Background: The United Kingdom has one of the highest rates of stillbirth in Europe, resulting in approximately 4,000 stillbirths every year. Potentially modifiable risk factors for late stillbirths are maternal age, obesity and smoking, but the population attributable risk associated with these risk factors is small.Recently the Auckland Stillbirth Study reported that maternal sleep position was associated with late stillbirth. Women who did not sleep on their left side on the night before the death of the baby had double the risk compared with sleeping on other positions. The population attributable risk was 37%. This novel observation needs to be replicated or refuted.
Methods/design: Case control study of late singleton stillbirths without congenital abnormality. Controls are women with an ongoing singleton pregnancy, who are randomly selected from participating maternity units booking list of pregnant women, they are allocated a gestation for interview based on the distribution of gestations of stillbirths from the previous 4 years for the unit. The number of controls selected is proportional to the number of stillbirths that occurred at the hospital over the previous 4 years.
Data collection: Interviewer administered questionnaire and data extracted from medical records.
Sample size: 415 cases and 830 controls. This takes into account a 30% non-participation rate, and will detect an OR of 1.5 with a significance level of 0.05 and power of 80% for variables with a prevalence of 57%, such as non-left sleeping position.
Statistical analysis: Mantel-Haenszel odds ratios and unconditional logistic regression to adjust for potential confounders.
Discussion: The hypotheses to be tested here are important, biologically plausible and amenable to a public health intervention. Although this case-control study cannot prove causation, there is a striking parallel with research relating to sudden infant death syndrome, where case-control studies identified prone sleeping position as a major modifiable risk factor. Subsequently mothers were advised to sleep babies prone ("Back to Sleep" campaign), which resulted in a dramatic drop in SIDS. This study will provide robust evidence to help determine whether such a public health intervention should be considered.
Trial registration number: NCT02025530.