First-year results are here: P4P demonstration shows savings
HHAs report positive feedback, improvement in quality
More than $15 million in savings was shared with 166 home health agencies throughout the country as a result of performance in the first year of the Centers for Medicare & Medicaid Services' Home Health Pay for Performance (P4P) demonstration.
The demonstration project, which started in 2007 and ended in December 2009, was designed to show the impact of financial incentives on the quality of care provided to patients in traditional fee-for-service Medicare and the overall Medicare costs. A total of 570 agencies volunteered to participate in the program and were randomly assigned to either a treatment group or control group. Savings in the program are shared with treatment group agencies that either maintained high levels of quality or made significant improvements in quality of care. Performance was measured using seven home health quality measures computed from the Outcome-Based Quality Improvement data set and publicly reported on the Home Health Compare website.
After evaluation of the first year's data, three of the four regions of the country showed a decrease in Medicare cost of care, says Henry Goldberg, BA, principal associate of health policy for Cambridge, MA-based Abt Associates and director of the pay-for-performance demonstration project. "Top performers in the treatment group in the three regions that showed a savings shared the $15 million," he explains. These incentive payments are based on savings in the first year of the project, he points out. "Claims data for the second year savings analysis were not available until August, so I hope to have the report and savings calculation completed in the fall," he adds.
Each agency's results in the quality measures are compared to results of other agencies in the treatment group within the region to identify recipients of the bonuses, explains Goldberg. "Incentive payments do seem to have an impact, even though the savings represented in this demonstration project is slightly less than 1%," he says. Even a 1% savings throughout all of home health Medicare spending can be significant, so Goldberg believes that the project is showing positive results for P4P. "We'll know more once the data for year two is evaluated, because we will see if agencies continue to find ways to reduce the cost of care and improve quality," he points out.
While Abt Associates is evaluating cost savings and identifying high performers within the treatment group, a group at University of Colorado, Denver, is comparing the treatment group results to the control group to identify best practices and to clarify the effect on staffing and agency costs as a result of P4P.
"There were no huge differences in the outcomes or the approach to care between the treatment and the control group, but this is not a classic treatment vs. control group study," explains Eugene Nuccio, PhD, senior instructor in the division of healthcare policy and research at the University of Colorado, Denver. "All of the agencies participating in the project volunteered and were most likely high-performing agencies with an emphasis on quality improvement before the project," he says.
Agencies were randomly assigned to the control or the treatment groups, and the treatment group of agencies did not receive any additional information throughout the project than the control group. "We told the treatment group agencies that they would receive a bonus payment if their region reduced costs to Medicare, and if they were high performers in the quality categories," Nuccio says. "We did not maintain continuous contact with agencies in the treatment group to let them know how they were doing, and we did not share best practice ideas with them," he adds. Feedback from agencies does indicate a desire for regular communication throughout the process, so they have an opportunity to identify areas in which they need to improve in a timely manner, he says.
Low-cost impact on agencies
In surveys of 41% of the treatment group agencies and 36% of the control group agencies, Nuccio discovered that concerns about the burden that P4P will place on agencies are largely unfounded." Cost impact of P4P implementation for agencies in the project [was] minimal," says Nuccio. More than 53% of the agencies reported cost increases of less than 1%, and slightly more than 22% reported increases between 1% and 5%, he says. Staffing also was not an issue, with the increase in staffing similar between the two groups. "One negative issue identified by agencies is the availability of physical and occupational therapists in their areas," he says. Because the early involvement of therapists in a home health patient's care can lead to better outcomes, home health managers want to add therapists to their staffs but have difficulty finding them, he explains.
Although outcomes for both groups were not significantly different, there were some differences in approach, says Nuccio. Survey results show that agencies in the treatment group emphasized:
enhanced communications among patients, physicians, and staff;
prevention and screening program;
telehealth;
changes in visit patterns.
Agencies in the control group emphasized:
staff productivity;
changes in patient visit mix;
wound care protocols;
electronic communications with discharge planners.
In addition to surveying almost 220 participating agencies, Nuccio's group also conducted focus groups with management and clinical staff of two highly performing agencies in each of the four regions for a total of eight agencies or 92 individuals. The focus group interviews were designed to identify the specific quality activities that high-performing agencies use to achieve superior outcomes, explains Nuccio. The agencies involved in the focus groups had a number of strategies in common, he says. These included:
strong, effective, purposeful leadership;
single, integrated system approach;
strong quality culture;
multidisciplinary teams with a patient-centric perspective.
Technology was often mentioned as a factor in improving cost-effectiveness and outcomes, but the decision to purchase technology was made with the goal of improving patient care, says Nuccio. For example, wound care is one area of technology that has far-reaching effects, he points out. Although the number of patients with wounds for any agency may not be high, a wound care patient requires intensive care and frequent visits, he explains. Any dressings, or wound care systems or protocols that reduce the number of visits or enable a nurse to care for complex wounds, can produce better outcomes for the patient and reduce costs, he says. "Several agencies mentioned that they control the costs of educating staff members on new wound care protocols by having the manufacturer's representative conduct the training," he says. "This is a cost-effective way to obtain training, and you know you are getting a qualified trainer," he adds.
Although the bonus payments were an incentive for agencies to perform well, Nuccio points out that not all agencies stressed the bonus to their employees. "There are some high-performing agencies that have not told their employees that they are in the treatment group and eligible for the incentive bonus," says Nuccio. "At these agencies, and many others in the demonstration project, it is just business as usual, with a continuous effort to provide good patient care," he says. "In general, most of the agency managers have the attitude that doing the right thing for the patient from the start of care not only improves outcomes, but is more cost-efficient anyway."
Sources
Henry Goldberg, BA, Principal Associate, Health Policy, Abt Associates, 55 Wheeler Street, Cambridge, MA 02148. Telephone: (617) 349-2482. E-mail: henry_Goldberg@abtassoc.com.
Eugene Nuccio, PhD, Senior Instructor, Division of Health Care Policy and Research, University of Colorado, Denver, 13611 East Colfax Avenue, Suite 100, Aurora, CO 80045-5701. Telephone: (303) 724-2479. E-mail: Eugene.nuccio@ucdenver.edu.