Preparing for the bedside-optimizing a postpartum depression risk prediction model for clinical implementation in a health system.

OBJECTIVE: We developed and externally validated a machine-learning model to predict postpartum depression (PPD) using data from electronic health records (EHRs). Effort is under way to implement the PPD prediction model within the EHR system for clinical decision support. We describe the pre-implementation evaluation process that considered model performance, fairness, and clinical appropriateness. MATERIALS AND METHODS: We used EHR data from an academic medical center (AMC) and a clinical research network database from 2014 to 2020 to evaluate the predictive performance and net benefit of the PPD risk model. We used area under the curve and sensitivity as predictive performance and conducted a decision curve analysis. In assessing model fairness, we employed metrics such as disparate impact, equal opportunity, and predictive parity with the White race being the privileged value. The model was also reviewed by multidisciplinary experts for clinical appropriateness. Lastly, we debiased the model by comparing 5 different debiasing approaches of fairness through blindness and reweighing. RESULTS: We determined the classification threshold through a performance evaluation that prioritized sensitivity and decision curve analysis. The baseline PPD model exhibited some unfairness in the AMC data but had a fair performance in the clinical research network data. We revised the model by fairness through blindness, a debiasing approach that yielded the best overall performance and fairness, while considering clinical appropriateness suggested by the expert reviewers. DISCUSSION AND CONCLUSION: The findings emphasize the need for a thorough evaluation of intervention-specific models, considering predictive performance, fairness, and appropriateness before clinical implementation.

Visualizing machine learning-based predictions of postpartum depression risk for lay audiences.

OBJECTIVES: To determine if different formats for conveying machine learning (ML)-derived postpartum depression risks impact patient classification of recommended actions (primary outcome) and intention to seek care, perceived risk, trust, and preferences (secondary outcomes). MATERIALS AND METHODS: We recruited English-speaking females of childbearing age (18-45 years) using an online survey platform. We created 2 exposure variables (presentation format and risk severity), each with 4 levels, manipulated within-subject. Presentation formats consisted of text only, numeric only, gradient number line, and segmented number line. For each format viewed, participants answered questions regarding each outcome. RESULTS: Five hundred four participants (mean age 31 years) completed the survey. For the risk classification question, performance was high (93%) with no significant differences between presentation formats. There were main effects of risk level (all P < .001) such that participants perceived higher risk, were more likely to agree to treatment, and more trusting in their obstetrics team as the risk level increased, but we found inconsistencies in which presentation format corresponded to the highest perceived risk, trust, or behavioral intention. The gradient number line was the most preferred format (43%). DISCUSSION AND CONCLUSION: All formats resulted high accuracy related to the classification outcome (primary), but there were nuanced differences in risk perceptions, behavioral intentions, and trust. Investigators should choose health data visualizations based on the primary goal they want lay audiences to accomplish with the ML risk score.

Perceptions of patient-reported outcome data access and sharing among patients with heart failure: ethical implications for research.

AIMS: In the face of growing expectations for data transparency and patient engagement in care, we evaluated preferences for patient-reported outcome (PRO) data access and sharing among patients with heart failure (HF) using an ethical framework. METHODS AND RESULTS: We conducted qualitative interviews with a purposive sample of patients with HF who participated in a larger 8-week study that involved the collection and return of PROs using a web-based interface. Guided by an ethical framework, patients were asked questions about their preferences for having PRO data returned to them and shared with other groups. Interview transcripts were coded by three study team members using directed content analysis. A total of 22 participants participated in semi-structured interviews. Participants were mostly male (73%), White (68%) with a mean age of 72. Themes were grouped into priorities, benefits, and barriers to data access and sharing. Priorities included ensuring anonymity when data are shared, transparency with intentions of data use, and having access to all collected data. Benefits included: using data as a communication prompt to discuss health with clinicians and using data to support self-management. Barriers included: challenges with interpreting returned results, and potential loss of benefits and anonymity when sharing data. CONCLUSION: Our interviews with HF patients highlight opportunities for researchers to return and share data through an ethical lens, by ensuring privacy and transparency with intentions of data use, returning collected data in comprehensible formats, and meeting individual expectations for data sharing.

Characterizing atrial fibrillation symptom improvement following de novo catheter ablation.

AIMS: Atrial fibrillation (AF) symptom relief is a primary indication for catheter ablation, but AF symptom resolution is not well characterized. The study objective was to describe AF symptom documentation in electronic health records (EHRs) pre- and post-ablation and identify correlates of post-ablation symptoms. METHODS AND RESULTS: We conducted a retrospective cohort study using EHRs of patients with AF (n = 1293), undergoing ablation in a large, urban health system from 2010 to 2020. We extracted symptom data from clinical notes using a natural language processing algorithm (F score: 0.81). We used Cochran's Q tests with post-hoc McNemar's tests to determine differences in symptom prevalence pre- and post-ablation. We used logistic regression models to estimate the adjusted odds of symptom resolution by personal or clinical characteristics at 6 and 12 months post-ablation. In fully adjusted models, at 12 months post-ablation patients, patients with heart failure had significantly lower odds of dyspnoea resolution [odds ratio (OR) 0.38, 95% confidence interval (CI) 0.25-0.57], oedema resolution (OR 0.37, 95% CI 0.25-0.56), and fatigue resolution (OR 0.54, 95% CI 0.34-0.85), but higher odds of palpitations resolution (OR 1.90, 95% CI 1.25-2.89) compared with those without heart failure. Age 65 and older, female sex, Black or African American race, smoking history, and antiarrhythmic use were also associated with lower odds of resolution of specific symptoms at 6 and 12 months. CONCLUSION: The post-ablation symptom patterns are heterogeneous. Findings warrant confirmation with larger, more representative data sets, which may be informative for patients whose primary goal for undergoing an ablation is symptom relief.

Characterising symptom clusters in patients with atrial fibrillation undergoing catheter ablation.

OBJECTIVE: This study aims to leverage natural language processing (NLP) and machine learning clustering analyses to (1) identify co-occurring symptoms of patients undergoing catheter ablation for atrial fibrillation (AF) and (2) describe clinical and sociodemographic correlates of symptom clusters. METHODS: We conducted a cross-sectional retrospective analysis using electronic health records data. Adults who underwent AF ablation between 2010 and 2020 were included. Demographic, comorbidity and medication information was extracted using structured queries. Ten AF symptoms were extracted from unstructured clinical notes (n=13 416) using a validated NLP pipeline (F-score=0.81). We used the unsupervised machine learning approach known as Ward's hierarchical agglomerative clustering to characterise and identify subgroups of patients representing different clusters. Fisher's exact tests were used to investigate subgroup differences based on age, gender, race and heart failure (HF) status. RESULTS: A total of 1293 patients were included in our analysis (mean age 65.5 years, 35.2% female, 58% white). The most frequently documented symptoms were dyspnoea (64%), oedema (62%) and palpitations (57%). We identified six symptom clusters: generally symptomatic, dyspnoea and oedema, chest pain, anxiety, fatigue and palpitations, and asymptomatic (reference). The asymptomatic cluster had a significantly higher prevalence of male, white and comorbid HF patients. CONCLUSIONS: We applied NLP and machine learning to a large dataset to identify symptom clusters, which may signify latent biological underpinnings of symptom experiences and generate implications for clinical care. AF patients' symptom experiences vary widely. Given prior work showing that AF symptoms predict adverse outcomes, future work should investigate associations between symptom clusters and postablation outcomes.

Health literacy, numeracy, graph literacy, and digital literacy: An overview of definitions, evaluation methods, and best practices.

Health literacy is an important skill for people receiving care. Those with limited literacy face disparities in their care and health outcomes when strategies for addressing literacy are not used when delivering health information. In this article, we introduce the importance of considering health literacy, defining it and related concepts including numeracy, graph literacy, and digital literacy, and discuss open questions about measuring health literacy in clinical care. Finally, we present best practices, including assuming "universal precautions," carefully considering wording, leveraging visualizations, recognizing cultural differences in interpretation, guidance on pilot testing, and considering digital literacy when developing electronic materials.

Women's perspectives on the use of artificial intelligence (AI)-based technologies in mental healthcare.

This study aimed to evaluate women's attitudes towards artificial intelligence (AI)-based technologies used in mental health care. We conducted a cross-sectional, online survey of U.S. adults reporting female sex at birth focused on bioethical considerations for AI-based technologies in mental healthcare, stratifying by previous pregnancy. Survey respondents (n = 258) were open to AI-based technologies in mental healthcare but concerned about medical harm and inappropriate data sharing. They held clinicians, developers, healthcare systems, and the government responsible for harm. Most reported it was "very important" for them to understand AI output. More previously pregnant respondents reported being told AI played a small role in mental healthcare was "very important" versus those not previously pregnant (P = .03). We conclude that protections against harm, transparency around data use, preservation of the patient-clinician relationship, and patient comprehension of AI predictions may facilitate trust in AI-based technologies for mental healthcare among women.

Mobile Integrated Health Interventions for Older Adults: A Systematic Review.

Background and Objectives: Mobile integrated health (MIH) interventions have not been well described in older adult populations. The objective of this systematic review was to evaluate the characteristics and effectiveness of MIH programs on health-related outcomes among older adults. Research Design and Methods: We searched Ovid MEDLINE, Ovid EMBASE, CINAHL, AgeLine, Social Work Abstracts, and The Cochrane Library through June 2021 for randomized controlled trials or cohort studies evaluating MIH among adults aged 65 and older in the general community. Studies were screened for eligibility against predefined inclusion/exclusion criteria. Using at least 2 independent reviewers, quality was appraised using the Downs and Black checklist and study characteristics and findings were synthesized and evaluated for potential bias. Results: Screening of 2,160 records identified 15 studies. The mean age of participants was 67 years. The MIH interventions varied in their focus, community paramedic training, types of assessments and interventions delivered, physician oversight, use of telemedicine, and post-visit follow-up. Studies reported significant reductions in emergency call volume (5 studies) and immediate emergency department (ED) transports (3 studies). The 3 studies examining subsequent ED visits and 4 studies examining readmission rates reported mixed results. Studies reported low adverse event rates (5 studies), high patient and provider satisfaction (5 studies), and costs equivalent to or less than usual paramedic care (3 studies). Discussion and Implications: There is wide variability in MIH provider training, program coordination, and quality-based metrics, creating heterogeneity that make definitive conclusions challenging. Nonetheless, studies suggest MIH reduces emergency call volume and ED transport rates while improving patient experience and reducing overall health care costs.

Returning study results to research participants: Data access, format, and sharing preferences.

INTRODUCTION: Research participants have a growing expectation for transparency with their collected information; however, there is little guidance on participant preferences for receiving health information and how researchers should return this information to participants. METHODS: We conducted a cross-sectional online survey with a representative sample of 502 participants in the United States. Participants were asked about their preferences for receiving, sharing, and the formatting of health information collected for research purposes. RESULTS: Most participants wanted their health information returned (84 %) to use it for their own knowledge and to manage their own health. Email was the most preferred format for receiving health data (67 %), followed by online website (44 %), and/or paper copy (32 %). Data format preferences varied by age, education, financial resources, subjective numeracy, and health literacy. Around one third of Generation Z (25 %), Millennials (30 %), and Generation X (29 %) participants preferred to receive their health information with a mobile app. In contrast, very few Baby Boomers (12 %) and none from the Silent Generation preferred the mobile app format. Having a paper copy of the data was preferred by 38 % of participants without a college degree compared to those with a college degree. Preferences were highest for sharing all health information with doctors and nurses (77 %), and some information with friends and family (66 %). CONCLUSION: Study findings support returning research information to participants in multiple formats, including email, online websites, and paper copy. Preferences for whom to share information with varied by stakeholders and by sociodemographic characteristics. Researchers should offer multiple formats to participants and tailor data sharing options to participants' preferences. Future research should further explore combinations of individual characteristics that may further influence data sharing and format preferences.

'Replace uncertainty with information': shared decision-making and decision quality surrounding catheter ablation for atrial fibrillation.

AIMS: As a first step in developing a decision aid to support shared decision-making (SDM) for patients with atrial fibrillation (AF) to evaluate treatment options for rhythm and symptom control, we aimed to measure decision quality and describe decision-making processes among patients and clinicians involved in decision-making around catheter ablation for AF. METHODS AND RESULTS: We conducted a cross-sectional, mixed-methods study guided by an SDM model outlining decision antecedents, processes, and outcomes. Patients and clinicians completed semi-structured interviews about decision-making around ablation, feelings of decision conflict and regret, and preferences for the content, delivery, and format of a hypothetical decision aid for ablation. Patients also completed surveys about AF symptoms and aspects of decision quality. Fifteen patients (mean age 71.1 ± 8.6 years; 27% female) and five clinicians were recruited. For most patients, decisional conflict and regret were low, but they also reported low levels of information and agency in the decision-making process. Most clinicians report routinely providing patients with information and encouraging engagement during consultations. Patients reported preferences for an interactive, web-based decision aid that clearly presents evidence regarding outcomes using data, visualizations, videos, and personalized risk assessments, and is available in multiple languages. CONCLUSION: Disconnects between clinician efforts to provide information and bolster agency and patient experiences of decision-making suggest decision aids may be needed to improve decision quality in practice. Reported experiences with current decision-making practices and preferences for decision aid content, format, and delivery can support the user-centred design and development of a decision aid.

Detecting early physiologic changes through cardiac implantable electronic device data among patients with COVID-19.

Background: Cardiac implantable electronic devices (CIEDs) may enable early identification of COVID-19 to facilitate timelier intervention. Objective: To characterize early physiologic changes associated with the onset of acute COVID-19 infection, as well as during and after acute infection, among patients with CIEDs. Methods: CIED sensor data from March 2020 to February 2021 from 286 patients with a CIED were linked to clinical data from electronic health records. Three cohorts were created: known COVID-positive (n = 20), known COVID-negative (n = 166), and a COVID-untested control group (n = 100) included to account for testing bias. Associations between changes in CIED sensors from baseline (including HeartLogic index, a composite index predicting worsening heart failure) and COVID-19 status were evaluated using logistic regression models, Wilcoxon signed rank tests, and Mann-Whitney U tests. Results: Significant differences existed between the cohorts by race, ethnicity, CIED device type, and medical admissions. Several sensors changed earlier for COVID-positive vs COVID-negative patients: HeartLogic index (mean 16.4 vs 9.2 days [P = .08]), respiratory rate (mean 8.5 vs 3.9 days [P = .01], and activity (mean 8.2 vs 3.5 days [P = .008]). Respiratory rate during the 7 days before testing significantly predicted a positive vs negative COVID-19 test, adjusting for age, sex, race, and device type (odds ratio 2.31 [95% confidence interval 1.33-5.13]). Conclusion: Physiologic data from CIEDs could signal early signs of infection that precede clinical symptoms, which may be used to support early detection of infection to prevent decompensation in this at-risk population.

Faces Scales for Anxiety and Anger: A National Study of Measurement Properties.

Importance: Faces scales are used worldwide to assess pain, but robust faces scales for anxiety and anger do not exist. These scales are urgently needed, because an estimated two-thirds of patients have difficulty reading written questionnaires.Objective: To develop and evaluate measurement properties of faces scales to monitor two mental health symptoms in US adults (anxiety and anger) in accordance with the COnsensus-based Standards for health Measurement INstruments (COSMIN).Methods: The development process included population identification, scale generation, and pretesting. The evaluation process included assessment of content validity, construct validity, criterion validity, test-retest reliability, and measurement error using 5 order-randomized, positively controlled online survey studies conducted between April and June 2020. We recruited national purposive samples of US adults representative on age, gender, and race. For each faces scale, participants assessed relevance, comprehensibility, and comprehensiveness (study 1, n = 300), strength-of-association (study 2, n = 300), convergent validity against the visual analog scale (VAS; study 3, n = 305), convergent validity against the Patient-Reported Outcomes Measurement Information System (PROMIS) questionnaires (study 4, n = 1,000), and test-retest reliability and measurement error (study 5, n = 853).Results: The anxiety and anger faces scales showed high relevance (95%-96%), comprehensibility (93%-97%), comprehensiveness (94%-97%), and strength-of-association (74%-96%). We found very high agreement with the VAS (ρ = 0.94-0.95) and high agreement with PROMIS questionnaires (ρ = 0.74-0.79). Scales showed adequate test-retest reliability (intraclass correlation = 0.70-0.78) and measurement error (standard error of measurement = 1.14-1.22).Conclusions: Faces scales to monitor anxiety and anger show adequate measurement properties, including content validity, construct validity, criterion validity, test-retest reliability, and measurement error. The recommended use is non-diagnostic monitoring of anxiety and anger, particularly when mental health is an ancillary but important outcome of treatment.

Building trust in research through information and intent transparency with health information: representative cross-sectional survey of 502 US adults.

OBJECTIVE: Participation in healthcare research shapes health policy and practice; however, low trust is a barrier to participation. We evaluated whether returning health information (information transparency) and disclosing intent of data use (intent transparency) impacts trust in research. MATERIALS AND METHODS: We conducted an online survey with a representative sample of 502 US adults. We assessed baseline trust and change in trust using 6 use cases representing the Social-Ecological Model. We assessed descriptive statistics and associations between trust and sociodemographic variables using logistic and multinomial regression. RESULTS: Most participants (84%) want their health research information returned. Black/African American participants were more likely to increase trust in research with individual information transparency (odds ratio (OR) 2.06 [95% confidence interval (CI): 1.06-4.34]) and with intent transparency when sharing with chosen friends and family (3.66 [1.98-6.77]), doctors and nurses (1.96 [1.10-3.65]), or health tech companies (1.87 [1.02-3.40]). Asian, Native American or Alaska Native, Native Hawaiian or Pacific Islander, Multirace, and individuals with a race not listed, were more likely to increase trust when sharing with health policy makers (1.88 [1.09-3.30]). Women were less likely to increase trust when sharing with friends and family (0.55 [0.35-0.87]) or health tech companies (0.46 [0.31-0.70]). DISCUSSION: Participants wanted their health information returned and would increase their trust in research with transparency when sharing health information. CONCLUSION: Trust in research is influenced by interrelated factors. Future research should recruit diverse samples with lower baseline trust levels to explore changes in trust, with variation on the type of information shared.

Monitoring Symptoms of COVID-19: Review of Mobile Apps.

BACKGROUND: Mobile health (mHealth) apps have facilitated symptom monitoring of COVID-19 symptoms globally and have been used to share data with health care professionals and support disease prediction, prevention, management, diagnostics, and improvements in treatments and patient education. OBJECTIVE: The aim of this review is to evaluate the quality and functionality of COVID-19 mHealth apps that support tracking acute and long-term symptoms of COVID-19. METHODS: We systematically reviewed commercially available mHealth apps for COVID-19 symptom monitoring by searching Google Play and Apple iTunes using search terms such as "COVID-19," "Coronavirus," and "COVID-19 and symptoms." All apps underwent three rounds of screening. The final apps were independently assessed using the Mobile Application Rating Scale (MARS), an informatics functionality scoring system, and the Center for Disease Control and World Health Organization symptom guidelines. The MARS is a 19-item standardized tool to evaluate the quality of mHealth apps on engagement, functionality, aesthetics, and information quality. Functionality was quantified across the following criteria: inform, instruct, record (collect, share, evaluate, and intervene), display, guide, remind or alert, and communicate. Interrater reliability between the reviewers was calculated. RESULTS: A total of 1017 mobile apps were reviewed, and 20 (2%) met the inclusion criteria. The majority of the 20 included apps (n=18, 90%) were designed to track acute COVID-19 symptoms, and only 2 (10%) addressed long-term symptoms. Overall, the apps scored high on quality, with an overall MARS rating of 3.89 out of 5, and the highest domain score for functionality (4.2). The most common functionality among all apps was the instruct function (n=19, 95%). The most common symptoms included in the apps for tracking were fever and dry cough (n=18, 90%), aches and pains (n=17, 85%), difficulty breathing (n=17, 85%), tiredness, sore throat, headache, loss of taste or smell (n=16, 80%), and diarrhea (n=15, 75%). Only 2 (10%) apps specifically tracked long-term symptoms of COVID-19. The top 4 rated apps overall were state-specific apps developed and deployed for public use. CONCLUSIONS: Overall, mHealth apps designed to monitor symptoms of COVID-19 were of high quality, but the majority of apps focused almost exclusively on acute symptoms. Future apps should also incorporate monitoring long-term symptoms of COVID-19 and evidence-based educational materials; they should also include a feature that would allow patients to communicate their symptoms to specific caregivers or their own health care team. App developers should also follow updated technical and clinical guidelines from the Center for Disease Control and the World Health Organization.

Using Mobile Integrated Health and telehealth to support transitions of care among patients with heart failure (MIGHTy-Heart): protocol for a pragmatic randomised controlled trial.

INTRODUCTION: Nearly one-quarter of patients discharged from the hospital with heart failure (HF) are readmitted within 30 days, placing a significant burden on patients, families and health systems. The objective of the 'Using Mobile Integrated Health and Telehealth to support transitions of care among patients with Heart failure' (MIGHTy-Heart) study is to compare the effectiveness of two postdischarge interventions on healthcare utilisation, patient-reported outcomes and healthcare quality among patients with HF. METHODS AND ANALYSIS: The MIGHTy-Heart study is a pragmatic comparative effectiveness trial comparing two interventions demonstrated to improve the hospital to home transition for patients with HF: mobile integrated health (MIH) and transitions of care coordinators (TOCC). The MIH intervention bundles home visits from a community paramedic (CP) with telehealth video visits by emergency medicine physicians to support the management of acute symptoms and postdischarge care coordination. The TOCC intervention consists of follow-up phone calls from a registered nurse within 48-72 hours of discharge to assess a patient's clinical status, identify unmet clinical and social needs and reinforce patient education (eg, medication adherence and lifestyle changes). MIGHTy-Heart is enrolling and randomising (1:1) 2100 patients with HF who are discharged to home following a hospitalisation in two New York City (NY, USA) academic health systems. The coprimary study outcomes are all-cause 30-day hospital readmissions and quality of life measured with the Kansas City Cardiomyopathy Questionnaire 30 days after hospital discharge. The secondary endpoints are days at home, preventable emergency department visits, unplanned hospital admissions and patient-reported symptoms. Data sources for the study outcomes include patient surveys, electronic health records and claims submitted to Medicare and Medicaid. ETHICS AND DISSEMINATION: All participants provide written or verbal informed consent prior to randomisation in English, Spanish, French, Mandarin or Russian. Study findings are being disseminated to scientific audiences through peer-reviewed publications and presentations at national and international conferences. This study has been approved by: Biomedical Research Alliance of New York (BRANY #20-08-329-380), Weill Cornell Medicine Institutional Review Board (20-08022605) and Mt. Sinai Institutional Review Board (20-01901). TRIAL REGISTRATION NUMBER: Clinicaltrials.gov, NCT04662541.

Atrial Fibrillation and Stroke Symptoms in the REGARDS Study.

Background It is unknown if stroke symptoms in the absence of a stroke diagnosis are a sign of subtle cardioembolic phenomena. The objective of this study was to examine associations between atrial fibrillation (AF) and stroke symptoms among adults with no clinical history of stroke or transient ischemic attack (TIA). Methods and Results We evaluated associations between AF and self-reported stroke symptoms in the national, prospective REGARDS (Reasons for Geographic and Racial Differences in Stroke) cohort. We conducted cross-sectional (n=27 135) and longitudinal (n=21 932) analyses over 8 years of follow-up of REGARDS participants without stroke/transient ischemic attack and stratified by anticoagulant or antiplatelet agent use. The mean age was 64.4 (SD±9.4) years, 55.3% were women, and 40.8% were Black participants; 28.6% of participants with AF reported stroke symptoms. In the cross-sectional analysis, comparing participants with and without AF, the risk of stroke symptoms was elevated for adults with AF taking neither anticoagulants nor antiplatelet agents (odds ratio [OR], 2.22; 95% CI, 1.89-2.59) or antiplatelet agents only (OR, 1.92; 95% CI, 1.61-2.29) but not for adults with AF taking anticoagulants (OR, 1.08; 95% CI, 0.71-1.65). In the longitudinal analysis, the risk of stroke symptoms was also elevated for adults with AF taking neither anticoagulants nor antiplatelet agents (hazard ratio [HR], 1.41; 95% CI, 1.21-1.66) or antiplatelet agents only (HR, 1.23; 95% CI, 1.04-1.46) but not for adults with AF taking anticoagulants (HR, 0.86; 95% CI, 0.62-1.18). Conclusions Stroke symptoms in the absence of a stroke diagnosis may represent subclinical cardioembolic phenomena or "whispering strokes." Future studies examining the benefit of stroke symptom screening may be warranted.

Comparative Effectiveness of Behavioral Interventions for Cardiovascular Risk Reduction in Latinos: A Systematic Review.

BACKGROUND: Latinos, the fastest growing ethnic minority group in the United States, are at a high risk for cardiovascular disease (CVD). However, little is known about effective strategies to reduce CVD risk in this population. OBJECTIVE: The aim of this study was to systematically review and synthesize evidence from randomized controlled trials that examined the effectiveness of behavioral interventions to reduce CVD risk in Latinos living in the United States. METHODS: Four electronic databases were searched for relevant peer-reviewed English- and Spanish-language articles published between January 1, 2000, and December 31, 2019. Four reviewers independently completed article screening, data abstraction, and quality appraisal. At least 2 reviewers completed data abstraction and quality appraisal for each article, and a third reviewer was assigned to settle disagreements. Data on study characteristics and outcomes were abstracted. RESULTS: We retrieved 1939 articles. After applying inclusion/exclusion criteria, 17 articles were included. Most interventions were led by community health workers (n = 10); 2 family-based interventions were identified. None of the included studies was nurse led. Behavioral factors were assessed across all included studies, whereas only 4 studies reported on psychosocial outcomes. Improvements were observed in dietary habits and psychosocial outcomes. Findings for physical activity and biological outcomes were mixed. We identified no differences in outcomes based on intervention modalities used or the role of those who led the interventions. CONCLUSION: Existing evidence is mixed. Future research should assess the effectiveness of understudied treatment modalities (including nurse-led, mobile health, and family-based interventions) in reducing CVD risk in Latinos.

Systematic review of current natural language processing methods and applications in cardiology.

Natural language processing (NLP) is a set of automated methods to organise and evaluate the information contained in unstructured clinical notes, which are a rich source of real-world data from clinical care that may be used to improve outcomes and understanding of disease in cardiology. The purpose of this systematic review is to provide an understanding of NLP, review how it has been used to date within cardiology and illustrate the opportunities that this approach provides for both research and clinical care. We systematically searched six scholarly databases (ACM Digital Library, Arxiv, Embase, IEEE Explore, PubMed and Scopus) for studies published in 2015-2020 describing the development or application of NLP methods for clinical text focused on cardiac disease. Studies not published in English, lacking a description of NLP methods, non-cardiac focused and duplicates were excluded. Two independent reviewers extracted general study information, clinical details and NLP details and appraised quality using a checklist of quality indicators for NLP studies. We identified 37 studies developing and applying NLP in heart failure, imaging, coronary artery disease, electrophysiology, general cardiology and valvular heart disease. Most studies used NLP to identify patients with a specific diagnosis and extract disease severity using rule-based NLP methods. Some used NLP algorithms to predict clinical outcomes. A major limitation is the inability to aggregate findings across studies due to vastly different NLP methods, evaluation and reporting. This review reveals numerous opportunities for future NLP work in cardiology with more diverse patient samples, cardiac diseases, datasets, methods and applications.

Cardiac symptom burden and arrhythmia recurrence drives digital health use: results from the iHEART randomized controlled trial.

AIMS: Digital health can transform the management of atrial fibrillation (AF) and enable patients to take a central role in detecting symptoms and self-managing AF. There is a gap in understanding factors that support sustained use of digital health tools for patients with AF. This study identified predictors of Alivecor® KardiaMobile ECG monitor usage among patients with AF enrolled in the iPhone®Helping Evaluate Atrial fibrillation Rhythm through Technology (iHEART) randomized controlled trial. METHODS AND RESULTS: We analysed data from 105 English and Spanish-speaking adults with AF enrolled in the intervention arm of the iHEART trial. The iHEART intervention included smartphone-based electrocardiogram self-monitoring with Alivecor® KardiaMobile and triweekly text messages for 6 months. The primary outcome was use of Alivecor® categorized as: infrequent (≤5 times/week), moderate (>5 times and ≤11 times/week), and frequent (>11 times/week). We applied multinomial logistic regression modelling to characterize frequency and predictors of use. Of the 105 participants, 25% were female, 75% were White, and 45% were ≥65 years of age. Premature atrial contractions (PACs) [adjusted odds ratio (OR): 1.23, 1.08-1.40, P = 0.002] predicted frequent as compared to infrequent use. PACs (adjusted OR: 1.17, 95% confidence interval 1.06-1.30, P = 0.003), lower symptom burden (adjusted OR: 1.06, 1.01-1.11, P = 0.02), and less treatment concern (adjusted OR: 0.96, 0.93-0.99, P = 0.02) predicted moderate as compared to infrequent use. CONCLUSIONS: Frequent use of AliveCor® is associated with AF symptoms and potentially symptomatic cardiac events. Symptom burden and frequency should be measured and incorporated into analyses of future digital health trials for AF management.