Full adherence to cirrhosis quality indicators is associated with lower mortality in acute variceal bleeding: Nationwide audit.

BACKGROUND AND AIMS: Acute variceal bleeding (AVB) is a major complication in patients with cirrhosis. Using a nationwide AVB audit, we performed a nested cohort study to determine whether full adherence to the AVB quality indicator (QI) improves clinical outcomes in patients with cirrhosis and AVB. APPROACH AND RESULTS: We assessed real-world adherence to AVB QI among patients with cirrhosis admitted for AVB in all public hospitals in Singapore between January 2015 and December 2020. Full adherence was considered when all 5 QIs were fulfilled: prophylactic antibiotics, vasoactive agents, timely endoscopy, endoscopic hemostasis during index endoscopy, and nonselective beta-blockers after AVB. We compare 6-week mortality between the full adherence and suboptimal adherence groups using a propensity-matched cohort.A total of 989 patients with AVB were included. Full adherence to all AVB QI was suboptimal (56.5%). Analysis of the propensity-matched cohort with comparable baseline characteristics showed that full adherence was associated with a lower risk of early infection (20.0% vs. 26.9%), early rebleeding (5.2% vs. 10.2%), and mortality at 6 weeks (8.2% vs. 19.7%) and 1 year (21.3% vs. 35.4%) ( p <0.05 for all). While full adherence was associated with a lower 6-week mortality regardless of the MELD score, nonadherence was associated with a higher 6-week mortality despite a lower predicted risk of 6-week mortality. Despite high adherence to the recommended process measures, patients with CTP-C remain at a higher risk of rebleeding, 6-week and 1-year mortality. CONCLUSIONS: Full adherence to the AVB QI should be the target for quality improvement in patients with cirrhosis.

Glucagon-Like Peptide-1 Receptor Agonists and Risk for Suicidal Ideation and Behaviors in U.S. Older Adults With Type 2 Diabetes : A Target Trial Emulation Study.

BACKGROUND: A major concern has recently emerged about a potential link between glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and increased risk for suicidal ideation and behaviors based on International Classification of Diseases codes. OBJECTIVE: To investigate the association between GLP-1 RAs, compared with sodium-glucose cotransporter-2 inhibitors (SGLT2is) or dipeptidyl peptidase-4 inhibitors (DPP4is), and risk for suicidal ideation and behaviors in older adults with type 2 diabetes (T2D). DESIGN: Two target trial emulation studies comparing propensity score (PS)-matched cohorts for GLP-1 RAs versus SGLT2is and GLP-1 RAs versus DPP4is. SETTING: U.S. national Medicare administrative data from January 2017 to December 2020. PATIENTS: Older adults (≥66 years) with T2D; no record of suicidal ideation or behaviors; and a first prescription for a GLP-1 RA, SGLT2i, or DPP4i. MEASUREMENTS: The primary end point was a composite of suicidal ideation and behaviors. New GLP-1 RA users were matched 1:1 on PS to new users of an SGLT2i or DPP4i in each pairwise comparison. A Cox proportional hazards regression was used to estimate the hazard ratio (HR) and 95% CIs within matched groups. RESULTS: This study included 21 807 pairs of patients treated with a GLP-1 RA versus an SGLT2i and 21 402 pairs of patients treated with a GLP-1 RA versus a DPP4i. The HR of suicidal ideation and behaviors associated with GLP-1 RAs relative to SGLT2is was 1.07 (95% CI, 0.80 to 1.45; rate difference, 0.16 [CI, -0.53 to 0.86] per 1000 person-years); the HR relative to DPP4is was 0.94 (CI, 0.71 to 1.24; rate difference, -0.18 [CI, -0.92 to 0.57] per 1000 person-years). LIMITATIONS: Low event rate; imprecise estimates; unmeasured confounders, such as body mass index; and potential misclassification of outcomes. CONCLUSION: Among Medicare beneficiaries with T2D, this study found no clear increased risk for suicidal ideation and behaviors with GLP-1 RAs, although estimates were imprecise and a modest adverse risk could not be ruled out. PRIMARY FUNDING SOURCE: American Foundation for Pharmaceutical Education, Pharmaceutical Research and Manufacturers of America Foundation, National Institute on Aging, and National Institute of Diabetes and Digestive and Kidney Diseases.

Morbidity and Mortality of Hospital-Onset SARS-CoV-2 Infections Due to Omicron Versus Prior Variants : A Propensity-Matched Analysis.

BACKGROUND: Many hospitals have scaled back measures to prevent nosocomial SARS-CoV-2 infection given large decreases in the morbidity and mortality of SARS-CoV-2 infections for most people. Little is known, however, about the morbidity and mortality of nosocomial SARS-CoV-2 infections for hospitalized patients in the Omicron era. OBJECTIVE: To estimate the effect of nosocomial SARS-CoV-2 infection on hospitalized patients' outcomes during the pre-Omicron and Omicron periods. DESIGN: Retrospective matched cohort study. SETTING: 5 acute care hospitals in Massachusetts, December 2020 to April 2023. PATIENTS: Adults testing positive for SARS-CoV-2 on or after hospital day 5, after negative SARS-CoV-2 test results on admission and on hospital day 3, were matched to control participants by hospital, service, time period, days since admission, and propensity scores that incorporated demographics, comorbid conditions, vaccination status, primary diagnosis category, vital signs, and laboratory test values. MEASUREMENTS: Primary outcomes were hospital mortality and time to discharge. Secondary outcomes were intensive care unit (ICU) admission, need for advanced oxygen support, discharge destination, hospital-free days, and 30-day readmissions. RESULTS: There were 274 cases of hospital-onset SARS-CoV-2 infection during the pre-Omicron period and 1037 cases during the Omicron period (0.17 vs. 0.49 cases per 100 admissions). Patients with hospital-onset SARS-CoV-2 infection were older and had more comorbid conditions than those without. During the pre-Omicron period, hospital-onset SARS-CoV-2 infection was associated with increased risk for ICU admission, increased need for high-flow oxygen, longer time to discharge (median difference, 4.7 days [95% CI, 2.9 to 6.6 days]), and higher mortality (risk ratio, 2.0 [CI, 1.1 to 3.8]) versus matched control participants. During the Omicron period, hospital-onset SARS-CoV-2 infection remained associated with increased risk for ICU admission and increased time to discharge (median difference, 4.2 days [CI, 3.6 to 5.0 days]). The association with increased hospital mortality was attenuated but still significant (risk ratio, 1.6 [CI, 1.2 to 2.3]). LIMITATION: Residual confounding may be present. CONCLUSION: Hospital-onset SARS-CoV-2 infection during the Omicron period remains associated with increased morbidity and mortality. PRIMARY FUNDING SOURCE: Harvard Medical School Department of Population Medicine.

Comparative Effectiveness and Safety of Apixaban, Rivaroxaban, and Warfarin in Patients With Cirrhosis and Atrial Fibrillation : A Nationwide Cohort Study.

BACKGROUND: Apixaban, rivaroxaban, and warfarin have shown benefit for preventing major ischemic events, albeit with increased bleeding risk, among patients in the general population with atrial fibrillation (AF). However, data are scarce in patients with cirrhosis and AF. OBJECTIVE: To compare the effectiveness and safety of apixaban versus rivaroxaban and versus warfarin in patients with cirrhosis and AF. DESIGN: Population-based cohort study. SETTING: Two U.S. claims data sets (Medicare and Optum's de-identified Clinformatics Data Mart Database [2013 to 2022]). PARTICIPANTS: 1:1 propensity score (PS)-matched patients with cirrhosis and nonvalvular AF initiating use of apixaban, rivaroxaban, or warfarin. MEASUREMENTS: Primary outcomes included ischemic stroke or systemic embolism and major hemorrhage (intracranial hemorrhage or major gastrointestinal bleeding). Database-specific and pooled PS-matched rate differences (RDs) per 1000 person-years (PY) and Cox proportional hazard ratios (HRs) with 95% CIs were estimated, controlling for 104 preexposure covariates. RESULTS: Rivaroxaban initiators had significantly higher rates of major hemorrhagic events than apixaban initiators (RD, 33.1 per 1000 PY [95% CI, 12.9 to 53.2 per 1000 PY]; HR, 1.47 [CI, 1.11 to 1.94]) but no significant differences in rates of ischemic events or death. Consistently higher rates of major hemorrhage were found with rivaroxaban across subgroup and sensitivity analyses. Warfarin initiators also had significantly higher rates of major hemorrhage than apixaban initiators (RD, 26.1 per 1000 PY [CI, 6.8 to 45.3 per 1000 PY]; HR, 1.38 [CI, 1.03 to 1.84]), particularly hemorrhagic stroke (RD, 9.7 per 1000 PY [CI, 2.2 to 17.2 per 1000 PY]; HR, 2.85 [CI, 1.24 to 6.59]). LIMITATION: Nonrandomized treatment selection. CONCLUSION: Among patients with cirrhosis and nonvalvular AF, initiators of rivaroxaban versus apixaban had significantly higher rates of major hemorrhage and similar rates of ischemic events and death. Initiation of warfarin versus apixaban also contributed to significantly higher rates of major hemorrhagic events, including hemorrhagic stroke. PRIMARY FUNDING SOURCE: National Institutes of Health.

Sodium-Glucose Cotransporter-2 Inhibitors, Dulaglutide, and Risk for Dementia : A Population-Based Cohort Study.

BACKGROUND: Both sodium-glucose cotransporter-2 (SGLT2) inhibitors and glucagon-like peptide-1 receptor agonists (GLP-1 RAs) may have neuroprotective effects in patients with type 2 diabetes (T2D). However, their comparative effectiveness in preventing dementia remains uncertain. OBJECTIVE: To compare the risk for dementia between SGLT2 inhibitors and dulaglutide (a GLP-1 RA). DESIGN: Target trial emulation study. SETTING: Nationwide health care data of South Korea obtained from the National Health Insurance Service between 2010 and 2022. PATIENTS: Patients aged 60 years or older who have T2D and are initiating treatment with SGLT2 inhibitors or dulaglutide. MEASUREMENTS: The primary outcome was the presumed clinical onset of dementia. The date of onset was defined as the year before the date of dementia diagnosis, assuming that the time between the onset of dementia and diagnosis was 1 year. The 5-year risk ratios and risk differences comparing SGLT2 inhibitors with dulaglutide were estimated in a 1:2 propensity score-matched cohort adjusted for confounders. RESULTS: Overall, 12 489 patients initiating SGLT2 inhibitor treatment (51.9% dapagliflozin and 48.1% empagliflozin) and 1075 patients initiating dulaglutide treatment were included. In the matched cohort, over a median follow-up of 4.4 years, the primary outcome event occurred in 69 participants in the SGLT2 inhibitor group and 43 in the dulaglutide group. The estimated risk difference was -0.91 percentage point (95% CI, -2.45 to 0.63 percentage point), and the estimated risk ratio was 0.81 (CI, 0.56 to 1.16). LIMITATION: Residual confounding is possible; there was no adjustment for hemoglobin A1c levels or duration of diabetes; the study is not representative of newer drugs, including more effective GLP-1 RAs; and the onset of dementia was not measured directly. CONCLUSION: Under conventional statistical criteria, a risk for dementia between 2.5 percentage points lower and 0.6 percentage point greater for SGLT2 inhibitors than for dulaglutide was estimated to be highly compatible with the data from this study. However, whether these findings generalize to newer GLP-1 RAs is uncertain. Thus, further studies incorporating newer drugs within these drug classes and better addressing residual confounding are required. PRIMARY FUNDING SOURCE: Ministry of Food and Drug Safety of South Korea.

Differentiation of Prior SARS-CoV-2 Infection and Postacute Sequelae by Standard Clinical Laboratory Measurements in the RECOVER Cohort.

BACKGROUND: There are currently no validated clinical biomarkers of postacute sequelae of SARS-CoV-2 infection (PASC). OBJECTIVE: To investigate clinical laboratory markers of SARS-CoV-2 and PASC. DESIGN: Propensity score-weighted linear regression models were fitted to evaluate differences in mean laboratory measures by prior infection and PASC index (≥12 vs. 0). (ClinicalTrials.gov: NCT05172024). SETTING: 83 enrolling sites. PARTICIPANTS: RECOVER-Adult cohort participants with or without SARS-CoV-2 infection with a study visit and laboratory measures 6 months after the index date (or at enrollment if >6 months after the index date). Participants were excluded if the 6-month visit occurred within 30 days of reinfection. MEASUREMENTS: Participants completed questionnaires and standard clinical laboratory tests. RESULTS: Among 10 094 participants, 8746 had prior SARS-CoV-2 infection, 1348 were uninfected, 1880 had a PASC index of 12 or higher, and 3351 had a PASC index of zero. After propensity score adjustment, participants with prior infection had a lower mean platelet count (265.9 × 109 cells/L [95% CI, 264.5 to 267.4 × 109 cells/L]) than participants without known prior infection (275.2 × 109 cells/L [CI, 268.5 to 282.0 × 109 cells/L]), as well as higher mean hemoglobin A1c (HbA1c) level (5.58% [CI, 5.56% to 5.60%] vs. 5.46% [CI, 5.40% to 5.51%]) and urinary albumin-creatinine ratio (81.9 mg/g [CI, 67.5 to 96.2 mg/g] vs. 43.0 mg/g [CI, 25.4 to 60.6 mg/g]), although differences were of modest clinical significance. The difference in HbA1c levels was attenuated after participants with preexisting diabetes were excluded. Among participants with prior infection, no meaningful differences in mean laboratory values were found between those with a PASC index of 12 or higher and those with a PASC index of zero. LIMITATION: Whether differences in laboratory markers represent consequences of or risk factors for SARS-CoV-2 infection could not be determined. CONCLUSION: Overall, no evidence was found that any of the 25 routine clinical laboratory values assessed in this study could serve as a clinically useful biomarker of PASC. PRIMARY FUNDING SOURCE: National Institutes of Health.

Nirmatrelvir/Ritonavir Use and Hospitalizations or Death in a Previously Uninfected Nonhospitalized High-Risk Population With COVID-19: A Matched Cohort Study.

OBJECTIVE: To determine the association of nirmatrelvir/ritonavir (NMV/r) with hospitalization or death within 30 days as compared with untreated controls previously uninfected and nonhospitalized. METHODS: We used a matched cohort design using inverse probability of treatment weight (IPTW). Individuals prescribed NMV/r within 3 days of COVID-19 diagnosis were compared with IPTW-based untreated controls. Variables for IPTW included age, race, sex, body mass index, geographic location, vaccination status, and multiple comorbidities. Additional analyses were conducted on NMV/r-treated and propensity score-matched untreated controls. RESULTS: Among 7615 individuals prescribed NMV/r and 62 077 controls identified between 1 January 2022 and 25 February 2023, the risk of hospitalization/death was lower among NMV/r-treated persons vs untreated controls (243 vs 3468 events; absolute risk difference [ARD], -2.36 [95% CI, -2.57 to -2.14]). The difference was significant for those >60 and ≤60 years old (ARD, -3.86 [95% CI, -4.19 to -3.54] vs -0.27 [95% CI, -0.51 to -0.03]) and for persons asymptomatic and symptomatic (ARD, -7.09 [95% CI, -7.62 to -6.55] vs -1.46 [95% CI, -1.66 to -1.25]). Significant benefit was observed among individuals unvaccinated and vaccinated, with or without a booster dose. CONCLUSIONS: NMV/r is associated with a significant reduction in 30-day hospitalization or death among individuals previously uninfected and nonhospitalized.

Optimal glycaemic control and the reduced risk of colorectal adenoma and cancer in patients with diabetes: a population-based cohort study.

OBJECTIVE: Whether varying degrees of glycaemic control impact colonic neoplasm risk in patients with diabetes mellitus (DM) remains uncertain. DESIGN: Patients with newly diagnosed DM were retrieved from 2005 to 2013. Optimal glycaemic control at baseline was defined as mean haemoglobin A1c (HbA1c)<7%. Outcomes of interest included colorectal cancer (CRC) and colonic adenoma development. We used propensity score (PS) matching with competing risk models to estimate subdistribution HRs (SHRs). We further analysed the combined effect of baseline and postbaseline glycaemic control based on time-weighted mean HbA1c during follow-up. RESULTS: Of 88 468 PS-matched patients with DM (mean (SD) age: 61.5 (±11.7) years; male: 47 127 (53.3%)), 1229 (1.4%) patients developed CRC during a median follow-up of 7.2 (IQR: 5.5-9.4) years. Optimal glycaemic control was associated with lower CRC risk (SHR 0.72; 95% CI 0.65 to 0.81). The beneficial effect was limited to left-sided colon (SHR 0.71; 95% CI 0.59 to 0.85) and rectum (SHR 0.71; 95% CI 0.57 to 0.89), but not right-sided colon (SHR 0.86; 95% CI 0.67 to 1.10). Setting suboptimal glycaemic control at baseline/postbaseline as a reference, a decreased CRC risk was found in optimal control at postbaseline (SHR 0.79), baseline (SHR 0.71) and both time periods (SHR 0.61). Similar associations were demonstrated using glycaemic control as a time-varying covariate (HR 0.75). A stepwise greater risk of CRC was found (Ptrend<0.001) with increasing HbA1c (SHRs 1.34, 1.30, 1.44, 1.58 for HbA1c 7.0% to <7.5%, 7.5% to <8.0%, 8.0% to <8.5% and ≥8.5%, respectively). Optimal glycaemic control was associated with a lower risk of any, non-advanced and advanced colonic adenoma (SHRs 0.73-0.87). CONCLUSION: Glycaemic control in patients with DM was independently associated with the risk of colonic adenoma and CRC development with a biological gradient.

Association of glucagon-like peptide-1 receptor agonists with suicidal ideation and self-injury in individuals with diabetes and obesity: a propensity-weighted, population-based cohort study.

AIMS/HYPOTHESIS: Regulators worldwide are reviewing safety data on glucagon-like peptide-1 receptor agonists (GLP-1RA), following reports by the Icelandic Medicines Agency in July 2023 of suicidal ideation and self-injury (SIS) in individuals taking liraglutide and semaglutide. We aimed to assess the risk of SIS in new users of GLP-1RA when compared with sodium-glucose cotransporter 2 inhibitors (SGLT-2i) users, prescribed to treat type 2 diabetes in individuals with obesity. METHODS: This is a cohort study combining several population-wide databases and covering a Spanish population of five million inhabitants, including all adults with obesity who initiated treatment with either GLP-1RA or SGLT-2i for type 2 diabetes from 2015 to 2021. To estimate the comparative effect of GLP-1RA on the risk of SIS, we employed a new user, active comparator design and we carried out multivariable Cox regression modelling with inverse probability of treatment weighting (IPTW) based on propensity scores. We performed several stratified and sensitivity analyses. RESULTS: We included 3040 patients initiating treatment with GLP-1RA and 11,627 with SGLT-2i. When compared with patients treated with SGLT-2i, those in the GLP-1RA group were younger (55 vs 60 years old, p<0.001), had more anxiety (49.4% vs 41.5%, p<0.001), sleep disorders (43.2% vs 34.1%, p<0.001) and depression (24.4% vs 19.0%, p<0.001), and were more obese (35.1% of individuals with BMI ≥40 vs 15.1%, p<0.001). After propensity score weighting, standardised mean differences between groups were <0.1 for all covariates, showing adequate balance between groups at baseline after adjustment. In the main per-protocol analyses we found no evidence that GLP-1RA increased the incidence of SIS (HR 1.04; 95% CI 0.35, 3.14). Intention-to-treat analyses resulted in an HR of 1.36 (95% CI 0.51, 3.61). In analyses excluding individuals with no BMI information and using imputation for BMI missing values, respective HRs were 0.89 (95% CI 0.26, 3.14) and 1.29 (95% CI 0.42, 3.92). Stratified analyses showed no differences between subgroups. CONCLUSIONS/INTERPRETATION: Our findings do not support an increased risk of SIS when taking GLP-1RA in individuals with type 2 diabetes and obesity; however, the rarity of SIS events and the wide uncertainty of effect size (although null, effect may be compatible with a risk as high as threefold) calls for a cautious interpretation of our results. Further studies, including final evaluations from regulatory bodies, are called for to discard a causal link between GLP-1RA and suicidality.

Higher Risk of Mental Illness in Patients With Diagnosed and Untreated Unruptured Intracranial Aneurysm: Findings From a Nationwide Cohort Study.

BACKGROUND: We aimed to investigate the association between a diagnosis of untreated unruptured intracranial aneurysms (UIAs) and the development of mental illness. METHODS: This retrospective, propensity-score-matched cohort study was based on the nationwide South Korean database. The UIA diagnosis group included participants newly diagnosed with UIA between 2011 and 2019. For a well-matched control group, patients diagnosed with an acute upper respiratory infection but without UIA during the same period were selected through 1:4 matching based on propensity scores, which were calculated using age, sex, economic status, and comorbidities. The study's outcome measure encompassed the incidence of mental illnesses over a 10-year period, using International Classification of Diseases-Tenth Revision codes for anxiety, stress, depressive, bipolar, and eating disorders, insomnia, and alcohol or drug misuse. RESULTS: After propensity score matching, 85 438 participants with untreated UIAs (50.75% male; average age, 56.41 [±13.82] years; follow-up, 4.21 [±2.56] years) and 331 123 controls (49.44% males; average age, 56.69 [±12.92] years; follow-up, 7.48 [±2.12] years) were compared. Incidence rate of mental illness was higher in the UIA group (113.07 versus 90.41 per 1000 person-years; hazard ratio, 1.104 [95% CI, 1.089-1.119]). The risk of mental illness varied slightly by sex (males: hazard ratio, 1.131 [95% CI, 1.108-1.155]; females: hazard ratio, 1.082 [95% CI, 1.063-1.103]). Hazard ratios showed a U-shaped relationship with age, peaking in younger age groups, decreasing in middle-aged groups, and slightly increasing in older age groups, especially in patients with severe mental illness receiving psychotherapy. CONCLUSIONS: Our findings indicate a higher risk of mental illness in patients with UIA diagnosis in specific demographic groups, suggesting a possible psychological burden associated with UIAs. Clinicians treating cerebral aneurysms should be aware that the psychological burden caused by the diagnosis of UIA itself could contribute to mental illness and strive to provide comprehensive care for these patients.

Oral Antibiotics for Treatment of Gram-Negative Bacteremia in Solid Organ Transplant Recipients: A Propensity Score Weighted Retrospective Observational Study.

BACKGROUND: We assessed the safety and efficacy of oral antibiotic step-down therapy for uncomplicated gram-negative blood stream infections in solid-organ transplant recipients. METHODS: We identified all solid-organ transplant recipients within the Massachusetts General and Brigham and Women's Hospital systems from 2016 to 2021 with uncomplicated gram-negative bacteremia involving an organism susceptible to an acceptably bioavailable oral antibiotic agent. Using inverse probability of treatment-weighted models based on propensity scores adjusting for potential clinical confounders, we compared outcomes of those transitioned to oral antibiotics with those who continued intravenous (IV) therapy for the duration of treatment. Primary endpoints were mortality, bacteremia recurrence, and reinitiation of IV antibiotics. Secondary endpoints included length of stay, Clostridioides difficile infection, treatment-associated complications, and tunneled central venous catheter placement. RESULTS: A total of 120 bacteremia events from 107 patients met inclusion criteria in the oral group and 42 events from 40 patients in the IV group. There were no significant differences in mortality, bacteremia recurrence, or reinitiation of IV antibiotics between groups. Patients transitioned to oral antibiotics had an average length of stay that was 1.97 days shorter (95% confidence interval [CI], -.39 to 3.56 days; P = .005). Odds of developing C. difficile and other treatment-associated complications were 8.4 times higher (95% CI, 1.5-46.6; P = .015) and 6.4 times higher (95% CI, 1.9-20.9; P = .002), respectively, in the IV group. Fifty-five percent of patients in the IV group required tunneled catheter placement. There was no difference in treatment duration between groups. CONCLUSIONS: Oral step-down therapy was effective and associated with fewer treatment-related adverse events.

Characterizing Imbalance in the Tails of the Propensity Score Distribution.

Understanding characteristics of patients with propensity scores in the tails of the propensity score (PS) distribution has relevance for inverse-probability-of-treatment-weighted and PS-based estimation in observational studies. Here we outline a method for identifying variables most responsible for extreme propensity scores. The approach is illustrated in 3 scenarios: 1) a plasmode simulation of adult patients in the National Ambulatory Medical Care Survey (2011-2015) and 2) timing of dexamethasone initiation and 3) timing of remdesivir initiation in patients hospitalized for coronavirus disease 2019 from February 2020 through January 2021. PS models were fitted using relevant baseline covariates, and tails of the PS distribution were defined using asymmetric first and 99th percentiles. After fitting of the PS model in each original data set, values of each key covariate were permuted and model-agnostic variable importance measures were examined. Visualization and variable importance techniques were helpful in identifying variables most responsible for extreme propensity scores and may help identify individual characteristics that might make patients inappropriate for inclusion in a study (e.g., off-label use). Subsetting or restricting the study sample based on variables identified using this approach may help investigators avoid the need for trimming or overlap weights in studies.

Invited commentary: mixing multiple imputation and bootstrapping for variance estimation.

Multiple imputation (MI) is commonly implemented to mitigate potential selection bias due to missing data. The accompanying article by Nguyen and Stuart (Am J Epidemiol. 2024;193(10):1470-1476) examines the statistical consistency of several ways of integrating MI with propensity scores. As Nguyen and Stuart noted, variance estimation for these different approaches remains to be developed. One common option is the nonparametric bootstrap, which can provide valid inference when closed-form variance estimators are not available. However, there is no consensus on how to implement MI and nonparametric bootstrapping in analyses. To complement Nguyen and Stuart's article on MI and propensity score analyses, we review some currently available approaches on variance estimation with MI and nonparametric bootstrapping.

Multiple imputation for propensity score analysis with covariates missing at random: some clarity on "within" and "across" methods.

In epidemiology and the social sciences, propensity score methods are popular for estimating treatment effects using observational data, and multiple imputation is popular for handling covariate missingness. However, how to appropriately use multiple imputation for propensity score analysis is not completely clear. This paper aims to bring clarity on the consistency (or lack thereof) of methods that have been proposed, focusing on the "within" approach (where the effect is estimated separately in each imputed dataset and then the multiple estimates are combined) and the "across" approach (where typically propensity scores are averaged across imputed datasets before being used for effect estimation). We show that the within method is valid and can be used with any causal effect estimator that is consistent in the full-data setting. Existing across methods are inconsistent, but a different across method that averages the inverse probability weights across imputed datasets is consistent for propensity score weighting. We also comment on methods that rely on imputing a function of the missing covariate rather than the covariate itself, including imputation of the propensity score and of the probability weight. Based on consistency results and practical flexibility, we recommend generally using the standard within method. Throughout, we provide intuition to make the results meaningful to the broad audience of applied researchers.

A Cautionary Note on Using Propensity Score Calibration to Control for Unmeasured Confounding Bias When the Surrogacy Assumption Is Absent.

Conventional propensity score methods encounter challenges when unmeasured confounding is present, as it becomes impossible to accurately estimate the gold-standard propensity score when data on certain confounders are unavailable. Propensity score calibration (PSC) addresses this issue by constructing a surrogate for the gold-standard propensity score under the surrogacy assumption. This assumption posits that the error-prone propensity score, based on observed confounders, is independent of the outcome when conditioned on the gold-standard propensity score and the exposure. However, this assumption implies that confounders cannot directly impact the outcome and that their effects on the outcome are solely mediated through the propensity score. This raises concerns regarding the applicability of PSC in practical settings where confounders can directly affect the outcome. While PSC aims to target a conditional treatment effect by conditioning on a subject's unobservable propensity score, the causal interest in the latter case lies in a conditional treatment effect conditioned on a subject's baseline characteristics. Our analysis reveals that PSC is generally biased unless the effects of confounders on the outcome and treatment are proportional to each other. Furthermore, we identify 2 sources of bias: 1) the noncollapsibility of effect measures, such as the odds ratio or hazard ratio and 2) residual confounding, as the calibrated propensity score may not possess the properties of a valid propensity score.

Clinical Features and Outcomes of Hospitalized Adult Patients With Breakthrough COVID-19 Infections: A Propensity-Score-Matched Observational Study.

In this study, we aimed to evaluate the impact of vaccination on intensive care unit (ICU) admission and in-hospital mortality among breakthrough coronavirus disease 2019 (COVID-19) infections. A total of 3,351 adult patients hospitalized with COVID-19 in the Memorial Healthcare System (Hollywood, Florida) between June 1 and September 20, 2021, were included; 284 (8.5%) were fully vaccinated. A propensity-score-matched analysis was conducted to compare fully vaccinated patients with unvaccinated controls. Propensity scores were calculated on the basis of variables associated with vaccination status. A 1:1 matching ratio was applied using logistic regression models, ensuring balanced characteristics between the two groups. The matched samples were then subjected to multivariate analysis. Among breakthrough infections, vaccinated patients demonstrated lower incidences of ICU admission (10.3% vs. 16.4%; P = 0.042) and death (12.2% vs. 18.7%; P = 0.041) than the matched controls. Risk-adjusted multivariate analysis demonstrated a significant inverse association between vaccination and ICU admission (odds ratio = 0.52, 95% confidence interval: 0.31, 0.89; P = 0.019) as well as in-hospital mortality (odds ratio = 0.57, 95% confidence interval: 0.34, 0.94; P = 0.027). Vaccinated individuals experiencing breakthrough infections had significantly lower risks of ICU admission and in-hospital mortality. These findings highlight the benefits of COVID-19 vaccines in reducing severe outcomes among patients with breakthrough infections.

Confounder Adjustment Using the Disease Risk Score: A Proposal for Weighting Methods.

Propensity score analysis is a common approach to addressing confounding in nonrandomized studies. Its implementation, however, requires important assumptions (e.g., positivity). The disease risk score (DRS) is an alternative confounding score that can relax some of these assumptions. Like the propensity score, the DRS summarizes multiple confounders into a single score, on which conditioning by matching allows the estimation of causal effects. However, matching relies on arbitrary choices for pruning out data (e.g., matching ratio, algorithm, and caliper width) and may be computationally demanding. Alternatively, weighting methods, common in propensity score analysis, are easy to implement and may entail fewer choices, yet none have been developed for the DRS. Here we present 2 weighting approaches: One derives directly from inverse probability weighting; the other, named target distribution weighting, relates to importance sampling. We empirically show that inverse probability weighting and target distribution weighting display performance comparable to matching techniques in terms of bias but outperform them in terms of efficiency (mean squared error) and computational speed (up to >870 times faster in an illustrative study). We illustrate implementation of the methods in 2 case studies where we investigate placebo treatments for multiple sclerosis and administration of aspirin in stroke patients.

Efficacy of a hydrogel spacer for improving quality of life in patients with prostate cancer undergoing low-dose-rate brachytherapy alone or in combination with intensity-modulated radiotherapy: An observational study using propensity score matching.

BACKGROUND: It is unclear whether a hydrogel spacer can improve quality of life (QOL) in patients undergoing low-dose-rate brachytherapy (LDR-BT) alone or in combination with intensity-modulated radiotherapy (IMRT). METHODS: We enrolled patients with prostate cancer who underwent LDR-BT alone with (n = 186) or without (n = 348) a hydrogel spacer, or underwent LDR-BT in combination with IMRT with (n = 70) or without (n = 217) a hydrogel spacer. QOL was evaluated using Expanded Prostate Cancer Index Composite (EPIC) questionnaires at baseline and 1, 3, 6, 12, and 24 months after implantation. The groups were compared using propensity score matching analysis. RESULTS: Among patients who underwent LDR-BT alone, there were no differences regarding changes in urinary, bowel, sexual, or hormonal domain scores between the spacer and no-spacer groups; however, the dose at the bowel was significantly lower in the spacer group than in the no-spacer group. Among patients who underwent LDR-BT in combination with IMRT, there were no differences regarding changes in urinary, sexual, or hormonal domain scores between the spacer and no-spacer groups. However, the changes in the bowel domain score were significantly lower in the spacer group than in the no-spacer group (p < 0.001). CONCLUSIONS: A hydrogel spacer may not improve impaired urinary, bowel, or sexual QOL in patients undergoing LDR-BT alone. However, in patients undergoing LDR-BT in combination with IMRT, a hydrogel spacer can improve impaired bowel QOL but not sexual or urinary QOL.

Prostate cancer in transgender women: A propensity score-matched analysis of disease severity and survival.

BACKGROUND: Despite the rise in gender-affirming care, our understanding of prostate cancer (PCa) in transgender women (TGW) remains in its infancy. Health disparities and lack of PCa awareness and screening are possible barriers to providing quality care for this population. In addition, the implication of hormonal manipulation for the aggressiveness of PCa in TGW is yet to be determined. Here, this study sought to compare oncological characteristics and survival outcomes between transgender and cisgender (CG) patients with PCa via two national data sets. METHODS: The Veterans Affairs Informatics and Computing Infrastructure database (1999-2020) and the Surveillance, Epidemiology, and End Results-Medicare database (2010-2017) were reviewed. Demographic and clinical details were analyzed. Logistic regression analysis was performed on propensity score-matched groups to identify predictors of high-risk disease and metastasis in patients with PCa. Groups were matched 5:1 (CG:TGW) on the basis of age, race, year of diagnosis, and Charlson Comorbidity Index score. Primary outcomes included metastatic presentation, high-risk localized disease, overall survival (OS), and prostate cancer-specific mortality (PCSM). RESULTS: A total of 1194 patients were included (199 TGW; 995 CG). Associations between transgender identity and metastatic presentation (odds ratio [OR], 0.38; p = .2), high-risk localized disease (OR, 1.19; p = .50), or PCSM (hazard ratio [HR], 0.65; p = .3) were not detected. Transgender identity was associated with improved OS (HR, 0.67; p = .014). CONCLUSIONS: PCa-specific outcomes seem comparable between TGW and CG men, although the study was underpowered to detect modest differences. Further investigation into the incidence and outcomes of PCa in TGW is warranted.

Venetoclax plus a hypomethylating agent versus cytarabine, aclarubicin, and granulocyte colony-stimulating factor chemotherapy as a first-line therapy for newly diagnosed acute myeloid leukemia: A propensity score-matched analysis.

BACKGROUND: Both venetoclax plus a hypomethylating agent (VEN/HMA) and cytarabine, aclarubicin, and granulocyte colony-stimulating factor (CAG) are low-intensity regimens for older patients with acute myeloid leukemia (AML) that show good efficacy and safety. It is unknown how VEN/HMA compares with the CAG regimen for the treatment of newly diagnosed AML. METHODS: The outcomes of patients with newly diagnosed AML treated with VEN/HMA were compared with those of patients treated with a CAG-based regimen. Propensity score matching between these two cohorts at a 1:1 ratio was performed according to age at diagnosis, sex, Eastern Cooperative Oncology Group performance status, state of fitness, and European LeukemiaNet (ELN) 2022 risk stratification to minimize bias. RESULTS: A total of 84 of 96 patients in the VEN/HMA cohort were matched with 84 of 147 patients in the CAG cohort. VEN/HMA resulted in a better response than the CAG-based regimens, as indicated by a higher composite complete remission (CRc) rate (82.1% vs. 60.7%; p = .002) and minimal residual disease negativity rate (88.2% vs. 68.2%; p = .009). In patients with an ELN adverse risk, VEN/HMA was associated with a higher CRc rate compared to CAG (80.5% vs. 58.3%; p = .006). VEN/HMA was associated with longer event-free survival (EFS) (median EFS, not reached vs. 4.5 months; p = .0004), whereas overall survival (OS) was comparable between the two cohorts (median OS, not reached vs. 18 months; p = .078). CONCLUSIONS: The VEN/HMA regimen may result in a better response than CAG-based treatment in older patients with newly diagnosed AML.