Five ways to fix statistics

Trajectory classes of engagement with an alcohol text message intervention and predictors of intervention engagement

Digital interventions have increased our capacity to reach young adults who have hazardous alcohol use. Alcohol text message interventions have shown small effects in reducing hazardous drinking, leaving room for improvement. An important challenge to address in improving digital interventions is maintaining engagement, which reflects the "dose" of intervention received. This study aimed to identify trajectory classes of engagement with an alcohol text message intervention, and baseline predictors of the trajectory classes to determine "for whom" the digital intervention was more versus less engaging, to guide further intervention tailoring. This secondary analysis examined data from a study that compared five 12-week alcohol text message interventions designed to reduce hazardous drinking in young adults (aged 18-25; N = 1,131, 68% female) recruited from Emergency Departments in Western Pennsylvania. Engagement with the intervention was based on response (present/absent) to text message queries delivered twice per week during 2-week run-in and 12-week intervention. Repeated measures latent profile analysis identified five latent trajectory classes as having the best fit to the data: "High engagement" (55.1%), "Slow decrease, moderate engagement" (23.2%); "Mid-way decrease in engagement" (8.9%), "Steadily decreasing engagement" (8.1%); and "Fluctuating, moderate engagement" (4.6%). Females and individuals enrolled in college were overrepresented in the high engagement trajectory class, whereas individuals higher in impulsivity were more likely to be in decreasing (versus high) engagement trajectory classes. Methods to boost engagement using, for example, motivational enhancement for young adults with higher levels of impulsivity, at specific time points, such as the mid-point of the intervention, warrant consideration.

Teaching students to R3eason, not merely to solve problem sets: The role of philosophy and visual data communication in accessible data science education

Much guidance on statistical training in STEM fields has been focused largely on the undergraduate cohort, with graduate education often being absent from the equation. Training in quantitative methods and reasoning is critical for graduate students in biomedical and science programs to foster reproducible and responsible research practices. We argue that graduate student education should more center around fundamental reasoning and integration skills rather than mainly on listing 1 statistical test method after the other without conveying the bigger context picture or critical argumentation skills that will enable student to improve research integrity through rigorous practice. Herein, we describe the approach we take in a quantitative reasoning course in the R3 program at the Johns Hopkins Bloomberg School of Public Health, with an error-focused lens, based on visualization and communication competencies. Specifically, we take this perspective stemming from the discussed causes of irreproducibility and apply it specifically to the many aspects of good statistical practice in science, ranging from experimental design to data collection and analysis, and conclusions drawn from the data. We also provide tips and guidelines for the implementation and adaptation of our course material to various graduate biomedical and STEM science programs.

What can mathematical modelling contribute to a sociology of quantification?

Sociology of quantification has spent relatively less energies investigating mathematical modelling than it has on other forms of quantification such as statistics, metrics, or algorithms based on artificial intelligence. Here we investigate whether concepts and approaches from mathematical modelling can provide sociology of quantification with nuanced tools to ensure the methodological soundness, normative adequacy and fairness of numbers. We suggest that methodological adequacy can be upheld by techniques in the field of sensitivity analysis, while normative adequacy and fairness are targeted by the different dimensions of sensitivity auditing. We also investigate in which ways modelling can inform other instances of quantification as to promote political agency.

Person-centered patterns of substance use during the COVID-19 pandemic and their associations with COVID-related impacts on health and personal finances in young Black and White women

BACKGROUND

Population-level statistics on pandemic-related change in substance use can obscure patterns of use (e.g., polysubstance use) within individuals. This longitudinal study used a person-centered approach to identify subgroups with respect to patterns of substance use prior to and during the COVID-19 pandemic, and to examine profile correlates (e.g., socio-demographic characteristics), which can inform tailored intervention.

METHODS

The two youngest age cohorts of the Pittsburgh Girls Study (n = 938; 59.1 % Black, 40.9 % White; mean age= 26.2 (SD= 0.8)), a longitudinal community sample, provided data on past year frequency of cigarette/e-cigarette use, binge drinking (>4 drinks per occasion), and cannabis use prior to and during the pandemic, and perceived change in use. Latent profile analysis identified subgroups. Profile correlates were examined (e.g., sociodemographics, COVID-19 infection status and reported exposure, COVID-19 impacts on psychological health and finances).

RESULTS

Seven profiles were identified: "Low use", "Occasional binge drinking", "Cannabis use", "Cigarette/e-cigarette & binge drinking", "Occasional binge drinking & cannabis", "Binge drinking & cannabis", and "Polysubstance use". Black women were overrepresented in "Low use", which was associated with fewer pandemic effects on health. Profiles associated with more frequent binge drinking were more likely to report COVID-19 infection, whereas "Cannabis use" had lower reported infection prevalence. "Polysubstance use" had more COVID-related depression and income loss, on average, than "Low use".

CONCLUSIONS

Distinct subgroups representing single substance use, co-use, and polysubstance use prior to and during the pandemic were identified. The profiles show differential response to COVID-19 impacts, ranging from relative hardiness to specific needs to guide personalized treatment.

Emotion differentiation in early recovery from alcohol use disorder: Associations with in-the-moment affect and 3-month drinking outcomes

Background

Early recovery from alcohol use disorder (AUD) is commonly associated with high levels of negative affect, stress, and emotional vulnerability, which confer significant relapse risk. Emotion differentiation—the ability to distinguish between discrete emotions—has been shown to predict relapse after treatment for a drug use disorder, but this relationship has not been explored in individuals recovering from AUD.

Methods

The current study used thrice daily random and up to thrice daily self‐initiated ecological momentary assessment surveys (N = 42, observations = 915) to examine whether 1) moments of high affective arousal are characterized by momentary differences in emotion differentiation among individuals in the first year of a current AUD recovery attempt, and 2) individuals’ average emotion differentiation would predict subsequent alcohol use measured by the timeline follow‐back over a 3‐month follow‐up period.

Results

Multilevel models showed that moments (Level 1) of higher‐than‐average negative affect (p < 0.001) and/or stress (p = 0.033) were characterized by less negative emotion differentiation, while moments of higher‐than‐average positive affect were characterized by greater positive emotion differentiation (p < 0.001). At the between‐person level (Level 2), participants with higher stress overall had lower negative emotion differentiation (p = 0.009). Linear regression showed that average negative, but not positive, emotion differentiation was inversely associated with percent drinking days over the subsequent 3‐month follow‐up period (p = 0.042). Neither form of average emotion differentiation was associated with drinking quantity.

Conclusions

We found that for individuals in early AUD recovery, affective states are associated with acute shifts in the capacity for emotion differentiation. Further, we found that average negative emotion differentiation prospectively predicts subsequent alcohol use.

An observational analysis of the trope "A p-value of < 0.05 was considered statistically significant" and other cut-and-paste statistical methods

Appropriate descriptions of statistical methods are essential for evaluating research quality and reproducibility. Despite continued efforts to improve reporting in publications, inadequate descriptions of statistical methods persist. At times, reading statistical methods sections can conjure feelings of dèjá vu, with content resembling cut-and-pasted or "boilerplate text" from already published work. Instances of boilerplate text suggest a mechanistic approach to statistical analysis, where the same default methods are being used and described using standardized text. To investigate the extent of this practice, we analyzed text extracted from published statistical methods sections from PLOS ONE and the Australian and New Zealand Clinical Trials Registry (ANZCTR). Topic modeling was applied to analyze data from 111,731 papers published in PLOS ONE and 9,523 studies registered with the ANZCTR. PLOS ONE topics emphasized definitions of statistical significance, software and descriptive statistics. One in three PLOS ONE papers contained at least 1 sentence that was a direct copy from another paper. 12,675 papers (11%) closely matched to the sentence "a p-value < 0.05 was considered statistically significant". Common topics across ANZCTR studies differentiated between study designs and analysis methods, with matching text found in approximately 3% of sections. Our findings quantify a serious problem affecting the reporting of statistical methods and shed light on perceptions about the communication of statistics as part of the scientific process. Results further emphasize the importance of rigorous statistical review to ensure that adequate descriptions of methods are prioritized over relatively minor details such as p-values and software when reporting research outcomes.

Introduction to Bayesian statistics: a practical framework for clinical pharmacists

OBJECTIVES

Most pharmaceutical investigations have relied on p values to infer conclusions from their study findings. Central to this paradigm is the concept of null hypothesis significance testing. This approach is however fraught with overuse and misinterpretations. Several alternatives have already been proposed, yet uptake remains low. In this study, we aimed to discuss the pitfalls of p value-based testing and to provide readers with the basics to apply Bayesian statistics.

METHODS

Jeffreys's Amazing Statistical Package (JASP) was used to evaluate the effect of a clinical pharmacy (CP) intervention (opposed to usual care) on the number of emergency department (ED) visits without hospital admission. Basic Bayesian terminology was explained and compared with classical p value-based testing. In the study example, a Cauchy prior distribution was used to determine the effect size with a scale parameter r=0.707 at location=0 and Bayes factors (BF) were subsequently estimated. A robustness analysis was then performed to visualise the impact of different r values on the BF value.

RESULTS

A BF of 4.082 was determined, indicating that the observed data were about four times more likely to occur under the alternative hypothesis that the CP intervention was effective. The median effect size of the CP intervention on ED visits was found to be 0.337 with a 95% credible interval of 0.074 to 0.635. A robustness check was performed and all BF values were in favour of the CP intervention.

CONCLUSION

Bayesian inference can be an important addition to the statistical armamentarium of pharmacists, who should become more acquainted with the basic terminology and rationale of such testing. To prove our point, Jeffreys' approach was applied to a CP study example, using an easy-to-use software program JASP.

Use of the p-values as a size-dependent function to address practical differences when analyzing large datasets

Biomedical research has come to rely on p-values as a deterministic measure for data-driven decision-making. In the largely extended null hypothesis significance testing for identifying statistically significant differences among groups of observations, a single p-value is computed from sample data. Then, it is routinely compared with a threshold, commonly set to 0.05, to assess the evidence against the hypothesis of having non-significant differences among groups, or the null hypothesis. Because the estimated p-value tends to decrease when the sample size is increased, applying this methodology to datasets with large sample sizes results in the rejection of the null hypothesis, making it not meaningful in this specific situation. We propose a new approach to detect differences based on the dependence of the p-value on the sample size. We introduce new descriptive parameters that overcome the effect of the size in the p-value interpretation in the framework of datasets with large sample sizes, reducing the uncertainty in the decision about the existence of biological differences between the compared experiments. The methodology enables the graphical and quantitative characterization of the differences between the compared experiments guiding the researchers in the decision process. An in-depth study of the methodology is carried out on simulated and experimental data. Code availability at https://github.com/BIIG-UC3M/pMoSS.

Insights Into Transcriptome Profiles Associated With Wooden Breast Myopathy in Broilers Slaughtered at the Age of 6 or 7 Weeks

Transcriptomes associated with wooden breast (WB) were characterized in broilers at two different market ages. Breasts (Pectoralis major) were collected, 20-min postmortem, from male Ross 308 broilers slaughtered at 6 and 7 weeks of age. The breasts were classified as "non-WB" or "WB" based on palpation hardness scoring (non-WB = no abnormal hardness, WB = consistently hardened). Total RNA was isolated from 16 samples (n = 3 for 6 week non-WB, n = 3 for 6 week WB; n = 5 for 7 week non-WB, n = 5 for 7 week WB). Transcriptome was profiled using a chicken gene expression microarray with one-color hybridization technique, and compared between non-WB and WB samples of the same age. Among 6 week broilers, 910 transcripts were differentially expressed (DE) (false discovery rate, FDR < 0.05). Pathway analysis underlined metabolisms of glucose and lipids along with gap junctions, tight junction, and focal adhesion (FA) signaling as the top enriched pathways. For the 7 week broilers, 1,195 transcripts were identified (FDR < 0.05) with regulation of actin cytoskeleton, mitogen-activated protein kinase (MAPK) signaling, protein processing in endoplasmic reticulum and FA signaling highlighted as the enriched affected pathways. Absolute transcript levels of eight genes (actinin-1 - ACTN1, integrin-linked kinase - ILK, integrin subunit alpha 8 - ITGA8, integrin subunit beta 5 - ITGB5, protein tyrosine kinase 2 - PTK2, paxillin - PXN, talin 1 - TLN1, and vinculin - VCL) of FA signaling pathway were further elucidated using a droplet digital polymerase chain reaction. The results indicated that, in 6 week broilers, ITGA8 abundance in WB was greater than that of non-WB samples (p < 0.05). Concerning 7 week broilers, greater absolute levels of ACTN1, ILK, ITGA8, and TLN1, accompanied with a reduced ITGB5 were found in WB compared with non-WB (p < 0.05). Transcriptional modification of FA signaling underlined the potential of disrupted cell-cell communication that may incite aberrant molecular events in association with development of WB myopathy.

Power Analysis for Null Hypothesis Significance Testing

Before conducting a scientific study, a power analysis is performed to determine the sample size required to test an effect within allowable probabilities of Type I error (α) or Type II error (β). The power of a study is related to Type II error by 1-β. Most scientific studies set α=0.05 and power=0.80 as minimums. More conservative study designs will decrease α or increase power, which will require a larger sample size. The third and final parameter required for a power analysis is the effect size (ES). ES is a measure of the strength of the observation in the outcome of interest (ie, the dependent variable). ES must be estimated from pilot studies or published values. A small ES will require a larger sample size than a large ES. It is possible to detect statistically significant findings even for very small ES, if the sample size is sufficiently large. Therefore, it is also essential to evaluate whether ES is sufficiently large to be clinically meaningful.

Evidence from two cohorts for the frailty syndrome as an emergent state of parallel dysregulation in multiple physiological systems

Frailty is a clinical syndrome often present in older adults and characterized by a heightened vulnerability to stressors. The biological antecedents and etiology of frailty are unclear despite decades of research: frailty is associated with dysregulation in a wide range of physiological systems, but no specific cause has been identified. Here, we test predictions stemming from the hypothesis that there is no specific cause: that frailty is an emergent property arising from the complex systems dynamics of the broad loss of organismal homeostasis. Specifically, we use dysregulation of six physiological systems using the Mahalanobis distance approach in two cohorts of older adults to test the breadth, diffuseness, and nonlinearity of associations between frailty and system-specific dysregulation. We find clear support for the breadth of associations between frailty and physiological dysregulation: positive associations of all systems with frailty in at least some analyses. We find partial support for diffuseness: the number of systems or total amount of dysregulation is more important than the identity of the systems dysregulated, but results only partially replicate across cohorts. We find partial support for nonlinearity: trends are exponential but not always significantly so, and power is limited for groups with very high levels of dysregulation. Overall, results are consistent with-but not definitive proof of-frailty as an emergent property of complex systems dynamics. Substantial work remains to understand how frailty relates to underlying physiological dynamics across systems.

Effect Sizes, Power, and Biases in Intelligence Research: A Meta-Meta-Analysis

In this meta-study, we analyzed 2442 effect sizes from 131 meta-analyses in intelligence research, published from 1984 to 2014, to estimate the average effect size, median power, and evidence for bias. We found that the average effect size in intelligence research was a Pearson's correlation of 0.26, and the median sample size was 60. Furthermore, across primary studies, we found a median power of 11.9% to detect a small effect, 54.5% to detect a medium effect, and 93.9% to detect a large effect. We documented differences in average effect size and median estimated power between different types of intelligence studies (correlational studies, studies of group differences, experiments, toxicology, and behavior genetics). On average, across all meta-analyses (but not in every meta-analysis), we found evidence for small-study effects, potentially indicating publication bias and overestimated effects. We found no differences in small-study effects between different study types. We also found no convincing evidence for the decline effect, US effect, or citation bias across meta-analyses. We concluded that intelligence research does show signs of low power and publication bias, but that these problems seem less severe than in many other scientific fields.

Physician's feedback on a clinical pharmacy program on geriatric wards

OBJECTIVES

As clinical pharmacy (CP) services can improve drug use and clinical outcome in older inpatients, a dedicated CP program was installed at the geriatric wards of an academic hospital. The aim of this study was to evaluate and potentially improve the CP program, by obtaining physician's feedback.

METHODS

An anonymous e-questionnaire was sent to all physicians who were active between October 2014 and March 2018 on the acute geriatric wards (80 beds) of the University Hospitals Leuven, Belgium. Thematic content analysis was applied. Six themes were defined: satisfaction with the service, time allocation of the clinical pharmacists, content and clinical relevance of pharmaceutical interventions, communication, time savings for the treating physician and future perspectives.

RESULTS

A total of 45 physicians (59%) completed the e-questionnaire. All respondents were satisfied with the content of the provided pharmaceutical recommendations. A minority (44%) found that a 0.8 full-time equivalent clinical pharmacist presence was sufficient in terms of the expected workload. The provided CP interventions improved quality of care according to 38 (84%) physicians. Oral and written communication were considered necessary by 89% and 82% of physicians, respectively. On average, an estimated 30 minutes physician time (IQR: 15-60) per patient was saved as a result of the program. The majority (87%) preferred clinical pharmacist presence for discharge support in all geriatric patients.

CONCLUSION

Physician's satisfaction with the CP service was very high. CP services in geriatric inpatients were perceived to be clinically relevant as well as time-saving by the involved physicians.

Transcriptional Profiles of Skeletal Muscle Associated With Increasing Severity of White Striping in Commercial Broilers

Development of the white striping (WS) abnormality adversely impacts overall quality of broiler breast meat. Its etiology remains unclear. This study aimed at exploring transcriptional profiles of broiler skeletal muscles exhibiting different WS severity to elucidate molecular mechanisms underlying the development and progression of WS. Total RNA was isolated from pectoralis major of male 7-week-old Ross 308 broilers. The samples were classified as mild (n = 6), moderate (n = 6), or severe (n = 4), based on number and thickness of the white striations on the meat surface. The transcriptome was profiled using a chicken gene expression microarray with one-color hybridization technique. Gene expression patterns of each WS severity level were compared against each other; hence, there were three comparisons: moderate vs. mild (C1), severe vs. moderate (C2), and severe vs. mild (C3). Differentially expressed genes (DEGs) were identified using the combined criteria of false discovery rate ≤ 0.05 and absolute fold change ≥1.2. Differential expression of 91, 136, and 294 transcripts were identified in C1, C2, and C3, respectively. There were no DEGs in common among the three comparisons. Based on pathway analysis, the enriched pathways of C1 were related with impaired homeostasis of macronutrients and small biochemical molecules with disrupted Ca²⁺-related pathways. Decreased abundance of the period circadian regulator suggested the shifted circadian phase when moderate WS developed. The enriched pathways uniquely obtained in C2 were RNA degradation, Ras signaling, cellular senescence, axon guidance, and salivary secretion. The DEGs identified in those pathways might play crucial roles in regulating cellular ion balances and cell-cycle arrest. In C3, the pathways responsible for phosphatidylinositol 3-kinase-Akt signaling, p53 activation, apoptosis, and hypoxia-induced processes were modified. Additionally, pathways associated with a variety of diseases with the DEGs involved in regulation of [Ca²⁺], collagen formation, microtubule-based motor, and immune response were identified. Eight pathways were common to all three comparisons (i.e., calcium signaling, Ras-associated protein 1 signaling, ubiquitin-mediated proteolysis, vascular smooth muscle contraction, oxytocin signaling, and pathway in cancer). The current findings support the role of intracellular ion imbalance, particularly Ca²⁺, oxidative stress, and impaired programmed cell death on WS progression.

The cellular prion protein is a stress protein secreted by renal tubular cells and a urinary marker of kidney injury

Endoplasmic Reticulum (ER) stress underlies the pathogenesis of numerous kidney diseases. A better care of patients with kidney disease involves the identification and validation of ER stress biomarkers in the early stages of kidney disease. For the first time to our knowledge, we demonstrate that the prion protein PrP^C is secreted in a conventional manner by ER-stressed renal epithelial cell under the control of the transcription factor x-box binding protein 1 (XBP1) and can serve as a sensitive urinary biomarker for detecting tubular ER stress. Urinary PrP^C elevation occurs in patients with chronic kidney disease. In addition, in patients undergoing cardiac surgery, detectable urine levels of PrP^C significantly increase after cardiopulmonary bypass, a condition associated with activation of the IRE1-XBP1 pathway in the kidney. In conclusion, our study has identified PrP^C as a novel urinary ER stress biomarker with potential utility in early diagnosis of ongoing acute or chronic kidney injury.

Childhood obesity intervention studies: A narrative review and guide for investigators, authors, editors, reviewers, journalists, and readers to guard against exaggerated effectiveness claims

Being able to draw accurate conclusions from childhood obesity trials is important to make advances in reversing the obesity epidemic. However, obesity research sometimes is not conducted or reported to appropriate scientific standards. To constructively draw attention to this issue, we present 10 errors that are commonly committed, illustrate each error with examples from the childhood obesity literature, and follow with suggestions on how to avoid these errors. These errors are as follows: using self-reported outcomes and teaching to the test; foregoing control groups and risking regression to the mean creating differences over time; changing the goal posts; ignoring clustering in studies that randomize groups of children; following the forking paths, subsetting, p-hacking, and data dredging; basing conclusions on tests for significant differences from baseline; equating "no statistically significant difference" with "equally effective"; ignoring intervention study results in favor of observational analyses; using one-sided testing for statistical significance; and stating that effects are clinically significant even though they are not statistically significant. We hope that compiling these errors in one article will serve as the beginning of a checklist to support fidelity in conducting, analyzing, and reporting childhood obesity research.

Factoring a 2 x 2 contingency table

We show that a two-component proportional representation provides the necessary framework to account for the properties of a 2 × 2 contingency table. This corresponds to the factorization of the table as a product of proportion and diagonal row or column sum matrices. The row and column sum invariant measures for proportional variation are obtained. Geometrically, these correspond to displacements of two point vectors in the standard one-simplex, which are reduced to a center-of-mass coordinate representation, [Formula: see text]. Then, effect size measures, such as the odds ratio and relative risk, correspond to different perspective functions for the mapping of (δ, μ) to [Formula: see text]. Furthermore, variations in δ and μ will be associated with different cost-benefit trade-offs for a given application. Therefore, pure mathematics alone does not provide the specification of a general form for the perspective function. This implies that the question of the merits of the odds ratio versus relative risk cannot be resolved in a general way. Expressions are obtained for the marginal sum dependence and the relations between various effect size measures, including the simple matching coefficient, odds ratio, relative risk, Yule's Q, ϕ, and Goodman and Kruskal's τc|r. We also show that Gini information gain (IGG) is equivalent to ϕ2 in the classification and regression tree (CART) algorithm. Then, IGG can yield misleading results due to the dependence on marginal sums. Monte Carlo methods facilitate the detailed specification of stochastic effects in the data acquisition process and provide a practical way to estimate the confidence interval for an effect size.

Effect Sizes of Somatic Mutations in Cancer

A major goal of cancer biology is determination of the relative importance of the genetic alterations that confer selective advantage to cancer cells. Tumor sequence surveys have frequently ranked the importance of substitutions to cancer growth by P value or a false-discovery conversion thereof. However, P values are thresholds for belief, not metrics of effect. Their frequent misuse as metrics of effect has often been vociferously decried, even in cases when the only attributable mistake was omission of effect sizes. Here, we propose an appropriate ranking-the cancer effect size, which is the selection intensity for somatic variants in cancer cell lineages. The selection intensity is a metric of the survival and reproductive advantage conferred by mutations in somatic tissue. Thus, they are of fundamental importance to oncology, and have immediate relevance to ongoing decision making in precision medicine tumor boards, to the selection and design of clinical trials, to the targeted development of pharmaceuticals, and to basic research prioritization. Within this commentary, we first discuss the scope of current methods that rank confidence in the overrepresentation of specific mutated genes in cancer genomes. Then we bring to bear recent advances that draw upon an understanding of the development of cancer as an evolutionary process to estimate the effect size of somatic variants leading to cancer. We demonstrate the estimation of the effect sizes of all recurrent single nucleotide variants in 22 cancer types, quantifying relative importance within and between driver genes.

A non-parametric significance test to compare corpora

Classical null hypothesis significance tests are not appropriate in corpus linguistics, because the randomness assumption underlying these testing procedures is not fulfilled. Nevertheless, there are numerous scenarios where it would be beneficial to have some kind of test in order to judge the relevance of a result (e.g. a difference between two corpora) by answering the question whether the attribute of interest is pronounced enough to warrant the conclusion that it is substantial and not due to chance. In this paper, I outline such a test.

An Information Theoretic Approach to Model Selection: A Tutorial with Monte Carlo Confirmation

A reliance on null hypothesis significance testing (NHST) and misinterpretations of its results are thought to contribute to the replication crisis while impeding the development of a cumulative science. One solution is a data-analytic approach called Information-Theoretic (I-T) Model Selection, which builds upon Maximum Likelihood estimates. In the I-T approach, the scientist examines a set of candidate models and determines for each one the probability that it is the closer to the truth than all others in the set. Although the theoretical development is subtle, the implementation of I-T analysis is straightforward. Models are sorted according to the probability that they are the best in light of the data collected. It encourages the examination of multiple models, something investigators desire and that NHST discourages. This article is structured to address two objectives. The first is to illustrate the application of I-T data analysis to data from a virtual experiment. A noisy delay-discounting data set is generated and seven quantitative models are examined. In the illustration, it is demonstrated that it is not necessary to know the "truth" is to identify the one that is closest to it and that the most likely models conform to the model that generated the data. Second, we examine claims made by advocates of the I-T approach using Monte Carlo simulations in which 10,000 different data sets are generated and analyzed. The simulations showed that 1) the probabilities associated with each model returned by the single virtual experiment approximated those that resulted from the simulations, 2) models that were deemed close to the truth produced the most precise parameter estimates, and 3) adding a single replicate sharpens the ability to identify the most probable model.

The importance of sample size in marine megafauna tagging studies

Telemetry is a key, widely used tool to understand marine megafauna distribution, habitat use, behavior, and physiology; however, a critical question remains: "How many animals should be tracked to acquire meaningful data sets?" This question has wide-ranging implications including considerations of statistical power, animal ethics, logistics, and cost. While power analyses can inform sample sizes needed for statistical significance, they require some initial data inputs that are often unavailable. To inform the planning of telemetry and biologging studies of marine megafauna where few or no data are available or where resources are limited, we reviewed the types of information that have been obtained in previously published studies using different sample sizes. We considered sample sizes from one to >100 individuals and synthesized empirical findings, detailing the information that can be gathered with increasing sample sizes. We complement this review with simulations, using real data, to show the impact of sample size when trying to address various research questions in movement ecology of marine megafauna. We also highlight the value of collaborative, synthetic studies to enhance sample sizes and broaden the range, scale, and scope of questions that can be answered.

Objective classification and scoring of movement deficiencies in patients with anterior cruciate ligament reconstruction

Motion analysis systems are widely employed to identify movement deficiencies-e.g. patterns that potentially increase the risk of injury or inhibit performance. However, findings across studies are often conflicting in respect to what a movement deficiency is or the magnitude of association to a specific injury. This study tests the information content within movement data using a data driven framework that was taught to classify movement data into the classes: NORM, ACLOP and ACLNO OP, without the input of expert knowledge. The NORM class was presented by 62 subjects (124 NORM limbs), while 156 subjects with ACL reconstruction represented the ACLOP and ACLNO OP class (156 limbs each class). Movement data from jumping, hopping and change of direction exercises were examined, using a variety of machine learning techniques. A stratified shuffle split cross-validation was used to obtain a measure of expected accuracy for each step within the analysis. Classification accuracies (from best performing classifiers) ranged from 52 to 81%, using up to 5 features. The exercise with the highest classification accuracy was the double leg drop jump (DLDJ; 81%), the highest classification accuracy when considering only the NORM class was observed in the single leg hop (81%), while the DLDJ demonstrated the highest classification accuracy when considering only for the ACLOP and ACLNO OP class (84%). These classification accuracies demonstrate that biomechanical data contains valuable information and that it is possible to differentiate normal from rehabilitating movement patterns. Further, findings highlight that a few features contain most of the information, that it is important to seek to understand what a classification model has learned, that symmetry measures are important, that exercises capture different qualities and that not all subjects within a normative cohort utilise 'true' normative movement patterns (only 27 to 71%).

An environmental cleaning bundle and health-care-associated infections in hospitals (REACH): a multicentre, randomised trial

BACKGROUND

The hospital environment is a reservoir for the transmission of microorganisms. The effect of improved cleaning on patient-centred outcomes remains unclear. We aimed to evaluate the effectiveness of an environmental cleaning bundle to reduce health care-associated infections in hospitals.

METHODS

The REACH study was a pragmatic, multicentre, randomised trial done in 11 acute care hospitals in Australia. Eligible hospitals had an intensive care unit, were classified by the National Health Performance Authority as a major hospital (public hospitals) or having more than 200 inpatient beds (private hospitals), and had a health-care-associated infection surveillance programme. The stepped-wedge design meant intervention periods varied from 20 weeks to 50 weeks. We introduced the REACH cleaning bundle, a multimodal intervention, focusing on optimising product use, technique, staff training, auditing with feedback, and communication, for routine cleaning. The primary outcomes were incidences of health-care-associated Staphylococcus aureus bacteraemia, Clostridium difficile infection, and vancomycin-resistant enterococci infection. The secondary outcome was the thoroughness of cleaning of frequent touch points, assessed by a fluorescent marking gel. This study is registered with the Australian and New Zealand Clinical Trial Registry, number ACTRN12615000325505.

FINDINGS

Between May 9, 2016, and July 30, 2017, we implemented the cleaning bundle in 11 hospitals. In the pre-intervention phase, there were 230 cases of vancomycin-resistant enterococci infection, 362 of S aureus bacteraemia, and 968 C difficile infections, for 3 534 439 occupied bed-days. During intervention, there were 50 cases of vancomycin-resistant enterococci infection, 109 of S aureus bacteraemia, and 278 C difficile infections, for 1 267 134 occupied bed-days. After the intervention, vancomycin-resistant enterococci infections reduced from 0·35 to 0·22 per 10 000 occupied bed-days (relative risk 0·63, 95% CI 0·41-0·97, p=0·0340). The incidences of S aureus bacteraemia (0·97 to 0·80 per 10 000 occupied bed-days; 0·82, 0·60-1·12, p=0·2180) and C difficile infections (2·34 to 2·52 per 10 000 occupied bed-days; 1·07, 0·88-1·30, p=0·4655) did not change significantly. The intervention increased the percentage of frequent touch points cleaned in bathrooms from 55% to 76% (odds ratio 2·07, 1·83-2·34, p<0·0001) and bedrooms from 64% to 86% (1·87, 1·68-2·09, p<0·0001).

INTERPRETATION

The REACH cleaning bundle was successful at improving cleaning thoroughness and showed great promise in reducing vancomycin-resistant enterococci infections. Our work will inform hospital cleaning policy and practice, highlighting the value of investment in both routine and discharge cleaning practice.

FUNDING

National Health and Medical Research Council (Australia).

Best Practices for Quantification of Uncertainty and Sampling Quality in Molecular Simulations [Article v1.0]

The quantitative assessment of uncertainty and sampling quality is essential in molecular simulation. Many systems of interest are highly complex, often at the edge of current computational capabilities. Modelers must therefore analyze and communicate statistical uncertainties so that "consumers" of simulated data understand its significance and limitations. This article covers key analyses appropriate for trajectory data generated by conventional simulation methods such as molecular dynamics and (single Markov chain) Monte Carlo. It also provides guidance for analyzing some 'enhanced' sampling approaches. We do not discuss systematic errors arising, e.g., from inaccuracy in the chosen model or force field.

Fingerprinting Acute Digestive Diseases by Untargeted NMR Based Metabolomics

Precision medicine may significantly contribute to rapid disease diagnosis and targeted therapy, but relies on the availability of detailed, subject specific, clinical information. Proton nuclear magnetic resonance (¹H–NMR) spectroscopy of body fluids can extract individual metabolic fingerprints. Herein, we studied 64 patients admitted to the Florence main hospital emergency room with severe abdominal pain. A blood sample was drawn from each patient at admission, and the corresponding sera underwent ¹H–NMR metabolomics fingerprinting. Unsupervised Principal Component Analysis (PCA) analysis showed a significant discrimination between a group of patients with symptoms of upper abdominal pain and a second group consisting of patients with diffuse abdominal/intestinal pain. Prompted by this observation, supervised statistical analysis (Orthogonal Partial Least Squares–Discriminant Analysis (OPLS-DA)) showed a very good discrimination (>90%) between the two groups of symptoms. This is a surprising finding, given that neither of the two symptoms points directly to a specific disease among those studied here. Actually herein, upper abdominal pain may result from either symptomatic gallstones, cholecystitis, or pancreatitis, while diffuse abdominal/intestinal pain may result from either intestinal ischemia, strangulated obstruction, or mechanical obstruction. Although limited by the small number of samples from each of these six conditions, discrimination of these diseases was attempted. In the first symptom group, >70% discrimination accuracy was obtained among symptomatic gallstones, pancreatitis, and cholecystitis, while for the second symptom group >85% classification accuracy was obtained for intestinal ischemia, strangulated obstruction, and mechanical obstruction. No single metabolite stands up as a possible biomarker for any of these diseases, while the contribution of the whole ¹H–NMR serum fingerprint seems to be a promising candidate, to be confirmed on larger cohorts, as a first-line discriminator for these diseases.

Neural networks of aggression: ALE meta-analyses on trait and elicited aggression

There is considerable evidence that emotion dysregulation and self-control impairments lead to escalated aggression in populations with psychiatric disorders. However, convergent quantitative evidence on the neural network explaining how aggression arises is still lacking. To address this gap, peak activations extracted from extant functional magnetic resonance imaging (fMRI) studies were synthesized through coordinate-based meta-analyses. A systematic search in the PubMed database was conducted and 26 fMRI studies met the inclusion criteria. Three separate activation likelihood estimation (ALE) meta-analyses were performed on (1) individual differences in trait aggression (TA) studies, (2) individual differences in TA studies examining executive functioning, and (3) elicited aggression (EA) studies across fMRI behavioral paradigms. Ensuing clusters from ALE meta-analyses were further treated as seeds for follow-up investigations on consensus connectivity networks (CCN) delineated from meta-analytic connectivity modeling (MACM) and resting-state functional connectivity (RSFC) to further characterize their physiological functions. Finally, we obtained a data-driven functional characterization of the ensuing clusters and their networks. This approach offers a boarder view of the ensuing clusters using a boarder network perspective. In TA, aberrant brain activations were found only in the right precuneus. Follow-up analyses revealed that the precuneus seed was within the frontal-parietal network (FPN) associated with action inhibition, visuospatial processing and higher-level cognition. With further restricting to only experiments examining executive functioning, convergent evidence was found in the right rolandic operculum (RO), midcingulate cortex (MCC), precentral gyrus (PrG) and precuneus. Follow-up analyses suggested that RO, MCC and PrG may belong to a common cognitive control network, while the MCC seems to be the hub of this network. In EA, we only revealed a convergent region in the left postcentral gyrus. Follow-up CCN analyses and functional characterizations suggested that this region may also belong to the same cognitive control network found in the TA sub-analysis. Our results suggested that escalated aggression arises from abnormal precuneus activities within the FPN, disrupting the recruitment of other large-scale networks such as adaptive cognitive control network. Consequently, failure to recruit such a network results in an inability to generate adaptive responses, increasing the likelihood of acting aggressively.

Machine learning in human movement biomechanics: Best practices, common pitfalls, and new opportunities

Traditional laboratory experiments, rehabilitation clinics, and wearable sensors offer biomechanists a wealth of data on healthy and pathological movement. To harness the power of these data and make research more efficient, modern machine learning techniques are starting to complement traditional statistical tools. This survey summarizes the current usage of machine learning methods in human movement biomechanics and highlights best practices that will enable critical evaluation of the literature. We carried out a PubMed/Medline database search for original research articles that used machine learning to study movement biomechanics in patients with musculoskeletal and neuromuscular diseases. Most studies that met our inclusion criteria focused on classifying pathological movement, predicting risk of developing a disease, estimating the effect of an intervention, or automatically recognizing activities to facilitate out-of-clinic patient monitoring. We found that research studies build and evaluate models inconsistently, which motivated our discussion of best practices. We provide recommendations for training and evaluating machine learning models and discuss the potential of several underutilized approaches, such as deep learning, to generate new knowledge about human movement. We believe that cross-training biomechanists in data science and a cultural shift toward sharing of data and tools are essential to maximize the impact of biomechanics research.

Poor statistical reporting, inadequate data presentation and spin persist despite editorial advice

The Journal of Physiology and British Journal of Pharmacology jointly published an editorial series in 2011 to improve standards in statistical reporting and data analysis. It is not known whether reporting practices changed in response to the editorial advice. We conducted a cross-sectional analysis of reporting practices in a random sample of research papers published in these journals before (n = 202) and after (n = 199) publication of the editorial advice. Descriptive data are presented. There was no evidence that reporting practices improved following publication of the editorial advice. Overall, 76-84% of papers with written measures that summarized data variability used standard errors of the mean, and 90-96% of papers did not report exact p-values for primary analyses and post-hoc tests. 76-84% of papers that plotted measures to summarize data variability used standard errors of the mean, and only 2-4% of papers plotted raw data used to calculate variability. Of papers that reported p-values between 0.05 and 0.1, 56-63% interpreted these as trends or statistically significant. Implied or gross spin was noted incidentally in papers before (n = 10) and after (n = 9) the editorial advice was published. Overall, poor statistical reporting, inadequate data presentation and spin were present before and after the editorial advice was published. While the scientific community continues to implement strategies for improving reporting practices, our results indicate stronger incentives or enforcements are needed.

Pacing of Paleozoic macroevolutionary rates by Milankovitch grand cycles

Periodic fluctuations in past biodiversity, speciation, and extinction have been proposed, with extremely long periods ranging from 26 to 62 million years, although forcing mechanisms remain speculative. In contrast, well-understood periodic Milankovitch climate forcing represents a viable driver for macroevolutionary fluctuations, although little evidence for such fluctuation exists except during the Late Cenozoic. The reality, magnitude, and drivers of periodic fluctuations in macroevolutionary rates are of interest given long-standing debate surrounding the relative roles of intrinsic biotic interactions vs. extrinsic environmental factors as drivers of biodiversity change. Here, we show that, over a time span of 60 million years, between 9 and 16% of the variance in biological turnover (i.e., speciation probability plus species extinction probability) in a major Early Paleozoic zooplankton group, the graptoloids, can be explained by long-period astronomical cycles (Milankovitch "grand cycles") associated with Earth's orbital eccentricity (2.6 million years) and obliquity (1.3 million years). These grand cycles modulate climate variability, alternating times of relative stability in the environment with times of maximum volatility. We infer that these cycles influenced graptolite speciation and extinction through climate-driven changes to oceanic circulation and structure. Our results confirm the existence of Milankovitch grand cycles in the Early Paleozoic Era and show that known processes related to the mechanics of the Solar System were shaping marine macroevolutionary rates comparatively early in the history of complex life. We present an application of hidden Markov models to macroevolutionary time series and protocols for the evaluation of statistical significance in spectral analysis.

A Pilot Study on Posterior Polyethylene Tethers to Prevent Proximal Junctional Kyphosis After Multilevel Spinal Instrumentation for Adult Spinal Deformity

BACKGROUND

Proximal junctional kyphosis (PJK) is a common problem after multilevel spine instrumentation.

OBJECTIVE

To determine if junctional tethers reduce PJK after multilevel instrumented surgery for adult spinal deformity (ASD).

METHODS

ASD patients who underwent posterior instrumented fusion were divided into 3 groups: no tether (NT), polyethylene tether-only (TO; tied securely through the spinous processes of the uppermost instrumented vertebra [UIV] + 1 and UIV-1), and tether with crosslink (TC; passed through the spinous process of UIV+1 and tied to a crosslink between UIV-1 and UIV-2). PJK was defined as proximal junctional angle ≥ 10° and ≥ 10° greater than the corresponding preoperative measurement.

RESULTS

One hundred eighty-four (96%) of 191 consecutive patients achieved minimum 3-mo follow-up (mean = 20 mo [range:3-56 mo]; mean age = 66 yr; 67.4% female). There were no significant differences between groups based on demographic, surgical, and sagittal radiographic parameters. PJK rates were 45.3% (29/64), 34.4% (22/64), and 17.9% (10/56) for NT, TO, and TC, respectively. PJK rate for all tethered patients (TO + TC; 26.7% [32/120]) was significantly lower than NT (P = .011). PJK rate for TC was significantly lower than NT (P = .001). Kaplan-Meier analysis showed significant time-dependent PJK reduction for TC vs NT (log rank test, P = .010). Older age and greater change in lumbar lordosis were independent predictors of PJK, while junctional tethers had a significant protective effect.

CONCLUSION

Junctional tethers significantly reduced occurrence of PJK. This difference was progressive from NT to TO to TC, but only reached pairwise significance for NT vs TC. This suggests potential benefit of tethers to reduce PJK, and that future prospective studies are warranted.

Policy-Led Comparative Environmental Risk Assessment of Genetically Modified Crops: Testing for Increased Risk Rather Than Profiling Phenotypes Leads to Predictable and Transparent Decision-Making

We describe two contrasting methods of comparative environmental risk assessment for genetically modified (GM) crops. Both are science-based, in the sense that they use science to help make decisions, but they differ in the relationship between science and policy. Policy-led comparative risk assessment begins by defining what would be regarded as unacceptable changes when the use a particular GM crop replaces an accepted use of another crop. Hypotheses that these changes will not occur are tested using existing or new data, and corroboration or falsification of the hypotheses is used to inform decision-making. Science-led comparative risk assessment, on the other hand, tends to test null hypotheses of no difference between a GM crop and a comparator. The variables that are compared may have little or no relevance to any previously stated policy objective and hence decision-making tends to be ad hoc in response to possibly spurious statistical significance. We argue that policy-led comparative risk assessment is the far more effective method. With this in mind, we caution that phenotypic profiling of GM crops, particularly with omics methods, is potentially detrimental to risk assessment.

Dealing with Confounders in Omics Analysis

The Anna Karenina effect is a manifestation of the theory-practice gap that exists when theoretical statistics are applied on real-world data. In the course of analyzing biological data for differential features such as genes or proteins, it derives from the situation where the null hypothesis is rejected for extraneous reasons (or confounders), rather than because the alternative hypothesis is relevant to the disease phenotype. The mechanics of applying statistical tests therefore must address and resolve confounders. It is inadequate to simply rely on manipulating the P-value. We discuss three mechanistic elements (hypothesis statement construction, null distribution appropriateness, and test-statistic construction) and suggest how they can be designed to foil the Anna Karenina effect to select phenotypically relevant biological features.

Gene annotation bias impedes biomedical research

We found tremendous inequality across gene and protein annotation resources. We observed that this bias leads biomedical researchers to focus on richly annotated genes instead of those with the strongest molecular data. We advocate that researchers reduce these biases by pursuing data-driven hypotheses.

Conceptualizing and evaluating replication across domains of behavioral research

We discuss the authors' conceptualization of replication, in particular the false dichotomy of direct versus conceptual replication intrinsic to it, and suggest a broader one that better generalizes to other domains of psychological research. We also discuss their approach to the evaluation of replication results and suggest moving beyond their dichotomous statistical paradigms and employing hierarchical/meta-analytic statistical models.