Table 2.

HRs for sputum culture conversion from start of treatment for incident TB cases^* and excess alcohol use by DOT vs. SAT

	Total† (n = 60 034)‡	Total converted by 4 months n (%)	Time to culture conversion, days, median [IQR]	HR (95%CI)	Adjusted HR§
All
No excess alcohol use	46 925	43 311 (92.3)	43 [25–66]	1.23 (1.22–1.26)	1.20 (1.18–1.23)
Excess alcohol use	12 512	11 082 (88.6)	54 [31–76]	Reference	Reference
SAT
No excess alcohol use	3 392	2 907 (85.7)	45 [26–69]	1.14 (1.01–1.27)	1.11 (0.99–1.25)
Excess alcohol use	308	255 (82.8)	50 [31–76]	Reference	Reference
DOT
No excess alcohol use	30 041	27 751 (92.4)	43 [24–65]	1.27 (1.23–1.30)	1.22 (1.19–1.25)
Excess alcohol use	8 887	7 878 (88.6)	54 [31–75]	Reference	Reference

^*Limited to US cases aged ≥15 years, 1997–2012, with culture positivity, confirmed susceptibility to isoniazid, rifampin, and ethambutol, no known resistance to pyrazinamide, alive at the time of diagnosis, started on four-drug regimen, had documentation of negative sputum, had valid documentation date of conversion, culture converted before stopping treatment, culture converted within 4 months, responded either ‘yes’ or ‘no’ to excess alcohol use.

^†Columns may not sum to 100% due to missing data.

^‡Total with documented conversion.

^§Adjusted for race/ethnicity, sex, age and homelessness. Missing cases excluded from analysis.

HR = hazard ratio; TB = tuberculosis; IQR = interquartile range; CI = confidence interval; DOT = directly observed therapy; SAT =self-administered therapy.