The infant mortality rate (IMR) is the number of deaths of infants under one year of age per thousand live births in a specified year. Its denominator is live births rather than the population of infants, which makes it a quasi-probability of dying in the first year rather than a true rate, and its unit of a thousand reflects how uncommon infant death becomes once conditions improve. Sociologists use it less as a health statistic than as a social indicator — a compact reading of how a society feeds, houses, educates and attends to its most vulnerable members.
IMR must be distinguished from neighbouring measures. Neonatal mortality counts deaths in the first 28 days; post-neonatal mortality covers the remainder of the first year; under-five mortality extends to the fifth birthday; and the maternal mortality ratio concerns the mother. The internal split matters because the two components have different causes. Neonatal deaths arise overwhelmingly from prematurity, low birth weight, birth asphyxia and infection, and therefore from the mother's own nutrition, age at childbirth, birth interval and access to skilled delivery care. Post-neonatal deaths arise mainly from diarrhoea, pneumonia and undernutrition, and therefore from water, sanitation, feeding practices and immunisation. As IMR falls, the neonatal share rises — which is why countries that have solved the environmental causes find further reduction hardest.
Why sociologists treat it as the key indicator
Infant survival is the outcome of a chain of ordinary social arrangements rather than of any single input. Mosley and Chen's framework showed that socio-economic determinants act only through proximate determinants — maternal factors, environmental contamination, nutrient availability, injury and personal illness control — and this is precisely why the indicator is so informative: it aggregates conditions across the household. John Caldwell's comparative work isolated maternal education as the strongest single predictor, operating not through income but through changed behaviour: an educated mother recognises danger signs earlier, uses health services more assertively, distributes food within the household differently and is less overruled in decisions about the child. Female autonomy, in other words, converts into infant survival.
The corollary is that IMR is only loosely coupled to per capita income. Societies at similar income levels record very different rates depending on public provision of water, sanitation, immunisation and primary care, and on how far women can act on their own judgement. Rising income helps most when it passes through public services and least when it is captured by a small stratum. This is the sense in which IMR measures development rather than growth.
The sex differential
Biologically, male infants die at higher rates than female infants; a normal population therefore records a male disadvantage in infancy. In parts of India this advantage is reversed. Where discriminatory care operates — shorter breastfeeding of daughters, later and less persistent treatment-seeking when a girl falls ill, smaller shares of nutrition and of household health spending, quicker recourse to hospital for a son — girls' survival chances fall below boys'. Excess female mortality in infancy and early childhood is thus a distinct phenomenon from sex-selective abortion: the first kills after birth through neglect, the second prevents birth. Both express son preference, and together they account for much of the deficit Amartya Sen described in his account of missing women. NFHS rounds and Sample Registration System reports have repeatedly recorded a female disadvantage in child mortality in India even while overall infant mortality has declined — the qualitative pattern is well established, and the point to make in an answer is the reversal itself rather than any particular figure.
Kerala and the northern states
The internal Indian contrast is the standard demonstration that mortality is socially rather than purely economically determined. Kerala achieved infant mortality comparable to middle-income countries while remaining, for most of the twentieth century, a poor state in per capita terms. The explanations offered by Moni Nag, and later by Drèze and Sen, converge on public action and gender: near-universal female literacy, a dense network of accessible primary health facilities, high rates of institutional delivery and immunisation, matrilineal inheritance traditions in some communities, and public distribution of food. The large northern states — Uttar Pradesh, Bihar, Madhya Pradesh, Rajasthan — recorded far higher rates at comparable or higher incomes, alongside lower female literacy, earlier marriage and childbearing, weaker health infrastructure and more restricted female mobility. Tamil Nadu's later convergence on Kerala through determined public health investment strengthens the argument, since it shows the result is reproducible rather than a Keralan peculiarity. Both interstate and rural-urban differentials, and the gap by caste and by mother's schooling, follow the same logic.
Measurement and its problems
The apparent precision of IMR conceals real difficulty. Civil registration of births and deaths remains incomplete in parts of India, so the principal source is the Sample Registration System, a dual-record survey combining continuous enumeration with half-yearly retrospective survey; the NFHS supplies an independent estimate from women's birth histories. Both face known biases. Very early neonatal deaths may be recorded as stillbirths, or not reported at all where a death within days of birth is treated as a non-event; recall error grows with the interval since the birth; and the definition of a live birth is applied unevenly. Because IMR is a small number, modest under-reporting produces large proportional error, and small-area estimates are unstable. Sociologically the under-reporting is itself data: what a society declines to register tells us how it values the life in question.
For the UPSC answer
Define IMR exactly — under one year, per thousand live births — and immediately split it into neonatal and post-neonatal components, since the causal accounts and the policy remedies differ. Make the central argument the Caldwell one: maternal education and female autonomy, not income, drive infant survival, which is why IMR works as a summary indicator of development. Use Kerala against the large northern states as the illustration, adding Tamil Nadu to show the result can be reproduced by public action. Mention the reversed sex differential as evidence of discrimination in care, and close with the registration problems so the answer treats the number critically rather than reciting it.
References & further reading
- Registrar General of India. Sample Registration System Statistical Report (annual series). Government of India.
- International Institute for Population Sciences (2021). National Family Health Survey, NFHS-5. IIPS.
- Caldwell, J. C. (1979). Education as a Factor in Mortality Decline. Population Studies, 33(3).
- Mosley, W. H. and Chen, L. C. (1984). An Analytical Framework for the Study of Child Survival in Developing Countries. Population and Development Review, 10.
- Nag, M. (1983). Impact of Social and Economic Development on Mortality: A Comparative Study of Kerala and West Bengal. Economic and Political Weekly.
- Drèze, J. and Sen, A. (2013). An Uncertain Glory: India and Its Contradictions. Allen Lane.