Strategies for assessing the impact of loss to follow-up on estimates of neurodevelopmental impairment in a very preterm cohort at 2 years of age

Table 1 Loss to follow-up and missing data at two years of corrected age in a very preterm cohort

		Total surviving to 2 years old^a	Non-responders	Responders
Total sample		N = 1737	N = 796 (45.8%)	N = 941 (54.2%)
Missing data	Perinatal and SES variablea^b,c	251 (14.5%)	141 (17.7%)	110 (11.7%)
Missing data	Neurodevelopmental impairment at 2 years CA^b,d	N/A	N/A	104 (11.1%)
Portugal		N = 603	N = 196 (32.5%)	N = 407 (67.5%)
Missing data	Perinatal and SES variables^a,d	72 (11.9%)	37 (18.9%)	35 (8.6%)
Missing data	Neurodevelopmental impairment at 2 years CA^b,d	N/A	N/A	19 (4.7%)
United Kingdom		N = 1134	N = 600 (52.9%)	N = 534 (47.1%)
Missing data	Perinatal and SES variables^a,d	179 (15.8%)	104 (17.3%)	75 (14.0%)
Missing data	Neurodevelopmental impairment at 2 years CA^b,d	N/A	N/A	85 15.9%)

^aDeath excluded; Number of deaths before age 2: total:7, Portugal:2, United Kingdom:5 ^bSES socioeconomic status, CA corrected age ^csee text for data items ^dNeurodevelopmental assessment is a composite of motor disability and non-verbal cognitive disability

ISSN: 1471-2288