Background: Tuberculosis burden among the incarcerated population is generally higher than that of general population. Early diagnosis and prompt initiation of treatment are key strategies to contain disease transmission. The aim of this study was to determine the time to treatment initiation among inmates with new smear or Xpert MTB/RIF positive pulmonary tuberculosis and explore risk factors associated with delayed treatment initiation in prison settings.
Methods: We conducted a retrospective cohort study using routine health care data from prison settings in Kzrgyz Republic on new pulmonary tuberculosis patients confirmed by smear microscopy or GeneXpert MTB/RIF during 2014-2019. We computed delay in start of treatment-days from specimen collection to treatment initiation-for exposure variables. We dichotomized treatment delay using 10-day cut-off point,and used logistic regression to identify factors associated with treatment delay.
Results: Among 406 cases included into analysis, the median delay to treatment initiation was 7 days [IQR: 2-16 days]. Using 10-day cut-off, 189 (46.6%) patients had delayed treatment initiation. Treatment delay was negatively associated with smear positivity [adjusted OR (aOR) = 0.44, 95% CI 0.29-0.68] compared to smear negative patients, while patients with isoniazid resistant (aOR = 2.61, 95%CI 1.49-4.56) and rifampicin resistant tuberculosis (aOR = 4.14, 95%CI 2.56-6.77) had increased delay compared to patients who were sensitive for both rifampicin and isoniazid.
Conclusion: Timely diagnosis and effective treatment remain the cornerstone of TB control program populations in the general and in prison settings in particular. Prison authorities need to address all potential areas of delay in TB diagnosis and treatment to strengthen their TB control efforts so that prisons remain free of TB for detainees, prison staff and visitors. These include improved supply of TB drugs, early detection of TB cases and improved collaboration with the health authorities outside the prison system.