“Compare the quality of your local hospitals,” urged the ad that ran recently in newspapers around the country. The ad encouraged consumers to visit CMS’ Hospital Compare Web site and get answers to questions like “What percentage of patients always received help when they wanted it?”
Motivating patients to care about hospital quality seems intuitively like a positive step, but according to experts, the statistics like those promoted in the advertisement raise new issues for hospitals and hospitalists as administrators grapple with whether and how to use the surveys to measure physician performance.
Although widespread physician reimbursement for satisfaction may be still in the future, hospital administrators are already feeling the financial pressures, experts said. “I think increasingly you're seeing hospital executives actually having their compensation tied to patient satisfaction results,” said Joseph A. Miller, senior vice president for the Society of Hospital Medicine (SHM).
Experts predicted that patient satisfaction scores eventually will affect contract negotiations, incentive payments, bonuses and even the leadership of hospitalist staffs. Recent SHM surveys have found that about half of the hospitalist programs in the country have some of their compensation tied to quality metrics, and that statistic is likely to increase and include some patient satisfaction scores, Mr. Miller said.
With so much at stake, hospitalists and administrators are naturally cautious about how to proceed, and the data collected so far have generated significant controversy. Ongoing debates center on the significance of patient satisfaction measures, the validity of the surveys that collect the information and both the most likely and the most appropriate responses to the data.
Interest in patient satisfaction has been driven to a peak in recent months by CMS, which first released its Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) scores in March. The HCAHPS survey asks recently discharged patients to rate hospitals on 10 measures (see ), the results of which are updated quarterly on the Hospital Compare Web site.
The focus on patient satisfaction has deeper roots than just this new government effort, however, experts said. “The stars are sort of aligning. With consumer-directed health plans, more and more people are paying out of pocket. People are getting pickier about care,” said Michael B. Guthrie, MD, former hospital CEO and executive-in-residence at the University of Colorado-Denver's health administration program.
Third-party payers have been slowly but surely picking up on the public's interest, he added. “If you read the headlines, you see repeated examples of a health plan or CMS announcing some project or pilot where they're going to collect and measure patient satisfaction.”
“It started out as more of a patient advocacy issue. It clearly has moved into one of the key quality metrics,” said Deirdre Mylod, PhD, vice president of Press Ganey Associates, the largest vendor of patient satisfaction data. “Our highest reporting hospitals have been reporting patient satisfaction measures for years. Now most boards know that this is out in the community.”
That has put pressure on hospital administrators who are, in turn, looking to hospitalists as the most likely path to improving their satisfaction scores.
“You have a small cohort of doctors treating large volumes of their patients,” said Adam Singer, ACP Member, president of IPC The Hospitalist Company. “So you're starting to see hospitals wanting these scores to be improved by the hospitalists.”
An imperfect instrument
Dr. Singer is also a founder of the Phoenix Group, a think tank for private practice hospitalist companies, which recently studied the patient satisfaction issue. The group is issuing a white paper on the subject that expresses several concerns about such surveys.
“What HCAHPS studies is not so much whether [the patients] are happy with their doctor or not, but whether the patient was happy with the hospital experience. That is only in one part related to the hospitalist,” explained Dr. Singer.
Dr. Mylod agreed that the surveys' ability to assess hospitalists is limited. “You might want to know about the way you manage something, but that may not be the way that patients think about the world. Patients may not be able to separate out who the hospitalist is from other physicians who are involved in their care,” she said.
Hospitalists face additional disadvantages when their patients are compared with those cared for by other departments in the hospital. Hospitalists are more likely to care for patients who were admitted through the emergency department and those who have no primary care physician, both groups that tend to report lower satisfaction. Patients in the hospital for obstetrics, cardiology or orthopedics—all areas where hospitalists have little involvement—tend to have higher satisfaction.
“It's not fair to compare hospitalists' patients to all of the other patients. When you break out and look at the difference in specialties, the types of patients who would not get assigned to hospitalists tend to have higher satisfaction scores,” Dr. Mylod said.
It's also unfair to judge individual physician performance based on the survey data, said Dr. Singer. “There's absolutely no way to attach a physician to the data, but yet hospitals are doing that anyway. They are given the patient's name who filled out the survey and backing into who the doctor was,” he said.
Dr. Mylod acknowledged that the reverse engineering required by the current survey system is less than ideal. “Generally, you ask the patients about their beliefs or perceptions of the physician care and then what you do on the back end is code which of the patients had their care coordinated and overseen by hospitalists,” she said.
That system places a heavy reliance on perfect coding, which many hospitals lack. Research conducted by Dr. Mylod and a hospitalist at one facility found that about 80% of patients were coded correctly as to whether they received care from a hospitalist.
Picking out the performance of individual physicians would be even trickier, Dr. Mylod said, given the small numbers of surveys usually involved. HCAHPS requires hospitals to collect only 300 surveys a year, although many complete more than that.
The small numbers and narrow ranges of the surveys are another concern of hospitalists. “There in fact may be differences by doctor, but unless you've got some meaningful data that expresses that, I think it's unfair to focus on individual performance,” said Mr. Miller.
It's also difficult to respond to the small, individual performance statistics. “What do you want me to do—find the one person who didn't like me?” asked Dr. Singer. “It could really be one person who's unhappy that puts you in the first percentile.”
Because satisfaction scores tend to be fairly similar across the board, percentile ratings fluctuate dramatically on small absolute score differences, the physicians said. “It's kind of like being in a class where everybody got a 95 or above, so you got a 96 but you got a D,” said Ron Greeno, MD, chief medical officer of Cogent Healthcare and a member of the Phoenix Group.
Suspicion and concern
Dr. Greeno voiced some hospitalists' general suspicion of the patient surveying process. “We see funky scores being assigned to programs that are performing at a relatively high level. Let's at least make sure this is being done the right way, which means right now some things have to change, because that's probably not occurring,” he said.
It's the current practice, not the principle, of patient satisfaction surveys that they have concerns about, the critics noted. “I've never met a hospitalist who says ‘I don't want my performance measured.’ It's almost part of our field,” said Dr. Greeno.
But some aspects of the new measurement metrics are less of a natural fit for hospitalists, the experts noted, specifically the fact that the actions of anyone in the hospital—from nurses to janitors—can affect the patient's ratings and, thereby, the hospitalist's compensation.
“We're trained to be responsible for our own behavior, but we're not very comfortable with how our behavior is amalgamated with the behavior of others who are part of a team. I think there's going to be a lot of hand-wringing and dissatisfaction among physicians with any measures of this type because they are amalgams,” Dr. Guthrie said.
Some physicians may also doubt the importance of the surveyed measures as compared to other factors affecting patient outcomes. “You're training me to be an expert and now you're going to ask me to be nice about it at the same time? As physicians, we're not trained to deal with a lot of this stuff,” said Dr. Guthrie.
Hospitalists are also faced with competing quality priorities. “How important is it to have the doctor sit down versus better manage the medication reconciliation?” asked Dr. Singer.
Those goals don't have to be mutually exclusive, according to Dr. Mylod. “There's a lot of data to show that those things [satisfaction and other measures of quality and outcomes] are related. It's not that if you're nice to people, you're not also providing great clinical care,” she said.
Whatever their personal views on the value of patient satisfaction surveys, hospitalists will have to either get on board with the idea or risk negative consequences, the experts agreed.
“You end up being looked at as being very defensive if you start challenging the validity of the data instead of accepting it and trying to figure out how to improve. You end up in the mode where doctors get angry and frustrated, administrators get angry and frustrated, and the process breaks down,” Dr. Singer said.
He and other hospitalist leaders do hope to push surveyors, like Press Ganey, to develop methods that better assess the work of hospitalists. “There needs to be pressure put on the measuring organizations to create a survey process that ensures that the results actually reflect reality,” said Dr. Greeno.
They are also interested in alternate methods of assessing patients' satisfaction with their hospitalists. IPC and some other large hospitalist groups call every patient after discharge. “We explain who the doctor was in more detail and we try to identify any problems they're having and fix them, which we think leads to better patient satisfaction, and even if it didn't, it leads to better care,” said Dr. Singer.
At Brigham and Women's Hospital in Boston, in addition to standard patient surveying, the primary care physician who receives a discharged patient is surveyed. “If a patient's unhappy with our care or if there were any issues, generally that gets fed back to the primary care physician. It's sort of an indirect way to assess patient satisfaction,” said Sylvia McKean, FACP, medical director of the hospitalist program.
For most hospitalists, however, it makes more sense to work with existing survey programs in their hospitals than to “reinvent the wheel and design a new survey questionnaire,” said Mr. Miller. He recommended that every lead hospitalist get to know the person within the hospital who coordinates the surveys, which are usually conducted by an external vendor.
“You need to work with your contact in the hospital to reach the survey vendor and find out what type of reports they have available,” he said. For example, data that compare your patient satisfaction scores to nationwide rates for hospitalists would help to accurately assess your performance.
Once the data are collected, the next project is figuring out how to improve the results. “That takes some digestion and digging and becoming familiar with the instruments and what they measure,” Dr. Guthrie said. “And then asking fairly seriously of the teams that are at the bedside, ‘What do we do that might affect this? And what can we do that actually makes a difference?’ That requires some experimentation with different approaches.”
Good communication is likely to be a key to higher scores, said Dr. Mylod. “So much of what the patient believes about the people involved in their care comes down to the communication style. What words are you using? Are you giving them access to ask questions? They [hospitalists] can have a profound impact on what the patient thinks of them and what the patient thinks of the hospital.”
Cooperation among everyone in the hospital is also crucial, both to improving patient satisfaction and working out the issues around surveying, the experts agreed. “We as a collaborative team, doctors and hospital administrators, should work together to globally improve these numbers without penalizing one party or the other,” Dr. Singer said.