Retention in Care among Patients with Early HIV Disease in Haiti

Abstract

In September 2015, the World Health Organization updated their guidelines to recommend antiretroviral therapy (ART) for all people living with HIV. Countries are now in the process of implementing strategies to provide universal HIV treatment. We analyzed the rate of retention and time to ART eligibility (according to 2013 WHO guidelines) among 3,345 adult patients receiving positive HIV test results between February 1, 2003 and March 31, 2013 at the GHESKIO Clinic in Haiti, with WHO stage 1 or 2 disease and initial CD4 cell count >500 cells/mm3. Among the 3,345 patients, 2,423 (72%) were female, the median age was 33 years, 3,089 (92%) lived in Port-au-Prince, and 1,944 (58%) had attended no school or primary school only. The median initial CD4 cell count was 668 cells/mm3 (IQR: 572-834); over the subsequent 2 years, 1,485 patients (44%) were lost to follow-up and 7 (<1%) died pre-ART, 1,041 (31%) were retained in pre-ART care, and 819 (24%) initiated ART. In multivariate analysis, secondary education (aOR 1.27; 95% CI: 1.10-1.47), female gender (aOR: 1.28; 95% CI: 1.09-1.50), co-habitation (aOR: 1.31; 95% CI: 1.09-1.57), and residence in Port-au-Prince (aOR: 1.43; 95% CI: 1.09-1.88) were associated with retention in care. The median time from baseline CD4 count to ART eligibility was 1.7 years. Prior to the implementation of universal treatment, pre-ART attrition was high among patients who did not qualify for ART at presentation. Though implementing WHO recommendations for universal ART will require service expansion, it will likely result in improved retention for those at risk of being lost to follow-up.

Introduction

Over the past decade, the World Health Organization (WHO) has revised their guidelines 3 times to recommend antiretroviral therapy (ART) for HIV-infected patients at higher CD4 counts. Initially, the CD4 count threshold for treatment was <200 cells/mm³; in 2009, it was increased to ≤350 cells/mm³; in 2013, it was increased to ≤500 cells/mm³; and in 2015, the guidelines were changed to recommend ART for all HIV-infected patients, as earlier treatment is associated with improved outcomes and lower transmission. ^{1 –4} Countries are now evaluating the feasibility of implementing universal treatment. As of May 2016, only 14 countries had adopted this strategy. ⁵ Little has been published on the additional time that would be required on ART compared to the 2013 eligibility criteria. ^{6 –8}

We evaluated raes and predictors of 2-year retention among patients with CD4 count >500 cells/mm³ at initial testing, and time to qualifying for treatment among those who did not meet the 2013 WHO ART eligibility criteria at the Haitian Group for the Study of Kaposi’s Sarcoma and Opportunistic Infections (GHESKIO).

Methods

Study Setting and Pre-ART Care

HIV care in Haiti and at GHESKIO has been described in prior reports. ^9,10 Patients not yet qualifying for ART were scheduled for monthly clinician visits for the first 3 months and then seen every 2 to 3 months. Isoniazid prophylaxis was prescribed for patients who were positive for purified protein derivative, without active tuberculosis (TB), and trimethoprim-sulfamethoxazole prophylaxis was prescribed for all HIV-infected patients, regardless of CD4 count. During the pre-ART period, patients did not receive transportation subsidies or phone calls regarding upcoming or missed visits. Antiretroviral therapy eligibility criteria changed during the study period, from WHO stage 4 condition and/or CD4 count <200 cells/mm³ initially, to WHO stage 3 or 4 condition and/or CD4 count ≤350 cells/mm³ in 2009, to ART for all patients with CD4 count ≤500 cells/mm³ in 2013, and universal ART in 2016.

Adults ≥18 years of age who met the following 4 criteria were included in the study: (1) received positive HIV test results between February 1, 2003, and March 31, 2013; (2) had initial CD4 count >500 cells/mm³ and returned for test results; (3) were ART naive; and (4) did not qualify for ART according to 2013 WHO guidelines or initiate ART within 3 months after initial CD4 count.

Statistical Analysis

Demographic information and clinical data were analyzed via SAS version 9.3 (SAS Institute Inc. Cary, North Carolina) and R 3.0 (R Core Team (2015), Vienna, Austria). Patients were categorized into cohorts by year of HIV test; year 1 included patients tested from March 1, 2003, to February 28, 2004; year 10 included patients tested from March 1, 2012, to February 28, 2013. The last date of data availability was August 30, 2015.

Primary outcomes were 2-year retention in care and time from initial CD4 count to either initiating ART or qualifying for treatment according to 2013 WHO guidelines. Patients were considered retained in care if they had at least 1 visit between 18 and 24 months after the date of baseline CD4 count. Univariable and multivariable analyses were conducted to identify predictors of 2-year retention. Adjusted odds ratios were reported along with 95% Wald confidence intervals (CIs). Significance of linear trend in 2-year retention was tested using the Cochrane-Armitage test for trend. Kaplan-Meier analyses were conducted to evaluate the time to qualifying for ART by WHO 2013 guidelines. Institutional review board approval was obtained from all participating institutions.

Results

From March 2003 to February 2013, a total of 4109 patients had initial CD4 counts >500 cells/mm³ at GHESKIO, and 3687 patients (90%) returned for CD4 count results a median of 8 days after blood draw (interquartile range [IQR]: 3-19 days). The trend in returning for results declined over the 10-year enrollment period (P = .018) but was at least 87% across all study years. Among patients who returned for CD4 count results, 327 (9%) were excluded because they became eligible for ART initiation within 3 months after initial CD4 count. Fifteen additional patients were excluded because they were known to have transferred during the study period, leaving a study population of 3345 patients. Of these, 2423 (72%) were female, the median age was 33 years (IQR: 26-43), 3089 (92%) lived in Port-au-Prince, 2071 (62%) reported an annual income ≤$US125/year, and 1944 (58%) had attended no school or attended primary school only. A total of 2525 patients (75%) were self-referred to GHESKIO or referred by a family member. The distribution of patients with CD4 counts from 501 to 600, 601 to 700, 701 to 800, 801 to 900, and >900 cells/mm³ at baseline was 1089 (33%), 789 (24%), 502 (15%), 351 (10%), and 614 (18%) respectively. The median initial CD4 count was 668 cells/mm³ (IQR: 572-834).

Retention in Care

At 2 years after the date of CD4 count testing, 1485 patients (44%) were loss to follow-up (LTFU) and 7 (<1%) died pre-ART, 1041 (31%) were retained in pre-ART care, 706 (21%) initiated ART and were retained in care, and 113 (3%) had initiated ART and were subsequently LTFU (n = 112) or died (n = 1). The proportion of patients retained in care for 2 years did not change over time during the 10-year study period (P = .235). In univariable analysis, education, gender, and marital status were predictors of 2-year retention. In multivariate analysis, secondary education (adjusted hazard ratio [aOR] 1.27; 95% CI: 1.10-1.47), female gender (aOR: 1.28; 95% CI: 1.09-1.50), cohabitation (aOR: 1.31; 95% CI: 1.09-1.57), and residence in Port-au-Prince (aOR: 1.43; 95% CI: 1.09-1.88) were associated with retention in care.

Time to Qualifying for ART by 2013 WHO Guidelines

The median follow-up time was 2.8 years. During the study period (2003-2015), 1474 patients (44%) remained in care until they either initiated ART or met 2013 WHO ART eligibility criteria (see Figure 1). The median time from baseline CD4 count to ART qualification was 1.7 years overall, and 1.2, 1.8, 2.1, 3.2, and 3.0 for patients with CD4 counts from 501 to 600, 601 to 700, 701 to 800, 801 to 900, and >900 cells/mm³, respectively (see Figure 1).

Figure 1.

Kaplan-Meier survival curve of time to antiretroviral therapy qualification. ART, antiretroviral therapy

Discussion

We found that LTFU was very high among patients who did not qualify for ART at HIV diagnosis. The median time to qualifying for ART according to the 2013 WHO guidelines was 1.7 years; after 2 years, 44% of patients were LTFU prior to initiating ART. Over the study period (2003-2015), only 44% of the patients were retained in care until they met 2013 ART eligibility criteria.

Among patients with early HIV disease, there are few reports ^{6 –8} describing the time to CD4 count ≤500 cells/mm³. Fast et al. assessed time to reach this threshold in a cohort of adults with known time of incident HIV infection in east and southern Africa. Of the 554 patients enrolled within 6 months of HIV infection, the majority had CD4 count ≤500 cells/mm³ in less than a year. ⁶ Using Concerted Action on Seroconversion to AIDS and Death in Europe data for individuals with well-estimated dates of seroconversion, Lodi et al. ⁸ estimated a median time of 1.19 years from seroconversion to CD4 count ≤500 cells/mm³. In the Netherlands, Cori et al. analyzed the AIDS Therapy Evaluation in the Netherlands cohort and estimated a mean time of 3.32 years. ⁷

Other studies have also documented low rates of retention in pre-ART care among patients with early HIV disease. Rosen et al. reviewed 14 African studies and found a median of 46% of patients (IQR 31%-95%) were retained from staging to ART eligibility. ¹¹ We found that female gender, secondary education, married or cohabiting living status, and residence in Port-au-Prince were associated with retention in care. This is consistent with the findings from other studies in resource-poor settings; patients with higher socioeconomic status and those with shorter travel distances face fewer barriers to remaining in care. We believe that implementing the new WHO guidelines, with provision of ART to patients with early HIV disease, will improve retention in care. However, further study will be necessary to ensure that high-risk patients are retained in care with this strategy.

Limitations in human and financial resources will create the greatest barrier to the provision of universal ART. However, universal treatment will streamline the ART initiation process by eliminating ART screening requirements. Furthermore, first-line therapy in most resource-poor settings (efavirenz [EFV], tenofovir [TDF], and lamivudine [3TC]) costs <US$150 per year. In China, Wu et al. showed that a simplified test and treatment intervention of concurrent HIV and CD4 testing with immediate initiation of ART regardless of CD4 count was cost-effective after the initial year. ^12,13 Eaton et al. found that expanding ART coverage to all HIV-positive adults would be cost-effective in South Africa, Zambia, Vietnam, and India, over a 20-year period. ¹³

Initiating the WHO recommendation for universal ART may prove difficult in the short-term, however it will streamline ART initiation procedures and likely result in improved retention for those with early HIV infection. Currently, these patients are likely to be LTFU in the short time it takes for them to qualify for ART.