Operationalizing a real-time scoring model to predict fall risk among older adults in the emergency department.

Predictive models are increasingly being developed and implemented to improve patient care across a variety of clinical scenarios. While a body of literature exists on the development of models using existing data, less focus has been placed on practical operationalization of these models for deployment in real-time production environments. This case-study describes challenges and barriers identified and overcome in such an operationalization for a model aimed at predicting risk of outpatient falls after Emergency Department (ED) visits among older adults. Based on our experience, we provide general principles for translating an EHR-based predictive model from research and reporting environments into real-time operation.

Collaborative design and implementation of a clinical decision support system for automated fall-risk identification and referrals in emergency departments.

Of the 3 million older adults seeking fall-related emergency care each year, nearly one-third visited the Emergency Department (ED) in the previous 6 months. ED providers have a great opportunity to refer patients for fall prevention services at these initial visits, but lack feasible tools for identifying those at highest-risk. Existing fall screening tools have been poorly adopted due to ED staff/provider burden and lack of workflow integration. To address this, we developed an automated clinical decision support (CDS) system for identifying and referring older adult ED patients at risk of future falls. We engaged an interdisciplinary design team (ED providers, health services researchers, information technology/predictive analytics professionals, and outpatient Falls Clinic staff) to collaboratively develop a system that successfully met user requirements and integrated seamlessly into existing ED workflows. Our rapid-cycle development and evaluation process employed a novel combination of human-centered design, implementation science, and patient experience strategies, facilitating simultaneous design of the CDS tool and intervention implementation strategies. This included defining system requirements, systematically identifying and resolving usability problems, assessing barriers and facilitators to implementation (e.g., data accessibility, lack of time, high patient volumes, appointment availability) from multiple vantage points, and refining protocols for communicating with referred patients at discharge. ED physician, nurse, and patient stakeholders were also engaged through online surveys and user testing. Successful CDS design and implementation required integration of multiple new technologies and processes into existing workflows, necessitating interdisciplinary collaboration from the onset. By using this iterative approach, we were able to design and implement an intervention meeting all project goals. Processes used in this Clinical-IT-Research partnership can be applied to other use cases involving automated risk-stratification, CDS development, and EHR-facilitated care coordination.

Machine learning-assisted screening for cognitive impairment in the emergency department.

BACKGROUND/OBJECTIVES: Despite a high prevalence and association with poor outcomes, screening to identify cognitive impairment (CI) in the emergency department (ED) is uncommon. Identification of high-risk subsets of older adults is a critical challenge to expanding screening programs. We developed and evaluated an automated screening tool to identify a subset of patients at high risk for CI. METHODS: In this secondary analysis of existing data collected for a randomized control trial, we developed machine-learning models to identify patients at higher risk of CI using only variables available in electronic health record (EHR). We used records from 1736 community-dwelling adults age > 59 being discharged from three EDs. Potential CI was determined based on the Blessed Orientation Memory Concentration (BOMC) test, administered in the ED. A nested cross-validation framework was used to evaluate machine-learning algorithms, comparing area under the receiver-operator curve (AUC) as the primary metric of performance. RESULTS: Based on BOMC scores, 121 of 1736 (7%) participants screened positive for potential CI at the time of their ED visit. The best performing algorithm, an XGBoost model, predicted BOMC positivity with an AUC of 0.72. With a classification threshold of 0.4, this model had a sensitivity of 0.73, a specificity of 0.64, a negative predictive value of 0.97, and a positive predictive value of 0.13. In a hypothetical ED with 200 older adult visits per week, the use of this model would lead to a decrease in the in-person screening burden from 200 to 77 individuals in order to detect 10 of 14 patients who would fail a BOMC. CONCLUSION: This study demonstrates that an algorithm based on EHR data can define a subset of patients at higher risk for CI. Incorporating such an algorithm into a screening workflow could allow screening efforts and resources to be focused where they have the most impact.

Association between social isolation and outpatient follow-up in older adults following emergency department discharge.

OBJECTIVES: Follow-up with outpatient clinicians after discharge from the emergency department (ED) reduces adverse outcomes among older adults, but rates are suboptimal. Social isolation, a common factor associated with poor health outcomes, may help explain these low rates. This study evaluates social isolation as a predictor of outpatient follow-up after discharge from the ED. MATERIALS AND METHODS: This cohort study uses the control group from a randomized-controlled trial investigating a community paramedic-delivered Care Transitions Intervention with older patients (age≥60 years) at three EDs in mid-sized cities. Social Isolation scores were measured at baseline using the PROMIS 4-item social isolation questionnaire, grouped into tertiles for analysis. Chart abstraction was conducted to identify follow-up with outpatient primary or specialty healthcare providers and method of contact within 7 and 30 days of discharge. RESULTS: Of 642 patients, highly socially-isolated adults reported significantly worse overall health, as well as increased anxiety, depressive symptoms, functional limitations, and co-morbid conditions compared to those less socially-isolated (p<0.01). We found no effect of social isolation on 30-day follow-up. Patients with high social isolation, however, were 37% less likely to follow-up with a provider in-person within 7 days of ED discharge compared to low social isolation (OR:0.63, 95% CI:0.42-0.96). CONCLUSION: This study adds to our understanding of how and when socially-isolated older adults seek outpatient care following ED discharge. Increased social isolation was not significantly associated with all-contact follow-up rates after ED discharge. However, patients reporting higher social isolation had lower rates of in-person follow-up in the week following ED discharge.

Factors Influencing Emergency Care by Persons With Dementia: Stakeholder Perceptions and Unmet Needs.

BACKGROUND/OBJECTIVES: People with dementia (PwD) frequently use emergency care services. To mitigate the disproportionately high rate of emergency care use by PwD, an understanding of contributing factors driving reliance on emergency care services and identification of feasible alternatives are needed. This study aimed to identify clinician, caregiver, and service providers' views and experiences of unmet needs leading to emergency care use among community-dwelling PwD and alternative ways of addressing these needs. DESIGN: Qualitative, employing semistructured interviews with clinicians, informal caregivers, and aging service providers. SETTING: Wisconsin, United States. PARTICIPANTS: Informal caregivers of PwD (n = 4), emergency medicine physicians (n = 4), primary care physicians (n = 5), geriatric healthcare providers (n = 5), aging service providers (n = 6), and community paramedics (n = 3). MEASUREMENTS: Demographic characteristics of participants and data from semistructured interviews. FINDINGS: Four major themes were identified from interviews: (1) system fragmentation influences emergency care use by PwD, (2) informational, decision-making, and social support needs influence emergency care use by PwD, (3) emergency departments (EDs) are not designed to optimally address PwD and caregiver needs, and (4) options to prevent and address emergency care needs of PwD. CONCLUSION: Participants identified numerous system and individual-level unmet needs and offered many recommendations to prevent or improve ED use by PwD. These novel findings, aggregating the perspectives of multiple dementia-care stakeholder groups, serve as the first step to developing interventions that prevent the need for emergency care or deliver tailored emergency care services to this vulnerable population through new approaches. J Am Geriatr Soc 67:711-718, 2019.

The placemat protocol: Measuring preschoolers' healthy-meal schemas with pretend meals.

Nutrition instruction can lead to more healthful food choices among children, but little is known about preschoolers' healthy-meal schemas because there are few developmentally appropriate measures. This study validated the Placemat Protocol, a novel measure of preschooler healthy-meal schemas using realistic food models to assemble pretend meals. Preschoolers (N = 247, mean age 4 years 8 months) created 2 meals (preferred and healthy), completed measures of verbal nutrition knowledge and vocabulary, and were weighed and measured for BMI. Parents reported healthy eating guidance, child dietary intake, and family demographics. Children used an average of 5.1 energy-dense (ED) and 3.4 nutrient-dense (ND) foods for their preferred meal, but reversed the ratio to 3.1 ED and 5.1 ND foods for their healthy meal. Healthy meals contained fewer estimated kcal, less fat, less sugar, and more fiber than preferred meals. Meal differences held for younger children, children with lower verbal nutrition knowledge and vocabulary, and child subgroups at higher risk for obesity. Placemat Protocol data correlated with parent healthy eating guidance and child obesogenic dietary intake as expected. The Placemat Protocol shows promise for assessing developing healthy-meal schemas before children can fully articulate their knowledge on verbal measures.