Burden of multimorbidity in sub-saharan africa: Preliminary findings from three community studies

Context: Multimorbidity (MM) (co-occurrence of ≥2 chronic conditions (CCs)) is a global health challenge. Most evidence from MM research comes from studies in high-income countries. In many low and middle-income countries (LMIC), including low-income sub- Saharan Africa (SSA), the burden of CC is rising, however, the epidemiology of MM remains relatively unknown.

Objective: To 1) harmonise multi-site CCs data; 2) compare/understand variation in adult (18 years) prevalence of, and sociodemographic factors associated with, MM using community cohort data from Gambia/Malawi/Uganda; and 3) identify data gaps.

Study Design and Analysis: We used population –self report and/or measurement based survey data from Kiang West Longitudinal Population Study (KWLPS) in the Gambia (2004 onwards); the Health and Demographic Surveillance Site (HDSS) in Malawi (2007-16); General Population Cohort (GPC) in Uganda (2010-11)

Setting and Population: Adults (≥18 years) residing in defined urban/rural populations in three SSA countries.

Results: Self-report and/or measurement data available for: hypertension, diabetes and obesity in all three datasets; hypercholesterolaemia, HIV and asthma in two and epilepsy in one. To date we have investigated/compared availability of CC/sociodemographic data across the three datasets. The mean age for participants in KWLPS data-Gambia (N=7917) was 38 years (SD=19.3), with 60% females; prevalence of MM was 4.5% (192 participants out of 4270), with CC prevalence data missing in 46% of participants. The mean age for participants in HDSS data-Malawi (N=30574) was 35 years (SD=14.8), with 61.8% females. The prevalence of MM was 11.8% (2389 participants out of 20299), however CC prevalence data was missing for 33.6% of participants. Mean age for participants in GPC data-Uganda (N=7833) 34 years (SD=18.3), with 56.2% females; prevalence of MM was 7.2% (563 participants out of 7829).

Analyses in progress include, age and sex stratified prevalence of MM, and associations between socio-demographic and lifestyle factors and MM, using regression modelling.

Conclusions: This work has potential to enhance understanding of the burden, distribution and determinants of MM in LMICs and to inform policy response. However, robust data on a wider range of CCs are needed to better understand the impact and appropriate response to MM in LMIC of SSA.