«Abstract Recent economic research on relationship between health and wealth has noted that this relationship operates in both directions; higher ...»
Adult Mortality in India: The Health-wealth Nexus
Debasis Barik1, Sonalde Desai2, Reeve Vanneman3
Recent economic research on relationship between health and wealth has noted that
this relationship operates in both directions; higher incomes lead to greater access to
healthcare while healthier individuals are likely to earn more. In societies like India
that are in midst of epidemiological transition, a third factor may also be important.
Higher income individuals may more likely be afflicted by life-style diseases that increase mortality. Using unique panel data from IHDS of 2004-5 and 2011-12, we examine the relationship between household wealth in 2004-5 and probability of dying in the subsequent seven years for adults ages 15 and above. The results show that although wealth is likely to be associated with slightly higher prevalence of noncommunicable diseases, wealthier individuals are less likely to die even after controlling for these factors. Moreover, individuals in the top wealth quintile, even with diseases are less likely to die than their poorer peers.
Adult mortality in India Adult Mortality in India: The Health-wealth Nexus Debasis Barik1, Sonalde Desai2, Reeve Vanneman3 Addressing Socioeconomic Gradient in Health Disparities In most societies, people with higher social status enjoy good health and lower mortality. More schooling, higher incomes, prestigious jobs etc. provide knowledge and resources to live healthier and for a longer period. These disparities in health outcomes among various population sub-groups by their relative socio-economic position have increasingly drawn attention of researchers and policymakers alike (Deaton, 2002). Though education is widely perceived as the single most important socio-economic determinant of mortality (Antonovsky 1967, Kitagawa 1973, Preston and Taubman 1994, Elo and Preston 1996), Liu, Hermalin et al. (1998) found that the direct influence of education to lower mortality was very low and insignificant among the older Taiwanese. 83 percent of the total effect of education on mortality was indirect, mostly by means of health disparities. However, these disparities in health outcomes are neither consistent across countries, nor over time (Preston 1975) making it difficult to develop public policies to address these disparities.
Arguably the greatest challenge to understand the nature and causes of health disparities lies in the reciprocity of this relationship. While we can expect greater economic resources to translate into better nutrition and more access to health care, thereby reducing mortality, a large number of studies also document that poor health leads to unemployment and lower income (Grossman 1972, Smith 1999). For Adult mortality in India individuals suffering from chronic illnesses, treatment costs will also reduce their disposable income and divert other family members from productive activity.
Moreover, in low income countries like India, rising incomes have created a dual nutrition burden in which the poor continue to suffer from malnutrition but the rich are afflicted with obesity and a rising burden of cardio-vascular diseases (Ke-You and Da-Wei 2001, Sengupta and Syamala 2012, Sengupta, Angeli et al. 2014).
In this paper we address some of these challenges to examine the link between household wealth and mortality in India using prospective survey data from India Human Development Survey (IHDS), a nationally representative panel survey conducted in 2004-5 and 2011-12.
Health and Wealth: Correlation and Causation A strong relationship between economic deprivation and ill health was first scientifically documented by René Villermé, who compared mortality rates and poverty across the arrondissements of Paris in the 1820s, although references to the relationship can be found even in ancient Greek and Chinese texts (Deaton 2002). In England and Wales, systematic documentation of mortality by occupational class began as early as 1851 with the publication of Decennial Supplements to the Annual Report of the Registrar General. Social class differentials in mortality became the focus of systematic study in United States only in the latter half of the twentieth century with the publication of Kitagawa & Hauser’s path-breaking study of demographic and socioeconomic mortality differentials based on the 1960 Census matched to death certificates filed in May–August of the same year(Kitagawa 1973, Adult mortality in India Hummer, Rogers et al. 1998). Though there exists ample literature on the nexus between socio-economic status and health and mortality, research on the issue in an Asian context are visible only after the late 1990s (Liu, Hermalin et al. 1998, Liang, McCarthy et al. 2000, Zimmer and Amornsirisomboon 2001, Zimmer, Kaneda et al.
2007, Zimmer, Martin et al. 2007, Zimmer 2008, Chen, Yang et al. 2010).
In spite of the considerable body of evidence showing this correlation, the direction of causation is not clearly established. Economists usually argue that poor health restricts a family’s capacity to earn income or accumulate assets by limiting work or by raising medical expenses. The alternative pathway running from wealth to health is more popular among medical practitioners and public health researchers.
James P. Smith (1999) in his pioneering research “Healthy Bodies and Thick Wallet” concluded that the causal direction of the social health gradient is not clearly understood; it varies with age. In pre-retirement, health affects income; for older individuals, income affects health. But, several researchers have also expressed their concerns regarding the hazard associated with wealth. The hazard of wealth includes the excessive consumption of food, alcohol and tobacco, linked to physical inactivity and other lifestyle factors (Razzell and Spence 2006).
Whether socioeconomic status has a protective impact on health outcomes also depends on the pathways through which this socioeconomic status is able to spread a protective umbrella over individual health. Preston and Haines (Preston and Haines 1991) argued that at the turn of the 20th century even educated or better off mothers could do little to protect their children from death before the germ theory became well known. Similarly, studies in Sub Saharan Africa in late 20th century Adult mortality in India where communicable diseases have held sway, found that maternal education-child health linkages were the weakest (Hobcraft, McDonald et al. 1984, Hobcraft 1993, Desai and Alva 1998).
Challenge of Rising Prosperity As we noted earlier, a large number of studies have noted that higher incomes are associated with lower mortality (Kawachi, Kennedy et al. 1997, McDonough, Duncan et al. 1997, Ross, Wolfson et al. 2000, Mackenbach 2002, Muller 2002). But pathways through which this relationship operates is not always clear nor do we understand how social context shapes this relationship. For example, using the Asset and Health Dynamics among the Oldest-Old (AHEAD) panel, Adams et al. (Adams, Hurd et al. 2003) found no direct causal link from household SES to mortality among elderly Americans, but observed an association between SES and the incidence of gradual onset health conditions. They explain the weak link between SES and mortality among American elderly as mainly due to Medicare coverage and universal access to pensions which do not depend upon the ability to work in old age.
In India, higher incomes pose a very different challenge. Public health spending is miniscule (only 1.03 percent of the GDP); people mostly rely on their own spending capacity to combat ill-health with more than 80% of the illnesses being treated by private physicians (Barik and Desai 2014) and more than three-fifth of the total health care expenditure are met by households through out-of-pocket payments (World Bank 2011). This would suggest that higher income households should be Adult mortality in India able to get better care. However, there are reasons to doubt the strength of this relationship.
Much of India’s public health system is geared around providing primary care for communicable diseases (IDFC Ltd. 2014). However, with rising prosperity and associated obesity, cardiovascular diseases become more important (Venkatramana and Reddy 2002, Srinath Reddy, Shah et al. 2005, Ghosh 2006) and the health sector is poorly equipped to deal with these diseases. A study of doctors around New Delhi found that very few doctors, even private doctors, could identify symptoms of heart attack (Das and Hammer 2007). This lack of familiarity with diseases of the more prosperous could reduce the usual relationship between wealth and health. This brief review suggests that the strength of the relationship between socioeconomic status and mortality deserves to be empirically examined.
Nexus of Income, Life Style Diseases and Mortality in India India is the second-fastest-growing economy in the world. The Indian economy grew at an average rate of 7.25 percent in the first decade of the twenty first century (2000-10), resulting in rising per capita incomes and declining poverty.
Researchers have documented a sharp income growth in both rural and urban areas during this period (Mitra and Saxena 2013).
However, Gillespie and Kadiyala (2012) has argued that the high level economic growth in India was far less pro-poor than its other Asian counterparts to reduce social-ills like child malnutrition. These growing incomes have not led to better health outcomes. For example, studies of dietary diversity document declining Adult mortality in India diversity over time (Gaiha, Kaicker et al. 2013), anemia remains prevalent at almost all income levels (NFHS – III), and the proportion of individuals suffering from noncommunicable diseases has grown even as India has experienced a surge of economic growth. Cardiovascular diseases, stroke, diabetes, cancer are the four leading NCDs in India(Upadhyay 2012). India has the highest number of people with diabetes than any other country in the world (Ghaffar, Reddy et al. 2004), so it is often referred to as the diabetic capital of the world (IDF 2009).
At a global level, the infectious and parasitic health disorders in the past are now being replaced by chronic, non-communicable conditions as evident from the latest Global Burden of Disease report 2013 (IHME 2013). The share of noncommunicable diseases on total disability adjusted life years (DALY) has increased from 31% in 1990 to 43% in 2010. The steep rise in the prevalence of noncommunicable diseases has spread across regions where more developed regions are prone to a higher prevalence. These illnesses usually incapacitate a person for a longer period and claim a huge toll on the individual and the welfare of the family.
This issue is particularly critical for India since South Asian populations in the abroad have also shown very high rates of diabetes, high blood pressure and heart conditions (Gunarathne, Patel et al. 2009, Gupta, Wu et al. 2011). Coronary heart disease rates have been reported to be unusually high in several parts of the world in people originating from the Indian subcontinent (McKeigue, Miller et al. 1989). A UK study showed that men and women from India had the highest standardized mortality rates due to cardiovascular disease, and that young Indian men were at particularly high risk (Balarajan, Bulusu et al. 1984). Harding (2003) also noted that cardiovascular Adult mortality in India and cancer mortality of South Asian migrants increased with duration of residence in England and Wales. But, it is not clear whether the high burden of NCD among overseas South Asian populations abroad may be due to dietary/environmental factors or genetic predisposition.
So, on the one hand rising incomes place individuals in lifestyles that are more prone to sedentary life-style diseases such as diabetes, heart disease, and high blood pressure. On the other hand, rising incomes also make it possible to seek better health care. The Indian health system is mostly privately funded with more than 60 percent of all treatment costs borne by the family members from out-of-pocket spending.
Thus the burden of treatment cost is disproportionately distributed among various income classes ranging from less than a percent among the top quintile to 15 per cent among the lowest quintile (Barik and Desai 2014). Out of India’s small health care expenditure (less than 4 percent of GDP), only one-fourth is funded by the central and the state governments. Although some efforts are now being made to provide hospitalization coverage to the poor (CPR 2011, IDFC Ltd. 2014), only a few households have health insurance. A lack of access to good medical care in rural areas is particularly problematic and often requires considerable expenditure. Although there has been some increase in secondary and tertiary care units like Tehsil or District level hospitals and specialty hospitals like All India Institute of Medical Sciences (AIIMS), etc. in the last decade, the majority of rural India depends heavily on the usually poor performance of primary health centres and sub-centres even for emergency care. Access to either public or private specialized health care centres, concentrated in urban India, can be costly. This implies that even as higher incomes Adult mortality in India increase the risk of life-style related diseases, they also allow for better treatment of those diseases so income’s net impact on mortality remains subject to empirical examination.
Absence of Research on Adult Mortality in India Most of the research on mortality in India has focused on infant and child mortality (Singh, Pathak et al. 2011, Ghosh 2012, Kumar, Singh et al. 2013). However, adult mortality research in India still remains in its infancy. Earlier studies of adult mortality in India were more concentrated on the levels and trends (Dandekar 1972, Dyson 1984, Clark 1987) of mortality. Preston (1980) discussed only the major causes of mortality decline in some less developed countries including India, and focused mainly on the macro-level contributors of this decline such as per-capita national income and the prevalence of various diseases.
Unfortunately, India has lacked comprehensive data for the analysis of individual and household level predictors of adult mortality. India has a vital registration system to record vital statistics like birth, deaths, and marriage, but it is frequently incomplete, particularly in rural areas, and tends to produce dusty records that are difficult to analyze. Adult mortality statistics come mainly from the Sample Registration System (SRS), which is fairly complete but lacks socio-economic information about individuals.