Comparison of self-reports and biological measures for alcohol, tobacco, and illicit drugs consumption in psychiatric inpatients

BACKGROUND: Asking psychiatric in-patients about their drug consumption is unlikely to yield reliable results, particularly where alcohol and illicit drug use is involved. The main aim of this study was to compare spontaneous self-reports of drug use in hospitalized psychiatric patients to biological measures of same. A secondary aim was to determine which personal factors were associated with the use of tobacco, alcohol, and illicit drugs as indicated by these biological measures. METHODS: The consumption of substances was investigated using biological measures (urine cotinine, cannabis, opiates, cocaine, amphetamines and barbiturates; blood carbohydrate-deficient transferrin [CDT] and gamma-glutamyl transferase [GGT]) in 486 consecutively admitted psychiatric patients, one day following their hospitalization. Patients' self-reports of alcohol, tobacco and illicit drugs consumption were recorded. Socio-professional and familial data were also recorded. RESULTS: The results show a low correlation between biological measures and self-reported consumption of alcohol and illicit drugs. Fifty-two percent of the patients under-reported their consumption of illicit drugs (kappa=.47). Patients with schizophrenia and personality disorders were more likely to disclose their illicit drug consumption relative to patients suffering from mood disorders and alcohol dependence. Fifty-six percent of patients underreported alcohol use, as evaluated by CDT (kappa=.2), and 37% underreported when using the CDT+GGT measure as an indicator. Smoking appeared to be reported adequately. In the study we observed a strong negative correlation between cannabis use and age, a strong correlation between tobacco and cannabis use, and correlations between tobacco, cannabis and alcohol consumption. CONCLUSION: This study is the first to compare self-reports and biological measures of alcohol, tobacco and illicit drug uses in a large sample of inpatients suffering from various categories of psychiatric illnesses, allowing for cross-diagnosis comparisons.