An evaluation of the Dartmouth Assessment of Lifestyle Inventory and the Leeds Dependence Questionnaire for use among detained psychiatric inpatients

AIM: To examine the psychometric properties of the Dartmouth Assessment of Lifestyle Inventory (DALI) as a substance use disorder screening instrument, and the Leeds Dependency Questionnaire (LDQ) as a measure of severity of dependence among inpatients with severe mental disorders detained in hospital. DESIGN: Comparison of DALI scores, LDQ scores and a clinical assessment of substance use disorder. Statistical analysis calculated Cronbach's alpha, receiver operator characteristics, anova and multivariate linear regression. SETTING: An English medium secure psychiatric hospital. PARTICIPANTS: A random sample of 60 adult inpatients with one or more diagnoses of schizophrenia, personality disorder and mild to moderate learning difficulty. MEASUREMENTS: All participants completed the LDQ, DALI and a clinical assessment of substance use disorder using ICD-10 criteria. FINDINGS: The diagnostic accuracy of the DALI instrument was 74% for alcohol disorders and 83% for drug disorders. The reliability of the LDQ and concurrent validity with the DALI were good (a = 0.915, R2 = 0.439, F (2, 57) = 22.3, P < 0.001). CONCLUSIONS: The DALI's overall accuracy for screening inpatients is good, but it under-diagnosed substance use disorders, probably because the hospital environment moderates substance use. The results suggest that the LDQ has very good psychometric properties and is useful as a measure of substance dependence in this population. Furthermore, the validity of the LDQ in this clinical population provides evidence to suggest the same cognitive and behavioural processes of addiction operate among those inpatients with severe mental disorders detained in psychiatric hospital as in the general population.