Published in the April 2005 issue of Today’s Hospitalist
Editor’s note: An article in the March 2005 issue of Today’s Hospitalist profiled three hospitals that have launched pay-for-performance programs. In this month’s issue, we focus on strategies and practical considerations to build these programs.
With so much attention being paid to the pay-for-performance movement, hospitalist programs everywhere are beginning to feel pressure to explore ways to pay their physicians for quality. The trouble is that quality incentive programs raise so many questions “What measures do you focus on? How do you track and collect data? What’s a fair way to split bonuses among physicians? “that it can be difficult to know where to begin.
Several developments, however, promise to make it easier to develop quality incentives for physicians. Here’s a look at some of those trends, along with some advice from hospitalists who have already begun building pay-for-performance programs.
Standardized performance measures
If you’re thinking of creating a pay-for-performance program, start by taking a look at the long list of performance measures endorsed by the National Quality Forum (NQF). The organization has endorsed nearly 60 measures that assess the care of a wide range of conditions, from heart failure and pneumonia to acute myocardial infarction and surgical infections.
Steven A. Nahm, vice president for physician services at The Camden Group, a Los Angeles-based consulting company that has helped create and set up numerous quality incentive programs for hospitalists, says these measures are an excellent blueprint.
“In the first six to 12 months of a new program,” he explains, “we look at the measures used by organizations like the Centers for Medicare and Medicaid Services (CMS), the Joint Commission, and the National Quality Forum (NQF).”
The NQF reviews performance measures developed by other groups and endorses those that meet its criteria. Other large organizations like Medicare and the Joint Commission on Accreditation of Healthcare Organizations have taken these measures and incorporated them into their quality incentive programs.
Why are national performance measures so important? If you use these measures in creating pay-for-performance programs, you stand to save a lot of time when it comes to collecting “and reporting “data.
If you use an NQF-endorsed performance measure on pneumonia care, for example, you’ll be collecting data in a format that will likely be recognized by Medicare and the Joint Commission. When those organizations ask for data showing how you care for pneumonia, you’ll already be collecting that information.
Even better, as other health plans and insurers start demanding information on how you care for pneumonia, many will use NQF-endorsed measures. If you’re already using an NQF measure “or one from Medicare or the Joint Commission “as your guide, you should be able to give other payers the same data you’re already collecting.
Mr. Nahm points out that while there is no shortage of interesting goals to include in a quality incentive program, “You’re very limited by the administrative ability to track and report that information. You have to be able to track and record, and do it without too much of an administrative burden.”
There is another bonus to using NQF-endorsed measures: By collecting the same information as other hospitals, you can compare your results to institutions around the country.
Cliff Hall, MD, director of the inpatient care service at Good Samaritan Regional Medical Center in Corvallis, Ore., says that’s a big advantage of the pay-for-performance program Mr. Nahm helped his six-physician group create two years ago.
“We can compare our data to the performance of other hospitals, including the family physicians and internists who are not part of the hospitalist program,” Dr. Hall explains. “We can also compare our performance to hospitals nationwide.”
Even if you start with measures that have been endorsed by the NQF and are being used by Medicare, you still have many performance measures to choose from. How do you decide exactly which measures you’ll track?
Mr. Nahm says that besides looking at NQF-endorsed standards, he also asks whether the state quality improvement organization is collecting any performance data. If your hospital is already collecting information on heart failure patients, for example, it makes sense to include a version of that measure in your hospitalist quality incentive program.
Another guiding principle of quality incentives programs is to keep it simple. You don’t want to pick goals that are so ambitious that your program fails to meet them “and your physicians receive no reward for their hard work. Instead, you want to walk a fine line of setting a goal that physicians can meet “but not too easily.
“Your target should be ambitious yet attainable,” says Winthrop Whitcomb, MD, founder of the hospitalist service at Mercy Medical Center in Springfield, Mass. “You don’t want the bar to be so high that there is no chance of attaining the bonus, yet you don’t want it to be so low that the incentive program does not result in meaningful change.”
When the hospitalist program at Mercy rolled out its first pay-for-performance program for hospitalists two years ago, Dr. Whitcomb looked at both high-volume diagnoses and NQF-endorsed measures being used by the Joint Commission, which are known as “Core Measures.” And while he decided to focus on the Joint Commission’s measures, Dr. Whitcomb implemented a stripped down version of the measures to help make sure the program’s physicians met the goals.
The Joint Commission’s performance measure for pneumonia care, for example, asks for data on a wide range of indicators. Among other things, the measure calls for examining how well physicians assess pneumonia patients’ oxygen level, how often they administer a pneumococcal vaccine, whether they draw blood cultures and administer appropriate antibiotics within four hours, and whether they counsel patients on smoking cessation. Mercy’s quality incentive program, however, focused on a much narrower “and less ambitious “goal: screening for and giving the pneumococcal vaccine to appropriate patients with pneumonia.
Mercy similarly streamlined the Joint Commission’s measures for heart failure patients. The hospitalist program tracked how often hospitalists gave ACE inhibitors to heart failure patients with a left ventricular ejection fraction of less than 40 percent, and how often they assessed left ventricular function in patients diagnosed with heart failure. Dr. Whitcomb decided to not focus on the two other indicators that are part of the Joint Commission’s heart failure measure.
Eliminating some quality indicators is only one way to simplify the goals of a quality incentive program. The pay-for-performance program for hospitalists at Good Samaritan incorporates more elements of the NQF-endorsed measures examining hospitalists’ care of stroke, heart failure, AMI and CAP. Those four measures, however, account for only 20 percent of the program’s bonus payment. If hospitalists don’t ace the clinical performance measures, they can make it up in other areas.
Dr. Hall explains that the other 80 percent of the hospitalists’ quality bonus comes from measures of what he called “more mundane” tasks, such as whether hospitalists complete admission notes and discharge summaries in a timely fashion, satisfaction surveys of patient and staff, participation in key hospital committees, and participation in quality improvement activities.
One of physicians’ biggest concerns about pay-for-performance programs is whether the incentive they receive is truly a bonus, or whether hospitals are simply taking money out of their paycheck and giving it back to them “but only after they’ve met a new set of conditions.
Mr. Nahm suggests that hospitals fund a bonus pool through hospital funds, not by taking money from physician payroll. There is generally not enough money in the payroll fund to create a significant bonus, he explains. “If you want to recruit good hospitalists,” he adds, “you need those funds to pay a good base compensation.”
He suggests that hospitals set aside a small amount of money, maybe $30 per case, and put it into a bonus pool. The program at Good Samaritan, which Mr. Nahm helped create, puts aside $25 for every patient treated by the hospitalists employed by the hospital.
Mr. Nahm says that hospital administrators typically understand the importance of not pulling money for bonus pools from the funds used for hospitalist pay. “They understand that if they cut salaries,” he explains, “they won’t attract hospitalists.”
He also notes that because many hospitalist programs already withhold a small amount of physician pay for productivity incentives, it’s often difficult to back out any more money for other initiatives.
Bonus payments raise another question: How do you divide money from the bonus pool? Dr. Whitcomb, who is also director of clinical performance improvement at Mercy, says there are key differences between awarding productivity incentives, which reward individual performance, and quality incentives, which reward the performance of the overall group. That’s why he splits quality bonuses evenly among the hospitalists in his group.
“For quality bonuses,” Dr. Whitcomb says, “you don’t have to worry about RVUs and productivity. You live and die by the group’s performance, because quality is a group effort. Productivity, on the other hand, is a completely different issue, so it should be more individually based.”
In addition, the hospitalist program at Mercy paid its quality bonus every six months. “You want to pay out the bonus frequently enough to change behavior and reinforce positive behavior change,” Dr. Whitcomb says, “so you don’t want to pay it out once a year.” He also wanted to make sure that the patient population on which the bonus was calculated was big enough to make for a meaningful analysis.
Giving everyone the same dollar amount encourages teamwork, says Dr. Hall from Good Samaritan. “The group has to meet all the performance criteria,” he explains. “If we have an outlier who has a significant problem in terms of getting dictations done or giving aspirin for acute coronary syndrome, that person’s performance affects the entire group.”
Measures to avoid
Finally, Mr. Nahm says there are some areas you should avoid when building a quality incentive program. He notes that hospital administrators sometimes want to include readmission rates, length of stay and resource utilization in quality incentive programs. But hospitalist programs, Mr. Nahm says, should try to keep those factors out of the pay-for-performance picture.
In his view, it simply doesn’t make sense to reward “or penalize “hospitalists for readmission rates. “Quite often the ability of a hospitalist to control or influence readmission rates is limited,” Mr. Nahm explains. “So much depends on patient compliance and the extent to which primary care physicians and other specialists follow up after the patient has been discharged.”
Instead of tying a quality bonus to readmission rates, Mr. Nahm suggests using a less quantitative measure, like requiring that hospitalists lead or serve on a committee that will identify ways to reduce readmission rates.
“Maybe you find a way to give underserved patients access to outpatient pharmaceuticals or make sure that good discharge processes are in place to make follow-up visits with a primary care physician or specialist,” Mr. Nahm says. “Those are the types of things that we want to make sure happen, and you can have hospitalists champion that cause.”
Hospital administrators sometimes want to reward hospitalists for gains they make in reducing length of stay and resource utilization. Mr. Nahm points out, however, that any program that shares this type of savings with physicians is considered gainsharing and is illegal under Medicare. He adds that Medicare has recently endorsed several non-hospitalist gainsharing arrangements, though under very strictly and narrowly defined rules.
Legal or not, tying physician bonuses directly to financial gains makes many physicians nervous. That’s why hospitalists like Dr. Hall suggest using incentives to tackle quality, not financial, issues. While those efforts may help reduce overall costs, physicians should worry about the care of patients, not the finances of the hospital.
“Nothing in our program focuses on reducing length of stay or reducing service,” Dr. Hall says. “Everything is quality-driven. No one can look at this and say you’re being rewarded for withholding care or speeding up care and discharging people sooner. Everything in the program is beneficial to both the system and the patients.”
Edward Doyle is Editor of Today’s Hospitalist.
For more information
A complete list of quality indicators being used by the CMS and Joint commission, along with more details on these measures, is online. Click on the link for Specifications Manual for National Hospital Quality Measures.