When faced with the possibility of discontinuation, patients, their family members and physicians fear recurrence of the disorder, and might also fear antidepressant discontinuation symptoms [see eg. previous post] .... The importance and prevalence of antidepressant discontinuation contrast largely with the limited scientific evidence and clinical attention for this topic. ... [D]iscussions ... are particularly driven by opinions or personal experiences rather than evidence.
There are methodological problems of antidepressant discontinuation studies, not least that blinding strategies are often insufficient to properly address placebo and nocebo aspects (see eg. previous post). The review notes that antidepressant discontinuation symptoms even occur in blinded antidepressant arms when antidepressants are continued. Evidence is needed about the underlying mechanisms of antidepressant withdrawal (see eg. previous post).
As the also helpful NICE draft guideline recommendations make clear:
Dependence is characterised by tolerance (the need for increasing doses to maintain the same effect) and withdrawal symptoms if the dose is reduced or the medicine is stopped abruptly. Addiction also features tolerance and withdrawal but has the additional characteristics of cravings, lack of control, overuse and continued use despite harm. There is considerable debate in relation to these definitions, and in practice, the terms are often used interchangeably.
Antidepressants do not cause tolerance in the same way as alcohol and opioids. Nor should people who uses antidepressants be seen as drug addicts, in the sense of drug abusers, as antidepressants are not stimulants as such and do not create cravings (see previous post). As I've said several times before, psychological dependence on antidepressants should not be minimised (see eg. previous post). The exact role of any physical dependence needs further clarification.