Reducing bias in cancer research: application of propensity score matching.
PubMedPublished 1 January 2008Open access
Bryce B. Reeve, Ashley Wilder Smith, Neeraj K. Arora, Ron D. Hays
Citations51
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
Results indicated that propensity score matching is an efficient and useful way to create a matched case-control study out of a large cohort study, and allows confidence in the strength of the observed outcomes of the study.
Abstract
In cancer observational studies, differences between groups on confounding variables may have a significant effect on results when examining health outcomes. This study demonstrates the utility of propensity score matching to balance a non-cancer and cancer cohort of older adults on multiple relevant covariates. This approach matches cases to controls on a single indicator, the propensity score, rather than multiple variables. Results indicated that propensity score matching is an efficient and useful way to create a matched case-control study out of a large cohort study, and allows confidence in the strength of the observed outcomes of the study.
Keywords
MathematicsEconomics, Econometrics and Finance
PubMedThe MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection.
33,975 Citations1992John E. Ware, Cathy D. Sherbourne
A 36-item short-form survey designed for use in clinical practice and research, health policy evaluations, and general population surveys to survey health status in the Medical Outcomes Study is constructed.
BiometrikaThe central role of the propensity score in observational studies for causal effects
30,963 Citations1983Paul R. Rosenbaum, Donald B. Rubin
Statistics in MedicinePropensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group
5,302 Citations1998Ralph B. D’Agostino
This tutorial discusses the uses of propensity score methods for bias reduction, gives references to the literature and illustrates the uses through applied examples.
The American StatisticianConstructing a Control Group Using Multivariate Matched Sampling Methods That Incorporate the Propensity Score
4,830 Citations1985Paul R. Rosenbaum, Donald B. Rubin
The use of multivariate matching methods are illustrated in an observational study of the effects of prenatal exposure to barbiturates on subsequent psychological development.
Annals of Internal MedicineEstimating Causal Effects from Large Data Sets Using Propensity Scores
2,879 Citations1997Donald B. Rubin
Propensity score methods generalize subclassification in the presence of many confounding covariates, such as age, region of the country, and sex, in a study of smoking and mortality.
Statistics in MedicineA critical appraisal of propensity‐score matching in the medical literature between 1996 and 2003
1,250 Citations2007Peter C. Austin
Recommendations are provided for the analysis and reporting of studies that employ propensity‐score matching and common errors included using the log‐rank test to compare Kaplan–Meier survival curves in the matched sample, and failing to account for the matched nature of the data.
American Journal of EpidemiologyInvited Commentary: Propensity Scores
1,056 Citations1999Marshall M. Joffe, Peter Rosenbaum
The authors present a new result of using propensity scores in case-cohort studies and review the uses and limitations of propensity scores and provide a brief outline of associated statistical theory.
Cambridge University Press eBooksMatching Using Estimated Propensity Scores: Relating Theory to Practice
1,032 Citations2006Neal Thomas
These results delineate the wide range of settings in which matching on estimated linear propensity scores performs well, thereby providing useful information for the design of matching studies and applying theoretical approximations to practice.
American Journal of EpidemiologyResults of Multivariable Logistic Regression, Propensity Matching, Propensity Adjustment, and Propensity-based Weighting under Conditions of Nonuniform Effect
730 Citations2005Tobias Kurth, Alexander M. Walker +5 more
Five methods for evaluating the effect of tissue plasminogen activator on death among 6,269 ischemic stroke patients registered in a German stroke registry were compared: multivariable logistic regression, propensity score-matched analysis, regression adjustment with the propensity score, and two propensity Score-based weighted methods-one estimating the treatment effect in the entire study population (inverse-probability-of-treatment weights), another in the treated population (standardized-mortality-rat
Journal of the American Statistical AssociationUsing Multivariate Matched Sampling and Regression Adjustment to Control Bias in Observational Studies
709 Citations1979Donald B. Rubin
CirculationPropensity Scores in Cardiovascular Research
450 Citations2007Ralph B. D’Agostino
The propensity score for an individual, defined as the conditional probability of being treated given the individual’s covariates, can be used to balance the covariates in the 2 groups and thus reduce this bias.
Sociological Methodology6. Matching with Multiple Controls to Estimate Treatment Effects in Observational Studies
365 Citations1997Herbert L. Smith
BiometricsEfficiency of Case-Control Studies with Multiple Controls Per Case: Continuous or Dichotomous Data
250 Citations1975Hans Ury
The statistical efficiency of a study design which matches each case with k controls is compared with that of the standard matched-pairs setup and the Pitman efficiency is compared for large-sample dichotomous situations.
PubMedCancer, comorbidities, and health-related quality of life of older adults.
209 Citations2008Ashley Wilder Smith, Bryce B. Reeve +6 more
Results indicated that across most cancer types, cancer patients reported significantly more comorbid conditions and poorer physical and mental health compared with patients without cancer.
CancerHealth‐related quality of life of cancer and noncancer patients in Medicare managed care
166 Citations2003Frank Baker, Samuel C. Haffer +1 more
Data from the Health Care Financing Administration's (HCFA) Medicare Health Outcomes Survey of patients enrolled in managed care services through Medicare were analyzed and provided baseline estimates of quality of life of cancer survivors in comparison to a frequency age‐matched cohort of noncancer patients.
Medical CarePropensity Score Matching
144 Citations2003E. Michael Foster
Analyses using propensity score matching suggest that added services improve treatment outcomes, especially child functioning, however, at least for the services and outcomes considered, the marginal benefits to high levels of treatment are limited.
PubMedOverview of the SEER--Medicare Health Outcomes Survey linked dataset.
125 Citations2008Anita Ambs, Joan L. Warren +4 more
The SEER-MHOS data as a tool to study cancer among Medicare enrollees is described and some descriptive statistics from the database are presented in order to highlight the strengths of the database.
Health and Quality of Life OutcomesThe Medicare Health Outcomes Survey program: Overview, context, and near-term prospects
69 Citations2004Nathaniel Jones, Stephanie Jones +1 more
The HOS program represents an important vehicle for collecting outcomes data from Medicare beneficiaries and many stakeholders recommend the implementation of incentives to encourage the use of the data, while others identify the need to monitor the health status of plan disenrollees.
Scholarworks (University of Massachusetts Amherst)Consider Propensity Scores to Compare Treatments
37 Citations2020Lawrence M. Rudner, Peyton Johnette
PubMedMeasuring and improving health outcomes in Medicare: the Medicare HOS program.
31 Citations2004Samuel C. Haffer, Sonya E. Bowen
The importance of functional status assessment is discussed, how researchers and quality improvement professionals are using the data to explore functional status measurement issues are explained, policy and programmatic implications for CMS, and opportunities to improve health care practice are identified.