Improving Prediction of Survival for Extremely Premature Infants Born at 23 to 29 Weeks Gestational Age in the Neonatal Intensive Care Unit: Development and Evaluation of Machine Learning Models.

BACKGROUND: Infants born at extremely preterm gestational ages are typically admitted to the neonatal intensive care unit (NICU) after initial resuscitation. The subsequent hospital course can be highly variable, and despite counseling aided by available risk calculators, there are significant challenges with shared decision-making regarding life support and transition to end-of-life care. Improving predictive models can help providers and families navigate these unique challenges. OBJECTIVE: Machine learning methods have previously demonstrated added predictive value for determining intensive care unit outcomes, and their use allows consideration of a greater number of factors that potentially influence newborn outcomes, such as maternal characteristics. Machine learning-based models were analyzed for their ability to predict the survival of extremely preterm neonates at initial admission. METHODS: Maternal and newborn information was extracted from the health records of infants born between 23 and 29 weeks of gestation in the Medical Information Mart for Intensive Care III (MIMIC-III) critical care database. Applicable machine learning models predicting survival during the initial NICU admission were developed and compared. The same type of model was also examined using only features that would be available prepartum for the purpose of survival prediction prior to an anticipated preterm birth. Features most correlated with the predicted outcome were determined when possible for each model. RESULTS: Of included patients, 37 of 459 (8.1%) expired. The resulting random forest model showed higher predictive performance than the frequently used Score for Neonatal Acute Physiology With Perinatal Extension II (SNAPPE-II) NICU model when considering extremely preterm infants of very low birth weight. Several other machine learning models were found to have good performance but did not show a statistically significant difference from previously available models in this study. Feature importance varied by model, and those of greater importance included gestational age; birth weight; initial oxygenation level; elements of the APGAR (appearance, pulse, grimace, activity, and respiration) score; and amount of blood pressure support. Important prepartum features also included maternal age, steroid administration, and the presence of pregnancy complications. CONCLUSIONS: Machine learning methods have the potential to provide robust prediction of survival in the context of extremely preterm births and allow for consideration of additional factors such as maternal clinical and socioeconomic information. Evaluation of larger, more diverse data sets may provide additional clarity on comparative performance.

Chemical Entity Normalization for Successful Translational Development of Alzheimer's Disease and Dementia Therapeutics.

Background: Identifying chemical mentions within the Alzheimer's and dementia literature can provide a powerful tool to further therapeutic research. Leveraging the Chemical Entities of Biological Interest (ChEBI) ontology, which is rich in hierarchical and other relationship types, for entity normalization can provide an advantage for future downstream applications. We provide a reproducible hybrid approach that combines an ontology-enhanced PubMedBERT model for disambiguation with a dictionary-based method for candidate selection. Results: There were 56,553 chemical mentions in the titles of 44,812 unique PubMed article abstracts. Based on our gold standard, our method of disambiguation improved entity normalization by 25.3 percentage points compared to using only the dictionary-based approach with fuzzy-string matching for disambiguation. For our Alzheimer's and dementia cohort, we were able to add 47.1% more potential mappings between MeSH and ChEBI when compared to BioPortal. Conclusion: Use of natural language models like PubMedBERT and resources such as ChEBI and PubChem provide a beneficial way to link entity mentions to ontology terms, while further supporting downstream tasks like filtering ChEBI mentions based on roles and assertions to find beneficial therapies for Alzheimer's and dementia.

Modeling pandemic to endemic patterns of SARS-CoV-2 transmission using parameters estimated from animal model data.

The contours of endemic coronaviral disease in humans and other animals are shaped by the tendency of coronaviruses to generate new variants superimposed upon nonsterilizing immunity. Consequently, patterns of coronaviral reinfection in animals can inform the emerging endemic state of the SARS-CoV-2 pandemic. We generated controlled reinfection data after high and low risk natural exposure or heterologous vaccination to sialodacryoadenitis virus (SDAV) in rats. Using deterministic compartmental models, we utilized in vivo estimates from these experiments to model the combined effects of variable transmission rates, variable duration of immunity, successive waves of variants, and vaccination on patterns of viral transmission. Using rat experiment-derived estimates, an endemic state achieved by natural infection alone occurred after a median of 724 days with approximately 41.3% of the population susceptible to reinfection. After accounting for translationally altered parameters between rat-derived data and human SARS-CoV-2 transmission, and after introducing vaccination, we arrived at a median time to endemic stability of 1437 (IQR = 749.25) days with a median 15.4% of the population remaining susceptible. We extended the models to introduce successive variants with increasing transmissibility and included the effect of varying duration of immunity. As seen with endemic coronaviral infections in other animals, transmission states are altered by introduction of new variants, even with vaccination. However, vaccination combined with natural immunity maintains a lower prevalence of infection than natural infection alone and provides greater resilience against the effects of transmissible variants.

Leukotriene inhibitors with dexamethasone show promise in the prevention of death in COVID-19 patients with low oxygen saturations.

Introduction: COVID-19 is a major health threat around the world causing hundreds of millions of infections and millions of deaths. There is a pressing global need for effective therapies. We hypothesized that leukotriene inhibitors (LTIs), that have been shown to lower IL6 and IL8 levels, may have a protective effect in patients with COVID-19. Methods: In this retrospective controlled cohort study, we compared death rates in COVID-19 patients who were taking a LTI with those who were not taking an LTI. We used the Department of Veterans Affairs (VA) Corporate Data Warehouse (CDW) to create a cohort of COVID-19-positive patients and tracked their use of LTIs between November 1, 2019 and November 11, 2021. Results: Of the 1,677,595 cohort of patients tested for COVID-19, 189,195 patients tested positive for COVID-19. Forty thousand seven hundred one were admitted. 38,184 had an oxygen requirement and 1214 were taking an LTI. The use of dexamethasone plus a LTI in hospital showed a survival advantage of 13.5% (CI: 0.23%-26.7%; p < 0.01) in patients presenting with a minimal O2Sat of 50% or less. For patients with an O2Sat of <60 and <50% if they were on LTIs as outpatients, continuing the LTI led to a 14.4% and 22.25 survival advantage if they were continued on the medication as inpatients. Conclusions: When combined dexamethasone and LTIs provided a mortality benefit in COVID-19 patients presenting with an O2 saturations <50%. The LTI cohort had lower markers of inflammation and cytokine storm.

Identification of patient characteristics associated with survival benefit from metformin treatment in patients with stage I non-small cell lung cancer.

BACKGROUND: Non-small cell lung cancer (NSCLC) continues to be a major cause of cancer deaths. Previous investigation has suggested that metformin use can contribute to improved outcomes in NSCLC patients. However, this association is not uniform in all analyzed cohorts, implying that patient characteristics might lead to disparate results. Identification of patient characteristics that affect the association of metformin use with clinical benefit might clarify the drug's effect on lung cancer outcomes and lead to more rational design of clinical trials of metformin's utility as an intervention. In this study, we examined the association of metformin use with long-term mortality benefit in patients with NSCLC and the possible modulation of this benefit by body mass index (BMI) and smoking status, controlling for other clinical covariates. METHODS: This was a retrospective cohort study in which we analyzed data from the Veterans Affairs (VA) Tumor Registry in the United States. Data from all patients with stage I NSCLC from 2000 to 2016 were extracted from a national database, the Corporate Data Warehouse that captures data from all patients, primarily male, who underwent treatment through the VA health system in the United States. Metformin use was measured according to metformin prescriptions dispensed to patients in the VA health system. The association of metformin use with overall survival (OS) after diagnosis of stage I NSCLC was examined. Patients were further stratified according to BMI and smoking status (previous vs current) to examine the association of metformin use with OS across these strata. RESULTS: Metformin use was associated with improved survival in patients with stage I NSCLC (average hazard ratio, 0.82; P < .001). An interaction between the effect of metformin use and BMI on OS was observed (χ2 = 3268.42; P < .001) with a greater benefit of metformin use observed in patients as BMI increased. Similarly, an interaction between smoking status and metformin use on OS was also observed (χ2 = 2997.05; P < .001) with a greater benefit of metformin use observed in previous smokers compared with current smokers. CONCLUSIONS: In this large retrospective study, we showed that a survival benefit is enjoyed by users of metformin in a robust stage I NSCLC patient population treated in the VA health system. Metformin use was associated with an 18% improved OS. This association was stronger in patients with a higher BMI and in previous smokers. These observations deserve further mechanistic study and can help rational design of clinical trials with metformin in patients with lung cancer.

Differences in mRNA-1273 (Moderna) and BNT162b2 (Pfizer-BioNTech) SARS-CoV-2 vaccine immunogenicity among patients undergoing dialysis.

BACKGROUND: Differences in immunogenicity between mRNA SARS-CoV-2 vaccines have not been well characterized in patients undergoing dialysis. We compared the serologic response in patients undergoing maintenance hemodialysis after vaccination against SARS-CoV-2 with BNT162b2 (Pfizer-BioNTech) and mRNA-1273 (Moderna). METHODS: We conducted a prospective observational cohort study at 2 academic centres in Toronto, Canada, from Feb. 2, 2021, to July 20, 2021, which included 129 and 95 patients who received the BNT162b2 and mRNA-1273 SARS-CoV-2 vaccines, respectively. We measured SARS-CoV-2 immunoglobulin G antibodies to the spike protein (anti-spike), receptor binding domain (anti-RBD) and nucleocapsid protein (anti-NP) at 6-7 and 12 weeks after the second dose of vaccine and compared those levels with the median convalescent serum antibody levels from 211 controls who were previously infected with SARS-CoV-2. RESULTS: At 6-7 weeks after 2-dose vaccination, we found that 51 of 70 patients (73%) who received BNT162b2 and 83 of 87 (95%) who received mRNA-1273 attained convalescent levels of anti-spike antibody (p < 0.001). In those who received BNT162b2, 35 of 70 (50%) reached the convalescent level for anti-RBD compared with 69 of 87 (79%) who received mRNA-1273 (p < 0.001). At 12 weeks after the second dose, anti-spike and anti-RBD levels were significantly lower in patients who received BNT162b2 than in those who received mRNA-1273. For anti-spike, 70 of 122 patients (57.4%) who received BNT162b2 maintained the convalescent level versus 68 of 71 (96%) of those who received mRNA-1273 (p < 0.001). For anti-RBD, 47 of 122 patients (38.5%) who received BNT162b2 maintained the anti-RBD convalescent level versus 45 of 71 (63%) of those who received mRNA-1273 (p = 0.002). INTERPRETATION: In patients undergoing hemodialysis, mRNA-1273 elicited a stronger humoral response than BNT162b2. Given the rapid decline in immunogenicity at 12 weeks in patients who received BNT162b2, a third dose is recommended in patients undergoing dialysis as a primary series, similar to recommendations for other vulnerable populations.

Efficacy of Rituximab in Treatment-Resistant Focal Segmental Glomerulosclerosis With Elevated Soluble Urokinase-Type Plasminogen Activator Receptor and Activation of Podocyte ß3 Integrin.

INTRODUCTION: Severe, nonresponsive, primary focal segmental glomerular sclerosis (FSGS) can progress to end-stage kidney disease (ESKD) in <5 years. Soluble urokinase-type plasminogen activator receptor (suPAR) may contribute to podocyte effacement by activating podocyte ß3 integrin. It has been reported as a potential permeability factor and biomarker for primary FSGS. Rituximab was found to have efficacy in case reports and small series. Whether rituximab is efficacious in patients with treatment-resistant FSGS in the context of high suPAR levels and evidence of podocyte B3 integrin activation remains unknown. METHODS: In this nonblinded, open-label pilot study, the safety and efficacy of rituximab were evaluated in treatment-resistant adult patients with primary FSGS and a suPAR level > 3500 pg/ml with evidence of ß3 integrin activation. Rituximab (1 g) was given on days 1 and 15. The primary outcome was proteinuria at 12 months. RESULTS: Only 13 of 38 screened patients qualified for the study, of whom 9 consented to participate. The baseline proteinuria and glomerular filtration rate (GFR) levels were 7.70 ± 4.61 g/d and 67 ± 38 ml/min, respectively. A transient response at 6 months was noted in 2 patients without a parallel change in suPAR level. At 12 months, there was no statistically significant improvement in proteinuria level with all participants remaining nephrotic (7.27 ± 7.30 g/d). GFR level marginally declined to 60 ± 38 ml/min with one patient progressing to ESKD. There were 2 serious infections, an infusion-related reaction and leucopenia attributed to rituximab. CONCLUSION: Rituximab was ineffective when administered to adult patients with treatment-resistant primary FSGS with a high suPAR and evidence of podocyte activation.

Using Artificial Intelligence With Natural Language Processing to Combine Electronic Health Record's Structured and Free Text Data to Identify Nonvalvular Atrial Fibrillation to Decrease Strokes and Death: Evaluation and Case-Control Study.

BACKGROUND: Nonvalvular atrial fibrillation (NVAF) affects almost 6 million Americans and is a major contributor to stroke but is significantly undiagnosed and undertreated despite explicit guidelines for oral anticoagulation. OBJECTIVE: The aim of this study is to investigate whether the use of semisupervised natural language processing (NLP) of electronic health record's (EHR) free-text information combined with structured EHR data improves NVAF discovery and treatment and perhaps offers a method to prevent thousands of deaths and save billions of dollars. METHODS: We abstracted 96,681 participants from the University of Buffalo faculty practice's EHR. NLP was used to index the notes and compare the ability to identify NVAF, congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke or transient ischemic attack, vascular disease, age 65 to 74 years, sex category (CHA2DS2-VASc), and Hypertension, Abnormal liver/renal function, Stroke history, Bleeding history or predisposition, Labile INR, Elderly, Drug/alcohol usage (HAS-BLED) scores using unstructured data (International Classification of Diseases codes) versus structured and unstructured data from clinical notes. In addition, we analyzed data from 63,296,120 participants in the Optum and Truven databases to determine the NVAF frequency, rates of CHA2DS2­VASc ≥2, and no contraindications to oral anticoagulants, rates of stroke and death in the untreated population, and first year's costs after stroke. RESULTS: The structured-plus-unstructured method would have identified 3,976,056 additional true NVAF cases (P<.001) and improved sensitivity for CHA2DS2-VASc and HAS-BLED scores compared with the structured data alone (P=.002 and P<.001, respectively), causing a 32.1% improvement. For the United States, this method would prevent an estimated 176,537 strokes, save 10,575 lives, and save >US $13.5 billion. CONCLUSIONS: Artificial intelligence-informed bio-surveillance combining NLP of free-text information with structured EHR data improves data completeness, prevents thousands of strokes, and saves lives and funds. This method is applicable to many disorders with profound public health consequences.

Establishing Healthcare Worker Performance and Safety in Providing Critical Care for Patients in a Simulated Ebola Treatment Unit: Non-Randomized Pilot Study.

Improving the provision of supportive care for patients with Ebola is an important quality improvement initiative. We designed a simulated Ebola Treatment Unit (ETU) to assess performance and safety of healthcare workers (HCWs) performing tasks wearing personal protective equipment (PPE) in hot (35 °C, 60% relative humidity) or thermo-neutral (20 °C, 20% relative humidity) conditions. In this pilot phase to determine the feasibility of study procedures, HCWs in PPE were non-randomly allocated to hot or thermo-neutral conditions to perform peripheral intravenous (PIV) and midline catheter (MLC) insertion and endotracheal intubation (ETI) on mannequins. Eighteen HCWs (13 physicians, 4 nurses, 1 nurse practitioner; 2 with prior ETU experience; 10 in hot conditions) spent 69 (10) (mean (SD)) minutes in the simulated ETU. Mean (SD) task completion times were 16 (6) min for PIV insertion; 33 (5) min for MLC insertion; and 16 (8) min for ETI. Satisfactory task completion was numerically higher for physicians vs. nurses. Participants' blood pressure was similar, but heart rate was higher (p = 0.0005) post-simulation vs. baseline. Participants had a median (range) of 2.0 (0.0-10.0) minor PPE breaches, 2.0 (0.0-6.0) near-miss incidents, and 2.0 (0.0-6.0) health symptoms and concerns. There were eight health-assessment triggers in five participants, of whom four were in hot conditions. We terminated the simulation of two participants in hot conditions due to thermal discomfort. In summary, study tasks were suitable for physician participants, but they require redesign to match nurses' expertise for the subsequent randomized phase of the study. One-quarter of participants had a health-assessment trigger. This research model may be useful in future training and research regarding clinical care for patients with highly infectious pathogens in austere settings.

Evaluation of the SARS-CoV-2 Antibody Response to the BNT162b2 Vaccine in Patients Undergoing Hemodialysis.

Importance: Patients undergoing hemodialysis have a high mortality rate associated with COVID-19, and this patient population often has a poor response to vaccinations. Randomized clinical trials for COVID-19 vaccines included few patients with kidney disease; therefore, vaccine immunogenicity is uncertain in this population. Objective: To evaluate the SARS-CoV-2 antibody response in patients undergoing chronic hemodialysis following 1 vs 2 doses of BNT162b2 COVID-19 vaccination compared with health care workers serving as controls and convalescent serum. Design, Setting, and Participants: A prospective, single-center cohort study was conducted between February 2 and April 17, 2021, in Toronto, Ontario, Canada. Participants included 142 patients receiving in-center hemodialysis and 35 health care worker controls. Exposures: BNT162b2 (Pfizer-BioNTech) COVID-19 vaccine. Main Outcomes and Measures: SARS-CoV-2 IgG antibodies to the spike protein (anti-spike), receptor binding domain (anti-RBD), and nucleocapsid protein (anti-NP). Results: Among the 142 participants undergoing maintenance hemodialysis, 94 (66%) were men; median age was 72 (interquartile range, 62-79) years. SARS-CoV-2 IgG antibodies were measured in 66 patients receiving 1 vaccine dose following a public health policy change, 76 patients receiving 2 vaccine doses, and 35 health care workers receiving 2 vaccine doses. Detectable anti-NP suggestive of natural SARS-CoV-2 infection was detected in 15 of 142 (11%) patients at baseline, and only 3 patients had prior COVID-19 confirmed by reverse transcriptase polymerase chain reaction testing. Two additional patients contracted COVID-19 after receiving 2 doses of vaccine. In 66 patients receiving a single BNT162b2 dose, seroconversion occurred in 53 (80%) for anti-spike and 36 (55%) for anti-RBD by 28 days postdose, but a robust response, defined by reaching the median levels of antibodies in convalescent serum from COVID-19 survivors, was noted in only 15 patients (23%) for anti-spike and 4 (6%) for anti-RBD in convalescent serum from COVID-19 survivors. In patients receiving 2 doses of BNT162b2 vaccine, seroconversion occurred in 69 of 72 (96%) for anti-spike and 63 of 72 (88%) for anti-RBD by 2 weeks following the second dose and median convalescent serum levels were reached in 52 of 72 patients (72%) for anti-spike and 43 of 72 (60%) for anti-RBD. In contrast, all 35 health care workers exceeded the median level of anti-spike and anti-RBD found in convalescent serum 2 to 4 weeks after the second dose. Conclusions and Relevance: This study suggests poor immunogenicity 28 days following a single dose of BNT162b2 vaccine in the hemodialysis population, supporting adherence to recommended vaccination schedules and avoiding delay of the second dose in these at-risk individuals.

Longitudinal K-means approaches to clustering and analyzing EHR opioid use trajectories for clinical subtypes.

Identification of patient subtypes from retrospective Electronic Health Record (EHR) data is fraught with inherent modeling issues, such as missing data and variable length time intervals, and the results obtained are highly dependent on data pre-processing strategies. As we move towards personalized medicine, assessing accurate patient subtypes will be a key factor in creating patient specific treatment plans. Partitioning longitudinal trajectories from irregularly spaced and variable length time intervals is a well-established, but open problem. In this work, we present and compare k-means approaches for subtyping opioid use trajectories from EHR data. We then interpret the resulting subtypes using decision trees, examining how each subtype is influenced by opioid medication features and patient diagnoses, procedures, and demographics. Finally, we discuss how the subtypes can be incorporated in static machine learning models as features in predicting opioid overdose and adverse events. The proposed methods are general, and can be extended to other EHR prescription dosage trajectories.

How physicians change: Multisource feedback driven intervention improves physician leadership and teamwork.

BACKGROUND: Multisource feedback provides a method of quantitatively assessing and improving physician professionalism, interpersonal communication, teamwork, and leadership behaviors. We sought to determine whether tiered educational interventions can improve measurements of multisource feedback for physicians across specialties, and whether multisource feedback baseline measurements and improvements after intervention vary by specialty designation. METHODS: Multisource feedback assessments were performed on physicians from academic (34%) and community hospitals (66%) in the United States and Canada. PULSE 360 Survey data was obtained on 1,190 physicians from primary care (25%), surgical (46%), and other (29%) specialties. Physician respondents were 75% male and 24% female. Raters included administrators, colleagues, staff, and self-ratings with an average of 35.7 ratings per physician. A leadership teamwork index was measured before and after delivery of educational intervention. Three tiers of intervention were used depending on baseline leadership teamwork index score: (1) report only, (2) debriefing only, and (3) debriefing and development. RESULTS: Surgeons had a significantly lower baseline leadership teamwork index at 59.9, whereas primary care and specialists started with an leadership teamwork index of 67.1 and 65.9, respectively. Those who participated in a tier 3 intervention had the greatest change from an average baseline leadership teamwork index of 36.6 to 56.3 on follow-up. Surgeons experienced the largest mean increase of 9.1 leadership teamwork index points after intervention, whereas medicine specialists had a mean increase of 6.7 leadership teamwork index points. CONCLUSION: Baseline multisource feedback scores vary by specialty and improve based on feedback, goal-setting, coaching, and education. In particular, physicians who start with low scores have the greatest potential for leadership teamwork index improvement.

Assessing Opioid Use Patient Representations and Subtypes.

Precision medicine, diagnosis, treatment, and prevention that accounts for natural human variability, can be beneficial for complex populations. The opioid user population is heterogeneous, characterized by many disorders, medications, and procedures. Using Electronic Health Record data, we create a patient representation, finding similarities between structured data, and then cluster the patients into patient subtypes. These subtypes can then be used for subsequent analysis.

Rosacea Patients Are at Higher Risk for Obstructive Sleep Apnea: Automated Retrospective Research.

Using big data science we employ NLP and a novel interface the BMI Investigator to answer clinically meaninful questions. The use case presented is the association between Rosacea and Obstructive Sleep Apnea.

Principles and Practices of Establishing a Hospital-Based Ebola Treatment Unit.

Outbreaks of Ebola virus disease and high-risk transmissible infections are increasing and pose threats to health care workers and global health systems. Previous outbreaks offer lessons for health system preparedness and response, including establishment of hospital-based high-risk pathogen treatment units. Their creation demands early preparation and interprofessional coordination; infection prevention and control; case management training; prepositioning of supplies; conversion of existing structures to treatment units; and strengthening communication and research platforms. Hospital-based Ebola and high-risk pathogen treatment units may improve case detection, interrupt transmission, and improve staff safety and patient care.

Expanding Evolutionary Terminology Auditing with Historic Formal and Linguistic Intensions: A Case Study in SNOMED CT.

A method is described to use SNOMED CT's history mechanism as a means to compute how the formal and linguistic intensions of its concepts change over versions. As a result of this, it is demonstrated that the intended principle of concept permanence is not always adhered to. It is shown that the evolution of formal intensions can be monitored fully automatically and that the proposed procedure includes a method to suggest missing subsumers in a concept's transitive closure set by identifying mistakes that have been made in the past. Changes in linguistic intensions were found to be much more labor-intensive to identify. It is suggested that this could be improved if the history mechanism would come with more detailed motivations for change than the current and insufficiently used annotation to the effect that a fully specified name 'fails to comply with the current editorial guidance'.

Changes in implicit alcohol attitudes across adolescence, and associations with emerging alcohol use: Testing the reciprocal determinism hypothesis.

Implicit alcohol-related cognitions develop during adolescence and are thought to play an important role in the etiology of adolescent alcohol use. Rooted in reciprocal determinism, a developmental theory of alcohol-related cognitions, the current study sought to enhance our understanding of the development of automatic alcohol associations and their relationship with alcohol use. To provide a theoretically aligned test of reciprocal determinism, we used latent change score models to examine whether growth in automatic alcohol associations and alcohol use was related to each other (between-person effects) and whether each construct led to changes in the other over time (within-person effects). Adolescents (N = 378) completed 4 annual assessments, spanning early to middle adolescence. Automatic alcohol associations were assessed with a Single Category Implicit Association Test, and we used a quadruple processing tree model to extract a more "process pure" index of these associations. Alcohol use increased from early to middle adolescence, as negative automatic alcohol associations weakened over that same time period. Although there was no support for between-person associations, on the within-person level, weak negative automatic alcohol associations at Waves 2 and 3 were associated with increases in drinking at subsequent waves. Alcohol use did not significantly predict changes in automatic alcohol associations. Findings suggest the utility of distinguishing within- and between-person associations to understand the development of automatic alcohol associations and that automatic alcohol associations are prospectively associated with alcohol use and a potential target for intervention, one that becomes an increasingly salient influence on drinking as adolescence progresses. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Biomedical Informatics Investigator.

The BMI Investigator is a computer human interface built in .Net which allows simultaneous query of structured data such as demographics, administrative codes, medications (coded in RxNorm), laboratory test results (coded in LOINC) and formerly unstructured data in clinical notes (coded in SNOMED CT). The ontology terms identified using SNOMED are all coded as either positive, negative or uncertain assertions. They are then where applicable built into compositional expressions and stored in both a graph database and a triple store. The SNOMED CT codes are stored in a NOSQL database, Berkley DB, and the structured data is stored in SQL using the OMOP/OHDSI format. The BMI investigator also lets you develop models for cohort selection (data driven recruitment to clinical trials) and automated retrospective research using genomic criteria and we are adding image feature data currently to the system. We performed a usability experiment and the users identified some usability flaws which were used to improve the software. Overall, the BMI Investigator was felt to be usable by subject matter experts. Next steps for the software are to integrate genomic criteria and image features into the query engine.

Clinical Data Warehouse Query and Learning Tool Using a Human-Centered Participatory Design Process.

BMI Investigator (BMII) is an interactive web-based tool with a learning knowledge base, which provides a way for researchers to query structured, unstructured, genomic and image data contained in a data warehouse. We demonstrate how development of an efficient, usable, and learnable web interface for a diverse group of research stakeholders benefits from an iterative human-centered participatory design process utilizing a team of clinicians, students, programmers, and informatics experts.

Exploring the use of Option Grid™ patient decision aids in a sample of clinics in Poland.

BACKGROUND: Research on the implementation of patient decision aids to facilitate shared decision making in clinical settings has steadily increased across Western countries. A study which implements decision aids and measures their impact on shared decision making has yet to be conducted in the Eastern part of Europe. OBJECTIVE: To study the use of Option GridTM patient decision aids in a sample of Grupa LUX MED clinics in Warsaw, Poland, and measure their impact on shared decision making. METHOD: We conducted a pre-post interventional study. Following a three-month period of usual care, clinicians from three Grupa LUX MED clinics received a one-hour training session on how to use three Option GridTM decision aids and were provided with copies for use for four months. Throughout the study, all eligible patients were asked to complete the three-item CollaboRATE patient-reported measure of shared decision making after their clinical encounter. CollaboRATE enables patients to assess the efforts clinicians make to: (i) inform them about their health issues; (ii) listen to 'what matters most'; (iii) integrate their treatment preference in future plans. A hierarchical logistic regression model was performed to understand which variables had an effect on CollaboRATE. RESULTS: 2,048 patients participated in the baseline phase; 1,889 patients participated in the intervention phase. Five of the thirteen study clinicians had a statistically significant increase in their CollaboRATE scores (p<.05) when comparing baseline phase to intervention phase. All five clinicians were located at the same clinic, the only clinic where an overall increase (non-significant) in the mean CollaboRATE top score percentage occurred from baseline phase (M=60 %, SD=0.49; 95 % CI [57-63 %]) to intervention phase (M=62 %, SD=0.49; 95% CI [59-65%]). Only three of those five clinicians who had a statistically significant increase had a clinically significant difference. CONCLUSION: The implementation of Option GridTM helped some clinicians practice shared decision making as reflected in CollaboRATE scores, but most clinicians did not have a significant increase in their scores. Our study indicates that the effect of these interventions may be dependent on clinic contexts and clinician engagement.