Our findings suggest that women and their families made informed choices about care-seeking when health problems arose during pregnancy. The pattern of reported symptoms was plausible in terms of their nature and frequency. Symptoms suggesting serious morbidity were reported in fewer than 15% of cases. Institutional care-seeking was high across the board (> 80%) and slightly higher for trigger symptoms (>88%). Delays were uncommon, and more rapid consultation and treatment was described for trigger symptoms.
The results of the regression analyses were internally consistent. The strongest determinant of site of care was residential location, followed by socio-economic status. The findings for location are consistent with our knowledge of service distribution. In general, where public-sector tertiary hospitals were easily accessible, they were used preferentially. Use of private hospitals was lowest in F/North ward, the site of a tertiary public hospital with a good reputation, and highest in K/West, the wealthiest of the wards. Use of a general hospital was highest in H/East, which has a reputable municipal hospital. Public-sector maternity homes were used least in H/East (where there are none), and most in K/West, the site of a large maternity home generally perceived to offer quality care. Single-handed private practitioners were consulted across the board, but were perhaps more likely to be chosen for symptoms perceived as less pregnancy-related, such as diarrhoea and vomiting. Overall, the wealthier her family, the more likely a pregnant woman was to seek care at a private hospital.
The strengths of the study were that it was community-based and involved a sample of more than 10,000 women recruited over 2 years. Limitations included potential recall bias associated with the self-reporting of illness. The accuracy of diagnoses could not be verified and our classification of symptoms into common and trigger categories were based on respondents’ own descriptions.
Compared to antenatal morbidity, reproductive health problems among women in India's urban slums are common (Garg et al. 2002
, Bhanderi and Kannan 2010
), as are other conditions such as anaemia (Mayank et al. 2001
). Symptoms that are unrecognised, not thought to be serious or considered normal may lead to underreporting and limited care-seeking. Taboos also render some reproductive health problems invisible and women are often expected to endure them (Garg et al. 2002
). Responses to problems during pregnancy may differ because of a perceived risk to the unborn child, and this may help to explain the higher levels of care-seeking. For example, one study showed that women in a Delhi slum generally sought care for obstetric morbidity even though their ability to recognise symptoms of serious complications was poor (Mayank et al. 2001
Several studies affirm the urban preference for private-sector care (Aljunid 1995
, Gupta and Dasgupta 2000
, Bhatia and Cleland 2001
). Among the reasons for this are ease of accessibility, convenient opening hours and a perception that the quality of care is higher (Bennett 1996
, Barua 2005
, Habtom and Ruys 2007
). Although the use of private facilities is limited by the ability to pay (Shah More et al. 2009a
), the willingness to meet the costs may be explained by clients’ expectations that they will receive superior service and more courteous treatment (The World Bank 1996
, De Zoysa et al. 1998
, Kausar et al. 1999
, Gupta and Dasgupta 2000
). Our previous research showed an association between routine private-sector maternity care and rising socio-economic status (Shah More et al. 2009b
). In the present study, the least poor chose private hospitals for morbidity care over other types of facility, suggesting that financial constraints do play a part in influencing the choice of provider (Kausar et al. 1999
, Mahal et al. 2001
It is estimated that there are well over 1.25 million unqualified practitioners in India (Radwan 2005
). We do not know the qualifications of the private practitioners visited by women in our study. However, research across India suggests that most have minimal training, and that individuals trained in non-biomedical disciplines often practice allopathy (De Zoysa et al. 1998
, Barua 2005
, Duggal and Gangolli 2005
). Importantly, clients may not distinguish between qualified and unqualified practitioners (De Zoysa et al. 1998
Despite the negative publicity that government health facilities have received (Gupta and Dasgupta 2000
, Barua 2005
, De Costa and Diwan 2007
), our findings show that the public sector is still an important health-care provider. Utilisation of municipal hospitals and maternity homes was higher in areas where they were more easily accessible and provided the necessary level of care. Similar research in another low-income area of Mumbai reported that choice of health-care provider was mediated by accessibility, affordability and quality of services, and that a shortfall in the availability of public facilities left some with no option other than to seek private care (Dilip and Duggal 2004
). The demand for health-care services for Mumbai's expanding population exceeds the supply of public-health infrastructure — a fact acknowledged by the city's municipal administration (MCGM 2005
) — and it is plausible that a corresponding increase in the number of peripheral public facilities would result in greater utilisation (Dilip and Duggal 2004
). Despite their easy access, we think that the underutilisation of community-based primary health facilities is largely attributable to their limited services and personnel, poor perceptions of quality and the tendency to seek antenatal and delivery care with the same provider. Conversely, the greater use of larger municipal hospitals exacerbates the problems of crowded outpatient departments, queuing, shorter consultation times and the loss of time that results from travelling further from home. These very factors dissuade people from seeking care in a sector that exists to serve them.
When faced with a health problem during pregnancy, women did not seem to face major barriers to accessing care. Rather than waiting for their next antenatal consultation, most consulted a health provider and received treatment within 2 days. Few were referred to another provider. While this may be due to a poor referral system (Barua 2005
), or women's reluctance to accept referrals (De Zoysa et al. 1998
), we propose that most families’ care-seeking choices are based on rational decisions about health problems in pregnancy, and that they can usually access the necessary resources to seek care (Matthews et al. 2005
). Furthermore, they are able to navigate a complex health system that comprises multiple sources and types of medical care.
Our findings suggest that the urban poor recognise symptoms of obstetric complications and understand the need for health care. This is a good thing, particularly in the event of serious illness. However, more than 80% of women sought care for non-life-threatening conditions such as tiredness and backache. We are not in a position to judge this, but it seems likely that it reflects a broader picture of care-seeking for all illnesses. We think that the propensity to choose institutional care over self-treatment may reflect acculturation to life in a megacity such as Mumbai, a process which might be termed ‘modernisation through migration’ (Basu 1990
). An important question arising from the study is whether the proliferation of private health-care providers in poor urban areas contributes to the medicalisation of pregnancy. We plan to explore this possibility in future qualitative research with private providers.
That women in Mumbai's urban slums are able to choose from a wide range of health-care providers and that utilisation is high is encouraging. However, access to public facilities is uneven and whether high levels of care-seeking result in better health outcomes are matters for debate (Das et al. 2008
). Seeking care for minor illnesses that could successfully be treated without recourse to a health practitioner places additional financial and social burdens on the poorest and can delay treatment of those with more urgent conditions. Conversely, the main reasons for not seeking care suggest that at least some women do not recognise the danger signs of problems during pregnancy, or do not have sufficient mobility or social support to enable them to visit a health provider. In this respect, we are currently conducting a study to quantify women's agency and its effect on care-seeking behaviour, and we have been working with women's groups to improve maternal and newborn health practices and encourage appropriate health-care seeking (Shah More et al. 2008
). The important role that private providers play in the provision of health care for the urban poor needs greater recognition, and ways in which the private and public sectors might collaborate merit investigation. Further research is needed on provider activities and behaviour to facilitate effective regulation and improve quality of care. How and why expectant mothers and their families make their care-seeking choices in urban slums, where multiple providers and levels of care coexist, also require more detailed investigation.