Comparing self-report medication data from a longitudinal study on intellectual disability and national dispensing records

Background: Medication data are a valuable resource in epidemiological studies. As the most common data collection method of medication data is self-report, it is important to understand the accuracy of this in comparison with other methods such as dispensing records. The aim of this study was to compare the agreement between two different sources of medication data of older adults with intellectual disability (ID).

Methods: Self-report medication data were gathered from the Intellectual Disability Supplement to the Irish Longitudinal Study on Ageing and linked to national pharmacy dispensing records. The kappa statistic was used to measure agreement between the two data sources for psychotropic medication.

Results: The lowest agreement level was 'moderate' for the number of anxiolytics reported (kappa 0.56). The highest level of agreement was 'almost perfect' for the binary variable of antipsychotics (kappa 0.91). Other agreement results were 'substantial' or 'almost perfect'.

Conclusions: Good agreement was found between the Intellectual Disability Supplement to the Irish Longitudinal Study on Ageing medication dataset and national dispensing records. Self-report medication data appear to be a valid method of data collection in psychotropic medication use in adults with ID.