Qualitative approximations to causality: Non-randomizable factors in clinical psychology

Background Causal quests in non-randomized studies are unavoidable just because research questions are beyond doubt causal (e.g., aetiology). Large progress during the last decades has enriched the methodical toolbox. Aims Summary papers mainly focus on quantitative and highly formal methods. With examples from clinical psychology, we show how qualitative approaches can inform on the necessity and feasibility of quantitative analysis and may yet sometimes approximate causal answers. Results Qualitative use is hidden in some quantitative methods. For instance, it may yet suffice to know the direction of bias for a tentative causal conclusion. Counterfactuals clarify what causal effects of changeable factors are, unravel what is required for a causal answer, but do not cover immutable causes like gender. Directed acyclic graphs (DAGs) address causal effects in a broader sense, may give rise to quantitative estimation or indicate that this is premature. Conclusion No method is generally sufficient or necessary. Any causal analysis must ground on qualification and should balance the harms of a false positive and a false negative conclusion in a specific context.

Outcome of patients with chronic limb-threatening ischemia with and without revascularization

Summary: Background: Patients with chronic critical limb-threatening ischemia (CLTI) are at high risk of amputation and death. Despite the general recommendation for revascularization in CTLI in the guidelines, the underlying evidence for such a recommendation is limited. The aim of our study was to assess the outcome of patients with CLTI depending on the use of revascularization in a retrospective real-world cohort. Patients and methods: Administrative data of the largest German Health insurance (BARMER GEK) were provided for all patients that were hospitalized for the treatment of CLTI Rutherford category (RF) 5 and 6 between 2009 and 2011. Patients were followed-up until December 31st, 2012 for limb amputation and death in relation to whether patients did (Rx +) or did not have (Rx −) revascularization during index-hospitalization. Results: We identified 15,314 patients with CLTI at RF5 (n = 6,908 (45.1%)) and RF6 (n = 8,406 (54.9%)), thereof 7,651 (50.0%) underwent revascularization (Rx +) and 7,663 (50.0%) were treated conservatively (Rx −). During follow-up (mean 647 days; 95% CI 640–654 days) limb amputation (46.5% Rx− vs. 40.6% Rx+, P < 0.001) and overall mortality (48.2% Rx− vs. 42.6% Rx+, P < 0.001) were significantly lower in the subgroup Rx+. Conclusions: In a real-world setting, only half of CLTI were revascularized during the in-hospital treatment. Though, revascularization was associated with significantly better observed short- and long-term outcome. These data do not allow causal conclusion due to lack of data on the underlying reason for applied or withheld revascularization and therefore may involve a relevant selection bias.

Explanation implies causation?

Most researchers do not deliberately claim causal results in an observational study. But do we lead our readers to draw a causal conclusion unintentionally by explaining why significant correlations and relationships may exist? Here we perform a randomized study in a data analysis massive online open course to test the hypothesis that explaining an analysis will lead readers to interpret an inferential analysis as causal. We show that adding an explanation to the description of an inferential analysis leads to a 15.2% increase in readers interpreting the analysis as causal (95% CI 12.8% - 17.5%). We then replicate this finding in a second large scale massive online open course. Nearly every scientific study, regardless of the study design, includes explanation for observed effects. Our results suggest that these explanations may be misleading to the audience of these data analyses.

Stress, sense of coherence and subjective health in adolescents aged 13–18 years

Background: This study investigated the relationships between sex, age, stress and sense of coherence (SOC) and each of self-rated health (SRH) and subjective health complaints (SHCs) in adolescents. Methods: The study was based on a cross-sectional sample of 1239 adolescents aged 13–18 years. The participants reported scores on a questionnaire, including 12 items assessing SHC, the 13-item version of the Orientation to Life Questionnaire and the 30-item Adolescent Stress Questionnaire. Data were analysed with descriptive, comparative and hierarchical multiple regression analyses. Results: Boys scored significantly higher on SRH than girls, whereas girls scored higher on SHCs. In the multivariate model, stress was significantly associated with SHCs, but not with SRH. SOC was significantly positively related to SRH and negatively related to SHC; the associations were significantly stronger for girls than for boys. A significant moderation effect (protective role) of SOC was found on the relation between stress and SHC. Conclusions: No causal conclusion was possible, but the findings may be used as a basis for further investigation of the role of stress and SOC in longitudinal studies and intervention studies.

Comment on Nalliah and Allareddy (2014): Unwarranted conclusions

A PeerJ paper by Nalliah and Allareddy (2014) describes improvement in weaker dental students’ scores (and decline in stronger students’ scores) by use of a unique instructional method. I argue that the causal conclusion in their paper cannot be justified because of lack of a comparison group. Regression to the mean is a common confound in test-retest studies such as presented in their paper. Inclusion of a comparison group could be used to rule it out. Other minor issues are raised involving scaling and consistency in the data.

Comment on Nalliah and Allareddy (2014): Unwarranted conclusions

A PeerJ paper by Nalliah and Allareddy (2014) describes improvement in weaker dental students’ scores (and decline in stronger students’ scores) by use of a unique instructional method. I argue that the causal conclusion in their paper cannot be justified because of lack of a comparison group. Regression to the mean is a common confound in test-retest studies such as presented in their paper. Inclusion of a comparison group could be used to rule it out. Other minor issues are raised involving scaling and consistency in the data.

Redefining the Concept of HRIS

A concept is a logical construction that is established in accordance with a framework and is part of it (Ferrari, 1977). According to this definition, a concept is unchangeable within the framework it belongs to and is part of. The concept of gravity is unchangeable within Newtonian physics. However, when treated by Einstein’s relativity it is quite different. In the natural sciences, the frontiers of frameworks can be clearly and easily defined with axioms. In the social sciences, however, axioms are unlikely (Perelman, 1982), so the definitions of concepts are more flexible than one can assure. In the social sciences, social constructed frameworks generally evolve, so the concepts of which they are made up must also evolve. The human resource information system (henceforth HRIS) is a good example of a technological tool that evolves over time. Its definition evolves not only as technology evolves but also as the perception of its impact changes. In fact, the comprehension of HRIS as a concept can go from the naïve understanding (e.g., software applied to human resource management) to its broad conceptualization “in relation to” the greater organizational system and dynamics. Although the literature suggests that HRIS practices can influence organizational performance, such causal conclusion is still problematic. This article describes a study that focuses on both issues. It aims to deepen the comprehension of how information systems are, in general, used as a constituent tool of a greater phenomenon: namely, organizational change and development. Consequently, it aims to redefine the main concepts in this field.

Warrant, Causation, and the Atomism of Evidence Law

ABSTRACTThe epistemological analysis offered in this paper reveals that a combination of pieces of evidence, none of them sufficient by itself to warrant a causal conclusion to the legally required degree of proof, may do so jointly. The legal analysis offered here, interlocking with this, reveals thatDaubert’s requirement that courts screen each item of scientific expert testimony for reliability can actually impede the process of arriving at the conclusion most warranted by the evidence proffered.