The D²Rwanda mixed-methods study including a cluster-randomised controlled clinical trial

Diabetes mellitus prevalence has been estimated at 5.1% in Rwanda. Several factors, including an increase in screening and diagnosis programmes, the urbanization of the population, and changes in lifestyle are likely to contribute to a sharp increase in the prevalence of diabetes mellitus in the next decade. Patients with low health literacy levels are often unable to recognise the signs and symptoms of diabetes mellitus, and may access their health provider late, hence presenting with more complications.
The Rwandan health care system is facing a severe shortage in human resources. In response to the need for a better management of non-communicable diseases at primary health care level, a new type of community health workers was introduced: the home-based care practitioners (HBCPs). Approximately 200 HBCPs were trained and deployed in selected areas (“cells”) in nine hospitals across the country.
There is growing evidence for the efficacy of interventions using mobile devices in low- and middle-income countries. In Rwanda, there is an urgent call to using mobile health interventions for the prevention and management of non-communicable diseases. The D²Rwanda (Digital Diabetes in Rwanda) research project aims at responding to this call.
The overall objectives of the D²Rwanda project are: a) to determine the efficacy of an integrated programme for the management of diabetes in Rwanda, which would include monthly patient assessments by HBCPs and an educational and self-management mobile health patient tool, and; b) to qualitatively explore the ways these interventions would be enacted, their challenges and effects, and changes in the patients’ health behaviours and HBCPs’ work satisfaction.
The project employed a mixed-methods sequential explanatory design consisting of a one-year cluster randomised controlled trial with two interventions and followed by focus group discussions with patients and HBCPs.
The dissertation presents three studies from the D²Rwanda project. The first study aimed at describing the protocol of the research project, reporting the research questions, inclusion and exclusion criteria, primary and secondary outcomes, measurements, power calculation, randomisation methods, data collection, analysis plan, implementation fidelity and ethical considerations.
The aim of the second study was to report on the translation and cultural adaptation of the Problem Areas in Diabetes (PAID) questionnaire and the evaluation of its psychometric properties. First, the questionnaire was translated following a standard protocol. Second, 29 participants were interviewed before producing a final version. Third, we examined a sample of 266 adult patients living with diabetes to determine the psychometric characteristics of the questionnaire. The full scale showed good internal reliability (Cronbach’s α = 0.88). A four-factor model with subdimensions of emotional, treatment, food-related and social-support problems was found to be an adequately approximate fit (RMSEA = 0.056; CFI = 0.951; TLI = 0.943). The mean total PAID score of the sample was high (48.21). Important cultural and contextual differences were noted, urging a more thorough examination of conceptual equivalence with other cultures.
The third study aimed at reporting on the disease-related quality of life of patients living with diabetes mellitus in a non-representative sample in Rwanda and to identify potential predictors. This cross-sectional study was part of the baseline assessment of the clinical controlled trial. Between January and August 2019, 206 adult patients living with diabetes were recruited. Disease-specific quality of life was measured using the Kinyarwanda version of the Diabetes-39 (D-39) questionnaire, which was translated and cross-culturally adapted beforehand by the same group of researchers. A haemoglobin A1c (HbA1c) test was performed on all patients. Socio-demographic and clinical data were collected, including medical history, disease-related complications and comorbidities. “Anxiety and worry” and “sexual functioning” were the two most affected dimensions. Hypertension was the most frequent comorbidity (49.0% of participants). The duration of the disease and HbA1c values were not correlated with any of the D-39 dimensions. The five dimensions of quality of life were predicted differentially by gender, age, years of education, marital status, achieving a HbA1c of 7%, hypertension, presence of complications and hypoglycaemic episodes. A moderating effect was identified between use of insulin and achieving a target HbA1c of 7% in the “diabetes control” scale. Further prospective studies are needed to determine causal relationships.