Maternal geographic residence, local health service supply and birth outcomes.

Objective: To describe pregnancy complications, mode of delivery and neonatal outcomes by mother's residence.

Design: Register-based cohort study.

Setting: Geographical regions of Iceland.

Population: Live singleton births from 1 January 2000 to 31 December 2009 (n = 40 982) and stillbirths ≥22 weeks or weighing ≥500 g (n = 145).

Methods: Logistic regression was used to explore differences in outcomes by area of residence while controlling for potential confounders. Maternal residence was classified according to distance from Capital Area and availability of local health services.

Main Outcome Measures: Preterm birth, low birthweight, perinatal death, gestational diabetes and hypertension.

Results: Of the 40 982 infants of the study population 26 255 (64.1%) were born to mothers residing in the Capital Area and 14 727 (35.9%) to mothers living outside the Capital Area. Infants outside the Capital Area were more likely to have been delivered by cesarean section (adjusted odds ratio 1.28; 95% CI 1.21-1.36). A lower prevalence of gestational diabetes (adjusted odds ratio 0.68; 95% CI 0.59-0.78), hypertension (adjusted odds ratio 0.82; 95% CI 0.71-0.94) as well as congenital malformations (adjusted odds ratio 0.55; 95% CI 0.48-0.63) was observed outside the Capital Area. We observed neither differences in mean birthweight, gestation length nor rate of preterm birth or low birthweight across Capital Area and non-Capital Area. The odds of perinatal deaths were significantly higher (adjusted odds ratio 1.87; 95% CI 1.18-2.95) outside the Capital Area in the second half of the study period.

Conclusion: Lower prevalence of gestational diabetes and hypertension outside the Capital Area may be an indication of underreporting and/or lower diagnostic activity.