Effects of HIV and non-communicable disease comorbidity on healthcare costs and health experiences in people living with HIV in Zimbabwe

Abstract

Background: The effects of HIV and non-communicable disease (NCD) comorbidities on healthcare costs and health experiences have been documented in most high-income countries. However, little similar data are available for Zimbabwe and most countries in sub-Saharan Africa. Untreated or under-treated NCDs can potentially negate the gains achieved from the control of HIV.

Objectives: The study sought to determine the effects of HIV-NCD comorbidity on healthcare costs, health experiences and treatment options for people living with HIV (PLWH) in Zimbabwe.

Methods: A repeated-measures, quantitative study was conducted at six antiretroviral therapy (ART) sites in the Gweru District of Zimbabwe. Simple random sampling was used to enrol 100 PLWH concurrently diagnosed with hypertension and/or diabetes mellitus (cases). Cases were matched by age, sex and viral load to an equal number of PLWH without hypertension and/or diabetes mellitus (controls). Quantitative data were collected using an interviewer-administered questionnaire at monthly intervals for 6 months. The questionnaire survey sought to compare healthcare costs, health-related experiences and treatment options between cases and controls. Data were analysed using Stata Version 13.1®. A logistic model was used to examine other factors such as demographic, clinical and behavioural data that were assumed to be unchanged over the study period. A random-effects model, including costs and other covariates, was used to compare groups in the final analysis.

Results: Non-communicable disease status was associated with the length of time on ART. Cases spent significantly more on transport (p = 0.0001) and medication (adjusted odds ratio [AOR] = 4.4, 95% confidence interval [CI]: 3.2–7.3); spent more days without doing usual daily activities because of sickness (AOR = 4.2, 95% CI: 3.3–7.6) and were more likely to use alternative medication (AOR = 3.4, 95% CI: 2.3–4.6) when compared with controls. Unemployment, female gender, age of 60 years and above, and living in rural areas were associated with failure to purchase prescribed medication.

Conclusions: HIV-NCD comorbidity causes an additional burden to PLWH because of increased transport costs, NCD prescribed medication expenses and more productive days lost due to illness. The success of HIV programmes does not only rely on improving access to the diagnosis and treatment of HIV. Addressing the complications of HIV-related NCDs, and the long-term costs of ART and its occasional potential for harm will be essential if health outcomes in Zimbabweans living with HIV are to be optimised.