Psychometric validation of a patient-reported experience measure for older adults attending the emergency department: the PREM-ED 65 study

Abstract

Introduction: Optimising emergency department (ED) patient experience is vital to ensure care quality. However, there are few validated instruments to measure the experiences of specific patient groups, including older adults. We previously developed a draft 82-item Patient Reported Experience Measure (PREM-ED 65) for adults ≥65 attending the ED. This study aimed to derive a final item list and provide initial validation of the PREM-ED 65 survey.

Methods: A cross-sectional study involving patients in 18 EDs in England. Adults aged 65 years or over, deemed eligible for ED discharge, were recruited between May and August 2021 and asked to complete the 82-item PREM at the end of the ED visit and 7-10 days post discharge. Test-retest reliability was assessed 7-10 days following initial attendance. Analysis included descriptive statistics, including per-item proportions of responses, hierarchical item reduction, exploratory factor analysis (EFA), reliability testing and assessment of criterion validity.

Results: Five hundred and ten initial surveys and 52 retest surveys were completed. The median respondent age was 76. A similar gender mix (men 47.5% vs women 50.7%) and reason for attendance (40.3% injury vs 49.0% illness) was observed. Most participants self-reported their ethnicity as white (88.6%).Hierarchical item reduction identified 53/82 (64.6%) items for exclusion, due to inadequate engagement (n=33), ceiling effects (n=5), excessive inter-item correlation (n=12) or significant differential validity (n=3). Twenty-nine items were retained.EFA revealed 25 out of the 29 items demonstrating high factor loadings (>0.4) across four scales with an Eigenvalue >1. These scales were interpreted as measuring 'relational care', 'the ED environment', 'staying informed' and 'pain assessment'. Cronbach alpha for the scales ranged from 0.786 to 0.944, indicating good internal consistency. Test-retest reliability was adequate (intraclass correlation coefficient 0.67). Criterion validity was fair (r=0.397) when measured against the Friends and Families Test question.

Conclusions: Psychometric testing demonstrates that the 25-item PREM-ED 65 is suitable for administration to adults ≥65 years old up to 10 days following ED discharge.

Keywords: Health Services for the Aged; geriatrics; quality assurance.