O'Connor et al (Reference O'Connor, Heron and Golding2002) report the effects of antenatal anxiety on behavioural/emotional problems in 4-year-old children. Their analysis of the Avon Longitudinal Study of Parents and Children (ALSPAC), a longitudinal, prospective study of women, their partners and an index child (Reference Golding, Pembrey and JonesGolding et al, 2001) takes into account a number of important covariates, including postnatal anxiety, gestational age, birth weight, and socio-economic status. They have not, however, included any measures of parenting. This is of concern because there is now a substantial body of evidence to indicate a clear association between parenting and child emotional and behavioural problems. For example, there are now a number of empirically validated models depicting the developmental progression for conduct and behaviour problems. These show a clear association between parenting practices characterised by harsh and inconsistent discipline, little positive parental involvement with the child, poor monitoring and supervision, and behaviour and conduct problems in early childhood (Reference Patterson, DeBaryshe and RamseyPatterson et al, 1989). Indeed, work using structural equation models showed that parenting and family interaction variables accounted for 30-40% of the variance in child antisocial behaviour (Reference Patterson, DeBaryshe and RamseyPatterson et al, 1989).

It seems likely that parenting practices would not be adequately controlled for through the use of a socio-economic covariate owing to the fact that, although parenting practices are influenced by social and cultural factors such as class (Reference Hoff, Laursen, Tardif and BornsteinHoff et al, 2002), one of the most extensive epidemiological studies of childhood psychiatric disorders showed that social class was a poor predictor of child adjustment (Reference Rutter, Cox and TemplingRutter et al, 1975).

It seems likely that parenting exerts an independent effect on child outcomes such as emotional and behavioural adjustment. The ALSPAC data contain a number of measures of parenting, including, for example, a standardised instrument measuring parenting practices during toddlerhood. It would be useful if further analysis of this data-set were undertaken to establish whether these important findings are maintained when parenting is included in the model.