Assessing the effectiveness of Lay Health Workers intervention on the management of hypertension in Agincourt rural, South Africa

Abstract

Background Uncontrolled hypertension determined by systolic blood pressure greater than 140 mmHg and diastolic blood pressure greater than 90 mmHg is among the leading cause of death in the whole world. In Agincourt a sub district of Mpumalanga in South Africa, uncontrolled hypertension prevalence is very high approximately 40%. This figure is believed to be higher since almost half of the population does not know their blood pressure status. This might be due to poor management of chronic illness in this area. In this study, lay health workers (LHWs) were used in managing uncontrolled hypertension in Agincourt. The aim of this study was to assess the impact of using multilevel logistic regression models with random effects instead of the mixed effects model which was used in the primary study to assess the effectiveness of LHW intervention in the management of uncontrolled hypertension in Agincourt, South Africa.

Methods The current study was a secondary data analysis of a cluster randomized controlled trial for data collected in Agincourt from 2013 to 2015. It adopted the primary study design as well as the sample size. The data consisted of population of hypertension patients aged 18 years and above. About 4000 people were allocated in each study arm which means that at least 500 people were allocated in each cluster. Demographic characteristics of the participants were shown as frequencies and percentages. The adjusted chi-squared test was used and the p-values were used to identify the association between uncontrolled hypertension and other factors. The treatment effect of lay workers was estimated using propensity score matching. Patient factors associated with the effectiveness of LHW in Agincourt were identified using level 1 ( with individual level random effects) and level 2 (with both individual and cluster random effects) multilevel models.

Results The descriptives statistics were focusing on the participants who were recorded both at baseline and at end of the intervention. The number of participants who were recorded both at baseline and end of intervention were 810 out of 4000. In the control arm, about 250 participants were recorded at baseline and 213 at the end of the intervention. In the intervention arm, 185 participants were recorded at baseline and 162 at the end of the intervention. There was no much difference on the social demographic variables between the control and the intervention arms at baseline and at the end of the intervention. Multilevel logistic regression model with individual random effects(level 1 multilevel model) was the best model compared to the model with both individual and cluster random effects (level 2 multilevel model) using intraclass correlation coefficient. The model highlighted that age, obesity, gender and smoking history were highly associated with uncontrolled hypertension. The odds (OR = 14.75; 95% CI: 8.1669 - 26.6513) of the elderly (40 years and above) of having uncontrolled hypertension is approximately seven times than that of young aged (18 - 39 years). The odds (OR = 2.8629; 95% CI: 1.5412 - 5.3178) of having uncontrolled hypertension among those who are obese is approximately three times than that of those who are not. The odds (OR = 2.2616; 95% CI: 0.9758 - 5.2414) of males of having uncontrolled hypertension is approximately twice than that of females.

Conclusion The findings in this study reported that lay health workers intervention was not effective in reducing uncontrolled hypertension. The results align with those obtained in the primary study by Goudge et al. (2018). However, findings were made that LHW are important in undertaking some of the clinical duties such as measuring of blood pressure, booking of patient’s appointments, retrieval of patient’s files a day before the appointment and refilling afterwards, providing of health education and assisting nurses with pre-packing of medication (11).