Integration of HIV services with primary care in Yangon, Myanmar: a retrospective cohort analysis.
Tun NN., McLean A., Wilkins E., Hlaing M., Aung YY., Linn T., Ashley EA., Smithuis FM.
<b>Objectives: </b>Integration of HIV care with general healthcare may improve patient engagement. We assessed patient outcomes in four clinics offering HIV care integrated into primary care clinics in Yangon, Myanmar.<br><br><b>Methods: </b>We carried out a retrospective cohort analysis of 4551 patients who started antiretroviral therapy between 2009 and 2017. Mortality and disengagement from care were assessed using Cox regression.<br><br><b>Results: </b>People living with HIV presented late with low CD4 counts [median (25<sup>th</sup> , 75<sup>th</sup> percentile) = 178 (65, 308) from 4216 patients] and advanced HIV (69% with stage 3 or 4). Survival was 0.95 at 1 year and 0.90 at 5 years. Males were at a higher risk of mortality than females [unadjusted hazard ratio (uHR) = 1.6 (95% CI: 1.3-2.0). Patients linked to HIV care via antenatal care or partner/parent notification were at reduced risk of mortality [uHR = 0.4 (95% CI: 0.1-1.0) and uHR = 0.5 (95% CI: 0.3-0.7)] relative to patients who presented for HIV testing. The cumulative incidence of disengagement was 0.06 at 1 year and 0.15 at 5 years. Young adults had a higher risk of disengagement than did children and older patients. Women linked to HIV care via antenatal care services were at increased risk of disengagement relative to patients who came for HIV testing (uHR = 2.4; 95% CI: 1.7-3.4). Mortality and disengagement remained steady over calendar time as the programme scaled up.<br><br><b>Conclusions: </b>HIV care within a primary care model is effective to attain early linkage to care, with high survival. However, close attention should be given to disengagement from care, in particular for pregnant women.