Development of a Content Valid and Reliable Prehospital Environmental Falls Risk Assessment Tool for Older Adults

Introduction: The aging population reintroduces the need to establish early identification of falls risk as a means of primary and secondary prevention of falls. While there are several existing tools to assess environmental risk factors developed for consumers or home health providers, assessment of environmental falls risk by emergency medical services (EMS) providers represents a novel approach to primary and secondary prevention. The purpose of this study was to evaluate a content valid and reliable assessment of environmental fall risk to be performed in the prehospital setting. Methods: This was a mixed methods study, conducted from August, 2015 to September, 2017 in Mecklenburg County, NC, utilizing qualitative methodology to develop content valid items for an environmental falls risk assessment and quantitative methodology to assess those items for interrater reliability. Content validity was assessed using 2 expert panels. Expert Panel One was tasked with assessing validity of a construct to indicate an increased risk of an in-home fall for elderly individuals and expert Panel Two was responsible for assessing the likelihood of an EMS professional to identify a construct during their course of patient care. To assess reliability of the identified content valid items, 5 paramedics were recruited for interrater reliability (IRR) testing of the validated falls risk assessment tool. Each paramedic and their partner received education on documentation and deployment of the tool. Crews independently documented presence or absence of each item with pair agreement assessed using Cohen's kappa (κ). Results: A total of 87 items were identified for assessment through review of validated scales and relevant literature, with the content validation process reducing to 9 the number of items tested in the field for reliability. A total of 57 paired assessments were completed and included in analysis. One item returned almost perfect agreement (κ = 0.87), 5 items returned moderate agreement (κ = 0.41-0.54), with the remaining 3 items illustrating fair agreement (κ = 0.33-0.39). Conclusion: We developed a construct valid and reliable assessment of environmental falls risk to be performed in the prehospital setting. Further trials should be conducted using this tool to determine appropriate cut scores and deployment in the prehospital setting to help with primary and secondary fall prevention.