Is thioridazine safer, cheaper, and more effective than alternative antipsychotic drug treatments for anxiety, agitation, mania, and hypomania? We do not have enough evidence to answer this because thioridazine has been widely used for 30 years—before the days of rigorous randomised controlled trials. Lack of evidence is not evidence of no benefit. Conversely, there is only evidence of a handful of adverse cardiac events, some of which may not have been directly caused by thioridazine or may have been due to combination with other drugs. Although the reported cardiac deaths are lamentable, it seems that thioridazine is much safer than other effective drugs such as aspirin, which continues to be sold over the counter. Even if adverse events are under-reported, there is far more practical evidence of long term safety for thioridazine compared with much more expensive and much less tried and tested drugs. A recent study noted that many new drugs have a high rate of serious side effects which go undetected until late in postmarketing surveillance.² The Committee for the Safety of Medicines needs to balance these relative risks.

The manufacturers of thioridazine have raised few objections to the restricted use of thioridazine because it is an extremely cheap drug, costing only a few pence compared with newer antipsychotics that may be up to a hundred times more expensive, as well as having different and sometimes distressing side effects. The prescription costs of antipsychotic drugs has risen sharply in Scotland in the past two years.

We looked at the following issues before withdrawal of thioridazine:

(1)Initial diagnosis before thioridazine was prescribed

(2)Duration and dose of treatment

(3)Review of previous psychiatric treatment including attempts to reduce or stop thioridazine.

Results indicated that 28 out of 175 patients under the care of a psychiatrist specialising in learning disability received thioridazine. Twenty three patients received small doses (50-100 mg) for many years. The main indication for use was marked challenging behaviour (12 patients). Based on the above information, a decision was made to gradually reduce (12 patients) or suddenly stop thioridazine (16 patients).

As an alternative, a tranquilliser was prescribed for all 28 patients. Risperidone was the most commonly used atypical antipsychotic drug because of some evidence of its effectiveness in challenging behaviour.^1-2 A review of patients' mental state three months after the substitution of tranquillisers indicated that twenty one patients either remained stable or improved, and seven patients deteriorated. These seven patients subsequently improved, either by a change to a different psychotropic drug or by adjusting the dose of existing drugs. No patient needed to restart taking thioridazine.

Review of previous psychiatric treatment helped us to identify instances of rebound increase in behavioural problems when thioridazine was stopped. It also prompted us to avoid using neuroleptics that were either ineffective or caused unacceptable side effects. Evidence of rebound aggression on stopping neuroleptics was also observed by Mos et al.^1-3 This lead us to withdraw thioridazine gradually in most cases.