1 July 2002

Benzodiazepine addiction - trial of 2 methods of treatment from Finland

Treatment of out-patients with complicated benzodiazepine dependence: comparison of two approaches. Vorma H, Naukkarinen H, Sarna S, Kuoppasalmi K. Addiction (2002) 97: 851-859

Dear Colleagues,

These authors randomised 76 benzodiazepine dependent patients to receive 'traditional' dose taper treatment or an experimental protocol involving fixed rates of withdrawal and psychosocial supports.

The patients presented to four addiction clinics in Finland with the experimental condition being implemented by one such clinic. This presumably entailed some patients having their treatment moved and standardised to the experimental protocol.

There was daily or binge drinking in about 30% of subjects. About half had used 40mg diazepam equivalent daily or more. Patients used sedatives for an average of 7 years and over 80% had previously tried to quit. About one in seven had used recreational drugs in the previous year. 80% of each group has a current axis I disorder (anxiety, depression, alcohol abuse). Two thirds had a personality disorder, leaving only 5-10% without a formal psychiatric diagnosis. There were no significant differences between the randomized groups.

The experimental plan included a fixed diazepam or equivalent taper based on the patient's dose and proportionate reductions of about one tenth each week. The time taken for tapering was thus longer for higher doses. They also included a drug/alcohol diary, video film on withdrawals, sleep advice, printed material on reductions, 'assessment of benzodiazepine functions' and relaxation exercises. About a third of subjects in each group was prescribed fluoxetine for depression. 'Control' treatment involved traditional sedative reductions, 'supervised by a physician'.

Subjects had monthly urine and/or blood drug screens plus interviews including an AUDIT questionnaire after abstinence was attained, or at 12 months, where possible.

Judging by abstinence or 12 months of treatment attendance the experimental group had a 46% completion rate while the control group yielded 70%, a significant difference. One subject who had been accepted but dropped out, committed suicide some months later. The authors state that if they excluded patients who did not participate in treatment or left treatment 'for reasons unrelated to the treatment' the differences become insignificant. [But if patients did not participate in the treatment as offered, then would normally be considered 'failures' in any fair assessment of an experimental intervention.]

The reductions in benzo use were impressive, all tending to be better in the 'control' taper group (but apparently not significantly). Fully 27% achieved abstinence in the control group, 13% in the experimental. There were reductions in overall benzo use by 45-80% across the groups. One in six subjects demonstrated no reductions over 12 months despite the treatment interventions. It would appear that all of the patients were followed up at some point, a remarkable feat for benzo users, or any drug treatment subjects.

This study tells us that trying to standardise an approach to treatment in a way which is not based on sound research findings does not lead to improvements in outcomes. Indeed, it may be that the lack of prescribing flexibility even reduced successes for some patients in this difficult area.

Unlike heroin addiction, there is still no 'gold standard' treatment for tranquillizer addiction. But despite this, most in the field seem to approve some degree of diazepam (or similar) prescribing with reductions as tolerated in a setting of medical care and psychosocial supports. These authors did not use diazepam in every case, nor do they describe supervised or daily dosing which is now recommended in certain unstable cases (see Strang J. UK Dependency Treatment Guidelines 1999 p31-33).

comments by Andrew Byrne ..