Causes for Hospital Readmissions of Patients with Cancer
Chicago, IL—Hospitalizations and readmissions add substantial costs to healthcare. The annual cost of 30-day hospital readmissions in the United States is estimated to be $16 billion. In addition, bundled payment models may eliminate additional payment for readmissions in a specified period after discharge.
Several poster presentations at ASCO 2013 explored factors associated with readmission or unplanned hospitalizations in patients with cancer.
Cancer site, Medicare severity diagnosis-related group (MS-DRG), admission status, length of stay, and payer status are significantly associated with readmission in patients with gynecologic cancer, found Kristy K. Ward, MD, Gynecologic Oncology Fellow, Department of Reproductive Medicine, University of California, San Diego Moores Cancer Center, La Jolla. Dr Ward and colleagues queried the database of the University HealthSystem Consortium (with 217 academic medical centers) to identify readmissions among patients with gynecologic cancer.
For each risk factor found to be independently associated with readmission, the low-risk group was scored 0 and the high-risk group was scored 1. A risk of readmission score was created using findings from the multivariate analysis.
The overall readmission rate for patients with gynecologic cancer was 4.5%. Vulvar cancer, medical MS-DRG, urgent or emergent admission, length of stay of >4 days, and coverage by a public payer were each independently associated with readmission. The probability of readmission rose significantly with increasing risk score.
Patients with a risk of readmission score of 0 or 1 have a readmission rate of 3.9% compared with 10.7% for patients with a risk of readmission score of >1. “Just 2 of these risk factors put you at greater risk of readmission than the general population,” according to Dr Ward. “The risk of readmission score may be used to determine which patients you would expect to be readmitted; then you could allocate your resources better toward those patients.”
High hospital admission rates among patients with gastrointestinal (GI) cancer are driven by unplanned hospitalization and are potentially preventable, according to researchers from M.D. Anderson Cancer Center, Houston. They used Texas Cancer Registry and Medicare claims data for in-hospital admissions from 30,199 patients with GI cancer aged >66 years. The rate of unplanned hospitalization was 58%, and 77% of these hospitalizations occurred within the first year of the cancer diagnosis. The top 5 reasons for unplanned hospitalization were volume depletion, congestive heart failure, pneumonia, urinary tract infection, and septicemia. These are “diagnoses that are considered potentially preventable and should be a focus for intervention,” noted Joanna-Grace M. Manzano, MD, M.D. Anderson Cancer Center, Houston.
On multivariate analysis, unplanned hospitalization was more likely among patients with esophageal cancer (relative risk [RR], 1.16), gastric cancer (RR, 1.07), pancreatic cancer (RR, 1.04), and rectal cancer (RR, 1.03). Patients with regional and distant diseases were at higher risk for unplanned hospitalization (RR, 1.14 and 1.13, respectively).
Having ≥1 unplanned hospitalizations was associated with poorer survival. Of the entire cohort, 14% had ≥3 unplanned hospitalizations, which amounted to 49% of the unplanned hospitalizations.
Academic versus Nonacademic Hospitals
Readmission after an index hospitalization for cancer surgery is higher in nonteaching hospitals than in teaching hospitals, found Nina A. Bickell, MD, Codirector, Center for Health Equity and Community Engaged Research, Mount Sinai, NY, and colleagues.
As part of the Healthcare Cost and Utilization Project, common cancer hospitalizations in New York State in 2009 were identified and 30-day readmissions were assessed. From 21,945 index admissions for cancer surgery, the overall readmission rate was 9.3%, with 11.2% readmissions in nonteaching hospitals and 8.6% in teaching hospitals (P <.001).
Being male, undergoing surgery at a nonteaching hospital, black race, and certain comorbidities increased a patient’s risk of 30-day readmission for a preventable cause.
“Our hypothesis is that academic hospitals are more likely to have clinical protocols, and that may reduce readmission rates,” said Dr Bickell.