Objective To explore risk factors for preterm birth (PTB) between 24+0 and 34+0 gestational weeks in the UK's largest maternity unit, with a particular focus on low risk pregnancies and the effect of socioeconomic status.

Design Retrospective cohort study of routinely collected obstetric and neonatal data.

Setting 50 486 singleton pregnancies booked at the Liverpool Women's NHS Foundation Trust for all women delivering after 24+0 weeks gestation over a 7-year period from 2002 to to 2008.

Main Outcome Measure The primary outcome was preterm birth. Pregnancies were stratified into three groups: low risk; those complicated by medical problems; pregnancies in women with a history of preterm delivery. Multiple logistic regression and generalised additive models were used to explore the effect of covariates including area deprivation, smoking status, BMI, parity and ethnicity.

Results The proportion of PTB was significantly different in the three groups: 1.35% (95% CI 1.24 to 1.47, n=38 994) in the low risk group, compared to 6.55% (CI 6.09 to 7.03, n=10 760) in the medical disorder group and 9.2% (CI 7.39 to 11.61, n=732) in the previous preterm group. 64% of the women delivering at LWH were in the most deprived quintile relative to the English population. The unadjusted odds of preterm delivery in the most deprived quintile compared to the least was 1.60 (CI 1.28 to 2.00) in the uncomplicated group. In a multiple regression model, ever having smoked (OR 1.68 CI 1.35 to 2.08), underweight (OR 1.65 CI 1.005 to 2.56) and highest quintile of area deprivation (OR 1.59 CI 1.19 to 2.11) were associated with increased the risk of PTB. Being overweight decreased the risk of PTB (OR 0.76 CI 0.59 to 0.97). In the medical disorders group, age (OR 1.02 CI 1.011 to 1.04), highest quintile of area deprivation (OR 1.46 CI 1.14 to 1.88), underweight (OR 1.68 CI 1.09 to 2.51), ever having smoked (OR 1.19 CI 1.00 to 1.44), nulliparity (OR 1.37 CI 1.13 to 1.66) and black ethnic group (OR 1.61 CI 1.00 to 2.48) were associated with PTB.

Conclusions Preterm delivery contributes to inequalities in infant mortality. In a cohort of women with no identifiable risk factors for PTB at booking, deprivation of area of residence is associated with higher risk of PTB, even with adjustment for smoking and underweight, which are also important independent risk factors. Deprivation of area of residence needs to be considered when comparing obstetric outcomes in units around the UK.