The effectiveness of psychiatric interventions

¶ … Psychiatric and Psychotherapeutic Treatment

The effectiveness of psychiatry and psychotherapy has made the word treatment become a buzz word among those in the health care industry. Clinical researchers study outcome to determine treatment effectiveness. Health care payers and Behavioral Managed Care Organizations (BMCOs) are interested in outcome research in order to establish an accountable basis for making decisions about resource allocation. (Wiens, 1994, p. 46) And not only that, the general public has become more educated about treatment options and they want to see evidence that treatment is working and is appropriate for their individual circumstances. In addition, large companies want to see evidence that treatments for psychiatric and substance abuse problems work. In short, there are a lot of people interested in knowing which therapy works and why.

Research over the past forty years has established that psychotherapy works; indeed "it seems that psychotherapy is one of the best documented medical interventions in history." (Barlow, 1996, p. 1050)Two articles in the British Journal of Psychiatry address the related questions of clinically relevant outcome research in individual psychotherapy and assessment and treatment selection for psychotherapeutic treatment. While they both assert that psychotherapy "works," they state that this is akin to saying antibiotics "work." "It doesn't make sense to set up an efficacy horse race between therapies that treat different conditions, and such comparisons should be taken with a grain of salt." Clearly, then, the task has been to elucidate what kind of psychotherapy or psychiatric treatment works and for what disorders. (Barlow, 1996, pp. 1050-1059)

Prevalence of Mental Illness

One of the most remarkable events in psychiatric health care in the United States was the national deinstitutionalization of people with chronic mental illness (CMI). The number of patients in psychiatric hospitals declined from 557,000 in 1955 to approximately 112,000 in 1988. (Barlow, 1996, p. 1052)Although thousands of long-term patients left psychiatric hospitals, deinstitutionalization by itself failed to meet its proponents' high expectations. Many discharged patients developed a pattern of frequent readmissions for brief stays, whereas others were reinstitutionalized in nursing homes, boarding homes, and other facilities that were ill-equipped to deal with the unique needs of the increasing numbers of patients with CMI.

"Medical necessity refers to an evolving set of clinical criteria driven by level of pathology and functioning which attempts to guide the delivery of psychiatric treatment." (Weins, 1994, pp. 43-47) Medical necessity directs treatment toward "returning patients to reasonable levels of functioning." (Weins, 1994, p. 51) It seeks to ensure that treatment is delivered by licensed mental health professionals for a health-related (DSM IV) diagnosis, in a manner consistent with current professional standards and practice. It seeks to ensure that treatment is adequate and essential at a level of care corresponding to the severity of the impairment. To a somewhat lesser extent it targets prevention of relapse and recurrence. "What medical necessity does not strive for is therapeutic perfectionism, or so-called "definitive treatments," or changes in underlying theoretical constructs as the primary goals." (Weins, 1994, p. 53) So, while the last century saw the deinstitualization of many mentally ill patients, it didn't' necessarily see them leave with effective treatment.

Dissemination Research

Obtaining data that can tell care providers exactly what treatments to use with exactly what people is a challenge. One that involves determining what factors influence the recognition of disorders in people who are seen in clinical settings. Previous research has indicated that providers do not always do well in recognizing mental disorders and then treating them accordingly. A second aspect concerns identifying factors that influence the actual delivery and adherence to a treatment service once the disorder is recognized. (VandenBos, 1996, pp. 1005-1006)

It is not only necessary to deliver a specific treatment but to implement it correctly and ensure that the person receiving the treatment complies with the treatment approach. Studies have shown that this often does not occur. Basic behavioral research has much to offer in understanding influences on human behavior including, for example, "the expectations of families and consumers regarding treatment and the value they attach to any given treatment." (Oss, 1995, p.4)

How do we know if a new psychiatric treatment works? There are two primary considerations in assessing therapeutic value. First, are there alternative explanations for observed clinical changes. Establishing causal effect requires randomized, controlled, double-blind trials. In treatment research, these efficacy trials delineate specific treatment benefits. However, "internal validity is only half of the battle." (Oss, 1995, p.4) Although efficacious in clinical trials, psychotherapeutic and pharmacological treatments often do not fare well in actual clinical practice. "This requires measures of external validity, or as some term it, transportability, demonstrating that treatment benefits generalize to an actual clinical population." Besides specific treatment benefits, other extraneous variables (e.g., compliance, diagnostic accuracy, and treatment accessibility and acceptability) affect outcomes. Studies that assess transportability are referred to as effectiveness studies. "The term effectiveness has also been used to refer to studies of long-term functional and health outcomes as well as to estimates of resource utilization and costs." (Oss, 1995, p.4)

To improve the average level of treatment outcomes, we must first measure treatment quality and factors that affect quality. Most current approaches for evaluating quality can be classified according to whether they examine treatment outcomes or the process of care. Outcomes include important patient characteristics, such as quality of life and satisfaction with treatment. Processes of care consist of the treatments that patients actually receive, such as the antipsychotic medication and dosage prescribed by a psychiatrist. Studying patient outcomes can identify domains in which patients are doing poorly. For example, patients may have poor social or vocational functioning, high rates of substance abuse or frequent psychotic relapses. (VandenBos, 1996, pp. 1005-1006)

Psychotherapeutic Treatments

In the practice guidelines proposed by the American Psychiatric Association, psychotherapeutic management and psychotherapy are key components in any patient's treatment. A psychiatrist who proposed medications without psychotherapy would probably be stepping in the quicksand created by the managed care companies' concept of "medication management." (Barlow, 1996, p. 1050)

As we advance our knowledge of the best psychiatric interventions, we continue to evaluate the efficacy of psychotherapies for major depressive disorders. This evaluation can be complicated by several problems, as pointed out in the APA Practice Guideline for the Treatment of Patients with Major Depressive Disorder (1999): "For some types of psychotherapeutic interventions, very few clinical trials have been conducted." It is also difficult to truly compare treatment effect sizes because studies have compared psychotherapies with a variety of treatments (placebo, antidepressants, no control group and so forth), and studies may have poor protocols/methodologies. (APA, 1996, pp. 373-375) However, these studies do tell us some of what we need to know about the effectiveness of psychotherapy.

McKay and colleagues (2000), in challenging the four beliefs that create the basis for the psychiatric treatment of depression, present those beliefs in their most rigid forms in order to attack them. They suggest that psychiatry believes that antidepressants are: conclusively more effective than placebos; safe, with minimal side effects; necessary to redress a chemical imbalance caused by a genetic predisposition; and more effective than psychotherapy, especially for severe or recurrent depression. (McKay, 2000, pp. 70-78) Investigators have used four complementary strategies to prove antidepressants' effectiveness:

1. Comparing patients with bipolar disorder (both with and without psychotic features) with schizophrenic patients to assess the structure and function of various brain structures, circuits and neurotransmitter systems;

2. Investigating circuits implicated in the generation and maintenance of normal emotions to determine whether the circuits are dysfunctional in people with mood disorders;

3. Assessing brain regions involved in the pathology of localized disorders in which comorbid mood changes are often observed (e.g., Huntington's disease, stroke and so forth); and

4. Measuring regional metabolic blood flow or neurotransmitter/receptor changes as provoked by biochemical challenge tests or as resulting from the treatment of mood disorders by mood-stabilizing or antidepressant drugs. (McKay, 2000, pp. 70-78)

Another meta-analysis investigated whether combined psychotherapy and pharmacotherapy is superior to either treatment alone for outpatients with depression. The researchers reviewed 17 controlled studies (N = 1,009) reported between 1984 and 1994. In the analysis, studies were given different weights on the basis of the scientific quality of the design, and these weights were multiplied by weights based on the outcome of the study. The results indicated that combined active treatments (drug plus psychotherapy) were appreciably (53% of the weighted evidence) more effective than minimal contact plus placebo and moderately superior to pharmacotherapy alone (29% of the evidence) but only slightly superior to psychotherapy plus placebo (19% of the evidence), psychotherapy alone (18% of the evidence), or pharmacotherapy plus minimal contact (15% of the evidence). In other words, 82% of the weighted evidence indicated no advantage of combined treatment over psychotherapy alone. (McKay, 2000, pp. 70-78)

A close inspection of the data shows that, of the 4 studies that used a combined behavioral plus drug condition in comparison with a behavioral plus placebo medication, 97% of the evidence indicated no significant difference. Interestingly, 3% of the evidence favored the behavioral intervention when combined with the placebo rather than the tricyclic medication. (McKay, 2000, pp. 70-78)

Another meta-analysis reviewed nine randomized controlled studies (N = 542) that directly compared cognitive therapy and tricyclic medications in the treatment of nonbipolar depressed outpatients. On the basis of their analysis, they concluded that (a) cognitive therapy appears to be roughly comparable to medication in the treatment of the acute episode; (b) combined cognitive therapy and drug treatment does not appear to be clearly superior to either modality alone, although trends of potential synergistic enhancement justify additional studies with larger samples; and (c) treatment with cognitive therapy (with or without drugs) during the acute episode appears to reduce the risk of subsequent relapse after termination. (McKay, 2000, pp. 70-78)

Different types of psychotherapy may have different outcomes. The studies support the superior efficacy of combined psychotherapy and drug treatment. Interpersonal psychotherapy focuses on clarifying and resolving current interpersonal difficulties related to the depression. "All of the treatment conditions produced better outcomes than the nonscheduled control group; interpersonal therapy outperformed drug treatment on adjustment measures (e.g., mood, apathy, suicidal ideation, work, and interest), whereas the drug treatment was superior on vegetative symptom (e.g., sleep and appetite) measures." (McKay, 2000, p. 78) On the basis of the differential impact of the single treatments, these investigators concluded that the combined treatment outcome was additive, even though the combined treatment condition was not statistically superior to the single treatments on any outcome measures. (Barlow, 1996, pp. 1050-1052)

It is possible that the type of psychotherapy may make a difference. Psychodynamic insight-oriented treatment of depression has performed relatively poorly in several studies. The recent multisite National Institute of Mental Health (NIMH) collaborative study on the treatment of depression has been cited to suggest that drugs are superior to psychotherapy in the treatment of severe depression. Despite media reports to the contrary, drugs were not significantly better than either of the psychotherapies with severely depressed patients on any measures. (McKay, 2000, pp. 70-78) "Cognitive therapy was ranked best on 11 of the 13 outcome measures reported in the published tables." There was a slight advantage of the psychotherapies over drug treatment with the milder depressions. The treatment outcomes were not statistically different in cases of severe depression. There did appear to be a reduced risk for relapse among the cognitive -- behavioral therapy patients. Of all patients entering treatment, the cognitive -- behavioral condition had the highest percentage of patients recover, the highest percentage of patients recover without a subsequent major depressive relapse, and the highest percentage of patients recover without major depressive relapse or treatment seeking. (McKay, 2000, pp. 70-78)

Some investigators have argued that the relatively high relapse rate after drug treatment indicates that depression should be treated like a chronic medical disease requiring ongoing, long-term, high-dose medication treatment indefinitely. This logic appears tautological: Drug treatment results in a higher relapse rate than cognitive -- behavioral therapy; therefore, patients should be maintained on drugs to prevent relapse. Because psychotherapy alone is not offered to patients initially, the maintenance phase of treatment is essentially restricted to drug responders. Therefore, patient samples in these drug maintenance studies should not be considered representative of the general population of depressed patients. (Inglehart, 1996, pp. 343-350)

Conclusion

One might legitimately ask how medications have become the predominant treatment in the United States despite the considerable risks and side effects and despite the considerable evidence demonstrating the efficacy of psychotherapy. Factors contributing to high use of antidepressant medications include the higher rate of reimbursement by some third-party payers for medical interventions relative to psychotherapy (usually 80% vs. 50%) and the pressure from some managed care organizations to use a seemingly quicker, apparently cost-effective drug treatment. However, cognitive -- behavioral treatments appear to be quite effective when delivered in a group format, providing a safe, time-efficient, cost-effective alternative to standard individual drug treatment, especially when side effects, dropout rate, and long-term outcome are taken into consideration.