Download PDF

Unplanned readmissions in Belgian hospitals: importance of quality of in-hospital care and quality of care transitions

Publication date: 2016-06

Author:

Braet, Anja

Keywords:

hospital readmission, care transition, quality of care

Abstract:

Hospital readmissions – defined as new admissions to the hospital after hospital discharge within a specific time interval – occur frequently, are costly and can lead to negative outcomes for patients. Because a considerable proportion of unplanned readmissions are caused by suboptimal quality of care and are therefore potentially preventable, unplanned hospital readmissions are used as an indicator of quality of care. The overall aim of this research was to study how to reduce hospital readmissions that are due to substandard quality of in-hospital care or to substandard quality of the care transition from hospital to home. We used a mixed-methods approach to address four operational aims. The first aim was to explore unplanned hospital readmissions in Belgium, addressing the incidence of unplanned hospital readmissions, the identification of patient groups that are most frequently readmitted and the identification of risk factors for unplanned readmissions. The second aim was to identify discharge interventions that have been demonstrated to be effective in reducing hospital readmissions within three months of discharge, and to understand their effect on mortality, use of emergency departments (EDs) and patient satisfaction. The third aim was to understand the causes of readmissions related to suboptimal quality of in-hospital care. Finally, the fourth aim was to understand the causes of readmissions related to suboptimal quality of care transition from hospital to home. An exploratory cross-sectional study was conducted to understand the phenomenon of hospital readmissions in Belgium (first aim). We analysed the Belgian Hospital Discharge Dataset including data from 1,130,491 patients discharged in 2008. The overall unplanned readmission rate 30 days after discharge was 5.2%. The highest numbers of readmissions were found for patients admitted for COPD (14.7% readmission rate), heart failure (14.0%) and pneumonia (9.4%). Overall, the most common reasons for readmission were cardiovascular and pulmonary diagnoses (in 16.8% and 13.3% of all readmissions, respectively) and 10.4% of all readmissions were due to complications. We identified multiple factors that increase the risk of readmission: male gender, age, discharge against medical advice, severity of illness, number of comorbidities, multiple previous ED visits, discharge destination, discharge on Friday, length of stay and acuity at admission. Because multiple ED visits are an important risk factor for readmissions, these ED visits must trigger actions to coordinate care between health professionals together with patients and their family caregivers. Another important finding is that the risk of readmission increases with length of stay; thus delaying discharge for a patient who is ready for discharge should be avoided. We performed a systematic literature review to study the effectiveness of discharge interventions in reducing hospital readmissions (second aim) and included 51 studies. Discharge interventions were defined as interventions designed to ease the care transition from hospital to home or to prevent problems after hospital discharge and were performed – at least partly – by hospital professionals. We found that discharge interventions significantly reduced the risk of hospital readmission (by 23%) and improved patient satisfaction. However, they did not reduce the risk of ED visits nor mortality. Interventions starting during hospital stay and continuing after discharge and interventions that support patient-empowerment were most effective in reducing readmissions. Additionally, discharge planning – an intervention to prepare patients for discharge during their hospital stay – reduced readmissions up to three months after discharge. Complex, multi-component interventions were not superior in comparison to single-component interventions. We conducted a prospective cohort study to understand the causes of readmissions related to suboptimal in-hospital quality of care (third aim) and suboptimal quality of the care transitions from hospital to home (fourth aim) for three patient groups (patients with heart failure, pneumonia and total hip/knee arthroplasty). For each patient group, hospitals with high and low readmission rates were selected. To assess the impact of in-hospital quality of care we explored hospital-acquired adverse events (AEs) (injuries caused by medical care) across hospitals with strong divergent readmission rates. A total of 100 AEs were detected in the 296 patient records reviewed (30.1% of the patients had AEs). We found no association between hospital-acquired AEs and hospital readmissions. Additionally, no association was found between AEs and the presence of post-discharge events (mortality, visit to ED or readmission). The impact of quality of care transitions was assessed by evaluating five care transition elements: readiness for discharge, patient and caregiver education, general practitioner (GP) contributions to the discharge process, and timeliness and completeness of discharge summaries. Overall, the quality of care transitions offers room for improvement. We found that patients with heart failure in hospitals with high readmission rates were less prepared for discharge compared to patients in hospitals with low readmission rates. We also found that more post-discharge events occurred in patients of GPs who did not receive the discharge summaries in a timely manner. We conclude that unplanned hospital readmissions occur frequently for specific patient groups. The risk of readmission is affected by patient- and disease related factors, as well as by the number of previous ED visits and the length of hospital stay, with increasing risk for longer stays. Based on this study, we identify patient empowerment, communication with primary care and timely preparing patients for discharge as important domains to prevent unplanned hospital readmissions.