SEARCH · PubMed Health

Results for “Propensity Score”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources

PubMed2003

[An effective method to reduce bias between two compared groups: propensity score].

OBJECTIVE: Through introduction of principal theory and algorithm of propensity score to design SAS macro programs for binary data. METHODS: Propensity score method was used to compare the differences of character variables between two groups, and the association of DNR (Do Not Resuscitate) with the mortality of congestive heart failure was evaluated with different methods. RESULTS: Significant differences among the character variables between two groups were effectively balanced with stratification or matching method. The odds ratios of DNR with the in-hospital mortality rate of congestive heart failure were estimated identical with different algorithms and to find that the association of DNR to in-hospital mortality was highly significant. CONCLUSION: Propensity score was a good algorithm that could be used to analyze any kind of observational data for matching the effects among the character variables.

Algorithms↗

PubMed2005

[A Monte-Carlo study for propensity score methods].

OBJECTIVE: To evaluate the statistical property of the estimators on exposure effect and value of propensity score methods in practical use. METHODS: Simulation data, propensity score methods (PSM) were conducted to assess bias and efficiency, under both with/without model misspecification conditions. RESULTS: Under model misspecification conditions, propensity score methods had better robustness than model based methods. CONCLUSION: For large and complicated data, propensity score methods had more flexibility in practical use than model based methods.

Algorithms↗

PubMed2003

Comparison of logistic regression versus propensity score when the number of events is low and there are multiple confounders.

The aim of this study was to use Monte Carlo simulations to compare logistic regression with propensity scores in terms of bias, precision, empirical coverage probability, empirical power, and robustness when the number of events is low relative to the number of confounders. The authors simulated a cohort study and performed 252,480 trials. In the logistic regression, the bias decreased as the number of events per confounder increased. In the propensity score, the bias decreased as the strength of the association of the exposure with the outcome increased. Propensity scores produced estimates that were less biased, more robust, and more precise than the logistic regression estimates when there were seven or fewer events per confounder. The logistic regression empirical coverage probability increased as the number of events per confounder increased. The propensity score empirical coverage probability decreased after eight or more events per confounder. Overall, the propensity score exhibited more empirical power than logistic regression. Propensity scores are a good alternative to control for imbalances when there are seven or fewer events per confounder; however, empirical power could range from 35% to 60%. Logistic regression is the technique of choice when there are at least eight events per confounder.

Bias↗

PubMed2003

Utilization of the propensity score method: an exploratory comparison of proxy-completed to self-completed responses in the Medicare Health Outcomes Survey.

BACKGROUND: This research examined the use of the propensity score method to compare proxy-completed responses to self-completed responses in the first three baseline cohorts of the Medicare Health Outcomes Survey, administered in 1998, 1999, and 2000, respectively. A proxy is someone other than the respondent who completes the survey for the respondent. METHODS: The propensity score method of matched sampling was used to compare proxy and self-completed responses. A propensity score is a value that equals the estimated probability of a given individual belonging to a treatment group given the observed background characteristics of that individual. Proxy and self-completed responses were compared on demographics, the SF-36, chronic conditions, activities of daily living, and depression-screening questions. For each individual survey respondent, logistic regression was used to calculate the probability that this individual belonged to the proxy respondent group (propensity score). Pre and post adjustment comparisons were tested by calculating effect sizes. RESULTS: Differences between self and proxy-completed responses were substantially reduced with the use of the propensity score method. However, differences were still found in the SF-36, several demographics, several impaired activities of daily living, several chronic conditions, and one depression-screening question. CONCLUSION: The propensity score method helped to reduce differences between proxy-completed and self-completed survey responses, thereby providing an approximation to a randomized controlled experiment of proxy-completed versus self-completed survey responses.

Activities of Daily Living↗

PubMed2006

Anaemia and mortality in haemodialysis patients: interaction of propensity score for predicted anaemia and actual haemoglobin levels.

BACKGROUND: Haemoglobin levels in haemodialysis patients could represent unknown comorbidities, more severe levels of known comorbidities, as well as therapeutic choice. Thus, integrating factors predictive of anaemia with actual haemoglobin levels might improve prognostic discrimination. METHODS: We retrospectively studied 93,087 patients who started haemodialysis between 1998 and 2000. Clinical and treatment factors from months 4 through 9, derived from Medicare claims, were used to develop propensity scores for anaemia (mean haemoglobin <11 g/dl). Tertiles of propensity scores were interacted with five levels of actual mean haemoglobin to form 15 groups, ranging from low (anaemia) probability with (mean) haemoglobin <10 g/dl to high probability with haemoglobin >or=13 g/dl. Cox proportional hazards regression evaluated mortality and first hospitalization among these groups. RESULTS: The anaemia propensity score improved overall prognostic discrimination. Propensity score adjustment significantly improved prediction of mortality (P<0.0001) after covariate adjustments including haemoglobin. For mortality, the highest and lowest adjusted hazard ratios (AHR) appeared in these groups, respectively: high probability with haemoglobin <10 g/dl (AHR 1.64 [1.54, 1.75], P<0.0001), and low probability with haemoglobin 12 to <13 g/dl (AHR 0.79 [0.74, 0.85], P<0.0001). Higher haemoglobin levels were associated with lower mortality even after propensity score adjustment. Similar patterns resulted for first hospitalization; however, the interaction was significant only for hospitalization (P = 0. 0212). CONCLUSIONS: Integrating factors predictive of anaemia improves overall prognostic discrimination. Propensity score adjustment refines the prognostic association of haemoglobin levels in haemodialysis patients.

Adult↗

PubMed1995

Propensity score adjustment for pretreatment differences between hospitalized and ambulatory patients with community-acquired pneumonia. Pneumonia Patient Outcomes Research Team (PORT) Investigators.

A primary goal of the Pneumonia Patient Outcomes Research Team (PORT) multicenter cohort study is to identify a subgroup of patients with community-acquired pneumonia (CAP) who could be safely treated on an ambulatory basis. The medical outcomes of inpatients and outpatients are to be compared. Propensity score adjustment provides a unified way to control for pretreatment differences in the analysis of all the outcomes in this nonrandomized study by defining "comparable" patients as those with the same propensity score (i.e., the same probability of hospitalization). Data for 747 patients (35.5% hospitalized) with CAP in the Pneumonia PORT study illustrate the development and assessment of a propensity score adjustment. A classification tree algorithm defined seven propensity score strata with hospitalization probabilities ranging from 6.5% to 76.5%. Statistically significant pretreatment imbalances favoring the outpatients were found for 29 of 44 baseline variables considered; after stratification on the propensity score, only 13 of the 29 imbalances remained statistically significant at the 0.05 level. Post hoc stratification on the estimated propensity score consistently reduced, but did not completely eliminate, systematic baseline differences between ambulatory and hospitalized patients with CAP. Regression adjustment can be used in conjunction with propensity score stratification to adjust further for the remaining identified imbalances.

Adult↗

PubMed2005

The reverse propensity score to detect selection bias and correct for baseline imbalances.

The propensity score has been proposed, and for the most part accepted, as a tool to allow for the evaluation of medical interventions in the presence of baseline imbalances arising in the context of observational studies. The lack of an analogous tool to allow for the evaluation of medical interventions in the presence of potentially systematic baseline imbalances in randomized trials has required the use of ad hoc methods. This, in turn, leads to challenges to the conclusions. For example, much of the controversy surrounding recommendations for or against mammography for some age groups stems from the fact that all the randomized trials to study mammography had baseline imbalances, to some extent, in important prognostic covariates. While some of these trials used cluster randomization, baseline imbalances are prevalent also in individually randomized trials. We provide a systematic approach for evaluating medical interventions in the presence of potentially systematic baseline imbalances in individually randomized trials with allocation concealment. Specifically, we define the reverse propensity score as the probability, conditional on all previous allocations and the allocation procedure (restrictions on the randomization), that a given patient will receive a given treatment. We demonstrate how the reverse propensity score allows for both detection of and correction for selection bias, or systematic baseline imbalances.

Biometry↗

PubMed2005

An application of propensity score matching using claims data.

BACKGROUND: Confounding by indication is a common problem in pharmacoepidemiology, where predictors of treatment also have prognostic value for the outcome of interest. The tools available to the epidemiologist that can be used to mitigate the effects of confounding by indication often have limits with respect to the number of variables that can be simultaneously incorporated as components of the confounding. This constraint becomes particularly apparent in the context of a rich data source (such as administrative claims data), applied to the study of an outcome that occurs infrequently. In such settings, there will typically be many more variables available for control as potential confounders than traditional epidemiologic techniques will allow. METHODS: One tool that can indirectly permit control of a large number of variables is the propensity score approach. This paper illustrates the application of the propensity score to a study conducted in an administrative database, and raises critical issues to be addressed in such an analysis. In this example, the effect of statin therapy on the occurrence of myocardial infarction was examined, and numerous potential confounders of this association were adjusted simultaneously using a propensity score to form matched cohorts of statin initiators and non-initiators. RESULTS: The incidence of myocardial infarction observed in the statin treated cohort was lower than the incidence in the untreated cohort, and the magnitude of this effect was consistent with results from randomized placebo controlled clinical trials of statin therapy. CONCLUSIONS: This example illustrates how confounding by indication can be mitigated by the propensity score matching technique. Concerns remain over the generalizability of estimates obtained from such a study, and how to know when propensity scores are removing bias, since apparent balance between compared groups on measured variables could leave variables not included in the propensity score unbalanced and lead to confounded effect estimates.

Cohort Studies↗

PubMed2003

Outcome of mitral valve repair or replacement: a comparison by propensity score analysis.

BACKGROUND: There are no randomized trials comparing outcomes after mitral valve (MV) repair and replacement. Propensity scoring is a powerful tool that has the potential to reduce selection bias in nonrandomized studies. METHODS: From the BC Cardiac Registries, 2,060 patients presented for MV surgery, with or without CABG between 1991 and 2000. We then identified 322 MV repairs who were then matched by propensity score to an equal number of MV replacement patients. We compared survival and freedom from re-operation outcomes using Cox proportional hazards model analysis. Multivariable analysis was then used to compare outcomes in 358 MV repair patients with 352 MV replacement patients who had undergone chordal sparing surgery. RESULTS: The comparison groups generated using propensity scores were well balanced with respect to all collected baseline risk factors. Median follow-up time was 3.4 years. Patients undergoing MV repair had significantly improved survival (RR 0.46; 95% CI, 0.28 to 0.75) but a trend toward more re-operations (RR 2.11; 95% CI, 1.00 to 4.47) compared with patients undergoing replacement. Mitral valve repair patients still had better survival (RR 0.52; 95% CI, 0.32 to 0.85) compared with MV replacement patients who had undergone chordal sparing surgery. CONCLUSIONS: We used propensity score methods to reduce selection bias in a population-based cohort of patients undergoing MV repair/replacement. Repair was associated with better survival, but a trend to increased re-operation.

Disease-Free Survival↗

PubMed2005

A review of the application of propensity score methods yielded increasing use, advantages in specific settings, but not substantially different estimates compared with conventional multivariable methods.

OBJECTIVE: Propensity score (PS) analyses attempt to control for confounding in nonexperimental studies by adjusting for the likelihood that a given patient is exposed. Such analyses have been proposed to address confounding by indication, but there is little empirical evidence that they achieve better control than conventional multivariate outcome modeling. STUDY DESIGN AND METHODS: Using PubMed and Science Citation Index, we assessed the use of propensity scores over time and critically evaluated studies published through 2003. RESULTS: Use of propensity scores increased from a total of 8 reports before 1998 to 71 in 2003. Most of the 177 published studies abstracted assessed medications (N=60) or surgical interventions (N=51), mainly in cardiology and cardiac surgery (N=90). Whether PS methods or conventional outcome models were used to control for confounding had little effect on results in those studies in which such comparison was possible. Only 9 of 69 studies (13%) had an effect estimate that differed by more than 20% from that obtained with a conventional outcome model in all PS analyses presented. CONCLUSIONS: Publication of results based on propensity score methods has increased dramatically, but there is little evidence that these methods yield substantially different estimates compared with conventional multivariable methods.

Bias↗

PubMed2003

Propensity score matching: an illustrative analysis of dose response.

BACKGROUND: Health services researchers are often interested in the effect of a treatment or a service in situations in which randomization is difficult or impossible. One useful alternative involves propensity score methods, a means for matching members of different groups based on a range of characteristics. Under certain assumptions, comparisons of the matched groups reveal the impact of the treatment of interest. OBJECTIVES: This article reviews propensity score methods and illustrates their use in an analysis of dose response, the relationship between the volume of services received, and treatment outcomes. In mental health policy, this question is central to key issues such as parity. RESEARCH DESIGN: Data for the illustrative analysis are taken from a well-known study of children's mental health services. This analysis estimates the impact of outpatient therapy based on comparisons of individuals receiving different treatment doses. Those comparisons are adjusted for preexisting observed differences among the groups using propensity score methods. SUBJECTS: The study includes 301 participants aged 5 to 18 years treated at the study sites. MEASURES: The analyses are based on family characteristics and the mental health status of children and adolescents reported in interviews with parents as well as administrative data on service use. RESULTS: 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. CONCLUSIONS: These analyses illustrate the potential value of propensity score methods to health services researchers.

Adolescent↗

PubMed2006

Outcome of inappropriate empirical antibiotic therapy in patients with Staphylococcus aureus bacteraemia: analytical strategy using propensity scores.

Patients with Staphylococcus aureus bacteraemia (SAB) who received either inappropriate or appropriate empirical therapy were compared by using two risk stratification models: (1) a cohort study using a propensity score to adjust for confounding by empirical treatment assignment; and (2) a propensity-matched case-control study. Inappropriate empirical therapy was modelled on the basis of patient characteristics, and included in the multivariate model to adjust for confounding. For case-matching analysis, patients with inappropriate empirical therapy (cases) were matched to those with appropriate empirical therapy (controls) on the basis of the propensity score (within 0.03 on a scale of 0-1). In total, 238 patients with SAB were enrolled in the cohort study. Characteristics associated with inappropriate empirical therapy were methicillin resistance, underlying haematological malignancy, no history of colonisation with methicillin-resistant S. aureus, and a long hospital stay before SAB. These variables were included in the propensity score, which had an area under the receiver operating characteristics curve of 85%. In the cohort study, SAB-related mortality was 39% (45/117) for inappropriate empirical therapy vs. 28% (34/121) for appropriate empirical therapy (odds ratio (OR) 1.60; 95% CI 0.93-2.76). After adjustment for independent predictors for mortality and the propensity score, inappropriate empirical therapy was not associated with mortality (adjusted OR 1.39; 95% CI 0.62-3.15). In the matched case-control study (50 pairs), SAB-related mortality was 32% (16/50) for inappropriate empirical therapy and 28% (14/50) for appropriate empirical therapy (McNemar's test; p 0.85; OR 1.15; 95% CI 0.51-2.64). In conclusion, inappropriate empirical therapy resulted in only a slight tendency towards increased mortality in patients with SAB.

Aged↗

PubMed2004

Advanced statistics: the propensity score--a method for estimating treatment effect in observational research.

Observational studies assessing the effect of a particular treatment or exposure may be subject to bias, which can be difficult to eliminate using standard analytic techniques. Multivariable models are commonly used in observational research to assess the relationship between a certain exposure or treatment and an outcome, while adjusting for important variables necessary to ensure comparability between the groups. Large differences in the observed covariates between two study groups may exist in observational studies in which the investigator has no control over who was allocated to each treatment group, and these differences may lead to biased estimates of treatment effect. When there are large differences in important prognostic characteristics between the treatment groups, adjusting for these differences with conventional multivariable techniques may not adequately balance the groups, and the remaining bias may limit valid causal inference. Use of a propensity score, described as a conditional probability that a subject will be "treated" based on an observed group of covariates, may better adjust covariates between the groups and reduce bias. The purpose of this article is to describe the use of propensity scores to adjust for bias when estimating treatment effects in observational research and to compare use of this technique with conventional multivariable regression. The authors present three methods for integrating propensity scores into observational analyses using a database collected on head-injured trauma patients. The article details the methods for creating a propensity score, analyzing data with the score, and explores differences between propensity score methods and conventional multivariable methods, including potential benefits and limitations. Graphical representations of the analyses are provided as well.

Adolescent↗

PubMed2006

Propensity score analysis in observational studies: outcomes after abdominal aortic aneurysm repair.

OBJECTIVE: Large databases composed of well-designed prospectively collected cohort data provide an opportunity to examine and compare healthcare treatments in actual clinical practice settings. Because the analysis of these data often leads to a retrospective cohort design, it is essential to adequately adjust for lack of balance in patient characteristics when making treatment comparisons. We used matched propensity scoring in a cohort of patients undergoing elective aneurysm repair as an illustrative example of this important statistical method that adjusts for baseline characteristics and selection bias by matching covariables. METHODS: By using prospectively collected clinical data from the National Surgical Quality Improvement Program of the Department of Veterans Affairs, we studied 30-day mortality, 1-year survival, and postoperative complications in 1904 patients who underwent elective AAA repair (endovascular aneurysm repair [EVAR], n=717 (37.7%); open aneurysm repair, n=1187 [62.3%]) at 123 Veterans Health Administration's hospitals between May 1, 2001, and September 30, 2003. In bivariate analysis, patient characteristics and operative and hospital variables were associated with both type of surgery and outcomes of surgery. Therefore, the predicted probability of receiving EVAR was tabulated for all patients by using multiple logistic regression to control for 32 independent demographic and clinical characteristics and then stratified into 5 groups. Patients were matched within strata based on similar levels of the independent measures (a propensity score technique), creating a pseudo-randomized control design. The proportion of patients with the morbidity and mortality outcomes was then compared between the EVAR and open procedures within strata to control for selection. RESULTS: Patients undergoing EVAR had significantly lower unadjusted 30-day (3.1% versus 5.6%, P=.01) and 1-year mortality (8.7% versus 12.1%, P=.018) than patients undergoing open repair. By using propensity scoring, the proportions of EVAR patients experiencing 30-day mortality were equal or less than patients undergoing open procedures for all levels of probability and decreased as the probability of EVAR increased. Furthermore, propensity scoring also showed that patients having EVAR had lower 1-year mortality and experienced fewer perioperative complications. CONCLUSIONS: We used a propensity score approach to examine outcomes after elective AAA repair to statistically control for many factors affecting both treatment selection and outcome. Patients who underwent elective EVAR had substantially lower perioperative mortality and morbidity rates compared with patients having open repair, which was not explained solely by patient selection in an observational dataset.

Aged↗

PubMed2002

Propensity score modeling strategies for the causal analysis of observational data.

Propensity score methods are used to estimate a treatment effect with observational data. This paper considers the formation of propensity score subclasses by investigating different methods for determining subclass boundaries and the number of subclasses used. We compare several methods: balancing a summary of the observed information matrix and equal-frequency subclasses. Subclasses that balance the inverse variance of the treatment effect reduce the mean squared error of the estimates and maximize the number of usable subclasses.

Journal Article↗

PubMed2004

Application of a propensity score to adjust for channelling bias with NSAIDs.

PURPOSE: To compare the relative risks of upper GI haemorrhage (UGIH) in users of Newer versus Older, non-specific NSAIDs when adjusted for channelling bias by regression on individual covariates, a propensity score and both. METHODS: Cohort study of patients prescribed NSAIDs between June 1987 and January 2000. Exposure to Newer and Older non-specific NSAIDs was identified, and risk factors evaluated for each patient. Results of multiple covariate analyses and the propensity scoring technique to assess potential channelling bias in comparisons between Newer and Older non-specific NSAIDs were compared. RESULTS: This study included 7.1 thousand patient years (tpy) exposure to meloxicam, 1.6 tpy exposure to coxibs, and 628 tpy exposure to Older non-specific NSAIDs. Patients receiving Newer NSAIDs were older, more likely to have a history of GI symptoms, and at higher risk for GI complications. Adjusting for these risk factors reduced the relative risks of UGIH on meloxicam and coxibs versus Older non-specific NSAIDs to 0.84 (95%CI 0.60, 1.17) and 0.36 (0.14, 0.97) respectively. CONCLUSIONS: Channelling towards high GI risk patients occurred in the prescribing of Newer NSAIDs. Propensity scores highlighted the markedly different risk profiles of users of Newer and Older non-specific NSAID. Correcting for channelling bias, coxib exposure, but not meloxicam exposure, was associated with less UGIH than Older non-specific NSAID exposure. In the present study, corrections made by regression on a propensity score and on individual covariates were similar.

Adult↗

PubMed2006

Addressing the issue of channeling bias in observational studies with propensity scores analysis.

Randomized Clinical Trials (RCTs) remain the gold standard for determining the utility of pharmaceuticals especially from a safety and efficacy standpoint. However, restrictive entry criteria and stringent protocols can be barriers to generalizing RCT findings to real world practices and outcomes. Observational studies overcome these limitations of RCTs since they are representative of real world populations and practices. Nonetheless, attributing causality remains a major limitation in observational studies, due to the non-random assignment of subjects to treatment. Non-random assignment can lead to imbalances in risk-factors between the groups being compared and thus bias the estimates of the treatment effect. Non-random assignment can be particularly problematic in observational studies comparing older versus newer pharmaceuticals from similar therapeutic classes due to the phenomenon of channeling. Channeling occurs when drug therapies with similar indications are preferentially prescribed to groups of patients with varying baseline prognoses. In this manuscript we discuss the phenomenon of channeling and the use of a statistical technique known an propensity scores analysis which potentially adjusts for the effects of channeling. During the course of this manuscript we discuss tests for determining the quality of the derived propensity score, various techniques for utilizing propensity scores, and also the potential limitations of this technique. With the increasing availability of high quality pharmaceutical and medical claims data for use in observational studies, increased attention must be given to analytic techniques that adjust optimally for non-random assignment and resulting channeling bias. For research studies using observational study designs, propensity score analysis offers a reasonable solution to address the limitation of non-random assignment, especially when RCTs are too costly, time-consuming or not ethically feasible.

Bias↗

PubMed2006

Propensity scores and the surgeon.

BACKGROUND: Evidence-based surgery has been established as a cornerstone of good clinical practice, promising to improve the treatment of patients and the quality of surgical education. However, evidence-based surgery requires dedicated clinicians trained to perform methodologically sound clinical investigations. Statistical knowledge is therefore invaluable. Surgical studies often cannot be randomized. Propensity scores offer a powerful alternative to multivariable analysis in the assessment of observational, non-randomized surgical studies. Unfortunately, many surgeons are unaware of this important analytical approach that has gained increasing stature in medical research. Thus, propensity score analyses are not used often in surgical studies. OBJECTIVE: The purpose of this paper is to provide a comprehensive overview of propensity score analysis, allowing the surgeon to understand the role, advantages and limitations of propensity scores, boosting their development in surgical investigations.

Confounding Factors, Epidemiologic↗