Factors associated with diagnostic and referral delay of patients with epidemic kaposi sarcoma for care at the Uganda Cancer Institute, Kampala

Abstract

Background: Epidemic Kaposi sarcoma (EKS) is the most prevalent cancer in sub-Saharan Africa. In Uganda, EKS patients generally present with advanced stage disease, which is associated with poor treatment response and survival outcomes. Understanding barriers to care could guide interventions aimed at down-staging EKS presentation. Aims: To characterize delays in EKS treatment and describe the factors associated with them in Uganda. Methods: We recruited HIV positive adults with a histological diagnosis of KS referred to the Uganda Cancer Institute (UCI). Patient interviews and chart abstraction were carried out to determine the times at the different points along an EKS patients’ path to care and the factors associated with late presentation. We also carried out key informant interviews (KII) at HIV clinics to identify provider and health system factors that affect the above time gaps. Results: Of 180 patients enrolled, average age was 37 years (range, 19-76); 64% were men. At diagnosis, 129 (71%) participants had advanced tumor stage (T1). Over 95% (172) of participants reported knowing their HIV status, and 91% (165) were enrolled in HIV care at KS diagnosis. Only 22% of participants had some knowledge about EKS at cancer diagnosis. Median time from the participant noticing the first KS lesion to initiating treatment at the UCI was 224.5 days (IQR, 97, 450). Median delay times included: “Patient Delay” (from first seeing a KS lesion to presenting to a health worker) of 61 days (IQR, 14-186), “diagnostic delay” (from first presenting to health worker to receiving biopsy results) of 63 days (IQR, 9-200), “referral delay” (from diagnosis to referral to the UCI) of 1 day (IQR, 0-7), and “treatment delay” (from presentation at UCI to chemotherapy initiation) of 14.5 days (IQR, 6-28). A high baseline CD4 cell count (>200cells/μl) (IRR=1.3, p=0.001) and low household income (<100,000/=) (IRR=1.9, P=0.04) were associated with advanced (T1) disease at multivariate analysis. In the KII, a high patient-clinician ratio, inability to perform biopsies and lack of guidelines for referral of EKS cases were reported as contributors of delay.