Although former approach usually produce clearer causal relationship, it typically based on more stricter assumptions. From this perspective, second approach is not only simpler, but also impose less restrictions. There are different decomposition methods that have been developed in poverty and income inequality literatures over last few decades. Kakwani and Subbaro (1990); Jian and Tandulkat
(1990); Ravallion and Huppi (1991); Datt and Ravallion (1992); Shorrocks and Kolenikov 2001; and Dhongde (2002) propose different methods to decompose the changes in poverty (measured in headcount ratio, poverty gap and squared poverty gap) over time into change due to a rise in the mean income level (growth component) and the change due to changes in the distribution of income (distribution component). Such methods can be used to decompose the change in overall health outcomes over time into change due to a change in the mean health measures (within group component) and the change due to changes in the distribution of population (between group component).
Different decomposition methods are available to decompose socio-economic health