The consistent finding that there are no or only small differences in outcome between bona-fide treatment interventions in the addiction field (and also in other neighbouring fields such as psychotherapy in general) is gaining increasing support in the international literature. The paper by Miller & Moyers 1 represents an elaborate and well-articulated piece in that genre. However, while the support for the outcome equivalence paradox (or the Dodo bird verdict) seems to develop into a consensus, the same is not true concerning what the consequences of this situation should entail for the organization of future research on addiction treatment. Babor suggests, with reference to his experiences as one of the leading researchers within Project MATCH 2, that the failure ‘to explain treatment effects also suggest that we focus attention of how to best motivate people to enter treatment, regardless of the therapeutic modality employed’ 3. The rejection of a continued focus on establishing differences in treatment outcome is turned upside-down in a proposal for more elaborated specificity put forward by Magill & Longabaugh 4, when they write: ‘to the extent that future RCT yield results of no difference the need for specificity as to how the treatment works becomes more compelling’ (p. 876). Miller & Moyers, on their part, state that it seems unlikely that treatment outcome will be improved by searching for better specific ingredients and that future research should focus on therapist, relational and client variables. Although this position seems to be congruent with the ‘common factor ‘ approach 5, such an interpretation is hard to maintain due to the fact that Miller & Moyers also claim that terms such as ‘common’ and ‘non-specific’ factors are misnomers. They claim that it is a mistake to refer to an important treatment factor as ‘non-specific’, and that to the extent that such factors are present we should specify them and develop an understanding of how they the work as well as of how they can be measured and incorporated into therapist training. Here it appears that Miller & Moyers assert that any factor that is important is also a specific factor; in my view, that is at least one step too far. Specific factors are factors that are characteristic and theoretically central to a given treatment intervention. In contrast, common factors are factors that seem to be present among a majority of treatment interventions, and for this reason they are designated as common or contextual factors. By definition, common factors are incidental to treatment intervention theory, a fact that does not hinder that a given common factor may be highly important for treatment outcome at the same time as it is not unique. The most well-studied common factor, which for many decades has been a part of the ‘gold standard’ design in the medical field, is the placebo, present in the randomized double-blind placebo-controlled effect study. A treatment intervention is considered to be a placebo if its content is incidental, as opposed to characteristic in relation to the therapeutic theory studied (i.e. the sugar pill is incidental in a study of the effects of a given antibiotic). The rationale that underlies this design is that comparison conditions should be impossible to identify as different compared to the experimental condition, i.e. the contextual framing should be common to both. For these reasons I find it hard to agree with the suggestion of Miller & Moyers 1 that it is mistake to talk about common and non-specific factors. None.
Read full abstract