Dr. Peter Haddad's paper "Do antidepressants have any potential to cause addiction?" is reminiscent of the study in which Marks (1978) concluded that the dependence risk with benzodiazepines (BDZs) was "probably less than one case per 50 million months in therapeutic use". With help from Roche, Marks had trawled 17 years worth of medical literature for reports of addiction to BDZs and then proposed, as a denominator, the total number of prescriptions written (Medawar, 1992, 133-136, 148-150). Haddad does essentially the same:

"Comparing the 21 reports in Table Two to the hundreds of millions of patients treated with antidepressants over more than 40 years, suggests an incidence of addiction so low as to be clinically irrelevant."

Like Marks, Haddad's also fiddles with definitions, even to the extent of calling the problem 'addiction' - though this word has been studiously avoided in his prime source: DSM-IV. Haddad says: "The term 'addiction' is used in preference to 'dependence', as it is associated with less semantic confusion". His argument then goes something like this:

1. The terms "addiction" and "dependence" mean the same. See DSM-IV (but don't look too hard)

2. Antidepressants aren't drugs of addiction. See my paucity of data (and never mind withdrawal).

3. Therefore antidepressants aren't drugs of dependence.

Thus Haddad does not come close to answering the kinds of questions patients are increasingly asking about their problems coming off antidepressants - including some telling and distressing examples on this website. Haddad's view is that the withdrawal symptoms you see with SSRIs are typical of those found with anti-convulsants, beta-blockers, nitrates, diuretics, anti-hypertensives, sympathomimetics, heparin, antipsychotics, dopaminergic agents and lithium. Haddad's list does not include benzodiazepines. He appears either unaware or unconcerned that the WHO's adverse drug reaction database in Uppsala includes three SSRI antidepressants and three benzodiazepines, in the top ten drugs reported for withdrawal reactions.

Haddad does not believe that the question of dependence arises just because some people find SSRIs hard to stop: "This is misleading on several counts", he says - because "Addiction/dependence is a syndrome in which the hallmark is a compulsive pattern of drug use". Citing all the usual suspects, including himself, he then sweepingly asserts: "withdrawal symptoms do not usually prevent antidepressants being stopped. Most symptoms are mild, do not reach medical attention and those that do usually only require reassurance".

Oh yes, one other thing. All this mistaken talk about dependence is contributing to a vast burden of ill-health:

"Such views may contribute to the 25% of depressed individuals not seeking medical help … and the sizeable proportion who comply poorly with antidepressants"