@Article{info:doi/10.2196/63609,
author="Silva, Malpriya S. Sandun
and Wabe, Nasir
and Nguyen, D. Amy
and Seaman, Karla
and Huang, Guogui
and Dodds, Laura
and Meulenbroeks, Isabelle
and Mercado, Ibarra Crisostomo
and Westbrook, I. Johanna",
title="Development of a Predictive Dashboard With Prescriptive Decision Support for Falls Prevention in Residential Aged Care: User-Centered Design Approach",
journal="JMIR Aging",
year="2025",
month="Apr",
day="7",
volume="8",
pages="e63609",
keywords="falls prevention",
keywords="dashboard architecture",
keywords="predictive",
keywords="sustainability",
keywords="challenges",
keywords="decision support",
keywords="falls",
keywords="aged care",
keywords="geriatric",
keywords="older adults",
keywords="economic burden",
keywords="prevention",
keywords="electronic health record",
keywords="EHR",
keywords="intervention",
keywords="decision-making",
keywords="patient safety",
keywords="risks",
keywords="older people",
keywords="monitoring",
abstract="Background: Falls are a prevalent and serious health condition among older people in residential aged care facilities, causing significant health and economic burdens. However, the likelihood of future falls can be predicted, and thus, falls can be prevented if appropriate prevention programs are implemented. Current fall prevention programs in residential aged care facilities rely on risk screening tools with suboptimal predictive performance, leading to significant concerns regarding resident safety. Objective: This study aimed to develop a predictive, dynamic dashboard to identify residents at risk of falls with associated decision support. This paper provides an overview of the technical process, including the challenges faced and the strategies used to overcome them during the development of the dashboard. Methods: A predictive dashboard was co-designed with a major residential aged care partner in New South Wales, Australia. Data from resident profiles, daily medications, fall incidents, and fall risk assessments were used. A dynamic fall risk prediction model and personalized rule-based fall prevention recommendations were embedded in the dashboard. The data ingestion process into the dashboard was designed to mitigate the impact of underlying data system changes. This approach aims to ensure resilience against alterations in the data systems. Results: The dashboard was developed using Microsoft Power BI and advanced R programming by linking data silos. It includes dashboard views for those managing facilities and for those caring for residents. Data drill-through functionality was used to navigate through different dashboard views. Resident-level change in daily risk of falling and risk factors and timely evidence-based recommendations were output to prevent falls and enhance prescriptive decision support. Conclusions: This study emphasizes the significance of a sustainable dashboard architecture and how to overcome the challenges faced when developing a dashboard amid underlying data system changes. The development process used an iterative dashboard co-design process, ensuring the successful implementation of knowledge into practice. Future research will focus on the implementation and evaluation of the dashboard's impact on health processes and economic outcomes. International Registered Report Identifier (IRRID): RR2-https://doi.org/10.1136/bmjopen-2021-048657 ",
doi="10.2196/63609",
url="https://aging.jmir.org/2025/1/e63609"
}


@Article{info:doi/10.2196/60471,
author="Chalmer, Rosansky Rachel Beth
and Ayers, Emmeline
and Weiss, F. Erica
and Fowler, R. Nicole
and Telzak, Andrew
and Summanwar, Diana
and Zwerling, Jessica
and Wang, Cuiling
and Xu, Huiping
and Holden, J. Richard
and Fiori, Kevin
and French, D. Dustin
and Nsubayi, Celeste
and Ansari, Asif
and Dexter, Paul
and Higbie, Anna
and Yadav, Pratibha
and Walker, M. James
and Congivaram, Harrshavasan
and Adhikari, Dristi
and Melecio-Vazquez, Mairim
and Boustani, Malaz
and Verghese, Joe",
title="Improving Early Dementia Detection Among Diverse Older Adults With Cognitive Concerns With the 5-Cog Paradigm: Protocol for a Hybrid Effectiveness-Implementation Clinical Trial",
journal="JMIR Res Protoc",
year="2025",
month="Apr",
day="3",
volume="14",
pages="e60471",
keywords="cognitive assessment",
keywords="cognitive screening",
keywords="cognitive impairment",
keywords="mild cognitive impairment",
keywords="dementia",
keywords="dissemination and implementation science",
keywords="clinical trial protocol",
keywords="randomized controlled trial",
keywords="hybrid implementation-effectiveness trial",
abstract="Background: The 5-Cog paradigm is a 5-minute brief cognitive assessment coupled with a clinical decision support tool designed to improve clinicians' early detection of cognitive impairment, including dementia, in their diverse older primary care patients. The 5-Cog battery uses picture- and symbol-based assessments and a questionnaire. It is low cost, simple, minimizes literacy bias, and is culturally fair. The decision support component of the paradigm helps nudge appropriate care provider response to an abnormal 5-Cog battery. Objective: The objective of our study is to evaluate the effectiveness, implementation, and cost of the 5-Cog paradigm. Methods: We will enroll 6600 older patients with cognitive concerns from 22 primary care clinics in the Bronx, New York, and in multiple locations in Indiana for this hybrid type 1 effectiveness-implementation trial. We will analyze the effectiveness of the 5-Cog paradigm to increase the rate of new diagnoses of mild cognitive impairment syndrome or dementia using a pragmatic, cluster randomized clinical trial design. The secondary outcome is the ordering of new tests, treatments, and referrals for cognitive indications within 90 days after the study visit. The 5-Cog's decision support component will be deployed as an electronic medical record feature. We will analyze the 5-Cog's implementation process, context, and outcomes through the Consolidated Framework for Implementation Research using a mixed methods design (surveys and interviews). The study will also examine cost-effectiveness from societal and payer (Medicare) perspectives by estimating the cost per additional dementia diagnosis. Results: The study is funded by the National Institute of Neurological Disorders and Stroke of the National Institutes of Health (2U01NS105565). The protocol was approved by the Albert Einstein College of Medicine Institutional Review Board in September 2022. A validation study was completed to select cut scores for the 5-Cog battery. Among the 76 patients enrolled, the resulting clinical diagnoses were as follows: dementia in 32 (42\%); mild cognitive impairment in 28 (37\%); subjective cognitive concerns without objective cognitive impairment in 12 (16\%); no cognitive diagnosis assigned in 2 (3\%). The mean scores were Picture-Based Memory Impairment Screen 5.8 (SD 2.7), Symbol Match 27.2 (SD 18.2), and Subjective Motoric Cognitive Risk 2.4 (SD 1.7). The cut scores for an abnormal or positive result on the 5-Cog components were as follows: Picture-Based Memory Impairment Screen ?6 (range 0-8), Symbol Match ?25 (range 0-65), and Subjective Motoric Cognitive Risk >5 (range 0-7). As of December 2024, a total of 12 clinics had completed the onboarding processes, and 2369 patients had been enrolled. Conclusions: The findings of this study will facilitate the rapid adaptation and dissemination of this effective and practical clinical tool across diverse primary care clinical settings. Trial Registration: ClinicalTrials.gov NCT05515224; https://www.clinicaltrials.gov/study/NCT05515224 International Registered Report Identifier (IRRID): DERR1-10.2196/60471 ",
doi="10.2196/60471",
url="https://www.researchprotocols.org/2025/1/e60471"
}


@Article{info:doi/10.2196/65178,
author="West, Matthew
and Cheng, You
and He, Yingnan
and Leng, Yu
and Magdamo, Colin
and Hyman, T. Bradley
and Dickson, R. John
and Serrano-Pozo, Alberto
and Blacker, Deborah
and Das, Sudeshna",
title="Unsupervised Deep Learning of Electronic Health Records to Characterize Heterogeneity Across Alzheimer Disease and Related Dementias: Cross-Sectional Study",
journal="JMIR Aging",
year="2025",
month="Mar",
day="31",
volume="8",
pages="e65178",
keywords="Alzheimer disease and related dementias",
keywords="electronic health records",
keywords="large language models",
keywords="clustering",
keywords="unsupervised learning",
abstract="Background: Alzheimer disease and related dementias (ADRD) exhibit prominent heterogeneity. Identifying clinically meaningful ADRD subtypes is essential for tailoring treatments to specific patient phenotypes. Objective: We aimed to use unsupervised learning techniques on electronic health records (EHRs) from memory clinic patients to identify ADRD subtypes. Methods: We used pretrained embeddings of non-ADRD diagnosis codes (International Classification of Diseases, Ninth Revision) and large language model (LLM)--derived embeddings of clinical notes from patient EHRs. Hierarchical clustering of these embeddings was used to identify ADRD subtypes. Clusters were characterized regarding their demographic and clinical features. Results: We analyzed a cohort of 3454 patients with ADRD from a memory clinic at Massachusetts General Hospital, each with a specialist diagnosis. Clustering pretrained embeddings of the non-ADRD diagnosis codes in patient EHRs revealed the following 3 patient subtypes: one with skin conditions, another with psychiatric disorders and an earlier age of onset, and a third with diabetes complications. Similarly, using LLM-derived embeddings of clinical notes, we identified 3 subtypes of patients as follows: one with psychiatric manifestations and higher prevalence of female participants (prevalence ratio: 1.59), another with cardiovascular and motor problems and higher prevalence of male participants (prevalence ratio: 1.75), and a third one with geriatric health disorders. Notably, we observed significant overlap between clusters from both data modalities ($\chi$24=89.4; P<.001). Conclusions: By integrating International Classification of Diseases, Ninth Revision codes and LLM-derived embeddings, our analysis delineated 2 distinct ADRD subtypes with sex-specific comorbid and clinical presentations, offering insights for potential precision medicine approaches. ",
doi="10.2196/65178",
url="https://aging.jmir.org/2025/1/e65178"
}


@Article{info:doi/10.2196/66104,
author="Bai, Anying
and He, Shan
and Jiang, Yu
and Xu, Weihao
and Lin, Zhanyi",
title="Comparison of 3 Aging Metrics in Dual Declines to Capture All-Cause Dementia and Mortality Risk: Cohort Study",
journal="JMIR Aging",
year="2025",
month="Jan",
day="30",
volume="8",
pages="e66104",
keywords="gerontology",
keywords="geriatrics",
keywords="older adults",
keywords="older people",
keywords="aging",
keywords="motoric cognitive risk syndrome",
keywords="MCR",
keywords="physio-cognitive decline syndrome",
keywords="PCDS",
keywords="cognitive frailty",
keywords="CF",
keywords="frailty",
keywords="discrimination",
keywords="risk factors",
keywords="prediction",
keywords="dementia risk",
keywords="mortality risk",
abstract="Background: The utility of aging metrics that incorporate cognitive and physical function is not fully understood. Objective: We aim to compare the predictive capacities of 3 distinct aging metrics---motoric cognitive risk syndrome (MCR), physio-cognitive decline syndrome (PCDS), and cognitive frailty (CF)---for incident dementia and all-cause mortality among community-dwelling older adults. Methods: We used longitudinal data from waves 10-15 of the Health and Retirement Study. Cox proportional hazards regression analysis was employed to evaluate the effects of MCR, PCDS, and CF on incident all-cause dementia and mortality, controlling for socioeconomic and lifestyle factors, as well as medical comorbidities. Discrimination analysis was conducted to assess and compare the predictive accuracy of the 3 aging metrics. Results: A total of 2367 older individuals aged 65 years and older, with no baseline prevalence of dementia or disability, were ultimately included. The prevalence rates of MCR, PCDS, and CF were 5.4\%, 6.3\%, and 1.3\%, respectively. Over a decade-long follow-up period, 341 cases of dementia and 573 deaths were recorded. All 3 metrics were predictive of incident all-cause dementia and mortality when adjusting for multiple confounders, with variations in the strength of their associations (incident dementia: MCR odds ratio [OR] 1.90, 95\% CI 1.30?2.78; CF 5.06, 95\% CI 2.87?8.92; PCDS 3.35, 95\% CI 2.44?4.58; mortality: MCR 1.60, 95\% CI 1.17?2.19; CF 3.26, 95\% CI 1.99?5.33; and PCDS 1.58, 95\% CI 1.17?2.13). The C-index indicated that PCDS and MCR had the highest discriminatory accuracy for all-cause dementia and mortality, respectively. Conclusions: Despite the inherent differences among the aging metrics that integrate cognitive and physical functions, they consistently identified risks of dementia and mortality. This underscores the importance of implementing targeted preventive strategies and intervention programs based on these metrics to enhance the overall quality of life and reduce premature deaths in aging populations. ",
doi="10.2196/66104",
url="https://aging.jmir.org/2025/1/e66104"
}


@Article{info:doi/10.2196/53205,
author="Zhang, Jinbao
and Prunty, E. Jonathan
and Charles, C. Alison
and Forder, Julien",
title="Association Between Digital Front Doors and Social Care Use for Community-Dwelling Adults in England: Cross-Sectional Study",
journal="J Med Internet Res",
year="2025",
month="Jan",
day="2",
volume="27",
pages="e53205",
keywords="social care support",
keywords="long term care",
keywords="access",
keywords="front door",
keywords="easy-read",
keywords="self-assessment",
keywords="system navigation",
keywords="digital system",
keywords="digital technology",
keywords="internet",
abstract="Background: Requests for public social care support can be made through an online portal. These digital ``front doors'' can help people navigate complex social care systems and access services. These systems can be set up in different ways, but there is little evidence about the impact of alternative arrangements. Digital front-door systems should help people better access services, particularly low-intensity services (high-intensity care is likely to require a full in-person assessment). Objective: This study aimed to investigate the association between 2 primary digital front door arrangements, easy-read information, and self-assessment tools provided on official websites, and the type of social care support that is offered: ongoing low-level support (OLLS), short-term care (STC) and long-term care (LTC). Methods: Information on front door arrangements was collected from the official websites of 152 English local authorities in 2021. We conducted a cross-sectional analysis using aggregated service use data from official government returns at the local authority level. The independent variables were derived from the policy information collected, specifically focusing on the availability of online digital easy-read information and self-assessment tools for adults and caregivers through official websites. The dependent variables were the rates of using social care support, including OLLS, STC, and LTC, across different age groups: the adult population (aged 18 and older), younger population (aged between 18 and 64 years), and older population (aged 65 and older). Multivariate regression analysis was used to examine the association between digital front door arrangements and access to social care support, controlling for population size, dependency level, and financial need factors. Results: Less than 20\% (27/147) of local authorities provided an integrated digital easy-read format as part of their digital front door system with about 25\% (37/147) adopting digital self-assessment within their system. We found that local authorities that offered an integrated digital easy-read information format showed higher rates of using OLLS ($\beta$ coefficient=0.54; P=.03; but no statistically significant association with LTC and STC). The provision of an online self-assessment system was not associated with service use in the 1-year (2021) cross-sectional estimate, but when 2 years (2020 and 2021) of service-use data were analyzed, a significant positive association was found on OLLS rates ($\beta$ coefficient=0.41; P=.21). Notably, these findings were consistent across different age groups. Conclusions: These findings are consistent with our hypothesis that digital systems with built-in easy-read and self-assessment may make access to (low-intensity) services easier for people. Adoption of these arrangements could potentially help increase the uptake of support among those who are eligible, with expected benefits for their care-related well-being. Given the limited adoption of the digital front door by local authorities in England, expanding their use could improve care-related outcomes and save social care costs. ",
doi="10.2196/53205",
url="https://www.jmir.org/2025/1/e53205"
}


@Article{info:doi/10.2196/57899,
author="Uihlein, Adriane
and Beissel, Lisa
and Ajlani, Hanane Anna
and Orzechowski, Marcin
and Leinert, Christoph
and Kocar, Derya Thomas
and Pankratz, Carlos
and Schuetze, Konrad
and Gebhard, Florian
and Steger, Florian
and Fotteler, Liselotte Marina
and Denkinger, Michael",
title="Expectations and Requirements of Surgical Staff for an AI-Supported Clinical Decision Support System for Older Patients: Qualitative Study",
journal="JMIR Aging",
year="2024",
month="Dec",
day="17",
volume="7",
pages="e57899",
keywords="traumatology",
keywords="orthogeriatrics",
keywords="older adult",
keywords="elderly",
keywords="older people",
keywords="aging",
keywords="interviews",
keywords="mHealth",
keywords="mobile health",
keywords="mobile application",
keywords="digital health",
keywords="digital technology",
keywords="digital intervention",
keywords="CDSS",
keywords="clinical decision support system",
keywords="artificial intelligence",
keywords="AI",
keywords="algorithm",
keywords="predictive model",
keywords="predictive analytics",
keywords="predictive system",
keywords="practical model",
keywords="decision support",
keywords="decision support tool",
abstract="Background: Geriatric comanagement has been shown to improve outcomes of older surgical inpatients. Furthermore, the choice of discharge location, that is, continuity of care, can have a fundamental impact on convalescence. These challenges and demands have led to the SURGE-Ahead project that aims to develop a clinical decision support system (CDSS) for geriatric comanagement in surgical clinics including a decision support for the best continuity of care option, supported by artificial intelligence (AI) algorithms. Objective: This qualitative study aims to explore the current challenges and demands in surgical geriatric patient care. Based on these challenges, the study explores the attitude of interviewees toward the introduction of an AI-supported CDSS (AI-CDSS) in geriatric patient care in surgery, focusing on technical and general wishes about an AI-CDSS, as well as ethical considerations. Methods: In this study, 15 personal interviews with physicians, nurses, physiotherapists, and social workers, employed in surgical departments at a university hospital in Southern Germany, were conducted in April 2022. Interviews were conducted in person, transcribed, and coded by 2 researchers (AU, LB) using content and thematic analysis. During the analysis, quotes were sorted into the main categories of geriatric patient care, use of an AI-CDSS, and ethical considerations by 2 authors (AU, LB). The main themes of the interviews were subsequently described in a narrative synthesis, citing key quotes. Results: In total, 399 quotes were extracted and categorized from the interviews. Most quotes could be assigned to the primary code challenges in geriatric patient care (111 quotes), with the most frequent subcode being medical challenges (45 quotes). More quotes were assigned to the primary code chances of an AI-CDSS (37 quotes), with its most frequent subcode being holistic patient overview (16 quotes), then to the primary code limits of an AI-CDSS (26 quotes). Regarding the primary code technical wishes (37 quotes), most quotes could be assigned to the subcode intuitive usability (15 quotes), followed by mobile availability and easy access (11 quotes). Regarding the main category ethical aspects of an AI-CDSS, most quotes could be assigned to the subcode critical position toward trust in an AI-CDSS (9 quotes), followed by the subcodes respecting the patient's will and individual situation (8 quotes) and responsibility remaining in the hands of humans (7 quotes). Conclusions: Support regarding medical geriatric challenges and responsible handling of AI-based recommendations, as well as necessity for a holistic approach focused on usability, were the most important topics of health care professionals in surgery regarding development of an AI-CDSS for geriatric care. These findings, together with the wish to preserve the patient-caregiver relationship, will help set the focus for the ongoing development of AI-supported CDSS. ",
doi="10.2196/57899",
url="https://aging.jmir.org/2024/1/e57899"
}


@Article{info:doi/10.2196/59234,
author="Takura, Tomoyuki
and Yokoi, Hiroyoshi
and Honda, Asao",
title="Factors Influencing Drug Prescribing for Patients With Hospitalization History in Circulatory Disease--Patient Severity, Composite Adherence, and Physician-Patient Relationship: Retrospective Cohort Study",
journal="JMIR Aging",
year="2024",
month="Dec",
day="6",
volume="7",
pages="e59234",
keywords="medication adherence",
keywords="drug prescription switch",
keywords="generic drug",
keywords="logistic model",
keywords="long-term longitudinal study",
keywords="patient severity",
keywords="systolic blood pressure",
keywords="serum creatinine",
keywords="aging",
keywords="big data",
abstract="Background: With countries promoting generic drug prescribing, their growth may plateau, warranting further investigation into the factors influencing this trend, including physician and patient perspectives. Additional strategies may be needed to maximize the switch to generic drugs while ensuring health care system sustainability, focusing on factors beyond mere low cost. Emphasizing affordability and clarifying other prescription considerations are essential. Objective: This study aimed to provide initial insights into how patient severity, composite adherence, and physician-patient relationships impact generic switching. Methods: This study used a long-term retrospective cohort design by analyzing data from a national health care database. The population included patients of all ages, primarily older adults, who required primary-to-tertiary preventive actions with a history of hospitalization for cardiovascular diseases (ICD-10 [International Statistical Classification of Diseases, Tenth Revision]) from April 2014 to March 2018 (4 years). We focused on switching to generic drugs, with temporal variations in clinical parameters as independent variables. Lifestyle factors (smoking and drinking) were also considered. Adherence was measured as a composite score comprising 11 elements. The physician-patient relationship was established based on the interval between physician change and prescription. Logistic regression analysis and propensity score matching were used, along with complementary analysis of physician-patient relationships, proportion of days covered, and adherence for a subset of the population. Results: The study included 48,456 patients with an average follow-up of 36.1 (SD 8.8) months. The mean age was 68.3\thinspace(SD 9.9)\thinspaceyears; BMI, 23.4\thinspace(SD\thinspace3.4)\thinspacekg/m2; systolic blood pressure, 131.2\thinspace(SD\thinspace15)\thinspacemm Hg; low-density lipoprotein cholesterol level, 116.6\thinspace(SD\thinspace29.3)\thinspacemg/dL; hemoglobin A1c (HbA1c), 5.9\%\thinspace(SD\thinspace0.8\%); and serum creatinine level, 0.9\thinspace(SD\thinspace0.8)\thinspacemg/dL. Logistic regression analysis revealed significant associations between generic switching and systolic blood pressure (odds ratio [OR] 0.996, 95\% CI 0.993-0.999), serum creatinine levels (OR 0.837, 95\% CI 0.729-0.962), glutamic oxaloacetic transaminase levels (OR 0.994, 95\% CI 0.990-0.997), proportion of days covered score (OR 0.959, 95\% CI 0.948-0.97), and adherence score (OR 0.910, 95\% CI 0.875-0.947). In addition, generic drug rates increased with improvements in the HbA1c level band and smoking level (P<.01 and P<.001). The group with a superior physician-patient relationship after propensity score matching had a significantly higher rate of generic drug prescribing (51.6\%, SD 15.2\%) than the inferior relationship group (47.7\%, SD17.7\%; P<.001). Conclusions: Although physicians' understanding influences the choice of generic drugs, patient condition (severity) and adherence also impact this decision. For example, improved creatinine levels are associated with generic drug choice, while stronger physician-patient relationships correlate with higher rates of generic drug use. These findings may contribute to the appropriate prescription of pharmaceuticals if the policy diffusion of generic drugs begins to slow down. Thus, preventing serious illness while building trust may result in clinical benefits and positive socioeconomic outcomes. ",
doi="10.2196/59234",
url="https://aging.jmir.org/2024/1/e59234"
}


@Article{info:doi/10.2196/56923,
author="Cotter, M. Lynne
and Shah, Dhavan
and Brown, Kaitlyn
and Mares, Marie-Louise
and Landucci, Gina
and Saunders, Sydney
and Johnston, C. Darcie
and Pe-Romashko, Klaren
and Gustafson, David
and Maus, Adam
and Thompson, Kasey
and Gustafson, H. David",
title="Decoding the Influence of eHealth on Autonomy, Competence, and Relatedness in Older Adults: Qualitative Analysis of Self-Determination Through the Motivational Technology Model",
journal="JMIR Aging",
year="2024",
month="Oct",
day="30",
volume="7",
pages="e56923",
keywords="self-determination theory",
keywords="usability",
keywords="mobile technology model",
keywords="aging",
keywords="eHealth",
keywords="mobile health",
keywords="mHealth",
keywords="smart displays",
keywords="video calls",
keywords="older adult",
keywords="chronic conditions",
keywords="mobile phone",
abstract="Background: Older adults adopt and use eHealth systems to build autonomy, competence, and relatedness and engage in healthy behaviors. The motivational technology model posits that technology features, such as those on websites, smart displays, and mobile phones, must allow for navigability, interactivity, and customizability, which spur feelings of self-determination and intrinsic motivation. We studied ElderTree, an online system for older adults that provides on-demand videos of healthy living content, self-monitoring, and weekly researcher-hosted video meetings. Objective: We aimed to understand the theoretical crossover between the motivational technology model and self-determination theory using features of ElderTree to understand the usability of the technology and how it may support older adults' autonomy, competence, and relatedness. Methods: Drawing participants from a randomized controlled trial of a mobile health app for older adults with multiple chronic conditions, we conducted qualitative interviews with 22 older adults about their use of the app; the interviews were coded using qualitative thematic analysis. Results: Older adults did find that features within ElderTree such as content available on demand, good navigation, and weekly researcher-led video calls supported feelings of autonomy, competence, and relatedness, respectively. Individual differences such as a background using computers also influenced participants' experiences with the smart displays. Conclusions: Participants confirmed the features that increased internal motivation, such as interactivity correlating with feelings of relatedness, but they also found other ways to support autonomous health behavior change beyond narrow views of navigability, interactivity, and customization. ",
doi="10.2196/56923",
url="https://aging.jmir.org/2024/1/e56923"
}


@Article{info:doi/10.2196/52310,
author="Angonese, Giulia
and Buhl, Mareike
and Kuhlmann, Inka
and Kollmeier, Birger
and Hildebrandt, Andrea",
title="Prediction of Hearing Help Seeking to Design a Recommendation Module of an mHealth Hearing App: Intensive Longitudinal Study of Feature Importance Assessment",
journal="JMIR Hum Factors",
year="2024",
month="Aug",
day="12",
volume="11",
pages="e52310",
keywords="hearing loss",
keywords="mobile health",
keywords="mHealth",
keywords="older adults",
keywords="help seeking",
keywords="mobile study",
keywords="machine learning",
keywords="supervised classification",
keywords="feature importance",
keywords="profiling",
keywords="mobile phone",
abstract="Background: Mobile health (mHealth) solutions can improve the quality, accessibility, and equity of health services, fostering early rehabilitation. For individuals with hearing loss, mHealth apps might be designed to support the decision-making processes in auditory diagnostics and provide treatment recommendations to the user (eg, hearing aid need). For some individuals, such an mHealth app might be the first contact with a hearing diagnostic service and should motivate users with hearing loss to seek professional help in a targeted manner. However, personalizing treatment recommendations is only possible by knowing the individual's profile regarding the outcome of interest. Objective: This study aims to characterize individuals who are more or less prone to seeking professional help after the repeated use of an app-based hearing test. The goal was to derive relevant hearing-related traits and personality characteristics for personalized treatment recommendations for users of mHealth hearing solutions. Methods: In total, 185 (n=106, 57.3\% female) nonaided older individuals (mean age 63.8, SD 6.6 y) with subjective hearing loss participated in a mobile study. We collected cross-sectional and longitudinal data on a comprehensive set of 83 hearing-related and psychological measures among those previously found to predict hearing help seeking. Readiness to seek help was assessed as the outcome variable at study end and after 2 months. Participants were classified into help seekers and nonseekers using several supervised machine learning algorithms (random forest, na{\"i}ve Bayes, and support vector machine). The most relevant features for prediction were identified using feature importance analysis. Results: The algorithms correctly predicted action to seek help at study end in 65.9\% (122/185) to 70.3\% (130/185) of cases, reaching 74.8\% (98/131) classification accuracy at follow-up. Among the most important features for classification beyond hearing performance were the perceived consequences of hearing loss in daily life, attitude toward hearing aids, motivation to seek help, physical health, sensory sensitivity personality trait, neuroticism, and income. Conclusions: This study contributes to the identification of individual characteristics that predict help seeking in older individuals with self-reported hearing loss. Suggestions are made for their implementation in an individual-profiling algorithm and for deriving targeted recommendations in mHealth hearing apps. ",
doi="10.2196/52310",
url="https://humanfactors.jmir.org/2024/1/e52310",
url="http://www.ncbi.nlm.nih.gov/pubmed/39133539"
}


@Article{info:doi/10.2196/54872,
author="Liu, Chang
and Zhang, Kai
and Yang, Xiaodong
and Meng, Bingbing
and Lou, Jingsheng
and Liu, Yanhong
and Cao, Jiangbei
and Liu, Kexuan
and Mi, Weidong
and Li, Hao",
title="Development and Validation of an Explainable Machine Learning Model for Predicting Myocardial Injury After Noncardiac Surgery in Two Centers in China: Retrospective Study",
journal="JMIR Aging",
year="2024",
month="Jul",
day="26",
volume="7",
pages="e54872",
keywords="myocardial injury after noncardiac surgery",
keywords="older patients",
keywords="machine learning",
keywords="personalized prediction",
keywords="myocardial injury",
keywords="risk prediction",
keywords="noncardiac surgery",
abstract="Background: Myocardial injury after noncardiac surgery (MINS) is an easily overlooked complication but closely related to postoperative cardiovascular adverse outcomes; therefore, the early diagnosis and prediction are particularly important. Objective: We aimed to develop and validate an explainable machine learning (ML) model for predicting MINS among older patients undergoing noncardiac surgery. Methods: The retrospective cohort study included older patients who had noncardiac surgery from 1 northern center and 1 southern center in China. The data sets from center 1 were divided into a training set and an internal validation set. The data set from center 2 was used as an external validation set. Before modeling, the least absolute shrinkage and selection operator and recursive feature elimination methods were used to reduce dimensions of data and select key features from all variables. Prediction models were developed based on the extracted features using several ML algorithms, including category boosting, random forest, logistic regression, na{\"i}ve Bayes, light gradient boosting machine, extreme gradient boosting, support vector machine, and decision tree. Prediction performance was assessed by the area under the receiver operating characteristic (AUROC) curve as the main evaluation metric to select the best algorithms. The model performance was verified by internal and external validation data sets with the best algorithm and compared to the Revised Cardiac Risk Index. The Shapley Additive Explanations (SHAP) method was applied to calculate values for each feature, representing the contribution to the predicted risk of complication, and generate personalized explanations. Results: A total of 19,463 eligible patients were included; among those, 12,464 patients in center 1 were included as the training set; 4754 patients in center 1 were included as the internal validation set; and 2245 in center 2 were included as the external validation set. The best-performing model for prediction was the CatBoost algorithm, achieving the highest AUROC of 0.805 (95\% CI 0.778?0.831) in the training set, validating with an AUROC of 0.780 in the internal validation set and 0.70 in external validation set. Additionally, CatBoost demonstrated superior performance compared to the Revised Cardiac Risk Index (AUROC 0.636; P<.001). The SHAP values indicated the ranking of the level of importance of each variable, with preoperative serum creatinine concentration, red blood cell distribution width, and age accounting for the top three. The results from the SHAP method can predict events with positive values or nonevents with negative values, providing an explicit explanation of individualized risk predictions. Conclusions: The ML models can provide a personalized and fairly accurate risk prediction of MINS, and the explainable perspective can help identify potentially modifiable sources of risk at the patient level. ",
doi="10.2196/54872",
url="https://aging.jmir.org/2024/1/e54872"
}


@Article{info:doi/10.2196/52592,
author="Barton, J. Hanna
and Maru, Apoorva
and Leaf, A. Margaret
and Hekman, J. Daniel
and Wiegmann, A. Douglas
and Shah, N. Manish
and Patterson, W. Brian",
title="Academic Detailing as a Health Information Technology Implementation Method: Supporting the Design and Implementation of an Emergency Department--Based Clinical Decision Support Tool to Prevent Future Falls",
journal="JMIR Hum Factors",
year="2024",
month="Apr",
day="18",
volume="11",
pages="e52592",
keywords="emergency medicine",
keywords="clinical decision support",
keywords="health IT",
keywords="human factors",
keywords="work systems",
keywords="SEIPS",
keywords="Systems Engineering Initiative for Patient Safety",
keywords="educational outreach",
keywords="academic detailing",
keywords="implementation method",
keywords="department-based",
keywords="CDS",
keywords="clinical care",
keywords="evidence-based",
keywords="CDS tool",
keywords="gerontology",
keywords="geriatric",
keywords="geriatrics",
keywords="older adult",
keywords="older adults",
keywords="elder",
keywords="elderly",
keywords="older person",
keywords="older people",
keywords="preventative intervention",
keywords="team-based analysis",
keywords="machine learning",
keywords="high-risk patient",
keywords="high-risk patients",
keywords="pharmaceutical",
keywords="pharmaceutical sales",
keywords="United States",
keywords="fall-risk prediction",
keywords="EHR",
keywords="electronic health record",
keywords="interview",
keywords="ED environment",
keywords="emergency department",
abstract="Background: Clinical decision support (CDS) tools that incorporate machine learning--derived content have the potential to transform clinical care by augmenting clinicians' expertise. To realize this potential, such tools must be designed to fit the dynamic work systems of the clinicians who use them. We propose the use of academic detailing---personal visits to clinicians by an expert in a specific health IT tool---as a method for both ensuring the correct understanding of that tool and its evidence base and identifying factors influencing the tool's implementation. Objective: This study aimed to assess academic detailing as a method for simultaneously ensuring the correct understanding of an emergency department--based CDS tool to prevent future falls and identifying factors impacting clinicians' use of the tool through an analysis of the resultant qualitative data. Methods: Previously, our team designed a CDS tool to identify patients aged 65 years and older who are at the highest risk of future falls and prompt an interruptive alert to clinicians, suggesting the patient be referred to a mobility and falls clinic for an evidence-based preventative intervention. We conducted 10-minute academic detailing interviews (n=16) with resident emergency medicine physicians and advanced practice providers who had encountered our CDS tool in practice. We conducted an inductive, team-based content analysis to identify factors that influenced clinicians' use of the CDS tool. Results: The following categories of factors that impacted clinicians' use of the CDS were identified: (1) aspects of the CDS tool's design (2) clinicians' understanding (or misunderstanding) of the CDS or referral process, (3) the busy nature of the emergency department environment, (4) clinicians' perceptions of the patient and their associated fall risk, and (5) the opacity of the referral process. Additionally, clinician education was done to address any misconceptions about the CDS tool or referral process, for example, demonstrating how simple it is to place a referral via the CDS and clarifying which clinic the referral goes to. Conclusions: Our study demonstrates the use of academic detailing for supporting the implementation of health information technologies, allowing us to identify factors that impacted clinicians' use of the CDS while concurrently educating clinicians to ensure the correct understanding of the CDS tool and intervention. Thus, academic detailing can inform both real-time adjustments of a tool's implementation, for example, refinement of the language used to introduce the tool, and larger scale redesign of the CDS tool to better fit the dynamic work environment of clinicians. ",
doi="10.2196/52592",
url="https://humanfactors.jmir.org/2024/1/e52592",
url="http://www.ncbi.nlm.nih.gov/pubmed/38635318"
}


@Article{info:doi/10.2196/49154,
author="Holmqvist, Malin
and Johansson, Linda
and Lindenfalk, Bertil
and Thor, Johan
and Ros, Axel",
title="Older Persons' and Health Care Professionals' Design Choices When Co-Designing a Medication Plan Aiming to Promote Patient Safety: Case Study",
journal="JMIR Aging",
year="2023",
month="Oct",
day="5",
volume="6",
pages="e49154",
keywords="co-design",
keywords="engagement",
keywords="medications",
keywords="medication plan",
keywords="older people",
keywords="older adults",
keywords="participatory",
keywords="patient experience",
keywords="patient safety",
keywords="remote",
abstract="Background: Harm from medications is a major patient safety challenge among older persons. Adverse drug events tend to arise when prescribing or evaluating medications; therefore, interventions targeting these may promote patient safety. Guidelines highlight the value of a joint plan for continued treatment. If such a plan includes medications, a medication plan promoting patient safety is advised. There is growing evidence for the benefits of including patients and health care professionals in initiatives for improving health care products and services through co-design. Objective: This study aimed to identify participants' needs and requirements for a medication plan and explore their reasoning for different design choices. Methods: Using a case study design, we collected and analyzed qualitative and quantitative data and compared them side by side. We explored the needs and requirements for a medication plan expressed by 14 participants (older persons, nurses, and physicians) during a co-design initiative in a regional health system in Sweden. We performed a directed content analysis of qualitative data gathered from co-design sessions and interviews. Descriptive statistics were used to analyze the quantitative data from survey answers. Results: A medication plan must provide an added everyday value related to safety, effort, and engagement. The physicians addressed challenges in setting aside time to apply a medication plan, whereas the older persons raised the potential for increased patient involvement. According to the participants, a medication plan needs to support communication, continuity, and interaction. The nurses specifically addressed the need for a plan that was easy to gain an overview of. Important function requirements included providing instant access, automation, and attention. Content requirements included providing detailed information about the medication treatment. Having the plan linked to the medication list and instantly obtainable information was also requested. Conclusions: After discussing the needs and requirements for a medication plan, the participants agreed on an iteratively developed medication plan prototype linked to the medication list within the existing electronic health record. According to the participants, the medication plan prototype may promote patient safety and enable patient engagement, but concerns were raised about its use in daily clinical practice. The last step in the co-design framework is testing the intervention to explore how it works and connects with users. Therefore, testing the medication plan prototype in clinical practice would be a future step. ",
doi="10.2196/49154",
url="https://aging.jmir.org/2023/1/e49154",
url="http://www.ncbi.nlm.nih.gov/pubmed/37796569"
}


@Article{info:doi/10.2196/48128,
author="Hekman, J. Daniel
and Cochran, L. Amy
and Maru, P. Apoorva
and Barton, J. Hanna
and Shah, N. Manish
and Wiegmann, Douglas
and Smith, A. Maureen
and Liao, Frank
and Patterson, W. Brian",
title="Effectiveness of an Emergency Department--Based Machine Learning Clinical Decision Support Tool to Prevent Outpatient Falls Among Older Adults: Protocol for a Quasi-Experimental Study",
journal="JMIR Res Protoc",
year="2023",
month="Aug",
day="3",
volume="12",
pages="e48128",
keywords="falls",
keywords="emergency medicine",
keywords="machine learning",
keywords="clinical decision support",
keywords="automated screening",
keywords="geriatrics",
abstract="Background: Emergency department (ED) providers are important collaborators in preventing falls for older adults because they are often the first health care providers to see a patient after a fall and because at-home falls are often preceded by previous ED visits. Previous work has shown that ED referrals to falls interventions can reduce the risk of an at-home fall by 38\%. Screening patients at risk for a fall can be time-consuming and difficult to implement in the ED setting. Machine learning (ML) and clinical decision support (CDS) offer the potential of automating the screening process. However, it remains unclear whether automation of screening and referrals can reduce the risk of future falls among older patients. Objective: The goal of this paper is to describe a research protocol for evaluating the effectiveness of an automated screening and referral intervention. These findings will inform ongoing discussions about the use of ML and artificial intelligence to augment medical decision-making. Methods: To assess the effectiveness of our program for patients receiving the falls risk intervention, our primary analysis will be to obtain referral completion rates at 3 different EDs. We will use a quasi-experimental design known as a sharp regression discontinuity with regard to intent-to-treat, since the intervention is administered to patients whose risk score falls above a threshold. A conditional logistic regression model will be built to describe 6-month fall risk at each site as a function of the intervention, patient demographics, and risk score. The odds ratio of a return visit for a fall and the 95\% CI will be estimated by comparing those identified as high risk by the ML-based CDS (ML-CDS) and those who were not but had a similar risk profile. Results: The ML-CDS tool under study has been implemented at 2 of the 3 EDs in our study. As of April 2023, a total of 1326 patient encounters have been flagged for providers, and 339 unique patients have been referred to the mobility and falls clinic. To date, 15\% (45/339) of patients have scheduled an appointment with the clinic. Conclusions: This study seeks to quantify the impact of an ML-CDS intervention on patient behavior and outcomes. Our end-to-end data set allows for a more meaningful analysis of patient outcomes than other studies focused on interim outcomes, and our multisite implementation plan will demonstrate applicability to a broad population and the possibility to adapt the intervention to other EDs and achieve similar results. Our statistical methodology, regression discontinuity design, allows for causal inference from observational data and a staggered implementation strategy allows for the identification of secular trends that could affect causal associations and allow mitigation as necessary. Trial Registration: ClinicalTrials.gov NCT05810064; https://www.clinicaltrials.gov/study/NCT05810064 International Registered Report Identifier (IRRID): DERR1-10.2196/48128 ",
doi="10.2196/48128",
url="https://www.researchprotocols.org/2023/1/e48128",
url="http://www.ncbi.nlm.nih.gov/pubmed/37535416"
}


@Article{info:doi/10.2196/31812,
author="Khadjesari, Zarnie
and Houghton, Julie
and Brown, J. Tracey
and Jopling, Helena
and Stevenson, Fiona
and Lynch, Jennifer",
title="Contextual Factors That Impact the Implementation of Patient Portals With a Focus on Older People in Acute Care Hospitals: Scoping Review",
journal="JMIR Aging",
year="2023",
month="Feb",
day="3",
volume="6",
pages="e31812",
keywords="patient portal",
keywords="tethered personal health records",
keywords="acute care hospitals",
keywords="implementation",
keywords="scoping review",
abstract="Background: Older people are the highest users of health services but are less likely to use a patient portal than younger people. Objective: This scoping review aimed to identify and synthesize the literature on contextual factors that impact the implementation of patient portals in acute care hospitals and among older people. Methods: A scoping review was conducted according to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines. The following databases were searched from 2010 to June 2020: MEDLINE and Embase via the Ovid platform, CINAHL and PsycINFO via the EBSCO platform, and the Cochrane Library. Eligible reviews were published in English; focused on the implementation of tethered patient portals; included patients, health care professionals, managers, and budget holders; and aimed at identifying the contextual factors (ie, barriers and facilitators) that impact the implementation of patient portals. Review titles and abstracts and full-text publications were screened in duplicate. The study characteristics were charted by one author and checked for accuracy by a second author. The NASSS (Non-adoption, Abandonment, Scale-up, Spread, and Sustainability) framework was used to synthesize the findings. Results: In total, 10 systematic reviews published between 2015 and 2020 were included in the study. Of these, 3 (30\%) reviews addressed patient portals in acute care hospitals, and 2 (20\%) reviews addressed the implementation of patient portals among older people in multiple settings (including acute care hospitals). To maximize the inclusion of the literature on patient portal implementation, we also included 5 reviews of systematic reviews that examined patient portals in multiple care settings (including acute care hospitals). Contextual factors influencing patient portal implementation tended to cluster in specific NASSS domains, namely the condition, technology, and value proposition. Certain aspects within these domains received more coverage than others, such as sociocultural factors and comorbidities, the usability and functionality aspects of the technology, and the demand-side value. There are gaps in the literature pertinent to the consideration of the provision of patient portals for older people in acute care hospitals, including the lack of consideration of the diversity of older adults and their needs, the question of interoperability between systems (likely to be important where care involves multiple services), the involvement of lay caregivers, and looking beyond short-term implementation to ways in which portal use can be sustained. Conclusions: We identified important contextual factors that impact patient portal implementation and key gaps in the literature. Future research should focus on evaluating strategies that address disparities in use and promote engagement with patient portals among older people in acute care settings. ",
doi="10.2196/31812",
url="https://aging.jmir.org/2023/1/e31812",
url="http://www.ncbi.nlm.nih.gov/pubmed/36735321"
}


@Article{info:doi/10.2196/39386,
author="Adisso, Lionel {\'E}v{\`e}hou{\'e}nou
and Taljaard, Monica
and Stacey, Dawn
and Bri{\`e}re, Nathalie
and Zomahoun, Vignon Herv{\'e} Tchala
and Durand, Jacob Pierre
and Rivest, Louis-Paul
and L{\'e}gar{\'e}, France",
title="Shared Decision-Making Training for Home Care Teams to Engage Frail Older Adults and Caregivers in Housing Decisions: Stepped-Wedge Cluster Randomized Trial",
journal="JMIR Aging",
year="2022",
month="Sep",
day="20",
volume="5",
number="3",
pages="e39386",
keywords="shared decision-making",
keywords="home care",
keywords="nursing homes",
keywords="patient engagement",
abstract="Background: Frail older adults and caregivers need support from their home care teams in making difficult housing decisions, such as whether to remain at home, with or without assistance, or move into residential care. However, home care teams are often understaffed and busy, and shared decision-making training is costly. Nevertheless, overall awareness of shared decision-making is increasing. We hypothesized that distributing a decision aid could be sufficient for providing decision support without the addition of shared decision-making training for home care teams. Objective: We evaluated the effectiveness of adding web-based training and workshops for care teams in interprofessional shared decision-making to passive dissemination of a decision guide on the proportion of frail older adults or caregivers of cognitively-impaired frail older adults reporting active roles in housing decision-making. Methods: We conducted a stepped-wedge cluster randomized trial with home care teams in 9 health centers in Quebec, Canada. Participants were frail older adults or caregivers of cognitively impaired frail older adults facing housing decisions and receiving care from the home care team at one of the participating health centers. The intervention consisted of a 1.5-hour web-based tutorial for the home care teams plus a 3.5-hour interactive workshop in interprofessional shared decision-making using a decision guide that was designed to support frail older adults and caregivers in making housing decisions. The control was passive dissemination of the decision guide. The primary outcome was an active role in decision-making among frail older adults and caregivers, measured using the Control Preferences Scale. Secondary outcomes included decisional conflict and perceptions of how much care teams involved frail older adults and caregivers in decision-making. We performed an intention-to-treat analysis. Results: A total of 311 frail older adults were included in the analysis, including 208 (66.9\%) women, with a mean age of 81.2 (SD 7.5) years. Among 339 caregivers of cognitively-impaired frail older adults, 239 (70.5\%) were female and their mean age was 66.4 (SD 11.7) years. The intervention increased the proportion of frail older adults reporting an active role in decision-making by 3.3\% (95\% CI --5.8\% to 12.4\%, P=.47) and the proportion of caregivers of cognitively-impaired frail older adults by 6.1\% (95\% CI -11.2\% to 23.4\%, P=.49). There was no significant impact on the secondary outcomes. However, the mean score for the frail older adults' perception of how much health professionals involved them in decision-making increased by 5.4 (95\% CI ?0.6 to 11.4, P=.07) and the proportion of caregivers who reported decisional conflict decreased by 7.5\% (95\% CI ?16.5\% to 1.6\%, P=.10). Conclusions: Although it slightly reduced decisional conflict for caregivers, shared decision-making training did not equip home care teams significantly better than provision of a decision aid for involving frail older adults and their caregivers in decision-making. Trial Registration: ClinicalTrials.gov NCT02592525; https://clinicaltrials.gov/show/NCT02592525 ",
doi="10.2196/39386",
url="https://aging.jmir.org/2022/3/e39386",
url="http://www.ncbi.nlm.nih.gov/pubmed/35759791"
}


@Article{info:doi/10.2196/39335,
author="Berridge, Clara
and Turner, R. Natalie
and Liu, Liu
and Karras, W. Sierramatice
and Chen, Amy
and Fredriksen-Goldsen, Karen
and Demiris, George",
title="Advance Planning for Technology Use in Dementia Care: Development, Design, and Feasibility of a Novel Self-administered Decision-Making Tool",
journal="JMIR Aging",
year="2022",
month="Jul",
day="27",
volume="5",
number="3",
pages="e39335",
keywords="Alzheimer disease",
keywords="advance care planning",
keywords="dyadic intervention",
keywords="technology",
keywords="remote monitoring",
keywords="artificial intelligence",
keywords="older adult",
keywords="seniors",
keywords="human-computer interaction",
keywords="aging",
keywords="elderly population",
keywords="digital tool",
keywords="educational tool",
keywords="dementia care",
keywords="ethics",
keywords="informed consent",
abstract="Background: Monitoring technologies are used to collect a range of information, such as one's location out of the home or movement within the home, and transmit that information to caregivers to support aging in place. Their surveilling nature, however, poses ethical dilemmas and can be experienced as intrusive to people living with Alzheimer disease (AD) and AD-related dementias. These challenges are compounded when older adults are not engaged in decision-making about how they are monitored. Dissemination of these technologies is outpacing our understanding of how to communicate their functions, risks, and benefits to families and older adults. To date, there are no tools to help families understand the functions of monitoring technologies or guide them in balancing their perceived need for ongoing surveillance and the older adult's dignity and wishes. Objective: We designed, developed, and piloted a communication and education tool in the form of a web application called Let's Talk Tech to support family decision-making about diverse technologies used in dementia home care. The knowledge base about how to design online interventions for people living with mild dementia is still in development, and dyadic interventions used in dementia care remain rare. We describe the intervention's motivation and development process, and the feasibility of using this self-administered web application intervention in a pilot sample of people living with mild AD and their family care partners. Methods: We surveyed 29 mild AD dementia care dyads living together before and after they completed the web application intervention and interviewed each dyad about their experiences with it. We report postintervention measures of feasibility (recruitment, enrollment, and retention) and acceptability (satisfaction, quality, and usability). Descriptive statistics were calculated for survey items, and thematic analysis was used with interview transcripts to illuminate participants' experiences and recommendations to improve the intervention. Results: The study enrolled 33 people living with AD and their care partners, and 29 (88\%) dyads completed the study (all but one were spousal dyads). Participants were asked to complete 4 technology modules, and all completed them. The majority of participants rated the tool as having the right length (>90\%), having the right amount of information (>84\%), being very clearly worded (>74\%), and presenting information in a balanced way (>90\%). Most felt the tool was easy to use and helpful, and would likely recommend it to others. Conclusions: This study demonstrated that our intervention to educate and facilitate conversation and documentation of preferences is preliminarily feasible and acceptable to mild AD care dyads. Effectively involving older adults in these decisions and informing care partners of their preferences could enable families to avoid conflicts or risks associated with uninformed or disempowered use and to personalize use so both members of the dyad can experience benefits. ",
doi="10.2196/39335",
url="https://aging.jmir.org/2022/3/e39335",
url="http://www.ncbi.nlm.nih.gov/pubmed/35896014"
}


@Article{info:doi/10.2196/35929,
author="Singh, Hardeep
and Tang, Terence
and Steele Gray, Carolyn
and Kokorelias, Kristina
and Thombs, Rachel
and Plett, Donna
and Heffernan, Matthew
and Jarach, M. Carlotta
and Armas, Alana
and Law, Susan
and Cunningham, V. Heather
and Nie, Xin Jason
and Ellen, E. Moriah
and Thavorn, Kednapa
and Nelson, LA Michelle",
title="Recommendations for the Design and Delivery of Transitions-Focused Digital Health Interventions: Rapid Review",
journal="JMIR Aging",
year="2022",
month="May",
day="19",
volume="5",
number="2",
pages="e35929",
keywords="transitions",
keywords="health",
keywords="medical informatics",
keywords="aged",
keywords="mobile phone",
abstract="Background: Older adults experience a high risk of adverse events during hospital-to-home transitions. Implementation barriers have prevented widespread clinical uptake of the various digital health technologies that aim to support hospital-to-home transitions. Objective: To guide the development of a digital health intervention to support transitions from hospital to home (the Digital Bridge intervention), the specific objectives of this review were to describe the various roles and functions of health care providers supporting hospital-to-home transitions for older adults, allowing future technologies to be more targeted to support their work; describe the types of digital health interventions used to facilitate the transition from hospital to home for older adults and elucidate how these interventions support the roles and functions of providers; describe the lessons learned from the design and implementation of these interventions; and identify opportunities to improve the fit between technology and provider functions within the Digital Bridge intervention and other transition-focused digital health interventions. Methods: This 2-phase rapid review involved a selective review of providers' roles and their functions during hospital-to-home transitions (phase 1) and a structured literature review on digital health interventions used to support older adults' hospital-to-home transitions (phase 2). During the analysis, the technology functions identified in phase 2 were linked to the provider roles and functions identified in phase 1. Results: In phase 1, various provider roles were identified that facilitated hospital-to-home transitions, including navigation-specific roles and the roles of nurses and physicians. The key transition functions performed by providers were related to the 3 categories of continuity of care (ie, informational, management, and relational continuity). Phase 2, included articles (n=142) that reported digital health interventions targeting various medical conditions or groups. Most digital health interventions supported management continuity (eg, follow-up, assessment, and monitoring of patients' status after hospital discharge), whereas informational and relational continuity were the least supported. The lessons learned from the interventions were categorized into technology- and research-related challenges and opportunities and informed several recommendations to guide the design of transition-focused digital health interventions. Conclusions: This review highlights the need for Digital Bridge and other digital health interventions to align the design and delivery of digital health interventions with provider functions, design and test interventions with older adults, and examine multilevel outcomes. International Registered Report Identifier (IRRID): RR2-10.1136/bmjopen-2020-045596 ",
doi="10.2196/35929",
url="https://aging.jmir.org/2022/2/e35929",
url="http://www.ncbi.nlm.nih.gov/pubmed/35587874"
}


@Article{info:doi/10.2196/37522,
author="Gustafson, H. David
and Mares, Marie-Louise
and Johnston, C. Darcie
and Landucci, Gina
and Pe-Romashko, Klaren
and Vjorn, J. Olivia
and Hu, Yaxin
and Maus, Adam
and Mahoney, E. Jane
and Mutlu, Bilge",
title="Using Smart Displays to Implement an eHealth System for Older Adults With Multiple Chronic Conditions: Protocol for a Randomized Controlled Trial",
journal="JMIR Res Protoc",
year="2022",
month="May",
day="5",
volume="11",
number="5",
pages="e37522",
keywords="eHealth",
keywords="aged",
keywords="geriatrics",
keywords="multiple chronic conditions",
keywords="chronic pain",
keywords="smart displays",
keywords="smart speakers",
keywords="quality of life",
keywords="primary care",
keywords="health expenditures",
keywords="mobile phone",
abstract="Background: Voice-controlled smart speakers and displays have a unique but unproven potential for delivering eHealth interventions. Many laptop- and smartphone-based interventions have been shown to improve multiple outcomes, but voice-controlled platforms have not been tested in large-scale rigorous trials. Older adults with multiple chronic health conditions, who need tools to help with their daily management, may be especially good candidates for interventions on voice-controlled devices because these patients often have physical limitations, such as tremors or vision problems, that make the use of laptops and smartphones challenging. Objective: The aim of this study is to assess whether participants using an evidence-based intervention (ElderTree) on a smart display will experience decreased pain interference and improved quality of life and related measures in comparison with participants using ElderTree on a laptop and control participants who are given no device or access to ElderTree. Methods: A total of 291 adults aged ?60 years with chronic pain and ?3 additional chronic conditions will be recruited from primary care clinics and community organizations and randomized 1:1:1 to ElderTree access on a smart display along with their usual care, ElderTree access on a touch screen laptop along with usual care, or usual care alone. All patients will be followed for 8 months. The primary outcomes are differences between groups in measures of pain interference and psychosocial quality of life. The secondary outcomes are between-group differences in system use at 8 months, physical quality of life, pain intensity, hospital readmissions, communication with medical providers, health distress, well-being, loneliness, and irritability. We will also examine mediators and moderators of the effects of ElderTree on both platforms. At baseline, 4 months, and 8 months, patients will complete written surveys comprising validated scales selected for good psychometric properties with similar populations. ElderTree use data will be collected continuously in system logs. We will use linear mixed-effects models to evaluate outcomes over time, with treatment condition and time acting as between-participant factors. Separate analyses will be conducted for each outcome. Results: Recruitment began in August 2021 and will run through April 2023. The intervention period will end in December 2023. The findings will be disseminated via peer-reviewed publications. Conclusions: To our knowledge, this is the first study with a large sample and long time frame to examine whether a voice-controlled smart device can perform as well as or better than a laptop in implementing a health intervention for older patients with multiple chronic health conditions. As patients with multiple conditions are such a large cohort, the implications for cost as well as patient well-being are significant. Making the best use of current and developing technologies is a critical part of this effort. Trial Registration: ClinicalTrials.gov NCT04798196; https://clinicaltrials.gov/ct2/show/NCT04798196 International Registered Report Identifier (IRRID): PRR1-10.2196/37522 ",
doi="10.2196/37522",
url="https://www.researchprotocols.org/2022/5/e37522",
url="http://www.ncbi.nlm.nih.gov/pubmed/35511229"
}


@Article{info:doi/10.2196/33320,
author="Dupont, Charl{\`e}ss
and Smets, Tinne
and Monnet, Fanny
and Pivodic, Lara
and De Vleminck, Aline
and Van Audenhove, Chantal
and Van den Block, Lieve",
title="Publicly Available, Interactive Web-Based Tools to Support Advance Care Planning: Systematic Review",
journal="J Med Internet Res",
year="2022",
month="Apr",
day="20",
volume="24",
number="4",
pages="e33320",
keywords="advance care planning",
keywords="systematic review",
keywords="web-based tools",
keywords="health communication",
keywords="quality of online content",
abstract="Background: There is an increasing number of interactive web-based advance care planning (ACP) support tools, which are web-based aids in any format encouraging reflection, communication, and processing of publicly available information, most of which cannot be found in the peer-reviewed literature. Objective: This study aims to conduct a systematic review of web-based ACP support tools to describe the characteristics, readability, and quality of content and investigate whether and how they are evaluated. Methods: We systematically searched the web-based gray literature databases OpenGrey, ClinicalTrials.gov, ProQuest, British Library, Grey Literature in the Netherlands, and Health Services Research Projects in Progress, as well as Google and app stores, and consulted experts using the following eligibility criteria: web-based, designed for the general population, accessible to everyone, interactive (encouraging reflection, communication, and processing of information), and in English or Dutch. The quality of content was evaluated using the Quality Evaluation Scoring Tool (score 0-28---a higher score indicates better quality). To synthesize the characteristics of the ACP tools, readability and quality of content, and whether and how they were evaluated, we used 4 data extraction tables. Results: A total of 30 tools met the eligibility criteria, including 15 (50\%) websites, 10 (33\%) web-based portals, 3 (10\%) apps, and 2 (7\%) with a combination of formats. Of the 30 tools, 24 (80\%) mentioned a clear aim, including 7 (23\%) that supported reflection or communication, 8 (27\%) that supported people in making decisions, 7 (23\%) that provided support to document decisions, and 2 (7\%) that aimed to achieve all these aims. Of the 30 tools, 7 (23\%) provided information on the development, all of which were developed in collaboration with health care professionals, and 3 (10\%) with end users. Quality scores ranged between 11 and 28, with most of the lower-scoring tools not referring to information sources. Conclusions: A variety of ACP support tools are available on the web, varying in the quality of content. In the future, users should be involved in the development process of ACP support tools, and the content should be substantiated by scientific evidence. Trial Registration: PROSPERO CRD42020184112; https://tinyurl.com/mruf8b43 ",
doi="10.2196/33320",
url="https://www.jmir.org/2022/4/e33320",
url="http://www.ncbi.nlm.nih.gov/pubmed/35442207"
}


@Article{info:doi/10.2196/34626,
author="Sch{\"o}pfer, C{\'e}line
and Ehrler, Frederic
and Berger, Antoine
and Bollondi Pauly, Catherine
and Buytaert, Laurence
and De La Serna, Camille
and Hartheiser, Florence
and Fassier, Thomas
and Clavien, Christine",
title="A Mobile App for Advance Care Planning and Advance Directives (Accordons-nous): Development and Usability Study",
journal="JMIR Hum Factors",
year="2022",
month="Apr",
day="20",
volume="9",
number="2",
pages="e34626",
keywords="usability",
keywords="mobile apps",
keywords="advance directives",
keywords="advance care planning",
keywords="mHealth",
keywords="mobile health",
keywords="palliative care",
keywords="mobile phone",
abstract="Background: Advance care planning, including advance directives, is an important tool that allows patients to express their preferences for care if they are no longer able to express themselves. We developed Accordons-nous, a smartphone app that informs patients about advance care planning and advance directives, facilitates communication on these sensitive topics, and helps patients express their values and preferences for care. Objective: The first objective of this study is to conduct a usability test of this app. The second objective is to collect users' critical opinions on the usability and relevance of the tool. Methods: We conducted a usability test by means of a think-aloud method, asking 10 representative patients to complete 7 browsing tasks. We double coded the filmed sessions to obtain descriptive data on task completion (with or without help), time spent, number of clicks, and the types of problems encountered. We assessed the severity of the problems encountered and identified the modifications needed to address these problems. We evaluated the readability of the app using Scolarius, a French equivalent of the Flesch Reading Ease test. By means of a posttest questionnaire, we asked participants to assess the app's usability (System Usability Scale), relevance (Mobile App Rating Scale, section F), and whether they would recommend the app to the target groups: patients, health professionals, and patients' caring relatives. Results: Participants completed the 7 think-aloud tasks in 80\% (56/70) of the cases without any help from the experimenter, in 16\% (11/70) of the cases with some help, and failed in 4\% (3/70) of the cases. The analysis of failures and difficulties encountered revealed a series of major usability problems that could be addressed with minor modifications to the app. Accordons-nous obtained high scores on readability (overall score of 87.4 on Scolarius test, corresponding to elementary school level), usability (85.3/100 on System Usability Scale test), relevance (4.3/5 on the Mobile App Rating Scale, section F), and overall subjective endorsement on 3 I would recommend questions (4.7/5). Conclusions: This usability test helped us make the final changes to our app before its official launch. ",
doi="10.2196/34626",
url="https://humanfactors.jmir.org/2022/2/e34626",
url="http://www.ncbi.nlm.nih.gov/pubmed/35442206"
}


@Article{info:doi/10.2196/32683,
author="Dimet-Wiley, Andrea
and Golovko, George
and Watowich, J. Stanley",
title="One-Year Postfracture Mortality Rate in Older Adults With Hip Fractures Relative to Other Lower Extremity Fractures: Retrospective Cohort Study",
journal="JMIR Aging",
year="2022",
month="Mar",
day="16",
volume="5",
number="1",
pages="e32683",
keywords="hip",
keywords="fracture",
keywords="mortality",
keywords="aging",
keywords="older adults",
keywords="elderly",
keywords="mortality risk",
keywords="electronic health record",
keywords="EHR",
keywords="survival probability",
keywords="postfracture mortality rate",
keywords="fall",
keywords="bone",
keywords="injury",
keywords="dementia",
keywords="diabetes",
keywords="type 2 diabetes",
keywords="trauma",
keywords="treatment",
keywords="comorbidity",
keywords="mobility",
abstract="Background: Hip fracture in older adults is tied to increased mortality risk. Deconvolution of the mortality risk specific to hip fracture from that of various other fracture types has not been performed in recent hip fracture studies but is critical to determining current unmet needs for therapeutic intervention. Objective: This study examined whether hip fracture increases the 1-year postfracture mortality rate relative to several other fracture types and determined whether dementia or type 2 diabetes (T2D) exacerbates postfracture mortality risk. Methods: TriNetX Diamond Network data were used to identify patients with a single event of fracture of the hip, the upper humerus, or several regions near and distal to the hip occurring from 60 to 89 years of age from 2010 to 2019. Propensity score matching, Kaplan-Meier, and hazard ratio analyses were performed for all fracture groupings relative to hip fracture. One-year postfracture mortality rates in elderly populations with dementia or T2D were established. Results: One-year mortality rates following hip fracture consistently exceeded all other lower extremity fracture groupings as well as the upper humerus. Survival probabilities were significantly lower in the hip fracture groups, even after propensity score matching was performed on cohorts for a variety of broad categories of characteristics. Dementia in younger elderly cohorts acted synergistically with hip fracture to exacerbate the 1-year mortality risk. T2D did not exacerbate the 1-year mortality risk beyond mere additive effects. Conclusions: Elderly patients with hip fracture have a significantly decreased survival probability. Greatly increased 1-year mortality rates following hip fracture may arise from differences in bone quality, bone density, trauma, concomitant fractures, postfracture treatments or diagnoses, restoration of prefracture mobility, or a combination thereof. The synergistic effect of dementia may suggest detrimental mechanistic or behavioral combinations for these 2 comorbidities. Renewed efforts should focus on modulating the mechanisms behind this heightened mortality risk, with particular attention to mobility and comorbid dementia. ",
doi="10.2196/32683",
url="https://aging.jmir.org/2022/1/e32683",
url="http://www.ncbi.nlm.nih.gov/pubmed/35293865"
}


@Article{info:doi/10.2196/26486,
author="Goh, Huat Kim
and Wang, Le
and Yeow, Kwang Adrian Yong
and Ding, Yoong Yew
and Au, Yi Lydia Shu
and Poh, Niang Hermione Mei
and Li, Ke
and Yeow, Lin Joannas Jie
and Tan, Heng Gamaliel Yu",
title="Prediction of Readmission in Geriatric Patients From Clinical Notes: Retrospective Text Mining Study",
journal="J Med Internet Res",
year="2021",
month="Oct",
day="19",
volume="23",
number="10",
pages="e26486",
keywords="geriatrics",
keywords="readmission risk",
keywords="artificial intelligence",
keywords="text mining",
keywords="psychosocial factors",
abstract="Background: Prior literature suggests that psychosocial factors adversely impact health and health care utilization outcomes. However, psychosocial factors are typically not captured by the structured data in electronic medical records (EMRs) but are rather recorded as free text in different types of clinical notes. Objective: We here propose a text-mining approach to analyze EMRs to identify older adults with key psychosocial factors that predict adverse health care utilization outcomes, measured by 30-day readmission. The psychological factors were appended to the LACE (Length of stay, Acuity of the admission, Comorbidity of the patient, and Emergency department use) Index for Readmission to improve the prediction of readmission risk. Methods: We performed a retrospective analysis using EMR notes of 43,216 hospitalization encounters in a hospital from January 1, 2017 to February 28, 2019. The mean age of the cohort was 67.51 years (SD 15.87), the mean length of stay was 5.57 days (SD 10.41), and the mean intensive care unit stay was 5\% (SD 22\%).  We employed text-mining techniques to extract psychosocial topics that are representative of these patients and tested the utility of these topics in predicting 30-day hospital readmission beyond the predictive value of the LACE Index for Readmission. Results: The added text-mined factors improved the area under the receiver operating characteristic curve of the readmission prediction by 8.46\% for geriatric patients, 6.99\% for the general hospital population, and 6.64\% for frequent admitters. Medical social workers and case managers captured more of the psychosocial text topics than physicians. Conclusions: The results of this study demonstrate the feasibility of extracting psychosocial factors from EMR clinical notes and the value of these notes in improving readmission risk prediction. Psychosocial profiles of patients can be curated and quantified from text mining clinical notes and these profiles can be successfully applied to artificial intelligence models to improve readmission risk prediction. ",
doi="10.2196/26486",
url="https://www.jmir.org/2021/10/e26486",
url="http://www.ncbi.nlm.nih.gov/pubmed/34665149"
}


@Article{info:doi/10.2196/24015,
author="Kraaijkamp, M. Jules J.
and van Dam van Isselt, F. El{\'e}onore
and Persoon, Anke
and Versluis, Anke
and Chavannes, H. Niels
and Achterberg, P. Wilco",
title="eHealth in Geriatric Rehabilitation: Systematic Review of Effectiveness, Feasibility, and Usability",
journal="J Med Internet Res",
year="2021",
month="Aug",
day="19",
volume="23",
number="8",
pages="e24015",
keywords="geriatric rehabilitation",
keywords="eHealth",
keywords="mHealth",
keywords="digital health",
keywords="effectiveness",
keywords="feasibility",
keywords="usability",
keywords="systematic review",
abstract="Background: eHealth has the potential to improve outcomes such as physical activity or balance in older adults receiving geriatric rehabilitation. However, several challenges such as scarce evidence on effectiveness, feasibility, and usability hinder the successful implementation of eHealth in geriatric rehabilitation. Objective: The aim of this systematic review was to assess evidence on the effectiveness, feasibility, and usability of eHealth interventions in older adults in geriatric rehabilitation. Methods: We searched 7 databases for randomized controlled trials, nonrandomized studies, quantitative descriptive studies, qualitative research, and mixed methods studies that applied eHealth interventions during geriatric rehabilitation. Included studies investigated a combination of effectiveness, usability, and feasibility of eHealth in older patients who received geriatric rehabilitation, with a mean age of ?70 years. Quality was assessed using the Mixed Methods Appraisal Tool and a narrative synthesis was conducted using a harvest plot. Results: In total, 40 studies were selected, with clinical heterogeneity across studies. Of 40 studies, 15 studies (38\%) found eHealth was at least as effective as non-eHealth interventions (56\% of the 27 studies with a control group), 11 studies (41\%) found eHealth interventions were more effective than non-eHealth interventions, and 1 study (4\%) reported beneficial outcomes in favor of the non-eHealth interventions. Of 17 studies, 16 (94\%) concluded that eHealth was feasible. However, high exclusion rates were reported in 7 studies of 40 (18\%). Of 40 studies, 4 (10\%) included outcomes related to usability and indicated that there were certain aging-related barriers to cognitive ability, physical ability, or perception, which led to difficulties in using eHealth. Conclusions: eHealth can potentially improve rehabilitation outcomes for older patients receiving geriatric rehabilitation. Simple eHealth interventions were more likely to be feasible for older patients receiving geriatric rehabilitation, especially, in combination with another non-eHealth intervention. However, a lack of evidence on usability might hamper the implementation of eHealth. eHealth applications in geriatric rehabilitation show promise, but more research is required, including research with a focus on usability and participation. ",
doi="10.2196/24015",
url="https://www.jmir.org/2021/8/e24015",
url="http://www.ncbi.nlm.nih.gov/pubmed/34420918"
}


@Article{info:doi/10.2196/22491,
author="Chen, Rai-Fu
and Cheng, Kuei-Chen
and Lin, Yu-Yin
and Chang, I-Chiu
and Tsai, Cheng-Han",
title="Predicting Unscheduled Emergency Department Return Visits Among Older Adults: Population-Based Retrospective Study",
journal="JMIR Med Inform",
year="2021",
month="Jul",
day="28",
volume="9",
number="7",
pages="e22491",
keywords="classification model",
keywords="decision tree",
keywords="emergency department",
keywords="older adult patients",
keywords="unscheduled return visits",
abstract="Background: Unscheduled emergency department return visits (EDRVs) are key indicators for monitoring the quality of emergency medical care. A high return rate implies that the medical services provided by the emergency department (ED) failed to achieve the expected results of accurate diagnosis and effective treatment. Older adults are more susceptible to diseases and comorbidities than younger adults, and they exhibit unique and complex clinical characteristics that increase the difficulty of clinical diagnosis and treatment. Older adults also use more emergency medical resources than people in other age groups. Many studies have reviewed the causes of EDRVs among general ED patients; however, few have focused on older adults, although this is the age group with the highest rate of EDRVs. Objective: This aim of this study is to establish a model for predicting unscheduled EDRVs within a 72-hour period among patients aged 65 years and older. In addition, we aim to investigate the effects of the influencing factors on their unscheduled EDRVs. Methods: We used stratified and randomized data from Taiwan's National Health Insurance Research Database and applied data mining techniques to construct a prediction model consisting of patient, disease, hospital, and physician characteristics. Records of ED visits by patients aged 65 years and older from 1996 to 2010 in the National Health Insurance Research Database were selected, and the final sample size was 49,252 records. Results: The decision tree of the prediction model achieved an acceptable overall accuracy of 76.80\%. Economic status, chronic illness, and length of stay in the ED were the top three variables influencing unscheduled EDRVs. Those who stayed in the ED overnight or longer on their first visit were less likely to return. This study confirms the results of prior studies, which found that economically underprivileged older adults with chronic illness and comorbidities were more likely to return to the ED. Conclusions: Medical institutions can use our prediction model as a reference to improve medical management and clinical services by understanding the reasons for 72-hour unscheduled EDRVs in older adult patients. A possible solution is to create mechanisms that incorporate our prediction model and develop a support system with customized medical education for older patients and their family members before discharge. Meanwhile, a reasonably longer length of stay in the ED may help evaluate treatments and guide prognosis for older adult patients, and it may further reduce the rate of their unscheduled EDRVs. ",
doi="10.2196/22491",
url="https://medinform.jmir.org/2021/7/e22491",
url="http://www.ncbi.nlm.nih.gov/pubmed/34319244"
}


@Article{info:doi/10.2196/16213,
author="Peng, Li-Ning
and Hsiao, Fei-Yuan
and Lee, Wei-Ju
and Huang, Shih-Tsung
and Chen, Liang-Kung",
title="Comparisons Between Hypothesis- and Data-Driven Approaches for Multimorbidity Frailty Index: A Machine Learning Approach",
journal="J Med Internet Res",
year="2020",
month="Jun",
day="11",
volume="22",
number="6",
pages="e16213",
keywords="multimorbidity frailty index",
keywords="machine learning",
keywords="random forest",
keywords="unplanned hospitalizations",
keywords="intensive care unit admissions",
keywords="mortality",
abstract="Background: Using big data and the theory of cumulative deficits to develop the multimorbidity frailty index (mFI) has become a widely accepted approach in public health and health care services. However, constructing the mFI using the most critical determinants and stratifying different risk groups with dose-response relationships remain major challenges in clinical practice. Objective: This study aimed to develop the mFI by using machine learning methods that select variables based on the optimal fitness of the model. In addition, we aimed to further establish 4 entities of risk using a machine learning approach that would achieve the best distinction between groups and demonstrate the dose-response relationship. Methods: In this study, we used Taiwan's National Health Insurance Research Database to develop a machine learning multimorbidity frailty index (ML-mFI) using the theory of cumulative diseases/deficits of an individual older person. Compared to the conventional mFI, in which the selection of diseases/deficits is based on expert opinion, we adopted the random forest method to select the most influential diseases/deficits that predict adverse outcomes for older people. To ensure that the survival curves showed a dose-response relationship with overlap during the follow-up, we developed the distance index and coverage index, which can be used at any time point to classify the ML-mFI of all subjects into the categories of fit, mild frailty, moderate frailty, and severe frailty. Survival analysis was conducted to evaluate the ability of the ML-mFI to predict adverse outcomes, such as unplanned hospitalizations, intensive care unit (ICU) admissions, and mortality. Results: The final ML-mFI model contained 38 diseases/deficits. Compared with conventional mFI, both indices had similar distribution patterns by age and sex; however, among people aged 65 to 69 years, the mean mFI and ML-mFI were 0.037 (SD 0.048) and 0.0070 (SD 0.0254), respectively. The difference may result from discrepancies in the diseases/deficits selected in the mFI and the ML-mFI. A total of 86,133 subjects aged 65 to 100 years were included in this study and were categorized into 4 groups according to the ML-mFI. Both the Kaplan-Meier survival curves and Cox models showed that the ML-mFI significantly predicted all outcomes of interest, including all-cause mortality, unplanned hospitalizations, and all-cause ICU admissions at 1, 5, and 8 years of follow-up (P<.01). In particular, a dose-response relationship was revealed between the 4 ML-mFI groups and adverse outcomes. Conclusions: The ML-mFI consists of 38 diseases/deficits that can successfully stratify risk groups associated with all-cause mortality, unplanned hospitalizations, and all-cause ICU admissions in older people, which indicates that precise, patient-centered medical care can be a reality in an aging society. ",
doi="10.2196/16213",
url="http://www.jmir.org/2020/6/e16213/",
url="http://www.ncbi.nlm.nih.gov/pubmed/32525481"
}


@Article{info:doi/10.2196/16678,
author="Tarekegn, Adane
and Ricceri, Fulvio
and Costa, Giuseppe
and Ferracin, Elisa
and Giacobini, Mario",
title="Predictive Modeling for Frailty Conditions in Elderly People: Machine Learning Approaches",
journal="JMIR Med Inform",
year="2020",
month="Jun",
day="4",
volume="8",
number="6",
pages="e16678",
keywords="predictive modeling",
keywords="frailty",
keywords="machine learning",
keywords="genetic programming",
keywords="imbalanced dataset",
keywords="elderly people",
keywords="classification",
abstract="Background: Frailty is one of the most critical age-related conditions in older adults. It is often recognized as a syndrome of physiological decline in late life, characterized by a marked vulnerability to adverse health outcomes. A clear operational definition of frailty, however, has not been agreed so far. There is a wide range of studies on the detection of frailty and their association with mortality. Several of these studies have focused on the possible risk factors associated with frailty in the elderly population while predicting who will be at increased risk of frailty is still overlooked in clinical settings. Objective: The objective of our study was to develop predictive models for frailty conditions in older people using different machine learning methods based on a database of clinical characteristics and socioeconomic factors. Methods: An administrative health database containing 1,095,612 elderly people aged 65 or older with 58 input variables and 6 output variables was used. We first identify and define six problems/outputs as surrogates of frailty. We then resolve the imbalanced nature of the data through resampling process and a comparative study between the different machine learning (ML) algorithms -- Artificial neural network (ANN), Genetic programming (GP), Support vector machines (SVM), Random Forest (RF), Logistic regression (LR) and Decision tree (DT) -- was carried out. The performance of each model was evaluated using a separate unseen dataset. Results: Predicting mortality outcome has shown higher performance with ANN (TPR 0.81, TNR 0.76, accuracy 0.78, F1-score 0.79) and SVM (TPR 0.77, TNR 0.80, accuracy 0.79, F1-score 0.78) than predicting the other outcomes. On average, over the six problems, the DT classifier has shown the lowest accuracy, while other models (GP, LR, RF, ANN, and SVM) performed better. All models have shown lower accuracy in predicting an event of an emergency admission with red code than predicting fracture and disability. In predicting urgent hospitalization, only SVM achieved better performance (TPR 0.75, TNR 0.77, accuracy 0.73, F1-score 0.76) with the 10-fold cross validation compared with other models in all evaluation metrics. Conclusions: We developed machine learning models for predicting frailty conditions (mortality, urgent hospitalization, disability, fracture, and emergency admission). The results show that the prediction performance of machine learning models significantly varies from problem to problem in terms of different evaluation metrics. Through further improvement, the model that performs better can be used as a base for developing decision-support tools to improve early identification and prediction of frail older adults. ",
doi="10.2196/16678",
url="http://medinform.jmir.org/2020/6/e16678/",
url="http://www.ncbi.nlm.nih.gov/pubmed/32442149"
}


@Article{info:doi/10.2196/13039,
author="Chen, Tao
and Dredze, Mark
and Weiner, P. Jonathan
and Hernandez, Leilani
and Kimura, Joe
and Kharrazi, Hadi",
title="Extraction of Geriatric Syndromes From Electronic Health Record Clinical Notes: Assessment of Statistical Natural Language Processing Methods",
journal="JMIR Med Inform",
year="2019",
month="Mar",
day="26",
volume="7",
number="1",
pages="e13039",
keywords="geriatrics",
keywords="clinical notes",
keywords="natural language processing",
keywords="information extraction",
keywords="conditional random fields",
abstract="Background: Geriatric syndromes in older adults are associated with adverse outcomes. However, despite being reported in clinical notes, these syndromes are often poorly captured by diagnostic codes in the structured fields of electronic health records (EHRs) or administrative records. Objective: We aim to automatically determine if a patient has any geriatric syndromes by mining the free text of associated EHR clinical notes. We assessed which statistical natural language processing (NLP) techniques are most effective. Methods: We applied conditional random fields (CRFs), a widely used machine learning algorithm, to identify each of 10 geriatric syndrome constructs in a clinical note. We assessed three sets of features and attributes for CRF operations: a base set, enhanced token, and contextual features. We trained the CRF on 3901 manually annotated notes from 85 patients, tuned the CRF on a validation set of 50 patients, and evaluated it on 50 held-out test patients. These notes were from a group of US Medicare patients over 65 years of age enrolled in a Medicare Advantage Health Maintenance Organization and cared for by a large group practice in Massachusetts. Results: A final feature set was formed through comprehensive feature ablation experiments. The final CRF model performed well at patient-level determination (macroaverage F1=0.834, microaverage F1=0.851); however, performance varied by construct. For example, at phrase-partial evaluation, the CRF model worked well on constructs such as absence of fecal control (F1=0.857) and vision impairment (F1=0.798) but poorly on malnutrition (F1=0.155), weight loss (F1=0.394), and severe urinary control issues (F1=0.532). Errors were primarily due to previously unobserved words (ie, out-of-vocabulary) and a lack of context. Conclusions: This study shows that statistical NLP can be used to identify geriatric syndromes from EHR-extracted clinical notes. This creates new opportunities to identify patients with geriatric syndromes and study their health outcomes. ",
doi="10.2196/13039",
url="http://medinform.jmir.org/2019/1/e13039/",
url="http://www.ncbi.nlm.nih.gov/pubmed/30862607"
}