I would like to comment on Lader [1] and support his scepticism about the risks of benzodiazepines. I would also like to suggest that the certainty surrounding the existence of benzodiazepine dependence may be misplaced [2–4]). Our attitudes to benzodiazepines are coloured by our morally jaundiced view of ‘drugs’ and their so-called abuse.

Twenty years ago a legal class action was started in England and Wales on behalf of patients who believed they had been seriously damaged by prescribed benzodiazepines; i.e. benzodiazepine dependence. Several hundred patients were involved and approximately 20 consultant psychiatrists examined the patients and reported on their clinical records. Along with most other consultant psychiatrists, I was convinced of the existence of such a disorder, and I supported patients and their groups who felt they were coming to grips with their seriously disabling benzodiazepine dependence. I was invited to become an expert witness for the plaintiffs.

After submitting a number of reports I was contacted by one of the patients' lawyers, who asked me why I had not yet found a case of dependence that could be put before the Courts. We had very full clinical National Health Service records that usually went back to the patients' birth, together with extensive hospital records. I explained that the patients' narrative was so often not supported by the clinical records.

The Courts require expert witnesses to be impartial, even if instructed by and paid by one of the parties to litigation. It followed that alternative explanations must be considered very seriously. I sent the lawyer a list of 24 alternative explanations and pointed out that I would have had to be able to rebut in Court each alternative explanation as well as support the patients' contention of harm if litigation were to succeed. The defendants independently developed a very similar list to organize their response.

Three immediate alternative explanations were (i) symptom recurrence; (ii) symptom emergence: new symptom emerged once the dose was reduced or stopped; and (iii) symptom misattribution. The patients were mistaken in blaming benzodiazepines for their current distress. It proved very difficult to rule out these alternative explanations or, indeed, the other 21.

The class action collapsed, but this litigation provided a unique opportunity to test forensically the existence of a clinical condition and the failure to demonstrate it must throw great doubt on our concepts of benzodiazepine dependence.

We must be very aware of the fallacy post hoc ergo propter hoc[after this therefore because of this]. While it may have seemed reasonable for patients and their clinical advisers to conclude that benzodiazepines produced their distress, they may have fallen victim to this fallacy. Indeed, are we still victims?

Perhaps it is time we got off our moral high horse in respect of benzodiazepine use and took a balanced view of the benefits and risks.


  1. Top of page
  2. Declaration of interest
  3. References