CHICAGO — With hospitals under pressure to improve efficiency by reducing patient length of stay, findings from a new study suggest that this effort does not necessarily lead to increased readmission rates.
That's what Dr. Jorge Go and his colleagues at the Iowa City Veterans Affairs Medical Center found in an analysis of all 3,709,103 medical admissions to 129 VA hospitals from 1997 to 2007. Approximately 18% of patients (692,599) were excluded for reasons such as death, terminal cancer, or transfer to other facilities, leaving 3,016,504 patients for the final analysis.
Among patients in the analysis, 97% were male and 72% were white; the mean age was 65 years, and 46% had three to five comorbidities. Patients were stratified by five common conditions: heart failure, chronic obstructive pulmonary disease (COPD), acute myocardial infarction, community acquired pneumonia, and gastrointestinal hemorrhage.
During the 10-year interval, there was a 25% reduction in average length of stay (LOS) and a 7% reduction in hospital readmissions, Dr. Go reported at the annual meeting of the Society of Hospital Medicine.
The unadjusted mean LOS decreased significantly for all diagnoses, from 6 days in 1997 to 4.5 days in 2006-2007. The biggest reduction was in MI and pneumonia cases, with LOS dropping by 2 full days.
The unadjusted readmission rate also declined significantly for all diagnoses, from 13.9% in 1997 to 12.7% in 2007. The biggest decrease was in COPD patients (17.3% to 14.3%), while readmissions remained constant among those with GI hemorrhage.
“Decreasing LOS and readmissions seem to reflect secular trends—that we're providing more efficient and better care,” Dr. Go said. “It's reassuring to show that increasing patient efficiency has not resulted in increased hospital readmission.”
A multivariate analysis that adjusted for age, sex, income level, comorbidities, admission source, and VA facility showed that the decrease in readmission risk occurred earlier for chronic conditions than for acute conditions.
Compared with 1997, the readmission rate began to decrease significantly among heart failure patients from 2002-2003 (odds ratio 0.93) to 2006-2007 (OR 0.90), and among COPD patients from 1998-1999 (OR 0.91) to 2006-2007 (OR 0.79).
In contrast, significant declines in readmission rates occurred only in 2006-2007 for the acute conditions of MI (OR 0.88) and GI hemorrhage (OR 0.87), with no improvement for pneumonia during the study period, Dr. Go reported.
Possible explanations for this finding are that medical advances and improvements in delivery of care have been greater for chronic conditions than for acute conditions, or that care for acute conditions was already very good and thus more difficult to improve, Dr. Go said in an interview.
The observed pattern of changes in readmission rates for chronic and acute diseases could be useful to policymakers using readmission rates as a quality improvement measure. The National Quality Forum already endorses using 30-day all-cause heart failure readmissions as a quality measure and plans to use it for other conditions as well, he said.
The researchers disclosed no relevant conflicts of interest.
It's reassuring that increasing patient efficiency has not resulted in increased hospital readmission. DR. GO