Building Programs by Outcomes (Not Fidelity to Guidelines)

Most products, clinical or otherwise, are designed and produced with detailed plans. A researcher once gave me an outline for a different approach to ensuring the quality of psychotherapy-based offerings. He proposed that the best way to build a quality program, using the most effective licensed therapists, is to retain those with positive outcomes after protracted measurement of the results for all candidates.

He contrasted this outcomes-based approach with the more standard one based on training in an evidence-based practice like cognitive behavior therapy (CBT). The proponents of CBT mistakenly think clinical techniques are key to results, and so they strive to mold highly proficient CBT therapists through training and measurement of fidelity to guidelines. Clinical outcomes are seen as a byproduct of properly applied techniques.

The Medical Integration Problem

The need for guidelines is a question raised by efforts to place psychotherapists in primary care. This is especially relevant when the goal is to reach more patients with brief interventions. Are full-length therapy sessions needed for an impact? Most integration models today involve therapy as usual based on primary care provider (PCP) referrals. Can we facilitate having therapists engage a larger segment of primary care practices?

An alternative primary care model proposes therapists work more independently in this setting. Therapists might help patients with signs of depression or drinking, those with unhealthy behaviors, or those not adhering to physician recommendations. Talking with patients in the exam room for 10 minutes or less prompts the question of what clinical protocols they should follow in this brief time.

While the 50-minute hour is traditional, it is plausible to think a skilled clinician can make an impact in less time. The question is not whether the clinical impact is equivalent. This is not about replacing psychotherapy. It is an attempt to create something new that can reach more people. It is an attempt to take advantage of a non-stigmatizing setting that people ideally visit at least annually for routine care.

In this alternative model, the impact of brief interventions is potentiated by three other factors. The PCP is embracing a new model for collaborating with the therapist to establish a culture of health focused on behavior change. The PCP is learning to prompt and reinforce steps toward behavior change. Finally, the therapist engages other resources, as needed, like digital therapeutics and outside psychotherapists.

This clinical model relies on the therapist’s judgment to use brief interventions judiciously and to add resources as indicated and accepted by the patient. Clinical change is thus not only due to brief interventions by the therapist, but also to the clinical culture and the blend of services helping patients achieve behavior change goals. Maximizing the measurement of clinical change must also be a priority.

Measurement and System Improvement

The measurement of clinical results is primarily seen as demonstrating the value of services to key stakeholders. Measurement can also be a stimulus for action, especially when feedback on clinical results is given to participants. The gold standard for measuring change in behavioral healthcare is the patient self-report questionnaire. They should be completed routinely and continuously for best results.

One of the benefits of moving more healthcare activities to a digital format is that brief questionnaires can be completed online between office visits. This administrative process has several advantages. Completion rates will be better for a routine clinical process than a unique request, and feedback to patients on scores and interpretation is easily facilitated, along with graphs of change over time.

Clinical results are useful at the level of each patient, each clinician, and the practice as a whole. The clinician-level results are especially helpful in validating the work of therapists in an environment with few set protocols and maximum clinical judgment. Clinical measures with normative data can be used for gauging how each clinician is performing against a benchmark.

Scores on clinical measures tend to fall along a bell curve. While two-thirds perform within the middle range (or average performance), a minority are below or above average. Clinicians with average or above-average scores will benefit from feedback on their performance and the ability to continuously learn what works in this unique setting. Below-average clinicians will need more help to merit retention.

An outcomes-based system is not only a valuable way to identify poorly performing clinicians, but it is cost-effective. Measuring results is cheaper than designing guidelines, training clinicians, and tracking adherence. Clinical guidelines or best practices may emerge over time with such a model. Yet until that day arrives, it is possible to monitor patients and clinicians while respecting clinical judgment.

This model requires enthusiastic clinicians and executives. Clinicians need to trust and value clinical results, while executives must negotiate and sustain good partnerships with their counterparts in primary care. It is a flexible yet rigorous model. It can reduce stigma, close the treatment gap, and improve population health. Its clinical services are both innovative and empirically grounded.

Ed Jones, PhD, is senior vice president for the Institute for Health and Productivity Management.