See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/311759916

Cholera Rapid Test with Enrichment Step Has Diagnostic Performance

Equivalent to Culture

Article  in  PLoS ONE · December 2016

DOI: 10.1371/journal.pone.0168257

CITATIONS READS
30 246

20 authors, including:

Lameck N. Ontweka Jean Rauzier

5 PUBLICATIONS   110 CITATIONS   
Institut Pasteur International Network
74 PUBLICATIONS   3,606 CITATIONS   
SEE PROFILE
SEE PROFILE

Lucy Anne Parker Joseph Francis Wamala

Universidad Miguel Hernández de Elche World Health Organisation WHO, Juba, South Sudan
77 PUBLICATIONS   1,569 CITATIONS    103 PUBLICATIONS   2,348 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Pandemic 2009 influenza View project

Medecins Sans Frontieres, Eswatini project View project

All content following this page was uploaded by Lucy Anne Parker on 14 May 2018.

The user has requested enhancement of the downloaded file.

 RESEARCH ARTICLE

Cholera Rapid Test with Enrichment Step Has

Diagnostic Performance Equivalent to Culture
Lameck N. Ontweka1,2, Lul O. Deng3, Jean Rauzier4, Amanda K. Debes5,
Fisseha Tadesse6, Lucy A. Parker1, Joseph F. Wamala7, Bior K. Bior3, Michael Lasuba3,
Abiem Bona But3, Francesco Grandesso6, Christine Jamet1, Sandra Cohuet6,
Iza Ciglenecki1, Micaela Serafini1, David A. Sack5, Marie-Laure Quilici4, Andrew
S. Azman8,1, Francisco J. Luquero5,6, Anne-Laure Page6*
1 Médecins Sans Frontières Operational Center Geneva, Geneva, Switzerland, 2 Amref Health Africa
Headquarters, Nairobi, Kenya, 3 National Public Health Laboratory, Ministry of Health, Juba, South Sudan,
4 Enteric Bacterial Pathogens Unit, National Reference Centre for Vibrios and Cholera, Institut Pasteur,
Paris, France, 5 Department of International Health, Johns Hopkins University, Baltimore, United States of
America, 6 Field epidemiology, Epicentre, Paris, France, 7 World Health Organization, Juba, South Sudan,
a11111 8 Department of Epidemiology, Johns Hopkins University, Baltimore, United States of America

* anne-laure.page@epicentre.msf.org

Abstract
OPEN ACCESS Cholera rapid diagnostic tests (RDT) could play a central role in outbreak detection and sur-
Citation: Ontweka LN, Deng LO, Rauzier J, Debes veillance in low-resource settings, but their modest performance has hindered their broad
AK, Tadesse F, Parker LA, et al. (2016) Cholera adoption. The addition of an enrichment step may improve test specificity. We describe the
Rapid Test with Enrichment Step Has Diagnostic
results of a prospective diagnostic evaluation of the Crystal VC RDT (Span Diagnostics,
Performance Equivalent to Culture. PLoS ONE 11
(12): e0168257. doi:10.1371/journal. India) with enrichment step and of culture, each compared to polymerase chain reaction
pone.0168257 (PCR), during a cholera outbreak in South Sudan. RDTs were performed on alkaline pep-
Editor: Vishnu Chaturvedi, Wadsworth Center, tone water inoculated with stool and incubated for 4–6 hours at ambient temperature. Chol-
UNITED STATES era culture was performed from wet filter paper inoculated with stool. Molecular detection of
Received: October 27, 2016 Vibrio cholerae O1 by PCR was done from dry Whatman 903 filter papers inoculated with
stool, and from wet filter paper supernatant. In August and September 2015, 101 consecu-
Accepted: November 28, 2016
tive suspected cholera cases were enrolled, of which 36 were confirmed by PCR. The
Published: December 19, 2016
enriched RDT had 86.1% (95% CI: 70.5–95.3) sensitivity and 100% (95% CI: 94.4–100)
Copyright: © 2016 Ontweka et al. This is an open specificity compared to PCR as the reference standard. The sensitivity of culture versus
access article distributed under the terms of the
PCR was 83.3% (95% CI: 67.2–93.6) for culture performed on site and 72.2% (95% CI:
Creative Commons Attribution License, which
permits unrestricted use, distribution, and 54.8–85.8) at the international reference laboratory, where samples were tested after an
reproduction in any medium, provided the original average delay of two months after sample collection, and specificity was 98.5% (95% CI:
author and source are credited. 91.7–100) and 100% (95% CI: 94.5–100), respectively. The RDT with enrichment showed
Data Availability Statement: All relevant data are performance comparable to that of culture and could be a sustainable alternative to culture
available in the Supporting Information files. confirmation where laboratory capacity is limited.
Funding: This study was funded by Médecins Sans
Frontières Operational Center Geneva. The funders
had staff (co-authors of this manuscript) who
played a role in study design, data collection and
analysis, and preparation of the manuscript. The
corresponding author had full access to all data
and had final responsibility for the decision to
submit for publication.

PLOS ONE | DOI:10.1371/journal.pone.0168257 December 19, 2016 1/8

 Evaluation of Cholera Rapid Test with Enrichment Step

Competing Interests: The authors have declared Introduction

that no competing interests exist.
Cholera continues to be a major public health problem for developing countries with an esti-
mated 2.8 million cholera cases and around 100,000 deaths each year worldwide [1]. Countries
with the highest incidence rates are in Africa, Southern Asia and the Caribbean, where surveil-
lance systems are often insensitive and unable to rapidly detect the transmission of epidemic
pathogens [2].
Rapid identification and confirmation of initial cases in the early phase of cholera epidemics
is critical for timely public health responses to control outbreaks. Diagnostic delays may result
in higher case numbers and case fatality rates, leading to an enormous health and economic
burden to affected countries. Currently, isolation of Vibrio cholerae O1 by stool culture is nec-
essary for cholera outbreak confirmation and remains the gold standard for diagnosis [2].
However, this procedure requires laboratory infrastructure, adequate transport procedures
and well trained staff. Moreover, the delay in obtaining results includes the 2 to 3-day duration
of the microbiological procedure, in addition to the time for transportation of the sample to
the closest laboratory. Culture sensitivity is also imperfect and can be affected by the delays in
transport to the laboratory, as well as prior consumption of antibiotics [3]. Polymerase chain
reaction (PCR) is becoming more commonly used to detect V. cholerae, with the advantages of
being faster and generally more sensitive than culture [3–5]. However, like culture, PCR
requires laboratory infrastructure and well trained staff, which are often not available at sites
where cholera outbreaks occur.
A rapid diagnostic test (RDT) that is accurate, simple, easy to use and interpret, and stored
at ambient temperature would be a useful tool for the early detection and confirmation of chol-
era outbreaks. Several rapid tests for cholera have been developed, based mostly on lateral flow
immunochromatographic techniques [6]. One of the most widely used RDTs for cholera,
Crystal VCTM (Span Diagnostics, Surat, India) or its prototype developed by Institut Pasteur
(IP), have been studied in different contexts, and have consistently shown high sensitivities
(92–97%), but moderate specificities (49–80%) when used directly on bulk stools and com-
pared to culture as the gold standard [6–11]. While the moderate specificity may be partly
explained by the imperfect nature of culture as the gold-standard, analyses including PCR or
using statistical models taking into account this bias only increased the specificity to 85% [4].
Alkaline peptone water (APW) is a commonly used enrichment medium for V. cholerae. In
initial assessments of the prototype developed by IP, the test was also evaluated on rectal swabs
inoculated in APW and incubated at room temperature for 4 hours [7,12]. This latter method
gave a much higher specificity, up to 97%, compared to using the RDT directly on bulk stool.
Both methods are now included in the package insert of Crystal VC: the “initial screening”
method on direct stool diluted in a sample processing reagent, and the “confirmation” method
on stool incubated in APW for 4 hours. So far, only one study conducted in Bangladesh has
evaluated the performance of both methods with the current version of the Crystal VC test,
and showed that the enriched “confirmation” method had both higher sensitivity and higher
specificity compared to the direct screening method [13]. However, in this study, the perfor-
mance of the direct test was not consistent with previously published studies, with low sensitiv-
ity (65%) and high specificity (92%); the reasons for these discrepancies were not discussed,
but could be due to the addition of a step of dilution of the stools in a sample preparation vial.
Another study confirmed the high specificity of RDT after enrichment in APW but sensitivity
could not be reliably estimated due to the low number of confirmed cholera cases [14].
In this study we evaluated the performance of Crystal VC used after a 4–6 hour enrichment
in APW compared to PCR as the gold standard method during a cholera outbreak in Juba,
South Sudan.

PLOS ONE | DOI:10.1371/journal.pone.0168257 December 19, 2016 2/8

 Evaluation of Cholera Rapid Test with Enrichment Step

Materials and Methods

Ethics
Ethical approval for this study was obtained from the John Hopkins Bloomberg School of Pub-
lic Health Institutional Review Board and the South Sudan Ministry of Health, Directorate of
Monitoring, Evaluation and Research. Written informed consent was obtained from all partic-
ipants enrolled in the study with parental or guardian consent for participants under 18 years
of age.

Study population
This study was conducted in Juba, the capital of South Sudan, during a cholera outbreak that
lasted for five months from May to September 2015. This study was nested in a vaccine effec-
tiveness study following a large oral cholera vaccination campaign [15]. The study population
consisted of all individuals aged one year and older presenting with three or more loose stools
in the preceding 24 hours at participating cholera treatment centres (CTCs) and oral rehydra-
tion posts around the city in August and September 2015.
Written informed consent to participate in the study was obtained once the patient was in
stable condition. If no consent was provided, the stool sample was tested for local surveillance
purposes, but the results were not used in the study.
We subsequently excluded all patients who reported to have taken the vaccine within 7 days
prior to sample collection, since rapid test can become positive in vaccine recipients for up to 6
days after vaccination [16].

Sample collection and processing

Upon admission, a stool sample was collected in a clean (non-disinfected), unused container.
Two drops of watery stool were transferred to a vial containing 3 ml of APW broth and kept at
ambient temperature for 4 to 6 hours.
Study staff soaked two 6-mm filter paper disks in each fresh watery stool sample and then
placed each filter paper into separate microtubes with two drops of normal saline. One micro-
tube was sent immediately to the National Public Health Laboratory (NPHL) in Juba, and the
other was stored at ambient temperature until the end of the study and sent to Institut Pasteur
(IP), Paris, France.
Using a Pasteur pipette, one to two drops of direct stool or enriched APW medium were
spotted onto a Whatman 903 Protein Saver Card (GE Healthcare Ltd., Forest Farm, Cardiff,
UK) and allowed to air-dry. Dry filter papers were packed individually with a desiccant bag
and stored at ambient temperature until processing. Dried filter papers were sent to Johns
Hopkins University (JHU) to be tested for V. cholerae using molecular methods.

Rapid test procedure

Rapid tests were performed at the CTC by three nurses, who were trained on the study proce-
dures (including rapid tests) for two days prior to the study start. RDT kits were stored at
ambient temperature.
For the enriched method, after the 4–6 hour incubation of APW at ambient temperature,
two drops of enriched medium were placed in the test tube and the dipstick was inserted. The
result was read after 15 minutes by trained study staff, and interpreted following the manufac-
turer’s recommendation. The test was considered positive if the control line and either line T2
(O1) or T1 (O139) or both (O1 and O139) showed pinkish red lines, negative if the control
line only showed a pinkish red line and invalid if the control line did not show any coloration.

PLOS ONE | DOI:10.1371/journal.pone.0168257 December 19, 2016 3/8

 Evaluation of Cholera Rapid Test with Enrichment Step

The staff reading the enriched test were not blinded to the results of the direct test, but were
blinded to the results of culture and PCR. A picture of each test was taken and results were re-
confirmed by the study co-investigators.
Rapid tests were also performed using two drops of direct stool. Since this procedure did
not strictly follow the manufacturer’s recommendations, which includes dilution in a sample
diluent buffer, we did not include the results in the main analysis and provide the correspond-
ing data in S1 Appendix.

Stool culture
Upon arrival in both laboratories, culture was performed from the wet filter papers by trained
laboratory technicians using standard methods including enrichment in APW [17]. Briefly, a
loopful of supernatant from the wet filter papers was cultured on thiosulfate-citrate-bile salt-
sucrose (TCBS) agar and, at NPHL, on MacConkey agar, as selective plating media, and on
blood agar or alkaline nutrient agar as nonselective plating media. In addition, the wet filter
papers were placed in APW solution and incubated for 6–8 hours at 37˚ C. After incubation,
one loopful of APW from the topmost portion of the broth, where V. cholerae preferentially
grows, was sub-cultured on selective and nonselective media as described above. Culture plates
were incubated overnight at a temperature of 35–37˚C. Screening of isolates was based on
morphological appearance on selective and nonselective media, Gram staining, motility and
oxydase testing. Identification was confirmed by serological testing performed using polyva-
lent O1- and O139-specific antisera (Bio-Rad, USA and Denka Seiken Co, Japan, respectively)
and serotype determination by monovalent Inaba and Ogawa antisera (Denka Seiken Co,
Japan), all performed on colonies grown on nonselective media.
The median delay between sample collection and culture was 1 day in Juba and 73 days
(range: 26 to 82 days) in Paris.

Polymerase Chain Reaction (PCR) analysis

In the absence of an internationally recognized standard method for molecular detection of V.
cholerae O1, PCR was performed in parallel in two laboratories both on different matrices and
different target genes.
At JHU, DNA from each dried filter paper specimen was extracted using chelex-100 (Bio-
Rad) and subsequent PCR amplification was performed for the detection of the outer mem-
brane protein of V. cholerae, ompW, for species confirmation; the cholera toxin A gene, ctxA,
to assess the toxigenic potential of the strains [18]; and the rfb gene for the identification of the
O1 or O139 serogroups [19]. All negative samples were further tested for the presence of 16S
rDNA to confirm DNA preservation techniques as described by Hasan et al. [20]. PCR was
performed according to methods previously described [14].
At IP, DNA was extracted from the wet paper supernatant and from the enrichment in
APW by boiling the samples for 10 minutes at 100˚C. PCR was performed to detect an inter-
genic spacer region specific of V. cholerae species (ISR gene) [21]. On samples positive for V.
cholerae, the rfb gene was amplified for the identification of V. cholerae serogroup O1 and
O139 as described by Hoshino [19]. On negative samples, PCR was performed on a 1/10 dilu-
tion of the target DNA to check for the presence of inhibitors.

Data analysis
The reference standard was considered as positive if the PCR result was positive for V. cholerae
O1 either at JHU or at IP. For the main analysis, the RDT was considered positive if the O1
line was positive and negative if the O1 line was negative, irrespective of the result of the O139

PLOS ONE | DOI:10.1371/journal.pone.0168257 December 19, 2016 4/8

 Evaluation of Cholera Rapid Test with Enrichment Step

line. Additional analysis was performed considering the RDT as positive if either the O1 or the
O139 line was positive, and negative if both lines were negative. Culture was considered posi-
tive if V. cholerae O1 was isolated from stool and negative if no V. cholerae was found, or if
non-O1 non-O139 V. cholerae was isolated.
Data analysis was performed using Stata 13 (Stata Corporation, College Station, Texas,
USA), with the estimates of diagnostic performance produced using the diagt command,
which displays summary statistics for diagnostic tests and provides exact binomial confidence
intervals for sensitivity, specificity, and predictive values.
For the comparison between rapid test and culture performance, we used McNemar’s chi-2
test and applied it globally, as well as separately on samples that were positive and negative by
the reference standard to assess for possible statistical differences in sensitivity and specificity.
We also used the Cohen’s kappa coefficient to assess global agreement between the assays.

Results
From August 9 to September 29, 2015, 110 patients attending the study health facilities were
screened for inclusion in the study. One patient was excluded because the stool sample was not
collected and five did not provide informed consent. In addition, three patients who reported
to have received a single dose of oral cholera vaccine within 7 days prior to consultation were
excluded. In total, 101 patients were included in these analyses, with a majority of males
(n = 59/99, 59.6%) and a median age of 26 years (interquartile range: 8–35). Of 83 patients
with clinical information available, most had severe dehydration (n = 51, 61.5%), 11 (13.3%)
reported having received antibiotics prior to admission and 16 (19.3%) received antibiotics at
the CTC before sample collection. Twelve (12.0%) reported to have taken a single dose of oral
cholera vaccine more than one week (range 8–35 days) prior to the consultation.
The RDT with enrichment was positive for O1 in 31, with a weak line in 2 of them (6.5%).
None of the enriched RDTs had a positive O139 reading. Culture was positive for V. cholerae
O1 in 31 patients at the Juba NPHL and in 26 at IP (Table 1). PCR was positive for V. cholerae
O1 from 31 wet filter papers at IP and from 35 dry filter papers at the JHU laboratory, resulting
in a PCR-positive result in 36 (35.6%) of the 101 specimens.
When compared to PCR as the reference standard, the enriched test showed moderate sen-
sitivity (86.1%) but very high specificity (100%), which resulted in a 100% positive predictive
value (Table 2). While on-site culture showed similar performance to the enriched rapid test

Table 1. Results of the enriched RDT and of culture at National Public Health Laboratory, Juba, and at
Institut Pasteur, Paris, compared to PCR results.
PCR V. cholerae O1 (reference standard)
Positive Negative Total
Enriched rapid test
Positive O1 31 0 31
Negative 5 64 69
Not done 0 1 1
Culture—NPHL
Positive 30 1 31
Negative 6 64 70
Culture—IP
Positive 26 0 26
Negative 10 65 75
Total 36 65 101
doi:10.1371/journal.pone.0168257.t001

PLOS ONE | DOI:10.1371/journal.pone.0168257 December 19, 2016 5/8

 Evaluation of Cholera Rapid Test with Enrichment Step

Table 2. Diagnostic performance of direct and enriched RDT, and of culture at National Public Health Laboratory, Juba, and at Institut Pasteur,
Paris, using PCR as the reference standard in all (N = 101) or patients without prior antibiotics (N = 80).
Sensitivity Specificity PPV NPV
% (95% CI) % (95% CI) % (95% CI) % (95% CI)
All
Enriched RDT 86.1 (70.5–95.3) 100 (94.4–100) 100 (88.8–100) 92.8 (83.9–97.6)
Culture NPHL 83.3 (67.2–93.6) 98.5 (91.7–100) 96.8 (83.3–99.9) 91.4 (82.3–96.8)
Culture IP 72.2 (54.8–85.8) 100 (94.5–100) 100 (86.8–100) 86.7 (76.8–93.4)
No prior antibiotics
Enriched RDT 87.5 (67.6–97.3) 100 (93.6–100) 100 (83.9–100) 94.9 (85.9–98.9)
Culture NPHL 87.5 (67.6–97.3) 98.2 (90.4–100) 95.5 (77.2–99.9) 94.8 (85.6–98.9)
Culture IP 70.8 (48.9–87.4) 100 (93.6–100) 100 (80.5–100) 88.9 (78.4–95.4)
doi:10.1371/journal.pone.0168257.t002

(exact McNemar chi-2 test p = 1 globally and for PCR-positive and negative samples sepa-
rately, kappa = 81.3%), delayed culture at IP showed a lower sensitivity of 72.2% (exact McNe-
mar chi-2 test p = 0.27 globally and for PCR-positive samples, p = 1 for PCR-negative samples,
kappa = 68.2%). Excluding patients with self-reported antibiotic intake either before or at the
CTC did not change the performance significantly (Table 2).

Discussion
This field evaluation of the enriched rapid test for diagnosis of cholera confirmed initial
reports suggesting that the APW incubation step ensures high specificity, which improves con-
fidence in the test results and could diminish the risk of false cholera outbreak alerts [13,14].
Incubation in APW for 4–6 hours is also the first step in laboratory methods used for the diag-
nosis of V. cholerae from faecal specimens by culture. The subsequent steps of sub-culture and
classical bacteriological identification are replaced here by the RDT for detection of V. cholerae
O1 or O139. When used on enriched specimens, the RDT proved that it was sufficiently sensi-
tive and specific for rapid and accurate detection and identification of V. cholerae, making this
method comparable to a simplified and rapid version of culture-based identification of V. cho-
lerae. This is confirmed by the fact that the test had similar performance as culture when per-
formed within 1 or 2 days of stool collection, as at the Juba NPHL laboratory, and even better
performance than culture in case of long delay between sampling and testing, as in the IP labo-
ratory, where culture was performed more than 2 months after sampling.
One possible disadvantage of the enriched method is that, like culture, it depends on the
presence of culturable organisms, while the direct test could also detect non-viable and non-
culturable organisms. Although it is reasonable to assume that under normal conditions of use
of the tests in the field, the absence of viable organisms in the stools of cholera patients is
unlikely, it is important to highlight that enrichment could be affected by problems linked to
sample collection and storage, such as containers with disinfectants, poor sampling or han-
dling practices with long delays or inappropriate temperature, or to prior antibiotic consump-
tion. Here, exclusion of patients with self-reported previous antibiotic consumption did not
significantly change the test performance, but numbers were too limited for a formal stratified
analysis, and the class of antibiotics and delay between intake and testing were not known.
There were several limitations to this study. First, the sample size was low, due to the fact
that the study was undertaken when the epidemic was already declining and after an OCV vac-
cination campaign. Second, although the RDT was also performed without enrichment, the
fact that we did not strictly follow the current manufacturer’s recommendations and did not
use the sample diluent buffer for the direct method prevented us from doing a formal

PLOS ONE | DOI:10.1371/journal.pone.0168257 December 19, 2016 6/8

 Evaluation of Cholera Rapid Test with Enrichment Step

comparison between these methods. The reasons for not strictly following the recommended
procedure was due to fear of a decreased sensitivity with the dilution of ~200 μL of stool in 1
mL diluent buffer and to habits with previous versions of the test. Indeed, it should be noted
that the initial version of Crystal VC recommended that the test be performed directly on liq-
uid stool, with no diluent, and all but one of the evaluations of Crystal VC published until now
used this previous method. Only one study published so far was done with the current version
of Crystal VC using the sample diluent buffer and showed the lowest sensitivity reported so far
(66%) and high specificity of 92% [13]. Thus, we cannot exclude the possibility that the dilu-
tion itself, rather than enrichment in APW, might increase the specificity. However, it seems
that dilution alone does not eliminate the problem with false positive O139 results with the
direct RDT method, as encountered here (see S1 Appendix), since this issue was also reported
after dilution in the sample diluent [13]. In addition, the results from the study in Bangladesh
suggests that the use of the sample diluent could reduce the RDT sensitivity to unacceptably
low levels.
In conclusion, our results show that the RDT used with a simple step of enrichment in
APW has performance similar to that of culture. This method could be a sustainable alterna-
tive to culture confirmation of cases in places where laboratory capacity is limited. Culture will
remain needed for phenotypic analysis, including antibiotic susceptibility testing, and further
molecular characterization, which is essential for worldwide disease surveillance.

Supporting Information
S1 Appendix. Results and performance of the RDT performed directly on stool.
(DOCX)
S1 Dataset. Study dataset.
(XLSX)

Acknowledgments
We would like to thank all the study staff who performed the field work at the participating
cholera treatment centres in Juba and other MSF staff who supported the logistics of this
study. We thank Andrew Nicholson for the laboratory work. We are also grateful to all study
participants.

Author Contributions
Conceptualization: IC ASA FJL ALP.
Formal analysis: LNO ASA FJL ALP.
Funding acquisition: CJ IC MS.
Investigation: LNO LOD JR AKD FT LAP BKB ML ABB.
Project administration: JFW FG CJ SC IC MS DAS MLQ.
Supervision: LAP DAS MLQ ASA FJL ALP.
Writing – original draft: LNO AKD MLQ ALP.
Writing – review & editing: ASA FJL IC SC DAS.

PLOS ONE | DOI:10.1371/journal.pone.0168257 December 19, 2016 7/8

 Evaluation of Cholera Rapid Test with Enrichment Step

References
1. Ali M, Nelson AR, Lopez AL, Sack DA. Updated global burden of cholera in endemic countries. PLoS
Negl Trop Dis. 2015; 9: e0003832. doi: 10.1371/journal.pntd.0003832 PMID: 26043000
2. World Health Organization. Cholera, 2014. Wkly Epidemiol Rec. 2015; 90: 517–28. PMID: 26433979
3. Alam M, Hasan NA, Sultana M, Nair GB, Sadique A, Faruque ASG, et al. Diagnostic Limitations to
Accurate Diagnosis of Cholera. J Clin Microbiol. American Society for Microbiology; 2010; 48: 3918–
3922. doi: 10.1128/JCM.00616-10 PMID: 20739485
4. Page A- L, Alberti KP, Mondonge V, Rauzier J, Quilici M- L, Guerin PJ. Evaluation of a rapid test for the
diagnosis of cholera in the absence of a gold standard. PLoS One. 2012; 7: e37360. doi: 10.1371/
journal.pone.0037360 PMID: 22666350
5. Sinha A, Sengupta S, Ghosh S, Basu S, Sur D, Kanungo S, et al. Evaluation of a rapid dipstick test for
identifying cholera cases during the outbreak. Indian J Med Res. 2012; 135: 523–528. PMID: 22664501
6. Dick MH, Guillerm M, Moussy F, Chaignat CL. Review of Two Decades of Cholera Diagnostics—How
Far Have We Really Come? PLoS Negl Trop Dis. 2012; 6.
7. Wang X- Y, Ansaruzzaman M, Vaz R, Mondlane C, Lucas MES, von Seidlein L, et al. Field evaluation of
a rapid immunochromatographic dipstick test for the diagnosis of cholera in a high-risk population. BMC
Infect Dis. 2006; 6: 17. doi: 10.1186/1471-2334-6-17 PMID: 16451731
8. Kalluri P, Naheed A, Rahman S, Ansaruzzaman M, Faruque ASG, Bird M, et al. Evaluation of three
rapid diagnostic tests for cholera: Does the skill level of the technician matter? Trop Med Int Heal. 2006;
11: 49–55.
9. Ley B, Khatib AM, Thriemer K, von Seidlein L, Deen J, Mukhopadyay A, et al. Evaluation of a rapid dip-
stick (crystal VC) for the diagnosis of Cholera in Zanzibar and a comparison with previous studies.
PLoS One. 2012; 7: 3–10.
10. Mukherjee P, Ghosh S, Ramamurthy T, Bhattacharya MK, Nandy RK, Takeda Y, et al. Evaluation of a
rapid immunochromatographic dipstick kit for diagnosis of cholera emphasizes its outbreak utility. Jpn J
Infect Dis. 2010; 63: 234–238. PMID: 20657061
11. Boncy J, Rossignol E, Dahourou G, Hast M, Buteau J, Stanislas M, et al. Performance and utility of a
rapid diagnostic test for cholera: notes from Haiti. Diagn Microbiol Infect Dis. 2013; 76: 521–3. doi: 10.
1016/j.diagmicrobio.2013.03.010 PMID: 23886437
12. Bhuiyan NA, Qadri F, Faruque ASG, Malek MA, Salam MA, Nato F, et al. Use of Dipsticks for Rapid
Diagnosis of Cholera Caused by Vibrio cholerae O1 and O139 from Rectal Swabs. J Clin Microbiol.
2003; 41: 3939–3941. doi: 10.1128/JCM.41.8.3939-3941.2003 PMID: 12904424
13. George CM, Rashid M, Sack DA, Sack RB, Saif-Ur-Rahman KM, Azman AS, et al. Evaluation of enrich-
ment method for the detection of Vibrio cholerae O1 using a rapid dipstick test in Bangladesh. Trop Med
Int Health. 2014; 19: 301–7. doi: 10.1111/tmi.12252 PMID: 24401137
14. Debes AK, Ateudjieu J, Guenou E, Ebile W, Sonkoua IT, Njimbia AC, et al. Clinical and Environmental
Surveillance for Vibrio cholerae in Resource Constrained Areas: Application During a 1-Year Surveil-
lance in the Far North Region of Cameroon. Am J Trop Med Hyg. 2016; 94: 537–43. doi: 10.4269/ajtmh.
15-0496 PMID: 26755564
15. Azman AS, Parker LA, Romunu J, Tadesse F, Grandesso F, Deng LL, et al. The Effectiveness of One
Dose of Oral Cholera Vaccine in Response to an Outbreak: A Case-Cohort Study. Lancet Glob Heal.;
4: e856–e863.
16. Martinez-Pino I, Luquero FJ, Sakoba K, Sylla S, Haile M, Grais RF, et al. Use of a cholera rapid diagnos-
tic test during a mass vaccination campaign in response to an epidemic in Guinea, 2012. PLoS Negl
Trop Dis. 2013; 7: e2366. doi: 10.1371/journal.pntd.0002366 PMID: 23967359
17. Dodin A, Fournier J-M. Diagnosis of the cholera vibrio. Laboratory methods for the diagnosis of cholera
vibrio and other vibrios. Institut P. Institut Pasteur (France); pp. 59–82.
18. Nandi B, Nandy RK, Mukhopadhyay S, Nair GB, Shimada T, Ghose AC. Rapid method for species-spe-
cific identification of Vibrio cholerae using primers targeted to the gene of outer membrane protein
OmpW. J Clin Microbiol. 2000; 38: 4145–51. PMID: 11060082
19. Hoshino K, Yamasaki S, Mukhopadhyay AK, Chakrabarti S, Basu A, Bhattachariya SK, et al. Develop-
ment and evaluation of a multiplex PCR assay for rapid detection of toxigenic Vibrio cholerae O1 and
O139. FEMS Immunol Med Microbiol. 1988; 20: 207–210.
20. Hasan NA, Chowdhury WB, Rahim N, Sultana M, Shabnam SA, Mai V, et al. Metagenomic 16S rDNA
Targeted PCR-DGGE in Determining Bacterial Diversity in Aquatic Ecosystem. Bangladesh J Microbiol.
2011; 27: 46–50.
21. Chun J, Huq A, Colwell RR. Analysis of 16S-23S rRNA intergenic spacer regions of Vibrio cholerae and
Vibrio mimicus. Appl Environ Microbiol. 1999; 65: 2202–8. PMID: 10224020

PLOS ONE | DOI:10.1371/journal.pone.0168257 December 19, 2016 8/8

View publication stats

Am. J. Trop. Med. Hyg., 94(3), 2016, pp. 537–543
doi:10.4269/ajtmh.15-0496
Copyright © 2016 by The American Society of Tropical Medicine and Hygiene

Clinical and Environmental Surveillance for Vibrio cholerae in Resource Constrained Areas:
Application during a 1-Year Surveillance in the Far North Region of Cameroon
Amanda K. Debes,* Jerome Ateudjieu, Etienne Guenou, Walter Ebile, Isaac Tadzong Sonkoua, Anthony Chebe Njimbia,
Peter Steinwald, Malathi Ram, and David A. Sack
Johns Hopkins University Bloomberg School of Public Health, Baltimore, Maryland; Department of Biomedical Sciences,
University of Dschang, Dschang, Cameroon; Meilleur Accès aux Soins de Santé (M.A. SANTE), Yaoundé, Cameroon;
Clinical Research Unit, Division of Health Operations Research, Ministry of Public Health, Yaoundé, Cameroon

Abstract. Biological confirmation of the presence of Vibrio cholerae in clinical and environmental samples is often
constrained due to resource- and labor-intensive gold standard methods. To develop low-cost, simple, and sustainable sur-
veillance techniques, we modified previously published specimen sampling and culture techniques and applied the use of
enriched dipstick testing in conjunction with the use of filter paper for DNA specimen preservation during clinical and
environmental surveillance in the Far North of Cameroon from August 2013 to October 2014. The enriched dipstick
methodology during routine use in a remote setting demonstrated a specificity of 99.8% compared with polymerase chain
reaction (PCR). The novel application of filter paper as a preservation method for cholera DNA specimens reduced the
need for cold chain storage and allowed for PCR characterization and confirmation of V. cholerae. The application of
basic technologies such as the enriched dipstick, the use of simplified gauze filtration for environmental sample collection,
and the use of filter paper for sample preservation enabled early case identification with reduced logistics and supply cost
while reporting minimal false-positive results. Simplified laboratory and epidemiological methodologies can improve the
feasibility of cholera surveillance in rural and resource-constrained areas, facilitating early case detection and rapid
response implementation.

INTRODUCTION Despite its comparative economic advantages over popula-

Accurate and reliable infectious disease burden data are
often minimally available or nonexistent in developing coun-
tries. Cholera disease burden data are similarly deficient due
to lack of surveillance in endemic areas as well as a fear of
economic repercussions from reporting outbreaks, ultimately
resulting in reduced reporting at district and national levels.1
Reporting is also constrained by logistics and costs associated
with transport of samples to a reliable microbiology labora-
tory. Of the 47 countries that reported cholera cases and
deaths to the World Health Organization (WHO) in 2013, 22
were in Africa.2 In spite of reporting the largest cholera dis-
ease burden in the world, there continues to be a poor under-
standing of the disease patterns in much of Africa because of
constraints in detecting and confirming cholera cases. In fact,
most cases that are reported are based on clinical case defi-
nitions without laboratory confirmation; significantly more
cases remain undetected and unreported every year because
of the inability to confirm a case. This presents a challenge
for proper disease surveillance and laboratory capacity require-
ments, especially in remote, rural areas and contributes to a
lack of understanding of cholera disease burden.
Although population-based surveillance provides the most
accurate information for disease burden, it is too resource
intensive for most cholera settings. Another option, sentinel
surveillance, may be used in resource-limited settings to moni-
tor a disease of interest at a specific site. This method requires
fewer resources than population-based surveillance to identify
the disease if the disease of interest is present.3 However, since
cholera tends to be sporadic, sentinel surveillance may not
provide representative data of the complete disease burden.
*Address correspondence to Amanda K. Debes, Department of
International Health, Johns Hopkins University Bloomberg School
of Public Health, 615 N Wolfe Street, Baltimore, MD 21205. E-mail:
adebes1@jhu.edu
537
tion-based surveillance, sentinel surveillance can also be diffi-
cult to sustain in low-resource settings given other health and
political priorities and the lack of funding to support a con-
certed effort to track a single disease.
A modified approach to sentinel surveillance has been suc-
cessfully implemented in Bangladesh where cholera has a more
consistent seasonal pattern. Since 1997, environmental and
clinical surveillance has been conducted in different locations
in Bangladesh on a reduced schedule of surveillance for three
successive days every 15 days. This strategy has shown that a
simplified surveillance methodology can aide in documenting
disease burden with reduced operational costs, while providing
data to better understand the seasonality and transmission pat-
terns of the disease.4
Diagnostic confirmation of Vibrio cholerae infection is chal-
lenging in rural and resource-limited settings. The accepted
method of V. cholerae diagnosis is through culture confirma-
tion, requiring well-equipped laboratories and trained labora-
tory personnel. This method is time consuming, resource
intensive, and prohibitively expensive for these settings. As a
result, cholera is often only diagnosed using a clinical defini-
tion recommended by the WHO, which classifies cholera case
as a patient aged 5 years or more who develops severe dehy-
dration or dies from acute watery diarrhea.5,6 More often,
cholera is not diagnosed, and the patient is treated without
identifying the etiologic agent. Commercially available dipstick
tests, such as Crystal VC™ (Arkray Healthcare Pvt Ltd.,
Surat, India), allow for rapid diagnosis. Unfortunately, previous
studies have reported low specificity when using the kits with
direct testing of stool samples per recommendation by the
package insert.7–9 We have applied a modified approach to
using the dipstick in both stool10–12 and environmental speci-
mens13 that has previously demonstrated improved specificity
of the dipstick.
In this study, we demonstrate the successful implementation
of sentinel surveillance using low-cost and rapid laboratory diag-
nostics in a low-resource setting. The simplified methodology
538 DEBES AND OTHERS

facilitates concurrent environmental and clinical surveillance

to provide crucial disease burden information, determine hot
spots for cholera activity, improve understanding of cholera
transmission patterns, and provide tools for early disease detec-
tion. Further, we report the novel use of filter paper technology
for preservation of specimens for DNA extraction and molecular
processing. Although culture methods are often considered to
be the gold standard, they are not 100% sensitive, particularly
if the patient has self-medicated with antibiotics before seeking
treatment at a health facility. Therefore, for our study, we con-
sidered a positive polymerase chain reaction (PCR) to be the
model gold standard of detection.

MATERIALS AND METHODS

Study design. Cameroon has a population of approximately

20 million people, divided into 10 regions with varied geogra-
phies and diverse climate zones (25). The Far North region
of Cameroon (FNC) is located in the Sahel desert and is
composed of a predominantly rural population, of which less
than half have access to improved drinking water.14 The inci-
dence of cholera disease is becoming increasingly common
in Cameroon, particularly in the FNC where 10 outbreaks
occurred from 1996 to 2014. Of all of these, the outbreak in
2010 was the most severe with almost 10,000 cases and
599 deaths (a 6.37% case fatality rate).15 These numbers
likely underestimate the true burden of disease due to lack of
surveillance and laboratory capacity.
Diarrhea surveillance was established at seven local health
facilities in the FNC, in and around Lake Chad, including
Kousseri Regional Hospital (Kousseri Health District), Mada
District Hospital (Mada Health District), Ngouma Integrated
Health Center (Makary Health District), Maltam Integrated
Health Center (Goulfey Health District), Blangoua Sub-
divisional Medical Center (SMC) (Mada Health District),
Darak SMC (Mada Health District), and Naga Integrated
Health Center (Mada Health District). These sites were selected
as being geographically representative of the Cameroonian
health facilities near Lake Chad. The selected sentinel sites
implemented a simplified sampling methodology in which each
of the seven sentinel site district health facilities enrolled
consenting diarrheal cases of any age into the study for a 3-day
period every 15 days. During the intervening days, any sus-
pected cholera patient meeting the WHO definition of cholera
was also requested to be enrolled in the study. Concurrently,
environmental sampling was conducted 1 day of every 15 days.
Study enrollment began in August 2013. Subjects with bloody
diarrhea and diarrhea lasting more than 7 days were excluded.
Clinical surveillance. Fecal specimens were collected from
consenting diarrhea subjects and screened for V. cholerae O1
and O139 using an enhanced dipstick method (Figure 1A;
Crystal VC) in which the specimen is tested via dipstick after
incubation for 6–8 hours in alkaline peptone water (APW;
in grams per liter: peptone, 10 g; sodium chloride, 10 g, pH
9.0–9.2). A positive result appears as two pink lines on the
dipstick; the upper line being the control band and the lower
band being the lipopolysaccharide specific to serogroup O1
positive band (Figure 1B and C). APW-enriched samples
that tested positive, as well as a minimum of 10% of negative
clinical samples, were inoculated into Cary-Blair transport
media (Becton Dickinson and Co., Franklin Lakes, NJ) for
storage and transport to the central reference laboratory in
FIGURE 1. Crystal VC kit. (A) 1) The dipstick is provided in an
individualized, humidity controlled package; 2) a clean test tube is
provided in the kit for each dipstick tested; 3) a Pasteur pipette is pro-
vided in the kit for each specimen to be transferred; and 4) a speci-
men processing vial containing enrichment media. (B) 1) A clean
dipstick; 2) a negative dipstick showing clearly the control line only;
3) a Vibrio cholerae O1–positive dipstick showing both the control line
and the O1-positive line. (C) A V. cholerae O1–positive dipstick as
it appears when testing in the kit-provided test tube; note that only
∼200 μL of specimen should be added to the tube as indicated by
the arrow.
the Kousseri health facility for microbiological confirmation.
Using a Pasteur pipette, one to two drops of the enriched
APW specimen were spotted onto a Whatman 903 211 Protein
Saver Card (GE Healthcare Ltd., Forest Farm, Cardiff, UK)
and allowed to air-dry. Filter paper was evaluated for viable
vibrios after overnight drying at room temperature, and all filter
paper specimens tested were culture negative. Filter paper
specimens were stored in individual plastic bags at room
temperature until DNA extraction and PCR processing
were implemented at a later date. The dipsticks and the
preparation of the Protein Saver Cards were carried out
in the individual hospitals by nurses, and the filter paper
samples were then sent to the central laboratory in
Kousseri for preservation. Dipsticks, conical tubes containing
APW-enriched specimens, and test tubes were treated as
biohazardous material and disinfected before disposal. A
detailed training manual for the detection of V. cholerae
O1 was created by several of the collaborating authors
during the study period, available at stopcholera.org.16
Environmental surveillance. Surface water samples were
collected once every 15 days from six sites near each of the
sentinel health facilities (total of 42 sites) to be tested for the
presence of V. cholerae O1 and O139. In a plastic jar, 3 L of
the surface water was collected and then filtered through sterile
gauze. The gauze (∼20 mL) was then combined with
 SIMPLIFIED CHOLERA SURVEILLANCE METHODS IN CAMEROON
equivalent volume (in mL) 2× APW (final concentration of 1×
APW) and incubated for 24 hours (±2 hours). The overnight
enriched solution was tested using the dipstick. All positive
and selected negative environmental samples were inoculated
into Cary-Blair transport media for storage until transported to
the central reference laboratory for microbiological confirma-
tion (Figure 2). Using a Pasteur pipette, one to two drops of
the APW-enriched specimens were preserved on a Whatman
903 Protein Saver Cards and allowed to air-dry. As with the
clinical specimens, filter paper specimens were stored in indi-
vidual plastic bags at room temperature until DNA extraction.
The processing of the environmental samples was carried out
by technicians at the individual hospitals. Enriched specimens
FIGURE 2. Procedure for detecting Vibrio cholerae O1 from envi-
ronmental source using enriched dipstick method. Figure 2 is a pic-
torial of the steps used to collect and assess V. cholerae O1 from
environmental samples. Step 1 is to collect 2–3 L water from envi-
ronmental sampling site. Step 2 shows that one should take medical
gauze and fold it, then roll it into a tube to be placed into the mouth
of the funnel device, ensuring that the gauze fits snug into the mouth
of the funnel so that it will not be displaced when funneling water.
The funnel can be a plastic bottle with the base cutoff or, if avail-
able, a funnel. Once the gauze filter is in place in the mouth of the
funnel, then 2–3 L water is filtered through the funnel. Step 3 is once
the water has all been filtered through the funnel, remove the gauze
filter and place into a 50-mL conical tube containing 20 mL of 2×
alkaline peptone water (APW). The APW is 2× to offset any water
retained in the filter, which would reduce the concentration of the
APW. Step 4 is to incubate the gauze in the 2× APW solution for
∼24 hours at ∼37°C (or between 25°C and 40°C room temperature
[RT] if incubator is not available). Step 5 is to use the Pasteur
pipette included in the Crystal VC dipstick kit and transfer ∼200 μL
of the enriched media to the test tube provided with the kit. Then
insert a new dipstick into the solution and allow to incubate at RT
for 15 minutes. Read the dipstick at 15 minutes to determine if it is
positive or negative (see Figure 1B above). Step 6 is to preserve the
specimen in Cary Blair and on precut individual Whatman filter
paper cards; all positives and a 10% sample of negatives were pre-
served on Cary Blair and filter paper for culture confirmation.
539
were decontaminated following the protocol for biohazardous
material; dipsticks, conical tubes containing APW-enriched
specimens, and test tubes were treated as biohazardous mate-
rial and disinfected before disposal.
Microbiological testing at the central laboratory. All posi-
tive clinical and environmental specimens and selected nega-
tive specimens were sent at routine (biweekly) intervals to the
central laboratory in Kousseri for confirmation. The specimens
were streaked directly onto thiosulfate citrate bile salt sucrose
(TCBS; Becton Dickinson and Co., Franklin Lakes, NJ) agar
and incubated for 24 hours at 37°C. Immediately after inocu-
lating, the first TCBS plate, a pre-labeled APW vial was inocu-
lated with the specimen and incubated for 6 hours at room
temperature. After 6-hour incubation, a second TCBS plate
was inoculated with the enriched specimen and incubated for
24 hours at 37°C. After 24-hour incubation, any cholera-like
colonies were selected with a sterile loop, resuspended in one
to two drops of phosphate-buffered saline (PBS), and tested
via dipstick. All dipstick-positive cultures, as well as any cultures
considered cholera suspect (demonstrating the morphology
of a cholera colony), were preserved in T1N1 agar (1% tryptone
and 1% NaCl) for further testing.
DNA extraction. The dried filter paper samples were sent
to Baltimore, MD, for PCR analysis. DNA extractions were
performed using methods similar to those previously pub-
lished.17 Each dried filter paper specimen was excised using
sterile scissors and placed into a pre-labeled tube. Of sterile
1× PBS, 1 mL was added to each sample tube and incubated
for 10 minutes at room temperature. One mL of sterile 1×
PBS was then added to each sample, and immediately
centrifuged (14,000 × g for 2 min) and the supernatant
discarded. To each sample, 1 mL sterile 1× PBS was added,
and then the sample was immediately centrifuged (14,000 × g
for 2 minutes) and the supernatant was discarded. Subsequently,
150 μL of a 2% (w/v) Chelex-100 solution (Bio-Rad, Hercules,
CA; catalog no. 1422832) followed by 50 μL of sterile
water was added to each sample. The samples were
placed in a heating block at 100°C for 8 minutes and then
centrifuged (14,000 × g for 2 minutes). The supernatant
was transferred to a new microcentrifuge tube and either
stored at −20°C or used in a PCR amplification reaction.
The quantity and quality of extracted DNA was determined
using an ultraviolet (UV) spectrophotometer (NanoDrop
2000, Thermo Scientific, Waltham, MA).
Polymerase chain reaction. As an initial screening step, a
multiplex PCR amplification was performed to determine
the presence of three Vibrio species in the DNA sample by
targeting the toxR genes of Vibrio vulnificus, Vibrio para-
haemolyticus, and V. cholerae. Following the previously
described PCR conditions, the universal forward primer
UtoxF was used in combination with species-specific primers
(Table 1) VvtoxR, VptoxR, and VctoxR to amplify products
of 297, 640, and 435 bp, respectively.18 If V. cholerae was
identified in the sample, a second multiplex PCR was then
performed to differentiate nontoxigenic and toxigenic V.
cholerae. The multiplex was performed as described by
Nandi and others with 15 pmol of primers targeting an outer
membrane protein gene, OmpW, which is a unique gene
conserved in the V. cholerae genome, in combination with
primers targeting cholera toxin A (ctxA) gene (Table 1) at
a concentration of 6.2 pmol.19 The amplified PCR products
for OmpW and ctxA were 588 and 301 bp, respectively.
540 DEBES AND OTHERS

TABLE 1
Primers for PCR assays
Primer name Sequence Amplicon (bp) Reference

UtoxF GASTTTGTTTGGCGYGARCAAGGTT – 18

VptoxR GGTTCAACGATTGCGTCAGAAG 297

18
VctoxR GGTTAGCAACGATGCGTAAG 640
VvtoxR AACGGAACTTAGACTCCGAC 435
19
CtxA-F CTCAGACGGGATTTGTTAGGCACG 302
CtxA-R TCTATCTCTGTAGCCCCTATTACG
19
OmpW-F CACCAAGAAGGTGACTTTATTGTG 588
OmpW-R GAACTTATAACCACCCGCG
20
O1F2-1 GTTTCACTGAACAGATGGG 192
O1R2-2 GGTCATCTGTAAGTACAAC
20
O139F2 AGCCTCTTTATTACGGGTGG 449
O139R2 GTCAAACCCGATCGTAAAGG
21
6968 GC (V6/V8) 5′–AACGCGAACCTTAC–3′ 457
L1401 3′–GCGTGTGTACAAGACCC–5′
PCR = polymerase chain reaction.

Subsequently, all V. cholerae-positive specimens were tested
to determine if they belonged to serogroup O1 or O139,
regardless of their toxigenic nature. This multiplex targets
unique regions in the rfb gene specific for the O1 and O139
serogroups. PCRs were conducted following methods described
by Hoshino and others20 with PCR optimization including an
increase in primer concentration to 2 μM for both O1- and
O139-specific primers (Table 1). On any negative samples,
16S rDNA PCR was performed to confirm DNA preservation
and extraction techniques by applying methods described
by Hasan and others.21 The primers, 6968-GC (V6/V8) and
L1401 (Table 1) selectively amplify the variable regions V6,
V7, and V8 of 16S rDNA genes from most bacteria producing
a 457-bp amplicon. The presence of PCR products was deter-
mined using standard 2% agarose gel electrophoresis con-
taining the intercalating agent ethidium bromide and visualized
under UV light.
Statistical methods. Clinical and environmental samples
collected between August 2013 and October 2014 were used
for these analyses. Descriptive statistics were used to plot the
number of confirmed cholera and diarrhea cases comparing
intensive versus routine surveillance for each study area over
the first year of the study. The environmental surveillance
data included the type of water source (pond, river, ditch,
well, sewage drain, or lake) and the date of collection. The
occurrence of vibrios in the environment was compared over
time by date, facility, and water source. We estimated the sig-
nificance of V. cholerae non-O1 detection over the first year
of the study, taking into account the differing water sources,
follow-up visits, and within-facility clustering of detection.
We used generalized estimating equations with a log link
function and an ar1 correlation matrix to account for cluster-
ing at the facility level. Statistical analyses were conducted
with Stata 13 (StataCorp LP, College Station, TX).22
RESULTS
From August 2013 through October 2014, a total of 1,042
patients were enrolled in the study among all seven study
sites. Of these patients, 619 were enrolled in the study during
intensive surveillance days, 423 were enrolled during routine
surveillance days (Figure 3). Figure 3 demonstrates that most
cases of acute diarrhea presenting both during the intensive
surveillance every 15 days and the routine surveillance were

FIGURE 3. Monthly enrollments of diarrhea and cholera cases. The graph above shows the enrollment of diarrheal cases by month, stratified
by surveillance type and cholera outcome.
 SIMPLIFIED CHOLERA SURVEILLANCE METHODS IN CAMEROON 541

TABLE 3
GEE results: Vibrio non-O1 detection
Factors OR (95% CI) P value

Month (January) Ref –

March 2.5 (1.9, 15.6) 0.001
April 4.8 (1.5, 14.7) 0.007
May 2.6 (0.9, 7.6) 0.091
June 1.9 (0.6, 6.2) 0.307
July 3.9 (1.2, 12.1) 0.021
Water source (river) Ref –
Well 3.0 (1.2, 7.3) 0.016
Sewage drain 10.4 (2.8, 38.1) < 0.001
Lake Chad 0.4 (0.1, 2.5) 0.298
Facility (Kousseri) Ref –
Blangoua 1.2 (0.3, 5.4) 0.837
Darak 9.7 (2.4, 38.3) 0.001
Naga 14.7 (3.8, 57.6) < 0.001
CI = confidence interval; GEE = generalized estimating equation; OR = odds ratio.

FIGURE 4. Percentage of diarrhea cases with severe dehydration

stratified by age group across health facilities. The graph demon-
strates the percentage of severe diarrheal cases enrolled at each
health facility participating in the surveillance efforts. Blangoua had
the highest numbers among all age groups.
not caused by V. cholerae. Though cholera was not common,
severe non-cholera diarrheal disease was prevalent in Blangoua
health facilities (Figure 4). Cholera was not detected in these
health facilities until October 2014 when an outbreak was iden-
tified in Darak, Cameroon. The outbreak continued through
October, with additional cases being detected in nearby
Blangoua. In total, 32 PCR-confirmed cholera cases were
enrolled: 30 patients had O1 V. cholerae and two had non-O1
cholera. Of these, 11 (34%) were children under 5 years of age
and 21 (66%) were males (Table 2).
Of the 1,011 water samples obtained from the environ-
mental sites for the seven health facilities during the study
period, 244 were V. cholerae non-O1 and non-O139 positive,
while none were V. cholerae O1 positive. (After this defined
period of surveillance, in November 2014, one environmental
sample was positive for V. cholerae O1 by dipstick.) The four
types of water sources sampled included rivers, wells, sewage
drains, and Lake Chad. Binomial regression analysis showed
an increased risk of Vibrio in April, May, and July as com-
pared with the risk in January (Table 3). At the facility
level, there was a significantly increased likelihood of Vibrio
detection at sites near Naga and Darak as compared with
Kousseri. Finally, as compared with Vibrio detection in
rivers, there is no significant increase of detection at sites
along Lake Chad. However, a significant increase of Vibrio
detection was seen in wells and sewage drains.
The sensitivity and specificity of the modified protocol for
the Crystal VC dipstick was confirmed by PCR confirmation
of 1,681 specimens; 673 clinical specimens were confirmed
via PCR. Of the 641 clinical specimens that were PCR nega-
tive, 638 were dipstick negative, representing a Crystal VC
dipstick specificity of 99.5% using the enriched methods
(Table 4). All 1,011 environmental samples tested within the
study time frame were both dipstick and PCR negative
(100% specificity). There were an insufficient number of posi-
tive environmental samples to confirm sensitivity. The sen-
sitivity, specificity, positive predictive value, and negative
predictive value of PCR as compared with culture confir-
mation of the 507 clinical specimens confirmed using both
methods was 90.9%, 99.3%, 95.2%, and 98.6%, respectively.
DISCUSSION
The use of simplified laboratory and epidemiological sur-
veillance methodologies in this study enabled rapid identifica-
tion of cases during an outbreak while significantly reducing
any false-positive test results. This study is also the first to
demonstrate the use of the enriched Crystal VC dipstick
method in a remote setting. The results of the dipstick test
were confirmed by a combination of culture and PCR for 673
clinical and 1,011 environmental samples, demonstrating a
specificity of 99.5% and 100%, respectively. This is signifi-
cantly higher than the specificity levels reported for direct
use of the dipstick in stool samples (49–79%).7–9 These data
validate the use of the enriched dipstick method to improve
TABLE 4
PPV and NPV, sensitivity and specificity of enriched Crystal VC dip-
stick as compared with the gold standard of PCR
Clinical samples

Enriched dipstick method PCR positive PCR negative Total

TABLE 2
Clinical surveillance PCR results Dipstick positive 25 3 28
Dipstick negative 7 638 645
No cholera (%) Cholera (%)
Total 32 641 673
Age group Male Female Total Male Female Total Estimate (%) 95% CI
Sensitivity 89.3 71.8–97.7
<5 116 (34) 95 (32) 211 (33) 8 (28) 3 (27) 11 (34) Specificity 98.9 97.8–99.6
5–15 67 (20) 54 (18) 121 (19) 10 (48) 2 (18) 12 (38) PPV 78.1 60.0–90.7
> 15 158 (46) 151 (50) 309 (48) 3 (14) 6 (55) 9 (28) NPV 99.5 98.6–99.9
Total 341 (53) 300 (47) 641 21 (66) 11 (34) 32
CI = confidence interval; NPV = negative predictive value; PCR = polymerase chain
PCR = polymerase chain reaction. reaction; PPV = positive predictive value.
542 DEBES AND OTHERS
specificity as initially demonstrated in the hospital setting in
Bangladesh10,12,13 and Mozambique.11 This is the first study to
demonstrate the use of this enriched method in remote and
rural settings with limited laboratory capacity.
This study presents the novel use of filter paper for environ-
mental and stool sample preservation for molecular screening.
This method proved to be a low-cost and low-maintenance
diagnostic for the field setting, eliminating the need for culture
confirmation and laboratory reagents. Although extensive
training of laboratory personnel was previously required to
conduct microbiological testing, this method enables non-
laboratory personnel to implement the technique in a field
setting. Furthermore, the filter paper does not need cold chain
for preservation and is considered non-biohazardous after
drying, thereby eliminating transportation issues.
This study is the first to investigate a simplified epidemio-
logical and laboratory surveillance methodology for use in
remote and rural settings and to successfully implement a
sustained clinical and environmental surveillance of cholera in
the FNC, particularly during a period of insecurity. Although
the surveillance approach was modeled after a similar meth-
odology used in Bangladesh,4 the methodology applied in
Cameroon was further simplified by the use of the enriched
dipstick test, the use of a low-cost gauze filtration device
for environmental sampling, and the preservation of dried,
enriched specimens on filter paper for PCR confirmation of
initial results. In 2014, after more than 1 year of surveillance,
the first cholera cases were identified during study surveil-
lance activities in Darak and Blangoua health districts, and
the cases were rapidly confirmed using the study-specific sim-
plified laboratory methodologies. This enabled rapid response
and interventions to limit the spread of the disease. Thus, not
only did these findings contribute to the assessment of a sur-
veillance strategy appropriate for remote and vulnerable set-
tings, but it also demonstrated the capability of this method
for early outbreak detection.
The surveillance results for the first year of this study did
not confirm our initial hypothesis that there is an environmen-
tal reservoir for cholera in Lake Chad that leads to occasional
cases. Nor did the regression analysis reveal a significant sea-
sonal trend in Vibrio presence in the environmental sites.
However, this report is based only on the first year of surveil-
lance. Given that outbreaks in the study area were only in
the beginning stages at the conclusion of this first report, con-
tinued surveillance over a longer period is needed.
There are several limitations to this study, the most impor-
tant being that the study site is situated in an area that con-
tinues to struggle with safety and security issues as a result
of terrorist activities. Safety issues resulted in the loss of data
in some study areas, particularly in Darak, which is located
on an island in Lake Chad. The presence of nontoxigenic
V. cholerae O1 in the environmental sources during the early
months of 2014 in Darak may have been an early warning
sign for the area. Unfortunately, the team was unable to
maintain regular surveillance, leading to an incomplete anal-
ysis of events before the toxigenic V. cholerae O1 outbreak
in the fall of 2014. The conventional culture techniques were
conducted on 507 samples in this study with only one culture
positive/PCR negative sample. This false-negative PCR spec-
imen, as well as many other samples, was collected by a
single health worker who received only one abbreviated day
of training amid the outbreak. The previously trained health-
care workers at the site had fled the area before the out-
break because of mounting insecurity in the area. Although
culture was completed at the central reference laboratory,
the initial dipstick testing and filter paper preservation were
completed at the site on case presentation. This highlights
the simplicity of the techniques and ease of implementation
in even the most challenging setting. There were no clinical
cases identified in our study area before October 2014. This
insufficient amount of positive results prevented us from
establishing an accurate sensitivity calculation for the enriched
dipstick methodology. Finally, only V. cholerae was confirmed
clinically and, therefore, we were unable to further character-
ize the cases of non-cholera diarrhea enrolled in the study.
In conclusion, the first year of this surveillance study dem-
onstrates the successful use and implementation of low-cost,
simplified epidemiological and laboratory methodologies for
surveillance in remote, rural, or vulnerable settings. The use
of an enriched dipstick protocol was successfully applied in a
field setting with improved specificity. Basic technologies, such
as the use of gauze filtration rather than more expensive
filtration methods, decrease supply and logistics costs and
allow for important environmental surveillance to provide
information about the burden of cholera in previously unde-
scribed areas. Finally, the novel application of dried filter paper
methodology to cholera DNA preservation demonstrates a
simplified method for assessment for Vibrio presence in stool
and environment.
Received July 3, 2015. Accepted for publication November 14, 2015.
Published online January 11, 2016.
Acknowledgments: We thank the staff of the Kousseri Health District,
Mada Health District, Makary Health District, and Goulfey Health
District for their participation in this study. We also thank the Far
North regional delegate for public health, Rebecca Djao, and The
Division of Health Operations Research, Cameroon Ministry of
Public Health, for their support in this project.
Financial support: This study was supported by the Delivering Oral
Vaccine Effectively (DOVE) project. DOVE is supported by the Bill
& Melinda Gates Foundation and administered through the Johns
Hopkins Bloomberg School of Public Health.
Disclaimer: The funder had no role in study design, data collection
and analysis, decision to publish, or preparation of the manuscript.
Authors’ addresses: Amanda K. Debes, Peter Steinwald, Malathi Ram,
and David A. Sack, Department of International Health, Johns Hop-
kins Bloomberg School of Public Health, Baltimore, MD, E-mails:
adebes1@jhu.edu, psteinw1@jhu.edu, mram1@jhu.edu, and dsack1
@jhu.edu. Jerome Ateudjieu, Department of Biomedical Sciences,
Faculty of Sciences, University of Dschang, Dschang, Cameroon,
E-mail: jateudj@yahoo.fr. Etienne Guenou, Walter Ebile, Isaac
Tadzong Sonkoua, and Antony Chebe Njimbia, Meilleure Acces aux
Soins de Santé (M.A. SANTE), Yaounde, Cameroon, E-mails:
etienneg83@yahoo.fr, walter.ebile@masante-cm.org, isaacsonkoua@
yahoo.fr, and chebeanthony@gmail.com.
This is an open-access article distributed under the terms of the
Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the
original author and source are credited.
REFERENCES
1. Griffith DC, Kelly-Hope LA, Miller MA, 2006. Review of
reported cholera outbreaks worldwide, 1995–2005. Am J Trop
Med Hyg 75: 973–977.
2. World Health Organization, 2014. Cholera, 2013. Wkly Epidemiol
Rec 89: 345–355.
 SIMPLIFIED CHOLERA SURVEILLANCE METHODS IN CAMEROON
3. Hampton LM, Zell ER, Schrag S, Cohen AL, 2012. Sentinel versus
population-based surveillance of pneumococcal conjugate vac-
cine effectiveness. Bull World Health Organ 90: 568–577.
4. Sack RB, Siddique AK, Longini IM Jr, Nizam A, Yunus M,
Islam MS, Morris JG Jr, Ali A, Huq A, Nair GB, Qadri F,
Faruque SM, Sack DA, Colwell RR, 2003. A 4-year study of
the epidemiology of Vibrio cholerae in four rural areas of
Bangladesh. J Infect Dis 187: 96–101.
5. Deen JL, von Seidlein L, Sur D, Agtini M, Lucas ME, Lopez
AL, Kim DR, Ali M, Clemens JD, 2008. The high burden of
cholera in children: comparison of incidence from endemic
areas in Asia and Africa. PLoS Negl Trop Dis 2: e173.
6. World Health Organization, 1993. Guidelines for Cholera Control.
Geneva, Switzerland: WHO.
7. Harris JR, Cavallaro EC, de Nobrega AA, Dos SBJC, Bopp C,
Parsons MB, Djalo D, Fonseca FG, Ba U, Semedo A, Sobel
J, Mintz ED, 2009. Field evaluation of Crystal VC® Rapid
Dipstick test for cholera during a cholera outbreak in Guinea-
Bissau. Trop Med Int Health 14: 1117–1121.
8. Ley B, Khatib AM, Thriemer K, von Seidlein L, Deen J,
Mukhopadyay A, Chang NY, Hashim R, Schmied W, Busch CJ,
Reyburn R, Wierzba T, Clemens JD, Wilfing H, Enwere G,
Aguado T, Jiddawi MS, Sack D, Ali SM, 2012. Evaluation of a
rapid dipstick (Crystal VC) for the diagnosis of cholera in Zanzibar
and a comparison with previous studies. PLoS One 7: e36930.
9. Mukherjee P, Ghosh S, Ramamurthy T, Bhattacharya MK,
Nandy RK, Takeda Y, Nair GB, Mukhopadhyay AK, 2010.
Evaluation of a rapid immunochromatographic dipstick kit for
diagnosis of cholera emphasizes its outbreak utility. Jpn J
Infect Dis 63: 234–238.
10. George CM, Rashid MU, Sack DA, Sack RB, Saif-Ur-Rahman
KM, Azman AS, Monira S, Bhuyian SI, Zillur Rahman KM,
Toslim Mahmud M, Mustafiz M, Alam M, 2014. Evaluation of
enrichment method for the detection of Vibrio cholerae O1
using a rapid dipstick test in Bangladesh. Trop Med Int Health
19: 301–307.
11. Wang XY, Ansaruzzaman M, Vaz R, Mondlane C, Lucas ME,
von Seidlein L, Deen JL, Ampuero S, Puri M, Park T, Nair
GB, Clemens JD, Chaignat CL, Rajerison M, Nato F, Fournier
JM, 2006. Field evaluation of a rapid immunochromatographic
dipstick test for the diagnosis of cholera in a high-risk popula-
tion. BMC Infect Dis 6: 17.
12. Bhuiyan NA, Qadri F, Faruque AS, Malek MA, Salam MA,
Nato F, Fournier JM, Chanteau S, Sack DA, Balakrish Nair
543
G, 2003. Use of dipsticks for rapid diagnosis of cholera caused
by Vibrio cholerae O1 and O139 from rectal swabs. J Clin
Microbiol 41: 3939–3941.
13. Chakraborty S, Alam M, Scobie HM, Sack DA, 2013. Adapta-
tion of a simple dipstick test for detection of Vibrio cholerae
O1 and O139 in environmental water. Front Microbiol 4: 320.
14. Ako AA, Shimada J, Eyong GE, Fantong WY, 2010. Access to
potable water and sanitation in Cameroon within the context
of Millennium Development Goals (MDGS). Water Sci Technol
61: 1317–1339.
15. International Federation of Red Cross and Red Crescent Societies
I, 2011. Cameroon: Cholera in Northern Cameroon. Available at:
http://reliefweb.int/report/cameroon/cholera-northern-cameroon-
dref-operation-final-report-mdrcm009. Accessed December
15, 2015
16. Debes A, Chakraborty S, Sack DA, 2015. Manual for detecting
Vibrio cholerae O1 from fecal samples using an enriched dip-
stick assay—a low-cost, simplified method of confirming cholera.
Available at: https://www.stopcholera.org/sites/cholera/files/stop_
cholera_fact_sheet_1_5_revised.pdf. Accessed October 20, 2015.
17. Kain KC, Lanar DE, 1991. Determination of genetic variation
within Plasmodium falciparum by using enzymatically ampli-
fied DNA from filter paper disks impregnated with whole
blood. J Clin Microbiol 29: 1171–1174.
18. Bauer A, Rorvik LM, 2007. A novel multiplex PCR for the iden-
tification of Vibrio parahaemolyticus, Vibrio cholerae and Vibrio
vulnificus. Lett Appl Microbiol 45: 371–375.
19. Nandi B, Nandy RK, Mukhopadhyay S, Nair GB, Shimada T,
Ghose AC, 2000. Rapid method for species-specific identifica-
tion of Vibrio cholerae using primers targeted to the gene of
outer membrane protein OmpW. J Clin Microbiol 38:
4145–4151.
20. Hoshino K, Yamasaki S, Mukhopadhyay AK, Chakraborty S, Basu
A, Bhattacharya SK, Nair GB, Shimada T, Takeda Y, 1998.
Development and evaluation of a multiplex PCR assay for rapid
detection of toxigenic Vibrio cholerae O1 and O139. FEMS
Immunol Med Microbiol 20: 201–207.
21. Hasan NA, Chowdhury WB, Rahim N, Sultana M, Shabnam SA,
Mai V, Ali A, Morris GJ, Sack RB, Huq A, Colwell RR, Endtz
HP, Cravioto A, Alam M, 2010. Metagenomic 16S rDNA
targeted PCR-DGGE in determining bacterial diversity in
aquatic ecosystem. Bangladesh J Microbiol 27: 46–50.
22. StataCorp, 2014. STATA Data Analysis and Statistical Software.
Available at: http://www.stata.com/.
 Vaccine 38 (2020) A167–A174

Contents lists available at ScienceDirect

Vaccine
journal homepage: www.elsevier.com/locate/vaccine

Non-vaccine strategies for cholera prevention and control: India’s

preparedness for the global roadmap
Madhuchhanda Das a, Harpreet Singh a, C.P. Girish Kumar b, Denny John c, Samiran Panda d,
Sanjay M. Mehendale a,⇑
a
Indian Council of Medical Research, New Delhi, India
b
ICMR-National Institute of Epidemiology, Chennai, India
c
Campbell Collaboration, New Delhi, India
d
ICMR-National AIDS Research Institute, Pune, India

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

Article history: Background: Recently World Health Organization’s Global Task Force on Cholera Control (GTFCC) has
Available online 20 August 2019 published a global roadmap for prevention and control of cholera. We review preparedness of existing
governmental non-vaccine programs and strategies for cholera prevention and control in India. We also
Keywords: describe strengths and gaps in the context of implementation of the global roadmap.
Cholera Methods: We reviewed published literature on non-vaccine based strategies for prevention and control of
Prevention cholera in India and analyzed strengths and weaknesses of Government of India’s major anti-cholera and
Control
ante-diarrhea initiatives under Integrated Disease Surveillance Program (IDSP), National Rural Health
WASH
Disease model
Mission (NRHM), and other disease surveillance platforms.
Results: The first strategy of the WHO global roadmap, namely, preparedness for early detection and out-
break containment, has been addressed by the IDSP. NRHM complements IDSP activities by focusing on
sanitation, hygiene, nutrition, and safe drinking water. We identified the need to adopt stricter case def-
initions and data validation protocols.
Multi-sectoral approach to prevent cholera occurrences and re-occurrences [the second suggested
strategy in the global roadmap], highlights identification of hotspots and implementing strategies based
on transmission dynamics. We recommend development of comprehensive models by integrating data
sources beyond the national programs to eliminate cholera hotspots in India.
Implementing the third proposed strategy in the global roadmap, coordinated technical support,
resource mobilization, and partnerships at local and global levels, has major challenges in India due to
structural issues related to health systems and health programs.
Conclusion: Even with a robust public health infrastructure, absence of a national cholera program might
have resulted in lack of specific focus and concerted efforts for cholera prevention and control in India. A
National Taskforce for Cholera Control must develop India-specific ‘National Cholera Prevention and
Response Road Map’ with an appropriate administrative and financially viable framework for its
implementation.
Ó 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://
creativecommons.org/licenses/by/4.0/).

1. Introduction been reported from 21 states and Union territories of which 12

Cholera has remained a public health problem in India for sev-
eral decades. However, there is no reliable nationwide data on its
prevalence [1]. Central Bureau of Health Intelligence (CBHI), and
Integrated Disease Surveillance Program (IDSP) of Government of
India publish reports on cholera incidence regularly. Cholera has
⇑ Corresponding author at: Indian Council of Medical Research, Post Box No.
4911, Ansari Nagar, New Delhi 110029, India.
E-mail address: sanjaymehendale.hq@icmr.gov.in (S.M. Mehendale).
states are reported to be endemic for Cholera [2]. A study esti-
mated an annual incidence of 675,188 cholera cases and 20,356
deaths (2008–2012) in India [3]. Between 2014 and 2016, 197 cho-
lera outbreaks were reported in the country [3]. ‘National Health
Profile’ published by CBHI reported total 718 cholera cases and 3
deaths in 2016 while 913 cholera cases and 04 deaths were
reported in the previous year [4,5]. Provisional data for 2017
reported 494 cholera cases and 3 deaths [4]. In 2016, IDSP reported
114 cholera outbreaks in the country [6].

https://doi.org/10.1016/j.vaccine.2019.08.010
0264-410X/Ó 2019 The Authors. Published by Elsevier Ltd.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
A168 M. Das et al. / Vaccine 38 (2020) A167–A174
Hence, there appears to be considerable discordance between
the cholera burden reported in published literature and data avail-
able with governmental systems and programs.
A recent watershed publication, ‘‘Ending Cholera—A Global
Roadmap to 2030”, by the Global Task Force on Cholera Control
(GTFCC) of World Health Organization (WHO) describes a vision
to reduce cholera-associated mortality by 90 percent by 2030 [7].
It highlights three strategies: (i) Early detection and quick response
to contain outbreaks, (ii) Targeted multi-sectoral approach to pre-
vent cholera recurrence in hotspots, and (iii) Coordinated technical
support, resource mobilization, and partnership at local and global
levels [7].
Prevention and treatment are the two critical components of
cholera control in any setting. Prevention encompasses use of Oral
Cholera Vaccines (OCVs) as well as non-vaccine interventions
focusing on Water, Sanitation and Hygiene (WASH). Both are com-
plementary and essential. However, despite in-country licensure,
OCVs are not yet adopted as part of the national immunization pro-
gram by the Government of India and strategies such as WASH,
early diagnosis, oral rehydration solution (ORS), and awareness
generation are emphasized. We present situation analysis in terms
of preparedness for prevention and control of cholera in India
focusing on non-vaccine strategies in the context of the GTFCC
roadmap. We have specifically examined the prevailing activities,
programs and the available infrastructure with a view to identify
future needs for cholera prevention and control in India.
2. Methods
The authors extensively deliberated and developed a matrix
defining the primary and secondary domains of prevention and
control of cholera in India. Electronic databases like NCBI-
PubMed, Cochrane, and CINAHL were searched using the keywords
‘‘cholera” and ‘‘prevention OR control” to identify all the published
literature including qualitative studies in India. We analyzed the
literature on non-vaccine prevention and control strategies for
cholera in the last 10 years. We also reviewed the strengths and
weaknesses of the major Government of India initiatives such as
IDSP, diarrhea control measures of the National Rural Health Mis-
sion (NRHM) and other disease surveillance platforms. A schematic
framework was developed to assess the level of preparedness for
cholera prevention and control in India within the bounds of
GTFCC roadmap.
3. Results and discussion
We developed a radial diagram showing WHO roadmap strate-
gies for control of cholera and then mapped our perceptions on
preparedness for cholera prevention and control in India in case
of various strategies, schemes and approaches as well as available
infrastructure to implement them. While doing this we have
retained alignment with the GTFCC for cholera prevention and con-
trol [Fig. 1a & b]. The colour codes and density reflect our percep-
tion and assessment of India’s level of preparedness after the
critical analysis. This schematic framework describes the existing
scenario and identifies future needs for cholera prevention and
control in India.
3.1. Early detection and outbreak containment
A robust outbreak control system could significantly reduce
mortality from cholera, particularly in countries with seasonal
recurrence of the disease [7]. Hence, the WHO roadmap lays
emphasis on containing outbreaks as soon as they are notified. Var-
ious components of an effective response include early case detec-
tion, quick response for containment, immediate community
engagement, early warning surveillance, readiness for supplies,
nation-wide network of laboratories and responsive health
systems.
3.1.1. Integrated surveillance system for early warning in India
An efficient surveillance system is critical in containment of any
diseases and for prioritization of areas for intervention. The
‘National Centre for Disease Control’ and NRHM functioning under
the ‘Ministry of Health and Family Welfare’ have created a mecha-
nism for early detection and early response to contain outbreaks in
India. The IDSP system connects villages to state-wide monitoring
systems through surveillance units. The Central Surveillance Unit
(CSU) is linked to the State Surveillance Units (SSU), and each
SSU is then connected to the District Surveillance Units (DSU). Each
DSU is linked to Peripheral Reporting Units (PRU) at the village
level. At present, 675 DSUs are operational across the country.
The IDSP network provides supervision and feedback from the
CSU to PRUs and ensures seamless data flow from PRUs to the
CSU for compilation and analysis.
The country-wide IDSP surveillance network has the potential
to detect emerging and re-emerging infectious diseases including
cholera in a timely manner. This network can also respond quickly
in post-disaster scenarios such as earthquakes, floods, and cyclones
and can initiate disease and outbreak surveillance immediately.
Information and communication technology has been made an
integral component of IDSP to enable networking, training, data
transfer, analysis and communication at districts, states and cen-
tral levels [8].
Since 2008, the IDSP portal has created a 24X7 call center facil-
ity to receive disease alerts from anywhere in the country on a toll-
free number. This facility is for reporting and verification of out-
break alerts, mitigating rumors, and initiating appropriate public
health actions.
The Indian Space Research Organization (ISRO) has set up a
satellite-based interactive network of 367 sites for training using
Education Satellite (EDUSAT) connectivity and presently covers
states like Maharashtra, Gujarat, Tamil Nadu, the North-Eastern
States, hilly states of Himachal Pradesh, Uttarakhand and Jammu
& Kashmir, and Islands [8]. Emulating ISRO, the ‘National Informat-
ics Centre’ has set up broadband-based connectivity at its 378 data
centers. These facilities can also be leveraged for efficient data cap-
ture and transmission to the IDSP CSU as alternative or additional
mechanisms. However, the call centers of IDSP and EDUSAT have
repeatedly experienced administrative and operational challenges
associated with the use of high-end technology. Their performance
has to become more dependable and consistent.
On a trial basis, less technology-intensive mechanism of short
messaging service (SMS) to capture data directly from the field
was made operational in 2008. However, it was observed that data
from the field was first being reviewed at the PHC level and filtered
data was being forwarded through the SMS. Although this was
done primarily to sanitize the content of information transmitted,
it was defeating the whole purpose of the instant SMS-alert system
to facilitate immediate response. The SMS-based data capture sys-
tem can be revitalized for wider, unfiltered data collection, rapid
confirmation and efficient linking with the IDSP system by imple-
menting validation and data authentication protocols [8].
3.1.2. Laboratory setup for detection of cholera in the Indian public
health system
Peripheral laboratories should rapidly confirm the diagnosis of
cholera and perform culture/ antibiotic susceptibility for confirmed
cholera cases. In the limited resource settings deploying trained
personnel for microbiological diagnosis of cholera in peripheral
health facilities is difficult. In such scenarios, rapid diagnostic tests
 M. Das et al. / Vaccine 38 (2020) A167–A174
Fig. 1. [A] Pictorial representation of the WHO Global Roadmap for Cholera Prevention and Control Strategies; [B] Pictorial representation of preparedness of India for
implementation of strategies for prevention and control of Cholera.
(RDTs) can be a cheaper and a practical screening option in periph-
eral laboratories like primary health centers. Although not replace-
ments for stool culture or PCR, RDTs can be employed for initial
case detection and early warning of an impending cholera out-
break. Further case confirmation using classical microbiological
procedures or PCR can be done at secondary/ tertiary health
facilities.
Under IDSP, 117 District Public Health Laboratories (DPHL)
headed by trained microbiologists have been established in 29
states of India for providing diagnostic support during epidemic
situations. Additionally, 107 state-level referral laboratories (SRL)
have been established at select medical colleges/institutions in
24 states. Unlike DPHLs, SRLs have the capacity to perform tests
for culture, sensitivity and serotyping of cholera and other
entero-pathogens.
The CSU receives information on lab-confirmed outbreaks from
SSUs through e-mail, via e-portals, or by fax every week [8]. The
data generated by laboratory-network is used for detecting and
reporting cholera outbreaks in India. Between 1997 and 2006, a
total of 68 cholera outbreaks occurred in 18 states and union terri-
tories [3]. During 2004–2005, cholera was reported in 15 Indian
states, which included 7 outbreaks [9]. Between 2003 and 2012,
37,037 cholera cases were reported by the CBHI. During 2010–
2012, 185 cholera outbreaks were reported to the IDSP. Significant
improvements in the reporting of cases and outbreaks is an impor-
tant public health achievement in the country [9]. A more recent
analysis of IDSP data for the period 2013–2015 documented 179
outbreaks from different parts of India [10].
To summarize, a strong laboratory network of public health lab-
oratories has been created in India that sends reports to the CSU.
However, nearly 75% of the Indian districts are yet to be covered
under DPHL. There is a gross mismatch in the number of cases
reported by the IDSP network and DPHLs, as well as SRLs. Triangu-
A169
Strongly prepared
Moderately prepared
Less prepared
Not discussed
lation of data from these sources for more effective mapping of
cholera in India is a major challenge. Initial cases of suspected cho-
lera are more likely to first seek medical attention at primary
health care facilities and hence they should be equipped with cho-
lera RDTs.
3.1.3. Rapid-response teams
Rapid response teams (RRTs) constitute a vital arm of IDSP as
part of preparedness for outbreaks and early response. A team of
an epidemiologist, microbiologist or laboratory specialist, and a
clinician is trained and deployed to investigate and respond to an
outbreak and report back to the national authorities immediately.
The RRTs are expected to initiate swift action to curtail further
spread of illness and prevent deaths by initiating emergency
response directed at WASH services, investigating risk sources
and working with affected communities to identify the urgent
interventions needed [8]. All diarrheal outbreaks reported to RRTs
are responded to. An analysis of acute diarrhoeal outbreaks in an
Indian state revealed a median time of 2.5 days to report to district
RRT (inter quartile range, 2–5 days) [11]. Delays in investigating
suspected cholera outbreak and delayed outbreak response have
been linked to high levels of morbidity and mortality.
Thus, the infrastructure and system for outbreak detection and
containment are in place in India. Additional work is required on
creating a critical mass of adequately trained skilled manpower
for RRT and their quick mobilization; improving IEC materials
and resources like drugs, vehicles, reagents etc.; and making the
technology more dependable for quick reporting and seamless
communication.
3.1.4. Preparedness and implementation of WASH
For many decades, developed countries in Europe and North
America have eliminated cholera by providing safe drinking water
A170 M. Das et al. / Vaccine 38 (2020) A167–A174
and advanced sanitation systems. In 2015, improved sanitation
facilities were available for 44% of the population in India (65%
urban and 34% rural population) [12]. Improved drinking-water
sources were available for 88% of the Indian population. Despite
this the 2017 UN-Water Global Analysis and Assessment of Sanita-
tion and Drinking-Water report documented 71.7/100,000 diar-
rheal deaths in under 5 years due to inadequate WASH facilities;
a significant challenge for India to attain sustainable development
goals [13]. The allotment of the equivalent of USD 3.554 billion in
the 2017 budget outlay of the Government of India towards WASH
may not be adequate. Despite laws and policies on (i) urban/ rural
sanitation and drinking water supply, (ii) hygiene promotion, and
(iii) water resources planning and management at the national
level; India continues to have a significant burden of diarrheal dis-
eases [2,3]. The reasons for this need to be explored and well laid-
out schemes and outcome-oriented initiatives should be sup-
ported. Funding for public health emergencies, public health prac-
tices, and community-level behavioral change needs to be backed
by appropriate research evidence.
3.1.5. WASH - safe water supply
The ‘Ministry of Drinking Water and Sanitation’ oversees the
implementation of the ‘National Rural Drinking Water Program’
in the country. Rural drinking water has been included by the
Government of India in rural infrastructure development plan
called ‘‘Bharat Nirman”, which aims at providing adequate safe
drinking water of good quality to all so far uncovered habitations.
In 2014, the Prime Minister of India launched the ‘‘Swachh Bharat
Abhiyan”, a program to create awareness about sanitation by erad-
icating open defecation, preventing water contamination, and
improving public health. Swachh Bharat Mission portal suggests
that 27 out of India’s 36 states and Union territories are now open
-defecation- free with 98.6% of Indian households having access to
toilets [14]. The National Annual Rural Sanitation Survey of 2018–
19 has reported that the number of Indians defecating in the open
has reduced to less than 50 million from 550 million in 2014 [15].
However, all these schemes have operational challenges and
ground-level implementation problems and their benefits will be
realized in the years to come. Discussion on the issues of sanitation
and open defecation on open platforms is a significant
development.
3.1.6. Monitoring of water quality
‘‘Prevention and Control of Water Pollution Act” was enacted to
restore and maintain water bodies in India in 1974. The ‘Central
Pollution Control Board’ has established a network of monitoring
stations on rivers across the country and initiated monitoring of
water quality in 1977–78 under the Global Environmental Moni-
toring System (GEMS). The present network has 2500 monitoring
stations in 28 states and 6 Union Territories. It covers 445 rivers,
154 lakes, 12 tanks, 78 ponds, 41 creeks or seawater inlets, 25
canals, 45 drains, 10 water treatment plants (raw water), and
807 wells [13]. Monitoring of Indian National Aquatic Resources
System (MINARS), and the Yamuna Action Plan (YAP) are addi-
tional complementary networks and GEMS, MINARS and YAP
jointly constitute the inland water-quality monitoring network.
The groundwater quality monitoring network has been extended
to 807 locations in the country [16].
Water samples are analyzed for 9 core parameters, 19 general
parameters, as well as trace metals at selected locations. Appropri-
ate corrective measures are suggested for restoring the water qual-
ity of the polluted water bodies. Major water quality concerns
include ‘Total Coliform’ and ‘Fecal Coliform’ counts, high Biochem-
ical/ Biological Oxygen Demand (BOD) and salinity resulting from
organic matter and chemical pollution. Monitoring in 2011 indi-
cated that organic pollution was a predominant cause of pollution
of aquatic resources. It was observed that nearly 63%, 19%, and 18%
of water bodies had BODs of less than 3 mg/l, 3–6 mg/l, and above
6 mg/l respectively. Drinking water sources need to be protected
from organic matter and fecal contamination through very system-
atic efforts in India.
3.1.7. Community engagement for behavioral changes and hygiene
practices
NRHM of India has implemented ‘Total Sanitation Campaign’ in
350 districts of the country and has proposed to extend coverage to
the remaining districts [17]. It attempts to comprehensively deal
with sanitation and hygiene, nutrition, and safe drinking water
through its District Plan for Health. The Village Health and Sanita-
tion Committees (VHSNC), consisting of opinion leaders at the vil-
lage level, have been set up. ‘Accredited social health activists’ from
villages mobilize the community and facilitate people’s access to
health and health-related services by working closely with
VHSNCs. The activities of VHSNCs with reference to WASH include
(i) Monitoring and facilitating access to essential public services
such as clean drinking water and clean toilets, and (ii) Organizing
local collective action for health promotion through voluntarism,
community mobilization and sensitization against poor environ-
mental hygiene practices.
The Government of India has focused on creating awareness
about sanitation, safe water, and prevention of open defecation
but in order to radically improve environmental hygiene it is
important that practices and behavior are adopted at individual,
family, community, and national levels.
3.1.8. Preparedness of health care system and training of health
workers
Strengthening of the existing health care facilities, establish-
ment of dedicated facilities such as ‘Cholera Treatment Centers’/
‘Cholera Treatment Units’ as also the training of health workers
are critical elements in prevention and control of cholera. It is also
important to have trained and dedicated staff for outbreak investi-
gation and containment.
Though India has a robust public health infrastructure, training
of the healthcare staff in urban, rural and tribal areas in case detec-
tion and dehydration management is essential to reduce cholera
associated mortality. Ability to quickly mobilize RRTs and to man-
age uniform year-round availability of the adequate stock of drugs
and fluids at the most peripheral level are likely to significantly
reduce cholera-related morbidity and mortality.
3.1.9. Prepositioning stocks
Improvement of efficiency of a rapid response for cholera out-
breaks containment through pre-positioning of resources for diag-
nostics, patient care, and emergency WASH intervention is highly
recommended. Maintenance of sufficient stocks of WASH supplies,
namely, rapid microbial test kits, chlorine tests, water disinfection
technologies including chlorine, water tanks, supplies like hygiene
kits, ORS, gloves, culture media, soap, rapid diagnostic kits, chlo-
rine tablets, bleaching powder, etc. at healthcare facilities at all
levels is required for effective control of cholera outbreaks.
3.2. Multi-sectoral approach to prevent cholera occurrence and re-
occurrence
Outbreaks of cholera are often observed in vulnerable popula-
tions following disasters, conflicts and famines [18,19]. The Global
Roadmap emphasizes identification of hotspots (geographical
regions or communities heavily affected by cholera), identification
and mapping of vulnerable populations, and studying transmission
dynamics [7]. These activities require multi-sectoral coordination
 M. Das et al. / Vaccine 38 (2020) A167–A174
and have been employed successfully for the prevention and con-
trol of cholera world-wide [20,21].
3.2.1. Identification of hotspots
‘Hotspots’ are the areas of increased transmission potential.
Their identification plays an important role in research, policy for-
mation, and public health practice [22]. Appropriate interventions
and preparedness at the identified hotspots is the responsibility of
the government. The latest technological tools like geographic
information system, global positioning system and Google Earth
application programming interface along with mobile phones
[23], have been globally employed for efficient identification of
hotspots.
There are very few studies about cholera hotspot identification
in India. Based on spatial clustering reports of district level cholera
cases during 2010–2015 obtained from the IDSP, 13 out of 36
states in India were classified as endemic for cholera and 78 out
of the 641 districts in 15 states were identified as ‘‘hotspots”
[10]. On the other hand, 111 districts in 9 states were identified
as ‘‘hotspots” from a model-based prediction. The risk for cholera
in a district was negatively associated with the proportion of liter-
ate people, households using treated water sources and ownership
of mobile telephones. Conversely, areas with poor sanitation and
drainage conditions and lower levels of urbanization were associ-
ated with a higher cholera risk [10]. Surveillance data on cholera
outbreaks from 2000 to 2011 from the health department of the
Chennai Corporation and population data (2001) from census were
used to identify cholera hotspots in the metropolis [24]. There are
also many reports of isolated cholera outbreaks beyond the hot-
spots in India [25–27]. A map depicting cholera hotspots in India
has been published recently [1].
3.2.2. Risk and vulnerability assessment
Vulnerability for cholera is dependent on environmental, phe-
notypic and genotypic factors [28]. Several studies have identified
poor environmental conditions such as unclean water, unhygienic
environment, and poor waste management [29,30] to be strongly
associated with outbreaks and endemicity of cholera. Reported
phenotypic factors defining vulnerability include malnutrition,
homelessness, poor housing, and destitution; while nucleotide
polymorphisms and epigenetic modifications are its genetic deter-
minants [31].
Though there have been reports of cholera outbreaks in various
parts of India [9], group-specific risk and individual vulnerabilities
have not been adequately studied. Similar to studies of host geno-
types on predicting susceptibility to tuberculosis [32], studies can
be done in the context of cholera as well.
3.2.3. Modelling and transmission dynamics
Modelling and transmission dynamics are being used exten-
sively for understanding the pathogenesis and the spread of com-
municable diseases [33] and also for evaluating risk factors and
the impact of control measures [34,35]. Models vary from simple
deterministic equations to complex stochastic frameworks with
input parameters for immunology, clonality and the population
structure of cholera cases. Transmission dynamics models have
been developed for cholera world-wide for understanding the role
of climate on transmissibility of the disease [36–39]. Majority of
the studies have used susceptible-infected-recovered (SIR) com-
partmental models with different parameters. A SIR compartmen-
tal model with parameters to estimate the impact of clean water,
vaccination and enhanced antibiotic distribution programs has
been reported [40]. A SIR model parameterized with high asymp-
tomatic ratio and rapid waning immunity has been used for
explaining 50 years of mortality data from 26 districts in West
Bengal [41]. The hyper-infectivity of bacterial strains has been pro-
A171
posed as a potential parameter for the impact of interventions on
the endemic cholera spread [42]. Recently, the weekly incidence
of suspected cases and fatality risk was used to parameterize the
family of logistic curves for describing the unbiased incidence in
the cholera outbreak in Yemen in 2017. Using logistics and gener-
alized logistics models, the cumulative incidence at the end of the
epidemic was estimated to be 790,778 (95% CI: 700,495, 914,442)
cases and 767,029 (95% CI: 690,877, 871,671) cases respectively
[38]. Mathematical models of cholera transmission have been pro-
posed with approaches for developing more detailed cholera out-
break models, including the addition of contaminated water
supplies, spatial effects, intra-household transmission, and inter-
ventions [43,44].
Models of cholera transmission need to be developed in India
and validated. Data from IDSP, state governments, and other data
sources needs to be integrated to develop a realistic predictive
model. Considering India’s population heterogeneity and strain
diversity, it is possible to develop a comprehensive SIR stochastic
framework model parameterized for several factors depicted in
Fig. 2. Though effective, some of the major challenges in developing
integrated models for cholera include limited capacity and inade-
quate trained manpower and non-uniform and non-validated data.
3.3. Coordinated technical support, resource mobilization and
partnership at local and global levels
The GTFCC created in 1992, has striven to be an effective and
well-coordinated platform for bringing together all multi-sector
technical partners from around the world to support countries in
their fight against cholera. The goal of the GTFCC is to support
national and inter-country capacities by providing a strong plat-
form for advocacy and communications, fund-raising, inter-
sectoral coordination, and technical assistance towards ending
cholera as a public health issue by 2030. Despite considerable
efforts, challenges such as a lack of acceptability of OCVs by policy/
decision-makers and integration with non-vaccine strategies such
as WASH need to be overcome. The Global Roadmap is an effort
to cover the last mile through support and cooperation from inter-
national and national-level stakeholders towards accelerating
action against cholera.
The comprehensive cholera control envisaged in the global
roadmap document requires significant financial investment by
the adopting countries. However, there is no mention of financial
assurance from the GTCC or WHO. Moreover, it is possible that
all countries may not have the technical skills, infrastructural back-
bone, and human/ financial resources to implement all the WHO-
recommended strategies. They are likely to face challenges in
designing appropriate strategies for their countries.
Various countries, including India, need to analyze their public
health scenarios to decide upon the most effective strategies that
require minimum financial commitment [Fig. 3].
It is important to establish an integrated surveillance network
with data from the IDSP, state-level surveillance units, and other
data sources to periodically estimate reliable prevalence estimates
of cholera in various districts, states, and regions of India. Strength-
ening of healthcare facilities with dedicated cholera units at all
levels, implementation of WASH, and community engagement to
encourage safe hygiene practices are the other critical strategies.
Establishing focused research priorities to foster multi-
disciplinary research on epidemiological, molecular, and clinical
aspects of cholera is essential. Establishing public-private partner-
ship ventures to stimulate the development of antimicrobials and
vaccines against cholera would also need coordinated action. India
requires a strong political commitment to eliminate cholera from
the country, as has been done for polio.
A172 M. Das et al. / Vaccine 38 (2020) A167–A174

Reinfecon

Genec resistance Resistant /

Herd Immunity Protecon
Vaccinaon

Birth Direct Contact Chemotherapy

Suscepble Infected Recovered
Immigraon Indirect Contact

Comorbidity
Cholera related
Contaminaon
Comorbidity
Bacterial growth/death

Fig. 2. Comprehensive model for Cholera transmission dynamics.

Fig. 3. Organogram of structural framework of the cholera control programme in India.

For achieving the target of cholera control by 2030, what is
urgently needed is a national taskforce for cholera control with a
national cholera prevention and response road map for India.
Given the fact that health is a state-level subject when it comes
to legislation, a steering committee comprising top-level central
and state health ministry officials should be constituted, which
can be overseen by an oversight committee consisting of the Union
Health Minister and State Health Ministers. It is only with the
patronage of both central and state governments and ownership
of the program by all stakeholders at various levels within and out-
side the Indian health system that the target of cholera control can
be achieved.
It is a limitation of the present analysis that experts outside the
current group of authors were not involved in the situation analy-
 M. Das et al. / Vaccine 38 (2020) A167–A174
sis. Their inclusion could have provided additional insights in
understanding the Indian scenario with reference to preparedness
for cholera control.
4. Conclusions
India has a robust public health infrastructure. Programs like
IDSP, NRHM etc. provide comprehensive surveillance data on var-
ious diseases; however, they lack specific focus for individual dis-
ease prevention and control. Further, there is serious deficiency in
both horizontal and vertical linkage among various government
surveillance programs. India has in place a surveillance platform
and health system mechanisms for detection of ‘‘hotspots” as well
as proper implementation of vaccine/non-vaccine interventional
modalities. However, there also exist considerable gaps and/or
shortfalls in these systems that are compounded by inability to
prove cholera etiology in a time-efficient manner, often resulting
into under-reporting of the disease. An appropriate administrative
and financially viable framework for cholera control in the country
needs to be created. Addressing these inadequacies will be critical
for significantly reducing cholera burden in India and enabling
achievement of cholera control by 2030.
Authors’ contribution
All authors participated in the preparation of the article and
have approved the final version of the manuscript.
Disclaimer
The findings and conclusions in this paper are those of the
authors and do not necessarily represent the official position of
ICMR.
Declaration of Competing Interest
The authors have no conflict of interest to declare.
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
References
[1] Ali M, Nelson AR, Lopez AL, Sack DA. Updated global burden of cholera in
endemic countries. PLoS Negl Trop Dis 2015;9:e0003832.
[2] Report of the ‘Fifth meeting of the Initiative against Diarrheal & Enteric
diseases in Asia- IDEA’; 2017. <http://www.cholera-network.org/wpcontent/
uploads/2017/08/5th-initiative-against-diarrheal-and-enteric-diseases-in-
asia-idea-2017-report.pdf> [accessed 26-09-2017].
[3] Kanungo S, Sah B, Lopez A, Sung J, Paisley A, Sur D, et al. Cholera in India: an
analysis of reports, 1997–2006. Bull World Health Organ 2010;88:185–91.
https://doi.org/10.2471/BLT.09.073460.
[4] Health status indicators in ’National Health Profile 2018’. Central Bureau of
Health Intelligence; 2018. <https://www.cbhidghs.nic.in/WriteReadData/
l892s/Chapter%203%20Health%20Status%20Indicators.pdf> [accessed 10-07-
2019].
[5] Health status indicators in ’National Health Profile 2017’. Central Bureau of
Health Intelligence; 2017. <https://www.cbhidghs.nic.in/WriteReadData/
l892s/Chapter%203.pdf> [accessed 10-07-2019].
[6] National Centre for Disease Control (NCDC): Annual Report 2016 -17. <https://
ncdc.gov.in/WriteReadData/linkimages/Annual%20report2016-17.pdf>
[accessed 10-07-2019].
[7] Global Task Force on Cholera Control. Ending cholera: a global roadmap to
2030. World Health Organization; 2017.
[8] Information and communication technology network. Integrated Disease
Surveillance Programme, Ministry of Health & Family Welfare, Government
of India. <https://idsp.nic.in/index4.php?lang=1&level=0&linkid=408&lid=
3691> [accessed 18-04-2019].
[9] Ramamurthy T, Sharma NC. Cholera Outbreaks in India. Curr Top Microbiol
Immunol 2014;379:49–85. https://doi.org/10.1007/82_2014_368.
[10] Ali M, Sen Gupta S, Arora N, Khasnobis P, Venkatesh S, Sur D, et al.
Identification of burden hotspots and risk factors for cholera in India: an
A173
observational study. PLoS ONE 2017;12:. https://doi.org/10.1371/journal.
pone.0183100e0183100.
[11] Debnath F, Ponnaiah M. Improved timeliness for reporting of acute diarrhoeal
disease under surveillance overtime: evaluation of integrated disease surveillance
programme in North 24 Parganas, West Bengal, India. Clinic Epidemiol Global
Health 2015;2018:163–7. https://doi.org/10.1016/j.cegh.2017.10.006.
[12] Progress on drinking water, sanitation and hygiene: 2017 update and SDG
baselines. Geneva: World Health Organization (WHO) and the United Nations
Children’s Fund (UNICEF); 2017. Licence: CC BY-NC-SA 3.0 IGO.
[13] UN-Water global analysis and assessment of sanitation and drinking-water
(GLAAS) 2017 report: financing universal water, sanitation and hygiene under
the sustainable development goals. Geneva: World Health Organization; 2017.
Licence: CC BY-NC-SA 3.0 IGO.
[14] Swachh Bharat Abhiyan: Why India’s toilet data is too good to be true; 2019.
<https://www.livemint.com/Politics/d0gb1cTpVnVwJaUQIPIaaP/Swachh-
Bharat-Abhiyan-Why-Indias-toilet-data-is-too-good-t.html> [accessed on 12-
04-2019].
[15] Independent Verification of Swachh Bharat Grameen confirms over 96% usage
of toilets. Ministry of Drinking Water & Sanitation; 2019. <http://pib.nic.in/
PressReleaseIframePage.aspx?PRID=1567486> [accessed on 12-04-2019].
[16] Board CPC. National Water Quality Monitoring Programme, <http://www.
cpcb.nic.in/divisionsofheadoffice/pams/NWMP.pdf> [accessed 16-11-2017].
[17] National Rural Health Mission (2005-2012), Mission Document. <https://
www.nhm.gov.in/images/pdf/communitisation/task-group-reports/mission-
document.pdf> [accessed 16-11-2017].
[18] White K. Rohingya in Bangladesh: an unfolding public health emergency.
Lancet 2017;390:1947. https://doi.org/10.1016/S0140-6736(17)32677-6.
[19] Härtl G. Cholera count reaches 500 000 in Yemen 2017, <http://www.who.
int/mediacentre/news/releases/2017/cholera-yemen-mark/en/> [accessed 19-
11-2017].
[20] Sun GQ, Xie JH, Huang SH, Jin Z, Li MT, Liu L. Transmission dynamics of
cholera: mathematical modeling and control strategies. Commun Nonlinear
Sci Numer Simul 2017;45:235–44. https://doi.org/10.1016/J.
CNSNS.2016.10.007.
[21] Siettos CI, Russo L. Mathematical modeling of infectious disease dynamics.
Virulence 2013;4:295–306. https://doi.org/10.4161/viru.24041.
[22] McKay HS, Lessler J, Moore SM, Azman AS. What is a Hotspot Anyway? Am J
Trop Med Hyg 2017;96:1270–3. https://doi.org/10.4269/ajtmh.16-0427.
[23] Fletcher-Lartey SM, Caprarelli G. Application of GIS technology in public
health: successes and challenges. Parasitology 2016;143:401–15. https://doi.
org/10.1017/S0031182015001869.
[24] Jayakumar K, Malarvannan S. Spatial mapping of cholera using GIS tools in
Chennai. Arch Appl Sci Res 2013;5:93–9.
[25] Mishra M, Kurhade A, Thakar Y, Vuma S. Occurrence of a cholera outbreak in
central India. Am J Infect Dis Microbiol 2015;3:141–3.
[26] Nijeeshi TP. Cholera outbreak reported in Malappuram. The Times of India. 14
July 2016, <https://timesofindia.indiatimes.com/city/thiruvananthapuram/
Cholera-outbreak-reported-in-Malappuram/articleshow/53215686.cms>
[accessed 11-12-2017].
[27] Rajiv G. Cholera alert in Kerala: 3 cases and 1 death reported. The Times of
India.3 Aug 2017, <https://timesofindia.indiatimes.com/city/
thiruvananthapuram/cholera-alert-in-kerala-3-cases-and-1-death-reported/
articleshow/59902844.cms> [accessed 11-11-2017].
[28] Wisner B, Adams J. Environmental health in emergencies and disasters: a
practical guide. Geneva: World Health Organization; 2002. <http://www.who.
int/iris/handle/10665/42561> [accessed 13-12-2017].
[29] Jutla A, Whitcombe E, Hasan N, Haley B, Akanda A, Huq A, et al. Environmental
factors influencing epidemic cholera. Am J Trop Med Hyg 2013;89:597–607.
https://doi.org/10.4269/ajtmh.12-0721.
[30] Mukherjee R, Halder D, Saha S, Shyamali R, Subhranshu C, Ramakrishnan R,
et al. Five pond-centred outbreaks of cholera in villages of West Bengal, India:
evidence for focused interventions. J Health Popul Nutr 2011;29:421–8.
[31] Centre for Disease Control. Planning for an emergency: strategies for
identifying and engaging at-risk groups. Atlanta(GA): CDC; 2015.
[32] van Tong H, Velavan TP, Thye T, Meyer CG. Human genetic factors in
tuberculosis: an update. Trop Med Int Health 2017;22:1063–71. https://doi.
org/10.1111/tmi.12923.
[33] Real LA, Biek R. Infectious disease modeling and the dynamics of transmission.
Curr Top Microbiol Immunol 2007;315:33–49.
[34] Kroiss SJ, Famulare M, Lyons H, McCarthy KA, Mercer LD, Chabot-Couture G.
Evaluating cessation of the type 2 oral polio vaccine by modeling pre- and
post-cessation detection rates. Vaccine 2017;35:5674–81. https://doi.org/
10.1016/j.vaccine.2017.08.048.
[35] Okuonghae D, Ikhimwin BO. Dynamics of a mathematical model for
tuberculosis with variability in susceptibility and disease progressions due
to difference in awareness level. Front Microbiol 2016;6:1530. https://doi.org/
10.3389/fmicb.2015.01530.
[36] Koelle K. The impact of climate on the disease dynamics of cholera. Clin
Microbiol Infect 2009;15:29–31. https://doi.org/10.1111/j.1469-
0691.2008.02686.x.
[37] Pascual M, Chaves L, Cash B, Rodó X, Yunus M. Predicting endemic cholera: the
role of climate variability and disease dynamics. Clim Res 2008;36:131–40.
https://doi.org/10.3354/cr00730.
[38] Nishiura H, Tsuzuki S, Yuan B, Yamaguchi T, Asai Y. Transmission dynamics of
cholera in Yemen, 2017: a real time forecasting. Theor Biol Med Model
2017;14:14. https://doi.org/10.1186/s12976-017-0061-x.
A174 M. Das et al. / Vaccine 38 (2020) A167–A174

[39] Abrams JY, Copeland JR, Tauxe RV, Date KA, Belay ED, Mody RK, et al. Real-time CSU: Central Surveillance Unit
modelling used for outbreak management during a cholera epidemic, Haiti, DPHL: District Public Health Laboratories
2010–2011. Epidemiol Infect 2013;141:1276–85. https://doi.org/10.1017/ DSU: District Surveillance Units
S0950268812001793. EDUSAT: Education Satellite
[40] Andrews JR, Basu S. Transmission dynamics and control of cholera in Haiti: an GEMS: Global Environmental Monitoring System
epidemic model. Lancet 2011;377:1248–55. https://doi.org/10.1016/S0140- GTFCC: Global Task Force on Cholera Control
6736(11)60273-0.
IDSP: Integrated Disease Surveillance Program
[41] King AA, Ionides EL, Pascual M, Bouma MJ. Inapparent infections and cholera
IDH: Infectious Diseases Hospital
dynamics. Nature 2008;454:877–80. https://doi.org/10.1038/nature07084.
[42] Hartley DM, Morris JG, Smith DL. Hyperinfectivity: a Critical element in the ISRO: Indian Space Research Organization
ability of V. cholerae to cause epidemics? PLoS Med 2005;3:. https://doi.org/ MINARS: Monitoring of Indian National Aquatic Resources System
10.1371/journal.pmed.0030007e7. NRHM: National Rural Health Mission:
[43] Chao DL, Longini IM, Morris JG. Modeling Cholera Outbreaks. In: Nair GB, OCVs: Oral Cholera Vaccines
Takeda Y, editors. Cholera outbreaks. Berlin/Heidelberg: Springer; 2013. p. PRU: Peripheral Reporting Units
195–209. https://doi.org/10.1007/82_2013_307. RRTs: Rapid Response Teams
[44] Fung IC-H. Cholera transmission dynamic models for public health SIR: Susceptible-Infected-Recovered
practitioners. Emerg Themes Epidemiol 2014;11:;1. https://doi.org/10.1186/ SMS: Short Message Service
1742-7622-11-1. SRL: State Based Referral Laboratory
SSU: State Surveillance Units
Glossary VHSNC: Village Health & Sanitation Committees
WASH: Water Sanitation and Hygiene
BOD: Biological Oxygen Demand WHO: World Health Organization
CBHI: Central Bureau of Health Intelligence YAP: Yamuna Action Plan