The Psychiatric Assessment Schedule for Adult with Developmental Disability (PAS-ADD) Checklist: reliability and validity of French version
Correspondence: Dr Giuliana Galli Carminati, Department of Mental Health and Psychiatry, HUG-University Hospital of Geneva, Geneva, Switzerland (e-mail: email@example.com).
The lack of psychometric measures of psychopathology especially in intellectual disabilities (ID) population was addressed by creation of the Psychiatric Assessment Schedule for Adult with Developmental Disability (PAS-ADD-10) in Moss et al. This schedule is a structured interview designed for professionals in psychopathology. The PAS-ADD Checklist was created as a screening tool designed for non-specialists in mental illness, such as families and care staff. The Checklist includes 29 symptoms items graded on a four-point scale. When the score passes the threshold, this indicates the need for further psychiatric assessment. In a study by Moss et al. and a replication by Sturmey et al., the PAS-ADD Checklist was validated and proved reliable as a screening tool for psychiatric disorders in a population of adults with ID. In this study, the French translation of the Checklist is compared with the English version and the psychometric properties are presented for outpatients.
The French version was created by translation and back translation. Acceptability, internal consistency, inter-rater reliability, factorial analysis and sensitivity/specificity were calculated.
Reliability analyses for sub-scales and threshold scales showed good (Cronbach's alpha coefficient greater than 0.7) to acceptable (alpha over 0.6) internal consistency. Cronbach's alpha was over 0.8 when the total scale was considered. Spearman Rank correlations, calculated for 45 pairs of raters on threshold scores, are above 0.66, which is a good sign of accordance between non-specialist raters. Sensitivity and specificity were computed for the number of participants who did and did not cross threshold and for whom a psychiatric disorder was or was not present. The sensitivity was 55% and specificity was 70%. The confirmatory factor analysis with a three-factor solution explained only 46.1%, which suggests a mediocre fit of the data to the model. Even if items have good saturation on each factor, it does not fit with the theoretical model.
The validity of the French version in this sample seems to be acceptable. Specificity was higher than those reported in the English version and sensitivity was lower. The French version was less successful in screening than English version, probably because of the low number of false negatives in this sample, which constitutes recruitment bias in a psychiatric sample. Nevertheless, the French version of the PAS-ADD Checklist is reliable as a general screening tool for psychiatric disorders.