Dataset Information

Evaluating possible confounding by prescriber in comparative effectiveness research.

ABSTRACT: In nonrandomized studies of comparative effectiveness of medications, the prescriber may be the most important determinant of treatment assignment, yet the majority of analyses ignore the prescriber. Via Monte Carlo simulation, we evaluated the bias of 3 approaches that utilize the prescriber in analysis compared against the default approach that ignores the prescriber. Prescriber preference instrumental variable (IV) analyses were unbiased when IV criteria were met, which required no clustering of unmeasured patient characteristics within prescriber. In all other scenarios, IV analyses were highly biased, and stratification on the prescriber reduced confounding bias at the patient or prescriber levels. Including a prescriber random intercept in the propensity score model reversed the direction of confounding from measured patient factors and resulted in unpredictable changes in bias. Therefore, we recommend caution when using the IV approach, particularly when the instrument is weak. Stratification on the prescriber may be more robust; this approach warrants additional research.

SUBMITTER: Franklin JM

PROVIDER: S-EPMC4347927 | biostudies-literature | 2015 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Evaluating possible confounding by prescriber in comparative effectiveness research.

Franklin Jessica M JM Schneeweiss Sebastian S Huybrechts Krista F KF Glynn Robert J RJ

Epidemiology (Cambridge, Mass.) 20150301 2

In nonrandomized studies of comparative effectiveness of medications, the prescriber may be the most important determinant of treatment assignment, yet the majority of analyses ignore the prescriber. Via Monte Carlo simulation, we evaluated the bias of 3 approaches that utilize the prescriber in analysis compared against the default approach that ignores the prescriber. Prescriber preference instrumental variable (IV) analyses were unbiased when IV criteria were met, which required no clustering ...[more]

PMID: 25643103

Similar Datasets

Project description:BackgroundUsing observational data to assess the relative effectiveness of alternative cancer treatments is limited by patient selection into treatment, which often biases interpretation of outcomes. We evaluated methods for addressing confounding in treatment and survival of patients with early-stage prostate cancer in observational data and compared findings with those from a benchmark randomized clinical trial.MethodsWe selected 14 302 early-stage prostate cancer patients who were aged 66-74 years and had been treated with radical prostatectomy or conservative management from linked Surveillance, Epidemiology, and End Results-Medicare data from January 1, 1995, through December 31, 2003. Eligibility criteria were similar to those from a clinical trial used to benchmark our analyses. Survival was measured through December 31, 2007, by use of Cox proportional hazards models. We compared results from the benchmark trial with results from models with observational data by use of traditional multivariable survival analysis, propensity score adjustment, and instrumental variable analysis.ResultsProstate cancer patients receiving conservative management were more likely to be older, nonwhite, and single and to have more advanced disease than patients receiving radical prostatectomy. In a multivariable survival analysis, conservative management was associated with greater risk of prostate cancer-specific mortality (hazard ratio [HR] = 1.59, 95% confidence interval [CI] = 1.27 to 2.00) and all-cause mortality (HR = 1.47, 95% CI = 1.35 to 1.59) than radical prostatectomy. Propensity score adjustments resulted in similar patient characteristics across treatment groups, although survival results were similar to traditional multivariable survival analyses. Results for the same comparison from the instrumental variable approach, which theoretically equalizes both observed and unobserved patient characteristics across treatment groups, differed from the traditional multivariable and propensity score results but were consistent with findings from the subset of elderly patient with early-stage disease in the trial (ie, conservative management vs radical prostatectomy: for prostate cancer-specific mortality, HR = 0.73, 95% CI = 0.08 to 6.73; for all-cause mortality, HR = 1.09, 95% CI = 0.46 to 2.59).ConclusionInstrumental variable analysis may be a useful technique in comparative effectiveness studies of cancer treatments if an acceptable instrument can be identified.

Project description:IntroductionTobacco smoking claims 700,000 lives every year in Europe and the cost of tobacco smoking in the EU is estimated between €98 and €130 billion annually; direct medical care costs and indirect costs such as workday losses each represent half of this amount. Policymakers all across Europe are in need of bespoke information on the economic and wider returns of investing in evidence-based tobacco control, including smoking cessation agendas. EQUIPT is designed to test the transferability of one such economic evidence base-the English Tobacco Return on Investment (ROI) tool-to other EU member states.Methods and analysisEQUIPT is a multicentre, interdisciplinary comparative effectiveness research study in public health. The Tobacco ROI tool already developed in England by the National Institute for Health and Care Excellence (NICE) will be adapted to meet the needs of European decision-makers, following transferability criteria. Stakeholders' needs and intention to use ROI tools in sample countries (Germany, Hungary, Spain and the Netherlands) will be analysed through interviews and surveys and complemented by secondary analysis of the contextual and other factors. Informed by this contextual analysis, the next phase will develop country-specific ROI tools in sample countries using a mix of economic modelling and Visual Basic programming. The results from the country-specific ROI models will then be compared to derive policy proposals that are transferable to other EU states, from which a centralised web tool will be developed. This will then be made available to stakeholders to cater for different decision-making contexts across Europe.Ethics and disseminationThe Brunel University Ethics Committee and relevant authorities in each of the participating countries approved the protocol. EQUIPT has a dedicated work package on dissemination, focusing on stakeholders' communication needs. Results will be disseminated via peer-reviewed publications, e-learning resources and policy briefs.

Project description:ObjectivesMany guidelines and regulations allow children and adolescents to be enrolled in research without the prospect of clinical benefit when it poses minimal risk. However, few systematic methods exist to determine when research risks are minimal. This situation has led to significant variation in minimal risk judgments, raising concern that some children are not being adequately protected. To address this concern, we describe a new method for implementing the widely endorsed "risks of daily life" standard for minimal risk. This standard defines research risks as minimal when they do not exceed the risks posed by daily life activities or routine examinations.MethodsThis study employed a conceptual and normative analysis, and use of an illustrative example.ResultsDifferent risks are composed of the same basic elements: Type, likelihood, and magnitude of harm. Hence, one can compare the risks of research and the risks of daily life by comparing the respective basic elements with each other. We use this insight to develop a systematic method, direct comparative analysis, for implementing the "risks of daily life" standard for minimal risk. The method offers a way of evaluating research procedures that pose the same types of risk as daily life activities, such as the risk of experiencing anxiety, stress, or other psychological harm. We thus illustrate how direct comparative analysis can be applied in practice by using it to evaluate whether the anxiety induced by a respiratory CO2 challenge poses minimal or greater than minimal risks in children and adolescents.ConclusionsDirect comparative analysis is a systematic method for applying the "risks of daily life" standard for minimal risk to research procedures that pose the same types of risk as daily life activities. It thereby offers a method to protect children and adolescents in research, while ensuring that important studies are not blocked because of unwarranted concerns about research risks.

Dataset Information

Evaluating possible confounding by prescriber in comparative effectiveness research.

Publications

Evaluating possible confounding by prescriber in comparative effectiveness research.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets