“There are three kinds of lies – lies, damned lies and statistics”
(Mark Twain: Chapters from My Autobiography)
A recent comment by India’s Prime Minister (PM) during an election speech comparing the infant mortality rate (IMR) in the tribal areas of Kerala state with those in Somalia kicked up a furore. A wounded Chief Minister of Kerala (from the opposing political party) has threatened to sue the PM, though the exact nature of the offence is not clear. Now that the electoral battle in Kerala has been lost and won, it is time we dispassionately analysed the contention of the PM and the implications for health policy in India. Let us first get to the numbers; at 60 deaths per 1000 live births in the tribal areas of Northern and Eastern Kerala, he felt that the area was not lagging far behind the African country of Somalia, which, as per the number he had, registered 85 deaths per 1000 live births in 2015. This is where statistics can be dangerous, and it does not need a Mark Twain to convey this message. Firstly, there seems to be no basis for concluding that the tribal areas of Kerala have an IMR of 60: whether this covers just the tribal population or the districts with a larger proportion of tribal population is not clear. Secondly, the PM’s information feeders seem to have culled the magic number of 85 from the latest country wise estimates of infant mortality released by the UN Inter-agency Group for Child Mortality Estimation (www.childmortality.org). The problem, as with all statistics, lies in the level of confidence reposed by the estimators in their own estimates. In the present case, three sets of numbers are given for each country: low, median and high. While the variation in these three numbers in countries like the United Kingdom with excellent reporting systems is minimal (3.0 to 3.5 to 4.1) and reasonable for a country like India (34.1 to 37.9 to 41.8), the range from the low to high figure is from 53.3 to 143.3, with a median figure of 85, for a country like Somalia with underdeveloped reporting systems. The UNICEF State of the World’s Children Report 2015 gives an IMR of 108 for Somalia and the CIA Fact Book places it at 98, showing that there is no unanimity on the number. With such a vast range of uncertainty regarding the numbers, it would be hazardous to plump for a number like 85 with any degree of confidence. The matter is further complicated when we compare the tribal population of Kerala with that of Somalia – 0.49 million versus 34 million. A small population, especially when it is largely comprised of poor tribals, will display higher figures of mortality in infants, given the prevalence of poverty and the poor reach of essential health services. The law of averages operates as the sizes of populations increase. To give one graphic example: just two infant deaths in a village with a population of 1000 (with an annual population growth rate of 2%) imply an IMR of about 100 per 1000 live births: which is why mortality statistics are never calculated below the district level. As statistics combine disadvantaged with more prosperous areas, these numbers come down, in the case of Kerala state to 12 per 1000 live births, which compares very favourably with many developed country figures.
The tragedy lies in the lessons that are not learnt from areas in Kerala like Wayanad, Idukki and Attappady, Palakkad (in the news a couple of years ago for infant deaths in significant numbers) and the mistakes committed through apathy and misgovernance across much larger swathes of India. Politicians would do well to remember the adage “Those who live in glass houses shouldn’t throw stones”. Of the nine states that are at the top of the high child malnutrition pecking order, seven are presently ruled by the party whose PM has spoken disparagingly about IMR levels in the tribal areas of Kerala, and all nine, including his own state of Gujarat, have or have had BJP governments (either on their own or in alliance with other parties). Barring two states where the BJP has recently come to power, its governments have had ample time to tackle the menace of child malnutrition, which is attributed by experts to contribute at least 45% of child deaths in India (and probably an even greater percentage of infant deaths, given that an overwhelming majority of under-5 children die before they cross the age of one). And yet, it is precisely these states which are the greatest contributors to infant mortality and child malnutrition. Don’t get me wrong: I am not in any way absolving other political parties which have ruled these states for many years without making a significant difference to the problem. The fault, dear Brutus, lies not in our stars, but in ourselves: in our dysfunctional systems, our cavalier disregard of data, our failure to focus on key geographical areas with a high child malnutrition burden and our failure to evolve a coherent, time bound public policy to effectively tackle the problem.
Let us start with our dismal grasp of the magnitude of the problem. Growth monitoring has always been one of the main components of the ICDS strategy right from its inception. Unfortunately, the monthly exercise of weighing of all under-5 children by the Anganwadi Worker (AWW) has been treated mostly as a routine task, with little or no importance being given to the use of this massive body of raw data. In the absence of weighing scales, weighing is sometimes not carried out; where weighing is done, there is no analysis of the data to chart out a meaningful course of remedial action in case of underweight children at any level, whether the anganwadi, ICDS project, district or state. Almost no state posts aggregated data, ICDS project wise, on the nutritional status of children online and it is doubtful if any administrator, at the project, district or state level, pays any attention to this data.
This blissful neglect of valuable data leads to governmental failure to identify and focus attention on the geographical regions requiring urgent, sustained intervention, be they remote tribal areas or congested urban slums. Aggregated data of monthly weights of children helps identify the specific localities (villages, hamlets, slums, etc.) that need to be focused on to reduce the burden of child malnutrition. The common budgetary approach of allocating funds to areas on a child population basis, without weightage for high burden malnutrition areas, discriminates against the latter. Poor infrastructure and inadequate staff in tribal areas lead to underutilisation of even allocated budgets. Resources of different departments are generally not combined in an innovative manner to deliver the crucial health and nutrition (both nutrition-specific and nutrition-sensitive) services to children and women that can reduce undernutrition and mortality. The new methodology of untying central fund releases to states is likely to see even further diminution in fund allocations to politically weak tribal regions of states and to urban slums. Public nutrition and health services for mothers and children are in short supply in urban slums. There are no systematic efforts to reach out to urban communities to develop their capacities to self-manage their nutrition and health issues. This limited attention given to identified high burden geographical areas is likely to see a continuation of high child malnutrition and mortality rates in these areas.
Resource misallocation to this critical area is aggravated by the absence of a clear cut vision on how to most effectively tackle the problem in the short run. India’s policy makers refuse to use height/length of under-5 children as a measure of nutrition status, in addition to weight (which has been used for nearly four decades). This would enable an immediate estimation of wasting (weight/height) status for taking action to improve the health and nutrition status of children suffering from severe acute malnutrition. Software exists to record both anthropometric measures digitally so that the wasting status of any child can be immediately established (a pilot project in Attappady, Kerala has proven the feasibility of such a digital approach to recording data using IVRS technology). Tackling moderate and severe wasting in India’s children (which goes upto 25% in many states) through inpatient and outpatient methods would significantly reduce malnutrition. But India’s ICDS and public health departments are unconvinced that they need to make this programme a key step to reduce child malnutrition and mortality. Adequate international evidence linking child malnutrition (especially wasting) to a higher incidence of mortality has had little to no impact on the thinking processes of the bulk of India’s medical professionals. Governments (at central and state levels) have failed to make such a programme the cornerstone of efforts to reduce malnutrition/mortality. The ICDS Commissionerates/Directorates are obsessed with centralised, contractor-dominated food supplies (rather than child feeding practices and micronutrient interventions), a policy which has drawn much critical comment from the Supreme Court and High Courts (the reasons are not difficult to fathom!). The resultant haphazard, ill-directed programmes to manage malnutrition, with no systematic measurement of nutrition outcomes and no focus on geographical areas most in need of such programmes, are the reason for India’s dismal world ranking in child nutrition indicators.
Finally, there is gross underutilisation of one of the most extensive systems set up anywhere in the world to deal with the issue of maternal and child nutrition — the ICDS. With anywhere from 50,000 to over 100,000 AWWs in each state of India, spread over almost every habitation in the country, this valuable human resource could well be the underpinning for a revolutionary transformation of the child malnutrition scenario in India. Unfortunately, with the ICDS largely functioning as a khichdi kitchen, these workers have never been empowered with the knowledge, skills and resources necessary to fulfil their innate potential. My experience in the nutrition sector in Maharashtra opened my eyes to the fantastic work they can do given the right working environment — upgraded knowledge/skills, access to resources, appreciation for their good work and the development of self-esteem for the important tasks they are undertaking. Even the huge public health system has no specific focus on the preventive aspects of health and good nutrition that could develop a generation of healthy girls and mothers, leading ipso facto to the birth of healthy, normal weight children.
For a country on the cusp of economic power and a growing global presence, it reflects poorly on India that she takes her place among the league of failed and failing nations in indices of child/infant mortality and undernutrition, whenever the exercise of evaluating each country’s performance in these areas is taken up. Latin America and East Asia have left us behind, as they made significant strides over the past few decades. Our immediate southern neighbour, Sri Lanka, is an object lesson to us that improvement in human development indicators can be achieved. Even within India, states like Goa, Kerala, Maharashtra and Tamil Nadu have performed far better than their counterparts in Northern and Eastern India in reducing IMR, though they still need to reduce wasting rates in under-5 children. If “Make in India” is to have any real meaning, children born in India need to have the guarantee of a healthy, disease-free, long and productive life.