Development and Validation of a Claims-Based Model to Predict Categories of Obesity.

We developed and validated a claims-based algorithm that classifies patients into obesity categories. Using Medicare (2007-2017) and Medicaid (2000-2014) claims data linked to 2 electronic health record (EHR) systems in Boston, Massachusetts, we identified a cohort of patients with an EHR-based body mass index (BMI) measurement (calculated as weight (kg)/height (m)2). We used regularized regression to select from 137 variables and built generalized linear models to classify patients with BMIs of ≥25, ≥30, and ≥40. We developed the prediction model using EHR system 1 (training set) and validated it in EHR system 2 (validation set). The cohort contained 123,432 patients in the Medicare population and 40,736 patients in the Medicaid population. The model comprised 97 variables in the Medicare set and 95 in the Medicaid set, including BMI-related diagnosis codes, cardiovascular and antidiabetic drugs, and obesity-related comorbidities. The areas under the receiver-operating-characteristic curve in the validation set were 0.72, 0.75, and 0.83 (Medicare) and 0.66, 0.66, and 0.70 (Medicaid) for BMIs of ≥25, ≥30, and ≥40, respectively. The positive predictive values were 81.5%, 80.6%, and 64.7% (Medicare) and 81.6%, 77.5%, and 62.5% (Medicaid), for BMIs of ≥25, ≥30, and ≥40, respectively. The proposed model can identify obesity categories in claims databases when BMI measurements are missing and can be used for confounding adjustment, defining subgroups, or probabilistic bias analysis.

A simulation-based bias analysis to assess the impact of unmeasured confounding when designing non-randomized database studies.

BACKGROUND: Unmeasured confounding is often raised as a source of potential bias during the design of non-randomized studies but quantifying such concerns is challenging. METHODS: We developed a simulation-based approach to assess the potential impact of unmeasured confounding during the study design stage. The approach involved generation of hypothetical individual-level cohorts using realistic parameters including a binary treatment (prevalence 25%), a time-to-event outcome (incidence 5%), 13 measured covariates, a binary unmeasured confounder (u1, 10%), and a binary measured 'proxy' variable (p1) correlated with u1. Strength of unmeasured confounding and correlations between u1 and p1 were varied in simulation scenarios. Treatment effects were estimated with, a) no adjustment, b) adjustment for measured confounders (Level 1), c) adjustment for measured confounders and their proxy (Level 2). We computed absolute standardized mean differences in u1 and p1 and relative bias with each level of adjustment. RESULTS: Across all scenarios, Level 2 adjustment led to improvement in balance of u1, but this improvement was highly dependent on the correlation between u1 and p1. Level 2 adjustments also had lower relative bias than Level 1 adjustments (in strong u1 scenarios: relative bias of 9.2%, 12.2%, 13.5% at correlations 0.7, 0.5, and 0.3, respectively versus 16.4%, 15.8%, 15.0% for Level 1, respectively). CONCLUSION: An approach using simulated individual-level data was useful to explicitly convey the potential for bias due to unmeasured confounding while designing non-randomized studies and can be helpful in informing design choices.

Prenatal Exposure to Opioids and Neurodevelopmental Disorders in Children: A Bayesian Mediation Analysis.

This study explores natural direct and joint natural indirect effects (JNIE) of prenatal opioid exposure on neurodevelopmental disorders (NDDs) in children mediated through pregnancy complications, major and minor congenital malformations, and adverse neonatal outcomes, using Medicaid claims linked to vital statistics in Rhode Island, United States, 2008-2018. A Bayesian mediation analysis with elastic net shrinkage prior was developed to estimate mean time to NDD diagnosis ratio using posterior mean and 95% credible intervals (CrIs) from Markov chain Monte Carlo algorithms. Simulation studies showed desirable model performance. Of 11,176 eligible pregnancies, 332 had ≥2 dispensations of prescription opioids anytime during pregnancy, including 200 (1.8%) having ≥1 dispensation in the first trimester (T1), 169 (1.5%) in the second (T2), and 153 (1.4%) in the third (T3). A significant JNIE of opioid exposure was observed in each trimester (T1, JNIE = 0.97, 95% CrI: 0.95, 0.99; T2, JNIE = 0.97, 95% CrI: 0.95, 0.99; T3, JNIE = 0.96, 95% CrI: 0.94, 0.99). The proportion of JNIE in each trimester was 17.9% (T1), 22.4% (T2), and 56.3% (T3). In conclusion, adverse pregnancy and birth outcomes jointly mediated the association between prenatal opioid exposure and accelerated time to NDD diagnosis. The proportion of JNIE increased as the timing of opioid exposure approached delivery.

Targeted Learning with an Undersmoothed Lasso Propensity Score Model for Large-Scale Covariate Adjustment in Healthcare Database Studies.

Lasso regression is widely used for large-scale propensity score (PS) estimation in healthcare database studies. In these settings, previous work has shown that undersmoothing (overfitting) Lasso PS models can improve confounding control, but it can also cause problems of non-overlap in covariate distributions. It remains unclear how to select the degree of undersmoothing when fitting large-scale Lasso PS models to improve confounding control while avoiding issues that can result from reduced covariate overlap. Here, we used simulations to evaluate the performance of using collaborative-controlled targeted learning to data-adaptively select the degree of undersmoothing when fitting large-scale PS models within both singly and doubly robust frameworks to reduce bias in causal estimators. Simulations showed that collaborative learning can data-adaptively select the degree of undersmoothing to reduce bias in estimated treatment effects. Results further showed that when fitting undersmoothed Lasso PS-models, the use of cross-fitting was important for avoiding non-overlap in covariate distributions and reducing bias in causal estimates.

Iterative Causal Forest: A Novel Algorithm for Subgroup Identification.

Precisely and efficiently identifying subgroups with heterogeneous treatment effects (HTEs) in real-world evidence studies remains a challenge. Based on the causal forest (CF) method, we developed an iterative CF (iCF) algorithm to identify HTEs in subgroups defined by important variables. Our method iteratively grows different depths of the CF with important effect modifiers, performs plurality votes to obtain decision trees (subgroup decisions) for a family of CFs with different depths, then finds the cross-validated subgroup decision that best predicts the treatment effect as a final subgroup decision. We simulated 12 different scenarios and showed that the iCF outperformed other machine learning methods for interaction/subgroup identification in the majority of scenarios assessed. Using a 20% random sample of fee-for-service Medicare beneficiaries initiating sodium-glucose cotransporter-2 inhibitors (SGLT2i) or glucagon-like peptide-1 receptor agonists (GLP1RA), we implemented the iCF to identify subgroups with HTEs for hospitalized heart failure. Consistent with previous studies suggesting patients with heart failure benefit more from SGLT2i, iCF successfully identified such a subpopulation with HTEs and additive interactions. The iCF is a promising method for identifying subgroups with HTEs in real-world data where the potential for unmeasured confounding can be limited by study design.

Development and Validation of Algorithms to Estimate Live Birth Gestational Age in Medicaid Analytic eXtract Data.

BACKGROUND: While healthcare utilization data are useful for postmarketing surveillance of drug safety in pregnancy, the start of pregnancy and gestational age at birth are often incompletely recorded or missing. Our objective was to develop and validate a claims-based live birth gestational age algorithm. METHODS: Using the Medicaid Analytic eXtract (MAX) linked to birth certificates in three states, we developed four candidate algorithms based on: preterm codes; preterm or postterm codes; timing of prenatal care; and prediction models - using conventional regression and machine-learning approaches with a broad range of prespecified and empirically selected predictors. We assessed algorithm performance based on mean squared error (MSE) and proportion of pregnancies with estimated gestational age within 1 and 2 weeks of the gold standard, defined as the clinical or obstetric estimate of gestation on the birth certificate. We validated the best-performing algorithms against medical records in a nationwide sample. We quantified misclassification of select drug exposure scenarios due to estimated gestational age as positive predictive value (PPV), sensitivity, and specificity. RESULTS: Among 114,117 eligible pregnancies, the random forest model with all predictors emerged as the best performing algorithm: MSE 1.5; 84.8% within 1 week and 96.3% within 2 weeks, with similar performance in the nationwide validation cohort. For all exposure scenarios, PPVs were >93.8%, sensitivities >94.3%, and specificities >99.4%. CONCLUSIONS: We developed a highly accurate algorithm for estimating gestational age among live births in the nationwide MAX data, further supporting the value of these data for drug safety surveillance in pregnancy. See video abstract at, http://links.lww.com/EDE/B989 .

Addressing Posttreatment Selection Bias in Comparative Effectiveness Research, Using Real-World Data and Simulation.

To examine methodologies that address imbalanced treatment switching and censoring, 6 different analytical approaches were evaluated under a comparative effectiveness framework: intention-to-treat, as-treated, intention-to-treat with censor-weighting, as-treated with censor-weighting, time-varying exposure, and time-varying exposure with censor-weighting. Marginal structural models were employed to address time-varying exposure, confounding, and possibly informative censoring in an administrative data set of adult patients who were hospitalized with acute coronary syndrome and treated with either clopidogrel or ticagrelor. The effectiveness endpoint included first occurrence of death, myocardial infarction, or stroke. These methodologies were then applied across simulated data sets with varying frequencies of treatment switching and censoring to compare the effect estimate of each analysis. The findings suggest that implementing different analytical approaches has an impact on the point estimate and interpretation of analyses, especially when censoring is highly unbalanced.

Synthetic Negative Controls: Using Simulation to Screen Large-scale Propensity Score Analyses.

The propensity score has become a standard tool to control for large numbers of variables in healthcare database studies. However, little has been written on the challenge of comparing large-scale propensity score analyses that use different methods for confounder selection and adjustment. In these settings, balance diagnostics are useful but do not inform researchers on which variables balance should be assessed or quantify the impact of residual covariate imbalance on bias. Here, we propose a framework to supplement balance diagnostics when comparing large-scale propensity score analyses. Instead of focusing on results from any single analysis, we suggest conducting and reporting results for many analytic choices and using both balance diagnostics and synthetically generated control studies to screen analyses that show signals of bias caused by measured confounding. To generate synthetic datasets, the framework does not require simulating the outcome-generating process. In healthcare database studies, outcome events are often rare, making it difficult to identify and model all predictors of the outcome to simulate a confounding structure closely resembling the given study. Therefore, the framework uses a model for treatment assignment to divide the comparator population into pseudo-treatment groups where covariate differences resemble those in the study cohort. The partially simulated datasets have a confounding structure approximating the study population under the null (synthetic negative control studies). The framework is used to screen analyses that likely violate partial exchangeability due to lack of control for measured confounding. We illustrate the framework using simulations and an empirical example.

Occurrence of inflammatory bowel disease in patients with chronic inflammatory skin diseases: a cohort study: Classification: Epidemiology.

BACKGROUND: Several studies have linked various chronic inflammatory skin diseases (CISDs) with inflammatory bowel disease (IBD) in a range of data sources with mixed conclusions. OBJECTIVES: We compared the incidence of IBD - ulcerative colitis (UC) and Crohn disease (CD) - in patients with a CISD vs. similar persons without a CISD. METHODS: In this cohort study using nationwide, longitudinal, commercial insurance claims data from the USA, we identified adults and children who were seen by a dermatologist between 2004 and 2020, and diagnosed with either psoriasis, atopic dermatitis, alopecia areata, vitiligo or hidradenitis suppurativa. Comparator patients were identified through risk-set sampling; they were eligible if they were seen by a dermatologist at least twice and not diagnosed with a CISD. Patient follow-up lasted until either IBD diagnosis, death, disenrolment or end of data stream, whichever came first. IBD events, UC or CD, were identified via validated algorithms: hospitalization or diagnosis with endoscopic confirmation. Incidence rates were computed before and after adjustment via propensity-score decile stratification to account for IBD risk factors. Hazard ratios (HR) and 95% confidence intervals (CIs) were estimated to compare the incidence of IBD in CISD vs. non-CISD. RESULTS: We identified patients with atopic dermatitis (n = 123 614), psoriasis (n = 83 049), alopecia areata (n = 18 135), vitiligo (n = 9003) or hidradenitis suppurativa (n = 6806), and comparator patients without a CISD (n = 2 376 120). During a median follow-up time of 718 days, and after applying propensity-score adjustment for IBD risk factors, we observed increased risk of both UC (HRUC 2·30, 95% CI 1·61-3·28) and CD (HRCD 2·70, 1·69-4·32) in patients with hidradenitis suppurativa, an increased risk of CD (HRCD 1·23, 1·03-1·46) but not UC (HRUC 1·01, 0·89-1·14) in psoriasis, and no increased risk of IBD in atopic dermatitis (HRUC 1·02, 0·92-1·12; HRCD 1·08, 0·94-1·23), alopecia areata (HRUC 1·18, 0·89-1·56; HRCD 1·26, 0·86-1·86) or vitiligo (HRUC 1·14, 0·77-1·68; HRCD 1·45, 0·87-2·41). CONCLUSIONS: IBD was increased in patients with hidradenitis suppurativa. CD alone was increased in patients with psoriasis. Neither UC nor CD was increased in patients with atopic dermatitis, alopecia areata or vitiligo. What is already known about this topic? Several studies have linked various chronic inflammatory skin diseases (CISDs) with inflammatory bowel disease (IBD) utilizing a range of data sources, with mixed conclusions. What does this study add? This large-scale, claims-based cohort study expands current knowledge by providing background rates for IBD across multiple CISDs using consistent methods and within a single, nationally representative patient population. We observed a relative increased risk of IBD in patients with hidradenitis suppurativa, but the overall incidence rate difference of IBD was generally low. Crohn disease alone was significantly increased in patients with psoriasis, and neither ulcerative colitis nor Crohn disease was increased in patients with atopic dermatitis, vitiligo or alopecia areata.

Machine learning for improving high-dimensional proxy confounder adjustment in healthcare database studies: An overview of the current literature.

PURPOSE: Supplementing investigator-specified variables with large numbers of empirically identified features that collectively serve as 'proxies' for unspecified or unmeasured factors can often improve confounding control in studies utilizing administrative healthcare databases. Consequently, there has been a recent focus on the development of data-driven methods for high-dimensional proxy confounder adjustment in pharmacoepidemiologic research. In this paper, we survey current approaches and recent advancements for high-dimensional proxy confounder adjustment in healthcare database studies. METHODS: We discuss considerations underpinning three areas for high-dimensional proxy confounder adjustment: (1) feature generation-transforming raw data into covariates (or features) to be used for proxy adjustment; (2) covariate prioritization, selection, and adjustment; and (3) diagnostic assessment. We discuss challenges and avenues of future development within each area. RESULTS: There is a large literature on methods for high-dimensional confounder prioritization/selection, but relatively little has been written on best practices for feature generation and diagnostic assessment. Consequently, these areas have particular limitations and challenges. CONCLUSIONS: There is a growing body of evidence showing that machine-learning algorithms for high-dimensional proxy-confounder adjustment can supplement investigator-specified variables to improve confounding control compared to adjustment based on investigator-specified variables alone. However, more research is needed on best practices for feature generation and diagnostic assessment when applying methods for high-dimensional proxy confounder adjustment in pharmacoepidemiologic studies.

Transparency of high-dimensional propensity score analyses: Guidance for diagnostics and reporting.

PURPOSE: The high-dimensional propensity score (HDPS) is a semi-automated procedure for confounder identification, prioritisation and adjustment in large healthcare databases that requires investigators to specify data dimensions, prioritisation strategy and tuning parameters. In practice, reporting of these decisions is inconsistent and this can undermine the transparency, and reproducibility of results obtained. We illustrate reporting tools, graphical displays and sensitivity analyses to increase transparency and facilitate evaluation of the robustness of analyses involving HDPS. METHODS: Using a study from the UK Clinical Practice Research Datalink that implemented HDPS we demonstrate the application of the proposed recommendations. RESULTS: We identify seven considerations surrounding the implementation of HDPS, such as the identification of data dimensions, method for code prioritisation and number of variables selected. Graphical diagnostic tools include assessing the balance of key confounders before and after adjusting for empirically selected HDPS covariates and the identification of potentially influential covariates. Sensitivity analyses include varying the number of covariates selected and assessing the impact of covariates behaving empirically as instrumental variables. In our example, results were robust to both the number of covariates selected and the inclusion of potentially influential covariates. Furthermore, our HDPS models achieved good balance in key confounders. CONCLUSIONS: The data-adaptive approach of HDPS and the resulting benefits have led to its popularity as a method for confounder adjustment in pharmacoepidemiological studies. Reporting of HDPS analyses in practice may be improved by the considerations and tools proposed here to increase the transparency and reproducibility of study results.

Propensity Score Weighting and Trimming Strategies for Reducing Variance and Bias of Treatment Effect Estimates: A Simulation Study.

To extend previous simulations on the performance of propensity score (PS) weighting and trimming methods to settings without and with unmeasured confounding, Poisson outcomes, and various strengths of treatment prediction (PS c statistic), we simulated studies with a binary intended treatment T as a function of 4 measured covariates. We mimicked treatment withheld and last-resort treatment by adding 2 "unmeasured" dichotomous factors that directed treatment to change for some patients in both tails of the PS distribution. The number of outcomes Y was simulated as a Poisson function of T and confounders. We estimated the PS as a function of measured covariates and trimmed the tails of the PS distribution using 3 strategies ("Crump," "Stürmer," and "Walker"). After trimming and reestimation, we used alternative PS weights to estimate the treatment effect (rate ratio): inverse probability of treatment weighting, standardized mortality ratio (SMR)-treated, SMR-untreated, the average treatment effect in the overlap population (ATO), matching, and entropy. With no unmeasured confounding, the ATO (123%) and "Crump" trimming (112%) improved relative efficiency compared with untrimmed inverse probability of treatment weighting. With unmeasured confounding, untrimmed estimates were biased irrespective of weighting method, and only Stürmer and Walker trimming consistently reduced bias. In settings where unmeasured confounding (e.g., frailty) may lead physicians to withhold treatment, Stürmer and Walker trimming should be considered before primary analysis.

Dupilumab and the risk of conjunctivitis and serious infection in patients with atopic dermatitis: A propensity score-matched cohort study.

BACKGROUND: Dupilumab is an effective treatment for moderate to severe atopic dermatitis (AD) with limited safety data in clinical practice. OBJECTIVE: To assess the 6-month risk of conjunctivitis and serious infections in patients with AD who initiated dupilumab. METHODS: In a cohort study using US claims data, we compared the risk of conjunctivitis and serious infections in patients with AD who initiated either dupilumab, methotrexate (MTX), cyclosporine, or mycophenolate. Relative risks (RRs) were computed after 1:1 propensity score matching. RESULTS: We identified 1775 dupilumab, 1034 MTX, 186 cyclosporine, and 257 mycophenolate users. The 6-month risk for any conjunctivitis was 6.5% for dupilumab, 3.3% for MTX, 4.8% for cyclosporine, and 1.2% for mycophenolate initiators. After PS matching, the RR of any conjunctivitis was increased in dupilumab users versus MTX (RR, 2.45; 95% confidence interval [CI], 1.47-4.08), versus cyclosporine (RR, 1.56; 95% CI, 0.69-3.50), and versus mycophenolate (RR, 7.00; 95% CI, 2.12-23.2). The risk of serious infection was 0.6% in dupilumab and 1.0% in MTX initiators (RR, 0.90; 95% CI, 0.37-2.20). LIMITATIONS: Analyses were based on few events, and differential surveillance is a concern. CONCLUSIONS: Although dupilumab shows a low risk of serious infections, it is associated with a clinically meaningful increase in conjunctivitis that needs to be managed in practice.

Greedy caliper propensity score matching can yield variable estimates of the treatment-outcome association-A simulation study.

PURPOSE: Greedy caliper propensity score (PS) matching is dependent on randomness, which can ultimately affect causal estimates. We sought to investigate the variation introduced by this randomness. METHODS: Based on a literature search to define the simulation parameters, we simulated 36 cohorts of different sizes, treatment prevalence, outcome prevalence, treatment-outcome-association. We performed 1:1 caliper and nearest neighbor (NN) caliper PS-matching and repeated this 1000 times in the same cohort, before calculating the treatment-outcome association. RESULTS: Repeating caliper and NN caliper matching in the same cohort yielded large variations in effect estimates, in all 36 scenarios, with both types of matching. The largest variation was found in smaller cohorts, where the odds ratio (OR) ranged from 0.53 to 10.00 (IQR of ORs: 1.11-1.67). The 95% confidence interval was not consistently overlapping a neutral association after repeating the matching with both algorithms. We confirmed these findings in a noninterventional example study. CONCLUSION: Caliper PS-matching can yield highly variable estimates of the treatment-outcome association if the analysis is repeated.

Consequences of Depletion of Susceptibles for Hazard Ratio Estimators Based on Propensity Scores.

We use simulated data to examine the consequences of depletion of susceptibles for hazard ratio (HR) estimators based on a propensity score (PS). First, we show that the depletion of susceptibles attenuates marginal HRs toward the null by amounts that increase with the incidence of the outcome, the variance of susceptibility, and the impact of susceptibility on the outcome. If susceptibility is binary then the Bross bias multiplier, originally intended to quantify bias in a risk ratio from a binary confounder, also quantifies the ratio of the instantaneous marginal HR to the conditional HR as susceptibles are depleted differentially. Second, we show how HR estimates that are conditioned on a PS tend to be between the true conditional and marginal HRs, closer to the conditional HR if treatment status is strongly associated with susceptibility and closer to the marginal HR if treatment status is weakly associated with susceptibility. We show that associations of susceptibility with the PS matter to the marginal HR in the treated (ATT) though not to the marginal HR in the entire cohort (ATE). Third, we show how the PS can be updated periodically to reduce depletion-of-susceptibles bias in conditional estimators. Although marginal estimators can hit their ATE or ATT targets consistently without updating the PS, we show how their targets themselves can be misleading as they are attenuated toward the null. Finally, we discuss implications for the interpretation of HRs and their relevance to underlying scientific and clinical questions. See video Abstract: http://links.lww.com/EDE/B727.

Use of Time-Dependent Propensity Scores to Adjust Hazard Ratio Estimates in Cohort Studies with Differential Depletion of Susceptibles.

Estimating hazard ratios (HR) presents challenges for propensity score (PS)-based analyses of cohorts with differential depletion of susceptibles. When the treatment effect is not null, cohorts that were balanced at baseline tend to become unbalanced on baseline characteristics over time as "susceptible" individuals drop out of the population at risk differentially across treatment groups due to having outcome events. This imbalance in baseline covariates causes marginal (population-averaged) HRs to diverge from conditional (covariate-adjusted) HRs over time and systematically move toward the null. Methods that condition on a baseline PS yield HR estimates that fall between the marginal and conditional HRs when these diverge. Unconditional methods that match on the PS or weight by a function of the PS can estimate the marginal HR consistently but are prone to misinterpretation when the marginal HR diverges toward the null. Here, we present results from a series of simulations to help analysts gain insight on these issues. We propose a novel approach that uses time-dependent PSs to consistently estimate conditional HRs, regardless of whether susceptibles have been depleted differentially. Simulations show that adjustment for time-dependent PSs can adjust for covariate imbalances over time that are caused by depletion of susceptibles. Updating the PS is unnecessary when outcome incidence is so low that depletion of susceptibles is negligible. But if incidence is high, and covariates and treatment affect risk, then covariate imbalances arise as susceptibles are depleted, and PS-based methods can consistently estimate the conditional HR only if the PS is periodically updated.

Serious infection risk in children with psoriasis on systemic treatment: A propensity score-matched population-based study.

BACKGROUND: Psoriasis is increasingly treated with systemic medications, yet their safety is not well characterized in children. OBJECTIVE: We sought to estimate the 6-month risk of serious infections in children with psoriasis treated with biologics, systemic nonbiologics, and phototherapy. METHODS: Using insurance claims data, we identified children aged <18 years with psoriasis and compared the frequency of serious infections in those initiating biologics, systemic nonbiologics, and phototherapy. Relative risks were estimated before and after 1:1 propensity score matching. RESULTS: Among 57,323 children with psoriasis, the 6-month risk of infection was 4.2 per 1000 patient-years in 722 biologic initiators, 5.1 in 988 systemic nonbiologic initiators, and 1.1 in 2657 phototherapy initiators. The relative risk (95% confidence interval) of infection in biologics vs nonbiologics was 0.67 (0.11-3.98), in biologics vs phototherapy was 1.50 (0.25-8.95), and in nonbiologics vs phototherapy was 5.00 (0.59-42.71). The background risk of infection in children with psoriasis was 1 per 1000, almost double the risk compared with children without psoriasis (relative risk, 1.84; 95% confidence interval, 1.15-1.97). CONCLUSIONS: We found no meaningful difference in infection risk between biologics vs nonbiologics and no robust difference between systemic users vs phototherapy. Independent of treatment, children with psoriasis had a higher risk of infection than those without psoriasis.

Restriction of Pharmacoepidemiologic Cohorts to Initiators of Medications in Unrelated Preventive Drug Classes to Reduce Confounding by Frailty in Older Adults.

Nonexperimental studies of the effectiveness of seasonal influenza vaccine in older adults have found 40%-60% reductions in all-cause mortality associated with vaccination, potentially due to confounding by frailty. We restricted our cohort to initiators of medications in preventive drug classes (statins, antiglaucoma drugs, and ß blockers) as an approach to reducing confounding by frailty by excluding frail older adults who would not initiate use of these drugs. Using a random 20% sample of US Medicare beneficiaries, we framed our study as a series of nonrandomized "trials" comparing vaccinated beneficiaries with unvaccinated beneficiaries who had an outpatient health-care visit during the 5 influenza seasons occurring in 2010-2015. We pooled data across trials and used standardized-mortality-ratio-weighted Cox proportional hazards models to estimate the association between influenza vaccination and all-cause mortality before influenza season, expecting a null association. Weighted hazard ratios among preventive drug initiators were generally closer to the null than those in the nonrestricted cohort. Restriction of the study population to statin initiators with an uncensored approach resulted in a weighted hazard ratio of 1.00 (95% confidence interval: 0.84, 1.19), and several other hazard ratios were above 0.95. Restricting the cohort to initiators of medications in preventive drug classes can reduce confounding by frailty in this setting, but further work is required to determine the most appropriate criteria to use.

Evaluating the use of bootstrapping in cohort studies conducted with 1:1 propensity score matching-A plasmode simulation study.

PURPOSE: Bootstrapping can account for uncertainty in propensity score (PS) estimation and matching processes in 1:1 PS-matched cohort studies. While theory suggests that the classical bootstrap can fail to produce proper coverage, practical impact of this theoretical limitation in settings typical to pharmacoepidemiology is not well studied. METHODS: In a plasmode-based simulation study, we compared performance of the standard parametric approach, which ignores uncertainty in PS estimation and matching, with two bootstrapping methods. The first method only accounted for uncertainty introduced during the matching process (the observation resampling approach). The second method accounted for uncertainty introduced during both PS estimation and matching processes (the PS reestimation approach). Variance was estimated based on percentile and empirical standard errors, and treatment effect estimation was based on median and mean of the estimated treatment effects across 1000 bootstrap resamples. Two treatment prevalence scenarios (5% and 29%) across two treatment effect scenarios (hazard ratio of 1.0 and 2.0) were evaluated in 500 simulated cohorts of 10 000 patients each. RESULTS: We observed that 95% confidence intervals from the bootstrapping approaches but not the standard approach, resulted in inaccurate coverage rates (98%-100% for the observation resampling approach, 99%-100% for the PS reestimation approach, and 95%-96% for standard approach). Treatment effect estimation based on bootstrapping approaches resulted in lower bias than the standard approach (less than 1.4% vs 4.1%) at 5% treatment prevalence; however, the performance was equivalent at 29% treatment prevalence. CONCLUSION: Use of bootstrapping led to variance overestimation and inconsistent coverage, while coverage remained more consistent with parametric estimation.