Criticality of predictors in multiple regression

Microhabitat conditions drive uncertainty of risk and shape neophobic responses in Trinidadian guppies, Poecilia reticulata

In response to uncertain risks, prey may rely on neophobic phenotypes to reduce the costs associated with the lack of information regarding local conditions. Neophobia has been shown to be driven by information reliability, ambient risk and predator diversity, all of which shape uncertainty of risk. We similarly expect environmental conditions to shape uncertainty by interfering with information availability. In order to test how environmental variables might shape neophobic responses in Trinidadian guppies (Poecilia reticulata), we conducted an in situ field experiment of two high-predation risk guppy populations designed to determine how the 'average' and 'variance' of several environmental factors might influence the neophobic response to novel predator models and/or novel foraging patches. Our results suggest neophobia is shaped by water velocity, microhabitat complexity, pool width and depth, as well as substrate diversity and heterogeneity. Moreover, we found differential effects of the 'average' and 'variance' environmental variables on food- and predator-related neophobia. Our study highlights that assessment of neophobic drivers should consider predation risk, various microhabitat conditions and neophobia being tested. Neophobic phenotypes are expected to increase the probability of prey survival and reproductive success (i.e. fitness), and are therefore likely linked to population health and species survival. Understanding the drivers and consequences of uncertainty of risk is an increasingly pressing issue, as ecological uncertainty increases with the combined effects of climate change, anthropogenic disturbances and invasive species.

Comparing the Relative Importance of Predictors of Intention to Use Bicycles

The use of bicycles for active commuting is an important target to reach because of the importance of increasing physical activity among the population and improving the air quality in cities. Among the models that have been utilized in previous studies, the Theory of Planned Behavior (TPB) has shown good results in terms of the total variance obtained. However, establishing the relative importance of the TPB variables is difficult. In the present study, which was carried out in the Italian context, the authors sought to establish the weight of the proposed variables based on the dominance analysis approach. Considering the initiatives, which the Italian government carries out, and the particular period in which the study was developed, the authors included two variables in addition to the classical factors: financial incentives and daily commuting habits. A survey was administered to 294 Italians (222 females and 72 males, from 18 to 77 years old) through social networks from July to September 2020. The results have shown how the main predictor of bicycle use was use habits, followed almost at the same level by financial incentives and attitude, while norms and perceived behavioral control (PBC) present low relative importance among the variables considered. Limits of the study have been discussed, and suggestions for future research have been proposed.

Frequencies of emergency department use and hospitalization comparing patients with different types of substance or polysubstance-related disorders

Background

This study measured emergency department (ED) use and hospitalization for medical reasons among patients with substance-related disorders (SRD), comparing four subgroups: cannabis-related disorders, drug-related disorders other than cannabis, alcohol-related disorders and polysubstance-related disorders, controlling for various clinical, sociodemographic and service use variables.

Methods

Clinical administrative data for a cohort of 22,484 patients registered in Quebec (Canada) addiction treatment centers in 2012-13 were extracted for the years 2009-10 to 2015-16. Using negative binomial models, risks of frequent ED use and hospitalization were calculated for a 12-month period (2015-16).

Results

Patients with polysubstance-related disorders used ED more frequently than other groups with SRD. They were hospitalized more frequently than patients with cannabis or other drug-related disorders, but less frequently than those with alcohol-related disorders. Patients with alcohol-related disorders used ED more frequently than those with cannabis-related disorders and underwent more hospitalizations than both patients with cannabis-related and other drug-related disorders. Co-occurring SRD-mental disorders or SRD-chronic physical illnesses, more years with SRD, being women, living in rural territories, more frequent consultations with usual general practitioner or outpatient psychiatrist, and receiving more interventions in community healthcare centers increased frequency of ED use and hospitalization, whereas both adverse outcomes decreased with high continuity of physician care. Behavioral addiction, age less than 45 years, living in more materially deprived areas, and receiving 1-3 interventions in addiction treatment centers increased risk of frequent ED use, whereas living in semi-urban areas decreased ED use. Patients 25-44 years old receiving 4+ interventions in addiction treatment centers experienced less frequent hospitalization.

Conclusion

Findings showed higher risk of ED use among patients with polysubstance-related disorders, and higher hospitalization risk among patients with alcohol-related disorders, compared with patients affected by cannabis and other drug-related disorders. However, other variables contributed substantially more to the frequency of ED use and hospitalization, particularly clinical variables regarding complexity and severity of health conditions, followed by service use variables. Another important finding was that high continuity of physician care helped decrease the use of acute care services. Strategies like integrated care and outreach interventions may enhance SRD services.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13011-021-00421-7.

Effect of Ozone, Clothing, Temperature, and Humidity on the Total OH Reactivity Emitted from Humans

People influence indoor air chemistry through their chemical emissions via breath and skin. Previous studies showed that direct measurement of total OH reactivity of human emissions matched that calculated from parallel measurements of volatile organic compounds (VOCs) from breath, skin, and the whole body. In this study, we determined, with direct measurements from two independent groups of four adult volunteers, the effect of indoor temperature and humidity, clothing coverage (amount of exposed skin), and indoor ozone concentration on the total OH reactivity of gaseous human emissions. The results show that the measured concentrations of VOCs and ammonia adequately account for the measured total OH reactivity. The total OH reactivity of human emissions was primarily affected by ozone reactions with organic skin-oil constituents and increased with exposed skin surface, higher temperature, and higher humidity. Humans emitted a comparable total mixing ratio of VOCs and ammonia at elevated temperature-low humidity and elevated temperature-high humidity, with relatively low diversity in chemical classes. In contrast, the total OH reactivity increased with higher temperature and higher humidity, with a larger diversity in chemical classes compared to the total mixing ratio. Ozone present, carbonyl compounds were the dominant reactive compounds in all of the reported conditions.

Dynamic waitlisted design for evaluating a randomized trial of evidence-based quality improvement of comprehensive women's health care implementation in low-performing VA facilities

Background

Women’s Health Services (WHS) in the Veterans Health Administration (VA) has long partnered with VA researchers to evaluate how VA care is organized for women veterans. This partnership has yielded substantial evidence of (1) variations in women veterans’ access to comprehensive healthcare services that contribute to disparities in quality and patient experience and (2) the positive impacts of gender-specific care models for women veterans’ quality and satisfaction. In an effort to provide support specifically to sites that were low-performing in women’s health, WHS and the VA Quality Enhancement Research Initiative co-funded an effort to roll out and evaluate evidence-based quality improvement (EBQI), an implementation strategy with demonstrated effectiveness in a prior cluster randomized trial in women’s health clinics.

Methods

We will identify 21 low-performing VA facilities through a combination of practice data, VA quality metrics (by gender), and other indicators. In partnership with WHS, an EBQI contractor will deliver the EBQI “package”—local consensus development and priority setting using stakeholder panels, multilevel stakeholder engagement, practice facilitation, local EBQI team training, and formative feedback—to participating sites. We propose a dynamic wait-listed design to evaluate the WHS plans for seven EBQI launches per year over 3 years. The goal is to evaluate (1) barriers and facilitators to achieving delivery of comprehensive women’s health care in low-performing VA facilities; (2) effectiveness of EBQI in supporting low-performing VA facilities to achieve improved practice features (e.g., level of comprehensive services available, care coordination arrangements, Patient Aligned Care Team (PACT) features implemented, environment of care improvements), provider/staff attitudes (e.g., improved gender awareness, women’s health knowledge and practice), quality of care, and patient experience; and (3) contextual factors, local implementation processes, and organizational changes over time.

Discussion

Access to comprehensive women’s health care reduces fragmentation of care, improves patient satisfaction, and results in better patient outcomes. We hypothesize that EBQI implementation will result in changes in leadership awareness and buy-in, multilevel engagement in problem-solving, an enhanced culture of quality improvement, structural changes in care, improved provider/staff attitudes, and better quality and patient experience.

Trial registration

ClinicalTrials.gov, NCT03238417. Registered 3 August 2017. Retrospectively registered, https://clinicaltrials.gov/ct2/show/study/NCT03238417

The spatial frequency of climatic conditions affects niche composition and functional diversity of species assemblages: the case of Angiosperms

Climatic conditions vary in spatial frequency globally. Spatially rare climatic conditions provide fewer suitable environments than common ones and should impose constraints on the types of species present locally and regionally. We used data on 467 North American angiosperms to test the effects of the spatial frequency of climatic conditions on ecological niche specialisation and functional diversity. We predicted that rare climates should favour generalist species that are able to inhabit a broader range of climatic conditions. Our results show that climate frequency filters species that differ in niche breadths and rare environments host species combinations with greater functional diversity. The proposed analytical approaches and hypotheses can be adapted to investigate different aspects of ecological assemblies and their biodiversity. We discuss different mechanisms regarding how spatial frequency of environments can affect niche composition and functional diversity. These should be useful while developing theoretical frameworks for generating a deeper understanding of its underpinnings.

History and Use of Relative Importance Indices in Organizational Research

The search for a meaningful index of the relative importance of predictors in multiple regression has been going on for years. This type of index is often desired when the explanatory aspects of regression analysis are of interest. The authors define relative importance as the proportionate contribution each predictor makes to R2, considering both the unique contribution of each predictor by itself and its incremental contribution when combined with the other predictors. The purposes of this article are to introduce the concept of relative importance to an audience of researchers in organizational behavior and industrial/organizational psychology and to update previous reviews of relative importance indices. To this end, the authors briefly review the history of research on predictor importance in multiple regression and evaluate alternative measures of relative importance. Dominance analysis and relative weights appear to be the most successful measures of relative importance currently available. The authors conclude by discussing how importance indices can be used in organizational research.

Letter to the Editor: On the term 'interaction' and related phrases in the literature on Random Forests

In an interesting and quite exhaustive review on Random Forests (RF) methodology in bioinformatics Touw et al. address—among other topics—the problem of the detection of interactions between variables based on RF methodology. We feel that some important statistical concepts, such as ‘interaction’, ‘conditional dependence’ or ‘correlation’, are sometimes employed inconsistently in the bioinformatics literature in general and in the literature on RF in particular. In this letter to the Editor, we aim to clarify some of the central statistical concepts and point out some confusing interpretations concerning RF given by Touw et al. and other authors.

An introduction to recursive partitioning: rationale, application, and characteristics of classification and regression trees, bagging, and random forests

Recursive partitioning methods have become popular and widely used tools for nonparametric regression and classification in many scientific fields. Especially random forests, which can deal with large numbers of predictor variables even in the presence of complex interactions, have been applied successfully in genetics, clinical medicine, and bioinformatics within the past few years. High-dimensional problems are common not only in genetics, but also in some areas of psychological research, where only a few subjects can be measured because of time or cost constraints, yet a large amount of data is generated for each subject. Random forests have been shown to achieve a high prediction accuracy in such applications and to provide descriptive variable importance measures reflecting the impact of each variable in both main effects and interactions. The aim of this work is to introduce the principles of the standard recursive partitioning methods as well as recent methodological improvements, to illustrate their usage for low and high-dimensional data exploration, but also to point out limitations of the methods and potential pitfalls in their practical application. Application of the methods is illustrated with freely available implementations in the R system for statistical computing.

The dominance analysis approach for comparing predictors in multiple regression

A general method is presented for comparing the relative importance of predictors in multiple regression. Dominance analysis (D. V. Budescu, 1993), a procedure that is based on an examination of the R2 values for all possible subset models, is refined and extended by introducing several quantitative measures of dominance that differ in the strictness of the dominance definition. These are shown to be intuitive, meaningful, and informative measures that can address a variety of research questions pertaining to predictor importance. The bootstrap is used to assess the stability of dominance results across repeated sampling, and it is shown that these methods provide the researcher with more insights into the pattern of importance in a set of predictors than were previously available.