Current status of family intervention science.

Looking at the field as a whole through metaanalysis, Shadish et al concluded (based on 162 studies) that marital and family therapies were significantly more effective than no treatment and at least as effective as other forms of psychotherapy. Although these reviews and others are positive, individual studies raise many questions. For instance, based on research findings, family treatments increasingly have become standard care for patients with schizophrenia. It remains unclear what degree and type of family involvement is needed for which patients at which stage of their disorder. In the area of anxiety and depression, there are too few studies to make any strong conclusion. Although investigators such as Barrett, Cobham, and Diamond have produced some positive results, the Lewinsohn and Clark studies fail to demonstrate the added benefit of family involvement. Although Brent's study showed CBT to reduce depression faster, family therapy and supportive therapy did just as well in the long run, and family conflict was a strong risk factor for relapse. In the area of anorexia, Russell and Robins produced strong results from family interventions, whereas Geist found no difference between different types of family interventions. Family treatments for obesity have been inconsistent. In a metaanalysis of 41 studies, parental involvement did not contribute significantly to outcomes. In the Epstein study, however, which included 5- and 10-year follow-up, the results of family intervention were impressive. Although many of these studies can be cited for various methodologic flaws, the most consistent problem is that sample sizes are too small to detect difference between two or more active treatments. The most consistent findings (and most well-done, large studies) that support the efficacy of family-based interventions are done with externalizing problems. Work groups led by Patterson, Eisenstadt, Webster-Stratton, Alexander, and Henggeler all have produced impressive reductions of oppositional and antisocial behavior. Clinical programs that treat these populations without using a family-based intervention as at least a component of a treatment package are seriously ignoring the findings of contemporary intervention science. Programs of research by Henggeler, Szapocznik, and Liddle demonstrate similarly impressive results for substance abusing adolescents. Although preliminary results from the Dennis et al study suggest that various treatment approaches may benefit this population. Family interventions have had less success in reducing ADHD symptoms, yet these psychosocial treatments have been essential in reducing much of the family and school behavior problems associated with this disorder. Many investigators would agree that a combined medication and family treatment approach may be the treatment of choice for children with ADHD. In fact, many studies across various disorders suggest that patients respond best to comprehensive treatment packages, of which a family treatment is at least one component. Although the data are promising, many challenges lie ahead. Although collectively many family intervention studies exist, many disorders lack enough rigorous and large-scale investigations to make any strong conclusions. Kazdin argues that sample sizes of 150 are essential to detect significant differences between active treatments, and few of the reviewed studies include these kinds of patient numbers. Furthermore, not enough committed and sophisticated family treatment researchers have carried out some of the major studies. For example, the Brent study on depression and the Barkley study of ADHD, although testing family approaches, lacked well-developed and published treatment manuals, a demonstration of the necessary expertise to supervise these treatments, and data about training and adherence to these models. Although the absence of expertise limits investigator allegiance biases, treatment development and modification are essential for tailoring family treatments to target family processes specific to each disorder. Investigators such as Patterson and Liddle have invested great effort in rigorously dismantling the treatment process, identifying and refining essential ingredients, and repackaging more potent treatment protocols. This process has paid off well. Programmatic treatment development is needed for many disorders to address myriad questions. What are the essential disorder-specific family processes that should be targeted by interventions? Hostility, criticism, communication, attachment and autonomy, attributional sets, and behavior management are important processes of family life, but each may have more relative importance for specific disorders. With a greater understanding of these processes, treatments could be tailored to target these mechanisms more efficiently and effectively. (ABSTRACT TRUNCATED)