Research-practice partnerships in education: Benefits, challenges, methodological considerations and key enablers for change
Background: Research-practice partnerships reflect collaborations between researchers and practice-based professionals to co-produce new research knowledge and are often cited as one way to narrow the widely recognized gap between research and practice. Within the context of education, research-practice partnerships draw upon the knowledge, expertise and experience of researchers and practice professionals with the aim of co-producing methodologically robust and educationally relevant research aligned with priorities of practice. Over the last decade, this methodological approach has been gaining traction, yet it is not without its challenges and methodological considerations, which need to be understood and carefully navigated as this trajectory seems set to continue. Aims: This article reviewed the benefits, challenges and methodological considerations associated with research-practice partnerships, and proposes four key enablers-infrastructure, funding, training and incentives-for research-practice partnerships to be optimally introduced, embedded and sustained. Methods: Drawing upon the research literature and diverse examples of infrastructure, funding, training and incentives across different European contexts, specifically Sweden, Norway, Germany, England and Scotland, this article aimed to synthesize academic and real-world insights to advance knowledge and thinking in this area. Conclusions: This article provides a contemporary account of key enablers that can initiate, develop, and sustain research-practice partnerships, drawing on insights from the research literature and illustrative examples of existing structures. In doing so, it aims to make a significant contribution to knowledge, thinking and potentially practice, as we work more collectively and productively together to reduce education inequalities and improve the educational experiences and outcomes of children and young people.
Research Culture and Integrity in Japan: A Qualitative Study
There have been numerous instances of research and funding fraud by Japanese researchers that have discredited Japanese research, and thus ministries, universities and other institutions have attempted to prevent research misconduct and promote research integrity through measures such as policy amendments and education. However, it is unclear whether those measures have been effective, because research misconduct continues to occur. In this study, we conducted focus group interviews with 25 participants, including both researchers and administrators, from various academic disciplines. The interviews were guided by seven semi-structured questions based on the Promoting Integrity as an Integral Dimension of Excellence in Research protocol. Our analysis of the resulting data identified 20 themes. Regarding research integrity, participants highlighted boundary ambiguity and the need for a new definition. In relation to the research culture, the following barriers were identified: the laboratory environment, job security for junior researchers and middle-senior researchers, and pressures faced by principal investigators. In terms of policies, participants observed that research integrity regulations were not widely disseminated, and proposed several methods to provide education, including informational materials. Regarding improvements to the current environment, participants mentioned performance evaluation, an open research environment, a supportive research environment, and sharing of previous cases. While numerous issues were raised, several solutions were also proposed. We found that gaining access to the collective knowledge of faculty, students, and administrators is the best way to foster a strong research culture and develop research integrity.
N-of-1 trials in clinical research: Methodological foundations, statistical approaches and implementation challenges
N-of-1 trials offer a unique and rigorous methodology for evaluating individualized treatment responses, particularly within the context of personalized medicine. This article provides a comprehensive explanation of the conceptual and methodological underpinnings of N-of-1 trials, with particular emphasis on statistical techniques and considerations critical for their design and analysis. Existing guidelines for planning and conducting these studies are summarized, along with a discussion of the practical and theoretical challenges to their implementation in clinical practice. We provide an overview for clinicians and researchers who may be unfamiliar with the design. As most of the existing guidance has focused on design and implementation considerations, we expand on the statistical analysis. We aim to support researchers and methodologists in understanding and advancing the methodological toolkit necessary for high-quality N-of-1 research.
Practical guide for determining whether healthcare providers are research participants in cluster randomised trials
Introduction: In cluster randomised trials (CRT), groups (rather than individuals) are randomised to intervention and control conditions. Since the publication of the Ottawa Statement on the Ethical Design and Conduct of CRTs, the accurate identification of research participants has continued to challenge researchers and research ethics committees. Objective: In this article, we focus on CRTs involving healthcare providers and provide a practical framework for applying Ottawa Statement criteria for identifying research participants. We illustrate key lessons with example CRTs. Framework: Study procedures should be analysed in relation to the study objective. A study intervention confers research participant status on healthcare providers if the study objective is to evaluate its effect and it is delivered to or targeted at healthcare providers. A data collection procedure confers research participant status on healthcare providers if it informs a study outcome used to achieve the study objective and it involves interactions between researchers and healthcare providers to collect their data, or the collection of healthcare providers' identifiable private information. Key lessons: In CRTs, healthcare providers may be research participants because of study interventions, data collection procedures, or both; conversely, they may simply be research collaborators. Some study interventions confer research participant status on both healthcare providers and patients. Collecting data on healthcare provider behaviour may confer research participant status on healthcare providers. Conclusions: Accurately identifying research participants in CRTs is essential to their ethical conduct. When healthcare providers are research participants, their rights and welfare should be protected in accordance with research ethics guidelines.
Addressing significant inequity
Achieving equity in health research requires sub-groups to have meaningful, if not equal, explanatory power, ideally through similiar sample sizes. Obtaining equal sample size, though, is often not possible. Small sub-group sizes increase the risk of false conclusions being drawn, which may reinforce inequities if results are misinterpreted (e.g., saying there is a difference between study arms when there is not and, conversely, saying there is no difference when there is). Here we provide examples of common pitfalls and potential considerations to guide researchers, reviewers and editors when analysing and interpreting sub-group data. We propose that researchers focus on presenting effect sizes and confidence intervals rather than statistical significance.
Statistics for the Clinician II: The T-test, Its Versions, and Its Alternatives
The t-test is a mainstay statistical test that allows researchers to use the observed averages (arithmetic means) of data samples to draw evidence-supported conclusions about the corresponding averages in the populations they represent. This article serves as an introduction to this common and useful technique for clinicians with a limited statistical background, who seek to deepen their understanding of the t-test, its reasoning, and its implications in clinical research. We describe the three basic forms of the t-test, its applications and alternatives, and some related concepts while avoiding technical concepts and jargon as much as possible.
Causally-interpretable random-effects meta-analysis
Recent work has made important contributions to the development of causally-interpretable meta-analysis. These methods transport treatment effects estimated in a collection of randomized trials to a target population of interest. Ideally, estimates targeted toward a specific population are more interpretable and relevant to policy-makers and clinicians. However, between-study heterogeneity that does not arise from differences in the distribution of treatment effect modifiers can create difficulties in synthesizing estimates across trials. We propose a conceptual framework and estimation procedures that attempt to account for such heterogeneity, and develop inferential techniques that aim to capture the accompanying excess variability in causal estimates. This framework also clarifies the kinds of treatment effects that are amenable to the techniques of generalizability and transportability.
Pairwise Comparisons of k Binomial Responses
Independent binomially distributed data arise in many contexts, such as clinical trials, quality control monitoring, and stratified sampling. Moreover, the scope is much larger because multinomially distributed data from a 2 by k contingency table can be viewed as k conditionally independent binomial random variables. A standard approach is to use Fisher's conditional test to test equality of the k underlying success probabilities. However, researchers often want to know where the important pairwise differences are. Thus, the closed method of pairwise comparisons is here combined with unconditional exact tests for 2 by 2 tables and Fisher's conditional test for larger tables to get p values exhibiting strong control of the Family-Wise Error Rate and excellent power properties. In clinical trials, studies are often multicenter, but the results here pertain only to single-site studies.
The role of reliability in experiments
We are concerned about an emphasis on reliability for analysis of psychology experiments. Experiments have two elements of sample size: the number of individuals and the number of replicate trials within a task, and that complicates reliability measures. To account for these elements, we distinguish among three levels of analysis: (1) A foundational level that centers task properties without recourse to either element of sample size. An example statistic is intraclass correlation which is the proportion of variances without reference to sample sizes. (2) An intermediate level that centers the number of trials but not the number of individuals. An example statistic on this level is reliability which describes variabilities with reference to numbers of trials but not numbers of individuals. A final level centers both the numbers of individuals and trials. An example quantity is the uncertainty in a correlation coefficient, which, ideally, reflects sample size limits in individuals and trials. Reliability describes an intermediate level - neither useful for communicating foundational task properties nor interpreting correlations. We advocate that researchers consider all three levels and highlight the role of hierarchical models in doing so.
Efficient interaction analysis in randomized controlled trials
Identifying treatment-covariate interaction is an initial step toward revealing treatment effect heterogeneity and advancing precision medicine. However, when the covariate of interest is continuous, the definition of treatment-covariate interaction can be ambiguous or rely on questionable model assumptions. To tackle this challenge, we introduce a model-free framework for interaction analysis in randomized controlled trials, advocating the adoption of a clearly defined target parameter for interaction and thereby avoiding assumptions about the functional form of the data-generating mechanism. To ensure feasibility for most randomized controlled trials encountered in practice, we study interaction analysis under covariate-adaptive randomization, including simple and stratified randomization, and minimization methods. We observe that the usual method for interaction analysis can either exaggerate or understate uncertainty, leading to asymptotically conservative or anti-conservative results. We modify the usual method with a consistent variance estimator. Furthermore, by acknowledging the dependence structure of treatment assignment induced by covariate-adaptive randomization, we compute the semiparametric efficiency bound and introduce a novel semiparametric efficient method for analyzing our specified interaction effect. We demonstrate that our semiparametric efficient inference procedure, equipped with nonparametric and machine learning techniques for adjusting baseline covariate information, is both efficient and widely applicable.
Publication Bias in Coronary Artery Disease Clinical Trials: A Bibliometric Review
Objective: The objective of our study was to determine the extent of publication bias of clinical trials pertaining to the treatment of coronary artery disease. Specifically, studies are more likely to be published if the results are positive and have a higher citation rate. Methods: The US National Library of Medicine Clinical Trials database was used to identify a total of 242 terminated and completed clinical trials with reported results since 2017. Studies were excluded if they had fewer than two treatment arms, fewer than five participants in any of the treatment arms, or no data analysis to determine significance. A trial was deemed "positive" if there was a statistically significant difference between the treatment arms that agreed with the hypothesis. A "negative" was not statistically significant and/or did not agree with the hypothesis of the trial. Data were collected on intervention, treatment arms, funding type, publication rates, citation rate, impact factor and H index of the journal. Results: Of the 144 trials analyzed, 86 had positive results and 58 had negative results. There was a statistically significant difference in the length of study and publication rate. The publication rate for positive studies was 92%, whereas for negative studies it was 50% (P<0.001). Conclusions: Positive outcomes had a higher publication rate compared with the negative results. Identification of this bias is critical in addressing the impact it could have on the practice of evidence-based medicine. If negative results are not being published, then this pushes physicians to rely on positive results more heavily, thereby skewing evidence-based decision making. This in turn impacts decisions that affect the quality of care.