of
variance. Metabolites measured in urine have long
half-lives (6-26 hrs) and are unique to NMP.

Metric 17:

Effect Biomarker

Medium

0.143

0.29

Metric 18:

Method Sensitivity

Medium

0.143

0.29

Metabolites measured with LC-MS/MS and a LOQ
of 0.2 mg/L.

Metric 19:

Biomarker Stability

Medium

0.143

0.29

Storage history not described, but do not have a high
likelihood of biomarker instability.

Metric 20:

Sample Contamination

Low

0.143

0.43

Blanks used for NMP metabolites, but no documen-
tation of steps used to ensure contamination free
from collection to measurement.

Metric 21:

Method Requirements

High

0.143

0.14

LC-MS/MS used for NMP metabolites

Metric 22:

Matrix Adjustment

High

0.143

0.14

Creatinine adjusted and unadjusted values provided
(Table 2).

Continued on next page . ..

Page 20 of 25

-------
. continued from previous page

Study Citation:

Data Type:
HERO ID:

Haufroid, V; Jaeger, VK; Jeggli, S; Eisenegger, R; Bernard, A; Friedli, D; Lison, D; Hotz, P (2014). Biological monitoring and
health effects of low-level exposure to N-methyl-2-pyrrolidone: a cross-sectional study International Archives of Occupational and
Environmental Health, 87(6), 663-674

Cross-sectional	Occupational	NMP	Liver	GGTlevels	Median-Hepatic

2654929

Domain

Metric

Rating^

MWF* Score

Comments^

Overall Quality Determination1"



Medium

2.0



Extracted



Yes







MWF = Metric Weighting Factor

High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value.

The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High.

Overall rating =

]T\ (Metric Score; X MWF;) / J] . MWFj

if any metric is unacceptable
(round to the nearest tenth) otherwise

where High: > 1 to < 1.7; Medium: > 1.7 to < 2.3; Low: > 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating
is crossed out and an arrow points to the new rating.

This metric met the criteria for high confidence as expected for this type of study.

Page 21 of 25

-------
Table 6: Bader et al., 2006: Evaluation of Irritation Outcomes

Study Citation:

Data Type:
HERO ID:

Bader, M; Rosenberger, W; Rebe, T; Keener, SA; Brock, TH; Hemmerling, HJ; Wrbitzky, R (2006). Ambient monitoring and biomoni-
toring of workers exposed to N-methyl-2-pyrrolidone in an industrial facility International Archives of Occupational and Environmental
Health, 79(5), 357-364

Cross-sectional Occupational NMP Irritation-Irritation

3539720

Domain

Metric

Rating^

MWF* Score

Comments^

Domain 1: Study Participation

Metric 1: Participant Selection

Metric 2: Attrition
Metric 3: Comparison Group

Low X 0.4 1.2 Participant selection not clear. 7 workers and 3

on-site examiners running the study in adhesive
bonding facility in Germany volunteered to physi-
cal exams, interviews and urine samples before/after
shifts. Number of eligible workers not stated.

Low X 0.4 1.2 Participation rates at study stages and inclu-

sion/exclusion criteria not stated.

Medium X 0.2 0.4 Workers served as own controls (pre/post shift) for

acute outcomes following a day of work after an
exposure-free weekend. Personal exposures 0.9-15.5
mg/m3 across workers, with duties including fore-
man, maintenance, and production worker. Ad-
ditionally, the 3 examiners conducting the study
(physician, study coordinator, and technician) were
exposed only to air contamination and included in
analysis. Note that for at least 1 worker, the "pre-
shift" sample/interview occurred 2 hours after their
shift began. Characteristics not reported.

Domain 2: Exposure Characterization

Metric 4: Measurement of Exposure

High

Metric 5: Exposure Levels

Low

X 0.4 0.4 Well established and detailed methods of direct ex-

posure measurements. Ambient air monitoring of
average workplace concentrations and short-term
peaks monitored by stationary and personal air mon-
itoring. NMP and metabolites (5-HNMP, 2-HMSI)
measured in spot urine tests. Dermal exposure noted
to occur (inconsistent PPE usage), but should be ac-
counted for in the biomonitoring data.

X 0.2 0.6 Exposure range in workers ranged from < LOD to

472 ug/g while exposure in examiners ranged from <
LOD to 123. The range and distribution are limited.

Continued on next page

Page 22 of 25

-------
. continued from previous page

Study Citation:

Data Type:
HERO ID:

Cross-sectional Occupational NMP Irritation-Irritation

3539720

Domain

Metric

Rating^

MWF* Score

Comments^

Metric 6: Temporality

Medium X 0.4 0.8 Temporality established for post-shift measure-

ments, but not for the pre-shift measurements. Two
pre-shift urine sample contained NMP and metabo-
lites. For worker 7 (pre-shift urine had NMP),
the pre-shift sample was taken 2 hours AFTER the
shift began. When considering these measurements
served as controls, it is problematic.

Domain 3: Outcome Assessment

Metric 7: Outcome Measurement or Characterization

Metric 8: Reporting Bias

Medium X 0.667 1.33 Examined before and after shifts by occupational

physician for irritation of the eyes, mucus mem-
branes and skin. Interviewed for related health ef-
fects. Bias is possible from both parties, due to
awareness of exposure, but no direct evidence of
misclassification. Presumably, the physician inter-
viewed/examined themselves, as a subject in the
study.

Low X 0.333 1.0 Interview/examination results presented qualita-

tively for selected participants. Outcomes stated for
"workers" and not directly linked to participant ex-
posure or biomonitoring data.

Domain 4: Potential Confounding/Variable Control
Metric 9: Covariate Adjustment

Metric 10: Covariate Characterization
Metric 11: Co-exposure Confounding

Low

Not Rated
Low

x 0.667

x 0.333

2 Workplace and tasks presented (vary across 7 work-

ers and 3 examiners), but no other characteristics
(age, sex. . . ). Comparison of pre-shift and post-shift
outcomes mediates the concern here though.

NA No covariates/confounders were assessed.

1 Solvents used in cleaning process - aromatic hy-

drocarbons, acetone, propylene glycol monomethyl
ether, and 3-methoxybutyl acetate. Residues in pro-
duction vessels - glutaric and succinic acid dimethyl
ester. Only low levels of acetone and aromatic hy-
drocarbons detected in air during cleaning proce-
dures, thus co-exposures deemed negligible. Due
to lapses in PPE, the glutaric and succinic acid
dimethyl ester could still be relevant and weren't
accounted for.

Domain 5: Analysis

Continued on next page

Page 23 of 25

-------
.. . continued from previous page

Study Citation: Bader, M; Rosenberger, W; Rebe, T; Keener, SA; Brock, TH; Hemmerling, HJ; Wrbitzky, R (2006). Ambient monitoring and biomoni-
toring of workers exposed to N-methyl-2-pyrrolidone in an industrial facility International Archives of Occupational and Environmental
Health, 79(5), 357-364

Data Type: Cross-sectional Occupational NMP Irritation-Irritation

HERO ID: 3539720

Domain

Metric

Rating^

MWF*

Score

Comments^

Metric 12:

Study Design and Methods

Unacceptable

0.143

0.02

Study design appropriate for monitoring acute ex-
posure outcomes. Descriptive outcome reporting did
not include any statistical methods (no means, me-
dians. . . ). Workers with common colds were not
excluded.

Metric 13:

Statistical Power

Unacceptable

0.071

0.01

Low statistical power (7 subjects, 3 controls, with
varied levels of exposure). Unable to determine if
effects related to exposure.

Metric 14:

Reproducibility of Analyses

Medium

0.071

0.14

Sufficient detail reported for only statistical method
applied (linear regression of air monitoring and post-
shift biomonitoring). No statistical method applied
to health outcomes.

Metric 15:

Statistical Models

Low

0.071

0.21

No models used to calculate risk estimates. Lin-
ear regression personal air monitoring results and
post-shift biomonitoring data (metabolite 5-HNMP
in urine) appropriate and transparent.

Domain 6: Other Considerations for Biomarker Selection and Measurement

Metric 16:

Use of Biomarker of Exposure

High

0.167

0.17

NMP and metabolites (5-HNMP, 2-HMSI) measured

in spot urine tests. Shown by linear regression to
correlate with ambient air exposure, and suspected
to also account for dermal exposure.

Metric 17:

Effect Biomarker

Not Rated

No biomarker of effect was measured.

Metric 18:

Method Sensitivity

Medium

0.167

0.33

LOD reported and sufficiently low to detect parent
and metabolites in 100%-40% of samples.

Metric 19:

Biomarker Stability

Medium

0.167

0.33

Storage duration and stability not noted. Stored at
4C during study, and -27C in the laboratory.

Metric 20:

Sample Contamination

Low

0.167

0.5

Blanks used for NMP metabolites, but no documen-
tation of steps used to ensure contamination free
from collection to measurement.

Metric 21:

Method Requirements

Medium

0.167

0.33

Analyzed with GC-MS.

Metric 22:

Matrix Adjustment

Medium

0.167

0.33

Only creatine-adjusted levels provided.

Overall Quality Determination

Unacceptable**

2.4

Extracted

Continued on next page .. .

Page 24 of 25

-------
... continued from previous page

Study Citation:

Data Type:
HERO ID:

Bader, M; Rosenberger, W; Rebe, T; Keener, SA; Brock, TH; Hemmerling, HJ; Wrbitzky, R (2006). Ambient monitoring and biomoni-
toring of workers exposed to N-methyl-2-pyrrolidone in an industrial facility International Archives of Occupational and Environmental
Health, 79(5), 357-364

Cross-sectional_Occupational_NMP_Irritation-Irritation
3539720

Domain

Metric

Rating^

MWF* Score

Comments^

* Consistent with our Application of Systematic Review in TSCA Risk Evaluations document, if a metric for a data source receives a score of Unacceptable (score = 4), EPA

will determine the study to be unacceptable. In this case, one or more of the metrics were rated as unacceptable. As such, the study is considered unacceptable and the score
is presented solely to increase transparency.

MWF = Metric Weighting Factor

High = 1; Medium = 2; Low = 3; Unacceptable = 4; N/A has no value.

The overall rating is calculated as necessary. EPA may not always provide a comment for a metric that has been categorized as High.

Overall rating =

]T\ (Metric Score; x MWF;) / J] . MWFj

if any metric is unacceptable

(round to the nearest tenth) otherwise

where High: > 1 to < 1.7; Medium: > 1.7 to < 2.3; Low: > 2.3 to < 3.0. If the reviewer determines that the overall rating needs adjustment, the original rating is
crossed out and an arrow points to the new rating.

^ This metric met the criteria for high confidence as expected for this type of study.

Page 25 of 25

-------