Abandon Statistical Significance

The multiplicity of analysis strategies jeopardizes replicability: lessons learned across disciplines

For a given research question, there are usually a large variety of possible analysis strategies acceptable according to the scientific standards of the field, and there are concerns that this multiplicity of analysis strategies plays an important role in the non-replicability of research findings. Here, we define a general framework on common sources of uncertainty arising in computational analyses that lead to this multiplicity, and apply this framework within an overview of approaches proposed across disciplines to address the issue. Armed with this framework, and a set of recommendations derived therefrom, researchers will be able to recognize strategies applicable to their field and use them to generate findings more likely to be replicated in future studies, ultimately improving the credibility of the scientific process.

The Impact of Coronavirus (COVID-19) Related Public-Health Measures on Training Behaviours of Individuals Previously Participating in Resistance Training: A Cross-Sectional Survey Study

Introduction

Understanding the impact of lockdown upon resistance training (RT), and how people adapted their RT behaviours, has implications for strategies to maintain engagement in similar positive health behaviours. Further, doing so will provide a baseline for investigation of the long-term effects of these public health measures upon behaviours and perceptions, and facilitate future follow-up study.

Objectives

To determine how the onset of coronavirus (COVID-19), and associated ‘lockdown’, affected RT behaviours, in addition to motivation, perceived effectiveness, enjoyment, and intent to continue, in those who regularly performed RT prior to the pandemic.

Methods

We conducted an observational, cross-sectional study using online surveys in multiple languages (English, Danish, French, German, Italian, Portuguese, Slovakian, Swedish, and Japanese) distributed across social media platforms and through authors’ professional and personal networks. Adults (n = 5389; median age = 31 years [interquartile range (IQR) = 25, 38]), previously engaged in RT prior to lockdown (median prior RT experience = 7 years [IQR = 4, 12]) participated. Outcomes were self-reported RT behaviours including: continuation of RT during lockdown, location of RT, purchase of specific equipment for RT, method of training, full-body or split routine, types of training, repetition ranges, exercise number, set volumes (per exercise and muscle group), weekly frequency of training, perception of effort, whether training was planned/recorded, time of day, and training goals. Secondary outcomes included motivation, perceived effectiveness, enjoyment, and intent to continue RT.

Results

A majority of individuals (82.8%) maintained participation in RT during-lockdown. Marginal probabilities from generalised linear models and generalised estimating equations for RT behaviours were largely similar from pre- to during-lockdown. There was reduced probability of training in privately owned gyms (~ 59% to ~ 7%) and increased probability of training at home (~ 18% to ~ 89%); greater probability of training using a full-body routine (~ 38% to ~ 51%); reduced probability of resistance machines (~ 66% to ~ 13%) and free weight use (~ 96% to ~ 81%), and increased probability of bodyweight training (~ 62% to ~ 82%); reduced probability of moderate repetition ranges (~ 62–82% to ~ 55–66%) and greater probability of higher repetition ranges (~ 27% to ~ 49%); and moderate reduction in the perception of effort experienced during-training (r = 0.31). Further, individuals were slightly less likely to plan or record training during lockdown and many changed their training goals. Additionally, perceived effectiveness, enjoyment, and likelihood of continuing current training were all lower during-lockdown.

Conclusions

Those engaged in RT prior to lockdown these behaviours with only slight adaptations in both location and types of training performed. However, people employed less effort, had lower motivation, and perceived training as less effective and enjoyable, reporting their likelihood of continuing current training was similar or lower than pre-lockdown. These results have implications for strategies to maintain engagement in positive health behaviours such as RT during-restrictive pandemic-related public health measures.

Pre-registration

https://osf.io/qcmpf.

Preprint

The preprint version of this work is available on SportRχiv: https://osf.io/preprints/sportrxiv/b8s7e/.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40279-021-01438-5.

Statistische Signifikanz – was der p-Wert aussagt

Der Begriff der statistischen Signifikanz geht oft mit Missverständnissen einher. Dies liegt u. a. daran, dass sich mit den theoretischen Konzepten und Kenngrößen, die hinter dem Signifikanztesten liegen, zu wenig auseinandergesetzt wird. Hierzu gehört beispielsweise der p-Wert, der aussagt, mit welcher Wahrscheinlichkeit der beobachtete oder ein extremerer Effekt gefunden würde, wenn dieser Effekt in Wirklichkeit gar nicht existieren würde. Anhand des Beispiels eines Münzwurfs soll das Konzept des Nullhypothesensignifikanztestens erläutert werden, um ein besseres Verständnis für die Interpretation von wissenschaftlichen Ergebnissen zu erzeugen.

Non-local Muscle Fatigue Effects on Muscle Strength, Power, and Endurance in Healthy Individuals: A Systematic Review with Meta-analysis

BACKGROUND

The fatigue of a muscle or muscle group can produce global responses to a variety of systems (i.e., cardiovascular, endocrine, and others). There are also reported strength and endurance impairments of non-exercised muscles following the fatigue of another muscle; however, the literature is inconsistent.

OBJECTIVE

To examine whether non-local muscle fatigue (NLMF) occurs following the performance of a fatiguing bout of exercise of a different muscle(s).

DESIGN

Systematic review and meta-analysis.

SEARCH AND INCLUSION

A systematic literature search using a Boolean search strategy was conducted with PubMed, SPORTDiscus, Web of Science, and Google Scholar in April 2020, and was supplemented with additional 'snowballing' searches up to September 2020. To be included in our analysis, studies had to include at least one intentional performance measure (i.e., strength, endurance, or power), which if reduced could be considered evidence of muscle fatigue, and also had to include the implementation of a fatiguing protocol to a location (i.e., limb or limbs) that differed to those for which performance was measured. We excluded studies that measured only mechanistic variables such as electromyographic activity, or spinal/supraspinal excitability. After search and screening, 52 studies were eligible for inclusion including 57 groups of participants (median sample = 11) and a total of 303 participants.

RESULTS

The main multilevel meta-analysis model including all effects sizes (278 across 50 clusters [median = 4, range = 1 to 18 effects per cluster) revealed a trivial point estimate with high precision for the interval estimate [- 0.02 (95% CIs = - 0.14 to 0.09)], yet with substantial heterogeneity (Q₍₂₇₇₎ = 642.3, p < 0.01), I² = 67.4%). Subgroup and meta-regression analyses showed that NLMF effects were not moderated by study design (between vs. within-participant), homologous vs. heterologous effects, upper or lower body effects, participant training status, sex, age, the time of post-fatigue protocol measurement, or the severity of the fatigue protocol. However, there did appear to be an effect of type of outcome measure where both strength [0.11 (95% CIs = 0.01-0.21)] and power outcomes had trivial effects [- 0.01 (95% CIs = - 0.24 to 0.22)], whereas endurance outcomes showed moderate albeit imprecise effects [- 0.54 (95% CIs = - 0.95 to - 0.14)].

CONCLUSIONS

Overall, the findings do not support the existence of a general NLMF effect; however, when examining specific types of performance outcomes, there may be an effect specifically upon endurance-based outcomes (i.e., time to task failure). However, there are relatively fewer studies that have examined endurance effects or mechanisms explaining this possible effect, in addition to fewer studies including women or younger and older participants, and considering causal effects of prior training history through the use of longitudinal intervention study designs. Thus, it seems pertinent that future research on NLMF effects should be redirected towards these still relatively unexplored areas.

Childhood maltreatment and cognitive functioning: the role of depression, parental education, and polygenic predisposition

Childhood maltreatment is associated with cognitive deficits that in turn have been predictive for therapeutic outcome in psychiatric patients. However, previous studies have either investigated maltreatment associations with single cognitive domains or failed to adequately control for confounders such as depression, socioeconomic environment, and genetic predisposition. We aimed to isolate the relationship between childhood maltreatment and dysfunction in diverse cognitive domains, while estimating the contribution of potential confounders to this relationship, and to investigate gene-environment interactions. We included 547 depressive disorder and 670 healthy control participants (mean age: 34.7 years, SD = 13.2). Cognitive functioning was assessed for the domains of working memory, executive functioning, processing speed, attention, memory, and verbal intelligence using neuropsychological tests. Childhood maltreatment and parental education were assessed using self-reports, and psychiatric diagnosis was based on DSM-IV criteria. Polygenic scores for depression and for educational attainment were calculated. Multivariate analysis of cognitive domains yielded significant associations with childhood maltreatment (η²p = 0.083, P < 0.001), depression (η²p = 0.097, P < 0.001), parental education (η²p = 0.085, P < 0.001), and polygenic scores for depression (η²p = 0.021, P = 0.005) and educational attainment (η²p = 0.031, P < 0.001). Each of these associations remained significant when including all of the predictors in one model. Univariate tests revealed that maltreatment was associated with poorer performance in all cognitive domains. Thus, environmental, psychopathological, and genetic risk factors each independently affect cognition. The insights of the current study may aid in estimating the potential impact of different loci of interventions for cognitive dysfunction. Future research should investigate if customized interventions, informed by individual risk profiles and related cognitive preconditions, might enhance response to therapeutic treatments.

Effect of an emergency department education and empowerment intervention on uncontrolled hypertension in a predominately minority population: The AHEAD2 randomized clinical pilot trial

OBJECTIVE

To determine whether an emergency department (ED) education and empowerment intervention coupled with early risk assessment can help improve blood pressure (BP) in a high-risk population.

METHODS

A hypertension emergency department intervention aimed at decreasing disparities (AHEAD2) is a 3-arm, single-site randomized pilot trial for feasibility in an urban academic ED. A total of 150 predominantly ethnic minorities with no primary care provider and severely elevated blood pressure (BP) (≥160/100 mm Hg) were enrolled over 10 months. Participants were randomized into 1 of 3 study arms: (1) enhanced usual care (EUC), (2) ED-initiated screening, brief intervention, and referral for treatment (ED-SBIRT), or (3) ED- SBIRT plus a 48-72 hours post-acute care hypertension transition clinic (ED-SBIRT+PACHT-c). Primary outcomes were change in systolic and diastolic BP (SBP and DBP) from baseline to 9 months. Secondary outcomes were BP control (BP <140/90 mm Hg), changes in hypertension knowledge, medication adherence, and limited bedside echocardiogram (LBE) findings.

RESULTS

SBP reduction from baseline to month 9 was -26.8 (95% confidence interval [CI]: -32.8, -20.7) mm Hg for ED-SBIRT, -23.4 (95% CI: -29.5, -17.3) mm Hg for ED-SBIRT+PACHT-c, and -18.9 (95% CI: -24.9, -12.9) mm Hg for EUC. DBP decreased by -12.5 (95% CI: -16.1, -9.0) mm Hg for ED-SBIRT, -11.3 (95% CI: -14.8, -7.7) mm Hg for ED-SBIRT+PACHT-c, and -8.4 (95% CI: -11.9, -4.9) mm Hg for EUC. A multicomponent intervention compared with EUC resulted in SBP decrease of -7.9 mm Hg (95% CI: -16.4, 0.6). At 9 months, hypertension was controlled for 29.3% (95% CI: 20.3, 38.3) of intervention and 23.5% (95% CI: 11.9, 35.2) of EUC participants. All groups saw improvements in hypertension knowledge, medication adherence, and LBEs, with greater improvements in intervention groups.

CONCLUSIONS

The study findings suggest that a multicomponent intervention comprising of ED education and empowerment coupled with early risk assessment may help improve BP in a high-risk population.

Understanding predictors of change in a day treatment setting for non-suicidal self-injury

OBJECTIVES

To examine change in non-suicidal self-injury (NSSI) frequency, quality of life, and functional impairment from admission to discharge in patients enrolled in partial hospitalization and intensive outpatient programmes (PHP/IOP) designed to treat NSSI. Demographic, clinical, and treatment-related predictors of changes were also examined.

DESIGN

Data were collected as part of routine clinical assessment procedures at admission and discharge from patients enrolled in a PHP/IOP programme designed to treat NSSI. The clinical assessment included measures examining quality of life, functional impairment, and NSSI behaviour.

METHODS

Paired t-tests were used to examine change in NSSI frequency, quality of life, and functional impairment. Reliable clinical change analyses were used to identify clinically significant change in quality of life and functional impairment. Multilevel mixed-effects regression was used to examine predictors of change for quality of life and functional impairment. Negative binomial regression was used to examine predictors of change for NSSI frequency.

RESULTS

From admission to discharge, NSSI frequency significantly decreased and quality of life and functional impairment evidenced clinically significant change. Age, race/ethnicity, and insurance type predicted change in functional impairment, while gender predicted change in quality of life. Urge to self-injure predicted change in NSSI frequency. Borderline symptoms predicted change across all outcome variables.

CONCLUSIONS

Patients who completed a day treatment programme for NSSI evidenced significant change in NSSI frequency, functional impairment, and quality of life at discharge; however, several demographic and clinical variables were associated with change.

PRACTITIONER POINTS

Patients who engage in NSSI show significant change from admission to discharge in a day treatment programme dedicated to the treatment of NSSI. Quality of life and functional impairment are important outcome variables to consider and evaluate in higher levels of care. It is important to consider demographic and clinical variables when creating a treatment plan for NSSI. Although BPD symptoms may be important to consider in day treatment for NSSI, interpersonal dysfunction, depressive symptoms, and mood lability may also affect change in symptoms.

Explaining Discrepancies Between Total and Segmental DXA and BIA Body Composition Estimates Using Bayesian Regression

INTRODUCTION/BACKGROUND

Few investigations have sought to explain discrepancies between dual-energy X-ray absorptiometry (DXA) and bioelectrical impedance analysis (BIA) body composition estimates. The purpose of this analysis was to explore physiological and anthropometric predictors of discrepancies between DXA and BIA total and segmental body composition estimates.

METHODOLOGY

Assessments via DXA (GE Lunar Prodigy) and single-frequency BIA (RJL Systems Quantum V) were performed in 179 adults (103 F, 76 M, age: 33.6 ± 15.3 yr; BMI: 24.9 ± 4.3 kg/m²). Potential predictor variables for differences between DXA and BIA total and segmental fat mass (FM) and lean soft tissue (LST) estimates were obtained from demographics and laboratory techniques, including DXA, BIA, bioimpedance spectroscopy, air displacement plethysmography, and 3-dimensional optical scanning. To determine meaningful predictors, Bayesian robust regression models were fit using a t-distribution and regularized hierarchical shrinkage "horseshoe" prior. Standardized model coefficients (β) were generated, and leave-one-out cross validation was used to assess model predictive performance.

RESULTS

LST hydration (i.e., total body water:LST) was a predictor of discrepancies in all FM and LST variables (|β|: 0.20-0.82). Additionally, extracellular fluid percentage was a predictor for nearly all outcomes (|β|: 0.19-0.40). Height influenced the agreement between whole-body estimates (|β|: 0.74-0.77), while the mass, length, and composition of body segments were predictors for segmental LST estimates (|β|: 0.23-3.04). Predictors of segmental FM errors were less consistent. Select sex-, race-, or age-based differences between methods were observed. The accuracy of whole-body models was superior to segmental models (leave-one-out cross-validation-adjusted R² of 0.83-0.85 for FM_TOTAL and LST_TOTAL vs. 0.20-0.76 for segmental estimates). For segmental models, predictive performance decreased in the order of: appendicular lean soft tissue, LST_LEGS, LST_TRUNK and FM_LEGS, FM_ARMS, FM_TRUNK, and LST_ARMS.

CONCLUSIONS

These findings indicate the importance of LST hydration, extracellular fluid content, and height for explaining discrepancies between DXA and BIA body composition estimates. These general findings and quantitative interpretation based on the presented data allow for a better understanding of sources of error between 2 popular segmental body composition techniques and facilitate interpretation of estimates from these technologies.

A Modified Sequential Probability Ratio Test

We describe a modified sequential probability ratio test that can be used to reduce the average sample size required to perform statistical hypothesis tests at specified levels of significance and power. Examples are provided for z tests, t tests, and tests of binomial success probabilities. A description of a software package to implement the test designs is provided. We compare the sample sizes required in fixed design tests conducted at 5% significance levels to the average sample sizes required in sequential tests conducted at 0.5% significance levels, and we find that the two sample sizes are approximately equal.

The case for formal methodology in scientific reform

Current attempts at methodological reform in sciences come in response to an overall lack of rigor in methodological and scientific practices in experimental sciences. However, most methodological reform attempts suffer from similar mistakes and over-generalizations to the ones they aim to address. We argue that this can be attributed in part to lack of formalism and first principles. Considering the costs of allowing false claims to become canonized, we argue for formal statistical rigor and scientific nuance in methodological reform. To attain this rigor and nuance, we propose a five-step formal approach for solving methodological problems. To illustrate the use and benefits of such formalism, we present a formal statistical analysis of three popular claims in the metascientific literature: (i) that reproducibility is the cornerstone of science; (ii) that data must not be used twice in any analysis; and (iii) that exploratory projects imply poor statistical practice. We show how our formal approach can inform and shape debates about such methodological claims.

Hierarchical Models for International Comparisons: Smoking, Disability, and Social Inequality in 21 European Countries

BACKGROUND

International comparisons of social inequalities in health outcomes and behaviors are challenging. Due to the level of disaggregation often required, data can be sparse and methods to make adequately powered comparisons are lacking. We aimed to illustrate the value of a hierarchical Bayesian approach that partially pools country-level estimates, reducing the influence of sampling variation and increasing the stability of estimates. We also illustrate a new way of simultaneously displaying the uncertainty of both relative and absolute inequality estimates.

METHODS

We used the 2014 European Social Survey to estimate smoking prevalence, absolute, and relative inequalities for men and women with and without disabilities in 21 European countries. We simultaneously display smoking prevalence for people without disabilities (x-axis), absolute (y-axis), and relative inequalities (contour lines), capturing the uncertainty of these estimates by plotting a 2-D normal approximation of the posterior distribution from the full probability (Bayesian) analysis.

RESULTS

Our study confirms that across Europe smoking prevalence is generally higher for people with disabilities than for those without. Our model shifts more extreme prevalence estimates that are based on fewer observations, toward the European mean.

CONCLUSIONS

We demonstrate the utility of partial pooling to make adequately powered estimates of inequality, allowing estimates from countries with smaller sample sizes to benefit from the increased precision of the European average. Including uncertainty on our inequality plot provides a useful tool for evaluating both the geographical patterns of variation in, and strength of evidence for, differences in social inequalities in health.

A Cautionary Note on Estimating Effect Size

An increasingly popular approach to statistical inference is to focus on the estimation of effect size. Yet this approach is implicitly based on the assumption that there is an effect while ignoring the null hypothesis that the effect is absent. We demonstrate how this common null-hypothesis neglect may result in effect size estimates that are overly optimistic. As an alternative to the current approach, a spike-and-slab model explicitly incorporates the plausibility of the null hypothesis into the estimation process. We illustrate the implications of this approach and provide an empirical example.

Computer-delivered personalized feedback intervention for hazardous drinkers with elevated anxiety sensitivity: A pilot randomized controlled trial

Hazardous drinkers with emotional vulnerabilities (e.g., elevated anxiety sensitivity) remain an underserved group. This study aimed to evaluate the feasibility, acceptability, and initial efficacy of a single session remotely-delivered personalized feedback intervention (PFI) targeting alcohol (mis)use and anxiety sensitivity among college students. Hazardous drinkers with elevated anxiety sensitivity (N = 125; 76.8% female; M_age = 22.14; 66.4% racial/ethnic minorities) were randomized to receive the integrated PFI (n = 63) or attention control (n = 62). Follow-up assessments were conducted one-week, one-month and three-months post-intervention. Latent growth curve modeling was used to test pilot outcomes. It was feasible to recruit and retain hazardous drinking students with elevated anxiety sensitivity through follow-up with no group differences in retention. The integrated PFI was rated as more acceptable than the control with medium/large differences (p's < 0.004; d's = 0.54-0.80). The integrated PFI group had statistically significantly greater change in primary outcomes: motivation, hazardous alcohol use, and anxiety sensitivity (p's < 0.05; d's = 0.08-0.37) with larger within-group effect sizes (d's = 0.48-0.61) than in control (d's = 0.26-0.54). Despite a small sample size, this one-session intervention offers promise among a high-risk group of drinkers with emotional vulnerabilities. The computer-based format may allow for mass distribution of a low-cost intervention in the future; however, follow-up testing in larger samples is needed.

How fragile are Mediterranean diet interventions? A research-on-research study of randomised controlled trials

INTRODUCTION

The Mediterranean diet (MD) is a traditional regional dietary pattern and a healthy diet recommended for the primary and secondary prevention of various diseases and health conditions. Results from the higher level of primary evidence, namely randomised controlled trials (RCTs), are often used to produce dietary recommendations; however, the robustness of RCTs with MD interventions is unknown.

METHODS

A systematic search was conducted and all MD RCTs with dichotomous primary outcomes were extracted from PubMed. The fragility (FI) and the reverse fragility index (RFI) were calculated for the trials with significant and non-significant comparisons, respectively.

RESULTS

Out of 27 RCTs of parallel design, the majority failed to present a significant primary outcome, exhibiting an FI equal to 0. The median FI of the significant comparisons was 5, ranging between 1 and 39. More than half of the comparisons had an FI <5, indicating that the addition of 1-4 events to the treatment arm eliminated the statistical significance. For the comparisons with an FI=0, the RFI ranged between 1 and 29 (Median RFI: 7). When the included RCTs were stratified according to masking, the use of a composite primary endpoint, sample size, outcome category, or dietary adherence assessment method, no differences were exhibited in the FI and RFI between groups, except for the RFI among different compliance assessment methods.

CONCLUSIONS

In essence, the present study shows that even in the top tiers of evidence hierarchy, research on the MD may lack robustness, setting concerns for the formulation of nutrition recommendations.

HLA-DRB1 molecules and the presentation of anchor peptides from RhD, RhCE, and KEL proteins

BACKGROUND

Antigens from the Rh and Kell systems are recognized as the most immunogenic in clinical practice. This study evaluated the possible molecular mechanisms involved in the interaction of antigenic peptides with the DRB1 molecules, which help to explain the high frequency of anti-K and association of D + C antibodies in transfusion and incompatible pregnancy.

STUDY DESIGN AND METHODS

We included 201 patients with antibodies against antigens from the Rh and Kell systems and compare them with 174,015 controls. HLA-DRB1 genotyping and in silico analysis were performed. The NetMHCIIpan software was used to identify RhD-, RhCE-, and KEL-derived anchor peptides that bind to DRB1 molecules.

RESULTS

HLA-DRB1*15 is associated with an increased risk of D, C, E, and K alloimmunization, while the HLA-DRB1*01 and *12 alleles are overrepresented in patients with anti-C and anti-D, respectively. In silico analysis showed that three polymorphic points (60I, 68S, and 103S) common to C and D antigens can be presented by several DRB1 molecules, including DRB1*15:01. The DRB1*09:01 molecule, although not showing statistical significance, was able to interact strongly with almost all five anchor peptides from the sequence containing the polymorphic determinants of E antigen, except 217-WMFWPSVNS-225.

CONCLUSION

The DRB1*15 molecule has specific physicochemical characteristics in residues 11P and 13R in the P4 pocket that can favor the response to various antigenic peptides. Anti-K alloimmunization is unrestricted for interaction with specific DRB1 molecules, which suggests that almost all individuals in our population have DRB1 molecules capable of binding to KEL-derived anchor peptides and produce anti-K when stimulated.

The Role of Perceived Energy and Self-Beliefs for Physical Activity and Sports Activity of Patients With Multiple Sclerosis and Chronic Stroke

Physical activity counteracts some of the negative consequences associated with chronic neurological diseases. Here, we describe the levels of physical activity (PA) and sports activity (Sport) in patients with multiple sclerosis (pMS, n = 59) and chronic stroke (pStroke, n = 67) and test compliance with the recommendation for health-promoting physical activity of the World-Health Organization (WHO). Secondly, we tested for differences between the groups of patients, and thirdly, we examined relationships between PA and Sport with psychological indicators of perceived energy (fatigue and vitality) and self-beliefs (self-efficacy and self-control). Psychological constructs were assessed with validated measures from different disciplines in Psychology. A statistical aim was to describe interpretations gained by (non-) parametric Bayesian and Null-Hypothesis-Significance Testing statistics (NHST) on the example of the conducted tests for differences and relationships. Descriptive analyses revealed that pMS and pStroke complied with recommendations of the WHO, but with large variance indicating that patient groups are not homogenous. Tests for differences showed that the PA difference between pMS and pStroke can be attributed to the higher proportion of women in the pMS sample as they engage more in household chores (important part of PA). Tests for relationships showed that for pStroke, vitality, self-control, and self-efficacy were positively related to the level of sports activity. Furthermore, pStroke who were sport active had lower fatigue and higher self-control and self-efficacy scores than sport inactive people. Although they address slightly different questions, the Bayesian and the NHST approach led to similar general conclusions.

Bottleneck analysis: Simple prediction of the precision of a planned case-control or cohort study based on healthcare registers

PURPOSE

In pharmacoepidemiological studies, the precision of effect estimates usually depends on the lowest number in the underlying two by two table. We denote this the "bottleneck count" (BNC). We describe how to translate the BNC into an achievable precision and provide empirical examples.

METHODS

First, we derive a theoretical prediction of the precision in a study where only the BNC determines precision. As an illustration, we calculated the expected precision of a null-effect study on retinoids and peptic ulcer bleeding, expressed as the upper/lower confidence limit ratio (ULCLR). Finally, we reviewed 126 effect estimates from the literature, analyzing the relationship between the predicted and achieved precision.

RESULTS

The log-log transformed ULCLR was shown to be a simple linear function of log(BNC). The expected annual number of retinoid-users experiencing a peptic ulcer bleeding was 9.8, yielding an estimated ULCLR for a 1-year study of 3.84. The literature review showed an inverse linear relationship between the logarithmic BNC and the log-log transformed ULCLR, which was largely independent of study design, effect measure and category of BNC. Achieved precision deviated little from predictions but was usually lower than predicted, particularly with low BNC.

CONCLUSION

The precision of a study can be predicted simply and with good accuracy from the BNC, which is useful for determining whether a study is worth pursuing or not.

The potential for overdiagnosis and underdiagnosis because of blood pressure variability: a comparison of the 2017 ACC/AHA, 2018 ESC/ESH and 2019 NICE hypertension guidelines

OBJECTIVE

To estimate the extent that BP measurement variability may drive over- and underdiagnosis of 'hypertension' when measurements are made according to current guidelines.

METHODS

Using data from the National Health and Nutrition Examination Survey and empirical estimates of within-person variability, we simulated annual SBP measurement sets for 1 000 000 patients over 5 years. For each measurement set, we used an average of multiple readings, as recommended by guidelines.

RESULTS

The mean true SBP for the simulated population was 118.8 mmHg with a standard deviation of 17.5 mmHg. The proportion overdiagnosed with 'hypertension' after five sets of office or nonoffice measurements using the 2017 American College of Cardiology guideline was 3-5% for people with a true SBP less than 120 mmHg, and 65-72% for people with a true SBP 120-130 mmHg. These proportions were less than 1% and 14-33% using the 2018 European Society of Hypertension and 2019 National Institute for Health and Care Excellence guidelines (true SBP <120 and 120-130 mmHg, respectively). The proportion underdiagnosed with 'hypertension' was less than 3% for people with true SBP at least 140 mmHg after one set of office or nonoffice measurements using the 2017 American College of Cardiology guideline, and less than 18% using the other two guidelines.

CONCLUSION

More people are at risk of overdiagnosis under the 2017 American College of Cardiology guideline than the other two guidelines, even if nonoffice measurements are used. Making clinical decisions about cardiovascular prediction based primarily on absolute risk, minimizes the impact of blood pressure variability on overdiagnosis.

A 50 kdyne contusion spinal cord injury with or without the drug SS-31 was not associated with major changes in muscle mass or gene expression 14 d after injury in young male mice

Spinal cord injury (SCI) leads to rapid muscle atrophy due to paralysis/paresis and subsequent disuse. SS-31 is a mitochondrial-targeting peptide that has shown efficacy in protecting skeletal muscle mass and function in non-SCI models of muscle wasting. We aimed to determine if SS-31 could prevent muscle loss after SCI. Male C57BL/6 mice aged 9 weeks underwent sham surgery or 50 kdyne contusion SCI and were administered daily injections of vehicle or 5 mg/kg SS-31 for 14 d. Both SCI groups had sustained losses in body mass compared to Sham animals and ~10% reductions in gastrocnemius, plantaris and tibialis anterior muscle mass after SCI with no clear effect of SS-31. Measurements of protein synthesis in the soleus and plantaris were similar among all groups. mRNA expression of atrophy-associated proinflammatory cytokines was also similar among all groups. There was elevation in MYH7 mRNA and a statistical reduction in MYH2 mRNA expression in the SCI+SS-31 animals compared to Sham animals. There was an SCI-induced reduction in mRNA expression of the E3 ligase FBXO32 (MAFbx), but no effect of SS-31. In summary, a 50 kdyne contusion SCI was able to reduce body mass but was not associated with substantial muscle atrophy or alterations in gene expression profiles associated with muscle health and function 14 d post-injury. SS-31 was not associated with protection against SCI-related changes in body or muscle mass, protein synthesis or gene expression in hindlimb muscles.

Potential Movement Corridors and High Road-Kill Likelihood do not Spatially Coincide for Felids in Brazil: Implications for Road Mitigation

The negative effects of roads on wildlife populations are a growing concern. Movement corridors and road-kill data are typically used to prioritize road segments for mitigation measures. Some research suggests that locations where animals move across roads following corridors coincide with locations where they are often killed by vehicles. Other research indicates that corridors and road-kill rarely occur in the same locations. We compared movement corridor and road mortality models as means of prioritizing road segments for mitigation for five species of felids in Brazil: tiger cats (Leopardus tigrinus and Leopardus guttulus were analyzed together), ocelot (Leopardus pardalis), jaguarundi (Herpailurus yagouaroundi), and puma (Puma concolor). We used occurrence data for each species and applied circuit theory to identify potential movement corridors crossed by roads. We used road-kill records for each species and applied maximum entropy to determine where mortality was most likely to occur on roads. Our findings suggest that movement corridors and high road mortality are not spatially associated. We suggest that differences in the behavioral state of the individuals in the species occurrence and road-kill data may explain these results. We recommend that the road segments for which the results from the two methods agree (~5300 km for all studied species combined at 95th percentile) should be high-priority candidates for mitigation together with road segments identified by at least one method in areas where felids occur in low population densities or are threatened by isolation effects.

A Novel Graphical Method for Data Presentation in Alcohol Systematic Reviews: The Interactive Harvest Plot

AIMS

To demonstrate a novel method for presenting and exploring data in systematic reviews of the alcohol literature.

METHODS

Harvest plots are a graphical method for displaying data on the overall pattern of evidence from a systematic review. They can display the direction of effects and risk of bias within studies for multiple outcomes in a single graphical chart. Using data from our previous meta-analysis on the association between personality disorder and alcohol treatment outcome, we extended the application of harvest plots by developing an interactive online harvest plot application.

RESULTS

Studies included in the review were heterogeneous in design. There were many different primary outcomes, and similar outcomes were often defined differently across studies. The interactive harvest plot allows readers to explore trends in the data across multiple outcomes, including the impact of within-study bias and year of publication. In contrast, meta-analysis on the same data was hampered by a lack of consistency in the way outcomes were measured, and incomplete reporting of effect sizes and their variance. This meant many studies included in the systematic review could not be meta-analysed.

CONCLUSIONS

Interactive harvest plots are a novel graphical method to present data from systematic reviews. They can supplement or even replace meta-analysis when the studies included in a systematic review use heterogeneous designs and measures, as is often the case in the alcohol literature.

Are developmentally missing teeth a predictive risk marker of malignant diseases in non-syndromic individuals? A systematic review

Background:

Different genes and loci that are associated with non-syndromic developmental tooth agenesis (TA) have the same causation pathway in the development of tumours including breast cancer (BC), epithelial ovarian cancer (EOC), colorectal cancer (CRC) and lung cancer (LC).

Objectives:

To assess the link between TA and the development of cancer.

Search sources:

This registered review included a comprehensive search of electronic databases (Cochrane Central Register of Controlled Trials [CENTRAL], LILACS, Scopus, Web of Science and Medline via Ovid) until 1 April 2020, supplemented by manual, grey literature and reference lists search. There was no restriction in term of date of publication, gender, race or type of hypodontia.

Data selection:

The primary outcome was the relationship between TA and cancer. The secondary outcome was to identify the genetic correlation between TA and cancer.

Data extraction:

Study selection, data extraction and risk of bias assessment were performed independently and induplicate by two reviewers, with disputes resolved by a third reviewer.

Results:

Eight studies with a moderate-high risk of bias were included in the final review, with a total of 5821 participants. Due to the heterogeneity among the included studies, the data were presented narratively. Limited studies reported a high prevalence of EOC (19.2%–20%) and CRC (82%–100%) in individuals with TA (depending on the study) compared to those without TA (3% for EOC and 0% for CRC). While others reported a weak correlation between EOC and CRC and TA ( P > 0.05). Weak evidence suggested a strong correlation between breast, cervical uterine and prostate cancers and TA ( P < 0.05).

Conclusions:

Though low-quality evidence suggests a link between TA and cancer, it was not possible to verify that TA can hold a predictive value as a marker for cancers. Further research is needed to confirm the association.

Registration:

PROSPERO (CRD42020139751).

Controllability over stressor decreases responses in key threat-related brain areas

Controllability over stressors has major impacts on brain and behavior. In humans, however, the effect of controllability on responses to stressors is poorly understood. Using functional magnetic resonance imaging (fMRI), we investigated how controllability altered responses to a shock-plus-sound stressor with a between-group yoked design, where participants in controllable and uncontrollable groups experienced matched stressor exposure. Employing Bayesian multilevel analysis at the level of regions of interest and voxels in the insula, and standard voxelwise analysis, we found that controllability decreased stressor-related responses across threat-related regions, notably in the bed nucleus of the stria terminalis and anterior insula. Posterior cingulate cortex, posterior insula, and possibly medial frontal gyrus showed increased responses during control over stressor. Our findings support the idea that the aversiveness of stressors is reduced when controllable, leading to decreased responses across key regions involved in anxiety-related processing, even at the level of the extended amygdala.

Long-Term Effects of Laparoscopic Sleeve Gastrectomy and Roux-en-Y Gastric Bypass on Body Composition and Bone Mass Density

BACKGROUND

Currently, the two most common bariatric procedures are laparoscopic sleeve gastrectomy (LSG) and laparoscopic Roux-en-Y gastric bypass (LRYGB). Long-term data comparing the two interventions in terms of their effect on body composition and bone mass density (BMD) are scarce.

OBJECTIVE

The aim of this study was to assess body composition and BMD at least 5 years after LSG and LRYGB.

SETTING

Department of Endocrinology and Nutrition, St. Claraspital Basel and St. Clara Research Ltd., Basel, Switzerland.

METHODS

Bariatric patients at least 5 years after surgery (LSG or LRYGB) were recruited, and body composition and BMD were measured by means of dual-energy X-ray absorptiometry. Data from body composition before surgery were included in the analysis. Blood samples were taken for determination of plasma calcium, parathyroid hormone, vitamin D3, alkaline phosphatase, and C-terminal telopeptide, and the individual risk for osteoporotic fracture assessed by the Fracture Risk Assessment Tool score was calculated. After surgery, all patients received multivitamins, vitamin D3, and zinc. In addition, LRYGB patients were prescribed calcium.

RESULTS

A total of 142 patients were included, 72 LSG and 70 LRYGB, before surgery: median body mass index 43.1, median age 45.5 years, 62.7% females. Follow-up after a median of 6.7 years. For LRYGB, the percentage total weight loss at follow-up was 26.3% and for LSG 24.1% (p = 0.243). LRYGB led to a slightly lower fat percentage in body composition. At follow-up, 45% of both groups had a T score at the femoral neck below -1, indicating osteopenia. No clinically relevant difference in BMD was found between the groups.

CONCLUSIONS

At 6.7 years after surgery, no difference in body composition and BMD between LRYGB and LSG was found. Deficiencies and bone loss remain an issue after both interventions and should be monitored.

No credible evidence for links between 2D:4D and COVID-19 outcomes: A probabilistic perspective on digit ratio, ACE variants, and national case fatalities

Research into COVID-19 susceptibility and outcomes are critical, but claims must be carefully evaluated to inform policy decisions. In a recent series of articles, Manning and Fink [1-3] use national-level data to describe associations between case-fatality ratios and male and female finger ratios (2D:4D), a suggested measure of prenatal androgen exposure, as well as angiotensin-converting enzyme (ACE) allele and genotype frequencies. The authors suggest that 2D:4D is linked with ACE variant prevalence, and that higher male 2D:4D is associated with higher case fatality ratios, and point to 2D:4D as a useful prognostic measure for COVID-19 susceptibility. A critical review and robust Bayesian analysis of the hypothesis is described here, finding no conclusive evidence of COVID-19 mortality and 2D:4D, nor associations between 2D:4D and ACE1 allele or ACE2 genotype frequency. This absence of evidence is present for data taken from the second wave of COVID-19 in October 2020. Problematic theoretical grounding, individual-level conclusions drawn from national-level data, and issues with statistical inference in the original articles are discussed. Taken together, the current data offer no clear utility of 2D:4D in determining COVID-19 outcomes.

Personalized beyond Precision: Designing Unbiased Gold Standards to Improve Single-Subject Studies of Personal Genome Dynamics from Gene Products

Background: Developing patient-centric baseline standards that enable the detection of clinically significant outlier gene products on a genome-scale remains an unaddressed challenge required for advancing personalized medicine beyond the small pools of subjects implied by "precision medicine". This manuscript proposes a novel approach for reference standard development to evaluate the accuracy of single-subject analyses of transcriptomes and offers extensions into proteomes and metabolomes. In evaluation frameworks for which the distributional assumptions of statistical testing imperfectly model genome dynamics of gene products, artefacts and biases are confounded with authentic signals. Model confirmation biases escalate when studies use the same analytical methods in the discovery sets and reference standards. In such studies, replicated biases are confounded with measures of accuracy. We hypothesized that developing method-agnostic reference standards would reduce such replication biases. We propose to evaluate discovery methods with a reference standard derived from a consensus of analytical methods distinct from the discovery one to minimize statistical artefact biases. Our methods involve thresholding effect-size and expression-level filtering of results to improve consensus between analytical methods. We developed and released an R package "referenceNof1" to facilitate the construction of robust reference standards. Results: Since RNA-Seq data analysis methods often rely on binomial and negative binomial assumptions to non-parametric analyses, the differences create statistical noise and make the reference standards method dependent. In our experimental design, the accuracy of 30 distinct combinations of fold changes (FC) and expression counts (hereinafter "expression") were determined for five types of RNA analyses in two different datasets. This design was applied to two distinct datasets: Breast cancer cell lines and a yeast study with isogenic biological replicates in two experimental conditions. Furthermore, the reference standard (RS) comprised all RNA analytical methods with the exception of the method testing accuracy. To mitigate biases towards a specific analytical method, the pairwise Jaccard Concordance Index between observed results of distinct analytical methods were calculated for optimization. Optimization through thresholding effect-size and expression-level reduced the greatest discordances between distinct methods' analytical results and resulted in a 65% increase in concordance. Conclusions: We have demonstrated that comparing accuracies of different single-subject analysis methods for clinical optimization in transcriptomics requires a new evaluation framework. Reliable and robust reference standards, independent of the evaluated method, can be obtained under a limited number of parameter combinations: Fold change (FC) ranges thresholds, expression level cutoffs, and exclusion of the tested method from the RS development process. When applying anticonservative reference standard frameworks (e.g., using the same method for RS development and prediction), most of the concordant signal between prediction and Gold Standard (GS) cannot be confirmed by other methods, which we conclude as biased results. Statistical tests to determine DEGs from a single-subject study generate many biased results requiring subsequent filtering to increase reliability. Conventional single-subject studies pertain to one or a few patient's measures over time and require a substantial conceptual framework extension to address the numerous measures in genome-wide analyses of gene products. The proposed referenceNof1 framework addresses some of the inherent challenges for improving transcriptome scale single-subject analyses by providing a robust approach to constructing reference standards.

"Just One More Rep!" - Ability to Predict Proximity to Task Failure in Resistance Trained Persons

In resistance training, the use of predicting proximity to momentary task failure (MF, i.e., maximum effort), and repetitions in reserve scales specifically, is a growing approach to monitoring and controlling effort. However, its validity is reliant upon accuracy in the ability to predict MF which may be affected by congruence of the perception of effort compared with the actual effort required. The present study examined participants with at least 1 year of resistance training experience predicting their proximity to MF in two different experiments using a deception design. Within each experiment participants performed four trials of knee extensions with single sets (i.e., bouts of repetitions) to their self-determined repetition maximum (sdRM; when they predicted they could not complete the next repetition if attempted and thus would reach MF if they did) and MF (i.e., where despite attempting to do so they could not complete the current repetition). For the first experiment (n = 14) participants used loads equal to 70% of a one repetition maximum (1RM; i.e., the heaviest load that could be lifted for a single repetition) performed in a separate baseline session. Aiming to minimize participants between day variability in repetition performances, in the second separate experiment (n = 24) they used loads equal to 70% of their daily isometric maximum voluntary contraction (MVC). Results suggested that participants typically under predicted the number of repetitions they could perform to MF with a meta-analytic estimate across experiments of 2.0 [95%CIs 0.0 to 4.0]. Participants with at least 1 year of resistance training experience are likely not adequately accurate at gauging effort in submaximal conditions. This suggests that perceptions of effort during resistance training task performance may not be congruent with the actual effort required. This has implications for controlling, programming, and manipulating the actual effort in resistance training and potentially on the magnitude of desired adaptations such as improvements in muscular hypertrophy and strength.

The generalizability crisis

Most theories and hypotheses in psychology are verbal in nature, yet their evaluation overwhelmingly relies on inferential statistical procedures. The validity of the move from qualitative to quantitative analysis depends on the verbal and statistical expressions of a hypothesis being closely aligned - that is, that the two must refer to roughly the same set of hypothetical observations. Here, I argue that many applications of statistical inference in psychology fail to meet this basic condition. Focusing on the most widely used class of model in psychology - the linear mixed model - I explore the consequences of failing to statistically operationalize verbal hypotheses in a way that respects researchers' actual generalization intentions. I demonstrate that although the "random effect" formalism is used pervasively in psychology to model intersubject variability, few researchers accord the same treatment to other variables they clearly intend to generalize over (e.g., stimuli, tasks, or research sites). The under-specification of random effects imposes far stronger constraints on the generalizability of results than most researchers appreciate. Ignoring these constraints can dramatically inflate false-positive rates, and often leads researchers to draw sweeping verbal generalizations that lack a meaningful connection to the statistical quantities they are putatively based on. I argue that failure to take the alignment between verbal and statistical expressions seriously lies at the heart of many of psychology's ongoing problems (e.g., the replication crisis), and conclude with a discussion of several potential avenues for improvement.

Risk Factors, Management, and Outcomes of Legionella Pneumonia in a Large, Nationally Representative Sample

BACKGROUND

American Thoracic Society/Infectious Diseases Society of America guidelines recommend against routine Legionella pneumophila testing, but recommend that hospitalized patients with community-acquired pneumonia receive empiric treatment covering Legionella. Testing, empiric treatment, and outcomes for patients with Legionella have not been well described.

RESEARCH QUESTION

Is testing for Legionella pneumophila appropriate, and could it impact treatment?

STUDY DESIGN AND METHODS

We conducted a large retrospective cohort analysis using Premier Healthcare Database data from 2010 to 2015. We included adults with a principal diagnosis code for pneumonia (or a principal diagnosis of respiratory failure or sepsis with secondary diagnosis of pneumonia) if they also received treatment for pneumonia on hospital days 1-3. We categorized Legionella-tested patients by test result, identified patient characteristics associated with testing and test result, and examined seasonal and regional patterns of Legionella pneumonia (LP) diagnoses. Empiric therapy for LP was defined as a macrolide, quinolone, or doxycycline, administered on each of the first two hospital days.

RESULTS

Of 166,689 eligible patients, 43,070 (26%) were tested for Legionella, and 642 (1.5%) tested positive. Although only 36% of tests were ordered from June to October, 70% of positive test results occurred during this time. Only 30% of patients with hyponatremia, 32% with diarrhea, and 27% in the ICU were tested. Of patients with positive test results, 495 of 642 (77%) had received empiric Legionella therapy. Patients with LP did not have more severe presentation. They had more frequent late decompensation, but similar mortality to patients without LP.

INTERPRETATION

Legionella is an uncommon cause of community-acquired pneumonia, occurring primarily from late spring through early autumn. Testing is uncommon, even among patients with risk factors, and many patients with positive test results failed to receive empiric coverage for LP.

Type and amount of help as predictors for impression of helpers

Impression of helpers can vary as a function of the magnitude of helping (amount of help) and of situational and motivational aspects (type of help). Over three studies conducted in Sweden and the US, we manipulated both the amount and the type of help in ten diverse vignettes and measured participants' impressions of the described helpers. Impressions were almost unaffected when increasing the amount of help by 500%, but clearly affected by several type of help-manipulations. Particularly, helpers were less positively evaluated if they had mixed motives for helping, did not experience intense emotions or empathy, or if helping involved no personal sacrifice. In line with the person-centered theory of moral judgment, people seem to form impressions of helpers primarily based on the presumed underlying processes and motives of prosociality rather than its consequences.

Effects of age on psychophysical measures of auditory temporal processing and speech reception at low and high levels

•

We found little evidence of greater age-related hearing declines at high sound levels.

•

There are age-related temporal-processing declines independent of hearing loss.

•

No evidence of age-related speech-reception deficits independent of hearing loss.

Age-related cochlear synaptopathy (CS) has been shown to occur in rodents with minimal noise exposure, and has been hypothesized to play a crucial role in age-related hearing declines in humans. It is not known to what extent age-related CS occurs in humans, and how it affects the coding of supra-threshold sounds and speech in noise. Because in rodents CS affects mainly low- and medium-spontaneous rate (L/M-SR) auditory-nerve fibers with rate-level functions covering medium-high levels, it should lead to greater deficits in the processing of sounds at high than at low stimulus levels. In this cross-sectional study the performance of 102 listeners across the age range (34 young, 34 middle-aged, 34 older) was assessed in a set of psychophysical temporal processing and speech reception in noise tests at both low, and high stimulus levels. Mixed-effect multiple regression models were used to estimate the effects of age while partialing out effects of audiometric thresholds, lifetime noise exposure, cognitive abilities (assessed with additional tests), and musical experience. Age was independently associated with performance deficits on several tests. However, only for one out of 13 tests were age effects credibly larger at the high compared to the low stimulus level. Overall these results do not provide much evidence that age-related CS, to the extent to which it may occur in humans according to the rodent model of greater L/M-SR synaptic loss, has substantial effects on psychophysical measures of auditory temporal processing or on speech reception in noise.

The relationship between isometric mid-thigh pull variables and athletic performance measures: empirical study of English professional soccer players and meta-analysis of extant literature

BACKGROUND

There is currently limited evidence available to support the use of the isometric mid-thigh pull (IMTP) within professional soccer. The aim of this study was to analyze the association between IMTP variables, with common markers of athletic performance capability.

METHODS

Eleven professional development soccer players (age: 20±2 years, stature: 1.82±0.10 m, mass: 76.4±12.8 kg) performed IMTP, 5 m and 10 m accelerations, maximal sprint speed (MSS), countermovement jump (CMJ), and the 505 change of direction test (COD).

RESULTS

Relative and absolute peak force (PF) and force at 50, 100, 150 and 200 ms values were measured during the IMTP. Relative F150, F200, PF displayed large to very large correlations with MSS (r=0.51, r=0.66, and r=0.76 respectively), while absolute PF also displayed a large correlation with MSS (r=0.57). Relative and absolute PF showed large correlations with CMJ height (r=0.54 and r=0.55 respectively). Relative F150 and F200 highlighted large correlations with COD ability (r=-0.68 and r=-0.60 respectively). Relative F200 and PF had a large negative correlation with 10 m acceleration (r=-0.55 and r=-0.53 respectively).

CONCLUSIONS

This study provides an important contribution to knowledge within the area of IMTP testing in professional soccer by evidencing the prominence of the isometric force generating capacity as an underpinning factor in relation to athletic capability.

Young people's use of e-cigarettes in Wales, England and Scotland before and after introduction of EU Tobacco Products Directive regulations: a mixed-method natural experimental evaluation

BACKGROUND

Young people's experimentation with e-cigarettes has increased in recent years, although regular use remains limited. EU Tobacco Products Directive (TPD) regulations introduced packet warnings, advertising restrictions, and regulated nicotine strength from 2016, in part due to concerns regarding use by young people. This paper examines e-cigarette use trajectories before and after TPD.

METHODS

E-cigarette use data were obtained from School Health Research Network/Health Behaviour in School-aged Children surveys in Wales and Smoking Drinking and Drug Use surveys in England. Data from Wales were analysed using segmented logistic regression, with before and after regression analyses of English data. Semi-structured group interviews included young people aged 14-16 years in Wales, England and Scotland in 2017 and 2018.

RESULTS

In Wales, ever use of e-cigarettes increased over time, but under a range of assumptions, growth did not appear to continue post-TPD. A small and non-significant change in trend was observed post-implementation (OR=0.96; 95%CI=0.91 to 1.01), which increased in size and significance after adjusting for ever smoking (OR=0.93; 95%CI=0.88 to 0.98). There was little increase in regular e-cigarette use from 2015 to 2017 in Wales. However, ever and regular use increased from 2014 to 2016 in England. Young people in all nations described limited interactions with components of TPD, while describing e-cigarette use as a 'fad', which had begun to run its course.

CONCLUSIONS

This study provides preliminary evidence that young people's e-cigarette experimentation may be plateauing in UK nations. The extent to which this arises from regulatory changes, or due to a fad having begun to lose its appeal among young people in the UK countries, remains unclear. These trends contrast to those observed in North America, where newer products whose EU market entry and marketing have been impacted by TPD, have gained traction among young people. Long-term monitoring of e-cigarette use trends and perceptions among young people remain vital.

Artificial Intelligence Algorithms for Analysis of Geographic Atrophy: A Review and Evaluation

Purpose

The purpose of this study was to summarize and evaluate artificial intelligence (AI) algorithms used in geographic atrophy (GA) diagnostic processes (e.g. isolating lesions or disease progression).

Methods

The search strategy and selection of publications were both conducted in accordance with the Preferred of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. PubMed and Web of Science were used to extract literary data. The algorithms were summarized by objective, performance, and scope of coverage of GA diagnosis (e.g. lesion automation and GA progression).

Results

Twenty-seven studies were identified for this review. A total of 18 publications focused on lesion segmentation only, 2 were designed to detect and classify GA, 2 were designed to predict future overall GA progression, 3 focused on prediction of future spatial GA progression, and 2 focused on prediction of visual function in GA. GA-related algorithms reported sensitivities from 0.47 to 0.98, specificities from 0.73 to 0.99, accuracies from 0.42 to 0.995, and Dice coefficients from 0.66 to 0.89.

Conclusions

Current GA-AI publications have a predominant focus on lesion segmentation and a minor focus on classification and progression analysis. AI could be applied to other facets of GA diagnoses, such as understanding the role of hyperfluorescent areas in GA. Using AI for GA has several advantages, including improved diagnostic accuracy and faster processing speeds.

Translational Relevance

AI can be used to quantify GA lesions and therefore allows one to impute visual function and quality-of-life. However, there is a need for the development of reliable and objective models and software to predict the rate of GA progression and to quantify improvements due to interventions.

Effects of age on electrophysiological measures of cochlear synaptopathy in humans

Age-related cochlear synaptopathy (CS) has been shown to occur in rodents with minimal noise exposure, and has been hypothesized to play a crucial role in age-related hearing declines in humans. Because CS affects mainly low-spontaneous rate auditory nerve fibers, differential electrophysiological measures such as the ratio of the amplitude of wave I of the auditory brainstem response (ABR) at high to low click levels (WIH/WIL), and the difference between frequency following response (FFR) levels to shallow and deep amplitude modulated tones (FFRS-FFRD), have been proposed as CS markers. However, age-related audiometric threshold shifts, particularly prominent at high frequencies, may confound the interpretation of these measures in cross-sectional studies of age-related CS. To address this issue, we measured WIH/WIL and FFRS-FFRD using highpass masking (HP) noise to eliminate the contribution of high-frequency cochlear regions to the responses in a cross-sectional sample of 102 subjects (34 young, 34 middle-aged, 34 older). WIH/WIL in the presence of the HP noise did not decrease as a function of age. However, in the absence of HP noise, WIH/WIL showed credible age-related decreases even after partialing out the effects of audiometric threshold shifts. No credible age-related decreases of FFRS-FFRD were found. Overall, the results do not provide evidence of age-related CS in the low-frequency region where the responses were restricted by the HP noise, but are consistent with the presence of age-related CS in higher frequency regions.

Improving practices and inferences in developmental cognitive neuroscience

The past decade has seen growing concern about research practices in cognitive neuroscience, and psychology more broadly, that shake our confidence in many inferences in these fields. We consider how these issues affect developmental cognitive neuroscience, with the goal of progressing our field to support strong and defensible inferences from our neurobiological data. This manuscript focuses on the importance of distinguishing between confirmatory versus exploratory data analysis approaches in developmental cognitive neuroscience. Regarding confirmatory research, we discuss problems with analytic flexibility, appropriately instantiating hypotheses, and controlling the error rate given how we threshold data and correct for multiple comparisons. To counterbalance these concerns with confirmatory analyses, we present two complementary strategies. First, we discuss the advantages of working within an exploratory analysis framework, including estimating and reporting effect sizes, using parcellations, and conducting specification curve analyses. Second, we summarize defensible approaches for null hypothesis significance testing in confirmatory analyses, focusing on transparent and reproducible practices in our field. Specific recommendations are given, and templates, scripts, or other resources are hyperlinked, whenever possible.

Semantic and cognitive tools to aid statistical science: replace confidence and significance by compatibility and surprise

BACKGROUND

Researchers often misinterpret and misrepresent statistical outputs. This abuse has led to a large literature on modification or replacement of testing thresholds and P-values with confidence intervals, Bayes factors, and other devices. Because the core problems appear cognitive rather than statistical, we review some simple methods to aid researchers in interpreting statistical outputs. These methods emphasize logical and information concepts over probability, and thus may be more robust to common misinterpretations than are traditional descriptions.

METHODS

We use the Shannon transform of the P-value p, also known as the binary surprisal or S-value s = -log2(p), to provide a measure of the information supplied by the testing procedure, and to help calibrate intuitions against simple physical experiments like coin tossing. We also use tables or graphs of test statistics for alternative hypotheses, and interval estimates for different percentile levels, to thwart fallacies arising from arbitrary dichotomies. Finally, we reinterpret P-values and interval estimates in unconditional terms, which describe compatibility of data with the entire set of analysis assumptions. We illustrate these methods with a reanalysis of data from an existing record-based cohort study.

CONCLUSIONS

In line with other recent recommendations, we advise that teaching materials and research reports discuss P-values as measures of compatibility rather than significance, compute P-values for alternative hypotheses whenever they are computed for null hypotheses, and interpret interval estimates as showing values of high compatibility with data, rather than regions of confidence. Our recommendations emphasize cognitive devices for displaying the compatibility of the observed data with various hypotheses of interest, rather than focusing on single hypothesis tests or interval estimates. We believe these simple reforms are well worth the minor effort they require.

The validity of small-sided games in predicting 11-vs-11 soccer game performance

Predicting performance in soccer games has been a major focus within talent identification and development. Past research has mainly used performance levels, such as elite vs. non-elite players, as the performance to predict (i.e. the criterion). Moreover, these studies have mainly focused on isolated performance attributes as predictors of soccer performance levels. However, there has been an increasing interest in finer grained criterion measures of soccer performance, as well as representative assessments at the level of performance predictors. In this study, we first determined the degree to which 7-vs-7 small-sided games can be considered as representative of 11-vs-11 games. Second, we assessed the validity of individual players’ small-sided game performance in predicting their 11-vs-11 game performance on a continuous scale. Moreover, we explored the predictive validity for 11-vs-11 game performance of several physiological and motor tests in isolation. Sixty-three elite youth players of a professional soccer academy participated in 11 to 17 small-sided games and six 11-vs-11 soccer games. In-game performance indicators were assessed through notational analysis and combined into an overall offensive and defensive performance measure, based on their relationship with game success. Physiological and motor abilities were assessed using a sprint, endurance, and agility test. Results showed that the small-sided games were faster paced, but representative of 11-vs-11 games, with the exception of aerial duels. Furthermore, individual small-sided game performance yielded moderate predictive validities with 11-vs-11 game performance. In contrast, the physiological and motor tests yielded small to trivial relations with game performance. Altogether, this study provides novel insights into the application of representative soccer assessments and the use of continuous criterion measures of soccer performance.

Mapping the relationships between joint stiffness, modeled muscle stiffness, and shear wave velocity

Shear wave velocity is commonly assessed to infer the muscular origins of changes in joint stiffness, but the exact relationship between shear wave velocity changes in muscle and joint stiffness changes remains unknown. Here, we systematically evaluated and quantified this relationship in the plantar flexors. Our results provide evidence for the ability of shear wave velocity to elucidate the muscular origins of joint stiffness changes.