Background
Health equity has been an important policy issue since the Alma-Ata Declaration of the World Health Organization (WHO). Since then all countries have been making efforts to reduce health inequities. Concerns have been expressed regarding the quality and availability of healthcare services in low and middle-income countries (LMICs) [
1,
2]. Accessibility and affordability of healthcare services are among the major healthcare challenges faced by developing countries [
3‐
5]. Financial barriers are key limitations to access healthcare services in LMICs since Out-of-Pocket Expenditure (OOPE)
1 constitutes a significant proportion of healthcare expenditure [
6‐
8]. Globally, about 1.3 billion people are deprived of access to effective and affordable health care. Majority of households spend more than 40% of their household income on medical treatment [
9,
10].
Possibilities are high of many households residing in LMICs being pushed into poverty when faced with high medical expenses, especially when it is combined with loss of income due to adverse health outcomes [
11,
12]. Measures to promote financial protection through universal health coverage (UHC) strategize to ensure that people would have access to health services without risking financial impoverishment [
13‐
15]. Health systems in many LMICs are financed through key sources such as, taxation, social security schemes, private health insurance and OOPE [
15].
More than half of the world’s population does not have access to formal social protection schemes [
16,
17]. Majority of the households who are unable to pay for using healthcare services either do not seek care or resort to short-term coping strategies such as minimizing food expenses, using past savings, and removing children from school to manage the financial shortfall [
18‐
20]. Short-term coping strategies may result in higher cost in the long run, leading to impoverishment and poverty. These households are not captured in poverty estimates, as high healthcare cost raises their expense above the threshold level and they are considered non-poor [
14].
Literature shows that the level of health payments also differs significantly with variations in certain specific characteristics of the households. Poor and disadvantaged sections of the population face more financial risks and need better financial mechanisms to avail healthcare services [
21,
22]. Literature also indicates differentials in healthcare spending among various regions as well as segments of population [
23,
24].
Indian scenario
Equity and justice in healthcare payment are integral parts of health policy draft in India [
25,
26]. Healthcare system in India is highly privatized and the main source of financing is OOPE [
27,
28]. OOPE does not provide any financial protection; as a result, such households incur heavy expenses in availing healthcare services [
29]. Expenditure on healthcare pushes a large number of families into poverty in India as they do not have sufficient spending power due to low level of income or sometimes, no fixed source of income [
30‐
32].
In India, more than 90% of the workforce, especially people who belong to socially and economically underprivileged sections of society, is engaged in informal economic activities.
2 As insurance facilities are available only to the workforce in the formal sector, majority of such households are not covered under any social protection scheme [
33,
34]. In the case of ill health, these households have to spend from their own pockets. Inadequate provisioning of health care facilities, coupled with a highly privatized health sector, further worsens the financial status of the poor and marginalized groups of the population [
35,
36]. It leads them into financial catastrophe and further deepens existing poverty [
36‐
39].
Level and pattern of OOPE are determined on the basis of socio-economic affiliations in India. Differences also prevail in the health spending and level of socio-economic inequalities among different Indian states. Different variables such as, socio-economic status, class, religious affiliation, place of residence, gender and age are used to classify the population [
35,
40,
41]. Among all these factors, economic status and caste affiliation are considered most important classificatory variables for analyzing socio-economic inequalities in health and health care expenditure in India [
42‐
45]. Socio-economic inequalities are highly unfavourable for healthcare, especially when society is more diversified, multi-ethnic, overpopulated and undergoing significant but unequal economic growth [
46].
There is evidence of wider inter-state differentials in public spending and health infrastructure across Indian states. The level of public spending on health in few of the backward states such as, Bihar, Jharkhand, and Odisha, is very low in comparison to Kerala, Punjab and other developed states [
47]. Studies also indicate increasing interstate inequalities in health spending in recent years. The difference between per capita OOPE among developed states such as, Kerala and Punjab, and backward states such as, Jharkhand, Chhattisgarh and Orissa, has increased, leading to greater divergence between these states [
48,
49].
Only a limited number of studies are available in the Indian contexts which have tried to examine the level of equity and regional variations in healthcare spending, by categorizing Indian states. We have classified the states on the basis of OOPE into three categories i.e., high, medium and low level of OOPE, and examined the pattern of inequalities in health spending among these states. This paper makes an effort to examine the pattern of health equity and regional disparities in healthcare spending among the Indian states by applying Andersen’s behavioural model of healthcare utilization.
Discussion
Equity in healthcare is one of the important and most desired goals to be achieved for any society. Inequalities in healthcare are measured on the basis of health outcome, utilization pattern and level of OOPE, between the Non-poor/Poor, Urban/Rural, Advantaged/Disadvantaged and other socio-economic groups of the population. Healthcare financing system should focus on achieving vertical equity (households of unequal ability should be treated unequally), horizontal equity (households of the same ability should be treated equally) and progressivity in healthcare expenditure [
55‐
57]. This paper focuses on the equity and regional perspective of healthcare expenditure among Indian states to uncover the linkages and the burden in the form of OOPE. Results indicate that burden of OOPE is not proportional among different subgroups of the population.
People residing in the urban areas, having higher economic status, belonging to non-Muslim and non-ST groups were spending significantly higher on healthcare than their counterparts. Literature also indicates that households that belong to the lower socio-economic status (Rural, STs, Muslims and lowest wealth quintile) were constantly experiencing poor health outcomes. This may be due to the fact that these people have minimum access to healthcare facilities or they were not in a position to pay for the use of healthcare services [
30,
58,
59].
There is also evidence of regional disparity in terms of healthcare spending among Indian states. It is evident from the results that in
Category A states (Kerala, Punjab, Maharashtra, Goa and Chandigarh), the average monthly expenditure on healthcare was highest followed by
Category B and
C. Findings from available literature also supports the fact that
Category A states are more affluent with higher per capita income as compared to the other categories of states (
B and
C).
Category C states constitute a higher share of tribal population as compared to the other two categories of states. Due to geographical isolation and inaccessibility, these states are more deprived in terms of availability of healthcare facilities [
25,
48,
49].
As in previous studies in the Indian context, our study also indicates progressivity in healthcare financing among Indian states. The incidence of OOPE was higher among
Category A states such as, Kerala, Punjab, Maharashtra, Goa, and Chandigarh. However,
Category A states face higher burdens of Non-communicable diseases (NCDs), causing them to spend more on healthcare, and resulting into a higher level of OOPE than their counterparts [
60‐
63].
There were noticeable differences in healthcare spending on the basis of economic status (richest /poorest) and social group affiliation (others /STs) among
Category A states. In
Category B states, inequalities are moderate among the above mentioned socio-economic groups. All these states (Karnataka, Andhra Pradesh, Tamil Nadu, Andaman and Nicobar, Lakshadweep, J&K, Delhi, Uttarakhand, Haryana, Himachal Pradesh, Rajasthan, Uttar Pradesh, Madhya Pradesh, Gujarat, Tripura and West Bengal) are middle-income states [
11,
63,
64].
Economic status and social group affiliation were important determinants of healthcare spending among C
ategory A and B states [
35,
55,
64]. Among
Category C states (Dadra Nagar Haveli, Daman & Diu, Sikkim, Nagaland, Assam, Manipur, Meghalaya, Mizoram, Arunachal Pradesh, Bihar, Jharkhand, Chhattisgarh and Orissa) the gap between the rich and the poor, and the disadvantaged social groups and others, is comparatively lower than it was in other categories of states [
25]. In the Indian context, income, class, caste and wealth quintile are considered the most powerful stratification variables in assessing socio-economic inequalities [
65].
Multivariate GLRM analysis shows that in
Category A and
B states, the role of religious affiliation and caste system is comparatively less important, while economic status is considered an important determinant of OOPE. Opposite trends have been observed in
Category C states where the role of caste system continues to be a predominating factor and significantly influences the spending pattern on health. This study brings into focus that even in the 21
st century, with all medical advancements and institutional reforms, social institutions still have a significant influence on the healthcare seeking behaviour and spending patterns of households [
66,
67]. In the Indian context, religion and caste significantly influence spending pattern among households. Results indicate significant variations in household expenditure on the basis of caste, especially among the STs [
33,
34].
Similar observations have been recorded for C
ategory B states where again economic status and social group affiliation are very important contributing factors of OOPE. Here also class concept (Economic status) is more dominant than caste but place of residence does not play an important role [
27,
68]. In
Category C states, religious affiliation and economic status both are considered equally important determinants of spending on health. Also, social group and place of residence are important determinants of OOPE among
Category C states.
In line with the Anderson model, our study also highlights the importance of predisposing, enabling and need factors. As per the categorization of states, the role of these factors in determining the level of OOPE varies significantly in the Indian context. Among
Category A states, enabling factors play a more dominant role than predisposing and need variables. A similar pattern was also observed among C
ategory B states. In
Category C states,
predisposition to use the services such as, caste and residence, still play a major role in the determination of the level of OOPE on healthcare than other factors such as, enabling and need factors [
69].
There is also evidence of the geographical concentration of the states on the basis of socio-economic inequalities and OOPE. Especially, C
ategory C states are geographically concentrated more either on the north-eastern or southern eastern side of the country. All these states, which fall in
Category C on the basis of OOPE, are backward states in terms of per capita income as well [
70‐
72]. While examining healthcare spending pattern and level of OOPE as per geographic concentration, it is not easy to state whether these variations are due to geographical factors or not. These variations may be due to differences in health-seeking behaviour among the population, and to some extent, the accessibility and availability of services. Instead, socioeconomic factors affect the need for health care and are more accountable for the variations in the level of OOPE [
73].
Conclusion
In this study, we have assessed the level of OOPE, and the socio-economic and regional variations, among all Indian states and UTs. In a developing country like India, where majority of the population spends on healthcare services from their own pockets, higher government spending on health is essential. This study brings into focus healthcare inequalities in India that are based on caste and social groups. The pattern of healthcare expenditure shows large variations in access to quality healthcare by the diverse socio-religious groups in the country. Special focus must be given to financing the health care needs of the disadvantaged sections of the population, as health expenses can push these households into greater risk of poverty through mobilizing funds to cater their healthcare needs. However, designing a financial protection mechanism requires a deeper understanding of both the absolute and relative amounts of the financial burden of OOPE on the households.
There is evidence of regional disparities in terms of the level of OOPE among Indian states. Developed states are spending more on healthcare and backward states are spending less. There is a need to look into why these backward states lag behind their counterparts. Our study brings into focus the issue of inter-state differentials in OOPE causing geographic variation and concentration in healthcare spending. It has been observed that the spending on healthcare was comparatively lower among all the backward or isolated states. Overall social security measures should be enhanced, but at the same time, more priority should be assigned to these disadvantaged states to reduce the burden of OOPE, looking at the regional differences. However, geographical differences cannot explain the OOPE differentials properly and more research is needed to understand why such variation occurs and what efforts are required to address these issues.
Any resulting policy changes should reflect the needs of the backward states and local communities. It was observed that there was an association between the prevalence of socio-economic inequalities and average monthly OOPE. Policy interventions are required from both centre and the states to increase budget allocation for health spending and for reducing the level of OOPE. We hope the findings of our study will be useful for policy makers, researchers and other stakeholders to formulate appropriate strategies for removing regional imbalances in terms of health spending among Indian states.
Acknowledgements
The authors are grateful to the Department of Humanities and Social Sciences, National Institute of Technology (NIT), Rourkela for their support and encouragement, which has helped in improving this paper. The authors also thank the journal editor and reviewers for their insightful comments which greatly helped improve this paper.