Long-term effects of transference interpretations: comparing results from a quasi-experimental and a naturalistic long-term follow-up study of brief dynamic psychotherapy.

A follow-up study, using a quasi-experimental non-equivalent groups comparison design, the so-called regression discontinuity design, reported a negative long-term effect of a high frequency of transference interpretations given to patients in brief dynamic psychotherapy who had been deemed highly suitable for such treatment. The major threat to internal validity of the findings was the possibility that the patients who were evaluated as highly suitable for dynamic psychotherapy might be the ones with the naturally worst outcomes, independent of therapy technique (selection maturation). Data from a similar naturalistic study with no quasi-experimental manipulation of the treatment technique revealed, however, that highly suitable and relatively less suitable patients had similar outcomes. This makes selection maturation a less plausible confound of the (negative) treatment effect estimated in the quasiexperimental study. A high frequency of transference interpretations in brief dynamic psychotherapy seems to be causally related to less favourable long-term dynamic change.