‘Trust and teamwork matter’: Community health workers' experiences in integrated service delivery in India

Abstract

A comprehensive and integrated approach to strengthen primary health care has been the major thrust of the National Rural Health Mission (NRHM) that was launched in 2005 to revamp India's rural public health system. Though the logic of horizontal and integrated health care to strengthen health systems has long been acknowledged at policy level, empirical evidence on how such integration operates is rare. Based on recent (2011–2012) ethnographic fieldwork in Odisha, India, this article discusses community health workers' experiences in integrated service delivery through village-level outreach sessions within the NRHM. It shows that for health workers, the notion of integration goes well beyond a technical lens of mixing different health services. Crucially, they perceive ‘teamwork’ and ‘building trust with the community’ (beyond trust in health services) to be critical components of their practice. However, the comprehensive NRHM primary health care ideology – which the health workers espouse – is in constant tension with the exigencies of narrow indicators of health system performance. Our ethnography shows how monitoring mechanisms, the institutionalised privileging of statistical evidence over field-based knowledge and the highly hierarchical health bureaucratic structure that rests on top-down communications mitigate efforts towards sustainable health system integration.

Keywords:

• integration
• community health workers
• trust
• teamwork
• India

Introduction

Beginning with the Alma Ata Declaration in 1978, the need for comprehensive and integrated approaches to health care has been well acknowledged. However, for a long time, such approaches had been enmeshed in polarised debates about vertical versus horizontal health programmes and/or comprehensive versus selective primary health care (Mills, 2005; Rifkin & Walt, 1986). The recent global calls to achieve universal health care and the Millennium Development Goals (MDGs) in low- and middle-income countries have revived, at least rhetorically, the emphasis on strengthening health systems through comprehensive and integrated approaches to health care. In line with these global developments and after a protracted focus on disease-specific programmes, India has recently embarked on a path to revamp its public health delivery system. The National Rural Health Mission (NRHM) in India was launched in 2005. It seeks to revitalise the primary health care approach and put in place several ‘architectural corrections’ in the basic health care system to guarantee equitable access to quality health care for the rural poor, women and children (NRHM, 2005). The NRHM sets out a horizontal and comprehensive approach to health care to undertake these architectural corrections. Some of the core strategies of this horizontal approach include: integration of vertical programmes and structures; integrating health with its broader determinants; mainstreaming traditional systems of medicine and revitalising local health traditions; decentralised health planning; effective community participation and ownership of health; and improved and effective public health management (Government of Odisha, 2007; NRHM, 2005). Notions of integration in NRHM thus flow from the comprehensive primary health care approach, that reinforce principles of community participation, provision of comprehensive care (curative, preventive and promotive) inter-sectoral collaboration, decentralisation and equity (NRHM Mission Document, 2005). Such an approach has been justified in light of India's race towards achieving the goals of MDGs 4 and 5 (reducing child mortality and improving maternal health, respectively).

World Health Organization (WHO, 2008a, p. 1) defines integrated service delivery as ‘management and delivery of health services so that clients receive a continuum of preventive and curative services, according to their needs, over time and across different levels of the health system’. Despite the wider policy acknowledgement of the need for such an integrated approach to health care, empirical evidence on how such integration operates on the ground is rather sparse and disparate (Atun, Jongh, Secci, Ohiri, & Adeyi, 2010; Wallace, Dietz, & Cairns, 2009; WHO, 2008a). Existing literature on global programmatic experiences with integration of health services tend to approach delivery of services as a technical and mechanistic process (Banerjee, Elamon, & Aggarwal, 2009; Partapuri, Steinglass, & Sequiera, 2012). Integrated service delivery in this sense is an ‘approach of combining services of multiple interrelated diseases to increase overall efficiency of the health system and patient convenience’ (Lenka & George, 2013, p. 1). For example, integration would include combining delivery of family planning messages during routine immunisation sessions or distribution of mosquito nets during post-partum visits. Often referred to as ‘service add on’ (Magtymova, 2007), benefits of integration are assessed through data on the uptake of the specific health services reflected in an increase in contraceptive prevalence, improved immunisation coverage or uptake of mosquito nets (Banerjee et al., 2009; Partapuri et al., 2012; USAID, fhi360 & PROGRESS, 2011). Feasibility of integration is weighed in terms of health system factors like availability of human resources, compatibility of services or supply chain management and infrastructure (Lenka & George, 2013). The focus on supply-side health system factors, though important, assumes that community demand relies unproblematically on providers delivering services. It is rightly argued that ‘commentaries on health sector and systems tend to focus on structural aspects and not on the actors who comprise the systems’ (Sheikh & George, 2010, p. 2). Ethnographic evidence suggests that the demand/uptake of health services is linked to a host of factors, such as the community's perceived vulnerability to a specific illness for which the health service is offered, previous experiences with other state health services, modes of health communication, interaction with health workers and broader political identities and perceptions of the state by the community (Leach & Fairhead, 2005; Mishra, Flikke, Nordfeldt, & Nyirenda, 2013a; Nichter, 1995). Exclusive focus on individual health services and sites of delivery ignores larger social processes at work at the intersection of supply and demand, as well as providers and local communities, thus reflecting little on how integration is achieved or even the sustainability of integration measures.

Using NRHM as a case study, this article offers a grounded perspective on integration of health services through capturing the everyday experiences of community health workers located at the interface between the formal health system and the local communities. Existing ethnographic studies show how the social experiences of health workers are important to capture in order to understand how policies are translated on the ground, communities are mobilised, care is delivered and raw data on the health status of populations and functioning of public health programmes are produced at the local level (Coutinho, Bisht, & Raje, 2000; George, 2010; Mishra, Hasija, & Roalkvam, 2013b; Walker & Gilson, 2004). This article adds to the literature and discusses community health workers’ experiences with integrating immunisation with broader maternal and child health services through the village outreach sessions called Health and Nutrition Days.

Methods

This article draws on 8 months of fieldwork conducted in 2011–2012 in one of the southern districts in the state of Odisha, India. The district is largely inhabited by indigenous communities (referred to as ‘adivasis’ and/or tribal by the communities themselves and others). The fieldwork was conducted as part of a larger ethnographic research study that sought to chart the growing focus on health system strengthening within global and national vaccination policies and explore how debates about health systems and strengthening are played out at global, national and local levels. This article is concerned with the local level, specifically, how health workers understand and translate their roles within the NRHM's emphasis on strengthening integrated and comprehensive primary health care. Additionally, the article draws on the author's prior fieldwork experiences focusing on local immunisation practices conducted in the same region between 2009 and 2010 (Mishra et al., 2013b).

The fieldwork was conducted by the author along with two junior researchers trained in anthropology. Data were collected through participant observation in two villages selected in the district, including during 8-monthly Health and Nutrition Days (outreach sessions) in each village; monthly supervisory meetings at the primary health centre which caters to the villages; training sessions of health workers; and panchayat (local government) level meetings. Additionally, we observed rituals/healing sessions in which health workers participated. We conducted open-ended in-depth interviews with 12 health workers, and accompanied these health workers in all their routine activities beyond the outreach sessions for a period of 6 months, including when they accompanied a pregnant woman to the health centre, took a malnourished child for treatment, distributed iron syrups to the villagers or accompanied a villager to the traditional healer. We also conducted interviews with 18 sub-district level health officers and with 43 villagers (both men and women). Data were analysed through a grounded theory approach (Strauss, 1987) by developing detailed coding and constantly validating responses through juxtaposing different sources of data (interviews with observations, data from interviews with health workers with responses from the communities). The coding was further refined to develop sub-themes and themes, and relationships between themes were further examined. This was analysed iteratively, by constantly going back and forth between these field data and secondary literature on the subject. Apart from ethical approval at the host University (University of Oslo, Norway), permission to undertake the study was obtained from the district health authorities in the state. Study protocols and findings were shared with them. Informed consent was obtained from all the participants.

Integration sites and actors: outreach sessions

At the community level, outreach sessions like Village Health and Nutrition Days (also known as immunisation days) are an important forum for provision of integrated and comprehensive care. The NRHM mandates that Health and Nutrition Days, hereafter referred to as outreach sessions, are organised once a month in each village. These sessions aim to serve as an important mechanism under NRHM for the convergence of all health-related activities through inter-sectoral collaboration, thus bringing together Department of Health and Family Welfare, Department of Women and Child Development and community representatives including the Panchayati Raj (local government) members in the village (NRHM Mission Document, 2005). As per the NRHM guidelines, these sessions offer a package of services pertaining to maternal and child health, communicable and non-communicable diseases; address social determinants of health including sanitation, nutrition and gender; and facilitate the collection of data on specific needs of vulnerable populations, vital events including disease outbreaks and audits of maternal and child deaths (NRHM, 2007). This package of services is offered through direct delivery (antenatal care [ANC], treatment of minor ailments, distribution of iron syrup); identification and referral (for symptoms of tuberculosis, malaria and others); information and counselling on several issues including early signs of communicable diseases, existing gender-based laws, importance of a clean environment, locally available nutritious diet and prevention of tobacco, etc. The NRHM envisages that these sessions provide an effective platform for delivering first-contact primary health care, thus maximising the points of interaction between the community and the formal health system.

Three sets of community level female health workers are principally involved in mobilising and delivering health services during these outreach sessions, though the NRHM guidelines envisage the involvement of panchayat members, primary schoolteachers, and traditional birth attendants in supporting and facilitating these sessions (NRHM, 2007). The three health workers are composed of an Auxiliary Nurse Midwife (ANM), an Accredited Social Health Activist (ASHA) and an Anganwadi (literally translated as ‘courtyard shelter’) worker (AWW). Each has different professional training and skills, performance incentives, professional trajectories and accountability mechanisms.

The ASHAs (who are addressed through this acronym by everyone, including the villagers) are village-level health workers who have been recruited since the start of the NRHM in 2005. One ASHA is responsible for a single village (population of about 1000). They are recruited on a voluntary basis from the village by the local government village representatives and are supposedly accountable to the community. ASHAs receive about 3 weeks of training over five rounds, focusing on their role as a link between the health system and the village. They are responsible for mobilising the community to access public health services, including immunisation, hospital birth and ANC; identifying and referring people affected by tuberculosis, leprosy and malaria; and promoting family planning programmes. Apart from this, they have a broader role of being health activists in the community by creating awareness about health and its determinants, mobilising the community towards local-level planning and increasing utilisation and accountability of existing health services (National Institute of Health and Family Welfare [NIHFW], Government of India, 2005). Although they receive no fixed salary, they are paid performance-based cash incentives, the amount of which differs according to the nature of the service (e.g., 700 Rs. or US$11 for mobilising and accompanying a pregnant woman for institutional delivery, 50 Rs., less than US$1, for organising an immunisation day).

Unlike the ASHAs who are recent recruits, the ANMs (also addressed through this acronym), have been a part of the health bureaucracy since the 1950s. An ANM serves four or five villages (a population of 5000), constituting a sub-centre, the lowest of the three-tier primary health care organisation in India. The ANMs’ roles have changed over time, from being focused initially on midwifery to include broader maternal and child health services, family planning services, nutrition and health education, immunisation, treatment of minor ailments and epidemic tours during outbreaks. Re-designated as multi-purpose workers, the ANMs are permanent government employees, and receive fixed salaries (approximately 10,000 Rs. or US$180 per month). They receive 1.5 to 2 years of training in midwifery, as well as periodic on-the-job training in different components of national health programmes and interventions. Unlike the ASHAs, the ANMs do not reside in the villages that they serve. Instead, they travel to these villages on different days, particularly for outreach sessions. The ANMs are accountable to the local health bureaucracy (report to a Medical Supervisor) in the primary health centre headed by the Medical Officer.

The AWWs, on the other hand, are recruited from the village (one AWW per one village, constituting of a population of about 1000) through the Integrated Child Development Service (ICDS) programme that the Government of India introduced in the mid-1970s. This is organised under the Department of Women and Child Development, rather than the health bureaucracy, which is administered under the Department of Health and Family Welfare. AWWs work at the intersection between the health and the education needs of children. They are responsible for new born care, as well as ensuring that all children below the age of six are immunised. They provide pre-school education to children between 3 and 5 years old and are also responsible for providing supplementary nutrition to both children below the age of six and pregnant and nursing women. AWWs receive monthly salaries, though well below that of ANMs (2000 Rs. or US$36). They report to the supervisor of the programme at the sub-district level. Marital status (married) and residence (residing in the village where she works) are important considerations for recruitment of both ASHAs and AWWs. As discussed below, these differences in their professional training, status and role are important considerations in ways they conduct outreach sessions and engage with the community.

While trying to operationalise the NRHM's strategy of integrating delivery of services through outreach sessions, health workers in our study emphasised how effective teamwork was critical to their practice of delivering care, although they acknowledged the challenges posed by differential training, salary, status and role.

Eliciting and sustaining trust with the community

Health workers’ efforts to provide integrated health services and care were not restricted to either individual outreach sessions or health services. More critically, it was about building relationships with the community, ‘an important trust building mechanism’ (Rowe & Calnan, 2006, p. 5). Thus, mobilisation, communication and delivery of services were embedded in the dynamics of relationship building at the interface between the health system (mediated by individual health works) and the community. An ASHA worker clarified what it means to be a good worker and to build a healthy relationship with the community:

An ASHA should have the ability to cooperate with others. Most importantly, more than money she should be interested in her work. If a child suffers from diarrhoea, an ASHA should accompany the patient to the primary health centre rather than just showing the way to the centre.

In a similar vein, another ASHA added:

Villagers trust me. They say ‘ASHA sister is there so there will not be any problem during delivery’. Once a woman developed labour at night. I arranged an auto-rickshaw and accompanied her to the referral hospital, as she wanted to be taken there (instead of the primary health centre). One of the important qualities of an ASHA is that one needs to talk to the people of the village regularly.

An ANM added, ‘we need to be there when the community members want our help, you know, be one of them’. Such qualities of ‘being there’, being sisterly, responding to the villager's needs and preferences are important elements of the meaning of trust that the health workers espouse. Health workers’ experiences suggest that building relations of trust with the community are not an ad hoc event but rather a continuous process of eliciting and sustaining such trust. Our field observations suggest that such trust cannot be taken for granted but needs to be nurtured. A host of factors are at play in the trust-building process, beginning with the social status (caste, ethnic, familial relations) of individual health workers, modes of communication, ability to cater to community health and non-health needs and the community's own prior experiences with other health interventions, including their perceptions of the state (see Roalkvam, this volume).

For the ASHA and AWW workers who mostly live in the villages they serve, creating demand for any health service implies eliciting trust in their roles, both as health workers and, by extension, state representatives, and as members of the community. For the tribal people in the villages studied, the state (referred to as the ‘Government’) is represented by the health workers. Vaccination or institutional delivery are not mere biomedical interventions, but rather reach the community as interventions provided by the state, such that for the local community, engaging with vaccines or ANC entails an engagement with the state. More than any other programme, due to its scope, approach and delivery mechanisms, the NRHM has made the state distinctly visible in local communities through the everyday material practices of immunisation days, recruitment of health workers, disbursement of cash incentives and distribution of iron syrup and other medicines. For tribal people, these material practices project the state as benevolent actors concerned about villagers’ vulnerability. This is reflected in the villagers’ statements: ‘The Government knows we are prone to diseases like diarrhoea, malaria and hence provides medicines through the ASHA sisters and the mobile clinics’; ‘The Government is helping our women to deliver in the hospital and provide cash incentives’. Communicating about state health services, as the workers shared (particularly the ASHAs and AWWs), was enmeshed in everyday interactions taking place during routine work in the river, agricultural fields and community rituals.

Health workers engaged in the process of building trust with the communities in several ways. For instance, they repackaged the merits of health services in terms of the community's existing health priorities, or ensured that these priorities were taken into account through relevant means. For example, ASHA workers communicated to the villagers that vaccination was ‘beneficial for overall health of children’ (instead of how each vaccine helps in preventing a specific disease) and even for curing ‘malaria, fits, diarrhoea’. Such repackaging is justified on the grounds of addressing the community's traditional concerns for protecting child health and the burden of common diseases. Further, they tried to ensure that the gynaecological complaints of all village women – rather than only the pregnant women who are targeted as part of the NRHM's focus on reducing maternal mortality – are addressed. In most of the cases we observed, the ASHA worker took to the doctor in the primary health centre (8 km away from one of the villages) both pregnant women who were eligible for ANC check-ups, as well as other women who had other complaints. The ASHA would not only accompany them, but also facilitated the interaction with the doctor, took care of the prescription and arranged the medicines from the chemist. By expanding their remit to cover all women and address their health needs, they seemed to successfully contribute to the process of building trust in health workers and formal state health services, as evidenced by a young married woman lamenting the death of her in-laws 7 years prior due to tuberculosis: ‘Had it been now, my in-laws would not have died, ASHA sister would have taken them to hospital and given medicines’. Similarly, another elderly woman attributed reduced mortality in her village to the ASHA's intervention: ‘Deaths, suicide cases have reduced in the village as ASHA gives medicines, (and) takes us to hospital when necessary’.

Apart from addressing the immediate health concerns of villagers, health workers also ensured that all state health services were moulded to accommodate local aetiologies of illness and remedies. Among the tribal communities of Odisha, health and illness are believed to be rooted in a number of factors including poverty, supernatural forces and lack of access to medicines. For example, the high number of stillbirths and the vulnerability of children to common illnesses are predominantly explained through the role of the dumas (ancestors) who need to be appeased periodically through rituals both at individual households and at community levels. A child (till he/she turns a year old) is considered to belong to the world of ancestors and becomes a social person only through these rituals. Thus, protection of child health involves a number of traditional rituals beginning with the protection of the foetus in the womb to safe delivery. While promoting ANC, immunisation and institutional delivery for pregnant women, ASHA and AWW workers actively participated in these traditional rituals. For the community, care has a larger connotation that involves integrating local aetiologies of illness, notions of health and healing modalities with that of the state-provided biomedical health services. Thus the traditional rituals of godh bhariba (a ritual aimed at protecting the foetus in the womb), the ritual tying of sacred thread and bila sukheiba (the ritual offerings to a tree representing the goddess responsible for ensuring healthy children) need not conflict with immunisation, distribution of nutritious meals and treatment for malnourishment in the Community Health Centre. The health workers promoted and subscribed to such notions by focusing as much on accompanying women to deliver in the public hospital (for which they are paid cash incentives) as on accompanying them to the female shaman to seek treatment for infertility and fits (two clinically under-diagnosed complaints in the region and for which there is no linked cash incentives). However, these efforts were confined to the community space and never shared with senior health officials, lest this might be seen as promoting quackery.

One of the ways NRHM seeks to promote integration is by revitalising local health traditions, though modalities of revitalisation are so far absent in the policy documents. Hence, for the officials in the health bureaucracy, all forms of traditional healing are branded as quackery. The potential role of health workers in initiating critical dialogue between different systems of medicine and providers (in the light of the community's preferences, notions of efficacy, common complaints for which traditional rituals are sought) is thus ignored (Scott & Shankar, 2010).

Another key element that health workers drew attention to in their experiences with integrating health services was the compatibility of curative and preventive services. Malaria, diarrhoea, skin-related diseases and generic symptoms like fever, headache and body ache are common complaints in the villages. There is a huge demand for medicines to cure these symptoms and diseases. Studies show how the credibility of primary health workers is linked to their ability to provide curative services and adequate supply of drugs (Nichter, 1995; Paralato & Favin, 1982). Health workers in our study reinforced the findings of these studies. The ANMs establish legitimacy largely through their ability to offer minor curative services. Villagers talk about a good or inefficient ANM in terms of her contribution to diagnosing and, where necessary, giving adequate information about referral service. During outreach sessions, villagers often came to the ANM with complaints of night blindness, fever or diarrhoea. The ASHAs and AWWs, on the other hand, ensured an adequate supply of drugs pertaining to more common complaints. The Government of Odisha facilitates the distribution of medicines through the primary health centres for combating the five priority diseases (tuberculosis, malaria, pneumonia, diarrhoea and leprosy).

The NRHM seeks to resolve the institutionalised dichotomy between curative and preventive services that has plagued the Indian public health system for a long time. The outreach sessions are important forums for resolution of this dichotomy by addressing curative, preventive and promotive services. However, the community's demand for preventive health services is largely mediated through their access to and experiences with curative services, particularly for the common diseases that become their immediate priority health needs (Mishra et al., 2013b; Paralato & Favin, 1982). Hence, adequate supply of medicines catering to these health needs is a major concern for the health workers, often voiced during monthly supervisory meetings. An ASHA worker explained:

The other day, the AWW shared that four children were suffering from diarrhoea and I should bring medicines. I did not have enough. How could we tell the mothers of these children to not bother about the diarrhoea or go to the PHC on their own but come to the outreach session the next day to get their children weighed?

Adequate availability of medicines with the ASHA/AWW is also a source of potential conflict among health workers and possible complaints against the ANM specifically, who is supposed to facilitate supply of drugs (at least for the ASHAs). Social relations of trust between the health workers and the community however go beyond the delivery of specific health services. Health workers provide information when demanded on government insurance schemes, for instance in the wake of a house fire, submission of relevant papers to the administrative office for availing of educational scholarships meant for eligible low-income families, etc.

Health workers thus emphasised values of cooperation, continuous and open communication, personal and professional motivation and empathy as critical elements of trust, values like those Gilson (2003, p. 1461) argues are important to produce health through a trust-based health system. It is around the acknowledgement of such values, communities talk about an ‘active/good/cooperative’ and ‘inactive/non-cooperative/indifferent/rude’ health worker. Our data indicate that relationships of trust and reciprocity work better where the ASHA worker is from the same community in a relatively ethnically homogenous village and is recruited through the involvement of village members. On the other hand, in villages where the ASHA does not reside in the village, or handles more than two or three villages and has been nominated by the ANM without village involvement, such trust was missing and the ASHA was regarded merely as a subordinate of the ANM. In a multi-caste or multi-tribe village, the caste and class status of the health worker and familial relations with other members of the village mediate in relationship-building and hence service delivery. During our visit to one such village, villagers even refused to lead us to the ASHA's residence and later complained about how indifferent and inactive an ASHA she was. We learned that this ASHA's recruitment had been controversial, as she was married to a rich contractor and a converted Christian, pitting her as someone who is privileged and unfit to relate to the concerns of the villagers. In multi-caste villages, the process of trust-building with the community is thus fragile and requires much more concerted and careful efforts on the part of both the community and the health workers.

Teamwork

In addition to trust, health workers also highlighted the role of effective teamwork in providing integrated and comprehensive care. The team dynamics we observed among the health workers are important not only for the success of individual episodes of outreach sessions but also its contribution to legitimating and sustaining their role as health workers. The vertical power hierarchy between the ANM on the one hand and the ASHA and AWW on the other is explicit though the ASHAs are supposed to be accountable to the local government/community. The ANMs’ power emanates from their training, salary and possession of technical skills to diagnose and treat minor ailments and ‘put injections’. The ASHA's ability to fulfil her job responsibilities directly depends on the ANM's support. For example, it is the ANM who distributes drugs to the ASHA; provides help and advice for pregnant women, even at odd hours; clears the papers to enable the ASHAs to avail of the cash incentive for taking women to institutional delivery; provides curative care to those whom the ASHA and AWW bring to the immunisation site. Though the AWW submits records to a supervisor in the ICDS, these records are validated and examined by the ANM in the village. One AWW explained, ‘we cross check our records (ANM and AWW) and ensure that we are presenting the same set of data to our respective supervisors. These will be discussed together in the sub-district and the district level, if there is discrepancy, we will be in trouble’. Further, the ANM validates and signs the document certifying the grade of a malnourished child for referral for which the AWW gets cash incentives. Both the technical and signing authority put the ANMs vertically above the ASHAs and AWWs. Such power is acknowledged positively on the one hand, and yet subtly resisted. Such resistance, however, is expressed off stage. For instance, ASHAs and AWWs complained to us that an ANM often ‘doesn't take rounds of the villages and enquire about people's health’; ‘makes mistakes in names while recording names of eligible children for vaccination’; ‘comes to the villages only for two fixed days – outreach sessions’; ‘keeps the medicines to herself and does not give it to us’; (then) villagers get upset when they realise that we do not have the stock’; and ‘doesn't cooperate to get the cash incentives’.

Such resentment can be attributed to a number of factors such as unequal professional growth trajectories, fixed salaries versus incentives and professional vulnerability in the absence of signing and other symbols of authority. Despite the resentment, all three of them realised the value of teamwork. As one ANM clarified, ‘Taking care of pregnant women is necessarily group work. Neither the ASHA nor the ANM alone can do much; it needs the cooperation and equal commitment from each one of us’. An ASHA worker added ‘We fall back on the ANM for help in clarifying the usage of any medicine, responding to villagers’ queries regarding post-vaccination effects and submitting our records’.

Values of cooperation were expressed in clarifying how each one needs the other. The ANMs depend on the ASHAs and AWWs. The ASHA's role is critical to mobilise women for ANC registration, institutional delivery and accessing the services in the outreach sessions. As we observed during an outreach session, one of the ANMs told an ASHA: ‘you should tell the villagers to take iron syrup. They will listen to you more than me’.

Strategies to build teamwork include keeping each other's contact phone numbers, meeting or talking to plan outreach sessions, trying to develop a common understanding of state health programmes, helping each other in filling out health and nutrition records and devising ways to cooperate to obtain cash incentives (using the system creatively to draw mutual benefits). Thus, during an outreach session, we observed that the health workers shared their common understanding of state health messages to discuss their adaptability to local contexts. For example, they communicated to the pregnant women how they could eat a nutritious diet that is locally available with them, gave examples of several such food items (and hence healthy diet need not be a platter of fruits which they cannot afford), and advised them that they could continue to work in the fields, provided they take care of their diet, monitor any warning signs and report to the health worker.

The teamwork is also important to demonstrate their performance as ‘good workers’ for the health system. Backstage preparation is undertaken to ensure that the onstage state event – the outreach session – is successfully conducted. This is because the outreach sessions provide the raw data for constituting the success of public health programmes and more importantly, the output indicators such as number of children immunised are part of the evaluation of the health workers’ performance, as explained below. These events have unequivocal sanctity marked through joint signatures of all three health workers (the proof that the events had taken place), records of names of pregnant women and children, weight measures and numbers of iron syrup bottles distributed, which give these sessions an objective, measurable character.

Our ethnographic data suggest that favourable team dynamics – such as visible cooperation, a proactive role played by each of the three workers and constructive supervision by the ANMs – often translated not merely into successful outreach sessions, but also community trust in state health services mediated by trust in health workers. Villagers recounted positive experiences of timely care (how the ASHA and the ANM coordinated to ensure that the woman in labour reached the hospital on time and was referred swiftly when complications arose). Similarly, negative experiences were also shared when such team dynamics failed. Practical strategies to establish team spirit and teamwork serve the larger purpose of being a good health worker both for the community and that of the health system, though these expectations often conflict with each other, as discussed below.

‘Tying their hands?’

Scott and Shankar (2010), in an article called ‘Tying their hands? Institutional obstacles to the success of the community health worker programme in rural North India’ show how the ASHA workers’ potential role as health activists are restricted by institutional obstacles like the rigid hierarchical health bureaucracy, and lack of flexible and creative remuneration structures. Our study findings share these concerns. While the health workers subscribe to larger notions of integration inherent in a primary health care approach, such efforts conflict with the narrow monitoring indicators used for health system performance, health system's privileging of statistical over experiential knowledge and its reliance on top-down channels of communication. This became clear during monthly supervisory meetings. Attended by the ASHAs and ANMs, such meetings are held once a month in the primary health centre, which caters to a population of 20,000 to 30,000. The medical officer presides over these meetings along with the medical supervisors. These meetings are an important forum for dissemination and monitoring of state health programmes by the senior officials in the primary health bureaucracy (the medical officers and supervisors). During these meetings, health workers submit the records of activities for the previous month gathered at the outreach sessions.

Though NRHM has the broad objective of strengthening primary health care and the rural public health system, in practice, there is a tendency among sub-district and district health officials to restrict NRHM efforts to control the infant mortality rate and the maternal mortality ratio only. This could be due to the pressure from higher up the health bureaucracy at the state and national levels to contribute to achieving the MDGs on maternal and child health and more directly to the fact that the health system performance is evaluated through indicators on maternal and child health. Thus, in all the monthly meetings, the ASHA workers and ANMs are asked to provide numerical evidence on number of pregnant women registered, offered ANC care and accompanied for institutional delivery (used as a proxy for maternal mortality), the number of malaria slides collected, the number of children immunised and the number of malnourished children referred further for treatment. This focus on indicator reporting means that such forums rarely provide any opportunity for the workers to share broader feedback from their practice. This is despite the fact that their experiences from the field would help to explain the mechanisms through which outputs are achieved (or not) and how. Health workers, for example, had insights on immunisation programme drop out in peak agricultural season, reasons for seeking treatment for epileptic fits from a traditional healer, and pregnant women being taken to a district hospital rather than the recommended primary health centres. These insights, however, never get communicated in the monthly meetings. For the health workers, such field-based knowledge and experience become ‘extras’ which need to be kept quiet or shared informally with peers, rather than evidence on whether and how NRHM is achieving its aims.

The supervisory meetings become a site for top-down communication exclusively, where senior officials either demand reports and records or make specific announcements on recent orders passed down to them from further up the health bureaucracy. Discussions in these meetings take the form of one-way communication: ‘Did immunisation take place last month? How many institutional deliveries? Any outbreak of diarrhoea? Did you collect malaria slides?’ Such top-down communication dilutes the significance of the forum of the supervisory meetings, which are an important point of encounter between top-down and bottom-up planning. The block level health system (the Primary Health Centre) thus fails to fulfil its key roles in responding to local needs and contexts in service provision as much as adapting national policy and guidelines to local circumstances. This is indeed a caricature of the NRHM's emphasis on decentralised health planning from the village level to the block, district, state and national levels. The monthly project meetings of the ICDS where the AWW submits the records (attended by the ICDS officers and supervisors along with the medical officers) followed a similar pattern.

The focus on numerical compliance, lack of scope for open communication and supervision result in demoralisation among health workers (Coutinho et al., 2000; George, 2009). These also encourage ‘gaming’ behaviour. Gaming refers to strategies to maximise performance in relation to the rewarded behaviour (Magrath & Nichter, 2012, p. 1780). In the context of village-level health workers, rewarded behaviour includes tasks that are incentivised and for which the health workers’ performance is evaluated. ‘Gaming’ strategies among the health workers in our study included falsification of evidence on identification of malnourished children for referral and the number of institutional deliveries. For instance, gaming behaviour could include recording a delivery as ‘institutional’ even when a woman delivered at home or on the way to the hospital, or recording a child as grade III malnourished even in the absence of systematic data. While conforming to the pressure of reporting practices, health workers sought to devise creative ways of dealing with individual situations – deciding to facilitate cash incentives for a woman who delivered at home on the ground that the family was poor. One of the consequences of such gaming strategies is that the health system actually has poor quality data about what is happening in the field.

Conclusion

Mainstream public health writings on delivery of integrated services tend to focus on the health services per se and modalities of their integration, assuming that effective supply chain management, infrastructure and human resources would achieve integration. These writings approach health service delivery itself as a technical and mechanistic process. Based on ethnographic evidence, this article shows how building social relations of trust and teamwork are critical to health workers’ efforts in delivering integrated services. The role of trust in health care has traditionally been examined in relation to doctor–patient relationships. However, recent anthropological literature has sought to bring social relations of trust to the centre stage in the study of health systems and policies (Gilson, 2003, 2005; Magrath & Nichter, 2012; Rowe & Calnan, 2006; Theide, 2005). Drawing on empirical evidence from a number of contexts, these studies have demonstrated that trust matters to health systems. Health workers in our study reinforced this conviction. These community health workers espouse an integrated approach to care by fostering relations of mutual trust, teamwork, cooperation, addressing community health and other needs, promoting a continuum of care from curative to preventive care and valuing the role of regular and effective communication with villagers and also amongst health workers themselves. These values are indeed the cornerstone of a primary health care ideology that promotes democracy, equity and participation (Nichter, 1986). These values are enshrined, at least rhetorically, in India's NRHM.

However the voices and experiences of health workers and other implementers are hardly taken into account and rarely thought to constitute evidence for public health policies and programmes. Anthropologists suggest an urgent need for consideration of process variables that could track how outcome variables are achieved (Magrath & Nichter, 2012). Such process evaluation needs to include inputs from different actors involved at different levels, including community health workers. We agree with Walker and Gilson (2004), who based on their study of nurses in South Africa, argue that discounting the perspectives and experiences of frontline health workers emanates from the methodological limitation of approaching policy implementation as a linear, top-down process.

The vision of comprehensive primary health care that the NRHM's health workers try to promote, and which is enshrined in policy documents, is in constant tension with other important elements of the Indian public health system: the narrow indicators used for health system performance; the highly hierarchical bureaucratic structure that rests on top-down communication and information; the institutionalised privileging of statistical evidence over field-based experiences. These features curb the potential role of community health workers as agents of social change, cultural mediators and health promotors through effective community participation. A narrow focus on achieving MDGs 4 and 5 through cash incentives, regularising outreach sessions, stricter monitoring of indicators relating to these goals might result in better immunisation coverage and higher rates of institutional delivery, but these need not guarantee a robust, sustainable health system – or indeed result in real improvement in health outcomes. A robust health system should put people, rather than simply ‘clients’ or ‘beneficiaries’ of health care interventions, and their health, not just the prevention and treatment of biomedically defined illness, at the centre stage. A recent WHO summary of African countries’ experiences with revitalisation of primary health care documents many similar challenges to fulfilling the primary health care agenda to those identified here, including reforming monitoring and evaluation systems, changing the vertical focus of health service delivery and enhancing greater community participation (WHO, 2008b).

Though the NRHM evokes the spirit of primary health care values, these are enforced through a health organisational bureaucracy that is deeply hierarchical. The NRHM is thus in danger of being looked upon as just another programme undermining the overhauling of public health ideologies in India. Bhatia and Rifkin (2010) rightly note that it is not only enough to revitalise primary health care, but also emphasise the need to reframe it in light of values of equity, community empowerment and determinants of health. This can be operationalised through an organisational culture that embodies these values.

Acknowledgements

The author is thankful to Katerini Storeng, Sidsel Roalkvam and the anonymous reviewers for detailed comments on earlier drafts.

Funding

This research has been funded by the Norwegian Research Council's Global Health and Vaccination Research programme (GLOBVAC) as part of a larger research project [grant number 196382] on Health systems strengthening within vaccination programmes: an ethnographic study.

Additional information

Funding

Funding: This research has been funded by the Norwegian Research Council's Global Health and Vaccination Research programme (GLOBVAC) as part of a larger research project [grant number 196382] on Health systems strengthening within vaccination programmes: an ethnographic study.