Monitoring the health sector

An initiative in Tamil Nadu to monitor the functioning of government sector health institutions may help improve their performance.

VENKATESH ATHREYA SHEELA RANI CHUNKATH

PRIMARY health care is now universally accepted as the most effective intervention to achieve significant improvement in the health status of a population, especially in respect of infant, child and maternal care. The Alma Ata declaration of 1978, signed by most nations of the world and health policy document (1983) of the Government of India, identify primary health care as a key strategic intervention for providing universal access to health care.

In Tamil Nadu, as in most other States, the government sector is the major provider of primary health care, particularly in rural areas. While 11 teaching hospitals, 26 district headquarters hospitals and 227 taluk and non-taluk hospitals in the State pr ovide secondary and tertiary care in the government health sector, rural primary health care needs are served by 1,404 primary health centres (PHCs) and 8,682 health sub-centres (HSCs). While the government health sector in its entirety employs around 8, 000 doctors and 28,000 paramedical personnel, approximately 2,263 doctors and 22,000 paramedical personnel serve in PHCs. Unlike in many other States, most PHCs in Tamil Nadu (94 per cent) function in Government-owned premises, nearly 40 per cent of them constructed in the last three years. All PHCs are electrified. Every block PHC has at least one roadworthy vehicle. Nearly one-fifth of the PHCs (250 out of 1,404) function round the clock, and most of them have ambulances.

In its budget for 1999-2000, the Tamil Nadu Government has provided Rs.1,051 crores for the health sector. This constitutes 5.6 per cent of the Government's total revenue expenditure.

Given the importance of the government sector in health and the scale of its expenditure, effective monitoring of institutional performance in this sector is an issue of considerable relevance. An important initiative has recently been taken in this rega rd by the Tamil Nadu Area Health Care Project (TNAHCP), a Government of Tamil Nadu project assisted by the Danish International Development Agency (DANIDA), which may have a wider national relevance.

The DANIDA-TNAHCP initiative seeks to put in place a comprehensive concurrent monitoring system for the entire government health sector that focusses on the performance of its health institutions at all levels of care: primary, secondary and tertiary. Th e objective of such an effort is two-fold: to facilitate course correction from time to time and to provide inputs for policy. A beginning has been made by developing and implementing a system for concurrent monitoring of the PHCs, focussing on instituti onal performance. A HSC serves a population of 5,000, on an average, and is run by a village health nurse (VHN), also called the "multi-purpose health worker (female)" (MPHW-F). She is to provide antenatal, natal and postnatal care, ensure immunisation o f pregnant women and infants and dispense medicine for minor ailments. There is also a multipurpose health worker (male), called the health inspector, for every 10,000 population.

At the PHC level, the staffing norms provide for two medical officers, an auxiliary nurse midwife, a male nursing assistant, a health inspector, and a pharmacist, besides skeletal support staff.

Traditionally, monitoring of rural primary health care delivery through PHCs and HSCs has tended to focus almost exclusively on the performance of village health nurses, and has been largely in terms of work related to immunisation and antenatal care. Da ta were of course collected on institutional deliveries, contraceptive use, sterilisations and medical termination of pregnancy (MTP). But the focus was not on the performance of the institutions. Rather, the standard practice was to consolidate data for the district as a whole and review a district's performance. Since such aggregated data did not distinguish between private and public sectors, there was no focus on assessing public sector health. Districts which had a good private sector presence woul d often perform better in terms of health indicators such as institutional deliveries and contraceptive performance and would accordingly get rewarded. Aggregating data of the performance of private and public sector institutions often led policymakers t o presume that a district's public sector health institutions were performing well when actually they were not.

Data on the number of outpatients treated and deliveries performed at PHCs are also being routinely collected, but without any great sense of urgency. Consolidated reports from the field would often take several months to reach the State-level public hea lth directorate, and the practice of verification/cross-checking was mostly non-existent. Information was often incomplete, except in the case of sterilisations and MTP, which were prioritised given the focus on population stabilisation. Thus, the practi ce of monitoring was perfunctory and tended to focus only on the performance of community outreach services by the VHNs. In particular, there was complete absence of monitoring of individual public sector health institutions in terms of their performance in respect of health services they were mandated to provide.

THE DANIDA-TNAHCP initiative specifically seeks to monitor the various services that a public health institution must provide. It seeks to monitor PHC performance, on a monthly basis, in respect of 12 different kinds of services that a PHC is mandated to provide. The services are outpatient (O.P.) services, deliveries, inpatient services, minor surgeries, ambulance services, anti-rabies vaccinations, administration of anti-snake venom and tetanus toxoid, contraceptive services, services relating to MTP, special clinics (such as antenatal clinics, under-5 clinics and ophthalmic clinics) and laboratory services.

Considerable preparatory work went into the initiative. First, a comprehensive format, readable by an optical mark reader (OMR) for monitoring PHC performance, had to be developed, field-tested and finalised. Field-level personnel charged with collection of data had to be trained to understand the content of the format, and to fill it correctly for scanning by the OMR. While all this preparation took several months, it also served to sensitise health officials and functionaries at several levels to the importance of institutional services monitoring (ISM).

For purposes of health administration, Tamil Nadu is divided into 42 health unit districts (HUDs). Primary health care services in a HUD are under the administration of the Deputy Director of Health Services (DDHS) while secondary health care institution s come under the Deputy Director of Medical Services (DD-Medical). The monthly data pertaining to every PHC in a HUD in respect of the 12 services listed earlier, along with PHC identification and staffing particulars, are entered both in a format readab le by an OMR and in a traditional format by a designated staff member of the PHC, and verified and signed by the medical officer in charge of the PHC*. The personnel in charge of statistics in the DDHS office ensure that the data and the forms are in ord er. Software has been developed at the DANIDA-TNAHCP directorate to process the collected data to produce a range of output tables and graphics relevant for course correction and policy formulation. The outputs are then shared in periodic review meetings with district health officials to enable an analysis of performance of PHCs in each district, identification of strengths and weaknesses, and evolving strategies for improvement.

This initiative was begun in April 1999. The data collected and outputs generated provide interesting insights and pointers. Discussion and sharing of the outputs at various levels of the State's health administration have helped sensitise health officia ls to the importance of ISM, and to a range of health policy issues. It must be added, however, that the reliability of the data gathered and processed cannot be taken for granted, and one would need to put in place mechanisms for sample cross-checks an d verification.

THE two main services provided at the PHCs are outpatient services and deliveries. During the first couple of months of the ISM process, a little more attention was given to these two important indicators.

Since data regarding the number of doctors in each PHC and in each district were available it was possible to analyse the distribution of doctors among the various districts and to correlate it with output. In April 1999, when the ISM process began, the number of outpatients treated per day per PHC was 68 and the number of outpatients, treated per day per PHC doctor was 42. Both figures have shown steady increase and stood at 93 and 58 respectively in November 1999. However, while the State average figu res showed a fairly steady rise, there were significant variations across districts, information on which helped focus administrative intervention to improve performance in districts where it was lagging. It must be pointed out, however, that data on out patients treated are generally subject to greater margins of error than those on deliveries.

A second important indicator of PHC performance is the number of deliveries that take place in the PHC. The overall picture in Tamil Nadu is that institutional deliveries (IDs) account for more than 80 per cent of all deliveries. Around half of these ins titutional deliveries take place in government medical institutions. However, despite their large number and wide geographical spread, PHCs account for only around 5 to 6 per cent of rural deliveries. What this implies is that district headquarters hospi tals and teaching institutions which should really be catering to more complicated deliveries end up being overloaded with normal deliveries which can easily be managed at the PHC level. An important policy objective is, therefore, to increase the number of deliveries at PHCs. The key findings in respect of deliveries that emerge from the ISM process are:

Only 46 per cent of PHCs were conducting deliveries in April 1999 and this figure rose to 52 per cent in November 1999. In absolute terms, the number of PHCs conducting deliveries increased from 646 in April 1999 to 727 in November 1999.

Between 1996-97 and 1998-99, the number of deliveries per PHC per month remained unchanged at 2.4. However, this has shown a steady increase in the last eight months reaching the figure of 3.1 in November 1999. The absolute number of deliveries in all PH Cs taken together has increased from 3,380 in March to 4,404 in November.

While it would perhaps be too early to see them as indicative of a definite trend, these encouraging overall results do suggest strongly that constant and concurrent monitoring can lead, even without additional input in terms of hardware and personnel, t o some improvement in performances possibly by enhancing the motivation to perform and inducing competitive pressures for emulation. It is obvious, though, that sustained performance improvement would require filling of gaps in physical and human resourc es.

The improvement in performance in terms of deliveries has not been uniform across districts. Some PHCs have shown significant improvement, while others lag behind. The availability of district-level and PHC-level data in this regard helps State-level and district-level officials to identify weaknesses and bottlenecks, and initiate corrective measures.

The process of monitoring the delivery of institutional services by PHCs has enabled the health directorates to respond to bottlenecks to PHC functioning identified through the ISM process. An important instance of this is the vacancy position pertaining to PHC medical officers. It was found from the ISM output that some backward districts had higher vacancy rates than some of the more advanced ones. The Government then moved quickly to appoint medical officers to PHCs on contract basis in the weaker di strict to address the problem in the short run. Simultaneously, medium-term and long-term policy measures are under discussion to ensure more rational deployment of medical personnel.

The data so far obtained from the ISM process suggest the need for policy intervention to address issues such as the need to ensure rational deployment of medical officers across districts while filling up vacant posts; the need to appoint more women med ical officers; sensitising rural health functionaries to the importance of maternal and child health services as well as delivery of curative services; promoting community involvement to enhance the degree of utilisation of PHCs; and to ensure availabili ty of contraceptive and MTP services in PHCs.

The logical follow-up would be to extend the process to institutions providing secondary and tertiary care. Local health committees involving elected local bodies and the Government health functionaries are to be set up as a pilot initiative in the DANID A-TNAHCP.

Venkatesh Athreya is Professor and Head, Department of Economics, Bharathidasan University, Tiruchirapalli. Sheela Rani Chunkath, IAS, is Commissioner for Maternal, Child Health and Welfare, and Project Director, DANIDA Tamil Nadu Area Health Care Project.

Views expressed in the article are of the authors and not necessarily of the organisations to which they belong.