Being Insider-Outsider: Public Policy, Social Identity, and Delivery of Healthcare Services in India

The pivotal role of community level workers in the delivery of public services is well-recognized. But, they often fail to provide equal opportunities to all ‘eligible’ beneficiaries to utilize a variety of public services. Although several predisposing household factors are held responsible for inequalities in access to the public services, in recent times, one factor that has been recognised as critical to such unequal access is the ‘exclusionary nature of social relations’ based on social identity embedded in the social life of village community. It is also argued that certain sections of the population are deprived of equal access to public services due to their social identity, which is different from service providers. However, the question remains–whether it is the social identity of users or providers of public services that is critical to unequal access to various services? What will be the extent of utilization of public services when the social identity of both users and providers of the services remain same? Do the social dynamics of the community life play any role in the delivery of public services? This essay addresses these questions in the context of delivery of integrated nutrition and healthcare services at the community level under the largest national flagship scheme of Integrated Child Development Services (ICDS). Drawing evidence from a larger sample survey of over 4000 household beneficiaries and 200 service providers, the essay sheds light on how the delivery of healthcare services is fraught with social injustice due to dominant socio-cultural norms around social identity despite the values of healthcare centres to cater to the health needs of all sections of society.


Introduction
Public policy prescribes specific norms and guidelines for the delivery of public services in order to fulfill a set of objectives. Functionaries in public institutions are expected to deliver public services in a specific manner, and address the needs of all sections of society. While majority of public institutions function within larger social contexts, there are institutions which operate within local social structures. Given that dominant socio-cultural norms, beliefs and practices at a societal level often become salient to influence the thoughts and behaviours of individuals (Hogg, 2015), this may have implications for the service providers working in close association with the community in particular.
The health sector of India has witnessed significant improvement in public health infrastructure in the forms of availability of public health centres and sub-centres, and community health centres. However, inequities in access to healthcare services continue to persist. Despite the expansion of public sector healthcare facilities, socioeconomically disadvantaged groups continue to derive lower benefits from various health services. Macro level data based on national level survey validates inter-group inequalities in various health outcomes (IIPS, 2017). It is often argued that this has much more to do with poor living conditions and lack of access to healthcare facilities (UNICEF, 2013). They are mainly affected due to inequities in the availability, utilization and affordability of health services (Baru et al., 2010).
In recognition of the concern of persistence of group inequalities in access to nutrition and healthcare facilities, while the National Health Policy (NHP) -2017 has a specific focus on most socio-economically vulnerable groups (NHFW, 2017), the National Nutrition Strategy (NNS) prioritizes interventions to address nutrition and healthcare needs of various target groups (MWCD, 2017). Evidently, public policy initiatives on nutrition and health include a 'multi-sectoral strategy' to strengthen the delivery of the integrated nutrition and health services. In this regard, the role of the national flagship scheme of Integrated Child Development Services (ICDS) is assumed most significant. Given that basic health care services are largely provided at public health centres and community health centres with the aim of reaching out to all sections of people, lower access of marginalised communities to healthcare facilities raises questions not only on the functioning of the service delivery system but also the health delivery behaviours of service providers.
In the context of village community, one factor that has been recognised as critical to the inequality in access to various public resources and services is the phenomenon of 'exclusionary nature of social relations' based on social identity of people. A body of literature around the social identity framework (Tajfel, 1982) establishes that in a stratified society, various social identities not only govern the intergroup social relations but also shape the behaviours of individuals towards each other primarily based on social categorization into 'in-group' and 'out-group' (Tajfel and Turner, 1986). Socio-psychologically, the need to maintain a distinct identity through downward comparison (Wills, 1981) serves as a psychological function for the differential behaviours towards each other (Hogg, 2015).
In the Indian context, caste has been a prominent social identity category and caste-based discrimination is one of the most complex human rights issues. In line with the social identity framework, the hierarchical structure associated with caste and identification of members with their caste group remains salient to influence individual behaviours significantly. The quest for necessitating caste identity frequently urges individuals to indulge in reproducing the discriminatory rules of the caste system (Jaspal, 2011) and this effects the structure of opportunities (Hoff and Pandey, 2006). The significant role of caste identity in accessing public resources and services, particularly in rural areas has been documented extensively. But, this is largely seen from the perspectives of high and low caste groups in general, with scant and limited focus on the social identity of the service providers in the delivery of public services. In the healthcare system at the community level, the public functionaries like the Auxiliary Nurse-Midwife (ANM) and Anganwadi Workers (AWW) form the major proportion of health workers cadre. Although, an ANM provides health services to women and children in public health sub-centres, she also plays a key role in providing healthcare at the community centres like AWC along with AWW as per the healthcare needs of the community. While ANM provides specific services to women and children, the AWW plays an important role in health services like weight monitoring, visiting pregnant women to understand healthcare needs, providing supplementary nutrition, visiting newborn children, referring for health services, and educating adolescent girls about healthcare. These workers, irrespective of their caste identities, are expected to deliver healthcare services in compliance with service delivery norms. However, in the delivery of all nutrition and health services under ICDS at a community level, the AWW plays a vital role in close association with target groups. Their performance determines the extent of utilization of these services. The question that this essay explores is: Are there any variations in the access to specific healthcare services across caste groups in villages with low caste and high caste AWW?
Following the universalization of ICDS in 2005, there has been an expansion of the nutritional and health intervention nets. A network of community-based centres, popularly known as Anganwadi Centres (AWC) remains the focal points for delivery of integrated nutrition and health services. Although, the operational aspects of AWC have been examined over the years, the equity and inclusion aspects in relation to delivery of various nutritional and health services remain a major concern (Borooah, 2012;Gill, 2012;NAC, 2011;Pal, 2016). Several studies clearly indicate not only poor health status of people from socially-marginalised groups but also lower access to healthcare facilities (Acharya, 2013;Acharya and Pal, 2017;Bansod, Salve and Jungari, 2022;Baraik and Kulkarni, 2014;Baru et al., 2010;Borooah, 2010Borooah, , 2018George, 2016aGeorge, , 2018James, 2016;Nayar, 2007;Raushan and Mamgain, 2021;Shah et al., 2006). Studies also indicate discriminatory practices in delivery of health services based on social identity (Acharya, 2010(Acharya, , 2013George, 2018;Thapa et al., 2021).
However, there are studies which indicate unequal access to nutritional and health services delivered under the ICDS (Borooah, 2012;MWCD, 1015;Maity, 2016;Mittal and Meenakshi, 2015;Sinha, 2006;Swain and Kumaran, 2012); besides discriminatory and exclusionary practices against certain sections of people based on their group identities (Diwakar, 2014;Mamgain and Diwakar, 2012;Mander and Kumaran;2006;Pal, 2016Pal, , 2021. Avoidance of home visits by the community level service providers based on social identities of healthcare seekers is a common practice of exclusion and discrimination (Acharya, 2010(Acharya, , 2013Pal, 2016;Shah et al., 2006). During delivery of various health services at institutional and household levels, while a few typical discriminatory behaviours of service providers are explicit, many occur in subtle manner.
The linkages between social identity and healthcare thus assume significance when we look at the group inequality in health status, deprivation of health facilities by certain groups, and evidence of exclusionary practices during delivery of health services, and consequently unequal access to health services. Despite considerable research on inequities in healthcare, a significant knowledge gap exists with regard to the nature of exclusion and discrimination faced by different marginalised groups in accessing nutrition and health services at a community level. The effect of the social identity of the service providers on delivering nutrition and health services under the ICDS remains a neglected issue. For example, the evidence on various forms of discriminatory practices or behaviours against low caste beneficiaries as a rule is linked to the social identity of the high caste service providers and community members. One of the most dominant views is that the high caste identity of the service providers and their unfair treatment against low caste people has implications for their unequal access to nutritional and healthcare services. But the question that arises is: In villages where the service provider is from a lower caste, will it affect the utilization of healthcare services and health-seeking behaviors of low and high caste groups differentially?
In recent times, few studies draw attention towards the social identity of healthcare providers in the context of denial of access to health services. George (2016b) while looking into inclusion in healthcare delivery system using macro level data concludes that denial of access to health services for marginalized groups occurs due to lack of accessibility to health services and the practice of discrimination by various healthcare personnel at public health centres like, doctors, nurses, assistants, ANM, etc. However, the argument is made based on macro level data on the social profile of healthcare personnel at public health centres which may not adequately reflect actual discriminatory behaviors. At the interpersonal level, the personnel like doctors and nurses might not have awareness of social identity of care seekers in many cases, to manifest any unfair or exclusionary behaviours. In another study, Verma and Acharya (2017) examine discrimination in the context of the caste identity of the ANM as health service providers from public health centres. It focuses on the experiences and perceptions of beneficiaries in relation to the social identity of the service providers and vice versa. The study indicates how the caste of health service providers and seekers shape their perceptions towards each other, leading to limited social interaction between them and inequity in healthcare services. However, these observations are based on a relatively small sample and case study approach. Further, the study has a specific focus on service providers like ANM who visit villages intermittently and households rarely, and health seekers use to have limited interaction with ANM for specific health services delivered mostly at care centres (AWC) or in a location away from a low caste locality.
This essay draws attention towards some research questions in the context of delivery of and access to healthcare services under the ICDS at the community level. These include: How do the social identities of the 'targeted' beneficiaries and service providers under ICDS matter in access to healthcare services? Do low and high caste households have equal access to healthcare services provided through communitybased centres like AWC? Are there any differences in access to healthcare services in the villages with low and high caste AWW? What challenges do the low caste AWW face in delivery of healthcare services in mixed-caste villages? Thus, special attention is given on the utilization of various healthcare services delivered by the service providers with different social identities, and understanding the challenges that low caste service providers face in delivering services to diverse social groups. An attempt is also made to understand institutional and household determinants of utilization of healthcare services, and psychosocial implications of differential access to healthcare services, and unfair behaviours of service providers. The essay, thus takes a holistic approach to understanding the complex linkages between the caste identity of care seekers and service providers, and access to healthcare services at a community level given dominant social-cultural norms based on social identities of people.

Methods and Data Base
This essay is primarily based on community level data collected during 2012-13 through a large sample survey of over 4000 households and 200 service providers (AWW) at AWC spread across nearly 200 villages in three states. A multi-stage sampling method was used to select mixed caste villages whereas purposive sampling was used to select sample households across social groups as beneficiaries of healthcare services under ICDS. Household data was primarily collected from women because of their overall knowledge about how they utilize services and a limited data from small children of households. Further, the service providers under ICDS interact largely with women and children while delivering various healthcare and nutritional services. Data was collected using a household schedule designed specifically for the study purpose. The household data were supplemented by over 200 focus group discussions (FGDs) with women from different social groups.
Service providers' data primarily included experiences and perspectives of AWW in relation to delivery of various services across social groups and challenges that they face during service delivery at a community level. Data was collected using an institution interview schedule to document facilities and functional aspects of AWC and details about experiences of main functionary (AWW). Besides, institutional data included observations of normal functioning of AWC. Thus, data sets included issues on the delivery and utilization of various healthcare services, besides other information about household beneficiaries. The analyses based on both quantitative and qualitative data have been structured around the specific research questions, mentioned earlier.

Key Findings
The findings discussed in the following sections are related to access to healthcare services across social groups at a community level under the ICDS scheme, experiences and perspectives of beneficiaries of healthcare services, access to healthcare services in the villages with low caste and high caste AWW and challenges that the service providers face during delivery of healthcare services, household and institutional determinants of healthcare services, and psycho-social implications of unequal access to health services.

Differential Access to Healthcare Services
This section addresses the question: What are the intergroup differentials in the access to healthcare services provided by AWC at the community level? The differentials examined across social groups (low and high caste) in access to different health services by target groups include children between ages 0-6 years, pregnant women and lactating mothers and adolescent girls. As per the ICDS norms, most health services are to be delivered at household levels. For this, visits of AWW to households/hamlets of different social groups remains critical. Results indicate that the frequency of visits of AWW to households vary across groups with a higher number of AWW visits to high caste households. The differentials in home visits would have bearings on access to specific health services. The results however show that low caste households have relatively higher utilization of overall health services as compared with high caste households. (see Table 1) With the exception of home visits by the service provider within one week of the birth of a child and growth monitoring of children between 0-3 years of age, low caste households show higher utilization of a majority of services for different target groups. The difference between percentage share of low caste and high caste beneficiaries was found the highest for services like supplementary food during a prior pregnancy (10 per cent), followed by weighing and monitoring the growth of children between 3-6 years (9 per cent), immunization of children between 3-6 years (8.2 per cent), and Janani Suraksha Yojna (JSY) entitlements (5.8 per cent). It was lowest for the service of immunization of children 0-3 years and health-related counseling for women. As per the norms, during pregnancy, women are entitled to supplementary nutrition for six months. Although the percentage share of women receiving supplementary nutrition in the form of 'take home rations' during pregnancy is conspicuously higher for low caste women (58 per cent) compared with high caste women (48 per cent), a majority of women beneficiaries (86 per cent) receive it for up to two months. However, a higher percentage of high caste women receive it beyond two months or for a higher number of days. For example, while about 13 per cent low caste women receive supplementary nutrition beyond 2 months, the figure for high caste women is about 19 per cent.
Results show that a relatively higher percentage of children between 0-3 years from low caste (80 per cent) are beneficiaries of at least one health service than high caste children (75 per cent). However, it varies for different services like care after birth, immunization, Vitamin A supplements, health check-ups and weight monitoring. While the percentage of children between 0-3 years who are beneficiaries of growth monitoring and immunization does not show group difference, a higher percentage of high caste households are beneficiaries of home visits by AWW within one week of childbirth, and a higher percentage of low caste households are beneficiaries of Vitamin A supplementation. Among the healthcare services, the percentage of beneficiaries was considerably higher for immunization and growth monitoring for both age groups of children between 0-3 years and 0-6 years. The fact is that immunization is a programme activity for children and growth monitoring is one of the regular activities. In contrast to no group difference for children between 0-3 years in access to immunization and growth monitoring, a considerably higher percentage of low caste children between 3-6 years have access to these services. The latter may be due to a higher percentage of registered low caste children in AWC, opening better opportunities for access to these services at an institutional level. Notably, adolescent girls have the worst access to nutrition and health-related services. Only about 6-9 per cent adolescent girls utilized services like nutrition and health education/counseling and IFA supplementation. The percentage of adolescent girl beneficiaries is found relatively higher among low castes. Contrary to studies showing a lower access of socially excluded groups to various public services, the low caste households have a higher utilization of nutrition and healthcare services provided through ICDS at the community level. However, the question remains as to whether the group differences indicating better access to many health services among low caste households, is 'supply-driven' or 'demand-driven'. When many of these services are highly required by low caste households and with lack of better alternatives, they might have higher demand and reach out for various services when compared with high caste households. This may be investigated in future research by looking at the processes of accessing various services. To some extent, this can be understood later from the analysis of differences in service delivery at two delivery points: institutional and household. Another critical issue is that although low caste households have higher utilisation of various health and nutrition services, are there identity-based exclusionary bias and discriminatory behaviors during delivery of services? In view of the limited scope of this essay for a detailed study of this aspect, some cases in point are discussed in the following section. Results based on FGDs indicate different forms of discriminations experienced by low caste beneficiaries in access to health services. According to household data, 62 per cent low caste households report that high caste AWW do not come inside/enter their houses. As shared by low caste women in some villages: 'High caste AWW rarely visits houses of low caste to extend assistance to pregnant and lactating mothers. In this village, a three-month pregnant low caste woman even did not get TT injection.' 'High caste AWW does not come to our hamlet rather distributes supplementary nutrition from her house.' 'Till today no member from our hamlet has seen AWW. There is [a] reason for not coming here. She is [the] daughter-in-law of [a] high caste who does not like to come to our house.' When responding to the question of whether the AWW touch children and women while providing health services like weighing and providing polio drop?, nearly half of the low caste households report in negative terms. This is also substantiated by statements made by low caste women during FGDs, as below: We do not like to visit ANM due to preferential treatments to high caste women. ANM does not visit low caste colonies but visits high caste colonies regularly whenever she comes to [the] village. While she provides health services to high caste households in their colonies, [she] asks AWW to call low caste women to high caste localities for health services.
Although ANM visits village just once every month, she normally performs her duty sitting at AWC. Both ANM and AWW sit at AWC only. Low caste women have to walk to AWC to visit them and get their health check-up even during the pregnancy, and also have to bring their small children for health services. Both ANM and AWW practice untouchability when they deliver health services like weight monitoring and body check-up.
Low caste women do not want to come to ANM, as they do not receive proper timely information, besides discrimination that they face in the form of preferential treatment. Despite the fact that we reach the AWC first, we are asked to sit or stand outside AWC for long whereas high caste women occupy space inside AWC. We remain vulnerable to aggressive gestures, abusive words and other caste-related derogatory remarks.
During the last stage of pregnancy and the first few weeks of delivery, women normally do not want to go to AWC for a check-up because of the risk of moving out and waiting for long periods for immunization. As ANM does not visit low caste localities and many women do not come to AWC meeting, many pregnant women and lactating mothers are deprived of timely immunization and other health-related services. Yet, ANM will sometimes visit the high caste localities for health check-ups due to pressure from a few high caste persons.
Low caste women also report similar experiences of discrimination when they go to the AWC to receive supplementary nutrition. Given the focus of the essay on health services, their experiential accounts in this regard are not discussed.

Social Identity, Service Providers and Utilisation of Healthcare Services
The Anganwadi Centres (AWC), one of the last rungs of the government system, is largely located within community. A majority of AWC functionaries belong to the village community and serve as community workers. They play key roles in providing healthcare and nutrition services nearer to home although many health services under the ICDS are delivered in coordination with other health workers from the formal primary health care system. But, unlike other health functionaries, AWW as a public community worker in the care centre (AWC), remains connected with each household in the community on a regular basis, and facilitates linking households to many health facilities. Although public health facilities have increased over the years, still people mainly from socio-economically marginalised groups in villages find it difficult to have easy access to these facilities. The AWW, as a local resident worker, facilitates the processes for better utilisation of health services. Within the public policy framework, when there is always a need to ensure accountability with the community, the role of AWW remains crucial given the delivery of several integrated public services under ICDS for different 'target' groups, including strengthening and facilitating access to healthcare services and also offering referral health services.
The critical question that has been a major focus of this essay is: Does the caste of the AWW, the main service provider in the community care centre (AWC), matter in access to and utilisation of health-related services? In specific terms: Are there differences in the utilisation of health services by the eligible household beneficiaries across social groups in the villages with low caste AWW (LC-AWW) and high caste AWW (HC-AWW)? This section presents some findings based on household data on utilization of health services in relation to the caste of the AWW in villages. As a practice, AWW play a role in providing health services in collaboration with ANM from the nearest health centre. But in case of certain health services, the AWW plays a significant role. The results reveal that the access to nutritional and health services by the households varies by the social identity of the AWW.
The data indicates a perceptible difference in the home visits of low and high caste AWW to households of different social groups to provide various services including health services. While 62 per cent households in villages with LC-AWW report frequent visits of AWW, 51 per cent households in villages with HC-AWW respond the same. The disaggregated data by the caste of households indicates that a higher percentage of high caste households (62 per cent) report the visits of LC-AWW compared with 52 per cent reporting visits of HC-AWW. This is further corroborated by visits of AWW for specific services, for example, visiting the newborn child at home. While 42 per cent households in villages with LC-AWW report AWW visits for this purpose, only 25 per cent households in villages with HC-AWW report the same. Both low and high caste households had similar responses on the visits of LC-AWW and HC-AWW for newborn child. The factor of 'avoidance of home visits' is found to be important behind the dislike towards the AWC. In villages with LC-AWW, only 7 per cent households attribute the non-visit of AWW as the cause of disliking the AWC, whereas 15 per cent households in villages with HC-AWW have similar responses. Further, a higher percentage of households (41 per cent) report visits of HC-AWW to similar caste and well-off households compared with LC-AWW (26 per cent). Thus, findings suggest that LC-AWW make more home visits to both low and high caste households than the HC-AWW. The views of the household respondents on AWW home visits to an extent corroborated with AWW self-reporting. A considerably higher percentage of LC-AWW (77 per cent) than HC-AWW (64 per cent) report their visits to individual households.
Thus, it is evident that more visits are made to high caste households by both high and low caste AWW either due to preference or pressure. HC-AWW frequently avoid visiting low caste households due to caste bias. The differences in the visits of LC-AWW and HC-AWW are however not found consistent with access to certain health services. For example, growth monitoring of children between the ages of 0-3 years and 3-6 years is found to be better in villages with HC-AWW. A higher percentage of low and high caste households receive growth monitoring for children between 0-3 years in villages with HC-AWW (61 per cent and 56 per cent, respectively) compared with LC-AWW (54 per cent and 49 per cent, respectively). The corresponding figures for children 3-6 years are 34 per cent and 37 per cent respectively in villages with HC-AWW, and 27 per cent and 22 per cent respectively for villages with LC-AWW.
In contrast, a higher percentage of households (31 per cent) in villages with LC-AWW report immunization of children of 0-6 years than villages with HC-AWW (24 per cent). A higher percentage of both low and high caste children are being immunized in villages with LC-AWW. While responding to whether the AWW examines persons by touching them while providing services like vaccination and weight monitoring, about 55 per cent households in villages with LC-AWW gave a positive response while 45 per cent households in villages with HC-AWW had a similar response. As reported, in many cases HC-AWW manage to provide these services with the help of other community members.
With the higher home visits by LC-AWW, a higher percentage of households across social groups show satisfaction with the behaviour of the LC-AWW (45 per cent) compared with HC-AWW (36 per cent). This is true for both low and high caste households. A higher percentage of low and high caste households (47 per cent and 44 per cent respectively) show satisfaction with the behaviour of LC-AWW compared with HC-AWW (35 per cent and 37 per cent respectively). Although the main reason for dissatisfaction is 'non-performance of duty' by AWW, a higher percentage of households (16 per cent) attributed 'unfair/dishonest practices' as reasons for dissatisfaction with HC-AWW compared with LC-AWW (7 per cent).
The equity and inclusion aspects of the public services to a great extent are determined by the performance of service providers. This however needs to be understood from the constraints and challenges that the service providers face in the delivery of public services because of their group identities. Understanding the challenges of public community workers like AWW in delivery of interlinked health and nutrition services, would have implications for delivery of public services at a community level. The data revealed that a considerably higher percentage of LC-AWW (56 per cent) than HC-AWW (37 per cent) feel the pressure to provide services to high caste households on priority basis. Similarly, another one-third of LC-AWW and 17 per cent of HC-AWW report pressure to provide services at the homes of high caste groups.
Overall, half of the AWW report facing some kind of problems during home visit. The reported problems are related to 'not seeking advice on immunization and family planning' (26 per cent), 'non-availability of many low caste parents during home visits due to working outside home' (20 per cent), and 'demanding for work which are not part of AWW duties' (16 per cent). About half of the AWW report that all parents did not show interest in sending their children to the AWC. Of these, a majority of AWW (57 per cent) report that children belonging to general caste and OBC (those are economically well-off) do not send their children to AWC. Out of these AWW, 48 per cent cite reasons such as 'feeling bad' due to the 'presence of other poor and low caste children', 'do not want to make their children sit with low caste children with dirtiness [sic] as a matter of social position and self-respect.' Further, the presence of LC-AWW in AWC also deters many high caste parents from sending their children. Consistently, a higher percentage of LC-AWW (76 per cent) reports it to be a problem compared with HC-AWW (49 per cent).
There are evidences to show that LC-AWW often have to work within social constraints because of the identity-based discriminatory practices in community life. However, only eleven per cent of LC-AWW report experiencing discrimination while providing certain services, may be due to fear of reporting such a sensitive issue. Some common forms of discrimination reported include: not being allowed to enter a high caste house to provide health services, maintaining physical distance, and refusing to accept services from them. As a few LC-AWW are of the view: 'In villages, discrimination based on caste is a common practice, if we go to a high caste home they never allow us to go inside.' Although the AWC remains a focal point of delivery of ICDS services, the AWW have to coordinate with other health functionaries like ANM, ASHA, nurse and doctor in providing health services like immunization, vaccination and polio drops. The responses of household respondents and AWW indicate that the visits of health officials have not been regular. The doctors' services at a community level are mostly confined to referral services. Their visits to villages have been non-existent. Thus, there has been a weak link between beneficiaries and high level health service providers. However, while a majority AWW report a good relationship with other health workers like ANM and ASHA, only 12 per cent report difficulties in coordinating with these health workers allegedly due to their non-cooperation.
At the community level, a village pradhan (head of village) as part of local self-government remains the focal point of contact between officials under the local governance system and people of the village community. He/she plays an important role in many decision-making processes so far as the delivery of public services is concerned. It is assumed that low caste village heads may have some influence on service delivery for low caste people.
An analysis of household responses in 75 mixed-caste villages with information about the social identity of the pradhans working since last one year reveals that in 16 villages with low caste pradhans, the level of awareness about health services and home visits by AWW is found higher than in villages with high caste pradhans. For example, 60 per cent of low caste households in villages with low caste pradhans report AWW home visits compared to 54 per cent household in villages with high caste pradhans. However, the presence of low caste pradhans does not guarantee better access to several health services by low caste households.
The actual benefits of health services in the villages with low and high caste pradhans provides mixed results. While villages with a high caste pradhan have better access to services like weight monitoring of children, and health counseling and supplementary nutrition for adolescent girls, in villages with a low caste pradhan, households have better access to the services like health counseling for women and follow-up visits for newborn children.

Factors Affecting Utilisation of Healthcare Services
The World Health Organization proclaims that health cannot be achieved by medical care alone, social factors are equally important (WHO, 2010). Although it puts down two broad categories of social determinants: structural conditions and health system, several factors operate around them at multiple levels. Importantly, while highlighting priority areas that need public policy attention, it draws attention towards the limited access to healthcare among certain sections of people due to exclusion (Acharya and Pal, 2018). In India, various health programmes, mainly system-driven, have not helped all Results (Table 2) show that the caste of the household does not play a significant role in the access to healthcare services by pregnant women and lactating mothers although high caste women have better access to health counseling. While low caste women demand for many health services as a better option for them, high caste women receive them as per their requirement and as reported by AWW earlier mostly due to preferential treatment. However, families with the membership of a member in any committees increase the chances of access to health counseling (β = 1.00), health check-ups (β = .90), and weight monitoring (β = .83) by pregnant women and lactating mothers. For those families with a lower level of education of the head, there is more likely utilisation of services like health counseling (β = -.76) and weight monitoring (β = -.42) by the women. Alongside, among institutional factors, the lesser the distance between household and AWC (care centre), more likely is the utilization of all three healthcare services. Understandably, higher the home visits of the service provider (AWW), higher the likelihood of utilization of all healthcare services. Similarly, higher is the hour of functioning of AWC, there is more likely utilization of services of health counseling (β = .79) and weight monitoring β = .59). In AWC with high caste AWW, there is a higher likelihood of health counseling by women (β = -.52), possibly due to the factor of social acceptance by women from both caste groups. From the results, it appears that institution-related factors to a greater extent determine the utilization of healthcare services by the target women compared with household factors. It also suggests that the supply side factors are important in the utilization of health care services by pregnant women and lactating mothers.
The results also indicate that factors like membership of family member, education level of head of family, distance of AWC, and caste of AWW play a significant role in utilization of health counseling by adolescent girls. These factors also significantly influence the utilisation of IFA supplementation. The likelihood of utilisation of health counseling and IFA supplementation by high caste adolescent girls increases with high caste service providers. Table 3 shows factors that determine the utilisation of healthcare services by children from both the 0-3 and 4-6 years age group. Caste plays an important role in growth monitoring of children between 0-3 years, indicating higher growth monitoring by high caste households (β = .30). The group-specific service like followup visits for children between 0-3 years is determined by the education of the head (β = -.55), membership (β = .53), distance of AWC from households (β = -1.64), caste of AWW (β = -.65), home visits of AWW (β = 1.74), and hours of functioning of AWC (β = .60). Factors which determine the utilisation of healthcare services, mainly growth monitoring of children in both 0-3 and 4-6 years are: membership of family member, caste of service providers, home visits of service provider and hours of functioning of AWC. The higher the education level of the head of the household, the less likely of growth monitoring of children between 4-6 years ((β = -.39), but the education does not play a significant role in growth monitoring of children between 0-3 years. It may be due to the fact that often, parents irrespective of their education, do not like to go for growth monitoring of small children due to certain beliefs associated with it. As evident from AWW responses, they sometimes face difficulty in providing services like weighing children and counseling to women on health-related issues due to social taboos. In many villages, high caste parents do not send their children to AWC when the AWW are from low caste. This is reinforced by the findings that greater the distance between household and care centre, the lesser likelihood of growth monitoring for children 0-3 years (β = .56). Further, higher the home visit of AWW, it is more likely that children will have access to various health services including weight monitoring. But, the notable aspect is that with the exception of the distance factor and home visits, no other factor plays a significant role in the immunization of children between 0-3 years, but almost all household and institutional factors determine access to immunization for 4-6 years. The difference may be attributed to the nature of health services like immunization being a programmatic service of the state for specific age groups. The results thus point to many factors-locational, operational and behavioralthat determine the utilization of health services by different target groups. Among household-related factors, membership and education level largely influence the utilisation of services. The caste identity of service providers plays an important role in the utilization of some health services by women and children. Contrary to beliefs that many clinical issues affect low caste, a variety of social determinates play a vital role in unequal access to healthcare services.

Inequity in Healthcare: Psychosocial Implications
When we talk of caste, inequity and discrimination in healthcare system, it brings the debate of how the dominant cultural practices of discrimination that creates psychosocial conditions effect one's mental and social well-being. Caste-based inequity impacts an individual's opportunity to have access to basic public resources and services, and in turn, upon their physical and psychological health. The basic premise is that inequality in utilization of any public services like health, damages the quality of social relations, one of the most important ways that inequality affects the quality of life (Wilkinson, 2005). The extents to which people are involved in local community life also confirm the socially corrosive effects of inequality (Wilkinson and Pickett, 2007). The inequality is sometimes linked with psychosocial health (Chandra, 2009). Despite public policy and institutional guidelines and principles in delivery of public services, the persistence of caste-based inequity creates dejected and deserted situations for low caste groups-both as consumers and providers of health services. Some disadvantages witnessed in public service delivery are common to all social groups, mainly due to failure on the supply side or gaps in the operational aspects. However, many disadvantages are a function of discriminations in delivery of services. One of the most perceptible evidence of discrimination is found in fewer AWW visits to low caste households either voluntarily or under pressure from others. This limits not only access to healthcare but also knowledge about health problems and information about possible cures. There are evidences of specific discrimination against low caste at the time of service delivery, for which the service providers are held responsible as they are expected to work as per professional ethics rather than community norms. These would not only effect the utilisation of services but also other psychological aspects in terms of loss of faith in institutional norms and health-seeking behaviours. Attention is drawn to the consequences of discriminatory practices to shed light as to how some of the disadvantages due to unequal access to health services besides behavioural manifestations of service providers are intensified for low caste.
The data confirms that most service infrastructures are located in places closer to high caste habitations. It creates not only physical distance between service delivery points and low caste households but also social distance between social groups within the community. With the distanced service delivery points (AWC), the physical distance becomes even wider by the occasional visits of service provider. Thus, the distance factor limits the move of both beneficiaries and providers. This isolates low caste groups physically and socially, and affects access to health services. The locational disadvantage coupled with other challenges due to fewer visits of service providers accentuates the problems of pregnant women and lactating mothers from low caste, forcing them travelling to AWC for essential services. As many do not have other provision of healthcare except AWC, they live with this reality, which deviates from the norms of public service delivery.
Very often, AWW instead of visiting the low caste households to pass information, prefer to call a meeting in a place outside their hamlet. For example, monthly health camps/programmes are held either at AWC or community places located in high caste localities. As a practice, AWW in collaboration with ANM provides health services at AWC or in a 'liking place' at a high caste locality. It is expected that the pregnant women and lactating mothers travel to these places to receive health services. This puts low caste women at additional disadvantages.
As evident from FGDs with low caste women, they sometimes avoid such meetings at high caste localities because of past experiences of exclusionary bias such as sitting away from women of high caste, waiting for longer, being subject to 'casteist remarks' and rude behaviour of service providers, exposing them to humiliating situations. These service delivery points become deterrents for low caste women, leading to lower access to vital information about health-related issues. Verma and Acharya (2017) find exclusionary practices have negative implications for the health-seeking behaviours of the low caste groups. However, the service delivery mechanism sometimes makes low caste women helpless as they cannot dare to complain against it, and even if they do, it may not help them much. As low caste women in a few FGDs express: 'We do not complain because we know, no action will be taken, rather she (AWW) will scold us.' 'To whom should we complain? No one listens. If AWW knows, she will fight.' 'AWW says that nothing will happen complaining against me. In turn, you will suffer. The supplementary nutrition material will stop.' 'People from our community do not complain, because we are poor. Rich people will start pressurizing. No one listens to poor.' The results also suggest that low caste households have relatively higher access to the institutional 'in-house services' (i.e. services provided in AWC) whereas high caste households to the 'out-house services'. The findings also indicate that the awareness level of low caste women about various aspects of health services under ICDS is lesser than high caste women. This 'awareness gap' is wider between social groups particularly on issues related to frequency of weighing children and referral services. It is clear that the social identity of service providers matters in the utilization of health services by different social groups. As observed, there is a lot of hesitation among high caste parents to send their children to the AWC where AWW are from lower castes (Mander and Kumaran, 2006;Pal, 2016). This type of exclusionary bias frequently eases the process of exclusion of low caste and creates social conditions that intensify 'feelings of inferiority' among low caste beneficiaries and threaten their dignity. When we talk about such humiliating behaviours, the role of service providers remains critical. But, sometimes this is reinforced by the indifferent behaviour of service providers themselves. Given that caste norms at a community level very often dominate the process of service delivery, this results in not only unequal access to healthcare services or limited access to certain services under discriminatory conditions, but also brings several adverse consequences that intensify social disadvantages.

Discussion and Conclusions
In recent times, inequalities in health status and access to healthcare services across socio-religious groups have been widely documented. While macro level official data points to the poor outcomes on several health indicators among the socio-economically marginalised sections of society, empirical evidences indicate their differential access to health services. In light of the persistence of group inequity in access to interrelated nutrition and health services in India, this essay looks into factors that affect utilization of healthcare services with a special focus on the social identity of users and providers of the healthcare services. Based on the experiential accounts of a large sample survey of household beneficiaries and providers of healthcare services, the essay identifies a few areas of concern that need special attention.
The essay reaffirms that the social identity of caste marginalizes low caste people in a multifarious manner in their access to healthcare services. Despite specific norms and guidelines for the implementation of various healthcare programmes, the dominant caste norms at a community level interfere with the delivery of healthcare services. It is not only the low caste identity of users of services but also of service providers that matters in differential utilization of healthcare services. The delivery services are largely influenced by the local social environment in which they live and work as public service providers. The service providers from low caste group even fail to do justice for low caste people under social compulsions. The service providers across caste groups remain indifferent to the day-to-day concerns of low caste beneficiaries, in turn bringing multifaceted challenges for them.
The public policy focuses on universal access to healthcare services. Still, healthrelated programmes have a specific social orientation in the delivery of healthcare services wherein marginalised groups always remain a priority in delivery of services. The unequal access to such services across social groups raises concerns over the supply side deficiencies, particularly the accountability and behavior of service providers. The major public policy concern remains on how to address the effects of the social identity of providers of health services at a community level in ensuring equality in utilization of the services when the local social structure is entrenched in a hierarchical system. While high caste service providers in a pursuit of maintaining high caste identity and autonomy, tend to show differential behaviours towards low caste users of health services, low caste service providers, on the other hand, under the influence of dominant social norms, face conflict in course of providing health services to different social groups. This minority face many challenges including the experience of differential treatment while delivering health services to different social groups. Being in a weak 'social position', they have to comply with dominant social norms in the course of delivery of health services under compulsions.
Low caste service providers are 'avoided' or 'neglected' by the high caste people during delivery of services while high caste providers themselves avoid visiting low caste households for service delivery. Given that high caste people can have access to alternative health services without taking any kind of help from low caste service providers, hence, this may not have much implication for them. However, for the low caste people, fewer visits/interactions of high caste service providers to low caste people matter. With the caste-based social order at a village level, sometimes both low caste service providers and beneficiaries are forced to accept the way high caste people or service providers accept it. Thus, the low caste service providers sometimes have to deliver health services under certain compulsions whereas high caste providers work with much liberty. It is interesting to note that when the low caste AWW faces difficulties in providing health services during her visits to high caste households because of her caste identity, the high caste AWW makes fewer visits to low caste households. They can refuse to visit low caste households as it may be accepted by the social norm. Thus, in mixed-caste villages, the healthcare behaviours of service providers with different social identities may not always be intended behaviours, but takes place due to the interplay of social identity and local social norms.
As a matter of fact, health policy framework and community level health system in India have not been well-found to address exclusionary practices based on social identity, affecting delivery of and access to healthcare services. Given that public health services are the only options for marginalized groups including low caste groups, when low caste service providers cannot ensure health services within their own community due to dominance of other socio-cultural norms, this is a challenge for achieving inclusion in the health services. Since, caste-based exclusions are entrenched in many mixed caste villages, social inclusion in delivery of healthcare services needs special policy attention. Given that providing health security to all has been a national priority, a special focus on health equity through community level services remains a critical area of public health policy interventions.