On making causal claims: A review and recommendations

Social scientists often estimate models from correlational data, where the independent variable has not been exogenously manipulated; they also make implicit or explicit causal claims based on these models. When can these claims be made? We answer this question by first discussing design and estimation conditions under which model estimates can be interpreted, using the randomized experiment as the gold standard. We show how endogeneity--which includes omitted variables, omitted selection, simultaneity, common methods bias, and measurement error--renders estimates causally uninterpretable. Second, we present methods that allow researchers to test causal claims in situations where randomization is not possible or when causal interpretation is confounded, including fixed-effects panel, sample selection, instrumental variable, regression discontinuity, and difference-in-differences models. Third, we take stock of the methodological rigor with which causal claims are being made in a social sciences discipline by reviewing a representative sample of 110 articles on leadership published in the previous 10 years in top-tier journals. Our key finding is that researchers fail to address at least 66 % and up to 90 % of design and estimation conditions that make causal claims invalid. We conclude by offering 10 suggestions on how to improve non-experimental research.

Scientific mindfulness: A foundation for future themes in international business.

We conceptualize new ways to qualify what themes should dominate the future international business and management (IB/IM) research agenda by examining three questions: Whom should we ask? What should we ask, and which selection criteria should we apply? What are the contextual forces? Our main findings are the following: (1) wider perspectives from academia and practice would benefit both rigor and relevance; (2) four key forces are climate change, globalization, inequality, and sustainability; and (3) we propose scientific mindfulness as the way forward for generating themes in IB/IM research. Scientific mindfulness is a holistic, cross-disciplinary, and contextual approach, whereby researchers need to make sense of multiple perspectives with the betterment of society as the ultimate criterion.

Médecines complémentaires: vers un consensus "evidence-based" à l'hôpital universitaire. [Complementary and alternative medicines (CAM): towards an "evidence-based" consensus in the University hospital]

Après la votation fédérale demandant de prendre en compte les médecines complémentaires, un consensus a été recherché dans quatorze services et unités du Centre hospitalier universitaire vaudois (CHUV). Confrontés aux données de la littérature (Plus de 2000 publications en "Evidence-based complementary medicine" depuis 1998), les soignants étaient tous surpris par l'ampleur des résultats cliniques disponibles actuellement. Tous identifiaient un besoin en formation et en informations sur le sujet. Une prise de position officielle de l'institution était aussi souhaitée, instituant l'enseignement et la recherche sur les médecines complémentaires et assurant la production d'informations rigoureuses et pertinentes pour la clinique. [Abstract] While a popular vote supported a new article on complementary and alternative medicines (CAM) in the Swiss Constitution, this assessment in 14 wards of the University Hospital of Lausanne, Switzerland, attempted at answering the question: How can CAM use be better taken into account and patients informed with more rigor and respect for their choices? Confronted with a review of the literature (> 2000 publications in "Evidence-based cornplementary medicine" since 1998), respondents declared their ignorance of the clinical data presently available on CAM. All were in favour of more teaching and information on the subject, plus an official statement from the Hospital direction, ensuring production and diffusion of rigorous and cJinically significant information on CAM.

On making causal claims: A review and recommendations

Social scientists often estimate models from correlational data, where the independent variable has not been exogenously manipulated; they also make implicit or explicit causal claims based on these models. When can these claims be made? We answer this question by first discussing design and estimation conditions under which model estimates can be interpreted, using the randomized experiment as the gold standard. We show how endogeneity--which includes omitted variables, omitted selection, simultaneity, common methods bias, and measurement error--renders estimates causally uninterpretable. Second, we present methods that allow researchers to test causal claims in situations where randomization is not possible or when causal interpretation is confounded, including fixed-effects panel, sample selection, instrumental variable, regression discontinuity, and difference-in-differences models. Third, we take stock of the methodological rigor with which causal claims are being made in a social sciences discipline by reviewing a representative sample of 110 articles on leadership published in the previous 10 years in top-tier journals. Our key finding is that researchers fail to address at least 66 % and up to 90 % of design and estimation conditions that make causal claims invalid. We conclude by offering 10 suggestions on how to improve non-experimental research.

The pneumonia severity index: a decade after the initial derivation and validation

The prognosis of community-acquired pneumonia ranges from rapid resolution of symptoms and full recovery of functional status to the development of severe medical complications and death. The pneumonia severity index is a rigorously studied prediction rule for prognosis that objectively stratifies patients into quintiles of risk for short-term mortality on the basis of 20 demographic and clinical variables routinely available at presentation. The pneumonia severity index was derived and validated with data on >50,000 patients with community-acquired pneumonia by use of well-accepted methodological standards and is the only pneumonia decision aid that has been empirically shown to safely increase the proportion of patients given treatment in the outpatient setting. Because of its prognostic accuracy, methodological rigor, and effectiveness and safety as a decision aid, the pneumonia severity index has become the reference standard for risk stratification of community-acquired pneumonia

Publication bias in animal research: a systematic review protocol.

BACKGROUND: Systematic reviews and meta-analyses of pre-clinical studies, in vivo animal experiments in particular, can influence clinical care. Publication bias is one of the major threats of validity in systematic reviews and meta-analyses. Previous empirical studies suggested that systematic reviews and meta-analyses have become more prevalent until 2010 and found evidence for compromised methodological rigor with a trend towards improvement. We aim to comprehensively summarize and update the evidence base on systematic reviews and meta-analyses of animal studies, their methodological quality and assessment of publication bias in particular. METHODS/DESIGN: The objectives of this systematic review are as follows: âeuro¢To investigate the epidemiology of published systematic reviews of animal studies until present. âeuro¢To examine methodological features of systematic reviews and meta-analyses of animal studies with special attention to the assessment of publication bias. âeuro¢To investigate the influence of systematic reviews of animal studies on clinical research by examining citations of the systematic reviews by clinical studies. Eligible studies for this systematic review constitute systematic reviews and meta-analyses that summarize in vivo animal experiments with the purpose of reviewing animal evidence to inform human health. We will exclude genome-wide association studies and animal experiments with the main purpose to learn more about fundamental biology, physical functioning or behavior. In addition to the inclusion of systematic reviews and meta-analyses identified by other empirical studies, we will systematically search Ovid Medline, Embase, ToxNet, and ScienceDirect from 2009 to January 2013 for further eligible studies without language restrictions. Two reviewers working independently will assess titles, abstracts, and full texts for eligibility and extract relevant data from included studies. Data reporting will involve a descriptive summary of meta-analyses and systematic reviews. DISCUSSION: Results are expected to be publicly available later in 2013 and may form the basis for recommendations to improve the quality of systematic reviews and meta-analyses of animal studies and their use with respect to clinical care.

Médecines complémentaires: vers un consensus evidence- based à l'hôpital universitaire [Complementary and alternative medicines (CAM): towards an evidence-based consensus in the university hospital].

While a popular vote supported a new article on complementary and alternative medicines (CAM) in the Swiss Constitution, this assessment in 14 wards of the University Hospital of Lausanne, Switzerland, attempted at answering the question: How can CAM use be better taken into account and patients informed with more rigor and respect for their choices? Confronted with a review of the literature (> 2000 publications in "Evidence-based complementary medicines" since 1998), respondents declared their ignorance of the clinical data presently available on CAM. All were in favour of more teaching and information on the subject, plus an official statement from the Hospital direction, ensuring production and diffusion of rigorous and clinically significant information on CAM.