Christian Schätti Zundel: Social and cultural features of vaccine acceptance and cost-effectiveness of an oral cholera mass vaccination campaign in Zanzibar

PhD thesis, Swiss Tropical and Public Health Institute, 2011

Classical cholera control is based on prevention—safe water, sanitation and education of people on the importance of hygiene and diarrheal diseases—health system preparedness and a timely response to provide appropriate treatment in the event of an outbreak. The World Health Organization (WHO) recommends vaccination with an oral cholera vaccine (OCV) as a supplement for prevention and control of epidemic and endemic cholera. Consideration of local cultural concepts of illness among potential vaccine recipients and how these may affect vaccine acceptance is crucial. To date, no published studies have examined the influence of social and cultural features of cholera on vaccine acceptance in African settings. The aims of the research presented here were to study social and cultural features of OCV acceptance and to assess the cost-effectiveness of the 2009 OCV mass campaign in Zanzibar.

The integrated-methods approach of cultural epidemiology was used to study local views of cholera-like illness and to examine their influence on OCV acceptance in endemic communities in Zanzibar before and after the OCV mass campaign. Public and private costs of illness due to cholera and costs of the vaccination campaign were estimated to assess the cost-effectiveness of using OCVs from a health care provider and a societal perspective.

A random sample of 356 unaffected adults from a periurban and a rural community was studied with a vignette-based, locally adapted semi-structured interview based on the Explanatory Model Interview Catalogue (EMIC). This descriptive study showed that cholera was more often recognized as serious illness that may be fatal without appropriate treatment than shigellosis. Features of distress were primarily related to the negative social and financial impact cholera can have on a patient’s life. Interference with work- or income-related activities was the most prominent category of distress. The most prominent somatic symptoms were related to dehydration and to general gastrointestinal features. Cholera was mainly attributed to a dirty environment and microbiological contamination while causes unrelated to the biomedical basis were also identified, but with less prominence. Even though rehydration of the patient (primarily in the periurban community) and use of herbal treatment and antibiotics (rural community) were the preferred self-treatment options, professional health facilities were universally recommended at both sites. This survey showed that cholera represented a significant perceived illness burden in periurban and rural Zanzibar.

Subsequent analysis showed that community willingness for a free OCV was almost universal (94%), but declined with increasing price to 61% if the OCV was offered at a low price (~USD 0.9), to 19% if offered at a medium price (~USD 4.5) and to 15% if offered at a high price (~USD 9). Logistic regression models including somatic symptoms (low and high price), social impact (low and medium price) and perceived causes (medium and high price) explained anticipated OCV acceptance better than models containing only sociodemographic characteristics. This showed that prevaccination assessments of community demand for OCV should not only consider the social epidemiology, but also examine local sociocultural features of cholera-like illness.

Since only 50% of the interviewed respondents had drunk two doses of the free OCV—with higher priority in the rural (59%) than in the periurban (41%) community (p<0.01)—study of social and cultural determinants of OCV uptake was deemed necessary. Similar to the previous study of determinants of anticipated OCV acceptance, this study showed that consideration of sociocultural features of illness explained uptake better than a purely social epidemiological analysis. Loss of appetite and nausea, both nonspecific features of cholera were negative determinants. Recognition of unconsciousness as a sign of serious dehydration and concern that cholera outbreaks could negatively impact the local health care system in the rural area were positive determinants of acceptance. Female gender, rural residence and older age were also positive determinants of OCV uptake.

A sample of 367 vaccinated and unvaccinated adults from the same two communities was studied in a postvaccination survey with a revised EMIC interview. Factors associated with uptake indicated a positive impact of the vaccination campaign and of sensitization activities on vaccine acceptance behavior. Analysis of barriers among unvaccinated people identified logistical issues as main reasons for the low community coverage, with people’s own busy daily schedules as the most prominent feature. Unlike communities opposed to cholera control or in settings where public confidence in vaccines is lacking, this study indicated a good campaign implementation and trust in the health system.

The incremental cost-effectiveness ratio (ICER) of USD 119,339 per disability-adjusted life-year averted exceeded three times the Tanzanian per capita gross domestic product; thus, use of OCVs was not considered a cost-effective strategy in comparison to the current practice based on decentralized cholera treatment centers in Zanzibar.

In conclusion, the research presented here suggests little community opposition to vaccination and good prospects to use OCVs for endemic cholera control in Zanzibar. From an economic perspective, prospects to use OCV mass campaigns under current conditions seem to be limited. However, at a subsidized purchase price and subsidized delivery costs of ~USD 1 each per immunized individual, OCV mass campaigns may become economically and financially feasible for cholera control in high-incidence areas of Zanzibar.