Recent empirical evidence of deterioration during both nonbehavioral and behavioral marital and family therapy is presented. While the frequency of patient worsening in marital-family therapy does not appear to exceed that previously found for individual psychotherapy, the acceptability of the evidence for negative effects in the treatment of systems may be greater than that which exists for individual treatment. After examining the empirical evidence of negative effects in family therapy and some of the factors that influence their occurrence, the authors discuss some conceptual issues relevant to a definition of worsening in marital and family therapy and present some methodological guidelines for the assessment of deterioration in therapy with family systems. It is concluded that the study of deterioration processes in family therapy may aid the understanding of family change processes more generally.

Wells and Dezen's revisited results of research on the outcomes of nonbehavioral family therapy are themselves revisited. While their conclusions are largely defensible in terms of conventional criteria for research design and for assessing change in psychotherapy, we question whether such standard criteria are sufficient for studying the outcomes of family therapy. Moreover, Wells and Dezen's preoccupation with therapeutic technology at the expense of relationship factors and of "objective" change measures at the expense of more inferential measures severely limits the clinical and conceptual meaningfulness of their review.

No abstract

Two recent published reviews of research on behavioral marriage therapy stimulated us to supplement these accounts with additional relevant data. First, we place research on behavioral couples therapy in the broader context of outcome research on nonbehavioral marital therapy. We then summarized the results of 23 studies of behavioral couples therapy not included in these previous reviews and conclude that these additional data on controlled and comparative studies do little to enhance the current empirical status of efficacy of behavioral marriage therapy and in no case do they establish the superiority of social learning approaches. It is concluded that an open mind to all sources of data on the efficacy of marital therapy is needed if the field is to make meaningful advances.

The Jacobson-Weiss critique of the Gurman-Knudson and Gurman-Kniskern discussions of behavioral marriage therapy (BMT), while scholarly, derives from such a narrow conceptual set that, with only minor exceptions, Jacobson and Weiss have failed basically to comprehend the essence of our theoretical and logical criticisms of BMT. Moreover, a careful reanalysis of the research cited by our critics as evidence for the efficacy of BMT reveals the strength of the empirical foundation of this approach to have been greatly exaggerated. While few of our concerns about a preominantly behavioral approach to marriage therapy have been ameliorated by Jacobson and Weiss' commentary, some success seems to have been achieved in terms of our original purpose: to stimulate critical thinking about the premises and process of marital therapy.