The objectives of this study were (1) to evaluate the prevalence of relational withdrawal behaviour in infants aged 14–18 months attending a public health centre in Paris, (2) to check some identified risk factors for relational withdrawal behaviour in this population.

A cross-sectional study was conducted in infants aged 14–18 months attending a child health screening centre during the year 2005.

A total of 640 children were included in the study. Thirteen percent of the 640 infants (n = 83, 95% CI [10.4%; 15.6%]) had an ADBB score at 5 and over 5 on the ADBB. There was a clear relationship between withdrawal behavior and having psychological difficulties as reported by parents, and between withdrawal and developmental delay. Withdrawal was also significantly associated with being a boy, with living in risk conditions (e.g. child being in joint custody, or with living in a foster family), with being adopted, or with being a twin. More withdrawn infants were taken care of at home.

Sustained relational withdrawal behaviour was linked with developmental disorders and psychopathology and not with SES, ethnical origin or rank of birth. The scale could be used in screening early psychopathology in infants aged 2–24 months of age.