PLoS ONE (Jan 2013)
Do diagnosis delays impact receipt of test results? Evidence from the HIV early infant diagnosis program in Uganda.
Abstract
BACKGROUND: There is scant evidence on the association between diagnosis delays and the receipt of test results in HIV Early Infant Diagnosis (EID) programs. We determine the association between diagnosis delays and other health care system and patient factors on result receipt. METHODS: We reviewed 703 infant HIV test records for tests performed between January 2008 and February 2009 at a regional referral hospital and level four health center in Uganda. The main outcome was caregiver receipt of the test result. The primary study variable was turnaround time (time between sample collection and result availability at the health facility). Additional variables included clinic entry point, infant age at sample collection, reported HIV status and receipt of antiretroviral prophylaxis for prevention of mother-to-child transmission. We conducted a pooled analysis in addition to separate analyses for each facility. We estimated the relative risk of result receipt using modified Poisson regression with robust standard errors. RESULTS: Overall, the median result turnaround time, was 38 days. 59% of caregivers received infant test results. Caregivers were less likely to receive results at turnaround times greater than 49 days compared to 28 days or fewer (ARR = 0.83; 95% CI = 0.70-0.98). Caregivers were more likely to receive results at the PMTCT clinic (ARR = 1.81; 95% CI = 1.40-2.33) and less likely at the pediatric ward (ARR = 0.54; 95% CI = 0.37-0.81) compared to the immunization clinic. At the level four health center, result receipt was half as likely among infants older than 9 months compared to 3 months and younger (ARR= 0.47; 95% CI = 0.25-0.93). CONCLUSION: In this study setting, we find evidence that longer turnaround times, clinic entry point and age at sample collection may be associated with receipt of infant HIV test results.