Evaluating Social Determinants of Health Variables in Advanced Analytic and Artificial Intelligence Models for Cardiovascular Disease Risk and Outcomes: A Targeted Review.

Introduction/Purpose: Predictive models incorporating relevant clinical and social features can provide meaningful insights into complex interrelated mechanisms of cardiovascular disease (CVD) risk and progression and the influence of environmental exposures on adverse outcomes. The purpose of this targeted review (2018-2019) was to examine the extent to which present-day advanced analytics, artificial intelligence, and machine learning models include relevant variables to address potential biases that inform care, treatment, resource allocation, and management of patients with CVD. Methods: PubMed literature was searched using the prespecified inclusion and exclusion criteria to identify and critically evaluate primary studies published in English that reported on predictive models for CVD, associated risks, progression, and outcomes in the general adult population in North America. Studies were then assessed for inclusion of relevant social variables in the model construction. Two independent reviewers screened articles for eligibility. Primary and secondary independent reviewers extracted information from each full-text article for analysis. Disagreements were resolved with a third reviewer and iterative screening rounds to establish consensus. Cohen's kappa was used to determine interrater reliability. Results: The review yielded 533 unique records where 35 met the inclusion criteria. Studies used advanced statistical and machine learning methods to predict CVD risk (10, 29%), mortality (19, 54%), survival (7, 20%), complication (10, 29%), disease progression (6, 17%), functional outcomes (4, 11%), and disposition (2, 6%). Most studies incorporated age (34, 97%), sex (34, 97%), comorbid conditions (32, 91%), and behavioral risk factor (28, 80%) variables. Race or ethnicity (23, 66%) and social variables, such as education (3, 9%) were less frequently observed. Conclusions: Predictive models should adjust for race and social predictor variables, where relevant, to improve model accuracy and to inform more equitable interventions and decision making.

Population Segmentation Using a Novel Socio-Demographic Dataset.

Appending market segmentation data to a national healthcare knowledge, attitude and behavior survey and medical claims by geocode can provide valuable insight for providers, payers and public health entities to better understand populations at a hyperlocal level and develop cohort-specific strategies for health improvement. A prolonged use case investigates population factors, including social determinants of health, in depression and develops cohort-level management strategies, utilizing market segmentation and survey data. Survey response scores for each segment were normalized against the average national score and appended to claims data to identify at-risk segment whose scores were compared with three socio-demographically comparable but not at-risk segments via Nonparametric Mann-Whitney U test to identify specific risk factors for intervention. The marketing segment, New Melting Point (NMP), was identified as at-risk. The median scores of three comparable segments differed from NMP in "Inability to Pay For Basic Needs" (121% vs 123%), "Lack of Transportation" (112% vs 153%), "Utilities Threatened" (103% vs 239%), "Delay Visiting MD" (67% vs 181%), "Delay/Not Fill Prescription" (117% vs 182%), "Depressed: All/Most Time" (127% vs 150%), and "Internet: Virtual Visit" (55% vs 130%) (all with p<0.001). The appended dataset illustrates NMP as having many stressors (e.g., difficult social situations, delaying seeking medical care). Strategies to improve depression management in NMP could employ virtual visits, or pharmacy incentives. Insights gleaned from appending market segmentation and healthcare utilization survey data can fill in knowledge gaps from claims-based data and provide practical and actionable insights for use by providers, payers and public health entities.

Digital Health Interventions to Enhance Prevention in Primary Care: Scoping Review.

BACKGROUND: Disease prevention is a central aspect of primary care practice and is comprised of primary (eg, vaccinations), secondary (eg, screenings), tertiary (eg, chronic condition monitoring), and quaternary (eg, prevention of overmedicalization) levels. Despite rapid digital transformation of primary care practices, digital health interventions (DHIs) in preventive care have yet to be systematically evaluated. OBJECTIVE: This review aimed to identify and describe the scope and use of current DHIs for preventive care in primary care settings. METHODS: A scoping review to identify literature published from 2014 to 2020 was conducted across multiple databases using keywords and Medical Subject Headings terms covering primary care professionals, prevention and care management, and digital health. A subgroup analysis identified relevant studies conducted in US primary care settings, excluding DHIs that use the electronic health record (EHR) as a retrospective data capture tool. Technology descriptions, outcomes (eg, health care performance and implementation science), and study quality as per Oxford levels of evidence were abstracted. RESULTS: The search yielded 5274 citations, of which 1060 full-text articles were identified. Following a subgroup analysis, 241 articles met the inclusion criteria. Studies primarily examined DHIs among health information technologies, including EHRs (166/241, 68.9%), clinical decision support (88/241, 36.5%), telehealth (88/241, 36.5%), and multiple technologies (154/241, 63.9%). DHIs were predominantly used for tertiary prevention (131/241, 54.4%). Of the core primary care functions, comprehensiveness was addressed most frequently (213/241, 88.4%). DHI users were providers (205/241, 85.1%), patients (111/241, 46.1%), or multiple types (89/241, 36.9%). Reported outcomes were primarily clinical (179/241, 70.1%), and statistically significant improvements were common (192/241, 79.7%). Results were summarized across the following 5 topics for the most novel/distinct DHIs: population-centered, patient-centered, care access expansion, panel-centered (dashboarding), and application-driven DHIs. The quality of the included studies was moderate to low. CONCLUSIONS: Preventive DHIs in primary care settings demonstrated meaningful improvements in both clinical and nonclinical outcomes, and across user types; however, adoption and implementation in the US were limited primarily to EHR platforms, and users were mainly clinicians receiving alerts regarding care management for their patients. Evaluations of negative results, effects on health disparities, and many other gaps remain to be explored.

Can de facto dosing practices bridge the knowledge gap in pediatric medication recommendations?

A knowledge gap exists in pediatric medication dosing recommendations due in part to the complexity of researching medication efficacy and safety in children. One possible resource resides in the electronic prescribing practices of pediatric clinicians. In this study, de facto pediatric weight-based levothyroxine dosing practices were studied as a potential source for pediatric medication clinical decision support. This was accomplished by extracting physical exam and prescription details from a well-used clinical data warehouse to calculate weight-based dosing practices, and comparing the results with established medication recommendations. Of the 854 prescriptions, 85.2% were under the recommended range, 9.37% were within the range and 5.39% were over the range. Thus, real world prescribing practices may differ from recommendations. Such information may be a valuable resource in pediatric clinical decision support, particularly where practice differs from recommendations, and can help close the knowledge gap where pediatric medication dosing information is sparse or unavailable.

A study of the age attribute in a query tool for a clinical data warehouse.

The RPDR, a clinical data warehouse with a user-friendly Querytool, allows researchers to perform studies on patient data. Currently, the RPDR represents age as the patient's age at the present time, which is problematic in situations where age at the time of the event is more appropriate. We will modify the Querytool to consider this by assessing the perception of age via survey, testing backend query solutions, and developing modifications based on these results.