Conference Coverage

Late-week discharges to home after CRC surgery prone to readmission

Key clinical point: Patients discharged home on a Thursday following surgery for primary colorectal cancer are more likely to be readmitted with 30 days than are patients discharged home on any other day of the week.

Major finding: The highest rate of readmission was for patients discharged on Thursday, at 12.4%, compared with lowest rate of 10.1% for patients discharged on Sunday.

Data source: Retrospective SEER-Medicare database review of records on 93,047 patients treated for colorectal cancer.

Disclosures: The study was internally supported. The authors reported having no relevant disclosures.



BOSTON – The day of the week a patient is discharged from the hospital may have an impact the likelihood of readmission.

Patients discharged home from the hospital on a Thursday after colorectal cancer surgery are more likely to be readmitted within 30 days than those discharged on any other day of the week, investigators found.

Anna Gustin Neil Osterweil/Frontline Medical News
Anna Gustin

In contrast, there were no significant day-dependent differences in readmission rates among patients discharged to a skilled nursing facility or acute rehabilitation program, although patients admitted to clinical facilities had higher overall readmission rates, reported Anna Gustin and coinvestigators at the Levine Cancer Institute at the Carolinas Medical Center in Charlotte, N.C.

“For a patient discharged on a Thursday, if you’re going to get an infection, it’s going to be probably during the weekend, when it’s difficult to contact your primary physician, and when other resources are not as readily available,” said Ms. Gustin, who conducts epidemiologic research at Levine Cancer Center and is also a pre-med student and Japanese major at Wake Forest University in Winston-Salem, N.C.

In a study presented in a poster session at the annual Society of Surgical Oncology Cancer Symposium, Ms. Gustin and her coauthors looked at factors influencing readmission rates among patients undergoing surgery for primary, nonmetastatic colorectal cancer resections.

They drew on the to evaluate outcomes for 93,04 SEER-(Surveillance, Epidemiology, and End Results) Medicare database seven patients aged 66 years and older treated for primary colorectal cancer from 1998 through 2009.

They looked at potential contributing factors such as patient demographics, socioeconomic status, length of stay, days of admission and discharge, and discharge setting (home or clinical facility).

They use multivariate logistic regression models to analyze readmission rates at 14 and 30 days after initial discharge.

Focusing on home discharges, they found that as the week progressed, there was a significant likelihood that a patient discharged home would be readmitted (P less then .001 by chi-square and Cochran-Armitage tests). As noted before, the highest rate of readmission was for patients discharged on Thursday, at 12.4%, compared with 10.1% for patients discharged on Sunday, the discharge day least likely to be associated with rehospitalization.

In multivariate analysis, factors significantly associated with risk for 30-day readmission included male vs. female (hazard ratio, 1.16), black vs. other race (HR, 1.22), length of stay 5, 6-7, or 8-10 vs. 12 or more days (HR, 0.48, 0.59, 0.77, respectively), Charlson comorbidity index score 0, 1 or 3 vs. 3 (HR, 0.59, 0.73, 0.82, respectively), and home discharge vs. other (HR, 0.66; all above comparisons significant as shown by 95% confidence intervals).

The authors concluded that although home discharge itself reduces the likelihood of readmission, “improvements in preparing patients for discharge to home are needed. Additional outpatient interventions could rescue patients from readmission.”

They also suggested reexamining staffing policies and weekend availability of resources for patients, and call for addressing disparities in readmissions based on race, sex, length of stay, and comorbidities.

The study was internally supported. The authors reported having no relevant disclosures.

Next Article:

   Comments ()