On October 2017, Global Task Force on Cholera Control (GTFCC) partners committed to reduce cholera deaths by 90% and to eliminate the disease in 20 countries by 2030, through a multi-sectoral approach (Anon, 2017). This new global strategy planned to combine long-term disease prevention in cholera hotspots with sustainable WaSH (water sanitation and hygiene) solutions and large-scale use of oral cholera vaccine (OCV), with the short-term strengthening of early detection of outbreaks and immediate and effective response through reactive OCV campaigns and rapid response teams (RRTs) (Anon, 2017). RRTs, also referred as mobile teams, have been successfully implemented against polio or Ebola outbreaks (Global Polio Eradication Initiative, 2017; World Health Organization (WHO), 2014). However, response interventions targeted to neighbours of cholera cases (case-area targeted interventions [CATIs]) using combinations of water, sanitation, and hygiene measures, and/or prophylactic antibiotics have rarely been documented, evaluated or promoted against cholera in the published literature (Voelckel, 1971; Piarroux and Bompangue, 2011; Deepthi et al., 2013; Taylor et al., 2015; Mwambi et al., 2016; Finger et al., 2018).

In practice, CATIs are supported by the frequent household transmission of Vibrio cholerae O1 (Weil et al., 2009; Weil et al., 2014; Taylor et al., 2015; Domman et al., 2018), the increased cholera risk among neighbours living within a few dozen meters of cases during the few days following disease onset (Debes et al., 2016; Azman et al., 2018), and the significant protection of household contacts of cholera patients by promoting hand washing with soap and treatment of water (George et al., 2016). A micro-simulation modelling study suggests that early CATIs can be more resource-efficient than mass interventions against cholera (Finger et al., 2018). However, CATI effectiveness has never been evaluated in a real-world setting.

Haiti has implemented CATIs as a national coordinated strategy against cholera since July 2013 (Rebaudet et al., 2019a). After the disease was accidentally imported in October 2010 (Piarroux et al., 2011), the country experienced a massive epidemic, with a total of 820,085 suspected cases and 9792 cholera-related deaths recorded by April 20, 2019 by the Haitian Ministry of Public Health and Population (MSPP) (http://mspp.gouv.ht/newsite/documentation.php, accessed Jul 1, 2019). In 2013, only 68% of Haitian households drank from improved water sources, 26% had access to improved sanitation facilities and 34% had water and soap available for hand washing (République d’Haïti, Ministère de la Santé Publique (MSPP), 2013). But little of the $1.5 billion USD designated by the Plan for the Elimination of Cholera in Haiti 2013–2022 to develop water and sanitation infrastructures has been expended or pledged so far (Ministry of Public Health and Population, National Directorate for Water Supply and Sanitation, 2013). Two pilot OCV reactive campaigns vaccinated approximately 100,000 people in 2012 and to date, additional campaigns have targeted about 10% of the Haitian population (Ivers, 2017; Poncelet, 2015). UNICEF thus backed the MSPP and the Haitian National Directorate for Water and Sanitation (DINEPA) to launch a complementary nationwide coordinated cholera alert-response strategy aiming to interrupt local cholera outbreaks at an early stage (Rebaudet et al., 2019a). This program planned to rapidly send multisectoral rapid response teams to every patient household and neighbourhood in order to identify additional cases, to decontaminate patient premises, to educate on risk factors and methods of prevention and management, to distribute soap and oral rehydration salts (ORS), to chlorinate water at the household level or directly at collection points, and to propose prophylactic antibiotics to close contacts of cholera cases.

This response CATI strategy was implemented gradually from mid-2013 and became an essential pillar of the fight when the national cholera elimination plan was updated in mid-2016 (République d’Haïti, Ministère de la Santé Publique et de la Population, Direction Nationale de l’Eau Potable et de l’Assainissement, 2016). Implementation of this strategy offers a unique opportunity to evaluate the effectiveness of CATIs against cholera outbreaks. Based on available data, we conducted a retrospective observational study over 3 years in the Centre department of Haiti addressing the outcome of local cholera outbreaks according to the response promptness and intensity. We present here the first effectiveness estimates for rapid and targeted response interventions against cholera.

Our quasi-experimental study, based on epidemiological and intervention records over three years in one administrative department of Haiti, showed that prompt and repeated response CATIs conducted by rapid response teams were significantly associated with shortening of cholera outbreaks and mitigating of outbreak case load. Of note, numerous suspected cholera outbreaks spontaneously ended before any response could be conducted. But when taking into account this significant confounding by indication (Remschmidt et al., 2015), the prompter the first complete CATI was implemented and the more complete CATIs were conducted, the fewer cases were recorded and the shorter the outbreak lasted.

While many mild suspected outbreaks may spontaneously end without any response intervention, prompt and repeated CATIs appear difficult to sustain during the largest outbreaks. As suggested by the slow increase in the number of CATIs observed during the study period, CATIs may be logistically complex to implement, and response teams can become overwhelmed when they try to simultaneously address a large number of cases (Finger et al., 2018; Rebaudet et al., 2019a). Such strategy certainly is most relevant at the beginning of epidemics, or during trough periods or tails of epidemics (Finger et al., 2018; Rebaudet et al., 2013).

Our study comes with a number of limitations. Because CATIs were not randomized, response effectiveness may have been biased by unmeasured confounders. As we observed a significant confounding by indication on the probability for an outbreak to receive a CATI response, we limited our analysis on outbreaks that were responded to. We subsequently did not observe any consistent residual difference of initial severity between classes of response promptness and response intensity. However, our models were adjusted for potential confounders and took into account the heterogeneity between localities. This quasi-experimental study was also stratified on response promptness and on response intensity, which yielded consistent response effectiveness estimates (Shadish et al., 2002).

Analyses may also have been biased by missing epidemiological data. Indeed, some patients do not seek care, even when they experience severe dehydration. Besides, stool sampling for confirmation culture was not systematic, which certainly led us to overlook several authentic outbreaks and mis-select clusters of non-cholera diarrhoeas. It may have led us to misdate several outbreak onset and outbreak end. Depending on the differential distribution of these potential biases among classes of response promptness and intensity, these limits could have led to over- or under-estimation of the effectiveness of prompt and intense CATIs. Nevertheless, our outbreak definition aimed to deal with those missing data and be specific in order to analyse CATI effectiveness on definite outbreaks. Like for many diseases, no standardized cholera outbreak criteria exists, and several definitions may be more or less suitable depending on interventions and analyses objectives (Brady et al., 2015). Our retrospectively defined outbreaks may be an approximate unit of analysis in terms of space, time and population, which may also have biased effectiveness results. We therefore conducted a sensitivity analysis using alternative definitions, including systematically lab-confirmed cholera outbreaks, which showed consistent and robust estimates (Appendix 3.1). We also used mixed models in order to take into account heterogeneity between localities in the random effect (Berridge and Crouchley, 2011). Additional CATI effectiveness studies at the household and at the administrative commune levels are underway in Haiti.

Our study analysed 3887 CATIs prospectively notified by rapid response teams to UNICEF. But some additional CATIs may have been omitted, while other CATIs remained unrecorded because they were implemented by the EMIRA alone. Nevertheless, many of their respective CATIs actually overlapped, and we thus believe our response database to be reasonably exhaustive. Conversely, only 16% of complete CATIs were conducted in a locality experiencing a current outbreak. The remaining CATIs were implemented in response to sporadic cases that did not meet outbreak definition criteria, as illustrated by much higher rates with less stringent outbreak definitions (Appendix 3.1). Sporadic CATIs may have prevented, delayed or attenuated the emergence of outbreaks. They may also be associated with the propensity of future outbreak response. We thus included the frequency of previous complete CATIs in our analysis but found no significant association with response promptness or intensity.

Our study aimed to assess the overall effectiveness of a CATI strategy. It neither aimed to estimate the respective effectiveness of each response components, nor the optimal radius of intervention, which would warrant dedicated field studies comparing different types of interventions. We thus chose a conservative definition of complete CATIs and performed a sensitivity analysis with alternative CATI definitions that exhibited consistent results (Appendix 3.2). Because nearly all CATIs included house decontamination, education and chlorine distribution, stratified analyses on these activities were not possible. However, three effectiveness estimates out of four appeared higher when all CATIs included antibiotic prophylaxis than when no CATI did. Several trials have also suggested that chemoprophylaxis has a protective effect among household contacts of people with cholera (Reveiz et al., 2011), and a micro-simulation model suggested that administration of antibiotics in CATIs could effectively avert secondary cases (Finger et al., 2018). But considering the risk of resistance selection (Mhalu et al., 1979; Dromigny et al., 2002), the selected distribution of antibiotic prophylaxis to close contacts is usually not recommended (Global Task Force on Cholera Control (GTFCC), 2018) and must, at the minimum, be used with caution and close monitoring of antibiotic susceptibility. In Haiti, all clinical V. cholerae O1 isolates have remained susceptible to doxycycline between 2013 and 2019 (Haitian Ministry of Public Health and Population, MSPP). As suggested by previous field or modelling studies (Ali et al., 2016; Parker et al., 2017b; Parker et al., 2017a; Finger et al., 2018), adding the administration of a single-dose OCV during CATIs could be an effective, but likely logistically complex, strategy.

Overall, our results suggest that case-area targeted interventions are significantly effective to mitigate and shorten local cholera outbreaks. Household water treatment, sanitation and hygiene promotion, as well as antibiotic prophylaxis theoretically prevent both human-to-human and environment-to-human cholera transmission pathways. Regardless of their respective role, which has been much debated (Morris, 2011; Kupferschmidt, 2017; Rebaudet et al., 2019b), our results thus confirm the relevance of promoting rapid response teams as a key component of the new global strategy for cholera control (Global Task Force on Cholera Control (GTFCC), 2017; Anon, 2017). Such findings need to be replicated in other settings and at other spatial and time scales. It will be critical to understand where CATIs should be prioritized, which radius is optimal, and which intervention components are most effective.

Data generated or analysed during this study are included in the manuscript and supporting files. Source data files have been provided for Figures 1 and 4.

1. 1. Ali M
   2. Debes AK
   3. Luquero FJ
   4. Kim DR
   5. Park JY
   6. Digilio L
   7. Manna B
   8. Kanungo S
   9. Dutta S
   10. Sur D
   11. Bhattacharya SK
   12. Sack DA

   (2016) Potential for controlling cholera using a ring vaccination strategy: re-analysis of data from a Cluster-Randomized clinical trial

   PLOS Medicine 13:e1002120.

   https://doi.org/10.1371/journal.pmed.1002120

   • PubMed
   • Google Scholar
2. 1. Andersen PK
   2. Gill RD

   (1982) Cox's Regression Model for Counting Processes: A Large Sample Study

   The Annals of Statistics 10:1100–1120.

   https://doi.org/10.1214/aos/1176345976

   • Google Scholar
3. 1. Anon

   (2017) Cholera: ending a 50-year pandemic

   The Lancet 390:1623.

   https://doi.org/10.1016/S0140-6736(17)32592-8

   • Google Scholar
4. 1. Azman AS
   2. Rudolph KE
   3. Cummings DA
   4. Lessler J

   (2013) The incubation period of cholera: a systematic review

   Journal of Infection 66:432–438.

   https://doi.org/10.1016/j.jinf.2012.11.013

   • PubMed
   • Google Scholar
5. 1. Azman AS
   2. Luquero FJ
   3. Salje H
   4. Mbaïbardoum NN
   5. Adalbert N
   6. Ali M
   7. Bertuzzo E
   8. Finger F
   9. Toure B
   10. Massing LA
   11. Ramazani R
   12. Saga B
   13. Allan M
   14. Olson D
   15. Leglise J
   16. Porten K
   17. Lessler J

   (2018) Micro-Hotspots of risk in urban cholera epidemics

   The Journal of Infectious Diseases 218:1164–1168.

   https://doi.org/10.1093/infdis/jiy283

   • Google Scholar
6. Book

   1. Berridge DM
   2. Crouchley R

   (2011) Multivariate Generalized Linear Mixed Models Using Redition

   CRC Press.

   https://doi.org/10.1201/b10850

   • Google Scholar
7. 1. Brady OJ
   2. Smith DL
   3. Scott TW
   4. Hay SI

   (2015) Dengue disease outbreak definitions are implicitly variable

   Epidemics 11:92–102.

   https://doi.org/10.1016/j.epidem.2015.03.002

   • PubMed
   • Google Scholar
8. Report

   1. Centers for Disease Control and Prevention (CDC)

   (1999) Laboratory Methods for the Diagnosis of Epidemic Dysentery and Cholera

   National Center for Infectious Diseases.

   http://www.cdc.gov/cholera/pdf/Laboratory-Methods-for-the-Diagnosis-of-Epidemic-Dysentery-and-Cholera.pdf

   • Google Scholar
9. 1. Debes AK
   2. Ali M
   3. Azman AS
   4. Yunus M
   5. Sack DA

   (2016) Cholera cases cluster in time and space in matlab, Bangladesh: implications for targeted preventive interventions

   International Journal of Epidemiology 38:2134–2139.

   https://doi.org/10.1093/ije/dyw267

   • Google Scholar
10. 1. Deepthi R
    2. Sandeep SR
    3. Rajini M
    4. Rajeshwari H
    5. Shetty A

    (2013) Cholera outbreak in a village in south India - Timely action saved lives

    Journal of Infection and Public Health 6:35–40.

    https://doi.org/10.1016/j.jiph.2012.05.003

    • PubMed
    • Google Scholar
11. 1. Domman D
    2. Chowdhury F
    3. Khan AI
    4. Dorman MJ
    5. Mutreja A
    6. Uddin MI
    7. Paul A
    8. Begum YA
    9. Charles RC
    10. Calderwood SB
    11. Bhuiyan TR
    12. Harris JB
    13. LaRocque RC
    14. Ryan ET
    15. Qadri F
    16. Thomson NR

    (2018) Defining endemic cholera at three levels of spatiotemporal resolution within Bangladesh

    Nature Genetics 50:951–955.

    https://doi.org/10.1038/s41588-018-0150-8

    • Google Scholar
12. 1. Dromigny JA
    2. Rakoto-Alson O
    3. Rajaonatahina D
    4. Migliani R
    5. Ranjalahy J
    6. Mauclére P

    (2002) Emergence and rapid spread of tetracycline-resistant Vibrio cholerae strains, Madagascar

    Emerging Infectious Diseases 8:336–338.

    https://doi.org/10.3201/eid0803.010258

    • PubMed
    • Google Scholar
13. 1. Eisenberg MC
    2. Kujbida G
    3. Tuite AR
    4. Fisman DN
    5. Tien JH

    (2013) Examining rainfall and cholera dynamics in Haiti using statistical and dynamic modeling approaches

    Epidemics 5:197–207.

    https://doi.org/10.1016/j.epidem.2013.09.004

    • PubMed
    • Google Scholar
14. 1. Finger F
    2. Bertuzzo E
    3. Luquero FJ
    4. Naibei N
    5. Touré B
    6. Allan M
    7. Porten K
    8. Lessler J
    9. Rinaldo A
    10. Azman AS

    (2018) The potential impact of case-area targeted interventions in response to cholera outbreaks: a modeling study

    PLOS Medicine 15:e1002509.

    https://doi.org/10.1371/journal.pmed.1002509

    • PubMed
    • Google Scholar
15. 1. Gaudart J
    2. Rebaudet S
    3. Barrais R
    4. Boncy J
    5. Faucher B
    6. Piarroux M
    7. Magloire R
    8. Thimothe G
    9. Piarroux R

    (2013) Spatio-temporal dynamics of cholera during the first year of the epidemic in Haiti

    PLOS Neglected Tropical Diseases 7:e2145.

    https://doi.org/10.1371/journal.pntd.0002145

    • PubMed
    • Google Scholar
16. 1. George CM
    2. Monira S
    3. Sack DA
    4. Rashid MU
    5. Saif-Ur-Rahman KM
    6. Mahmud T
    7. Rahman Z
    8. Mustafiz M
    9. Bhuyian SI
    10. Winch PJ
    11. Leontsini E
    12. Perin J
    13. Begum F
    14. Zohura F
    15. Biswas S
    16. Parvin T
    17. Zhang X
    18. Jung D
    19. Sack RB
    20. Alam M

    (2016) Randomized controlled trial of Hospital-Based hygiene and water treatment intervention (CHoBI7) to reduce cholera

    Emerging Infectious Diseases 22:233–241.

    https://doi.org/10.3201/eid2202.151175

    • PubMed
    • Google Scholar
17. Report

    1. Global Polio Eradication Initiative

    (2017) Stantard Operating Procedures. Responding to a Poliovirus Event or Outbreak. Part 1: General SOPs

    Geneva, Switzerland: World Health Organization (WHO).

    http://polioeradication.org/wp-content/uploads/2018/01/pol-sop-responding-polio-event-outbreak-part1-20180117.pdf

    • Google Scholar
18. Report

    1. Global Task Force on Cholera Control (GTFCC)

    (2017) Ending Cholera. a Global Roadmap to 2030

    World Health Organization (WHO).

    https://www.who.int/cholera/publications/global-roadmap.pdf

    • Google Scholar
19. Report

    1. Global Task Force on Cholera Control (GTFCC)

    (2018) Technical Note - Use of Antibiotics for the Treatment and Control of Cholera.

    World Health Organization (WHO).

    https://www.who.int/cholera/task_force/use-of-antibiotics-for-the-treatment-of-cholera.pdf?ua=1

    • Google Scholar
20. Report

    1. Institut Haitien de Statistique et d’Informatique (IHSI)

    (2015) Population Totale, Population De 18 Ans Et Plus, Ménages Et Densités Estimés En 2015

    Direction Des Statistiques Démographiques Et Sociales, Ministère De l’Economie Et Des Finances, République d’Haïti.

    http://ihsi.ht/pdf/projection/Estimat_PopTotal_18ans_Menag2015.pdf

    • Google Scholar
21. 1. Ivers LC

    (2017) Eliminating cholera transmission in Haiti

    New England Journal of Medicine 376:101–103.

    https://doi.org/10.1056/NEJMp1614104

    • PubMed
    • Google Scholar
22. 1. Kupferschmidt K

    (2017) Genomes rewrite cholera's global story

    Science 358:706–707.

    https://doi.org/10.1126/science.358.6364.706

    • PubMed
    • Google Scholar
23. 1. Mhalu FS
    2. Mmari PW
    3. Ijumba J

    (1979) Rapid emergence of El Tor Vibrio cholerae resistant to antimicrobial agents during first six months of fourth cholera epidemic in Tanzania

    Lancet 1:345–347.

    https://doi.org/10.1016/s0140-6736(79)92889-7

    • PubMed
    • Google Scholar
24. 1. Mickey RM
    2. Greenland S

    (1989) The impact of confounder selection criteria on effect estimation

    American Journal of Epidemiology 129:125–137.

    https://doi.org/10.1093/oxfordjournals.aje.a115101

    • Google Scholar
25. Report

    1. Ministry of Public Health and Population, National Directorate for Water Supply and Sanitation

    (2013) National Plan for the Elimination of Cholera in Haiti 2013-2022

    Port-au-Prince, Haiti: Ministry of Public Health and Population.

    http://www.paho.org/hq/index.php?option=com_docman&task=doc_view&gid=20326&Itemid=270&lang=en

    • Google Scholar
26. 1. Morris JG

    (2011) Cholera—Modern Pandemic Disease of Ancient Lineage

    Emerging Infectious Diseases 17:2099–2104.

    https://doi.org/10.3201/eid1711.111109

    • Google Scholar
27. 1. Mwambi P
    2. Mufunda J
    3. Lupili M
    4. Bangwe K
    5. Bwalya F
    6. Mazaba ML

    (2016)

    Timely response and containment of 2016 cholera outbreak in northern Zambia

    Medical Journal of Zambia 43:64–69.

    • Google Scholar
28. 1. Parker LA
    2. Rumunu J
    3. Jamet C
    4. Kenyi Y
    5. Lino RL
    6. Wamala JF
    7. Mpairwe AM
    8. Ciglenecki I
    9. Luquero FJ
    10. Azman AS
    11. Cabrol JC

    (2017a) Adapting to the global shortage of cholera vaccines: targeted single dose cholera vaccine in response to an outbreak in south Sudan

    The Lancet Infectious Diseases 17:e123–e127.

    https://doi.org/10.1016/S1473-3099(16)30472-8

    • PubMed
    • Google Scholar
29. 1. Parker LA
    2. Rumunu J
    3. Jamet C
    4. Kenyi Y
    5. Lino RL
    6. Wamala JF
    7. Mpairwe AM
    8. Muller V
    9. Llosa AE
    10. Uzzeni F
    11. Luquero FJ
    12. Ciglenecki I
    13. Azman AS

    (2017b) Neighborhood-targeted and case-triggered use of a single dose of oral cholera vaccine in an urban setting: feasibility and vaccine coverage

    PLOS Neglected Tropical Diseases 11:e0005652.

    https://doi.org/10.1371/journal.pntd.0005652

    • PubMed
    • Google Scholar
30. 1. Piarroux R
    2. Barrais R
    3. Faucher B
    4. Haus R
    5. Piarroux M
    6. Gaudart J
    7. Magloire R
    8. Raoult D

    (2011) Understanding the cholera epidemic, Haiti

    Emerging Infectious Diseases 17:1161–1168.

    https://doi.org/10.3201/eid1707.110059

    • PubMed
    • Google Scholar
31. 1. Piarroux R
    2. Bompangue D

    (2011)

    Encyclopedia of Infectious Diseases : Modern Methodologies

    639–653, Needs for an integrative approach of epidemics: The Example of Cholera, Encyclopedia of Infectious Diseases : Modern Methodologies, Wiley & Sons, Inc.

    • Google Scholar
32. 1. Poncelet JL

    (2015)

    Cholera in Haiti: successes and challenges in a low income country

    Wkly Epidemiol Rec 90:542–544.

    • PubMed
    • Google Scholar
33. 1. Rebaudet S
    2. Gazin P
    3. Barrais R
    4. Moore S
    5. Rossignol E
    6. Barthelemy N
    7. Gaudart J
    8. Boncy J
    9. Magloire R
    10. Piarroux R

    (2013) The dry season in Haiti: a window of opportunity to eliminate cholera

    PLOS Currents 5:.

    https://doi.org/10.1371/currents.outbreaks.2193a0ec4401d9526203af12e5024ddc

    • PubMed
    • Google Scholar
34. 1. Rebaudet S
    2. Bulit G
    3. Gaudart J
    4. Michel E
    5. Gazin P
    6. Evers C
    7. Beaulieu S
    8. Abedi AA
    9. Osei L
    10. Barrais R
    11. Pierre K
    12. Moore S
    13. Boncy J
    14. Adrien P
    15. Duperval Guillaume F
    16. Beigbeder E
    17. Piarroux R

    (2019a) The case-area targeted rapid response strategy to control cholera in Haiti: a four-year implementation study

    PLOS Neglected Tropical Diseases 13:e0007263.

    https://doi.org/10.1371/journal.pntd.0007263

    • PubMed
    • Google Scholar
35. 1. Rebaudet S
    2. Moore S
    3. Rossignol E
    4. Bogreau H
    5. Gaudart J
    6. Normand AC
    7. Laraque MJ
    8. Adrien P
    9. Boncy J
    10. Piarroux R

    (2019b) Epidemiological and molecular forensics of cholera recurrence in Haiti

    Scientific Reports 9:1164.

    https://doi.org/10.1038/s41598-018-37706-0

    • PubMed
    • Google Scholar
36. 1. Remschmidt C
    2. Wichmann O
    3. Harder T

    (2015) Frequency and impact of confounding by indication and healthy vaccinee Bias in observational studies assessing influenza vaccine effectiveness: a systematic review

    BMC Infectious Diseases 15:429.

    https://doi.org/10.1186/s12879-015-1154-y

    • PubMed
    • Google Scholar
37. Report

    1. République d’Haïti, Ministère de la Santé Publique (MSPP)

    (2013) Haïti. Enquête Mortalité, Morbidité Et Utilisation Des Services (EMMUS-V)

    Pétionville, Haïti: Institut Haïtien de l’Enfance.

    http://mspp.gouv.ht/site/downloads/Rapport%20preliminaire%20final%20EMMUS-V.pdf

    • Google Scholar
38. Report

    1. République d’Haïti, Ministère de la Santé Publique et de la Population, Direction Nationale de l’Eau Potable et de l’Assainissement

    (2016)

    Plan National D’élimination Du Choléra. Développement Du Moyen Terme. Juillet 2016 - Décembre 2018

    République d’Haïti.

    • Google Scholar
39. 1. Reveiz L
    2. Chapman E
    3. Ramon-Pardo P
    4. Koehlmoos TP
    5. Cuervo LG
    6. Aldighieri S
    7. Chambliss A

    (2011) Chemoprophylaxis in contacts of patients with cholera: systematic review and meta-analysis

    PLOS ONE 6:e27060.

    https://doi.org/10.1371/journal.pone.0027060

    • PubMed
    • Google Scholar
40. 1. Shadish WR
    2. Cook TD
    3. Campbell DT

    (2002)

    Experimental and Quasi-Experimental Designsdesigns for Generalizedgeneralized Causalcausal Inferenceinference

    103–134, Quasi-Experimental designs that either lack a control group or lack pretest observations on the outcome, Experimental and Quasi-Experimental Designsdesigns for Generalizedgeneralized Causalcausal Inferenceinference, Houghton Mifflin Company.

    • Google Scholar
41. 1. Sorichetta A
    2. Hornby GM
    3. Stevens FR
    4. Gaughan AE
    5. Linard C
    6. Tatem AJ

    (2015) High-resolution gridded population datasets for Latin America and the Caribbean in 2010, 2015, and 2020

    Scientific Data 2:e150045.

    https://doi.org/10.1038/sdata.2015.45

    • Google Scholar
42. 1. Taylor DL
    2. Kahawita TM
    3. Cairncross S
    4. Ensink JH

    (2015) The impact of water, sanitation and hygiene interventions to control cholera: a systematic review

    PLOS ONE 10:e0135676.

    https://doi.org/10.1371/journal.pone.0135676

    • PubMed
    • Google Scholar
43. 1. Voelckel J

    (1971)

    Assainissement et lutte anticholérique

    Med Trop Mars 31:133–144.

    • Google Scholar
44. 1. Weil AA
    2. Khan AI
    3. Chowdhury F
    4. Larocque RC
    5. Faruque AS
    6. Ryan ET
    7. Calderwood SB
    8. Qadri F
    9. Harris JB

    (2009) Clinical outcomes in household contacts of patients with cholera in Bangladesh

    Clinical Infectious Diseases 49:1473–1479.

    https://doi.org/10.1086/644779

    • PubMed
    • Google Scholar
45. 1. Weil AA
    2. Begum Y
    3. Chowdhury F
    4. Khan AI
    5. Leung DT
    6. LaRocque RC
    7. Charles RC
    8. Ryan ET
    9. Calderwood SB
    10. Qadri F
    11. Harris JB

    (2014) Bacterial shedding in household contacts of cholera patients in Dhaka, Bangladesh

    The American Journal of Tropical Medicine and Hygiene 91:738–742.

    https://doi.org/10.4269/ajtmh.14-0095

    • PubMed
    • Google Scholar
46. 1. Weiss DJ
    2. Nelson A
    3. Gibson HS
    4. Temperley W
    5. Peedell S
    6. Lieber A
    7. Hancher M
    8. Poyart E
    9. Belchior S
    10. Fullman N
    11. Mappin B
    12. Dalrymple U
    13. Rozier J
    14. Lucas TCD
    15. Howes RE
    16. Tusting LS
    17. Kang SY
    18. Cameron E
    19. Bisanzio D
    20. Battle KE
    21. Bhatt S
    22. Gething PW

    (2018) A global map of travel time to cities to assess inequalities in accessibility in 2015

    Nature 553:333–336.

    https://doi.org/10.1038/nature25181

    • PubMed
    • Google Scholar
47. Report

    1. World Health Organization (WHO)

    (2014) Ebola and Marburg Virus Disease Epidemics: Preparedness, Alert, Control, and Evaluation

    Geneva, Switzerland: World Health Organization.

    http://www.who.int/csr/resources/publications/ebola/manual_EVD/en/

    • Google Scholar

Acceptance summary:

This manuscript addresses an important and timely question around the effectiveness of case-area targeted interventions (CATIs) against cholera in Haiti. The cholera epidemic in Haiti has been swift and devastating. It has led to substantial scientific study of the spatio-temporal dynamics of the disease, but also to fundamental debates on transmission routes and implications for best approaches to control and possibly elimination. The detailed monitoring of CATIs in the Centre Department provides an opportunity for real, on-the-ground, assessment of this strategy. With a set of robust statistical analyses, the manuscript presents evidence for the effectiveness and importance of CATIs. It finds that a prompter response (measured in categories of days since the first case was reported) and a more intense response (measured in the number of CATIs per week and/or case of the outbreak) were effective at reducing the number of cases and duration of cholera outbreaks.

Decision letter after peer review:

Thank you for submitting your article "Estimating effectiveness of case-area targeted response interventions against cholera in Haiti" for consideration by eLife. Your article has been reviewed by three peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Neil Ferguson as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Virginia E. Pitzer.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Overall, we feel that the results are novel and important for justifying the continued (and expanded) use of CATIs in response to cholera. The statistical analysis was also felt to be through and well handled. However, we have concerns about the methodological approach, interpretation of findings and presentation of results.

Essential revisions:

1) Our major concern was the choice of CATIs to include. For instance, according to the manuscript, only 15% of the CATI interventions were done in response to an "outbreak" detected (a-posteriori). Thus it seems like CATI that weren't implemented within an outbreak were excluded from the analysis but it was maybe just the effect of those CATI that kept the number of cases small enough to stay below the authors definition of an "outbreak". This potentially adds important bias to the reported results. While this limitation might be inherent in a retrospective study, the manuscript needs to acknowledge this limitation more clearly. In particular, you should discuss how these excluded CATIs may affect the results.

2) If the data allow to differentiate between different types of CATI (for instance antibiotics) then an additional stratified analysis of CATIs with or without prophylactic antibiotics should be included. For instance, the findings of the study would be considerably strengthened by including an additional analysis of the association between the promptness and intensity of the response and the outbreak size and duration stratified by whether or not the CATIs included the distribution of prophylactic antibiotics to household members of identified cases. Based on the results presented in Appendix 3—table 4, it would seem that such an analysis would be feasible and straightforward to perform, and would have reasonable power to detect a potential difference in the effect estimates. Given the use of antibiotic prophylaxis to prevent cholera is somewhat controversial compared to the other components of the CATIs, it would be very interesting and potential impactful to have an estimate of its effectiveness from a real-world implementation setting.

3) More details on the CATIs would also be useful, for example: Do you know how many households or people have been targeted with exactly what intervention? Or in what geographical area around cases the interventions have been applied? The same applies to case data (you mention that some confirmed cases are not mentioned in the line list, and that many lab results are negative for V. cholerae).

4) Regarding the analysis of confounders and subsequent control for confounding in the adjusted models of intervention effectiveness: it's not clear why you went to all the trouble to analyse which covariates may be potential confounders of the relationship between response promptness (Table 1) and response intensity (Table 2) and outcomes of interest (outbreak size and duration), but then subsequently included all of the potential covariates in the adjusted models. By definition, any covariate that is not associated with both the exposure and outcome of interest is not a potential confounder. Furthermore, including covariates that are downstream of other confounders has the potential to inadvertently introduce bias into an analysis (see Hernan et al., Epidemiology, 2004). The sensitivity analysis presented in Appendix 3—table 5 shows that the results are very similar when only adjusting for the covariates that are significantly associated with the exposure and/or outcome. Nevertheless, these should be presented in the main text, and the model including all potential covariates could be included as a sensitivity analysis.

5) The presentation of results in the main text would be clearer and more concise if you focused the analysis of the association between the intensity of the response and the outbreak size and duration on only one of the exposure-outcome combinations. As is, you present the results for the association between response intensity as measured by the number of CATIs per week and the outcome as measured by the cumulative number of cholera cases in the outbreak in Table 5, and the association between response intensity as measured by the number of CATIs per case and the outcome as measured by the outbreak duration in Table 6. Why were these exposure-outcome combinations chosen and not the other two alternatives (e.g. the association between the number of CATIs per week and outbreak duration, and the association between the number of CATIs per case and outbreak size)? In general, the results were quite similar (although there was little difference between the effectiveness of having 0.25 to <0.5 CATIs per case and the baseline comparator of >0 to <0.25 CATIs per case). Therefore, just one of the results should be presented in the main text, and the other 3 exposure-outcome combinations presented in the Appendix, with only the qualitative differences described in the main text. Additionally, the notation used in the tables and figures for these categories is quite confusing and unnecessary.

6) Another concern is in regard to the interpretation of the potential impact of the under-diagnosis of cases and misclassification of non-cholera diarrhoea cases in paragraph four of the Discussion. Here, it is stated that both a high number of unrecorded cases and the misclassification of outbreaks due to inconsistent culture confirmation would lead to an underestimation of the effectiveness of CATIs. However, we do not believe there is any reason to conclude that the direction of the bias would be to underestimate the effect of the intervention in either instance. In order for the first statement to be true, cases would have to be differentially underdiagnosed more often when the CATIs are implemented earlier compared to when they were introduced >7 days after the outbreak started, or else cases would have to be underdiagnosed more often when the control efforts were more intense. Neither situation seems likely. While misclassification of non-cholera diarrhoea outbreaks could potentially attenuate the effectiveness estimate for the interventions (as suggested by the alternative outbreak definition "Outbreak Cu1" examined in Appendix 3—table 2), this was not true for most of the other outbreak definitions, and again is unlikely to be true more generally. At the very least, more justification should be provided for these statements.

7) The effectiveness of CATIs has implications well beyond the specific control question at hand, on the still poorly agreed importance of routes of transmission in cholera. In turn, these routes and in particular the role of an environmental reservoir continue to be the subject of substantial debate, and also influence views on approaches to control and sustain the gains of control in the long run. We understand that the paper is not meant to resolve these matters but it would be valuable to add some discussion of related implications even if briefly. These will broaden the scientific value of the findings.

8) There is concern about the choice of "outbreaks", which the authors define post-hoc based on reported incidence, as the unit of analysis. Some discussion would be useful of the implications if the unit of analysis is either an individual or a household, targeted with interventions or not.

9) Outbreaks that haven't been responded to by a targeted intervention have been excluded from the analysis because the authors have detected major bias ("confounding by indication"), e.g. larger outbreaks are more likely to be targeted by > 1 intervention. Doesn't the same bias exist between outbreaks that have been responded to more or less promptly or with a higher or lower intensity?

Essential revisions:

1) Our major concern was the choice of CATIs to include. For instance, according to the manuscript, only 15% of the CATI interventions were done in response to an "outbreak" detected (a-posteriori). Thus it seems like CATI that weren't implemented within an outbreak were excluded from the analysis but it was maybe just the effect of those CATI that kept the number of cases small enough to stay below the authors definition of an "outbreak". This potentially adds important bias to the reported results. While this limitation might be inherent in a retrospective study, the manuscript needs to acknowledge this limitation more clearly. In particular, you should discuss how these excluded CATIs may affect the results.

We thank the reviewers for raising the concern of CATIs conducted outside identified outbreaks.

To clarify this point, we included the number and proportion of CATIs during outbreaks in Appendix 3—table 1, in Appendix 3—table 3 and in Appendix 3—table 4.

CATIs conducted outside identified outbreaks were implemented in response to sporadic cases that did not meet outbreak definition criteria, as illustrated by the higher proportions of CATIs within outbreaks found when using less stringent outbreak definitions (up to 40% in scenario 2, Appendix 3—table 2).

However, effectiveness sensitivity analyses using alternative outbreak definitions that included higher proportions of CATIs, did not exhibit substantially different results (Appendix 3—table 2).

Sporadic CATIs may theoretically have prevented, delayed or attenuated the emergence of certain outbreaks. They may also be associated with the propensity of future outbreak response. In our revised effectiveness analyses, we thus included the frequency of previous complete CATIs conducted between the beginning of the study period and the outbreak onset as a new covariate (see Table 1 and 2). Because previous CATIs may be dependent of the previous number of cases to be responded to, we also included the previous incidence of suspected cholera cases recorded between the beginning of the study period and the outbreak onset as another new covariate. However, we found no significant association between these covariates and response promptness or intensity. We now address this concern in the revised Discussion section.

Finally, sporadic CATIs may prevent the occurrence of future outbreaks, or attenuate outbreaks that may thus end before being responded. Therefore, the real impact of CATIs against cholera may in fact be broader than suggested by our analyses, which specifically assessed the effectiveness of prompt and intense CATIs to reduce the size and duration of outbreaks that received a response.

2) If the data allow to differentiate between different types of CATI (for instance antibiotics) then an additional stratified analysis of CATIs with or without prophylactic antibiotics should be included. For instance, the findings of the study would be considerably strengthened by including an additional analysis of the association between the promptness and intensity of the response and the outbreak size and duration stratified by whether or not the CATIs included the distribution of prophylactic antibiotics to household members of identified cases. Based on the results presented in Appendix 3—table 4, it would seem that such an analysis would be feasible and straightforward to perform, and would have reasonable power to detect a potential difference in the effect estimates. Given the use of antibiotic prophylaxis to prevent cholera is somewhat controversial compared to the other components of the CATIs, it would be very interesting and potential impactful to have an estimate of its effectiveness from a real-world implementation setting.

Our study initially did not aim at evaluating the respective effectiveness of each activity included in CATIs. Besides, stratified analyses on house decontamination, education and chlorine distribution was not possible as nearly all CATIs included these activities.

However, we thank reviewers for suggesting us to include stratified analyses on antibiotic prophylaxis, which are now described in a separate part in the Results section. We conducted additional comparisons of outbreak size or outbreak duration according to the response promptness or to the response intensity, stratified by whether all complete CATIs or none of complete CATIs conducted within outbreaks included antibiotic prophylaxis. We found that most of effectiveness estimates appeared higher when all CATIs included antibiotic prophylaxis.

We also discuss these results, including the risk of bacterial resistance selection when using antibiotics as selected chemoprophylaxis.

Nevertheless, we would like to point out that the methodology used in this study is probably not the most appropriate to accurately measure the impact of an antibiotic prophylaxis delivered during CATIs. Other approaches, such as randomization of antibiotic distribution during CATIs, would help to better understand the additional impact of this measure. The circumstances in which we conducted our study did not allow such randomization. This is why we insisted at the end of the Discussion on the interest of conducting additional research to better define the optimal modalities for the implementation of CATIs.

3) More details on the CATIs would also be useful, for example: Do you know how many households or people have been targeted with exactly what intervention? Or in what geographical area around cases the interventions have been applied? The same applies to case data (you mention that some confirmed cases are not mentioned in the line list, and that many lab results are negative for V. cholerae).

Following the reviewers’ request, we added a supplementary Table describing baseline characteristics of suspected cholera cases, cholera stool cultures and CATIs (Appendix 1—table 1), including the number of households targeted by decontamination or chlorine distribution and the number of people who were educated and who received antibiotic chemoprophylaxis.

4) Regarding the analysis of confounders and subsequent control for confounding in the adjusted models of intervention effectiveness: it's not clear why you went to all the trouble to analyse which covariates may be potential confounders of the relationship between response promptness (Table 1) and response intensity (Table 2) and outcomes of interest (outbreak size and duration), but then subsequently included all of the potential covariates in the adjusted models. By definition, any covariate that is not associated with both the exposure and outcome of interest is not a potential confounder. Furthermore, including covariates that are downstream of other confounders has the potential to inadvertently introduce bias into an analysis (see Hernan et al., Epidemiology, 2004). The sensitivity analysis presented in Appendix 3—table 5 shows that the results are very similar when only adjusting for the covariates that are significantly associated with the exposure and/or outcome. Nevertheless, these should be presented in the main text, and the model including all potential covariates could be included as a sensitivity analysis.

We agree with the reviewers’ suggestion and therefore reintegrated adjusted estimates for covariates for which P-values were less than 0.25 at the initial univariate step in the main manuscript. We now only present the model including all potential covariates as a sensitivity analysis (Appendix 3—table 5). We also kept models minimizing the Akaike information criterion (AIC) as a sensitivity analysis (Appendix 3—table 5). Results between the three model groups are indeed very similar.

We also totally agree that including covariates that are downstream of other confounders has the potential to inadvertently introduce bias into an analysis. We believe that the four final models presented in the main text do not present such bias.

5) The presentation of results in the main text would be clearer and more concise if you focused the analysis of the association between the intensity of the response and the outbreak size and duration on only one of the exposure-outcome combinations. As is, you present the results for the association between response intensity as measured by the number of CATIs per week and the outcome as measured by the cumulative number of cholera cases in the outbreak in Table 5, and the association between response intensity as measured by the number of CATIs per case and the outcome as measured by the outbreak duration in Table 6. Why were these exposure-outcome combinations chosen and not the other two alternatives (e.g. the association between the number of CATIs per week and outbreak duration, and the association between the number of CATIs per case and outbreak size)? In general, the results were quite similar (although there was little difference between the effectiveness of having 0.25 to <0.5 CATIs per case and the baseline comparator of >0 to <0.25 CATIs per case). Therefore, just one of the results should be presented in the main text, and the other 3 exposure-outcome combinations presented in the Appendix, with only the qualitative differences described in the main text. Additionally, the notation used in the tables and figures for these categories is quite confusing and unnecessary.

We thank the reviewers for these comments and better explained in the Materials and method section why we used two different proxies of response intensity when comparing outbreak size or outbreak duration. Indeed, we had to avoid that cases or duration be included both within outcome and exposure variables: we could neither compare the number of cases accumulated from the 4th day of outbreak between classes of CATIs per case, nor compare the duration of outbreaks between classes of CATIs per week. We thus approximated response intensity by the number of complete CATIs per week ratio when comparing the number of cases accumulated from the 4th day of outbreak (CE₃), and by the number of complete CATIs per case ratio when comparing the duration of outbreak (CE₄).

Considering that we independently evaluated CATI effectiveness by comparing outbreak outcome defined by two different variables (outbreak size and outbreak duration), between classes of two exposure variables (response promptness and response intensity), we believed important to keep all four groups of results within the main manuscript. The fact that all effectiveness estimates were of the same order of magnitude suggests that our results are consistent.

Finally, we clarified the notation for classes of response intensity used in the tables and figures, as suggested by the reviewers.

6) Another concern is in regard to the interpretation of the potential impact of the under-diagnosis of cases and misclassification of non-cholera diarrhoea cases in paragraph four of the Discussion. Here, it is stated that both a high number of unrecorded cases and the misclassification of outbreaks due to inconsistent culture confirmation would lead to an underestimation of the effectiveness of CATIs. However, we do not believe there is any reason to conclude that the direction of the bias would be to underestimate the effect of the intervention in either instance. In order for the first statement to be true, cases would have to be differentially underdiagnosed more often when the CATIs are implemented earlier compared to when they were introduced >7 days after the outbreak started, or else cases would have to be underdiagnosed more often when the control efforts were more intense. Neither situation seems likely. While misclassification of non-cholera diarrhoea outbreaks could potentially attenuate the effectiveness estimate for the interventions (as suggested by the alternative outbreak definition "Outbreak Cu1" examined in Appendix 3—table 2), this was not true for most of the other outbreak definitions, and again is unlikely to be true more generally. At the very least, more justification should be provided for these statements.

We thank the reviewers for this comment and acknowledge that our CATI effectiveness may have been over or under estimated depending on the differential distribution of missing values between classes of response promptness and intensity. We modified our Discussion accordingly.

Despite missing data, we believe that our sensitivity analyses using very different outbreak definitions provide sufficient confidence that prompt and intense CATIs have the potential to effectively reduce cholera outbreak size and duration.

7) The effectiveness of CATIs has implications well beyond the specific control question at hand, on the still poorly agreed importance of routes of transmission in cholera. In turn, these routes and in particular the role of an environmental reservoir continue to be the subject of substantial debate, and also influence views on approaches to control and sustain the gains of control in the long run. We understand that the paper is not meant to resolve these matters but it would be valuable to add some discussion of related implications even if briefly. These will broaden the scientific value of the findings.

We agree with the reviewers that better understanding the respective importance of cholera transmission routes and the potential role of an environment reservoir can impact strategic choices in epidemiological surveillance and control. But as we believe that household water treatment, sanitation and hygiene promotion, as well as antibiotic prophylaxis may theoretically prevent both human-to-human and environment-to-human cholera transmission pathways, we initially did not discuss this important issue within the specific context of this paper dedicated to case-area-targeted interventions. Nevertheless, following the reviewers’ suggestion, we now discuss this important scientific issue in our conclusion.

8) There is concern about the choice of "outbreaks", which the authors define post-hoc based on reported incidence, as the unit of analysis. Some discussion would be useful of the implications if the unit of analysis is either an individual or a household, targeted with interventions or not.

Together with the reviewers, we are concerned that our effectiveness estimates may be biased by the fact that our retrospectively defined outbreaks may be an approximate unit of analysis in term of space, time and population. This is why we conducted a sensitivity analysis using alternative definitions, including systematically lab-confirmed cholera outbreaks, which showed consistent and robust estimates (Appendix 3.1). As suggested by the reviewers, another CATI effectiveness studies at the household level is underway in Haiti.

We believe that our Discussion section now better acknowledges this potential bias.

9) Outbreaks that haven't been responded to by a targeted intervention have been excluded from the analysis because the authors have detected major bias ("confounding by indication"), e.g. larger outbreaks are more likely to be targeted by > 1 intervention. Doesn't the same bias exist between outbreaks that have been responded to more or less promptly or with a higher or lower intensity?

We shared the same concern as the reviewers regarding the possibility of a residual confounding by indication in the group of outbreaks that received a CATI response.

However, proxies of outbreak initial severity (no. of cases and no. of positive cultures during the first three days) were not systematically higher in classes of better response promptness or intensity. In addition, our models were adjusted for potential confounders and took into account the heterogeneity between localities. This quasi-experimental study was also stratified on response promptness and on response intensity, which classes yielded consistent response effectiveness estimates.

Anyway, in the event a significant but hidden confounding by indication remained, this should have underestimated CATI effectiveness, which all appeared significantly positive.

Author details

1. Edwige Michel

   Ministry of Public Health and Population, Directorate of Epidemiology Laboratory and Research, Port-au-Prince, Haiti

   Contribution

   Data curation, Formal analysis, Investigation

   Competing interests

   No competing interests declared

2. Jean Gaudart

   Aix-Marseille Université, APHM, INSERM, IRD, SESSTIM, Hop Timone, BiSTIC, Biostatistics and ICT, Marseille, France

   Contribution

   Conceptualization, Software, Supervision, Validation, Methodology

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-9006-5729

3. Samuel Beaulieu

   United Nations Children's Fund, Port-au-Prince, Haiti

   Contribution

   Funding acquisition, Validation, Investigation

   Competing interests

   No competing interests declared

4. Gregory Bulit

   United Nations Children's Fund, Port-au-Prince, Haiti

   Contribution

   Funding acquisition, Validation, Investigation

   Competing interests

   No competing interests declared

5. Martine Piarroux

   Centre d'Épidémiologie et de Santé Publique des Armées, Service de Santé des Armées, Marseille, France

   Contribution

   Supervision, Investigation, Methodology

   Competing interests

   No competing interests declared

6. Jacques Boncy

   Ministry of Public Health and Population, National Laboratory of Public Health, Delmas, Haiti

   Contribution

   Funding acquisition, Investigation, Project administration

   Competing interests

   No competing interests declared

7. Patrick Dely

   Ministry of Public Health and Population, Directorate of Epidemiology Laboratory and Research, Port-au-Prince, Haiti

   Contribution

   Supervision, Funding acquisition, Project administration

   Competing interests

   No competing interests declared

8. Renaud Piarroux

   Sorbonne Université, Sorbonne Université, INSERM, Institut Pierre-Louis d’Epidémiologie et de Santé Publique (IPLESP), AP-HP, Hôpital Pitié-Salpêtrière, Paris, France

   Contribution

   Conceptualization, Supervision, Funding acquisition, Validation, Investigation, Methodology

   For correspondence

   renaud.piarroux@aphp.fr

   Competing interests

   No competing interests declared

9. Stanislas Rebaudet

   1. APHM, Hôpital Européen, Aix Marseille Université, INSERM, IRD, SESSTIM, IPLESP, Marseille, France
   2. Sorbonne Université, INSERM, Institut Pierre-Louis d’Epidémiologie et de Santé Publique, Paris, France

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Project administration

   For correspondence

   stanreb@gmail.com

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-5099-1947

• 1,417

  Page views

• 164

  Downloads

• 14

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.