In Reply: Dr. Salzer contends that the positive relationship between patient satisfaction and treatment gains is due to shared method variance. He concludes that there is not any evidence to accept the concept of consumer satisfaction as having face validity. We disagree with this conclusion for a number of reasons.
We found not only a strong relationship between self-reported patient improvement but also significant negative correlations between satisfaction and certain diagnoses, namely personality disorders and posttraumatic stress disorder. These diagnoses were determined by clinical staff using DSM-IV criteria. We also found that patient satisfaction was related to preadmission adjustment and whether the patient had been employed. We also did a blind chart review with two different clinicians to make judgments about patient satisfaction with service. Unlike Dr. Salzer, we believe the above methodology utilizes different measures of functioning, and therefore the results are not due to shared method variance.
Dr. Salzer also tries to explain away these findings by arguing that the study population was older and treated in a VA hospital setting, and therefore the findings are not generalizable. We have replicated these findings with younger patients being seen in an outpatient university setting (1). Dr. Salzer further argues that these findings are unique, which is simply not so (2).
At the heart of this disagreement is a philosophy that patients do not know what is good treatment. Dr. Salzer argues that patient satisfaction is not a proxy measure of treatment effectiveness and does not promote consumer input. Our research and experience shows otherwise. Patient satisfaction data are being used in many hospital settings in quality improvement efforts to increase consumer involvement in different aspects of care decisions. It is time that behavioral health providers understand that it is not just our expertise that is important, but also whether our services meet customer needs. Dr. Salzer's position is a step backward for behavioral health care.