Sputum for Tuberculosis Investigation among Clinic Attendees in Primary Health Clinics in Mangaung, South Africa

Abstract

Background: Tuberculosis (TB) is an infectious disease that can be prevented or cured if diagnosed early. TB can be fatal if not treated. In an effort to diagnose TB early, the WHO recommends universal TB screening in health facilities among clinic attendees. Any clinic attendee who has at least one symptom requires further investigations to confirm if they have TB and they are asked to produce sputum for TB tests. However, there are clinic attendees who fail to produce sputum despite the fact that they might have the disease. Therefore, this study will contribute to the efforts for eliminating TB by investigating factors associated with failure to produce sputum. Methods: This study is a secondary analysis of a cross sectional study that was carried out between November 2017 and June 2018 in three Primary Health Clinics (PHCs) in Mangaung South Africa (SA). If a clinic attendee had at least one symptom or abnormal chest x-ray, they were asked to produce sputum. All clinic attendees who were asked to produce sputum were included in this study, including those who failed to produce sputum. The outcome variable was a binary variable (failure to produce sputum). The main exposure variables were reason for being asked to produce sputum and HIV status. Univariable and multivariable logistic regression models were fitted to examine factors associated with failure to produce sputum. The analysis was done using STATA 17. Factors were considered significant at P<0.05. Findings: A total of 1356 clinic attendees were asked to produce sputum, of whom 530 (39.1%) failed to produce sputum. A higher proportion of females (48.1%) failed to produce sputum compared to males (27.0%). A highest proportion of those aged 18-29 (46.2%) failed to produce sputum as compared to the lowest proportion of those aged 60+ (33.3%). A higher proportion of those who were asked to produce sputum based on the symptom screen only (45.9%) failed to produce sputum compared to those who were asked to produce sputum due xii to both symptoms and the chest x-ray (11.0%) and those who were asked to produce sputum due to the chest x-ray only (33.7%). The proportion failing to produce sputum was higher among People living with HIV (PLHIV) on ART (43.9%) and lower among PLHIV not on ART (23.5%) compared to HIV negative (38.9%) or HIV unknown (31.9%). In univariable logistic regression analysis, one factor was positively associated with failure to produce sputum and four factors negatively associated with failure to produce sputum. Method of screening was positively associated with failure to produce sputum (P<0.001) while HIV/ART status (P=0.001), sex (P<0.001), having had TB previously (P=0.002) and being an ex-miner (P=0.004) were negatively associated with failure to produce sputum. In the multivariable logistic regression analysis, there were three factors associated with failure to produce sputum namely the method by which a clinic attendee was screened positive, sex and being a clinic attendee that was a TB contact. Conclusion: TB prevalence remains high in SA and the crowded health facilities with inadequate resources foster the spread of TB. If a high proportion of the clinic attendees fail to produce sputum this could lead to the spread of TB as some of those who fail to produce sputum may be infected. In addition, if an individual fail to produce sputum this may result in a delay in TB diagnosis which increases the risk of an unfavourable TB outcome. Therefore, the fact that some people fail to produce sputum warrants more work on alternative methods of TB diagnosis.