Public health infrastructure: What we need and what we have

By Sandhya Srinivasan

We need 7,415 community health centres per 100,000 population. We have less than half the number. Worse, at the healthcare facilities we do have, the basic staff is not in place. Only 38% of our primary health centres have all the required medical personnel. With the public health infrastructure in such a shambles, how can the poor count on government health centres

How can India 's poor count on government health centres when the infrastructure is in such a shambles? Dr Alpana Sagar, from the Centre of Social Medicine and Community Health in JNU, Delhi , asked this question at a meeting in Mumbai recently.

Dr Sagar presented data showing that after 1991 the government stopped planning for any significant growth in rural healthcare infrastructure. Figures on infrastructure development since the First Five-Year Plan (1951-56) indicate a drastic slowing down from the Eighth Plan (1992-1997) onwards (see table).

Establishment of infrastructure in India since First Five-Year Plan

Plan number

Plan period

Community health centres

Primary health centres















































According to the government's formula, we're supposed to have one sub-centre for every 5,000 people (3,000 in hilly areas), one primary health centre for every 30,000 people (20,000 in hilly areas) and one community health centre for every 120,000 people (80,000 in hilly areas). Dr Sagar used the government's formula to calculate the number and staffing at healthcare facilities in rural areas where about three-fourths of India 's 1.027 billion population live.

Rural health services infrastructure 2000-2001





1 per 20,000-30,000




1 per 3,000-5,000



Community health centres

1 per 100,000



We are far short of this requirement. The worst-off are community health centres. If we should have one for every 100,000 rural population, we need at least 7,415 CHCs, but we have less than half of what we should have.

In the 3,043 CHCs that we do have, only 440 have a paediatrician, only 704 have a physician, only 780 have a gynaecologist and 781 a surgeon. So not only is the infrastructure inadequate, we don't even have the staff to use the existing infrastructure. We need 76,622 midwife nurses (one per PHC and seven per CHC). We have planned only for 44,143 and only 27,336 -- barely half the requirement -- are in place.

In the case of other rural healthcare personnel too there are similar and dramatic differences between what we need, what is allocated, and what is actually in place : we have only 71,053 male multipurpose workers compared to the 13,73,311 planned and the 148,303 needed; 137,407 auxiliary nurse midwives compared to the 160,153 planned and 173,020 needed; 19,927 male health assistants compared to the 22,842 planned and the 24,717 needed; 19,855 lady health visitors compared to the 22,842 planned and the 24,717 needed; 21,118 pharmacists compared to the 25,885 planned and the 32,132 needed; and 13,262 lab technicians compared to the 25,885 planned and the 32,132 needed.

These figures give only a broad picture of the problem. They cannot tell us, for example, about whether the facilities are appropriately located. According to the NCAER, in nearly 20% of cases rural households travelled more than 10 km for treatment. In Meghalaya, in 54.56% of rural illness cases and in Orissa in 33.47% of rural illness cases, patients travelled more than 10 km. Even when patients do get to the health centre there is no guarantee that the staff will be present. According to a survey by the Jan Swasthya Abhiyan, only 38% of all PHCs have all the critical staff. A survey by the International Institute of Population Sciences found that only 69% of PHCs have at least one bed, and only 20% have a telephone.

The National Health Policy 2002 aims to increase usage of public health facilities from the current level of less than 20% to more than 75% by 2010. How realistic is this goal?

Kerala: The last bastion begins to fall

Kerala's striking health indices are partly attributed to a healthcare infrastructure developed by a government committed to healthcare. For example, in 1955-56, the state's revenue expenditure on health was 8.48% of total revenue expenditure, compared to the all-states average of 4.36%, and it has pretty much stayed that way. In 1994-95, the state's revenue expenditure on health was 7.44% of total revenue expenditure, compared to the all-states average of 2.63%. Even with fiscal crises since the 1980s, the state's per capita public expenditure on health remained higher than the all-states average. As a result, in the 1980s Kerala had an adequate network of well-staffed health facilities, though access was limited in some districts of the state, according to public health researcher V Raman Kutty.

Starting in the 1980s, there was an overall drop in the rate of growth in government health expenditure forced by the pressures of a fiscal crisis. According to Dr Raman Kutty, it would have been difficult to cut back on salaries so revenue expenditure -- for salaries rather than medical supplies -- increased at the cost of capital expenditure for new infrastructure.

This was accentuated after 1991. In a study of the impact of macroeconomic adjustment policies on access to healthcare, Dr D Narayana of the Centre for Development Studies, Trivandrum , notes that between 1981-82 and 1997-98, t he state's expenditure on medical and public health services, as a proportion of total expenditure, declined from 9.62% to 6.98%. Revenue expenditure on medical and public health services, as a proportion of total revenue expenditure, went from 9.74% to 8.7%. Capital expenditure on medical and public health services, as a percentage of total capital expenditure, plunged from 9.61% to 1.57%.

People relied increasingly on private services as the state's public services deteriorated in quality. Sixty per cent of people in rural areas avoided the government's primary health centres, citing lack of medicines and long distances.

As Dr Joy Elamon of the Kerala Health Studies Research Centre, Trivandrum , put it: "They used to say that if a PHC in Kerala was unmanned, a crowd would gather to demand that the doctor turn up. That's apparently no longer true."

The growth of the private sector

Health economists like Dr Rama Baru from the Centre of Social Medicine and Community Health, Jawaharlal Nehru University (JNU), New Delhi , link the growth of the private sector to the deterioration of public services.

Dr Baru notes that starting in the late-1970s, private nursing homes began growing in number, mostly in urban areas and in agriculturally prosperous states -- essentially wherever there was a market for their services. So nursing homes and hospitals started coming up in states like Kerala, Andhra Pradesh, Maharashtra , Punjab , Gujarat and Tamil Nadu. The trend was accentuated in the 1980s as a result of government policy. Not only did cutbacks as far back as the 1980s affect the quality of government services, but government doctors were more often found doing the more lucrative private practice. Meanwhile, the medical education system had created an army of doctors. And in the absence of regulation, the private sector was an attractive option, with the potential for unnecessary procedures, tests, kickbacks for referrals, and so on.

Indeed, a study in Ahmednagar, Maharashtra, found six private providers (registered and otherwise) for every 2,000 people in urban areas, but only one for every 2,000 people in rural areas -- naturally, the doctors went where the market was. Similarly, a study on urban and rural Nashik district in Maharashtra found that fairly inaccessible but prosperous villages had good dispensaries. The private sector is the major source of curative services in urban as well as rural areas. Remote villages were served almost solely by visiting 'quacks'. Slum settlements within the city were totally dependent on private services for treatment.

The setting up of the first corporate hospital in Chennai in 1983 marked a qualitative shift in the growth of the private sector, hitherto dominated by single-owner nursing homes. Another concurrent trend has been the import of medical equipment for stand-alone diagnostic centres as well as corporate hospitals. This high technology pushed up the cost of medical care. Dr Rama Baru quotes a study in Hyderabad that found hysterectomies cost between Rs 14,000 and 23,000 in corporate hospitals compared to Rs 4,000-6,000 in nursing homes; C-sections cost Rs 10,000-12,000 compared to Rs 6,000 at nursing homes.

The Apollo Group is now the largest private healthcare corporation in Asia and was followed into the healthcare market by several business groups in Andhra Pradesh, Tamil Nadu and Karnataka. Corporate hospitals started acquiring small hospitals. The next step, in the 1990s, has been collaborations with multinationals and with state governments.

The government actively supported the growth of the private sector, notes Dr K V Narayana from the Centre for Economic and Social Studies, Hyderabad . Customs duty for medical equipment was halved in the 1980s, when liberalisation actually started. Various other taxes were waived and land was given free if hospitals treated 10% of their in-patients and 40% of their outpatients free of charge. Hospitals could also get loans from private institutions.

The private sector outstrips the public

Thus by the 1990s, public hospitals were losing their place as the primary providers of in-patient care, writes Dr K V Narayana. He notes that between 1986-87 and 1995-96, the private sector's share of in-patient care grew in every state in the country except rural Orissa where it went from 11.9% to 9.4%. In 1995-96, 54.6% of in-patient rural healthcare and 56.9% of urban healthcare was in the private sector. In Andhra Pradesh, 77.5% of rural healthcare and 63.8% of urban care was in the private sector.

In outpatient care, the private sector grew in every state. All-India, 81% of outpatient rural care was in the private sector; 82.6% in urban areas. The exceptions were rural Andhra Pradesh and urban Assam and Bihar . As much as 95.3% of care in rural Uttar Pradesh -- a poor state with some of the worst health indices in the country -- was in the private sector.

In 1992-93, 57.3% of institutional deliveries in India were in the public sector, and 42.7% in the private sector. By 1998-99, 49.5% of deliveries were in private hospitals, 48.4% in public hospitals and 2.1 in voluntary sector hospitals.

The shift was more dramatic in Andhra Pradesh where in 1992-93, private hospitals took in 58.3% of deliveries and 41.7% were in public hospitals. By 1998-99, 70.5% of deliveries were in private hospitals and just 25.1% in public hospitals.

The World Bank's health sector reforms in Andhra Pradesh encourage a public-private partnership at the expense of the public sector, according to Dr Narayana. The project discourages upgrading PHCs if there is a private service in the vicinity.

Finally, is the growth of private hospitals reducing the pressure on public services? No, Dr Narayana answers. Public hospital administrators must fight for funds; many senior specialists have left for corporate hospitals where they'll be paid much more, and public hospitals have become 'dumping grounds' for unwanted cases (see box).

Public hospitals in poor states

According to a survey in the early-1990s by the National Council for Applied Economic Research, people were more likely to go to government doctors in Himachal Pradesh , Jammu and Kashmir , Orissa, Rajasthan and Pondicherry . In general, says economist T N Krishnan, public hospitals are more important in backward states -- even if they are grossly inadequate, with barely 40 hospital beds per 100,000 population in states like Uttar Pradesh and Orissa where up to 80% of in-patients are treated at public hospitals. The reason: "High levels of poverty and low incomes presumably restrict the demand for private healthcare," he writes.

Further, though 55-60% of people get admitted to general wards (up to 95% in backward states), only about 20% get free treatment -- less than 3% in Haryana and Punjab and about 7% in Uttar Pradesh). Presenting an analysis of NSS data Dr Krishnan notes that the cost of care at both private and public hospitals is higher in more affluent states and lower in backward ones.

Even in public hospitals, the burden of treatment will depend on a number of factors -- cost of treatment, type of illness, various sources of finance, effect on the family income, etc. Dr Krishnan calculates that for the bottom 10% income class, "While (the ratio of treatment cost to the annual per capita expenditure) is below 30% in Kerala, Tamil Nadu and West Bengal, the burden of treatment in government hospitals in the rural sector varies between 100 and 230% in Bihar, Assam, Punjab, Rajasthan, Haryana and Uttar Pradesh." The burden of private care for the same group "exceeds 100% in all states except Assam , West Bengal , Kerala and Tamil Nadu, implying thereby that it would have entailed either incurring debts to pay for treatment or decline in the overall consumption levels of the household." The burden of public care is much less for urban people -- except in Uttar Pradesh where it is 350%.

In other words, the burden of treatment will worsen poverty. There is a direct link between access to healthcare and poverty. Dr Krishnan notes that this picture should signal that privatisation of healthcare would be suicidal. The priority must be to provide access to universal healthcare in public services.

(Sandhya Srinivasan is a freelance health writer and executive editor of The Indian Journal of Medical Ethics. Email : This email address is being protected from spambots. You need JavaScript enabled to view it.)

InfoChange News & Features, June 2005