Unplanned readmissions of elderly patients.

OBJECTIVE

This study was undertaken to determine the prevalence of unplanned readmissions in Geriatric Medicine in Aberdeen and to examine their nature in order to establish how many of them were avoidable.

DESIGN

Patients living within Aberdeen who were discharged from Care of the Elderly Assessment Wards in one hospital between 1 August 1994 and 31 January 1995 were identified. Any subsequent unplanned readmissions of this population to any local hospital within 28 days were identified and formed the study's sample. Subsequent comparison with the non-readmitted population yielded readmission rates. The nature of each episode was investigated by obtaining a wide range of data (e.g. from medical and nursing notes) soon after readmission and by the audit team subsequently identifying principal and associated causative factors. A questionnaire was also sent to the discharging consultant and the patient's GP seeking opinions on whether readmission was avoidable and these were weighted equally with the audit team's opinion in order to establish avoidability.

SETTING

The patients were all discharged from the nine Care of the Elderly Assessment wards at Woodend Hospital in Aberdeen and readmitted to any NHS hospital within Aberdeen.

SUBJECTS

The subjects were patients living within Aberdeen who satisfied the above criteria.

RESULTS

There were 109 episodes of readmission resulting from 713 discharges, making a readmission rate of 15.3%. The readmitted population was elderly with multiple medical problems; 50% lived alone. In 87% of cases the principal causative factor in readmission was medical, most commonly involving relapse of illness. The remaining 13% were 'social' in nature. Response rates to the questionnaire by GP's and Consultants were excellent (96% and 99% respectively) and 34 cases emerged where either party, or both, thought readmission to be avoidable. These cases were subject to further review by the audit team and in this way 16 of the 34 cases were judged to be avoidable. The main area for improvement was considered to be pre-discharge assessment of home circumstances.

CONCLUSION

The majority of unplanned readmissions were medical in nature and unavoidable. The use of rates of unplanned readmission as a measure of clinical outcome in care of the elderly is unsatisfactory as they do not accurately reflect the quality of in-patient care.