Matching on provider is risky.

OBJECTIVES To illustrate that matching on provider may exacerbate, not remove, bias. STUDY DESIGN AND SETTING The degree of confounding bias depends in part on the proportions of treatment variation that can be ascribed to confounders and to instruments, respectively. This commentary raises the specific example of bias by matching on hospital induced in a study of coronary artery bypass graft surgery patients and illustrates the effect of matching on provider in a constructed example. RESULTS Matching on provider removes a "benign" source of treatment variability, leaving unmeasured confounders as potentially the most important determinants of treatment. CONCLUSIONS Researchers need to articulate the presumed source of pseudorandom variation in observational studies and need to take care not to reduce their effect through unnecessary control.

• Publication year

  2013

• Venue

  Journal of Clinical Epidemiology

• Publication date

  2013-08-01

• Fields of study

  Medicine

• Identifiers
  DOI 10.1016/j.jclinepi.2013.02.012PMID 23849156
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar, PubMed

• No claims are published for this paper.

• No concepts are published for this paper.

• Invited commentary: understanding bias amplification.

  2011cited by this paper
• Effects of adjusting for instrumental variables on bias and precision of effect estimates.

  2011cited by this paper
• Confounding Control in Healthcare Database Research: Challenges and Potential Approaches

  2010cited by this paper
• On a Class of Bias-Amplifying Variables that Endanger Effect Estimates

  2010cited by this paper
• Aprotinin during Coronary-Artery Bypass Grafting and Risk of Death

  2009cited by this paper
• Aprotinin during coronary-artery bypass grafting and risk of death.

  2008cited by this paper
• Simultaneous assessment of short-term gastrointestinal benefits and cardiovascular risks of selective cyclooxygenase 2 inhibitors and nonselective nonsteroidal antiinflammatory drugs: an instrumental variable analysis.

  2006cited by this paper
• Pair-matching-a reappraisal of a popular technique.

  1977cited by this paper

• Assessment of the E-value in the presence of bias amplification: a simulation study

  2024cites this paper
• Risk of anastomotic leakage with two-row versus three-row manual circular staplers in colorectal anastomosis: a U.S. cohort study

  2023cites this paper
• Bias amplification of unobserved confounding in pharmacoepidemiological studies using indication‐based sampling

  2022cites this paper
• Constructing Several Comparison Groups

  2020cites this paper
• Modern Algorithms for Matching in Observational Studies

  2020cites this paper
• Long-term risk of hepatocellular carcinoma mortality in 23220 hospitalized patients treated with micafungin or other parenteral antifungals.

  2019cites this paper
• Balancing Multiple Goals In Observational Study Design

  2017cites this paper
• Propensity Scores in Pharmacoepidemiology: Beyond the Horizon

  2017cites this paper
• Instrumental variables as bias amplifiers with general outcome and confounding

  2017cites this paper
• DQe-v: A Database-Agnostic Framework for Exploring Variability in Electronic Health Record Data Across Time and Site Location

  2017cites this paper
• Association between statins and infections among patients with diabetes: a cohort and prescription sequence symmetry analysis

  2016cites this paper
• Constructed Second Control Groups and Attenuation of Unmeasured Biases

  2016influential citation
• Use and Interpretation of Propensity Scores in Aging Research: A Guide for Clinical Researchers

  2016cites this paper
• Short-term risk of liver and renal injury in hospitalized patients using micafungin: a multicentre cohort study.

  2016cites this paper
• Comparative effectiveness of injectable paliperidone palmitate versus oral atypical antipsychotics: early postmarketing evidence.

  2015cites this paper
• Antibiotic Exposure and Juvenile Idiopathic Arthritis: A Case–Control Study

  2015cites this paper
• Visualizing Anomalies in Electronic Health Record Data: The Variability Explorer Tool

  2015cites this paper
• Evaluating Possible Confounding by Prescriber in Comparative Effectiveness Research

  2015cites this paper
• Antihyperglycaemic treatment patterns, observed glycaemic control and determinants of treatment change among patients with type 2 diabetes in the United Kingdom primary care: a retrospective cohort study

  2014cites this paper
• Methods for comparative effectiveness research/patient-centered outcomes research: from efficacy to effectiveness.

  2013cites this paper