Table 1.

Number of deaths prevented among 100,000 people in India with direct observation of therapy and increased use of case finding

	Direct observation of therapy	Increase in TB case-finding
Population denominator	100 000	100 000
Incidence of smear-positive TB	0.00043*	0.00032†
Proportion of the diseased population eligible for the intervention: best practice goal minus current practice levels	All those identified are estimated to be subject to direct observation	0.40 (0.284)‡
Baseline risk (case fatality)	7.3% [17]	27% [17]
RRR from intervention	0.06 [12]	0.73 [17]
NEPP (deaths from TB prevented in the next year)	0.188§ (95% CI 0.014–0.294)	1.79¶ (95% CI 1.01–2.69)

NEPP, number of events prevented in the population; RRR, relative risk reduction; TB, tuberculosis.

The examples are based on a population of 100 000, but in reality the actual number and composition of the population to which the results are to be applied would be identified.

*Based on 57% DOTS case detection rate, or 43 per 100 000 per year [13]

†The estimated true incidence rate is 75 per 100 000[13], so 32 per 100 000 is the potential for case finding.

‡40% of cases identified by case detection; proportion of identified cases who will complete treatment = 71%, thus 0.40 × 0.71 = 0.284.

§100 000 × 0.00043 × 0.073 × 0.06 = 0.188.

¶100 000 × 0.00032 × 0.284 × 0.27 × 0.73 = 1.79.