ABSTRACT
ABSTRACT
The percentage of India's national budget allocated to the health sector remains one of the lowest in the world, and healthcare expenditures are largely out-of-pocket (OOP). Currently, efforts are being made to expand health insurance coverage as one means of addressing health disparity and reducing catastrophic health costs. In this review, we document reasons for rising interest in health insurance and summarize the country's history of insurance projects to date. We note that most of these projects focus on in-patient hospital costs, not the larger burden of out-patient costs. We briefly highlight some of the more popular forms that government, private, and community-based insurance schemes have taken and the results of quantitative research conducted to assess their reach and cost-effectiveness. We argue that ethnographic case studies could add much to existing health service and policy research, and provide a better understanding of the life cycle and impact of insurance programs on both insurance holders and healthcare providers. Drawing on preliminary fieldwork in South India and recognizing the need for a broad-based implementation science perspective (studying up, down and sideways), we identify six key topics demanding more in-depth research, among others: (1) public awareness and understanding of insurance; (2) misunderstanding of insurance and how this influences health care utilization; (3) differences in behavior patterns in cash and cashless insurance systems; (4) impact of insurance on quality of care and doctor-patient relations; (5) (mis)trust in health insurance schemes; and (6) health insurance coverage of chronic illnesses, rehabilitation and OOP expenses.
Introduction
In recent years, health planners in India have advocated for the expansion of health insurance as an essential component of the country's healthcare reform and poverty reduction agenda (Planning Commission and United Nations 2011; Reddy 2012; Forgia and Nagpal 2012). One of the most ambitious plans in Indian healthcare reform has been a call for ‘universal healthcare for all by 2020’ (Reddy et al. 2011b), now extended to year 2022 (Devadasan et al. 2014). Reaching this goal would include implementing universal health insurance, which been seen as a potential way of reducing health disparities and OOP health expenditure (Bennett, Ozawa, and Rao 2010; Reddy et al. 2011b; Reddy 2012). Currently, numerous public, private, and community-based insurance schemes have come to coexist and even merge with each other, a situation that is hardly surprising in a country as diverse as India. The newest draft of National Health Policy (NHP) was revealed in December 2014, promising ‘universal ability of free, comprehensive primary healthcare services as an entitlement’ (Sharma 2015, 317). However, some have already pointed out that NHP overemphasizes the role of the private sector in healthcare delivery and financing, and view the proposed plan as a step away from universal health coverage (Sharma 2015).
What forces have brought health insurance to the attention of Indian policy-makers? In sum, they include (1) high burden of ill health; (2) low public spending on healthcare, (3) high private (especially OOP) healthcare expenditure; and (4) limited coverage by the existing health insurance schemes. India is currently going through a stage of health transition characterized by high burdens of ill health caused by nutritional deficiencies, infectious disease, and non-communicable disease (NCD), which are a long-term sequela of poverty, as well as an outcome of defective modernization associated with such trends as high tobacco use, overconsumption of high caloric fat-rich diets, and low levels of exercise among the more affluent. The latter factors contribute to rising rates of cardiovascular disease, diabetes, and cancer (Reddy et al. 2005).1 The lion's share of health problems, however, lies with the impoverished 75% of India's population who live on less than $2 a day (Subramanian et al. 2013). Within this population, rates of mortality and morbidity are uneven and subject to factors including gender, caste, socioeconomic status, and education, all of which have a significant impact on health risks and access to healthcare (Duggal 2007; Subramanian et al. 2008).
Despite a significant demand for community-based health services, national public spending on healthcare in India remains quite low.2 This has led to substantial private spending on health, especially in terms of OOP expenditure on medicines (Dror, van Putten-Rademaker, and Koren 2008; Shahrawat and Rao 2012; Karan, Selvaraj, and Mahal 2014). There are significant differences between India's states in the amount, distribution, use, and effect of public spending in healthcare, but even in states that spend more on healthcare OOP expenses remain high.3 On the providers’ side, low public healthcare spending has had a negative impact on the quality of care delivered at primary health centers and district level hospitals (Rao and Choudhury 2012). On the other hand, the private sector has been thriving, providing 80% of outpatient and 60% of inpatient care (Sharma 2015). Even the poorest and least educated people in both rural and urban settings consult private practitioners more than government practitioners and spend about twice as much on treatment from them than from government practitioners (Bhatia and Cleland 2001; Devadasan et al. 2006; Purohit and Siddiqui 1994; Madhukumar, Sudeepa, and Gaikwad 2012).
Health-related debt has pushed many low and middle-income households into poverty. It has been estimated that OOP expenses are directly responsible for the deepening of poverty in both rural and urban areas, pushing between 32 million and 39 million Indians into poverty every year (van Doorslaer et al. 2007; Shahrawat and Rao 2012; Balarajan, Selvaraj, and Subramanian 2011; Berman, Ahuja, and Bhandari 2010). Households with elderly and chronically ill members are especially susceptible to impoverishment due to health expenditure (Mohanty et al. 2014). In Kerala, for example, the loss of household income per illness episode has been estimated to be over four times greater for the poor elderly in comparison with the rich elderly households (Mukherjee and Levesque 2012). Hospitalization is often presumed to be the most important cause of health related impoverishment in India, but research has revealed that drug expenditure is actually the largest component of OOP payments, accounting for 61% to 88% of the total OOP spending (Garg and Karan 2009; Peters et al. 2002; Roy and Hill Howard 2007; Shahrawat and Rao 2012). Contributing to this expenditure is liberal prescription of drugs by practitioners (Nichter 1996; Porter and Grills 2015) and widespread over-the-counter medication use by both the poor and middle class (Basak and Sathyanarayana 2010; Ghosh 2011; Kamat and Nichter 1998).4 It has been argued that universal health coverage (UHC) implementers should focus explicitly on medicines as one of the most important drivers of quality, safety, equity and cost of care (Wagner, Quick and Ross-Degnan 2014; Bigdeli et al. 2015).
In the field of public health and health policy making it has been suggested that health insurance could provide an important safety net for low to middle income citizens by reducing emergency healthcare expenses for all social classes (Reshmi et al. 2007; Kasirajan 2012).5 Yet up until fairly recently only a small percentage of India's population has been insured. The Employees’ State Insurance Scheme (ESIS, introduced in 1948) and the Central Government Health Scheme (CGHS, introduced in 1954) cover only about 10% of India's population working as public employees (Shiva Kumar et al. 2011; Selvaraj and Karan 2012). During the last several years, private insurance companies have started offering health insurance next to life and property insurance, and a number of community-based schemes have been implemented, although the actual impact of many of them is not yet known.
At national and state levels, there is much debate about how best to proceed as a means of providing some level of coverage to as many people in India as possible. While some policy makers have called for insurance coverage in cases of catastrophic medical events, others have called for broader policies to reduce OOP expenditure as a means of preventing those living on the margin from falling into poverty (Garg and Karan 2009; Sodhi and Rabbani 2014). Still others have argued for the funding of more preventive and promotive health programs to decrease the country's NCD burden (Patel et al. 2011).
This paper examines challenges to implementing health insurance in India and makes a case for anthropological studies of the social life of insurance schemes (Dao and Nichter 2015) as a means of providing fresh insights to the emerging interdisciplinary field of health policy and service research (Gilson et al. 2011; Mills 2012; Hafner and Shiffman 2013). Medical anthropologists and anthropologists of global health and development have long been investigating the expansion of biomedical technologies, including issues of access and affordability (Lock and Nguyen 2010), and of health-related intervention programs (e.g. Pigg 2013). Health insurance as one such technology and also an intervention is yet to be addressed properly through ethnography everywhere. In the United States of America (USA), anthropologists have only recently been called upon to pay attention to this topic, as health reform has become a pressing political issue (Horton et al. 2014). A similar appeal has been made for anthropologists working in low- and middle-income countries such as Vietnam and Thailand where the idea of UHC has been pushed for by international bodies such as the World Health Organization (Dao and Nichter 2015). When health policies are encouraged by national or international, public or private institutions, a so-called ‘implementation gap’ may arise. This is what happened, for example, in the case of an initiative to organize community auto-financing of healthcare in Burkina Faso (Ridde 2008).6 But studying money and health managing practices anthropologically might shed light on how people think about new technologies, such as health insurance, as well as how they use, appropriate or reject them.7
In this paper, we take an initial step towards exploring health insurance in India, with the aim of encouraging further anthropological investigation of the issue. We first provide a short critical review of three types of insurance schemes already introduced in the country, namely public, private, and community based, as a way of laying out the health insurance landscape in India. In this part, we review the available literature, which mostly comes from the fields of public health and health policy. As Pigg (2013, 133) suggests, it is important that ethnographers of (global) health ask, ‘What is going on?’, before asking if or why something is good or bad. With that in mind, we wish to go beyond the negative or positive evaluations of various health insurance schemes in India, and argue for a critical investigation of the interaction between various insurance interventions and the people and institutions involved with insurance at various levels. What follows is a series of ethnographic observations on public and practitioner response to health insurance. These observations are gleaned from interviews carried out with multiple stakeholders in the South Indian state of Kerala in 2014. On the basis of these observations a research agenda for future in-depth ethnographic case studies of health insurance acceptability and implementation is laid out. Among other issues, the article argues for case studies that provide a more ‘population near’ and highly nuanced understanding of the public perceptions, (mis)trust and expectations of insurance schemes, and how they influence quality of care and practitioner-patient relations. The questions we raise about health insurance in India are provided as a springboard for further engaged deliberations in the areas of health financing and policy making, in India and elsewhere.8
Overview of current insurance schemes in India
Insurance schemes introduced by the central and state governments
In 2003, the Indian government launched Universal Health Insurance (UHI) to protect families living below the poverty line (BPL) by partially subsidizing insurance premiums. While the scheme has been criticized for high premiums, slow uptake (by 2008–2009 only 3.7 million people were covered), and for covering only hospitalization, it has provided valuable lessons for future government-sponsored health insurance schemes (Ahuja 2004; Ahuja and De 2004; Ahuja and Narang 2005; Bhat and Saha 2004; Gupta and Trivedi 2005; Forgia and Nagpal 2012; Rao 2004). Among the most important realizations is that in order to be successful, UHI needs to be supplemented by interventions at various health system levels; this, in turn, should be based on local research and policy analysis, rather than trying to ‘copy’ or ‘import’ insurance systems from other countries (Duran, Kutzin, and Menabde 2014).
State insurance programs, in practice based on public–private partnerships, have also been encouraged. In Andhra Pradesh, one example of private–public partnership, Rajive Aarogyarsi, was launched in 2007 to benefit BPL families who possess a ration card (Yellaiah 2013). The scheme is overseen by a trust that has selected a private insurance company through an open, competitive bidding process and established a network of private and public hospitals to provide free secondary and tertiary healthcare. By 2013, 87% of BPL families in the state have been covered, which is a notable success (Yellaiah 2013). The scheme does not reduce OOP for outpatient visits, but it does reduce unpredictable expense related to hospitalizations that are more likely to suddenly plunge a household into economic crises (Wagstaff and Bergkvist 2011). Ethnographic research has highlighted significant barriers in accessing health-care among the poor rural population in this state, reducing the impact an insurance scheme such as Aarogyasri could have (Narasimhan et al. 2014; Chakrabarti and Shankar 2015).
Several schemes launched by the government in recent years have fostered a rapid expansion in insurance coverage. Between 2007 and 2010, the proportion of the Indian population with some kind of health insurance roughly tripled to 25% or 302 million people (Burns 2014, 106). However, critics have argued that state-sponsored schemes, as currently designed, encourage the growth of tertiary care corporate hospitals in rural areas; they suggest that a more integrated model, including primary care, would be more beneficial to the users and more cost-effective for the government (Rao et al. 2011; Shukla, Shatrugna, and Srivatsan 2011; Selvaraj and Karan 2012; Sodhi and Rabbani 2014). It has recently been proposed that despite its diversity, the private health sector could be mobilized as a potential resource to contribute to public health goals if proper regulation and monitoring were in place (Garg and Nagpal 2014).9
Private health insurance
In 1999, the Insurance Regulatory and Development Authority Act (IRDA) liberalized and opened the market in India to foreign insurance companies. Private–public collaboration in health insurance, advocates claimed, would encourage competition between different care plans and permit patient choice of providers affording greater flexibility (WHO 2000). Due to the investment of a number of foreign firms, health insurance is now the second largest business in the non-life insurance sector (Sen, Pickett, and Burns 2014). But despite the efforts of 40 private health insurance providers operating in India in 2010, only 2% of the population had private insurance in that year (Thomas and Vel 2011). While some see India as a lucrative business opportunity for health insurance in the future (Yellaiah 2013), insurance schemes so far have proved to be a risky business venture due to a low level of insurance awareness as well as poor healthcare infrastructure in rural areas (Vellakkal 2009a; Vellakkal and Ebrahim 2013; Ahmed 2013; Gupta 2006; Thomas and Vel 2011).
Private health insurers recognize India as a potential market due to its increasing purchasing power, growing demand for healthcare, an expanding competitive private healthcare market, and rising rates of chronic disease (Burns 2014; Burns, Srinivasan, and Vaidya 2014). Because of this profit-oriented approach, private health insurers target mainly the middle-class population. This is also evident from the cost of insurance policies, with annual premiums typically starting at Rs 4000 (about US$65) per member and covering only inpatient treatment for a maximum of Rs 0.4 million (about US$6500) annually (Thomas and Vel 2011).
Besides noting the limits in terms of coverage (inpatient only) and target population (middle-class), critics argue that the market principles guiding private healthcare insurance lead to greater health disparity and rising health costs for the poor, serving to undermine national health equity goals (Selvaraj and Karan 2012; Reddy et al. 2011a).10 They call for regulation of benefit packages, restrictions on risk-selection procedures, and greater protection of customers (Mahal 2002).
Community-based health insurance schemes (CHI)
CHI is ‘any non-for profit insurance scheme aimed primarily at the informal sector and formed on the basis of a collective pooling of health risks, and in which the members participate in its management’ (Devadasan et al. 2006, 225). CHIs generally function in association with the existing community organizations such as trade unions and cooperative societies. Most schemes follow a ‘linked model’ whereby the NGOs act as intermediaries between the community and formal insurance companies and managerial functions are taken by professionals instead of volunteers. Studies of CHI introduced in some other countries in Asia and Africa suggest that they help increase access to healthcare for low-income populations, and reduce OOP expenditure for health (Jakab and Krishnan 2001). Since they are based on the principle of social solidarity, CHIs also tend to be better accepted by local populations (Ahuja and Narang 2005; Bhat and Saha 2004; Devadasan et al. 2006).
In India, over 115 CHI schemes have been initiated by local NGOs in support of broader development programs (Aggarwal 2010; Michielsen et al. 2011; Devadasan et al. 2006). One of the most prominent and long-standing CHIs was launched in India in 1992 by the Self-Employed Women's Association (SEWA; Ranson et al. 2007). Between 2003 and 2005, the mean socioeconomic status of those insured by SEWA increased significantly relative to the non-insured population in the same sub-districts (Ranson et al. 2007). However, there has been a significant difference in enrolment among rural and urban areas (Desai 2009; Ranson et al. 2007; Sinha et al. 2006). In the South Indian state of Karnataka, the Yeshasvini insurance scheme, based on a partnership between public, private, and cooperative sectors, was launched in 2010 and covers over three million people, but has only been able to persist thanks to government support and donations from private and public bodies (Aggarawal 2010).
Lessons learnt from Rashtriya Swasthya Bima Yojna (RSBY)
The insurance scheme that has received the most attention in India is RSBY. The Indian government launched this ‘cashless’ scheme in 2008 to cover BPL families’ hospitalization expenses tracked through the use of a smart card against only a small enrolment fee (Shiva Kumar et al. 2011; Rao and Choudhury 2012). RSBY has been described as offering the most inclusive and comprehensive social health protection in India so far (Michielsen et al. 2011). Insurance is provided to families of up to five members, covering surgical procedures and hospital admissions for up to Rs 30,000 (about US$500) yearly (Dasgupta et al. 2013). The central government covers 75% of the yearly insurance premium and the rest is covered by the state governments. The initial goal was to cover the entire country by 2012–2013 (Reddy et al. 2011b), and currently all 29 states of India have implemented RSBY in at least one of their districts (with great variation in the number of covered districts among states), covering over 37 million BPL families.11
India's previous health minister, Harsh Vardhan, cited the program as an example to follow and extend in scope (Bagcchi 2014). A study in Udupi district of Karnataka state has, for example, shown that the majority of RSBY subscribers have found the scheme beneficial and were willing to renew their enrolment (Kamath et al. 2014). Studies also show that in the large majority of the states, monthly healthcare expenditure for inpatient treatment has either reduced or remained constant since the introduction of the scheme (Sinha and Chatterjee 2014).
Other evaluations of RSBY, however, raise issues begging further inquiry.12 One criticism is that the scheme focuses too exclusively on BPL populations and does not take into account the families that are above the poverty line (APL) but may nevertheless struggle or become impoverished due to healthcare expenses (Selvaraj and Karan 2012). This narrow approach has been linked with the early genesis of RSBY guided by particular developmental goals and institutions, whereby efficiency, competition and individual choice were considered most important (Virk and Atun 2015). As a response to such criticisms, the coverage has recently been extended to APL families in Kerala (Sinha and Chatterjee 2014). Second, CHI schemes such as RSBY have created an increasingly complex administrative system, which results in higher management costs requiring increases in premiums to stay solvent (Selvaraj and Karan 2012). Third, the claims ratio among enrolled families, reaching only up to 15%, indicates poor utilization of the scheme benefits (Sodhi and Rabbani 2014).
Furthermore, while there has been a slight decline in OOP expenditures, there has been a sharp increase in hospitalization expenditure in post-insurance years (Selvaraj and Karan 2012). As Selvaraj and Karan (2012) argue, the reason behind this is the narrow focus of RSBY on secondary and tertiary healthcare, encouraging inpatient care where outpatient care could suffice. An analysis of insurance claims in public hospitals in Chhattisgarh, for example, revealed that common conditions, which would normally be treated in outpatient care, such as diarrhea and respiratory infections, were often treated through expensive hospitalizations (Dasgupta et al. 2013).
Finally, evaluations by Dasgupta and colleagues (2013) and Salvaraj and Karan (2012) suggest that rather than strengthening public institutions, schemes such as RSBY actually contribute to a weak public health system and an unregulated growing private health system. Instead of protecting poor and middle-income populations against catastrophic healthcare expenses, these insurance schemes are pushing them further into poverty. Data on RSBY from across the country indicate that despite their health insurance coverage, the financial burden of OOP increased faster among the poorest 20% of the population compared with the richest 20% (Karan, Selvaraj and Mahal 2014). Because most of the expenses are due to non-institutional treatment, which is not covered by RSBY and other schemes, poor families tend to avoid treatment altogether or opt for low quality treatment. While RSBY has proved relatively successful in terms of enrolment, utilization, and impact in some states, it has not made a significant impact on a national level (Sinha and Chatterjee 2014).
Ethnographic observations
Methods
Our explorations of health insurance in India started with one of the authors, Mark Nichter, carrying out initial ethnographic observations on the topic intermittingly between 2001 and 2012, in the course of broad-based research on healthcare utilization and provision on the border between Karnataka and Kerala. All three authors followed up these observations for six months in 2014 in several districts across Kerala, the state with the highest density of public and private medical facilities and high rates of utilization of health services (Levesque et al. 2006). Most of our research took place in urban and semi-urban areas, which are most prevalent in Kerala, a densely populated state characterized by small and medium towns (Levesque et al. 2006).
We conducted open-ended interviews with a total sample of 63 community members hailing from a mix of social classes, nine public and 11 private practitioners as well as three healthcare administrators and 10 representatives of private and public insurance providers. We chose a purposeful sample of community members who had recently been placed in economic distress due to medical expenses related to an illness they or a family member had suffered. They were interviewed in the privacy of their own homes. The community members we interviewed about general impressions were long-time key informants of two authors, Mark Nichter and Tanja Ahlin. Additionally, Mark Nichter and Gopukrishnan Pillai, who is a medical doctor originating from Kerala, interviewed healthcare practitioners and administrators at well-known and well-attended health clinics. These study participants were experienced with having to navigate the insurance system, were known to the authors from their previous research, and were willing to discuss the topic confidentially. We interviewed the research participants one or more times. We informed them of the exploratory nature of this research and that their participation was voluntary and confidential. Following this, we obtained oral informed consent for participation from the interviewees.
According to Dao and Nichter (2015), ethnographic investigations of the ‘social life of insurance’ are a necessary complement to the quantitative studies conducted to date. Toward that end, fieldwork was conducted in Kerala to shed light on both public impressions of health insurance and the manner in which these perceptions influence the engagement of insurance, and how insurance influences healthcare providers’ practices. As Kerala has been hailed as one of the most progressive Indian states, with high rates of urbanization and education as well as better living conditions and healthcare access compared with the poorer states (Levesque et al. 2006), this study is limited in terms of generalizing the findings across India. We do not claim to offer a comprehensive study of all concerns related to health insurance in India. Instead, we shed light on some issues that we believe can and should also be addressed in other Indian states, as the literature review suggests that the problems related to health insurance are similar across states with different demographic characteristics. However, we encourage further research in the various states to pay attention to specific local contexts. Based on short extracts from our interviews and observations, we summarize some of our more salient observations and highlight additional themes for further investigation. Rather than a full ethnographic account, the material presented in the continuation should be viewed as indicating several points of entry for more in-depth investigations by means of ethnography, attentive to all that is less visible, ignored or uncertain (Pigg 2013).
Pertinent examples of issues with health insurance
Public awareness and understanding of insurance
Several reviewed studies have documented that public awareness of health insurance in India is poor (Ahmed 2013; Platteau and Ontiveros 2013; Reshmi et al. 2007, 2012; Vellakkal and Ebrahim 2013). Due to the efforts of private insurance agents, general public awareness of health insurance in Kerala as well as elsewhere in India has been raising (Vellakkal 2009a). Discussions regarding the problems of implementing new technologies, including those related to managing money and health, in any country or community often see ‘the culture’, or lack of necessary education, as a ‘barrier’, due to which people have difficulties learning how to use these new technologies (cf. Pigg 2013, 129). However, our fieldwork so far indicates that the understanding of what benefits one is entitled to, when and where, often remains confusing not only to the poor and illiterate, but also to middle class educated citizens. Such miscomprehension may be fostered by the promotional language of the insurance sellers that is hard to grasp for the general public of different educational backgrounds. For example, an employee of a government-sponsored insurance scheme juxtaposed life and health insurance as one likely reason for misunderstandings:
Health insurance is … far more complicated than the traditional life insurance that the general population is more familiar with. This is a major reason for lack of interest initially and disagreement later on.… Sometimes even those who wanted to read the detailed terms and conditions could not do so as the policy certificate was made available only some days after the agreement is signed. There is a cooling off period during which time an insurance contract can be terminated without prejudice, but very few people have availed [themselves of] this and if an agent tells [about it] he may lose a commission after a hard sell. Health insurance is problematic. People buy and then do not renew.
Indeed, some of our other informants expressed a wish for more transparent explanations of differences between life and health insurance.
Three questions emerge in relation to raising awareness about health insurance: (1) who should raise awareness about insurance; (2) what are the issues that need to be addressed; and (3) how should that be approached? A college-educated informant with insurance had this to say on the topic:
I don't think the common man knows what exactly [health insurance] is all about.… I have an insurance coverage with a particular scheme. I don't know what the terms are, I have not found out.… I am able to read and write [but] what about the people who may be illiterate … how much time they will spend [on finding out the terms and conditions]? They wouldn't know much. So the average person does not know. I think that it's the duty of socially committed organizations to create that awareness, not insurance agents.
According to the informants, information on health insurance should be provided by trusted parties such as respected healthcare workers, local NGOs, and members of the local government, called panchayat (see also Ahmed, 2013; Madhukumar, Sudeepa and Gaikwad, 2012). Regarding the content of the explanations needed, one of the main issues to address would be the basic principle of risk pooling. A doctor who runs a small hospital spoke about his efforts to describe insurance in relation to credit and debit cards that are now becoming more common in urban areas. He was led to do so because people, in his words, confuse insurance with having a credit card:
Patients and their family members come into my hospital with an insurance card and treat it like a credit card and expect all expenses will be paid, or that they can charge everything and pay later but with no interest charge.… I have to tell them no, the insurance card is like a debit card that can only be used for certain things. Once you run out of your funds there is no credit. I have this posted on signs in the accounts department and near reception so there is no misunderstanding. This misunderstanding is common.
Published online:
01 February 2016Figure 1. Posted in a nursing home on Kerala-Karnataka border, this sign warns against the common misunderstanding that health insurance can be used like a credit card. (Photograph by Mark Nichter.)
![]({"type":"image","src":"/na101/home/literatum/publisher/tandf/journals/content/canm20/2016/canm20.v023.i01/13648470.2015.1135787/20170213/images/medium/canm_a_1135787_f0001_oc.jpg"})
Figure 1. Posted in a nursing home on Kerala-Karnataka border, this sign warns against the common misunderstanding that health insurance can be used like a credit card. (Photograph by Mark Nichter.)
As to how insurance could be better explained to the masses, it would make sense to draw comparisons as well as make distinctions with familiar monetary practices. For example, some informants suggested insurance could be explained through local institutions like the rotating credit systems or ‘chit funds’.13
Further, here is an observation of a senior official at the government-owned Life Insurance Corporation of India (LIC) that brings out the issue of unrealistic and misplaced expectations of the public:
When I took charge of operations in 2010, health insurance was a relatively new idea. We had to explain to people regularly. But now there is wider understanding of the concept of health insurance, though not the details. People are on the whole unwilling to take health insurance, even though they acknowledge the risks and raising costs of healthcare. Why? Because there have been many instances reported of consumers feeling denied of insurance benefits they feel are due to them. As a result of this, LIC health insurance has gone out of favor with people and business is below expectation. A major reason for this is a mismatch between customers’ expectations and poor understanding of actual terms of the contract. This is mainly due to the failure of LIC to adequately train and monitor insurance agents who are intermediaries employed on commission basis to convince people to buy insurance. The agents do not convey exceptions to insurance coverage adequately to customers. The people on their part have not been diligent in understanding the terms of the contract, partly due to the existing trust they had for LIC as a public sector insurer.
Poor understanding of insurance mechanisms may continue even after people acquire health insurance policies. This may result in the insurance beneficiaries having unrealistic or misplaced expectations about what kind of treatment is included in the coverage (e.g., outpatient or inpatient healthcare) and what are the limits of insurance in terms of the amount of covered expenses.
Impact of misunderstanding on health seeking behavior and treatment
Misunderstandings about health insurance schemes, for example thinking about them in terms of credit cards, can lead people into far more debt than they may have incurred if they did not have insurance. This occurs when they seek out more expensive care than they can afford without realizing that their insurance coverage at a hospital is limited. A healthcare practitioner at another NGO clinic explained a common way this happens:
If a man on the street buys into health insurance for Rs 35 and gets coverage for Rs 30,000 for a full year, what does he often do? When he has an illness, because the money is available and because the treatment can be anywhere, he is tempted to go to the best hospital. And there he finds out that after the first three days of treatment, the money is over. So now he borrows money – more than once! And even after borrowing from the relatives he finds that there is no end to it. Then he is discharged and goes to a government hospital. Now, he is no longer able to afford food … The limited financial input from the insurance actually encourages [people] to seek treatment from the expensive [private] hospitals. Then they find out that they are unable to pay the whole bill in order to complete the treatment. So they borrow and fall in the debt trap. Very often they get halfway through the treatment and then stop because they cannot afford to continue to pay for the rest of the treatment out of pocket.
We found that some clinics limited the amount of insurance coverage a person could draw upon during any one hospitalization as a matter of policy, possibly to educate patients that coverage was not in fact free but against a balance. For example, a policy put in place at one clinic allowed patients to pay a part of their bill by insurance (up to a limit) and then the rest out of pocket. The following case illustrates this ad hoc procedure:
My husband has been ill for [the] past two years, due to which he has had to be hospitalized repeatedly. We are beneficiaries of RSBY. However, the entire money is not allowed to be withdrawn at one go. If all of the money on the card for the entire family is allowed to be withdrawn at once, then it could create problems later on. So only a maximum of Rs 5000 out of a total Rs 30,000 can be issued during one hospitalization event. At the private medical college hospital where [my husband] is admitted in the general ward, any expense beyond this must be borne out of pocket. Our expenses have been far beyond that covered by RSBY… We have for the most part been helped by friends and family over the past two years. But we are facing financial ruin due to my husband's prolonged illness. The banks won't lend to us anymore so we have to borrow money at usurious rates.
The decision of capping was taken by the clinic administrators in order to prevent the money available under the RSBY insurance scheme to the beneficiaries too quickly. Such a limitation, however, is not a part of the original RSBY scheme employed across India, in which the critical limitation is rather that the beneficiary card can only be used once in 24 hours so as to prevent corrupt practices (Dasgupta et al. 2013).14
Differences in behavior patterns in cash and cashless insurance systems
The introduction of a cashless system, such as the smart card used in RSBY, may have an impact on patients’ perceptions that through direct payment of some kind they are establishing a bond with practitioners. In many parts of India it has long been customary to gift or indirectly pay government and indigenous practitioners for ‘free’ services as an incentive to poorly paid practitioners or to forge personal relationships (Nichter 1996). This is seen less as corruption and more as socially acceptable relationship building, a form of interaction through which a trusting relationship between the doctor and the patient becomes established and is therefore perceived as ethical by the parties involved.
In their book, Poor Economics, Banerjee and Duflo (2011) point out that people often choose things that cost them more due to perceptions of quality and what the product or service means socially and to a transaction. Will removal of direct payment for services and subsidized costs lead to patient concern that lack of an immediate incentive will jeopardize their ability to receive best care? Will they continue to choose private healthcare over public precisely because it is more expensive and perceived to be of higher value (Bhatia and Cleland 2001)? Our preliminary research suggests that unless perceptions of quality of care and medications at public institutions change, this is likely to occur.
Impact of insurance on quality of care
Researchers will need to investigate how cashless health insurance impacts on quality of care and practitioner decision-making. The following observations of one doctor about RSBY captures what several of his colleagues in the private sector saw as a likely future:
It is not true to say that the Mediclaim model has no caps on health expenditure. They might not reveal to patients always, but they do dictate to hospitals how much they can spend per procedure or disease. So, in case if the money allotted only covers four days worth of hospitalization, then patient is likely to be discharged at end of four days even if the doctor would normally have advised a couple days more of treatment or further investigations. Thus quality of care is likely to deteriorate if health insurance becomes the norm.
Our interviews with doctors indicated that nearly all thought diagnostic tests and Caesarean sections would increase substantially if covered by insurance, both because of practitioners’ wish to increase their earnings and because of higher demand from patients (see also Ajeet and Nankishore 2013; Bogg et al. 2010). It is not clear, however, whether the caps on expenditure for Caesarean sections would prevent excessive diagnostics and bring down costs, or increase the amount of serious health complications. Community members, even those with low levels of education, suspected doctors profited from ordering tests and were aware that they may not be needed, but viewed them as indicators of both quality of care and doctors’ interest in them (Nichter and Van Sickle 2002). It is important to note that perception of quality of care does not equal quality of care, and a higher number of tests and surgeries does not automatically translate into high quality of care; it may even have the opposite effect as unnecessary surgeries and tests may lead to unwanted side-effects. The subjective meaning of quality of care is a worthy research topic in its own right and directly tied to perceptions of the merits of health insurance coverage.
(Mis)trust in health insurance schemes
Misunderstanding of insurance as well as provision of information about it from insurance sellers, who work for commission, often leads to mistrust among the people. For example, there seems to be little understanding of risk pooling as a population-based principle underlying insurance. While insurance experts’ calculate risk in terms of numerical probability, lay-consumers find the logic of insurance based on contracts and continuous payments suspicious (Golomski 2013). We found a sense of ambivalence about insurance coverage and little concern for being overcharged. Many informants felt that it was likely they were being cheated, but that since it was not their money, they were not much concerned. One of our informants explained:
The insurance scheme is no doubt a moneymaker for the hospitals. How are we common folk able to know what they have put down on the file by means of medicines and tests! Only when the money is cut from our card do we know the billed amount. Who has the inclination to ask about such things during treatment? People are just grateful to be receiving any support at all from government. They are ecstatic that not only don't they pay any money to hospital, but instead even receive a sum of Rs 100 at time of discharge for bus fare back home. So they don't mind whatever the hospital writes on file. 'It's government money, why do we care what they do with it?' is the common attitude. But then the amount is depleted and not available next time it is needed.
Many of those interviewed stated they have little faith in health insurance schemes and prefer to pay OOP for services as needed. As justification they cited stories about negative experiences of family or friends with other types of insurance. Another issue worth examination is the extent to which trust in government-supported insurance schemes is related to the much larger issue of trust in the state, and the extent to which the state is seen as fulfilling its mandate to protect its citizens. Because of widespread suspicions of corruption in public institutions (see also Berger 2014), officials dealing with claims might be reluctant to approve claims, even when they are clearly legitimate. As a judge in charge of a district consumer court explained:
The organizational set-up of insurance bureaucracy could contribute to disputes. For example, in [the] public sector, officials in charge of granting relief have to reckon with [the] possibility of enquiry from anti-corruption police. They might feel compelled to offer only the minimum amount or contest claims … even if they know they are legitimate. [They] prefer that the matter be settled in court so as not to be accused of taking a bribe in future. Similarly, private companies have engaged parallel investigation mechanisms to verify claims, and these reports invariably tend to be conservative.
In the long run, this adds up to the perception that insurance companies are ‘cheats’ that do not want to pay and have to be forced to pay even for legitimate cases. Needless to say, this undermines trust. Further ethnographic research is also needed on insurance perception spillover from one type of insurance to the next as well as how trust has been established in successful health insurance schemes, but lost in others.
Different sectors of the population may view insurance in different ways, leading them to be more or less inclined to participate in schemes or pay premiums (Reshmi et al. 2012). While some segments of the population may have little trust in insurance as a resource, other segments may see insurance as an entitlement and associate the right to insurance with citizenship. The issues of trust seem to play a pivotal role in insurance enrolment, expectations, and use, and this begs further ethnographic exploration. The study of trust could be started by first examining other kinds of insurance (e.g. life or property insurance), and especially those contexts in which particular insurance schemes have been more successful than others. Anthropologists could also look at various other local practices around India that are based on trust to see what are the specific mechanisms that foster or endanger trust.
Health insurance coverage: chronic illnesses, rehabilitation and OOP expenses
Insurance coverage in India, as noted in the introduction of this paper, primarily addresses the catastrophic costs of hospitalization, while the major OOP spending is for outpatient care and medicines. Families often slip into debt regardless of insurance coverage for medical procedures as a result of the costs of essential follow up care. This is an issue Selvaraj and Karan (2012, 67) have highlighted in their critique of the narrow focus of ‘compartmentalized care’ to the neglect of continuity of care. Lack of support enabling treatment adherence and continuity of care can in turn lead to a much higher probability of future hospitalizations. Consider the case of the following child being cared for by his grandmother:
My grandson was born with both knees twisted inwards. I secured an insurance card and took him to a private medical college hospital as advised by a neighbor. There, they surgically extended his knees. The bill for this procedure was paid from the insurance card. However, all future expenses for care had to be met out of pocket. We have been advised physiotherapy for at least 3 months. [The] daily fee is Rs 150 plus travel fare. This is not paid for by the card. We depend on remittances from his mother for sustenance. How can I meet this expense?
Most public insurance schemes also do not cover the costs of medicine and outpatient treatment for those with chronic illnesses, even though studies suggest that many people would prefer to acquire health insurance packages covering both inpatient and outpatient treatment (Dror et al. 2014). The following case of an elderly man illustrates the dilemma which many families with chronically ill members face: they have to choose between medicine and basic necessities such as food. The case also speaks to how doctors try to work the system for the benefit of their patients:
I have had diabetes and hypertension for the last 10 years. My doctor is good. But I cannot purchase medicines he prescribes routinely or [visit] the clinic as directed because the expense is too great for my wife and I to bear. We live on very limited means and buying medicine means less money for food. The first two years I tried to do the needful and purchase the medicines, but I fell into debt. So now I purchase the medicine and visit the clinic when symptoms are more. If I develop wounds on my legs or feet that look serious, the doctor sometimes admits me to the hospital for a few days. My card will not pay for medicines, but it will pay for my stay in the hospital. The doctor gives me samples of medicines to take home.
In this example, the doctor admitted the patient in order to treat and stabilize his condition. The doctor was aware of the patient's dire economic situation and inability to purchase necessary medicines. This case of a doctor working the system on the patient's behalf provides a good counterbalance to the many cases reported of clinics gaming the system by ordering unnecessary diagnostic tests and operations such as Caesarean sections as a means of earning profit (Desai 2009; Ajeet and Nandkishore 2013).
Chronic disease and mental illness are expected to increase substantially in India in the next decade (Chatterjee 2012; Patel et al. 2011; Reddy et al. 2005; Vellakkal 2009a, 2009b). A key issue to be addressed by insurance providers is the extent to which routine monitoring of chronic illness proves to be a cost effective means of reducing expensive downstream costs of acute disease episodes. A community-based study of chronic illness in Kerala suggests that a major cause of non-adherence for both diabetes and hypertension is the financial burden of follow-up care and lack of insurance coverage (Devi and Nichter 1998; Devi 2008). Working along with colleagues in the health service research field, anthropologists might investigate patterns of (non)use of clinic services if and when they are made available to those with chronic diseases.
Other areas recommended for further research
There are a number of other aspects of health insurance that call for further investigation by anthropologists who could contribute to a more in-depth understanding of the life cycle and impact of the existing schemes. Important questions to look into are who buys insurance, for whom, and why. Some research has indicated that nuclear families and families with fewer members are more likely to buy insurance policies (Madhukumar, Sudeepa, and Gaikwad 2012) and socio-economic factors including literacy, religion, occupation, and gender are important (Reshmi et al. 2007, 2012; Rai and Ravi 2011; Chakrabarti and Shankar 2015). Research is needed on both motivations for and ambivalence about health insurance experienced in different types of households. Insurance schemes often target families, yet little is known about who encourages enrolment in insurance schemes and for what reasons.
Further, what is the impact of health insurance on self-care and the household production of health (Gilson et al. 2011; Olson 1994; Schumann and Mosley 1994)? Will insurance coverage influence self-care practices, such as purchasing medicines over the counter, and reduce delay in seeking healthcare, and if so, for what type of complaints? Also, will impoverished women feel more empowered to seek healthcare for themselves if they do not feel they are a drain on a household's limited resources?
Other promising lines of research include the ways in which health insurance programs impact patient referral patterns and introduce audit systems that change the way healthcare is managed. Finally, future research will need to examine insurance coverage for Indian indigenous medical systems such as Ayurveda, Yoga, Unani, and Siddha (AYUSH), and when these systems of medicine are turned to for disease management as well as promotive health and care for the elderly.
Conclusion
Assessment of the real world impact of health insurance requires a holistic, multi-disciplinary analysis that captures its complexity and is sensitive to the larger social, political, and economic context in which policy is created and introduced. A health policy and systems research approach promotes such ‘systems thinking’ and explores why and how certain programs work for some, but not for others (Gilson et al. 2011; Mishra 2013). Such research must be people-centered and attentive to human agency, social relations, cultural values, and trust (Sheikh, Ranson, and Gilson 2014). At the same time, it should also be policy-minded and attentive to processes, structures, and power relations that constitute the field in which a policy is both constructed and negotiated.
In the case of health insurance, research needs to take stock of those factors that lead citizens to enroll in and drop out of insurance programs as well as to use insurance in particular ways. It should also investigate the impact of insurance on the way healthcare is administered at multiple levels. Medical anthropology could play a vital role in this research agenda, as is evident from a few of the many lines of potential research proposed in this paper. Among other issues, we highlighted the following six key areas for in-depth research: (1) public awareness and understanding of insurance; (2) impact of misunderstanding on health seeking behavior and treatment; (3) differences in behavior patterns in cash and cashless insurance systems; (4) impact of insurance on quality of care; (5) (mis)trust in health insurance schemes; and (6) health insurance coverage, especially in relation to chronic illnesses, rehabilitation and OOP expenses. These topics emerged from our interviews time and again, and, combined with the literature review, they indicate where most misunderstandings and tensions occur. A better grasp of the enumerated areas could potentially lead to improvements in providing, explaining and implementing health coverage, especially for those with limited resources. We conclude with a call for organizational ethnographies of insurance that include what Laura Nader (2008) has referred to as ‘up, down, and sideways’ perspectives. This refers to the study of those in power (policy makers and influential stakeholders), those subjected to the directives of those in power (parties charged with implementing programs downstream as well as program recipients), and those who are motivated to frame, fund, and publicize research on insurance for a myriad of purposes.
Acknowledgements
The authors would like to thank Arima Mishra, Reshmi Nair and the anonymous reviewers for their comments on earlier versions of this paper.
Disclosure statement
No potential conflict of interest was reported by the authors.
Ethics
The authors followed the American Anthropological Association ethics guidelines. Oral consent for interviews was obtained from each individual study participant.
Conflict of interests
None.
Notes
1. On the positive side, India has also achieved success in certain healthcare areas; for example, substantial improvements have been made in antenatal and infant care (Balarajan, Selvaraj, and Subramanian 2011).
2. The estimated government public health expenditure was stagnant at around 1% of GDP between 1990 and 2006, increasing only to 1.2% of GDP in 2009 (Tandon and Cashin 2010; Rao and Choudhury 2012; Gupta 2006). In December 2014, the government announced a cut of nearly 20% in its healthcare budget for 2015/16. It is as yet unclear how this will affect the planned launching of a universal healthcare program, providing all citizens with free drugs and diagnostic treatments as well as insurance benefits (Kalra 2014).
3. In Kerala, the amount of public financing of healthcare is three times higher than in Bihar (Shiva Kumar et al. 2011), yet OOP expenditure reaches almost 7% of total household expenditure, which is one of the highest levels in the country (Garg and Karan 2009).
4. In 1999–2000, OOP payments constituted 4.8% of total household consumption expenditures on a national level, with significant differences between states (from 2% in Assam to as much as 7% in Kerala), and between rural and urban areas (Garg and Karan 2009).
5. Studies in other countries indicate the benefits of health insurance indeed include reduced financial burdens, improved access to care and better health outcomes (Kenney et al. 2014).
6. To add a historical perspective, see Börner (2011) for an analysis of the implementation of national health insurance policies and the tensions this raised with the local communities in charge of social benefits in Germany and the United Kingdom.
7. We thank one of the anonymous reviewers for bringing our attention to this point.
8. As Stacy L. Pigg (2013, 132) notes, there have been decades of discussion regarding the accountability and purpose of anthropology, which were made explicit through the dichotomy of applied and theoretical work. Here, we employ and encourage ‘engaged anthropology’ (Nichter 2006) as a possible way to overcome this dilemma.
9. Others have argued that over-reliance on the private sector, as promoted by the latest NHP, would encourage hospital care before integrated care, which would combine primary, secondary and tertiary facilities for the optimum patient wellbeing (Sharma 2015).
10. A similar argument, namely that health insurance is only one, and perhaps not even the most important, aspect of addressing the problem of national health disparities, has been made in the context of the Affordable Care Act in the USA (Stewart, Hardcastle and Zelinsky 2014).
11. The data are from the official website of the RSBY scheme; see http://www.rsby.gov.in/Overview.aspx.
12. Impact assessment of health insurance schemes has been shown to be sensitive to methodology and the data used for the analysis; this has been pointed out as a reason for contrasting evaluations of RSBY (Vellakkal and Ebrahim 2013).
13. Chit funds are a mix of auctioning and credit with varying interest rates depending on auction each month.
14. A study on benefit-package design of community-based health microinsurance showed that people preferred caps, especially if they would apply to households rather than their individual members (Dror et al. 2014). According to Dror and colleagues (2014), this so-called ‘family floater’ would mean that within a cap per family and per event, one household member could claim more than one event per year. This practice would encourage circulation of funds within households; it would also add to the welfare of the most vulnerable family members as well as to the non-claiming ones as they would be released from the obligation to provide financially for treatments of their close relatives.