According to the third National Family Health Survey in India, conducted in 2006, the private sector remains a primary source for health care for patients of all socioeconomic levels [1]. A large proportion of patients with tuberculosis (TB) in countries with high TB prevalence first seek care with a private practitioner [2]. In India, it is estimated that 46% of patients with TB seek care in the private sector [3]. However, TB diagnosis and treatment practices among private practitioners in India vary widely and are not formally regulated by the national TB programme [4].

In 2002 India’s Revised National Tuberculosis Control Programme (RNTCP) introduced guidelines for involvement of Private Practitioners (PP) through an initiative called ‘Public Private Mix’ which encompasses training activities and formal collaboration with the RNTCP [5]. RNTCP provides training and encourages PP to adhere to the International Standards of TB Care (ISTC) [6]. PP may also collaborate formally with the RNTCP by serving as a referral facility, diagnostic centre, or treatment centre after entering into a formal memorandum of understanding with the RNTCP.

Despite ongoing efforts to train PP on standard clinical practices, a recent non-representative survey from Mumbai suggests that practices of PP have not improved markedly from those of a decade earlier. [7] In Pune Municipal Corporation (PMC), the urban capital city of the South Indian State of Maharashtra, RNTCP training has been offered and promoted by RNTCP since 2003. However, the impact of efforts made by the RNTCP to influence the quality of TB care in the private sector in Pune remains largely unknown. We sought to assess the influence of RNTCP training on PP in Pune by comparing reported clinical practices relative to ISTC standards for providers based on whether or not they had been exposed to RNTCP training.

Of 268 providers approached, 249 were interviewed (93% participation rate) (Table 1). Of these, 175 (70%) did not have an allopathic medical (Bachelor of Medicine & Bachelor of Surgery, [MBBS]) degree and 154 (62%) had more than 10 years of professional experience. 90% of PP had also evening hours clinic which was assessed to know relationship if any of clinic hours with exposure to training. A total of 136/249 (55%) had been exposed to RNTCP training and 142/249 (57%) were involved with RNTCP. More than 90% of Private Practitioners either work as referring centre or DOT centre. As a referral centre, they refer presumptive TB patients/TB patients for diagnosis/treatment.

Drugs (Patient wise boxes) are made available from programme to DOT centre. Patients come to DOT Centre and takes medicine under observation of Private Practitioners.

Less than 10% Private Practitioners are involved in other schemes like designated Microscopy centres or sputum collection & transport centres.

Proportion of PP with qualification other than MBBS participating in study were 70%. This proportion is similar to that in community where PP possessing qualification other than MBBS are more in number. In contrast to providers with MBBS degrees, providers with other medical qualifications were more likely to have been exposed to RNTCP training (OR 2.4, 95% CI 1.4–4.4).

Providers who are involved with RNTCP were more likely to have been exposed to RNTCP training than providers not involved with RNTCP (OR 5.1, 95% CI 2.8–9.1). There was no association between exposure to RNTCP training and provider sex, years of experience, or clinic hours of operation (Table 2).

Of the 249 participants, 63% (158) indicated that they use microscopy of two sputum specimens for diagnosis of new pulmonary TB. Many providers reported use of additional diagnostic methods for diagnosis of new pulmonary TB as well: 90% (223 out of 249) used chest X-ray, 92% (229/249) used erythrocyte sedimentation rate (ESR), 48% (120/249) used serological tests, and 55% (137/249) used tuberculin skin tests. Odds of reporting investigative methods for diagnosis of pulmonary TB that are in agreement with ISTC guidelines did not differ significantly by exposure to RNTCP training (Table 3).

Response rates varied for questions regarding diagnostic and treatment practices, although differences based on RNTCP training did not reach statistical significance based on 95% CI. (Table 4). PP who had been exposed to RNTCP training tended to adhere to international standards for diagnosis of new pulmonary TB and previously treated pulmonary TB. For new pulmonary TB, 94/136 (69%) of PP who were exposed to RNTCP training adhered to diagnostic standards compared to 64/113 (57%) of providers who were not exposed to RNTCP training (OR 1.7, 95% CI 0.99–2.98). For previously treated pulmonary TB, 99/126 (78%) of PP exposed to RNTCP training adhered to diagnosis standards compared to 69/98 (70%) of providers not exposed to RNTCP training (OR 1.54, 95% CI 0.80–2.96). For diagnosis of extrapulmonary TB, proportions of PP adherent to international standards were similar irrespective of exposure to RNTCP training (Table 3). Associations between RNTCP training and self-report of diagnostic or treatment practices that are in accordance with international standards remained non-significant after stratification by qualification (data not shown).

For treatment of new cases of pulmonary TB, 48/136 (35%) of PP exposed to RNTCP training adhered to treatment standards compared to 36/113 (32%) of PP without exposure to RNTCP training (OR 1.17, 95% CI 0.66–2.05). Overall, 34% (84/249) of PP reported use of only 4 first line drugs for treatment of new pulmonary TB and 38% (94/249) indicated that in addition to four first line drugs they would use one or more second line anti-TB drugs. Overall, 26% (65/249) of PP reported use of fluoroquinolones for new TB. Use of second line drugs to treat new pulmonary TB was greater for PP who had been exposed to RNTCP training compared to those not exposed, though this association did not reach statistical significance (OR 1.5, 95% CI 0.9–2.7).

In a systematic random sample of private practitioners in an urban district in India, we found serious gaps between self-reported diagnostic and treatment practices and international standards, irrespective of prior RNTCP training. Particularly striking was that only 34% of respondents reported use of the international standard anti-TB treatment regimens. It was also noted that a large proportion (26%) of PP reported addition of one or more second-line anti-TB drugs to the first line regimen. This is concerning because inappropriate use of second-line anti-TB drugs for treatment of drug-sensitive TB may promote the emergence of more complicated forms of drug-resistant TB [9].

After a decade of RNTCP training to promote implementation of ISTC among private providers, just over half of private providers reported exposure to RNTCP training. However, we did find that PP who are involved with RNTCP were more likely to have been exposed to RNTCP training than PP not involved with RNTCP, suggesting that training programmes may be one of the interventions for involving PP in collaboration with the RNTCP.

Though policies and contents of the training changed in context with treatment of smear negative cases, it did not affect the study as management of smear negative TB was not part of the study.

The proportion of private providers who have participated in RNTCP activities was similar to that reported from surveys in Kerala and Delhi, but in these settings 40% and 60% of private providers follow RNTCP or international guidelines, respectively [10]–[11].

In general, we found associations between exposure to RNTCP training and self-reported clinical practices consistent with international standards, though these associations did not meet statistical significance. However, this may be due in part to insufficient sample size and power to detect differences at the levels observed, which differed from our sample size estimates. This finding should be interpreted with some caution; a similar study from another large city in South India, Chennai Corporation, did find a significant improvement in self-reported TB management practices after exposure to Public Private Mix (PPM) activities [12].

In the past, very few private practitioners used sputum microscopy among the initial diagnostic tests for suspected pulmonary TB. [8] We found that 63% of private practitioners report use of sputum microscopy for diagnosis of new pulmonary TB, in accordance with ISTC. Our data on use of smear microscopy are consistent with findings from other recent studies, which report a high proportion of providers using smear microscopy as an initial diagnostic tool [11], [12], [14]. However, a large proportion of private practitioners also report use of other tests which are not recommended by the RNTCP for diagnosis in adults, such as tuberculin skin testing or erythrocyte sedimentation rate (ESR) and serological tests. Dozens of distinct commercial serological tests are marketed in many parts of the world specially in developing countries with weak regulatory systems. [15] Serological diagnostic tests for TB were recently banned in India, which may change this particular practice in the future.

Most studies conducted among PP in India show wide variation in self-reported diagnosis and treatment practices, but these have been limited by a lack of representativeness in the provider sample [12], [13], [15], [16]. This study had several limitations. First, only PP registered with professional organizations were eligible for inclusion, so we were not able to assess the clinical practices of those providers not belonging to professional organizations. Assessment of clinical practices relied on self-report and not all participants provided responses to each clinical scenario, therefore we cannot exclude reporting bias. History of participation in RNTCP was by self-report and we were not able to validate this information. We asked providers about participation in training activities at any time during an eight year period; because of this long recall period we cannot exclude recall bias in reporting of activities. In addition, the duration of time between training of PPs and this study was not taken into account in assessing management practices. Some policies and training content changed over the time period of interest; however the primary RNTCP training module published in 2002 was used throughout the period of interest. We were unable to stratify responses by time since training so could not assess any possible effect of waning knowledge after a prolonged gap since training. Finally, we were not able to measure the extent to which trainings were implemented as intended or to verify their content. What providers reported may not correlate to what they did; substantial differences have been reported between self-reported provider practices and objective practice quality assessments.

Strategies that have been used successfully in India to improve adherence by private practitioners to international standards of care include timely provision of professional support services for retrieval of patients lost to follow up and documentation of treatment services, efficient logistics management, and reinforcement of education in standards of care through one-to-one contact between private practitioners and staff of the national TB control programme [12]. Training and sensitization programmes may need to be tailored to individual providers to address specific gaps in practice and modified to incorporate the choices of private providers.

Despite a decade of effort to train PP, a high proportion of private practitioners in Pune report TB management practices not in accordance with international guidelines. Because private practitioners play a significant role in the diagnosis and treatment of TB, approaches to engage the private sector in India are critical. Training and education should focus on raising the minimum standard of care provided to patients diagnosed and treated in the private sector, and reducing the inappropriate use of second line drugs that could lead to more cases of drug resistant TB.