The Additive Effects of Household Smoking and Housing Tenure on Birthweight Distribution in a Birth Cohort

Objective:to study the additive effects of smoking and housing tenure on birthweight distribution and risk of birthweight <2500g, <3000g and <3500g Setting:Coventry Study design:whole year birth cohort Outcomes: birthweight normal distribution curves with medians (50th centile), 10th and 90th centiles; birthweight <2500g, <3000g and <3500g Participants:2600 infants enrolled into the CoventryCohort Study by the Family Health Visitor at the birth visit Results:birthweight data were available on 2584 infants. The median for the whole cohort was 3340g (3316, 3364) with a 10th centile of 2560g (2536, 2584) anda 90th centile of 4000g (3976, 4024). Medians, 10th and 90th centiles were significantly lower among infants born in smoking households and rented housing: smoking households 3260g (3224, 3294), 2520g (2486, 2554) and 3920g (3886, 3954); rented housing 3250g (3212, 3288), 2470 g (2432, 2508) and 3910g (3874, 3946). Households which are both rented and smoking have the lowest median birthweight (3210g [3162, 3258]), 10th centile (2440g [2392, 2488]) and 90th centile (3856g [3808, 3904]). Risks of birthweight <2500g, <3000g and <3500g show a clear gradient with the highest risk amongsmoking and rented households (odds ratios 1.78 (1.15, 2.81); 1.99(1.52, 2.61); 2.13(1.70, 2.70)) after adjustment for ethnicity, marital status, gestational age and maternal education. Conclusions:these data show a shift to the left of the normal birthweight distribution among infants born into smoking households and those living in rented accommodation – an accepted low income proxy in the UK. An additive effect of smoking and rented housing is shown resulting in a further shift to the left of the distribution. The additive effects of smoking and living in rented housing on the risk of birthweight below optimum levels adjusted for ethnicity, marital status, maternal education and gestational age are demonstrated