High-dimensional Iterative Causal Forest (hdiCF) for Subgroup Identification Using Health Care Claims Data.

We recently developed a machine-learning subgrouping algorithm, iterative causal forest (iCF), to identify subgroups with heterogeneous treatment effects (HTEs) using predefined covariates. However, such predefined covariates may miss or poorly define important features leading to inaccurate subgrouping. To address such limitations, we developed a new semi-automatic subgrouping algorithm, hdiCF, which adapts methodology from high-dimensional propensity score for feature recognition in claims data. The hdiCF algorithm has 3 steps: 1) high-dimensional feature identification by International Classification of Diseases, Current Procedural Terminology, and Anatomical Therapeutic Chemical codes (in/outpatient diagnoses, procedures, prescriptions) and creation of ordinal variables by frequency of occurrence; 2) propensity score trimming and high-dimensional feature preparation; 3) iCF implementation to identify subgroups. We applied hdiCF in a 20% random sample of fee-for-service Medicare beneficiaries who initiated sodium-glucose cotransporter-2 inhibitors (SGLT2i) or glucagon-like peptide-1 receptor agonists to identify subgroups with HTEs for incidence of hospitalized heart failure. HdiCF findings were consistent with studies suggesting SGLT2i to be more beneficial for patients with pre-existing heart failure or chronic kidney disease. HdiCF is not dependent on prior hypotheses about HTEs and identifies subgroups with markers for potential HTEs in real-world evidence studies where active-comparator, new-user study designs limit the potential for unmeasured confounding.

A Systematic Approach to Evaluating Instrumental Variable Assumptions: Applied Example of Glucose-lowering Medications and Risk for Hospitalized Heart Failure in Older Adults.

One obstacle to adopting instrumental variable (IV) methods in pharmacoepidemiology is their reliance on strong, unverifiable assumptions. We can falsify IV assumptions by leveraging the causal structure, which can strengthen or refute their plausibility and increase the validity of effect estimates. We illustrate a systematic approach to evaluate calendar time IV assumptions in estimating the known effect of thiazolidinediones on hospitalized heart failure. Using cohort entry time before and after 09/2010, when the U.S. Food and Drug Administration issued a safety communication as a proposed IV, we estimated IV and propensity score-weighted 2-year risk differences (RDs) using Medicare data (2008-2014). We (i) performed inequality tests, (ii) identified the negative control IV/outcome using causal assumptions, (iii) estimated RDs after narrowing the calendar time range and excluding patients likely associated with unmeasured confounding, (iv) derived bounds for RDs, and (v) estimated the proportion of compliers and their characteristics. The findings revealed that IV assumptions were violated and RDs were extreme, but the assumptions became more plausible upon narrowing the calendar time range and restricting the cohort by excluding prevalent heart failure (the strongest measured predictor of outcome). Systematically evaluating IV assumptions could help detect bias in IV estimators and increase their validity.

Use of Sensitivity Analyses to Assess Uncontrolled Confounding from Unmeasured Variables in Observational, Active Comparator Pharmacoepidemiologic Studies: A Systematic Review.

Understanding the potential for, direction, and magnitude of uncontrolled confounding is critical for generating informative real-world evidence. Many sensitivity analyses are available to assess robustness of study results to residual confounding, but it is unclear how researchers are using these methods. We conducted a systematic review of published active comparator cohort studies of drugs or biologics to summarize use of sensitivity analyses aimed at assessing uncontrolled confounding from an unmeasured variable. We reviewed articles in five medical and seven epidemiologic journals published between January 1, 2017, and June 30, 2022. We identified 158 active comparator cohort studies, 76 from medical and 82 from epidemiologic journals. Residual, unmeasured, or uncontrolled confounding was noted as a potential concern in 93% of studies, but only 84 (53%) implemented one or more sensitivity analysis to assess uncontrolled confounding from an unmeasured variable. The most common analyses were E-values among medical journal articles (21%) and restriction on measured variables among epidemiologic journal articles (22%). Researchers must rigorously consider the role of residual confounding in their analyses and the best sensitivity analyses for assessing this potential bias.

Longitudinal trajectories of a claims-based frailty measure during adjuvant chemotherapy in women with stage I-III breast cancer.

BACKGROUND: Frailty is a dynamic syndrome characterized by reduced physiological reserve to maintain homeostasis. Prospective studies have reported frailty worsening in women with breast cancer during chemotherapy, with improvements following treatment. We evaluated whether the Faurot frailty index, a validated claims-based frailty measure, could identify changes in frailty during chemotherapy treatment and identified predictors of trajectory patterns. METHODS: We included women (65+ years) with stage I-III breast cancer undergoing adjuvant chemotherapy in the SEER-Medicare database (2003-2019). We estimated the Faurot frailty index (range: 0-1; higher scores indicate greater frailty) at chemotherapy initiation, 4 months postinitiation, and 10 months postinitiation. Changes in frailty were compared to a matched noncancer comparator cohort. We identified patterns of frailty trajectories during the year following chemotherapy initiation using K-means clustering. RESULTS: Twenty-one thousand five hundred and ninety-nine women initiated adjuvant chemotherapy. Mean claims-based frailty increased from 0.037 at initiation to 0.055 4 months postchemotherapy initiation and fell to 0.049 10 months postinitiation. Noncancer comparators experienced a small increase in claims-based frailty over time (0.055-0.062). We identified 6 trajectory patterns: a robust group (78%), 2 resilient groups (16%), and 3 nonresilient groups (6%). Black women and women with claims for home hospital beds, wheelchairs, and Parkinson's disease were more likely to experience nonresilient trajectories. CONCLUSIONS: We observed changes in a claims-based frailty index during chemotherapy that are consistent with prior studies using clinical measures of frailty and identified predictors of nonresilient frailty trajectories. Our study demonstrates the feasibility of using claims-based frailty indices to assess changes in frailty during cancer treatment.

Glucagon-Like Peptide-1 Receptor Agonists Do Not Increase Aspiration During Upper Endoscopy in Patients With Diabetes.

BACKGROUND & AIMS: Glucagon-like peptide-1-receptor agonists (GLP1-RAs) have been associated with greater retention of gastric contents, however, there is minimal controlled, population-based data evaluating the potential adverse effects of GLP1-RA in the periprocedural setting. We aimed to determine if there is increased risk of aspiration and aspiration-related complications after upper endoscopy in patients using GLP1-RAs. METHODS: We used a nationwide commercial administrative claims database to conduct a retrospective cohort study of patients aged 18 to 64 with type 2 diabetes who underwent outpatient upper endoscopy from 2005 to 2021. We identified 6,806,046 unique upper endoscopy procedures. We compared claims for aspiration and associated pulmonary adverse events in the 14 days after upper endoscopy between users of GLP1-RAs, dipeptidyl peptidase 4 inhibitors (DPP4is), and chronic opioids. We adjusted for age, sex, Charlson Comorbidity score, underlying respiratory disease, and gastroparesis. RESULTS: We found that pulmonary adverse events after upper endoscopy are rare, ranging from 6 to 25 events per 10,000 procedures. When comparing GLP1-RAs with DPP4i, crude relative risks of aspiration (0.67; 95% CI, 0.25-1.75), aspiration pneumonia (0.95; 95% CI, 0.40-2.29), pneumonia (1.07; 95% CI, 0.62-1.86), or respiratory failure (0.75; 95% CI, 0.38-1.48) were not higher in patients prescribed a GLP1-RA. When comparing GLP1-RAs with opioids, crude relative risks were 0.42 (95% CI, 0.15-1.16) for aspiration, 0.60 (95% CI, 0.24-1.52) for aspiration pneumonia, 0.30 (95% CI, 0.19-0.49) for pneumonia, and 0.24 (95% CI, 0.13-0.45) for respiratory failure. These results were consistent across several sensitivity analyses. CONCLUSIONS: GLP1-RA use is not associated with an increased risk of pulmonary complications after upper endoscopy compared with DPP4i use in patients with type 2 diabetes.

Patterns of chemotherapy receipt among patients with hormone receptor-positive, HER2-negative breast cancer.

BACKGROUND: Breast cancer chemotherapy utilization not only may differ by race and age, but also varies by genomic risk, tumor characteristics, and patient characteristics. Studies in demographically diverse populations with both clinical and genomic data are necessary to understand potential disparities by race and age. METHODS: In the Carolina Breast Cancer Study Phase 3 (2008-2013), chemotherapy receipt (yes/no) and regimen type were assessed in association with age and race among hormone receptor (HR) positive and HER2-negative tumors (n = 1862). Odds ratios were estimated for the association between demographic factors and chemotherapy receipt. RESULTS: Monotonic decreases in frequency of adjuvant chemotherapy receipt were observed over time during the study period, while neoadjuvant chemotherapy was stable. Younger age was associated with chemotherapy receipt (OR [95% CI]: 2.9 [2.4, 3.6]) and with anthracycline-based regimens (OR [95% CI]: 1.7 [1.3, 2.4]). Participants who had Medicaid (OR [95% CI]: 1.8 [1.3, 2.5]), lived in rural settings (OR [95% CI]: 1.4 [1.0, 2.0]), or were Black (OR [95% CI]: 1.5 [1.2, 1.8]) had slightly higher odds of chemotherapy, but these associations were non-significant with adjustment for stage and grade. Associations between younger age and chemotherapy receipt were strongest among women who did not receive genomic testing. CONCLUSIONS: While race was not strongly associated with chemotherapy receipt, younger age remains a strong predictor of chemotherapy receipt, even with adjustment for clinical factors and among women who receive genomic testing.

Visualizing External Validity: Graphical Displays to Inform the Extension of Treatment Effects from Trials to Clinical Practice.

BACKGROUND: In the presence of effect measure modification, estimates of treatment effects from randomized controlled trials may not be valid in clinical practice settings. The development and application of quantitative approaches for extending treatment effects from trials to clinical practice settings is an active area of research. METHODS: In this article, we provide researchers with a practical roadmap and four visualizations to assist in variable selection for models to extend treatment effects observed in trials to clinical practice settings and to assess model specification and performance. We apply this roadmap and visualizations to an example extending the effects of adjuvant chemotherapy (5-fluorouracil vs. plus oxaliplatin) for colon cancer from a trial population to a population of individuals treated in community oncology practices in the United States. RESULTS: The first visualization screens for potential effect measure modifiers to include in models extending trial treatment effects to clinical practice populations. The second visualization displays a measure of covariate overlap between the clinical practice populations and the trial population. The third and fourth visualizations highlight considerations for model specification and influential observations. The conceptual roadmap describes how the output from the visualizations helps interrogate the assumptions required to extend treatment effects from trials to target populations. CONCLUSIONS: The roadmap and visualizations can inform practical decisions required for quantitatively extending treatment effects from trials to clinical practice settings.

Performance of a Claims-Based Frailty Proxy Using Varying Frailty Ascertainment Lookback Windows.

BACKGROUND: Frailty is an aging-related syndrome of reduced physiological reserve to maintain homeostasis. The Faurot frailty index has been validated as a Medicare claims-based proxy for predicting frailty using billing information from a user-specified ascertainment window. OBJECTIVES: We assessed the validity of the Faurot frailty index as a predictor of the frailty phenotype and 1-year mortality using varying frailty ascertainment windows. RESEARCH DESIGN: We identified older adults (66+ y) in Round 5 (2015) of the National Health and Aging Trends Study with Medicare claims linkage. Gold standard frailty was assessed using the frailty phenotype. We calculated the Faurot frailty index using 3, 6, 8, and 12 months of claims prior to the survey or all-available lookback. Model performance for each window in predicting the frailty phenotype was assessed by quantifying calibration and discrimination. Predictive performance for 1-year mortality was assessed by estimating risk differences across claims-based frailty strata. RESULTS: Among 4253 older adults, the 6 and 8-month windows had the best frailty phenotype calibration (calibration slopes: 0.88 and 0.87). All-available lookback had the best discrimination (C-statistic=0.780), but poor calibration. Mortality associations were strongest using a 3-month window and monotonically decreased with longer windows. Subgroup analyses revealed worse performance in Black and Hispanic individuals than counterparts. CONCLUSIONS: The optimal ascertainment window for the Faurot frailty index may depend on the clinical context, and researchers should consider tradeoffs between discrimination, calibration, and mortality. Sensitivity analyses using different durations can enhance the robustness of inferences. Research is needed to improve prediction across racial and ethnic groups.

Clarifying the causal contrast: An empirical example applying the prevalent new user study design.

PURPOSE: The prevalent new user design extends the active comparator new user design to include patients switching to a treatment of interest from a comparator. We examined the impact of adding "switchers" to incident new users on the estimated hazard ratio (HR) of hospitalized heart failure. METHODS: Using MarketScan claims data (2000-2014), we estimated HRs of hospitalized heart failure between patients initiating GLP-1 receptor agonists (GLP-1 RA) and sulfonylureas (SU). We considered three estimands: (1) the effect of incident new use; (2) the effect of switching; and (3) the effect of incident new use or switching, combining the two population. We used time-conditional propensity scores (TCPS) and time-stratified standardized morbidity ratio (SMR) weighting to adjust for confounding. RESULTS: We identified 76 179 GLP-1 RA new users, of which 12% were direct switchers (within 30 days) from SU. Among incident new users, GLP-1 RA was protective against heart failure (adjHRSMR = 0.74 [0.69, 0.80]). Among switchers, GLP-1 RA was not protective (adjHRSMR = 0.99 [0.83, 1.18]). Results in the combined population were largely driven by the incident new users, with GLP-1 RA having a protective effect (adjHRSMR = 0.77 [0.72, 0.83]). Results using TCPS were consistent with those estimated using SMR weighting. CONCLUSIONS: When analyses were conducted only among incident new users, GLP-1 RA had a protective effect. However, among switchers from SU to GLP-1 RA, the effect estimates substantially shifted toward the null. Combining patients with varying treatment histories can result in poor confounding control and camouflage important heterogeneity.

The Effect of Sodium Glucose Cotransporter-2 Inhibitors on Hemoglobin A1c Variability and Acute Kidney Injury: A Causal Mediation Analysis.

PURPOSE: The role of lower hemoglobin A1c (HbA1c) variability in the effect of sodium glucose cotransporter-2 inhibitors (SGLT2i) on acute kidney injury (AKI) remains unclear. We compared AKI risk between SGLT2i and dipeptidyl peptidase 4 inhibitors (DPP4i) initiators. Additionally, we aimed to explore the extent to which SGLT2i's influence on AKI risk is mediated by reducing long-term HbA1c variability. METHODS: Using 2018-2022 year data in Yinzhou Regional Health Care Database, we included adult, type 2 diabetes patients who were new users of SGLT2i or DPP4i. The effect of SGLT2i versus DPP4i on AKI, HbA1c variability, and AKI through HbA1c variability was compared using inverse probability of treatment weighted Cox proportional hazards models, median regression models, and causal mediation analysis. RESULTS: With a median follow-up of 1.76 years, 19 717 adults (for SGLT2i, n = 6008; for DPP4i, n = 13 709) with type 2 diabetes were included. The adjusted hazard ratio for SGLT2i versus DPP4i was 0.79 (95% confidence interval [CI] 0.64-0.98) for AKI. The adjusted differences in median HbA1c variability score (HVS) and HbA1c reduction were -16.67% (95% CI: -27.71% to -5.62%) and -1.98% (95% CI: -14.34% to 10.38%), respectively. Furthermore, lower AKI risk associated with SGLT2i was moderately mediated (22.77%) through HVS. The results remained consistent across various subgroups and sensitivity analyses. CONCLUSIONS: Compared to DPP4i, lower AKI risk associated with SGLT2i is moderately mediated through HbA1c variability. These findings enhance our understanding of the effect of SGLT2i on AKI and underscore the importance of considering HbA1c variability in diabetes treatment and management.

Hazard Ratios and Alternative Effect Measures: An Applied Illustration.

PURPOSE: Although the limitations of hazard ratios (HRs) for quantifying treatment effects in right-censored data have been widely discussed, HRs are still preferentially reported over other, more interpretable effect measures. This may stem from the fact that there are few applied examples that directly contrast the HR and its interpretation with alternative effect measures. METHODS: We analyzed data from two randomized clinical trials comparing panitumumab plus standard-of-care chemotherapy (SOCC) with SOCC alone as first- and second-line treatment for metastatic colorectal cancer. We report the effect of treatment with panitumumab on progression-free survival (PFS) using a Cox proportional hazards model to estimate the HR and the Kaplan-Meier estimator of cumulative incidence (risk). Further analyses included examining the cumulative incidence curves; kernel-smoothed, non-parametric hazards curves; fitting the Cox model with a continuous time variable; and estimating restricted mean survival as well as median survival. RESULTS: The HR was 0.82 (95% confidence interval [CI]: 0.71, 0.93), while the risk ratio (or relative risk [i.e., ratio of the cumulative incidence among the treated versus comparator]) was 0.99 (95% CI: 0.96, 1.02). These two measures suggest apparently different conclusions: either a treatment benefit or no effect. Through subsequent analyses, we demonstrated that, while the cumulative incidence of the outcome was similar by the end of follow-up regardless of treatment, the panitumumab treated group experienced longer PFS than those randomized to SOCC. Substantial nonproportional hazards were evident with panitumumab treatment reducing the hazard of progression/mortality during the first ~1.75 years but associated with an increased hazard of progress/mortality thereafter. DISCUSSION: This example underscores the difficulties in interpreting HRs, particularly in the setting of qualitative violations of proportional hazards, and the value of quantifying treatment effects via multiple effect measures.

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.

Translation of a Claims-Based Frailty Index From the International Classification of Diseases, Ninth Revision, Clinical Modification to the Tenth Revision.

The Faurot frailty index (FFI) is a validated algorithm that uses enrollment and International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM)-based billing information from Medicare claims data as a proxy for frailty. In October 2015, the US health-care system transitioned from the ICD-9-CM to the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). Applying the Centers for Medicare and Medicaid Services General Equivalence Mappings, we translated diagnosis-based frailty indicator codes from the ICD-9-CM to the ICD-10-CM, followed by manual review. We used interrupted time-series analysis of Medicare data to assess the comparability of the pre- and posttransition FFI scores. In cohorts of beneficiaries enrolled in January 2015-2017 with 8-month frailty look-back periods, we estimated associations between the FFI and 1-year risk of aging-related outcomes (mortality, hospitalization, and admission to a skilled nursing facility). Updated indicators had similar prevalences as pretransition definitions. The median FFI scores and interquartile ranges (IQRs) for the predicted probability of frailty were similar before and after the International Classification of Diseases transition (pretransition: median, 0.034 (IQR, 0.02-0.07); posttransition: median, 0.038 (IQR, 0.02-0.09)). The updated FFI was associated with increased risks of mortality, hospitalization, and skilled nursing facility admission, similar to findings from the ICD-9-CM era. Studies of medical interventions in older adults using administrative claims should use validated indices, like the FFI, to mitigate confounding or assess effect-measure modification by frailty.

A method for comparing multiple imputation techniques: A case study on the U.S. national COVID cohort collaborative.

Healthcare datasets obtained from Electronic Health Records have proven to be extremely useful for assessing associations between patients' predictors and outcomes of interest. However, these datasets often suffer from missing values in a high proportion of cases, whose removal may introduce severe bias. Several multiple imputation algorithms have been proposed to attempt to recover the missing information under an assumed missingness mechanism. Each algorithm presents strengths and weaknesses, and there is currently no consensus on which multiple imputation algorithm works best in a given scenario. Furthermore, the selection of each algorithm's parameters and data-related modeling choices are also both crucial and challenging. In this paper we propose a novel framework to numerically evaluate strategies for handling missing data in the context of statistical analysis, with a particular focus on multiple imputation techniques. We demonstrate the feasibility of our approach on a large cohort of type-2 diabetes patients provided by the National COVID Cohort Collaborative (N3C) Enclave, where we explored the influence of various patient characteristics on outcomes related to COVID-19. Our analysis included classic multiple imputation techniques as well as simple complete-case Inverse Probability Weighted models. Extensive experiments show that our approach can effectively highlight the most promising and performant missing-data handling strategy for our case study. Moreover, our methodology allowed a better understanding of the behavior of the different models and of how it changed as we modified their parameters. Our method is general and can be applied to different research fields and on datasets containing heterogeneous types.

Healthcare system engagement and algorithm-identified cancer incidence following initiation of a new medication.

PURPOSE: Implausibly high algorithm-identified cancer incidence within a new user study after medication initiation may result from increased healthcare utilization (HU) around initiation ("catch-up care") that increases diagnostic opportunity. Understanding the relationships between HU prior to and around initiation and subsequent cancer rates and timing is important to avoiding protopathic bias. METHODS: We identified a cohort of 417 458 Medicare beneficiaries (2007-2014) aged ≥66 initiating an antihypertensive (AHT) after ≥180 days of non-use. Initiators were stratified into groups of 0, 1, 2-3, and ≥4 outpatient visits (OV) 60-360 days before initiation. We calculated algorithm-identified colorectal cancer (aiCRC) rates stratified by OVs and time since AHT initiation: (0-90, 91-180, 181-365, 366-730, and 731+ days). We summarized HU -360/+60 days around AHT initiation by aiCRC timing: (0-29, 30-89, 90-179, and ≥180 days). RESULTS: AiCRC incidence (311 per 100 000 overall) peaked in the first 0-90 days, was inversely associated with HU before initiation, and stabilized ≥180 days after AHT initiation. Catch-up care was greatest among persons with aiCRCs identified <30 days in follow-up. Catch-up care magnitude decreased as time to the aiCRC date increased, with aiCRCs identified ≥180 days after AHT initiation exhibiting similar HU compared with the full cohort. CONCLUSION: Lower HU before-and increased HU around AHT initiation-seem to drive excess short-term aiCRC incidence. Person-time and case accrual should only begin when incidence stabilizes. When comparison groups within a study differ by HU, outcome-detection bias may exist. Similar observations may exist in other settings when typical HU is delayed (e.g., cancer screening during SARS-CoV-2).

Impacts of Initial Prescription Length and Prescribing Limits on Risk of Prolonged Postsurgical Opioid Use.

BACKGROUND: In response to concerns about opioid addiction following surgery, many states have implemented laws capping the days supplied for initial postoperative prescriptions. However, few studies have examined changes in the risk of prolonged opioid use associated with the initial amount prescribed. OBJECTIVE: The objective of this study was to estimate the risk of prolonged opioid use associated with the length of initial opioid prescribed and the potential impact of prescribing limits. RESEARCH DESIGN: Using Medicare insurance claims (2007-2017), we identified opioid-naive adults undergoing surgery. Using G-computation methods with logistic regression models, we estimated the risk of prolonged opioid use (≥1 opioid prescription dispensed in 3 consecutive 30-d windows following surgery) associated with the varying initial number of days supplied. We then estimate the potential reduction in cases of prolonged opioid use associated with varying prescribing limits. RESULTS: We identified 1,060,596 opioid-naive surgical patients. Among the 70.0% who received an opioid for postoperative pain, 1.9% had prolonged opioid use. The risk of prolonged use increased from 0.7% (1 d supply) to 4.4% (15+ d). We estimated that a prescribing limit of 4 days would be associated with a risk reduction of 4.84 (3.59, 6.09)/1000 patients and would be associated with 2255 cases of prolonged use potentially avoided. The commonly used day supply limit of 7 would be associated with a smaller reduction in risk [absolute risk difference=2.04 (-0.17, 4.25)/1000]. CONCLUSIONS: The risk of prolonged opioid use following surgery increased monotonically with increasing prescription duration. Common prescribing maximums based on days supplied may impact many patients but are associated with relatively low numbers of reduced cases of prolonged use. Any prescribing limits need to be weighed against the need for adequate pain management.

Core concepts in pharmacoepidemiology: Confounding by indication and the role of active comparators.

Confounding by indication poses a significant threat to the validity of nonexperimental studies assessing effectiveness and safety of medical interventions. While no different from other forms of confounding in theory, confounding by indication often requires specific methods to address the bias it creates in addition to common epidemiological adjustment or restriction methods. Clinical indication influencing treatment prescription is patient-specific and complex, making it challenging to measure within nonexperimental research. Restriction of the study population to patients with the indication for treatment would effectively mitigate confounding by indication and bring about comparability between exposure and comparator populations with respect to probability of the exposure. Active comparators are often an effective practical solution to restrict the study population in this manner when indication cannot be measured accurately. This article discusses various forms of confounding by indication, the utility of active comparators for nonexperimental studies of treatment effects, and the active comparator, new user (ACNU) study design to implicitly condition on indication. Considerations for selecting active comparators and conducting an ACNU study design are discussed to enable increased adoption of these methods, improve quality of nonexperimental studies, and ultimately strengthen our evidence base for intended and unintended treatment effects in relevant target populations.

Considerations for observational study design: Comparing the evidence of opioid use between electronic health records and insurance claims.

PURPOSE: Pharmacoepidemiology studies often use insurance claims and/or electronic health records (EHR) to capture information about medication exposure. The choice between these data sources has important implications. METHODS: We linked EHR from a large academic health system (2015-2017) to Medicare insurance claims for patients undergoing surgery. Drug utilization was characterized based on medication order dates in the EHR, and prescription fill dates in Medicare claims. We compared opioid use documented in EHR orders to prescription claims in four time periods: 1) Baseline (182 days before surgery); 2) Perioperative period; 3) Discharge date; 4) Follow-up (90 days after surgery). RESULTS: We identified 11 128 patients undergoing surgery. During baseline, 34.4% (EHR) versus 44.1% (claims) had evidence of opioid use, and 56.9% of all baseline use was reflected only in one data source. During the perioperative period, 78.8% (EHR) versus 47.6% (claims) had evidence of use. On the day of discharge, 59.6% (EHR) versus 45.5% (claims) had evidence of use, and 51.8% of all discharge use was reflected only in one data source. During follow-up, 4.3% (EHR) versus 10.4% (claims) were identified with prolonged opioid use following surgery with 81.4% of all prolonged use reflected only in one data source. CONCLUSIONS: When characterizing opioid exposure, we found substantial discrepancies between EHR medication orders and prescription claims data. In all time periods assessed, most patients' use was reflected only in the EHR, or only in the claims, not both. The potential for misclassification of drug utilization must be evaluated carefully, and choice of data source may have large impacts on key study design elements.

Alternative analytic and matching approaches for the prevalent new-user design: A simulation study.

PURPOSE: To describe the creation of prevalent new user (PNU) cohorts and compare the relative bias and computational efficiency of several alternative analytic and matching approaches in PNU studies. METHODS: In a simulated cohort, we estimated the effect of a treatment of interest vs a comparator among those who switched to the treatment of interest using the originally proposed time-conditional propensity score (TCPS) matching, standardized morbidity ratio weighting (SMRW), disease risk scores (DRS), and several alternative propensity score matching approaches. For each analytic method, we compared the average RR (across 2000 replicates) to the known risk ratio (RR) of 1.00. RESULTS: SMRW and DRS yielded unbiased results (RR = 0.998 and 0.997, respectively). TCPS matching with replacement was also unbiased (RR = 0.999). TCPS matching without replacement was unbiased when matches were identified starting with patients with the shortest treatment history as initially proposed (RR = 0.999), but it resulted in very slight bias (RR = 0.983) when starting with patients with the longest treatment history. Similarly, creating a match pool without replacement starting with patients with the shortest treatment history yielded an unbiased estimate (RR = 0.997), but matching with the longest treatment history first resulted in substantial bias (RR = 0.903). The most biased strategy was matching after selecting one random comparator observation per individual that continued on the comparator (RR = 0.802). CONCLUSIONS: Multiple analytic methods can estimate treatment effects without bias in a PNU cohort. Still, researchers should be wary of introducing bias when selecting controls for complex matching strategies beyond the initially proposed TCPS.

A systematic review of whether COVID-19 randomized controlled trials reported on demographic and clinical characteristics.

PURPOSE: We aim to assess the reporting of key patient-level demographic and clinical characteristics among COVID-19 related randomized controlled trials (RCTs). METHODS: We queried English-language articles from PubMed, Web of Science, clinicaltrials.gov, and the CDC library of gray literature databases using keywords of "coronavirus," "covid," "clinical trial" and "randomized controlled trial" from January 2020 to June 2021. From the search, we conducted an initial review to rule-out duplicate entries, identify those that met inclusion criteria (i.e., had results), and exclude those that did not meet the definition of an RCT. Lastly, we abstracted the demographic and clinical characteristics reported on within each RCT. RESULTS: From the initial 43 627 manuscripts, our final eligible manuscripts consisted of 149 RCTs described in 137 articles. Most of the RCTs (113/149) studied potential treatments, while fewer studied vaccines (29), prophylaxis strategies (5), and interventions to prevent transmission among those infected (2). Study populations ranged from 10 to 38 206 participants (median = 100, IQR: 60-300). All 149 RCTs reported on age, 147 on sex, 50 on race, and 110 on the prevalence of at least one comorbidity. No RCTs reported on income, urban versus rural residence, or other indicators of socioeconomic status (SES). CONCLUSIONS: Limited reporting on race and other markers of SES make it difficult to draw conclusions about specific external target populations without making strong assumptions that treatment effects are homogenous. These findings highlight the need for more robust reporting on the clinical and demographic profiles of patients enrolled in COVID-19 related RCTs.