Effect of COVID- 19 Period Compared to Non COVID-19 Period on the Odds of Pulmonary Tuberculosis at Ndola Teaching Hospital in Zambia

Introduction: There is a belief by researchers that COVID-19 pandemic threatens to reverse recent progress in reducing the global burden of TB disease. This is as result of extreme pressures on health systems, exacerbated by COVID-19, people with TB are likely to face decreased access to diagnostic and treatment services, which might also result in adverse outcomes.

Objective: To estimate the odds of Pulmonary Tuberculosis before and during COVID- 19 period at Ndola Teaching Hospital using the effect size of the odds ratio.

Design and Methods: This was an analytical cross-sectional study on the secondary data from Ndola Teaching Hospital Information system data base. We estimated the odds of Pulmonary Tuberculosis among patients with respiratory tract infections who were referred to Ndola Teaching Hospital during non-COVID-19 period compared to COVID-19 period (2020 March, to 2021 March). The Wilcoxon rank sum test for continuous variables and the Chi squared test for categorical variables were applied to compare the study population by Tuberculosis status. The main research question was addressed by establishing the association between the periods in which the patient was reviewed and the Pulmonary Tuberculosis diagnosis status, adjusted for potential confounders using the logistic regression model.

Results: The results revealed that COVID-19 period compared to non-COVID-19 period reduced the odds of Pulmonary Tuberculosis (OR:0.649, 95% CI 0.448, 0.940, p-value= 0.022).The results further showed that a unit increase in age was associated with reduced odds of Pulmonary Tuberculosis(OR: 0.983,95% CI:0.972, 0.994, p-value= 0.002), equally, there was an indication that being female compared to being male reduced the odds of Pulmonary Tuberculosis(OR:0.987, 95% CI: 0.689,1.413, p-value= 0.942) albeit the estimates were not statistically significant and staying in an urban area compared to living in the sub-urban area had a protective effect against Pulmonary Tuberculosis (OR: 0.358,95% CI: 0.220, 0.583, PV: <0.0001).

Conclusion: Period in which the patient was reviewed is significantly predictive of the odds of Pulmonary Tuberculosis at univariate analysis however at multivariable analysis the effect was insignificant. The reduced effect of the COVID-19 period may be a manifest of screening efforts for TB being reduced to deal with the effects of the pandemic by health providers.

Keywords: Odds, Pulmonary Tuberculosis, COVID-19 period, Non-COVID-19 period.