Open access Original research BMJ Open: first published as 10.1136/bmjopen-2020-038464 on 10 December 2020. Downloaded from Use of surveys to evaluate an integrated oral cholera vaccine campaign in response to a cholera outbreak in Hoima district, Uganda Godfrey Bwire,1 Mellisa Roskosky,2 Anne Ballard,2 W Abdullah Brooks,2 Alfred Okello,3 Florentina Rafael,4 Immaculate Ampeire,5 Christopher Garimoi Orach,3 David A Sack 2 To cite: Bwire G, Roskosky M, ABSTRACT Strengths and limitations of this study Ballard A, et al. Use of Objectives To evaluate the quality and coverage of the surveys to evaluate an campaign to distribute oral cholera vaccine (OCV) during ► The cluster survey of households in communities integrated oral cholera vaccine a cholera outbreak in Hoima, Uganda to guide future campaign in response to a targeted for vaccination efficiently documented ac- campaigns of cholera vaccine. cholera outbreak in Hoima tual vaccine coverage in the target population. Design Survey of communities targeted for vaccination to district, Uganda. BMJ Open ► The cluster surveys of households identified mild determine vaccine coverage rates and perceptions of the 2020;10:e038464. doi:10.1136/ adverse events not identified during the campaign vaccination campaign, and a separate survey of vaccine bmjopen-2020-038464 and identified the need to emphasise the second staff who carried out the campaign. dose, especially among less educated groups. ► Prepublication history for Setting Hoima district, Uganda. this paper is available online. ► Surveys of the vaccination staff immediately fol- Participants Representative clusters of households To view these files, please visit lowing each round identified certain weaknesses in residing in the communities targeted for vaccination and the journal online (http:// dx. doi. staff orientation as well as constraints to their job staff members who conducted the vaccine campaign. org/ 10. 1136/ bmjopen- 2020- performance in the field. 038464). Results Among 209 households (1274 individuals) ► The household surveys obtained data from a single included in the coverage survey, 1193 (94%; 95% CI 92% spokesperson for the household rather than from Received 11 March 2020 to 95%) reported receiving at least one OCV dose and each individual which might have introduced some Revised 17 November 2020 998 (78%; 95% CI 76% to 81%) reported receiving two Accepted 19 November 2020 uncertainty in the household data. doses. Among vaccinated individuals, minor complaints http://bmjopen.bmj.com/ ► Evaluation of the vaccination staff was carried were reported by 71 persons (5.6%). Individuals with through surveys and would have benefited by direct ‘some’ education (primary school or above) were more observation of the training and the field performance. knowledgeable regarding the required OCV doses © Author(s) (or their compared with non- educated (p=0.03). Factors negatively employer(s)) 2020. Re- use associated with campaign implementation included in several countries in sub- Saharan Africa permitted under CC BY. community sensitisation time, staff payment and problems where cholera outbreaks also negatively Published by BMJ. with field transport. Although the campaign was carried 1 affect development due to associated high Department of Community out quickly, the outbreak was over before the campaign 2 3 Health, Ministry of Health, started. Most staff involved in the campaign (93%) were economic burden. Between 2010 and 2016, on September 28, 2021 by guest. Protected copyright. Kampala, Uganda knowledgeable about cholera control; however, 29% did an average 141 918 incident cases annually 2Department of International not clearly understand how to detect and manage adverse were reported from sub-Saharan African 4 Health, Johns Hopkins events following immunisation. countries, including Uganda. In Uganda, Bloomberg School of Public Conclusion The campaign achieved high OCV coverage, cholera outbreaks occurred as both endemic Health, Baltimore, Maryland, but the surveys provided insights for improvement. To USA and epidemic disease. Epidemic disease achieve high vaccine coverage, more effort is needed for 3Makerere University, College occurred in northern and eastern Uganda 5 of Health Sciences, Kampala, community sensitisation, and additional resources for staff districts and are thought to be worsened Uganda transportation and timely payment for campaign staff is by contamination of water due to poor sani- 4 required. Pretest and post- test assessment of staff training Department of Infectious tation.6 Cholera outbreaks especially occur Hazard Management, World can identify and address knowledge and skill gaps. Health Organization, Geneva, in districts along the international borders Switzerland with the Democratic Republic of the Congo 5Ministry of Health, Uganda INTRODUCTION (DRC), South Sudan and Kenya and along National Expanded Program on Cholera, a preventable and treatable disease, the Great Lakes.5 7 These districts include Immunization, Kampala, Uganda is characterised by profuse watery diar- Hoima, where cholera is endemic.5 8–10 Correspondence to rhoea caused by infection of the intestine There has been debate in the public health 1 Dr David A Sack; with the bacterium Vibrio cholerae. Cholera community on best practices for endemic dsack1@ jhu. edu is a major cause of morbidity and mortality and epidemic cholera disease control, with Bwire G, et al. BMJ Open 2020;10:e038464. doi:10.1136/bmjopen-2020-038464 1 Open access BMJ Open: first published as 10.1136/bmjopen-2020-038464 on 10 December 2020. Downloaded from some preferring to focus on water, sanitation and hygiene was used to deliver two doses of vaccine to an estimated (WASH) interventions, and others advocating for oral 360 000 people, including pregnant women, over the age cholera vaccine (OCV) for both endemic and epidemic of 1 year residing in the four targeted subcounties. disease control.11 In part, this has been facilitated by a To carry out the campaign, the MoH organised all relative lack of experience with OCV and concern that activities including logistics, community mobilisation and excess reliance on vaccine might negatively affect essen- implementation, coordinating ground activities through tial infrastructural development and hygienic practices. an assigned point person. Many stakeholders contrib- WHO recommends an integrated approach to cholera uted to the campaign including the Hoima district local prevention where WaSH interventions are complemented government, WHO, UNICEF, UNHCR and Médecins by vaccine campaigns which provide OCV to persons sans Frontiers. Prior to the campaign, the stakeholders living in areas considered high risk.2 12 These vaccine met to define and coordinate their complementary tasks. campaigns may be either preventive, in which the vaccine The epidemic curve based on a line list of cases and is targeted to cholera hotspots, or reactive in which the deaths by date and stated nationality is shown on figure 1. campaign is implemented in response to an outbreak or Over the course of the outbreak, 2122 cases with 44 a humanitarian emergency.13 deaths (case fatality rate (CFR), 2.1%) were reported. Two WHO- prequalified currently OCVs are available Sixty- six per cent (1410) of the cases and 64% (28) of the from the global stockpile: Shanchol (Shantha Biotech- deaths occurred during the first 2 weeks of the outbreak. nics, India) and Euvichol (Eubiologics, Korea).2 The stan- Many of the cases and deaths (1276 and 32, or 60% and dard immunisation schedule consists of two doses given 73%, respectively) occurred among persons who were at an interval of at least 2 weeks to all persons in the target from DRC, and the refugees developed cholera symp- area above 1 year of age. While there is increasing use of toms soon after arrival in Uganda. Among the 44 deaths OCV to control outbreaks, preventive use is constrained reported, 25 (57%) occurred in the community, not in due to inadequate vaccine supply.14 Since creation of a the health facility. Nineteen of the fatal cases were treated global OCV stockpile in July 2013, several OCV campaigns at the health facility; the CFR for facility- treated patients had been successfully implemented13 14 but it is still was 0.9%. Although the emergency vaccination campaign important to document national campaign experiences intended to control the outbreak, because the outbreak as well as monitoring and evaluation activities, to contin- was so sudden and so short lived, the campaign could ually improve the effectiveness and efficiency of vaccine only be initiated after the outbreak had already declined. campaigns. The Ugandan Ministry of Health (MoH) had prepared Rationale plans for OCV campaigns in the areas identified as While vaccines are commonly used in Uganda, espe- cholera hotspots starting in the western districts of cially through the longstanding programme (Expanded Uganda (including Hoima), near the border with DRC Programme on Immunisation), this was the first OCV http://bmjopen.bmj.com/ and close to, or adjoining Lake Albert. These hotspot use in Uganda, and there was no prior experience to districts and their specified subcounties were confirmed guide responders and implementers. Thus, this study was during a national cholera workshop in Kampala on 29 carried out with the aim to document campaign activities January 2018–31 January 2018. This workshop led to the and to monitor and evaluate its procedures and outcomes development of an application for OCV to the Global