Health inequality caused by urbanization in developing countries
My speciality is health economics, an interdisciplinary field at the intersection of economics, public health, and epidemiology. Health economics is widely recognized in Europe and the United States, and interest is gradually expanding in Japan too, given the link between health economics and policy formation.
My specific research theme is health inequality in developing countries such as those in Asia, where lifestyles are changing due to economic growth and urbanisation, and problems such as obesity are spreading throughout society. The major causes of death are also changing, from infectious diseases to non-infectious diseases such as diabetes. Moreover, the health situation varies depending on socio-economic conditions, and countermeasures have not caught up with the changes. In other words, there is inequality in health status.
I have been analysing health data from developing countries from various perspectives, measuring the effectiveness of existing health-related policy, and making policy recommendations for mitigating inequality.
Did India’s cash benefits programme increase the infant survival rate there?
In my latest study, I evaluated the effectiveness of an Indian cash benefit programme for mothers. Of all nations, India has the highest number of infant deaths, accounting for more than 20% of the world total. Women in India generally have to work until just before childbirth, and it is difficult for them to get proper nutrition and medical treatment, even during pregnancy. That situation is linked to India’s high infant mortality rate.
In response to that situation, during 2010–2011 India introduced Indira Gandhi Matriya Sahayog Yojana (IGMSY), a cash benefit programme, in some districts. In that programme, mothers receive cash three times: 1500 rupees in the sixth month of pregnancy, 1500 rupees in the third month after childbirth, and 1000 rupees in the sixth month after childbirth, if they satisfy the conditions, which include notification to the government office, examination during pregnancy, participation in counselling, and vaccination of the newborn.
I evaluated IGMSY’s contribution to the improvement of infant survival. To assess the effectiveness of the programme, I investigated whether there was a difference between infant survival in the programme implementation area and other areas, before and after the implementation of the programme. I designated infants born in the implementation area as “treatment district group” and those born in non-implementation areas as “control district group” and then estimated the changes in survival rate for both groups, both before and after the implementation of the programme.
The data that I used to verify the effects were taken from the 2015/16 edition of the National Family Health Survey (NFHS), a survey of health and socio-economic conditions throughout India. I used the data for infants born in 2010 and 2011 (before the implementation of the programme) and in 2012 and later (after the implementation of the programme) and analysed the data for the treatment districts and control districts. The results are shown in Figure 1.
The survival rates for the treatment and control districts before IGMSY implementation were not significantly different. On the other hand, there was a considerable difference between the survival rates for the two groups after the implementation; it can be seen in the figure that improvement in the treatment districts was greater in the control districts. Figure 2 presents that result graphically to show how many infants were saved per 1,000 infants.
The “difference-in-differences” technique was used to remove the fixed effect of time and region from the effect of IGMSY. A machine learning model called “Random Survival Forest” (created for survival analysis) was used in the estimations.
A significant change in infant survival rate was found. After seven months of age, the treatment effect becomes even larger. Even though the IGMSY programme cash benefits ended six months after birth, the effectiveness of the programme continued beyond that point. I can conclude that measures such as cash benefits, counselling, and vaccinations for babies had a cumulative effect.
The effect of socio-economic attributes such as infant gender and caste on IMGSY results was analysed from four perspectives (Fig. 3).
Figure 3-A shows infant rate improvement by gender. As can be seen in the figure, improvement was much greater for boys than for girls. The culture of India, where boys are treated better than girls, may have influenced the effectiveness.
Figure 3-B shows the effect across mother’s education level on child mortality. Higher effects were seen for less educated mothers, who had minimal general knowledge at the outset, so there was more room for growth; also, it is thought that IGMSY counselling was effective.
Figure 3-C shows the effect of caste on improvement rate. Stronger effects were seen in high caste infants. In the case of low caste families, the literacy rate may have been too low, and those families likely reside in areas with poor access to medical care. Those effects may have prevented mothers from accessing the programme in the middle term, which would constitute a barrier to achieving the maximum expected effect.
Figure 3-D shows the results by area of residence. Stronger effects can be seen for residents of urban areas, where there are more hospitals and better-developed transportation networks than in rural areas, which could have had an impact on the child mortality results.
The results of this study indicate that the implementation of IGMSY saved the lives of a considerable number of infants. However, while the child mortality rate improved for relatively wealthy families, that may not have been the case for the poorest families. I plan to analyse such inequality in more detail in the future.
Future study: the relationship between health inequality and childhood environment
Aside from measuring the effects of IGMSY, I would like to conduct several other studies. As I explained earlier, inequality resulting from contracting non-infectious diseases such as diabetes and hypertension is now widespread in many countries. I am very interested in how much those differences can be explained by environmental factors beyond the control of the subjects. In the future, I would like to analyse those issues not only for the case of developing countries, but also for developed countries including Japan.
Interview and composition: MATSUMOTO Kaori
In cooperation with: Waseda University Graduate School of Political Science J-School