IntroductionAdaptation is a key aspect of implementation science; interventions frequently need adaptation to better fit their delivery contexts and intended users and recipients. As digital health interventions are rapidly developed and expanded, it is important to understand how such interventions are modified. This paper details the process of engaging end-users in adapting the PREVENT digital health intervention for rural adults and systematically reporting adaptations using the Framework for Reporting Adaptations and Modifications-Enhanced (FRAME). The secondary objective was to tailor FRAME for digital health interventions and to document potential implications for equity.MethodsPREVENT's adaptations were informed by two pilot feasibility trials and a planning grant which included advisory boards, direct clinic observations, and qualitative interviews with patients, caregivers, and healthcare team members. Adaptations were catalogued in an Excel tracker, including a brief description of the change. Pilot coding was conducted on a subset of adaptations to revise the FRAME codebook and generate consensus. We used a directed content analysis approach and conducted a secondary data analysis to apply the revised FRAME to all adaptations made to PREVENT (n = 20).ResultsAll but one adaptation was planned, most were reactive (versus proactive), and all adaptations preserved fidelity to PREVENT. Adaptations were made to content and features of the PREVENT tool and may have positive implications for equity that will be tested in future trials.ConclusionEngaging rural partners to adapt our digital health tool prior to implementation with rural adults was critical to meet the unique needs of rural, low-income adult patients, fit the rural clinical care settings, and increase the likelihood of generating the intended impact among this patient population. The digital health expansion of FRAME can be applied prospectively or retrospectively by researchers and practitioners to plan, understand, and characterize digital health adaptations. This can aid intervention design, scale up, and evaluation in the rapidly expanding area of digital health.

Violent injury survivors are at risk for revictimization. The St. Louis area hospital-based violence intervention program (HVIP), Life Outside of Violence (LOV), is the first multisystem, region-wide HVIP in the United States. To describe the LOV program during its pilot phase and evaluate violent reinjury 1 year after index injury among LOV participants. Pilot observational cohort study of violently injured patients. Data were queried from the program's multisystem data repository, which contains individual encounter-level data for all violent injury visits from the 2 adult and 2 pediatric LOV-partner level 1 trauma hospitals. Patients eligible for LOV who were violently injured between 15 August 2018 and 31 December 2022 and enrolled in LOV, matched to control participants of nonenrolled LOV-eligible patients selected from the data repository using propensity score matching. Participation in the LOV program. Sociodemographic characteristics, predictors of LOV enrollment, and returning to a LOV partner hospital with a violent reinjury within 1 year of index injury. The probability of violent reinjury and 95% CIs were compared between LOV participants and control participants using Kaplan-Meier estimates. 233 of 3744 eligible patients enrolled in LOV. Of 198 LOV-enrolled participants matched to 388 nonenrolled control participants, Kaplan-Meier estimates for 1-year probability of reinjury were 7.6% (95% CI, 3.8% to 11.2%) among LOV participants and 7.4% (CI, 4.8% to 10.0%) in control participants. This pilot study cannot provide precise estimates of LOV intervention efficacy. Although no informative evidence of differences in reinjury probability for LOV participants was seen, findings suggest that the overwhelming risk in which our patients are immersed cannot be overcome by an approach scaled for individual-level impact. Missouri Foundation for Health.

Digital health tools are positive for delivering evidence-based care. However, few studies have applied rigorous frameworks to understand their use in community settings. This study aimed to identify implementation determinants of the Automated Heart-Health Assessment (AH-HA) tool within outpatient oncology settings as part of a hybrid effectiveness-implementation trial. A mixed-methods approach informed by the Consolidated Framework for Implementation Research (CFIR) examined barriers and facilitators to AH-HA implementation in four NCI Community Oncology Research Program (NCORP) practices participating in the WF-1804CD AH-HA trial. Provider surveys were analyzed using descriptive statistics. Interviews with providers (n = 15) were coded using deductive (CFIR) and inductive codes by trained analysts. The CFIR rating tool was used to rate each quote for (i) valence, defined as a positive (+) or negative (-) influence, and (ii) strength, defined as a neutral (0), weak (1), or strong (2) influence on implementation. All providers considered discussing cardiovascular health with patients as important (61.5%, n = 8/13) or somewhat important (38.5%, n = 5/13). The tool was well-received by providers and was feasible to use in routine care among cancer survivors. Providers felt the tool was acceptable and usable, had a relative advantage over routine care, and had the potential to generate benefits for patients. Common reasons clinicians reported not using AH-HA were (i) insufficient time and (ii) the tool interfering with workflow. Systematically identifying implementation determinants from this study will guide the broader dissemination of the AH-HA tool across clinical settings and inform implementation strategies for future scale-up hybrid trials.

BackgroundMost survivors of cancer have multiple cardiovascular risk factors, increasing their risk of poor cardiovascular and cancer outcomes. The Automated Heart-Health Assessment (AH-HA) tool is a novel electronic health record clinical decision support tool based on the American Heart Association’s Life’s Simple 7 cardiovascular health metrics to promote cardiovascular health assessment and discussion in outpatient oncology. Before proceeding to future implementation trials, it is critical to establish the acceptability of the tool among providers and survivors. ObjectiveThis study aims to assess provider and survivor acceptability of the AH-HA tool and provider training at practices randomized to the AH-HA tool arm within WF-1804CD. MethodsProviders (physicians, nurse practitioners, and physician assistants) completed a survey to assess the acceptability of the AH-HA training, immediately following training. Providers also completed surveys to assess AH-HA tool acceptability and potential sustainability. Tool acceptability was assessed after 30 patients were enrolled at the practice with both a survey developed for the study as well as with domains from the Unified Theory of Acceptance and Use of Technology survey (performance expectancy, effort expectancy, attitude toward using technology, and facilitating conditions). Semistructured interviews at the end of the study captured additional provider perceptions of the AH-HA tool. Posttreatment survivors (breast, prostate, colorectal, endometrial, and lymphomas) completed a survey to assess the acceptability of the AH-HA tool immediately after the designated study appointment. ResultsProviders (n=15) reported high overall acceptability of the AH-HA training (mean 5.8, SD 1.0) and tool (mean 5.5, SD 1.4); provider acceptability was also supported by the Unified Theory of Acceptance and Use of Technology scores (eg, effort expectancy: mean 5.6, SD 1.5). Qualitative data also supported provider acceptability of different aspects of the AH-HA tool (eg, “It helps focus the conversation and give the patient a visual of continuum of progress”). Providers were more favorable about using the AH-HA tool for posttreatment survivorship care. Enrolled survivors (n=245) were an average of 4.4 (SD 3.7) years posttreatment. Most survivors reported that they strongly agreed or agreed that they liked the AH-HA tool (n=231, 94.3%). A larger proportion of survivors with high health literacy strongly agreed or agreed that it was helpful to see their heart health score (n=161, 98.2%) compared to survivors with lower health literacy scores (n=68, 89.5%; P=.005). ConclusionsQuantitative surveys and qualitative interview data both demonstrate high acceptability of the AH-HA tool among both providers and survivors. Although most survivors found it helpful to see their heart health score, there may be room for improving communication with survivors who have lower health literacy. Trial RegistrationClinicalTrials.gov NCT03935282; http://clinicaltrials.gov/ct2/show/NCT03935282 International Registered Report Identifier (IRRID)RR2-https://doi-org.wake.idm.oclc.org/10.1016/j.conctc.2021.100808

12019 Background: Cardiovascular disease causes significant morbidity and mortality among US survivors. To enhance guideline recommended cardiovascular health (CVH) discussions during survivorship care, our team developed the EHR-based AH-HA clinical decision support tool, which displays modifiable CVH factors and cancer treatments with cardiotoxic potential. This tool significantly improved delivery of guideline-concordant CVH discussions, the primary study outcome; here we report AH-HA impacts on 12-month CVH improvements. Methods: The Wake Forest NCORP Research Base coordinated this practice-randomized clinical trial (NCT# 03935282) comparing AH-HA and usual care (UC) practices. Participants were survivors ≥ 6 months post-potentially curative treatment for breast, prostate, colorectal, endometrial cancers, or lymphoma, scheduled for routine follow-up. The AH-HA tool, based on the American Heart Association (AHA) Life’s Simple 7, aimed to enhance CVH awareness and action by providers and survivors. At AH-HA practices, providers used the tool with survivors during an outpatient oncology visit. CVH data were collected from the EHR and survivors (diet quality & physical activity) at baseline and 12 months. Outcomes included Simple 7 total CVH score (0-100, using AHA algorithm) and meaningful change in individual CVH factors (Table). Generalized estimating equations calculated rates of clinically meaningful improvements in CVH factors at 12 months by group, adjusting for cancer type and clustering within practice. Results: 645 survivors (82.3% breast cancer; 96.0% female; 83.9% white non-Hispanic, 7.8% Black, 3.7% Hispanic) enrolled at 9 practices (5 UC and 4 AH-HA). The total CVH score did not significantly change from baseline or between groups (Table 1, p>.05). Within the AH-HA arm, 20.3% of survivors achieved 5% weight reduction compared to 12.6% in UC; physical activity also improved more in the AH-HA arm, but not significantly. Rates were similar between arms for diet, blood pressure, and hemoglobin A1c. Conclusions: In addition to facilitating guideline concordant CVH discussions, AH-HA shows promise for encouraging weight loss among survivors. It is notable that a brief intervention delivered as part of standard oncology care impacted weight reduction. Clinical trial information: NCT03935282 . 12 month cardiovascular health outcomes among post-treatment survivors. CVH Outcomes AH-HA 4 practices, (n=281) Usual Care 5 practices, (n=342) Adj P-value Improvement, % Yes BMI (5% weight ↓) 20.3 12.6 0.02 Blood Pressure (5 mm ↓) 53.1 52.0 0.75 Physical Activity (+ 30 mins) 40.9 34.3 0.14 Healthy Diet Score (0-1 to 2-5, or 2-3 to 4-5 components) 21.7 18.9 0.45 Cholesterol (20% ↓) 8.6 9.6 # A1c (0.5% ↓) 8.9 8.8 0.97 Change in total CVH Score adjusted for baseline (positive=improvement) 0.9 -0.5 0.28 #Model did not converge due to small sample size.

INTRODUCTION:Neurofibromatosis type 1 (NF1) is a rare genetic disorder affecting multiple organ systems with significant clinical heterogeneity. Managing individuals with NF1 is challenging due to variability in disease progression and outcomes and limited early risk assessment tools. OBJECTIVE:This study aims to develop an effective, generalizable, user-friendly clinical entity extraction pipeline for identifying NF1-related phenotypes from unstructured clinical notes to enhance research and risk-modeling efforts. We compare the benefits of rule-based natural language processing (NLP) vs large language models (LLMs) for this purpose. MATERIALS AND METHODS:Four phenotype extraction pipelines (3 LLM-based vs 1 rule-based) were developed to automatically extract selected NF1-relevant phenotypes. Subject matter experts manually reviewed clinical notes, generating a gold-standard annotation dataset for evaluation. In Phase 1, notes authored by a single NF1 physician were used to guide pipeline development and refinement. In Phase 2, notes from a second NF1 physician were used to assess pipeline generalizability, followed by further refinement to accommodate differences in physician terminology. RESULTS:With refinement, the rule-based model had higher distributions of F1 scores than the LLMs in both Phase 1 and Phase 2. However, the LLMs demonstrated better generalizability between physicians without refinement, showing lesser performance decreases (4.4%-5.1%) when transitioning from Phase 1 to Phase 2 without refinement, compared to an 8.8% decrease for the rule-based model. CONCLUSION:We highlight trade-offs between the effectiveness of rule-based NLP vs generalizability and ease of implementation of LLMs for clinical entity extraction, with implications for pipeline portability across providers and institutions.

Peak oxygen consumption (V̇O2peak) is used to predict outcomes and time to transplantation in patients with heart failure with reduced ejection fraction (HFrEF); V̇O2peak also has predictive utility in patients with adult congenital heart disease (ACHD). However, the predictive value of a given V̇O2peak on cardiac events in patients with ACHD compared to HFrEF, especially after adjustment for age and sex, is unclear. Therefore, we performed a longitudinal cohort study comparing patients with ACHD to patients with HFrEF. The cohorts were sex and age matched (±10 years). V̇O2peak tests were conducted from 1993 to 2012. Cardiac events included death, cardiac transplantation, and LVAD placement. Events were obtained via electronic medical record, SSDI, and phone interview. Cox proportional-hazard regression analyses were used to evaluate relationships of event-free survival with predictor variables. Patients with ACHD (N = 137) and HFrEF (N = 137) had median follow-up times of 19.0 years (14.8 to 21.1) and 14.5 years (13.4 to 15.6), respectively. In multivariable models, Higher V̇O2peak was associated with lower risk for a cardiac event, independent of age and sex, in both ACHD (HR 0.89, 95% CI 0.83 to 0.96, p = 0.002) and HFrEF (HR 0.86, 95% CI 0.82 to 0.91, p <0.001). Male sex was associated with greater risk of a cardiac event HFrEF (HR 1.90, 95% CI 1.24 to 2.90, p = 0.003) but not in ACHD group. After multivariable adjustment (Beta-blockers, sex, and V̇O2peak), having ACHD conferred a 71% lower risk of cardiac events compared to a HFrEF diagnosis (HR 0.29, 95% CI 0.18 to 0.47, p <0.001). V̇O2peak independently predicts event-free survival among adults with ACHD or HFrEF and has clinical utility in outpatient settings. Patients with ACHD have a better prognosis after multivariable adjustment including V̇O2peak compared to HFrEF.

Objective:To implement the BREASTChoice decision tool into the electronic health record and evaluate its effectiveness.Background: BREASTChoice, is a multilevel decision tool that (1) educates patients about breast reconstruction, (2) estimates personalized risk of complications, (3) clarifies patient preferences, and (4) informs clinicians about patients' risk and preferences.Methods:A multisite randomized controlled trial enrolled adult women with stage 0 to III breast malignancy undergoing mastectomy. Participants were randomized to BREASTChoice or a control website. A survey assessed knowledge, preferences, decisional conflict, shared decision-making, preferred treatment, and usability. We conducted intent-to-treat (ITT), per-protocol (PP) analyses (those randomized to BREASTChoice who accessed the tool), and stratified analyses.Results:A total of 23/25 eligible clinicians enrolled. A total of 369/761 (48%) contacted patients enrolled and were randomized. Patients' average age was 51 years; 15% were older than 65. BREASTChoice participants had higher knowledge than control participants (ITT: mean 70.6 vs 67.4, P=0.08; PP: mean 71.4 vs 67.4, P=0.03), especially when stratified by site (ITT: P=0.04, PP: P=0.01), age (ITT: P=0.04, PP: P=0.02), and race (ITT: P=0.04, PP: P=0.01). BREASTChoice did not improve decisional conflict, match between preferences and treatment, or shared decision-making. In PP analyses, fewer high-risk patients using BREASTChoice chose reconstruction. BREASTChoice had high usability.Conclusions: BREASTChoice is a novel decision tool incorporating risk prediction, patient education, and clinician engagement. Patients using BREASTChoice had higher knowledge; older adults and those from racially minoritized backgrounds especially benefitted. There was no impact on other decision outcomes. Future studies should overcome implementation barriers and specifically examine decision outcomes among high-risk patients.

BACKGROUND:Violent injury survivors are at risk for revictimization. The St. Louis area hospital-based violence intervention program (HVIP), Life Outside of Violence (LOV), is the first multisystem, region-wide HVIP in the United States. OBJECTIVE:To describe the LOV program during its pilot phase and evaluate violent reinjury 1 year after index injury among LOV participants. DESIGN:Pilot observational cohort study of violently injured patients. SETTING:Data were queried from the program's multisystem data repository, which contains individual encounter-level data for all violent injury visits from the 2 adult and 2 pediatric LOV-partner level 1 trauma hospitals. PARTICIPANTS:Patients eligible for LOV who were violently injured between 15 August 2018 and 31 December 2022 and enrolled in LOV, matched to control participants of nonenrolled LOV-eligible patients selected from the data repository using propensity score matching. INTERVENTION:Participation in the LOV program. MEASUREMENTS:Sociodemographic characteristics, predictors of LOV enrollment, and returning to a LOV partner hospital with a violent reinjury within 1 year of index injury. The probability of violent reinjury and 95% CIs were compared between LOV participants and control participants using Kaplan-Meier estimates. RESULTS:233 of 3744 eligible patients enrolled in LOV. Of 198 LOV-enrolled participants matched to 388 nonenrolled control participants, Kaplan-Meier estimates for 1-year probability of reinjury were 7.6% (95% CI, 3.8% to 11.2%) among LOV participants and 7.4% (CI, 4.8% to 10.0%) in control participants. LIMITATION:This pilot study cannot provide precise estimates of LOV intervention efficacy. CONCLUSION:Although no informative evidence of differences in reinjury probability for LOV participants was seen, findings suggest that the overwhelming risk in which our patients are immersed cannot be overcome by an approach scaled for individual-level impact. PRIMARY FUNDING SOURCE:Missouri Foundation for Health.

Objective:Dimensionality reduction techniques aim to enhance the performance of machine learning (ML) models by reducing noise and mitigating overfitting. We sought to compare the effect of different dimensionality reduction methods for comorbidity features extracted from electronic health records (EHRs) on the performance of ML models for predicting the development of various sub-phenotypes in children with Neurofibromatosis type 1 (NF1). Materials and Methods:EHR-derived data from pediatric subjects with a confirmed clinical diagnosis of NF1 were used to create 10 unique comorbidities code-derived feature sets by incorporating dimensionality reduction techniques using raw International Classification of Diseases codes, Clinical Classifications Software Refined, and Phecode mapping schemes. We compared the performance of logistic regression, XGBoost, and random forest models utilizing each feature set. Results:XGBoost-based predictive models were most successful at predicting NF1 sub-phenotypes. Overall, features based on domain knowledge-informed mapping schema performed better than unsupervised feature reduction methods. High-level features exhibited the worst performance across models and outcomes, suggesting excessive information loss with over-aggregation of features. Discussion:Model performance is significantly impacted by dimensionality reduction techniques and varies by specific ML algorithm and outcome being predicted. Automated methods using existing knowledge and ontology databases can effectively aggregate features extracted from EHRs. Conclusion:Dimensionality reduction through feature aggregation can enhance the performance of ML models, particularly in high-dimensional datasets with small sample sizes, commonly found in EHRs health applications. However, if not carefully optimized, it can lead to information loss and data oversimplification, potentially adversely affecting model performance.

Many patients with serious illness prioritize comfort over the prolongation of life in the final days and weeks prior to death. Goals-of-care discussions (GOCDs) can provide patients with the opportunity to express their preferred end-of-life experience, prevent aggressive and often futile interventions, improve patient satisfaction, and reduce unnecessary costs. Clinicians may feel uncomfortable initiating these conversations, however, given the lack of relevant training and difficulty identifying patients at elevated risk of mortality. The authors developed an intervention to promote GOCDs that combines mortality risk estimation using AI, clinician training, person-to-person alert messages, streamlined clinical workflows, and enhanced palliative care capacity. The program was implemented across 8 of 10 adult hospitals in the BJC HealthCare system, and data were collected between December 22, 2020 (at launch in the first hospital sites) and December 31, 2024. During this time, more than 300 clinicians were trained through the program, and they identified 13,976 patients as candidates for GOCDs. Clinicians exhibited a high response rate (93%) to the patient eligibility alert. Among patients without a prior documented GOCD at the time of messaging, 54% of responding physicians opted to initiate a GOCD with the patient, and another 24% requested palliative care to initiate the GOCD. Systemwide improvements were observed across several metrics, including a fivefold increase in GOCDs, from 1.2% of 146,257 encounters in 2021 to 6.7% of 167,681 encounters in 2024, and a 63% increase in the proportion of encounters with palliative care consults, from 2.2% of 146,257 encounters in 2021 to 3.6% of 167,681 encounters in 2024. The Vizient Mortality Index (a ratio of observed to expected mortality) also decreased by 32% during this time frame (where a lower score indicates that fewer patients died than would have been expected), from 0.92 in 2021 to 0.62 in 2024. Keys to implementation included identifying a staff member at each site to orchestrate alert messaging and ensuring adequate palliative care staffing at each site. This case study demonstrates the importance of pairing accurate mortality predictions with systems and resources that enable clinicians to act on these predictions, including engaging with alerts and improving through comprehensive clinical training.

BACKGROUND:Patients with end-stage kidney disease (ESKD) face significant barriers to accessing hospice services in the United States, where Medicare Local Coverage Determination (LCD) guidelines significantly contribute to establishing hospice eligibility, ostensibly by identifying patients with an estimated life expectancy of six months or less. OBJECTIVE:Researchers sought to determine whether LCD guidelines for ESKD accurately identified patients with a six-month prognosis. DESIGN:This study utilized a retrospective cohort design. SETTING/SUBJECTS:Medicare beneficiary data from a large midwestern Accountable Care Organization, collected between October 2017 and May 2024, were included in study analyses. Data were included for decedent patients who had chronic kidney disease at the time of death, met LCD guidelines for ESKD at the time of death, and had laboratory testing completed in the 180 days prior to death. DATA ANALYSIS:Data were analyzed via nonparametric maximum likelihood survival analyses to assess the length of time the patients had met LCD guidelines for ESKD prior to death. RESULTS:Among 769 patients who met LCD guidelines for ESKD at the time of death, the probability of a patient meeting LCD guidelines for ESKD for at least one month prior to death was 0.28 (95% confidence interval [CI]: 0.24, 0.32), the probability of meeting these guidelines for at least six months prior to death was 0.05 (95% CI: 0.03, 0.09), and the median time of meeting these guidelines was nine days prior to death. CONCLUSIONS:Results indicated that hospice LCD guidelines inaccurately predicted six-month life expectancy for patients with ESKD, suggesting they may be inappropriate for use in determining hospice eligibility.

The Automated Heart-Health Assessment (AH-HA) tool is a novel electronic health record clinical decision support tool based on the American Heart Association’s Life’s Simple 7 cardiovascular health (CVH) metrics to promote CVH assessment and discussion in outpatient oncology. Before proceeding to future implementation trials, it is critical to establish the acceptability of the tool among providers and survivors. We assessed provider and survivor acceptability of the AH-HA tool and provider training at practices randomized to the AH-HA tool arm within WF-1804CD. Providers (physicians, nurse practitioners, physician assistants) completed a survey to assess acceptability of the AH-HA training, immediately following training. Providers also completed surveys to assess AH-HA tool acceptability and potential sustainability. Tool acceptability was assessed after 30 patients were enrolled at the practice with both a survey developed for the study as well as with domains from the Unified Theory of Acceptance and Use of Technology (UTAUT) survey (Performance Expectancy, Effort Expectancy, Attitude Toward using Technology, and Facilitating Conditions). Semi-structured interviews at the end of the study captured additional provider perceptions of the AH-HA tool. Post-treatment survivors (breast, prostate, colorectal, endometrial, and lymphomas) completed a survey to assess acceptability of the AH-HA tool immediately after the designated study appointment. Providers (n=15) reported high overall acceptability of the AH-HA training (mean=5.8, SD=1.0) and tool (mean=5.5, SD =1.4); provider acceptability was also supported by UTAUT scores (e.g., Effort Expectancy mean=5.6, SD=1.5). Qualitative data also supported provider acceptability of different aspects of the AH-HA tool (e.g., It helps focus the conversation and give the patient a visual of continuum of progress). Providers were more favorable about using the AH-HA tool for post-treatment survivorship care. Enrolled survivors (n=245) were an average of 4.4 years post-treatment (SD =3.7). Most survivors reported that they strongly agreed/agreed that they liked the AH-HA tool (94.3%, n=231). A larger proportion of survivors with high health literacy strongly agreed/agreed that it was helpful to see their heart health score (98.2%, n=161) compared to survivors with lower health literacy scores (89.5%, n=68; p=0.005). Quantitative surveys and qualitative interview data both demonstrate high acceptability of the AH-HA tool among both providers and survivors. Although most survivors found it helpful to see their heart health score, there may be room for improving communication with survivors who have lower health literacy. Assessing Effectiveness and Implementation of an EHR Tool to Assess Heart Health Among Survivors (AH-HA) NCT03935282 https://clinicaltrials.gov/study/NCT03935282?term=NCT03935282&rank=1 RR2-https://doi-org.wake.idm.oclc.org/10.1016/j.conctc.2021.100808

Background This pilot study examined the preliminary effectiveness of the PREVENT digital intervention that supports health care teams in delivering health behavior counseling on cancer survivors’ motivation to change behavior, their physical activity and food intake behaviors, and cardiovascular health (CVH). Methods Clinicians (physicians, nurse practitioners) at three urban cancer survivorship clinics were trained to use PREVENT. Patients were randomized to the PREVENT intervention or a wait-list routine care control group. Eligibility criteria for patients included: between ages 12–39, overweight or obese, were at least 6-months post-active cancer treatment, and had sufficient English proficiency. Results Fifty-five participants were enrolled; 27 were randomized to the PREVENT intervention and 28 to wait-list routine care control. The majority of the participants (82%) identified as non-Hispanic white, with an average age of 19.8 (SD ± 5.2) years. Patients that received the PREVENT intervention had greater increases in their self-efficacy, vigorous activity and number of food recommendations met than those who received routine clinical care. Changes in willingness, knowledge, and CVH outcomes were not significant. Conclusions The PREVENT digital intervention may provide improvements in preventive behaviors among AYA cancer survivors by supporting care teams with delivering evidence-based, tailored behavior change recommendations and resources to support patient health. Trial registration This trial ( NCT04623190 ) was registered on 11/02/2022.

Outcomes 1. Participants will self-report the ability to name LCD criteria for chronic renal failure.2. Participants will be able to evaluate the prognostic utility of LCD guidelines compared to receiving an expected six months of hospice care. Key Message Chronic renal failure (CRF) disproportionately impacts Black Americans. Our data demonstrate that Medicare criteria for hospice service eligibility at six months among patients with CRF has poor prognostic utility and patients experienced less than one month of hospice care, which may further racial disparities in healthcare delivery in this population. Introduction Chronic renal failure (CRF) is a leading cause of death in the United States impacting Black Americans at more than three times the rate of white Americans. [1] Medicare local coverage determination (LCD) criteria are designed to help identify patients eligible for hospice care. For CRF, these guidelines require that a patient is not seeking dialysis, renal transplant, or is discontinuing dialysis, and has a creatinine clearance (GFR) < 15ml/min or serum creatinine > 8.0 mg/dl. [2] Objective To determine whether LCD guidelines for hospice referrals due to CRF accurately identify patients with a six-month prognosis. Methods Nonparametric maximum likelihood survival analysis was used to assess the time period during which patients met LCD guidelines for CRF stratified by dialysis utilization. We included patients with CRF who met LCD guidelines at the time of death and utilized accountable care organization outpatient data from BJC HealthCare in Saint Louis from January 2017 to February 2020. Results 221 patients met LCD criteria for CRF at the time of death, of which 94 (42.5%) had dialysis services recorded while 127 (57.5%) did not. The probability of a patient meeting LCD criteria for at least one month was 0.21 (95% confidence interval [CI]: 0.15, 0.26), decreasing to 0.04 (95% CI: 0.01, 0.07) for at least six months. All patients experienced a median time of 7 days for meeting criteria compared to 12 days for patients who did not utilize dialysis. Conclusion LCD guidelines failed to accurately identify CRF patients with a six-month prognosis. The low sensitivity of the CRF LCD identified in this preliminary evaluation may lead to inappropriate denials of hospice eligibility and disproportionately reduce access to hospice care for Black Americans. A data-driven approach is needed to develop more equitable guidelines. Keywords Ethical / Legal Aspects of Care; Disease specific management

PURPOSE:Guidelines recommend cardiovascular (CV) risk assessment and counseling for cancer survivors. This study evaluated the automated heart-health assessment (AH-HA) clinical decision support tool to promote provider-patient CV health (CVH) discussions in outpatient oncology. METHODS:The AH-HA trial (WF-1804CD), coordinated by the Wake Forest National Cancer Institute Community Oncology Research Program Research Base, randomized practices to the AH-HA tool or usual care (UC) and enrolled survivors receiving routine care ≥6 months after curative cancer treatment. The tool displayed American Heart Association Life's Simple 7 CVH factors (BMI, physical activity, diet, smoking status, blood pressure, cholesterol, and glucose), populated from the electronic health record (EHR), alongside cancer treatments received with cardiotoxic potential. The primary end point was survivor-reported discussion of nonideal or missing CVH factors. A mixed-effects logistic regression model assessed the effect of AH-HA on CVH discussions, adjusting for practice. RESULTS:Five UC and four AH-HA practices enrolled 645 survivors (82% breast, 8% endometrial, 5% colorectal, and 5% lymphoma, prostate, or multiple types) from October 1, 2020, to February 28, 2023. Most survivors were female (96%; 84% White/non-Hispanic, 8% Black; 3% Hispanic). Nearly all survivors (98%) in AH-HA practices reported a discussion for ≥1 nonideal or missing CVH factor compared with 55% in UC (P < .001). The average number of survivor-reported factors discussed was higher in AH-HA compared with UC (mean, 4.06 v 1.27; P < .001), as were EHR-documented discussions (3.83 v 0.77; P = .03). Survivors in AH-HA practices were also significantly more likely to report a recommendation to see a primary care provider (39%) compared with UC practices (25%, P = .02). Reported recommendations to see a cardiologist were low (approximately 6%) and did not differ between groups. CONCLUSION:The AH-HA tool was effective at promoting CVH discussions during routine follow-up care for survivors and recommendations to consult primary care.

Alcohol use and smoking are common substance-use behaviors with well-established negative health effects, including decreased brain health. We examined whether alcohol use and smoking were associated with the same neuroimaging-derived brain measures. We further explored whether the effects of alcohol use and smoking on the brain were additive or interactive. We leveraged a cohort of 36,309 participants with neuroimaging data from the UK Biobank. We used linear regression to determine the association between 354 neuroimaging-derived brain measures and alcohol use defined as drinks per week, pack years of smoking, and drinks per week × pack years smoking interaction. To assess whether the brain associations with alcohol are broadly similar or different from the associations with smoking, we calculated the correlation between z-scores of association for drinks per week and pack years smoking. Results indicated overall moderate positive correlation in the associations across measures representing brain structure, magnetic susceptibility, and white matter tract microstructure, indicating greater similarity than difference in the brain measures associated with alcohol use and smoking. The only evidence of an interaction between drinks per week and pack years smoking was seen in measures representing magnetic susceptibility in subcortical structures. The effects of alcohol use and smoking on brain health appeared to be additive rather than multiplicative for all other brain measures studied. 97% (224/230) of associations with alcohol and 100% (167/167) of the associations with smoking that surpassed a p value threshold are in a direction that can be interpreted to reflect reduced brain health. Our results underscore the similarity of the adverse associations between use of these substances and neuroimaging derived brain measures.