Balancing versus modelling in weighted analysis of non-randomised studies with survival outcomes: A simulation study.

Weighting methods are widely used for causal effect estimation in non-randomised studies. In general, these methods use the propensity score (PS), the probability of receiving the treatment given the covariates, to arrive at the respective weights. All of these "modelling" methods actually optimize prediction of the respective outcome, which is, in the PS model, treatment assignment. However, this does not match with the actual aim of weighting, which is eliminating the association between covariates and treatment assignment. In the "balancing" approach, covariates are thus balanced directly by solving systems of numerical equations, explicitly without fitting a PS model. To compare modelling, balancing and hybrid approaches to weighting we performed a large simulation study for a binary treatment and a survival outcome. For maximal practical relevance all simulation parameters were selected after a systematic review of medical studies that used PS methods for analysis. We also introduce a new hybrid method that uses the idea of the covariate balancing propensity score and matching weights, thus avoiding extreme weights. In addition, we present a corrected robust variance estimator for some of the methods. Overall, our simulations results indicate that balancing approach methods work worse than expected. However, among the considered balancing methods, entropy balancing consistently outperforms the variance balancing approach. All methods estimating the average treatment effect in the overlap population perform well with very little bias and small standard errors even in settings with misspecified propensity score models. Finally, the coverage using the standard robust variance estimator was too high for all methods, with the proposed corrected robust variance estimator improving coverage in a variety of settings.

Vaccination Training for Pharmacy Undergraduates as a Compulsory Part of the Curriculum?-A Multicentric Observation.

In order to increase vaccination rates, the Government of Germany introduced vaccination against influenza and COVID-19 into the regular care administered by pharmacists. However, vaccination training is yet not integrated into the German pharmacy curriculum. Therefore, the Institute for Clinical Pharmacy and Pharmacotherapy in Duesseldorf had developed an innovative vaccination course using high-fidelity simulation for students. To investigate the acceptance further, the course was carried out at three different German universities (Bonn, Duesseldorf, Greifswald). Students were asked to give their self-assessment before and after and satisfaction only after the training course. Responses from 33 participants from the University of Bonn, 42 from the University of Duesseldorf and 49 from the University of Greifswald were analyzed. Every participant at the respective universities showed a significant increase in their self-assessment and indicated a high level of satisfaction with the course. The results also did not differ significantly between the respective universities. Consequently, the results lead to the hypothesis that the satisfaction of pharmacy students with this kind of training using high-fidelity simulation is very high and attractive, and can be recommended for other German universities. The integration of such vaccination training into the German pharmacy curriculum might be a future step.

Impact of a Digital Tool on Pharmacy Students' Ability to Perform Medication Reviews: A Randomized Controlled Trial.

Digital Medication Review Tools (DMRTs) are increasingly important in pharmacy practice. To ensure that young pharmacists are sufficiently competent to perform medication reviews after graduation, the introduction of DMRTs teaching in academic education is necessary. The aim of our study was to demonstrate the effect of DMRTs use on pharmacy students' performance when conducting a medication review (MR) in a randomized controlled pre-post design. Forty-one pharmacy students were asked to complete a MR within 60 min, followed by a 10-min consultation with (intervention group) and without a DMRT (control group). The MR performance was subdivided into four categories: communication skills, subjective and objective patient data, assessment, and plan. Performance was assessed using objective structured clinical examinations (OSCEs) and analytical checklists. With the use of DMRTs, the overall performance was improved by 17.0% compared to the control group (p < 0.01). Improvement through DMRTs was seen in the subcategories "Assessment" and "Plan". Furthermore, pharmacy students liked using DMRTs and felt more confident overall. Our study results demonstrate that DMRTs improve the performance of MRs, hence DMRTs should become an integral part of pharmacy curriculum. Consequently, digitally enabled pharmacists using DMRTs will be better prepared for their professional careers in pharmacy practice.

Development and Assessment of Innovative High-Fidelity Simulation Vaccination Course Integrating Emergency Cases for Pharmacy Undergraduates-A Randomized Controlled Study.

Recently, pharmacists in Germany were allowed to administer influenza and COVID-19 vaccines for people aged 12 years and older in order to increase vaccination coverage rates. In order to adapt the pharmacy curriculum for clinical practice, an innovative, vaccination training course using a high-fidelity simulator (HFS) was developed, implementing clinical scenarios to manage adverse events. In a randomized controlled trial using a pre and post design with pharmacy undergraduates, the intervention group interacted with an HFS, while the control group was trained with low-fidelity injection pads. Before and after the respective training, each participant went through an objective structured clinical examination (OSCE) and completed a self-assessment questionnaire and knowledge quiz. Both training methods showed a significant increase in skills, but there was also a significant greater increase in the intervention group when compared to the control group, particularly with respect to the vaccination process. Furthermore, every individual in the intervention group improved from the pre- to post-training OSCEs. Therefore, HFS has been proven to be an appropriate tool to train pharmacy students for the purposes of vaccine administration and to prepare for future challenges. Particularly, recognizing and managing adverse reactions can be addressed in a very effective way.

Efficient gene-environment interaction testing through bootstrap aggregating.

Gene-environment (GxE) interactions are an important and sophisticated component in the manifestation of complex phenotypes. Simple univariate tests lack statistical power due to the need for multiple testing adjustment and not incorporating potential interplay between several genetic loci. Approaches based on internally constructed genetic risk scores (GRS) require the partitioning of the available sample into training and testing data sets, thus, lowering the effective sample size for testing the GxE interaction itself. To overcome these issues, we propose a statistical test that employs bagging (bootstrap aggregating) in the GRS construction step and utilizes its out-of-bag prediction mechanism. This approach has the key advantage that the full available data set can be used for both constructing the GRS and testing the GxE interaction. To also incorporate interactions between genetic loci, we, furthermore, investigate if using random forests as the GRS construction method in GxE interaction testing further increases the statistical power. In a simulation study, we show that both novel procedures lead to a higher statistical power for detecting GxE interactions, while still controlling the type I error. The random-forests-based test outperforms a bagging-based test that uses the elastic net as its base learner in most scenarios. An application of the testing procedures to a real data set from a German cohort study suggests that there might be a GxE interaction involving exposure to air pollution regarding rheumatoid arthritis.

A cross-cohort replicable and heritable latent dimension linking behaviour to multi-featured brain structure.

Identifying associations between interindividual variability in brain structure and behaviour requires large cohorts, multivariate methods, out-of-sample validation and, ideally, out-of-cohort replication. Moreover, the influence of nature vs nurture on brain-behaviour associations should be analysed. We analysed associations between brain structure (grey matter volume, cortical thickness, and surface area) and behaviour (spanning cognition, emotion, and alertness) using regularized canonical correlation analysis and a machine learning framework that tests the generalisability and stability of such associations. The replicability of brain-behaviour associations was assessed in two large, independent cohorts. The load of genetic factors on these associations was analysed with heritability and genetic correlation. We found one heritable and replicable latent dimension linking cognitive-control/executive-functions and positive affect to brain structural variability in areas typically associated with higher cognitive functions, and with areas typically associated with sensorimotor functions. These results revealed a major axis of interindividual behavioural variability linking to a whole-brain structural pattern.

Chronic air pollution-induced subclinical airway inflammation and polygenic susceptibility.

BACKGROUND: Air pollutants can activate low-grade subclinical inflammation which further impairs respiratory health. We aimed to investigate the role of polygenic susceptibility to chronic air pollution-induced subclinical airway inflammation. METHODS: We used data from 296 women (69-79 years) enrolled in the population-based SALIA cohort (Study on the influence of Air pollution on Lung function, Inflammation and Aging). Biomarkers of airway inflammation were measured in induced-sputum samples at follow-up investigation in 2007-2010. Chronic air pollution exposures at residential addresses within 15 years prior to the biomarker assessments were used to estimate main environmental effects on subclinical airway inflammation. Furthermore, we calculated internally weighted polygenic risk scores based on genome-wide derived single nucleotide polymorphisms. Polygenic main and gene-environment interaction (GxE) effects were investigated by adjusted linear regression models. RESULTS: Higher exposures to nitrogen dioxide (NO2), nitrogen oxides (NOx), particulate matter with aerodynamic diameters of ≤ 2.5 µm, ≤ 10 µm, and 2.5-10 µm significantly increased the levels of leukotriene (LT)B4 by 19.7% (p-value = 0.005), 20.9% (p = 0.002), 22.1% (p = 0.004), 17.4% (p = 0.004), and 23.4% (p = 0.001), respectively. We found significant effects of NO2 (25.9%, p = 0.008) and NOx (25.9%, p-value = 0.004) on the total number of cells. No significant GxE effects were observed. The trends were mostly robust in sensitivity analyses. CONCLUSIONS: While this study confirms that higher chronic exposures to air pollution increase the risk of subclinical airway inflammation in elderly women, we could not demonstrate a significant role of polygenic susceptibility on this pathway. Further studies are required to investigate the role of polygenic susceptibility.

The Role of Polygenic Susceptibility on Air Pollution-Associated Asthma between German and Japanese Elderly Women.

Polygenic susceptibility likely influences individual responses to air pollutants and the risk of asthma. We compared the role of polygenic susceptibility on air pollution-associated asthma between German and Japanese women. We investigated women that were enrolled in the German SALIA cohort (n = 771, mean age = 73 years) and the Japanese Shika cohort (n = 847, mean age = 67 years) with known asthma status. Adjusted logistic regression models were used to assess the associations between (1) particulate matter with a median aerodynamic diameter ≤ 2.5µm (PM2.5) and nitrogen dioxide (NO2), (2) polygenic risk scores (PRS), and (3) gene-environment interactions (G × E) with asthma. We found an increased risk of asthma in Japanese women after exposure to low pollutant levels (PM2.5: median = 12.7µg/m3, p-value < 0.001, NO2: median = 8.5µg/m3, p-value < 0.001) and in German women protective polygenic effects (p-value = 0.008). While we found no significant G × E effects, the direction in both groups was that the PRS increased the effect of PM2.5 and decreased the effect of NO2 on asthma. Our study confirms that exposure to low air pollution levels increases the risk of asthma in Japanese women and indicates polygenic effects in German women; however, there was no evidence of G × E effects. Future genome-wide G × E studies should further explore the role of ethnic-specific polygenic susceptibility to asthma.

Core body temperatures during final stages of life-an evaluation of data from in-hospital decedents.

Temperature-based methods are widely accepted as the gold standard for death time estimation. In the absence of any other information, the nomogram method generally assumes that a person died with a core body temperature of approximately 37.2 °C. Nevertheless, several external and internal factors may alter the body temperature during agony. A retrospective medical record analysis was carried out on in-hospital death cases from two consecutive years of surgical intensive care units to determine the effects of factors influencing the core body temperature at the point of death. Data from 103 case files were included in the statistical data evaluation. The body temperature fluctuated between and within individuals over time. No clear correlation to certain death groups was observed. Even primary cardiac deaths showed broad intervals of temperatures at the point of death. Men seem to die with higher body temperatures than women. The presented data highlight potential biases for death time estimations when generally assuming a core body temperature of 37.2 °C. In conclusion, the estimation of the time of death should include various methods, including a non-temperature-dependent method. Any uncertainties regarding the body temperature at point of death need to be resolved (e.g. by identifying fever constellations) and elucidated if elimination is not possible.

Cycling under the influence of alcohol-criminal offenses in a German metropolis.

INTRODUCTION: Real or simulated cycling tests under the influence of alcohol might be biased by laboratory settings. Accident analyses consider incidents with injuries only. Herein, criminal offenses consisting of drunk cycling are evaluated in detail to fill this gap. MATERIAL AND METHODS: All police-recorded cases of cycling under the influence of alcohol that took place in Düsseldorf, Germany, from 2009 to 2018 were identified. A total of 388 respective prosecutor's files were available for analyses. RESULTS: Mean blood alcohol concentrations were approximately 2 g/kg in both men and women. Men were overrepresented (6:1). Almost 60% of the cases were recorded between Friday and Sunday (the "weekend"). The average blood alcohol concentration (BAC) at night (01:00-05:59) was 0.39 g/kg lower than that during the day (06:00-17:59). Drinking after cycling allegations appear almost irrelevant among (German) cyclists. On average, the legal outcomes show 33 daily rates (median: 30). Additionally, the presented data raise doubts about whether the utilized medical tests or the ways in which they are carried out reliably discriminate between different grades of intoxication. Negative tests did not exclude high BACs, nor did positive tests correlate well with BACs. DISCUSSION/CONCLUSION: In practice, CUI is seen with BACs above 1.60 g/kg in most cases. BACs below 1.60 g/kg either seem to be a minor problem or they have been incompletely addressed thus far. In summary, to be prosecuted, drunk cyclists have to ride their bikes in either a highly insecure or rude manner or they must cause an accident.

Evaluation of tree-based statistical learning methods for constructing genetic risk scores.

BACKGROUND: Genetic risk scores (GRS) summarize genetic features such as single nucleotide polymorphisms (SNPs) in a single statistic with respect to a given trait. So far, GRS are typically built using generalized linear models or regularized extensions. However, these linear methods are usually not able to incorporate gene-gene interactions or non-linear SNP-response relationships. Tree-based statistical learning methods such as random forests and logic regression may be an alternative to such regularized-regression-based methods and are investigated in this article. Moreover, we consider modifications of random forests and logic regression for the construction of GRS. RESULTS: In an extensive simulation study and an application to a real data set from a German cohort study, we show that both tree-based approaches can outperform elastic net when constructing GRS for binary traits. Especially a modification of logic regression called logic bagging could induce comparatively high predictive power as measured by the area under the curve and the statistical power. Even when considering no epistatic interaction effects but only marginal genetic effects, the regularized regression method lead in most cases to inferior results. CONCLUSIONS: When constructing GRS, we recommend taking random forests and logic bagging into account, in particular, if it can be assumed that possibly unknown epistasis between SNPs is present. To develop the best possible prediction models, extensive joint hyperparameter optimizations should be conducted.

E-scooter driving under the acute influence of alcohol-a real-driving fitness study.

PURPOSE: To assess the effects of alcohol on the ability to drive an e-scooter, driving tests reflecting real-life situations accompanied by medical examinations focusing on balance were conducted at different blood alcohol concentrations (BACs). METHODS: Fifty-seven subjects who consumed alcohol (28 female, 29 male) and 6 consistently sober subjects (3 female, 3 male) participated in the study. Alcohol was administered on a fixed schedule, and the individual drinking quantity was individually calculated in advance using the Widmark formula. Repeated runs through a fixed course were performed. Following each ride, a blood sample was taken for BAC determination, and medical tests were performed. RESULTS: Even at low BACs (0.21-0.60 g/kg), subjects showed a significant decrease in driving performance, to approximately 60% of the initial level. Differences in driving performance at different BAC ranges were observed for different obstacles, especially for the narrowing track, gate passage, slalom, and driving in circles obstacles. Furthermore, worse Romberg and Unterberger test results were correlated with worse driving performance. It cannot be assumed that learning effects during the study had a relevant effect, as shown in the comparison of the driving performance of the alcohol-consuming group with that of the control group. Sex-specific differences were not found. DISCUSSION: Significant deteriorations in driving performance at BACs below 1.10 g/kg confirmed alcohol-related risk potential when using e-scooters. At this time, these findings may lead to the assumption of "relative driving impairment" in Germany. The Romberg and Unterberger tests could be considered a complementary investigation method for the assessment of e-scooter driving impairment. CONCLUSION: Even at rather low BACs between 0.21 and 0.40 g/kg, there was a significant deterioration in driving performance under the influence of alcohol compared to sober, which highlights the negative effects of alcohol on e-scooter driving.

Confound-leakage: confound removal in machine learning leads to leakage.

BACKGROUND: Machine learning (ML) approaches are a crucial component of modern data analysis in many fields, including epidemiology and medicine. Nonlinear ML methods often achieve accurate predictions, for instance, in personalized medicine, as they are capable of modeling complex relationships between features and the target. Problematically, ML models and their predictions can be biased by confounding information present in the features. To remove this spurious signal, researchers often employ featurewise linear confound regression (CR). While this is considered a standard approach for dealing with confounding, possible pitfalls of using CR in ML pipelines are not fully understood. RESULTS: We provide new evidence that, contrary to general expectations, linear confound regression can increase the risk of confounding when combined with nonlinear ML approaches. Using a simple framework that uses the target as a confound, we show that information leaked via CR can increase null or moderate effects to near-perfect prediction. By shuffling the features, we provide evidence that this increase is indeed due to confound-leakage and not due to revealing of information. We then demonstrate the danger of confound-leakage in a real-world clinical application where the accuracy of predicting attention-deficit/hyperactivity disorder is overestimated using speech-derived features when using depression as a confound. CONCLUSIONS: Mishandling or even amplifying confounding effects when building ML models due to confound-leakage, as shown, can lead to untrustworthy, biased, and unfair predictions. Our expose of the confound-leakage pitfall and provided guidelines for dealing with it can help create more robust and trustworthy ML models.

EMDIA Case Series-Effective Medication Therapy Management (MTM) for Diabetes Type 2 Patients-A Proof of Concept Study.

Background: A 2016 meta-analysis of pharmaceutical care for patients with diabetes mellitus showed that the following four components were most effective: (a) individual goal setting, (b) sending feedback to the physician, (c) reviewing the medication, and (d) reviewing blood glucose measurements. Methods: To formulate a hypothesis regarding the effect of these four pharmaceutical care components on glycemic control in patients with diabetes mellitus and the feasibility of these components in practice. Ten patients with type 2 diabetes were included in the case series and received medication therapy management over four months. Results: The four care components were feasible in everyday practice and could be implemented within one patient visit. The average visits were 49 and 28 min at the beginning and end of the study, respectively. The glycated hemoglobin values did not change over the study period, though the fasting blood glucose decreased from 142 to 120 mg/dl, and the number of unsolved drug-related problems decreased from 6.9 to 1.9 per patient by the study end. Conclusions: This case series supports the hypothesis that community pharmacists can implement structured pharmaceutical care in everyday pharmacy practice for patients with type 2 diabetes mellitus.

Blended Learning on Blood Pressure Measurement: Investigating Two In-Class Strategies in a Flipped Classroom-Like Setting to Teach Pharmacy Students Blood Pressure Measurement Skills.

For reliable blood pressure measurement, various potential sources of inaccuracies need to be considered to avoid incorrect decision-making. Pharmacy students should be sensitized and taught the skill accordingly. One strategy to teach students' blood pressure measurement skills might be through a blended learning approach in a flipped classroom-like setting. With a randomized two-arm study among pharmacy students in their eighth semester, the required extent of in-class session in the scope of a blended learning approach in a flipped classroom-like setting was evaluated. Participants' self-confidence and self-perceived proficiency were evaluated through a survey, and participants' blood pressure measurement performance was assessed by objective structured clinical examination (OSCE). Participants' satisfaction with, and perception of, the flipped classroom were also surveyed. The extended in-class activities did not result in a significantly higher increase of participants' OSCE score and self-assessment score when compared to the brief in-class session. Both in-class sessions yielded a significant increase in the OSCE scores as well as in the self-assessment scores. Moreover, the teaching approaches were predominantly well-received by the students. The use of both flipped classroom-like approaches improved pharmacy students' blood pressure measurement performance, though the brief in-class session was sufficient. Students' self-confidence/self-perceived proficiency in blood pressure measurement skills increased similarly in both settings.

Training Pharmacy Students in Self-Medication Counseling Using an Objective Structured Clinical Examination-Based Approach.

INTRODUCTION: Pharmacists play an important role in ensuring the safe, effective, and rational use of drugs in self-medication. Given the potential risks of self-medication, adequate training on self-medication counseling should be provided to pharmacy students during their academic education. Objective structured clinical examinations (OSCEs) could be used to train pharmacy students in these skills. This study evaluated the efficacy of an OSCE-based approach for training pharmacy students in self-medication counseling and communication skills. METHODS: This randomized controlled study was conducted among pharmacy students using a pre-post design. The intervention group completed OSCE-based self-medication training, while the control group collected counseling-relevant information from summaries of product characteristics of over-the-counter drugs. The counseling and communication skills of both groups before and after training were assessed by completing OSCEs. The participants completed a self-assessment questionnaire on self-confidence and self-perceived proficiency before each OSCE encounter and a satisfaction survey at the end of the seminar. RESULTS: Students were generally satisfied with the seminar. While the OSCE-trained group demonstrated significantly greater increases in counseling skills and self-confidence and self-perceived proficiency than the control group, both groups had similar increases in communication skills. CONCLUSION: The present study suggests that applying OSCEs as a learning tool for self-medication counseling is beneficial for improving students' counseling skills as well as self-confidence and self-perceived proficiency. These results support the inclusion of OSCEs in pharmacy education and highlight its potential to bridge gaps between knowledge and practice.

A Connectivity-Based Psychometric Prediction Framework for Brain-Behavior Relationship Studies.

The recent availability of population-based studies with neuroimaging and behavioral measurements opens promising perspectives to investigate the relationships between interindividual variability in brain regions' connectivity and behavioral phenotypes. However, the multivariate nature of connectivity-based prediction model severely limits the insight into brain-behavior patterns for neuroscience. To address this issue, we propose a connectivity-based psychometric prediction framework based on individual regions' connectivity profiles. We first illustrate two main applications: 1) single brain region's predictive power for a range of psychometric variables and 2) single psychometric variable's predictive power variation across brain region. We compare the patterns of brain-behavior provided by these approaches to the brain-behavior relationships from activation approaches. Then, capitalizing on the increased transparency of our approach, we demonstrate how the influence of various data processing and analyses can directly influence the patterns of brain-behavior relationships, as well as the unique insight into brain-behavior relationships offered by this approach.

Estimation of time since death after a post-mortem change in ambient temperature: Evaluation of a back-calculation approach.

The temperature-based "Henssge method" is widely applied for death time estimation. For cases with a sudden post-mortem (pm) change in ambient temperature (e.g., by bringing the deceased into a cooling chamber), a mathematical approach has already been proposed [1] that enables estimation of the time of death by back-calculation of body temperature. This approach was evaluated under clinically controlled conditions. Twenty-five individuals who died in a neurosurgical intensive care unit were brought to cooling storage after approximately 3 h pm. Body temperature was repeatedly measured on the ward and in cooling storage over a period of 9 h pm. Back-calculation of body temperature was carried out on the basis of the proposed mathematical approach for cases with pm changes in ambient temperature; the results were compared to the known body temperatures. In many cases, the back-calculated and true body temperatures differed widely. Bodies regularly cooled down slower after being brought into cooling storage than the back-calculations indicated. The sudden change in ambient temperature could only be addressed roughly by the proposed method of back-calculation. In conclusion, the evaluated approach for addressing pm changes in ambient temperature should only be applied with great caution.

Efficacy of an Objective Structured Clinical Examination Training Approach for Training Pharmacy Students in Diabetes Mellitus Counseling: A Randomized Controlled Trial.

Pharmacists' tasks are multifaceted and include, for example, vital counseling and communication skills. Objective Structured Clinical Examinations (OSCEs) could be used to train pharmacy students in these skills. Our study sought to determine the efficacy of our OSCE training approach for training pharmacy students' counseling and communication skills on diabetes mellitus compared to a control group. This randomized controlled study was conducted with pharmacy students using a pre-post-design. The intervention group completed diabetes OSCE training, while the control group solved diabetes patient cases using subjective, objective, assessment, and plan notes. Before and after the respective training, both groups completed OSCEs evaluating counseling and communication skills. Before each OSCE encounter, the participants completed a self-assessment questionnaire and, upon completion of the seminar, filled out a satisfaction survey. The OSCE-trained group demonstrated a significantly greater increase in counseling and communication skills and self-confidence than the control group. Both groups were generally satisfied with the seminar. These results demonstrate that our OSCE training approach allows for the effective training of pharmacy students' diabetes counseling and communication skills and suggests the inclusion of such a skill-based approach more widely in pharmacy students' education.