Are We Getting Better?
Psychotherapy Dose-Response Effect: A Clinician’s Comments


Hansen (2002) and his colleagues find that clinicians are unable to be as effective in treatment because they are deprived access to a sufficient number of sessions. Using clinical significance methodology they found that only 35% of patients were better at treatment completion, most often after only three sessions. While their recommendation to extend treatment length may be appropriate for many patients, treatment that patients perceive as successful can also be accomplished in fewer sessions for others. Before changing practices, we will need outcomes data that are individualized, subjective and identify specific markers of poor prognosis that can be easily assessed early in treatment. Lacking these data, in most cases, I still prefer to empower patients to decide when treatment is complete.

I was glad to discover additional evidence supporting the enterprise of psychotherapy and its effectiveness in this paper by Hansen, Lambert and Forman (This Issue). This adds to the substantial and encouraging empirical support for the role of psychotherapy in health care. I am afraid, however, that this particular study has little relevance to my practice. I am aware that many researchers and clinicians are lobbying payors and policy makers to provide access to longer-term treatment for their patients. In addition, they seek to persuade patients that there are benefits of continuing in treatment beyond the point at which these patients would otherwise choose to discontinue treatment. But I believe that universal application of this recommendation is ill advised. I consider my practice to be effective even though on average, I see my patients for fewer sessions than this study recommends.

I recognize that I may be in the minority among my psychologist colleagues in psychotherapy “dosing.” This possibility has prompted me to question my practice behaviors. However, while I could comment on my introspection in this regard, I believe that it may be more interesting to comment on why I do not intend to make significant changes in my psychotherapy practice despite these findings.

Since the average length of treatment of my patients is in the range of five to eight visits per episode of treatment, the paper’s findings would suggest that only about 20 percent of my patients are improved at termination. Of course, I do not perceive this to be accurate. Perhaps an analogy will help convey my experience at reviewing this paper.

If a travel agent told me that according to a national average, half of all travelers are in route 2.5 days but most of her clients traveled for 1.75 days, I would expect that she could explain to me how her work and/or clientele differ from the national averages. She might be arranging more flights than other modes of transportation relative to her peers. It might be that her clients’ destinations were closer than the national average. It could be that she plans only a portion of many of her clients’ trips where most agents plan whole trips. It would be silly of me to conclude that based on the national database, most of her clients were not reaching their destinations.

I believe that it is equally unreasonable to conclude that based on the outcomes literature, most of my patients do not reach their treatment goals because they are not remaining in treatment for 20 sessions. Might we instead need to inspect the length of treatment needed and offered in terms of the outcomes sought and desired rather than in terms of some assumed constant criteria of “benefit” for all patients? Perhaps by explaining the development of my practice behaviors I may illuminate what distinguishes the work of like-minded colleagues and myself in “mode” of treatment, patient characteristics and scope of treatment. This may also explain why I believe that my style of treatment is brief but effective.

Do I use a different “mode of transportation?” During my career I, like many, shifted from doing primarily insight-oriented therapy to more active cognitive and behavioral interventions. But maybe more importantly, over twenty years of clinical work, I have adjusted the style of my interactions with patients; by “style” I mean something distinct from drawing on techniques from schools of psychotherapy. One important example to highlight is that I now adjust treatment plans to the wants and expectations of my patients instead of imposing my treatment goals on them. Yet, I have become quite active, particularly in initial sessions, in order to clarify the goals of treatment and instill hope for change. I find it absolutely crucial to have agreement on clear, realistic goals at the outset of treatment. With this hope of, and focus on change, we engage in the most expedient process to meet these goals.

Many have had the experience of avoiding health practitioners due to concern that we will be pressured to engage in more extensive treatment than we desire. A few years ago I was resisting seeing a chiropractor for back pain for this reason. I had heard of friends getting hooked on chiropractic care. They described how it all began with an x-ray that showed previously undetected (and non-symptomatic) spinal abnormalities. These findings led them to engage in months of adjustments two to three times per week to correct these problems. In addition, they were encouraged to consume several expensive homeopathic remedies for holistic health. But my pain grew so I made an appointment. I was preparing myself to explain that I only wanted to address this pain, not re-orient my life. Imagine my relief when this doctor told me that no x-ray was needed and gave me an explanation of the problem and how it likely occurred. He had his physical therapist do a few pain-relieving modalities of treatment for me over a few sessions, then sent me off with some exercises. He instructed me to return if I did not continue to improve, predicting that I would soon be symptom free – and I was.

A good number of the patients I see approach psychotherapy as I approached chiropractic treatment. They are in distress, want resolution as quickly as possible and are concerned that therapists will not support this plan but instead direct them into meandering self-exploration. It is my experience that the “style” of treatment I have described more closely matches the desires, if not expectations, of most of my patients. It is not unusual for patients to report to me that they were pleased by the amount of progress they could make so quickly and relieved that I was not pressuring them to engage in an ambiguous long-term psychotherapy commitment.

Are we there yet? Early in my career, my colleagues and I lamented about patients who were therapy “dropouts.” We discussed how these patients were not ready for therapy and expected that they would make little progress toward their goals until they prepared themselves for psychotherapy. My understanding of these “dropouts” has changed. Supported by some anecdotal data, I believe that many very brief treatments are successful. But my definition of “success” has also changed. I believe that access to limited information or coping tools, and change in hopefulness can be sufficient without significant change in symptoms or life functioning.

In my experience, most people leave treatment when they reach a point where they have been restored to hope for self-sufficiency, or have made progress and come to a place of relative wellbeing (or believe they will not progress) and their evaluation of the costs of treatment come to be greater than the expected benefits of treatment. Some will accomplish this in only one visit. If a patient’s decision to terminate treatment clearly leaves them susceptible to a serious problem, I will carefully query their rationale. Otherwise, I will support their decision, review their accomplishments, predict the challenges they may face and let them know that I remain available to them as needed. I have found that giving patients maximum control of the discharge plan in treatment increases the likelihood that they will consider returning in the future as needed.

I often discontinue treatment at the point the patient comes to feel more optimistic and hopeful—even self-efficacious—about their problems. This can mean termination before symptoms have significantly changed and typically before a patient has reached all of her or his goals. If we have been successful in developing hope for change, if they have made significant progress toward their goals and if they have developed coping skills for self-sufficiency, I find that patients continue to progress toward their long-term goals for change without needing to meet with me on a regular basis. I don’t need to see everyone through to full functioning. When patients leave with hope, skills and future access to me and/or others as needed, most patients have a good prognosis for full recovery. Our aftercare plan always includes access to return to treatment on an as needed basis. To continue the travel analogy, I can get off the train in Denver; I don’t need to accompany them all the way home to New York.

Are the patients getting better? While I can describe my impressions of patient progress in my practice, we know that clinician perceptions of change often do not match reality very well. However, some of my practice is in a behavioral health group in California that psychologist colleagues and I founded five years ago. This group is composed of psychologists, psychiatrists and master’s level psychotherapists. We have a variety of primary modalities of psychotherapy represented, but all are focused on actively pursuing change for patients. One of our contracting agencies does collect outcome data on our patients and reports it to us.

Our average number of psychotherapy visits for this patient group is 6.7 visits. According to an outcomes instrument (adapted from the work of Dr. Lambert) our group is obtaining patient change scores that are above the aggregate expected level of change for all patients in the national database. For patients with more severe diagnoses, the group’s change scores have an even more positive deviation from the national averages.

Are these patients different in type, diagnosis and/or treatment expectations? We know that there are regional differences in length of treatment, patients in the West having shorter treatment duration than those in the East (Psychotherapy Finances, 2000). It is also true that our mode of treatment impacts the patient populations we serve. Some patients have self-selected to see other therapists because of their discomfort with a focus on specific treatment goals and expectations to make progress toward these goals. This reduces the representation of chronically ill and character disordered patients. As a result, the dose-response data from our group may be quite different from other psychologists’ practices and from this study’s patient samples.

All of the patients we see through our group are enrolled in a managed care company plan and are utilizing these benefits. The results reported by Hansen, Lambert and Forman suggest that managed health care companies authorize an inadequate number of treatment sessions. The above referenced survey data indicates that patients under managed care do tend to have shorter treatment duration than those who are in self-pay or indemnity programs (Psychotherapy Finances, 2000). While some might believe this confirms the conclusion that managed care companies do not authorize sufficient sessions to complete treatment, it is also possible that there are other factors that influence length of treatment. Perhaps briefer treatment better matches the expectations of patients and providers who are engaging in treatment covered by managed care companies. Most of the patients I see under managed care reimbursement have no expectation of receiving long-term psychotherapy treatment. Instead, most are seeking to resolve the problems they are currently facing as quickly as possible. Further, I expect that psychotherapists working actively with managed-care companies are more likely to be committed to brief treatment in their core beliefs regarding what is effective psychotherapy.

While there was a time when managed care reviewers denied access to adequate care, for a few years now, I have been able to obtain authorization from managed care companies for sufficient visits to complete treatment, even for greater than 30 sessions. While it does represent a hassle to me to have to convey information to justify the need for more visits, no longer am I refused additional visits.

What research data would be most useful to affect my treatment “dosing?” First, from my experience, too many patients are improved from single session encounters to ignore these data. Second, we know that many patients make progress after treatment has been discontinued. We need to measure progress at periodic intervals after treatment termination in order to assess the potential for continuing progress or decline. Of course, this would also be required to properly measure the outcomes for single session cases.

Finally, to influence my decisions regarding length of treatment, I would need to review research using outcome measures that are individualized and subjective. Individualized outcome measures would provide the opportunity to assess the outcome of treatment with regard to progress toward the specific goals identified by the patient versus comparisons with broad, universal treatment goals or scaled patient traits. Subjective outcome measures would allow the opportunity for the patient to report their personal perception of change and the sufficiency of that change in their life, versus assessing their level of functioning against normative data. Objective data on patient change allows us to make global determinations about treatment effectiveness. However, to assess optimal dosing of an individual’s treatment, patients’ own experiences are crucial. We need to know if in the perception of the patient, they made sufficient progress toward their personally identified goal(s). In other words, I trust patients to tell us when they are done.

I would welcome unobtrusive measures that identify critical patient markers, evident while in active treatment, that indicate poor ultimate outcomes. Indeed, in another study, Lambert and his colleagues reported a positive result from relaying such information to therapists (Lambert et al., 2001) though they only reported trends in functioning level from successive session measures. I look forward to having access to indicators of poor prognosis that would lead to prescriptive recommendations patients could embrace, increasing the likelihood of obtaining outcomes patients would consider a success. If I had these data, then obtained confirming experience in my practice, I would be quite willing to educate these patients regarding the likely benefits of continuing in treatment.

I applaud this study’s demonstration of the effectiveness of longer-term psychotherapy. If this research and extensions of it can increase access to higher “dosages” of psychotherapy by changing public policy and authorization criteria for psychotherapy management businesses, it appears many patients will enjoy better outcomes. Clinicians are deeply indebted to our academic colleagues for the creative and laborious work that has validated the enterprise of psychotherapy. Research is the foundation for our treatment. I appreciate the opportunity to take part in the dialog that will improve psychotherapy effectiveness research design and its application in practice.


Hanson et al

Lambert, M.J., Whipple, J. L., Smart, D.W., Vermeersch, D.A., Nielsen, S. L., Hawkins, E. J. (2001). The effects of providing therapists with feedback on patient progress during psychotherapy: Are outcomes enhanced? Psychotherapy Research, 11 49-68.
Psychotherapy Finances. (2000). 2000 Fee, Practice and Managed Care Survey, 10, (318), 10.