Lessons Learned in Performance Measurement

As scorekeeping begins for new Medicare penalties for hospitals with excessive numbers of patients returning shortly after they are discharged, a new Dartmouth Atlas Project report shows little progress over a five-year period in reducing these hospital readmissions and improving care coordination for Medicare patients. On the contrary, readmission rates for some conditions have increased nationally and for many regions and hospitals, including some of America's most elite academic medical centers. The report shows that roughly one in six Medicare patients wind up back in the hospital within a month after being discharged for a medical condition.

In an examination of the records of 10.7 million hospital discharges for Medicare patients, researchers found striking variation in 30-day readmission rates across regions and academic medical centers. Researchers also found that more than half of Medicare patients discharged home do not see a primary care clinician within two weeks of leaving the hospital, and that facilities and regions with general patterns of high use of hospitals for medical conditions were frequently the same places with high readmission rates, an indication that some communities are more likely than others to rely on the hospital as a site of care across the board.

"The report highlights widespread and systematic failures in coordinating care for patients after they leave the hospital," said David C. Goodman, MD, MS, lead author and co-principal investigator for the Dartmouth Atlas Project, and director of the Center for Health Policy Research at the Dartmouth Institute for Health Policy and Clinical Practice. "Irrespective of the cause, unnecessary hospital readmissions lead to more tests and treatments, more time away from home and family, and higher health care costs."

The readmission rate to a hospital is increasingly seen as a marker of a local health care system's ability to coordinate care for patients across care settings, and readmissions are often a sign of inadequate discharge planning and the lack of effective community-based care. Centers for Medicare & Medicaid Services (CMS) has estimated the cost of avoidable readmissions at more than $17 billion a year. In hopes of decreasing these costs, Medicare plans to reduce payments for readmissions, exposing hospitals to considerable financial risks. In fiscal year 2013, hospitals face a penalty equal to 1 percent of their total Medicare billings if an excessive number of patients are readmitted. The penalty rises to 2 percent in 2014 and 3 percent in 2015.

"The need to develop more efficient systems of care that include discharge planning and care coordination is clear," said Elliott S. Fisher, MD, MPH, report author and co-principal investigator of the Dartmouth Atlas Project and director of the Center for Population Health at the Dartmouth Institute for Health Care Policy and Clinical Practice. "The report shows the opportunity for improvement, and the importance of aligning efforts to reduce readmissions with other policy and payment initiatives."

Trends over time in 30-day readmission rates

Nationally, there was relatively little change in 30-day readmission rates from 2004 to 2009, regardless of the cause of the initial hospitalization. Readmission rates following surgery were 12.7 percent in both 2004 and 2009, while readmission rates for medical conditions rose slightly from 15.9 percent in 2004 to 16.1 percent in 2009.

Similarly, most regions across the country did not experience significant reductions in readmissions from 2004 to 2009. Readmissions decreased after medical discharges in 11 regions, with Bismarck, N.D. experiencing the largest decrease, from 16.3 percent in 2004 to 14 percent in 2009. There was an increase in readmissions in 27 regions, the highest in Aurora, Ill., which increased from 14.3 percent in 2004 to 18 percent in 2009. Readmission rates after surgeries varied as well, with 28 regions experiencing a decrease, most notably in Elyria, Ohio, which decreased from 19 percent in 2004 to 15.2 percent in 2009. White Plains, N.Y. was among the 18 regions with increases in readmission rates following surgical discharges, with an increase from 13 percent in 2004 to 17.4 percent in 2009.

Only seven academic medical centers had significant changes in 30-day readmission rates following medical discharge from 2004 to 2009. Northwestern Memorial Hospital in Chicago, Ill. showed the most decrease, from 19.9 percent in 2004 to 16.7 percent in 2009, while the University of Connecticut Health Center in Farmington, Conn. increased from 13.1 percent to 17.9 percent. Among patients discharged after surgery, 11 academic medical centers experienced significant changes in 30-day readmission rates between 2004 and 2009. The University of Missouri Hospital and Clinic in Columbia, Mo. decreased from 19.7 percent of patients in 2004 to 14.5 percent in 2009, while Montefiore Medical Center in the Bronx, N.Y. increased from 15.6 percent to 19.4 percent.

Regional variation in 30-day readmission rates

In 2009, the percentage of patients readmitted to the hospital within 30 days of an initial discharge varied markedly for both medical and surgical discharges across regions of the country. Among patients who first visited the hospital for medical treatment, 16.1 percent were readmitted to the hospital within 30 days. The highest rates occurred in Michigan, including Pontiac (18.9%), Royal Oak (18.8%), Dearborn (18.0%) and Detroit (17.9%), while far lower rates were found in Utah, including Ogden (11.5%), Provo (13.0%) and Salt Lake City (13.6%). For patients who were discharged from the hospital after having surgery, 12.7 percent were readmitted to the hospital within 30 days. However, there was more than twofold variation in these rates in regions across the U.S., from Rapid City, S.D. (7.5%) to the Bronx, N.Y. (19.0%).

Primary care follow-up after discharge

Overall, 42.9 percent of patients who were released to go home from the hospital after medical treatment had a primary care visit within two weeks in 2009. Patients in New Orleans, La. were far less likely to see a primary care clinician after discharge home, with 25.6 percent having a visit to a primary care clinician within two weeks of medical treatment in a hospital, compared to 61.4 percent of patients in Lincoln, Neb.

Among academic medical centers, the range of variation was somewhat higher. Less than 20 percent of patients discharged from New York University Medical Center in Manhattan, N.Y. saw a primary care clinician within two weeks of a medical discharge, while the rate was nearly three times higher at the Mayo Clinic's St. Mary's Hospital in Rochester, Minn.

These findings highlight the pervasive problems with patient care after hospital discharge, and underscore the importance of primary care systems in reducing avoidable hospitalizations. While there are many different reasons for higher readmission rates across regions and hospitals, prior research has documented the failings of current care coordination and the high proportion of readmissions that can be avoided by improving care.

"It's very important that patients and health care providers communicate clearly so that all questions are answered and everyone understands what will happen when the patient leaves the hospital," said Risa Lavizzo-Mourey, MD, MBA, president and CEO of the Robert Wood Johnson Foundation, a longtime funder of the Dartmouth Atlas Project. "Everyone—patients, doctors, nurses, caregivers—has a role to play in ensuring quality care and avoiding another hospital stay. They need to work together to create a plan for how care will proceed when the patient returns home. This should include a clear understanding of the patient's medical problems, a schedule for follow-up appointments, a list of medications and instructions for taking them."

As part of the Care About Your Care initiative, the Dartmouth Atlas Project and Robert Wood Johnson Foundation have co-produced a companion to the report with tips for patients when they leave the hospital. Additional information can be found at www.dartmouthatlas.org.

Researchers studied 100% of fee-for-service Medicare beneficiaries with full Part A and Part B coverage during the study periods. Hospital claims from short-term acute or critical access hospitals were identified among the study population for each cohort, with the first period of index discharges as July 1, 2003 to June 30, 2004 and the second as July 1, 2008 to June 30, 2009. Because of the way hospitals are paid under Medicare in Maryland, readmission rates for Maryland hospital referral regions were suppressed. Data was adjusted for differences in age, sex and race.

About the Dartmouth Atlas Project
For more than 20 years, the Dartmouth Atlas Project has documented glaring variations in how medical resources are distributed and used in the United States. The project uses Medicare data to provide information and analysis about national, regional, and local markets, as well as hospitals and their affiliated physicians. This research has helped policymakers, the media, health care analysts and others improve their understanding of our health care system and forms the foundation for many of the ongoing efforts to improve health and health systems across America.