Using machine-learning methods to predict in-hospital mortality through the Elixhauser index: A Medicare data analysis.

Accurate in-hospital mortality prediction can reflect the prognosis of patients, help guide allocation of clinical resources, and help clinicians make the right care decisions. There are limitations to using traditional logistic regression models when assessing the model performance of comorbidity measures to predict in-hospital mortality. Meanwhile, the use of novel machine-learning methods is growing rapidly. In 2021, the Agency for Healthcare Research and Quality published new guidelines for using the Present-on-Admission (POA) indicator from the International Classification of Diseases, Tenth Revision, for coding comorbidities to predict in-hospital mortality from the Elixhauser's comorbidity measurement method. We compared the model performance of logistic regression, elastic net model, and artificial neural network (ANN) to predict in-hospital mortality from Elixhauser's measures under the updated POA guidelines. In this retrospective analysis, 1,810,106 adult Medicare inpatient admissions from six US states admitted after September 23, 2017, and discharged before April 11, 2019 were extracted from the Centers for Medicare and Medicaid Services data warehouse. The POA indicator was used to distinguish pre-existing comorbidities from complications that occurred during hospitalization. All models performed well (C-statistics >0.77). Elastic net method generated a parsimonious model, in which there were five fewer comorbidities selected to predict in-hospital mortality with similar predictive power compared to the logistic regression model. ANN had the highest C-statistics compared to the other two models (0.800 vs. 0.791 and 0.791). Elastic net model and AAN can be applied successfully to predict in-hospital mortality.

Exploring HIV concern in a population of Dominican American women midlife and older.

BACKGROUND: The feminization and ethnic diversification of HIV infection, has resulted in a call for gender- and culture-specific prevention strategies for at-risk groups including Latinos in the United States. The steadily changing demographic profile of the AIDS epidemic challenges prevention strategies to remain relevant and up-to-date, particularly in populations of women midlife and older where an understanding of risk remains under explored. As the CDC requests country-specific HIV risk profiles for Latino communities in the US, understanding the socio-economic, behavioral and personal risk reasons of HIV risk for older Dominican women is critical for prevention. METHODS: We conducted focus group discussions informed by the Theory of Gender and Power (TGP). The three constructs of the TGP: 1) Affective influences/social norms; 2) Gender-specific norms and. 3) Power and Authority guided the thematic analysis and identified themes that described the socio-cultural and contextual reasons that that contribute to perceptions of HIV risk. RESULTS: Sixty Dominican American women ages 57-73 participated in our focus group discussions. Sexual Division of Labour: 1) Economic Dependence; 2) Financial Need and 3) Education and Empowerment. Sexual Division of Power: 4) HIV Risk and 5) Relationship Dynamics. Cathexis: Affective Influences/Social Norms: 6) HIV/AIDS Knowledge and 7) Prevention and Testing. Importantly, participants were concerned about partner fidelity when visiting the Dominican Republic, as the country accounts for the second highest HIV rates in the Caribbean. CONCLUSIONS: Our results confirm previous findings about perceptions of HIV risk and provide additional insight into aging-related aspects of HIV risk for Latino women midlife and older.

A look at the increasing demographic representation within behavioral medicine.

Meeting the behavioral medicine research and clinical needs of an increasingly diverse United States population is an issue of national concern. We examine the trends in the demographic representation of the behavioral medicine scientific workforce through an analysis of the training grants funded by National Institutes of Health for the field of behavioral medicine from 1980 to 2018. We report the topics of these training grants, and we depict the demographic representation of the training leaders. We provide the demographic representation of the trainees, and of the first authors of publications reported within those training grants. Finally, we report the topics addressed in these behavioral medicine publications, to determine if topic diversity increased as the behavioral medicine scientific workforce diversified. Visualizations are presented that tell a story of how we have, and have not, advanced representation within the field of behavioral medicine. Best practices for launching future successful behavioral medicine scientists are then presented, to ensure optimal representation and diversification occurs in our workforce, our science, and our delivery of our clinical care.

Nurse Documentation of Sexual Orientation and Gender Identity in Home Healthcare: A Text Mining Study.

Health disparities have been documented in the lesbian, gay, bisexual, and transgender population, but more research is needed to better understand how to address them. To that end, this observational study examined what is documented about sexual orientation and gender identity in narrative home care nurses' notes in an electronic health record. Lexical text mining approaches were used to examine a total of 862 715 clinical notes from 20 447 unique patients who received services from a large home care agency in Manhattan, New York, and extracted notes were qualitatively reviewed to build a lexicon of terms for use in future research. Forty-two notes, representing 35 unique patients, were identified as containing documentation of the patient's sexual orientation or gender identity. Documentation of sexual orientation or gender identity was relatively infrequent, compared to the estimated frequency of lesbian, gay, bisexual, and transgender people in the US population. Issues related to fragmentary language emerged, and variety in phrasing and word frequency was identified between different types of notes and between providers. This study provides insight into what nurses in home healthcare document about sexual orientation and gender identity and their clinical priorities related to such documentation, and provides a lexicon for use in further research in the home care setting.

The Influence of Daily Stress on Sedentary Behavior: Group and Person (N of 1) Level Results of a 1-Year Observational Study.

OBJECTIVE: The purpose of this study, which used mobile technologies to continuously collect data for 1 year, was to examine the association of psychological stress with objectively measured sedentary behavior in adults at both the group (e.g., nomothetic approach) and individual (e.g., idiographic approach) level. METHODS: Data were collected in an observational study of healthy adults (n = 79) residing in the New York City metro area who were studied for 365 days from 2014 to 2015. Sedentary behavior was objectively measured via accelerometry. A smartphone-based electronic diary was used to assess level of stress ("Overall, how stressful was your day?" 0-10 scale) and sources of stress. RESULTS: The end-of-day stress rating was not associated with total sedentary time (B = -1.34, p = .767) at the group level. When specific sources of stress were evaluated at the group level, argument-related stress was associated with increased sedentariness, whereas running late- and work-related stress were associated with decreased sedentariness. There was a substantial degree of interindividual variability in the relationship of stress with sedentary behavior. Both the level and sources of stress were associated with increased sedentariness for some, decreased sedentariness for others, and had no effect for many (within-person variance p < .001). CONCLUSIONS: These findings suggest that the influence of stress on sedentary behavior varies by source of stress and from person to person. A precision medicine approach may be warranted to target reductions in sedentary time, although further studies are needed to confirm the observed findings in light of study limitations including a small sample size and enrollment of participants from a single, urban metropolitan area.

Visualization of Data Regarding Infections Using Eye Tracking Techniques.

OBJECTIVE: To evaluate ease of use and usefulness for nurses of visualizations of infectious disease transmission in a hospital. DESIGN: An observational study was used to evaluate perceptions of several visualizations of data extracted from electronic health records designed using a participatory approach. Twelve nurses in the master's program in an urban research-intensive nursing school participated in May 2015. METHODS: A convergent parallel mixed method was used to evaluate nurses' perceptions on ease of use and usefulness of five visualization conveying trends in hospital infection transmission applying think-aloud, interview, and eye-tracking techniques. FINDINGS: Subjective data from the interview and think-aloud techniques indicated that participants preferred the traditional line graphs in simple data representation due to their familiarity, clarity, and easiness to read. An objective quantitative measure of eye movement analysis (444,421 gaze events) identified a high degree of participants' attention span in infographics in all three scenarios. All participants responded with the correct answer within 1 min in comprehensive tests. CONCLUSIONS: A user-centric approach was effective in developing and evaluating visualizations for hospital infection transmission. For the visualizations designed by the users, the participants were easily able to comprehend the infection visualizations on both line graphs and infographics for simple visualization. The findings from the objective comprehension test and eye movement and subjective attitudes support the feasibility of integrating user-centric visualization designs into electronic health records, which may inspire clinicians to be mindful of hospital infection transmission. Future studies are needed to investigate visualizations and motivation, and the effectiveness of visualization on infection rate. CLINICAL RELEVANCE: This study designed visualization images using clinical data from electronic health records applying a user-centric approach. The design insights can be applied for visualizing patient data in electronic health records.

Refining a self-assessment of informatics competency scale using Mokken scaling analysis.

Healthcare environments are increasingly implementing health information technology (HIT) and those from various professions must be competent to use HIT in meaningful ways. In addition, HIT has been shown to enable interprofessional approaches to health care. The purpose of this article is to describe the refinement of the Self-Assessment of Nursing Informatics Competencies Scale (SANICS) using analytic techniques based upon item response theory (IRT) and discuss its relevance to interprofessional education and practice. In a sample of 604 nursing students, the 93-item version of SANICS was examined using non-parametric IRT. The iterative modeling procedure included 31 steps comprising: (1) assessing scalability, (2) assessing monotonicity, (3) assessing invariant item ordering, and (4) expert input. SANICS was reduced to an 18-item hierarchical scale with excellent reliability. Fundamental skills for team functioning and shared decision making among team members (e.g. "using monitoring systems appropriately," "describing general systems to support clinical care") had the highest level of difficulty, and "demonstrating basic technology skills" had the lowest difficulty level. Most items reflect informatics competencies relevant to all health professionals. Further, the approaches can be applied to construct a new hierarchical scale or refine an existing scale related to informatics attitudes or competencies for various health professions.

A Data Mining Approach for Examining Predictors of Physical Activity Among Urban Older Adults.

The current study applied innovative data mining techniques to a community survey dataset to develop prediction models for two aspects of physical activity (i.e., active transport and screen time) in a sample of urban, primarily Hispanic, older adults (N=2,514). Main predictors for active transport (accuracy=69.29%, precision=0.67, recall=0.69) were immigrant status, high level of anxiety, having a place for physical activity, and willingness to make time for physical activity. The main predictors for screen time (accuracy=63.13%, precision=0.60, recall=0.63) were willingness to make time for exercise, having a place for exercise, age, and availability of family support to access health information on the Internet. Data mining methods were useful to identify intervention targets and inform design of customized interventions.

Using social network analysis to examine collaborative relationships among PhD and DNP students and faculty in a research-intensive university school of nursing.

The nursing profession has seen a dramatic rise in the number of schools offering both DNP and PhD nursing programs. Information is limited on the impact of this parallel approach in doctoral education on the quality and scope of scholarly interactions or institutional culture.The authors studied collaboration characteristics across the DNP and PhD programs of a research-intensive university school of nursing, before and after programmatic enhancements. An IRB-approved online survey was delivered to faculty and students of both programs at baseline and one year after curricular changes. Response rates were 70% and 74%, respectively. The responses were analyzed by using social network analysis and descriptive statistics to characterize the number and strength of connections between and within student groups, and between students and faculty. At baseline, the flow of communication was centralized primarily through faculty. At Time 2, density of links between students increased and network centralization decreased, suggesting more distributed communication. This nonlinear quantitative approach may be a useful addition to the evaluation strategies for doctoral education initiatives.

Application of Topic Modeling on Artificial Intelligence Studies as a Foundation to Develop Ethical Guidelines in African American Dementia Caregiving.

We applied natural language processing and topic modeling to publicly available abstracts and titles of 263 papers in the scientific literature mentioning AI and demographics (corpus 1 before Covid-19, corpus 2 after Covid-19) extracted from the MEDLINE database. We found exponential growth of AI studies mentioning demographics since the pandemic (Before Covid-19: N= 40 vs. After Covid-19: N= 223) [forecast model equation: ln(Number of Records) = 250.543*ln(Year) + -1904.38, p = 0.0005229]. Topics related to diagnostic imaging, quality of life, Covid, psychology, and smartphone increased during the pandemic, while cancer-related topics decreased. The application of topic modeling to the scientific literature on AI and demographics provides a foundation for the next steps regarding developing guidelines for the ethical use of AI for African American dementia caregivers.

Analyzing Twitter-Based Social Networks of Support Communities for Hispanic and African American Family Caregivers of Persons with Dementia.

We applied social network analysis to compare Hispanic and Black dementia caregiving networks on Twitter that were established as part of a clinical trial from January 12, 2022, to October 31, 2022. We extracted Twitter data from our caregiver support communities (N=1980 followers, 811 enrollees) via the Twitter API and used social network analysis software to compare friend/follower interactions within each Hispanic and Black caregiving network. Analysis of the social networks revealed that enrolled family caregivers without prior social media competency had overall low connectedness compared to both enrolled and non-enrolled caregivers with social media competency, who were more integrated into the communities that developed through the clinical trial, partly due to their ties to external dementia caregiving groups. These observed dynamics will help to guide further social media-based interventions and also support the observation that our recruitment strategies effectively enrolled family caregivers with various levels of social media competency.

Comparing Emotional Valence Scores of Twitter Messages from Human Coding and Machine Learning Algorithms Among Hispanic and African American Family Caregivers of Persons with Dementia.

We compared emotional valence scores as determined via machine learning approaches to human-coded scores of direct messages on Twitter from our 2,301 followers during a Twitter-based clinical trial screening for Hispanic and African American family caregivers of persons with dementia. We manually assigned emotional valence scores to 249 randomly selected direct Twitter messages from our followers (N=2,301), then we applied three machine learning sentiment analysis algorithms to extract emotional valence scores for each message and compared their mean scores to the human coding results. The aggregated mean emotional scores from the natural language processing were slightly positive, while the mean score from human coding as a gold standard was negative. Clusters of strongly negative sentiments were observed in followers' responses to being found non-eligible for the study, indicating a significant need for alternative strategies to provide similar research opportunities to non-eligible family caregivers.

Using Artificial Intelligence to Develop a Lexicon-Based African American Tweet Detection Algorithm to Inform Culturally Sensitive Twitter-Based Social Support Interventions for African American Dementia Caregivers.

We extracted 3,291,101 Tweets using hashtags associated with African American-related discourse (#BlackTwitter, #BlackLivesMatter, #StayWoke) and 1,382,441 Tweets from a control set (general or no hashtags) from September 1, 2019 to December 31, 2019 using the Twitter API. We also extracted a literary historical corpus of 14,692 poems and prose writings by African American authors and 66,083 items authored by others as a control, including poems, plays, short stories, novels and essays, using a cloud-based machine learning platform (Amazon SageMaker) via ProQuest TDM Studio. Lastly, we combined statistics from log likelihood and Fisher's exact tests as well as feature analysis of a batch-trained Naive Bayes classifier to select lexicons of terms most strongly associated with the target or control texts. The resulting Tweet-derived African American lexicon contains 1,734 unigrams, while the control contains 2,266 unigrams. This initial version of a lexicon-based African American Tweet detection algorithm developed using Tweet texts will be useful to inform culturally sensitive Twitter-based social support interventions for African American dementia caregivers.

Comparing Themes Extracted via Topic Modeling and Manual Content Analysis: Korean-Language Discussions of Dementia on Twitter.

We randomly examined Korean-language Tweets mentioning dementia/Alzheimer's disease (n= 12,413) posted from November 28 to December 9, 2020, without limiting geographical locations. We independently applied Latent Dirichlet Allocation (LDA) topic modeling and qualitative content analysis to the texts of the Tweets. We compared the themes extracted by LDA topic modeling to those identified via manual coding methods. A total of 16 themes were detected from manual coding, with inter-rater reliability (Cohen's kappa) of 0.842. The proportions of the most prominent themes were: burdens of family caregiving (48.50%), reports of wandering/missing family members with dementia (18.12%), stigma (13.64%), prevention strategies (5.07%), risk factors (4.91%), healthcare policy (3.26%), and elder abuse/safety issues (1.75%). Seven themes whose contents were similar to themes derived from manual coding were extracted from the LDA topic modeling results (perplexity: -6.39, coherence score: 0.45). Our findings suggest that applying LDA topic modeling can be fairly effective at extracting themes from Korean Twitter discussions, in a manner analogous to qualitative coding, to gain insights regarding caregiving for family members with dementia, and our approach can be applied to other languages.

Comparing Emotional Valence Scores of Twitter Posts from Manual Coding and Machine Learning Algorithms to Gain Insights to Refine Interventions for Family Caregivers of Persons with Dementia.

We randomly extracted Korean-language Tweets mentioning dementia/Alzheimer's disease (n= 12,413) from November 28 to December 9, 2020. We independently applied three machine learning algorithms (Afinn, Syuzhet, and Bing) using natural language processing (NLP) techniques and qualitative manual scoring to assign emotional valence scores to Tweets. We then compared the means and distributions of the four emotional valence scores. Visual examination of the graphs produced indicated that each method exhibited unique patterns. The aggregated mean emotional valence scores from the NLP methods were mostly neutral, vs. slightly negative for manual coding (Afinn 0.029, 95% CI [-0.019, 0.077]; Syuzhet 0.266, [0.236, 0.295]; Bing -0.271, [-0.289, -0.252]; manual coding -1.601, [-1.632, -1.569]). One-way analysis of variance (ANOVA) showed no statistically significant differences among the four means after normalization. These findings suggest that the application of NLP can be fairly effective in extracting emotional valence scores from Korean-language Twitter content to gain insights regarding family caregiving for a person with dementia.

Application of Natural Language Processing to Learn Insights on the Clinician's Lived Experience of Electronic Health Records.

We interviewed six clinicians to learn about their lived experience using electronic health records (EHR, Allscripts users) using a semi-structured interview guide in an academic medical center in New York City from October to November 2016. Each participant interview lasted approximately one to two hours. We applied a clustering algorithm to the interview transcript to detect topics, applying natural language processing (NLP). We visualized eight themes using network diagrams (Louvain modularity 0.70). Novel findings include the need for a concise and organized display and data entry page, the user controlling functions for orders, medications, radiology reports, and missing signals of indentation or filtering functions in the order page and lab results. Application of topic modeling to qualitative interview data provides far-reaching research insights into the clinicians' lived experience of EHR and future optimal EHR design to address human-computer interaction issues in an acute care setting.

Applying Participatory Design Mixed-Methods Approaches to Refine Twitter-Based Social Support Interventions for African Americans and Hispanic Family Caregivers of Persons with Dementia.

We applied mixed-methods to refine our first version of the Twitter message library (English 400, translated into Spanish 400) for African Americans and Hispanic family caregivers for a person with dementia. We conducted a series of expert panels to collect quantitative and qualitative data using surveys and in-depth interviews. Using mixed methods to ensure unbiased results, the panelists first independently scored them (1 message/5 panelist) on a scale of 1 to 4 (1: lowest, 4: highest), followed by in-depth interviews and group discussions. Survey results showed that the average score was 3.47, indicating good to excellent (SD 0.35, ranges from 1.8 to 4). Quantitative surveys and qualitative interviews showed different results in emotional support messages.

Analyzing Topics and Sentiments from Twitter to Gain Insights to Refine Interventions for Family Caregivers of Persons with Alzheimer's Disease and Related Dementias (ADRD) During COVID-19 Pandemic.

We randomly extracted Tweets mentioning dementia/Alzheimer's caregiving-related terms (n= 58,094) from Aug 23, 2019, to Sep 14, 2020, via an API. We applied a clustering algorithm and natural language processing (NLP) to publicly available English Tweets to detect topics and sentiment. We compared emotional valence scores of Tweets from before (through the end of 2019) and after the beginning of the COVID-19 pandemic (2020-). Prevalence of topics related to caregiver emotional distress (e.g., depression, helplessness, stigma, loneliness, elder abuse) and caregiver coping (e.g., resilience, love, reading books) increased, and topics related to late-stage dementia caregiving (e.g., nursing home placement, hospice, palliative care) decreased during the pandemic. The mean emotional valence score significantly decreased from 1.18 (SD 1.57; range -7.1 to 7.9) to 0.86 (SD 1.57; range -5.5 to 6.85) after the advent of COVID-19 (difference -0.32 CI: -0.35, -0.29). The application of topic modeling and sentiment analysis to streaming social media provides a foundation for research insights regarding mental health needs for family caregivers of a person with ADRD during COVID-19 pandemic.

Applying Social Network Analysis to Compare Dementia Caregiving Networks on Twitter in Hispanic and Black Communities.

We applied social network analysis (SNA) on Tweets to compare Hispanic and Black dementia caregiving networks. We randomly extracted Tweets mentioning dementia caregiving and related terms from corpora collected daily via the Twitter API from September 1 to December 31, 2019 (initial corpus: n = 2,742,539 Tweets, random sample n = 549,380 English Tweets, n= 185,684 Spanish Tweets). After removing bot-generated Tweets, we first applied a lexicon-based demographic inference algorithm to automatically identify Tweets likely authored by Black and Hispanic individuals using Python (n = 114,511 English, n = 1,185 Spanish). Then, using ORA, we computed network measures at macro, meso, and micro levels and applied the Louvain clustering algorithm to detect groups within each Hispanic and Black caregiving network. Both networks contained a similar proportion of dyads and triads (Hispanic 88.2%, Black 88.9%), while the Black caregiving network included a slightly larger proportion of isolates (Hispanic 0.8%, Black 4.0%). This study provides useful baseline information on the composition of existing large groups and small groups. In addition, this work provides useful guidance for future recruitment strategies and the design of social support interventions regarding emotional needs for Hispanic and Black dementia caregivers.

Application of persuasive systems design principles to design a self-management application user interface for Hispanic informal dementia caregivers: user preferences and perceptions.

OBJECTIVE: We designed an mHealth application (app) user interface (UI) prototype informed by participatory design sessions, persuasive systems design (PSD) principles, and Lorig and Holman's self-management behavior framework to support self-management activities of Hispanic informal dementia caregivers and assessed their perceptions and preferences regarding features and functions of the app. MATERIALS AND METHODS: Our observational usability study design employed qualitative methods and forced choice preference assessments to identify: (1) the relationship between user preferences for UI features and functions and PSD principles and (2) user preferences for UI design features and functions and app functionality. We evaluated 16 pairs of mHealth app UI prototype designs. Eight paper-based paired designs were used to assess the relationship between PSD principles and caregiver preferences for UI features and functions to support self-management. An Apple iPad WIFI 32GB was used to display another 8 paired designs and assess caregiver preferences for UI functions to support the self-management process. RESULTS: Caregivers preferred an app UI with features and functions that incorporated a greater number of PSD principles and included an infographic to facilitate self-management. Moreover, caregivers preferred a design that did not depend on manual data entry, opting instead for functions such as drop-down list, drag-and-drop, and voice query to prioritize, choose, decide, and search when performing self-management activities. CONCLUSION: Our assessment approaches allowed us to discern which UI features, functions, and designs caregivers preferred. The targeted application of PSD principles in UI designs holds promise for supporting personalized problem identification, goal setting, decision-making, and action planning as strategies for improving caregiver self-management confidence.