Science of the Total Environment 613–614 (2018) 379–387

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Health-based ingestion exposure guidelines for Vibrio cholerae: Technical

basis for water reuse applications☆
Annetta P. Watson a, Anthony Q. Armstrong a,⁎, George H. White b, Brandolyn H. Thran b
a
Environmental Sciences Division, Oak Ridge National Laboratory, P.O. Box 2008, Bldg 1507, MS 6407, Oak Ridge, TN 37831-6407, United States
b
U.S. Army Public Health Center, 5158 Blackhawk Road, Aberdeen Proving Ground, MD 21010-5403, United States

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• Generation of non-potable graywater

reuse exposure guidelines for Vibrio
cholerae.
• Dose-response modeling for diarrheal
illness following V. cholerae exposure
• Novel risk-based water concentrations
proposed for military graywater reuse
applications.

a r t i c l e i n f o a b s t r a c t

Article history: U.S. military and allied contingency operations are increasingly occurring in locations with limited, unstable or
Received 9 June 2017 compromised fresh water supplies. Non-potable graywater reuse is currently under assessment as a viable
Received in revised form 29 August 2017 means to increase mission sustainability while significantly reducing the resources, logistics and attack vulnera-
Accepted 29 August 2017
bilities posed by transport of fresh water. Development of health-based (non-potable) exposure guidelines for
Available online 14 September 2017
the potential microbial components of graywater would provide a logical and consistent human-health basis
Editor: Jay Gan for water reuse strategies. Such health-based strategies will support not only improved water security for contin-
gency operations, but also sustainable military operations.
Keywords: Dose-response assessment of Vibrio cholerae based on adult human oral exposure data were coupled with oper-
Graywater ational water exposure scenario parameters common to numerous military activities, and then used to derive
Water security health risk-based water concentrations. The microbial risk assessment approach utilized oral human exposure
Vibrio cholerae V. cholerae dose studies in open literature. Selected studies focused on gastrointestinal illness associated with ex-
Microbial risk perimental infection by specific V. cholerae serogroups most often associated with epidemics and pandemics (O1
and O139). Nonlinear dose-response model analyses estimated V. cholerae effective doses (EDs) aligned with gas-
trointestinal illness severity categories characterized by diarrheal purge volume. The EDs and water exposure as-
sumptions were used to derive Risk-Based Water Concentrations (CFU/100 mL) for mission-critical illness
severity levels over a range of water use activities common to military operations.
Human dose-response studies, data and analyses indicate that ingestion exposures at the estimated ED1 (50 CFU)
are unlikely to be associated with diarrheal illness while ingestion exposures at the lower limit (200 CFU) of the

☆ Notice: This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains
and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or
reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of
federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).
⁎ Corresponding author.
E-mail address: armstrongaq@ornl.gov (A.Q. Armstrong).

http://dx.doi.org/10.1016/j.scitotenv.2017.08.297
0048-9697/© 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
380 A.P. Watson et al. / Science of the Total Environment 613–614 (2018) 379–387
estimated ED10 are not expected to result in a level of diarrheal illness associated with degraded individual capa-
bility. The current analysis indicates that the estimated ED20 (approximately 1000 CFU) represents initiation of a
more advanced stage of diarrheal illness associated with clinical care.
© 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license
1. Introduction
1.1. Background
Developing health-based exposure guidelines for waterborne gas-
trointestinal pathogens are an essential component to implementing
sustainable water use and reuse practices that will be protective against
illness caused by microbial pathogens. Such exposure guidelines are a
source of increasing interest to communities and institutions that are
evaluating water supply enhancement strategies that reduce both
fresh water consumption as well as demand on septic treatment facili-
ties (Ludwig, 2015; NAS, 2016; Church et al., 2015; Jahne et al., 2017;
Schoen et al., 2017). Included among the many users evaluating these
options is the U.S. Department of the Army, which is evaluating applica-
tions for fixed and contingency facilities as well as related operations
(USAPHC, 2017; Umble, 2014; AEPI, 2011).
U.S. military and allied contingency operations are increasingly occur-
ring in locations with limited, unstable or compromised fresh water sup-
plies. Non-potable graywater reuse is currently under assessment as a
viable means to increase mission sustainability while significantly reduc-
ing the resources, logistics and attack vulnerabilities posed by transport of
fresh water (USAPHC, 2017; USAPHC, 2014a, 2014b). Water scarcities
create military health vulnerabilities that could be mitigated in part by
placing greater emphasis on sustainable practices that reduce potable
and fresh water demand. Such practices include re-using graywater
from sinks, showers and laundries for non-potable uses such as
showering, laundry, heat-casualty cooling, and vehicle washing
(USAPHC, 2014a, 2014b; Ferdinado, 2013; Church et al., 2015; Schoen
and Garland, 2017; Barker et al., 2017). While it is known that potable-
quality waters are not required for certain military operations such as
dust suppression, fire-fighting and the washing of vehicles, there are, at
present, no health-based or risk-based exposure guidelines for pathogen-
ic components of graywater directly applicable to decision-making for
non-potable use in the field (USAPHC, 2014a; USAPHC, 2017).
Gastrointestinal system infections and accompanying gastroenteri-
tis are a significant cause of morbidity and degraded military operation-
al effectiveness in active-duty components of the U.S. Armed Forces
(MSMR, 2013a, 2013b; Rha et al., 2016; Hyams et al., 1991; Chapman
et al., 2011; Arness et al., 2000). Thus, in any consideration of incidental
ingestion of reused graywater, emphasis should be placed on microbes
known to be responsible for incapacitating gastrointestinal illness and
adverse performance effects. Such emphasis is the focus of this
assessment.
Causative microbes potentially associated with gastrointestinal ill-
ness from incidental graywater ingestion include enteric bacteria and
viruses as well as certain protozoans (Schoen and Garland, 2017). To
build upon previous estimates of non-potable reuse water quality
(Church et al., 2015; Schoen and Garland, 2017), the authors chose an
example enteropathogenic microbe for which there are ample human
data collected from well-conducted, open-literature challenge studies.
In addition, the human subject data for the example microbe needed
to be age-consistent with the active-duty composition of the U.S.
Armed Forces. The selected microbe meeting those criteria is Vibrio
cholerae, the causative pathogen of cholera, and a potential water con-
taminant of interest to the military (DA, 2010). As a waterborne disease
primarily transmitted by ingestion of contaminated water and/or food,
cholera generates military concern due to its swift and adverse health
(http://creativecommons.org/licenses/by/4.0/).
effects on exposed populations and the resulting potential for rapid
mission impact (DA, 2010).
Further, this pathogen is considered a “Category B” bioterrorism or-
ganism of concern (DHHS/CDC, 2014a). Contamination with V. cholerae
in food and water supplies is a high-consequence event for individual,
unit, and installation health as well as for storm and disaster recovery
(DA, 2005; DHHS/CDC, 2014b; Colwell, 1996). V. cholerae occurrence
in water and other aquatic media could create a biosecurity issue that
is worthy of detailed assessment (Casadevall and Relman, 2010).
1.2. Vibrio cholerae pathotypes and illness characteristics
Of the N 200 V. cholerae serogroups, only the O1 and O139
serogroups are known to cause epidemic or pandemic cholera.
Serogroup O1 has been associated with pandemics since 1899,
serogroup O139 has been associated with pandemics since 1992, and
both serogroups have been identified in recent cholera outbreaks
(Colwell, 1996; UN, 2011; WHO, 2017). Serotyping has now been large-
ly supplanted by combined genotypic and phenotypic analyses in deter-
mining pathogenic isolates of serogroups O1 and O139 (e.g., Hasan et al.,
2012; Mutreja et al., 2011; Safa et al., 2009; Li et al., 2002). The present
derivation of health-based exposure guidelines focuses only on dose-
response data for the O1 and O139 serogroups.
V. cholerae O1 and O139 serogroups produce the cholera toxin (CT)
that alters epithelial cell membrane permeability in the small intestine,
resulting in an uncontrolled secretion of water and electrolytes into the
large intestine and colon (Colwell, 2002; USAMRIID, 1998). Virulence is
a function of intestinal colonization and adherence factors (Merrell
et al., 2002; Kaper et al., 1995), which can enhance CT effects. Onset is
usually abrupt and can be associated with vomiting. The American Pub-
lic Health Association (APHA, 2008) indicates that, in most cases, infec-
tion is usually asymptomatic or can cause mild diarrhea. In symptomatic
cases, severe purging results in rapid dehydration, acidosis, circulatory
collapse, electrolyte imbalance, and renal failure, all of which can lead
to shock and a swift death (Colwell, 2002; Ministry of Health and
Population in Haiti/CDC, 2011; DHHS/CDC, 2014b). In severely
dehydrated cases (cholera gravis; associated with passage of ≥1 L diar-
rheal feces/h) (Morris, 2011), death may occur within a few hours and
the case-fatality rate may exceed 50% unless rehydration therapy is ini-
tiated (APHA, 2008; Ministry of Health and Population in Haiti/CDC,
2011).
Cholera is communicable and contagious as long as stools contain vi-
able V. cholerae (considered “culture-positive”); this can occur several
days post-recovery. Contagion takes place through personal contact
with culture-positive stool material or fluids (common examples in-
clude inadequate management of body waste or unclean hands after
toileting). A person who is in a “carrier state” appears well but remains
infectious and can unintentionally transfer V. cholerae to food, contact
surfaces, and sewer or septic systems for multiple months post-
recovery (APHA, 2008; DA, 2010).
1.3. Study purpose
This paper introduces a health-based exposure guidelines derivation
process specific to gastrointestinal illness resulting from single-event
oral exposure to the candidate waterborne enteropathogen
(V. cholerae). Data qualification processes and logic, diarrheal illness
 A.P. Watson et al. / Science of the Total Environment 613–614 (2018) 379–387
criteria selection, and dose-response analyses incorporated into the
derivation are all addressed in the development of risk-based water
concentrations (RBWCs) for various incidental water ingestion sce-
narios common to military operational settings. A primary objective
of this analysis is to characterize cholera dose-response in a manner
that provides sound analytical support for implementing appropriate
protective actions that are intended to minimize adverse effects to
adult (approximately 18–40 years of age) military personnel in an oper-
ational setting. No elders (N60 years of age) or minors (b18 years of
age) are considered part of the potentially exposed population. Focus
of the present work is on expected nonlethal outcomes that can result
in significant and mission-critical performance decrements among
active-duty forces. The incorporation of pathogen-specific health-
based exposure guidelines into water reuse protocols or capability as-
sessments is expected to greatly improve health protection and risk
management decisions by military authorities, and is in keeping with
recent recommendations to apply a microbial risk assessment process
for predicting pathogen risks associated with non-potable water reuse
(Schoen and Garland, 2017).
2. Materials and methods
2.1. Serogroup selection
Dose-response data employed in the current assessment were se-
lected to ensure representation of the human illness spectrum observed
during recognized disease outbreaks as well as to be directly relevant to
water reuse assessment. As a consequence, the exposure studies exam-
ined are focused on human subjects exposed to the most virulent path-
ogenic V. cholerae serogroups (O1 or O139) isolated from cholera
epidemics or pandemics. Alternate serogroups are not as virulent as
O1 and O139 and are not associated with cholera epidemics or pan-
demics (e.g., serogroups O75, O141) (Onifade et al., 2011; Dalsgaard
et al., 2001).
2.2. Human exposure studies
Overall, the most complete data characterizing cholera illness pro-
gression are those collected from well-conducted human-subject chal-
lenge trials primarily for vaccine development and published over a
period of approximately 30 years (1971–1999). All trials evaluated
followed strict experimental protocols for adult human challenge inges-
tion exposures under controlled conditions with informed consent and
medical oversight. Consequently, only the human ingestion-challenge
experimental exposure dataset has been examined during the present
assessment; laboratory animal and human-outbreak studies have
been excluded. A major advantage of this approach is the high relevance
of studies consistently designed to examine, in a controlled and
quarantined manner, consequences of adult human subject ingestion
exposure to virulent pathogenic strains. An additional advantage of
this dataset is the long-term adherence to a stable human study proto-
col established by Hornick et al. (1971) and Cash et al. (1974). Due to
the cross-study protocol adherence, results of V. cholerae exposure stud-
ies performed by multiple investigators and research teams can be
readily compared, contrasted, and combined based on common data
utility relevance.
Ingestion exposure studies selected for inclusion in this assessment
exhibit human dose-response illness categories ranging from asymp-
tomatic to severe. Due to the progressive and well-characterized lethal
nature of unchecked cholera, each experimental human study involved
medical intervention in the form of either oral or IV rehydration solu-
tions after individual participants met well-documented study criteria
for clinical dehydration and/or clinical diarrhea. While no lethal
human cases were identified in the literature review, severe cases
could have rapidly proceeded to a fatal outcome without swift (within
hours) intervention by sufficient volumes of oral or IV electrolyte
381
solutions (DHHS/CDC, 2014b; Morris, 2011). Medical interventions
taken to protect subjects during these studies did not interfere with ob-
servance of dose-response relationships for early-phase illness.
In the human study data for cholera, the potential for illness severity
is categorized by purged stool volume as a quantifiable surrogate indi-
cator for the degree of dehydration and electrolyte loss. It is well-
recognized that dehydration and electrolyte loss are major root causes
of casualty and death in cholera cases (Ministry of Health and
Population in Haiti/CDC, 2011; APHA, 2008). Clinical illness categories
employed by experimental investigators (Hornick et al., 1971, Sack
et al., 1998 and Cohen et al., 1999) to segregate cholera dose-response
on the basis of diarrheal stool volume were utilized for the dose-
response analyses. Thus, the resulting clinical cholera illness categories
employed in the current analyses are defined by total purge volume as
follows:
• Severe as N 5 kg or N 5 L purge volume,
• Moderate as 3–5 kg or 3–5 L purge volume,
• Mild as b3 kg or b3 L purge volume.
The literature also noted several asymptomatic categories defined as
the presence or absence of cholera-positive stool culture and the pres-
ence or absence of antitoxin, lactoferrin or vibriocidal response
measured as evidence of V. cholerae exposure (Hornick et al., 1971;
Morris et al., 1995; Cash et al., 1974; Levine et al., 1981, 1984, 1988;
Sack et al., 1998; Silva et al., 1996).
3. Results and discussion
3.1. Characterization of key studies by illness category
Selected human data characterized by illness severity (categorized
by purge volume) and presence or absence of infection as defined by an-
tibody and serological markers are summarized in Table 1. Classification
of illness severity based on purge volume relies on well-documented
experimental decision criteria for allocating interventions and measur-
ing recovery as defined by Hornick et al. (1971). The purge volume clin-
ical protocol summarized in Table 1 was established in the early 1970s
by Hornick et al. (1971) and Cash et al. (1974) in their studies of
serogroup O1 biotypes. This protocol is robust and stable, and continued
to be followed by later investigators investigating both O1 and O139
serogroup biotypes for the next 28 years (see Cohen et al., 1999). The
long-term incorporation of the Hornick et al. (1971) protocol into sub-
sequent experimental design for both pandemic serogroups (O1 and
O139) serves to facilitate cross-study comparability.
It is acknowledged that, under military operational field conditions
outside of a controlled clinical setting and without ready intervention,
even clinically “mild” diarrheal stool volumes (defined earlier as b3 kg
or b3 L) (Table 1) are likely to be detrimental to individual and unit ca-
pability. Progression to more severe clinical categories and person-to-
person transmission may follow, especially if appropriate treatment is
absent or limited. As purge stool volume increases with greater illness
severity, experience has shown that additional personnel and supply re-
sources are necessarily diverted from other applications (Ministry of
Health and Population in Haiti/CDC, 2011); in an emergency or deploy-
ment situation, such diversions to care for sick individuals and maintain
contagion control are expected to result in an overall degraded unit ca-
pability and readiness.
Further, either large purge volumes or lengthy diarrheal purge dura-
tion leads to dehydration and loss of critical electrolytes, which continue
to be an area of universal health concern (Kenefick et al., 2012). Even
mild to moderate dehydration can induce multiple harmful physiological
and cognitive changes such as visual errors as well as impaired neuro-
muscular control and alertness which have been associated with self-
injury (Szinnai et al., 2005). The cognitive and performance effects that
accompany dehydration levels associated with cholera illness are a
382 A.P. Watson et al. / Science of the Total Environment 613–614 (2018) 379–387

Table 1 The availability of such detailed “no illness” characterization al-

Summary of signs, symptoms and gastrointestinal illness categories in human test subjects lows evaluation of the outcomes that may arise from individual
(M/F aged 18–40) following oral exposure to V. cholerae (in water and/or beef bouillon).
doses of V. cholerae since the dataset exhibits a distribution of ill-
Diarrheal Illness Category Vibrio cholerae Exposure ness responses (Table 1 and Fig. 1); it is clear that the same
(signs and symptoms) ingested dose is associated with generation of a variety of illness
Serogroup and strain Reference
categories. For example, oral doses of 10 5 or 10 6 CFU of cholera
Severe Illness O1 Inaba 569B or Hornick et al. 1971;
Diarrheal stool O1 Ogawa 395 Cash et al. 1974 serogroups O1and O139 are associated with severe, moderate, and
(N5 kg or N5 L)a O1 Ogawa 395 Levine et al. 1981 mild illness as well as the categories of “no (diarrheal) illness” char-
O1 El Tor Inaba Levine et al. 1988; acterized by Sack et al. (1998), Hornick et al. (1971), Cash et al.
Sack et al. 1998 (1974), Silva et al. (1996) and Levine et al. (1981).
O139 Bengal Morris et al. 1995
O139 Cohen et al. 1999
It is noted that various levels of diarrheal illness clinically character-
Moderate Illness O1 El Tor Inaba Sack et al. 1998; ized in the cholera literature by purge volume have not yet undergone
Diarrheal stool Morris et al. 1996 sufficiently rigorous evaluation to allow clear categorization based on
(3-5 kg or 3-5 L) O139 Bengal Morris et al. 1995 military operationally significant severity. Parameters that will likely
O139 Cohen et al. 1999;
contribute to this determination include treatability, levels of dehydra-
Silva et al. 1996
Mild Illness O1 El Tor Inaba Morris et al. 1996; tion expected at various purge volumes, relationship between task per-
Diarrheal stool Taylor et al. 1994; formance degradation and purge volume, and level of medical and/or
(b3 kg or 3 L) Sack et al. 1998 “buddy” support required to provide treatment and hydration assis-
O1 Ogawa 395 Levine et al. 1981 tance (and thereby generate a diversion of personnel and resources
O1 El Tor Silva et al. 1996
O139 Bengal Morris et al. 1995
from mission-critical applications).
O139 Cohen et al. 1999
Illness (unreported diarrheal stool volume) 3.2. Dose-response analysis and modeling
Culture positive stools O1 Inaba 569B or Hornick et al. 1971;
O1 Ogawa 395 Cash et al. 1974
In this assessment, diarrheal illness data were combined to allow
O1 El Tor Ogawa Levine et al. 1982; 1984
O1 El Tor Inaba Morris et al. 1996; focus on presence or absence of diarrheal illness in healthy populations
Levine et al. 1984, 1988 exposed to known quantities of the major epidemic and pandemic
O139 Bengal Morris et al. 1995 V. cholerae serogroups (e.g., combined serogroups O1 and O139;
No Diarrheal Illness/Asymptomatic Table 2). Organization of dose-response data into the categories of “di-
Evidence of Infection O1 Inaba 569B Hornick et al. 1971 arrheal illness” vs “no diarrheal illness” allows inclusion of all available
Culture-positive stool or positive O1 Ogawa 395 Hornick et al. 1971 diarrheal illness data from Table 1. Thus, the “diarrheal illness” category
serological response O1 Inaba 569B Cash et al. 1974
encompasses the entire purge volume and illness spectrum reported for
O1 Ogawa Levine et al. 1984
O1 Inaba Levine et al. 1984 each individual study, and the “no diarrheal illness” category includes
O1 Ogawa 395 Levine et al. 1981; all asymptomatic categories. Percent diarrheal illness is provided as a
Sack et al. 1998 mean and standard deviation when more than a single data set was
O1 El Tor Inaba Levine et al. 1988 available (Table 2).
O1 El Tor Silva et al. 1996
O139 Bengal Morris et al. 1995
The “diarrheal illness” combined serogroups O1 and O139 data from
O139 Cohen et al. 1999 the human subject studies are plotted in Fig. 2. The sigmoidal shape of
No Evidence of Infection O1 Inaba 569B or Hornick et al. 1971; the plotted human subject diarrheal illness data (Fig. 2) indicates that
Culture-negative stool or O1 Ogawa 395 Cash et al. 1974 an ingested dose between 102 and 103 CFU (O1 or O1 + O139) elicits
negative serological response O1 Ogawa Levine et al. 1984
a diarrheal response, and that the plateau of diarrheal response is
O1 Inaba Levine et al. 1984
O1 El Tor Inaba Sack et al. 1998 achieved at an approximate ingested dose of 105 to 106 CFU. Data in
O139 Bengal Morris et al. 1995 the low-dose (e.g., 101 to 103 CFU) and high-dose (e.g., ≥ 107 CFU)
O139 Silva et al. 1996 ranges are limited and due to small sample size; the high-dose data
* Experimental illness severity categorization follows purge stool volume, culture, and se- are not plotted in Fig. 2 given that the dose-response plateau had been
rological definitions established by Hornick et al. (1971), Cash et al. (1974), Sack et al. achieved at 105 to 106 CFU.
(1998) and Cohen et al. (1999). Severe illness also further characterized by these authors Based on the plotted data, dose-response relationships were in-
as “cholera diarrhea” or severe diarrhea requiring IV therapy.
vestigated for the diarrheal illness dichotomous data by fitting non-
linear mathematical models to observable dose-response data.
Initial dose-response models evaluated were the exponential, β-
Poisson, and logistic models as described in Quantitative Microbial
Risk Assessment (Haas et al., 1999) and EPA Benchmark Dose Soft-
significant area of military readiness concern, particularly if the affected
ware and Guidance (USEPA, 2015). The diarrheal illness incidence
individuals or populations have rapid response and/or security
data for both the O1 serogroup and combination of O1 + O139
responsibilities.
serogroups best fit a nonlinear model that does not assume a stan-
In addition to “Severe,” “Moderate,” and “Mild” illness categories,
dard slope but rather optimizes the fit of the slope to the data (a var-
Table 1 includes further description of the “No” (diarrheal) illness cate-
iable slope model). After review of initial model parameters and
gories. These human study data, in which each subject received a
estimates of fitting the models to the data, the logistic model was se-
known ingestion exposure to V. cholerae, are sufficiently well character-
lected to define dose-response relationships outside the observable
ized to support subcategories of infection status and include:
range (i.e., low-dose extrapolation) or between observable doses.
• Asymptomatic (for diarrheal illness) and may exhibit signs of infec- The logistic model is a variable slope model and is based on the em-
tion such as elevated serum antibodies, positive antitoxin, vibriocidal pirical dose-response relationship equations presented below (Eqs.
or lactoferrin response and/or the presence of positive stool cultures (1) and (2)). The logistic model defines a response as a function of
(evidence of infection). the minimum response, the maximum response, the concentration
• Asymptomatic (for diarrheal illness) with negative antitoxin, (or dose) required to evoke a response half-way between the mini-
vibriocidal or lactoferrin response and/or culture-negative stool (no mum and maximum (the EC50), and a parameter that describes the
evidence of infection). steepness of the curve (slope factor). Response at any dose is defined
 A.P. Watson et al. / Science of the Total Environment 613–614 (2018) 379–387 383

Fig. 1. Plot of Exposure vs. Response for all challenged individuals exhibiting cholera illness (severe, moderate or mild) as well as those individuals who are asymptomatic or exhibit no
evidence of infection (% challenged populations vs. oral exposure in 10× CFU) for combined studies of O1 + O139 serogroups V. cholerae.

by an equation with four parameters: minimum response, maximum
response, the log (EC 50 ), and the slope factor (also called the Hill
Slope).
EC 50 ¼ Log10 ðECF Þ−ð1⁄ðHill SlopeÞLog10 ½ F⁄ð100−F Þ
h
i
Y ¼ Bottom þ ðTop−BottomÞ= 1 þ 10ðLog10 EC 50 ÞHill Slope
Log EC50 is the log of the median Effective Concentration, log ECF is
the log of the Effective Concentration Factor (the user-defined percent
response of interest selected from the curve), the Hill Slope (or slope fac-
tor) is a measure of the steepness of the dose-response curve, and F is
the percentage response of interest (as 1%, 10%, etc.) between the Bot-
tom and Top of the distribution. The EC50 for this analysis is the
Eq: 1 V. cholerae dose in CFU that elicits gastrointestinal illness (as defined
by purge volume) in 50% of the exposed population.
Results from the logistic dose-response modeling for V. cholerae
Eq: 2
human subject data are plotted in Fig. 3. The slope was optimized to
fit to the diarrheal illness response data resulting in a slope of 1.2 for
the O1 serogroup data and 1.1 for the combined O1 + O139 serogroups.
Evaluation of Figs. 2 and 3 illustrates the dose-response comparability

Table 2
Incidence of adult human diarrheal illness (combined studies; V. cholerae serogroups O1 + O139).

Diarrheal illness status Dose (CFU)

101 102 103 104 5 × 104 105 106 107 108 109 1010 1011
a a b k a c c c c c
Diarrheal illness (combined severity categories) 0/2 NA 1/6 4/5 3/5 6/8 64/81 0/4 4/6 0/2 0/1 2/2c
9/15c 34/40b 4/5d
2/4l 5/5k 35/38e
6/6f
7/8g
23/24g
5/5h
6/6i
18/25j
18/22l
14/15k
Total diarrheal illness 0/2 NA 1/6 15/24 3/5 45/53 200/235 0/4 4/6 0/2 0/1 2/2
% diarrheal illness (O1 + O139) [SD]⁎ 0 NA 17 62.5 60 85 85 0 67 0 0 100
[21] [18] [10]

No diarrheal illness⁎⁎ 2/2a NA 5/6a 1/5b 2/5k 2/8a 17/81c 4/4c 2/6c 2/2c 1/1c 0/2c
6/15c 6/40b 1/5d
2/4l 0/5k 3/38e
0/6f
1/8g
1/24g
0/5h
0/6i
7/25j
4/22l
1/15k
Total no diarrheal illness 2/2 NA 5/6 9/24 2/5 8/53 35/235 4/4 2/6 2/2 1/1 0/2
% no diarrheal illness (O1 + O139) [SD] 100 NA 83 37.5 40 15 15 100 33 100 100 0
[21] [18] [10]
⁎ Percent diarrheal illness provided as a mean and standard deviation ([SD]).
⁎⁎ Includes data for asymptomatic with culture-positive stool and/or positive serological marker, and asymptomatic with culture-negative stool and/or negative serological marker.
a
Source: Hornick et al., 1971, b Source: Sack et al., 1998, c Source: Cash et al., 1974, d Source: Levine et al., 1981, e Source: Levine et al., 1988, f Source: Morris et al., 1996,
g
Source: Silva et al., 1996, h Source: Taylor et al., 1994, i Source: Levine et al., 1982, j Source: Levine et al., 1984, k Source: Cohen et al., 1999, l Source: Morris et al., 1995.
384 A.P. Watson et al. / Science of the Total Environment 613–614 (2018) 379–387
Fig. 2. Plot of human subject data for diarrheal illness response following ingestion of
V. cholerae. Data from studies with serogroup O1 and combined with data from
serogroup O139. (Note: Data points for ingested dose of 1 and 2 log CFU rest on the x-axis).
between the O1 data and the combined O1 + O139 data at same or sim-
ilar exposures and supports the logic of combining diarrheal illness ef-
fects data for these two serogroups to allow a more robust assessment
and improve statistical power. Biological basis for the combinatorial ap-
proach is provided by recent genomic sequence analyses of O1 and
O139 isolates, which identified a close genetic relationship between
these two serogroups; serogroup O139 is sometimes described as an al-
tered O1 El Tor (Delores and Satchell, 2013; Mutreja et al., 2011). The
decision to combine serogroups for modeling efforts is thus substantiat-
ed by genetic comparability.
Tabulated results from the logistic dose-response model for
V. cholerae human subject data are presented in Table 3. Table 3 presents
the EDs for the diarrheal illness endpoint based on the logistic dose-
response model as well as the 95% confidence intervals for each ED es-
timate following ingestion of a single-bolus dose of V. cholerae. Key
goodness-of-fit parameters from the logistic dose-response model in-
clude a R square of 0.99, Shapiro-Wilk value of 0.98, and P value of
0.96 indicating an acceptable representation of the data by the model.
The modeling assessment and analyses of V. cholerae dose-response re-
lationships derived from human data supports development of
Fig. 3. Dose-response plot of human subject data for diarrheal illness response following
ingestion of V. cholerae modeled with nonlinear logistic variable slope equation. Dose-
response curves for serogroup O1 studies and combined serogroups O1 + O139 studies.
ingestion exposure guideline values (risk-based water concentrations,
RBCWs) considered protective of “diarrheal illness” defined by the
ED1, ED10 and ED20 values in Table 3.
It is noted that others have evaluated dose-response relationships
for several of the cholera diarrheal illness datasets included in this as-
sessment. Haas et al. (1999) developed a dose-response curve for
V. cholerae strain Inaba 569B infectivity based on oral exposure in
humans from the Hornick et al. (1971) study. Analyses by Haas et al.
(1999) indicated the β-Poisson model provided the better fit for this
one dataset and estimated an ID50 (Infectious Dose) of 243 CFU. The
work by Haas et al. (1999) represents an early quantitative dose-
response model fit to human data on V. cholerae.
The Center for Advancing Microbial Risk Assessment (CAMRA) in
2012 examined the Hornick et al. (1971) and Cash et al. (1974) datasets
using several different response parameters (infection, diarrhea with
culture-positive stool, diarrheal illness, and cholera diarrheal illness)
to analyze dose relationships by Exponential and β-Poisson dose-
response models (Weir, 2012). The Weir (2012) analysis of these
datasets determined that the β-Poisson model provided a better fit
when compared to the Exponential model. In addition to the estimated
ingestion median infectious dose (ID50) of 243 CFU for O1 Inaba 569B
infection (measured by a culture-positive stool), Weir (2012) also esti-
mated an ingestion ID50 of 6820 CFU for diarrheal illness. However, the
Weir (2012) analyses of diarrheal illness data with the β-Poisson model
displayed wide variability, with ID50's spanning several orders of
magnitude.
Please note that this paper's authors are well aware that quantitative
microbial risk assessment can be adequately performed by multiple
means. Detailed information reviewing suitable methods for microbial
risk assessment can be found in publications of the Interagency
Microbiological Risk Assessment Guideline Workgroup (2012) and
The Center for Advancing Microbial Risk Assessment (Weir, 2012).
The goal of the current assessment was not to compare and contrast
various models, but to identify an acceptable fit for the combined
human exposure dataset and diarrheal illness endpoint to develop a
technically sound process and assessment tailored for military opera-
tional exposure guidelines.
3.3. Dose evaluation
Since the diarrheal illness category includes cases exhibiting a range
of severity (mild to severe; see Table 1), it is acknowledged that such di-
chotomous categorization of data is not ideally suited for estimating ex-
posures responsible for low hazard severity. The direct development of
a minimal-effects estimate was hindered by the small number of test
subjects at V. cholerae doses b104 CFU since many available studies
employed a large challenge dose to clearly determine vaccine efficacy
against clinically measurable effects. Nevertheless, a low hazard severity
may be judged to describe a diarrheal illness threshold at the 1% effec-
tive dose level (ED1 of 50 CFU with a 95% confidence interval of 9–
Table 3
Nonlinear Logistic dose-response model results of diarrheal illness following ingestion of
V. cholerae (serogroups O1 + O139) in human subject studies.
Response level Ingested dose (CFU) for diarrheal illness
Serogroups O1 + O139 95% confidence intervals
ED1 51 9–294
ED10 496 192–1280
ED16 813 367–1800
ED20 1070 526–2180
ED25 1410 744–2660
ED50 3990 2500–6380
ED75 11,300 6280–20,400
ED80 14,900 7720–28,600
ED84 22,300 10,300–48,300
ED90 32,000 13,300–77,600
 A.P. Watson et al. / Science of the Total Environment 613–614 (2018) 379–387
294 CFU; Table 3), and thus provide an estimate of a no to minimal effect
value for this pathogen.
A 10% effective dose (ED10 of 500 CFU, CI of 192–1280 CFU; Table 3)
for diarrheal illness (combined serogroups) indicates 10% of the ex-
posed population may exhibit some level of diarrheal illness within
the bounds of this estimate. Comparison with dose response data and
modeling (Figs. 1 and 3) and the results of Weir (2012) indicates that
the ED10 of 500 CFU is associated with some diarrheal illness, as well
as asymptomatic and no-infection cases. Therefore, the authors consider
this ED10 (500 cfu) to represent a reasonable estimate of mild diarrheal
illness. Further, the dose-response analyses and reported case data for
asymptomatic or no infection cases indicate adult exposures near the
lower bound of the ED10 (approximately 200 CFU) are not associated
with illness requiring outpatient care or degraded individual capability.
A limited potential for contagion could be generated at the ED10 upper
bound of 1280 CFU due to the possible initiation of diarrheal illness
(Figs. 2 and 3).
Dose comparisons (Figs. 1–3) indicate that the effective doses of 16–
25% (ED16 to ED25 and their CIs: Table 3) represent initiation of more
cases with diarrheal illness as well as a potential for cases exhibiting a
more advanced stage of diarrheal illness. Effects data for the combined
serogroups O1 and O139 are not precise, and biological variability is ev-
ident. The current modeling analysis cannot distinguish differences in
diarrheal illness response between exposures associated with the
upper bound of an ED10 (1280 CFU), an ED16 (8.1 × 102 CFU; CI of
367–1800 CFU), ED20 (1.1 × 103 CFU; CI of 526–2180 CFU) or ED25
(1.4 × 103 CFU; CI of 744–2660 CFU). The range represented by the
ED10 upper bound and the ED16 to ED25 for diarrheal illness approxi-
mates 103 CFU, which is proposed as a reasonable estimate of the
more adverse “diarrheal illness” severity category. This value approxi-
mates the estimated ED20 (1.1 × 103 CFU) (Table 3).
Further consideration of greater exposures identified by the model-
ing analysis (Table 3), such as the ED50 (3990 CFU; CI of 2500–
6380 CFU) as potential health exposure endpoints is not under current
consideration due to the potential for generating clinically moderate
(N3 but b 5 L purge stool) as well as clinically severe (N 5 L purge
stool) diarrheal illness which would also be highly operationally signif-
icant (see Figs. 1–3). It is noted that the modeled ED50 and CI are consis-
tent with data characterizing clinically moderate and severe illness in
the human subject data of Morris et al. (1995), Cohen et al. (1999),
Hornick et al. (1971) and Sack et al. (1998). The severity of the potential
diarrheal illness outcome at these exposures would seriously compro-
mise the exposed population and can be expected to generate a high po-
tential for contagion. In addition, examination of dose-response at
ingested doses between 103 and 104 CFU (Fig. 3) highlights the steep di-
arrheal illness response in this range (approximately 60% at 104), which
is not health-protective, and would compromise both individual and
unit mission capability.
3.4. Risk-based water concentrations (RBWCs)
To enhance the utility of our findings, risk-based water concentra-
tions (RBWCs) were developed for a variety of single-event adult
Table 4
Estimated risk-based water concentrations (RBWCs) for single-dose V. cholerae military incidental water ingestion exposure assumptions.
Total ingested dose
Effective dose
(CFU)
ED20 1000
ED10 200
ED1 50
a
Dose and water concentration values are provided in V. cholerae colony forming units (CFUs). Derivation based on data representative of a single dose and one exposure event. Effective
doses (rounded) as defined in Table 3; 200 CFU is the lower bound of the ED10 (CI of 192–1280 CFU).
b
Presented values are not incidental water consumption standards, but are intended for application to risk assessment in this article.
385
ingestion rates based on military operational exposure scenarios
(Table 4). Incidental water ingestion is considered during three expo-
sure events; showering, splashing, and heat casualty cooling. Incidental
ingestion during showering at a rate of 10–30 mL/event (USAPHC,
2014b) represents a frequent military event considered a possible can-
didate for deployed force water reuse. Incidental ingestion resulting
from splashing (4.0 mL/event; USEPA, 2011) is comparable to that en-
countered by personnel performing military vehicle washing activities
(Sinclair et al., 2016), and considered another good candidate for
water reuse in a deployed setting (Table 4). Incidental ingestion expo-
sure during heat casualty cooling at an estimated rate of 3.5 mL/event
(USAPHC, 2014b) is an assumed exposure event associated with this
use of water during a deployment operation or training exercise,
where quantities of fresh water may not be present. Other daily inciden-
tal water ingestion volumes can be calculated or scaled for other mili-
tary exposure and use case scenarios. For these selected exposure
scenarios, the availability of well-documented health- and risk-based
concentrations developed from a robust set of human exposure data
are crucial to managing health risk over a range of water-use (or
reuse) activities in military operational settings.
While the risk-based concentrations presented in Table 4 assume
single-event exposures, it is acknowledged that individuals may obtain
water from a particular source and may use that source more than once
throughout a given duration. Multiple exposures may occur, so the sin-
gle acute exposure assumption of this paper may not be fully appropri-
ate in all cases. A number of exposure-related questions remain
unanswerable at this point; therefore, guidelines developed during
this assessment must be applied with care, and different outcomes
may result from multiple exposures (higher frequency) over a longer
duration. Some immunity (of varying duration) has been observed fol-
lowing exposure to some cholera strains (Morris et al., 1995; Taylor
et al., 1994); indeed, multiple exposures may not result in a full comple-
ment of health outcomes after initial illness. There is also ample evi-
dence of asymptomatic infection (Table 1, Fig. 1). An exposure that
generates clinical illness in most of the exposed population may not re-
sult in observable illness in some individuals and is a key factor in the
generation of “carriers.”
3.5. Applications
A risk-management application in which this paper's dose-response
analysis (Figs. 2 and 3, Tables 2 and 3) is coupled with USEPA (2011), DA
(2010) and literature values of incidental ingestion water volume as-
sumptions (e.g., from showering, splashing during the washing of mili-
tary vehicles, heat casualty cooling) allows estimation of the associated
V. cholerae water concentrations as an aid to incidental non-potable
water-use decision making. As an example, Table 4 presents estimated
V. cholerae CFU concentrations for three incidental water ingestion sce-
narios. From a risk management perspective, if no precautions are taken
to reduce or eliminate incidental water ingestion, cautious application
of the dose-response analysis would recommend no showering,
splashing or heat casualty body cooling in reclaimed waters containing
V. cholerae CFUs at the ED20 concentration (1000 CFU/event from
Scenario-specific RBWCs (CFU/100 mL)a,b
Shower Splash Heat casualty cooling
(10–30 mL/event) (4 mL/event) (3.5 mL/event)
3330–10,000 25,000 28,570
670–2000 5000 5710
170–500 1250 1430
386 A.P. Watson et al. / Science of the Total Environment 613–614 (2018) 379–387
incidental ingestion exposure to waters containing 3.3 × 103 to 28.6
× 103 CFU/100 mL) due to the potential for adverse gastrointestinal ef-
fects (including diarrheal illness) in an unacceptable percentage of the
exposed population. It is noted that waters containing these ED20 con-
centrations for incidental exposure would be more appropriate for
water-reuse applications in military operational settings such as dust
suppression where incidental exposure is not likely or could be avoided.
At 200 CFU V. cholerae (the 95% lower confidence interval on the
ED10 of 500), incidental ingestion water concentrations for showering,
splash, and heat casualty cooling scenarios are calculated to range
from 0.7 to 5.7 × 103 CFU/100 mL (Table 4). Given the observed variabil-
ity of human dose-response to V. cholerae ingestion (Fig. 1), caution is
warranted during development of incidental water-use decisions at
200 CFU/event; limited ingestion exposures at this concentration are
recommended. However, if there is an immediate need (e.g., to treat a
heat casualty), it would be reasonable to utilize cooling waters contain-
ing ED10 V. cholerae concentrations while taking precautions to prevent
potential incidental ingestion.
At the ED1, incidental ingestion water concentrations for the military
operational scenarios considered range from 0.2 × 103 to 1.4
× 103 CFU/100 mL (Table 4). Current dose-response data and modeling
analyses indicate that these values are unlikely to be associated with di-
arrheal illness and would be reasonable for these applications.
This application of ED estimates for V. cholerae illustrates the utility
of the current assessment as a risk management resource. Command
decisions regarding force strength and level of medical support neces-
sary for mission success as well as reclaimed water treatment and sup-
ply requirements would all be supported by the information provided in
Table 4.
4. Conclusions
The development of health-based exposure guidelines for water-
borne pathogens such as V. cholerae is an essential component to the
generation and implementation of sustainable water use and reuse
practices that help prevent debilitating gastrointestinal diseases prob-
lematic for U.S. and allied forces during contingency, training and garri-
son operations within the continental United States and abroad. The
incidental ingestion Risk Based Water Concentrations for V. cholerae de-
veloped and presented here (Table 4) complement existing knowledge
and serve as health-based pathogen exposure estimates for develop-
ment of water use or reuse risk management protocols (USACHPPM,
2009; USAPHC, 2014a, 2014b).While the present evaluation is derived
from analysis of human challenge exposure data for the virulent and
pandemic V. cholerae serogroups (O1 and O139), the logic and approach
developed here could also be applied to other enteric pathogens of mis-
sion concern such as enterotoxigenic E. coli, Shigella spp., and
noroviruses (Hyams et al., 1991; Arness et al., 2000; Chapman et al.,
2011).
The range of likely military water-use scenarios evaluated for inci-
dental ingestion exposure include showering (a frequent military
event considered a possible candidate for water reuse), water splash
(considered comparable to that encountered by personnel performing
military vehicle washing activities, an additional application under con-
sideration for water reuse) and heat casualty cooling (associated with
deployment operations and training exercises where quantities of
fresh water may not be present). For each of these incidental exposure
scenarios, estimated V. cholerae water concentrations in CFU/100 mL
are summarized for ED1, ED10 (lower-bound) and ED20 endpoints in
Table 4.
Comparison with the human dose-response studies and data
(Tables 1–3; Figs. 1–3) supports a determination that the estimated
ED1 is unlikely to be associated with diarrheal illness, while the estimat-
ed ED10 (lower bound) is unlikely to be associated with a level of diar-
rheal illness requiring outpatient care. In addition, the ED10 (lower
bound) is unlikely to be associated with degraded individual capability.
The current analysis indicates that the estimated ED20 represents initia-
tion of a more advanced stage of diarrheal illness associated with clinical
care; in addition, the steep diarrheal illness dose response can be ex-
pected to initiate at ingestion doses of ED20 (Figs. 2 and 3). Estimated
ED1, ED10 (lower bound), and ED20 effect levels are considered key de-
cision points for managing health risk over a range of water reuse activ-
ities in military operational settings. Military health risk assessors, as
well as military medical and operational planners responsible for gener-
ating water use or reuse protocols, have expressed interest in such
pathogen-specific tools.
Acknowledgments
This research was sponsored by the U.S. Army Public Health Center
under U.S. Department of Energy Proposal No. 2207-M135-A1 and per-
formed in part at Oak Ridge National Laboratory, Oak Ridge, TN. Oak
Ridge National Laboratory is managed by UT-Battelle, LLC, for the U.S.
Department of Energy under Contract DE-AC05-00OR22725.
The current work emerged from the exposure guideline conceptual
and technical foundations developed by the Army Public Health Center
team of Dr. Brandolyn Thran, Dr. Gabriel Intano, Mr. Stephen Comaty
and Mr. Matthew McAtee.
The authors further wish to express appreciation for the excellent
technical assistance provided by Ms. Debra Stewart, The Institute for En-
vironmental Modeling, Department of Ecology and Evolutionary Biolo-
gy, University of Tennessee, Knoxville, TN.
The views expressed in this presentation are those of the
author(s) and do not necessarily reflect the official policy of the Depart-
ment of Defense, Department of the Army, U.S. Army Medical Depart-
ment, Department of Energy or the U.S. Government. Mention of trade
names or commercial products does not constitute endorsement or rec-
ommendation of use.
References
AEPI (Army Environmental Policy Institute), 2011. Army Water Security Strategy. Decem-
ber 2011. http://www.aepi.army.mil/docs/whatsnew/ArmyWaterStrategy.pdf
(accessed 11.10.2016).
APHA (American Public Health Association), 2008. Control of communicable diseases
manual. In: Heymann, D.L. (Ed.), Official Report of the APHA. American Public Health
Association, 800 I Street, Washington, D.C 20001-3710.
Arness, M.K., Feighner, B.H., Canham, M.L., et al., 2000. Norwalk-like viral gastroenteritis
outbreak in U.S. Army trainees. Emerg. Infect. Dis. 6 (2), 204–207.
Barker, S.F., O'Toole, J., Sinclair, M.L., et al., 2017. Endotoxin health risk associated with
high pressure cleaning using reclaimed water. Microbial Risk Anal. 5, 65–70.
Casadevall, A., Relman, D.A., 2010. Microbial threat lists: obstacles in the quest for
biosecurity? Nat. Rev. 8, 149–154.
Cash, R.A., Music, S.I., Libonati, J.P., et al., 1974. Response of man to infection with Vibrio
cholerae. Clinical, serologic, and bacteriologic response to a known inoculum. J Infect
Dis 129, 45–52.
Chapman, A.S., Witkop, C.T., Escobar, J.D., et al., 2011. Norovirus outbreak associated with
person-to-person transmission, U.S. Air Force Academy, July, 2011. MSMR 18 (11), 2–5.
Church, J., Verbyla, M.E., Lee, W.H., et al., 2015. Dishwashing water recycling system and
related water quality standards for military use. Sci. Total Environ. 529, 275–284.
Cohen, M.B., Giannella, R.A., Losonsky, G.A., et al., 1999. Validation and characterization of
a human volunteer challenge model for cholera by using frozen bacteria of the new
Vibrio cholerae epidemic serotype, O139. Infect. Immun. 67 (12), 6346–6349.
Colwell, R.R., 1996. Global climate and infectious disease: the cholera paradigm. Science
274, 2025–2031.
Colwell, R.R., 2002. A voyage of discovery: cholera, climate and complexity. Environ.
Microbiol. 4 (2), 67–69.
DA (Department of Army), 2005. Potential Military Chemical/Biological Agents and Com-
pounds. Field Manual (FM) 3-11.9/MCRP 3-37.1B/NTRP 3-11.32/AFTTP(I) 3-2.55.
Headquarters, Departments of the Army, Marine Corps, Navy and Air Force. US
Army Training and Doctrine Command, Fort Monroe, VA.
DA (Department of Army), 2010. Sanitary Control and Surveillance of Field Water Sup-
plies. Technical Bulletin, Medical 577 (TB MED 577/NAVMED P-5010-10/AFMAN
48-138_IP). Headquarters, Departments of the Army, Marine Corps, Navy and Air
Force, Washington, DC.
Dalsgaard, A., Serichantalergs, O., Forslund, A., Lin, W., Mekalanos, J., Mintz, E., Shimada, T.,
Wells, J.G., 2001. Clinical and environmental isolates of Vibrio cholerae serogroup
O141 carry the CTX phage and the genes encoding the toxin-coregulated pili.
J. Clin. Microbiol. 39 (11), 4086–4092.
Delores, J., Satchell, K.J.F., 2013. Analysis of Vibrio cholerae genome sequences reveals
unique rtxA variants in environmental strains and an rtxA-null mutation in recent al-
tered El Tor isolates. MBio 4, 1–9.
 A.P. Watson et al. / Science of the Total Environment 613–614 (2018) 379–387
DHHS/CDC (U.S. Department of Health and Human Services/Centers for Disease Control
and Prevention), 2014a. National Select Agent Registry. Select Agents and Toxins Ex-
clusions. http://www.selectagents.gov/SelectAgentsandToxinsExclusions.html
(accessed 11.23.2014).
DHHS/CDC (U.S. Department of Health and Human Services/Centers for Disease Control
and Prevention), 2014b. Cholera-Vibrio cholerae infection. www.cdc.gov/cholera/
(accessed 11.23.2014).
Ferdinado, L., June 6, 2013. Net zero provides energy security, saves resources, protects
soldiers. ARNEWS Available at. http://www.army.mil/article/104959 (accessed
11.03.2016).
Haas, C.N., Rose, J.B., Gerba, C.P., 1999. Quantitative Microbial Risk Assessment. John Wiley
and Sons Inc., New York, New York.
Hasan, N., Choi, S.Y., Eppinger, M., et al., 2012. Genomic diversity of 2010 Haitian cholera
outbreak strains. PNAS PLUS 109 (29), E2010–E2017.
Hornick, R.B., Music, S.I., Wenzel, R., et al., 1971. The broad street pump revisited: re-
sponse of volunteers to ingested cholera vibrios. Bull. NY Acad. Med. 47 (10),
1181–1191.
Hyams, K.C., Bourgeois, A.L., Merrell, B.R., et al., 1991. Diarrheal disease during Operation
Desert Shield. NEJM 325 (20), 1423–1428.
Interagency Microbiological Risk Assessment Guideline Workgroup, 2012. Microbial Risk
Assessment Guideline: Pathogenic Microorganisms With Focus on Food and Water.
Pub. No. USDA/FSIS/2012-001 and EPA/100/J12/001. Food Safety and Inspection Ser-
vice, U.S. Department of Agriculture and U.S. Environmental Protection Agency,
Washington, DC http://www.fsis.usda.gov/wps/portal/fsis/topics/science/risk-assess-
ments/microbial-risk-assessment-guideline/microbial-risk-assessment-guideline
(accessed 10.04.2016).
Jahne, M.A., Schoen, M.E., Garland, J.L., Ashbolt, N.J., 2017. Simulation of enteric pathogen
concentrations in locally-owned greywater and wastewater for microbial risk assess-
ments. Microbial Risk Anal. 5, 44–52.
Kaper, J.B., Morris Jr., J.G., Levine, M.M., 1995. Cholera. Clin. Microbiol. Rev. 8 (1), 48–86.
Kenefick, R.W., Cheuvront, S.N., Leon, L.R., et al., 2012. Dehydration and rehydration. In:
Auerbach, P.S. (Ed.), Wilderness Medicine, 6th ed. Elsevier Mosby, Philadelphia,
Pennsylvania. Report Number MISC 10-37. Thermal and Mountain Medicine Division,
U. S. Army Research Institute of Environmental Medicine, U.S. Department of the
Army, Natick, Massachusetts, pp. 71–82.
Levine, M.M., Black, R.E., Clements, M.L., et al., 1981. Volunteer studies in development of
vaccines against cholera and enterogenic Escherichia coli: a review. In: Holme, T.,
Holmgren, J., Meison, M.H., Mölby, R. (Eds.), Acute Enteric Infections in Children:
New Prospects for Treatment and Prevention. Elsevier/North-Holland Biomedical
Press, Amsterdam, pp. 443–459.
Levine, M.M., Black, R.E., Clements, M.L., et al., 1982. The pathogenicity of
nonenterotoxigenic Vibrio cholerae serogroup O1 biotype El Tor isolated from sewage
water in Brazil. J Infect Dis 145 (3), 296–299.
Levine, M.M., Black, R.E., Clements, M.L., et al., 1984. Evaluation in humans of attenuated
Vibrio cholerae El Tor Ogawa strain Texas Star-SR as a live oral vaccine. Infect. Immun.
43 (2), 515–522.
Levine, M.M., Kaper, J.B., Herrington, D., et al., 1988. Volunteer studies of deletion mutants
of Vibrio cholerae O1 prepared by recombinant techniques. Infect. Immun. 56 (1),
161–167.
Li, M., Sumida, T., Morris, J.G., et al., 2002. Evidence for the emergence of non-O1 and non-
O139 Vibrio cholerae strains with pathogenic potential by exchange of O-antigen bio-
synthesis regions. Infect. Immun. 70, 2441–2453.
Ludwig, A., 2015. Builder's Greywater Guide: Installation, Standards, and Science for
Builders, Landscapers, Regulators, Policymakers, Researchers, and Homeowners (In-
cludes Current Codes). A Supplement to the book “Create an Oasis with Greywater.”
23rd Printing. Oasis Design, Santa Barbara, CA http://oasisdesign.net/greywater/
(accessed 11.09.2016).
Merrell, D.S., Butler, S.M., Qadri, F., et al., 2002. Host-induced epidemic spread of the chol-
era bacterium. Nature 417 (6), 642–645.
Ministry of Health and Population in Haiti/U.S. Centers for Disease Control and Prevention
(CDC), 2011. Haiti cholera training manual: a full course for healthcare providers Ver-
sion 2.1, 2011. www.cdc.gov/haiticholera/pdf/haiticholera_trainingmanual_en.pdf.
(accessed 10.04.2016).
Morris, J.G., 2011. Cholera-modern pandemic disease of ancient lineage. Emerg. Infect.
Dis. 17, 2099–2104.
Morris Jr., J.G., Losonsky, G., Johnson, J.A., et al., 1995. Clinical and immunological
characteristics of Vibrio cholerae O139 Bengal infection in North American volunteers.
J Infect Dis 171 (4), 903–908.
Morris Jr., J.G., Sztein, M.B., Rice, E.W., et al., 1996. Vibrio cholerae O1 can assume a
chlorine-resistant rugose survival form that is virulent for humans. J Infect Dis 174,
1364–1368.
MSMR (Medical Surveillance Monthly Report), 2013a. Absolute and relative morbidity
burdens attributable to various illnesses and injuries, U.S. Armed Forces, 2012.
MSMR 20 (4), 5–10.
387
MSMR (Medical Surveillance Monthly Report), 2013b. Gastrointestinal infections, active
component, U.S. Armed Forces, 2002–2012. MSMR 20 (10), 7–11.
Mutreja, A., Kim, D.W., Thomson, N.R., et al., 2011. Evidence for several waves of global
transmission in the seventh cholera pandemic. Nature 477, 462–465.
NAS (National Academy of Sciences, Engineering and Medicine), 2016. Using Graywater
and Stormwater to Enhance Local Water Supplies: An Assessment of Risks, Costs
and Benefits. Committee on the Beneficial Use of Graywater and Stormwater;
Water Science and Technology Board; Division on Earth and Life Studies; National
Academies of Sciences, Engineering and Medicine. National Academy of Sciences,
Washington, DC.
Onifade, T.J.M., Hutchinson, R., Van Zile, K., Bodager, D., Baker, R., Blackmore, C., 2011.
Toxin producing Vibrio cholerae O75 outbreak, United States, March to April 2011.
Euro Surveill. 16 (20), 19870.
Rha, B., Lopman, B.A., Alcala, A.N., Porter, C., 2016. Incidence of Norovirus-associated med-
ical encounters among active duty United States military personnel and their depen-
dents. PLoS One 11 (4), e0148505.
Sack, D.A., Tacket, C.O., Cohen, M.B., et al., 1998. Validation of a volunteer model of cholera
with frozen bacteria as the challenge. Infect. Immun. 66 (5), 1968–1972.
Safa, A., Nair, G.B., Kong, R.Y.C., 2009. Evolution of new variants of Vibrio cholerae O1.
Trends Microbiol. 18 (1), 46–54.
Schoen, M.E., Garland, J., 2017. Review of pathogen treatment reductions for onsite non-
potable reuse of alternative source waters. Microbial Risk Anal. 5, 25–31.
Schoen, M.E., Ashbolt, N.J., Jahne, M.A., Garland, J., 2017. Risk-based enteric pathogen tar-
gets for non-potable and direct potable use of roof runoff, stormwater, and
greywater. Microbial Risk Anal. 5, 32–43.
Silva, T.M.J., Schleupner, M.A., Tacket, C.O., et al., 1996. New evidence for an inflammatory
component in diarrhea caused by selected new, live, attenuated cholera vaccines and
by El Tor and O139 Vibrio cholerae. Infect. Immun. 64 (6), 2362–2364.
Sinclair, M., Roddick, F., Nguyen, T., et al., 2016. Measuring water ingestion from spray ex-
posures. Water Res. 99, 1–6.
Szinnai, G., Schachinger, H., Arnaud, M.J., et al., 2005. Effect of water deprivation on
cognitive-motor performance in healthy men and women. Am. J. Phys. Regul. Integr.
Comp. Phys. 289, R275–R280.
Taylor, D.N., Killeen, K.P., Hack, D.A., et al., 1994. Development of a live, oral, attenuated
vaccine against El Tor cholera. J Infect Dis 170 (6), 1518–1523.
Umble, A., 2014. Engineering Angle: Fit-for-purpose Water Reuse and the Road Toward
Water Security. July 29, 2014 Guest Column - Water Online. http://
www.wateronline.com/doc/engineering-angle-fit-for-purpose-water-reuse-and-
the-road-toward-water-security-0001 (accessed 11.09.2016).
UN (United Nations), 2011. Final Report of the Independent Panel of Experts on the Chol-
era Outbreak in Haiti. https://www.un.org/News/dh/infocus/haiti/UN-cholera-re-
port-final.pdf (accessed 11.10.2016).
USACHPPM (U.S. Army Center for Health Promotion and Preventive Medicine), 2009.
Health-Based Detection Goals for Monitoring Biological Agent in Drinking Water.
USACHPPM Project No. 31-EC-093Y-09. U.S. Army Center for Health Promotion and
Preventive Medicine, Aberdeen Proving Ground, Maryland.
USAMRIID (U.S. Army Medical Research Institute of Infectious Diseases), 1998. Medical
Management of Biological Casualties. 3rd ed. U. S. Army Medical Research Institute
of Infectious Diseases, Fort Detrick, Maryland.
USAPHC (U.S. Army Public Health Center), 2017. Assessment of the Applicability of Pub-
lished Water Reuse Guidelines for Contingency Operations. PHIP-39-06-0417. Envi-
ronmental Health Risk Assessment Division, Army Public Health Center, Aberdeen
Proving Ground, MD.
USAPHC (U.S. Army Public Health Command), 2014a. Water Reuse in Contingency Oper-
ations: A Study for the Army Study Program Management Office. Health Risk Analysis
and Recommendations for a Comprehensive Risk Management Approach to Quality
Assurance. Technical Guide (TG) 364a. US Army Public Health Command, Aberdeen
Proving Ground, MD.
USAPHC (U.S. Army Public Health Command), 2014b. Microbial Risk Assessment for Un-
restricted Wastewater Reuse during Army Deployments. Public Health Information
Paper PHIP 39-01-0514. US Army Public Health Command, Aberdeen Proving
Ground, MD.
USEPA (U.S. Environmental Protection Agency), 2011. Exposure Factors Handbook. 2011
edition. EPA/600/R-09/052F. U.S. Environmental Protection Agency, National Center
for Environmental Assessment, Washington, DC http://www.epa.gov/ncea/efh
(accessed 10.04.2016).
USEPA (U.S. Environmental Protection Agency), 2015. Benchmark Dose Software and
Technical Guidance, Version 2.6. https://www.epa.gov/bmds (accessed 7.12.2015).
Weir, M., 2012. Vibrio cholerae: Dose Response Models. Center for Advanced Microbial
Risk Assessment (CAMRA) http://www.camra.msu.edu/ (accessed 10.04.2016).
WHO (World Health Organization), 2017. Global epidemics and impact of cholera. http://
www.who.int/topics/cholera/impact/en/ (accessed 01.12.2017).