Key Principle #1: Causal inference requires careful consideration of confounding d Preferred variable selection methods 1. Historical confounder definition with purposeful variable selection 2. Causal models using directed acyclic graphs d Variable selection methods that do not adequately control for confounding 3. P value-or model-based methods 4. Methods based on b-coefficient changes 5. Selection of variables to identify "independent predictors" d Do not present all of the effect estimates from a model designed to test a single causal association (Table 2 fallacy) Key Principle #2: Interpretation of results should not rely on the magnitude of P values d P values should rarely be presented in isolation d Present effect estimates and measures of variability with or without P values d Variability around effect estimates should inform conclusions d A conclusion of "no association" should require exclusion of meaningful effect sizes d Avoid the word "significant" in favor of more specific language.Key Principle #3: Results should be presented in a granular and transparent fashion d Use the STROBE statement and checklist d Model tables after the STROBE explanation and elaboration document (30) d Visual presentation of quantitative results B Present individual data points when possible B Avoid excessive lines, text, grids, and abbreviations B Continuous data should not be presented in bar charts with standard error bars ("plunger plots") B Use color-blind-friendly palettes

Control of Confounding and Reporting of Results in Causal Inference Studies. Guidance for Authors from Editors of Respiratory, Sleep, and Critical Care Journals