Table 5.

Confounder-adjusted association between Axshya SAMVAD exposure and catastrophic costs due to TB diagnosis (outcome) using log binomial regression after accounting for clustering in districts, Axshya SAMVAD study, India, 2016–2017(N = 451)^a.

	Axshya SAMVAD	Non-Axshya SAMVAD
	% (outcome/total)	% (outcome/total)	PR (95% CI)	aPR (95% CI)b
Assuming average household size of 4.8	10.3 (23/224)	11.5 (26/227)	0.89 (0.56, 1.42)	0.68 (0.69, 0.97)c
Sensitivity analysis
Assuming average household size of 3.9	12.9 (29/224)	14.5 (33/227)	0.89 (0.56, 1.41)	0.72 (0.53, 0.97)c
Assuming average household size of 5.5	8.5 (19/224)	10.1 (23/227)	0.84 (0.47, 1.49)	0.69 (0.50, 0.94)c

TB: tuberculosis; SAMVAD: sensitisation and advocacy in marginalised and vulnerable areas of the district; Axshya SAMVAD: an active case finding strategy under project Axshya implemented by The Union, South East Asia office, New Delhi, India, across 285 districts of India; Non-Axshya SAMVAD: patients detected through passive case findings; PR: prevalence ratio; aPR: adjusted prevalence ratio.

^aCosts due to TB diagnosis were more than 20% of pre-TB annual household income.

^bComplete case analysis was performed; model building (log binomial) by backward stepwise method. Age, sex, monthly income per capita, education, history of weight loss and distance of residence from microscopy centre were the confounders adjusted for.

^cStatistically significant.