Independent Predictors of Repeat Emergency Room Presentations: Insights from a Cohort of 1066 Consecutive Patients with Non-Cardiac Chest Pain Generating 4770 Repeat Presentations

Abstract

Background and importance: Chest pain (CP) is one of the most frequent presentations to the emergency department (ED), a large proportion of which is non-cardiac chest pain (NCCP). Repeat attendances to ED are common and impose considerable burden to overstretched departments. Objective: Our aim was to determine drivers for repeat ED presentations using NCCP as the primary cause of index presentation. Design, setting and participants: This was a retrospective cohort study of 1066 consecutive presentations with NCCP to a major urban hospital ED in North England. Index of Multiple Deprivation (IMD), a postcode-derived validated index of deprivation, was computed. Charlson comorbidity index (CCI) was determined by reference to known comorbidity variables. Repeat presentation to ED to any national hospital was determined by a national linked database (population 53.5 million). Independent predictors of ED representation were computed using logistic regression analysis. Results: Median age was 43 (IQR 28–59), and 50.8% were male. Furthermore, 27.8%, 8.1% and 3.8% suffered from chronic obstructive pulmonary disease (COPD), hypertension and diabetes mellitus, respectively. The most frequent diagnoses, using ICD-10 coding, were non-cardiac chest pain (55.1%), followed by respiratory conditions (14.7%). One-year incidence of adjudicated myocardial infarction, urgent or emergency coronary revascularisation and all-cause death was 0.6%, 2% and 5.3%, respectively. There was a total of 4770 ED repeat presentations 1 year prior to or following index presentation with NCCP in this cohort. Independent (multivariate) predictors for frequent re-presentation (defined as ≥2 representations) were a history of COPD (OR [odds ratio] 2.06, p = 0.001), previous MI (OR3.6, p = 0.020) and a Charlson comorbidity index ≥1 (OR 1.51, p = 0.030). The frequency of previous MI was low as only 3% had sustained a previous MI. Conclusions: This analysis indicates that COPD and complex health care needs (represented by high CCI), but not socio-economic deprivation, should be health policy targets for lessening repeat ED presentations. What is already known on this topic: Repeat presentations with non-ischaemic chest pain are common, placing a considerable burden on emergency departments. What this study adds: COPD and complex health care needs, denoted by Charlson comorbidity index, are implicated as drivers for repeat presentation to accident and emergency department. Socio-economic deprivation was not an independent predictor of re-presentation. How might this study affect research, practice, or policy: Community-based support for COPD and complex health care needs may reduce frequency of ED attendance.

Keywords

• acute chest pain
• hospitalisation
• epidemiology