Rates of refusal of clinical autopsies among HIV-positive decedents and an overview of autopsies in Uganda

Background: Human immunodeficiency virus (HIV)-related mortality remains high in sub-Saharan Africa. Clinical autopsies can provide invaluable information to help ascertain the cause of death. We aimed to determine the rate and reasons for autopsy refusal amongst families of HIV-positive decedents in Uganda. Methods: We consented the next-of-kin for post-mortem examinations among Ugandan decedents with HIV from 2017-2020 at Kiruddu National Referral Hospital. For those who refused autopsies, reasons were recorded. Results: In this analysis, 165 decedents with HIV were included from three selected wards at Kiruddu National Referral Hospital. Autopsy was not performed in 45% of the deceased patients; the rate of autopsy refusal was 36%. The most common reasons for autopsy refusal were time constraints (30%), family satisfaction with clinical diagnosis (15%), fear of disfigurement of the remains (15%), and lack of perceived benefit (15%). By seeking consent from multiple family members and clearly explaining to them the purpose of performing the autopsy, we found a reduction in the rate of autopsy refusal among relatives of the deceased patients at this hospital compared to previous studies at the same site (36% vs. 60%). Conclusions: We found lower rates of autopsy refusal compared to previous studies at the same site. This underscores the importance of clearly explaining the purpose of autopsies as they increase active sensitization about their relevance and dispel myths related to autopsies among the general population. Good, culturally sensitive, and timely explanations to the family of the benefits of autopsy increase the rate of obtaining permission. Building capacity for performing autopsies by training more pathologists and increasing laboratory resources to decrease the turn-around-time for autopsy reports and extending these services to peripheral health facilities could improve autopsy acceptance rates.