The efficiency of doubly robust estimators of the average causal effect (ACE) of a treatment can be improved by including in the treatment and outcome models only those covariates which are related ...to both treatment and outcome (i.e., confounders) or related only to the outcome. However, it is often challenging to identify such covariates among the large number that may be measured in a given study. In this article, we propose GLiDeR (Group Lasso and Doubly Robust Estimation), a novel variable selection technique for identifying confounders and predictors of outcome using an adaptive group lasso approach that simultaneously performs coefficient selection, regularization, and estimation across the treatment and outcome models. The selected variables and corresponding coefficient estimates are used in a standard doubly robust ACE estimator. We provide asymptotic results showing that, for a broad class of data generating mechanisms, GLiDeR yields a consistent estimator of the ACE when either the outcome or treatment model is correctly specified. A comprehensive simulation study shows that GLiDeR is more efficient than doubly robust methods using standard variable selection techniques and has substantial computational advantages over a recently proposed doubly robust Bayesian model averaging method. We illustrate our method by estimating the causal treatment effect of bilateral versus single-lung transplant on forced expiratory volume in one year after transplant using an observational registry.
Unbiased estimation of causal effects with observational data requires adjustment for confounding variables that are related to both the outcome and treatment assignment. Standard variable selection ...techniques aim to maximize predictive ability of the outcome model, but they ignore covariate associations with treatment and may not adjust for important confounders weakly associated to outcome. We propose a novel method for estimating causal effects that simultaneously considers models for both outcome and treatment, which we call the bilevel spike and slab causal estimator (BSSCE). By using a Bayesian formulation, BSSCE estimates the posterior distribution of all model parameters and provides straightforward and reliable inference. Spike and slab priors are used on each covariate coefficient which aim to minimize the mean squared error of the treatment effect estimator. Theoretical properties of the treatment effect estimator are derived justifying the prior used in BSSCE. Simulations show that BSSCE can substantially reduce mean squared error over numerous methods and performs especially well with large numbers of covariates, including situations where the number of covariates is greater than the sample size. We illustrate BSSCE by estimating the causal effect of vasoactive therapy vs. fluid resuscitation on hypotensive episode length for patients in the Multiparameter Intelligent Monitoring in Intensive Care III critical care database.
Abstract Objective Severe, late functional tricuspid regurgitation is characterized by annulus dilation, right ventricular enlargement, and papillary muscle displacement with leaflet tethering. ...However, the early stages of mild tricuspid regurgitation and its progression are poorly understood. This study examined structural heart changes in mild, early tricuspid regurgitation. Methods Sequential patients undergoing cardiac computed tomography and transthoracic echocardiography with tricuspid regurgitation were identified and evaluated. The tricuspid annulus area and chamber volumes were measured by computed tomography angiography and categorized by tricuspid regurgitation severity. Results Patients (n = 622) were divided into 3 groups by tricuspid regurgitation severity: no/trace (n = 386), mild (n = 178), and moderate/severe tricuspid regurgitation (n = 58). Annulus area was highly dependent on and proportional to regurgitation severity and correlated with both right/left atrial enlargement. Annulus area most strongly correlated with right and left atrial volume, and the annulus shape changed from elliptical to circular in moderate/severe tricuspid regurgitation. Mild tricuspid regurgitation was associated with less right/left atrial enlargement than significant tricuspid regurgitation, normal right ventricular size, and annular dilation. Significant tricuspid regurgitation was associated with annular dilation, circularization, and right ventricular enlargement. Mild and significant tricuspid regurgitation were differentiated by annulus area and indexed right ventricular volume. Conclusions Tricuspid annular dilation and right/left atrial enlargement comprise early events in mild functional tricuspid regurgitation. Atrial enlargement occurs before right ventricular dilation, which occurs late, when tricuspid regurgitation is severe. Atrial volume and tricuspid annular dilation are early and sensitive indicators of tricuspid regurgitation significance.
A single generalizable metric that accurately predicts early dropout from digital health interventions has the potential to readily inform intervention targets and treatment augmentations that could ...boost retention and intervention outcomes. We recently identified a type of early dropout from digital health interventions for smoking cessation, specifically, users who logged in during the first week of the intervention and had little to no activity thereafter. These users also had a substantially lower smoking cessation rate with our iCanQuit smoking cessation app compared with users who used the app for longer periods.
This study aimed to explore whether log-in count data, using standard statistical methods, can precisely predict whether an individual will become an iCanQuit early dropout while validating the approach using other statistical methods and randomized trial data from 3 other digital interventions for smoking cessation (combined randomized N=4529).
Standard logistic regression models were used to predict early dropouts for individuals receiving the iCanQuit smoking cessation intervention app, the National Cancer Institute QuitGuide smoking cessation intervention app, the WebQuit.org smoking cessation intervention website, and the Smokefree.gov smoking cessation intervention website. The main predictors were the number of times a participant logged in per day during the first 7 days following randomization. The area under the curve (AUC) assessed the performance of the logistic regression models, which were compared with decision trees, support vector machine, and neural network models. We also examined whether 13 baseline variables that included a variety of demographics (eg, race and ethnicity, gender, and age) and smoking characteristics (eg, use of e-cigarettes and confidence in being smoke free) might improve this prediction.
The AUC for each logistic regression model using only the first 7 days of log-in count variables was 0.94 (95% CI 0.90-0.97) for iCanQuit, 0.88 (95% CI 0.83-0.93) for QuitGuide, 0.85 (95% CI 0.80-0.88) for WebQuit.org, and 0.60 (95% CI 0.54-0.66) for Smokefree.gov. Replacing logistic regression models with more complex decision trees, support vector machines, or neural network models did not significantly increase the AUC, nor did including additional baseline variables as predictors. The sensitivity and specificity were generally good, and they were excellent for iCanQuit (ie, 0.91 and 0.85, respectively, at the 0.5 classification threshold).
Logistic regression models using only the first 7 days of log-in count data were generally good at predicting early dropouts. These models performed well when using simple, automated, and readily available log-in count data, whereas including self-reported baseline variables did not improve the prediction. The results will inform the early identification of people at risk of early dropout from digital health interventions with the goal of intervening further by providing them with augmented treatments to increase their retention and, ultimately, their intervention outcomes.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, UILJ, UKNU, UL, UM, UPUK
Background & Aims We recently identified a polymorphism upstream of interleukin ( IL ) -28B to be associated with a 2-fold difference in sustained virologic response (SVR) rates to pegylated ...interferon-alfa and ribavirin therapy in a large cohort of treatment-naive, adherent patients with chronic hepatitis C virus genotype 1 (HCV-1) infection. We sought to confirm the polymorphism's clinical relevance by intention-to-treat analysis evaluating on-treatment virologic response and SVR. Methods HCV-1 patients were genotyped as CC, CT, or TT at the polymorphic site, rs12979860. Viral kinetics and rates of rapid virologic response (RVR, week 4), complete early virologic response (week 12), and SVR were compared by IL-28B type in 3 self-reported ethnic groups: Caucasians (n = 1171), African Americans (n = 300), and Hispanics (n = 116). Results In Caucasians, the CC IL-28B type was associated with improved early viral kinetics and greater likelihood of RVR (28% vs 5% and 5%; P < .0001), complete early virologic response (87% vs 38% and 28%; P < .0001), and SVR (69% vs 33% and 27%; P < .0001) compared with CT and TT. A similar association occurred within African Americans and Hispanics. In a multivariable regression model, CC IL-28B type was the strongest pretreatment predictor of SVR (odds ratio, 5.2; 95% confidence interval, 4.1–6.7). RVR was a strong predictor of SVR regardless of IL-28B type. In non-RVR patients, the CC IL-28B type was associated with a higher rate of SVR (Caucasians, 66% vs 31% and 24%; P < .0001). Conclusions In treatment-naive HCV-1 patients treated with pegylated interferon and ribavirin, a polymorphism upstream of IL-28B is associated with increased on-treatment and sustained virologic response and effectively predicts treatment outcome.
Background
For many childhood cancers, survival is lower among non‐Hispanic blacks and Hispanics in comparison with non‐Hispanic whites, and this may be attributed to underlying socioeconomic ...factors. However, prior childhood cancer survival studies have not formally tested for mediation by socioeconomic status (SES). This study applied mediation methods to quantify the role of SES in racial/ethnic differences in childhood cancer survival.
Methods
This study used population‐based cancer survival data from the Surveillance, Epidemiology, and End Results 18 database for black, white, and Hispanic children who had been diagnosed at the ages of 0 to 19 years in 2000‐2011 (n = 31,866). Black‐white and Hispanic‐white mortality hazard ratios and 95% confidence intervals, adjusted for age, sex, and stage at diagnosis, were estimated. The inverse odds weighting method was used to test for mediation by SES, which was measured with a validated census‐tract composite index.
Results
Whites had a significant survival advantage over blacks and Hispanics for several childhood cancers. SES significantly mediated the race/ethnicity–survival association for acute lymphoblastic leukemia, acute myeloid leukemia, neuroblastoma, and non‐Hodgkin lymphoma; SES reduced the original association between race/ethnicity and survival by 44%, 28%, 49%, and 34%, respectively, for blacks versus whites and by 31%, 73%, 48%, and 28%, respectively, for Hispanics versus whites ((log hazard ratio total effect – log hazard ratio direct effect)/log hazard ratio total effect).
Conclusions
SES significantly mediates racial/ethnic childhood cancer survival disparities for several cancers. However, the proportion of the total race/ethnicity–survival association explained by SES varies between black‐white and Hispanic‐white comparisons for some cancers, and this suggests that mediation by other factors differs across groups.
Socioeconomic status mediates the association between race/ethnicity and childhood cancer survival, though to varying degrees across cancers. The proportion of the total effect explained by socioeconomic status varies by race/ethnicity for some cancers.
See also pages 3975‐8.
Non-adherence has been well recognized for years to be a common issue that significantly impacts clinical outcomes and health care costs. Medication adherence is remarkably low even in the controlled ...environment of clinical trials where it has potentially complex major implications. Collection of non-adherence data diverge markedly among cardiovascular randomized trials and, even where collected, is rarely incorporated in the statistical analysis to test the consistency of the primary endpoint(s). The imprecision introduced by the inconsistent assessment of non-adherence in clinical trials might confound the estimate of the calculated efficacy of the study drug. Hence, clinical trials may not accurately answer the scientific question posed by regulators, who seek an accurate estimate of the true efficacy and safety of treatment, or the question posed by payers, who want a reliable estimate of the effectiveness of treatment in the marketplace after approval. The Non-adherence Academic Research Consortium is a collaboration among leading academic research organizations, representatives from the U.S. Food and Drug Administration and physician-scientists from the USA and Europe. One in-person meeting was held in Madrid, Spain, culminating in a document describing consensus recommendations for reporting, collecting, and analysing adherence endpoints across clinical trials. The adoption of these recommendations will afford robustness and consistency in the comparative safety and effectiveness evaluation of investigational drugs from early development to post-marketing approval studies. These principles may be useful for regulatory assessment, as well as for monitoring local and regional outcomes to guide quality improvement initiatives.
Objective: This study determined the characteristics of engagement and whether engagement in an adaptive preventive intervention (API) was associated with reduced binge drinking and alcohol-related ...consequences. Method: Incoming students were recruited for a sequential multiple assignment randomized trial (SMART; N = 891, 62.4% female, 76.8% non-Hispanic White) with an assessment-only control group. The API occurred during the first semester of college, with outcomes assessed at the end of the semester. The API involved two stages. Stage 1 included universal intervention components (personalized normative feedback PNF and self-monitoring). Stage 2 bridged heavy drinkers to access additional resources. We estimated the effect of engagement in Stage 1 only and in the whole API (Stages 1 and 2) among the intervention group, and the effect of the API versus control had all students assigned an API engaged, on alcohol-related outcomes. Results: Precollege binge drinking, intention to pledge a fraternity/sorority, and higher conformity motives were most associated with lower odds of Stage 1 engagement. Action (readiness to change) and PNF engagement were associated with Stage 2 engagement. API engagement was associated with significant reductions in alcohol-related consequences among heavy drinkers. Compared to the control, we estimated the API would reduce the relative increase in alcohol-related consequences from baseline to follow-up by 25%, had all API students engaged. Conclusions: Even partial engagement in each component of the "light-touch" API rendered benefits. Analyses suggested that had all students in the intervention group engaged, the API would significantly reduce the change in alcohol-related consequences over the first semester in college.
What is the public health significance of this article?
This study focuses on the characteristics associated with engaging in a brief adaptive preventive intervention and concludes that engagement is associated with reduced alcohol-related consequences among first-year college students. Identifying characteristics associated with intervention engagement (e.g., intending to pledge into a fraternity/sorority, precollege drinking motives and behaviors, readiness to change) can inform efforts to improve intervention implementation and thereby improve outcomes.
End‐stage renal disease (ESRD) is a risk after kidney donation. We sought, in a large cohort of kidney donors, to determine the causes of donor ESRD, the interval from donation to ESRD, the role of ...the donor/recipient relationship, and the trajectory of the estimated GFR (eGFR) from donation to ESRD. From 1/1/1963 thru 12/31/2015, 4030 individuals underwent living donor nephrectomy at our center, as well as ascertainment of ESRD status. Of these, 39 developed ESRD (mean age ± standard deviation SD at ESRD, 62.4 ± 14.1 years; mean interval between donation and ESRD, 27.1 ± 9.8 years). Donors developing ESRD were more likely to be male, as well as smokers, and younger at donation, and to have donated to a first‐degree relative. Of donors with a known cause of ESRD (n = 25), 48% was due to diabetes and/or hypertension; only 2 from a disease that would have affected 1 kidney (cancer). Of those 25 with an ascertainable ESRD cause, 4 shared a similar etiology of ESRD with their recipient. Almost universally, thechange of eGFR over time was stable, until new‐onset disease (kidney or systemic). Knowledge of factors contributing to ESRD after living kidney donation can improve donor selection and counseling, as well as long‐term postdonation care.
In a single‐center study with long‐term living kidney donor follow‐up, the authors find that end‐stage renal disease is rare, occurs late postdonation (median, 30 years), and is commonly due to new‐onset disease. See Wainright et al's companion article (page 1129) and Gill's editorial (page 1041).
Background: Raw bioimpedance parameters (eg, 50-kHz phase angle PA and 200-kHz/5-kHz impedance ratio IR) have been investigated as predictors of nutrition status and/or clinical outcomes. However, ...their validity as prognostic measures depends on the availability of appropriate reference data. Using a large and ethnically diverse data set, we aimed to determine if ethnicity influences these measures and provide expanded bioimpedance reference data for the U.S. population. Methods: The National Health and Nutrition Examination Survey (NHANES) is an ongoing compilation of studies conducted by the U.S. Centers for Disease Control and Prevention designed to monitor nutrition status of the U.S. population. The NHANES data sets analyzed were from the years 1999–2000, 2001–2002, and 2003–2004. Results: Multivariate analysis showed that PA and IR differed by body mass index (BMI), age, sex, and ethnicity (n = 6237; R2 = 41.2%, P < .0001). Suggested reference cut-points for PA stratified by age decade, ethnicity, and sex are provided. Conclusion: Ethnicity is an important variable that should be accounted for when determining population reference values for PA and IR. We have provided sex-, ethnicity-, and age decade–specific reference values from PA for use by future studies in U.S. populations. Interdevice differences are likely to be important contributors to variability across published population-specific reference data and, where possible, should be evaluated in future research. Ultimately, further validation with physiologically relevant reference measures (eg, dual-energy x-ray absorptiometry) is necessary to determine if PA/IR are appropriate bedside tools for the assessment of nutrition status in a clinical population.
PDF
