Factors affecting unplanned readmissions from community hospitals to acute hospitals: a prospective observational study

<p><b>INTRODUCTION</b>While the readmission rate from community hospitals is known, the factors affecting it are not. Our aim was to determine the factors predicting unplanned readmissions from community hospitals (CHs) to acute hospitals (AHs).</p><p><b>MATERIALS AND METHODS</b>This was an observational prospective cohort study, involving 842 patients requiring post-acute rehabilitation in 2 CHs admitted from 3 AHs in Singapore. We studied the role of the Cumulative Illness Rating Scale (CIRS) organ impairment scores, the Mini-mental State Examination (MMSE) score, the Shah modified Barthel Index (BI) score, and the triceps skin fold thickness (TSFT) in predicting the rate of unplanned readmissions (UR), early unplanned readmissions (EUPR) and late unplanned readmissions (LUPR). We developed a clinical prediction rule to determine the risk of UR and EUPR.</p><p><b>RESULTS</b>The rates of EUPR and LUPR were 7.6% and 10.3% respectively. The factors that predicted UR were the CIRS-heart score, the CIRS-haemopoietic score, the CIRS-endocrine / metabolic score and the BI on admission. The MMSE was predictive of EUPR. The TSFT and CIRS-liver score were predictive of LUPR. Upon receiver operator characteristics analysis, the clinical prediction rules for the prediction of EUPR and UR had areas under the curve of 0.745 and 0.733 respectively. The likelihood ratios of the clinical prediction rules for EUPR and UR ranged from 0.42 to 5.69 and 0.34 to 3.16 respectively.</p><p><b>CONCLUSIONS</b>Patients who have UR can be identified by the admission BI, the MMSE, the TSFT and CIRS scores in the cardiac, haemopoietic, liver and endocrine/metabolic systems.</p>

Bridging the gap between primary and specialist care--an integrative model for stroke

Stroke is a major cause of death and disability in Singapore and many parts of the world. Chronic disease management programmes allow seamless care provision across a spectrum of healthcare facilities and allow appropriate services to be brought to the stroke patient and the family. Randomised controlled trials have provided evidence for efficacious interventions. After the management of acute stroke in a stroke unit, most stable stroke patients can be sent to their family physician for continued treatment and rehabilitation supervision. Disabled stroke survivors may need added home-based services. Suitable community resources will need to be harnessed. Clinic-based stroke nurses may enhance service provision and coordination. Close collaboration between the specialist and family physician would be needed to right-site patients and also allow referrals in either direction where necessary. Barriers to integration can be surmounted by trust and improved communication. Audits would allow monitoring of care provision and quality care enhancement. The Wagner model of chronic care delivery involves self-management support, shared clinical information systems, delivery system redesign, decision support, healthcare organisation and community resources. The key and critical feature is the need for an informed, activated (or motivated) patient, working in collaboration with the specialist and family physician, and a team of nursing and allied healthcare professionals across the continuum of care. The 3-year Integrating Services and Interventions for Stroke (ISIS) project funded by the Ministry of Health will test such an integrative system.