Objectives: Most life threatening obstetric complications require hospital treatment to avert maternal mortality. Some assume that in developing countries hospital service for the poor must be in government hospitals and that a large proportion of deliveries needs to be in these hospitals to provide timely access to emergency care. This presents a major problem in countries like India, where almost all rural deliveries are at home and accessible government hospitals generally do not provide surgical treatment for obstetric emergencies. The study's objective was to determine obstetric outcomes, patterns and costs of obstetric care in a part of rural Maharashtra, India, where obstetric outcomes appear relatively good even though most deliveries are at home and government hospitals do not provide emergency obstetric care (EmOC).
Methods: 2905 pregnancies were identified and followed to term to learn the number and types of complications, where these complications were treated, how many women received EmOC and how these services affected outcome.
Results: Eighty-five percent of 2861 deliveries after 24 weeks were at home. A total of 14.4% of deliveries after 24 weeks had identified complications. Of these complicated deliveries, 78.9% were in a hospital. Forty-eight percent of hospital deliveries were in a private hospital, 35% in our project hospital and 18% in a government hospital. Hospitalized patients with obstetric complications constituted 11.4% of all deliveries. The cesarean section rate for all deliveries was 2.0%. Twenty-two of the cesareans were in private hospitals, 32 in our hospital and four in a government hospital. Hospital case fatality (deaths of mothers with identified complications) was 0.3%. Overall case fatality was 0.5%. However, there were only two maternal deaths from obstetric causes (70 per 100,000 live births), making these rates less than robust. The perinatal mortality rate was 36 per thousand live and still births. These outcome and process indicators are better than those reported in most of India, but both maternal deaths could have been prevented by early referral to hospital and 64% of perinatal deaths were to infants delivered at home.
Conclusions: A network of private clinics with a voluntary, low cost hospital is providing effective EmOC in a remote rural area at very low per capita cost in the absence of easily accessible government service and with only 15% of deliveries in hospitals. Charges are low but low per capita cost is primarily due to intelligent self-selection of patients who need hospital care. Even though overall cost is low, cost is still an important barrier for many poor families. Improving the purchasing power of poor families through insurance or subsidy could be a more effective way to improve EmOC than trying to improve inadequate government facilities.