Better picture of patient experience requires more data, wiser use of it

The focus for competition in healthcare is moving inexorably toward meeting patients' needs with greater efficiency. As logical as this goal might be, measurement of the outcomes that matter to patients is an imperfect science, and improving those outcomes is a difficult challenge.

In this context, some clinicians have expressed concerns that measurement of patient experience might have perverse unintended consequences—and actually worsen care.

They argue that sample sizes are too small and somehow biased, and that patients do not understand real quality, and might push for more tests and medications, thus raising costs. For many issues, these critics note, the individual physician is not the right unit of analysis. They assert that incentives to improve on these measures might encourage suboptimal care, and that the pressure created by these measures drives physicians into retirement.

I recently asked one angry but thoughtful emergency medicine physician if he knew of any doctors who had a personal financial incentive to change patients' pain scores. Or if he knew any doctors who actually gave narcotics or antibiotics to every patient who asked for them in an attempt to raise their patient-experience scores. Or if he knew of any physicians who had left the practice of medicine because they believed these measures were causing them to do the wrong thing. He agreed that he could not demonstrate that any of these assertions were true—but, he said, “It feels like they are true.”

I get that. It's not easy for good, hard-working clinicians to take criticism. It's also true that data obtained directly from patients are always incomplete, and that every quality-related measure—even mortality—could have perverse consequences if carried to an extreme.

So how do we resolve the tension between our need to improve and the imperfections in the data? The answer is improving the measures, getting more data and using the data appropriately.

“Hard” clinical outcomes such as death and complications are the most important measures, but there are other issues that matter to patients. Only they can tell us if their needs are being met. Beyond physical pain, there is confusion about what is happening next in their care, concern that clinicians are not talking to each other, and anxiety about whether their real concerns are being heard.

My colleagues and I describe unmet needs as “suffering,” because that word better reflects what patients experience than “satisfaction.” (“Patient” comes from the Latin word for suffering.) Still, we have much to learn about measuring trust, fear, peace of mind and management of hope—aspects of care that are often considered part of the art of medicine, but could be made more reliable.

If being graded is painful, being graded on a small sample of patients is enraging. To drive improvement rather than generate rage, we need more data. The closer we come to collecting patient-experience data on every patient, the better.

More data requires going electronic. Start with getting e-mail addresses and surveying patients after every doctor visit and hospitalization. With this approach, clinicians get feedback from hundreds of patients, not a few. Even if those patients may never be a perfectly representative sample, the data become more compelling.

Since measures will never be perfect, and data will always be incomplete, provider organizations must be thoughtful about how they use the information. The issue should always be, “How do we get better than we currently are?”—not, “Who do we punish or shame?”

When data are used to drive improvement rather than to punish, methodological demands are lower. Less time is spent arguing about sample size and risk adjustment, and more time is spent on how to improve.

Financial incentives are a double-edged sword. They attract attention but generate endless questions about the measures, the thresholds, the sample size, and whether patient mix has been fully taken into account. These questions have no satisfactory answers. Therefore, peer pressure and public transparency are increasingly how many organizations are trying to drive improvement.

Measurement of patient experience is imperfect and always will be. But the right course is to make measurement better and more complete and use the information wisely.