Harman JS, Kelleher KJ, Mulsant BH, Reynolds CF; Academy for Health Services Research and Health Policy. Meeting.
Abstr Acad Health Serv Res Health Policy Meet. 2001; 18: 54.
University of Pittsburgh School of Medicine, 3811 O'Hara Street, Suite 430, Pittsburgh, PA 15213, Phone: (412) 624-1872, Fax: (412) 624-2360, E-mail: harmanjs@msx.upmc.edu
RESEARCH OBJECTIVE: In efforts to control costs, many health plans have begun to use clinical practice guidelines regarding length of stay (LOS) to limit use of inpatient services. Numerous commentaries have raised concerns regarding the impact of these guidelines on patients, especially since many are not evidence based. The objective of this research is to describe the distribution of LOS for treatment of depression and to determine the role that factors such as age, race, and type of insurance have on LOS.STUDY DESIGN: This study used data on all hospital discharges in Pennsylvania over a three-year period. Differences in mean LOS by year, age, race, diagnosis, comorbid conditions, and type of insurance were examined in both a bivariate and multivariate context. The multivariate analysis used gamma regression with a log link to determine the independent effect of each variable.POPULATION STUDIED: All individuals hospitalized in Pennsylvania with a primary diagnosis of depression in 1996, 1997, and 1998 (N = 28,257).PRINCIPAL FINDINGS: The mean LOS for depression decreased by 21% between 1996 and 1998 (8.2 days to 6.8 days). LOS was longer for children and elderly patients as compared to adults (9.3 days and 9.8 days vs. 6.5 days). LOS was shorter for patients covered by managed care vs. other forms of insurance (5.8 days vs. 7.8 days). There was no difference in LOS by race. These observed differences in LOS remained evident in the multivariate analysis. The mean LOS for 1998 was 15% shorter than in 1996 after controlling for all other variables. The mean LOS was 49% longer for children and 28% longer for elderly patients as compared to LOS for adults. Patients with managed care had 8% shorter LOS than patients with private fee-for-service (FFS) insurance, while patients with Medicare had 41% longer LOS as compared to patients with private FFS insurance. A 9% increase in readmission rates for depression was seen over the three-year time period.CONCLUSIONS: There was a dramatic decrease in the LOS for depression over a very short period of time. This decrease in LOS is most likely due to cost containment efforts and the proliferation of managed care. The mean LOS for patients in managed care was the shortest when compared to all other sources of insurance. Elderly patients, who are more likely to have comorbid conditions, and children, who have fewer alternative services, had longer LOS.IMPLICATIONS FOR POLICY, DELIVERY, OR PRACTICE: These findings are important for two main reasons. First, they provide empirical data on actual LOS for depression that can be used as a basis for clinical practice guidelines. Differences that are seen by age, type of depression diagnosis, and other factors should be considered when creating and implementing practice guidelines. Second, the large differences in LOS seen by type of insurance demonstrate the substantial impact of factors that are not fully related to illness severity on LOS. The concurrent increase in recidivism indicates that potentially poor outcomes are resulting from shorter LOS. This suggests that caution should be used when implementing practice guidelines for LOS.PRIMARY FUNDING SOURCE: This research was funded by NIMH grants P30 MH52247, P30 MH30915, and R25 MH60473
Publication Types:
Keywords:
- Adult
- Aged
- Child
- Costs and Cost Analysis
- Fee-for-Service Plans
- Hospitals
- Humans
- Insurance, Health
- Managed Care Programs
- Medicare
- Pennsylvania
- Practice Guidelines as Topic
- economics
- hsrmtgs
Other ID:
UI: 102273252
From Meeting Abstracts