A VIEW FROM THE OFFICE
IS PATIENT SATISFACTION A GOOD THING FOR PATIENTS?
Almost all health care organizations are obsessed by patient satisfaction as an outcome measure. But think about it. Does a satisfied patient necessarily mean that the care was either good or effective? Patient satisfaction is well known to reduce law suits (if they like the doctor they are less likely to sue him/her), but it is also well known that patients who have liked their doctor over time are less likely to sue even if an egregious mistake was made. Patient satisfaction is really just a subjective measure of a brief interaction. It is more sensitive to process problems like long waits, rude staff, doctor in a hurry, etc., which cumulatively can undermine the quality of care. But actually there is very little a patient can discern about whether his/her doctor was right, efficient, or cost-effective in his/her care. So is it enough just to measure the subjective quality of the experience? Do we have to give up on trying to understand anything about the 'quality' of the care? These questions are worth much further thought. This recent article provides that and comes to a surprising conclusion.(1)
These authors conducted a prospective cohort study of 51,946 adult respondents to the 2000-2007 national Medical Expenditure Panel Survey. Patient satisfaction was assessed using 5 items from the Consumer Assessment of Health Plans Survey. The positive correlate of patient satisfaction (in the highest quartile group) was a lower odds of any emergency department visit (OR 0.92; 95% CI, 0.84-1.00), which is a good thing. On the other hand, however, this group had higher odds of an inpatient admission (OR 1.12), 8.8% higher total health expenditures, 9.1% higher prescription drug expenses, and higher mortality (OR 1.26, 95% CI, 1.05-1.53).
These findings are not completely out of line with limited prior research. Among a vulnerable older population, patient satisfaction had no association with the technical quality of geriatric care;(4) other evidence suggests that satisfaction has little or no correlation with Health Plan Employer Data and Information Set (HEDIS) quality metrics.(2,3) [Original References 3, 7, and 8] There are other data to suggest that physicians whose compensation is more strongly linked with patient satisfaction are more likely to deliver discretionary services (e.g. advanced imaging for acute low back pain).(5)
It is not certain what is going on here. Physicians may be just trying to keep patients happy by ordering tests, medications, and hospitalizations that are not really necessary. On the other hand, sicker patients may develop closer long-term, more satisfied, relationships with physicians who order appropriate tests and services. But the 26% greater mortality in this study strongly suggests that this is not an idle or trivial question to pursue. We don't want to kill our patients with kindness.
The authors comment: "In the ideal vision of patient-centered care, physicians deliver evidence-based care in accord with the preferences of informed patients, thereby improving satisfaction and health outcomes, while using health resources efficiently. However, patient-centered communication requires longer visits and may be challenging for many physicians to implement...Relaxing patient satisfaction incentives may encourage physician to prioritize the benefits of truthful therapeutic discourse, despite the risks of dissatisfying some patients."