Monitoring treatment progress with psychological assessment instruments can prove to be quite valuable, especially with patients who are seen over relatively long periods of time. If the treatment is inefficient, inappropriate or otherwise not resulting in the expected effects, changes in the treatment plan can be formulated and deployed. These adjustments may reflect the need for (a) more intensive or aggressive treatment (e.g., increased number of psychotherapeutic sessions each week, addition of a medication adjunct); (b) less intensive treatment (e.g., reduction or discontinuation of medication, transfer from inpatient to outpatient care); or (c) a different therapeutic approach (e.g., a change from humanistic therapy to cognitive-behavioral therapy). Regardless, any modifications require later reassessment of the patient to determine if the treatment revisions have affected patient progress in the expected direction. This process may be repeated any number of times. These in-treatment reassessments also can provide information relevant to the decision of when to terminate treatment.
Methods for determining if statistically and clinically significant change has occurred from one point in time to another have been developed and can be used for treatment monitoring. Many of these methods are the same as those that can be used for outcomes assessment and are discussed later in this chapter. In addition, the reader is also referred to an excellent discussion of analyzing individual and group change data in F. L. Newman and Dakof (1999) and F. L. Newman and Tejeda (1999).
Patient profiling is yet another approach to monitoring therapeutic change that can prove to be more valuable than looking at simple changes in test scores from one point in time to another. Patient profiling involves the generation of an expected curve of recovery over the course of psychotherapy based on the observed recovery of similar patients (Howard, Moras, Brill, Martinovich, & Lutz, 1996; Leon, Kopta, Howard, & Lutz, 1999). An individual recovery curve is generated from selected clinical characteristics (e.g., severity and chronicity of the problem, attitudes toward treatment, scores on treatment-relevant measures) present at the time of treatment onset. This curve will enable the clinician to determine if the patient is on the expected track for recovery through the episode of care. Multiple measurements of the clinical characteristics during the course of treatment allow a comparison of the patient's actual score with that which would be expected from similar individuals after the same number of treatment sessions. The therapist thus knows when the treatment is working and when it is not working so that any necessary adjustments in the treatment strategy can be made.
Aside from its obvious treatment value, treatment monitoring data can support decisions regarding the need for continued treatment. This holds true whether the data are nothing more than a set of scores from a relevant measure (e.g., a symptom inventory) administered at various points during treatment, or are actual and expected recovery curves obtained by the Howard et al. (1996) patient profiling method. Expected and actual data obtained from patient profiling can easily point to the likelihood that additional sessions are needed or would be significantly beneficial for the patient. Combined with clinician impressions, these data can make a powerful case for the patient's need for additional treatment sessions or, conversely, for treatment termination.
As well as the need for supporting decisions regarding additional treatment sessions for patients already in treatment, there are indications that patient profiling may also be useful in making initial treatment-related decisions. Leon et al. (1999) sought to determine whether patients whose actual response curve matched or exceeded (i.e., performed better than) the expectancy curve could be differentiated from those whose actual curve failed to match their expectancy curve on the basis of pretreatment clinical characteristics. They first generated patient profiles for 821 active outpatients and found a correlation of .57 (p < .001) between the actual and expected slopes. They then used half of the original sample to develop a discriminate function that was able to significantly discriminate (p < .001) patients whose recovery was predictable (i.e., those with consistent actual and expected curves) from those whose recovery was not predictable (i.e., those with inconsistent curves). The discriminant function was based on 15 pretreatment clinical characteristics (including the subscales and items of the Mental Health Index, or MHI; Howard, Brill, Lueger, O'Mahoney, & Grissom, 1993) and was cross-validated with the other half of the original sample. In both subsamples, lower levels of symptomatology and higher levels of functioning were associated with those in the predictable group of patients.
The implications of these findings are quite powerful. According to Leon et al. (1999),
The patient profiling-discriminant approach provides promise for moving toward the reliable identification of patients who will respond more rapidly in psychotherapy, who will respond more slowly in psychotherapy, or who will demonstrate a low likelihood of benefiting from this type of treatment.
The implications of these possibilities for managed mental health care are compelling. . . . [A] reliable prediction system— even for a proportion of patients—would improve efficiency, thereby reducing costs in the allocation and use of resources for mental health care. For instance, patients who would be likely to drain individual psychotherapeutic resources while achieving little or no benefit could be identified at intake and moved into more promising therapeutic endeavors (e.g., medication or group psychotherapy). Others, who are expected to succeed but are struggling could have their treatment reviewed and then modified in order to get them back on track. . . . Patients who need longer term treatment could justifiably get it because the need would be validated by a reliable, empirical methodology. (p. 703)
Intuitively, one would expect that patient profiling information would result in positive outcomes for the patient. Is this really the case, though? Lambert et al. (1999) sought to answer this question by conducting a study to determine if patients whose therapists receive feedback about their progress
(experimental group) would have better outcomes and better treatment attendance (an indicator of cost-effective psychotherapy) than those patients whose therapists did not receive this type of feedback (control group). The feedback provided to the experimental group's therapists came in the form of a weekly updated numerical and color-coded report based on the baseline and current total scores of the Outcome Questionnaire (OQ-45; Lambert et al., 1996) and the number of sessions that the patient had completed. The feedback report also contained one of four possible interpretations of the patient's progress (not making expected level of progress, may have negative outcome or drop out of treatment, consider revised or new treatment plan, reassess readiness for change).
The Lambert et al. (1999) findings from this study were mixed and lend only partial support for benefits accruing from the use of assessment-based feedback to therapists. They also suggested that information provided in a feedback report alone is not sufficient to maximize its impact on the quality of care provided to a patient; that is, the information must be put to use. The use of feedback to therapists appears to be beneficial, but further research in this area is called for.
Notwithstanding whether it is used as fodder for generating complex statistical predictions or for simple point-in-time comparisons, psychological test data obtained for treatment monitoring can provide an empirically based means of determining the effectiveness of mental health and substance abuse treatment during an episode of care. Its value lies in its ability to support ongoing treatment decisions that must be made using objective data. Consequently, it allows for improved patient care while supporting efforts to demonstrate accountability to the patient and interested third parties.
Was this article helpful?