HOT SPRINGS, VA. – A shorter length of stay appears to be associated with a higher risk of readmission after colectomy, but only in patients who are older and have more preoperative comorbidities and perioperative complications, results of a large database review demonstrate.
Dr. Timothy Pawlik and his colleagues from Johns Hopkins Hospital, Baltimore, conducted a retrospective study of the Surveillance, Epidemiology, and End Results (SEER) database of 149,622 Medicare patients with a primary diagnosis of colorectal cancer who underwent colectomy from 1986 to 2005. The goal was to determine trends in readmission rates during the first 30 days.
"How do we begin to remedy the demonization of legitimate readmissions either politically, medically, or both?"
Increasingly, payers – especially Medicare – are penalizing hospitals for failing to prevent readmissions, which cost about $40 billion annually. Moreover, up to three-quarters of readmissions may be avoidable. In the meantime, many hospitals are fast-tracking patients for discharge after colorectal surgery.
In the SEER data, the mean age was 75 years in the 1986-1990 period and 77 years in the later period of 2001-2005. Patients were sicker in the later time period, with 63% having a Charlson comorbidity score of 3 or greater, compared with only 53% in the earlier time frame.
Most patients (38%) had a right colectomy; 23% had a sigmoid procedure, 13% had a rectal procedure, and 11% had a left colectomy. Forty-six percent of patients had lymph node metastasis, and 17% had an emergent procedure.
Morbidity for the entire study period was 37%; most complications were gastrointestinal or related to bleeding or postoperative infection. Again, there was a significant difference between the earlier and the later time frames. Perioperative morbidity was 27% in the 1986-1990 time frame, vs. 40% in 2001-2005. Mortality, at 4%, remained stable over time.
The mean length of stay decreased from 14 to 10 days from the early time to the later time period. And, over time, the percentage of patients discharged to home decreased, while discharges to skilled nursing facilities increased.
Overall, there were 17,000 readmissions, for a rate of 11%. Readmissions also increased, from 10% early on to 14% in 2001-2005. During that later time frame, patients had a 46% increased risk of readmission. Almost half of the readmissions occurred within the first 7 days after surgery, primarily for complications, dehydration, or infection. The mortality rate associated with readmission was 8%.
Multivariate analysis showed that the factors most likely to impact readmission were multiple comorbidities or a history of any perioperative complication, said Dr. Pawlik. Early discharge alone was not associated with a higher risk of readmission.
The study shows that "ongoing initiatives to reduce risk of readmissions and the associated costs, morbidity, and mortality are needed," said Dr. Pawlik.
It also "provides data that we all know to be true: namely, that the patients we are operating on have more comorbidities and that length of stay increases the rate of readmissions," said Dr. Susan Galandiuk, the discussant at the meeting. However, payers have been using length of stay and readmissions as quality surrogates, said Dr. Galandiuk, professor of surgery at the University of Louisville, Ky.
Not readmitting an elderly patient with many comorbidities and complications "would be a quality of care issue, and not the other way around," she added. The problem now is, "How do we begin to remedy the demonization of legitimate readmissions either politically, medically, or both?" asked Dr. Galandiuk.
Dr. Pawlik agreed that readmissions should not necessarily be held to be all bad. "Blaming readmissions on length of stay is a gross oversimplification of what’s going on," he added. Surgeons are clearly operating on older and sicker patients, and are doing more complicated operations that may result in a higher rate of morbidity – all this "in a culture of being asked to send people home earlier," he said.
"Our data clearly show that early discharge is feasible in some patients, but it needs to be used judiciously, especially in an older population," said Dr. Pawlik.
Dr. Pawlik and Dr. Galandiuk reported no conflicts.