Purpose
– The purpose of this paper is threefold: first, to measure the health inequalities among malnourished children; second, to decompose the health inequalities to identify key socioeconomic predictors for child malnutrition; and third, to assess the change in the proportional contribution of key predictors over time.

Design/methodology/approach
– The study has used data of National Family Health Survey (NFHS) conducted in 1992-1993, 1998-1999 and 2005-2006. The information on anthropometric indicators for children below three years of age is provided; however the study is restricted to “weight-for-age,” as it is considered to be a comprehensive indicator of child nutritional status. In the first stage of analysis, health inequalities are measured among malnourished children using concentration indices (CI) for each round of NFHS. In second stage, the inequalities are decomposed to estimate the proportional contribution of socioeconomic predictors. In the third stage, change in the relative contribution of socioeconomic predictors over three rounds is assessed to suggest target-specific policies and programs.

Findings
– The results highlight a slow decline of only seven percentage points in the proportion of malnourished children in India during 1992-2006. The increasing values of CIs from -0.13 (1992) to -0.18 (2006) demonstrates the concomitant rise in economic inequalities among malnourished children. The results of decomposition analysis point toward household poor economic status and mother’s illiteracy as the major contributor of inequalities during 1992-2006. During the study period, the economic status explained 50, 65 and 59 percent of inequalities, whereas mother’s illiteracy explained 40, 30 and 29 percent of inequalities, respectively. Overall, the contributors to health inequalities remained the same over time with a change in their relative contribution.

Research limitations/implications
– The present study is focussed on three rounds of NFHS data conducted at different time period and so it has certain limitations which should be kept in mind while interpreting the results. The study has revealed mother’s education and economic status of the household as the major contributory factors toward child health inequalities. However, one should not forget that the level and quality of education has undergone tremendous change from 1992 to 2006 which the authors could not capture considering the availability of data in the form of years of schooling. Second, since the NFHS-1 has collected the information about the caste groups in only three categories, i.e. schedule caste, schedule tribe and others; the authors have to pool the other backward caste groups with the general caste groups. Third, the authors have used the broad classification of place of residence, i.e. rural and urban area to analyze the inequalities assuming the uniform level of development across the urban regions; however there exists huge disparities within urban areas which leave scope for further research. Fourth, though, the authors have estimated the wealth based inequalities, but NFHS does not provide the absolute level of wealth and so the authors have computed the proxy measure for wealth based on the household assets which has been extensively used in the field of research. Despite these limitations, the authors however believe that the present research work has appropriately decomposed the inequalities among malnourished children and have revealed the changes in the proportional contribution of socioeconomic factors over time.

Practical implications
– The decomposition analysis brought into light that average health indicators are insufficient for determining the right approach to health intervention programs. Health policy interventions have to focus ideally on both health averages and within and between group inequalities based on varying contributions of socioeconomic determinants.

Social implications
– Concentrated efforts along with the inter-sectoral concurrence, good nutrition governance, effective investment and unequal distribution of resources are pre-requisites to ameliorate the level and existing inequalities in child malnutrition in India.

Originality/value
– The distinctiveness of this study can be primarily found in the use of all three rounds of NFHS data to estimate health inequalities among underweight children. The study has also decomposed the health inequalities to estimate and analyze the change in relative contribution of socioeconomic predictors for each round to facilitate the formulation of target-specific policies and programs.