6 Reasons Why Patient Survey Data Doesn’t Tell the True Story

Share on FacebookTweet about this on TwitterPin on PinterestShare on Google+Share on LinkedInEmail this to someone

Perceptions of patients have changed. They’re not just “patients” anymore — they’re also “customers.”

And that small semantic difference has big implications for the healthcare industry. One of them is the abundant use of customer-satisfaction surveys.

Enshrined into official healthcare policy in 2010, patient satisfaction now carries serious implications for providers. Hospitals that fail to sufficiently satisfy their patients see significant cuts in their reimbursement for Medicare and Medicaid services.

Hospitals have reacted the way any business would when it faces a serious threat to its revenues. They have started to devote more time and more resources toward satisfying patients, hoping that these efforts will drive up volume, and secure reimbursement from the Centers for Medicare and Medicaid Services (CMS).

Predictably, patient satisfaction has become one of the most-discussed, most-debated, most consulted-on issues in healthcare. It’s become a hallmark metric of the age.

I’m not sure that’s necessarily a good thing. Here’s why…

1) The Surveys Have Problems

HCAHPS, the CMS-mandated survey, suffers from a dismal response rate — as low as a 20% national average. Low response rates contribute to statistical uncertainties. Smaller pools of numbers mean more extreme — and less accurate — representations of any given provider’s reviews within the community.

Additional problems stem from self-selection. Certain types of patients who choose to complete surveys have an outsized say in patient feedback. For example, older, sicker patients who see female providers are far more likely to respond to surveys than their peers. So, male providers who see young patients will see depressed survey responses, and get less accurate feedback.

And that’s if a facility administers surveys in a uniform way across the board for every patient. The truth is, they tend not to. The way that surveys are given and assertiveness in follow-up frequently change, driving down the reliability of these surveys even more.

2) Patients Are Terrible at Evaluating What’s Important in Healthcare

A 2017 JAMA Study confirmed that patient ratings of specialist physicians have zero correlation with objective measures of quality, or with peer assessment of performance.

Worse, a UC Davis study found that some patients have it precisely backwards. The study found that the most satisfied patients are more likely to be readmitted, have steeper costs, and are more likely to die in the hospital than their unsatisfied peers.

If this wasn’t enough to confirm that patients don’t know best, last year The Atlantic dove deep into the patient satisfaction responses to see exactly what patients were concerned with. Turns out, their complaints fixated on issues that ranged from petty and irrelevant — “This hospital doesn’t have Splenda” — to the tragicomically callous — “My roommate was dying all night, and his breathing was very noisy.”

Perhaps this is what comes when patients are treated as “customers.”

3) Surveys Lead to Misallocated Hospital Resources

But really, hospitals have no choice. If they want patients to come, and if they want to fend off CMS penalties, they have to keep patients satisfied.

This explains the rise of valet-parking, custom-order room-service meals, flat-screen TVs in every room, and special lounges for privileged loyalty patients.

These frills might make a hospital stay more pleasant. It’s hard to argue, however, that they do much for quality of care. Are they really the best possible use of a hospital’s limited resources?

What if the money spent on fancy new TVs went toward hiring new nurses instead? How much would that improve employee morale and, ultimately, care outcomes?

4) Surveys Degrade Doctors and Nurses

This points to another major problem — that the emphasis on the customer comes at the expense of clinical workers.

Many doctors and nurses, already overburdened, must now go through exercises to develop their ‘customer service skills.’ While there’s some merit in focusing clinician attention on a patient’s experience, some of these programs are downright demeaning. “Scripting,” for example, entails nurses and physicians rote-memorizing pat responses to patient complaints.

And if they deviate from script, the consequences can be severe. Health systems have started using patient satisfaction scores as a factor in calculating salaries for doctors and nurses. As if the stresses of clinical work weren’t demanding enough!

Small wonder that 78% of clinicians report that these surveys negatively affect their job satisfaction — with 28% of them saying they’ve considered hanging up their stethoscopes for good.

5) Surveys Can Sabotage Care

The effects of patient surveys can even interfere with care itself.

Sometimes, patients can have impulsive demands that don’t at all relate to their treatment. Satisfaction surveys give doctors and nurses an incentive to pander.

This has started to affect clinical decisions. Doctors, for example, may be willing to give patients a medically unnecessary test, or a certain drug, in order to keep the patient satisfied, even if the patient doesn’t need them. One striking example: in 2014, 48% of doctors reported prescribing inappropriate narcotic pain medication, because they had patient satisfaction scores in mind.  

Joshua Fenton, who conducted the UC Davis study above, points out that this isn’t how it should work. “Patients should be satisfied with their physicians, but ideally it’s because their physicians guide them toward the best care,” he says, not just because the physician gives patients what they want.

6) Surveys Distract from the Metrics that Matter

The Atlantic piece concludes with a bombshell about patient satisfaction. The author, Alexandra Robbins, used objective outcomes data from Medicare to find some of the country’s worst-performing hospitals. These facilities produce more mortalities, more readmissions, and more serious complications than their peers.

“And yet,” Robbins writes,  “two-thirds of these poorly performing hospitals scored higher than the national average on the key HCAHPS question; their patients reported that ‘YES, [they would definitely recommend the hospital].’”

This is the real problem of patient satisfaction.

HCAHPS scores divert attention from the real criteria we should judge healthcare systems by. At the end of the day, healthcare is about outcomes, not about ‘customer service’, and the metrics we use to evaluate hospitals should reflect that.