Directed Acyclic Graphs (DAGs), which offer systematic representations of causal relationships, have become an established framework for the analysis of causal inference in epidemiology; often being used to determine covariate adjustment sets for minimizing confounding bias. DAGitty is a popular web application for drawing and analysing DAGs. Here we introduce the R package ÔdagittyÕ, which provides access to all of the capabilities of the DAGitty web application within the R platform for statistical computing, and also offers several new functions. We describe how the R package ÔdagittyÕ can be used to: evaluate whether a DAG is consistent with the dataset it is intended to represent; enumerate Ôstatistically equivalentÕ but causally different DAGs; and identify exposure-outcome adjustment sets that are valid for causally different but statistically equivalent DAGs. This functionality enables epidemiologists to detect causal misspecifications in DAGs and make robust inferences that remain valid for a range of different DAGs. AvailabilityThe R package ÔdagittyÕ is available through the comprehensive R archive network (CRAN) at https://cran.r-project.org/web/packages/dagitty/. The source code is available on github at https://github.com/jtextor/dagitty. The web application ÔDAGittyÕ is free software, licensed under the GNU general public license (GPL) version 2 and is available at http://dagitty.net/.

Background Directed acyclic graphs (DAGs) are an increasingly popular approach for identifying confounding variables that require conditioning when estimating causal effects. This review examined the use of DAGs in applied health research to inform recommendations for improving their transparency and utility in future research. Methods Original health research articles published during 1999–2017 mentioning ‘directed acyclic graphs’ (or similar) or citing DAGitty were identified from Scopus, Web of Science, Medline and Embase. Data were extracted on the reporting of: estimands, DAGs and adjustment sets, alongside the characteristics of each article’s largest DAG. Results A total of 234 articles were identified that reported using DAGs. A fifth (n = 48, 21%) reported their target estimand(s) and half (n = 115, 48%) reported the adjustment set(s) implied by their DAG(s). Two-thirds of the articles (n = 144, 62%) made at least one DAG available. DAGs varied in size but averaged 12 nodes [interquartile range (IQR): 9–16, range: 3–28] and 29 arcs (IQR: 19–42, range: 3–99). The median saturation (i.e. percentage of total possible arcs) was 46% (IQR: 31–67, range: 12–100). 37% (n = 53) of the DAGs included unobserved variables, 17% (n = 25) included ‘super-nodes’ (i.e. nodes containing more than one variable) and 34% (n = 49) were visually arranged so that the constituent arcs flowed in the same direction (e.g. top-to-bottom). Conclusion There is substantial variation in the use and reporting of DAGs in applied health research. Although this partly reflects their flexibility, it also highlights some potential areas for improvement. This review hence offers several recommendations to improve the reporting and use of DAGs in future research.

Some researchers have recently questioned the validity of associations between birth weight and health in later life. They argue that these associations might be due in part to inappropriate statistical adjustment for variables on the causal pathway (such as current body size), which creates an artifactual statistical effect known as the "reversal paradox." Computer simulations were conducted for three hypothetical relations between birth weight and adult blood pressure. The authors examined the effect of statistically adjusting for different correlations between current weight and birth weight and between current weight and adult blood pressure to assess their impact on associations between birth weight and blood pressure. When there was no genuine relation between birth weight and blood pressure, adjustment for current weight created an inverse association whose size depended on the magnitude of the positive correlations between current weight and birth weight and between current weight and blood pressure. When there was a genuine inverse relation between birth weight and blood pressure, the association was exaggerated following adjustment for current weight, whereas a positive relation between birth weight and blood pressure could be reversed after adjusting for current weight. Thus, researchers must consider the reversal paradox when adjusting for variables that lie within causal pathways.

OBJECTIVEContinuous glucose monitoring (CGM) is increasingly used to assess glucose control in diabetes. The objective was to examine how analysis of glucose data might improve our understanding of the role temporal glucose variation has on largefor-gestational-age (LGA) infants born to women with diabetes. RESEARCH DESIGN AND METHODSFunctional data analysis (FDA) was applied to 1.68 million glucose measurements from 759 measurement episodes, obtained from two previously published randomized controlled trials of CGM in pregnant women with diabetes. A total of 117 women with type 1 diabetes (n = 89) and type 2 diabetes (n = 28) who used repeated CGM during pregnancy were recruited from secondary care multidisciplinary obstetric clinics for diabetes in the U.K. and Denmark.LGA was defined as birth weight ‡90th percentile adjusted for sex and gestational age. RESULTSA total of 54 of 117 (46%) women developed LGA.LGA was associated with lower mean glucose (7.0 vs. 7.1 mmol/L; P < 0.01) in trimester 1, with higher mean glucose in trimester 2 (7.0 vs. 6.7 mmol/L; P < 0.001) and trimester 3 (6.5 vs. 6.4 mmol/L; P < 0.01). FDA showed that glucose was significantly lower midmorning (0900-1100 h) and early evening (1900-2130 h) in trimester 1, significantly higher early morning (0330-0630 h) and throughout the afternoon (1130-1700 h) in trimester 2, and significantly higher during the evening (2030-2330 h) in trimester 3 in women whose infants were LGA. CONCLUSIONSFDA of CGM data identified specific times of day that maternal glucose excursions were associated with LGA. It highlights trimester-specific differences, allowing treatment to be targeted to gestational glucose patterns. LGA, LGA infants are themselves at increased risk of developing obesity, diabetes, and cardiovascular disease in later life (9-13). Maternal hyperglycemia has long been considered the principal determinant of LGA and the factor most amenable to intervention (14,15). However, the prevalence of LGA remains high even in diabetic pregnancies that are considered clinically "well controlled" where selfmonitored capillary blood glucose (SMBG) or HbA 1c measurements indicate that clinical management has been successful in normalizing maternal glucose levels (4)(5)(6)16). This suggests either that something other than glucose levels is responsible for LGA in these women or that SMBG and HbA 1c measurements fail to detect the variation in glucose levels that is capable of causing LGA.This has led to substantial interest in the potential role that continuous glucose monitoring (CGM) might play in improving the clinical assessment and management of glycemic control. Nonetheless, the sheer volume of data these devices produce (288 glucose measurements per day) and the complexity of the underlying signals these data contain mean that CGM data have proved challenging to analyze and interpret. To address this, some analysts have recommended using a wide range of summary statistical indices (such as calculating average glucose levels over specified time periods...