Propensity score weighting for causal subgroup analysis.
Ontology highlight
ABSTRACT:
SUBMITTER: Yang S
PROVIDER: S-EPMC8360075 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Propensity score weighting is sensitive to model misspecification and outlying weights that can unduly influence results. The authors investigated whether trimming large weights downward can improve the performance of propensity score weighting and whether the benefits of trimming differ by propensity score estimation method. In a simulation study, the authors examined the performance of weight trimming following logistic regression, classification and regression trees (CART), boosted CART, and random forests to estimate propensity score weights. Results indicate that although misspecified logistic regression propensity score models yield increased bias and standard errors, weight trimming following logistic regression can improve the accuracy and precision of final parameter estimates. In contrast, weight trimming did not improve the performance of boosted CART and random forests. The performance of boosted CART and random forests without weight trimming was similar to the best performance obtainable by weight trimmed logistic regression estimated propensity scores. While trimming may be used to optimize propensity score weights estimated using logistic regression, the optimal level of trimming is difficult to determine. These results indicate that although trimming can improve inferences in some settings, in order to consistently improve the performance of propensity score weighting, analysts should focus on the procedures leading to the generation of weights (i.e., proper specification of the propensity score model) rather than relying on ad-hoc methods such as weight trimming.
| S-EPMC3069059 | biostudies-other
Project description:To address issues with measured and unmeasured confounding in observational studies, we developed a unified approach to using an instrumental variable in more flexible ways to evaluate treatment effects. The approach is based on an instrumental propensity score conditional on baseline variables, which can then be incorporated in matching, regression, subclassification, or weighting along with various parametric, semiparametric, or nonparametric methods for the assessment of treatment effects. Therefore, the application of the instrumental propensity score allows different methods for outcome effect evaluations in addition to standard 2-stage least square models while controlling for unmeasured confounders. Several properties of the instrumental propensity score are discussed. The approach is then illustrated using subclassification along with a semiparametric density ratio model and empirical likelihood. This method allows us to evaluate distributional and subgroup treatment effects in addition to the overall average treatment effect. Simulation studies showed that the method works well. We applied our method to a study of the effects of attending a Catholic school versus a public school and found that attending a Catholic school had significant beneficial effects on subsequent wages among a subgroup of subjects.
| S-EPMC5976917 | biostudies-literature
Project description:Background Extrathyroidal extension is a major risk factor for poor prognosis in papillary thyroid cancer. However, the effect of different degrees of extrathyroidal extension on prognosis remains controversial. We performed a retrospective study to elucidate how the extent of extrathyroidal extension in papillary thyroid cancer affected the clinical prognosis of patients and its covariates. Methods The study included 108,426 patients with papillary thyroid cancer. We categorized the extent of extension into none, capsule, strap muscles, and other organs. Three causal inference methods for retrospective studies, namely, inverse probability of treatment weighting, standardized mortality ratio weighting, and propensity score matching analysis, were used to minimize potential selection bias. Kaplan–Meier analysis and univariate Cox regression analyses were applied to analyze the precise effect of ETE on survival in papillary thyroid cancer patients. Results In the Kaplan–Meier survival analysis, only extrathyroidal extension into or beyond the strap muscles was statistically significant for both overall survival (OS) and thyroid cancer-specific survival (TCSS). In univariate Cox regression analyses before and after matching or weighting based on causal inference, extrathyroidal extension into soft tissues or other organs is a high-risk factor for both overall survival and thyroid cancer-specific survival. Sensitivity analysis revealed that lower overall survival was observed in patients with older age (≥55) and larger tumor size (>2 cm) of papillary thyroid cancer with extrathyroidal extension into or beyond the strap muscles. Conclusions Our study indicates that extrathyroidal extension into soft tissues or other organs is a high-risk factor in all papillary thyroid cancer. Even though invasion into the strap muscles did not seem to be a marker for poor prognosis, it still impaired the overall survival of patients with older age (≥55 years old) or larger tumor size (>2 cm). Further investigation is needed to confirm our results and to clarify further risk factors independent of extrathyroidal extension.
| S-EPMC10244725 | biostudies-literature
Project description:IntroductionDespite being genetically inherited, it is unclear how non-genetic factors (e.g., substance use, employment) might contribute to the progression and severity of Huntington's disease (HD).MethodsWe used propensity score (PS) weighting in a large (n = 2914) longitudinal dataset (Enroll-HD) to examine the impact of education, employment status, and use of tobacco, alcohol, and recreational and therapeutic drugs on HD progression. Each factor was investigated in isolation while controlling for 19 other factors to ensure that groups were balanced at baseline on potential confounders using PS weights. Outcomes were compared several years later using doubly robust models.ResultsOur results highlighted cases where modifiable (non-genetic) factors - namely light and moderate alcohol use and employment - would have been associated with HD progression in models that did not use PS weights to control for baseline imbalances. These associations did not hold once we applied PS weights to balance baseline groups. We also found potential evidence of a protective effect of substance use (primarily marijuana use), and that those who needed antidepressant treatment were likely to progress faster than non-users.ConclusionsOur study is the first to examine the effect of non-genetic factors on HD using a novel application of PS weighting. We show that previously-reported associated factors - including light and moderate alcohol use - are reduced and no longer significantly linked to HD progression after PS weighting. This indicates the potential value of PS weighting in examining non-genetic factors contributing to HD as well as in addressing the known biases that occur with observational data.
| S-EPMC7949328 | biostudies-literature
Project description:Propensity score methods (e.g., matching, weighting, subclassification) provide a statistical approach for balancing dissimilar exposure groups on baseline covariates. These methods were developed in the context of data with no hierarchical structure or clustering. Yet in many applications the data have a clustered structure that is of substantive importance, such as when individuals are nested within healthcare providers or within schools. Recent work has extended propensity score methods to a multilevel setting, primarily focusing on binary exposures. In this paper, we focus on propensity score weighting for a continuous, rather than binary, exposure in a multilevel setting. Using simulations, we compare several specifications of the propensity score: a random effects model, a fixed effects model, and a single-level model. Additionally, our simulations compare the performance of marginal versus cluster-mean stabilized propensity score weights. In our results, regression specifications that accounted for the multilevel structure reduced bias, particularly when cluster-level confounders were omitted. Furthermore, cluster mean weights outperformed marginal weights.
| S-EPMC5157938 | biostudies-literature
Project description:BackgroundThe diagnostic process is a key element of medicine but it is complex and prone to errors. Infectious diseases are one of the three categories of diseases in which diagnostic errors can be most harmful to patients. In this study we aimed to estimate the effect of initial misdiagnosis of the source of infection in patients with bacteraemia on 14 day mortality using propensity score methods to adjust for confounding.MethodsData from a previously described longitudinal cohort of patients diagnosed with monobacterial bloodstream infection (BSI) at the Leiden University Medical Centre (LUMC) between 2013 and 2015 were used. Propensity score matching and inversed probability of treatment weighting (IPTW) were applied to correct for confounding. The average treatment effect on the treated (ATT), which in this study was the average effect of initial misdiagnosis on the misdiagnosed (AEMM), was estimated. Methodological issues that were encountered when applying propensity score methods were addressed by performing additional sensitivity analyses. Sensitivity analyses consisted of varying caliper in propensity score matching and using different truncated weights in inversed probability of treatment weighting.ResultsData of 887 patients were included in the study. Propensity scores ranged between 0.015 and 0.999 and 80 patients (9.9%) had a propensity score > 0.95. In the matched analyses, 35 of the 171 misdiagnosed patients died within 14 days (20.5%), versus 10 of the 171 correctly diagnosed patients (5.8%), yielding a difference of 14.6% (7.6%; 21.6%). In the total group of patients, the observed percentage of patients with an incorrect initial diagnosis that died within 14 days was 19.8% while propensity score reweighting estimated that their probability of dying would have been 6.5%, if they had been correctly diagnosed (difference 13.3% (95% CI 6.9%;19.6%)). After adjustment for all variables that showed disbalance in the propensity score a difference of 13.7% (7.4%; 19.9%) was estimated. Sensitivity analyses yielded similar results. However, performing weighted analyses without truncation yielded unstable results.ConclusionThus, we observed a substantial increase of 14 day mortality in initially misdiagnosed patients. Furthermore, several patients received propensity scores extremely close to one and were almost sure to be initially misdiagnosed.
| S-EPMC11007998 | biostudies-literature
Project description:BackgroundPerioperative hemoglobin drop after noncardiac surgery is associated with acute kidney injury (AKI). However, opinion on the tolerable difference in postoperative hemoglobin drop in patients with different preoperative hemoglobin levels does not reach a consensus. This study aimed to identify hemoglobin drop thresholds for AKI after noncardiac surgery stratified by preoperative hemoglobin levels.MethodThis was a single-center retrospective cohort study for elective noncardiac surgery from January 1, 2012, to December 31, 2018. The endpoint was the occurrence of AKI 7 days postoperatively in the hospital. The generalized additive model described the non-linear relationship between hemoglobin drop and AKI occurrence. The minimum P-value approach identified cut-off points of hemoglobin drop within postoperative 7 days for patients with or without preoperative anemia. Stratified by preoperative anemia, hemoglobin drop's odds ratio as continuous, quintile and dichotomous variables by various cut-off points for postoperative AKI were calculated in multivariate logistic regression models before and after propensity score weighting (PSW).ResultsOf the 35,631 surgery, 5.9% (2105 cases) suffered postoperative AKI. Non-linearity was found between hemoglobin drop and postoperative AKI occurrence. The thresholds and corresponding odds ratio of perioperative hemoglobin drop for patients with and without preoperative anemia were 18 g/L (1.38 (95%CI 1.14 -1.62), P < .001; after PSW: 1.42 (95%CI 1.17 -1.74), P < .001) and 43 g/L (1.81 (95%CI 1.35-2.27), P < .001; after PSW: 2.88 (95%CI 1.85-4.50), P < .001) respectively. Overall thresholds and corresponding odds ratio were 43 g/L (1.82 (95%CI 1.42-2.21)), P < .001; after PSW: 3.29 (95%CI 2.00-5.40), P < .001). Sensitivity analysis showed similar results. Heterogeneity subgroup analysis showed that intraoperatively female patients undergoing intraperitoneal surgery without colloid infusion seemed to be more vulnerable to higher hemoglobin drop. Further analysis showed a possible linear relationship between preoperative hemoglobin and perioperative hemoglobin drop thresholds. Additionally, this study found that the creatinine level changed simultaneously with hemoglobin level within five postoperative days.ConclusionsHeterogeneity of hemoglobin drop endurability exists after noncardiac non-kidney surgery. More care and earlier intervention should be put on patients with preoperative anemia.
| S-EPMC9188693 | biostudies-literature
Project description:We estimated the causal effects of use of an online smoking cessation community on 30-day point prevalence abstinence at 3 months.Participants (N = 492) were adult current smokers in the enhanced Internet arm of The iQUITT Study, a randomized trial of Internet and telephone treatment for smoking cessation. All participants accessed a Web-based smoking-cessation program that included a large, established online community. Automated tracking metrics of passive (e.g., reading forum posts, viewing member profiles) and active (e.g., writing forum posts, sending private messages) community use were extracted from the site at 3 months. Self-selected community use defines the groups of interest: "None," "Passive," and "Both" (passive + active). Inverse probability of treatment weighting corrected for baseline imbalances on demographic, smoking, psychosocial, and medical history variables. Propensity weights estimated via generalized boosted models were used to calculate Average Treatment Effects (ATE) and Average Treatment effects on the Treated (ATT).Patterns of community use were: None = 198 (40.2%), Passive = 110 (22.4%), and Both = 184 (37.4%). ATE-weighted abstinence rates were: None = 4.2% (95% CI = 1.5-6.9); Passive = 15.1% (95% CI = 8.4-21.9); Both = 20.4% (95% CI = 13.9-26.8). ATT-weighted abstinence rates indicated even greater benefits of community use.Community users were more likely to quit smoking at 3 months than nonusers. The estimated benefit from use of online community resources was even larger among subjects with high propensity to use them. No differences in abstinence emerged between passive and passive/active users. Results suggest that lurking in online communities confers specific abstinence benefits. Implications of these findings for online cessation communities are discussed.
| S-EPMC4681311 | biostudies-literature
Project description:In a previous study of the Child-Parent Centers (CPC) education program, preschool participation was linked to a 4.6 percentage point reduction (26%) in depressive symptoms at ages 22-24 over the matched comparison group enrolling the usual programs. The present study reanalyzed these data in the Chicago Longitudinal Study to address potential attrition bias since more than a quarter of the sample was missing on the outcome. Using inverse probability weighting (IPW) involving 32 predictors of sample retention, findings for the 1,142 participants growing up in high-poverty neighborhoods indicated that CPC participation was associated with a 7.1 percentage point reduction (95% CI = [-9.7, -5.4]) in one or more depressive symptoms (39% reduction over the comparison group). Although this marginal effect was within the confidence interval of the original study (95% CI = [-9.5, 0.3]), the 54% increase in the point estimate is substantial and of practical significance, suggesting underestimation in the prior study. Alternative analysis of different predictors and IPW models, including adjustments for program selection and attrition together, yielded similar results. Findings indicate that high-quality early childhood programs continue to be an important strategy for the prevention of depression and its debilitating effects on individuals and families.
| S-EPMC8127666 | biostudies-literature
Project description:Many spatial phenomena exhibit interference, where exposures at one location may affect the response at other locations. Because interference violates the stable unit treatment value assumption, standard methods for causal inference do not apply. We propose a new causal framework to recover direct and spill-over effects in the presence of spatial interference, taking into account that exposures at nearby locations are more influential than exposures at locations further apart. Under the no unmeasured confounding assumption, we show that a generalized propensity score is sufficient to remove all measured confounding. To reduce dimensionality issues, we propose a Bayesian spline-based regression model accounting for a sufficient set of variables for the generalized propensity score. A simulation study demonstrates the accuracy and coverage properties. We apply the method to estimate the causal effect of wildland fires on air pollution in the Western United States over 2005-2018.
| S-EPMC10790180 | biostudies-literature