Factors associated with retention among a cohort of patients classified as pre-anti-retroviral treatment (pre-ART) in HIV care and treatment program in South Africa

Abstract

Background Early enrolment and retention of people living with HIV in care and treatment services is one of the important pillars for a successful care and treatment program. Retaining patients in HIV care has been a noted challenge in sub-Saharan Africa. This study sought to identify baseline socio-demographic and clinical factors that influence retention of adult patients enrolled in HIV chronic care from pre-ART. The findings and conclusions drawn from this study will inform policy and programme decisions that improve retention. Methods: This was a retrospective clinical cohort study using routinely collected data from a record review conducted in 15 public health clinics in Swaziland in the period from 1st April 2012 – 30th June 2012. Baseline clinical and socio-demographic characteristics of a cohort of patients (n=510) enrolled for pre-ART were used. The primary outcome was retention at three, six, and 12 months. For descriptive analysis, frequencies and proportions, and medians and interquartile ranges were calculated. Bivariate and multivariate logistic regression model and survival analysis were conducted to measure retention rates with associated factors and also to control for confounders. In both cross tabulation and multivariate analysis, a p-value of <0.05 was considered statistically significant in explaining observed differences. Data analyses were performed using Stata 13. Results: Women were the majority (78%) of patients enrolling for HIV chronic care (starting with pre-ART) and 37% were pregnant. The median CD4 count at enrolment was 357 cells/mm3 (IQR 216 - 546) and 89% had no clinical complaints at time of enrolment and were either in WHO clinical stage I and II. About a quarter (24%) of patients were lost within the first three months after the initial clinical visits. About half, of the patients 52%, 49% and 46%, were retained in HIV care at 3, 6 and 12 months respectively post-enrolment. The probability of being retained decreased with time. At 12 months patients had a 50% chance of being retained [St= 0.50, 95% CI (0.45-0.54)]. The socio-demographic and clinical factors that were significantly associated with retention among men were baseline CD4 count and age. Men who had a CD4 count greater than 500 cells /mm3 were less likely to be retained in care compared to men with CD4 count < 350 cells/mm3 (odds ratio (OR) = 0.38, (95% CI 0.61 – 0.94, P-value, 0.04) at three months and, OR= 0.22, (95% CI 0.05– 0.93, P-value, 0.04), at 12 months). Men who were 30 years and older were more likely to be retained in HIV care compared to younger men less than 30 years (. OR = 3.4, (95% CI 1.32 – 8.70, p= 0.01) at three months and, OR= 4.05 (95% CI 1.51-10.80, p= 0.005) at 6 months and OR= 3.17 (95% CI 1.23- 8.16, p= 0.02) at 12 months. Among females, baseline CD4 count, marital status and not being pregnant were significantly associated with retention. Women who were single were less likely to be retained in HIV care compared to those who were married at all retention points: at three months, OR = 0.43 (95% CI 0.26 – 0.71, p = 0.001), at six months OR= 0.62, (95% CI 0.38 – 1.00, p = 0.05) and at 12 months, OR 0.56= 95% CI 0.38 – 1.00, p = 0.05). Women with CD4 count greater than 500 cells / mm3 were 47% less likely (OR 0.53, 95% CI 0.30 – 0.91, P=0.02) to be retained in HIV care compared to women with CD4 count < 350 cells/ml. At three months non-pregnant women were 1.69 times more likely (OR 1.69, 95% CI 0.99 – 2.6, P=0.056) to be retained in HIV care compared to pregnant women. Conclusion and recommendations Retention rates among patients enrolling for HIV care starting with pre-ART were found to be relatively low, with only one in two retained in care. Loss-to- follow-up was the common reason for not being retained in care. Most patients were lost during the first three months especially soon after the day of enrolment. Baseline CD4 count and age among men, and baseline CD4 count, marital and pregnancy status among women, were the factors associated with retention in HIV care after 12 months. Patients most likely to be lost from care were those assessed as ineligible to initiate ART with a CD4 count >500 cell/mm3, young men, single and pregnant women. To improve retention rates, adopting and implementing the Treat All policy, where all people living with HIV are eligible to initiate ART regardless of CD4 count is recommended. In addition, building the capacity of health care workers to provide quality psychosocial assessment, counselling and support is essential. More focus should be placed on retaining patients not initiated on ART in the first three months. Real-time electronic HIV care data systems with unique patient identifiers may also improve data quality and tracking of patient outcomes. Further research is also needed to evaluate the effectiveness of the current national patient follow-up strategies and Treat All policy in improving retention rates in HIV care.