Involving private practitioners in the Indian tuberculosis programme: a randomised trial.

OBJECTIVES: To assess a multicomponent intervention to improve private practitioners (PPs) involvement in referral of presumptive pulmonary TB (PTB) cases to the Revised National TB Control Programme (RNTCP) for sputum examination. METHODS: Randomised controlled trial. We randomly allocated all 189 eligible PPs in Tumkur city, South India, to intervention or control arm. The intervention, implemented between December 2014 and January 2016, included two sets of activities, one targeted at health system strengthening (building RNTCP staff capacity to collaborate with PPs, provision of feedback on referrals through SMS) and one targeted at intervention PPs (training in RNTCP, provision of referral pads and education materials and monthly visits to PPs by RNTCP staff). Crude and adjusted referral and PTB case-finding rate ratios were calculated with negative binomial regression. RESULTS: PPs referred 836 individuals (548 from intervention and 169 from control arm PPs) of whom 176 were diagnosed with bacteriologically confirmed PTB. The proportion (95% confidence interval) of referring PPs [0.59 (0.49, 0.68) vs. 0.42 (0.32, 0.52) in the intervention and control arm, respectively], mean referral rate per PP-year [(5.7 (3.8, 8.7) vs. 1.8 (1.2, 2.8)] and smear-positive PTB case-finding rate per PP-year [(1.5 (0.9, 2.2) vs. 0.6 (0.3, 0.9)] were significantly higher in the intervention than the control arm. Stratifying by qualification, a statistically significant difference in the above indicators remained only among GPs and internists. Overall, surgeons, paediatricians and gynaecologists referred few patients. PP referrals contributed to 20% of the sputum smear positive PTB cases detected by RNTCP in Tumkur city (14% were from intervention arm PPs). CONCLUSIONS: We demonstrated the effectiveness of a health system-oriented intervention to improve PP's referrals of presumptive PTB cases to RNTCP.