Ghana Study Highlights 3 Factors Linked to HIV ART Adherence

High interpersonal support, living in a rural area, and not disclosing HIV status to partners were all linked to higher antiretroviral therapy (ART) adherence among people living with HIV (PLWH) in a new study published in AIDS Research and Therapy.

“The findings highlight the need to develop and implement more novel and robust interpersonal support services for PLWH,” the study authors said. “Moreover, there is a need for assessment of the availability of tangible interpersonal support services during clinical encounters and the need for follow-up on patients with no tangible interpersonal support to improve ART adherence among PLWH.”

The authors’ goal was to determine whether there is an association between ART adherence and factors like interpersonal support and depression scores among PLWH living in the Volta region of Ghana.

To assess this, they conducted a cross-sectional survey among 159 PLWH—after exclusions due to missing data—who were aged 18 and older and receiving at an ART clinic between November 2021 and March 2022.

The survey consisted of 3 components: a 6-item simplified ART adherence scale, a 20-item Center for Epidemiologic Studies Depression Scale, and a 12-item Interpersonal Support Evaluation List-12. A chi-squared test was used to examine the relationship between these variables, along with additional demographic factors, and the status of ART adherence. The authors subsequently constructed a stepwise multivariable logistic regression model to help understand how these factors may influence ART adherence.

The authors found that ART adherence among this cohort was 34%.

While nearly a quarter (23%) of participants met the threshold for depression, there was no significant association found between depression and adherence in the multivariate analysis (P = .25). The average (SD) depression level was 9.11 (8.79), which was considered low risk for depression in the study.

On the other hand, 48.1% of participants reported high social support, with an average total interpersonal support score of 18.67 (6.61). Scores of 17 or greater were considered high, and high support was found to be significantly associated with adherence (adjusted odds ratio [aOR], 3.45; 95% CI, 1.09-5.88; P = .033).

Additional factors associated with adherence were identified in the multivariable model. These included not disclosing HIV status (aOR, 2.17; 95% CI, 1.03-4.54; P = .044) and not residing in an urban area (aOR, 0.24; 95% CI, 0.11-0.52; P = .00037).

Less than half (47.7%) of participants disclosed their HIV status to their partner. The authors found an adherence rate of 42% among PLWH with a non-disclosed status, and 23% among PLWH that have disclosed their status to their partners (P = .019).

"This may mean that the PLWH are not able to receive the needed support from their spouse," the authors noted. "Therefore, it is plausible interpersonal support helps sustain treatment linkage and promotes sustained adherence to ART that otherwise would have been interrupted."

They also found 66.7% adherence among PLWH residing in rural areas and 33.3% adherence among PLWH residing in urban areas (P = .0017).

“Unlike rural residents who have more community support in Ghana, urban lifestyle is often individualistic with minimal communal support,” the authors said. “Therefore, urban PLWH may lack the support network needed to remind and support them to adhere to ART—as underscored by our findings, high social support is associated with adherence. This coupled with the general stress and high cost of living for urban residents may impact their adherence to ART.”

No associations were found between adherence and the following variables:

• relationship
• gender
• age
• level of education
• monthly income
• depression score

While monthly income was not found to be an independent factor, the authors noted that income was more likely to be reported by participants who were adherent (P = .0013).

The study had several limitations, including the use of purposive sampling, the small sample size, potential response bias in self-report data, limited generalizability due to recruitment from a single clinic, inability to establish causal relationships with a cross-sectional design, and a substantial amount of missing data that could introduce bias. However, multiple imputation was employed to handle the missing data, and the authors said these findings provide insights for future studies using larger randomly selected samples to enhance validity.

Reference

Nutor JJ, Gyamerah AO, Alhassan RK, et al. Influence of depression and interpersonal support on adherence to antiretroviral therapy among people living with HIV. AIDS Res Ther. Published online June 29, 2023. doi:10.1186/s12981-023-00538-8