[The new preschool examination in Baden-Wuerttemberg: what determinants influence the school medical evaluation special need for language promotion in childhood development?].

Background And Aims: The new examination before primary school enrollment in Baden-Wuerttemberg aims at detecting problems in infant development with regard to later school success in time to initiate supporting measures, especially to improve the language skills of children with other native languages. By a 2-level process composed of a screening of language skills (HASE and KVS) and an additional test (SETK 3-5) of children who did not pass the screening, the school physicians attested special needs for language promotion in the kindergarten. This study looked for associated risks of children with special needs for language promotion. The degree of test quality of the 2-level process for identifying special needs for language promotion was determined.

Methods And Results: This cross-sectional analysis explored findings of n=80,781 children in the new examination before primary school enrollment of the data-set of Baden-Wuerttemberg (children with school enrollment 2011). 56,352 children (69.8%) were speaking German, 24,429 children (30.2%) had other family languages. 20,461 children (25.3%) had special needs for language promotion in the kindergarten. A logistic regression model to determine main risks of special needs for language promotion was developed. Main effects were other native languages (OR 5.1 [4.8; 5.2]), problems in subitising (OR 2.8 [2.7; 3.0]) and language development lags in the questionnaire of the nursery school teachers (OR 3.5 [3.3; 3.7]). Protective effects were an elevated graduation of the mother (OR 0.7 [0.7; 0.7]) or the father (OR 0.8 [0.7; 0.8]). Risk scores of the effects were defined. The corresponding predictive probability to different levels of risk scores was calculated. The true positive rate of the screening of language skills (HASE/KVS) in regard to special needs for language promotion was 0.95, the true negative rate was 0.72 and the -positive predictive value was 0.53. The school physician's findings of special needs for language promotion acted as gold standard. With the additional test (SETK 3-5) the positive predictive value improved to 0.9, if at least one of the subtests of the SETK 3-5 was not passed. The risk score-level corresponded with the pretest-probability and the consecutive positive predictive value of the screening of language skills.

Conclusions: This study showed an adequate degree of test quality of the 2-level process in the new examination before primary school enrollment in Baden-Wuerttemberg (screening of language skills and additional test, if the screening is not passed). In addition children with special needs for language promotion had associated risks. Risk scores, that have been defined, offer an information tool to the school physicians concerning the positive predictive value of the screening of language skills without additional testing.