SEARCH · PubMed Health

Results for “Propensity Score”

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

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

At least 379 records · Page 21Linked to original sources

PubMed2004

The effect of practice budgets on patient waiting times: allowing for selection bias.

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.

Budgets↗

PubMed2006

Effectiveness of ceftriaxone plus doxycycline in the treatment of patients hospitalized with community-acquired pneumonia.

BACKGROUND: Limited data exist on the effectiveness of ceftriaxone plus doxycycline in the treatment of patients hospitalized with community-acquired pneumonia (CAP). METHODS: We performed a retrospective cohort study of all adults hospitalized for pneumonia between January 1999 and July 2001 at an academic medical center. Outcomes were compared for patients with CAP treated with ceftriaxone plus doxycycline versus other appropriate initial empiric antibiotic therapies. Outcomes were adjusted with the use of a propensity score to account for differences in patient characteristics and illness severity between groups. RESULTS: A total of 216 patients were treated with ceftriaxone plus doxycycline and 125 received other appropriate initial empiric antibiotic therapies. After adjustment, use of ceftriaxone plus doxycycline was associated with reduced inpatient mortality (OR = 0.26, 95% CI: 0.08-0.81) and 30-day mortality (OR = 0.37, 95% CI: 0.17-0.81), but not with length of stay or readmission rates. Analysis of a subset of the sample that excluded patients admitted from nursing homes, patients admitted to the ICU, and patients diagnosed with aspiration also showed reduced inpatient mortality with the use of ceftriaxone plus doxycycline. CONCLUSIONS: The use of ceftriaxone plus doxycycline as an initial empiric therapy for patients hospitalized with CAP appears safe and effective, and its potential superiority should be evaluated prospectively.

Adult↗

PubMed2006

Sensitivity analysis and external adjustment for unmeasured confounders in epidemiologic database studies of therapeutics.

BACKGROUND: Large health care utilization databases are frequently used to analyze unintended effects of prescription drugs and biologics. Confounders that require detailed information on clinical parameters, lifestyle, or over-the-counter medications are often not measured in such datasets, causing residual confounding bias. OBJECTIVE: This paper provides a systematic approach to sensitivity analyses to investigate the impact of residual confounding in pharmacoepidemiologic studies that use health care utilization databases. METHODS: Four basic approaches to sensitivity analysis were identified: (1) sensitivity analyses based on an array of informed assumptions; (2) analyses to identify the strength of residual confounding that would be necessary to explain an observed drug-outcome association; (3) external adjustment of a drug-outcome association given additional information on single binary confounders from survey data using algebraic solutions; (4) external adjustment considering the joint distribution of multiple confounders of any distribution from external sources of information using propensity score calibration. CONCLUSION: Sensitivity analyses and external adjustments can improve our understanding of the effects of drugs and biologics in epidemiologic database studies. With the availability of easy-to-apply techniques, sensitivity analyses should be used more frequently, substituting qualitative discussions of residual confounding.

Anti-Inflammatory Agents, Non-Steroidal↗

PubMed2006

Insights into different results from different causal contrasts in the presence of effect-measure modification.

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.

Bias↗

PubMedDate not supplied

Approaches to combat with confounding by indication in observational studies of intended drug effects.

There has been a resurgence of controversy about the usefulness of observational data to study the efficacy of drugs. Nearly every week a researcher makes some criticism of clinical trials or justifies observational research into intended effects, with other researchers offering a contradictory viewpoint. Literature reviews are not useful in this regard because the contradictory studies will not usually be carried out. Some methods are discussed which may have potential utility in the study of intended effects. There may be a marginal role for statistical techniques such as propensity scores and confounder scores. More promising techniques may include ecological analyses, restriction of subjects and blinded prospective review. Because it is currently unknown when the observational study of drug efficacy is possible, we should arguably always carry out a study of the determinants of prescribing first, and possibly consider using the various techniques that are outlined in this article.

Confounding Factors, Epidemiologic↗

PubMed2002

Using inverse weighting and predictive inference to estimate the effects of time-varying treatments on the discrete-time hazard.

We estimate the effects of non-randomized time-varying treatments on the discrete-time hazard, using inverse weighting. We consider the special monotone pattern of treatment that develops over time as subjects permanently discontinue an initial treatment, and assume that treatment selection is sequentially ignorable. We use a propensity score in the hazard model to reduce the potential for finite-sample bias due to inverse weighting. When the number of subjects who discontinue treatment at any given time is small, we impose scientific restrictions on the potentially observable discontinuation hazards to improve efficiency. We use predictive inference to account for the correlation of the potential hazards, when comparing outcomes under different durations of initial treatment.

Antidepressive Agents↗

PubMed2003

An integrated population-averaged approach to the design, analysis and sample size determination of cluster-unit trials.

While the mixed model approach to cluster randomization trials is relatively well developed, there has been less attention given to the design and analysis of population-averaged models for randomized and non-randomized cluster trials. We provide novel implementations of familiar methods to meet these needs. A design strategy that selects matching control communities based upon propensity scores, a statistical analysis plan for dichotomous outcomes based upon generalized estimating equations (GEE) with a design-based working correlation matrix, and new sample size formulae are applied to a large non-randomized study to reduce underage drinking. The statistical power calculations, based upon Wald tests for summary statistics, are special cases of a general power method for GEE.

Adolescent↗

PubMed2007

Sensitivity of the hazard ratio to nonignorable treatment assignment in an observational study.

In non-randomized studies, estimation of treatment effects generally requires adjustment for imbalances in observed covariates. One such method, based on the propensity score, is useful in many applications but may be biased when the assumption of strongly ignorable treatment assignment is violated. Because it is not possible to evaluate this assumption from the data, it is advisable to assess the sensitivity of conclusions to violations of strong ignorability. Lin et al. (Biomet. 1998; 54:948-963) have implemented this idea by investigating how an unmeasured covariate may affect the conclusions of an observational study. We extend their method to assess sensitivity of the treatment hazard ratio to hidden bias under a range of covariate distributions. We derive simple formulas for approximating the true from the apparent treatment hazard ratio estimated under a specific survival model, and assess the validity of these formulas in simulation studies. We demonstrate the method in an analysis of SEER-Medicare data on the effects of chemotherapy in elderly colon cancer patients.

Aged↗

PubMed2000

Changes in injection risk behavior associated with participation in the Seattle needle-exchange program.

We studied the influence of the Seattle, Washington, needle-exchange program on sharing of drug injection equipment (syringes, drug cookers, filtration cotton) to identify potential gaps in risk reduction and to understand in greater detail the lack of an association between exchange use and risk of hepatitis B or C virus transmission. In a cohort of 2,208 injection drug users who completed a 1-year follow-up visit, we measured the association between needle-exchange use at study enrollment (ever vs. never) and injection risk behavior at the follow-up. Control for confounding was carried out using both logistic regression and propensity score analytic methods to estimate the adjusted odds ratio (AOR). In both univariate and multivariate analyses, needle-exchange use was associated with a lower likelihood of injection with a used syringe (AOR = 0.7, 95% confidence limit 0.5, 0.9). There was no association between exchange use and cooker or cotton sharing (AOR = 0.8, 95% confidence limit 0.6, 1.1) or between exchange use and use of a common syringe to divide drugs (AOR = 0.9). This analysis suggests that risk reduction measures adopted by users of the Seattle exchange may not be sufficient to prevent transmission of all blood-borne viruses, including hepatitis C virus. Greater awareness of the infection risk associated with these practices may help curb this type of equipment sharing and ultimately prevent disease transmission.

Adult↗

PubMed2005

Erythropoietin is an effective clinical modality for reducing RBC transfusion in joint surgery.

PURPOSE: Efficacy trials of preoperative erythropoietin therapy (PET) recommend a dosing schedule that cannot always be adhered to in everyday clinical practice. Consequently, we instituted a flexible dosing schedule and routinely offered it to anemic patients [hemoglobin (Hb)] < or = 130 g x L(-1)) undergoing total joint arthroplasty (TJA). The purpose of this observational, cohort study was to assess the effectiveness of this practice in reducing red blood cell (RBC) transfusion. METHODS: After obtaining Institutional Ethics Board approval, data were collected prospectively on all patients who underwent TJA at our institution from July 1999 to June 2003. Patients with baseline Hb < or = 130 g x L(-1) were offered PET as follows: one to three sc injections (20,000 IU for those < or = 70 kg, and 40,000 IU for those > 70 kg) every three to seven days before surgery. Since treatment was not randomly assigned, multivariable logistic regression analysis and propensity score case-control matching were used to adjust for baseline differences in patient demographics and perioperative risk factors for RBC transfusion. The adjusted relationship between PET and RBC transfusion was then determined. RESULTS: Of the 1,782 patients that underwent TJA during the study period, 770 (47.9%) had a Hb < 130 g x L(-1). Of these patients, 214 received PET and their RBC transfusion rate was 16.4%, whereas the transfusion rate was 56.1% in those who did not receive PET (P < 0.0001). The adjusted odds ratio of RBC transfusion with PET was 0.33 (95% confidence interval = 0.21-0.49). CONCLUSION: PET, used as part of routine clinical practice, is an effective blood conservation modality.

Adult↗

PubMed2003

Change in patients' body weight after 12 months of treatment with glimepiride or glibenclamide in Type 2 diabetes: a multicentre retrospective cohort study.

AIMS/HYPOTHESIS: Our study compared the effects of glimepiride or glibenclamide treatment on body weight over 12 months of treatment in patients with Type 2 diabetes in routine outpatient practice. METHODS: This new retrospective study design used data from physicians in a restricted manner (retrolective). Data from case report forms from 520 patients from 91 randomly selected centres were assessed and covariance analysis performed. RESULTS: The influence of practice and patient characteristics on treatment assignment was low, reflecting the design of randomised controlled trials. Mean weight loss and reduction in body mass index from baseline to study endpoint were greater with glimepiride than with glibenclamide (-2.04+/-3.99 kg vs -0.58+/-3.65 kg, p<0.001; -0.71+/-1.38 kg/m(2) vs -0.20+/-1.28 kg/m(2), p<0.001). Duration of treatment at baseline influenced treatment outcome, but propensity score, sex, age and fasting blood glucose at baseline did not. Both glimepiride and glibenclamide led to decreases in fasting blood glucose (-2.43+/-0.24 mmol/l vs -3.03+/-0.24 mmol/l; p<0.001 vs baseline) and HbA(1c) (-1.23+/-0.09% vs -1.26+/-0.09%; p<0.001 vs baseline). Both treatments were associated with a decrease in serum total cholesterol and low density lipoprotein cholesterol. Triglycerides were lower in the glibenclamide group and high density lipoprotein cholesterol was higher in the glimepiride group only. CONCLUSIONS/INTERPRETATION: Initial treatment of Type 2 diabetes with glimepiride was associated with a significantly greater decrease in body weight and body mass index than treatment with glibenclamide, while providing equivalent glycaemic control.

Blood Glucose↗

PubMed2006

[Risk stratification of patients with diabetes mellitus undergoing coronary artery bypass grafting--a comparison of statistical methods].

Among the coronary bypass patients from our Datamart database, we found a prevalence of 29.6% of diagnosed diabetics. 5.2% of the patients without a diagnosis of diabetes mellitus and a fasting plasma glucose level > 125 mg/dl were defined as undiagnosed diabetics. The objective of this paper was to compare univariate methods and techniques for risk stratification to determine, whether undiagnosed diabetes is per se a risk factor for increased ventilation time and length of ICU stay, and for increased prevalence of resuscitation, reintubation and 30-d mortality for diabetics in heart surgery. Univariate comparisons reveals that undiagnosed diabetics needed resuscitation significantly more often and had an increased ventilation time, while the length of ICU stay was significantly reduced. The significantly different distribution between the diabetics groups of 11 from 32 attributes examined, demands the use of methods for risk stratification. Both risk adjusted methods regression and matching confirm that undiagnosed diabetics had an increased ventilation time and an increased prevalence of resuscitation, while the length of ICU stay was not significantly reduced. A homogeneous distribution of the patient characteristics in the two diabetics groups could be achieved through a statistical matching method using the propensity score. In contrast to the regression analysis, a significantly increased prevalence of reintubation in undiagnosed diabetics was found. Based on an example of undiagnosed diabetics in heart surgery, the presented study reveals the necessity and the possibilities of techniques for risk stratification in retrospective analysis and shows how the potential of data collection from daily clinical practice can be used in an effective way.

Coronary Artery Bypass↗

PubMed2000

Hemispheric stroke following cardiac surgery: a case-control estimate of the risk resulting from ipsilateral asymptomatic carotid artery stenosis.

A case-control study was undertaken to determine if asymptomatic carotid artery stenosis (ACS) is independently associated with ipsilateral hemispheric stroke following cardiac surgery (CS). All CS patients (3069) who were at two hospitals between 1989 and 1994 were reviewed. Cases (31) selected for this study were those with hemispheric stroke within 30 days following CS. Controls (69) were taken from those without hemispheric stroke. Case-control analysis demonstrated that ACS of 50-90% and of 80-90% increased the risk of ipsilateral stroke 5.2-fold (95% confidence interval [CI] = 1.5-16.3, p = 0.01) and 24.3-fold (CI = 2.6-114.9, p = 0.002), respectively. Other variables with significant odds ratios (OR) were age > or =65 years (OR = 4.0, CI = 1.3-10.5, p = 0.01), peripheral vascular disease (OR = 3.4, CI = 1.3-8.8, p = 0.02), hypertension (OR = 3.0, CI = 1.2-7.0, p = 0.02), and female gender (OR = 3.0, CI = 1.2-7.1, p = 0.04). A second conservative analysis for missing data demonstrated a significant association for ACS of 80-90% alone (OR = 13.1, CI = 1.5-60.9, p = 0.01). This association remained significant after multivariate adjustment with propensity score stratification. ACS (80-90%) appears to be independently associated with ipsilateral hemispheric stroke following CS when evaluated against the present study variables. This finding supports the need for a properly conducted prospective natural history study, including an evaluation of aortic arch atherosclerosis, to determine the clinical relevance of this observation.

Aged↗

PubMed2026

Gout and allopurinol adherence in newly diagnosed patients and the risk of fractures: a nationwide cohort study.

INTRODUCTION / OBJECTIVES: The association between gout and fractures remains controversial due to the competing effects of uric acid's antioxidant properties and gout-induced chronic inflammation. Our study aimed to evaluate the risk of fractures in newly diagnosed gout patients and to analyze the impact of Allopurinol medication adherence on this risk. METHODS: This nationwide cohort study utilized the National Health Insurance Service-National Health Screening Cohort (NHIS-HEALS) database (2002-2019). We identified 4,107 patients newly diagnosed with gout who remained on continuous allopurinol therapy during the 24-month medication assessment period and matched them 1:3 with 12,321 non-gout controls using propensity score matching. Allopurinol adherence was assessed via the Medication Possession Ratio (MPR) over a 24-month period and categorized into three groups: MPR < 0.3, 0.3 ≤ MPR < 0.8, and MPR ≥ 0.8. Multivariable Cox proportional hazards regression was used to calculate adjusted hazard ratios (aHR) and 95% confidence intervals (CI) for fractures (vertebral, hip, and distal radius). RESULTS: Gout patients demonstrated a significantly higher risk of overall fractures compared to the non-gout group (aHR 5.52; 95% CI 4.19-7.26). A significant inverse linear relationship was observed between allopurinol adherence and fracture risk (P for trend < .001). The risk was highest in the low-adherence group (MPR < 0.3; aHR 5.91; 95% CI 4.40-7.93) and relatively lower in the high-adherence group (MPR ≥ 0.8; aHR 4.77; 95% CI 2.94-7.72). Consistent trends were observed for vertebral (aHR 5.68) and hip (aHR 4.22) fractures. These associations remained consistent across all subgroups, including age, sex, and comorbidities. CONCLUSION: Newly diagnosed gout is associated with a substantially increased risk of fractures. Higher adherence to Allopurinol therapy is correlated with a significant reduction in this risk, suggesting that consistent urate-lowering therapy may mitigate gout-related bone fragility. Key Points • Newly diagnosed gout patients have a significantly higher risk of major osteoporotic fractures compared to the non-gout population. • A significant inverse linear relationship exists between Allopurinol adherence and fracture risk, with the highest adherence group (MPR ≥ 0.8) showing a relatively lower risk. • Consistent urate-lowering therapy (ULT) may mitigate bone fragility in gout patients by reducing systemic inflammatory burden caused by urate crystal deposition.

Humans↗

PubMed2006

Comparing costs and effectiveness of different starting points for highly active antiretroviral therapy in HIV-positive patients. Evidence from the ICONA cohort.

We evaluated the costs and effectiveness of starting highly active antiretroviral therapy (HAART) at different points during the course of HIV infection, defined on the basis of CD4 T-lymphocytes counts. The study considered 3,250 HAART-naive patients of the Italian Cohort Naive Antiretrovirals (ICONA), enrolled and followed between 1997 and 2002. In correspondence to the thresholds of 500, 350, and 200 CD4 cells/mm(3), we selected immediate and deferred groups accounting for lead-time bias. The effects of immediate vs. deferred treatment on AIDS-free survival and direct health costs were estimated stratifying on the propensity score of immediate HAART initiation. The incremental cost-effectiveness ratio (ICER) and the cost-effectiveness acceptability curve were also obtained. Although immediate HAART initiation did not affect incidence AIDS and death at high CD4 levels, starting HAART with 200-349 CD4 cells/mm(3) rather than deferring it below 200 CD4 cells/mm(3), proved to be cost-effective.

Antiretroviral Therapy, Highly Active↗

PubMed2006

At-risk high school students in the "Gaining Early Awareness and Readiness" Program (GEAR UP): academic and behavioral outcomes.

The purpose of this study was to assess the effects of a GEAR UP intervention with at-risk high school students at a large urban high school in Florida. The goals of the GEAR UP program were to improve academic performance, decrease behavior-related problems, and reduce truancy and absenteeism. The study design consisted of a three-group comparison of the 447 students in GEAR UP: the No Participation Group, the Low Participation Group, and the High Participation Group. Participation levels were calculated for each category of activity (academic, behavior-related, and social), and propensity scoring was used to match the groups on sociodemographic characteristics and other differentiating variables. Results indicated that race (i.e., African American) and sex (i.e., female) are associated with high participation in program activities. Also, students who spent a substantial amount of time on academic activities improved their GPAs over a semester, and students who took advantage of behavior-related services and participated in social activities significantly reduced disciplinary referrals (p < .05). Suggestions for program refinement resulting from the study are discussed.

Absenteeism↗

PubMed2006

Medicaid coverage and access to publicly funded opiate treatment.

This observational study examines changes in access to methadone maintenance treatment following Oregon's decision to remove substance abuse treatment from the Medicaid benefit for an expansion population. Access was compared before and after the benefit change for two cohorts of adults addicted to opiates presenting for publicly funded treatment. Propensity score analysis helped model some selective disenrollment from Medicaid that occurred after the benefit change. Logistic regression was used to compare access to methadone by cohort controlling for client characteristics. Opiate users presenting for publicly funded treatment after the change were less than half as likely (OR = 0.40) to be placed in an opiate treatment program compared to the prior year. Further analysis revealed that those with no recent treatment history were less likely to present for treatment after the benefit change. These results have implications for states considering Medicaid cuts, especially if the anticipated increases in illegal activity, emergency room utilization, unemployment, and mortality can be demonstrated.

Adolescent↗

PubMed1994

Surgical treatment of endometriosis-associated infertility: meta-analysis compared with survival analysis.

OBJECTIVE: Our purpose was to evaluate the role of surgery in the treatment of endometriosis associated with infertility. STUDY DESIGN: We used a prospective cohort analysis of pregnancy rates and variables affecting pregnancy rates for surgical, medical, and no treatment. Our studies were combined with those reported by Hughes et al. (Fertil Steril 1993; 59:963-70), and the meta-analysis was expanded to include additional comparisons. Treatment was performed by a single surgeon in a referral reproductive endocrinology and surgery private practice. Results from 579 women with endometriosis and infertility in our study and the meta-analysis of 25 studies by Hughes et al. were examined. Interventions consisted of no treatment, medical treatment, or surgical treatment by laparoscopy or laparotomy. The main outcome measure was pregnancy rates. RESULTS: For minimal and mild disease, no treatment, laparoscopy, and laparotomy had equivalent 3-year estimated cumulative life-table pregnancy rates (67% +/- 12%, 68% +/- 4%, and 74% +/- 8%, respectively) that were higher than medical treatment pregnancy rates (Breslow p = 0.003). For moderate and severe disease, all but 11 patients were treated surgically. The 3-year estimated cumulative life-table pregnancy rates were 62% + 6% [corrected] for 120 laparoscopy cases and 44% + 6% [corrected] for 102 laparotomy cases (Breslow p = 0.054). For endometriomas, 48 laparoscopy patients had a 3-year estimated cumulative life-table pregnancy rate of 52% +/- 9% and 52 laparotomy patients had a 3-year estimated cumulative life-table pregnancy rate of 46% +/- 9% (Breslow p = 0.48). For 28 patients with complete cul-de-sac obliteration, the 3-year estimated cumulative life-table pregnancy rates were 30% +/- 14% after laparoscopy and 24% +/- 12% after laparotomy (Breslow p = 0.084). Comparison of our results with the expanded meta-analysis revealed deficiencies in the design of meta-analysis studies and the impact of our using life-table pregnancy rates controlled for factors influencing outcome (survival analysis with fixed covariates) rather than the simple pregnancy rates used in the meta-analysis. Benefits of sophisticated statistical techniques, including propensity scores, to adjust for noncomparability of groups in prospective cohort studies were identified. CONCLUSION: Both our study and the meta-analysis show that either no treatment or surgery is superior to medical treatment for minimal and mild endometriosis associated with infertility. For moderate and severe disease, surgery is usually used. In these patients experienced surgeons utilizing good clinical judgment can achieve results at operative laparoscopy at least equivalent to those at laparotomy, even in cases involving endometriomas and complete cul-de-sac obliteration. Prospective randomized trials should be performed to confirm these findings.

Cohort Studies↗