What Factors Predict Differences in Infant and Perinatal Mortality in Primary Care Trusts in England? A Prognostic Model

Abstract

Primary care services in the United Kingdom are managed by 303 Primary Care Trusts (PCTs), each one covering a separate local area. PCTs determine the need of health services in a local community and are responsible for providing them. There is considerable variation between rates of both infant and perinatal mortality in many PCTs and rates are high in some. Those with the worst health and deprivation indicators have been designated as having “Spearhead” status. Although Spearhead PCTs have inferior health outcomes for infant and perinatal mortality, it is unclear whether these poor outcomes result from population characteristics, such as ethnicity and deprivation, inadequate service provision, and/or lack of health service funding for maternity services. This study used prognostic multivariable mixed models to identify PCTs with higher than expected rates of perinatal and infant mortality and investigated factors that might explain this observed variability between PCTs. Data were obtained from each PCT on the number of infant and perinatal deaths, deprivation, ethnicity, Spearhead status, maternal age, and PCT spending on maternal services. Spearhead status was designated in 88 (29%) PCTs. Measures of deprivation, ethnicity, and maternal age were included in the final models for infant and perinatal mortality. The primary study outcome measures were rates of perinatal and infant mortality across PCTs. The final fitted model accounted for 70% of the between-PCT variability for infant mortality and 80.5% of the between-PCT heterogeneity in perinatal mortality. PCT spending on maternal services had no predictive effect on the rates of either perinatal or infant mortality. The rate of perinatal mortality was substantially higher than the expected rate in 2 PCTs. Neither of these trusts had Spearhead status. The investigators conclude from these findings that social deprivation, ethnicity, and maternal age are predictors of both infant and perinatal mortality. However, higher PCT spending on maternity services does not appear to explain differences in rates of infant and perinatal mortality.