Abstract
Background
Between April and June 2012, a reactive cholera vaccination campaign was conducted in Haiti using an oral inactivated bivalent whole-cell vaccine (BivWC).
Methods
We conducted a case-control study to estimate field effectiveness of the vaccine. Cases had acute watery diarrhea, sought treatment at one of three participating cholera treatment units from October 24, 2012 through March 9, 2014, and had a stool sample positive for cholera by culture. For each case, four controls (individuals who did not seek treatment for acute watery diarrhea) were matched by location of residence, calendar time, and age. We also conducted a bias-indicator case-control study to assess the likelihood of bias in the vaccine effectiveness (VE) study.
Findings
During the study period, 114 eligible individuals presented with acute watery diarrhea and were enrolled. 47 were analyzed as cases in the VE case-control study and 42 as cases in the bias-indicator study. In multivariable analyses, VE was 63% [95% confidence interval (CI): 8%–85%] by self-reported vaccination and 58% [95% CI: 13%–80%] for verified vaccination. Neither self-reported nor verified vaccination was significantly associated with non-cholera diarrhea (VE: 18% [95% CI: −208%–−78%] by self-report and −21% [95%CI: −238%–57%] for verified vaccination).
Interpretation
BivWC oral cholera vaccine was effective in protecting against cholera in Haiti during the study period –from 4 through 24 months after vaccination. Vaccination is an important component of epidemic cholera control efforts.
Funding
National Institutes of Health, Delivering Oral Vaccines Effectively project, Department of Global Health and Social Medicine at Harvard Medical School.
BACKGROUND
Cholera epidemics continue to cause major morbidity and mortality globally, and recent large outbreaks in Haiti, Zimbabwe, and Sierra Leone1–3 demonstrate the urgent need for improved control measures to save lives and to reduce human suffering. Two oral cholera vaccines(OCVs) are pre-qualified for use by the World Health Organization (WHO)—a bivalent inactivated vaccine containing killed whole cells of V. cholerae O1 and V. cholerae O139 (BivWC), marketed as Shanchol™, and an inactivated vaccine containing killed whole cells of V. cholerae O1 with recombinant B-subunit of cholera toxin (WC/rBS), marketed as Dukoral™. Although OCVs have been found to be safe and effective in large field trials,4–7 and WHO has recommended consideration of their use in epidemics, OCVs are not routinely used in the setting of cholera outbreaks.8 This is a result of a number of factors,9 but not least is the lack of data on the effectiveness of the vaccines under field conditions, especially during the complex situation of epidemic cholera.10 In 2012, large-scale reactive OCV campaigns, conducted in Guinea and Haiti using BivWC, have contributed to increased understanding of the use of OCV in epidemic settings.11, 12
In response to a major cholera epidemic in Haiti that began in 2010, we conducted an oral cholera vaccination campaign using BivWC in partnership with the Haitian Ministry of Health from April through June 2012, vaccinating 45,417 people, of which 90.8% received both doses of the two-dose vaccination schedule.12, 13 We subsequently undertook a case-control study to evaluate the field effectiveness of the vaccine.
METHODS
Study setting
Ten months after an earthquake near the capital city of Port-au-Prince resulted in a massive humanitarian disaster,14 a major cholera epidemic began in central Haiti in October 2010 and rapidly spread throughout entire country within a month.3 The Haitian National Public Health Laboratory identified Vibrio cholerae serogroup O1, biotype El Tor as the cause of the epidemic by culture of stool specimen.3 By May 2014, there were 702,892 reported cases, of which 393,450 were hospitalized, and 8,561 died, with unmeasured social and financial cost.15, 16 Cholera had never been reported in Haiti prior to this outbreak. At the time that the vaccination campaign began, the cholera epidemic had been ongoing for 17 months in Haiti, and reported weekly case incidence remained as high as 123.5 cases per 10,000 population in some regions of the country.15
The study was conducted in the Artibonite Department of Haiti, in three health centers that provide primary health services to the regions of Bocozel and Grand Saline. Bocozel and Grand Saline are rural, rice-growing regions of the country that are irrigated by branches of the Artibonite river, without which they are dry and desert-like. Located approximately 120 km from Port-au-Prince, they have a combined population of approximately 55,000 people, and were targeted by the OCV campaign between April and June 2012. The areas are served by a 150-bed ministry of health hospital, L’Hopital St Nicolas, which is supported by the non-governmental organization Partners In Health, and by two dispensaries, one in each of the isolated communities of Drouin and Bocozel. An estimated 76·7–92·7% of the Bocozel community, and 62·5% of the Grand Saline community were vaccinated against cholera in the 2012 campaign.12 The vaccination campaign is described in detail elsewhere.12
Study design and participants
We conducted two matched case-control studies. The primary study aimed to examine the effectiveness of oral cholera vaccination. The second case-control study was a bias-indicator study, in which we assessed the likelihood of bias in the primary case-control study by examining the relationship between vaccination and non-cholera diarrhea.17 To be eligible for inclusion in either study, cases and controls were required to be residents of Bocozel or Grand Saline at the time of the study initiation, to have been eligible for the vaccination campaign (i.e. ≥ 12 months of age, not pregnant, and living in the region at the time of the vaccine campaign), and to be able to provide informed consent for participation. “Resident” was defined as eating and sleeping in a household in the location for most days of the week.
We recruited study participants at the health centers, and we trained community health workers in the region to conduct surveillance and refer acute watery diarrhea cases to the health center (after initiating oral rehydration as appropriate). We enrolled individuals with acute watery diarrhea (defined as three or more loose, non-bloody, liquid stools in a 24-hour period with an onset of three days or fewer prior to presentation) who sought treatment at any of the three study sites and met all eligibility criteria. Participants were asked to provide a stool sample for testing by the Crystal VC™ rapid test (Span Diagnostics, India, hereafter referred to as ‘rapid test’), and by culture. Cases were later classified as either cholera cases or non-cholera cases based on the results of rapid test and culture.
Cases in the vaccine effectiveness study were individuals with a stool sample positive for Vibrio cholerae O1 by culture. Controls were individuals who did not seek treatment for diarrhea between the first day of study enrollment and their corresponding case’s symptom onset date. Four controls were matched to each case by location of residence, calendar time, age (1–4 years, 5–15 years, >15 years).18 When more than one eligible control was available in a household, an individual of the same gender was selected when possible. In rural Haiti, households are often grouped in a cluster of multigenerational families called “lakou”. In choosing controls, study workers approached the home nearest to the case’s home, excluding homes within the same lakou, as we anticipated that exposure to the cholera vaccine was likely to be highly correlated within the lakou. Study workers then moved to the next closest residence and repeated until four controls were enrolled. Study enrollment began four months after the vaccination campaign ended.19
Cases in the bias-indicator study were individuals with acute watery diarrhea with a stool sample testing negative for cholera by both rapid test and culture. Controls were selected in the same manner as in the primary case-control study. Because the vaccine was not expected to provide protection against non-cholera diarrhea, in the absence of bias we expected a null association between vaccination and non-cholera diarrhea in the bias-indicator case-control study.17, 20, 21
Laboratory testing
Trained staff used standard laboratory procedures to perform rapid tests and stool cultures from study cases, and results were recorded in the study register. The stool samples were collected in sterile containers, and rapid tests were performed immediately according to the manufacturer’s protocol. An additional specimen was transported in Cary-Blair media to the Haitian National Public Health Laboratory in Port-au-Prince for subsequent culture on thiosulfate-citrate-bile salts-sucrose. Identification of V. cholerae serogroup O1 at the serotype level was performed using a standard slide agglutination method.22
Data collection
To collect data on socio-demographic characteristics, risk factors for cholera, and self-reported vaccination, study workers conducted interviews with participants at the cholera treatment unit. In instances where the participant was a child or was unavailable for interview, guardians or a family member proxy responded to questions on behalf of the participant. Individuals who reported to have received at least one dose of the vaccine were asked to produce their vaccine card as verification at a later home visit. Vaccination registries were used to verify vaccination status for individuals who reported vaccination but could not produce a vaccine card, or who reported no vaccination. A study worker reviewed the clinical charts of confirmed cholera cases.
Statistical analyses
We analyzed data using conditional logistic regression analyses, adjusting for matching factors. Models included a measure of vaccination status (self-reported vaccination with at least one dose, number of self-reported doses, or documented vaccination with at least one dose per vaccine card or registry review). Multivariable models adjusted for sex and age as a continuous variable because matching on age was conducted using broad age categories. We additionally adjusted for variables that were associated with both vaccination and cholera (or non-cholera diarrhea in the bias-indicator study) at a p-value of <0·20 (i.e., potential confounders). Field effectiveness of the vaccine was calculated using the vaccine effectiveness formula: vaccine effectiveness = (1 – relative risk).23 A likelihood ratio test was used to examine whether there was a linear dose-response relationship between the number of vaccine doses received and effectiveness. We examined whether cholera vaccine effectiveness varied by time since vaccination (before versus on or after July 1, 2013, approximately one year after the completion of the vaccination campaign), age group (less than versus greater than or equal to five years old), prior admission to a cholera treatment unit, and severity of dehydration. To do this we included an interaction term between the potential modifier and any self-reported vaccination. Due to small sample sizes in subgroups, we reported ratios adjusted for matching factors only.
Ethical approval for this study was obtained from Partners Institutional Review Board (Boston, MA), and the Haitian National Bioethics Committee (Port-au-Prince, Haiti). The vaccination campaign itself was a public health campaign directed by the Ministry of Health and Population of Haiti, implemented by Partners In Health and GHESKIO, and approved by the Haitian Bioethics Committee; informed consent for vaccination was not required during the campaign, but all research study participants signed informed consent prior to participation. For those that were unable to consent, (for example cases that were too unwell), a healthcare proxy was permitted to provide consent to participate.
Role of the funding source
The funders had no role in the study design, execution, analysis, reporting or decision to submit the manuscript. The corresponding author had full access to all data in the study and had final responsibility for the decision to submit for publication.
RESULTS
From October 24, 2012 through March 9, 2014, 114 eligible individuals presented with acute watery diarrhea and were enrolled. Of these, 48 were confirmed to have cholera by culture and therefore were cases in the primary case-control study of cholera vaccine effectiveness. One case was subsequently excluded from analysis of vaccine effectiveness when electronic interview data was lost due to a technology malfunction. Forty-two individuals were confirmed not to have cholera by a negative rapid test and culture, and therefore were analyzed as cases in the secondary bias-indicator study (Figure 1). Cholera and non-cholera diarrhea cases in the study were geographically dispersed throughout the study area, with most recruited cases residing in Bocozel and southern Grand Saline. Among the 48 cholera cases, presentation to the clinic tended to occur within a day of symptom onset (Table 1), and over half of cases had severe (grade C) dehydration. Median stay at the cholera treatment unit was 3 days (interquartile range 2–4 days). Figure 2 displays the timeline of presentation of cholera and non-cholera study cases during the study period. Cholera cases in the study were generally more frequent at times that corresponded to the rainy season.
Figure 1.
Overview of case enrollment and cholera test results in a case-control study of oral cholera vaccine effectiveness in Haiti
Table 1.
Clinical presentation and treatment of cholera cases
| Cholera diarrhea cases N=48 |
Non-cholera diarrhea cases N=41a |
|
|---|---|---|
| Variable | n (%) | n (%) |
| Time from symptom onset to admission (days)b | 0 [0 – 1] | 0 [0 – 1] |
| Serotype | ||
| Ogawa | 37 (77) | Not applicable |
| Inaba | 11 (23) | Not applicable |
| Dehydration stage at presentation | ||
| A (Mild) | 0 (0) | 10 (24) |
| B (Moderate) | 22 (46) | 25 (61) |
| C (Severe) | 26 (54) | 6 (15) |
| Treatment received at clinic | ||
| Oral rehydration solution | 46 (96) | 36 (95) |
| Intravenous fluids | 46 (96) | 29 (71) |
| Antibiotics | 9 (19) | 2 (5) |
| Amount of oral rehydration solution given in clinic (liters)b,c | 10 [6 – 14] | 2 [2 – 4.5] |
| Amount of intravenous fluid given in clinic (liters)b,d | 15 [9 – 20] | 3 [2 – 6] |
| Admitted overnight to the cholera treatment unit | 48 (100) | 29 (71) |
| Duration of stay at cholera treatment unit (days)b | 3 [2 – 4] | 1 [0 – 1] |
| Outcome | ||
| Discharged | 45 (94) | 38 (93) |
| Transferred | 1 (2) | 0 (0) |
| Died | 1 (2) | 0 (0) |
| Left against medical advice | 1 (2) | 3 (7) |
Clinical data available was missing for one of 42 non-cholera diarrhea cases.
Median [interquartile range]
Among those given oral rehydration solution(46 cholera diarrhea cases, 36 noncholera diarrhea cases)
Among those given intravenous fluids (46 cholera diarrhea cases, 29 non-cholera diarrhea cases)
Figure 2.
Cases of cholera and non-cholera acute watery diarrhea enrolled in one of two case-control studies in Haiti, October 24, 2012–March 9, 2014
Cases classified based on culture results; where culture was not available (n=8), based on rapid test result. Depicted rainy seasons based on average rainfall, not actual rainfall.
Vaccine effectiveness case-control study
Characteristics of cases and controls are shown in Table 2. In the primary case-control study, cholera cases were less likely than controls to be female, and less likely to have completed the interview themselves. They were more likely to have electricity in their home and to have a latrine as their main toilet type. Seventy percent of cases and 89% of controls self-reported vaccination, and we verified vaccination in 57% of cases and 78% of controls, respectively. Among those vaccinated, the majority received both vaccine doses (Table 3). In univariate analyses adjusting for matching factors, both any self-reported vaccination and any vaccination verified through vaccine card or registry were associated with a statistically significant reduction in the risk of cholera (Table 3). In multivariable analyses vaccine effectiveness was 63% [8%–85%] by self-report and 58% [13%–80%] for vaccination verified through the card or registry. We examined whether protection increased with each dose of the vaccine. While the small number of individuals receiving one dose did not experience a statistically significantly lower rate of cholera relative to those who were unvaccinated, we could not rule out the possibility of a linear relationship between number of doses and effectiveness based on the likelihood ratio test (likelihood ratio p-value: 0·67). We did not identify an interaction between the vaccine and age group, time since vaccination, severity of dehydration, or prior admission to a cholera treatment unit (Table 4).
Table 2.
Characteristics of cholera cases, non-cholera diarrhea cases, and controls
| Cholera vaccine effectiveness case- control study |
Bias-indicator case-control study | |||||
|---|---|---|---|---|---|---|
| Variable | Cholera cases n (%) (N=47) |
Controls n (%) (N=188) |
p- valuea |
Non-cholera diarrhea cases (N=42) |
Controls (N=168) |
p- valuea |
| Female | 16 (34) | 97 (52) | 0.010 | 26 (62) | 102 (61) | 0.823 |
| Ageb,c | 27 [13 – 43] | 30·5 [9·5 – 44·5] | 31.5 [20 – 45] | 34 [18.5 – 48] | ||
| Participant responded to interview (versus proxy) | 30 (64) | 136 (72) | 0.012 | 35 (83) | 133 (79) | 0.264 |
| Earthen floor in home (versus cement or wood) | 36 (77) | 141 (75) | 0.773 | 35 (83) | 112 (67) | 0.021 |
| Ever attended school | 22 (47) | 112 (60) | 0.088 | 28 (67) | 110 (65) | 0.877 |
| House has electricity | 8 (17) | 19 (10) | 0.045 | 6 (14) | 23 (14) | 0.860 |
| Household sizeb | 5 [3 – 7] | 5 [4 – 6] | 0.424 | 5 [3 – 6] | 5 [3 – 7] | 0.877 |
| Agriculture is main income generating activity | 29 (62) | 109 (58) | 0.585 | 22 (52) | 100 (60) | 0.350 |
| Main toilet is latrine (versus unimproved or open defecation) | 29 (62) | 90 (48) | 0.043 | 26 (62) | 95 (58) | 0.475 |
| Ever admitted overnight to a cholera treatment unitd | 6 (17) | 20 (14) | 0.655 | 5 (13) | 19 (12) | 0.910 |
| Household member with cholera in the last week | 2 (4) | 3 (2) | 0.283 | 2 (5) | 2 (1) | 0.152 |
| Household member ever spent a night in cholera treatment unit | 17 (36) | 50 (27) | 0.192 | 17 (40) | 57 (34) | 0.405 |
| Household member with diarrhea in the last weeke | 8 (17) | 18 (10) | 0.117 | 4 (10) | 11 (7) | 0.491 |
P-value from univariable conditional logistic regression adjusting for matching factors
Median [interquartile range]
Cases and controls were matched by age; therefore, no pvalue is provided.
For cholera case-control study, data available for 35 cases and 140 controls. In non-cholera diarrhea case-control study, data available for 39 cases and 160 controls.
For cholera case-control study, data available for 46 cases.
Table 3.
Effectiveness of the oral cholera vaccine in rural Haiti
| Cases n (%) |
Controls n (%) |
Crude RRa [95% CI] |
Adjusted RR [95% CI] |
Vaccine effectiveness, % [95% CI] |
p- value |
|
|---|---|---|---|---|---|---|
| Cholera vaccine effectiveness case-control study | N=47 | N=188 | ||||
| Vaccinated, self-report | 33 (70) | 167 (89) | 0·27 [0·12 – 0·61] | 0·37 [0·15 – 0·92]b | 63 [8 – 85] | 0·031 |
| Number of self-reported doses | ||||||
| None | 14 (30) | 21 (11) | Reference | Reference | ||
| One | 3 (6) | 19 (10) | 0·20 [0·05 – 0·87] | 0·33 [0·07 – 1·62]b | 67 [−62 – 93] | 0·170 |
| Two | 30 (64) | 148 (79) | 0·28 [0·13 – 0·63] | 0·38 [0·15 – 0·94]b | 62 [6 – 85] | 0·036 |
| Proof of vaccination (card or registry record) | 27 (57) | 147 (78) | 0·35 [0·17 – 0·72] | 0·42 [0·20 – 0·87]c | 58 [13 – 80] | 0·020 |
| Bias-indicator case-control study | N=42 | N=168 | ||||
| Vaccinated, self-report | 39 (93) | 158 (94) | 0·83 [0·22 – 3·09] | 0·82 [0·22 – 3·08]c | 18 [−208 – 78] | 0·773 |
| Number of self-reported doses | ||||||
| None | 3 (7) | 10 (6) | Reference | Reference | ||
| One | 7 (17) | 11 (6) | 2·50 [0·47 – 13·25] | 2·53 [0·48 – 13·37]c | −153 [−1237 – 52] | 0·275 |
| Two | 32 (77) | 147 (88) | 0·73 [0·19 – 2·78] | 0·72 [0·19 – 2·74]c | 28 [−174 – 81] | 0·631 |
| Proof of vaccination (card or registry record) | 36 (86) | 137 (82) | 1·39 [0·52 – 3·70] | 1·21 [0·43 – 3·38]d | −21 [−238 – 57] | 0·718 |
Adjusted for matching factors
Adjusted for matching factors, female sex, age (continuous), electricity in the home, main toilet type, and whether the participant completed the interview (versus a proxy)
Adjusted for matching factors, female sex and age (continuous)
Adjusted for matching factors, female sex, age (continuous), and earthen floor in the household
Table 4.
Subgroup analyses of effectiveness of an oral killed whole cell bivalent cholera vaccine in rural Haiti
| Cases n (%) N=47 |
Controls n (%) N=188 |
CrudeRR for self-reported vaccinationa [95% CI] |
Vaccine effectiveness, % [95% CI] |
p-value for interaction |
|
|---|---|---|---|---|---|
| Age | |||||
| <5years | 9 (19) | 31 (16) | 0·50 [0·03 – 9·5] | 50 [−850 – 97] | 0·703 |
| ≥5years | 38 (81) | 157 (84) | 0·28 [ 0·12 – 0·64] | 72 [36 – 88] | |
| Dehydration stage at presentation | |||||
| B (Moderate) | 21 (45) | -- | 0·28 [0·08 – 0·97] | 72 [3 – 92] | 0·945 |
| C (Severe) | 26 (55) | -- | 0·27 [0·09 – 0·77] | 73 [23 – 91] | |
| Time since vaccination | |||||
| <1 year | 12 (26) | 49 (26) | 0·13 [0·02 – 0·68] | 87 [ 32 – 98] | 0·286 |
| ≥1 year | 35 (74) | 139 (74) | 0·36 [0·14 – 0·90] | 64 [10 – 86] | |
| Prior admission to a cholera treatment unitb | |||||
| Yes | 6 (17) | 20 (14) | 1·03 [0·08 – 12·61] | −3 [−1161 – 92] | 0·353 |
| No | 29 (83) | 120 (86) | 0·29 [0·11 – 0·79] | 71 [21 – 89] |
Adjusted for matching factors
For cholera case-control study, data available for 35 cases and 140 controls.
Bias-indicator case-control study
Nearly all cases and controls in the bias-indicator study self-reported receipt of at least one dose of vaccine (93% and 94%, respectively; Table 4). Vaccination with at least one dose of vaccine was verified in 86% of cases and 82% of controls. Neither self-reported nor verified vaccination was significantly associated with non-cholera diarrhea in univariate or multivariable analyses (vaccine effectiveness in multivariable analyses: 18% [95% CI: −208%–78%] by self-report and − 21% [95% CI: −238%–57%] for verified vaccination; Table 4).
DISCUSSION
We found that BivWC vaccine was effective in reducing the rate of cholera amongst vaccine recipients in rural Haiti during the study period – from 4 to 24 months after vaccination began. We found a vaccine effectiveness that was similar to the estimate of protective effectiveness of the same vaccine when administered in a historically cholera-endemic region: Sur et al reported a two-year vaccine effectiveness of 67% in India.6 This may be because the vaccine was administered in Haiti after many months of protracted transmission.
Our findings address the critical knowledge gap surrounding the use of oral cholera vaccines in reactive vaccination in populations that lack endemic exposure to cholera.8 Such populations are most at risk of explosive epidemics once cholera emerges. A study from the reactive oral cholera vaccination campaign in Guinea in 2012 recently demonstrated that BivWC offered significant early protection against cholera at 6 months(effectiveness 86·6%; 95% confidence interval 56·7–95·7, P = 0·001).11 Our study similarly suggests that reactive vaccination is effective, and in this case, in a population with no historical exposure or immunity to cholera, in the midst of the worst cholera epidemic of the past decade. Furthermore, it extends the findings of the Guinea study, showing that the benefit of vaccination extends beyond six months out to two years post-vaccination. Our results are also consistent with our previous finding that seroconversion rates amongst Haitian vaccine recipients were robust and comparable to those of Bangladeshis that had also received two doses of the vaccine.24
Analysis of data from a placebo-controlled trial of WC-rBS vaccine in Bangladesh demonstrated that vaccination provided significant herd protection to neighboring non-vaccinated individuals.25 In our study, community coverage in the region targeted by the vaccination campaign, which is also where the study sites were located, was between 62·5–92·7%.12 This is more than is believed to be required for significant herd immunity.25 We also previously demonstrated that the vaccination campaign in Haiti was associated with significant improvements in cholera knowledge and practices related to waterborne disease,26 suggesting additional indirect benefits associated with vaccination.
This was an observational field study, and it is worth noting some limitations. We assessed vaccine exposure through self-report, and we verified vaccination through documentation of the vaccine card or registration in the vaccine registry. Neither of these assessments is perfect: self-reported vaccination may be influenced by a cholera episode (and therefore differ by case or control status), while lost vaccine cards and misspelled names may lead to underestimates of vaccination coverage. However, despite this, we observed consistent, robust estimates of vaccine effectiveness on the order of 58–63%, using two vaccination assessments with errors that are unlikely to be correlated. The small number of cases in this study limited power for subgroup and dose response analyses, which merit examination in larger studies. We assessed the likelihood of bias in the vaccine effectiveness case-control study and found no association between vaccination and non-cholera diarrhea, supporting the validity of our study findings.
Now in its fourth year, the cholera epidemic in Haiti has decelerated in terms of case incidence, but continues to cause significant morbidity and mortality. Over 58,500 cases were registered in 2013 alone, and 6,071 cases and 30 deaths occurred in the first five months of 2014.15 The epidemic subsequently spread from Haiti to other countries in the region, including the Dominican Republic, Cuba, and Mexico.28 The rapid evolution of this epidemic and its toll on human life as well as on the health system of Haiti have reinforced the need for improved methods for controlling cholera in the country. Investing in water and sanitation infrastructure in Haiti will have widespread impact on improving health, and is the cornerstone of a binational plan to eliminate cholera from Hispaniola.29 In addition to intermediate- and long-term water and sanitation intervention show ever, new approaches to disease control are urgently needed. Our study contributes to mounting evidence that oral cholera vaccines have an important role to play as a component of comprehensive, integrated cholera control efforts in Haiti.
Conclusion
Cholera vaccination with BivWC in the midst of a cholera epidemic in rural Haiti, where cholera had been recently introduced, was associated with a significant protective effect during the first 24 months following the vaccination campaign. Oral cholera vaccination should be strongly considered for use as a complementary tool in the response to cholera epidemics, including in the protracted epidemic in Haiti.
Panel: Research in context.
Systematic review
Prior to planning the study, we searched PubMed with the search terms “oral cholera vaccine”, “cholera”, “vaccine effectiveness”, “vaccine efficacy”, and “case control” for articles. This search yielded 147 records from which we identified 12 relevant studies, many of which were also included in a Cochrane review of OCV studies.30 We also identified relevant articles and studies from agencies and resources such as the World Health Organization, the Haitian Ministry of Health website (articles in French), NIH RePORTER, and ClinicalTrials.gov. Three additional papers were published after our study began. Two case control studies were conducted in Mozambique and Zanzibar using a recombinant cholera-toxin B subunit vaccine (rBS-WC), and showed 78% and 79% protection against cholera, respectively.20, 31 In the Mozambique study, there was no difference in effectiveness between those under or over five years of age.20 In trials conducted in endemic settings, the same oral cholera bivalent whole-cell (BivWC) vaccine that we studied demonstrated protection with 65% effectiveness in India5 and approximately 50% effectiveness in Bangladesh32 at three-year follow-up. In outbreak settings, a killed whole cell vaccine produced in Vietnam (ORC-VAX) had 76% protection33, and BivWC had 86.6% protection at 6 months in Guinea.11 In Peru, rBS-WC OCV was found to have protective efficacy at 24-month follow-up ranging from 61–72%, with better protection in those 15 years and older.34 In the same campaign in India mentioned above, a follow-up was published demonstrating that the vaccine sustained a cumulative 65% protective efficacy at 5 years.4
Interpretation
In our case-control study of BivWC oral cholera vaccine in a protracted epidemic in Haiti, vaccination was protective against cholera up to two years of follow-up. Our study adds to the literature by demonstrating that OCV is effective in an epidemic setting, in addition to endemic settings like India and Bangladesh, where the majority of evidence on OCV effectiveness has been collected. Our findings in the context of the existing literature suggest that OCV can be used as a complementary tool in the response to cholera outbreaks, and can reduce the disease burden of cholera in protracted epidemic settings. Continued follow-up should be conducted to ascertain longer-term vaccine effectiveness in epidemics, especially in populations such as that in our study, that do not have historic exposure to endemic cholera. Further research is also warranted to understand the effectiveness of a single dose of vaccine, the safety and effectiveness of the vaccine in pregnancy, and the heat-stability of the vaccine, all of which would improve the ease with which mass vaccination campaigns could be undertaken.
Acknowledgements
The study was supported by grants from the National Institutes of Health, (NIAID R01 AI099243; JBH, LCI, JET, MFF), the Delivering Oral Vaccines Effectively project at Johns Hopkins University, which is funded by the Bill & Melinda Gates Foundation (LCI, JET),and the Global Health Research Core of the Department of Global Health and Social Medicine (MFF). We thank the study participants, the Ministry of Health and Population of Haiti, the Abundance Fund, and the staff of ZanmiLasante. We are also grateful to the American Red Cross and private donors who funded the oral cholera vaccination campaign.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Contributors
LCI conceived the study, designed the protocol, contributed to analysis and interpretation of data, and drafted the first version of the manuscript. MFF designed the protocol, analyzed the data, contributed to interpretation of data, and contributed to the first version of the manuscript. IJH was responsible for data collection and contributed to interpretation of the data. JET contributed to design of the study, collection of data and interpretation of findings. JB, JB contributed to the analysis of data. JBH, MM, CPA, JGJ, RT all contributed to the design of the study and to interpretation of data. All authors contributed significantly to the manuscript revisions and approved the final version to be published.
Competing interests
All authors declare that they have no competing interests.
References Cited
- 1.Outbreak news. Cholera, Sierra Leone. Releve epidemiologique hebdomadaire / Section d'hygiene du Secretariat de la Societe des Nations = Weekly epidemiological record / Health Section of the Secretariat of the League of Nations. 2012;87(36):337–338. [PubMed] [Google Scholar]
- 2.Cholera outbreak Zimbabwe. Releve epidemiologique hebdomadaire / Section d'hygiene du Secretariat de la Societe des Nations = Weekly epidemiological record / Health Section of the Secretariat of the League of Nations. 2009;84(7):50–52. [Google Scholar]
- 3.Barzilay EJ, Schaad N, Magloire R, et al. Cholera surveillance during the Haiti epidemic--the first 2 years. The New England journal of medicine. 2013;368(7):599–609. doi: 10.1056/NEJMoa1204927. [DOI] [PubMed] [Google Scholar]
- 4.Bhattacharya SK, Sur D, Ali M, et al. 5 year efficacy of a bivalent killed whole-cell oral cholera vaccine in Kolkata, India: a cluster-randomised, double-blind, placebo-controlled trial. The Lancet infectious diseases. 2013;13(12):1050–1056. doi: 10.1016/S1473-3099(13)70273-1. [DOI] [PubMed] [Google Scholar]
- 5.Sur D, Kanungo S, Sah B, et al. Efficacy of a low-cost, inactivated whole-cell oral cholera vaccine: results from 3 years of follow-up of a randomized, controlled trial. PLoS neglected tropical diseases. 2011;5(10):e1289. doi: 10.1371/journal.pntd.0001289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sur D, Lopez AL, Kanungo S, et al. Efficacy and safety of a modified killed-whole-cell oral cholera vaccine in India: an interim analysis of a cluster-randomised, double-blind, placebo-controlled trial. Lancet. 2009;374(9702):1694–1702. doi: 10.1016/S0140-6736(09)61297-6. [DOI] [PubMed] [Google Scholar]
- 7.van Loon FP, Clemens JD, Chakraborty J, et al. Field trial of inactivated oral cholera vaccines in Bangladesh: results from 5 years of follow-up. Vaccine. 1996;14(2):162–166. doi: 10.1016/0264-410x(95)00122-h. [DOI] [PubMed] [Google Scholar]
- 8.World Health Organization. Meeting of the Strategic Advisory Group of Experts on immunization, Oct 2009 - conclusions and recommendations. Weekly Epidemiological record. 2009;84(50):517–531. [PubMed] [Google Scholar]
- 9.von Seidlein L, Jiddawi M, Grais RF, Luquero F, Lucas M, Deen J. The value of and challenges for cholera vaccines in Africa. The Journal of infectious diseases. 2013;208(Suppl 1):S8–S14. doi: 10.1093/infdis/jit194. [DOI] [PubMed] [Google Scholar]
- 10.Deen J, Ali M, Sack D. Methods to assess the impact of mass oral cholera vaccination campaigns under real field conditions. PloS one. 2014;9(2):e88139. doi: 10.1371/journal.pone.0088139. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Luquero FJ, Grout L, Ciglenecki I, et al. Use of Vibrio cholerae vaccine in an outbreak in Guinea. The New England journal of medicine. 2014;370(22):2111–2120. doi: 10.1056/NEJMoa1312680. [DOI] [PubMed] [Google Scholar]
- 12.Ivers LC, Teng JE, Lascher J, et al. Use of oral cholera vaccine in Haiti: a rural demonstration project. Am J Trop Med Hyg. 2013;89(4):617–624. doi: 10.4269/ajtmh.13-0183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ivers LC, Farmer PE, Pape WJ. Oral cholera vaccine and integrated cholera control in Haiti. Lancet. 2012;379(9831):2026–2028. doi: 10.1016/S0140-6736(12)60832-0. [DOI] [PubMed] [Google Scholar]
- 14.Pape JW, Johnson WD, Jr, Fitzgerald DW. The earthquake in Haiti--dispatch from Port-au-Prince. The New England journal of medicine. 2010;362(7):575–577. doi: 10.1056/NEJMp1001015. [DOI] [PubMed] [Google Scholar]
- 15.Ministère de la Santé et de la Population H. Rapport de cas [French], 19 May, 2014. 19 May, 2014 ed. 2014. [Google Scholar]
- 16.Ministère de la Santé et de la Population H. Rapport de cas [French], 1 Janvier, 2014. 1 January, 2014 ed. 2014. [Google Scholar]
- 17.Shapiro ED. Case-control studies of the effectiveness of vaccines: validity and assessment of potential bias. The Pediatric infectious disease journal. 2004;23(2):127–131. doi: 10.1097/01.inf.0000109248.32907.1d. [DOI] [PubMed] [Google Scholar]
- 18.Uyr H. Efficiency of case-control studies with multiple controls per case: continuous or dichotomous data. Biometrics. 1975;31:643–649. [PubMed] [Google Scholar]
- 19.Edmond Y, Randolph S, Richard G. The Lakou System: A Cultural, Ecological Analysis of Mothering in Rural Haiti. The Journal of Pan African Studies. 2007;2(1):19. [Google Scholar]
- 20.Lucas ME, Deen JL, von Seidlein L, et al. Effectiveness of mass oral cholera vaccination in Beira, Mozambique. The New England journal of medicine. 2005;352(8):757–767. doi: 10.1056/NEJMoa043323. [DOI] [PubMed] [Google Scholar]
- 21.Lipsitch M, Tchetgen Tchetgen E, Cohen T. Negative controls: a tool for detecting confounding and bias in observational studies. Epidemiology. 2010;21(3):383–388. doi: 10.1097/EDE.0b013e3181d61eeb. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.World Health Organization. A manual for the laboratory identification and antimicrobial susceptibility testing of bacterial pathogens of public health importance in developing world. WHO/CDS/CSR/RMD/2003.6 Geneva, Switzerland: World Health Organization; 2003. [Google Scholar]
- 23.Halloran ME, Struchiner CJ, Longini IM., Jr Study designs for evaluating different efficacy and effectiveness aspects of vaccines. American journal of epidemiology. 1997;146(10):789–803. doi: 10.1093/oxfordjournals.aje.a009196. [DOI] [PubMed] [Google Scholar]
- 24.Charles RC, Hilaire IJ, Mayo-Smith LM, et al. Immunogenicity of a killed bivalent (O1 and O139) whole cell oral cholera vaccine, Shanchol, in Haiti. PLoS neglected tropical diseases. 2014;8(5):e2828. doi: 10.1371/journal.pntd.0002828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Ali M, Emch M, von Seidlein L, et al. Herd immunity conferred by killed oral cholera vaccines in Bangladesh: a reanalysis. Lancet. 2005;366(9479):44–49. doi: 10.1016/S0140-6736(05)66550-6. [DOI] [PubMed] [Google Scholar]
- 26.Aibana O, Franke MF, Teng JE, Hilaire J, Raymond M, Ivers LC. Cholera vaccination campaign contributes to improved knowledge regarding cholera and improved practice relevant to waterborne disease in rural haiti. PLoS neglected tropical diseases. 2013;7(11):e2576. doi: 10.1371/journal.pntd.0002576. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Emch M, Ali M, Park JK, Yunus M, Sack DA, Clemens JD. Relationship between neighbourhood-level killed oral cholera vaccine coverage and protective efficacy: evidence for herd immunity. Int J Epidemiol. 2006;35(4):1044–1050. doi: 10.1093/ije/dyl100. [DOI] [PubMed] [Google Scholar]
- 28.Pan American Health Organization. Epidemiologic update: Cholera. 2014 Jun 27; 2014. [Google Scholar]
- 29.Ministère de la Santé et de la Population H. Plan d'elimination du cholera en Haitim 2013–2022 [French] Haiti: Port-au-Prince; 2012. [Google Scholar]
- 30.Sinclair D, Abba K, Zaman K, Qadri F, Graves PM. Oral vaccines for preventing cholera. The Cochrane database of systematic reviews. 2011;(3):CD008603. doi: 10.1002/14651858.CD008603.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Khatib AM, Ali M, von Seidlein L, et al. Effectiveness of an oral cholera vaccine in Zanzibar: findings from a mass vaccination campaign and observational cohort study. The Lancet infectious diseases. 2012;12(11):837–844. doi: 10.1016/S1473-3099(12)70196-2. [DOI] [PubMed] [Google Scholar]
- 32.Clemens JD, Sack DA, Harris JR, et al. Field trial of oral cholera vaccines in Bangladesh: results from three-year follow-up. Lancet. 1990;335(8684):270–273. doi: 10.1016/0140-6736(90)90080-o. [DOI] [PubMed] [Google Scholar]
- 33.Anh DD, Lopez AL, Thiem VD, et al. Use of oral cholera vaccines in an outbreak in Vietnam: a case control study. PLoS neglected tropical diseases. 2011;5(1):e1006. doi: 10.1371/journal.pntd.0001006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Taylor DN, Cardenas V, Sanchez JL, et al. Two-year study of the protective efficacy of the oral whole cell plus recombinant B subunit cholera vaccine in Peru. The Journal of infectious diseases. 2000;181(5):1667–1673. doi: 10.1086/315462. [DOI] [PubMed] [Google Scholar]