Home About us Contact
|
Home About us Contact | ||
|
|
||
Propensity Score (propensity + score)
Terms modified by Propensity Score Selected AbstractsUtilizing Propensity Scores to Estimate Causal Treatment Effects with Censored Time-Lagged DataBIOMETRICS, Issue 4 2001Kevin J. Anstrom Summary. Observational studies frequently are conducted to compare long-term effects of treatments. Without randomization, patients receiving one treatment are not guaranteed to be prognostically comparable to those receiving another treatment. Furthermore, the response of interest may be right-censored because of incomplete follow-up. Statistical methods that do not account for censoring and confounding may lead to biased estimates. This article presents a method for estimating treatment effects in nonrandomized studies with right-censored responses. We review the assumptions required to estimate average causal effects and derive an estimator for comparing two treatments by applying inverse weights to the complete cases. The weights are determined according to the estimated probability of receiving treatment conditional on covariates and the estimated treatment-specific censoring distribution. By utilizing martingale representations, the estimator is shown to be asymptotically normal and an estimator for the asymptotic variance is derived. Simulation results are presented to evaluate the properties of the estimator. These methods are applied to an observational data set of acute coronary syndrome patients from Duke University Medical Center to estimate the effect of a treatment strategy on the mean 5-year medical cost. [source] Statin treatment after a recent TIA or stroke: is effectiveness shown in randomized clinical trials also observed in everyday clinical practice?ACTA NEUROLOGICA SCANDINAVICA, Issue 1 2010H. F. Lingsma Lingsma HF, Steyerberg EW, Scholte op Reimer WJM, van Domburg R, Dippel DWJ, the Netherlands Stroke Survey Investigators. Statin treatment after a recent TIA or stroke: is effectiveness shown in randomized clinical trials also observed in everyday clinical practice? Acta Neurol Scand: 2010: 122: 15,20. © 2009 The Authors Journal compilation © 2009 Blackwell Munksgaard. Aim and background,,, The benefit of statin treatment in patients with a previous ischemic stroke or transient ischemic attack (TIA) has been demonstrated in randomized clinical trials (RCT). However, the effectiveness in everyday clinical practice may be decreased because of a different patient population and less controlled setting. We aim to describe statin use in an unselected cohort of patients, identify factors related to statin use and test whether the effect of statins on recurrent vascular events and mortality observed in RCTs is also observed in everyday clinical practice. Methods,,, In 10 centers in the Netherlands, patients admitted to the hospital or visiting the outpatient clinic with a recent TIA or ischemic stroke were prospectively and consecutively enrolled between October 2002 and May 2003. Statin use was determined at discharge and during follow-up. We used logistic regression models to estimate the effect of statins on the occurrence of vascular events (stroke or myocardial infarction) and mortality within 3 years. We adjusted for confounders with a propensity score that relates patient characteristics to the probability of using statins. Results,,, Of the 751 patients in the study, 252 (34%) experienced a vascular event within 3 years. Age, elevated cholesterol levels and other cardiovascular risk factors were associated with statin use at discharge. After 3 years, 109 of 280 (39%) of the users at discharge had stopped using statins. Propensity score adjusted analyses showed a beneficial effect of statins on the occurrence of the primary outcome (odds ratio 0.8, 95% CI: 0.6,1.2). Conclusion,,, In our study, we found poor treatment adherence to statins. Nevertheless, after adjustment for the differences between statin users and non-statin users, the observed beneficial effect of statins on the occurrence of vascular events within 3 years, although not statistically significant, is compatible with the effect observed in clinical trials. [source] Postmastectomy radiation therapy for lymph node-negative, locally advanced breast cancer after modified radical mastectomyCANCER, Issue 1 2008Analysis of the NCI Surveillance, End Results database, Epidemiology Abstract BACKGROUND. The role of postmastectomy radiotherapy (PMRT) for lymph node-negative locally advanced breast carcinoma (T3N0M0) after modified radical mastectomy (MRM) with regard to improvement in survival remains an area of controversy. METHODS. The 1973,2004 National Cancer Institute (NCI) Surveillance, Epidemiology, and End Results (SEER) database was examined for patients with T3N0M0 ductal, lobular, or mixed ductal and lobular carcinoma of the breast who underwent MRM, treated from 1988-2003. Patients who were men, who had positive lymph nodes, who survived ,6 months, for whom breast cancer was not the first malignancy, who had nonbeam radiation, intraoperative or preoperative radiation were excluded. The average treatment effect of PMRT on mortality was estimated with a propensity score case-matched analysis. RESULTS. In all, 1777 patients were identified; 568 (32%) patients received PMRT. Median tumor size was 6.3 cm. The median number of lymph nodes examined was 14 (range, 1,49). Propensity score matched case-control analysis showed no improvement in overall survival with the delivery of PMRT in this group. Older patients, patients with ER, disease (compared with ER+), and patients with high-grade tumors (compared with well differentiated) had increased mortality. CONCLUSIONS. The use of PMRT for T3N0M0 breast carcinoma after MRM is not associated with an increase in overall survival. It was not possible to analyze local control in this study given the limitations of the SEER database. The impact of potential improvement in local control as it relates to overall survival should be the subject of further investigation. Cancer 2008. © 2008 American Cancer Society. [source] Using propensity scores to estimate the effects of insecticides on stream invertebrates from observational data,ENVIRONMENTAL TOXICOLOGY & CHEMISTRY, Issue 7 2009Lester L. Yuan Abstract Analyses of observational data can provide insights into relationships between environmental conditions and biological responses across a broader range of natural conditions than experimental studies, potentially complementing insights gained from experiments. However, observational data must be analyzed carefully to minimize the likelihood that confounding variables bias observed relationships. Propensity scores provide a robust approach for controlling for the effects of measured confounding variables when analyzing observational data. Here, we use propensity scores to estimate changes in mean invertebrate taxon richness in streams that have experienced insecticide concentrations that exceed aquatic life use benchmark concentrations. A simple comparison of richness in sites exposed to elevated insecticides with those that were not exposed suggests that exposed sites had on average 6.8 fewer taxa compared to unexposed sites. The presence of potential confounding variables makes it difficult to assert a causal relationship from this simple comparison. After controlling for confounding factors using propensity scores, the difference in richness between exposed and unexposed sites was reduced to 4.1 taxa, a difference that was still statistically significant. Because the propensity score analysis controlled for the effects of a wide variety of possible confounding variables, we infer that the change in richness observed in the propensity score analysis was likely caused by insecticide exposure. [source] A propensity score approach to correction for bias due to population stratification using genetic and non-genetic factorsGENETIC EPIDEMIOLOGY, Issue 8 2009Huaqing Zhao Abstract Confounding due to population stratification (PS) arises when differences in both allele and disease frequencies exist in a population of mixed racial/ethnic subpopulations. Genomic control, structured association, principal components analysis (PCA), and multidimensional scaling (MDS) approaches have been proposed to address this bias using genetic markers. However, confounding due to PS can also be due to non-genetic factors. Propensity scores are widely used to address confounding in observational studies but have not been adapted to deal with PS in genetic association studies. We propose a genomic propensity score (GPS) approach to correct for bias due to PS that considers both genetic and non-genetic factors. We compare the GPS method with PCA and MDS using simulation studies. Our results show that GPS can adequately adjust and consistently correct for bias due to PS. Under no/mild, moderate, and severe PS, GPS yielded estimated with bias close to 0 (mean=,0.0044, standard error=0.0087). Under moderate or severe PS, the GPS method consistently outperforms the PCA method in terms of bias, coverage probability (CP), and type I error. Under moderate PS, the GPS method consistently outperforms the MDS method in terms of CP. PCA maintains relatively high power compared to both MDS and GPS methods under the simulated situations. GPS and MDS are comparable in terms of statistical properties such as bias, type I error, and power. The GPS method provides a novel and robust tool for obtaining less-biased estimates of genetic associations that can consider both genetic and non-genetic factors. Genet. Epidemiol. 33:679,690, 2009. © 2009 Wiley-Liss, Inc. [source] The Effect of High Nursing Surveillance on Hospital CostJOURNAL OF NURSING SCHOLARSHIP, Issue 2 2008Leah L. Shever Purpose: The purpose of this study was to determine the cost of one nursing treatment, surveillance, for older, hospitalized adults at risk for falling. Design: An observational study using information from data repositories at one Midwestern tertiary hospital. The inclusion criteria included patients age >60 years, admitted to the hospital between July 1, 1998 and June 31, 2002, at risk for falls or received the nursing treatment of fall prevention. Methods: Data came from clinical and administrative data repositories that included Nursing Interventions Classification (NIC). The nursing treatment of interest was surveillance and total hospital cost associated with surveillance was the dependent variable. Propensity-score analysis and generalized estimating equations (GEE) were used as methods to analyze the data. Independent variables related to patient characteristics, clinical conditions, nurse staffing, medical treatments, pharmaceutical treatments, and other nursing treatments were controlled for statistically. Findings: The total median cost per hospitalization was $9,274 for this sample. The median cost was different (p = 0.050) for patients who received high versus low surveillance. High surveillance delivery cost $191 more per hospitalization than did low surveillance delivery. Conclusion: Propensity scores were applied to determine the cost of surveillance among hospitalized adults at risk for falls in this observational study. The findings show the effect of high surveillance delivery on total hospital cost compared to low surveillance delivery and provides an example of a useful method of determining cost of nursing care rather than including it in the room rate. More studies are needed to determine the effects of nursing treatments on cost and other patient outcomes in order for nurses to provide cost-effective care. Propensity scores were a useful method for determining the effect of nursing surveillance on hospital cost in this observational study. Clinical Relevance: The results of this study along with possible clinical benefits would indicate that frequent nursing surveillance is important and might support the need for additional nursing staff to deliver frequent surveillance. [source] Weaknesses of goodness-of-fit tests for evaluating propensity score models: the case of the omitted confounder,PHARMACOEPIDEMIOLOGY AND DRUG SAFETY, Issue 4 2005Sherry Weitzen PhD Abstract Purpose Propensity scores are used in observational studies to adjust for confounding, although they do not provide control for confounders omitted from the propensity score model. We sought to determine if tests used to evaluate logistic model fit and discrimination would be helpful in detecting the omission of an important confounder in the propensity score. Methods Using simulated data, we estimated propensity scores under two scenarios: (1) including all confounders and (2) omitting the binary confounder. We compared the propensity score model fit and discrimination under each scenario, using the Hosmer,Lemeshow goodness-of-fit (GOF) test and the c-statistic. We measured residual confounding in treatment effect estimates adjusted by the propensity score omitting the confounder. Results The GOF statistic and discrimination of propensity score models were the same for models excluding an important predictor of treatment compared to the full propensity score model. The GOF test failed to detect poor model fit for the propensity score model omitting the confounder. C-statistics under both scenarios were similar. Residual confounding was observed from using the propensity score excluding the confounder (range: 1,30%). Conclusions Omission of important confounders from the propensity score leads to residual confounding in estimates of treatment effect. However, tests of GOF and discrimination do not provide information to detect missing confounders in propensity score models. Our findings suggest that it may not be necessary to compute GOF statistics or model discrimination when developing propensity score models. Copyright © 2004 John Wiley & Sons, Ltd. [source] Application of a propensity score to adjust for channelling bias with NSAIDs,PHARMACOEPIDEMIOLOGY AND DRUG SAFETY, Issue 6 2004S. V. Morant Abstract Purpose To compare the relative risks of upper GI haemorrhage (UGIH) in users of Newer versus Older, non-specific NSAIDs when adjusted for channelling bias by regression on individual covariates, a propensity score and both. Methods Cohort study of patients prescribed NSAIDs between June 1987 and January 2000. Exposure to Newer and Older non-specific NSAIDs was identified, and risk factors evaluated for each patient. Results of multiple covariate analyses and the propensity scoring technique to assess potential channelling bias in comparisons between Newer and Older non-specific NSAIDs were compared. Results This study included 7.1 thousand patient years (tpy) exposure to meloxicam, 1.6,tpy exposure to coxibs, and 628,tpy exposure to Older non-specific NSAIDs. Patients receiving Newer NSAIDs were older, more likely to have a history of GI symptoms, and at higher risk for GI complications. Adjusting for these risk factors reduced the relative risks of UGIH on meloxicam and coxibs versus Older non-specific NSAIDs to 0.84 (95%CI 0.60, 1.17) and 0.36 (0.14, 0.97) respectively. Conclusions Channelling towards high GI risk patients occurred in the prescribing of Newer NSAIDs. Propensity scores highlighted the markedly different risk profiles of users of Newer and Older non-specific NSAID. Correcting for channelling bias, coxib exposure, but not meloxicam exposure, was associated with less UGIH than Older non-specific NSAID exposure. In the present study, corrections made by regression on a propensity score and on individual covariates were similar. Copyright © 2004 John Wiley & Sons, Ltd. [source] Statins Decrease Adverse Outcomes in Coronary Artery Bypass for Extensive Coronary Artery Disease as well as Left Main Coronary StenosisCARDIOVASCULAR THERAPEUTICS, Issue 2 2010Hui-Li Gan The aim of this study was to evaluate the effects of preoperative and postoperative statins on coronary artery bypass grafting (CABG) for extensive coronary artery disease as well as left main coronary stenosis (LMS). The data of 626 cases of extensive coronary artery disease as well as LMS patients in Anzhen Hospital between January 1998 and March 2008 for CABG procedure were retrospectively analyzed, and were classified as preoperative statin therapy group (Group A, n = 320) or preoperative no statin therapy group (Group B, n = 306). Propensity scores were estimated to determine the probability of inclusion into statin therapy group, resulting in the successful matching of 267 pairs. The incidence of in-hospital death, and atrial fibrillation or flutter and disabling stroke was higher in Group B than in Group A. The actuarial freedom from late events at 5 yrs were 98.75%± 0.73% for the postoperative statin therapy group and 88.33%± 3.71% for the postoperative no statin therapy group respectively, P= 0.000. The logistic regression revealed that CRP (>5.0 mg/L), and elevated Troponin I, and emergent procedure, and preoperative IABP support, and EF < 40% were the independent risk factors, and preoperatively statins was the protective factor for the perioperative death; and the Cox proportional hazard also revealed that preoperative IABP support and preoperative cardiac arrest, and EF < 40% were independent risk factors, and postoperatively statins were the protective factor for the late cardiac events. Preoperative statin therapy could provide protective effect in the perioperative period. Postoperative statin usage could provide protective effect on the late cardiac events. [source] Increased Mortality Associated With Low Use of Clopidogrel in Patients With Heart Failure and Acute Myocardial Infarction Not Undergoing Percutaneous Coronary InterventionCONGESTIVE HEART FAILURE, Issue 5 2010Scott Harris DO We studied the association of clopidogrel with mortality in acute myocardial infarction (AMI) patients with heart failure (HF) not receiving percutaneous coronary intervention (PCI). Background. Use of clopidogrel after AMI is low in patients with HF, despite the fact that clopidogrel is associated with absolute mortality reduction in AMI patients. Methods. All patients hospitalized with first-time AMI (2000 through 2005) and not undergoing PCI within 30 days from discharge were identified in national registers. Patients with HF treated with clopidogrel were matched by propensity score with patients not treated with clopidogrel. Similarly, 2 groups without HF were identified. Risks of all-cause death were obtained by the Kaplan,Meier method and Cox regression analyses. Results. We identified 56,944 patients with first-time AMI. In the matched cohort with HF (n=5050) and a mean follow-up of 1.50 years (SD=1.2), 709 (28.1%) and 812 (32.2%) deaths occurred in patients receiving and not receiving clopidogrel treatment, respectively (P=.002). The corresponding numbers for patients without HF (n=6092), with a mean follow-up of 2.05 years (SD=1.3), were 285 (9.4%) and 294 (9.7%), respectively (P=.83). Patients with HF receiving clopidogrel demonstrated reduced mortality (hazard ratio, 0.86; 95% confidence interval, 0.78,0.95) compared with patients with HF not receiving clopidogrel. No difference was observed among patients without HF (hazard ratio, 0.98; 95% confidence interval, 0.83,1.16). Conclusions. Clopidogrel was associated with reduced mortality in patients with HF who do not undergo PCI after their first-time AMI, whereas this association was not apparent in patients without HF. Further studies of the benefit of clopidogrel in patients with HF and AMI are warranted.,Bonde L, Sorensen R, Fosbol EL, et al. Increased mortality associated with low use of clopidogrel in patients with heart failure and acute myocardial infarction not undergoing percutaneous coronary intervention: a nationwide study. J Am Coll Cardiol. 2010;55:1300,1307. [source] The added risk of opioid problem use among treatment-seeking youth with marijuana and/or alcohol problem useADDICTION, Issue 4 2010Geetha A. Subramaniam Abstract Objectives To determine the added risk of opioid problem use (OPU) in youth with marijuana/alcohol problem use (MAPU). Methods A total of 475 youth (ages 14,21 years) with OPU + MAPU were compared to a weighted sample of 475 youth with MAPU only (i.e. no OPU) before and after propensity score matching on gender, age, race, level of care and weekly use of marijuana/alcohol. Youth were recruited from 88 drug treatment sites participating in eight Center for Substance Abuse Treatment-funded grants. At treatment intake, participants were administered the Global Appraisal of Individual Need to elicit information on demographic, social, substance, mental health, human immunodeficiency virus (HIV), physical and legal characteristics. Odds ratios with confidence intervals were calculated. Results The added risk of OPU among MAPU youth was associated with greater comorbidity; higher rates of psychiatric symptoms and trauma/victimization; greater needle use and sex-related HIV risk behaviours; and greater physical distress. The OPU + MAPU group was less likely to be African American or other race and more likely to be aged 15,17 years, Caucasian; report weekly drug use at home and among peers; engage in illegal behaviors and be confined longer; have greater substance abuse severity and polydrug use; and use mental health and substance abuse treatment services. Conclusions These findings expand upon the existing literature and highlight the substantial incremental risk of OPU on multiple comorbid areas among treatment-seeking youth. Further evaluation is needed to assess their outcomes following standard drug treatment and to evaluate specialized interventions for this subgroup of severely impaired youth. [source] A propensity score approach to correction for bias due to population stratification using genetic and non-genetic factorsGENETIC EPIDEMIOLOGY, Issue 8 2009Huaqing Zhao Abstract Confounding due to population stratification (PS) arises when differences in both allele and disease frequencies exist in a population of mixed racial/ethnic subpopulations. Genomic control, structured association, principal components analysis (PCA), and multidimensional scaling (MDS) approaches have been proposed to address this bias using genetic markers. However, confounding due to PS can also be due to non-genetic factors. Propensity scores are widely used to address confounding in observational studies but have not been adapted to deal with PS in genetic association studies. We propose a genomic propensity score (GPS) approach to correct for bias due to PS that considers both genetic and non-genetic factors. We compare the GPS method with PCA and MDS using simulation studies. Our results show that GPS can adequately adjust and consistently correct for bias due to PS. Under no/mild, moderate, and severe PS, GPS yielded estimated with bias close to 0 (mean=,0.0044, standard error=0.0087). Under moderate or severe PS, the GPS method consistently outperforms the PCA method in terms of bias, coverage probability (CP), and type I error. Under moderate PS, the GPS method consistently outperforms the MDS method in terms of CP. PCA maintains relatively high power compared to both MDS and GPS methods under the simulated situations. GPS and MDS are comparable in terms of statistical properties such as bias, type I error, and power. The GPS method provides a novel and robust tool for obtaining less-biased estimates of genetic associations that can consider both genetic and non-genetic factors. Genet. Epidemiol. 33:679,690, 2009. © 2009 Wiley-Liss, Inc. [source] Linkage mapping methods applied to the COGA data set: Presentation Group 4 of Genetic Analysis Workshop 14GENETIC EPIDEMIOLOGY, Issue S1 2005E. Warwick Daw Abstract Presentation Group 4 participants analyzed the Collaborative Study on the Genetics of Alcoholism data provided for Genetic Analysis Workshop 14. This group examined various aspects of linkage analysis and related issues. Seven papers included linkage analyses, while the eighth calculated identity-by-descent (IBD) probabilities. Six papers analyzed linkage to an alcoholism phenotype: ALDX1 (four papers), ALDX2 (one paper), or a combination both (one paper). Methods used included Bayesian variable selection coupled with Haseman-Elston regression, recursive partitioning to identify phenotype and covariate groupings that interact with evidence for linkage, nonparametric linkage regression modeling, affected sib-pair linkage analysis with discordant sib-pair controls, simulation-based homozygosity mapping in a single pedigree, and application of a propensity score to collapse covariates in a general conditional logistic model. Alcoholism linkage was found with ,2 of these approaches on chromosomes 2, 4, 6, 7, 9, 14, and 21. The remaining linkage paper compared the utility of several single-nucleotide polymorphism (SNP) and microsatellite marker maps for Monte Carlo Markov chain combined oligogenic segregation and linkage analysis, and analyzed one of the electrophysiological endophenotypes, ttth1, on chromosome 7. Linkage was found with all marker sets. The last paper compared the multipoint IBD information content of several SNP sets and the microsatellite set, and found that while all SNP sets examined contained more information than the microsatellite set, most of the information contained in the SNP sets was captured by a subset of the SNP markers with ,1-cM marker spacing. From these papers, we highlight three points: a 1-cM SNP map seems to capture most of the linkage information, so denser maps do not appear necessary; careful and appropriate use of covariates can aid linkage analysis; and sources of increased gene-sharing between relatives should be accounted for in analyses. Genet. Epidemiol. 29(Suppl. 1):S29,S34, 2005. © 2005 Wiley-Liss, Inc. [source] The effect of practice budgets on patient waiting times: allowing for selection biasHEALTH ECONOMICS, Issue 10 2004Mark Dusheiko Abstract Under the UK fundholding scheme, general practices could elect to hold a budget to meet the costs of some types of elective surgery (chargeable admissions) for their patients. It was alleged that patients of fundholding practices had shorter waits for elective surgery than the patients of non-fundholders. Comparison of waiting times between fundholding and non-fundholding practices are potentially confounded by selection bias as fundholding was voluntary. We estimate the effect of a practice's fundholding status on the waiting times of its patients using both cross-sectional methods (OLS, propensity score, instrumental variables, Heckman selection correction and Heckman heterogenous effects estimators) and difference in differences methodologies to correct for selection bias. The estimated effect of fundholding status was to significantly reduce the waiting times for chargeable admissions of the patients of fundholders by 4.1,6.6% (or 4,7 days) with the instrumental variables and Heckman selection correction estimators yielding the highest estimates. We also find that patients of fundholding practices had shorter waits (by 3.7% or 2 days) for non-chargeable elective admissions, suggesting that fundholders were able to obtain shorter waits for all types of elective admissions. Copyright © 2004 John Wiley & Sons, Ltd. [source] Time to Send the Preemie Home?HEALTH SERVICES RESEARCH, Issue 2p1 2009Additional Maturity at Discharge, Outcomes, Subsequent Health Care Costs Objective. To determine whether longer stays of premature infants allowing for increased physical maturity result in subsequent postdischarge cost savings that help counterbalance increased inpatient costs. Data Sources. One thousand four hundred and two premature infants born in the Northern California Kaiser Permanente Medical Care Program between 1998 and 2002. Study Design/Methods. Using multivariate matching with a time-dependent propensity score we matched 701 "Early" babies to 701 "Late" babies (developmentally similar at the time the earlier baby was sent home but who were discharged on average 3 days later) and assessed subsequent costs and clinical outcomes. Principal Findings. Late babies accrued inpatient costs after the Early baby was already home, yet costs after discharge through 6 months were virtually identical across groups, as were clinical outcomes. Overall, after the Early baby went home, the Late,Early cost difference was $5,016 (p<.0001). A sensitivity analysis suggests our conclusions would not easily be altered by failure to match on some unmeasured covariate. Conclusions. In a large integrated health care system, if a baby is ready for discharge (as defined by the typical criteria), staying longer increased inpatient costs but did not reduce postdischarge costs nor improve postdischarge clinical outcomes. [source] The Relationship of Post-acute Home Care Use to Medicaid Utilization and ExpendituresHEALTH SERVICES RESEARCH, Issue 3 2002Susan M. C. Payne Research Objectives: To describe the use of post-acute home care (PAHC) and total Medicaid expenditures among hospitalized nonelderly adult Medicaid eligibles and to test whether health services utilization rates or total Medicaid expenditures were lower among Medicaid eligibles who used PAHC compared to those who did not. Study Population: 5,299 Medicaid patients aged 18,64 discharged in 1992,1996 from 29 hospitals in the Cleveland Health Quality Choice (CHQC) project. Data Sources: Linked Ohio Medicaid claims and CHQC medical record abstract data. Data Extraction: One stay per patient was randomly selected. Design: Observational study. To control for treatment selection bias, we developed a model predicting the probability (propensity) a patient would be referred to PAHC, as a proxy for the patient's need for PAHC. We matched 430 patients who used Medicaid-covered PAHC ("USE") to patients who did not ("NO USE") by their propensity scores. Study outcomes were inpatient re-admission rates and days of stay (DOS), nursing home admission rates and DOS, and mean total Medicaid expenditures 90 and 180 days after discharge. Principal Findings: Of 3,788 medical patients, 12.1 percent were referred to PAHC; 64 percent of those referred used PAHC. Of 1,511 surgical patients, 10.9 percent were referred; 99 percent of those referred used PAHC. In 430 pairs of patients matched by propensity score, mean total Medicaid expenditures within 90 days after discharge were $7,649 in the USE group and $5,761 in the NO USE group. Total Medicaid expenditures were significantly higher in the USE group compared to the NO USE group for medical patients after 180 days (p<.05) and surgical patients after 90 and 180 days (p<.001). There were no significant differences for any other outcome. Sensitivity analysis indicates the results may be influenced by unmeasured variables, most likely functional status and/or care-giver support. Conclusions: Thirty-six percent of the medical patients referred to PAHC did not receive Medicaid-covered services. This suggests potential underuse among medical patients. The high post-discharge expenditures suggest opportunities for reducing costs through coordinating utilization or diverting it to lower-cost settings. Controlling for patients' need for services, PAHC utilization was not associated with lower utilization rates or lower total Medicaid expenditures. Medicaid programs are advised to proceed cautiously before expanding PAHC utilization and to monitor its use carefully. Further study, incorporating non-economic outcomes and additional factors influencing PAHC use, is warranted. [source] A Novel Interdisciplinary Analgesic Program Reduces Pain and Improves Function in Older Adults After Orthopedic SurgeryJOURNAL OF AMERICAN GERIATRICS SOCIETY, Issue 1 2009R. Sean Morrison MD OBJECTIVES: To examine the effect of a multicomponent intervention on pain and function after orthopedic surgery. DESIGN: Controlled prospective propensity score,matched clinical trial. SETTING: New York City acute rehabilitation hospital. PARTICIPANTS: Two hundred forty-nine patients admitted to rehabilitation after hip fracture repair (n=51) or hip (n=64) or knee (n=134) arthroplasty. INTERVENTION: Pain assessment at rest and with physical therapy (PT) by staff using numeric rating scales (1 to 5). Physician protocols for standing analgesia and preemptive analgesia before PT were implemented on the intervention unit. Control unit patients received usual care. MEASUREMENTS: Pain, analgesic prescribing, gait speed, transfer time, and percentage of PT sessions completed during admission. Pain and difficulty walking at 6, 12, 18, and 24 weeks after discharge. RESULTS: In multivariable analyses intervention patients were significantly more likely than controls to report no or mild pain at rest (66% vs 49%, P=.004) and with PT (52% vs 38%, P=.02) on average for the first 7 days of rehabilitation, had faster 8-foot-walk times on Days 4 (9.3 seconds vs 13.2 seconds, P=.02) and 7 (6.9 vs 9.2 seconds, P=.02), received more analgesia (23.6 vs 15.6 mg of morphine sulfate equivalents per day, P<.001), were more likely to receive standing orders for analgesia (98% vs 48%, P<.001), and had significantly shorter lengths of stay (10.1 vs 11.3 days, P=.005). At 6 months, intervention patients were less likely than controls to report moderate to severe pain with walking (4% vs 15%, P=.02) and that pain did not interfere with walking (7% vs 18%, P=.004) and were less likely to be taking analgesics (35% vs 51%, P=.03). CONCLUSION: The intervention improved postoperative pain, reduced chronic pain, and improved function. [source] Do Palliative Consultations Improve Patient Outcomes?JOURNAL OF AMERICAN GERIATRICS SOCIETY, Issue 4 2008David Casarett MD OBJECTIVES: To determine whether inpatient palliative consultation services improve outcomes of care. DESIGN: Retrospective telephone surveys conducted with family members of veterans who received inpatient or outpatient care from a Department of Veterans Affairs (VA) medical facility in the last month of life. SETTING: Five VA Medical Centers or their affiliated nursing homes and outpatient clinics. PARTICIPANTS: Veterans had received inpatient or outpatient care from a participating VA in the last month of life. One family member completed each survey. MEASUREMENTS: The telephone survey assessed nine aspects of the care the patient received in his or her last month of life: the patient's well-being and dignity (4 items), adequacy of communication (5 items), respect for treatment preferences (2 items), emotional and spiritual support (3 items), management of symptoms (4 items), access to the inpatient facility of choice (1 item), care around the time of death (6 items), access to home care services (4 items), and access to benefits and services after the patient's death (3 items). RESULTS: Interviews were completed with 524 respondents. In a multivariable linear regression model, after adjusting for the likelihood of receiving a palliative consultation (propensity score), palliative care patients had higher overall scores: 65 (95% confidence interval (CI)=62,66) versus 54 (95% CI=51,56; P<.001) and higher scores for almost all domains. Earlier consultations were independently associated with better overall scores (,=0.003; P=.006), a difference that was attributable primarily to improvements in communication and emotional support. CONCLUSION: Palliative consultations improve outcomes of care, and earlier consultations may confer additional benefit. [source] SOME PRACTICAL GUIDANCE FOR THE IMPLEMENTATION OF PROPENSITY SCORE MATCHINGJOURNAL OF ECONOMIC SURVEYS, Issue 1 2008Marco Caliendo Abstract Propensity score matching (PSM) has become a popular approach to estimate causal treatment effects. It is widely applied when evaluating labour market policies, but empirical examples can be found in very diverse fields of study. Once the researcher has decided to use PSM, he is confronted with a lot of questions regarding its implementation. To begin with, a first decision has to be made concerning the estimation of the propensity score. Following that one has to decide which matching algorithm to choose and determine the region of common support. Subsequently, the matching quality has to be assessed and treatment effects and their standard errors have to be estimated. Furthermore, questions like ,what to do if there is choice-based sampling?' or ,when to measure effects?' can be important in empirical studies. Finally, one might also want to test the sensitivity of estimated treatment effects with respect to unobserved heterogeneity or failure of the common support condition. Each implementation step involves a lot of decisions and different approaches can be thought of. The aim of this paper is to discuss these implementation issues and give some guidance to researchers who want to use PSM for evaluation purposes. [source] Hepatic resection compared to percutaneous ethanol injection for small hepatocellular carcinoma using propensity score matchingJOURNAL OF GASTROENTEROLOGY AND HEPATOLOGY, Issue 10 2007Yong Beom Cho Abstract Background:, Several surgical and non-surgical therapeutic modalities have been used for the treatment of hepatocellular carcinoma (HCC). There have been several studies comparing hepatic resection (HR) and percutaneous ethanol injection (PEI) for the treatment of HCC. However, there is still disagreement about the best treatment modality. Methods:, From 130 patients undergoing HR, 116 patients were individually matched to 116 controls from 249 patients undergoing PEI using propensity score matching to overcome possible biases in non-randomized study. Survival analyses were undertaken to compare these propensity score-matched groups. Results:, After matching by propensity score, the major clinical outcomes in the HR (n = 116) and the PEI (n = 116) groups were found to be similar. The 1-, 3- and 5-year overall survival rates were higher in the HR group (94.8%, 76.5% and 65.6%) compared to the PEI group (95.7%, 73.5% and 49.3%) (P = 0.059). The cumulative 1-, 3- and 5-year disease-free survival rates showed the same trend (HR: 76.1%, 50.6% and 40.6%; PEI: 62.6%, 25.5% and 19.1%) (P < 0.001). However, when stratified by Child,Pugh classification, it was no longer the case in the Child B patients. Single intrahepatic recurrence was the most common pattern of tumor recurrence after both treatments. Conclusions:, Patients undergoing HR had a better survival profile than those undergoing PEI. However, when considering which technique to use for optimal HCC management, the individual patient's hepatic function must be considered. [source] Another look into the effect of premarital cohabitation on duration of marriage: an approach based on matchingJOURNAL OF THE ROYAL STATISTICAL SOCIETY: SERIES A (STATISTICS IN SOCIETY), Issue 1 2009Stefano Mazzuco Summary., The paper proposes an alternative approach to studying the effect of premarital cohabitation on subsequent duration of marriage on the basis of a strong ignorability assumption. The approach is called propensity score matching and consists of computing survival functions conditional on a function of observed variables (the propensity score), thus eliminating any selection that is derived from these variables. In this way, it is possible to identify a time varying effect of cohabitation without making any assumption either regarding its shape or the functional form of covariate effects. The output of the matching method is the difference between the survival functions of treated and untreated individuals at each time point. Results show that the cohabitation effect on duration of marriage is indeed time varying, being close to zero for the first 2,3 years and rising considerably in the following years. [source] Bronchopulmonary dysplasia and brain white matter damage in the preterm infant: a complex relationshipPAEDIATRIC & PERINATAL EPIDEMIOLOGY, Issue 6 2009Luigi Gagliardi Summary We analysed the relationship between bronchopulmonary dysplasia (BPD) and brain white matter damage (WMD) in very preterm infants, adjusting for common risk factors and confounders. We studied a cohort of infants <32 weeks gestational age (GA) and <1500 g, admitted to 12 hospitals in Northern Italy in 1999,2002. The association between BPD and WMD was estimated by generalised estimating equations and conditional logistic models, adjusting for centre, GA, propensity score for prolonged ventilation and other potential confounders. Directed acyclic graphs (DAG) were used to depict the underlying causal structure and guide analysis. Of the 1209 infants reaching 36 weeks, 192 (15.8%) developed BPD (supplemental oxygen at 36 weeks) and 88 (7.3%) ultrasound-defined WMD (cystic periventricular leukomalacia). In crude analysis, BPD was a strong risk factor for WMD [odds ratio (OR) = 5.9]. With successive adjustments, the OR progressively decreased to 3.88 when adjusting for GA, to 2.72 adding perinatal risk factors, and further down to 2.16 [95% confidence interval 1.1, 3.9] when ventilation was also adjusted for. Postnatal factors did not change the OR. Significant risk factors for WMD, in addition to BPD, were a low GA, a lower Apgar score, a higher illness severity score, ventilation and early-onset sepsis, while antenatal steroids, being small for GA, and surfactant were associated with a reduced risk. In conclusion, our data suggest that BPD is associated with an increased risk of WMD; most of the effect is due to shared risk factors and causal pathways. DAGs helped clarify the complex confounding of this scenario. [source] Corticosteroids for acute chest syndrome in children with sickle cell disease: Variation in use and association with length of stay and readmission,AMERICAN JOURNAL OF HEMATOLOGY, Issue 1 2010Amy Sobota Acute chest syndrome (ACS) causes significant morbidity and mortality in sickle cell disease. The role of corticosteroids is unclear. The objectives of our study were to examine the variation between hospitals in their use of corticosteroids for ACS, describe characteristics associated with corticosteroids, and investigate the association between corticosteroids, length of stay, and readmission. We performed a retrospective examination of 5,247 hospitalizations for ACS between January 1, 2004, and June 30, 2008, at 32 hospitals in the Pediatric Health Information System database. We used multivariate regression to examine the variability in the use of corticosteroids adjusting for hospital case mix, identify factors associated with corticosteroid use, and evaluate the association of corticosteroids with length of stay and 3-day readmission rates controlling for propensity score. Corticosteroid use varied greatly by hospital (10,86% among all patients, 18,92% in patients with asthma). Treatment with corticosteroids was associated with comorbid asthma (OR 3.9, 95% CI: 3.2,4.8), inhaled steroids (OR 1.4, 95% CI: 1.1,1.7), bronchodilators (OR 3.2, 95% CI: 2.5,4.2), nitric oxide (OR 2.4, 95% CI: 1.2,5.0), oxygen (OR 2.3, 95% CI: 1.8,2.9), ICU (OR 1.7, 95% CI: 1.3,2.3), ventilation (OR 2.0, 95% CI: 1.4,2.8), APR-DRG severity level (OR 1.4, 95% CI: 1.2,1.6), and discharge year (OR 0.86, 95% CI: 0.80,0.92). Corticosteroids were associated with an increased length of stay (25%, 95% CI: 14,38%) and a higher 3-day readmission rate (OR 2.3, 95% CI: 1.6,3.4), adjusted for confounding. Hospitals vary greatly in the use of corticosteroids for ACS, even in patients with asthma. Clear evidence of the efficacy and toxicity of corticosteroid treatment in ACS may reduce variation in care. Am. J. Hematol. 2010. © 2009 Wiley-Liss, Inc. [source] Asthma and allergy medication use and costs among pediatric primary care patients on asthma controller therapyPEDIATRIC ALLERGY AND IMMUNOLOGY, Issue 8 2006Vasilisa Sazonov-Kocevar As observational studies in children initiating GINA-Step 3 therapies are scarce, we evaluated outcomes and costs in a primary care cohort. Two-yr retrospective cohort study included French children (age: 6,14) continuously followed in BKL-Thalès database who received ,2 consecutive prescriptions for GINA-Step 3 therapy (=addition of montelukast or other controllers (,other'), such as increasing inhaled-corticosteroid dose (hICS), adding long-acting , agonist (LABA), or ICS + LABA). After matching on gender and propensity score, medication use [rescue (short-acting , agonists), acute (antibiotics (AB), oral corticosteroids (OCS)), allergy (antihistamines, nasal steroids) and other respiratory] was estimated via mean number of prescriptions and mean cost (per child/per month), and cost trends. During 12-month follow-up, children adding montelukast (n = 71) vs. ,other' (n = 213) had similar asthma rescue/acute and allergy medication use. Subgroup with asthma and allergic rhinitis (A + AR) adding montelukast used less OCS and AB (p = 0.014). Two-yr cost trends suggest stable asthma/allergy medication use in montelukast group (,0.83) compared with increase in ,other' (,5.39), which was driven by nasal steroid use [,0.32 (,other') vs. ,,0.04 (montelukast), p = 0.0013]. In subgroup with A + AR decline in asthma/allergy medication use in montelukast group (,,0.47) vs. increase in ,other' (,11.05), p = 0.015, was driven by differences in AB and OCS (p = 0.04) and nasal steroid use (p = 0.001). Concomitant asthma/allergy medication use was similar in children adding montelukast or ,other' controllers (hICS, LABA, ICS + LABA), while children with allergic rhinitis on montelukast used less AB. Concomitant medication costs after addition of montelukast remained stable, while ,other' group experienced increase, especially in children with concomitant allergic rhinitis. [source] Semi-automated risk estimation using large databases: quinolones and clostridium difficile associated diarrhea,PHARMACOEPIDEMIOLOGY AND DRUG SAFETY, Issue 6 2010Robertino M. Mera Abstract Purpose The availability of large databases with person time information and appropriate statistical methods allow for relatively rapid pharmacovigilance analyses. A semi-automated method was used to investigate the effect of fluoroquinolones on the incidence of C. difficile associated diarrhea (CDAD). Methods Two US databases, an electronic medical record (EMR) and a large medical claims database for the period 2006,2007 were evaluated using a semi-automated methodology. The raw EMR and claims datasets were subject to a normalization procedure that aligns the drug exposures and conditions using ontologies; Snowmed for medications and MedDRA for conditions. A retrospective cohort design was used together with matching by means of the propensity score. The association between exposure and outcome was evaluated using a Poisson regression model after taking into account potential confounders. Results A comparison between quinolones as the target cohort and macrolides as the comparison cohort produced a total of 564,797 subjects exposed to a quinolone in the claims data and 233,090 subjects in the EMR. They were matched with replacement within six strata of the propensity score. Among the matched cohorts there were a total of 488 and 158 outcomes in the claims and the EMR respectively. Quinolones were found to be twice more likely to be significantly associated with CDAD than macrolides adjusting for risk factors (IRR 2.75, 95%CI 2.18,3.48). Conclusions Use of a semi-automated method was successfully applied to two observational databases and was able to rapidly identify a potential for increased risk of developing CDAD with quinolones. Copyright © 2010 John Wiley & Sons, Ltd. [source] Confounder summary scores when comparing the effects of multiple drug exposures,PHARMACOEPIDEMIOLOGY AND DRUG SAFETY, Issue 1 2010Suzanne M. Cadarette PhD Abstract Purpose Little information is available comparing methods to adjust for confounding when considering multiple drug exposures. We compared three analytic strategies to control for confounding based on measured variables: conventional multivariable, exposure propensity score (EPS), and disease risk score (DRS). Methods Each method was applied to a dataset (2000,2006) recently used to examine the comparative effectiveness of four drugs. The relative effectiveness of risedronate, nasal calcitonin, and raloxifene in preventing non-vertebral fracture, were each compared to alendronate. EPSs were derived both by using multinomial logistic regression (single model EPS) and by three separate logistic regression models (separate model EPS). DRSs were derived and event rates compared using Cox proportional hazard models. DRSs derived among the entire cohort (full cohort DRS) was compared to DRSs derived only among the referent alendronate (unexposed cohort DRS). Results Less than 8% deviation from the base estimate (conventional multivariable) was observed applying single model EPS, separate model EPS or full cohort DRS. Applying the unexposed cohort DRS when background risk for fracture differed between comparison drug exposure cohorts resulted in ,7 to,+,13% deviation from our base estimate. Conclusions With sufficient numbers of exposed and outcomes, either conventional multivariable, EPS or full cohort DRS may be used to adjust for confounding to compare the effects of multiple drug exposures. However, our data also suggest that unexposed cohort DRS may be problematic when background risks differ between referent and exposed groups. Further empirical and simulation studies will help to clarify the generalizability of our findings. Copyright © 2009 John Wiley & Sons, Ltd. [source] Relationship between thiazolidinedione use and cardiovascular outcomes and all-cause mortality among patients with diabetes: a time-updated propensity analysis,PHARMACOEPIDEMIOLOGY AND DRUG SAFETY, Issue 6 2009Zeina A. Habib MD Abstract Purpose To investigate the association of the thiazolidinediones (TZDs), rosiglitazone, and pioglitazone, together and individually on the risk of cardiovascular outcomes and all-cause mortality, using time-updated propensity score adjusted analysis. Methods We conducted a retrospective cohort study in a large vertically integrated health system in southeast Michigan. Cohort inclusion criteria included adult patients with diabetes treated with oral medications and followed longitudinally within the health system between 1 January 2000 and 1 December 2006. The primary outcome was fatal and non-fatal acute myocardial infarction (AMI). Secondary outcomes included hospitalizations for congestive heart failure (CHF), fatal, and non-fatal cerebrovascular accidents (CVA) and transient ischemic attacks (TIA), combined coronary heart disease (CHD) events, and all-cause mortality. Results 19,171 patients were included in this study. Use of TZDs (adjusted hazard ratio (aHR) with propensity adjustment (PA), 0.92; 95% confidence interval (CI) 0.73,1.17), rosiglitazone (aHR with PA, 1.06; 95%CI 0.66,1.70), and pioglitazone (aHR with PA, 0.91; 95%CI 0.69,1.21) was not associated with a higher risk of AMI. However, pioglitazone use was associated with a reduction in all-cause mortality (aHR with PA, 0.60; 95%CI 0.42,0.96). Compared with rosiglitazone, pioglitazone use was associated with a lower risk of all outcomes assessed, particularly CHF (p,=,0.013) and combined CHD events (p,=,0.048). Conclusions Our findings suggest that pioglitazone may have a more favorable risk profile when compared to rosiglitazone, arguing against a singular effect for TZDs on cardiovascular outcomes. Copyright © 2009 John Wiley & Sons, Ltd. [source] Evaluating uses of data mining techniques in propensity score estimation: a simulation study,PHARMACOEPIDEMIOLOGY AND DRUG SAFETY, Issue 6 2008DrPH, Soko Setoguchi MD Abstract Background In propensity score modeling, it is a standard practice to optimize the prediction of exposure status based on the covariate information. In a simulation study, we examined in what situations analyses based on various types of exposure propensity score (EPS) models using data mining techniques such as recursive partitioning (RP) and neural networks (NN) produce unbiased and/or efficient results. Method We simulated data for a hypothetical cohort study (n,=,2000) with a binary exposure/outcome and 10 binary/continuous covariates with seven scenarios differing by non-linear and/or non-additive associations between exposure and covariates. EPS models used logistic regression (LR) (all possible main effects), RP1 (without pruning), RP2 (with pruning), and NN. We calculated c-statistics (C), standard errors (SE), and bias of exposure-effect estimates from outcome models for the PS-matched dataset. Results Data mining techniques yielded higher C than LR (mean: NN, 0.86; RPI, 0.79; RP2, 0.72; and LR, 0.76). SE tended to be greater in models with higher C. Overall bias was small for each strategy, although NN estimates tended to be the least biased. C was not correlated with the magnitude of bias (correlation coefficient [COR],=,,0.3, p,=,0.1) but increased SE (COR,=,0.7, p,<,0.001). Conclusions Effect estimates from EPS models by simple LR were generally robust. NN models generally provided the least numerically biased estimates. C was not associated with the magnitude of bias but was with the increased SE. Copyright © 2008 John Wiley & Sons, Ltd. [source] Insights into different results from different causal contrasts in the presence of effect-measure modification,PHARMACOEPIDEMIOLOGY AND DRUG SAFETY, Issue 10 2006Til Stürmer Abstract Purpose Both propensity score (PS) matching and inverse probability of treatment weighting (IPTW) allow causal contrasts, albeit different ones. In the presence of effect-measure modification, different analytic approaches produce different summary estimates. Methods We present a spreadsheet example that assumes a dichotomous exposure, covariate, and outcome. The covariate can be a confounder or not and a modifier of the relative risk (RR) or not. Based on expected cell counts, we calculate RR estimates using five summary estimators: Mantel-Haenszel (MH), maximum likelihood (ML), the standardized mortality ratio (SMR), PS matching, and a common implementation of IPTW. Results Without effect-measure modification, all approaches produce identical results. In the presence of effect-measure modification and regardless of the presence of confounding, results from the SMR and PS are identical, but IPTW can produce strikingly different results (e.g., RR,=,0.83 vs. RR,=,1.50). In such settings, MH and ML do not estimate a population parameter and results for those measures fall between PS and IPTW. Conclusions Discrepancies between PS and IPTW reflect different weighting of stratum-specific effect estimates. SMR and PS matching assign weights according to the distribution of the effect-measure modifier in the exposed subpopulation, whereas IPTW assigns weights according to the distribution of the entire study population. In pharmacoepidemiology, contraindications to treatment that also modify the effect might be prevalent in the population, but would be rare among the exposed. In such settings, estimating the effect of exposure in the exposed rather than the whole population is preferable. Copyright © 2006 John Wiley & Sons, Ltd. [source] Weaknesses of goodness-of-fit tests for evaluating propensity score models: the case of the omitted confounder,PHARMACOEPIDEMIOLOGY AND DRUG SAFETY, Issue 4 2005Sherry Weitzen PhD Abstract Purpose Propensity scores are used in observational studies to adjust for confounding, although they do not provide control for confounders omitted from the propensity score model. We sought to determine if tests used to evaluate logistic model fit and discrimination would be helpful in detecting the omission of an important confounder in the propensity score. Methods Using simulated data, we estimated propensity scores under two scenarios: (1) including all confounders and (2) omitting the binary confounder. We compared the propensity score model fit and discrimination under each scenario, using the Hosmer,Lemeshow goodness-of-fit (GOF) test and the c-statistic. We measured residual confounding in treatment effect estimates adjusted by the propensity score omitting the confounder. Results The GOF statistic and discrimination of propensity score models were the same for models excluding an important predictor of treatment compared to the full propensity score model. The GOF test failed to detect poor model fit for the propensity score model omitting the confounder. C-statistics under both scenarios were similar. Residual confounding was observed from using the propensity score excluding the confounder (range: 1,30%). Conclusions Omission of important confounders from the propensity score leads to residual confounding in estimates of treatment effect. However, tests of GOF and discrimination do not provide information to detect missing confounders in propensity score models. Our findings suggest that it may not be necessary to compute GOF statistics or model discrimination when developing propensity score models. Copyright © 2004 John Wiley & Sons, Ltd. [source] | ||||||||||||||||||||||||||||