Data-driven approaches to care delivery: Actionable informatics in the DoD's primary care behavioral health program.

INTRODUCTION: Transforming administrative health care data into meaningful metrics has been critical to the implementation of the Department of Defense's Primary Care Behavioral Health (PCBH) program. METHODS: Data from clinical encounters with PCBH providers are used to develop metrics of program performance collaboratively. Metrics focus on describing the PCBH program and patients, provider fidelity to the model, and provider performance. These metrics form two key deliverables: a monitoring dashboard for program managers and a training dashboard for expert trainers conducting site visits. RESULTS: Behavioral health consultants (BHCs) conducted nearly 200,000 encounters with more than 100,000 unique patients in fiscal year 2019 at more than 170 locations in 6 countries and 37 states. Administrative data derived from these encounters were used to create a variety of metrics that describe practice and performance at both the provider and program levels. These metrics are delivered through a variety of analytic products to stakeholders who use that information to make data-driven decisions about program direction and provider training. DISCUSSION: We discuss examples of program management decisions and expert trainer actions based on these dashboards, highlighting the benefits of continued collaboration between analysts and program managers. Specifically, excerpts from several dashboards illustrate how penetration and productivity metrics yield specific, tailored action plans to improve care delivery and provider performance. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Genetic Interaction Network Interpretation: A Tidy Data Science Perspective.

As practitioners, we aim to provide a consolidated introduction of tidy data science along with routine packages for relational data representation and interpretation, with the focus on analytics related to human genetic interactions. We describe three showcases (also made available at https://23verse.github.io/gini ), all done so via the R one-liner, in this chapter defined as a sequential pipeline of elementary functions chained together achieving a complex task. We guide the readers through step-by-step instructions on (case 1) performing network module analysis of genetic interactions, followed by visualization and interpretation; (case 2) implementing a practical strategy of how to identify and interpret tissue-specific genetic interactions; and (case 3) carrying out interaction-based tissue clustering and differential interaction analysis. All showcases demonstrate simplistic beauty and efficient nature of this analytics. We anticipate that mastering a dozen of one-liners to efficiently interpret genetic interactions is very timely now; opportunities for computational translational research are arising for data scientists to harness therapeutic potential of human genetic interaction data that are ever-increasingly available.

Systematic review of education and practical guidance on regression modeling for medical researchers who lack a strong statistical background: Study protocol.

In the last decades, statistical methodology has developed rapidly, in particular in the field of regression modeling. Multivariable regression models are applied in almost all medical research projects. Therefore, the potential impact of statistical misconceptions within this field can be enormous Indeed, the current theoretical statistical knowledge is not always adequately transferred to the current practice in medical statistics. Some medical journals have identified this problem and published isolated statistical articles and even whole series thereof. In this systematic review, we aim to assess the current level of education on regression modeling that is provided to medical researchers via series of statistical articles published in medical journals. The present manuscript is a protocol for a systematic review that aims to assess which aspects of regression modeling are covered by statistical series published in medical journals that intend to train and guide applied medical researchers with limited statistical knowledge. Statistical paper series cannot easily be summarized and identified by common keywords in an electronic search engine like Scopus. We therefore identified series by a systematic request to statistical experts who are part or related to the STRATOS Initiative (STRengthening Analytical Thinking for Observational Studies). Within each identified article, two raters will independently check the content of the articles with respect to a predefined list of key aspects related to regression modeling. The content analysis of the topic-relevant articles will be performed using a predefined report form to assess the content as objectively as possible. Any disputes will be resolved by a third reviewer. Summary analyses will identify potential methodological gaps and misconceptions that may have an important impact on the quality of analyses in medical research. This review will thus provide a basis for future guidance papers and tutorials in the field of regression modeling which will enable medical researchers 1) to interpret publications in a correct way, 2) to perform basic statistical analyses in a correct way and 3) to identify situations when the help of a statistical expert is required.

Artificial Intelligence in mental health and the biases of language based models.

BACKGROUND: The rapid integration of Artificial Intelligence (AI) into the healthcare field has occurred with little communication between computer scientists and doctors. The impact of AI on health outcomes and inequalities calls for health professionals and data scientists to make a collaborative effort to ensure historic health disparities are not encoded into the future. We present a study that evaluates bias in existing Natural Language Processing (NLP) models used in psychiatry and discuss how these biases may widen health inequalities. Our approach systematically evaluates each stage of model development to explore how biases arise from a clinical, data science and linguistic perspective. DESIGN/METHODS: A literature review of the uses of NLP in mental health was carried out across multiple disciplinary databases with defined Mesh terms and keywords. Our primary analysis evaluated biases within 'GloVe' and 'Word2Vec' word embeddings. Euclidean distances were measured to assess relationships between psychiatric terms and demographic labels, and vector similarity functions were used to solve analogy questions relating to mental health. RESULTS: Our primary analysis of mental health terminology in GloVe and Word2Vec embeddings demonstrated significant biases with respect to religion, race, gender, nationality, sexuality and age. Our literature review returned 52 papers, of which none addressed all the areas of possible bias that we identify in model development. In addition, only one article existed on more than one research database, demonstrating the isolation of research within disciplinary silos and inhibiting cross-disciplinary collaboration or communication. CONCLUSION: Our findings are relevant to professionals who wish to minimize the health inequalities that may arise as a result of AI and data-driven algorithms. We offer primary research identifying biases within these technologies and provide recommendations for avoiding these harms in the future.

A COVID-19 e o papel dos sistemas de informação e das tecnologias na atenção primária / COVID-19 and the role of information systems and technologies at the first level of care / COVID-19 y el rol de los sistemas de información y las tecnologías en el primer nivel de atención

Atenção centrada na resposta à COVID-19: identificar, informar, conter, manejar e encaminhar. Os sistemas de informação em saúde ­ por meio do acesso oportuno a dados devidamente desagregados, a correta integração dos sistemas nacionais e locais, a saúde digital e o uso das tecnologias da informação (TIC) de uso frequente ­ facilitam a identificação eficaz, informação e análise de casos e contatos; a busca e detecção de casos em tempo hábil; e a identificação e seguimento da população de risco, dos casos e de seus contatos. A contenção é fortalecida com as plataformas de seguimento e monitoramento de casos, contatos, quarentena e isolamento social. Por sua vez, esses sistemas possibilitam a difusão maciça a toda a sociedade dos comunicados sobre medidas preventivas. As plataformas de teleconsulta, monitoramento remoto de pacientes e comunicação a distância permitem à atenção primária manejar a assistência médica e facilitam o seguimento domiciliar das pessoas com COVID-19. Esses mesmos mecanismos, integrados aos prontuários eletrônicos e aos sistemas locais e nacionais de informação, permitem e facilitam as referências, em âmbito hospitalar, dos pacientes com sinais e sintomas graves ou com fatores de risco.

Atención centrada en la respuesta a la COVID-19: identificar, reportar, contener, manejar y referir. Los sistemas de información para la salud ­a través del acceso oportuno a datos correctamente desagregados, la correcta integración de los sistemas nacionales y locales, la salud digital y la utilización de las tecnologías de la información (TIC) de uso frecuente­ facilitan la identificación eficaz, el reporte y análisis de casos y contactos; la búsqueda y detección tempranas de casos; y la identificación y el seguimiento de la población de riesgo, los casos y sus contactos. La contención se ve fortalecida con las plataformas de seguimiento y monitoreo de casos, contactos, cuarentena y aislamiento social. Estos sistemas permiten a su vez la difusión masiva a toda la sociedad de las comunicaciones sobre medidas preventivas. Las plataformas de teleconsulta, monitoreo remoto de pacientes y comunicación a distancia permiten al primer nivel de atención el manejo de la asistencia médica y facilitan el seguimiento domiciliario de las personas con COVID-19. Estos mismos mecanismos, integrados con los registros electrónicos de salud y los sistemas locales y nacionales de información, permiten y facilitan las referencias al nivel hospitalario de los pacientes con signos y síntomas graves o factores de riesgo.

Care centered on the response to COVID-19: Identify, report, contain, manage, and refer. Information systems for health­through timely access to correctly disaggregated data, proper integration of national and local systems, digital health, and the application of widely used information and communication technologies (ICTs)­facilitate the effective identification, reporting, and analysis of cases and contacts; early search for and detection of cases; and identification and monitoring of at-risk populations, cases, and contacts. Containment is strengthened through platforms for follow-up and monitoring of cases, contacts, quarantine, and social isolation. These systems, in turn, enable mass dissemination of information on preventive measures to all of society. Platforms for telemedicine visits, remote monitoring of patients, and remote communication enable health workers at the first level of care to manage medical care and facilitate home monitoring of people with COVID-19. These same mechanisms, together with electronic health records and local and national information systems, facilitate hospital referrals of patients with severe signs and symptoms or risk factors.

Advancing injury and violence prevention through data science.

INTRODUCTION: The volume of new data that is created each year relevant to injury and violence prevention continues to grow. Furthermore, the variety and complexity of the types of useful data has also progressed beyond traditional, structured data. In order to more effectively advance injury research and prevention efforts, the adoption of data science tools, methods, and techniques, such as natural language processing and machine learning, by the field of injury and violence prevention is imperative. METHOD: The Centers for Disease Control and Prevention's (CDC) National Center for Injury Prevention and Control has conducted numerous data science pilot projects and recently developed a Data Science Strategy. This strategy includes goals on expanding the availability of more timely data systems, improving rapid identification of health threats and responses, increasing access to accurate health information and preventing misinformation, improving data linkages, expanding data visualization efforts, and increasing efficiency of analytic and scientific processes for injury and violence, among others. RESULTS: To achieve these goals, CDC is expanding its data science capacity in the areas of internal workforce, partnerships, and information technology infrastructure. Practical Application: These efforts will expand the use of data science approaches to improve how CDC and the field address ongoing injury and violence priorities and challenges.

Por que a desagregação de dados é essencial durante pandemias / Por qué es importante la desagregación de datos durante una pandemia / Why Data Disaggregation is key during a pandemic

O que é desagregação de dados? O termo dados desagregados se refere à separação das informações coletadas em unidades menores para revelar tendências e padrões subjacentes. Os dados compilados podem vir de diversas fontes (setores público e privado e organizações nacionais e internacionais) e ter diversas variáveis, ou "dimensões". Para melhor entender uma situação, os dados são agrupados por dimensão, como idade, sexo, área geográfica, escolaridade, etnia ou outras variáveis socioeconômicas. Por que precisamos de dados desagregados durante uma pandemia? Quando ocorre uma pandemia, uma resposta adequada e eficiente requer a identificação e caracterização dos fatores que desaceleram ou aceleram a transmissão e das populações mais vulneráveis. Dados desagregados de alta qualidade, acessíveis, seguros, atuais, abertos e confiáveis são fundamentais a fim de gerar informações valiosas para a tomada de decisões em tempo real. Por exemplo, para determiner se uma intervenção (como a autotriagem em massa) é efetiva, precisamos saber a proporção da população que foi testada. Isso pode exigir análise por idade, área geográfica e/ou outras variáveis socioeconômicas...

¿Qué significa la desagregación de datos? La desagregación de datos se refiere a la separación de la información recabada en unidades más pequeñas para dilucidar las tendencias y los patrones subyacentes. Los datos recabados pueden provenir de múltiples fuentes (los sectores público y privado, y organizaciones nacionales e internacionales) y tener múltiples variables o "dimensiones". Para mejorar la comprensión de una situación, los datos se agrupan por dimensión, como edad, sexo, zona geográfica, educación, etnicidad u otras variables socioeconómicas. ¿Por qué necesitamos datos desagregados durante una pandemia? Cuando hay una pandemia, una respuesta apropiada y eficaz requiere que determinemos y caractericemos los factores que enlentecen o aceleran la transmisión y los grupos poblacionales que son más vulnerables. Los datos desagregados de alta calidad, accesibles, fiables, oportunos, abiertos y fidedignos son fundamentales para generar información valiosa para la toma de decisiones en tiempo real. Por ejemplo, a fin de determinar si una intervención (como el autotamizaje masivo) es eficaz, tenemos que saber qué proporción de la población ha sido objeto de la prueba. Esto puede requerir un análisis por edad, zona geográfica u otras variables socioeconómicas...

Data Disaggregation is the separation of compiled information into smaller units to elucidate underlying trends and patterns. High quality, accessible, trusted, timely, open, and reliable disaggregated data is critical to generating valuable information for decision-making in real time.

Data-driven analysis using multiple self-report questionnaires to identify college students at high risk of depressive disorder.

Depression diagnosis is one of the most important issues in psychiatry. Depression is a complicated mental illness that varies in symptoms and requires patient cooperation. In the present study, we demonstrated a novel data-driven attempt to diagnose depressive disorder based on clinical questionnaires. It includes deep learning, multi-modal representation, and interpretability to overcome the limitations of the data-driven approach in clinical application. We implemented a shared representation model between three different questionnaire forms to represent questionnaire responses in the same latent space. Based on this, we proposed two data-driven diagnostic methods; unsupervised and semi-supervised. We compared them with a cut-off screening method, which is a traditional diagnostic method for depression. The unsupervised method considered more items, relative to the screening method, but showed lower performance because it maximized the difference between groups. In contrast, the semi-supervised method adjusted for bias using information from the screening method and showed higher performance. In addition, we provided the interpretation of diagnosis and statistical analysis of information using local interpretable model-agnostic explanations and ordinal logistic regression. The proposed data-driven framework demonstrated the feasibility of analyzing depressed patients with items directly or indirectly related to depression.

Data science and machine learning in anesthesiology.

Machine learning (ML) is revolutionizing anesthesiology research. Unlike classical research methods that are largely inference-based, ML is geared more towards making accurate predictions. ML is a field of artificial intelligence concerned with developing algorithms and models to perform prediction tasks in the absence of explicit instructions. Most ML applications, despite being highly variable in the topics that they deal with, generally follow a common workflow. For classification tasks, a researcher typically tests various ML models and compares the predictive performance with the reference logistic regression model. The main advantage of ML lies in its ability to deal with many features with complex interactions and its specific focus on maximizing predictive performance. However, emphasis on data-driven prediction can sometimes neglect mechanistic understanding. This article mainly focuses on the application of supervised ML to electronic health record (EHR) data. The main limitation of EHR-based studies is in the difficulty of establishing causal relationships. However, the associated low cost and rich information content provide great potential to uncover hitherto unknown correlations. In this review, the basic concepts of ML are introduced along with important terms that any ML researcher should know. Practical tips regarding the choice of software and computing devices are also provided. Towards the end, several examples of successful ML applications in anesthesiology are discussed. The goal of this article is to provide a basic roadmap to novice ML researchers working in the field of anesthesiology.

Portfolio Analysis of Research Grants in Data Science Funded by the National Heart, Lung, and Blood Institute.

Leveraging emerging opportunities in data science to open new frontiers in heart, lung, blood, and sleep research is one of the major strategic objectives of the National Heart, Lung, and Blood Institute (NHLBI), one of the 27 Institutes/Centers within the National Institutes of Health (NIH). To assess NHLBI's recent funding of research grants in data science and to identify its relative areas of focus within data science, a portfolio analysis from fiscal year 2008 to fiscal year 2017 was performed. In this portfolio analysis, an efficient and reliable methodology was used to identify data science research grants by utilizing several NIH databases and search technologies (iSearch, Query View Reporting system, and IN-SPIRE [Pacific Northwest National Laboratory, Richland, WA]). Six hundred thirty data science-focused extramural research grants supported by NHLBI were identified using keyword searches based primarily on NIH's working definitions of bioinformatics and computational biology. Further analysis characterized the distribution of these grants among the heart, lung, blood, and sleep disease areas as well as the subtypes of data science projects funded by NHLBI. Information was also collected for data science research grants funded by other NIH institutes/centers using the same search and analysis methodology. The funding comparison among different NIH institutes/centers highlighted relative data science areas of emphasis and further identified opportunities for potential data science areas in which NHLBI could foster research advances.

Dynamic, data-driven typologies of long-term smoking, cessation, and their correlates: Findings from the International Tobacco Control (ITC) Netherlands Survey.

RATIONALE: Efforts towards tobacco control are numerous, but relapse rates for smoking cessations remain high. Behavioral changes necessary for continuous cessation appear complex, variable and subject to social, biological, psychological and environmental determinants. Currently, most cessation studies concentrate on short-to midterm behavioral changes. Besides, they use fixed typologies, thereby failing to capture the temporal changes in smoking/cessation behaviors, and its determinants. OBJECTIVE: To obtain long-term, data-driven longitudinal patterns or profiles of smoking, cessation, and related determinants in a cohort of adult smokers, and to investigate their dynamic links. METHODS: The dataset originated from the International Tobacco Control (ITC) Netherlands Project, waves 2008 to 2016. Temporal dynamics of smoking/cessation, psychosocial constructs, and time-varying determinants of smoking were extracted with Group-Based Trajectory Modeling technique. Their associations were investigated via multiple regression models. RESULTS: Substantial heterogeneity of smoking and cessation behaviors was unveiled. Most respondents were classified as persistent smokers, albeit with distinct levels of consumption. For a minority, cessation could be sustained between 1 and 8 years, while others showed relapsing or fluctuating smoking behavior. Links between smoking/cessation trajectories with those of psychosocial and sociodemographic variables were diverse. Notably, changes in two variables were aligned to behavioral changes towards cessation: decreasing number of smoking peers and attaining a higher self-perceived control. CONCLUSION: The unveiled heterogeneity of smoking behavior over time and the varied cross-dependencies between smoking data-driven typologies and those of underlying risk factors underscore the need of individually tailored approaches for motivational quitting.

Reproducible big data science: A case study in continuous FAIRness.

Big biomedical data create exciting opportunities for discovery, but make it difficult to capture analyses and outputs in forms that are findable, accessible, interoperable, and reusable (FAIR). In response, we describe tools that make it easy to capture, and assign identifiers to, data and code throughout the data lifecycle. We illustrate the use of these tools via a case study involving a multi-step analysis that creates an atlas of putative transcription factor binding sites from terabytes of ENCODE DNase I hypersensitive sites sequencing data. We show how the tools automate routine but complex tasks, capture analysis algorithms in understandable and reusable forms, and harness fast networks and powerful cloud computers to process data rapidly, all without sacrificing usability or reproducibility-thus ensuring that big data are not hard-to-(re)use data. We evaluate our approach via a user study, and show that 91% of participants were able to replicate a complex analysis involving considerable data volumes.

Healthcare data scientist qualifications, skills, and job focus: a content analysis of job postings.

OBJECTIVE: Growth in big data and its potential impact on the healthcare industry have driven the need for more data scientists. In health care, big data can be used to improve care quality, increase efficiency, lower costs, and drive innovation. Given the importance of data scientists to U.S. healthcare organizations, I examine the qualifications and skills these organizations require for data scientist positions and the specific focus of their work. MATERIALS AND METHODS: A content analysis of U.S. healthcare data scientist job postings was conducted using an inductive approach to capture and categorize core information about each posting and a deductive approach to evaluate skills required. Profiles were generated for 4 job focus areas. RESULTS: There is a spectrum of healthcare data scientist positions that varies based on hiring organization type, job level, and job focus area. The focus of these positions ranged from performance improvement to innovation and product development with some positions more broadly defined to address organizational-specific needs. Based on the job posting sample, the primary skills these organizations required were statistics, R, machine learning, storytelling, and Python. CONCLUSIONS: These results may be useful to organizations as they deepen our understanding of the qualifications and skills required for data scientist positions and may aid organizations in identifying skills and knowledge areas that have been overlooked in position postings.

Data analytics of call log data to identify caller behaviour patterns from a mental health and well-being helpline.

This work presents an analysis of 3.5 million calls made to a mental health and well-being helpline, seeking to answer the question, what different groups of callers can be characterised by specific usage patterns? Calls were extracted from a telephony informatics system. Each call was logged with a date, time, duration and a unique identifier allowing for repeat caller analysis. We utilized data mining techniques to reveal new insights into help-seeking behaviours. Analysis was carried out using unsupervised machine learning (K-means clustering) to discover the types of callers, and Fourier transform was used to ascertain periodicity in calls. Callers can be clustered into five or six caller groups that offer a meaningful interpretation. Cluster groups are stable and re-emerge regardless of which year is considered. The volume of calls exhibits strong repetitive intra-day and intra-week patterns. Intra-month repetitions are absent. This work provides new data-driven findings to model the type and behaviour of callers seeking mental health support. It offers insights for computer-mediated and telephony-based helpline management.

Características asociadas al trauma dentoalveolar en incisivos superiores / Characteristics associated with dentoalveolar trauma in upper incisor teeth

Introducción: los traumatismos de los dientes anteriores son eventos que suceden con frecuencia, por su gran impacto social y psicológico deben ser considerados como un tema de trascendental importancia, lo cual motivó realizar el estudio. Objetivo: identificar características asociadas con las fracturas dentarias en incisivos superiores permanentes en estudiantes que asistieron al servicio estomatológico de la escuela primaria Lazo de la Vega del municipio Marianao en el año 2009. Método: Se realizó un estudio analítico, en el que se incluyeron los 235 escolares matriculados en la escuela del mismo nombre, entre 7 y 12 años de edad que asistieron a consulta. Se estudiaron las variables edad, sexo, presencia de hábitos bucales deformantes, tipo de fractura dentaria e incisivo traumatizado. Se calculó la frecuencia absoluta, el riesgo a través de la tasa por 100, el riesgo relativo (RR) a través de la razón entre riesgos y se aplicó el estadígrafo X² de Pearson para la asociación entre variables. Resultados: el grupo de edad más afectado es el de 10 a 12 años con un 23,4 por ciento, con predominio del sexo masculino dado por un RR de 2,47 veces más que el femenino. La fractura no complicada de corona aparece con mayor frecuencia con un 61,8 por ciento, el diente mayormente afectado es el incisivo central superior izquierdo con un 51,4 por ciento. La diferencia de riesgo entre los grupos de edades no son significativas, aunque en relación con el sexo las diferencias son muy significativas. Conclusiones: Hubo franco predominio de las fracturas no complicadas y de los incisivos centrales superiores, en particular el izquierdo(AU)

Introduction: traumas of front teeth are frequent events which, because of their social and psychological impact, should be considered a topic of paramount importance. This fact motivated the authors to conduct the study. Objectives: to identify the characteristics associated with dental fractures in permanent upper incisors in students who went to the Stomatology Service of Lazo de la Vega Primary School in Marianao Municipality in the year 2009. Method: an analytical study was conducted in 235 students from 7 to 12 years of age who were enrolled in the school mentioned above who presented to the consultation. The variables studied were: age, sex, presence of deforming buccal habits, type of dental fracture and traumatized incisor teeth. Absolute frequency, the rate/100 risk, and the relative risk (RR) through the ratio among risks were calculated and the Pearson's X2 statistics was applied for the association among variables. Results: the most affected age group was from 10 to 12 years with a 23.4 por ciento, predominating the male sex due to a RR 2.47 times higher than the female sex. The non complicated crown fracture was the most frequent one with a 61.8 por ciento, the most affected tooth is the left upper central incisive, with a 51.4 por ciento. The risk difference between the age groups was not significant, although in relation to the sex, the differences are highly significant. Conclusions: there was a marked predominance of the uncomplicated fractures and the upper central incisors, particularly the left ones(AU)