Baker D; Association for Health Services Research. Meeting.
Abstr Book Assoc Health Serv Res Meet. 1997; 14: 169.
MetroHealth Medical Center, Case Western Reserve University, Cleveland, OH 44109-1998, USA.
RESEARCH OBJECTIVES: To identify clinical characteristics associated with readmission for heart failure within 30 days of discharge and to examine variations in hospital readmission rates, adjusting for clinical characteristics. STUDY DESIGN: The sample included consecutive admissions for congestive heart failure to 30 Cleveland area hospitals between July 1, 1992 and March 31, 1995. Patients were identified from the Ohio Medicare MEDPRO Part A data file on the basis of ICD-9 principal diagnosis codes (428.x, 402.x1, 404.0x, 404.1x, 404.9x, and 398.91). MEDPRO data was then merged with the Medicare denominator file to obtain Medicare Part B coverage and with data collected by Cleveland Health Quality Choice, a regional program that routinely collects nearly 200 clinical data elements from patients' hospital records. A total of 21,464 admissions were matched from the three data sets on the basis of six common elements (birth date, gender, hospital, admission and discharge dates, and principal diagnosis code). We then created patient-hospital combinations, and randomly selected a single admission if a patient was admitted more than once to a specific hospital, leaving 13,604 admissions for analysis. For each admission, we determined whether readmission to any hospital occurred within 30-days of discharge, and used logistic regression to determine the independent effects on readmission of clinical characteristics, severity of illness, insurance coverage, home health care, and the hospital where the index admission occurred. PRINCIPAL FINDINGS: 852 patients (6.3%) died within the first 30 days without being readmitted for heart failure and were excluded from the analysis. Of the remaining 12,752 patients, 825 (6.5%) were admitted again with a principal diagnosis of heart failure within 30-day readmission for heart failure. For example, the odds ratio (OR) of readmission for patients with the worst cardiac function (left ventricular ejection fraction 0.20 or less) was 2.76 (95% CI 1.57 to 4.90) compared to patients with the least impairment. Other risk factors included: impaired renal function (OR 1.52, 95% CI 1.19 to 1.92); chest x-ray showing an enlarged heart (OR 1.22, 95% CI 1.02 to 1.46), pleural effusion (OR 1.17, 95% CI 1.00 to 1.37), or pulmonary edema (OR 1.22, 95% CI 1.00 to 1.49). Several co-morbidities were also associated with an increased risk of 30-day readmission; insulin-requiring diabetes mellitus (OR 1.28, 95% CI 1.08 to 1.53), ischemic heart disease (OR 1.43, 95% CI 1.23 to 1.67), a cardiac pacemaker (OR 1.65, 95% CI 1.30 to 2.09), and chronic obstructive lung disease (OR 1.44, 95% CI 1.21 to 1.73). After adjusting for patients' age and the variables listed above, patients who lacked Part B Medicare coverage were more likely to be readmitted with heart failure within 30 days of discharge (OR 1.97, 95% CI 1.56 to 2.50). Discharge to home health care was also a risk factor for hospital readmission within 30 days (OR 1.30, 95% CI 1.09 to 1.55). Heart failure readmission rates varied from 3.3% to 9.7% across the 30 hospitals in the study. After adjusting for the variables above, there were significant differences in readmission rates by hospital. The adjusted risks of 30-day readmission at hospital J and hospital Y were significantly lower (OR 0.47, 95% CI 0.25 to 0.88; OR 0.54, 95% CI 0.30 to 0.98) than the average risk of readmission (as represented by the 10 hospitals in the middle tertile of readmission risk). Conversely, the 30-day readmission rates at hospital O and hospital CC were significantly higher (OR 1.52, 95% CI 1.03 to 2.25; OR 1.57, 95% CI 1.06 to 2.32) than the average. CONCLUSIONS: Worse disease severity, comorbid illness, lack of Medicare Part B coverage, and discharge to home health care all increased the risk of 30-day readmission for heart failure. Patients who lack Medicare Part B coverage may have more barriers to receiving adequate outpatient treatment. Home health care may incrase hospitalizations by increasing detection of patients with moderate decompensation that may have resolved without admission. Alternatively, this result may be due to unmeasured differences in the functional status of patients discharged to home health care. Finally, there is a three-fold variation in the 30-day risk of readmission for heart failure across Cleveland hospitals that is not accounted for by differences in clinical characteristics. The number of hospital readmissions for heart failure may be a useful indicator of quality of care that incorporates both inpatient (e.g., patient status at discharge) and outpatient quality of care. RELEVANCE TO CLINICAL PRACTICE AND POLICY: Monitoring readmission rates as a quality indicator may encourage hospitals to implement disease management programs and improve health outcomes. Home health care programs may need additional training and experience in managing heart failure patients as outpatients to prevent them from increasing rather than ....
Publication Types:
Keywords:
- Case-Control Studies
- Comorbidity
- Heart Failure
- Home Care Services
- Hospitalization
- Hospitals
- Hospitals, Teaching
- Humans
- Medicare
- Myocardial Ischemia
- Odds Ratio
- Ohio
- Patient Discharge
- Patient Readmission
- Quality Indicators, Health Care
- Risk Factors
- economics
- hsrmtgs
Other ID:
UI: 102233653
From Meeting Abstracts