One of the key weaknesses of the “consumer-driven health plan” movement is that consumers have little access to quality information about individual doctors.
Now a team of researchers led by Dana Gelb Safran, a researcher at a health institute affiliated with Tufts-New England Medical Center in Boston, has published results of a study suggesting that quality evaluators can get reasonably reliable physician quality data from a collection of 45 completed patient satisfaction survey questionnaires and very reliable quality data from a collection of about 300 patient questionnaires.
The Safran paper, which deals only with patient satisfaction and not with measures of the technical quality of care, such as how likely comparable patients are to die or end up in the hospital, appears in the January issue of the Journal of General Internal Medicine.
The researchers drew on a sample of 13,000 adult patients of 215 “generalist physicians,” or primary care doctors, at 67 practices in Massachusetts.
The researchers ended up with an average of 58 completed questionnaires per physicians. Questions asked about matters such as satisfaction with physician referrals and the politeness of patients’ own doctors’ office staff.
The researchers used a tool called the “Spearman Brown Prophecy Formula” to estimate roughly how well correlated the average patient response was for each satisfaction measure and about how many responses would be needed to get reliable results for each physician.
The researchers also used another statistical tool, the Central Limit Theorem, to estimate how many physicians might end up getting lower grades than they deserved.
The results suggest that members of Massachusetts Medicaid managed care plans are about as happy with their doctors as members of the state’s widely respected commercial plans are, and that the Medicaid plan members are actually happier with their doctors’ efforts to “treat the whole person” and promote wellness.
The results also suggest that patient satisfaction depends mostly on the physicians and the specific offices that the physicians practice in, and very little with the health plans that the patients use to pay for the physicians’ services, Safran and her colleagues write in their paper.
Safran and her colleagues concede that the variables included in the questionnaire seemed to be responsible for only part of patients’ satisfaction, but they argue that focusing on those variables in an effort to improve the quality of care is legitimate.
Questioning whether physicians should focus on traits such as politeness of office staff or clarity of physician communications “seems analogous to questioning whether clinicians ought to focus on particular known clinical factors, such as blood cholesterol levels, even when these factors only account for a modest share of disease risk,” the researchers write. “As with health, the influences on quality are multifactorial. Because there is not likely to be a single element that substantially determines any dimension of quality, we must identify factors that show meaningful influence and are within the delivery system’s purview.”
Average performance scores in this study varied by about 20 points on a scale of 100, and that suggests that physicians could make meaningful improvements in care simply by working to narrow the difference, the researchers write.
The researchers estimate collecting the kind of data they collected would cost an average of just 50 cents per adult patient for each round of surveys.