Thursday, April 14
8:30 a.m.-10:00 a.m.
Session 5:
Epidemiology and Quantitative
Microbial Risk Assessment
91
-------
U.S. EPA's 2016 Recreational Waters Conference
Acute Illness Associated with Ocean Exposure
and Fecal Indicator Bacteria during Dry and
Wet Weather: A Longitudinal Cohort Study of
Surfers in San Diego, California
Ayse Ercumen
University of California—Berkeley
Abstract
Background: Fecal indicator bacteria levels
suggest that freshwater runoff following rain-
storms increases ocean contamination, but little
is known about whether ocean recreators are at
higher risk of acute infections in those condi-
tions. Southern California receives nearly all of
its annual rainfall during the winter months,
and surfers enter the ocean year-round—even
after rainstorms.
Methods: We enrolled 654 surfers in the
San Diego area through on-beach and online
recruitment during the winters of 2013-14 and
2014-15. We collected surf activity (date, loca-
tion, times) and illness symptoms (gastroin-
testinal illness, sinus infections, ear infections,
infected wounds) every 7 days using a smart-
phone- and Web-based application. We clas-
sified surf sessions within 0-3 days of rainfall
> 0.1 inch in 24 hours as wet weather exposure,
and estimated adjusted incidence rate ratios
(IRRs) to compare ocean exposure during dry or
wet periods to unexposed periods. At two sen-
tinel beaches, we collected daily water samples,
measured Enterococcus levels, and matched
daily geometric mean values to individual surf
sessions at those beaches.
Results: Surfers contributed 33,377 days
of observation. Of 10,081 surf sessions, 1,327
followed wet weather. Compared with unex-
posed periods, exposure to seawater during
dry weather increased incidence rates of all
outcomes (e.g., gastrointestinal illness IR R=1.30
[0.95, 1.76], earache/infection IRR=1.86 [1.27,
2.73]); exposure during wet weather further
increased rates (gastrointestinal illness IR R=1.41
[0.92, 2.17], earache/infection IRR=3.28 [1.96,
5.50]). Enterococcus levels were associated only
with illness following wet weather.
Conclusions: Ocean exposure increased
rates of acute illness among surfers during the
winter months, and exposure following rain-
storms further increased illness rates.
Biosketch
Dr. Ayse Ercumen is an epidemiologist at
the University of California at Berkeley School
of Public Health. Dr. Ercumen received her
doctorate degree and master of public health
from University of California at Berkeley
in Epidemiology, and also has a bachelor's
degree in environmental engineering from
Massachusetts Institute of Technology and a
master's degree from University of California at
Berkeley in the same field. Her research inter-
ests include drinking and recreational water
quality, waterborne infections from exposure to
recreational waters, diarrhea transmission and
soil-transmitted helminth infections in low-
income settings, health impact evaluations of
water, sanitation and hygiene (WASH) interven-
tions and waterborne disease associated with
piped water distribution networks.
92
-------
f*"!
Day Two: Session 5
J
iL2a
Development of Site-Specific REC Criteria for a
Tidal Creek using Peer-Reviewed, Stakeholder-
Led QMRA Process
Dustin Bambic
Paradigm Environmental
Abstract
Tecolote Creek is an impaired tidal creek
within the limits of the City of San Diego,
California. The creek flows through a steep
open canyon that is inhabited by a range of
wildlife and flows into Mission Bay a heavily
visited beach area. Several years of microbial
source tracking showed very low prevalence of
human sources and motivated a stakeholder-
led process to use QMRA for development
of site-specific REC criteria. The stakeholder
group includes stormwater, health, regulatory,
academic, and nongovernmental agencies. The
findings of the process are peer-reviewed by
a committee of leading experts in REC water
quality, and the group has had close interac-
tions with the U.S. Environmental Protection
Agency's Office of Science and Technology and
Office of Research and Development. From
summer 2013 through the present, the monitor-
ing program has included the quantification
of fecal indicator bacteria, source identifiers
(HF183, HumBac, and GB124), pathogenic bac-
teria, protozoa, and human pathogenic viruses
(norovirus, enterovirus, and adenovirus) using
state-of-the-art approaches (e.g., digital droplet
PCR). The source identifier data collected dur-
ing the study have led to successful elimination
of bacteria sources. Over the course of the study,
analytical advancements greatly improved virus
detection limits, which is key to risk assessment.
This could be the first effort in the United States
to strive to develop site-specific criteria for a
flowing freshwater system, and the stakeholder
process can serve as a model for other efforts.
The presentation will highlight key lessons
learned that will be intriguing to a wide array
of audience members.
Biosketch
Mr. Dustin Bambic is a director and
professional hydrologist with Paradigm
Environmental. He has led projects across
the United States related to recreational water
quality, including special studies, total maxi-
mum daily load implementation plans, and
watershed modeling efforts. Mr. Bambic holds
bachelor's degrees in physics and mathematics
from Western Kentucky University and mas-
ter's degrees in hydrologic science and envi-
ronmental engineering from the University of
California-Davis.
93
-------
U.S. EPA's 2016 Recreational Waters Conference
Preliminary Evidence for Asymptomatic
Norovirus Infection Transmission Associated
with Swimming at a Tropical Beach
Timothy Wade, PhD
U.S. Environmental Protection Agency
Abstract
Swimming in fecally contaminated water
bodies can result in gastrointestinal (GI) infec-
tions. However, the pathogenic microorganisms
responsible are often unidentified because stud-
ies rely on self-reported symptoms. Noroviruses
are considered a likely cause because they are
resistant to conventional wastewater treatment
and can survive in the environment. Symptoms
among swimmers usually occur within a few
days of exposure, consistent with the short
incubation period of noroviruses. In the summer
of 2009, we conducted an epidemiology study in
Puerto Rico, where we previously reported no
association between swimming and GI symp-
toms. We collected saliva samples from a subset
of participants (N=1300) using an oral swab on
the day of the beach visit (SI), after 10-12 days
(S2), and after approximately 3 weeks (S3) and
tested them for IgG antibody responses to two
common noroviruses (Norwalk and VA387). An
immunoconversion, indicating a potential new
infection, was defined a fourfold increase in
norovirus-specific median fluorescence intensity
(MFI) from the SI to the S2 sample with the S3
sample remaining at least two times above the
baseline (SI) MFI. Approximately 4.7% (N=61)
immunoconverted to norovirus. Swimmers who
immersed their heads in water had a higher
rate of immunoconversion (5.5%) compared to
nonswimmers (2.0%) (OR=3.32, 95% CI 1.2-9.5).
Immunoconversion to norovirus was not associ-
ated with increased GI symptoms, indicating
these infections were asymptomatic. This is the
first epidemiology study to show an association
between norovirus infection and swimming
exposure; however, these preliminary findings
must be verified with additional sensitivity
analyses.
Note: This abstract does not reflect EPA
policy.
Biosketch
Dr. Timothy Wade is an epidemiolo-
gist with the U.S. Environmental Protection
Agency (EPA) in the Office of Research and
Development (ORD) at the Human Studies
Division in Chapel Hill, North Carolina. He
received his master of public health degree
in epidemiology and biostatistics in 1998 and
his doctorate degree in epidemiology in 2002
from the University of California at Berkeley.
Dr. Wade joined EPA ORD in 2003 as a post-
doctoral researcher and has been chief of the
Epidemiology Branch since 2010. His research
focuses on quantifying and measuring the
health effects of waterborne contaminants. He
has authored and coauthored more than 75
peer-reviewed manuscripts and book chapters
and is a section editor for Current Environmental
Health Reports. Dr. Wade was the lead investi-
gator on a series of epidemiological studies at
beach sites across the United States and a series
of arsenic studies in Inner Mongolia, China. His
additional research interests include the devel-
opment and application of salivary immunolog-
ical biomarkers for environmental pathogens,
microbial risk assessment, infectious disease
modeling, and modeling the health effects of
waterborne exposures. Dr. Wade also is an
assistant adjunct professor in the Epidemiology
Division at the University of North Carolina at
Chapel Hill.
94
-------
f*"!
Day Two: Session 5
J
iL2a
Evaluation of the Dry and Wet Weather
Recreational Health Risks in a Semi-Enclosed
Marine Embayment in Southern California
Sunny Jiang, PhD
University of California—Irvine
Abstract
Recreational beach water quality is cur-
rently regulated by the level of fecal indicator
bacteria (FIB) for recreational health protection.
The validity of these indicators, however, has
been questioned in recent years. The poor corre-
lations between FIB and human pathogens can
lead to "underprotection" or "overprotection"
of public health, which can impact the human
health and socioeconomic values of the region.
A new source-apportionment quantitative
microbial risk assessment (QMRA) approach
to risk management, which is based on site-
specific conditions, is proposed and was tested
at Baby Beach in southern California. The small,
semienclosed beach has been suffering from
chronic elevated levels of Enterococcus, espe-
cially during poststorm conditions. The results
of the study show that the median illness risks
are meeting the U.S. Environmental Protection
Agency recreational water quality criteria of 36
illness cases per 1,000 bathers 100 percent of the
time during dry weather conditions, and over
93 percent of the time during wet weather when
the stormwater is contaminated by 5 percent
sewage. The results imply that complying with
current FIB water criteria places unnecessary
burdens on the recreational water manager
without necessarily managing the recreational
water illness (RWI) rate. Optimizing the risks
and benefits of recreational beaches requires
balancing the RWI with the socioeconomic
value of the beaches. A health risk-based
approach as implemented in this study can be
an important complement to a better health risk
management of a nonpoint source recreational
beach.
Biosketch
Dr. Sunny Jiang is a professor of envi-
ronmental engineering at the University of
California, Irvine (UCI). She received her master
of science and doctoral degrees in marine
science from the University of South Florida.
She has been a professor at UCI since 1998 and
teaches more than 150 students per year. During
her career at UCI, she has focused her research
on microbiological water quality, water treat-
ment, and quantitative microbial risk assess-
ment. Dr. Jiang has authored over 70 research
publications and served on committees to guide
the nation's decisions on water reuse.
95
-------
INTRODUCTION
Recreational health rieke are currently regulated baeed on FIB level in water
Enterococcue (ENT) ie used for marine recreational beach
• ENT level « Illness risk
Such relationship is developed and ONLY
APPLICABLE Tor recreaflonal water impacted
by wastewater effluent
In recreational water receiving non-point
source runoff, ENT level is not proportional to
illness risk
Storm
H drains _____
U.S. EPA's 2016 Recreational Waters Conference
EVALUATION OF THE DRY AND WET
WEATHER RECREATIONAL HEALTH
RISKS IN A
SEMI ENCLOSED MARINE EMBAYMENT
IN SOUTHERN CALIFORNIA
K e a h - y i>ri gj. ¦¦ m1, S t e - I a S h a o 2,
Stanley Grant1, i>unny Jiang-
1U M IVE RS IT Y OF CALIFORNIA,
ZC 0 U N T Y OF ORANGE
ENT SOURCE APPORTIONMENT QMRA
QMRA that involves estimating the target pathogen(s) density in water based on the ENT level and source
Ihazard identification
i.e. ENT level in water = lOOCFU/lOOmL: Based on
the land use (source h acking work) of the watershed;
4?
V
I
EXPOSCBJE ASSESSMENT
• vonunt or water ingested during recreational
DOSE-RESPONSE ASSESSMENT
• Dose-response models for each target
pathogen in eavli ENT suuivc
Target pathogen density - -
1. /(50CFU/100mL Plant ENT), no pothogono
2. /t40CKJ.T00niL Oull ENT;, only SaimmeUa and Camcylohaaer
3 /(lOCFU/lOOmL Human ENT), all possible pathogens "
J lino w i-Wc attributed to JEN 1 cortfee
Ote+SH Hltiess mfe flue te utl ftit LN'l
&0IITCC9
Patliogeii-
indirating ENT
Gtarrija
*ENT source from human indicates the
highest amount of hunt an pathogens
ENT SOURCES MATTER
• ENT in water ie uced to indicate the preeence of human pathogens, but ENT of
different sources and origins indicate the different suite of pathogens.
Veroi dtuxriart of puifwxem Uulicuted by ENT
• Stormwater may include a combination of these ENf sources depending on the land use
ENT that do not
indicate me presence
of pathogen
HYPOTHESIS
•Water recreational health risks impacted by stormwater
can vary widely depending on the contamination sources
of Ihe sLormwaler.
STUDY SITE
• Baby Beach at Dana Point,
California
¦ 414 acres watershed
¦ Semi-enclosed beach
receiving only
stormwater, but no
¦ rinsed dnum in the past
due to high FIB level
Enterococcus (ENT) level of the
water monitored weekly since 1995
along the four sampling sites
96
-------
Day Two: Session 5
ENT CONCENTRATION OVERTIME
Overall B0P12.
'i
14. and 16 Enterococcus co
lie installation nt tfnrnnvatfr fitter
vaults at the end nt stnrm drain
and diversion of dry weather run-
off in Kali 2IIII5 has resulted in a
decline of ENT concentration.
Wet wealtief B0P12,13,14, and 15 Enterococcus concentraton
n'; \ii
Urban runoff is a significant ENT •
source of Baby Beacli water
DRY WEATHER ENT SOURCE IDENTIFICATION AND
APPORTIONMENT
Two ENT sources were identified :
1. Bathers' .shed(lings (characterized as primary sewage) - , V3TY with season (Based on
2. Seahircfc'tes K ^
lifeguards' log)
_
Season
Apr-Oct
(Dry/Summer)
Nov-Mar
(Wet/Winter)
ENT source and
appui liuiiuteul
Bather s shedding
90%
65%
Sftahmts' teres
10%
1S%
METHODS
Z7/VT5
n':, = x x Pl x r* x v
X)s ' the (inse of each reference pathogen in
r*JT- ihp FNT level in the recreational v.
) specific ENT source
fer oontnluited Itv a specific source (S)
g*: is the density of ENT in fee«s (wuet mass) (CFI l/g) or in sewage (CF1 l/L)
is the reference pathogen level in feces (wet mass) or sewage (ti of pathogens/g or L or genome copies/g or L)
P* : is the fraction of human-infectious pathogenic strains in the reference pathogen of a soecific source
/5 : is the prevalence of infection in the non-human source
V: is the volume of water ingested {ml).
The illness risk due to earh referent pathogen (77/,) js calculated liased nn the published
infecliun UustHespunse models
The illness risk due to all the reference pathogen in a single ENT source- ijj — 1 -no-*e)
i ne overall mness nsK aue to multiple tN i sources: /^ -1 - (1 - /£)
DRY WEATHER ENT APPORTIONMENT
E
o
Sm<
A?
-------
U.S. EPA's 2016 Recreational Waters Conference
WET WEATHER RISK
102-
£
$10'
1 8 I S
>V> °\° > A" «V> > «V>
. *b V <3 r£>
% of stormwater ENT from sewage
CONCLUSIONS
• Source apportionment QMRA confirms that health risks vary significantly
with the proportion and source of the ENT in the water
• Exceedlngthe ENT numerical concentration standard is not always
indlcatlngthe violation or the health risk standard
• Health risks arc moat sensitive to the presence of ENT of human sewage
origin
• Enforcing ENT numerical concentration may not be the best approach to
manage the recreational health risk
ACKNOWLEDGEMENTS
Financial ana
technical 3upport3
O UCIrvine
PublicWorks
WET WEATHER ENT
cumulative distribution or wet tNf samples
Wei weather
ENT data
were
compiled from
1999 to 2011
ENT level (k>g,.CFU/100mL)
- , ^
I SB I I B
98
-------
f*"!
Day Two: Session 5
J
iL2a
Human-Associated Fecal qPCR Measurements
and Simulated Risk of GI Illness in Recreational
Waters Contaminated with Raw Sewage
Alexandria Boehm, PhD
Stanford University
Abstract
We used quantitative microbial risk assess-
ment to simulate the risk of gastrointestinal (GI)
illness associated with swimming in waters
containing different concentrations of human-
associated fecal markers from raw sewage,
HF183 and HumM2. The volume/volume ratio
of raw sewage to ambient water was determined
by comparing marker concentrations in recre-
ational water to concentrations in raw sewage
from 54 geographic locations across the United
States. Concentrations of reference GI pathogens
in raw sewage, volumes ingested by swim-
mers, dose-response functions, and fractions
of infected that become ill were adopted from
previous studies. Simulated GI risk increased
with concentration of the human quantitative
polymerase chain reaction markers in recre-
ational waters. A benchmark illness rate of
30 GI illnesses per 1000 swimmers occurred at
median concentrations of 4200 copies of HF183
and 2800 copies of HumM2 per 100 mL of rec-
reational water. This study establishes a risk-
based approach for interpreting concentrations
of human fecal markers in ambient waters.
Biosketch
Dr. Alexandria (Ali) Boehm is a profes-
sor in civil and environmental engineering at
Stanford University. She received her master
of science and doctoral degrees in civil and
environmental engineering from University of
California, Irvine and her bachelor of science
degree with honors in engineering and applied
science from California Institute of Technology.
Dr. Boehm has been working on beach contami-
nation issues for the past 15 years and is actively
engaged in research on water and health in
developing countries, stormwater contamina-
tion, coastal water quality more generally, and
biomonitoring of macro-organisms in water
using environmental DNA.
99
-------
U.S. EPA's 2016 Recreational Waters Conference
Question & Answer Session
Question 1
(Unknown): This question is mostly for Sunny [Jiang]. I followed your logic in that you were able to
show for your beach given either stormwater or bather shedding as a pollution source, which is differ-
ent than sewage. How does that translate into beach management decisions? Say a sample exceeds the
standard, can you tell at each beach what the number is specific to that beach that managers can use?
Answer 1
Sunny Jiang: Its very site specific. With EPA's new criteria, it has allowed you to look at the
risk as a management approach. If you can show for a specific area you have a number, but
the risk isn't about that specific number, there is the potential to implement that as a man-
agement strategy. In EPA's criteria they are talking about a model approach. So you can
convert that number to 150, which equals the 236.
Comment 1
Phil Scanlan: I gather from all speakers that having sewage in water is not a good idea. In
New Jersey we had the largest sewage problem in the nation. I have a book; I'd love all the
managers to get a free copy.
Question 2
(Unknown): We have done a lot of work in Milwaukee, and for 50 percent of our urban creeks it is
difficult to correlate concentrations to certain flows. Do you have a frame work? We only find a sewage
signal after heavy rains, not during dry weather.
Answer 2
Dustin Bambic: We try to sample the storm drains as much we can. Find a source upstream. A
lot of GIS [geographic information system] work came out of Santa Barbara. Looking at storm
drains is a good place to start, but we haven't found that smoking gun.
100
------- |