Content area
Abstract
This dissertation explores several aspects of the integration of traditional and biomedical maternity care in Guatemala: (1) the distribution of providers across communities and the degree of collaboration between providers; (2) the effect of integration on the content of care provided by midwives; and (3) the determinants of women's choice of provider and the decision to integrate traditional and biomedical care. The data come primarily from the 1995 Guatemalan Survey of Family Health, which includes interviews with 2,872 women of reproductive age in 60 rural communities, as well as interviews with community informants and providers in each community. Data were collected for each of the last two live births in the five years prior to the survey (n = 3,350). The results show that midwives continue to be the major source of pregnancy-related care in rural Guatemala. Most midwives report making referrals, but only about one-third do so on a regular basis. Referrals in the opposite direction, from biomedical providers to midwives, are less common, yet about one-fifth of health centers and posts report making such referrals. Midwife-training programs appear to have a large positive effect on midwife referrals, but this effect appears to be due, at least in part, to establishing a more collaborative relationship between providers. Unlike referrals, there is no evidence that training improves the quality of care provided by midwives. Most midwives continue to use traditional practices, but there is indication that some of these practices are declining in use, while other biomedical practices are being adopted, sometimes with potentially serious consequences. Social and cultural factors such as ethnicity and education, and indicators of need for obstetrical care are more important than enabling factors in explaining use of biomedical care and whether traditional and biomedical care is combined. Financial resources and access to health services do appear to be important in determining choice of a biomedical provider; government-sponsored facilities are the main option for poor, rural Guatemalans. Multilevel models demonstrate that unobserved characteristics of the family and community continue to have a large effect on health care choices, even after controlling for this rich set of observed covariates.
