Rawlance Ndejjo

PhD summary

Cardiovascular disease (CVD), the number one cause of death globally, disproportionately affects low- and middle- income countries (LMICs) creating a new burden for their unprepared health systems amidst resource constraints. Community-wide interventions targeting CVD risk factors can support CVD prevention and control efforts in LMICs. However, there is need for more evidence on the effectiveness of community-based interventions for CVD prevention and an exploration of their implementation processes in real-world settings. This evidence will inform implementation and scale-up of community level CVD prevention programmes in LMIC contexts. Objectives The objectives of this research were to: synthesize evidence on the effectiveness of community-based interventions for CVD prevention among adults in LMICs (Paper I); determine community knowledge and describe the distribution of lifestyle practices for CVD prevention (Paper II & III); explore the acceptability of a community programme to increase knowledge and improve lifestyle practices for CVD prevention (Paper IV); document barriers and facilitators in implementation of the community programme (Paper V); and examine the factors influencing uptake of lifestyle practices for CVD prevention following implementation of the CVD prevention programme (Paper VI). Methods This thesis employed an embedded mixed methods study within a quasi-experimental design to answer the research questions. Besides a systematic review that was conducted to synthesise evidence on effectiveness of community-based interventions for CVD prevention among adults in LMICs, the remaining objectives were answered using primary data collected in Mukono and Buikwe districts in Uganda. Quantitative data were collected at baseline among 4372 respondents from 3689 randomly selected households providing information on the level of knowledge and distribution of CVD risk factors within the community. For analysis, descriptive statistics and generalized linear modelling controlling for clustering were conducted with the aid of Stata 13.0 or 15.0. The community intervention that involved community health workers (CHWs) to educate community members, screen for their CVD risk and counsel or refer them to health facilities was implemented in five parishes within 20 selected villages. Qualitative data were collected from community members and community health workers to inform exploration of the acceptability of the intervention, barriers and facilitators of implementation and factors influencing uptake of healthy lifestyle practices. Thematic analysis following semantic or latent approaches were employed in analysing qualitative data with the aid of either Atlas ti 6.0.15 or NVIVO 12.6. Results Results from the systematic review revealed that community-based interventions successfully improved population knowledge on CVD and its risk factors and influenced physical activity and dietary practices for CVD prevention. However, evidence was inconsistent for smoking cessation and reduced alcohol consumption. The most effective interventions involved health education, community mobilisation and lifestyle counselling (Paper I). To inform the community CVD prevention programme, we collected baseline data from 4372 respondents and found that only 776 (17.7%) were knowledgeable on CVD prevention. The factors associated with CVD knowledge were post-primary education [adjusted prevalence ratio (APR) = 1.55 (95% CI: 1.18 - 2.02), p = 0.002]; formal employment [APR = 1.69 (95% CI: 1.40 - 2.06), p <0.001]; and high socio-economic index [APR = 1.35 (95% CI: 1.09 - 1.67), p = 0.004]. Households that owned a mobile phone [APR = 1.35 (95% CI: 1.07 - 1.70), p = 0.012] and respondents who had ever received advice on healthy lifestyles [APR = 1.38 (95% CI: 1.15 - 1.67), p = 0.001] had significantly high knowledge (Paper II). Mapping of the prevalence of selected CVD risk factors using the baseline data indicated substantial gender and small area geographic heterogeneity which was masked on aggregate analysis. Patterns and clustering were observed for hypertension, physical inactivity, smoking, alcohol consumption and risk factor combination (Paper III). Evidence on effective interventions together with gaps identified in knowledge and CVD risk practices informed the design of our intervention which was acceptable to both CHWs and community members. CHWs had implemented similar community programmes and were eager to participate in the intervention while community members looked forward to health services being brought nearer to them. CHWs anticipated challenges in mobilising the community but anticipated to be trained, supported, and supervised while the community was eager to receive sufficient information on CVD prevention (Paper IV). In exploration of the barriers and facilitators during intervention implementation, the CHWs noted that the intervention was complex, they were sometimes mistrusted by their community whose awareness of CVD was low and had other competing demands. On the other hand, the availability of inputs and protective equipment, being trained, frequent support supervision and engagements and working with available community structures including leaders and groups facilitated the intervention (Paper V). Following implementation, we found variations in uptake of healthy lifestyle practices for CVD prevention with most changes reported for dietary behaviour and physical activity than alcohol consumption, smoking behaviours, and stress reduction like our review findings. The barriers to uptake of healthy lifestyle practices were related to accessibility challenges, limited resources, limited capability or skills, low risk perception and effects of COVID-19 restrictions such as lockdown. In contrast, the facilitators of practices uptake were knowledge, personal determination, competence to change, health benefits, social support, and reinforcement of messages (Paper VI). Conclusions Evidence-based, acceptable, and well-designed community-based interventions led by community health workers can support uptake of healthy lifestyles that are key for CVD prevention. These interventions hold promise for LMICs to deal with CVD. Study strengths and limitations This thesis provides evidence of effectiveness of community-based interventions for CVD prevention through a systematic review that informed the planned interventions. The thesis was informed by theories and frameworks and triangulation was used as a principle to present perspectives of CHWs and the community. The research could have been influenced by social desirability and information biases, however, the in-depth nature of the inquiry lessened the impact of these. The study was also only conducted in two districts, which although are similar to many rural and semi-urban areas in Uganda, may limit generalisability of findings.