Factors associated with patient and health care system delay in diagnosis and treatment for TB in sub‐Saharan African countries with high burdens of TB and HIV

Abstract

To identify factors causing delayed diagnosis and treatment for tuberculosis in high tuberculosis (TB)/HIV burden African countries. We searched Ovid Medline, CINAHL, PsychInfo, Scopus and Sabinet-South African journals and reference lists, Google, Google Scholar and Google SA for reports in English, ≥1992, of original data from sub-Saharan countries on patient or system delay in TB diagnosis or treatment with populations ≥15 years old. Two reviewers extracted data independently for each study. We categorized independent variables as predisposing, enabling or reinforcing (PRECEDE model). Meta-analysis was conducted for factors associated with delay in ≥2 studies. Of 20 eligible studies, 12 assessed both sources of delay; 1, system delay; and 7, patient delay only. Most were cross-sectional surveys (k = 13) with samples of consecutive patients (k = 13) and bivariate analyses (k = 11). Starting and endpoints for patient delay were consistent, but not system delay. Patient characteristics were studied frequently; HIV stigma and enabling factors were studied infrequently, although the last were most often associated with delay. Consulting traditional healers first--usually by rural residents--consistently led to patient delay; OR = 3.45 (1.91-6.21). Travel time for the return visit was consistently associated with system delay OR = 1.87 (1.378-2.531). We recommend partnerships with traditional healers and research emphasizing HIV and system factors, standard definitions of delay and qualitative and cohort studies to identify enabling and reinforcing factors related to delay.