Immortal Time Bias in Observational Studies Yadav, Kabir; Lewis, Roger J
JAMA : the journal of the American Medical Association,
02/2021, Volume:
325, Issue:
7
Journal Article
Peer reviewed
This JAMA Guide to Statistics and Medicine explains immortal time bias, an error in estimating the association between an exposure and an outcome that results from misclassification or exclusion of ...time intervals; explains how this misclassification or exclusion can occur; and presents approaches to minimize or avoid immortal time bias.
Confounding by Indication in Clinical Research Kyriacou, Demetrios N; Lewis, Roger J
JAMA : the journal of the American Medical Association,
11/2016, Volume:
316, Issue:
17
Journal Article
Peer reviewed
Kyriacou and Lewis stress that the possibility of confounding by other factors must be considered in the assessment of the effect of a treatment or potential risk factor--termed an exposure--on a ...patient outcome. The primary goal of clinical research is to obtain valid measures of the effects of treatments or potential risk factors on patient outcomes. Because confounding distorts the true relationship between the exposure of interest and the outcome, investigators attempt to control confounding to provide valid measures of the observed associations or treatment effects. In particular, randomized clinical trials (RCTs) use randomized treatment assignment to balance potential confounding factors--whether measured, unmeasured, or unknown--that might affect the outcome to ensure that those factors are unrelated to the assigned intervention. Thus, RCTs do not typically require use of statistical methods to adjust for confounding, as the randomization process is meant to limit all forms of confounding.
The Propensity Score Haukoos, Jason S; Lewis, Roger J
JAMA : the journal of the American Medical Association,
10/2015, Volume:
314, Issue:
15
Journal Article
Peer reviewed
Open access
The article elaborates upon the technique of propensity score stratification used in which participants are divided into groups based on propensity scores and the association between the treatment of ...interest and the outcome of interest is estimated within each spectrum. This method is expected to reduce the bias in estimated treatment effects and allow investigators to reduce the likelihood of confounding when analysing nonrandomized, observational data.
Decision Curve Analysis Fitzgerald, Mark; Saville, Benjamin R; Lewis, Roger J
JAMA : the journal of the American Medical Association,
01/2015, Volume:
313, Issue:
4
Journal Article
Peer reviewed
The article discusses the advantages offered by decision curve analysis (DCA), a method of evaluating the benefits of a diagnostic test across a range of patient preferences for accepting risk of ...undertreatment and overtreatment to facilitate decisions about test selection and use. The use of DCA by Siddiqui and colleagues to evaluate3 prostate biopsy strategies is highlighted.
When assessing the clinical utility of therapies intended to improve subjective outcomes, the amount of improvement that is important to patients must be determined. The smallest benefit of value to ...patients is called the minimal clinically important difference (MCID). The MCID is a patient-centered concept, capturing both the magnitude of the improvement and also the value patients place on the change. Using patient-centered MCIDs is important for studies involving patient-reported outcomes, for which the clinical importance of a given change may not be obvious to clinicians selecting treatments. The MCID defines the smallest amount an outcome must change to be meaningful to patients. Here, McGlothin and Lewis explain the use of MCID method.
Futility in Clinical Trials Wendelberger, Barbara; Lewis, Roger J
JAMA : the journal of the American Medical Association,
08/2023, Volume:
330, Issue:
8
Journal Article
Peer reviewed
This JAMA Guide to Statistics and Methods discusses the early stopping of clinical trials for futility due to lack of evidence supporting the desired benefit, evidence of harm, or practical issues ...that make successful completion unlikely.
The article discusses the key aspects of the Bayesian approach, which allows for the integration or updating of prior information with newly obtained data to yield a final quantitative summary of the ...information. Some of the limitations of prior information are highlighted.
Missing data are common in clinical research, particularly for variables requiring complex, time-sensitive, resource-intensive, or longitudinal data collection methods. However, even seemingly ...readily available information can be missing. Here, Newgard et al tells why these methods are used and cite ways by which data may be missing.
Lewis and Viele revisit the analogy between clinical trials and diagnostic tests by interpreting a negative trial as a negative test for efficacy. Clinical trials are designed to detect hypothesized ...treatment effects in the same way diagnostic tests are designed to detect an abnormality or target illness. A trial's statistical power, namely the probability that it can detect the treatment effect given that the hypothesized effect exists, is analogous to a diagnostic test's sensitivity, namely the probability a positive test result is obtained when the target condition is present. Just as the sensitivity of a diagnostic test can be affected by the severity of illness, an effect called "spectrum bias," the power of a clinical trial, is strongly influenced by the magnitude of the true treatment benefit associated with the experimental treatment.
The concept of a "pragmatic" clinical trial was first proposed nearly 50 years ago as a study design philosophy that emphasizes answering questions of most interest to decision makers. Decision ...makers, whether patients, physicians, or policy makers, need to know what they can expect from the available diagnostic or therapeutic options when applied in day-to-day clinical practice. This focus on addressing real-world effectiveness questions influences choices about trial design, patient population, interventions, outcomes, and analysis. Here, Sox and Lewis illustrate the issues that clinicians should consider in deciding whether a trial result is likely to apply to their patients.