Geographic distribution and predictors of diagnostic delays among possible TB patients in Uganda.

Abstract

Understanding the geographic distribution and factors associated with delayed TB diagnosis may help target interventions to reduce delays and improve patient outcomes. We conducted a secondary analysis of adults undergoing TB evaluation within a public health demonstration project in Uganda. Using Global Moran's I (GMI) and Getis-Ord GI* statistics, we evaluated for residential clustering and hotspots associated with patient-related and health system-related delays. We performed multivariate logistic regression to identify individual predictors of both types of delays. Of 996 adults undergoing TB evaluation (median age: 37 years, IQR 28-49), 333 (33%) experienced patient delays, and 568 (57%) experienced health system delays. Participants were clustered (GMI 0.47-0.64, P ⩽ 0.001) at the sub-county level, but there were no statistically significant hotspots for patient or health system delays. Married individuals were less likely to experience patient delays (OR 0.6, 95% CI 0.48-0.75; P < 0.001). Those aged 38-57 years (OR 1.2, 95% CI 1.07-1.38; P = 0.002) were more likely than those aged ⩾58 years to experience patient delays. Knowledge about TB (OR 0.8, 95% CI 0.63-0.98; P = 0.03) protected against health system delays. We did not identify geographic hotspots for TB diagnostic delays. Instead, delays were associated with individual factors such as age, marital status and TB knowledge.