David B, Einstadter D, Husak SS, Cebul RD; AcademyHealth. Meeting (2003 : Nashville, Tenn.).
Abstr AcademyHealth Meet. 2003; 20: abstract no. 770.
Feinberg School of Medicine, Northwestern University, Medicine, 676 N St. Clair, Chicago, IL 60201 Tel. (312) 695-0917 Fax (312) 695-0951
RESEARCH OBJECTIVE: Length of hospital stay (LOS) declined dramatically over the last decade, but it is not known whether this had adverse consequences. This study examined 1) trends in 30-day post-discharge mortality and all-cause readmissions from 1991-97 and 2) whether patients whose LOS was much shorter than expected had higher 30-day post-discharge mortality and readmission rates. STUDY DESIGN: We conducted a time-series analysis of 30-day post-discharge mortality and all-cause readmission rates for patients with 6 medical diagnoses, using data from the Cleveland Health Quality Choice Program and Medicare files (MEDPAR) to track death and readmission within 30 days of discharge. Trends in 30-day post-discharge mortality and readmission rates were determined using multivariate logistic regression models with adjustment for admission severity of illness. Expected LOS was calculated using previously published risk-adjustment models, and the ratio of Observed/Expected (O/E) LOS calculated. The adjusted relative risk of death and readmission for patients in the lowest decile of O/E LOS was determined, using patients in deciles 5 and 6 as the reference category. Analyses were stratified by whether a do-not-resuscitate (DNR) order was written within 2 days of admission (early), later during hospitalization, or not at all. POPULATION STUDIED: Medicare patients discharged alive from hospitals in Northeast Ohio after a first admission for acute myocardial infarction (AMI; N=8612), congestive heart failure (CHF; N=22,203), gastrointestinal hemorrhage (GIH; N=8178), chronic obstructive pulmonary disease (COPD; N=10,566), pneumonia (N=21,113), or stroke (12,773). PRINCIPAL FINDINGS: Between 1991-97, the adjusted relative risk (ARR) of post-discharge mortality rose significantly for patients with AMI (ARR 1.65, 95% CI 1.24 - 2.19) and stroke (ARR 1.59, 95% CI 1.27 - 1.97). There was no change in post-discharge mortality for CHF, GI hemorrhage, COPD, and pneumonia. Trends in post-discharge mortality differed according to whether a DNR order was written. For patients who did not have a DNR order, the risk of death during the 30 days after discharge remained stable from 1991-97. In contrast, the risk adjusted post-discharge mortality rate for patients with early DNR orders increased for all diagnoses. Markedly shorter than expected LOS was associated with higher than expected risk-adjusted mortality for patients with early DNR orders but not for others (no DNR and late DNR). Risk-adjusted readmission rates remained stable from 1991-97, except for a 15% (95% CI 3-30%) increase for patients with CHF. Markedly shorter than expected length of stay was not associated with higher readmission rates. CONCLUSIONS: The dramatic decline in length of stay from 1991-97 was not associated with worse post-discharge outcomes for patients without DNR orders. Post-discharge mortality increased among patients with early DNR orders, but only a small portion of this trend was explained by patients being discharged more rapidly than previously. IMPLICATIONS FOR POLICY, DELIVERY OR PRACTICE: Overall, financial pressures to reduce length of stay do not appear to have jeopardized patients' health, and the financial benefits of shorter length of stay are not diminished from higher readmission rates. Providers should closely examine the quality of care received by patients with DNR orders.
Publication Types:
Keywords:
- Heart Failure
- Hospitalization
- Hospitals
- Humans
- Logistic Models
- Medicare
- Myocardial Infarction
- Ohio
- Patient Discharge
- Pneumonia
- Pulmonary Disease, Chronic Obstructive
- Risk Adjustment
- Stroke
- economics
- mortality
- trends
- hsrmtgs
Other ID:
UI: 102275738
From Meeting Abstracts