Prevalence and determinants of hypertension in older adults in Ghana : A systematic review and analysis of a longitudinal study

Bosu, William and Zucchelli, Eugenio and Reilly, Siobhan (2020) Prevalence and determinants of hypertension in older adults in Ghana : A systematic review and analysis of a longitudinal study. PhD thesis, Lancaster University.

Text (2021BosuPhD)
Bosu_Thesis_HTN_in_older_adults_in_Ghana.docx - Published Version
Download (16MB)

Abstract

Introduction Ghana is undergoing rapid epidemiological and demographic transitions. The proportion of the population aged 60 years and older is projected to reach 12.6% in 2050 from 5.4% in 2010 while the urban population increased from 23.3% in 1960 to 50.9% in 2010. These developments have contributed to a changing disease profile with an increasing burden of chronic noncommunicable diseases (NCDs). The share of total deaths in all ages due to cardiovascular diseases almost doubled from 10.8% in 1990 to 18.2% in 2017. Although hypertension is a major risk factor for cardiovascular diseases, its determinants have not been adequately explored in adults aged 50 years and older in Ghana. ObjectivesThe study sought to assess the prevalence of hypertension, its severity and its determinants of hypertension in community-dwelling older adults at and over two time points. It also estimated the level of awareness, treatment experience and blood pressure (BP) control among people with hypertension as well as the clustering of risk factors among the study participants. MethodsThe study involved two components. One component estimated the prevalence of hypertension in older adults in Ghana within the framework of a systematic review and meta-analysis of community-based studies on the prevalence of hypertension in older adults in Africa published between 1980 and May 2018. The review involved the use of appropriate search terms in major databases to maximise yield. The pooled prevalence was estimated using a random effects model. Measures to assess the quality of the results included independent assessment of the risk of bias in the included studies, sensitivity analysis involving the exclusion of one study at a time to assess its effect on the pooled prevalence and the assessment of the presence of reporting bias. The other component involved the analysis of publicly available datasets of two waves of a nationally representative Study on global AGEing and adult health (SAGE) in Ghana conducted in 2007/2008 and in 2015 respectively. A series of single level (individual), multilevel (individuals nested in households which are in turn nested in enumeration area) as well as ordinal (to capture different grades of hypertension) regression models were fitted to the data with hypertension defined alternatively as a binary and an ordinal outcome variable. The different categories of the ordinal outcome variable corresponded to increasing levels of hypertension. All models included a wide range of demographic, socioeconomic and lifestyle-related explanatory variables. Statistical analyses were first conducted separately for Wave 1 (W1) and Wave 2 (W2) participants and then secondly, by pooling together the W1 and W2 datasets for longitudinal data analysis. Main findingsThe systematic review involved 34 studies which enrolled 43,025 individuals from 15 African countries. There were four studies from Ghana involving 10,519 older adults. The pooled prevalence of hypertension among the older adults from Ghana was 44.0% [95% confidence interval (CI) 34.0%-54.0%]. In the secondary data analyses, the prevalence of hypertension declined from 57.1% in W1 to 35.1% in W2. Of those with hypertension, 25.1% in W1 and 12.5% in W2 had severe or grade 3 hypertension with blood pressure >180/110 mmHg. About 13.2% and 21.6% of the variation in the prevalence of hypertension in W1 and W2 respectively could be attributed to unobserved between-households and between-enumeration area differences, after accounting for the predictor variables in the model.Most of the major determinants that were independently associated with hypertension were also associated with severe hypertension in the separate study wave analyses as well as in the longitudinal data analyses. Residence in the Greater Accra Region (GAR) [odds ratio (OR) 2.94, 95% CI 2.04-4.25) relative to residence in the northern regions]; urban residence (OR 1.24, 95% CI 1.04-1.48 relative to rural residence); being unmarried or having never been married (OR 1.50, 95% CI 1.27-1.78 relative to being currently married); being overweight (OR 2.23, 95% CI 1.67-2.98 relative to underweight) or obese (OR 2.52, 95% CI 1.77-3.58) were positive predictors of hypertension over time. The odds of hypertension increased with increasing body mass index category. In contrast, having recently worked for at least two days in the past week (OR 0.70, 95% CI 0.59-0.83 relative to not currently working) and having participated in study W2 (OR 0.28, 95% CI 0.23-0.34) were negative predictors of hypertension. Of the older adults with hypertension, 24.1% in W1 and 27.6% in W2 were aware of their condition from a previous diagnosis while only 4.8% and 2.3% respectively waves had their blood pressure controlled. Older adults in Ghana had an average of 2.5 concurrent NCD risk factors. Clustering of risk factors was common with 83.0% of the older adults in W1 and 71.4% of those in W2 having two or more concurrent risk factors. ConclusionsThis is one of the few studies in Africa to explore the determinants of hypertension in older adults through longitudinal data analysis and multilevel techniques. The study showed a high prevalence of hypertension in older adults which appears to have declined over time. The reasons for the decline could be related to the assessment of hypertension through single visit blood pressure measurements, a protocol which is known to overestimate the prevalence of hypertension. The significant proportion of severe hypertension may indicate longstanding disease and could be associated with end organ damage. The determinants of hypertension from the secondary data analyses were generally consistent with those from the systematic review. The clustering of risk factors as well as the low awareness and poor control of hypertension call for the intensification of efforts to prioritize the health of older adults towards healthy ageing. The findings of this study call for a concerted integrated multisectoral approach to tackling hypertension which might involve interventions across the life course; prioritising the Greater Accra Region (where the greatest risk occurs); raising awareness; and implementing weight reduction programmes. Yet, further research might be needed to better define the trends in hypertension, to explain the change in the prevalence of hypertension observed over time and to determine how best to deliver the interventions to this older age group.