Acta Pædiatrica ISSN 0803–5253 REGULAR ARTICLE A validation study of the Norwegian version of the Ages and Stages Questionnaires J¨ org Richter (jorg.richter@r-bup.no) 1 , Harald Janson 2 1.Centre for Child and Adolescent Mental Health, Eastern and Southern Norway, Oslo, Norway 2.Norwegian Centre for the studies of Conduct Problems and Innovative Practice Ltd (Behavioral Centre), Oslo, Norway Keywords Ages and Stages Questionnaires, Early childhood development, Gender comparison, Norwegian representative data, Preterm birth Correspondence Prof. Dr. J¨ org Richter, Centre for Child and Adolescent Mental Health, Eastern and Southern Norway, P.O. 4623 Nydalen, N-0405 Oslo, Norway. Tel: +47 2258 6043 | Fax: +47 2258 6001 | E-mail: jorg.richter@r-bup.no Received 9 October 2006; revised 8 January 2007; accepted 15 January 2007 DOI:10.1111/j.1651-2227.2007.00246.x Abstract Aim: To explore the construct validity of the ASQ in relation to premature birth; maternal age, completeness of the family and level of education; and the child’s gender. Methods: Norwegian mothers’ evaluations of 1172 children who completed age appropriate ASQs twice were statistically analysed by means of univariate and multivariate methods. Results: The developmental level of girls is on average higher than that of boys at both assessments and there is a substantial increase between follow-up assessments in all ASQ scores except for gross motor skills. The mother’s level of education; whether the child was born prematurely or at full-term; and the interaction between the child’s gender and his/her mother’s level of education are important factors relating to the child’s developmental status in both assessments. Conclusion: The results that were derived by means of the Ages and Stages Questionnaires concerning the impact of maternal socio-demographical factors, the child’s gender, and premature birth upon early childhood development largely confirm the findings in the literature supporting the construct validity of the ASQ. The ASQ was found to be an effective diagnostic tool of developmental delay and/or disturbances. The development of gender dependent norms is recommended. Various instruments have been developed in order to de- tect possible developmental disturbances in premature, low birth weight or otherwise disabled children with the aim of facilitating an early rehabilitation if necessary. One widely used method of assessing the development of infants are the Bayley Scales of Infant Development (BSID; Ref. 1), con- sisting of a mental scale, a motor scale and a behaviour rat- ing scale. This method has to be administered by examiners who are experienced clinicians specifically trained in BSID test procedures. It was standardized on infants, toddlers and preschoolers between one and 42 months of age. Other screening tools often applied include the various versions of the Denver Developmental Questionnaire (2), a parent-answered questionnaire that can be used from birth up to the age of 6. The child’s development is assessed, for example, by means of the Denver II Pre-screening Devel- opmental Questionnaire (PDQ-II) on the basis of his/her ability to: move; use visual and hand skills; employ social and self-help skills; and utilize speaking and understanding skills. Four age-related forms are available. Parents found the Revised Pre-screening Developmental Questionnaire (R- PDQ) interesting and easy to complete, whilst healthcare professionals found it economical and easy to interpret (3). In the case of abnormal results, it is recommended that a re-evaluation be performed 1 month later. More recently, the Ages and Stages Questionnaires (ASQ - 4) were developed. Nineteen age-specific questionnaires are available. Each questionnaire contains thirty items de- signed to assess the infant’s development in the areas of com- munication, gross motor, fine motor, problem solving, and personal-social skills. The parents or other caregivers are asked whether the child performs the described behaviour based on three alternatives: ‘yes’, ‘sometimes’ and ‘not yet’. It is reported to have excellent psychometric properties and one of its advantages is that it requires much less time from the paediatrician than instruments that require direct exam- ination (4). However, there are conflicting reports of single- point accuracy which appears to depend upon the setting in which the ASQ is used. In relation to the Canadian version of the ASQ, utilized in a community setting with the ques- tionnaire being completed in the waiting room of a busy clinic, Rydz et al. (5) reported a sensitivity of 0.67, a speci- ficity of 0.39, a positive predictive value of 34% and a nega- tive predictive value of 71%. However, Skellern, Rogers and O’Callagham (6) found a higher sensitivity of 0.90, a speci- ficity of 0.77, a positive predictive value of 40% and a neg- ative predictive value of 98% in a mail study conducted in Australia. Accordingly, the former concluded that the ASQ does not meet the requisite standard for screening tests, whereas the latter concluded that the high ASQ negative predictive value supports its use as a screening tool in prema- ture children. In a smaller sample, Danish colleagues found a significant correlation between ASQ scores and IQ. Further- more, the ASQ score of extremely preterm children ranged significantly below the score of term children (7). It is well known that the development of children is de- termined and moderated by many different biological and socio-cultural factors that are in a continuous interaction. The gender of the child; the level of the mother’s education; her age; and whether the child is living with single parent 748 C 2007 The Author(s)/Journal Compilation C 2007 Foundation Acta Pædiatrica/Acta Pædiatrica 2007 96, pp. 748–752