Behavioral Intervention Team Improving Outcomes in Older Adults with Mental Illness: Retrospective Review of Outcomes on Med-Surg Units
Lucy Easler,Juliet Glover,Nell Payne,Jessica Anderson,Beth Smith,Shilpa Srinivasan,Deborah Hurley
DOI: https://doi.org/10.1016/j.jagp.2024.01.198
IF: 7.996
2024-02-21
American Journal of Geriatric Psychiatry
Abstract:Introduction Mentally ill older adults often exhibit behavioral disturbances, such as removing medical devices or wandering while hospitalized. Challenging behaviors can lead to sentinel events such as falls, elopements, or violence (Sledge et al., 2015; Wittink et al., 2020). Prevalence of psychiatric diagnoses in medical inpatients is between 20-40% (Chan et al., 2018; Jansen et al, 2018; Sledge et al., 2015), and more recently, has been estimated as high as 50% (Robbins-Welty & Gagliardi, 2023). These co-morbidities negatively impact mortality, length of stay (LOS), cost of care, use of restraints, and readmission rates (Chan et al., 2018; Jansen et al., 2018; Wittink et al., 2020). Patients with chronic mental illnesses have a higher risk for medical illnesses, with increased morbidity and mortality, and more frequent hospitalizations. Complicated alcohol withdrawal, neuroleptic malignant syndrome, cardiovascular emergencies, and severe eating disorders frequently require care (such as IV medications, wound care, and cardiac monitoring) that exceeds the capabilities of inpatient psychiatric units (Chan et al., 2018; Robbins-Welty & Gagliardi, 2023; Wittink et al., 2020). At the same time, behavioral symptoms create excessive demands on the medical team, and may adversely impact medical team engagement (Sledge et al., 2015). The traditional consultation-liaison model, though not standardized across all facilities, is the current approach to behavioral care needs in many healthcare systems. This approach is reactive, and consultations are sought when challenging behaviors or psychiatric symptoms emerge. Lack of consistency in care leads to missed opportunities, late requests, and delayed discharges. The interdisciplinary Behavioral Intervention Team (BIT) was implemented on the campus of Prisma Health Baptist Hospital in Columbia, SC to address and anticipate the needs for behavioral health support on medical units which often extended beyond a consult with a psychiatrist. Methods After obtaining approval from Prisma Health Institutional Review Board, a retrospective review was conducted to evaluate the correlation between the BIT interventions and patient safety events, high-risk behaviors, and lengths of stay in older adults (age > 55 years) on a medical unit. Two time periods were examined: April 1st, 2021 to September 30th, 2021 (six months without BIT) and October 1st, 2021 to March 31st, 2022 (six months with BIT). Medical records were reviewed and data was collected including demographic information, diagnoses, and length of stay. Documentation of events in both safety reports and the electronic medical record was reviewed and categorized (Figure 1). Qualitative analysis of the Field Notes was guided by Paul Riceours' methodology. Researchers completed a naïve reading of each entry in order to gain an initial understanding of the text. After interpretation, researchers conducted structural analysis, splitting texts into units of meaning and units of significance, requiring researchers to divide, re-divide, and re-align texts via a dialectal and iterative process to the emergence of patterns, sub-themes, and main themes surrounded by contextual fabric. The last phase of analysis involved critical interpretation and discussion against the backdrop of the health care environment (Ricoeur, 1976; Simony et al., 2018). Descriptive statistics was used to summarize patient factors, clinical factors, and outcomes. Mean and standard deviation was calculated for all continuous variables, and counts and proportions for categorical variables. For normally distributed continuous variables, group means was compared using t-tests or an Analysis of Variance (ANOVA). For non-normally distributed continuous variables, the Wilcoxon Rank Sum or the Kruskal-Wallis tests was used and compared to t-test/ANOVA results. Normality was evaluated using the Shapiro-Wilk test and visual inspection of histogram plots with a normal curve overlay. For categorical variables, a Chi-square test or Fisher's exact test was used to evaluate group level differences. Patient, clinical, and other risk factor variables associated with research aims and hypotheses was evaluated with appropriate correlation analyses and/or regression models to determine associations between factors and outcomes, obtaining parameter estimates with 95-percent confidence intervals and p-values. Statistical significance was based on associated p-values (p<.05). All analyses was conducted in SAS Enterprise Guide 8.3. Results Preliminary results reveal that the majority of admitted patients were age 60 or above and presented with metabolic derangements or infection/sepsis. Of patients with a pre-existing psychiatric diagnosis, -Abstract Truncated-
psychiatry,geriatrics & gerontology,gerontology