Table 2

Logistic regression models with difference-in-differences analyses of the effect of the urban perinatal health programme on perinatal mortality, preterm birth and small-for-gestational age (SGA)

Independent variablesPerinatal mortalityPreterm birthSGA
OR95% CIOR95% CIOR95% CI
Trend per year0.94390.9362 to 0.95170.99700.9944 to 0.99970.98090.9787 to 0.9831
Difference intervention and control group0.86010.7534 to 0.98191.09621.0505 to 1.14401.13131.0935 to 1.1703
Programme effect (change in trend per year from 2010 onwards)1.05350.9889 to 1.12230.98090.9619 to 1.00040.99280.9772 to 1.0086
Low SES1.37611.2917 to 1.46601.20171.1764 to 1.22751.44171.4170 to 1.4668
Ethnicity non-Dutch1.26901.3462 to 1.19631.00681.0264 to 0.98761.29271.3134 to 1.2722
Parity 0 (reference)1.00001.00001.0000
Parity 1–20.75600.7120 to 0.80260.61140.5998 to 0.62320.53400.5254 to 0.5427
Parity 3+1.28231.1541 to 1.42480.82180.7894 to 0.85540.48710.4687 to 0.5062
Age <25 (reference)1.00001.00001.0000
Age 25–340.94010.8654 to 1.02120.97860.9529 to 1.00500.85540.8377 to 0.8735
Age≥351.22381.1112 to 1.34791.07621.0423 to 1.11110.93390.9100 to 0.9585
Baseline rate*8.67.8 to 9.47169 to 74155151 to 159
  • Perinatal mortality is defined as stillbirth from 24 weeks onwards plus early neonatal mortality. The total number of observations for the perinatal mortality model is 942 614. Preterm is defined as born before a gestational age of 37 weeks and SGA is defined as a birth weight below the 10th percentile corrected for gestational age and sex. The total number of observations for the preterm birth and SGA models is 937 639.

  • *For perinatal mortality per 1000 live and stillbirths, for preterm birth and SGA per 1000 live births.

  • SES, socioeconomic status.