The Calculus of M-Estimation in R with geex.

M-estimation, or estimating equation, methods are widely applicable for point estimation and asymptotic inference. In this paper, we present an R package that can find roots and compute the empirical sandwich variance estimator for any set of user-specified, unbiased estimating equations. Examples from the M-estimation primer by Stefanski and Boos (2002) demonstrate use of the software. The package also includes a framework for finite sample, heteroscedastic, and autocorrelation variance corrections, and a website with an extensive collection of tutorials.

Is fasting safe? A chart review of adverse events during medically supervised, water-only fasting.

BACKGROUND: Evidence suggests that fasting, during which only water is consumed, results in potentially health promoting physiological effects. However, peer-reviewed research assessing the safety of water-only fasting is lacking. To address this, we conducted a chart review to describe adverse events (AEs) that occurred during medically supervised, water-only fasting. METHODS: Electronic charts from patient visits to a residential medical facility from 2006 to 2011 were reviewed. Patients who were at least 21 years of age and water-only fasted for ≥2 consecutive days with a refeeding period equal to half of the fast length were included. Out of 2539 charts, 768 visits met our inclusion and exclusion criteria. AEs were abstracted from chart notes and classified according to CTCAE (v4.03) and MedDRA (v12.1) terminology. Descriptive analysis of AEs is reported. RESULTS: During the protocol period, the highest grade AE (HGAE) in 555 visits was a grade 2 event or lower, in 212 visits it was a grade 3 event, in 1 visit it was a grade 4 event, and there were no grade 5 events. There were 2 (0.002%) visits with a serious adverse event (SAE). The majority of AEs identified were mild (n = 4490, 75%) in nature and known reactions to fasting. CONCLUSIONS: To our knowledge, this is the most comprehensive analysis of AEs experienced during medically supervised, water-only fasting conducted to date. Overall, our data indicate that the majority of AEs experienced were mild to moderate and known reactions to fasting. This suggests that the protocol used in this study can be safely implemented in a medical setting with minimal risk of a SAE.

A Recipe for inferference: Start with Causal Inference. Add Interference. Mix Well with R.

In causal inference, interference occurs when the treatment of one subject affects the outcome of other subjects. Interference can distort research conclusions about causal effects when not accounted for properly. In the absence of interference, inverse probability weighted (IPW) estimators are commonly used to estimate causal effects from observational data. Recently, IPW estimators have been extended to handle interference. Tchetgen Tchetgen and VanderWeele (2012) proposed IPW methods to estimate direct and indirect (or spillover) effects that allow for interference between individuals within groups. In this paper, we present inferference, an R package that computes these IPW causal effect estimates when interference may be present within groups. We illustrate use of the package with examples from political science and infectious disease.

Lipid Testing Trends in the US Before and After the Release of the 2013 Cholesterol Treatment Guidelines.

BACKGROUND: The 2013 ACC/AHA cholesterol treatment guidelines removed the recommendation to treat adults at risk of cardiovascular disease to goal levels of low-density lipoprotein cholesterol (LDL-C). We anticipated that the frequency of LDL-C testing in clinical practice would decline as a result. To test this hypothesis, we evaluated the frequency of LDL-C testing before and after the guideline release. METHODS: We used the MarketScan® Commercial and Medicare Supplemental claims data (1/1/2007-12/31/2016) to identify four cohorts: 1) statin initiators (any intensity), 2) high-intensity statin initiators, 3) ezetimibe initiators, and 4) patients at very high cardiovascular risk (≥2 hospitalizations for myocardial infarction or ischemic stroke, with prevalent statin use). Rates of LDL-C testing by calendar year quarter were estimated for each cohort. To estimate rates in the absence of a guideline change, we fit a time-series model to the pre-guideline rates and extrapolated to the post-guideline period, adjusting for covariates, seasonality, and time trend. RESULTS: Pre- and post-guideline rates (LDL-C tests per 1,000 persons per quarter) were 248 and 235, respectively, for 3.9 million statin initiators; 263 and 246 for 1.3 million high-intensity statin initiators; 277 and 261 for 323,544 ezetimibe initiators; and 180 and 158 for 42,108 very high-risk patients. For all cohorts, observed post-guideline rates were similar to model-predicted rates. On average, the difference between observed and predicted rates was 8.5 for patients initiating any statin; 2.6 for patients initiating a high-intensity statin; 11.4 for patients initiating ezetimibe, and -0.5 for high-risk patients. CONCLUSION: We observed no discernible impact of the release of the 2013 ACC/AHA guidelines on LDL-C testing rates. Rather, there was a gradual decline in testing rates starting prior to the guideline change and continuing throughout the study period. Our findings suggest that the guidelines had little to no impact on use of LDL-C testing.

Downstream Effects of Upstream Causes.

The United States Environmental Protection Agency considers nutrient pollution in stream ecosystems one of the U.S.' most pressing environmental challenges. But limited independent replicates, lack of experimental randomization, and space- and time-varying confounding handicap causal inference on effects of nutrient pollution. In this paper the causal g-methods are extended to allow for exposures to vary in time and space in order to assess the effects of nutrient pollution on chlorophyll a - a proxy for algal production. Publicly available data from North Carolina's Cape Fear River and a simulation study are used to show how causal effects of upstream nutrient concentrations on downstream chlorophyll a levels may be estimated from typical water quality monitoring data. Estimates obtained from the parametric g-formula, a marginal structural model, and a structural nested model indicate that chlorophyll a concentrations at Lock and Dam 1 were influenced by nitrate concentrations measured 86 to 109 km upstream, an area where four major industrial and municipal point sources discharge wastewater.