scholarly journals Generalizing evidence from randomized trials using inverse probability of sampling weights

2018 ◽  
Vol 181 (4) ◽  
pp. 1193-1209 ◽  
Author(s):  
Ashley L. Buchanan ◽  
Michael G. Hudgens ◽  
Stephen R. Cole ◽  
Katie R. Mollan ◽  
Paul E. Sax ◽  
...  
Keyword(s):  
Randomized Trials ◽  
Sampling Weights ◽  
Inverse Probability

scholarly journals Inverse probability weighted estimating equations for randomized trials in transfusion medicine

2013 ◽  
Vol 32 (25) ◽  
pp. 4380-4399 ◽  
Author(s):  
Richard J. Cook ◽  
Ker-Ai Lee ◽  
Meaghan Cuerden ◽  
Cecilia A. Cotton
Keyword(s):  
Randomized Trials ◽  
Estimating Equations ◽  
Transfusion Medicine ◽  
Inverse Probability ◽  
Inverse Probability Weighted

scholarly journals Central sensitization predicts greater fatigue independently of musculoskeletal pain

Rheumatology ◽  
2019 ◽  
Vol 58 (11) ◽  
pp. 1923-1927 ◽  
Author(s):  
Katie L Druce ◽  
John McBeth
Keyword(s):  
Confidence Interval ◽  
Musculoskeletal Pain ◽  
Central Sensitization ◽  
Ratio Test ◽  
Sampling Weights ◽  
Inverse Probability ◽  
Study Participants ◽  
Anxiety Depression ◽  
The Relationship

Abstract Objectives To test whether central sensitization was associated with greater fatigue, independently of musculoskeletal pain. Methods 2477 prospective cohort study participants completed a baseline questionnaire comprising the Chalder Fatigue Scale (CFQ), pain, demographics, physical activity, anxiety, depression and medication use. In a clinical assessment of 290 (11.7%) participants, central sensitization was measured by the wind-up ratio test at the hand (WUR-H) and foot (WUR-F). Bioelectric impedance determined proportion body fat. All participants were followed up 12 months later, at which time they completed the CFQ. Linear regression, with inverse probability sampling weights, tested the relationship between WUR at baseline and CFQ at 12 months, adjusted for baseline CFQ, demographics, lifestyle factors, mental health and baseline pain. Results At baseline, the median interquartile range WUR-H and WUR-F were similar (2.3 (1.5, 4.0) and 2.4 (1.6, 3.9) respectively) and did not differ by sex (difference WUR-H: −0.29, 95% confidence interval −1.28–0.71; WUR-F: −0.57 (−1.50–0.36) or age(WUR-H: −0.53, −1.49–0.43; WUR-F:−0.08, −0.98–0.82). WUR-H scores (β = 0.11, 95% confidence interval: 0.07–0.16) and WUR-F scores (0.13, 0.08–0.17) were positively associated with CFQ scores at follow-up, independently of baseline CFQ and other covariates. These associations were not explained by baseline pain. Conclusion Fatigue was predicted by central sensitization, independently of the presence of pain. For those seeking to treat fatigue, the benefit of interventions that reduce central sensitization should be investigated.

Adjusting for adherence in randomized trials when adherence is measured as a continuous variable: An application to the Lipid Research Clinics Coronary Primary Prevention Trial

2020 ◽  
Vol 17 (5) ◽  
pp. 570-575
Author(s):  
Kerollos Nashat Wanis ◽  
Arin L Madenci ◽  
Miguel A Hernán ◽  
Eleanor J Murray
Keyword(s):  
Primary Prevention ◽  
Dose Response ◽  
Randomized Trials ◽  
Continuous Variable ◽  
Probability Weighting ◽  
Lipid Research ◽  
Primary Prevention Trial ◽  
Inverse Probability ◽  
Inverse Probability Weights ◽  
Percentage Points

Background: Clinicians and patients may be more interested in per-protocol effect estimates than intention-to-treat effect estimates from randomized trials. However, per-protocol effect estimates may be biased due to insufficient adjustment for prognostic factors that predict adherence. Adjustment for this bias is possible when appropriate methods, such as inverse probability weighting, are used. But, when adherence is measured as a continuous variable, constructing these weights can be challenging. Methods: In the placebo arm of the Lipid Research Clinics Coronary Primary Prevention Trial, we estimated the 7-year cumulative incidence of coronary heart disease under 100% adherence and 0% adherence to placebo. We used dose-response discrete-hazards models with inverse probability weighting to adjust for pre- and post-randomization covariates. We considered several continuous distributions for constructing the inverse probability weights. Results: The risk difference estimate for 100% adherence compared with 0% adherence ranged from −7.7 to −6.1 percentage points without adjustment for baseline and post-baseline covariates, and ranged from −1.8 to 2.2 percentage points with adjustment using inverse probability weights, depending on the dose-response model and inverse probability weight distribution used. Conclusions: Methods which appropriately adjust for time-varying post-randomization variables can explain away much of the bias in the “effect” of adherence to placebo. When considering continuous adherence, investigators should consider several models as estimates may be sensitive to the model chosen.

Correcting for selective participation in cohort studies using auxiliary register data without identification of non-participants

2019 ◽  
pp. 140349481989078 ◽  
Author(s):  
Carl Bonander ◽  
Anton Nilsson ◽  
Göran M.L. Bergström ◽  
Jonas Björk ◽  
Ulf Strömberg
Keyword(s):  
Cohort Studies ◽  
Bias Correction ◽  
Propensity Scores ◽  
Target Population ◽  
Population Based ◽  
Register Data ◽  
Sensitive Data ◽  
Sampling Weights ◽  
Inverse Probability ◽  
Study Results

Aims: Selective participation may hamper the validity of population-based cohort studies. The resulting bias can be alleviated by linking auxiliary register data to both the participants and the non-participants of the study, estimating propensity scores for participation and correcting for participation based on these. However, registry holders may not be allowed to disclose sensitive data on (invited) non-participants. Our aim is to provide guidance on how adequate bias correction can be achieved by using auxiliary register data but without disclosing information that could be linked to the subset of non-participants. Methods: We show how existing methods can be used to estimate generalisation weights under various data disclosure scenarios where invited non-participants are indistinguishable from uninvited ones. We also demonstrate how the methods can be implemented using Nordic register data. Results: Inverse-probability-of-sampling weights estimated within a random sample of the target population in which the non-respondents are disclosed are equivalent in expectation to analogous weights in a scenario where the non-participants and uninvited individuals from the population are indistinguishable. To minimise the risk of disclosure when the entire population is invited to participate, investigators should instead consider inverse-odds-of-sampling weights, a method that has previously been suggested for transporting study results to external populations. Conclusions: Generalisation weights can be estimated from auxiliary register data without disclosing information on invited non-participants.

scholarly journals A Comparison of Covariate Adjustment Approaches Under Model Misspecification In Individually Randomized Trials

2022 ◽  
Author(s):  
Mia S. Tackney ◽  
Tim Morris ◽  
Ian White ◽  
Clemence Leyrat ◽  
Karla Diaz-Ordaz ◽  
...  
Keyword(s):  
Sample Size ◽  
Randomized Trials ◽  
Model Misspecification ◽  
Likelihood Estimation ◽  
Small Sample ◽  
Analysis Of Covariance ◽  
Small Samples ◽  
Linear Interaction ◽  
Inverse Probability ◽  
Omitted Covariate

Abstract Adjustment for baseline covariates in randomized trials has been shown to lead to gains in power and can protect against chance imbalances in covariates. For continuous covariates, there is a risk that the the form of the relationship between the covariate and outcome is misspecified when taking an adjusted approach. Using a simulation study focusing on small to medium-sized individually randomized trials, we explore whether a range of adjustment methods are robust to misspecification, either in the covariate-outcome relationship or through an omitted covariate-treatment interaction. Specifically, we aim to identify potential settings where G-computation, Inverse Probability of Treatment Weighting ( IPTW ), Augmented Inverse Probability of Treatment Weighting ( AIPTW ) and Targeted Maximum Likelihood Estimation ( TMLE ) offer improvement over the commonly used Analysis of Covariance ( ANCOVA ). Our simulations show that all adjustment methods are generally robust to model misspecification if adjusting for a few covariates, sample size is 100 or larger, and there are no covariate-treatment interactions. When there is a non-linear interaction of treatment with a skewed covariate and sample size is small, all adjustment methods can suffer from bias; however, methods that allow for interactions (such as G-computation with interaction and IPTW ) show improved results compared to ANCOVA . When there are a high number of covariates to adjust for, ANCOVA retains good properties while other methods suffer from under- or over-coverage. An outstanding issue for G-computation, IPTW and AIPTW in small samples is that standard errors are underestimated; development of small sample corrections is needed.

Sign in / Sign up

Export Citation Format

Share Document