A case management system was ineffective in reducing hospital readmissions for general medical inpatients
ACP J Club. 1995 Mar-April;122:52. doi:10.7326/ACPJC-1995-122-2-052
Fitzgerald JF, Smith DM, Martin DK, Freedman JA, Katz BP. A case manager intervention to reduce readmissions. Arch Intern Med. 1994 Aug 8;154:1721-9.
To examine the effectiveness of a case management system to increase outpatient primary care contacts and reduce subsequent hospital readmissions among patients discharged from inpatient services.
1-year randomized controlled trial.
University-affiliated Veterans Affairs (VA) medical center.
Of 1068 eligible patients, 668 (63%) (mean age 65 y) were included; they were discharged from a general medicine inpatient service between November 1988 and October 1990, were ≥ 45 years of age, received primary care in the hospital's clinics, lived in the primary service area of the hospital, and had access to a telephone. Patients determined by the resident physician to have < 60 days to live were excluded. Follow-up was 98%.
At hospital discharge, patients were stratified by age group (45 to 54 y, 55 to 64 y, 65 to 74 y, or ≥ 75 y) and allocated to usual care that included mailed reminders of upcoming clinic appointments (n = 335) or to visits with a nurse case manager who counseled patients about their medical problems, facilitated access to usual care, identified and fulfilled unmet social and medical needs with standard or alternative sources of care, and arranged all appointments by mailed reminder and telephone contact (n = 333). Patients in the intervention group were also given a telephone number with 24-hour access to the case manager or a study physician.
Main outcome measures
Outpatient clinic visits, emergency department visits, hospital readmissions and days spent in the hospital, nursing home admissions and days spent in the nursing home, and mortality.
Patients who had a case manager made more outpatient clinic visits (P = 0.02) and had longer nursing home stays (P = 0.04) than did patients receiving usual care (Table). The groups did not differ in the number of visits to other VA or non-VA outpatient clinics or to emergency departments or in nonelective hospital readmissions, admissions to nursing homes, or mortality.
When compared with usual care, a case management system did not reduce hospital readmissions, emergency department visits, or number of days spent in the hospital for patients discharged from a general medicine inpatient service.
Source of funding: Department of Veterans Affairs.
For article reprint: Dr. J.F. Fitzgerald, Department of Medicine, Indiana University School of Medicine, IN, USA. FAX 317-630-8686.
Table. Case management system vs usual care for general medical inpatients*
|Outcomes at 1 y||Case management system||Usual care||Mean difference (95% CI)|
|Outpatient clinic visits (mean visits/patient per month)||0.30||0.26||0.04 (0.006 to 0.07)|
|Nursing home stay (mean days/patient per month)||0.64||0.22||0.42 (0.29 to 0.81)|
*Mean difference and CI calculated from data in article.
Interest in reducing the use of costly services has encouraged the development of programs that foster coordinated and managed care. Case management has been widely used in this regard, but few studies have been done to critically evaluate its effectiveness.
The results obtained by Fitzgerald and colleagues appear to question the effectiveness of case management in decreasing the use of certain expensive services. The increased use of outpatient clinic visits may show improvement in access, but the number of hospital readmissions did not decrease. The population of chronically ill patients who were already enrolled in a comprehensive system of care and the brief 12-month follow-up period may have contributed to a shortfall of expected benefit.
Aggressive case management programs designed to reduce costs have also failed to substantially decrease use of inpatient services (1). In addition, subpopulations of patients may receive different benefits from similar types of interventions (2).
Case management of chronically ill patients represents a complex intervention in a complex system, and it is not surprising that attempts to define outcomes are difficult (3). As with other attempts to restructure the health care system, improving access and comprehensiveness may conflict with efforts directed at cost containment. Health care providers should work with payers to develop programs that enhance (or at least maintain) access to necessary services while attempting to curtail use that does not add value to care.
Barbara E. Barnes, MD
University of PittsburghPittsburgh, Pennsylvania, USA