The Hindu Business Line : Public health spending and outcomes in States

In the previous edition of Macroscan, C. P. Chandrasekhar and Jayati Ghosh discussed the inadequate levels of Central Government expenditure on health, which has led to increased private expenditure. In this edition, patterns of State government spending are examined, to consider whether there is any apparent relationship between public health expenditure and health outcomes in four States.

Health is a Concurrent subject under the Indian Constitution, but State governments are dominantly responsible for most health provision, both curative and public health aspects.

While there are some specific central intervention, especially in various "Missions" as well as high-end curative facilities, the bulk of the health provision that affects most of the citizenry is the result of spending by State governments.

State governments account for about two-thirds and the Centre about one-third of the total public spending on health.

However, as Table 1 shows, there are large variations in this ratio across States, as well as in per capita spending on health by State governments.

It is apparent that low public expenditure can play a role in leading to higher levels of private expenditure, even in poorer States where presumably per capita consumption expenditure is already lower. Further, there are very wide rural-urban differentials in public spending, with per capita spending in urban areas generally much higher than in rural areas for most States (with the exception of Assam, Punjab and Rajasthan).

Obviously, there is no direct and clear linkage that can be made between government health expenditure and health outcomes of the people.

However, certain broad tendencies can certainly be identified. In particular, the unacceptably slow improvement in a range of basic health indicators such as life expectancy at birth, infant mortality and maternal mortality, can be partially traced to inadequate public expenditure and intervention.

Furthermore, the expenditure on public health does have a direct impact on certain health indicators such as the spread of communicable diseases.

And, as Table 2 shows, while there may not be a direct and regular relationship, it is certainly the case that there is a broad overall correlation between per capita health spending of the State government and certain basic health indicators.

In Table 2, States are placed in descending order according to the government's per capita expenditure on health and family welfare taken together.

Some broad features are evident: The richer States tend to have higher per capita spending, and generally tend to have lower infant mortality rates (IMRs), greater incidence of full antenatal care and safe deliveries, and better nutrition indicators.

However, it is not the case that the richer States all have higher per capita health expenditure — rather, some of them, such as Gujarat and Haryana, have among the lowest per capita expenditure on health.

Conversely, Rajasthan, with relatively low per capita income, has relatively higher per capita government spending on health.

While there is no exact correlation between per capita health spending and some basic indicators such as life expectancy, IMR and Under-5 MR, there is a broad relationship along expected lines.

However, it should be noted that (barring Kerala) most indicators even in the high-spending States are relatively poor by international standards.

NFHS-3 results

The recently released preliminary results of the NFHS-3 allow for a consideration of some health indicators over time in four of the States considered here: Gujarat, Maharashtra, Orissa and Punjab. Chart 1 provides trends in per capita expenditure on health and family welfare by these State governments, in constant 1993-94 prices, and the subsequent charts provide some information on health outcomes in these States based on successive NFHS (National Family Health Survey) reports.

As is evident from Chart 1, real per capita spending on health and family welfare showed a volatile pattern in these States, stagnating or even declining slightly at the start of this period, increasing over the late 1990s and then once again stagnating or declining.

There have been relatively sharp falls in Maharashtra and Gujarat from the turn of the decade, uneven pattern in Maharashtra and stagnation in Orissa.

While the real per capita expenditure on health and family welfare remained the highest in Punjab throughout the period, the relative ranking of Gujarat and Maharashtra fluctuated, and Orissa continued to have the lowest though the gap narrowed with the middle two States.

However, as Chart 2 shows, this is only partly reflected in IMRs in these States. While that in Orissa was the highest, it also showed the sharpest decline, whereas the IMR of Punjab has been comparable with that of Maharashtra with much lower per capita spending and has also shown a greater variation over time.

Disturbing trend in immunisation. - G Moorthy

But the most direct impact of public spending appears to be felt in a very significant indicator — the proportion of children in the age group 12-23 months who have undergone the full required immunisations, that is BCG plus 3 polio plus 3 DPT plus measles.

The shocking fact that emerges from Chart 3 is that, apart from Orissa (where as we have seen, government health expenditure in real per capita terms has increased slightly in the period between NFHS-2 and NFHS-3), immunisation rates have actually fallen in other States.

The decline is very dramatic in the case of Maharashtra and Punjab, but is also clearly evident for Gujarat.

In fact, it turns out that even in Orissa, only measles coverage has improved and coverage of the other vaccines has worsened. In Punjab, Gujarat and Haryana, the decline is essentially because of the much reduced coverage of both DPT and polio.

It is worth noting that polio, which was supposed to be eradicated from India, has witnessed a recent upsurge, and that the incidence of both diphtheria and tetanus has increased even though these are both easily and completely preventable through immunisation.

The low and declining extent of immunisation coverage therefore suggests greater vulnerability of the population to diseases which are entirely preventable.

This worsening of one of the most basic indicators of public health appears to be the direct result of reduced government expenditure which, in turn, has reduced the spread of and access to vaccination among the general populations, and particularly in the rural areas.

This suggests that while infant mortality rates may not directly reflect the trends in government health expenditure so clearly (although there are broad relationships that can be discerned), immunisation coverage appears to be directly (and even almost immediately) affected by public spending in this area.

The reduced expenditure can, therefore, have alarming and completely unnecessary adverse effects upon the health of the population, which will eventually reflect not in mortality per se but in increased morbidity and reduced capabilities.

(Concluded)