Alliance for Health Policy and Systems Research

Alliance for Health Policy and Systems Research Research to Policy Abstract Evolving a Public-Private Mix for DOTS in rural India: Early outcomes and lessons Rasalpurkar S,Juvekar SK, Morankar SN, Sheikh K, Rangan S, Porter J, & Uplekar M rkmutatkar@hotmail.com drsnm@pn3.vsnl.net.in Problem In the past year, researchers from a NGO have been involved in initiating and coordinating a project to involve private medical practitioners (PPs) in the DOTSbased Revised National TB Control Programme. This PPM initiative was implemented in one TB Unit with a population of 350,794, 14 public health facilities and 100 PPs in rural Pune, one of the more developed districts of Maharashtra, India. The NGO helped to develop a mutually acceptable referral system by sensitising public and private health providers and attempted to bridge gaps between them through joint meetings. This process resulted in the flow of 56 referrals from 23 PPs to the RNTCP over a 3-month period, and 4 PPs became DOT providers for 11 patients. Demonstrable links between the private and public sector were thus created. PPs appeared more willing to collaborate than public health personnel. The collaboration was found to be dependent on the external facilitating agency and the need to devise a suitable exit strategy early in the process was perceived. The paper elaborates on lessons learnt during this process and examines roles and responsibilities of all actors in developing a sustainable PPM for DOTS implementation. Morankar, Involving Private Practitioners in National Tuberculosis Programme in Rural Areas of Western India: an Action-Research to aid

Policy Development. Center for Health Research and Development, Maharashtra Association of Anthropological Sciences, Pune, India. Background The Public Private Mix (PPM) Project in Pune District, was the only rural PPM project supported by the Alliance for Health Policy and Systems Research, along with the urban PPM projects in Vietnam, Kenya and Delhi. The project was initiated in April 2001 with the objective of developing a sustainable model for involvement of private practitioners (PPs) in the Revised National Tuberculosis Control Project (RNTCP) in Pune district in collaboration with the District Tuberculosis Centre (DTC), Pune. Using a participatory approach, the project was successful in developing tools for involvement of PPs in a PPM initiative in the RNTCP in the Bhor TB Unit area. A mutually acceptable referral system was set up, which allowed PPs to use the diagnostic services of the RNTCP, refer suspected patients for diagnosis and management and act as Directly Observed Treatment (DOT) provider for their confirmed patients. Findings At the end of the project, a total of 77 suspects had been referred by 30 PPs, of which 51 had been confirmed as TB cases and started on treatment. Seven PPs were acting as DOT provider for 12 patients and three patients had successful treatment outcome. Return

IJTLD-03-03-0092 Title (R1): Tuberculosis Control in Rural India: Lessons from a Public-Private Collaboration Authors: S Rangan i, S Juvekar i, S Rasalpurkar i, S Morankar i, Joshi A ii, J Porter iii Running Head: Public-Private collaboration for TB control Word Count (excluding References, Tables and Figures): 3886 Keywords: TB control, public-private mix, rural, India Summary (276 words): The Revised National TB Control Programme (RNTCP) in India encourages public-private partnerships to assist in the rapid expansion of the Directly Observed Treatment, Short course strategy. A model partnership to link rural private medical practitioners (PMP) with the public health services was developed in a rural TB Unit (population of 350,794) in Pune district of Maharashtra State in India, through a collaborative project, which included 100 of the 113 PMPs in the study area, the district health and tuberculosis staff and a nongovernmental organisation (NGO) that facilitated the process. Participatory research methods were used to plan and implement the partnership. The process of creation and implementation of the partnership was analysed using both quantitative and qualitative research methods. Five months after implementation of the model, (30%) of PMPs had referred patients to the RNTCP, which contributed to 30% of the cases detected in the TB Unit. A plan was then created for the NGO to withdraw from the partnership and for its role to be taken up by the public health services. A monitoring visit by the NGO six months later, showed that referrals from the private sector to the public health services were continuing, albeit to a lesser extent, but there was a breakdown of the communication and the documentation systems created to link the two sectors. The project highlights the importance of organisational and individual commitment to these partnerships, the key roles of the District Health Officer and the District Tuberculosis Officer in guiding and supporting these i Centre for Health Research and Development (CHRD), a unit of the Maharashtra Association of Anthropological Sciences (MAAS), Pune, India ii Member Secretary, District TB Control Society, Pune (Maharashtra) between June 2001 and May 2002 iii London School of Hygiene and Tropical Medicine

initiatives from the public sector, the potential role of process and outcome indicators in monitoring partnerships and the important role of NGOs as intermediaries and facilitators. Introduction India s integrated National Tuberculosis (TB) Programme (NTP) was established in 1962. 1 It was based on primary health care principles and remained entirely dependent on the general health services for its success and survival. The early successes of the programme were countered by falling investment in the delivery of health services in the 1970s at the community level. 2,3,4 This was accompanied by a rapid growth of the private medical sector, prompted and promoted by several other factors, including the willingness of the growing middle class dissatisfied with the public sector to pay for private health care. 5 For TB patients, the comparative advantage of attending a private medical practitioner (PMP) can be easily recognised: easier accessibility, shorter waiting times, availability of doctors and drugs, considerate staff attitudes and sometimes credit or payment in instalments. 6 It is estimated that the private sector in India, which comprises close to 80% of the eight million registered practitioners in the country, handles close to a sixth of the world s TB patients. 5 In 1991, a landmark study described the TB management practices of the private general practitioners in Mumbai and observed problems with both diagnosis and treatment. 7,8 A number of studies carried out since then, have described the characteristics of the PMPs and the quality of services provided in the private medical sector, all of which could negatively affect the epidemiological impact desired by the global TB control efforts. 9,10,11 They underscore the need for a systematic approach to the management of TB by the PMPs and indicate the importance of including the private medical sector in the development of the health care delivery system in India. 12,13,14 Partnership in the health sector has been described by WHO as a process of bringing together a set of actors for the common goal of improving the health of populations based on mutually agreed roles and principles. 15 Since the mid 90s, the World Health Organization (WHO) has been supporting and undertaking studies to document the need, feasibility and effectiveness of the involvement of PMPs in national TB control efforts. 11,14,16,17,18 As part of the development of the Revised National TB Control Programme (RNTCP) in India, the

Central TB Division (CTD) iv has begun to encourage the development of public-private partnerships to assist in the rapid expansion of the DOTS strategy. 19 Efforts in different parts of the world are also currently underway, to document the various methods used in initiating and sustaining the public-private partnerships, most of these in the urban context. 17,18,20,21 Several of these experiments have demonstrated success, though documentation regarding the processes and outcome of such partnerships are few. 20 This paper is based on a initiative to involve PMPs in the local RNTCP and reports on the creation of a model for linking the public and private sectors in one TB Unit (TU) v, in a rural area of Maharashtra State in India. The processes used to link PMPs with the public sector in the project are described and the lessons learnt through this experiment discussed. Methods The study was conducted at the Bhor TU in the Pune District of Maharashtra State in Western India between April 2001 and March 2002. The research work involved both quantitative and qualitative research methods undertaken by the NGO, which facilitated the process of creating the linkage and monitoring of the project. At the end of March 2002, the NGO withdrew its facilitation and monitoring support, as planned, and handed this over to the RNTCP. A further monitoring visit was conducted by the NGO in September 2003. Organisation of the RNTCP in the study area Information on the study area was collected from local sources including the population census of 2001, reports from the TU and the State TB Office. Additional information on the RNTCP, including technical and operational guidelines were obtained from the Central TB Division in Delhi. 21,22 iv It is suggested that PMPs could improve TB care by ensuring prompt referral of patients with cough for three weeks or more for sputum smears, by providing reassurance that TB can be cured, by giving only RNTCPrecommended drug regimens, and by starting treatment with rifampicin-containing regimens under observation. In addition, it is suggested that some PMPs might opt to serve as treatment providers for patients, have their laboratories included in the quality control network of the RNTCP and/or serve as microscopy centres for the RNTCP v The sub-district level unit of TB control within the RNTCP

Creation of the Public/Private Partnership Model (April October 2001) A participatory approach was used to develop the partnership and to create the public-private model. Initially, three meetings were held between June and July 2001, to familiarize the PMPs with the principles and the working of the RNTCP (technical and operational guidelines of the programme), to enable them to meet the staff who ran the public health facilities, to realize the opportunity presented to them to collaborate with the RNTCP and to avail themselves of the free TB diagnosis and treatment offered by the RNTCP for their patients. 21,22 Public/private partnership process Qualitative information: A series of three follow-up meetings were held between July 2001 and March 2002 to create the processes for the study and to determine the methods for evaluation. The first follow-up meeting was used to develop the referral system. This included the creation of the tools to help formalize, document and monitor the partnership, eg the letter of agreement to participate, the referral and back referral forms vi (See Figure 1). The model was then introduced into the field in October 2001. The second meeting was held six weeks after the model had been introduced. It provided an opportunity for feedback on what was happening, to modify the logistics relating to the referral system, and to discuss the future monitoring of the project. In the third meeting held during the last month of the project, there was for a general discussion on the achievements of the project and about the long-term sustainability of the public-private partnership. The events of all the meetings were recorded in log books and analyzed qualitatively for themes. 23 Participant and non-participant observation was used to collect further information on the meetings. 24 These observations frequently overlapped with unstructured interviews. 25 Quantitative information: The project staff from the NGO monitored the referrals from the PMPs for five months after implementation of the referral system in October by visiting the PMPs and the public health facilities. The data regarding the management of the patients by the RNTCP were collected from the duplicate back referral forms used by the RNTCP. Validation of these data was conducted through examination of the laboratory and district TB vi The back referral form was the system devised within the model to provide feedback on the referred patient to the referring PMP by the RNTCP (See Fig 1)

registers of the RNTCP. Data on patients, number and type of referrals, were entered and analyzed using Microsoft Excel 2000. Public/private partnership indicators In addition to the use of the qualitative information collected from the meetings, the process of the partnership was also studied through the use of indicators. Process indicators (eg number of PMPs agreeing to participate, number of PMPs acting as DOT providers) were used to monitor the process of linkage of the two sectors. In addition, outcome indicators (eg number of confirmed TB patients taking DOT from PMPs) were created to help determine the overall outcome and success of the project. Six month monitoring visit In September 2002, six months after the withdrawal of the NGO, a monitoring exercise was undertaken by the NGO staff to collect data on the referrals from the private sector to the RNTCP. Visits were made to the PMPs and to the public health facilities. A total of 25 PMPs (those who had referred at least one patient to the programme during the monitoring period through the formal referral system) and 44 non-participating PMPs (those who had not referred any patients to the programme through the formal referral system), were interviewed using a semi structured interview schedule with open-ended questions. 24 The questions addressed: the demographic characteristics of the PMPs practice (eg qualification, number of patients, address), the management of TB patients, and perception with regard to the partnership. In addition, a quantitative analysis was conducted of the referral and back referral forms and linked to examination of the laboratory and district TB registers. Results Organization of the RNTCP in the study area At the field level of the health care infrastructure for the RNTCP in Maharashtra, TB diagnosis and treatment is provided through the primary health care infrastructure consisting of the rural hospitals (RH), primary health centres (PHC) and its outreach programme consisting of sub-centres and community level health workers. Administrative control of the staff at the RHs and PHCs remain with the Civil Surgeon and the District Health Officer (DHO). The responsibility for planning and financial management of the TB programme is given to the District TB Control Society (DTCS), of which the District TB Officer (DTO) is

the member secretary. The DTO and his team at the District TB Centre (DTC) are charged with the implementation of the programme in the entire district through the primary health infrastructure. A supervision and monitoring unit called the TB Unit (TU) has been added on at the sub-district level (population of 500,000), consisting of a senior treatment supervisor and a senior TB laboratory supervisor appointed by the DTCS under the control of a local Medical Officer in charge of the TU. Within the Bhor TU (population 350,794, 2001 census), there were 113 PMPs, three X-ray centres (located in RHs), four microscopy centres located in three RHs and one PHC and 66 treatments centres located in PHCs. The Public-Private Partnership Model The intervention model and the tools for the referral and documentation system emerged during the follow up meetings (See Figure 1). The referral system was as follows: if a PMP agreed to participate in the partnership, s/he began by signing the letter of agreement and then used the referral form to send her/his patients to the RNTCP for sputum microscopy and categorization. On this form, s/he also indicated whether s/he would like to act as the Directly Observed Treatment (DOT) provider and therefore to receive the patient s treatment card and box of medications. Some PMPs were sceptical about the partnership; one 50 year old allopathic PMP commented signing of an agreement will be a futile exercise..i do not believe that the public health services will change their ways of working. The RNTCP staff agreed to send the PMP feedback on the patient regarding the result of the sputum examination, the diagnosis, treatment advised and follow up needed, by using the back referral form. In addition, if the PMP wanted to be a DOT provider, a health worker would be sent to visit them with the patient s treatment box and card. The Public-Private Partnership Process Eighty-seven of the 113 PMPs practicing in the area were personally met by the project staff of the NGO and invited to attend conveniently organized orientation meetings in their areas of residence/practice. Forty-six (53%) of these 87 PMPs attended the initial orientation meetings. A further 13 PMPs were met after the meeting and the partnership explained, taking the total number of contacted PMPs to 100. The remaining 13 PMPs were not available despite repeated attempts to contact them. A major concern of the PMPs was regarding the handling of private patients by the staff of the public health facilities. An allopathic practitioner with a 25-year old practice remarked, Patients referred by PMPs

should be treated properly by the staff of the public health services particularly during the diagnosis period. They should never be sent back without doing the necessary tests In the second round of follow-up meetings, both PMPs and the public health staff shared their experiences with the partnership. For example, the PMPs felt that they needed to meet with the Medical Officers of PHCs every month to discuss the clinical and operational issues and problems. The RNTCP and the public health staff unanimously agreed that there was a need for somebody from the programme to visit the PMPs and get feedback through regular visits and that the male and female multipurpose workers were felt to be the appropriate persons for this undertaking this responsibility. The response of the public health services in the process of initiation and implementation of this model through the meetings and field-work, was poor. Though the District Tuberculosis Officer (DTO) and his team were enthusiastic and supportive of the partnership, this did not translate into cooperation from the public health staff in the Microscopy and Treatment Centres because the project was not seen to be a priority by the District Health Officer (DHO), who was their supervisor. The project staff, hence, faced several administrative delays in obtaining permission to visit and sensitise the staff of the public health facilities and to introduce the referral tools. This resulted in the NGO staff shouldering the responsibility of ensuring maintenance of documentation at both the public facilities and the PMPs during the entire period of the initial five months of monitoring, from November 2001 to March 2002. Though the roles and responsibilities of the RNTCP and the public health functionaries were discussed and agreed upon in the creation of the model, these were not formally documented and staff members were not informed appropriately. This was principally due to a lack of commitment and cooperation from the DHO. At the last follow-up meeting, all the participants (PMPs, RNTCP staff, public health staff, and NGO staff) expressed an interest in the continuation of the project. The NGO was asked to continue to act as the facilitating agency. An allopathic practitioner declared, This initiative will not continue if there is no one to keep the interest of the PMPs and the PHC staff alive.

The DTO agreed to support the continuation of the public-private model by offering to print the referral forms. However, two months after withdrawal of the NGO, the DTO was transferred and the Senior Treatment Supervisor in charge of the TU quit his job, resulting in a lack of direct support to the partnership. The Public-Private Partnership Indicators Table 1 provides the process indicators of this public-private partnership. A third (33%) of the PMPs contacted, agreed to formally participate in the partnership by signing a letter of agreement to participate and a majority (91%) of these PMPs actually referred patients. The key outcome indicators after five months are provided in Table 2. The partnership contributed 30% to the case detection by the TU in the five-month monitoring period, though there was no observed change in case detection in the TU when compared to the same period in the previous year. Six month monitoring visit Six months after its withdrawal from the partnership in March 2002, the NGO conducted a monitoring visit. This showed an overall reduction in participation by the PMPs. Almost all PMPs had stopped using the referral forms for reasons such as the forms getting misplaced or running out of forms, and were using their letterheads to refer patients. The public health facilities were erratic in sending back referral forms to the PMPs and documentation regarding referrals by five participating PPs was missing at the public health facilities. In the enquiry conducted among the PMPs, four of the 44 non-participating PMPs also reported referring their patients routinely to the public health facilities, but since they had failed to use the referral forms, there was no documentation. Table 3 provides the comparative data on referrals during the study period and six months after withdrawal of the NGO, showing a reduction in all the indicators monitored. There was no difference in the age and educational profile of the 25 participating and 44 nonparticipating PMPs interviewed, though it was seen that distance from a public health facility was one of the prime reasons for their inability to participate. Seventeen (68%) of the participating PMPs interviewed were practising within one kilometre from a public health facility, while 21 (48%) of the non-participating PMPs interviewed were located more than five kilometres from a public health facility. PMPs were not participating because they had

not received the expected response from the RNTCP for their first referral and because of the inconvenience imposed on their patients by the RNTCP (eg multiple visits for diagnosis and inconvenient location of health facilities in the TU). Only 6 (24%) of the 25 participating PMPs interviewed, reported receiving any visits from the public health functionaries. Thirteen of these PMPs (52%) had not faced any problems with the partnership, while 12 (48%) complained about the delays between referral, diagnosis and starting of treatment at the public health facilities. Ten (40%) also expressed disappointment of the public health facilities at the lack of feedback regarding referred patients. Regular meetings between partners, as organised by the NGO, were identified by 23 (92%) of the PMPs as the key to improving and sustaining the partnership. It was clear from the interviews that the withdrawal of the NGO from its facilitating role without giving adequate time for complete integration of the project in the RNTCP, and the poor commitment of the local administration to sustain the partnership had been responsible for the breakdown of communication between the partners, thereby negatively impacting on the partnership. Discussion The public-private mix (PPM) approach in TB control has been acknowledged as one of the important ways to reach the global targets for case detection, which have been eluding TB programmes and posing the biggest barrier to achieving the global TB control targets for 2005, as accepted by the Amsterdam Declaration. 27 Attempts are being made in several countries to implement the PPM and assess its impact on case detection, though most of these have been on a pilot basis and in urban areas. 28 This project was an exploratory attempt by an NGO to link the public and private sectors in the creation of a model for TB control in a rural area. Although only 30% of the PMPs in the area actively participated by referring patients, a system was developed that could have encouraged a more permanent link between these different organizations so that TB control could be effectively, and wholly achieved in the area. There were several important lessons from the partnership. First, the process is difficult to initiate and even more difficult to sustain. All organisations and individuals need to be

committed to making the partnership work, particularly the key actors like the DTO and DHO as well as the managers of the NGO. Without commitment from key staff, who can make changes in the large bureaucratic systems of the public sector, these initiatives will fail. The second lesson is that the public health system structure and management is crucial to sustaining public-private partnerships. A project of this kind highlights the problems and difficulties contained within the public health infrastructure, and provides an opportunity to address them. The general health services staff (all managers and field level staff) need to show initiative and be involved right from the planning stage of the project. This will ensure that the capacities and constraints of the general health services, through which the RNTCP is implemented, are considered as the operational model is being developed. The DTO and DHO are crucial in this process. The reasons reported by PMPs for non-participation in the partnership (eg. poor/non response from the public health services towards their referrals, and the inconvenience imposed on their patients by the RNTCP requirements) point to the need for adequately preparing the public health services so as to provide the expected response to the PMPs. The breakdown in the referral and documentation systems and the dialogue between the private and public sectors after the withdrawal of the NGO shows that longer periods of facilitation may be essential to allow integration of the partnership in the RNTCP. There is, however, also a need for more discussion and greater commitment from the local health administration, so that the facilitating NGO s role could be incorporated into the management structure of the TB programme. The distance from facilities seems to be a strong indicator of non-participation and needs to be explored further and addressed in any public-private partnership. This is particularly true for the rural context, where physical access is a problem both for providers and patients. Identifying and accrediting private facilities for diagnosis may be one way of solving this, though this has also to be accompanied by increased involvement of the community-based public health functionaries. With the introduction of the guidelines for the involvement of the private sector in the RNTCP in India, there are reports and plans for the use of accredited private laboratories, though these are yet to be formally evaluated. 19,26,29 There is also a need to look at other factors, which influence referral of private patients to the public health

facilities and understand the profile of private patients accessing TB care in the public health sector. The third lesson is that methods for following this type of endeavour need to be created and process and outcome indicators, as used in this partnership, developed as a potential way forward. An approach to determining the success of a programme of this type is to interrogate process indicators. These indicators help to measure the smooth functioning of the links between public and private sectors and NGOs, user satisfaction with the system of referrals, and finally the general quality of patient care. The indicators may be more useful in determining programme effectiveness in the long term than by simply comparing the number of TB patients enrolled at different times of the year as new initiatives get underway. The need to involve PMPs and NGOs in the RNTCP in India is recognized and guidelines for this involvement defined. 19 NGOs can play an important role in facilitating and monitoring and most of the ongoing initiatives at linking the public and private sectors have included a facilitating agency for initiating the project and for subsequent co-ordination and monitoring, but there needs to be clarification about their long term contribution to the partnership. 26 There have been reports of successful collaborations between the public and private health providers in urban areas but little from rural populations. For example, Murthy and colleagues have reported a successful experiment in Hyderabad city, where PMPs are involved in TB diagnosis and treatment through an intermediary organization, which, in this case, is a specialist hospital in the city, without much involvement of the local RNTCP structures. 20 At what point does the public sector take control of this role; indeed, should they take control of this role or should there simply be an ongoing process of linkages and support through other organizations, are questions yet to be answered. At the end of the project, it was evident that the continuous monitoring by the NGO during the project period had prevented the integration of the public-private model into the RNTCP. As long as both the PMPs and the staff of the public health facilities felt that the agency was present, they did not perceive the need to take over these responsibilities from the agency and to integrate this role into the RNTCP. The NGO acted as a catalyst and creator of this project, but the long-term sustainability of these projects depends on the management and organisation of the government structures and their ability to initiate collaborations with different organisations and philosophies.

Practical tools to assist in the integration of the public-private partnership model into the RNTCP are now urgently needed. What is also important is developing the capacity of the public health providers to initiate and forge collaborations with the PMPs through publicprivate partnership training modules. These may include issues about collaboration as well as the development of suitable local operational partnership models, which take into consideration local resources and needs. The key to sustaining such partnerships in rural areas, thus, appears to be strengthening the RNTCP to respond to the diagnostic, treatment and monitoring demands of the partnership, and increasing the involvement and cooperation of the local general health services to ensure sustainable relationship between the partners. Acknowledgements The authors would like to thank all the participants in the project from the public and private sectors in Bhor TU and Pune district. The project would not have been possible but for the financial support provided by the Alliance for Health Policy and Systems Research, Geneva and the technical support provided by WHO, Geneva, particularly by Dr Mukund Uplekar, PPM Task Manager, Stop TB. We are also grateful to Dr AK Chakraborty, Epidemiology Analyst, Bangalore, and Dr Karina Kielmann, Medical Anthropologist, London School of Hygiene and Tropical Medicine for their comments and suggestions on the earlier drafts of the paper. References 1. Banerji D, Andersen S. A sociological study of awareness of symptoms among persons with pulmonary tuberculosis. Bull Wld Hlth Org 1963; 29: 665. 2. Ogden J, Rangan S, Lewin S, ed. Tuberculosis control in India - a state-of-the-art review. London School of Hygiene and Tropical Medicine, London, 1999: pp 22. 3. Krishna Murthy VV. Evaluation of the performance of the National Tuberculosis Programme during the VII Plan. Ind J Tub 1993; 40(4): 129. 4. Rao Sujata. Economic aspects of tuberculosis control in India. Radical J Hlth 1995; 1(4): 264.

5. Uplekar M. Involving the private medical sector in Tuberculosis Control. In Porter JDH, Grange JM, ed. Tuberculosis an Interdisciplinary Perspective. Imperial College Press, London, 1999: pp 193. 6. Brugha R, Zwi AB. Tuberculosis treatment in the public and private sectors - potential for collaboration. In Porter JDH, Grange JM, ed. Tuberculosis an Interdisciplinary Perspective. Imperial College Press, London, 1999: pp 167. 7. Uplekar MW & Shepard DS. Treatment of tuberculosis by private general practitioners in India. Tubercle 1991; 72: 284. 8. Uplekar M & Rangan S. Private doctors and tuberculosis control in India. Tubercle Lung Dis 1993; 74: 333. 9. Uplekar MW, Juvekar SK, Parande SD, Dalal DB, Khanvilkar SS, Vadair AS, Rangan SG. Tuberculosis management in private practice and its implications. Ind J Tub 1996; 43: 19. 10. Singla N, Sharma PP, Singla R, Jain RC. Survey of knowledge, attitudes and practices for tuberculosis among general practitioners in private clinics in India. Int J Tuberc Lung Dis 1998; 2(5): 384. 11. Uplekar M, Juvekar S, Morankar S, Rangan S, Nunn P. Tuberculosis patients and practitioners in private clinics in India. Int J Tuberc Lung Dis 1998; 2: 324. 12. Bhat R. The private /public mix in health care in India. Hlth Pol Plan 1993; 8(1): 43. 13. Swan M, Zwi A. Private practitioners and public health: close the gap or increase the distance. Department of Public Health and Policy, Departmental Publication No 24: London School of Hygiene and Tropical Medicine, London, 1997. 14. Pathania V, Almeida J, Kochi A. TB patients and private-for-profit health care providers in India. WHO/TB/97.223.1997. World Health Organization, Geneva, Switzerland, 1997. 15. Kickbusch I, Quick J. Partnerships for health in the 21 st Century. World Health Statistics Quarterly 1998; 51(1): 68-74. 16. World Health Organization / World Bank. Draft Report on Protocol Development Workshop to Re-assess the Role of the Private Sector in the Management of Tuberculosis in High Prevalence Areas, held by The Operational Research Unit, Tuberculosis Programme, World Health Organization, Geneva in Collaboration with The Foundation

for Research in Community Health, Bombay, India, 27-30th September 1994. WHO, Geneva. (Unpublished), 1994. 17. World Health Organization. Involving Private Practitioners in Tuberculosis Control: Issues, Interventions and Emerging Policy Framework. WHO/CDS/TB/2001.285. World Health Organization, Geneva, 2001. 18. Uplekar M, Pathania V, Raviglione M. Private practitioners and public health: weak links in tuberculosis control. The Lancet 2001; 358 (9285): 912. 19. Central TB Division. Involvement of Private Practitioners in the Revised National Tuberculosis Control Programme. Central TB Division, Directorate General of Health Services, Ministry of Health and Family Welfare, New Delhi, 2002. 20. Murthy KJR, Frieden TR, Yazdani A, Hreshikesh P. Public private partnership in tuberculosis control: experience in Hyderabad, India. Int. J Tuberc Lung Dis 2001; 5(4): 354. 21. Central TB Division. 1997. Revised National Tuberculosis Control Programme: Technical Guidelines for Tuberculosis Control. Directorate General of Health Services, Ministry of Health, Government of India, New Delhi, India. 22. Central TB Division. 1997. Revised National Tuberculosis Control Programme: Operational Guidelines for Tuberculosis Control. Directorate General of Health Services, New Delhi, India. 23. Silverman D. Doing qualitative research. London. Sage Publications 2000. 24. Bernard HR. Research methods in cultural anthropology. London. Sage Publications 1988. 25. Hurtig AK, Pande SB, Baral S, Newell J, Porter JDH, Bam DS. Linking private and public sectors in tuberculosis treatment in Kathmandu Valley, Nepal. Hlth Pol Plan 2002; 17(1): 78-89. 26. Rangan S. The Public-Private Mix in India s Revised National Tuberculosis Control Programme: an update. J Ind Med Assoc 2003; 101(3): 159. 27. Dye C, Watt CJ, Bleed D. Low access to a highly effective therapy: a challenge for international tuberculosis control. Bull World Health Organ 2002; 80: 437.

28. Uplekar M. Involving private health care providers in delivery of TB care: global strategy. Tuberculosis 2003; 83: 156-164 29. Rangan S, Ambe G, Borremans N, Zallocco D, Porter J. The Mumbai experience in building field-level partnerships for DOTS implementation. Tuberculosis 2003; 83:165-172. Table 1: The Public-Private Partnership Process Indicators Process Indicators Number Total Percentage of PMPs PMPs (N) Contacted by the project staff (out of those practicing 100* 113 88 in the area) Oriented initially (out of those invited) 46 87* 53 Formally agreeing to collaborate by signing letter of agreement (out of those contacted): Agreeing to only refer patients (out of those signing) Agreeing to refer patients and act as DOT provider 33 17 16 100 33 33 33 52 48 (out of those signing the letter of agreement) Participating by referring suspects (out of those 30 33 91 agreeing to collaborate) Acting as DOT providers (out of those participating by referring suspects) 7 30 23 Note: * Those PMPs who could be contacted before the orientation meeting were those that were oriented to the partnership initially, 13 were subsequently met and oriented in their clinics, thereby taking the total number of contacted PMPs to 100

Table 2: Outcome Indicators of the Public-Private Partnership Outcome Indicators Number Total Percentage Population covered by the project (out of TU 350,794 350,794 100 population) Total symptomatics referred by PMPs during 77 - - the monitoring period* Patients diagnosed as TB (out of referrals 51 77 66 from PMPs during the monitoring period) New sputum positives (out of diagnosed TB 18 51 35 cases referred by PMPs during the monitoring period) Patients receiving DOT from PMPs (out of 12 51 23 diagnosed TB patients referred by PMPs during the monitoring period) Contribution by PMPs to the case detection 51 167 30 by the RNTCP in the TU during the monitoring period Contribution of sputum positive cases by PMPs to the sputum positive case detection by the TU during the monitoring period 18 79 23 Note: * Monitoring Period: 5 months (November 2001-March 2002)

Table 3: Monitoring of the Public-Private Partnership Indicators Period of Monitoring % reduction in November 2001 March 2002 April 2002 August 2002 participation/ performance Number Total % Number Total % PMPs referring patients 30 100 30 20 100 20 10 (out of those contacted) PMPs acting as DOT 7 30 23 2 20 1 22 provider (out of those referring patients) Number of symptomatics 77 43 referred Number of confirmed 51 77 66 24 43 56 10 patients (out of referred symptomatics) Number of confirmed 12 51 24 3 24 13 9 patients taking DOT from PMPs (out of confirmed TB patients) Contribution of PMP referrals to total case finding in the TU (during the 5 month monitoring period) 51 167 30 24 112 21 9

Figure 1: The Public Private Partnership Model TB Symptomatic Back referral form through patient or Health Worker Treatment Box and Card with Health Worker in case PMP desires to be DOT provider Back referral form through patient or Health Worker Non -TB Private Medical Practitioner (PMP) PHC / RH Referral Form with patient Diagnosis through RNTCP TB Categorization by RNTCP Patients unable to afford private treatment Referral Form with patient Private Lab/X-ray Clinic Private Diagnosis Patients able to afford private treatment Private Treatment Treatment with RNTCP PHC: Primary Health Centre RH: Rural Hospital RNTCP:Revised National TB Control Programme