(P37) Designing a New Behavioral Reporting Tool for Long-Term-Care-Residents with Dementia

Objective/Aim: We aimed to develop a new behavioral reporting tool for long-term care (LTC) residents that allows for establishing a behavioral baseline, longitudinal assessment of residents with behaviors, and assessing behavioral outcomes after interventions.

Background: Behavioral and psychological symptoms of dementia (BPSD) are expected to develop in up to 90% of dementia patients. There is no gold standard for assessing BPSD in LTC.

Quality Improvement

Methods: Phase 1- Needs assessment: We reviewed behavioral logs of all units in the Long Island State Veterans Home(New York) for four months to determine the most common reasons and percentage completion. We searched PubMed and Embase for validated BPSD instruments, the most commonly reported BPSD, and non-pharmacological interventions.
Phase 2 – Designing the tool. 2.1 Reason for initiating the tool: we established new and readmission, behaviors, unit transfers, peer conflict, or psychotropic dose adjustment as reasons to start the tool based on Phase 1 findings. The charge nurse will review all logs. If there is no improvement with non-pharmacological measures, a multidisciplinary team (MD, NP, psychiatry, psychologist, and social worker) will review the logs. 2.2 Behavior and Intervention code: we included 20 behaviors and "No behavior" to choose from and 14 different interventions. All terms are in lay language. 2.3

Assessment: We included the time, trigger (if noted), and frequency of BPSD during the shift, the intervention, and the response. Residents must be reassessed every shift (day, evening, night) and documented even in the absence of behaviors for up to 7days.
Phase 3 – Feedback: We described the new tool (Tool A), Cohen-Mansfield Agitation Inventory (Tool B), and Direct Observation System (Tool C) to MDs, physician assistants (PAs), and registered nurses (RNs). We asked to complete the three tools for any patient on the unit who exhibited behaviors during that day, followed by an anonymous survey comparing the tools in different categories. We did a descriptive statistical analysis of the responses.

Results: We designed a new tool that establishes a behavior baseline and assesses BPSD and response to interventions in LTC residents. Our tool is designed to be longitudinal, with observations from different providers at different times of the day, to establish patterns and decrease the risk of recall and observation bias. A novel aspect of our tool is the inclusion of non-pharmacological measures as part of the assessment. Sixteen providers were surveyed; 62% were RN, 25% were MDs, and 13% were PAs.

Conclusions: Our tool was preferred by MDs and RNs for documentation, establishing patterns, and addressing behaviors. However, it is less likely to be completed entirely and more challenging to learn than other tools, indicating that further improvement is required to make it more straightforward. Our report system is a promising tool for assessing BPSD and intervention response in LTC residents. We recognize that validation of the tool is very important, and we plan to do so in the future. We created this tool in recognition of the limitations of the behavioral tools available. We believe that sharing our thought processes and methods with other physicians and members of the PALTC community, even in the absence of validation, is also important.