By Brenda L. Mooney
WASHINGTON — When it comes to hospital stays, longer is not necessarily better in reducing 30-day readmission rates.
That’s according to a review of records for the more than 4 million patients hospitalized at 129 acute care VA hospitals from 1997 to 2010. For the study, published recently in the Annals of Internal Medicine, investigators from VA’s Health Services Research & Development (HSR&D) service examined trends in hospital length-of-stay (LOS), 30-day readmission rates and 90-day mortality for all medical diagnoses combined, as well as five specific common diagnoses: heart failure, chronic obstructive pulmonary disease (COPD), acute myocardial infarction, community-acquired pneumonia and gastrointestinal hemorrhage.
Their research determined that, while LOS decreased by 27% in adjusted analysis over the time period, the relative readmission rates did not rise. In fact, they decreased by 16%.
Furthermore, all-cause mortality at 30 and 90 days decreased by about 3% annually during the 14-year span.
The greatest LOS reductions over the 14-year period were for acute myocardial infarction, 2.85 days, and community-acquired pneumonia, 2.22 days.
Readmissions, meanwhile, dropped from 22.6% to 19.8% for acute myocardial infarction and 17.9% to 14.6% for COPD.
“This indicates that LOS reductions have not, thus far, adversely affected the likelihood of hospital readmission,” according to the authors.
The VA’s success with lowering LOS can be a valuable example for non-federal hospitals, according to the authors, because, “hospitals are driven to improve efficiency as skyrocketing costs have outpaced reimbursement and face incentives to reduce LOS as part of the Medicare prospective payment system, diagnostic-related group — based third-party payments, and closed system HMO care models.”
Readmission within 30 days can be costly, according to the study background, which notes that, in studies of Medicare patients, 30-day readmission rates range from 8% to 21%, depending on diagnosis, and cost about $17.4 billion annually.
Finding the appropriate length-of-stay can be a delicate balance, however.
The study found that patients with longer LOS also have higher readmission rates—a 3% increase for each additional day of stay—which is likely to be due to unmeasured severity that affects both readmission and LOS.
In addition, “there is emerging concern that excessive length of stay (LOS) reductions may be harmful because discharge before medical stability can result in hospital readmission or use of emergency department services,” according to HSR&D.
Hospitals with mean risk-adjusted LOS lower than the average across all hospitals posted a higher readmission rate, according to the research, with a 6% increased risk for each day lower than expected.Reasons for LOS Improvement
Study authors note several possible reasons that LOS improved at the VA without adversely impacting the 30-day hospital readmission rate.
One possible explanation is inefficiencies in care at VA facilities, according to the report, which cited a 1982 General Accounting Office report suggesting that 43% of days in VA hospitals were “medically avoidable.”
The authors also noted internal studies finding that reduction in LOS in VA hospitals had not kept up with private-sector hospitals, although it was improving.
In a more positive take, the study cited VA’s Flow Improvement Inpatient Initiative, begun in 2006 to improve hospital flow.
“Although the focus was on inpatient flow, these efforts may have resulted in
changes in transitions of care that also improved readmission rates,” according to the study, which also made note of VA’s “extensive medication reconciliation efforts at the time of hospital transitions of care, an initiative shown to reduce readmissions.”
The use of a hospitalist model of care at the VA also could have contributed to the better outcomes, according to the authors. The study pointed out that, while only 10% of VA hospitals employed hospitalists in 1997, more than 65% did so by 2007.
“The use of hospitalists for inpatient care has been shown to reduce LOS by up to 15% and improve other measures of quality, although no studies have reported reduced readmissions,” according to the study.
While hospital readmission rates can be a valuable metric in some ways, according to an accompanying editorial written by two researchers from the Center for Health Services Research in Primary Care at the Durham, NC, VAMC, “hospital readmission rates may be a poor measure of quality of care because of the complexity of factors that cause them and the poor correlation among those factors.
“It’s not only the quality of care during the index hospitalization or the quality of the handoff to post-discharge care that influences readmission rates,” writes Eugene Z. Oddone, MD, MHSc, of the Duke University School of Medicine and Morris Weinberger, PhDm, of the Gillings School of Global at the Public Health, University of North Carolina at Chapel Hill, NC. “Rather, many important factors affect when and how often patients are hospitalized, including access to post-discharge care, ability to purchase evidence-based medications or services prescribed at discharge, disease and disease severity, socioeconomic status, community resources, and social support.”
Study authors agreed that LOS and readmission rates alone do not provide a complete picture.
“Recent discussions about proposed modifications to the Medicare prospective payment system, such as payments by acute care episode (that is, bundled payments), raise important questions about identification of appropriate rates of hospital readmission,” they write. “For chronic illnesses, such as COPD and cancer, repeated hospital admissions may represent appropriate care. Even patients with terminal conditions who are enrolled in hospice care are readmitted for symptom management to improve quality of life.
“Thus, it is neither possible nor desirable to expect complete elimination of 30-day readmissions, and efforts to reduce readmissions may have unintended consequences. Reporting hospital readmission rates should be coupled with rates of return to the emergency department, use of home care services and outpatient clinic care to identify shifting of resource use.”
Accounting for nearly a third of all cancer diagnoses, prostate cancer is the most frequently diagnosed cancer in the VHA, where past research has suggested that the malignancy is caught earlier than in other healthcare systems.
In the past five years, 10 new system therapies have been approved for renal cell carcinoma (RCC), the most common type of kidney cancer.