Journal of International Development J. Int. Dev. (2009) Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/jid.1574

PUBLIC SECTOR INSTITUTIONS, POLITICS AND OUTSOURCING: REFORMING THE PROVISION OF PRIMARY HEALTHCARE IN PUNJAB, PAKISTAN IRAM A. KHAN* Office of the Auditor General of Pakistan, Islamabad, Pakistan

Abstract: Lodhran/Rahim Yar Khan (RYK) model of primary healthcare is a home-grown experiment in Pakistan. The model envisaged transferring the management of primary healthcare facilities to a public sector NGO. The paper finds that through the re-alignment of interest groups, the model was replicated in several other districts of the province. It also examines the evolution of stakeholders’ behaviour, interest, position and influence in its implementation over a period of time. The study concludes that sustainability of the model remains doubtful unless public sector health bureaucracy is restructured and made to work effectively. Copyright # 2009 John Wiley & Sons, Ltd. Keywords: public policy; outsourcing; health planning; NGOs; Pakistan

1

INTRODUCTION

The last two decades have seen a fundamental change in the shape of public organisations and the governance and management of societal problems, with governments becoming increasingly dependent on outside agents for the realisation of their policy objectives. This paradigm shift in policy is the result of a belief in the efficiency of private entrepreneurs to rejuvenate an ailing and inefficient public sector (Hood, 1991) as well as the emergence of complex and specialized social demands that governments are finding difficult to meet on their own. This is true for both the developed and developing countries. Outsourcing or contracting out has become a popular and preferred choice for the policy makers today. However, as Mills et al. (2002) point out, outsourcing is the beginning of a logical destination that eventually terminates in comprehensive restructuring of the public sector.

*Correspondence to: Dr Iram A. Khan, Office of the Auditor General of Pakistan, Audit House, Constitution Avenue, Islamabad, Pakistan. E-mail: [email protected]

Copyright # 2009 John Wiley & Sons, Ltd.

I. A. Khan This research paper presents an experiment in outsourcing the delivery of primary healthcare in Punjab, Pakistan to a public sector NGO. The need for outsourcing arose because of the state failure to manage adequate provision and quality of primary healthcare in the province (actually in the country as a whole). Investment in healthcare has been low in Pakistan, declining as a percentage of GDP from 0.58 in 1999–2000 to 0.51 in 2005– 2006 (GOP, 2006). Even within this low expenditure ratios, priority has been accorded to hospitals, medical colleges and curative services in the urban areas while primary healthcare and rural health services have been ignored (SBP, 2004; Zaidi, 1988). Outsourcing the primary healthcare is an attempt to revive the confidence of the general public, especially of the rural areas, on state provision. The rest of the paper is structured as follows. Introduction is followed by review of literature in the context of this study. The third section discusses Lodhran/Rahim Yar Khan (RYK) model of primary healthcare delivery. The penultimate section analyses the role of interest groups and political alignments in primary healthcare reform within the framework of stakeholder analysis. It also discusses the significance of outsourcing to a public sector NGO and suggests future course of action. Conclusion sums up the discussion.

2

CONTEXT OF THE DEBATE

Prompted by fiscal, social, administrative, political and ideological challenges, governments since the mid-1980s have shifted away from traditional line departments and experimented with new forms of service delivery including outsourcing or contracting (Peters and Savoie, 2000). Outsourcing is a purchasing mechanism used to acquire specified services of a defined quality at an agreed price from a specific private provider for a specific period of time (Mills and Broomberg, 1998). Outsourcing in basic public services, including healthcare, is possible where state sets the conditions, specifies the products and policy objectives it wants to achieve rather than directly delivering them through its bureaucratic machinery and can monitor outputs and outcomes (Klijn, 2002). An outside agency, preferably from the private sector, carries out the implementation of policy guidelines (Klijn, 2002; Milward and Provan, 2000; Rhodes, 1997). This may, however, lead to fragmentation of policy and decision making functions. Kadzamira et al. (2004) mentions this general complaint from the providers outside government that it makes policies and takes decisions with little or no involvement from them. Marek et al. (2005) point out some of the advantages of outsourcing. These are mobilisation of additional resources and borrowing private sector management approaches to make the public sector efficient and effective. They also spell out three basic conditions for successful contracting. These are knowledge of the services to be contracted, capacity to manage contracts and sufficient funding to cover the economic cost of the service at the projected level of demand. Mills et al. (2001) find certain capacity weaknesses in government while designing and awarding contracts. These are information asymmetries with respect to cost and quality, absence of budgetary frameworks and financial control mechanisms to monitor expenditure against plans, vague performance indicators and ambiguity about the roles, responsibilities of different organizations. Since outsourcing is possible through the interplay of shifting partnerships of both the public and private sector actors, and decisions are made in different arenas by network of Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

Reforming the Primary Healthcare in Punjab, Pakistan organisations (Castells, 2000), this may lead to interesting products or policy outcomes (Osborne, 2000; Kickert et al., 1997; Rhodes, 1997). They, however, ignore the fact that this may also create tensions resulting from divergence rather than convergence of outcomes. These tensions may also emerge from different degree of emphasis given to efficiency and equity by the public and private sectors. Mills and Broomberg (1998) discuss the extent of contracting and in no case, they do find full range of services in the provision of primary healthcare to be outsourced. Mostly there were cases of selective contracting, especially in areas where more qualified and organised workforce was not available (Gilson et al., 1997). Marek et al. (2005) also conclude that the scope is mostly limited to clinical or non-clinic purchase of services from the private providers to complement public provision. This is despite the pressure by donor agencies that have been spreading out word about outsourcing. Batley and Larbi (2004), Mills et al. (2001), De Beyer et al. (2000) and Pfeiffer (2003) discuss the donor pressure, especially in African countries, to extend the range of activities that are contracted out. This failure to adopt outsourcing at a systemic level shows the complexity involved in such efforts. Monitoring is crucial to protect against potential opportunistic behaviour of a contractor (Kamensky and Morales, 2006; Goldsmith and Eggers, 2004; Deakin and Michie, 1997; Williamson, 1996). Effective monitoring will ensure proper translation of policy guidelines, and goals and targets into action and precludes the possibility of an agent taking advantage from its principal. Travis and Cassels (2006) underscore the need for a key role to be played by the public sector in the design and supervision of monitoring programme. Effective Monitoring leads to decrease in total family expenditure by nearly 40% from $18 to $11 per capita per year (Soeters and Griffiths, 2003). There is also evidence that contracting mechanisms do not work well in situations where monitoring arrangements are weak (Ayeni, 2002; Mills et al., 2001). Literature review by Liu et al. (2008) on the effectiveness and impact of contracting out primary healthcare services finds that it is not possible to determine the systemic effects of outsourcing. They can either be positive or negative. The review suggests that although contracting has improved access to services in many cases, the effects on other performance indicators such as equity, quality and efficiency are not clear. However, the context of contracting out and how the intervention has been designed are likely to influence the chances of success. Loevinsohn and Harding (2005) discuss the pros and cons of contracting out to non-state actors as a means for improving health care delivery. The analysis is done with the help of ten case studies from around the globe. They conclude that contracting with nongovernmental entities is likely to provide better results than government provision of the same services. Loevinsohn (2008) is a toolkit that describes the best practices associated with contracting out healthcare services to non-state providers in the context of developing countries. The theme is performance-based budgeting whereby a series of objectives and indicators are identified that serve as benchmarks for assessing the extent of success/failure in the contracting out arrangement. The toolkit draws on lessons of experience based on 14 case studies of contracting healthcare services and concludes that performance based contracting can lead to rapid improvements in the coverage and quality of publicly financed health services. Palmer et al. (2006) bring another perspective to this debate especially within the context of fragile states, which neither have the capacity to ensure effective delivery of healthcare nor effective monitoring if a service is outsourced. In such weak states, donors contract out Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

I. A. Khan services in response to lack of government infrastructure and the need to expand services rapidly. As a result, paradoxically, the weaker the country’s government capacity, the more likely it is that contracting is adopted. The notable example given is that of Afghanistan. Zaidi (1999) also reinforces this point when he concludes that stronger and more effective the developmentalist state (such as Singapore, South Korea), less the need for such development-oriented NGOs. However, when state and market institutions are inadequate, NGOs emerge to fill the gap. The outsourcing of primary healthcare facilities to NGOs in Pakistan also shows the absence of an effective developmentalist state.

3

REFORM OF PRIMARY HEALTHCARE IN PUNJAB, PAKISTAN

Constitutionally, health is a provincial subject in Pakistan. Federal Government is mandated with policy-making, coordination, research, training and seeking foreign assistance while the provincial and district health governments are responsible for the delivery and management of health services. This makes it possible for the provinces to pursue their own reform agendas in managing the delivery of healthcare. Punjab, the largest province, took the lead in early 1990s in reforming its healthcare sector. The overall reform agenda in the 1990s was generally introduced within the overall administrative and institutional context of health department under a new nomenclature District Health Government Scheme. It proposed the introduction of an ‘internal market’ for the financing and management of health services in both the rural and urban areas. Basic health units (BHUs) and rural health centres were to be contracted out to private health physicians and autonomy was to be granted to certain hospitals with permission to levy user charges (Gera, 2003). Some hospitals were made autonomous and were headed by chief executives. However, these reforms could not be fully implemented primarily because of lack of capacity and training to exercise authority at the lower tiers. The scheme, although technically in place, has been abandoned for all practical purposes. The introduction of local government system in 2001 completely overshadowed it. Collins et al. (2002) sum up several of these initiatives in a chronological order (Table 1). A new reform initiative, called Lodhran/RYK model1, was implemented after a successful pilot that outsourced the primary healthcare infrastructure to an institutional set up outside government bureaucracy. Although initiated in one district, it has been Table 1.

Earlier initiatives for reform in the health sector

Date

Initiatives

1990 1993–1994 1996 1997 1998 1998

Delegation of financial authority to deputy district health officers (officer in charge at tehsil level) Sheikhupura PHC pilot project District health management teams (DHMTs) established in 16 out of 34 districts District health authorities (DHAs) established in Jhelum (July 1997) and Multan (March 1998) Semi-autonomous hospitals established in the tertiary sector District health government (DHG) developed, approved by the Chief Minister in November 1998. Since then, recruitment and training of district level chief executives and rules for DHG

1 The pilot project was initiated in a small town Lodhran and was later replicated in district Rahim Yar Khan (RYK) at a much larger scale. That is why, it is called Lodhran/RYK model of primary healthcare.

Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

Reforming the Primary Healthcare in Punjab, Pakistan replicated in 11 more districts of Punjab. At the heart of this reform, effort was a shift from government by control to government by contract (Khaleghian Das Gupta, 2005).

3.1

Lodhran/RYK Model of Primary Healthcare

A beginning was made in 1999 in a small town called Lodhran when a public sector NGO was entrusted with the responsibility of re-engineering the business processes and running one cluster of three BHUs on behalf of district administration. This experiment although rudimentary and on a small scale, proved a success. Relative success of the programme convinced the provincial government to take over the initiative and launch it as the ‘Chief Minister’s Initiative on Primary Healthcare’ (CMIPHC). District Rahim Yar Khan, whose district Nazim (mayor) endorsed this programme and extended political and administrative support to it was chosen as a pilot district for the project. Under this initiative, one Provincial Support Unit (PSU) was established with grant-inaid by the provincial government to Punjab Rural Support Programme (PRSP), a public sector NGO. Terms of partnership agreement was signed between PRSP and the district government for a period of 5 years on 15 April, 2003. On signing the agreement, 104 BHUs along with their allocated budgets were transferred to PRSP that was given complete administrative and financial autonomy in their management in the district. The district government agreed to transfer primary healthcare budget to PRSP as one-line transfer to ensure financial and management flexibility for operations. PRSP agreed to remain within the existing health budgets and render accounts of the operation to district governments within a period of 3 months at the end of every financial year. Under CMIPHC, 104 BHUs have been consolidated in the form of 35 clusters of two/ three within a range of 15–20 km. Clustering of BHUs has been done in the case of nonavailability of doctors so that all BHUs are served by a medical officer at least twice or thrice a week (PRSP, n.d.). There is one doctor posted for each cluster and covers all BHUs according to a timetable. He is paid Rs.30 000 per month in case of three BHUs and Rs.24 000 per month if he is supervising two BHUs. This package is significantly higher than Rs.12 000 salary offered by the provincial government, where one doctor is posted for one BHU. In addition to that, he is allowed an interest free loan of Rs.100 000 to buy a vehicle to ensure his mobility (SBP, 2004). He resides in the focal BHU, which is decided on the basis of better residential facilities and is not allowed private practice during the period of his contract. He is the administrative head of BHUs cluster and is responsible for its overall discipline and management (Ali, 2004). He is allowed complete independence and flexibility in utilising budget at his disposal. With the attachment of two social mobilisers with the BHU team, it has been possible to involve the community in the BHU affairs and give them a sense of ownership. In some instances, the community has itself contributed funds to improve the BHUs. There was not a single female medical officer (FMO) working in the BHUs of the district. Realising that FMOs cannot reside in the rural areas, they have been allowed to stay at tehsil headquarters and visit 5 BHUs and one girls’ school in a week. In addition to a special salary package of Rs.37 000 per month, they have been given an interest free loan of Rs.150 000 for the purchase of a car. The result is that eight FMOs were working in Rahim Yar Khan district and visiting 40 BHUs every week, by mid-2005. Provincial and district support units (DSUs) have been established to manage the affairs of BHUs in the province and districts. The support units have been funded by the provincial Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

I. A. Khan government and have taken over the administration of BHUs, maintenance of stock and budget from the district government. Again, depending on circumstances, some two/three districts have been grouped under one DSU, while instances of one DSU per district are also there. District support manager (DSM) is the lynchpin at the district level. DSMs are selected by the project director with the help of a committee constituted by the governing body. Some have been hired from the market, while the others are from the civil service and have been posted in CMIPHC on secondment. In addition to government salary, they are paid a project allowance that is more than double their salary. This brings their remuneration package close to what market would have offered. While a district health officer (DHO) had to work in a rigid bureaucratic set up and reported to several tiers in the provincial government hierarchy, a DSM only reports to the project director. A DHO also had to rely on and coordinate with different government agencies for the smooth functioning of his office. A DSM, on the other hand, has been provided sufficient independence in administrative and financial matters to run and manage BHUs on his own. Unlike the DHO, he appoints all professional2, support3 and auxiliary4 staff at BHUs. He also has powers for the local purchase of medicines and supplies up to Rs.50 000. Similarly, he can incur expenditure up to Rs.20 000 for the purchase of capital items, commodities and furniture and fixtures. This limit increases to Rs.100 000 if the purchase is made with the help of a committee constituted by the PSU. For the administrative approval of the estimates of repair works on buildings, a DSM has powers up to Rs.200 000 per health facility in a committee constituted by the PSU (PPHI, 2007). These powers are meant to ensure that he is not constrained in any manner and has the requisite powers and flexibility to discharge his responsibilities. A weakness of the provincial health management system was its poor monitoring, which permitted rent-seeking behaviour. The provincial and district support officers visit 60 or more BHUs in a month (SBP, 2004). During each visit, feedback is obtained from the doctor, staff and patients regarding the working of a BHU, and all records and stocks are inspected. Monthly review meetings (MRMs), which all the medical officers in the district attend, are an open forum for discussing administrative and financial issues. The performance of equipment and operational matters such as coordination with district and local administration also figure in it. There have been a number of instances when doctors were fired or repatriated on the basis of monitoring and evaluation. In district Chakwal5, the services of six doctors working on district government contract were terminated as they were found absent during monitoring visits. Three doctors who were permanent government employees were transferred back to district/provincial governments and five doctors resigned because they could not keep pace with the working style of PRSP (PRSP, n.d.). DSMs were subject to an even stricter monitoring regime. They were under close scrutiny of the project management and district health administration whom they had replaced. Due to careful selection, handsome monetary package and continuous monitoring, no DSM was fired in CMIPHC.

2 Professional staff includes Medical Officers, Administrative, Finance, Audit, Statistics, Planning and Monitoring staff. 3 Support Staff means computer operators, assistants and paramedical staff. 4 Auxiliary staff means watchmen, drivers, guards, gardeners etc. 5 One of the 11 districts, where this model has been replicated.

Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

Reforming the Primary Healthcare in Punjab, Pakistan 4

PERFORMANCE OF LODHRAN/RYK MODEL

Although this research paper does not intend to undertake a comprehensive analysis of the performance of Lodhran/RYK model, it will still be useful to assess its performance with the help of some preliminary data. Data have been collected for the following variables from district governments and PRSP publications:  Number of patients visiting BHUs for the period May 2002 to July 2005 in district Rahim Yar Khan.  Number of patients visiting BHUs for the period January 2003 to July 2005 in 10 other districts of Punjab.  Comparison of cost per patient, excluding medicines, under district government and CMIPHC.  Comparison of cost of medicines per patient under district government and CMIPHC. Secondary data collected from PRSP database also feed into the provincial health information management system (HMIS). Therefore, it is expected that its authenticity will be confirmed by health professionals at the district and provincial levels. The overall trend is also confirmed by other independent evaluations (Loevinsohn et al., 2006; Nayyar-Stone et al., 2006; Batley et al., 2004; SBP, 2004). There has been a positive public response due to the availability of doctors, up gradation of primary healthcare facilities, community involvement and free supply of medicines. In district Rahim Yar Khan, the number of patients visiting BHUs has jumped from less than 50 000 in May 2002, when these were managed by the district government, to 120 000 and 146 000 in the corresponding months in 2003 and 2004, respectively, after the introduction of the model (Figure 1). This is despite variations in different months due to seasonal factor. Similar changes were seen in other districts that adopted the Lodhran/RYK model. This is evident from Figure 2. This indicates that the primary underlying problem was the absence of doctors and medicines. When these were made available, the public confidence grew and they preferred BHUs to quacks and private practitioners, who provided a poor quality of service and also charged a fee for it. Loevinsohn et al., (2006) further reports that compared with Bahawalpur (19%), a contiguous district, women in Rahim Yar Khan were twice (37%) as

Figure 1.

OPD record in Rahim Yar Khan district. Source: PRSP database. This figure is available in colour online at www.interscience.wiley.com/journal/jid

Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

I. A. Khan

Figure 2.

OPD visits for 10 PRSP districts of Punjab. This figure is available in colour online at www.interscience.wiley.com/journal/jid

likely to use a BHU when ill. This preference is due to the presence of FMOs at BHU level in district Rahim Yar Khan. A comparison of cost per patient and cost of medicines per patient between district government and PRSP for the corresponding periods shows that the average cost is much lower for PRSP. The comparison is given in Tables 2 and 3.

Table 2. District

Comparison of cost per patient 2004–2005 Cost Per patient under district government

Cost per patient under CMIPHC

103.16 134.83 202.78 118.72 108.38 137.11 122.64

44.24 61.84 60.59 49.98 48.14 58.12 51.45

Rahim Yar Khan Chakwal Vehari Lahore Faisalabad Sahiwal Average Cost Source: PRSP documents.

Table 3. Comparison of cost of medicines per patient 2004–2005 District

Cost of medicines per patient under district government

Cost of medicines per patient under CMIPHC

26.19 45.32 45.03 16.98 25.11 20.18 27.57

11.23 20.78 13.45 7.15 11.15 8.55 11.57

Rahim Yar Khan Chakwal Vehari Lahore Faisalabad Sahiwal Average Cost Source: PRSP documents. Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

Reforming the Primary Healthcare in Punjab, Pakistan Table 4. District Rahim Yar Khan Chakwal Vehari Lahore Faisalabad Sahiwal Total

Savings from budgets transferred from district governments for 2004–2005 Funds received in Balance due in Expenditure in Savings in Rupees Rupees (in million) Rupees (in million) Rupees (in million) (in million) 55.482 35.158 38.230 18.28 70.178 21.838 239.168

3.204 Nil Nil Nil Nil Nil 3.204

45.254 16.081 21.403 15.678 51.74 14.442 164.598

13.432 19.077 16.827 2.602 18.439 7.396 77.775

Source: PRSP documents.

Dataset shows that PRSP incurred a much lower cost in both the categories. The cost was not only lower on average but also lower for each individual district. Despite increase in the number of patients, this efficient utilisation of resources has allowed PRSP to save funds. The details are listed in Table 4. These savings were transferred to ‘reserved for improvements’ established in every district for long-term investments in the primary health sector. 5

THE POLITICAL CONTEXT OF HEALTHCARE REFORM

Although the Lodhran/RYK model may not be unique for contracting out the provision of primary healthcare in the world, it is a home-grown phenomenon that has not been promoted or financed by external agencies (Loevinsohn et al., 2006). Its impact is also significant since it provides primary healthcare services to more than 25 million people. This is going to be even greater now that it is being replicated in other provinces and regions of the country. This is a ‘bold and creative reform, something that is not common in South Asia’ (Loevinsohn et al., 2006, p. 28). 5.1

Stakeholder Analysis

The paper undertakes a stakeholder analysis (Brugha and Varvasovszky, 2000; Varvasovszky and Brugha, 2000) and tries to understand the behaviours, intentions, inter-relations, influences, resources and interests of different actors and organisations for launching, implementing and achieving goals of the Lodhran/RYK model. This tool also helps in identifying potential allies and building alliances or removing threats (Blair et al., 1996). For this purpose, a detailed analysis of different stakeholders and actors in the launch and operation of the initiative was undertaken. An historical account of change in position (whether supportive or otherwise) and strategy to deal with the stakeholder has also been recorded. A new management model cannot be initiated unless official government policy is linked to it. It was the alignment of many interests that could ensure the beginning and continuation of Lodhran/RYK model. The chief coordinator of this initiative was advisor to the chief minister (CM), Punjab in 1999 and was the one to design and run this programme in Lodhran on a pilot basis. He was able to convince the CM to adopt the initiative and launch it as the CMIPHC. Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

I. A. Khan Firstly, the provincial and district health offices and, more crucially, the doctor’s lobby might have opposed this idea. However, their resistance was overcome when the initiative was launched in the name of Chief Minister of the province. Secondly, relatively small beginning of the initiative in a remote district in rural Punjab was not likely to hurt their vested interests. Its focus on primary rather than tertiary healthcare and its rural rather than urban bias also kept its profile low. Thirdly, the provincial government was already recruiting doctors for these remotely located health facilities on contract basis. The transfer of these facilities to PRSP increased the doctors’ salaries that compensated for less favourable terms and conditions of service. As for the district government, Rahim Yar Khan was the home constituency of the chief coordinator of the initiative, and district Nazim (mayor) was closely related to him. Without whole-hearted support from the district government, it would not have been possible to start the initiative at the district level. Provincial bureaucracy has the potential to derail the reform effort. However, the personal involvement of the advisor to CM and initial upscaling of the programme in his home district, Rahim Yar Khan, convinced them to support the idea. Implementation of the model by a public sector NGO headed by an ex-civil servant was also a factor to win them over. District health bureaucracy felt threatened but had little choice when the provincial government and more so the district Nazim himself supported the idea. However, there has been a re-alignment of political interests in the Punjab province after the local government elections in 2005. Political parting of ways between the CM, Punjab and chief coordinator of the initiative meant that the former who had owned and launched the programme now opposes it. The provincial bureaucracy, at the behest of the CM, Punjab started creating hurdles in the release of funds, etc. to PRSP. The provincial support officer, CMIPHC, being a government employee, was transferred and health secretary was directed to critically review the status of the initiative being implemented in different districts. To neutralise this opposition, the president of Pakistan has been convinced by the chief coordinator of the programme to launch it from 2006 as the ‘President’s Primary Healthcare Initiative (PPHI)’ in the whole country including Punjab. Due to the direct involvement of the president, Punjab government could not rollback the programme although it has not been replicated in other districts either. Federal Ministry of Health, despite its failure to curtail the expansion of the initiative in other provinces, has managed to keep control over federally administered preventive healthcare programmes, which have not been integrated into it. The result is that fragmentation of the healthcare system in the country has persisted, and curative and preventive services are not interlinked at the delivery level. Table 5 shows the evolution of stakeholders’ interest, position and influence in the Lodhran/RYK model. It also provides a historical glimpse into the stakeholders’ reactions to the initiative.

5.2

A Public Sector NGO Made It Possible

Punjab Rural Support Programme (PRSP), a public sector NGO, managed the primary healthcare initiative in Punjab. It was established by the provincial government in 1998 and was funded through an endowment grant. According to Batley et al. (2004), PRSP considers itself a government-organized non-governmental organization and a community mobilization arm of the Government of Punjab. Hood (1997) calls such organisations quasi-non-government organisations that function as part of the state. PRSP, being a nonCopyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

President is not involved since the model is being implemented in a province. President approves replication of this model in the whole country in 2006 under the ‘PPHI’. Being the provincial chief executive, he is responsible for making final decision for the launch of the initiative. Being politically aligned against the chief coordinator of the initiative, he creates hurdles in its future expansion; rather attempts to roll back the existing arrangements. Since the model is being implemented in another province, they are not concerned. Being the provincial chief executive, he is responsible for making final decision for the launch of the initiative. CMIPHC is his brainchild. He can have personal sense of achievement as well as gain political mileage from its success.

President

Copyright # 2009 John Wiley & Sons, Ltd.

Chief coordinator

CMs, other provinces

CM, Punjab

Involvement in the issue

Actors

High High

High

Low-–medium

High High

Low

High

Medium

Low

High

High

Medium

Low

Low

Influence/ power

Low

Interest in the issue

Supportive Supportive

Supportive

Disinterested

Opposed

Supportive

Supportive

Disinterested

Position

Characteristics

Low–medium Low–medium

Low

Low

Low

Low

Low

Low

Impact on the actor

(Continues)

Involve Involve

Involve

None

Defend

Involve

Involve

—

Strategy vis-a`-vis position

Table 5. A historical analysis of stakeholders’ characteristics for Lodhran/RYK model (shaded rows give present status while others indicate status prior to 2005—before the elections of the local government)

Reforming the Primary Healthcare in Punjab, Pakistan

J. Int. Dev. (2009) DOI: 10.1002/jid

Copyright # 2009 John Wiley & Sons, Ltd.

Since the model is being implemented in a province, the ministry is not concerned. The model now includes federal territories and envisages incorporation of federally administered healthcare programmes and projects into the Initiatives. Manages and runs the health activities in the province. Coordinate the programme at district level. Earlier, they were running the health show in the district. Being head of the district, he is responsible and accountable for all administrative decisions including provision of healthcare. Depending on whether they side with the chief coordinator or the CM, they support/oppose the programme. Protects the interests of doctors.

Federal Ministry of Health

Implements the initiatives.

Benefit from the smooth operation of the initiative.

PRSP

Poor

Doctors lobby

District Nazims

Provincial Health Department, Punjab District health departments

Involvement in the issue

Actors

Medium–high Medium–high High High Low Low

Medium–high

High

Medium–high Medium-High High High High High

Medium–high

High

Medium–high Medium-High Low Low

Medium

High

Medium–high Medium–high High High

Low

Neutral Neutral Supportive Supportive Non-mobilised Non-mobilised

Supportive/ opposed

Supportive

Supportive Opposed Marginal Marginal

Opposed

Disinterested

Position

Characteristics Influence/ power

Low

Interest in the issue

Table 5. (Continued)

Low–medium Low–medium Medium–High Medium–High High High

Low

Low

Low–medium Low–medium High High

Medium–high

Low

Impact on the actor

Collaborate Collaborate Implements Implements No voice No voice

Involve/defend

Involve

Involve Defend Monitor Monitor/defend

Defend

None

Strategy vis-a`-vis position

I. A. Khan

J. Int. Dev. (2009) DOI: 10.1002/jid

Reforming the Primary Healthcare in Punjab, Pakistan profit limited company registered under the Companies Ordinance, is governed by the corporate governance guidelines issued by the companies’ regulator rather than rigid rules and regulations formulated by the government. This distinct legal status gives PRSP managerial flexibility and autonomy to contract personnel on terms and conditions different from the government and monitor them more efficiently, while being a public sector NGO provided a comfort level to the provincial and district governments to transfer to it the assets and budgets of BHUs without inviting competition. Loevinsohn et al., (2006) has also pointed out the absence of competition in the selection of PRSP and notes that the NGO was ‘well known to government officials’ (p. 31). Batley et al. (2004) convey the general feeling in government circles that ‘you can ‘‘trust’’ the PRSP as it is a form of quasi-governmental organization itself’’(p. 34). PRSP, being a non-conventional arm of the Punjab Government, was also the reason for setting up programme office in the CM secretariat. Provincial support officer was also transferred from bureaucracy. He worked under PRSP, but was attached with the CM Secretariat in a marriage of convenience where he had the working flexibility of a nonprofit NGO but the clout of the office of CM. Only a public sector NGO could have enjoyed this privilege. Another interesting aspect of outsourcing is that PRSP was not paid any management or performance fee for undertaking this assignment. Rather, Punjab government agreed to fund the establishment of provincial and District Support Units under PRSP and also agreed to transfer its handpicked officers on secondment there. An NGO from the private sector could not have been treated like one of the family. The agreement signed between the district government and PRSP is quite vague and does not list the set of indicators on the basis of which the performance of the latter could have been monitored. The relationship between Punjab government and PRSP seems more informal and trust-based like the one reported by Moran and Batley (2004) in many African countries between government and church-based hospitals and dispensaries. Russell et al. (1997) and Smithson et al. (1997) report instances where governments have given licences with block grants (Zimbabwe) or subsidies to church health facilities (Ghana) without any serious specification of the service to be provided in return. The case of handing over primary healthcare facilities to PRSP does not seem to be any different. With the expansion of the initiatives to other provinces and regions of the country, similar management model has been replicated there. Provincial governments in Sindh, Baluchistan and NWFP have contracted out most of their primary healthcare facilities to similar public sector NGOs in their respective provinces. This process started under PPHI in 2006. However, due to political uncertainty in the country and national elections in 2008, it has gained momentum only recently. It also needs to be seen how it will impact the capacity of the NGOs to be flexible and innovative. There is a justified apprehension that they may lose their ‘‘touch’’ and become rigid and bureaucratic like traditional line departments (Palmer et al., 2006; Batley et al., 2004; Kaul, 2000). They may also lose a local approach, which makes them effective deliverers.

6

FUTURE OF THE INITIATIVES

When Lodhran/RYK model was launched, no explicit exit strategy was planned. However, when the President’s (now renamed People’s) Primary Healthcare Initiative was launched, it was decided to restructure the provision of primary healthcare at the level of district governments. It was categorically stated that funding for the provincial and District Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

I. A. Khan Support Units would be required for a period of 3 years only. At the end of 3 years, a new organisational structure will be in place to take back the primary healthcare facilities from the NGOs. The task to re-engineer the health departments was given to National Reconstruction Bureau (NRB), which had prepared a blueprint for local government system for the country in 2001. However, even after a lapse of more than 2 years, NRB has not made anything public that would suggest that it has developed a new model for primary healthcare delivery. It so appears that this claim to restructure public sector healthcare delivery system will remain a dream only. This does not augur well for the initiative, which although based on sound management principles, relies on elusive political support. The PPHI although approved in September 2005, could not be launched during the first 16 months. The re-alignment of political forces in Punjab before and after the elections in 2008 had a significant effect on the pace of the initiative. Before the elections, the Chief Coordinator and the CM got into opposing camps. This had jeopardised the very existence of the whole project. However, after the elections, situation changed in Punjab. With a new supportive CM in power, the primary healthcare initiative has got a new lease of life in the province. However, these ups and downs due to political expediency clearly show the dilemma of relying on political bosses. It is therefore crucial that the state apparatus is overhauled and revamped to manage the health facilities in an efficient and effective manner. Otherwise, we can revert back to the original position where provincial and district governments ran the facilities with sad consequences. This is evident from the case of district Lodhran, where the first beginnings of this model were made. After running the BHU cluster for nearly 3 years, district government decided to terminate the contract with the NGO and run the BHUs on its own. What happened has been depicted in Figure 3. Figure 3 shows that soon after taking over the BHUs by the NGO, number of OPD visits went up and remained high as long as the BHUs were run by it. When the district government took them back in 2003, the visits declined dramatically. This shows the importance of revamping and re-engineering the government apparatus. One silver lining in the cloud exists in the management model itself. The programme offices at the provincial level work in the CMs’ secretariats and, although they are part of NGOs, are administratively distinct from them. It is possible to transfer these offices to the provincial governments and allow them to function with the same level of management flexibility and financial autonomy that they enjoy now. They can even be converted into

Figure 3.

District Lodhran: continuity is important. Source: PRSP database.

Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

Reforming the Primary Healthcare in Punjab, Pakistan non-profit legal entities under a corporate governance framework, similar to the one that governs PRSP and other public sector NGOs. All primary healthcare facilities can function under them on a permanent basis. The activities of provincial health departments should be scaled down proportionately. Keeping in view the incapacity of traditional line ministries and departments, the government may also decide to set up a separate organisation to monitor the performance of primary healthcare delivery. However, these are early days for the initiative and it may not be politically feasible to talk of this strategy. Since the government has not formulated any plans to revamp and re-engineer its health service delivery functions and processes, it is quite likely that the public sector NGOs may continue to run this initiative on a permanent basis.

7

CONCLUSIONS

Lodhran/RYK model primary healthcare is a home-grown effort that was conceived and developed in response to the pathetic conditions of basic healthcare in rural Punjab. It has been envisioned to re-engineer the primary healthcare services through the transfer of these facilities to a public sector NGO. The NGO has worked as an extended arm of the government. In that sense, there is as such no privatisation of the primary healthcare and the state continues to play its welfare role with a new face and under a new management model. Although working successfully, it may not be possible to continue with the model due to shifting political alignments. It is, therefore, important that provincial and district health bureaucracies are restructured and made efficient and effective. However, despite a clear commitment, no steps have been taken to prepare a blueprint for this purpose. One possible alternative is to convert existing programme offices into non-profit corporate entities and allow to manage primary healthcare facilities with the same level of flexibility and autonomy they are enjoying now.

ACKNOWLEDGEMENTS The author would like to thank the anonymous referee for his/her useful comments on an earlier draft of this paper.

REFERENCES Ali M. 2004. Identification of best practices in devolution. CIDA Devolution Support Programme, Lahore. Ayeni V. 2002. Public Sector Reform in Developing Countries: A Handbook of Commonwealth Experiences. Commonwealth Secretariat: London. Batley R, Larbi GA. 2004. The Changing Role of Government: The Reform of Public Services in Developing Countries. MacMillan: Palgrave. Batley R, Hussein M, Khan A, Mumtaz Z, Palmer N, Sansom K. 2004. Pakistan: Study of Non-State Providers of Basic Services. Department for International Development: London. Blair JD, Fottler MD, Whitehead CJ. 1996. Diagnosing the stakeholder bottom line for medical group practices. Key stakeholders’ potential to threaten and/or cooperate. Medical Group Medical Journal 43(2): 40, 42–48, 50–51. Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

I. A. Khan Brugha R, Varvasovszky Z. 2000. Stakeholder analysis: a review. Health Policy and Planning 15(3): 239–246. Castells M. 2000. The Rise of the Network Society: Economy, Society and Culture. Blackwell Publishing: Cambridge. Collins CD, Omar M, Tarin E. 2002. Decentralization, health care and policy process in the Punjab, Pakistan in the 1990s. International Journal of Health Planning Management 17: 123–146. DOI: 10.1002/hpm.657 De Beyer JAS, Preker A, Feachem RG. 2000. The role of the World Bank in international health: Renewed commitment and partnership. Social Science & Medicine 50(2): 169–176. Deakin S, Michie J. 1997. Contract, Co-operation, and Competition: Studies in Economics, Management and Law. Oxford: Oxford University Press. Gera N. 2003. Pakistan’s Health Care Under Structural Adjustment. The Lahore Journal of Economics 8(2): 65–85. Gilson L, Adusei J, Arhin D, Hongoro C, Mujinja P, Sagoe K. 1997. Should governments in developing countries contract out clinical services to NGOs? In Private Health Providers in Developing Countries: Serving the Public Interest? Bennett S, McPake B, Mills A (eds). Zed Books: London. Goldsmith S, Eggers WD. 2004. Governing by network: the new shape of the public sector. Brookings Institution Press: Washington. GOP. 2006. Economic Survey of Pakistan. Ministry of Finance, Government of Pakistan, Islamabad, Pakistan. Hood C. 1997. Which contract state? Four perspectives on over-outsourcing for public services. Australian Journal of Public Administration 56(3): 120–131. Hood C. 1991. A public management for all seasons? Public Administration 69: 3–19. Kadzamira E, Moran D, Mulligan J, Ndirenda N, Nthara K, Reed B, Rose P. 2004. Malawi: Study of Non-State Providers of Basic Services. IDD, University of Birmingham, Birmingham. Kamensky JM, Morales A (eds). 2006. Competition, Choice, and Incentives in Government Programs. Rowman & Littlefield: Lanham, Maryland. Kaul M. 2000. Management Reforms in Government: A Review of International Practices and Strategies. CAPAM: London. Khaleghian P, Das Gupta M. 2005. Public Management and the Essential Public Health Functions. World Development 33(7): 1083–1099. DOI:10.1016/j.worlddev.2005.04.001 Kickert WJM, Klijn E-H, Koppenjan, Joop FM. 1997. Managing Complex Networks: Strategies for the Public Sector. Sage Publications: London. Klijn E-H. 2002. Governing networks in the hollow state: Contracting out, process management or a combination of the two? Public Management Review 4(2): 149–165. Liu Xingzhu, Hotchkiss DR, Bose S. 2008. The effectiveness of contracting-out primary health care services in developing countries: a review of the evidence. Health Policy and Planning 23: 1–13. DOI:10.1093/heapol/czm042 Loevinsohn B. 2008. Performance-Based Contracting for Health Services In Developing Countries: A Toolkit. World Bank: Washington, DC. Loevinsohn B, Haq I, Cuffinhal A, Pande A. 2006. Partnering with NGOs to Strengthen Management: An External Evaluation of the Chief Minister’s Initiative on Primary Health Care in Rahim Yar Khan District, Punjab, Discussion Paper Series, Report No. 13. World Bank, Washington, DC. Loevinsohn B, Harding A. 2005. Buying results? Contracting for health service delivery in developing countries. The Lancet 366(9486): 676–681. DOI:10.1016/S0140-6736(05)67140-1 Marek T, O’farrell C, Yamamoto C, Zable I. 2005. Trends and opportunities in public–private partnerships to improve health service delivery in Africa. World Bank: Washington, DC. Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid

Reforming the Primary Healthcare in Punjab, Pakistan Mills A, Bennett S, Russell S. 2001. The Challenge of Health Sector Reform: What Must Governments Do? Palgrave: Basingstoke/New York. Mills A, Broomberg J. 1998. Experiences of contracting health services: an overview of the literature. Macroeconomics, Health and Development Series No. 33, World Health Organisation, Geneva. Mills A, Brugha R, Hanson K, Mcpake B. 2002. Approaches for improving service delivery in the non state sector: what is the evidence on what works, where and why? Paper presented at the ‘‘Making Services Work for the Poor’’ World Development Report (WDR) 2003/04 Workshop. 4– 5 November, Eynsham Hall, Oxfordshire. Milward HB, Provan KG. 2000. Governing the hollow state. Journal of Public Administration Research and Theory 10(2): 359–379. Moran D, Batley R. 2004. Literature review of non-state provision of basic services. International Development Department, School of Public Policy, The University of Birmingham, Birmingham, UK. Nayyar-Stone R, Ebel R, Ignatova S, Rashid K, Hatry H, Peterson G. 2006. Assessment Report. Pakistan Devolution Support Project. The Urban Institute, Washington, DC. Osborne SP. 2000. Public-Private Partnerships: Theory and Practice in International Perspective. Routledge: London. Palmer N, Strong L, Wali A, Sondorp E. 2006. Contracting out health services in fragile states. British Medical Journal 332(7543): 718–721. DOI:10.1136/bmj.332.7543.718 Peters BG, Savoie DJ. 2000. Governance in the 21st Century: Revitalizing the Public Service. Montreal: McGill-Queen’s Press. Pfeiffer J. 2003. International NGOs and primary health care in Mozambique: the need for a new model of collaboration. Social Science & Medicine 56(4): 725–738. PPHI. 2007. Manual of Operations for the People’s Primary Healthcare Initiative. Federal Support Unit, People’s Primary Healthcare Initiative, Lahore, Pakistan. PRSP. n.d. A Guideline for District Support Managers. Punjab Rural Support Programme, Lahore, Pakistan. Rhodes RAW. 1997. Understanding Government. Buckingham; Open University Press. Russell S, Kwaramba P, Hongoro C, Chikandi S. 1997. Reforming the Health Sector in Zimbabwe: Does Government have the Capacity? The Role of Government in Adjusting Economies Paper 20. International Development Department, School of Public Policy, University of Birmingham, Birmingham. SBP. 2004. Making health services work for the poor in Pakistan: Rahim Yar Khan primary healthcare pilot project. In The State of Pakistan’s Economy: First Quarterly Report for the Year 2003–04 of the Central Board of State Bank of Pakistan. State Bank of Pakistan: Karachi. Smithson P, Asamoa-Baah A, Mills A. 1997. The case of the health sector in Ghana. Role of Government in Adjusting Economies Paper 26, International Development Department, School of Public Policy, University of Birmingham, Birmingham. Soeters R, Griffiths F. 2003. Improving government health services through contract management: a case from Cambodia. Health Policy and Planning 18(1): 74–83. DOI: 10.1093/heapol/18.1.74 Travis P, Cassels A. 2006. Safe in their hands? Engaging private providers in the quest for public health goals. Bulletin of World Health Organisation, 84(6): 427. Varvasovszky Z, Brugha R. 2000. A stakeholder analysis. Health Policy and Planning 15(3): 338– 345. Williamson OE. 1996. The Mechanisms of Governance. Oxford University Press: Oxford. Zaidi SAr. 1988. The Political Economy of Health Care in Pakistan. Vanguard Books: Lahore. Zaidi SA. 1999. NGO failure and the need to bring back the state. Journal of International Development 11(2): 259–271. Copyright # 2009 John Wiley & Sons, Ltd.

J. Int. Dev. (2009) DOI: 10.1002/jid