Scholarly article on topic 'Signalling, status and inequities in maternal healthcare use in Punjab, Pakistan'

Signalling, status and inequities in maternal healthcare use in Punjab, Pakistan Academic research paper on "Sociology"

CC BY-NC-SA
0
0
Share paper
Academic journal
Social Science & Medicine
OECD Field of science
Keywords
{Pakistan / "Maternal health services" / Caste / "Status consumption" / "Status symbol" / "Public sector healthcare" / "Private sector healthcare" / Midwife}

Abstract of research paper on Sociology, author of scientific article — Zubia Mumtaz, Adrienne Levay, Afshan Bhatti, Sarah Salway

Abstract Despite rising uptake of maternal healthcare in Pakistan, inequities persist. To-date, attempts to explain and address these differentials have focused predominantly on increasing awareness, geographic and financial accessibility. However, in a context where 70% of healthcare is private sector provided, it becomes pertinent to consider the value associated with this good. This study examined patterns of maternal healthcare use across socioeconomic groups within a rural community, and the meanings and values attached to this behaviour, to provide new insight into the causes of persistent inequity. A 10-month qualitative study was conducted in rural Punjab, Pakistan in 2010/11. Data were generated using 94 in-depth interviews, 11 focus group discussions and 134 observational sessions. Twenty-one pregnant women were followed longitudinally as case studies. The village was comprised of distinct social groups organised within a caste-based hierarchy. Complex patterns of maternal healthcare use were found, linked not only to material resources but also to the apparent social status associated with particular consumption patterns. The highest social group primarily used free public sector services; their social position ensuring receipt of acceptable care. The richer members of the middle social group used a local private midwife and actively constructed this behaviour as a symbol of wealth and status. Poorer members of this group felt pressure to use the afore-mentioned midwife despite the associated financial burden. The lowest social group lacked financial resources to use private sector services and opted instead to avoid use altogether and, in cases of complications, use public services. Han, Nunes, and Dreze's (2010) model of status consumption offers insight into these unexpected usage patterns. Privatization of healthcare within highly hierarchical societies may be susceptible to status consumption, resulting in unforeseen patterns of use and persistent inequities. To-date these influences have not been widely recognised, but they deserve greater scrutiny by researchers and policy-makers given the persistence of the private sector.

Academic research paper on topic "Signalling, status and inequities in maternal healthcare use in Punjab, Pakistan"

SOCIAL

SCIENCE

MEDICINE

Contents lists available at SciVerse ScienceDirect

Social Science & Medicine

journal homepage: www.elsevier.com/locate/socscimed

Signalling, status and inequities in maternal healthcare use in Punjab, Pakistan^

CrossMark

Zubia Mumtaz3,*, Adrienne Levaya, Afshan Bhattib, Sarah Salwayc

a School of Public Health, University of Alberta, 3-309 Edmonton Clinic Health Academy, 11405 - 87 Ave, Edmonton, AB T6G 1C9, Canada bReal Medicine Foundation Pakistan # 70, Nazimuddin Road, F-7/4 Islamabad, Pakistan

c School of Health and Related Research, University of Sheffield, Regent Court, 30 Regent Street, Sheffield, S1 4DA, UK

ARTICLE INFO

Article history:

Available online 21 June 2013

Keywords: Pakistan

Maternal health services Caste

Status consumption Status symbol Public sector healthcare Private sector healthcare Midwife

ABSTRACT

Despite rising uptake of maternal healthcare in Pakistan, inequities persist. To-date, attempts to explain and address these differentials have focused predominantly on increasing awareness, geographic and financial accessibility. However, in a context where 70% of healthcare is private sector provided, it becomes pertinent to consider the value associated with this good. This study examined patterns of maternal healthcare use across socioeconomic groups within a rural community, and the meanings and values attached to this behaviour, to provide new insight into the causes of persistent inequity. A 10-month qualitative study was conducted in rural Punjab, Pakistan in 2010/11. Data were generated using 94 in-depth interviews, 11 focus group discussions and 134 observational sessions. Twenty-one pregnant women were followed longitudinally as case studies. The village was comprised of distinct social groups organised within a caste-based hierarchy. Complex patterns of maternal healthcare use were found, linked not only to material resources but also to the apparent social status associated with particular consumption patterns. The highest social group primarily used free public sector services; their social position ensuring receipt of acceptable care. The richer members of the middle social group used a local private midwife and actively constructed this behaviour as a symbol of wealth and status. Poorer members of this group felt pressure to use the aforementioned midwife despite the associated financial burden. The lowest social group lacked financial resources to use private sector services and opted instead to avoid use altogether and, in cases of complications, use public services. Han, Nunes, and Dreze's (2010) model of status consumption offers insight into these unexpected usage patterns. Privatization of healthcare within highly hierarchical societies may be susceptible to status consumption, resulting in unforeseen patterns of use and persistent inequities. To-date these influences have not been widely recognised, but they deserve greater scrutiny by researchers and policy-makers given the persistence of the private sector.

© 2013 The Authors. Published by Elsevier Ltd. All rights reserved.

Introduction

Pakistan is one of six countries contributing to over half of all maternal deaths worldwide (Hogan et al., 2010). In its efforts to meet the targets of Millennium Development Goal 5 (MDG5), the government has focused on improving the delivery of both facility-based and domiciliary maternal healthcare by strengthening district health services' technical and managerial skills and creation of

q This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike License, which permits noncommercial use, distribution, and reproduction in any medium, provided the original author and source are credited.

* Corresponding author. Tel.: +1 780 492 7709.

E-mail addresses: zubia_mumtaz@yahoo.com, zubia.mumtaz@ualberta.ca (Z. Mumtaz), adrienne.levay@ualberta.ca (A. Levay), s.salway@sheffield.ac.uk (S. Salway).

demand. It is assumed these will lead to the provision of optimal quality of care (Government of Pakistan, 2006). Consequently, some improvement in service availability and levels of use has occurred, particularly antenatal care usage, which doubled from 32% to 61% between 1995/6 and 2006/7 (DHS, 2008).

However, as services are becoming more available, inequities in use between the rich and poor are also increasing (DHS, 1992,2008; Mahmood, 2010). The 2006/7 Demographic and Health Survey found that 92% of women in the highest wealth quintile reported seeking ANC compared to 37% of women in lowest quintile; 74% of women in the highest wealth quintile delivered in a health facility compared to just 12% of women in the lowest quintile (DHS, 2008).

Inequities between the rich and poor in maternal healthcare usage are well documented across the global south and a large body of research seeks to understand and address the underlying factors that restrict uptake of services (Amin, Shah, & Becker, 2010; Aremu,

0277-9536/$ — see front matter © 2013 The Authors. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.socscimed.2013.06.013

Lawoko, & Dalal, 2011; Barros et al., 2012; Thind, Mohani, Banerjee, & Hagigi, 2008). To-date, this work has focused predominantly on supply-side factors, particularly increased geographical accessibility and quality of services (Thind et al., 2008), and some demand-side factors, including, levels of awareness and knowledge and ability to pay (Amin et al., 2010; Aremu et al., 2011).

While this work has provided useful insight and direction to policy and practice interventions, it appears to paint a partial picture of the factors that shape and constrain maternal healthcare use. This incomplete understanding is suggested by the disappointing and unanticipated outcomes of some interventional approaches aimed at addressing these obstacles. For instance in Pakistan, a 10-district maternal healthcare intervention that focused on improving service quality and creating demand, reported an increase in facility deliveries across the districts from 62% to 74% among the highest wealth quintile but no improvement among the lowest wealth quintile (18 to 17%) (Mahmood, 2010). Furthermore, the National Maternal, Newborn, and Child Health (MNCH) program has aimed to improve service delivery in the public sector, assuming it will be used by the poor. However, national data shows that 84% of women who sought delivery care from the public sector were non-poor (Mahmood & Bashir, 2012).

This backdrop suggests the need for more careful exploration of the social, economic and cultural contexts within which maternal healthcare seeking behaviour is situated. Early anthropological work has usefully described the gender context (Winkvist & Akhtar, 2000) as well as the existing social structures (Mohmand & Ghazdar, 2007) in Pakistan in general. Further, some studies have sought to illuminate the ways in which these structures and processes shape and constrain access to healthcare, with useful recent contributions addressing women's gendered position within households, limited decision-making authority (Bhatti & Jeffery, 2012; Mumtaz & Salway, 2009), and issues of social accessibility (Mumtaz & Salway, 2005). The present paper adds to this important body of work by pursuing a line of enquiry that has to-date received little attention within the maternal healthcare literature; namely the meaning and value attached to maternal healthcare consumption.

Much of the research on maternal healthcare use in the global south assumes that use of such services is fundamentally linked to their inherent utility in relation to enhancing the health of the mother and/or unborn child. Though not all studies are explicit about their theoretical underpinnings, these approaches tend to align with psychological models of healthcare utilization that foreground the interplay of perceived threats to health and benefits of (or need for) particular health behaviours on the one hand with the obstacles (or disincentives) to such behaviour on the other hand, for instance the Health Belief Model (Rosenstock, Stretcher, & Becker, 1988) or Andersen and Newman's health-seeking model (Andersen & Newman, 2005).

Such models have been widely applied within health services research, but can be critiqued for their failure to consider the possibility that healthcare behaviours and services may be adopted - at least in part — for reasons other than their perceived health-enhancing effect.

The consumption of goods and services for reasons other than their direct utility, particularly for their symbolic meaning, has been widely documented within economic and sociological literature (Richins, 2011; Witt, 2010). Further, the links between consumption, social position and status distinctions have long been the subject of empirical observation and theoretical debate (Aydin, 2006; Bourdieu, 1984). Heterogeneity in use of goods or services is a common feature of consumption. While some suggest that this fragmentation reflects individual taste and style (Pakulsky & Waters, 1996), others argue that fragmentation appears along socio economic lines (Aydin, 2006; Manza & Brooks, 1998) and

serves to maintain social distance between hierarchically structured groups.

In general terms, the growing commodification of healthcare is widely recognized (Henderson & Petersen, 2002; Pelligrino, 1999)and its potential to encourage unnecessary or even harmful patterns of healthcare use has been documented in varied contexts. For example in Vietnam, ultrasounds in pregnancy have become a highly profitable service, with some women reporting having gone for one or more ultrasounds per month (Gammeltoft & Nguyen, 2007). Furthermore, patterns of healthcare consumption that suggest motivational factors beyond inherent health-enhancing utility have been noted in some settings. For instance, a recent commentary in the United States suggests that use of midwives by the wealthy elite in Manhattan has become a status symbol, as well as being associated with a more natural birth experience (Pergament, 2012). The potential relevance of status signalling processes to understanding patterns of maternal healthcare utilization clearly deserves greater attention.

An exploration of the patterns and meanings of maternal healthcare consumption is particularly pertinent to Pakistan. Privatization of the health sector has been aggressively pursued in Pakistan since the 1990s as part of its structural adjustment program (World Bank, 1998). Consequently, as of2005,74% of the total health expenditure was from the private sector (World Health Organization, 2013). Evidence also suggests that private sector services are perceived to be superior to public sector care (Irfan & Ijaz, 2011). In addition, Pakistan is recognized as a highly unequal society with ideologies and social processes acting in consort to perpetuate rigid gender and socio economic hierarchies (Easterly, 2001; Mumtaz, Salway, Shanner, Bhatti, & Laing, 2011). Furthermore, although there have been significant public sector investments in improving maternal healthcare services, concerns remain regarding low uptake among the poorer sections of the population (Mahmood, 2010). The present study sought to provide new insight into the factors shaping maternal healthcare utilization by examining in detail the patterns of use across socioeconomic groups within one community, and the meanings and values attached to this behaviour.

Methods

The findings presented in this paper form part of a larger study aimed at developing a detailed understanding of the influence of caste and gender hierarchies on maternal health. A ten month study (May 2010—February 2011) was conducted in Ganji village; district Chakwal, northern Punjab. Given that no one village can represent an entire country, Ganji (population 1229) was selected because it was socio economically heterogeneous. District Chakwal is relatively well-developed and can be considered representative of conditions in Northern Punjab. Land-holdings are small compared to southern Punjab. Consequently men seek work beyond the village, commonly in the military and related industry. Poverty levels are therefore somewhat lower compared to southern Punjab. A critical ethnographic approach was used (Geertz, 2000; Wainwright, 1997). Four inter-related phases of data generation were involved: (1) Familiarization and rapport building (four social mapping exercises, home visits and a demographic survey of all village households. The survey collected basic socio-demographic data of all the village residents and included education, occupation, caste, landownership and maternal health services use in the preceding five years); (2) Exploration of social norms, everyday practices and their relation to childbearing (observations and informal interviews, mapping of behaviours and decision-making processes for 18 pregnant women, in-depth interviews with 34 young women, 27 older women, 20 young men, and 13 older men). These methods also enabled us to document behaviours and decision-making processes around seeking maternal healthcare as

well as experiences of exclusion based on social and economic characteristics; (3) Detailed exploration of five maternal deaths that had occurred in the preceding five years (interviews with husbands, mothers-in-law, other women, neighbours and healthcare providers associated with the dead woman around the time of her death); and (4) Respondent validation (11 focus group discussions of 6—10 individuals each, organised along age, gender and caste lines; repeated conversations, 61 in total, with two key informants). Respondents were purposively selected on the basis of experiences that we believed would illuminate the ways in which processes of social exclusion operate and impact upon maternal health while also ensuring that we included a wide range of people varying in terms of age, gender, socio economic position and caste.

Data were collected by a team of five researchers including AB who lived in the village for 10-months. The Primary Investigator (PI), ZM also lived in the village for three months. ZM and AB have worked together for over a decade, conducting village ethnographies with the objective to map how gender and class inequities impact on women's reproductive health. As politically concerned researchers, these authors collect and analyse data through a feminist-inspired critical lens that aims to map how interests and perspectives are controlled by the powerful. Both ZM and AB speak the local language, Potohari, and are familiar with the local socio-cultural context. Research was conducted in the local language and digitally recorded if permission was granted. Rarely when permission was not given, the research team recapped the interview in field notes recorded digitally twice per day. All data (in-depth interviews, focus group discussions and observation notes) were translated and transcribed verbatim into English. The PI double-checked a random sample of transcripts for accuracy and translation.

Most of the findings presented in this manuscript are based on observation/participation because the nature of the findings, that maternal health services use is a status symbol, is not something people explicitly state. Any information that is meant to reflect the 'status' of the person is usually signalled by behaviours, body-language, indirect comments, and consumption practices. These behaviours can only be observed and a key strength of our research is that the researchers lived amongst the villagers, embedded in their lives. In this traditional ethnographic approach, observations are a valid and powerful tool.

Data analysis occurred concurrently with data collection and creation of a database of transcribed notes. The researchers immersed themselves in the data to gain intimate knowledge of the cases and reflect carefully on the meaning to be derived from them (Thorne, Kirkham, & O'Flynn-Magee, 2008). Data were open-coded and manually recategorised into domains, which were then analysed to extract themes by ZM and AB who worked closely together throughout the data analysis process. Narratives from different sources were merged to describe typical experiences and behaviours, although the atypical were also accounted for and alternate explanations of the phenomena were carefully considered. It was the exploration of unexpected and atypical findings of the larger study exploring the experiences of social exclusion and inequities in maternal healthcare use that led to development of this paper. The findings were explored for congruency by triangulation of different data sources (observations, in-depth and group interviews). Ethics clearance was obtained from the National Bioethics Committee, Pakistan and the University of Alberta, Human Ethics Research, Health Panel B. The village name and respondent identities have been changed.

Findings

The caste and class system in the village

The fundamental social order of the village was found to be based on a caste system. There were five main castes (known locally

as zaat or quam) in the village based on an occupational hierarchy and linked to purity classifications.

The Chaudhrys and Rajas, as the highest caste, were also traditionally the owners of large tracts of land. Some of them were employed in the armed forces, police, or the government indicating their access to sources of power. They were rich in, using Bourdieu's conceptualisation, social capital (Bourdieu, 1984). Not only did land ownership and government employment give them economic security, power and prestige, but they were also found to be socially constructed as 'khandaani' people, that is, people who have high moral standards, whose word can be trusted, and who never beg.

At the bottom of the zaat hierarchy were the Kammis. Traditionally landless, this caste was socially constructed to perform low-status tasks such as butchers, barbers and shoemakers. Some Kammis had a social contract with higher caste families, called a seph, through which they were expected to perform tasks all year round for small amounts of cash and in-kind remuneration. Chronic inter-generational poverty was their defining characteristic. More importantly, they were socially constructed as inferior; being portrayed as polluting, having a low level of 'zameer' (conscience, virtue and moral character), and as cowards.

The Mirzas and Miannes occupied a middling social position. They owned small tracts of land, usually owned their homes and were often educated (some more so than the Rajas and Chaudhrys). They mostly worked in the private sector, though some were employed in the army and police. They appeared to be the most agile and responsive to global market forces, with a number of them having migrated to Europe and the United States.

Our data suggested status consumption was a common practice in the village and it varied by caste. As the highest caste, the Rajas and Chaudhrys drew upon their lineage and caste to convey their status. For instance, the wife of the largest landowner dressed rather simply and wore limited gold jewellery, although the family residence was a huge 5-roomed haveli (a mansion with historical significance). The Raja and Chaudhry men who had pucci (permanent) government jobs presented their employment as a sign of their humility, that they work although they do not need to given their land ownership. At the other end, the Kammis, invisibilised and demeaned by virtue of their membership in the lowest caste were too poor to engage in status consumption. Nonetheless, if a Kammi family's economic status improved, they made a visible attempt to demonstrate their new-found wealth. For example, a couple of lower caste Kammi families who had done well economically (through remunerations received from sons in the Middle East) were conspicuously well-dressed with lots of gold jewellery that was worn even during a family funeral, a highly inappropriate practice. It was, however, the middle caste Mirzas and Miannes who had the most need to demonstrate status. They were aware of their middling caste, but had the economic resources to attempt to emulate the higher caste Rajas and Chaudhry's. The richer Mirzas in particular made every attempt to demonstrate status consumption. For example, one Mirza women gave a bhans (an expensive oxen, worth about Rs 120,000 or US$ 1500) as a gift to her grandson. Although traditionally natal grandparents are expected to give gifts to the bride or groom, this gift was, by most standards, expensive.

Maternal health services in the village

A range of maternal healthcare services were available in this community. The public healthcare sector consisted of two fully functional ('sarkari ') Rural Health Centres (RHC), about 15 min drive on either side of the village. As part of MDG5 initiatives, the United Nations Fund for Population Activities, in partnership with the Punjab Health Department, had staffed the RHCs with resident female physicians and ensured availability of reasonably functional

labour rooms. The Basic Health Unit (BHU), a first-level care facility about 10 minutes drive away, had a female physician available twice a week and a labour room that was accessible during morning working hours. Emergency obstetric services were available in a district hospital about a one hour drive away and in three large teaching hospitals, four hours drive away. These public sector facilities were supposed to be free of cost, but patients had to pay for transport and food, and also buy the necessary medicines and surgical supplies. The total cost of a scheduled C-section in the district hospital during the fieldwork period was approximately Rs 7000—10,000 (75—100 USD). In addition, there were two military hospitals for military personnel and their families within a 90 min drive. Some of the villagers were employed in large public sector organizations, such as the Oil and Gas Development Corporation and had access to industry-specific health facilities.

The private sector consisted of a trained midwife, a few dais (traditional birth attendants) and a number of physicians, most of whom were within an hour's drive of the village. The private services of the midwife and physicians consisted of small two to three bed facilities, a labour room, some with a small operating theatre equipped for C-sections. The midwife referred complicated cases to a private physician with operating facilities. The physicians referred their more complicated cases to the public sector hospitals, the exact facility depending on the severity of the complication. A normal vaginal delivery attended by the trained midwife cost Rs 2200 (23 USD) and by a physician Rs 5500 (56 USD), excluding medicines or other costs. A C-sec-tion by a private physician cost about Rs 25,000 (255 USD). These rates are high given the average monthly cash earnings were around Rs 5000. The dais had small practices in terms of numbers of deliveries conducted per month. They catered primarily to the ultra-poor Kammi women conducting home deliveries.

Patterns of maternal healthcare use

We found that while a majority of village women used biomedical antenatal care and childbirth services, there was vast heterogeneity in the amount of healthcare sought. For instance, some women sought antenatal care in excess of the World Health Organisation's recommended five visits. A number of pregnant women had monthly ultrasounds and discussed with us if this frequency was harmful.

We observed Zarina (a Mirza woman) seek three antenatal visits in two days from three different providers, one of whom was a fee-paying private physician. She discarded the iron supplements provided free of cost from the mobile antenatal clinic (provided by the United Nations) and then purchased the same product from the village pharmacist. [Field notes, 22nd May 2010)

Nasina (a Mirza woman) was observed accessing three different providers during her pregnancy, one a free public sector physician, one a private-sector gynaecologist, and a private midwife. [Field notes 29 June, 2nd, 7th, 10th and 21st July, and 10th August 2010]

In contrast to this rather excessive use, other women had very limited or no access to maternal healthcare services.

Zainab (a Kammi woman), after suffering physical abuse by her in-laws, gave birth to twins. She bled to death during childbirth. A minimally trained Lady Health Worker was consulted during childbirth but no one was called to help postpartum as she hemorrhaged. Zainab's marital family did not seek the required care because they did not have the financial resources to do so, nor did they think she was worth spending what little they had.

Zainab's husband, who is primarily responsible for her care, was unemployed and her natal family was very poor and could not provide the necessary resources nor advocate on her behalf. [Field notes 23rd, 26th, 29th, 30th May, 15th June, 6th, 7th, 16th July, 17th August 2010]]

Salma (a Kammi woman), pregnant with her fifth child, had uncontrolled hypertension. Her husband could only spare Rs 200 (2.4 USD) for the childbirth and Salma was willing to deliver alone at home rather go to a public sector facility. So the research team took her to the midwife and bore the costs of the care. [Field notes: 8th, 17th, 19th, 24th August, 2011]

This heterogeneous use of services was clearly fragmented along socio-economic and caste lines. However, the patterns of use were found to be complex, as described further below.

The majority of the higher caste Raja and Chaudhry women were found to use the public sector services, which are intended to be free or low cost. Overall, the socio-demographic survey found that 36% of the 112 births in the village in the preceding 5 years took place in these public sector facilities. Fifty-five percent (17/31) of these users were higher caste Raja and Chaudhry women. Much of this public sector use was related to entitlements to care in military hospitals, through Raja/Chaudhry menfolk's work-related benefits. However, an unexpected finding was that even those Raja and Chaudhry women without institutional medical benefits opted to use the public sector RHCs and the district hospital services. These high caste women appeared to enjoy a kind of special access to these services by virtue of their social capital; referred to as 'ponch' (special access) by lower caste women. Raja and Chaudhry women who used the public services considered the care received to be of good quality, while lower caste women had vastly different opinions and experiences to report of the same services (as described in more detail below).

Sajida, the daughter-in-law of the largest and richest landlord of the village sought antenatal care from the nearby Rural Health Center since a relative was the Lady Health Worker there and she ensured that Sajida had privileged access to the physician. She was later referred to the district hospital for a possible C-section.. [Field notes 15th, 16th May, 4th, 14th, 15th June, 30thJuly, 30th Sep, 2010]

In contrast to the high castes, the Mirzas and Miannes, who had weaker social and material resources, were found to generally use private sector maternal healthcare services. Sixty seven percent of all births in these groups in the last five years occurred in a private sector facility, with a further 5% paying for the services of a dai. Their preferred provider was one particular midwife, with 62% of all private-sector births being attended by this one provider. The young women rated this midwife's quality of care highly although it did not always meet best-practice guidelines. Aspects of her delivery care that were considered to be favourable included: hastening labour with injections, discharging patients within an hour of birth and delivering on a charpoy (traditional bed) instead of the high and narrow delivery table. The older women were happy with her care too because she did not refer women for C-sections as frequently as the physicians did, a practice that meant a lower chance of high delivery costs. Although the women's narratives suggested, and our observations corroborated, that this midwife had a high threshold for risk, much higher than the physicians, the women perceived it as a helpful practice.

Turning now to the low caste Kammis, most of the women in this social group reported actively avoiding use of public sector services, the option one might assume to be the most appropriate given their poor financial position. A key reason given for this non-use was the poor treatment they received in these facilities.

'This is how it happens, Shimraz's son was born in Sukhru (the RHC). Ask her what happened with her; how they dealt with her. They made her suffer. Like slaughtering a goat. This is how it happened. It is good for us poor people to use government hospitals, but government hospitals are butchers.' (Kammi woman, mother of five, FGD)

Our analysis of their childbirth histories showed that some Kammi women had in fact used the public sector services (29% of births in the past 5 years). However, in 57% of cases the women had not used any biomedical care, either during the antenatal period or at the time of delivery. In most cases, a relative or a dai had attended their childbirth at home.

Abda, a Kammi woman pregnant with a third child decided to seek antenatal carefor her first time in her life. She requested the research team to accompany her to reduce the possibility of mistreatment. The physician and the staff at the RHC recognised Abda's social status, as reflected in her and her mother-in-law's dress and despite the researcher's presence, completely ignored her, focusing instead on purchasing eggs and jewellery from hawkers. Abda, eight months pregnant, stood in one corner for two hours before the physician examined her. [Field notes: 11th, 27th November 2010]

Status signalling, conspicuous consumption and use of maternal health services

The discussion so far has focused on the patterns of care across social groups and the perceived advantages associated with their choice of provider as expressed by women and their family members. However, though financial resources and perceptions of quality of care were important factors shaping maternal healthcare choices, they were not the whole story. Indeed, a key theme that emerged from our data was that maternal healthcare use in our field site — in particular the choice of childbirth attendant — was a status symbol for some groups. Drawing on Han, Nunes, and Dreze's (2010) model of wealth and status, we turn now to consider the heterogeneity in type and amount of service use across social groups through a status consumption lens.

Drawing on earlier classic works by Veblen (1899) and Bourdieu (1984) Han et al.'s (2010) model divides consumers into four groups or classes on the basis of their ability to pay and their need to signal their social status (see Fig. 1). Patricians possess significant wealth along with strong social and cultural capital. They are primarily concerned with signalling horizontally to other Patricians rather than needing to disassociate themselves from the other three lower social groups, and this tends to involve subtle signals and inconspicuous consumption. The Parvenus (from the Latin word 'to arrive' or 'reach') possess resources, but also have a high need to demonstrate their social status. The Poseur (from the French word meaning 'a person who pretends to be what he or she is not') is highly motivated to consume for the sake of status even if they cannot afford it ("pecuniary emulation"). The Proletarian — a term commonly used to identify those from a lower social or economic class — are not motivated to consume for the sake of status or simply cannot afford to consume.

While developed to explain patterns of branding and consumption of luxury goods in the Western context, this simple model appears to have some utility in explaining maternal healthcare consumption in our rural, Pakistani village.

Considering first the Mirzas and Miannes. As noted above, these families occupied a middle rank position within the caste hierarchy. Our data revealed that the consumption of private maternal healthcare had taken on the value of a status symbol among this social group within the village. The majority of the Mirzas and Miannes had sought both antenatal and delivery care from the

Hans Model of Wealth and Need for Status (2010)

Have-nots

Fig. 1. An adaptation of Han et al.'s (2010) model, taxonomy based on the two dimensions of wealth and need for status to understand status consumption.

popular private midwife. Conversations with both young and old women and our observations suggested that seeking care from this midwife was constructed as symbol of status. In fact, so important was it to been seen seeking care from this provider that her care was sought even for potentially complicated cases. For example, Farzana had pregnancy-induced hypertension (180/150 mm hg) and Aliya was severely anaemic. Both were diagnosed by a physician during a free mobile antenatal clinic and advised to deliver under the attendance of a physician, but they chose to use the midwife's services. The status-symbolism operated both at the individual and family level. At the individual level, the young pregnant woman receiving care from the midwife — which implied family financial outlays on her behalf — constructed this as a reflection of her value and position within her marital home. At the family level, decision-makers who provided this private midwife's care for their daughters-in-law constructed this act as reflective of their household's status amongst their extended family (biradari), and village society more generally. In contrast, the use of public maternal healthcare services was shunned by this social group and actively constructed as inferior and a marker of low social status and impoverishment.

Am I a lawaris (left on the roadside) that you will take me to a 'sarkari hospital' (a public sector hospital) Naila, 28 years old Mirza woman, primigravida in response to her mother-in-law's suggestion that she use the RHC for childbirth (in-depth interview).

Kauser Bibi wanted Sadia (Mirza women), her daughter-in-law to deliver in the public RHC since this facility had improved greatly in the past year with the presence of a female physician available at all times. But Sadia's mother could not bear the shame of it. 'Is our daughter a yateem (an orphan) that you want her to deliver in Sukhru (the RHC)'. [Field notes: 24th, 25th, 30th May 2010, 4th June 2010]

This construction of the private midwife as a status symbol aligns with Han et al.'s (2010) model, with the Mirzas and Miannes being akin to the parvenus and poseurs; two groups that work hard to confer status upon themselves. Of note is the fact that this status symbolism went to a less expensive provider as opposed to the more expensive private physicians. This again aligns with the Han et al. (2010) model whereby the poseurs in particular purchased less expensive, but 'louder' items from the same product group within a particular luxury brand. These poseurs, in an attempt to

Patrician (Rajas & Chaudhrys) Parvenu (richer Mirzas)

Proletarian (Kammis) Poseur (Poorer Mirzas & Miannes)

Low High

Need for status

confer status through consumption of a luxury item they could not afford, turned to fake copies of these luxury products.

Kishwer Bib, a poor Mirza woman, could not afford the midwife's care for her daughter-in-law, but was forced to do so for not complying with the norm would lead to a loss of status. Over several conversations, it emerged that she was saving for the childbirth using the money she received from the Benazir Income Support Program, a cash transfer program for poor women. We later learnt from her daughter-in-law's mother that they were helping Kishwer Bibi also draw upon a local food bank secretly because this is a stigmatizing practice. [Field notes date 24th, 25th May 2010, 4th, 29th June, 10th, 16thJuly, 9th, 12th August, 4th September2010,22ndJanuary 2011].

Regardless of all the above complexities, it was clear that older women in the village (both from these middle castes as well as the other groups) generally remained unconvinced about the utility of routine biomedical antenatal or delivery care. The narratives of these older women, who were ostensibly responsible for the care of their pregnant daughters-in-law, indicated that they viewed pregnancy and childbirth as normal phenomena and that they considered seeking biomedical care as a behayii (brazen) fashion. Therefore, rather than being motivated by a strong recognition of its inherent value, decisions to purchase private maternal healthcare were prompted by the perceived status benefits of complying with this current norm and the potential loss of status that would occur should they not. Mothers-in-law who did not provide this maternal healthcare for their daughters-in-law feared being called kanjoos (miserly) and phohar (one without foresight, incompetent). As Rafia, a mother-in-law said 'zamana dada ye' (the world is very harsh, referring to the pressure they were under to provide care for their daughters-in-law). When asked in a focus group discussion why the midwife was the preferred provider for many, one respondent answered sarcastically,' to give money'; clearly alluding to the fact that it was the demonstration of the money spent rather than the inherent value of the midwife's services that was the key reason for her use.

The Rajas and Chaudhrys can be considered analogous to the Patricians in the Han et al. (2010) model, being strong in both material and social resources, and having a low need to signal their status. In common with the model, our fieldwork found that the Raja and Chaudhry families did not tend to engage in conspicuous consumption and expressed their social position through more subtle cues. The type of maternal healthcare provider was not constructed as a status symbol in this group and they predominantly opted to use the public sector services. Use of these free maternal health services was also a logical choice given that an acceptable standard of care was assured either through entitlement to military services or through their status-related 'special access' (ponch) within the standard government facilities. Interestingly, older women in these groups also remained unconvinced of the inherent need for routine antenatal and delivery care and were found to raise objections if even a small financial cost was incurred.

Asma, the daughter-in-law of the largest landlord was referred to the District hospital for a possible C-section. To avoid a C-section and associated costs ofa hospital care (medicines, blood, transport, food), the mother and sister-in-law instead started shopping for a provider willing to deliver Asma normally. However, Asma's elder brother-in-law finally stepped in and made the decision to take her to the district hospital. The baby was ultimately born vaginally but the total cost of the transport, food and other incidentals came to about Rs. 7000 (90 USD). This rather small amount, given the family's wealth, was an amount that made the mother- and sisters-in-law very unhappy since they considered it a fuzuul (unnecessary) expense. [Field notes: 16th, 15th, 30th June, 30th July, 26th Sep, 2011].

Finally, the Kammis are analogous to the Proletarians in the Han et al. (2010) model. This group did not possess the material resources to engage in conspicuous consumption. Although they knew of the availability of biomedical care, most did not seek any care because they expressed fear of mistreatment within the public sector (although some did seek care at the government centres in cases of obstetric complications) and could not afford the private sector. Unlike the other three groups discussed above, use of maternal health services had not yet become a normative practice among the Kammi women. Such care was deemed unnecessary and they did not face any peer pressure. In fact when Abda decided to seek antenatal care from the nearby RHC, her mother-in-law told us that they had sought and received permission from the 'jinns' (spirits) to do so.

Discussion

The purpose of this paper was to examine the value and meaning attached to maternal healthcare use within an increasingly privatized healthcare system, and to raise the possibility that consumers' decisions may be influenced by factors other than their perceptions of the health benefits of such services.

Our data revealed complex patterns of maternal healthcare consumption across social groups within our rural, Punjabi setting. The highest social group — wealthy high caste families — who were rich in terms of both economic and social capital were most likely to use public sector services. The middling social groups were found to primarily pay for private sector services. Meanwhile, the lowest social groups who were financially and socially excluded tended not to use any maternal healthcare services, despite the physical proximity of government facilities.

Our analysis revealed that anticipated benefits from different providers were an important element in decisions about the type and quantity of maternal healthcare sought across the groups. Thus, the highest castes' choice of public sector services was strongly influenced by the value they placed on avoiding financial outlays for this type of care and the expectation they had of respectful and competent care. In contrast, the lowest castes anticipated very poor treatment within the public sector and prohibitive costs within the private sector. Coupled with a widespread perception that routine maternal healthcare had no inherent health-promoting value and was therefore unnecessary, these women commonly consumed no maternal healthcare at all. Meanwhile, the middling castes highly valued many aspects of care provided by their favoured midwife, several of which could be seen to be linked not only to ensuring the woman's comfort but was also affordable.

However, this was not the whole story. Rather, the use of private maternal healthcare by the Mirzas and Miannes was clearly also an act of conspicuous consumption, intended to enhance their social status. A number of themes in our data give support to this claim. First, the meaning attached to consuming private maternal health services, expressed both explicitly and implicitly by a range of actors, went beyond any recognised health-promoting utility. Indeed, many older women in these groups continued to verbalise their scepticism of the inherent value of such care to the mother or child. Instead, there was a prominent discourse centred on such consumption reflecting: the younger woman's worth and position within her marital home; the household decision-makers' competence and caring nature; and the family being financially robust. Consuming public sector services was, in contrast, constructed as an indication of incompetence, inferiority and impoverishment. Second, we found that this consumption was actively displayed, women both young and old, emphasising their use of the midwives services proudly. Third, we identified several cases where poorer Miannes and Mirzas reported feeling pushed into consuming

private maternal health services despite their limited financial means. Using the services of the midwife appeared to have become the expected norm for this group. These findings are akin to economic literature that shows how the consumption of particular brands creates identities, particularly those associated with a sense of achievement and social position (Mead, Baumeister, Stillman, Rawn, & Vohs, 2011; O'Cass & McEwen, 2006).

It is unclear from our available data why and how maternal healthcare consumption should have become a vehicle for expressing status within this context. Indeed, the fieldwork revealed several other common patterns of conspicuous consumption that were doing similar status work' for these groups, including clothing and jewellery. However, it seems likely that the increasingly privatized nature of healthcare within Pakistan has played a role here. Earlier work in Pakistan suggests that a large proportion of people view private sector care as superior to that offered by the public sector (Irfan & Ijaz, 2011), and the findings presented above show that, aside from the highest social groups, village residents expected (and had received) poor treatment and care within government facilities. It seems plausible that the combination of an increasingly commoditised healthcare sector, and a significant group of people who can both afford to pay for private maternal healthcare and are uncertain of receiving decent care within the public sector, has contributed to the construction of this behaviour as a status symbol for these groups.

Whether the patterns of maternal healthcare use illustrated in our village are repeated elsewhere requires further exploration. However, our findings tally with national survey work showing that 82% of users of public sector health services were non poor' (Mahmood & Bashir, 2012). Optimistically, the status consumption of maternal health services could be considered a positive force in that young women and their families are starting to use biomedical care. At the same time, however, status consumption can have a detrimental effect on people as it adversely forces them to use a type of care that they financially cannot afford when they could have received care from a cheaper source. Furthermore, the substandard care on offer to middle and lower social groups within the public sector remains unchallenged.

These findings have important policy implications, in particular for the current trend towards privatization of maternal health services. Future policies and program designs need to critically examine the consequences of privatization of services, both for meanings that become attached to a service that is essentially a public good and for equity.

The findings presented here have several limitations. First, the study was based in one village and as such there must be caution in generalizing the findings to other settings within or beyond Pakistan. Furthermore, there are aspects of the complex interplay of social, gender and economic factors that we were unable to completely unpack with our available data. In particular, our data suggest that the value and meanings attached to maternal healthcare use were dynamic and contested, varying along gender and generational lines within caste groups, as well as between castes. It seems likely that shifting relations of power in the context of social and economic changes and increasing western influence may be instrumental in the emerging patterns of maternal healthcare use, but these require further elucidation. The paper has provided an alternate and novel framework for understanding inequities in maternal healthcare use.

Conclusion

Stubborn inequities in maternal healthcare use in Pakistan are undermining progress towards achievement of the MDG5. The

current paper has borrowed from the discipline of economics to identify an important avenue for investigation, namely the values and symbolic meanings attached to the consumption of maternal healthcare. Our findings suggest that when healthcare is commo-ditised via the private sector within highly hierarchical societies, uptake may be susceptible to status consumption, resulting in unforeseen patterns of use and persistent inequities. These influences have not been widely recognized, but deserve greater scrutiny by researchers and policy makers given that expansion of the private sector is likely to continue.

Acknowledgements

The authors would like to thank the women and men of village of Ghanji, district Chakwal, Pakistan for their generosity of time and for welcoming us into their homes for 10 months. We also thank our Research Associates Ayesha Farooq, Nighat Yasmeen and Manzhar Iqbal. The study was funded by the Canadian Institution of Health Research Grant no 21376-CIHR. Zubia Mumtaz is currently funded by the Alberta Innovates Health Solutions through its Alberta Heritage Foundation for Medical Research Population Health Investigator Awards.

References

Amin, R., Shah, N. M., & Becker, S. (2010). Socioeconomic factors differentiating maternal and child health-seeking behavior in rural Bangladesh: a cross-sectional analysis. International Journal for Equity in Health, 9, 9—19. Andersen, R., & Newman, J. F. (2005). Societal and individual determinants of medical care utilization in the United States. Milbank Memorial Quarterly, 83(4). Aremu, O., Lawoko, S., & Dalal, K. (2011). Neighborhood socioeconomic disadvantage, individual wealth status and patterns of delivery care utilization in Nigeria: a multilevel discrete choice analysis. International Journal of Women's Health, 3(1), 167—174. Aydin, K. (2006). Social stratification and consumption patterns in Turkey. Social

Indicators Research, 75(3), 463—50 . Barros, A. J., Ronsmans, C., Axelson, H., Loaiza, E., Bertoldi, A. D., Franca, G. V. A., et al. (2012). Equity in maternal, newborn, and child health interventions in countdown to 2015: a retrospective review of survey data from 54 countries. The Lancet, 379, 1225—1233. Bhatti, F., & Jeffery, R. (2012). Girls' schooling and transition to marriage and motherhood: exploring the pathways to young women's reproductive agency in Pakistan. Comparative Education, 48(2), 149—166. Bourdieu, P. (1984). Distinction: A social critique of the judgment of taste (Richard

Nice, Trans.). Cambridge, MA: Harvard University Press. Easterly, W. (2001). The political economy of growth without development: A case study of Pakistan. Development Research Group, World Bank. Retrieved from http://williameasterly.files.wordpress.com/2010/08/9_-easterly_thepolitical economyofgrowthwithoutdevelopmentpakistan.pdf. Gammeltoft, T., & Nguyln, H. T. T. (2007). The commodification of obstetric ultrasound scanning in Hanoi, Viet Nam. Reproductive Health Matters, 15(29), 163—171.

Geertz, C. (2000). Available light. Anthropological reflections on philosophical topics.

Princeton: Princeton University Press. Government of Pakistan. (2006). PC-1 national maternal, newborn and child health

program (MNCH). Islamabad: Federal Ministry of Health. Han, Y. J., Nunes, J. C., & Dreze, X. (2010). Signaling status with luxury goods: the

roles of brand prominence. Journal of Marketing, 47(July), 15—30. Henderson, S., & Petersen, A. (2002). Consuming health: The commodification of

health care. New York: Routledge. Hogan, M., Foreman, K., Naghavi, M., Wang, M., Makela, S., Lopez, A. D., et al. (2010). Maternal mortality for 181 countries, 1980—2008. The Lancet, 375(9726), 1609—1623. Irfan, S. M., & Ijaz, A. (2011). Comparison of service quality between private and public hospitals: empirical evidences from Pakistan. Journal of Quality and Technology Management, 7(1), 1—22. Mahmood, A. Improving maternal and neonatal health: measuring the impact of the PAIMAN project in ten districts in Pakistan. Comparing baseline and endline survey findings (2005—2010). Presented at the PAIMAN end-of-Project national Dissemination meeting, population Council, Islamabad, November 11, 2010. Mahmood, N., & Bashir, S. (2012). Applying an equity lens to maternal health care practices in Pakistan. PIDE Working papers 2012:83. Islamabad, Pakistan: Pakistan Institute of Development Economics. Manza, J., & Brooks, C. (1998). Social cleavages and political change. Oxford: Oxford University Press.

Mead, N. L., Baumeister, R. F., Stillman, T. F., Rawn, C. D., & Vohs, K. D. (2011). Social exclusion causes people to spend and consume strategically in the service of

affiliation. Journal of Consumer Research, 37(5), 902—919. http://dx.doi.org/ 10.1086/656667.

Mohmand, S., & Ghazdar, H. (2007). Social structures in rural Pakistan (No. TA 4319-PAK). Islamabad: Asian Development Bank, Pakistan Resident Mission. Retrieved from http://www2.adb.org/Documents/Reports/Consultant/37711-PAK/Social-Structures-Rural-Pak.pdf.

Mumtaz, Z., & Salway, S. (2005). "I never go anywhere": extracating the links between women's mobility and uptake of reproductive health services in Pakistan. Social Science & Medicine, 60(7), 1767—1779.

Mumtaz, Z., & Salway, S. (2009). Understanding gendered influences on women's reproductive health in Pakistan: moving beyond the autonomy paradigm. Social Science & Medicine, 68,1349—1356.

Mumtaz, Z., Salway, S., Shanner, L., Bhatti, A., & Laing, L. (2011). Maternal deaths in Pakistan: intersection of gender, caste, and social exclusion. BMC International Health and Human Rights, 11(Suppl. 2), S4. http://dx.doi.org/10.1186/1472-698X-11-S2-S4.

DHS, National Institute of Population Studies (NIPS), & IRD/Macro International Inc.. (1992). Pakistan demographic and health survey 1990/1991. Islamabad, Pakistan: DHS.

National Institute of Population (NIPS) [Pakistan] Studies, & Macro International Inc. (2008). Pakistan demographic and health survey 2006—07. Islamabad, Pakistan: National Institute of Population Studies and Macro International Inc.

O'Cass, A., & McEwen, H. (2006). Exploring consumer status and conspicuous consumption. Journal of Consumer Behaviour, 4(1), 25—39.

Pakulsky, J., & Waters, M. (1996). Death of class. London: Sage Publications.

Pelligrino, E. D. (1999). The commodification of medical and health care: the moral consequences of a paradigm shift from a professional to a market ethic. Journal of Medical Philosophy, 24(3), 243—266.

Pergament, D. (2012, June 15). The midwife becomes a status symbol for the Hip — NYTimes.com. New York Times — Fashion & Style. Retrieved October 30, 2012,

from http://www.nytimes.com/2012/06/17/fashion/the-midwife-becomes-a-status-symbol-for-the-hip.html?_r=0.

Richins, M. L. (2011). Materialism, transformation expectations, and spending: implications for credit use. Journal of Public Policy & Marketing, 30(2), 141-156.

Rosenstock, I. M., Stretcher, V. J., & Becker, M. H. (1988). Social learning theory and the health belief model. Health Education & Behaviour, 15(2), 175-183.

Thind, A., Mohani, A., Banerjee, K., & Hagigi, F. (2008). Where to deliver? Analysis of choice of delivery location from a national survey in India. BMC Public Health, 8. Retrieved from http://www.scopus.com/inward/record.url?eid=2-s2.0-39749155664&partnerlD=40&md5=7dcb4dfa12e1ba527d05fda4ecd99ead.

Thorne, S., Kirkham, S., & O'Flynn-Magee, K. (2008). The analytic challenge in interpretive description. International Journal of Qualitative Methods, 3(1), 1-11.

Veblen, T. (1899). The theory of the leisure class. New York: Penguin.

Wainwright, D. (1997). Can sociological research be qualitative, critical and valid? The Qualitative Report, 3(2). Retrieved from http://www.nova.edu/ssss/QR/ QR3-2/wain.html.

Winkvist, A., & Akhtar, H. Z. (2000). God should give daughters to rich families only: attitudes towards childbearing among low-income women in Punjab, Pakistan. Social Science & Medicine, 51(1), 73-81. http://dx.doi.org/10.1016/S0277-9536(99)00440-2.

Witt, U. (2010). Symbolic consumption and the social construction of brand characteristics. Structural Change and Economic Dynamics, 21,17-25.

World Bank. (1998). Pakistan towards a health sector Strategy: Document of the World Bank. Washington: World Bank.

World Health Organization. (2013). Global health expenditure database. Table of key indicators, sources and methods by country indicators: Pakistan-national expenditure on health. Retrieved January 24, 2013, from http://apps.who.int/nha/ database/StandardReport.aspx?lD=REP_WEB_MlNl_TEMPLATE_WEB_VERSlON &C0UNTRYKEY=8470 .