Journal of Clinical Epidemiology
Volume 59, Issue 5 , Pages 437.e1-437.e24, May 2006

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

  • Til Stürmer

      Affiliations

    • Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA
    • Division of Preventive Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA
    • Corresponding Author InformationCorresponding author. Tel.: 617-278-0627; fax: 617-232-8602.
    • ,
  • Manisha Joshi

      Affiliations

    • Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA
    • ,
  • Robert J. Glynn

      Affiliations

    • Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA
    • Division of Preventive Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA
    • ,
  • Jerry Avorn

      Affiliations

    • Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA
    • ,
  • Kenneth J. Rothman

      Affiliations

    • Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA
    • Department of Epidemiology, Boston University School of Public Health, Boston, MA, USA
    • ,
  • Sebastian Schneeweiss

      Affiliations

    • Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA

Accepted 15 June 2005. published online 14 October 2005.

Article Outline

Abstract 

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.

Keywords: Propensity score, Epidemiology, Confounding, Bias, Statistical methods, Clinical effectiveness

 

Back to Article Outline

1. Introduction 

Randomized controlled trials are considered the gold standard for assessing the efficacy of medications, medical procedures, or clinical strategies. Nevertheless, particularly for research on the prevention of chronic disease, randomized trials are often infeasible because of their size, time, and budget requirements, as well as questionable generalizability or ethical constraints [1].

On the other hand, nonexperimental studies of interventions have frequently been criticized because of their potential for selection bias. This concern reached a crescendo with the disparity in estimated effects of hormone replacement therapy from randomized trials and nonexperimental studies [2]. This imbroglio highlighted the need to develop and apply improved methods to reduce bias in nonexperimental studies in which selection bias or confounding is likely to occur [3].

The use of multivariate confounder scores to combine many covariates into a single variable can be traced back to Miettinen in 1976 [4]. In 1983, Rosenbaum and Rubin [5] developed the concept of propensity scores (PS) estimated at baseline to control for selection bias in cohort studies. This technique has become popular to control confounding bias in epidemiologic studies that assess the outcomes of drugs and medical procedures. Propensity scores estimate the predicted probability (propensity) of use of a given drug or procedure in a particular subject, based on his or her characteristics when the treatment is chosen. In principle, the effect of the treatment can then be measured among patients who have the same predicted propensity of treatment, thus controlling for confounding [5].

Use of PS to reduce bias is especially appealing because, under the assumption that all relevant predictors of treatment have been adequately captured, subjects with the same PS should have the same chance of receiving treatment. Propensity scores are therefore often conceptualized as mimicking randomized trials, although they do so only with respect to factors that have been adequately measured. Randomization, in contrast, removes bias from both measured and unmeasured factors. Propensity scores allow simultaneous control for confounding by several variables in situations where conventional multivariable models might not be appropriate, owing to the small number of outcomes. Propensity scores, however, are frequently used in settings where the outcome is common; their value in this situation is not yet clear. We sought to review the application of PS in the medical literature and to assess its practical value.

Back to Article Outline

2. Propensity scores 

2.1. Background 

A propensity score can be defined as the probability of exposure to, for example, a treatment, given observed covariates [5]. The score is usually estimated using a multivariable logistic regression model, but can be estimated with a variety of multivariable scoring functions. In a logistic model, the scores range from 0 to 1 and reflect the estimated probability, based on the subject's characteristics, that the subject will receive the treatment of interest, such that individuals with the same estimated PS have the same chance of receiving treatment. Any two subjects with the same PS can have different values for specific covariates, but overall the covariates entered in the PS model will tend to be balanced for treated and untreated subjects with similar PS. This balance of covariates can easily be checked and the performance of PS to achieve this goal can be clearly communicated (e.g., by presenting the distribution of covariates in exposed and unexposed separately, stratified by quintiles of the PS).

By estimating the PS and analyzing the data within homogeneous levels of PS, in theory one can achieve a virtual randomization, in which comparable patients are separated into the exposed and unexposed groups. Because PS are estimated using measured data, however, they cannot control for unmeasured or imperfectly measured variables. Therefore, residual systematic bias cannot be excluded.

Once PS are estimated, they can be used in various ways to control for selection bias or confounding in nonexperimental cohort studies. Possible implementations include matching on the PS, stratified analysis using PS as the stratification variable, and combinations of these two approaches with conventional multivariable outcome modeling. In theory, within each PS stratum, some patients will have received the treatment of interest and others will not. In practice, however, this is not always the case (see Fig. 1), and how one uses PS in analyses can make a difference.

  • View full-size image.
  • Fig. 1. 

    Nonoverlap of the propensity score distributions among exposed and unexposed subjects. In this example, subjects with low propensity scores are never exposed, and subjects with high propensity scores are always exposed.

2.2. Alternative ways to apply propensity scores 

2.2.1. Matching 

One strategy is to match each exposed subject to one or more unexposed subjects with a similar PS, thus avoiding the complexity of matching within multiple strata [6]. A variety of matching methods are available to identify unexposed subjects with PS similar to those of exposed subjects [7], [8]. Effective balancing is achieved by any matching procedure that produces good agreement between the mean PS in exposed and unexposed subjects [9]. Selecting an equal number of exposed and unexposed subjects within categories of the PS (frequency matching), instead of individual matching, enables the inclusion of exposed subjects for whom no exact unexposed match can be found, but may introduce bias stemming from nonoverlapping ranges for exposed and unexposed subjects at the extremes of the distribution of the PS [6]. This bias can be avoided by restricting analyses to the range of PS common to both exposed and unexposed patients: that is, excluding unexposed patients with a PS lower than the lowest PS observed in exposed patients and excluding exposed patients with a PS higher than the highest PS observed in unexposed patients. Plotting the PS distribution for exposed and unexposed subjects is an easy diagnostic for nonoverlap (see Fig. 1).

2.2.2. Subclassification on propensity score 

As an alternative to matching, one can include all available subjects in an analysis and control for the PS. This can be achieved by simple stratification or modeling of the PS–disease association (e.g., as a continuous covariate). Often, a model with indicator variables for quintiles of the observed PS is used [10], but control for confounding may be better when PS are modeled as continuous variables [11]. Again, inclusion of all subjects might introduce bias due to inclusion of subjects with a PS from outside the mutual range of scores among exposed and unexposed. Similar to matching, this bias may be reduced by excluding unexposed patients with a PS lower than the lowest PS observed in exposed patients and vice versa.

2.2.3. Combinations with traditional multivariate outcome modeling 

An issue of some controversy in use of PS is whether better control of confounding can be achieved, and hence better estimates of the effect of the treatment on the outcome can be obtained, by including the PS along with other important predictors of the outcome [12], [13]. In theory, confounding can be controlled and a treatment effect can be estimated validly if only one of the two models—the treatment model (PS) or the outcome model (conventional multivariate modeling)—is specified correctly. Strategies that include both approaches, and that thus involve possibly redundant control of confounding, have been called doubly robust [14]. The theory behind these methods is complex, however, and software tools with adequate documentation are not yet available.

Back to Article Outline

3. Literature search and abstracting 

We identified studies in which the propensity score was used through PubMed and Science Citation Index. Initially, a keyword search was performed through PubMed, identifying studies including the term propensity. This broad search yielded 5,311 unduplicated references published from 1983 through December 31, 2003. After review of the abstracts, we identified 167 articles that used propensity score methods in the study of medical interventions and health outcomes (excluding articles focusing solely on methodological or statistical aspects, editorials, review articles or letters, and articles in languages other than English).

To increase the sensitivity of our search, we also searched for articles that cited one of the important propensity score methods articles [5], [6], [13], [15], [16], [17]. This search yielded another 73 articles. All these reports were obtained and read by one of the authors. We excluded 48 articles that did not include analysis of data (28), randomized clinical trials (9), case-control studies (2), and articles primarily analyzing cost-effectiveness (6) or practice patterns (3).

Back to Article Outline

4. Results 

Our search revealed 58 substantive medical research studies that used PS in 2003 [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], [70], [71], [72], [73], [74], [75], 38 in 2002 [76], [77], [78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96], [97], [98], [99], [100], [101], [102], [103], [104], [105], [106], [107], [108], [109], [110], [111], [112], [113], 28 in 2001 [114], [115], [116], [117], [118], [119], [120], [121], [122], [123], [124], [125], [126], [127], [128], [129], [130], [131], [132], [133], [134], [135], [136], [137], [138], [139], [140], [141], 6 in 2000 [142], [143], [144], [145], [146], [147], 5 in 1999 [148], [149], [150], [151], [152], 5 in 1998 [153], [154], [155], [156], [157], and a total of 5 before 1998 [158], [159], [160], [161], [162]. Additional articles found through a citation search of the significant methods articles written about PS, using Science Citation Index, yielded 13 medical research studies that used PS in 2003 [163], [164], [165], [166], [167], [168], [169], [170], [171], [172], [173], [174], [175], 13 in 2002 [176], [177], [178], [179], [180], [181], [182], [183], [184], [185], [186], [187], [188], 11 in 2001 [189], [190], [191], [192], [193], [194], [195], [196], [197], [198], [199], 3 in 2000 [200], [201], [202], 1 in 1999 [203], 3 in 1998 [204], [205], [206], and a total of 3 before 1998 [207], [208], [209]. We present the number of studies with results based on PS methods published in each of these years in Fig. 2.

  • View full-size image.
  • Fig. 2. 

    Number of medical research studies listed in PubMed and Science Citation Index using propensity score methods to control for confounding, according to year of publication.

After further review of articles, 15 articles were excluded from further analysis as the outcomes were continuous and it was not possible to calculate an odds ratio or risk ratio [59], [86], [88], [111], [120], [124], [137], [150], [152], [160], [162], [193], [195], [196], [206]. The final selection of studies abstracted comprised 70 articles from 2003, 48 articles from 2002, 33 articles from 2001, 9 from 2000, 4 from 1999, 7 from 1998, and 6 articles before 1998.

For all selected reports published through 2003, we abstracted the following items: (i) the number of variables used to predict treatment and outcome, respectively; (ii) the unadjusted (crude) estimate for the treatment-outcome association; (iii) the estimates for the treatment-outcome association adjusted by use of PS matching, PS adjustment, and/or multivariable outcome models, including models without PS and with PS as well as covariates; (iv) the predictive value of the PS as assessed by the area under the receiver operating characteristic (ROC) curve (equivalent to the c-statistic in logistic regression); and (v) the percent of exposed participants that could be matched to unexposed participants (where applicable). We extracted or calculated odds ratios or relative risks whenever adequate data were presented.

Studies published in 2003 (Table E1) and six corresponding tables for the years 2002 and prior years appear as online supplemental tables (available at www.Elsevier.com on the journal's website). The following results are based on all 177 substantive studies reporting on dichotomous exposures and outcomes published through 2003. Among the medical specialties covered in these reports were cardiology, including cardiac and vascular surgery (N = 90); general internal medicine (N = 34); oncology (N = 20); nephrology (N = 9); psychiatry (N = 4); and rheumatology (N = 2). The treatments studied included medications (N = 60); surgical interventions (N = 51); catheterization (N = 13); other medical procedures, including care after myocardial infarction and in end-stage renal failure; lifestyle factors; or a wide variety of other comparisons. The main outcome assessed was mortality (N = 118). Other outcomes included myocardial infarction (N = 6), stroke (N = 3), and a wide variety of other outcomes (including complications of infection, gastrointestinal events, and emergency hospitalizations).

The number of exposed subjects (or unexposed subjects, if this number was smaller) ranged from 61 to >1,380,000, and the number of outcomes ranged from 23 to 285,965. In 109 studies, the number of exposed subjects was larger than the number of subjects who experienced the outcome; in 13 studies, it was smaller.

To estimate the PS, 2 to 112 variables were used (in those reports in which this information was presented), compared with 1 to 45 used in multivariable outcome models. Direct comparison of the number of variables was possible in 90 studies, of which only 51 used more variables to estimate the PS than to estimate the corresponding outcome model; 27 used fewer variables to do so. Sixty-five studies had fewer than 8 outcomes for each variable entered into the PS model; this is a setting for which the use of PS methods has been shown to be advantageous compared with conventional outcome modeling [210]. In 60% of studies (96 of 161), the number of outcomes would have been sufficient to enter all variables used in the propensity score model in the corresponding outcome model.

The area under the ROC curve or c-statistic was presented for 73 studies. It ranged from .56 to .93, indicating poor to good predictive power. The lowest predictive value (c = .56) was achieved predicting the annual volume of patients treated by admitting physicians (in a study assessing its association with mortality in acute myocardial infarction) [199]; the highest (c = .94) was achieved when predicting revascularization in coronary artery disease [43] and thrombolysis in patients with stroke [47]. Very high values (c > .90) were reported in six additional studies for treatments including statins [68], amiodarone after acute myocardial infarction [87], chemotherapy in colon cancer [85], heart valve repair vs. replacement [122], bilateral thoracic artery bypass [148], and a hospital comparison [90].

Fifty-one studies used matching on the PS as either the main analytic strategy or as one of several analytic strategies presented. The percentage of exposed participants that could be matched to an unexposed participant was presented for 49 studies, and ranged from 26% to 100% (median = 90%).

Most studies showed clear evidence of confounding, with substantial changes in the point estimate after adjustment. Whether PS methods or conventional outcome models were used to control for confounding, however, seemed to matter little in most of those 69 studies in which such a comparison was possible. These included 10 studies in which the authors made a qualitative statement that (mostly PS) analyses showed “similar” results.

In 20% of studies (14 of 69), however, there was a greater than 20% difference in the point estimate obtained from the conventional outcome model compared with any propensity score method presented [22], [24], [34], [51], [52], [62], [73], [100], [102], [105], [107], [121], [123], [192]. We used this arbitrary cut-point as a marker of a substantial difference in results. Of these, five [22], [100], [107], [121], [123] showed results not meeting our 20% criterion for at least one of the analytic strategies using PS. In four of these studies, the PS strategy not meeting criterion was when the PS was added to the conventional multivariable outcome model [22], [100], [107], [123]. In the study by Foody et al. [121], the result matched on the PS did not meet our criterion. This left the remaining 13% of studies (9 of 69) in which all PS analyses presented showed a substantial difference compared with conventional outcome models.

Back to Article Outline

5. Discussion 

The number of studies using PS methods, though not yet large, is climbing rapidly [211]. According to the authors of many of these studies, the main reason for using PS methods was better control for confounding compared with conventional multivariable outcome modeling. We found no empirical evidence, however, that PS analyses controlled confounding more effectively than did conventional outcome modeling in the majority of the studies where results from both methods were presented. Potentially meaningful differences in the control of confounding were observed in <15% of studies. Because the true underlying association is unknown, it remains unclear whether these differences are due to better control for confounding using the PS or whether adjusting for an inaccurate PS distorted results in some studies [212], [213]. The use of PS as the only analytic technique applied comes at the price of losing potentially useful information about predictors of outcome. It therefore seems desirable to use PS only if a reduction in bias or an improvement in efficiency can be achieved.

Cook and Goldman [214] compared the performance of tests of significance under the null hypothesis (i.e., assuming no difference between treatments) for PS and for traditional multivariable outcome models using simulations. Propensity scores appeared to produce valid results in most circumstances, but were biased in situations with very strong treatment–confounder associations.

In some practical situations, the choice of analytic method will be limited. Because 10 events per covariate is usually considered to be a minimum requirement for stable estimates in multivariable models [215], [216], PS analyses combining multiple covariates into a single score are especially desirable if the treatment is common and the outcome is rare [217], [218]. A recent simulation study comparing PS with multivariable outcome models concluded that PS performed better in situations with less than 8 outcomes per covariate [210]. Apart from this specific condition (relevant in 65 of the 161 studies presenting the necessary information), there is little if any practical guidance for researchers regarding when the use of PS will produce different—and, in particular, better— estimates compared with conventional multivariable outcome models.

Propensity scores are used to reduce bias. Drake [219] observed that the magnitude and direction of bias resulting from omitting an important confounder from analysis was similar in multivariable outcome modeling and in estimating the treatment–outcome relation controlling for PS. This observation suggests that PS may not be superior to conventional multivariable outcome models in controlling bias from unobserved confounders.

Several strategies for using PS are currently being applied in medical research, and often results from more than one of these strategies are reported in a single report. Individual matching on a PS has intuitive appeal, and in those studies that used matching, the proportion of exposed subjects that could be matched ranged from 26% to 100%. Excluding a large proportion of exposed subjects because of a lack of unexposed matches, however, may severely alter the composition of the study population.

Because comparisons may be valid within that altered population, we would therefore not call this issue a bias. Nevertheless, it is essential to appreciate and to describe clearly the differences between this altered population and the original study population. On the other hand, including subjects with a PS outside the overlapping range, such as using conventional outcome modeling or PS methods including nonoverlapping ranges, can lead to bias due to model extrapolation or smoothing. Such subjects might include, for example, patients with absolute indications or contraindications to treatment, who should not be included in any treatment comparison [220], but are usually not recognized using conventional multivariable outcome modeling. Because this is a clear advantage of PS, a graphical exploration similar to Fig. 1 could be used as a routine procedure before doing any multivariable outcome modeling in treatment comparisons. Unfortunately, systematic comparisons of the different strategies to apply PS with respect to validity and efficiency with specific attention to exclusion of participants and nonlinear associations between the PS and the outcome are sparse [221].

Variable selection in constructing PS is at present an ad hoc process that lacks guidelines and well-understood model diagnostics. The area under the ROC curve or c-statistic (from logistic regression) to quantify the predictive power of a model is a well established concept in clinical epidemiology [222]. Its value when assessing the performance of PS to control confounding is unclear, however. Indeed, a very high c-statistic can indicate considerable nonoverlap in PS distributions between exposed and unexposed as shown in Fig. 1.

Some authors argue that variables that only predict treatment choice but are not associated with the study outcome should not be included in the PS [223]. By definition, these are not confounders, but they may increase the area under the ROC curve and thereby erroneously imply a high validity of the PS analysis.

A practical way to assess the value of the PS model in controlling for confounding is to check the balance of important risk factors for the outcome between exposed and unexposed within levels of the estimated PS. This method has the advantage of being driven by substantive knowledge, rather than statistics, and the results can easily be communicated to the reader in a table. It allows direct assessment of comparability of exposed and unexposed by the reader (a clear advantage to using PS methods compared with the black box of the conventional outcome model).

Our review of the application of propensity score methods in the medical literature has several limitations. We may well have missed some studies under our specific search strategy, but this problem should not affect the comparison over time. Important information in understanding similarities and differences between the analytic approaches, including description of the types of variables, variable selection procedures, and measures of model adequacy, could not be abstracted systematically, because these are rarely presented with sufficient detail in published reports.

In conclusion, methods using propensity scores may be good candidates for improving inference in nonexperimental studies, but a better understanding of the benefits and limitations of these methods in practical circumstances is needed. Meanwhile, propensity scores, like any other method, should not be automatically regarded as a preferable and sole method to control for confounding in nonexperimental research, but rather as a promising addition.

Back to Article Outline

Acknowledgments 

The project was funded by a grant from the National Institute on Aging (R01 AG023178).

Back to Article Outline

Appendix. 

Table E1. Substantive studies in 2003 using propensity score methods (PS)
ExposureOutcome Multivariable outcome modelinga
AuthorN nVariables in model, no. nVariables in model, no.CrudePS matchingPS adjustmentWithout PSPS and covariatesAUCb% matchedc
Allen-Ramey, Duong, Goodman et al. [21]960Inhaled steroids153?Hospitalizations?n/a?———1.9 (0.7–4.6)?n/a
Bhatt, Chew, Lincoff et al. [26]10,480Revascularization in acute CAD4,175?Mortality∼64015?——0.8 (0.6–1.0)?0.84n/a
Horwitz, Berlin, Sauer et al. [165]18,821Glycoprotein inhibitors2,52523Bleeding215n/a1.9 (1.4–2.6)—1.4 (1.0–2.0)——?n/a
Rubins, Nelson, Noorbaloochi et al. [64]16,470Lipid lowering drugs7,01231Mortality6,066n/a0.5 (0.5–0.5)—0.8 (0.7–0.9)—0.8 (0.7–0.8)0.83n/a
Seeger, Walker, Williams et al. [68]8,288Statins4,14458MI325n/a2.1 (1.5–3.0)0.7 (0.5–0.9)———0.9270
Vakili, Kaplan, Slater et al. [72]10,847Glycoprotein inhibitors in PTCA2,18317Mortality, CABG, MI507n/a2.1 (1.7–2.6)—1.5 (1.2–1.9)—1.6 (1.2–2.0)?n/a
Schwarz, Smith, Keny et al. [67]61Intra-OP radiation in pancreatic CA30?Mortality∼46n/a“sign. diff.”—1.5 (0.8–2.9)—1.7 (P = .3)?n/a
Ishani, Ibrahim, Gilbertson et al. [49]4,046Residual renal function42417Graft failure443131.0 (0.8–1.4)——1.0 (0.7–1.3)1.0 (0.7–1.3)?n/a
Cho, Bhatt, Marso et al. [34]7,897Catheterization in acute MI3,361?Mortality59040.5—0.7 (0.6–0.8)0.5 (0.1–0.8)—0.74n/a
Harbarth, Garbino, Pugin et al. [46]904Inappropriate AB therapy211?Mortality250202.0 (1.4–2.8)———1.8 (1.2–2.6)?n/a
Leon, Solomon, Mueller et al. [56]285Antidepressant therapy?11Recovery?n/a?—1.9 (1.3–2.7)——?n/a
Urbach, Bell, Swanstrom et al. [174]1,919Bile bypass in pancreatic CA94523Mortality1,845241.2 (1.1–1.4)—1.2 (1.1–1.3)1.2 (1.1–1.3)—0.62n/a
Ascione, Narayan, Rogers et al. [23]250Off vs. on pump CABG7438Mortality50n/a2.5 (1.1–5.0)—2.0 (0.8–5.0)——?n/a
Boening, Friedrich, Hedderich et al. [27]650Off vs. on pump CABG13314Mortality3n/a?2.7 (0.2–31)———?54
Calafiore, Di Mauro, Canosa et al. [30]1,266Off vs. on pump CABG51024Mortality46n/a?0.5 (0.3–1.0)——0.3 (0.2–0.5)??
Hamamoto, Bando, Kobayashi et al. [45]379Heart valve surgery80?Mortality536?———1.0 (0.5–2.0)?n/a
Kurlansky, Williams, Traad et al. [54]987Arterial vs. venous graft CABG41326Mortality40325??—0.8 (0.6–1.0)—??
Srinivasan, Grayson, Pullan et al. [69]?Preop aspirin in CABG17010Mortality5n/a?0.7———0.68?
Stamou, Kapetanakis, Lowery et al. [70]511Minimally invasive valve surgery5617Blood transfusion144n/a0.4 (0.2–0.8)0.8 (0.4–1.8)———?100
Schumacher, Schneider, Reich et al. [66]689Mistletoe in BRCA219?BRCA relapse30?0.5 (0.2–1.3)———0.3 (0.1–0.8)?n/a
Aronow, Novaro, Lauer et al. [22]2,126Lipid lowering drugs after coronary intervention1757Persistent use after 6mo62873.1 (2.8–3.3)2.3 (2.0–2.5)2.8 (2.5–3.1)3.2 (2.9–3.4)3.2 (2.9–3.4)0.7391
Stenestrand, Wallentin L [173]6,891Fibrinolytic therapy in acute MI2,99422Mortality & stroke2,476290.9—0.8 (0.8–0.9)0.9 (0.8–0.9)0.9 (0.8–0.9)0.66n/a
Murray, Singer, Dawson et al. [170]11,150Rehabilitation vs. no rehabilitation group4,738112Discharge4,404360.3 (0.3–0.4)1.38 (1.29–1.47)——1.5 (1.4–1.6)0.8259
Bukhari, Wiles, Lunt et al. [29]335DMARD in RA8210RA progressionn/an/a2.0 (1.4–2.8)—1.5 (1.0–2.2)——?n/a
Abramov, Tamariz, Fremes et al. [19]2,214Pre OP renal function507?Mortality?n/a?—2.0 (1.1–3.8)1.9 (1.3–2.8)—?n/a
Abramov, Tamariz, Serrick et al. [20]1,820Pulsatile perfusion in CABG905?Cardiovascular accident?4?——1.9 (1.1–3.3)2.2 (1.1–4.4)?n/a
Martelli, Miceli, De Palo et al. [168]671Axillary dissection in BRCA1726Mortality (BRCA)607“no sign. diff.”———1.0 (0.7–1.4)0.72n/a
Chan, Bhatt, Chew et al. [32]1,552Statins before PTCA61522Mortality & MI?30.5 (P = .003)———0.4 (P = .04)0.71n/a
Hachamovitch, Hayes, Friedman et al. [43]10,627Revasc in CAD67112Mortality146122.3 (1.4–3.7)——??0.94n/a
Landesberg, Mosseri, Wolf et al. [55]407Revasc in CAD74?Mortality?80.8 (0.5–1.3)—0.4 (0.2–0.7)0.5 (0.3–0.9)—?n/a
Magee, Coombs, Peterson et al. [58]204,602Off vs. on pump CABG17,96919Mortality∼5,860270.8 (0.7–0.9)0.8 (0.7–1.0)—0.8 (0.7–0.8)—?94
McGuire, Anstrom, Peterson [169]9,619Clinical alert2,644?% PTCA after vs. before?120.9 (P = .06)———1.2 (1.0–1.3)?n/a
Moss, Humphries, Gao et al. [60]1,981Mitral valve repair vs. replacement33843Mortality?n/a?0.5 (0.3–0.8)———?95
Trichon, Glower, Shaw et al. [71]2,757Surgery in mitral regurg278?Mortality∼1,400?0.5?—0.6 (0.5–0.7)—??
Yu, Platt, Lanken et al. [175]1,010Pulmonary artery catheter27518Mortality40431.3 (1.0–1.7)0.8 (0.5–1.3)1.0 (0.6–1.7)——0.8551
Brener, Ellis, Schneider et al. [28]10,471Abciximab in PTCA4,81621Mortality1,504250.9 (0.8–1.0)——1.0 (0.9–1.1)?0.83n/a
Calafiore, Di Mauro, Canosa et al. [30]4,381Off vs. on pump CABG∼96010Mortality44n/a?0.4 (0.2–0.8)———??
Karthik, Musleh, Grayson et al. [50]828Off vs. on pump CABG41120Mortality31n/a0.9 (0.5–1.9)—0.8 (0.4–1.9)——0.8n/a
MacDonald, Morant, Goldstein, Burke et al. [167]∼1,380,000Coxib vs. NSAID∼1,60018GI bleeding8,526180.6 (0.2–1.5)—0.4 (0.2–1.1)0.4 (0.1–1.0)0.4 (0.1–1.0)?n/a
Dendukuri, Normand, McNeil [36]37,788Angiography in acute MI9,84916Mortality?n/a“no sign. diff.”“no sign. diff.”———?61
O'Neill, Iturria, Link et al. [171]97Surgery for brain Meta2325(sic!)Mortality803“no sign. diff.”——1.2 (0.7–2.0)1.0 (0.5–1.8)0.89n/a
Chan, Moliterno, Berger et al. [33]4,809Clopidogrel in PTCA3328Mortality88>100.5 (0.3–0.9)———0.5 (0.3–1.0)0.64n/a
Johnson, Jin, Quan et al. [166]11,854Treatment in CHF546?Mortality2,68118?—“analogous”0.5 (0.4–0.7)—?n/a
Peterson, Pollack, Roe et al. [62]60,770Glycoprotein inhib in acute MI15,37923Mortality∼4,860?0.3 (0.3–0.4)0.9 (0.8–1.0)—0.6 (0.6–0.7)0.9 (0.8–1.0)0.7892
Young-Xu, Chan, Liao et al. [75]371Satin use14024Well being (depression)?250.7 (0.5–0.9)——0.6 (0.4–0.9)0.6 (0.4–0.9)0.86n/a
Beddhu, Samore, Roberts et al. [25]2,920Timing of dialysis1,44411Mortality1,07021?—1.5 (1.3–1.6)1.4 (1.3–1.4)1.1 (1.1–1.2)?n/a
Reddan, Szczech, Tuttle et al. [172]4,584Revasc in CAD1,380?Mortality?14?———0.4 (0.3–0.5)?n/a
Winkelmayer, Owen, Levin et al. [74]3,014Late referral in ESRF1,03948Mortality1,429??1.4 (1.2–1.6)—1.4 (1.2–1.5)1.3 (1.2–1.5)0.68100
Ezri, Weisenberg, Khazin et al. [37]644CABG200?Difficult laryngoscopy43?2.0 (P = .02)———“no sign. assoc.”?n/a
Krzyzanowska, Weeks, Earle [53]1,696Treatment in pancreatic CA7465 (?)Mortality≫50%3?—0.50.5 (0.5–0.6)—?n/a
Hall, Summers, Obenchain [44]11,443Insulin (Lispro vs. regular)3,34125Health care utilization (hosp)678n/a?0.7 (0.6–0.8)———?55
Abbott, Trespalacios, Agodoa [18]993AV fistula in dialysis25712CHF462??——“almost same”0.9 (0.6–1.1)0.70n/a
Banbury, White, Blackstone et al. [24]357Vacuum assisted venous return1506Blood usage9440.4 (0.2–0.6)——0.3 (0.2–0.6)0.2 (0.1–0.5)0.79n/a
Flameng, Herijgers, Dewilde et al. [38]1,098Cardioplegy in cardiac surgery50436Mortality7460.6 (0.4–1.0)——0.4 (0.2–0.8)0.4 (0.2–0.9)?n/a
Gillinov, Faber, Houghtaling et al. [40]679Mitral valve repair2328Mortality30n/a0.6 (0.3–1.2)———1.0 (0.4–2.4)0.86n/a
Gillinov, Blackstone, Cosgrove et al. [163]813Mitral valve repair vs. double replacement29513Mortality52110.8 (.4–1.4)————.81n/a
Koch, Khandwala, Nussmeier et al. [52]15,597Female gender in CABG3,59668Mortality25791.7 (1.3–2.2)0.9 (0.5–1.7)—1.3 (1.0–1.8)“not sign.”?26
Murthy, Law, Whooley et al. [61]921Atrial fib after esophagectomy1984Mortality42n/a—4.5 (2.0–9.7)———0.773
Rice, Adelstein, Chidel et al. [63]83Chemo in esophageal CA3115Mortality∼54n/a0.70 (no treated cases)———?65
Schmitz, Weinreich, White et al. [65]582Intraaortic filtration in CABG2789Stroke4660.4“no diff.”—“no diff.”0.4 (0.2–0.8)0.6481
Girou, Brun-Buisson, Taille et al. [41]479Noninvasive ventilation1666Mortality5340.5 (0.3–0.9)——0.4 (0.2–0.8)0.4 (0.2–0.8)?n/a
Gregg, Cauley, Stone et al. [164]3,009Start of physical activity811?Mortality∼40011?—“similar”0.5 (0.4–0.7)—?n/a
Grzybowski, Clements, Parsons et al. [42]19,917Revasc in acute MI4,70523Mortality5,173n/a0.3 (0.3–0.3)0.5 (0.4–0.6)———0.8083
Vikram, Buenconsejo, Hasbun et al. [73]513Valve surgery23012Mortality13170.4 (0.3–0.6)0.5 (0.2–0.9)0.5 (0.2–0.9)0.4 (0.2–0.5)—0.8647
Macaubas, de Klerk, Holt [57]407Antenatal cytokines110?Asthma138n/a0.7 (0.5–1.2)—0.6 (0.4–1.0)——?n/a
Frolkis, Pothier, Blackstone et al. [39]29,244Ventricular ectopy1,08030Mortality1,862302.4 (2.0–2.9)1.5 (1.1–1.9)—1.6 (1.3–1.9)—0.8099
Katzan, Cebul, Husak et al. [51]11,286Pneumonia in stroke63538Mortality644206.1——4.4 (3.7–5.2)3.0 (2.4–3.7)0.83n/a
Christakis, Iwashyna [35]195,553Hospice use in widowers30,91628Mortality4,191n/a?0.9 (0.8–1.0)———?100
Heuschmann, Berger, Misselwitz et al. [47]13,440Thrombolysis in stroke38414Mortality486n/a2.8 (2.0–3.9)1.7 (1.0–2.8)———0.9498
Hughes, Evans, Lightfoot et al. [48]1,228Transfusion in infection5877Mortality218123.6 (2.6–5.0)———0.9 (0.5–1.5)?n/a

Studies were identified from PubMed search and Science Citation Index. Quotations indicate qualitative statement only, data not presented.

aNumbers presented are odds ratios, relative risks, or incidence rate ratios with their corresponding 95% confidence intervals (in brackets).

bArea under the receiver operating characteristics (ROC) curve (c-statistic) from logistic regression model used to estimate the PS.

cPercent of all exposed subjects for whom an unexposed subject could be matched on the PS; exposed subjects without match are discarded from analysis; n/a = not applicable (no matching).

Table E2. Substantive studies in 2002 using propensity score methods (PS)
ExposureOutcome Multivariable outcome modelinga
AuthorN nVariables in model, no. nVariables in model, no.CrudePS matchingPS adjustmentWithout PSPS and covariatesAUCb% matchedc
Goldstein, Racz, Hannan [177]61,873Combined vs. staged PTCA23,46213Mortality∼300140.8 (P = .08)—“no diff.”1.1 (0.9–1.4)—?n/a
Kilborn, Rathore, Gersh et al. [87]17,597Amiodarone after MI55014Mortality5,583101.1 (1.0–1.2)———1.0 (0.9–1.2)>0.78n/a
Mukherjee, Mahaffey, Moliterno [93]5,200Low molecular weight heparin in acute CAD8051030-day revascularization2,192n/a0.7 (0.6–0.8)—0.7 (0.6–0.8)——0.68n/a
Shavelle, Parsons, Sada et al. [100]20,137Early PTCA in MI2,40210Mortality1,415100.71.0—0.8 (0.6–1.0)0.9 (0.7–1.1)—58 (4-digit matching)
Wright, Williams, Cramner et al. [188]213Elevated troponin-I532Mortality19?3.1 (1.3–7.6)—3.2 (1.0–10)2.9 (1.2–7.3)—?n/a
Mehta, McDonald, Gabbai et al. [91]215Timing of nephrology consultation in ARF617Mortality10283.1 (1.7–5.8)——2.5 (1.1–5.9)2.0 (0.8–5.1)—n/a
Shishehbor, Baker, Blackstone et al. [102]3,423Low educational level87428Attenuated heart rate recovery1,141132.7 (2.3–3.1)1.7 (1.4–2.1)—2.4 (2.0–2.8)—0.7191
Sin, McAlister [103]11,942ß-Blocker in CHF1,16216Mortality6,7578?———0.7 (0.7–0.8)?n/a
Weiss, Saynina, McDonald et al. [112]125,892Implanted defibrillator7,789∼33Mortality23,297n/a0.4 (0.4–0.5)0.6 (0.5–0.6)———0.8498
Jasmer, Saukkonen, Blumberg et al. [179]589Rifampicin & pyrazinamide vs. Isoniazid28210Hepatotoxicity18n/a8.5 (1.9–77)—7.8 (1.7–71)——?n/a
Sundararajan, Mitra, Jacobson et al. [106]4,7685-FU in CRC?2,480?Mortality?5?———0.7 (0.6–0.7)?n/a
Ferraris, Ferraris, Joseph et al. [176]2,606Aspirin before CABG706?Blood transfusion571171.3 (1.0–1.6)—“sign in quintile 1”1.4 (1.1–1.8)—?n/a
Grunkemeier, Payne, Jin et al. [84]7,955CABG (off- vs. on-pump)9905Stroke6032.0 (1.1–3.5)—2.7 (1.3–5.8)2.7 (1.3–5.6)2.60.84n/a
Magee, Jablonski, Stamou et al. [89]8,449CABG (on- vs. off-pump)1,98324Mortality238241.7 (1.2–2.5)1.9 (1.2–3.1)—1.8 (1.2–2.7)——79–83
Stamou, Jablonski, Pfister et al. [104]10,389CABG (off- vs. on-pump)2,32015Stroke22882.2 (1.4–3.2)1.8 (1.0–3.1)—1.6 (1.0–2.7)—0.7972
Umana, Miller, Mitchell [109]189Surgery in aortic dissection674Mortality1838“similar”—“nearly identical”“no important bearing”—?n/a
Carmeli, Eliopoulos, Mozaffari et al. [77]880Vancomycin resistant enterococci23312Mortality81?3.5———2.1 (1.1–4.4)0.80n/a
Soumerai, McLaughlin, Ross-Degnan et al. [185]2,659Thrombolytic therapy73510Mortality∼5001?———1.5 (1.1–2.1)?n/a
Patel, Spitznagel, Piccirillo [96]532Radiation in head & neck cancer135∼13Mortality251162.6—“approx. same as crude”2.2 (1.3–3.8)—0.76n/a
Shireman, Braman [101]1,506Prophylactic therapy for resp. sync. virus1376Hospitalizations?n/a?0.5 (0.2–1.1)———?100
Rahme, Pettitt, LeLorier [98]48,185Acetaminophen vs. NSAIDs21,20719GI events343?2.5———“around 1.0”—n/a
Chan, Bhatt, Chew et al. [78]5,052Statins after PCI1,33725Mortality166330.7 (0.5–1.0)—0.6 (0.4–1.0)0.7 (0.4–1.0)0.7 (0.4–1.0)0.83n/a
Mukamal, Maclure, Muller et al. [181]1,900Heavy tea consumption26636Mortality228290.7 (0.5–1.0)—0.6 (0.4–0.9)0.6 (0.4–0.8)—?n/a
Teufelsbauer, Prusa, Wolff et al. [107]454Stent vs. surgery for aortic dissection2068Mortality∼68120.9 (0.6–1.6)—2.0 (1.1–3.5)2.7 (1.4–5.0)2.7 (1.4–5.0)0.78n/a
Cole, Loughlin, Ajene et al. [80]4,678Zanamivir for influenza2,341?Influenza complications∼6,361?—1.2 (1.0–1.3)————99.8
Patel, Grayson, Jackson et al. [182]10,941Off vs. on pump CABG84312Mortality∼250130.5 (0.3–0.9)———0.6 (0.3–1.1)?n/a
Majahalme, Kim, Bruckman et al. [90]586US vs. Finish hospital27214Atrial fibrillation after CABG197n/a0.7 (0.5–1.0)—“no assoc” (P = .9)——0.91n/a
Chan, Quinn, Bhatt et al. [79]4,553ß-Blocker after PTCA2,05628Mortality∼230350.6 (0.5–0.8)—0.7 (0.5–0.9)0.6 (0.5–0.9)0.6 (0.5–0.9)0.68n/a
Elad, French, Shavelle et al. [81]7,358Early PTCA in MI1,6319Mortality435100.50.7—0.7 (0.5–0.9)0.7 (0.5–1.0)—90 (5 to 1 digits)
Pereira, Lauer, Bashir et al. [97]157Aortic valve replacement6820Mortality6170.7 (p<0.0001)———0.2 (0.1–0.4)0.8657 (freq)
Schomig, Mehilli, Holle et al. [184]4,520Statins after PTCA (w stent)9355Mortality∼14590.5 (0.3–0.7)———0.5 (0.4–0.7)?n/a
Winkelmayer, Glynn, Mittleman et al. [113]2,503Hemo- vs. peritoneal dialysis53754Mortality1,27725———1.6 (1.1–2.4)1.5 (0.9–2.5)0.82n/a
Iwashyna, Lamont [85]3,3575-FU in colon cancer1,52311Mortality∼1,770?—0.80.7 (0.6–0.8)“similar”—> 0.8398.3
Neugut, Fleischauer, Sundararajan et al. [94]1,807Adjuvant chemotherapy in rectal cancer67318Mortality685?0.9—0.8 (0.7–1.0)“equal to PS”——n/a
Turner, Laine, Cohen et al. [187]47,260Mental health care10,44018Dental care∼27,500241.1——1.5 (1.4–1.6)“minimal diff.”?n/a
Ferguson, Coombs, Peterson [82]99,942CABG (internal thoracic artery)23,70028Mortality4,557280.6 (0.6–0.7)0.7 (0.7–0.8)“similar”0.9 (0.8–0.9)——99
Sabik, Gillinov, Blackstone et al. [99]3,712Off vs. on pump CABG48117Mortality38?1.5 (0.7–3.5)0.5 (0.1–2.7)“entirely consist.”—“entirely consist.”0.7284
Umana, Lai, Mitchell et al. [108]189Surgery for aortic dissection674Mortality∼758“no sign. diff.”“no sign. diff.”—“no sign. diff.”——100 (freq)
Ferguson, Coombs, Peterson [83]629,877ß-Blocker in CABG285,96531Mortality∼19,200300.8 (0.8–0.8)1.0 (0.9–1.0)—0.9 (0.9–1.0)—0.680
Hu, Bronner, Willett et al. [178]84,688Frequent fish consumption??MI193160.80.6 (0.4–0.9)—0.7 (0.5–0.9)—??
Mehta, Pascual, Soroko et al. [92]552Diuretic use vs. nonuse22621Mortality294101.4 (1.0–1.9)?—1.7 (1.1–2.6)1.7 (1.1–2.6)?n/a
Newby, Kristinsson, Bhapkar et al. [95]12,365Early statins after ACS3,95230Mortality225?0.6 (0.4–0.8)—1.0 (0.7–1.5)0.9 (0.7–1.3)1.1 (0.8–1.6)0.77n/a
Tanasescu, Leitzmann, Rimm et al. [186]∼17,780Exercise (highest vs. lowest quintile)∼8,89029Coronary heart disease700290.60.7 (0.6–0.9)—0.7 (0.6–0.8)—??
Vincent, Baron, Reinhart et al. [110]3,534Red blood cells1,63811Mortality6144—1.3 (P = .02)—1.4 (1.0–1.8)—0.6439(?) (6 to 1 digits)
Stenestrand, Wallentin [105]21,912CABG & PTCA in acute MI2,55433Mortality1,835450.6—0.5 (0.4–0.7)0.4 (0.3–0.6)0.5 (0.4–0.6)—n/a
Ayanian, Landrum, Guadagnoli et al. [76]35,520Post MI care10,86436Mortality4,984n/a0.6 (0.5–0.6)0.8 (0.7–0.8)————93.9
Schneeweiss, Walker, Glynn et al. [183]37,362Reference drug pricing5,35323Emergency hospitalization>30,0005——1.3—1.20.62n/a
Kimmel, Berlin, Kinman et al. [180]20,364NSAID (parenteral Ketorolac)10,14514MI6350.4 (0.2–0.7)—0.4 (0.2–0.7)0.4 (0.2–0.7)—?n/a

Studies were identified from PubMed search and Science Citation Index. Quotations indicate qualitative statement only, data not presented.

aNumbers presented are odds ratios, relative risks, or incidence rate ratios with their corresponding 95% confidence intervals (in brackets).

bArea under the receiver operating characteristics (ROC) curve (c-statistic) from logistic regression model used to estimate the PS.

cPercent of all exposed subjects for whom an unexposed subject could be matched on the PS; exposed subjects without match are discarded from analysis; n/a = not applicable (no matching).

Table E3. Substantive studies in 2001 using propensity score methods (PS)
ExposureOutcome Multivariable outcome modelinga
AuthorN nVariables in model, no. nVariables in model, no.CrudePS matchingPS adjustmentWithout PSPS and covariatesAUCb% matchedc
Rathore, Gersh, Weinfurt et al. [135]1,954Reperfusion in acute MI171?Mortality50361.0 (0.8–1.3)———1.1 (0.8–1.4)?n/a
Foody, Cole, Blackstone et al. [121]4,353Smoking2,166?Mortality356271.4 (1.1–1.7)2.1 (1.5–2.8)—2.4 (1.9–3.1)3.0 (2.2–4.1)0.7660 (freq)
Mast, Gersing, Anstrom et al. [128]5,437SSRI2927Heart valve regurgitation64270.8 (0.6–1.1)—0.9——0.82n/a
Peterson, Topol, Roe et al. [133]7,065Heparin in acute CAD88233Mortality or MI966111.2 (1.0–1.4)———1.2 (0.9–1.6)0.69n/a
Shlipak, Browner, Noguchi et al. [198]9,181ACE inhibitor & ß–blocker3,309?Mortality2,842270.6—“similar”0.7 (0.6–0.8)—?n/a
Sernyak, Desai, Stolar et al. [136]45,917Clozapine1,41518Completed suicide33n/a?0.7 (0.3–1.5)———?100 (freq)
Ascione, Nason, Al-Ruzzeh et al. [189]253Off- vs. On-pump CABG51?Acute renal failure3550.3 (0.1–1.1)——0.4 (0.2–1.0)“if anything, greater”?n/a
Carmeli, Castro, Eliopoulos et al. [116]3,753Ceftriaxone vs. ampicillin–sulbactam1,30813Nosocomial infection17251.6 (P = .02)———1.6 (1.1–2.1)0.80n/a
Wiles, Lunt, Barrett et al. [141]384Early DMARD in RA14724Disability?22.1 (1.1–3.7)—0.9 (0.4–1.7)—0.7 (0.3–1.4)?n/a
Mitra, Schnabel, Neugut et al. [129]3,080Intensive surveillance program5816Early stage BRCA522n/a2.2 (1.4–3.6)—1.5 (0.9–2.5)——?n/a
O'Day, Boasberg, Kristedja et al. [194]80High vs. low dose Tamoxifen358Melanoma response42n/a0.8 (0.3–1.9)—0.8 (0.3–2.0)——?n/a
Beuth, Ost, Pakdaman et al. [115]649Oral enzyme in cancer2398Symptoms (e.g. pain)61n/a1.2—0.7 (0.3–1.9)——?n/a
Ellis, Brener, Lincoff et al. [119]6,200ß-Blocker before PTCA2,9268Elevated CK815?1.0 (0.8–1.1)———“no diff” (p = 1.0)0.62n/a
Margolis, Kantor, Santanna et al. [127]26,599Platelet releasate in ulcers6,25214Wound healing11,468n/a1.4 (1.4–1.5)—1.4 (1.3–1.4)——0.90n/a
Osswald, Blackstone, Tochtermann et al. [132]859Complete vs. incomplete revasc CABG133∼30Mortality137?1.9 (1.2–2.9)—“no residual bias”1.8 (1.1–2.8)“no residual bias”0.81n/a
Ioannidis, Galanos, Katritsis et al. [125]1,697Bilateral vs. single internal thor. CABG83017Mortality46?0.7 (P = .3)—1.1 (P = .8)0.9 (P = .9)—?n/a
Suero, Marso, Jones et al. [139]25,620PTCA in total occlusion2,00727Mortality∼2,862n/a?1.0 (P = .9)———?100 (indiv)
Normand, Landrum, Guadagnoli et al. [131]37,788Angiography17,304102Mortality13,513n/a?0.5 (0.5–0.5)———0.8457 (stratified)
Earle, Tsai, Gelber et al. [118]6,232Chemo in lung cancer2,01212Mortality5,0324(?)?—0.80.8 (0.8–0.9)—?n/a
Cen, Glower, Landolfo et al. [190]1,139Biologic vs. mechanical valve4957Mortality∼45010?———0.8 (0.6–1.1)?n/a
Gillinov, Wierup, Blackstone et al. [122]482Valve repair vs. replacement85∼70Mortality548?———0.1 (0.1–0.2)0.92n/a
Mehta, Bruckman, Das et al. [192]473Left ventricular mass11521Mortality25715 (5.4–41)——23 (7.2–74)38 (9.3–154)?n/a
Gum, Thamilarasan, Watanabe et al. [123]6,174Aspirin2,31034Mortality276271.1 (0.9–1.4)0.5 (0.4–0.7)—0.7 (0.5–0.9)0.6 (0.4–0.8)0.8358 (5 to 1)
Mukamal, Maclure, Muller et al. [130]1,217Daily alcohol consumption32128Mortality12670.4 (0.3–0.6)—0.7 (0.4–1.0)—0.7 (0.5–1.1)0.81n/a
Polanczyk, Rohde, Goldman et al. [134]4,059Right heart cath22120Cardiac events171n/a4.9 (3.3–7.4)1.6 (0.9–2.8)———0.8597 (indiv)
Stenestrand, Wallentin L [138]19,599Early statins after acute MI5,52842Mortality1,526430.4 (0.3–0.5)——0.7 (0.6–0.9)0.8 (0.6–0.9)0.81n/a
Tu, Austin, Chan BT [199]52,568Patient volume25,07411Mortality7,397251.6 (1.5–1.6)—1.2 (p<0.001)∼1.5—0.56n/a
Welch, Zalenski, Frederick et al. [140]253,634Normal ECG in acute MI30,759?Mortality∼27,384210.5 (0.4–0.5)0.6 (0.5–0.6)0.6 (0.6–0.6)—0.6 (0.6–0.6)?95
Aronow, Topol, Roe et al. [114]20,809Lipid lowering therapy3,65335Mortality668>350.5 (0.4–0.6)—0.7 (0.5–0.9)0.6 (0.5–0.8)0.7 (0.5–1.0)0.87n/a
Keating, Weeks, Landrum et al. [126]792Oncologist consultation17431Breast conserving surgery225n/a?—1.2 (0.8–1.8)——?n/a
Dammann, Allred, Kuban et al. [117]799Postnatal hypocarbia18214White matter echo-lucency48?1.9 (1.1–3.6)———1.7 (0.8–3.2)?n/a
Kachele, Kordy, Richard et al. [191]399Duration of psychotherapy19720Therapy success145n/a0.9 (0.6–1.4)—0.9 (ns)——?n/a
Schnuelle, Berger, de Boer et al. [197]1,742Catecholamines in donor1275Renal graft survival in recipient65211?——0.9 (0.7–1.0)0.9 (0.8–1.0)?n/a

Studies were identified from PubMed search and Science Citation Index. Quotations indicate qualitative statement only, data not presented.

aNumbers presented are odds ratios, relative risks, or incidence rate ratios with their corresponding 95% confidence intervals (in brackets).

bArea under the receiver operating characteristics (ROC) curve (c-statistic) from logistic regression model used to estimate the PS.

cPercent of all exposed subjects for whom an unexposed subject could be matched on the PS; exposed subjects without match are discarded from analysis; n/a = not applicable (no matching).

Table E4. Substantive studies in 2000 using propensity score methods (PS)
ExposureOutcome Multivariable outcome modelinga
AuthorN nVariables in model, no. nVariables in model, no.CrudePS matchingPS adjustmentWithout PSPS and covariatesAUCb% matchedc
Auerbach, Hamel, Davis et al. [200]1,298Care by cardiologists vs. generalists55525Mortality492211.0 (0.8–1.3)— 0.8 (0.7–1.0)?n/a
Murdoch, Cohen, Bellamy [144]4,182Pulmonary artery catheterization1,8497Mortality1,08725.2 (4.4–6.0)—1.1 (0.9–1.3)——0.88n/a
Hannan, Racz, Arani et al. [201]19,792Stent placement vs. balloon angioplasty7,197?Mortality35210.5 (.3–.7)————?n/a
Weiss, Gruver, Kaul et al. [146]1,033Referral to cardiologists vs. noncardiologists42816Yield of 2-dimensional echocardiogram41422.4 (1.8–3.0)—2.0 (1.5–2.8) 0.69n/a
Potosky, Legler, Albertsen et al. [145]1,591Prostatectomy1,15626Urinary function10572.9 (1.6–5.1)———3.2 (1.7–6.2)?n/a
Holman, Li, Kiefe et al. [143]7,581CABG5928Mortality 11?1.0 (.7–1.5)——0.9 (0.7–1.1)0.7992.9
Yau, El-Ghoneimi, Armstrong et al. [147]573Mitral valve repair vs. replacement1425Mortality31?0.7 (0.5–1.1)————0.81n/a
Hagan, Thiede [142]1,582Needle exchange use3468Syringe sharing78420.6 (0.5–.8)——0.7 (0.5–0.9)0.7 (0.5–0.9)?n/a
Petersen, Normand, Daley et al. [202]31,735Outcome of MI in VHA vs. Medicare patients2,48629Mortality10,090291 (2.5–2.9)?——0.94 (0.84–1.05)0.8891.1

Studies were identified from PubMed search and Science Citation Index. Quotations indicate qualitative statement only, data not presented.

aNumbers presented are odds ratios, relative risks, or incidence rate ratios with their corresponding 95% confidence intervals (in brackets).

bArea under the receiver operating characteristics (ROC) curve (c-statistic) from logistic regression model used to estimate the PS.

cPercent of all exposed subjects for whom an unexposed subject could be matched on the PS; exposed subjects without match are discarded from analysis; n/a = not applicable (no matching).

Table E5. Substantive studies in 1999 using propensity score methods (PS)
ExposureOutcome Multivariable outcome modelinga
AuthorN nVariables in model, no. nVariables in model, no.CrudePS matchingPS adjustmentWithout PSPS and covariatesAUCb% matchedc
Nakamura, Moss, Brown et al. [149]1,779Long acting nitrates97830Mortality125131.7 (1.2–2.6)—1.5 (P = .07)1.6 (1.1–2.4)—?n/a
Shwartz, Saitz, Mulvey et al. [203]8,011Acupuncture vs. residential detoxification1,104?Readmission2,68660.4 (0.3–0.5)——0.7 (0.5–1.0)0.6 (0.4–0.9)0.9067
Lytle, Blackstone, Loop et al. [148]10,124Bilateral thoracic artery CABG2,001>36Mortality781>151.4 (1.2–1.5)1.7 (1.5–1.9)——1.3 (1.1–1.5)0.9197
Shepardson, Youngner, Speroff et al. [151]13,337Do not resuscitate order2,8989Mortality∼1,320744 (37–52)———34 (27–42)?n/a

Studies were identified from PubMed search and Science Citation Index. Quotations indicate qualitative statement only, data not presented.

aNumbers presented are odds ratios, relative risks, or incidence rate ratios with their corresponding 95% confidence intervals (in brackets).

bArea under the receiver operating characteristics (ROC) curve (c-statistic) from logistic regression model used to estimate the PS.

cPercent of all exposed subjects for whom an unexposed subject could be matched on the PS; exposed subjects without match are discarded from analysis; n/a = not applicable (no matching).

Table E6. Substantive studies in 1998 using propensity score methods (PS)
ExposureOutcome Multivariable outcome modelinga
AuthorN nVariables in model, no. nVariables in model, no.CrudePS matchingPS adjustmentwithout PSPS and covariatesAUCb% matchedc
Regueiro, Hamel, Davis et al. [156]866Pulmonologist vs. generalist35433Mortality116181.6 (1.1–2.3)——1.6 (1.0–2.6)1.6 (1.0–2.5)0.78n/a
Fraenkel, Zhang, Chaisson et al. [205]442HRT635Raynaud49n/a2.6 (1.2–5.3)—2.5 (1.2–5.3)——?n/a
Intrator, Berg [155]324Home health care1296Rehospitalization97n/a0.7 (0.3–1.6)—0.7 (0.4–1.0)——?n/a
Barosi, Ambrosetti, Centra et al. [154]549Splenectomy in myelofibrosis8721Blast transformation7812?—P = .0082.6 (1.4–5.0)—?n/a
Smith, Reiber, Psaty et al. [157]247Diltiazem vs. β-blocker59?Mortality54n/a1.6 (0.8–3.3)—1.1 (0.5–2.4)——?n/a
Cooper, Chak, Connors [204]3,801Endoscopy in upper GI bleed1,56136Mortality109?1.0 (0.7–1.5)—1.2 (0.8–1.7)1.1 (0.7–1.7)—?n/a
Barker, Chang, Gutin et al. [153]173Resection of recurrent Glioblastoma322Mortality∼1702“No sign. diff”——0.7 (0.4–1.0)0.6 (0.4–1.0)?n/a

Studies were identified from PubMed search and Science Citation Index. Quotations indicate qualitative statement only, data not presented.

aNumbers presented are odds ratios, relative risks, or incidence rate ratios with their corresponding 95% confidence intervals (in brackets).

17Area under the receiver operating characteristics (ROC) curve (c-statistic) from logistic regression model used to estimate the PS.

cPercent of all exposed subjects for whom an unexposed subject could be matched on the PS; exposed subjects without match are discarded from analysis; n/a = not applicable (no matching).

Table E7. Substantive studies in 1997 and before, using propensity score methods (PS)
ExposureOutcome Multivariable outcome modelinga
AuthorN nVariables in model, no. nVariables in model, no.CrudePS matchingPS adjustmentWithout PSPS and covariatesAUCb% matchedc
Myers, Gersh, Fischer et al. [161]856Medical vs. early surgical therapy41352MI12761.9 (1.3–2.8)————?n/a
Fiebach, Cook, Lee et al. [209]467Admission to step down vs. coronary care unit5816Mortality47?0.3 (0.1–1.2)—0.9 (.5–1.8)——?n/a
Connors, Speroff, Dawson et al. [158]5,735Right heart catheterization2,18433Mortality2,817101.4 (1.2–1.5)1.2 (1.0–1.5)——1.2 (1.1–1.3)0.8392.3
Lieberman, Lang, Cohen et al. [159]1,733Epidural vs. nonepidural74215Cesarean delivery19824.3 (3.1–6.0)———3.7 (2.4–5.7)?n/a
King, Barnhart, Kosinski et al. [207]366Randomized vs. registry patients16822Mortality3?0.6 (0.1–6.7)————?n/a
Soumerai, McLaughlin, Spiegelman et al. [208]3,737β-Blocker therapy1,38214Mortality?2?—0.6 (0.5–0.7)0.6 (0.5–0.7)—?n/a

Studies were identified from PubMed search and Science Citation Index. Quotations indicate qualitative statement only, data not presented.

aNumbers presented are odds ratios, relative risks, or incidence rate ratios with their corresponding 95% confidence intervals (in brackets).

bArea under the receiver operating characteristics (ROC) curve (c-statistic) from logistic regression model used to estimate the PS.

cPercent of all exposed subjects for whom an unexposed subject could be matched on the PS; exposed subjects without match are discarded from analysis; n/a = not applicable (no matching).

Back to Article Outline

References 

  1. Concato J, Shah N, Horwitz RI. Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med. 2000;342:1887–1892
  2. Writing group for the Women's Health Initiative investigators . Risk and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. JAMA. 2002;288:321–333
  3. Walker AM. Confounding by indication. Epidemiology. 1996;7:335–336
  4. Miettinen OS. Stratification by a multivariate confounder score. Am J Epidemiol. 1976;104:609–620
  5. Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70:41–55
  6. Rosenbaum PR, Rubin DB. The bias due to incomplete matching. Biometrics. 1985;41:103–116
  7. Gu XS, Rosenbaum PR. Comparison of multivariate matching methods: structures, distances, and algorithms. J Comput Graph Stat. 1993;2:405–420
  8. Parsons LS. Reducing bias in a propensity score matched-pair sample using greedy matching techniques. In: Proceedings of the 26th Annual SAS Users Group International Conference. April 22–25, 2001. Long Beach California. Cary, NC: SAS Institute; 2001:214–26 [Available at: http://www2.sas.com/proceedings/sugi26/p214-26.pdf. Accessed January 2005].
  9. Rubin DB, Thomas N. Matching using estimated propensity scores: relating theory to practice. Biometrics. 1996;52:249–264
  10. Cochran WG. The effectiveness of adjustment by subclassification in removing bias in observational studies. Biometrics. 1968;24:205–213
  11. Brenner H, Blettner M. Controlling for continuous confounders in epidemiologic research. Epidemiology. 1997;8:429–434
  12. Rubin DB, Thomas N. Combining propensity score matching with additional adjustment for prognostic covariates. J Am Stat Assoc. 2000;95:573–585
  13. D'Agostino RB. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med. 1998;17:2265–2281
  14. Brookhart MA, van der Laan MJ. A semiparametric model selection criterion with applications to the marginal structural model. Comput Stat Data Anal 2006;50:475–98.
  15. Joffe MM, Rosenbaum PR. Invited commentary: propensity scores. Am J Epidemiol. 1999;150:327–333
  16. Rosenbaum PR, Rubin DB. Reducing bias in observational studies using sub-classification on the propensity score. J Am Stat Assoc. 1984;79:516–524
  17. Rubin DB. Estimating causal effects from large data sets using propensity scores. Ann Intern Med. 1997;127:757–763
  18. Abbott KC, Trespalacios FC, Agodoa LY. Arteriovenous fistula use and heart disease in long-term elderly hemodialysis patients: analysis of United States Renal Data System Dialysis Morbidity and Mortality Wave II. J Nephrol. 2003;16:822–830
  19. Abramov D, Tamariz M, Fremes S, et al. Impact of preoperative renal dysfunction on cardiac surgery results. Asian Cardiovasc Thorac Ann. 2003;11:42–47
  20. Abramov D, Tamariz M, Serrick CI, et al. The influence of cardiopulmonary bypass flow characteristics on the clinical outcome of 1820 coronary bypass patients. Can J Cardiol. 2003;19:237–243
  21. Allen-Ramey FC, Duong PT, Goodman DC, et al. Treatment effectiveness of inhaled corticosteroids and leukotriene modifiers for patients with asthma: an analysis from managed care data. Allergy Asthma Proc. 2003;24:43–51
  22. Aronow HD, Novaro GM, Lauer MS, et al. In-hospital initiation of lipid-lowering therapy after coronary intervention as a predictor of long-term utilization: a propensity analysis. Arch Intern Med. 2003;163:2576–2582
  23. Ascione R, Narayan P, Rogers CA, Lim KH, Capoun R, Angelini GD. Early and midterm clinical outcome in patients with severe left ventricular dysfunction undergoing coronary artery surgery. Ann Thorac Surg. 2003;76:793–799
  24. Banbury MK, White JA, Blackstone EH, Cosgrove DM. Vacuum-assisted venous return reduces blood usage. J Thorac Cardiovasc Surg. 2003;126:680–687
  25. Beddhu S, Samore MH, Roberts MS, et al. Impact of timing of initiation of dialysis on mortality. J Am Soc Nephrol. 2003;14:2305–2312
  26. Bhatt DL, Chew DP, Lincoff AM, et al. Effect of revascularization on mortality associated with an elevated white blood cell count in acute coronary syndromes. Am J Cardiol. 2003;92:136–140
  27. Boening A, Friedrich C, Hedderich J, Schoettler J, Fraund S, Cremer JT. Early and medium-term results after on-pump and off-pump coronary artery surgery: a propensity score analysis. Ann Thorac Surg. 2003;76:2000–2006
  28. Brener SJ, Ellis SG, Schneider J, Apperson-Hansen C, Topol EJ. Abciximab-facilitated percutaneous coronary intervention and long-term survival–a prospective single-center registry. Eur Heart J. 2003;24:630–638
  29. Bukhari MA, Wiles NJ, Lunt M, et al. Influence of disease-modifying therapy on radiographic outcome in inflammatory polyarthritis at five years: results from a large observational inception study. Arthritis Rheum. 2003;48:46–53
  30. Calafiore AM, Di Mauro M, Canosa C, et al. Myocardial revascularization with and without cardiopulmonary bypass in multivessel disease: impact of strategy on midterm outcome. Ann Thorac Surg. 2003;76:32–36
  31. Calafiore AM, Di Mauro M, Canosa C, Di Giammarco G, Iaco AL, Contini M. Early and late outcome of myocardial revascularization with and without cardiopulmonary bypass in high risk patients (EuroSCORE > or = 6). Eur J Cardiothorac Surg. 2003;23:360–367
  32. Chan AW, Bhatt DL, Chew DP, et al. Relation of inflammation and benefit of statins after percutaneous coronary interventions. Circulation. 2003;107:1750–1756
  33. Chan AW, Moliterno DJ, Berger PB, et al. Triple antiplatelet therapy during percutaneous coronary intervention is associated with improved outcomes including one-year survival: results from the Do Tirofiban and ReoProGive Similar Efficacy Outcome Trial (TARGET). J Am Coll Cardiol. 2003;42:1188–1195
  34. Cho L, Bhatt DL, Marso SP, et al. An invasive strategy is associated with decreased mortality in patients with unstable angina and non-ST-elevation myocardial infarction: GUSTO IIb trial. Am J Med. 2003;114:106–111
  35. Christakis NA, Iwashyna TJ. The health impact of health care on families: a matched cohort study of hospice use by decedents and mortality outcomes in surviving, widowed spouses. Soc Sci Med. 2003;57:465–475
  36. Dendukuri N, Normand SL, McNeil BJ. Impact of cardiac service availability on case-selection for angiography and survival associated with angiography. Health Serv Res. 2003;38:21–40
  37. Ezri T, Weisenberg M, Khazin V, et al. Difficult laryngoscopy: incidence and predictors in patients undergoing coronary artery bypass surgery versus general surgery patients. J Cardiothorac Vasc Anesth. 2003;17:321–324
  38. Flameng WJ, Herijgers P, Dewilde S, Lesaffre E. Continuous retrograde blood cardioplegia is associated with lower hospital mortality after heart valve surgery. J Thorac Cardiovasc Surg. 2003;125:121–125
  39. Frolkis JP, Pothier CE, Blackstone EH, Lauer MS. Frequent ventricular ectopy after exercise as a predictor of death. N Engl J Med. 2003;348:781–790
  40. Gillinov AM, Faber C, Houghtaling PL, et al. Repair versus replacement for degenerative mitral valve disease with coexisting ischemic heart disease. J Thorac Cardiovasc Surg. 2003;125:1350–1362
  41. Girou E, Brun-Buisson C, Taille S, Lemaire F, Brochard L. Secular trends in nosocomial infections and mortality associated with noninvasive ventilation in patients with exacerbation of COPD and pulmonary edema. JAMA. 2003;290:2985–2991
  42. Grzybowski M, Clements EA, Parsons L, et al. Mortality benefit of immediate revascularization of acute ST-segment elevation myocardial infarction in patients with contraindications to thrombolytic therapy: a propensity analysis. JAMA. 2003;290:1891–1898
  43. Hachamovitch R, Hayes SW, Friedman JD, Cohen I, Berman DS. Comparison of the short-term survival benefit associated with revascularization compared with medical therapy in patients with no prior coronary artery disease undergoing stress myocardial perfusion single photon emission computed tomography. Circulation. 2003;107:2900–2907
  44. Hall JA, Summers KH, Obenchain RL. Cost and utilization comparisons among propensity score-matched insulin lispro and regular insulin users. J Manag Care Pharm. 2003;9:263–268
  45. Hamamoto M, Bando K, Kobayashi J, et al. Durability and outcome of aortic valve replacement with mitral valve repair versus double valve replacement. Ann Thorac Surg. 2003;75:28–33
  46. Harbarth S, Garbino J, Pugin J, Romand JA, Lew D, Pittet D. Inappropriate initial antimicrobial therapy and its effect on survival in a clinical trial of immunomodulating therapy for severe sepsis. Am J Med. 2003;115:529–535
  47. Heuschmann PU, Berger K, Misselwitz B, et al. Frequency of thrombolytic therapy in patients with acute ischemic stroke and the risk of in-hospital mortality: the German Stroke Registers Study Group. Stroke. 2003;34:1106–1113
  48. Hughes MG, Evans HL, Lightfoot L, et al. Does prior transfusion worsen outcomes from infection in surgical patients?. Surg Infect (Larchmt). 2003;4:335–343
  49. Ishani A, Ibrahim HN, Gilbertson D, Collins AJ. The impact of residual renal function on graft and patient survival rates in recipients of preemptive renal transplants. Am J Kidney Dis. 2003;42:1275–1282
  50. Karthik S, Musleh G, Grayson AD, et al. Effect of avoiding cardiopulmonary bypass in non-elective coronary artery bypass surgery: a propensity score analysis. Eur J Cardiothorac Surg. 2003;24:66–71
  51. Katzan IL, Cebul RD, Husak SH, Dawson NV, Baker DW. The effect of pneumonia on mortality among patients hospitalized for acute stroke. Neurology. 2003;60:620–625
  52. Koch CG, Khandwala F, Nussmeier N, Blackstone EH. Gender and outcomes after coronary artery bypass grafting: a propensity-matched comparison. J Thorac Cardiovasc Surg. 2003;126:2032–2043
  53. Krzyzanowska MK, Weeks JC, Earle CC. Treatment of locally advanced pancreatic cancer in the real world: population-based practices and effectiveness. J Clin Oncol. 2003;21:3409–3414
  54. Kurlansky PA, Williams DB, Traad EA, et al. Arterial grafting results in reduced operative mortality and enhanced long-term quality of life in octogenarians. Ann Thorac Surg. 2003;76:418–426
  55. Landesberg G, Mosseri M, Wolf YG, et al. Preoperative thallium scanning, selective coronary revascularization, and long-term survival after major vascular surgery. Circulation. 2003;108:177–183
  56. Leon AC, Solomon DA, Mueller TI, et al. A 20-year longitudinal observational study of somatic antidepressant treatment effectiveness. Am J Psychiatry. 2003;160:727–733
  57. Macaubas C, de Klerk NH, Holt BJ, et al. Association between antenatal cytokine production and the development of atopy and asthma at age 6 years. Lancet. 2003;362:1192–1197
  58. Magee MJ, Coombs LP, Peterson ED, Mack MJ. Patient selection and current practice strategy for off-pump coronary artery bypass surgery. Circulation. 2003;108S1:II9–14
  59. Martin S, Kolb H, Beuth J, van Leendert R, Schneider B, Scherbaum WA. Change in patients' body weight after 12 months of treatment with glimepiride or glibenclamide in Type 2 diabetes: a multicentre retrospective cohort study. Diabetologia. 2003;46:1611–1617
  60. Moss RR, Humphries KH, Gao M, et al. Outcome of mitral valve repair or replacement: a comparison by propensity score analysis. Circulation. 2003;108S1:II90–II97
  61. Murthy SC, Law S, Whooley BP, Alexandrou A, Chu KM, Wong J. Atrial fibrillation after esophagectomy is a marker for postoperative morbidity and mortality. J Thorac Cardiovasc Surg. 2003;126:1162–1167
  62. Peterson ED, Pollack CV, Roe MT, et al. Early use of glycoprotein IIb/IIIa inhibitors in non-ST-elevation acute myocardial infarction: observations from the National Registry of Myocardial Infarction 4. J Am Coll Cardiol. 2003;42:45–53
  63. Rice TW, Adelstein DJ, Chidel MA, et al. Benefit of postoperative adjuvant chemoradiotherapy in locoregionally advanced esophageal carcinoma. J Thorac Cardiovasc Surg. 2003;126:1590–1596
  64. Rubins HB, Nelson DB, Noorbaloochi S, Nugent S. Effectiveness of lipid-lowering medications in outpatients with coronary heart disease in the Department of Veterans Affairs System. Am J Cardiol. 2003;92:1177–1182
  65. Schmitz C, Weinreich S, White J, et al. Can particulate extraction from the ascending aorta reduce neurologic injury in cardiac surgery?. J Thorac Cardiovasc Surg. 2003;126:1829–1838
  66. Schumacher K, Schneider B, Reich G, et al. Influence of postoperative complementary treatment with lectin-standardized mistletoe extract on breast cancer patients: a controlled epidemiological multicentric retrolective cohort study. Anticancer Res. 2003;23:5081–5087
  67. Schwarz RE, Smith DD, Keny H, et al. Impact of intraoperative radiation on postoperative and disease-specific outcome after pancreatoduodenectomy for adenocarcinoma: a propensity score analysis. Am J Clin Oncol. 2003;26:16–21
  68. Seeger JD, Walker AM, Williams PL, Saperia GM, Sacks FM. A propensity score-matched cohort study of the effect of statins, mainly fluvastatin, on the occurrence of acute myocardial infarction. Am J Cardiol. 2003;92:1447–1451
  69. Srinivasan AK, Grayson AD, Pullan DM, Fabri BM, Dihmis WC. Effect of preoperative aspirin use in off-pump coronary artery bypass operations. Ann Thorac Surg. 2003;76:41–45
  70. Stamou SC, Kapetanakis EI, Lowery R, Jablonski KA, Frankel TL, Corso PJ. Allogeneic blood transfusion requirements after minimally invasive versus conventional aortic valve replacement: a risk-adjusted analysis. Ann Thorac Surg. 2003;76:1101–1106
  71. Trichon BH, Glower DD, Shaw LK, et al. Survival after coronary revascularization, with and without mitral valve surgery, in patients with ischemic mitral regurgitation. Circulation. 2003;108S1:II103–II110
  72. Vakili BA, Kaplan RC, Slater JN, et al. A propensity analysis of the impact of platelet glycoprotein IIb/IIIa inhibitor therapy on in-hospital outcomes after percutaneous coronary intervention. Am J Cardiol. 2003;91:946–950
  73. Vikram HR, Buenconsejo J, Hasbun R, Quagliarello VJ. Impact of valve surgery on 6-month mortality in adults with complicated, left-sided native valve endocarditis: a propensity analysis. JAMA. 2003;290:3207–3214
  74. Winkelmayer WC, Owen WF, Levin R, Avorn J. A propensity analysis of late versus early nephrologist referral and mortality on dialysis. J Am Soc Nephrol. 2003;14:486–492
  75. Young-Xu Y, Chan KA, Liao JK, Ravid S, Blatt CM. Long-term statin use and psychological well-being. J Am Coll Cardiol. 2003;42:690–697
  76. Ayanian JZ, Landrum MB, Guadagnoli E, Gaccione P. Specialty of ambulatory care physicians and mortality among elderly patients after myocardial infarction. N Engl J Med. 2002;347:1678–1686
  77. Carmeli Y, Eliopoulos G, Mozaffari E, Samore M. Health and economic outcomes of vancomycin-resistant enterococci. Arch Intern Med. 2002;162:2223–2228
  78. Chan AW, Bhatt DL, Chew DP, et al. Early and sustained survival benefit associated with statin therapy at the time of percutaneous coronary intervention. Circulation. 2002;105:691–696
  79. Chan AW, Quinn MJ, Bhatt DL, et al. Mortality benefit of beta-blockade after successful elective percutaneous coronary intervention. J Am Coll Cardiol. 2002;40:669–675
  80. Cole JA, Loughlin JE, Ajene AN, Rosenberg DM, Cook SE, Walker AM. The effect of zanamivir treatment on influenza complications: a retrospective cohort study. Clin Ther. 2002;24:1824–1839
  81. Elad Y, French WJ, Shavelle DM, Parsons LS, Sada MJ, Every NR. Primary angioplasty and selection bias inpatients presenting late (>12 h) after onset of chest pain and ST elevation myocardial infarction. J Am Coll Cardiol. 2002;39:826–833
  82. Ferguson TB, Coombs LP, Peterson ED. Internal thoracic artery grafting in the elderly patient undergoing coronary artery bypass grafting: room for process improvement?. J Thorac Cardiovasc Surg. 2002;123:869–880
  83. Ferguson TB, Coombs LP, Peterson ED. Preoperative beta-blocker use and mortality and morbidity following CABG surgery in North America. JAMA. 2002;287:2221–2227
  84. Grunkemeier GL, Payne N, Jin R, Handy JR. Propensity score analysis of stroke after off-pump coronary artery bypass grafting. Ann Thorac Surg. 2002;74:301–305
  85. Iwashyna TJ, Lamont EB. Effectiveness of adjuvant fluorouracil in clinical practice: a population-based cohort study of elderly patients with stage III colon cancer. J Clin Oncol. 2002;20:3992–3998
  86. Jenkins NP, Keevil BG, Hutchinson IV, Brooks NH. Beta-blockers are associated with lower C-reactive protein concentrations in patients with coronary artery disease. Am J Med. 2002;112:269–274
  87. Kilborn MJ, Rathore SS, Gersh BJ, Oetgen WJ, Solomon AJ. Amiodarone and mortality among elderly patients with acute myocardial infarction with atrial fibrillation. Am Heart J. 2002;144:1095–1101
  88. Luedtke R, Weber U, Fischer I, Friese KH, Moeller H. An example on the value of non-randomisation in clinical trials in complementary medicine. Forsch Komplementarmed Klass Naturheilkd. 2002;9:105–109
  89. Magee MJ, Jablonski KA, Stamou SC, et al. Elimination of cardiopulmonary bypass improves early survival for multivessel coronary artery bypass patients. Ann Thorac Surg. 2002;73:1196–1202
  90. Majahalme S, Kim MH, Bruckman D, Tarkka M, Eagle KA. Atrial fibrillation after coronary surgery: comparison between different health care systems. Int J Cardiol. 2002;82:209–218
  91. Mehta RL, McDonald B, Gabbai F, et al. Nephrology consultation in acute renal failure: does timing matter?. Am J Med. 2002;113:456–461
  92. Mehta RL, Pascual MT, Soroko S, Chertow GM. Diuretics, mortality, and nonrecovery of renal function in acute renal failure. JAMA. 2002;288:2547–2553
  93. Mukherjee D, Mahaffey KW, Moliterno DJ, et al. Promise of combined low-molecular-weight heparin and platelet glycoprotein IIb/IIIa inhibition: results from Platelet IIb/IIIa Antagonist for the Reduction of Acute coronary syndrome events in a Global Organization Network B (PARAGON B). Am Heart J. 2002;144:995–1002
  94. Neugut AI, Fleischauer AT, Sundararajan V, et al. Use of adjuvant chemotherapy and radiation therapy for rectal cancer among the elderly: a population-based study. J Clin Oncol. 2002;20:2643–2650
  95. Newby LK, Kristinsson A, Bhapkar MV, et al. Early statin initiation and outcomes in patients with acute coronary syndromes. JAMA. 2002;287:3087–3095
  96. Patel U, Spitznagel E, Piccirillo J. Multivariate analyses to assess treatment effectiveness in advanced head and neck cancer. Arch Otolaryngol Head Neck Surg. 2002;128:497–503
  97. Pereira JJ, Lauer MS, Bashir M, et al. Survival after aortic valve replacement for severe aortic stenosis with low transvalvular gradients and severe left ventricular dysfunction. J Am Coll Cardiol. 2002;39:1356–1363
  98. Rahme E, Pettitt D, LeLorier J. Determinants and sequelae associated with utilization of acetaminophen versus traditional nonsteroidal antiinflammatory drugs in an elderly population. Arthritis Rheum. 2002;46:3046–3054
  99. Sabik JF, Gillinov AM, Blackstone EH, et al. Does off-pump coronary surgery reduce morbidity and mortality?. J Thorac Cardiovasc Surg. 2002;124:698–707
  100. Shavelle DM, Parsons L, Sada MJ, French WJ, Every NR. Is there a benefit to early angiography in patients with ST-segment depression myocardial infarction? An observational study. Am Heart J. 2002;143:488–496
  101. Shireman TI, Braman KS. Impact and cost-effectiveness of respiratory syncytial virus prophylaxis for Kansas Medicaid's high-risk children. Arch Pediatr Adolesc Med. 2002;156:1251–1255
  102. Shishehbor MH, Baker DW, Blackstone EH, Lauer MS. Association of educational status with heart rate recovery: a population-based propensity analysis. Am J Med. 2002;113:643–649
  103. Sin DD, McAlister FA. The effects of beta-blockers on morbidity and mortality in a population-based cohort of 11,942 elderly patients with heart failure. Am J Med. 2002;113:650–656
  104. Stamou SC, Jablonski KA, Pfister AJ, et al. Stroke after conventional versus minimally invasive coronary artery bypass. Ann Thorac Surg. 2002;74:394–399
  105. Stenestrand U, Wallentin L. Early revascularisation and 1-year survival in 14-day survivors of acute myocardial infarction: a prospective cohort study. Lancet. 2002;359:1805–1811
  106. Sundararajan V, Mitra N, Jacobson JS, Grann VR, Heitjan DF, Neugut AI. Survival associated with 5-fluorouracil-based adjuvant chemotherapy among elderly patients with node-positive colon cancer. Ann Intern Med. 2002;136:349–357
  107. Teufelsbauer H, Prusa AM, Wolff K, et al. Endovascular stent grafting versus open surgical operation in patients with infrarenal aortic aneurysms: a propensity score-adjusted analysis. Circulation. 2002;106:782–787
  108. Umana JP, Lai DT, Mitchell RS, et al. Is medical therapy still the optimal treatment strategy for patients with acute type B aortic dissections?. J Thorac Cardiovasc Surg. 2002;124:896–910
  109. Umana JP, Miller DC, Mitchell RS. What is the best treatment for patients with acute type B aortic dissections–medical, surgical, or endovascular stent-grafting?. Ann Thorac Surg. 2002;74:S1840–S1843
  110. Vincent JL, Baron JF, Reinhart K, et al. Anemia and blood transfusion in critically ill patients. JAMA. 2002;288:1499–1507
  111. Weber U, Luedtke R, Friese KH, Fischer I, Moeller H. A non-randomised pilot study to compare complementary and conventional treatments of acute sinusitis. Forsch Komplementarmed Klass Naturheilkd. 2002;9:99–104
  112. Weiss JP, Saynina O, McDonald KM, McClellan MB, Hlatky MA. Effectiveness and cost-effectiveness of implantable cardioverter defibrillators in the treatment of ventricular arrhythmias among Medicare beneficiaries. Am J Med. 2002;112:519–527
  113. Winkelmayer WC, Glynn RJ, Mittleman MA, Levin R, Pliskin JS, Avorn J. Comparing mortality of elderly patients on hemodialysis versus peritoneal dialysis: a propensity score approach. J Am Soc Nephrol. 2002;13:2353–2362
  114. Aronow HD, Topol EJ, Roe MT, et al. Effect of lipid-lowering therapy on early mortality after acute coronary syndromes: an observational study. Lancet. 2001;357:1063–1068
  115. Beuth J, Ost B, Pakdaman A, et al. Impact of complementary oral enzyme application on the postoperative treatment results of breast cancer patients: results of an epidemiological multicentre retrolective cohort study. Cancer Chemother Pharmacol. 2001;47S:S45–S54
  116. Carmeli Y, Castro J, Eliopoulos GM, Samore MH. Clinical isolation and resistance patterns of and superinfection with 10 nosocomial pathogens after treatment with ceftriaxone versus ampicillin-sulbactam. Antimicrob Agents Chemother. 2001;45:275–279
  117. Dammann O, Allred EN, Kuban KC, et al. Hypocarbia during the first 24 postnatal hours and white matter echolucencies in newborns < or = 28 weeks gestation. Pediatr Res. 2001;49:388–393
  118. Earle CC, Tsai JS, Gelber RD, Weinstein MC, Neumann PJ, Weeks JC. Effectiveness of chemotherapy for advanced lung cancer in the elderly: instrumental variable and propensity analysis. J Clin Oncol. 2001;19:1064–1070
  119. Ellis SG, Brener SJ, Lincoff AM, et al. beta-blockers before percutaneous coronary intervention do not attenuate postprocedural creatine kinase isoenzyme rise. Circulation. 2001;104:2685–2688
  120. Feng D, D'Agostino RB, Silbershatz H, et al. Hemostatic state and atrial fibrillation (the Framingham Offspring Study). Am J Cardiol. 2001;87:168–171
  121. Foody JM, Cole CR, Blackstone EH, Lauer MS. A propensity analysis of cigarette smoking and mortality with consideration of the effects of alcohol. Am J Cardiol. 2001;87:706–711
  122. Gillinov AM, Wierup PN, Blackstone EH, et al. Is repair preferable to replacement for ischemic mitral regurgitation?. J Thorac Cardiovasc Surg. 2001;122:1125–1141
  123. Gum PA, Thamilarasan M, Watanabe J, Blackstone EH, Lauer MS. Aspirin use and all-cause mortality among patients being evaluated for known or suspected coronary artery disease: A propensity analysis. JAMA. 2001;286:1187–1194
  124. Hayashi K, Hasegawa K, Kobayashi S. Effects of angiotensin-converting enzyme inhibitors on the treatment of anemia with erythropoietin. Kidney Int. 2001;60:1910–1916
  125. Ioannidis JP, Galanos O, Katritsis D, et al. Early mortality and morbidity of bilateral versus single internal thoracic artery revascularization: propensity and risk modeling. J Am Coll Cardiol. 2001;37:521–528
  126. Keating NL, Weeks JC, Landrum MB, Borbas C, Guadagnoli E. Discussion of treatment options for early-stage breast cancer: effect of provider specialty on type of surgery and satisfaction. Med Care. 2001;39:681–691
  127. Margolis DJ, Kantor J, Santanna J, Strom BL, Berlin JA. Effectiveness of platelet releasate for the treatment of diabetic neuropathic foot ulcers. Diabetes Care. 2001;24:483–488
  128. Mast ST, Gersing KR, Anstrom KJ, Krishnan KR, Califf RM, Jollis JG. Association between selective serotonin-reuptake inhibitor therapy and heart valve regurgitation. Am J Cardiol. 2001;87:989–993
  129. Mitra N, Schnabel FR, Neugut AI, Heitjan DF. Estimating the effect of an intensive surveillance program on stage of breast carcinoma at diagnosis: a propensity score analysis. Cancer. 2001;91:1709–1715
  130. Mukamal KJ, Maclure M, Muller JE, Sherwood JB, Mittleman MA. Prior alcohol consumption and mortality following acute myocardial infarction. JAMA. 2001;285:1965–1970
  131. Normand ST, Landrum MB, Guadagnoli E, et al. Validating recommendations for coronary angiography following acute myocardial infarction in the elderly: a matched analysis using propensity scores. J Clin Epidemiol. 2001;54:387–398
  132. Osswald BR, Blackstone EH, Tochtermann U, et al. Does the completeness of revascularization affect early survival after coronary artery bypass grafting in elderly patients?. Eur J Cardiothorac Surg. 2001;20:120–125
  133. Peterson JG, Topol EJ, Roe MT, et al. PURSUIT Investigators; Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy. Prognostic importance of concomitant heparin with eptifibatide in acute coronary syndromes. Am J Cardiol. 2001;87:532–536
  134. Polanczyk CA, Rohde LE, Goldman L, et al. Right heart catheterization and cardiac complications in patients undergoing noncardiac surgery: an observational study. JAMA. 2001;286:309–314
  135. Rathore SS, Gersh BJ, Weinfurt KP, Oetgen WJ, Schulman KA, Solomon AJ. The role of reperfusion therapy in paced patients with acute myocardial infarction. Am Heart J. 2001;142:516–519
  136. Sernyak MJ, Desai R, Stolar M, Rosenheck R. Impact of clozapine on completed suicide. Am J Psychiatry. 2001;158:931–937
  137. Sernyak MJ, Rosenheck R, Desai R, Stolar M, Ripper G. Impact of clozapine prescription on inpatient resource utilization. J Nerv Ment Dis. 2001;189:766–773
  138. Stenestrand U, Wallentin L. Early statin treatment following acute myocardial infarction and 1-year survival. JAMA. 2001;285:430–436
  139. Suero JA, Marso SP, Jones PG, et al. Procedural outcomes and long-term survival among patients undergoing percutaneous coronary intervention of a chronic total occlusion in native coronary arteries: a 20-year experience. J Am Coll Cardiol. 2001;38:409–414
  140. Welch RD, Zalenski RJ, Frederick PD, et al. Prognostic value of a normal or nonspecific initial electrocardiogram in acute myocardial infarction. JAMA. 2001;286:1977–1984
  141. Wiles NJ, Lunt M, Barrett EM, et al. Reduced disability at five years with early treatment of inflammatory polyarthritis: results from a large observational cohort, using propensity models to adjust for disease severity. Arthritis Rheum. 2001;44:1033–1042
  142. Hagan H, Thiede H. Changes in injection risk behavior associated with participation in the Seattle needle-exchange program. J Urban Health. 2000;77:369–382
  143. Holman WL, Li Q, Kiefe CI, et al. Prophylactic value of preincision intra-aortic balloon pump: analysis of a statewide experience. J Thorac Cardiovasc Surg. 2000;120:1112–1119
  144. Murdoch SD, Cohen AT, Bellamy MC. Pulmonary artery catheterization and mortality in critically ill patients. Br J Anaesth. 2000;85:611–615
  145. Potosky AL, Legler J, Albertsen PC, et al. Health outcomes after prostatectomy or radiotherapy for prostate cancer: results from the Prostate Cancer Outcomes Study. J Natl Cancer Inst. 2000;92:1582–1592
  146. Weiss JP, Gruver C, Kaul S, Harrell FE, Sklenar J, Dent JM. Ordering an echocardiogram for evaluation of left ventricular function: level of expertise necessary for efficient use. J Am Soc Echocardiogr. 2000;13:124–130
  147. Yau TM, El-Ghoneimi YA, Armstrong S, Ivanov J, David TE. Mitral valve repair and replacement for rheumatic disease. J Thorac Cardiovasc Surg. 2000;119:53–60
  148. Lytle BW, Blackstone EH, Loop FD, et al. Two internal thoracic artery grafts are better than one. J Thorac Cardiovasc Surg. 1999;117:855–872
  149. Nakamura Y, Moss AJ, Brown MW, Kinoshita M, Kawai C. Multicenter Myocardial Ischemia Research Group. Long-term nitrate use may be deleterious in ischemic heart disease: A study using the databases from two large-scale postinfarction studies. Am Heart J. 1999;138:577–585
  150. Schofer J, Rau T, Schluter M, Mathey DG. Restenosis after stenting of matched occluded and non-occluded coronary arteries: should there be a difference?. Eur Heart J. 1999;20:1175–1181
  151. Shepardson LB, Youngner SJ, Speroff T, Rosenthal GE. Increased risk of death in patients with do-not-resuscitate orders. Med Care. 1999;37:727–737
  152. Takizawa T, Haga M, Yagi N, et al. Pulmonary function after segmentectomy for small peripheral carcinoma of the lung. J Thorac Cardiovasc Surg. 1999;118:536–541
  153. Barker FG, Chang SM, Gutin PH, et al. Survival and functional status after resection of recurrent glioblastoma multiforme. Neurosurgery. 1998;42:709–720
  154. Barosi G, Ambrosetti A, Centra A, et al. Italian Cooperative Study Group on Myeloid with Myeloid Metaplasia. Splenectomy and risk of blast transformation in myelofibrosis with myeloid metaplasia. Blood. 1998;91:3630–3636
  155. Intrator O, Berg K. Benefits of home health care after inpatient rehabilitation for hip fracture: health service use by Medicare beneficiaries, 1987–1992. Arch Phys Med Rehabil. 1998;79:1195–1199
  156. Regueiro CR, Hamel MB, Davis RB, Desbiens N, Connors AF, Phillips RS, et al. A comparison of generalist and pulmonologist care for patients hospitalized with severe chronic obstructive pulmonary disease: resource intensity, hospital costs, and survival. Am J Med. 1998;105:366–372
  157. Smith NL, Reiber GE, Psaty BM, et al. Health outcomes associated with beta-blocker and diltiazem treatment of unstable angina. J Am Coll Cardiol. 1998;32:1305–1311
  158. Connors AF, Speroff T, Dawson NV, et al. SUPPORT Investigators. The effectiveness of right heart catheterization in the initial care of critically ill patients. JAMA. 1996;276:889–897
  159. Lieberman E, Lang JM, Cohen A, D'Agostino R, Datta S, Frigoletto FD. Association of epidural analgesia with cesarean delivery in nulliparas. Obstet Gynecol. 1996;88:993–1000
  160. Stone RA, Obrosky DS, Singer DE, Kapoor WN, Fine MJ. Pneumonia Patient Outcomes Research Team (PORT) Investigators. Propensity score adjustment for pretreatment differences between hospitalized and ambulatory patients with community-acquired pneumonia. Med Care. 1995;33S:AS56–AS66
  161. Myers WO, Gersh BJ, Fisher LD, et al. Medical versus early surgical therapy in patients with triple-vessel disease and mild angina pectoris: a CASS registry study of survival. Ann Thorac Surg. 1987;44:471–486
  162. Myers WO, Gersh BJ, Fisher LD, et al. Coronary Artery Surgery Study. Time to first new myocardial infarction in patients with mild angina and three-vessel disease comparing medicine and early surgery: a CASS registry study of survival. Ann Thorac Surg. 1987;43:599–612
  163. Gillinov AM, Blackstone EH, Cosgrove DM, et al. Mitral valve repair with aortic valve replacement is superior to double valve replacement. J Thorac Cardiovasc Surg. 2003;125:1372–1387
  164. Gregg EW, Cauley JA, Stone K, et al. Relationship of changes in physical activity and mortality among older women. JAMA. 2003;289:2379–2386
  165. Horwitz PA, Berlin JA, Sauer WH, Laskey WK, Krone RJ, Kimmel SE. Bleeding risk of platelet glycoprotein IIb/IIIa receptor antagonists in broad-based practice (results from the Society for Cardiac Angiography and Interventions Registry). Am J Cardiol. 2003;91:803–806
  166. Johnson D, Jin Y, Quan H, Cujec B. Beta-blockers and angiotensin-converting enzyme inhibitors/receptor blockers prescriptions after hospital discharge for heart failure are associated with decreased mortality in Alberta, Canada. J Am Coll Cardiol. 2003;42:1438–1445
  167. MacDonald TM, Morant SV, Goldstein JL, Burke TA, Pettitt D. Channelling bias and the incidence of gastrointestinal haemorrhage in users of meloxicam, coxibs, and older, non-specific non-steroidal anti-inflammatory drugs. Gut. 2003;52:1265–1270
  168. Martelli G, Miceli R, De Palo G, et al. Is axillary lymph node dissection necessary in elderly patients with breast carcinoma who have a clinically uninvolved axilla?. Cancer. 2003;97:1156–1163
  169. McGuire DK, Anstrom KJ, Peterson ED. Influence of the Angioplasty Revascularization Investigation National Heart, Lung, and Blood Institute Diabetic Clinical Alert on practice patterns: results from the National Cardiovascular Network Database. Circulation. 2003;107:1864–1870
  170. Murray PK, Singer M, Dawson NV, Thomas CL, Cebul RD. Outcomes of rehabilitation services for nursing home residents. Arch Phys Med Rehabil. 2003;84:1129–1136
  171. O'Neill BP, Iturria NJ, Link MJ, Pollock BE, Ballman KV, O'Fallon JR. A comparison of surgical resection and stereotactic radiosurgery in the treatment of solitary brain metastases. Int J Radiat Oncol Biol Phys. 2003;55:1169–1176
  172. Reddan DN, Szczech LA, Tuttle RH, et al. Chronic kidney disease, mortality, and treatment strategies among patients with clinically significant coronary artery disease. J Am Soc Nephrol. 2003;14:2373–2380
  173. Stenestrand U, Wallentin L. Fibrinolytic therapy in patients 75 years and older with ST-segment-elevation myocardial infarction: one-year follow-up of a large prospective cohort. Arch Intern Med. 2003;163:965–971
  174. Urbach DR, Bell CM, Swanstrom LL, Hansen PD. Cohort study of surgical bypass to the gallbladder or bile duct for the palliation of jaundice due to pancreatic cancer. Ann Surg. 2003;237:86–93
  175. Yu DT, Platt R, Lanken PN, et al. Relationship of pulmonary artery catheter use to mortality and resource utilization in patients with severe sepsis. Crit Care Med. 2003;31:2734–2741
  176. Ferraris VA, Ferraris SP, Joseph O, Wehner P, Mentzer RM. Aspirin and postoperative bleeding after coronary artery bypass grafting. Ann Surg. 2002;235:820–827
  177. Goldstein CL, Racz M, Hannan EL. Impact of cardiac catheterization-percutaneous coronary intervention timing on in-hospital mortality. Am Heart J. 2002;144:561–567
  178. Hu FB, Bronner L, Willett WC, et al. Fish and omega-3 fatty acid intake and risk of coronary heart disease in women. JAMA. 2002;287:1815–1821
  179. Jasmer RM, Saukkonen JJ, Blumberg HM, et al. Short-course rifampin and pyrazinamide compared with isoniazid for latent tuberculosis infection: a multicenter clinical trial. Ann Intern Med. 2002;137:640–647
  180. Kimmel SE, Berlin JA, Kinman JL, et al. Parenteral ketorolac and risk of myocardial infarction. Pharmacoepidemiol Drug Saf. 2002;11:113–119
  181. Mukamal KJ, Maclure M, Muller JE, Sherwood JB, Mittleman MA. Tea consumption and mortality after acute myocardial infarction. Circulation. 2002;105:2476–2481
  182. Patel NC, Grayson AD, Jackson M, et al. The effect off-pump coronary artery bypass surgery on in-hospital mortality and morbidity. Eur J Cardiothorac Surg. 2002;22:255–260
  183. Schneeweiss S, Walker AM, Glynn RJ, Maclure M, Dormuth C, Soumerai SB. Outcomes of reference pricing for angiotensin-converting-enzyme inhibitors. N Engl J Med. 2002;346:822–829
  184. Schomig A, Mehilli J, Holle H, et al. Statin treatment following coronary artery stenting and one-year survival. J Am Coll Cardiol. 2002;40:854–861
  185. Soumerai SB, McLaughlin TJ, Ross-Degnan D, Christiansen CL, Gurwitz JH. Effectiveness of thrombolytic therapy for acute myocardial infarction in the elderly: cause for concern in the old-old. Arch Intern Med. 2002;162:561–568
  186. Tanasescu M, Leitzmann MF, Rimm EB, Willett WC, Stampfer MJ, Hu FB. Exercise type and intensity in relation to coronary heart disease in men. JAMA. 2002;288:1994–2000
  187. Turner BJ, Laine C, Cohen A, Hauck WW. Effect of medical, drug abuse, and mental health care on receipt of dental care by drug users. J Subst Abuse Treat. 2002;23:239–246
  188. Wright RS, Williams BA, Cramner H, et al. Elevations of cardiac troponin I are associated with increased short-term mortality in noncardiac critically ill emergency department patients. Am J Cardiol. 2002;90:634–636
  189. Ascione R, Nason G, Al-Ruzzeh S, Ko C, Ciulli F, Angelini GD. Coronary revascularization with or without cardiopulmonary bypass in patients with preoperative nondialysis-dependent renal insufficiency. Ann Thorac Surg. 2001;72:2020–2025
  190. Cen YY, Glower DD, Landolfo K, et al. Comparison of survival after mitral valve replacement with biologic and mechanical valves in 1139 patients. J Thorac Cardiovasc Surg. 2001;122:569–577
  191. Kachele H, Kordy H, Richard M. Research group TR-EAT. Therapy amount and outcome of inpatient psychodynamic treatment of eating disorders in Germany: Data from a multicenter study. Psychotherapy Res. 2001;11:239–257
  192. Mehta RH, Bruckman D, Das S, et al. Implications of increased left ventricular mass index on in-hospital outcomes in patients undergoing aortic valve surgery. J Thorac Cardiovasc Surg. 2001;122:919–928
  193. Novaro GM, Tiong IY, Pearce GL, Lauer MS, Sprecher DL, Griffin BP. Effect of hydroxymethylglutaryl coenzyme a reductase inhibitors on the progression of calcific aortic stenosis. Circulation. 2001;104:2205–2209
  194. O'Day SJ, Boasberg PD, Kristedja TS, et al. High-dose tamoxifen added to concurrent biochemotherapy with decrescendo interleukin-2 in patients with metastatic melanoma. Cancer. 2001;92:609–619
  195. Popiela T, Kulig J, Hanisch J, Bock PR. Influence of a complementary treatment with oral enzymes on patients with colorectal cancers–an epidemiological retrolective cohort study. Cancer Chemother Pharmacol. 2001;47S:S55–S63
  196. Rumsfeld JS, Magid DJ, Plomondon ME, et al. Predictors of quality of life following acute coronary syndromes. Am J Cardiol. 2001;88:781–784
  197. Schnuelle P, Berger S, de Boer J, Persijn G, van der Woude FJ. Effects of catecholamine application to brain-dead donors on graft survival in solid organ transplantation. Transplantation. 2001;72:455–463
  198. Shlipak MG, Browner WS, Noguchi H, Massie B, Frances CD, McClellan M. Comparison of the effects of angiotensin converting-enzyme inhibitors and beta blockers on survival in elderly patients with reduced left ventricular function after myocardial infarction. Am J Med. 2001;110:425–433
  199. Tu JV, Austin PC, Chan BT. Relationship between annual volume of patients treated by admitting physician and mortality after acute myocardial infarction. JAMA. 2001;285:3116–3122
  200. Auerbach AD, Hamel MB, Davis RB, et al. SUPPORT Investigators; Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments. Resource use and survival of patients hospitalized with congestive heart failure: differences in care by specialty of the attending physician. Ann Intern Med. 2000;132:191–200
  201. Hannan EL, Racz MJ, Arani DT, McCallister BD, Walford G, Ryan TJ. A comparison of short- and long-term outcomes for balloon angioplasty and coronary stent placement. J Am Coll Cardiol. 2000;36:395–403
  202. Petersen LA, Normand SL, Daley J, McNeil BJ. Outcome of myocardial infarction in Veterans Health Administration patients as compared with Medicare patients. N Engl J Med. 2000;343:1934–1941
  203. Shwartz M, Saitz R, Mulvey K, Brannigan P. The value of acupuncture detoxification programs in a substance abuse treatment system. J Subst Abuse Treat. 1999;17:305–312
  204. Cooper GS, Chak A, Connors AF, Harper DL, Rosenthal GE. The effectiveness of early endoscopy for upper gastrointestinal hemorrhage: a community-based analysis. Med Care. 1998;36:462–474
  205. Fraenkel L, Zhang Y, Chaisson CE, Evans SR, Wilson PW, Felson DT. The association of estrogen replacement therapy and the Raynaud phenomenon in postmenopausal women. Ann Intern Med. 1998;129:208–211
  206. Van den Eeden SK, Glasser M, Mathias SD, Colwell HH, Pasta DJ, Kunz K. Quality of life, health care utilization, and costs among women undergoing hysterectomy in a managed-care setting. Am J Obstet Gynecol. 1998;178:91–100
  207. King SB, Barnhart HX, Kosinski AS, et al. Emory Angioplasty versus Surgery Trial Investigators. Angioplasty or surgery for multivessel coronary artery disease: comparison of eligible registry and randomized patients in the EAST trial and influence of treatment selection on outcomes. Am J Cardiol. 1997;79:1453–1459
  208. Soumerai SB, McLaughlin TJ, Spiegelman D, Hertzmark E, Thibault G, Goldman L. Adverse outcomes of underuse of beta-blockers in elderly survivors of acute myocardial infarction. JAMA. 1997;277:115–121
  209. Fiebach NH, Cook EF, Lee TH, et al. Outcomes in patients with myocardial infarction who are initially admitted to stepdown units: data from the Multicenter Chest Pain Study. Am J Med. 1990;89:15–20
  210. Cepeda MS, Boston R, Farrar JT, Strom BL. Comparison of logistic regression versus propensity score when the number of events is low and there are multiple confounders. Am J Epidemiol. 2003;158:280–287
  211. Weitzen S, Lapane KL, Toledano AY, Hume AL, Mor V. Principles for modeling propensity scores in medical research: a systematic literature review. Pharmacoepidemiol Drug Saf. 2004;13:841–853
  212. Greenland S, Robins JM. Confounding and misclassification. Am J Epidemiol. 1985;122:495–506
  213. Robins JM, Greenland S. The role of model selection in causal inference from nonexperimental data. Am J Epidemiol. 1986;123:392–402
  214. Cook EF, Goldman L. Performance of tests of significance based on stratification by a multivariate confounder score or by a propensity score. J Clin Epidemiol. 1989;42:317–324
  215. Harrell FE, Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med. 1996;15:361–387
  216. Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49:1373–1379
  217. Braitman LE, Rosenbaum PR. Rare outcomes, common treatments: analytic strategies using propensity scores. Ann Intern Med. 2002;137:693–695
  218. Cepeda MS. The use of propensity scores in pharmacoepidemiologic research. Pharmacoepidemiol Drug Saf. 2000;9:103–104
  219. Drake C. Effects of misspecification of the propensity score on estimators of treatment effect. Biometrics. 1993;49:1231–1236
  220. Dehejia R, Wahba S. Causal effect in non-experimental studies: reevaluating the evaluation of training programs. J Am Stat Assoc. 1999;94:1053–1062
  221. Stürmer T, Schneeweiss S, Brookhart MA, Rothman KJ, Avorn J, Glynn RJ. Analytic Strategies to adjust confounding using Exposure Propensity Scores and Disease Risk Scores: Nonsteroidal Antiinflammatory Drugs and Short-Term Mortality in the Elderly. Am J Epidemiol. 2005;161:891–898
  222. Hosmer DW, Lemenshow S. Applied logistic regression. 2nd ed.. New York: Wiley; 2000;
  223. Perkins SM, Tu W, Underhill MG, Zhou XH, Murray MD. The use of propensity scores in pharmacoepidemiologic research. Pharmacoepidemiol Drug Saf. 2000;9:93–101

PII: S0895-4356(05)00224-6

doi:10.1016/j.jclinepi.2005.07.004

Journal of Clinical Epidemiology
Volume 59, Issue 5 , Pages 437.e1-437.e24, May 2006

Article Tools

* Image Usage & Resolution