Attending physicians in an emergency department may benefit from clinical guidance when working under time pressure with an increasing number of fragile patients in an understaffed clinical environment. One in two readmitted patients were seen within one week, one in three patients came back with similar diagnosis, and four areas worth the effort of raised focus were identified when caring for older patients. Our findings corroborate the need to pay special attention to polypharmacy, multimorbidity, disability and marital status during early discharge planning.
Our data illustrated that taking eight or more systemic prescription drugs increased readmission likelihood by more than four times. A lower yet markedly raised risk was seen among patients prescribed five or more medications (T. Alarcón et al. 1999). These findings suggest that the more drugs the older patient is prescribed, the more likely the patient is to benefit from comprehensive geriatric assessment and care (G. Ellis et al. 2011). Patients prescribed fewer drugs may benefit, too, as different types of prescription medication may impact patients differently. In line with these considerations, we identified a noteworthy number of patients being prescribed risk drugs, such as benzodiazepines, associated with readmission in our data. Such drugs may be a topic for special focus (American Geriatrics Society 2019 Updated AGS Beers Criteria).
Extending findings that comorbidities raised the risk of readmission (V. Calsolaro et al. 2019; M. Zanocchi et al. 2006; Pedersen, L.H., Gregersen, M., Barat, I. et al 2018), we found that raised CCI doubled the risk of early readmission in older patients following also short-term hospitalisation. This finding was statistically significant even with the limitations of CCI and supports a clinical focus on multimorbidity among older patients discharged from Emergency Departments.
Disability, as defined by residence in nursing homes or receiving residential care prior to admission, was associated with a 62% higher risk for readmission corroborating previous findings (A. Tanderup et al. 2018). These readmissions may be speculated to relate to an early response by out-of-hospital nursing staff to deteriorating health related to premature discharge in our group of older patients with a maximum of 48h hospitalisation. A previous study suggested a higher prevalence of illness in this group (A. Tanderup et al. 2018), and our findings need to be examined in detail not possible in our study.
Unwed patients were more frequently readmitted to the Emergency Department in keeping with previous reports (Pedersen, L.H., Gregersen, M., Barat, I. et al 2018), and in our data, readmission was more frequent in unwed patients than in patients with disability. Hence, being married may suggest support and care by the spouse, while the observation and response to deterioration seen with nursing home residence or out-of-hospital nurse visits could be lacking. It may thus be suggested that early discharge should be supported by subsequent timely home visits (Pedersen, L.H., Gregersen, M., Barat, I. et al 2018). This is further supported by our finding that readmission occurred within the first week in nearly half of the patients.
A limitation of the present study was the inherent risk of selection- and information bias in retrospective case-control studies, alongside the inability to detail the timeline of exposure and outcome. However, the 3:1 number of potential patients for controls compared to the cases suggests these limitations have a low impact on the findings. Moreover, the validity of the reported number of drugs taken was based on redeemed prescriptions. Additionally, our study was a single-centre study in a publicly funded healthcare system, and the results may not be generalisable outside such a setting. Finally, we focussed on older patients with short-term hospitalisation in an Emergency Department, and our findings and conclusion should not be applied to other settings or time frames.
In conclusion, receiving multiple systemic prescription drugs, having multiple comorbidities, disability, and unwed status increased the risk of early readmittance in older patients discharged within 48 hours from an Emergency Department. Therefore, the examined demographic and geriatric factors emphasised in this study should raise alertness in daily clinical practice and may guide the busy attending physician in Emergency Departments when planning early discharge of older patients.