Research note: The legacy of colonial medicine in Central Africa

In an effort to develop their colonies, France implemented  an extensive medical campaign for sleeping sickness between the 1920s and 1950s. These consisted of forced medical examinations and injections of drugs which had dangerous and sometimes lethal side effects. Such terrible experiences may have create a lasting impact in those communities that were exposed to modern (Western) medicine for the first time. In fact, they could explain why in large parts of Africa there is a very low feeling of trust in modern medicine which has hindered the potential growth of health care in these areas. Also, research shows that when there is enough availability of health care, the demand remains puzzlingly low. 

The article by Lowes and Montero (2021) aims at researching this. They hypothesize that colonial medical campaigns may have had a series of unintended effects on both beliefs about modern medicine and the success of health interventions. In the following, I summarize their main findings. 

Linking colonial medical campaigns and current outcomes

Lowes and Montero (2021) test their hypothesis by examining how historical colonial medical campaigns influence present-day vaccination rates, trust in medicine and the success of the World Bank health projects.  

First, they construct a dataset on medical campaigns using data from French military archives for five countries, namely The colonial governments of Cameroon, Central African Republic, Chad, Gabon, and the Republic of Congo. The historical reports consist of detailed descriptions of the places visited by the sleeping sickness medical teams and the types of treatments administered between 1921 and 1956. The figure below shows the frequencies of the visits in the region. 

Second, they match these data with information from the Demographic and Health Surveys (DHS) from recent years. The authors construct a vaccination index for children under five years old and also obtain information on whether the child has completed (all, multiple or none of) the nine different vaccines: polio, tuberculous, diphtheria, tetanus pertussis and measles. 

Third, they proxy for trust in medicine by whether an individual consents to a free and non-invasive blood test (either for anemia or HIV). A refusal to the blood test is considered as a proxy for mistrust in modern medicine.

At last, they used data from AidData on the location of World Bank projects approved between 1995 and 2014 to examine their successful implementation. The World Bank classifies its projects in five sectors: “health, central government administration, general public administration, other social services, railways, and roads and highways”. Since the World Bank rates their projects from “highly unsatisfactory” to “highly satisfactory”, they were able to compare the results for health projects with the data on the intensity of the sleeping sickness campaigns.

Main findings

Results show that a greater exposure to the sleeping sickness campaigns is associated with lower vaccinations rates for children. To be precise, when an area has an average of 15 years being visited by the campaigns, it is associated with a 5.8 percentage point decrease in the vaccination index. Second, their analysis show that increased exposure to the sleeping sickness campaigns is correlated with lower levels of trust in medicine today. When an area has been visited for 15 years by the medical campaigns it increases refusal of blood tests by 5.4 percentage points. The figure below show the latter result graphically. 

Both results cannot identify the causal effect of medical campaigns on vaccination and trust in medicine, since there might be another variable which determines the outcomes of both of our examined variables. To address this concern, the authors take an instrumental variable approach. The instrument used is the log suitability for cassava compared to the log suitability for millet. The results obtained with this empirical strategy are then similar and thus strengthen a causal interpretation of their findings. 

At last, the authors find that greater exposure to the campaigns is correlated with less successful health projects. However the projects in other domains are not negatively affected by the exposure to the medical campaigns. The size of this effect is as big as the rate changing from moderately satisfactory to moderately unsatisfactory.

Lowes and Montero have found sizeable and significant results for vaccination rates, trust in medicine and the successfulness of World Bank health interventions. But how can we further explain these results? They discuss two mechanisms that might be at play. First, they show that the exposure to the medical campaigns only affects trust in medicine, and not other trust-related measures. Second, they suggest that mistrust in medicine is mostly transmitted vertically (horizontal transmission is not ruled out) because the effect of exposure to the medical campaigns is twice as large for an individual’s own ethnic group as compared to others’ exposure. 

Conclusion

These results remind us of the importance historical events can have on present-day events. The colonial medical campaigns still influence the health of communities they visited on an individual and communal level, through their negative consequences on vaccination rates and trust in modern medicine. This can have serious complications for the overall health individuals living in these areas, since they are less likely to seek medical help even when it is available to them. Thus, building the demand for medical help will need more attention on rebuilding trust in modern medicine, instead of solely focusing on informing these communities about their availability.

---