HCAHPS Report Card: A+ or C-?
Robin Young • Thu, April 6th, 2017
In 2015 Medicare paid $1.4 billion to hospitals based on the 32 question Hospital Consumer Assessment of Healthcare Providers and System (HCAHPS) survey.
Of the roughly 3,500 hospitals participating in the HCAHPS program, more than 1,600 got a check—and about 1,900 didn’t.
For trivia buffs…the state with the highest average HCAHPS score is South Dakota.
The worst is…Washington, DC. The District of Columbia has, by far, the worst HCAHPS scores in the United States. Next worse is Nevada.
The HCAHPS patient survey has measured thousands of hospitals, tens of thousands of physicians and nurses and millions of patients since its national roll out eight years ago.
Recently some critics have argued that asking certain questions can have an untended and potentially harmful effect on physicians or patients.
Others have argued that richer hospitals, that can afford to invest in patient-care systems, are snaring a bigger share of the incentive dollars while hospitals with tighter budgets are structurally at an HCAHPS disadvantage.
The “HCAHPS” Effect
Funny thing about measuring people. They tend to interact with the survey process.
In a Western Electric factory in 1924 just outside of Chicago, the plant’s manager commissioned a study to measure the effect of lighting on worker productivity. He hypothesized that better illumination would increase worker productivity.
By increments, the researchers turned up the lights. And production increased.
Could, wondered the researchers, the opposite be true? So they turned down the lights. Productivity, however, kept rising. Even when the workers were in near darkness their production rose.
This effect, which has become known as the “Hawthorne Effect” refers to the confounding that occurs in a study when the consequences of subject’s performance affects what the subject does. In the case of this Western Electric plant, the workers liked the attention from the researchers. The study itself transformed an otherwise tedious job into something meaningful and interesting.
HCAHPS as U.S. Healthcare’s “Hawthorne Effect”
Like the Western Electric factory workers of 1924, healthcare providers are acutely aware they are being measured and questions have emerged as to whether the act of being measured is itself affecting physician, nurse and staff behavior.
One example is HCAHPS two pain questions—# 13: How often was your pain well controlled? And #14: How often did hospital staff do everything they could to help you with your pain?
In July 2016 Centers for Medicare and Medicaid Services (CMS) announced a proposal to remove pain management questions from the HCAHPS survey to eliminate financial incentives for overprescribing pain medication. This proposal was included in an overall program to fight opioid abuse.
Apparently, CMS worried that clinicians feel pressure to overprescribe opioids because scores on the HCAHPS survey pain management questions are tied to Medicare payments to hospitals.
But, of course, the point of the HCAHPS and its system of financial incentives and penalties is to change behavior.
So, how’s the HCAHPS survey doing?
If users were to give it a mid-term grade, what would it earn?
Dr. Paul Cleary
To get an answer we hiked up to New Haven and interviewed Yale’s recently retired Dean of the Yale School of Public Health, Paul Cleary, Ph.D. Dr. Cleary is currently Director, Center for Interdisciplinary Research on AIDS (CIRA) and the Anna M. R. Lauder Professor of Public Health.
Dr. Cleary, who has toiled in the fields of patient-care surveys for 30 or 40 years had a fast response to the HCAHPS grade question:
“It gets an A+”
“The first goal of HCAHPS was to create awareness of the importance and relevance of the patient experience. I don’t mean the food, I don’t mean what they thought about the parking or the signage…but were they treated appropriately.”
“For example, were they told the purposes of their medication?”
“So, HCAHPS is not about whether a patient is happy or not happy (everyone’s unhappy with the food). It’s about, as even the gruffest orthopedic surgeon would say, that ‘my work isn’t going to have the impact it should have if the patient doesn’t understand why they’re taking medicine, anti-coagulant or rehab.’”
“Goal one of the HCAHPS survey is to make patient centered care part of the discourse. Some people hate it. Some people love it. But patient centered care has become one of the cardinal aspects of high quality care.”
“I would say that Goal One has been a success.”
Patient Satisfaction vs Quality of Care
Cleary also made the point that HCAHPS is designed to drill down into the nature and quality of communication between care providers and their patient and to, hopefully, present hospitals with actionable information.
“If HCAHPS tells a hospital ‘17% of your patients said that no one explained the purpose of their medications’ the hospital administrator doesn’t have to be a psychometrician to know if that’s a good number or a bad number.” Said Cleary. “If you think 17% is a bad number, then you figure out what you want to do about it.”
Cleary also pointed out that several studies of patient assessments of care have shown that patients tend to rate their experience higher based on the nature and content of provider communication more so, in fact, than with the particular treatment itself.
“To my knowledge, there is no published evidence indicating that providers obtain higher CAHPS scores by providing inappropriate care.”
Telling Patients to Stop Smoking
One test of the notion that physicians may receive higher patient assessment scores by soft pedaling unpleasant advice is the Winpenny smoking study.
About a year ago, a research team led by Eleanor Winpenny, Ph.D., a fellow at the UK Centers for Research Consortium, looked at 26,432 completed surveys by smokers aged 65+ to see if there was a connection between telling patients to stop smoking and their responses to a survey very much like HCAHPS.
The results of Winpenny et al.’s study was published in Health Service Research on April 2016.
Somewhat counterintuitively the investigators found that 11 of 12 patient experience measures were significantly more positive among smokers who were always advised to quit smoking than those advised to quit less frequently.
Quoting directly from the paper: “There was an attenuated but still significant and positive association of advice to quit smoking with both physician rating and physician communication, after controlling for other measures of care experiences.”
Conclusion: Telling patients to stop smoking will tend to result in higher patient assessment scores.
HCAHPS Effect on Hospitals
There have been hundreds of studies about patient assessment surveys. Here are two which are fairly representative of the overall conclusions so far.
The first study (Accelerating Improvement and Narrowing Gaps: Trends in Patients' Experiences with Hospital Care Reflected in HCAHPS Public Reporting) by Marc Elliott et al., looked at surveys from 4.8 million adult inpatients and 3,541 U.S. hospitals.
They found that HCAHPS scores increased by 2.8 percentage points from 2008 to 2011.
Their conclusion was that, five years after HCAHPS public reporting began, patient’s hospital care had improved—especially among initially low-scoring hospitals. The greatest improvement was in for-profit and larger (200 or more beds) hospitals.
Another study, this time by Claire Senot, Ph.D., assistant professor of management science at the A. B. Freeman School of Business at Tulane University analyzed six years of data from 3,474 U.S. acute care hospitals and found that, together, patient-centered care and adherence to evidence-based standards of care were associated with lower readmission rates.
1% Increase in HCAHPS = 5% Reduced Readmissions
Senot and her colleagues found that a 1% increase in the HCAHPS score was associated with an almost 5% decrease in readmissions, although both patient experience scores and adherence to guidelines were independently associated with higher costs per discharge. They also examined communication and response to patient needs separately and found that the higher the quality of technical quality, the lower the cost of improving the quality of communication.
They suggested that hospitals can reduce the trade-off between reducing readmissions and controlling costs by giving a higher priority to aspects of quality such as communication.
CMS and Poorer Hospitals
Because of CMS’ concerns over HCAHPS’ pain questions the survey’s designers are redrafting those questions. Having said that, however, a couple of recent studies (Maher et al. 2015; Nota et al. 2015) found that higher CAHPS scores are associated with less, not more, opioid use. Also, one has to wonder why any orthopedic physician would cut that particular corner at all.
A tougher question, however, is the issue of patient mix and financial resources.
Hospitals with more disadvantaged patients have tended to receive lower care experience scores. One of the most common explanations for this is that such hospitals have fewer resources to improve care than hospitals that treat a different mix of patients.
Indeed, a recent analyses of trends in HCAHPS scores found that larger for-profit hospitals improved more than other hospitals (Elliott, Cohea, et al. 2015).
One of the strategies to address this issue is the Medicare HVBP (Hospital Value-Based Purchasing) program which assigns points for improvement as well as performance, although as some commentators have written, it is still not enough to compensate for their limited resources.
HCAHPS has put patient-centered care into the fabric of U.S. orthopedic care.
The survey itself, which took decades to design, is delivering actionable information to hospitals. Changing healthcare quality is extremely difficult. But the early signs are that this information is nudging hospitals to improve patient care and thereby reduce readmissions.
What’s HCAHPS’ grade?