Background Children who are born preterm and those who are socially disadvantaged tend to have increased morbidity in childhood. Few studies are large enough and have sufficient follow-up to explore the size of these separate effects across childhood. We explored whether birth before full term (gestation 39–41 weeks) and social disadvantage represent a double jeopardy by comparing the size of these separate effects on rates of hospital admission from birth to 10 years.

Methods We linked routine data from birth registration and hospital admissions for all live, singleton births in England in 2005–2006. Children (n=1,018,136) were followed up from discharge from their birth admission to age 10 years, death or study end (31st March 2015). Negative binomial regression models were used to estimate adjusted rate ratios (aRR) with 95% confidence intervals (CI) for the effect of gestational age (<32, 32–33, 34–36, 37–38, 39–41, 42 weeks) and area deprivation (quintiles) at birth on hospital admissions rates, stratified by age (<1, 1–4, 5–10 years). Population attributable fractions (PAFs) were estimated using the aRR for gestational age and area deprivation. All analysis was conducted in Stata version 14.

Results Hospital admission rates increased strongly with prematurity and area deprivation, even after adjustment for other factors. The effect of gestation was most marked in infancy, when even being born at ‘early term’ (gestation 37–38 weeks) had a stronger effect than living in the most deprived areas (aRR=1.37, 95% CI 1.35–1.39 for 37–38 versus 39–41 weeks; aRR=1.26, 95% CI 1.24–1.28 for the most versus least deprived areas). The effect of gestation persisted though was less marked at age 5–10 years, although even being born ‘late preterm’ (gestation 34–36 weeks) had a stronger effect than living in the most deprived areas (aRR=1.39, 95% CI 1.35–1.43 for 34–36 versus 39–41 weeks; aRR=1.29, 95% CI 1.26–1.31 for the most versus least deprived areas). In infancy, the PAF for birth before full term (<39 weeks) was larger than the PAF for living in the poorest quintile. At age 5–10 years, the PAF for birth before full term was only slightly smaller than the PAF for living in the poorest quintile

Conclusion Gestational age and social deprivation were independent risk factors for hospital admissions throughout childhood. The impact of being born even a few weeks early is of similar magnitude to living in the most deprived areas. Children born early and living in more deprived areas have a double jeopardy.