Article Text
Abstract
Background The proportion of women in England having a caesarean has increased from 9% to 24.8% between 1980 and 2012. There is increasing attention focused on the short and long term effects on the mother and the baby and efforts to have an increased understanding of associated influences. Previous research has adjusted for population case-mix (Bragg al 2010), but there has been less research on the relative impact of organisational factors which is the aim of this study.
Methods A retrospective cross-sectional analysis was performed using routinely collected data from Hospital Episode Statistics dataset with over 660,000 deliveries in 144 NHS trusts in England. The dataset included all women giving birth between April 2010 and March 2011. The outcome was whether the delivery was by caesarean section. Clinical risk factors were calculated using the NICE intrapartum guideline criteria for women at increased risk of complications at the end of pregnancy. Exploratory statistical analysis of caesarean rates for each trust was performed to identify significant variation in rates (both planned and emergency). A multilevel logistic regression model was used to estimate the likelihood of women having a caesarean section given their maternal (level 1) characteristics (age, ethnicity, parity, socioeconomic deprivation), clinical risk factors (NICE Intrapartum guideline criteria) and organisational (level 2) factors (staffing, trust configuration, number of delivery beds, number of deliveries, teaching and Foundation Trust status, CNST rating and women’s experience).
Results In 2010/11, 48% of women in England were categorised as low risk according to NICE Intrapartum Guidelines at the end of pregnancy. Among 660,000 deliveries, 24.8% of women were delivered by caesarean section. Unadjusted rates of caesarean section among the NHS trusts ranged from 36.1% down to 15.2%. Following adjustment for socio-demographic and clinical risk factors, adjusted rates were of a similar range but rates for individual trusts changed, and the impact of organisational factors were examined.
Discussion Once adjusted for socio-demographic and clinical risk factors, it is possible to see how caesarean rates for trusts differ. We discuss the relative impact of organisational characteristics, and what requires further exploration. Routinely collected data can provide information about the type of births that women are experiencing, and provide the means to adjust trust data to take account of the profile of women giving birth in each location, and organisational characteristics. This allows trusts to benchmark its services for quality improvement and provides baseline data for exploring reasons for high and low performing outliers.