Adjustment For Baseline Covariates Research Articles

Abstract 4136804: Effect of Obesity on Arrhythmia Recurrence and Complications following Pulmonary Vein Isolation for Paroxysmal Atrial Fibrillation in the REAL-AF Registry

Introduction: Obesity is an important risk factor for atrial fibrillation and its prevalence in the United States is rising. The effect of obesity on outcomes following pulmonary vein isolation in a contemporary setting has not been extensively evaluated. Research Questions: - Is higher body mass index (BMI) associated with increased risk of atrial arrhythmia recurrence following pulmonary vein isolation? - Is higher BMI associated with increased risk of post-procedural complications? Method: - The REAL-AF (Real-world Experience of Catheter Ablation for the Treatment of Symptomatic Paroxysmal and Persistent Atrial Fibrillation; ClinicalTrials.gov Identifier: NCT04088071) Registry is a multicenter, prospective observational registry of patients undergoing radiofrequency catheter ablation of atrial fibrillation. We included 2666 patients with paroxysmal atrial fibrillation undergoing index ablation from January 2018 through April 2023. We excluded patients with missing covariates. - The primary outcome was freedom from atrial arrhythmia following ablation. Rhythm monitoring was standardized using patch monitor at 6 month and 1 year. Time-to-event analysis was performed using a 3-month blanking period. Results: - Baseline characteristics are shown in Table 1. The median follow-up was 382 days (IQR: 363-410). - There was a trend toward higher rate of atrial arrhythmia following ablation for patients in the highest BMI group compared with the reference group (BMI<25) ( BMI≥40: HR = 1.41 (0.94-2.12), p= 0.092). There was no difference in the rate of atrial arrhythmia for other BMI groups ( BMI 25 - <30: HR = 0.90 (0.65-1.23), p = 0.5; BMI 30 - <35: HR = 0.89 (0.64-1.23), p = 0.5; BMI 35 - <40: HR = 0.87 (0.57-1.32), p = 0.5). This trend persisted following adjustment for baseline covariates ( BMI 25 - <30: HR = 1.01 (0.77-1.33), p > 0.9; BMI 30 - <35: HR = 0.97 (0.72-1.32), p = 0.9; BMI 35 - <40: HR = 1.01 (0.66-1.55), p > 0.9; BMI≥40: HR = 1.57 (0.96-2.57), p =0.071). - First-pass isolation rates progressively decreased with higher BMI (Table 2). - Complication rates were low across all BMI groups (Table 2). Conclusions: - Severely obese patients with BMI ≥40 experience higher risk of atrial arrhythmia following radiofrequency catheter ablation of paroxysmal atrial fibrillation compared with normal weight individuals. This corresponds with lower rates of first-pass isolation at higher BMI. - Patients across BMI categories experience similarly low rates of procedural complication.

Long-term clinical outcomes of remote monitoring for implantable cardioverter-defibrillators in Singapore.

Remote monitoring (RM) for cardiac implantable electronic devices is on the rise and has been shown to reduce the burden of in-clinic follow-up visits. We aimed to investigate the long-term clinical outcomes of RM versus no RM. This was a prospective, single-centre cohort study of consecutive patients with an implantable cardioverter-defibrillator (ICD) or cardiac resynchronisation therapy-defibrillator (CRT-D) followed up from 2018 to 2023. Patients who received non-ICD devices were excluded. In general, RM was offered to all patients, but uptake depended on patient preference. For data analysis, patients were stratified according to whether RM was used. The primary outcome was all-cause mortality; secondary outcomes were hospitalisation for heart failure and device therapy (shocks and electrical storm). Of 551 patients, 284 (51.5%) received RM and 267 (49.5%) did not. Baseline demographics were similar between the two arms. All-cause mortality was significantly lower in RM versus non-RM patients (hazard ratio [HR] 0.45, 95% confidence interval [CI] 0.33-0.60, P <0.001), as was hospitalisation for heart failure (HR 0.39, 95% CI 0.25-0.59, P <0.001); these remained significant after adjustment for baseline covariates. More patients on RM received appropriate antitachycardia pacing (ATP) (17.6% vs. 10.7%, P = 0.035) and appropriate shocks (24.1% vs. 14.7%, P = 0.017). The incidences of inappropriate ATP, inappropriate shocks and electrical storm were similar. More patients on RM underwent pulse generator change (34.1% vs. 10.1%, P <0.001). Remote monitoring was associated with significantly lower mortality in both ICDs and CRT-Ds and in primary and secondary indications, as well as fewer heart failure hospitalisations. This supports current guidelines recommending the use of RM in all patients with ICD or CRT-D.

Can algorithms Replace Expert Knowledge for Causal Inference? A Case Study on Novice Use of Causal Discovery.

With growing interest in causal inference and machine learning among epidemiologists, there is increasing discussion of causal discovery algorithms for guiding covariate selection. We present a case study of novice application of causal discovery tools and attempt to validate the results against a well-established causal relationship. As a case study, we attempted causal discovery of relationships relevant to the effect of adherence on mortality in the placebo arm of the Coronary Drug Project (CDP) dataset. We used four algorithms available as existing software implementations and varied several model inputs. We identified 15 adjustment sets from 17 model parameterizations. When applied to a baseline covariate adjustment analysis, these 15 adjustment sets returned effect estimates with similar magnitude and direction of bias as prior published results. When using methods to control for time-varying confounding, there was generally more residual bias than compared to expert-selected adjustment sets. Although causal discovery algorithms can perform on par with expert knowledge, we do not recommend novice use of causal discovery without the input of experts in causal discovery. Expert support is recommended to aid in choosing the algorithm, selecting input parameters, assessing underlying assumptions, and finalizing selection of the adjustment variables.

A Nationwide Target Trial Emulation Assessing the Risk of Antidepressant-Induced Mania Among Patients With Bipolar Depression.

Antidepressants are commonly used to treat bipolar depression but may increase the risk of mania. The evidence from randomized controlled trials, however, is limited by short treatment durations, providing little evidence for the long-term risk of antidepressant-induced mania. The authors performed a target trial emulation to compare the risk of mania among individuals with bipolar depression treated or not treated with antidepressants over a 1-year period. The authors emulated a target trial using observational data from nationwide Danish health registers. The study included 979 individuals with bipolar depression recently discharged from a psychiatric ward. Of these, 358 individuals received antidepressant treatment, and 621 did not. The occurrence of mania and bipolar depression over the following year was ascertained, and the intention-to-treat effect of antidepressants was analyzed by using Cox proportional hazards regression with adjustment for baseline covariates to emulate randomized open-label treatment allocation. The fully adjusted analyses revealed no statistically significant associations between treatment with an antidepressant and the risk of mania in the full sample (hazard rate ratio=1.08, 95% CI=0.72-1.61), in the subsample concomitantly treated with a mood-stabilizing agent (hazard rate ratio=1.16, 95% CI=0.63-2.13), and in the subsample not treated with a mood-stabilizing agent (hazard rate ratio=1.16, 95% CI=0.65-2.07). Secondary analyses revealed no statistically significant association between treatment with an antidepressant and bipolar depression recurrence. These findings suggest that the risk of antidepressant-induced mania is negligible and call for further studies to optimize treatment strategies for individuals with bipolar depression.

#1881 Association of circulating autoantibodies to erythropoietin receptor with anaemia in patients with type 2 diabetes and chronic kidney disease

Abstract Background and Aims Anaemia is a risk factor for cardiovascular and kidney prognosis in patients with type 2 diabetes and it is important to identify patients at high risk of anemia. We have previously reported that autoantibodies to erythropoietin receptor (anti-EPOR antibodies) may functionally neutralize the activity of erythropoietin and are associated with haemoglobin (Hb) in patients with lupus nephritis, and hypo responsiveness to erythropoiesis-stimulating agents (ESAs) in patients undergoing haemodialysis. This study assessed the association between anti-EPOR antibodies and anaemia in patients with type 2 diabetes and chronic kidney disease. Method This cohort study included patients with type 2 diabetes and chronic kidney disease who were diagnosed between 1989 and 2014 at our hospital. The presence of anti-EPOR antibodies at baseline were assessed by ELISA in stocked serum. To investigate the association between anti-EPOR antibodies and anaemia, odds ratios (ORs) and 95% confidence intervals (CIs) were estimated using logistic regression analysis adjusted for multiple covariates. We then estimated hazard ratios (HR) and 95% CIs using Cox regression models to investigate the association between anti-EPOR antibodies and the risk of iron deficiency, defined as transferrin saturation below 20% in those with follow-up measurements of iron metabolism. Results The mean age was 63 years, the mean eGFR was 42 ml/min/1.73 m2 and the median urinary protein to creatinine ratio (UPCR) was 1.8 g/gCr. Anti-EPOR antibodies were detected in 26 out of 110 patients (24%). At baseline, most patients with anti-EPOR antibodies had anaemia (n=24, 92%) and severe anaemia with Hb&amp;lt;10 g/dl (n=16, 62%) and used ESAs (n=16, 62%). The presence of anti-EPOR antibodies was independently associated with severe anaemia (OR 9.3; 1.2-70.1, p=0.03) after adjustment for baseline covariates including HbA1c, eGFR, UPCR, iron, and CRP. During a mean follow-up of 6.5 years, 6 of 10 patients (60%) developed iron deficiency in anti-EPOR antibodies positive patients, while 3 of 20 patients (15%) in negative patients. Patients with anti-EPOR antibodies were associated with an higher risk of iron deficiency compared to those without (HR 76.0; 95% CI 1.2-4721.7, p=0.04). Conclusion In patients with type 2 diabetes and chronic kidney disease, the presence of anti-EPOR antibodies was associated with a risk of severe anaemia and iron deficiency independently of kidney function. These results suggest that anti-EPOR antibodies may be useful in predicting the risk of anaemia and establishing therapeutic strategies of anaemia in people with type 2 diabetes.

Age is not a sole predictor of outcomes in octogenarians undergoing complex endovascular aortic repair

ObjectiveTo examine the perioperative, postoperative, and long-term outcomes of fenestrated/branched endovascular aneurysm repair (F/BEVAR) in octogenarians compared with nonoctogenarians. MethodsA multicenter, retrospective cohort study was conducted using the Vascular Quality Improvement database, which prospectively captures information on patients who undergo vascular surgery across 1021 academic and community hospitals in North America. All patients who underwent F/BEVAR endovascular aortic repair from 2012 to 2022 were included. Patients were stratified into two groups: those aged <80 years and those aged ≥80 years at the time of the procedure. The preoperative, intraoperative, and postoperative factors were compared between the two groups. The primary outcome was long-term all-cause mortality; secondary outcomes included aortic-specific mortality and aortic-specific reintervention. ResultsA total of 6007 patients (aged <80 years, n = 4860; aged ≥80 years, n = 1147) who had undergone F/BEVAR procedures were included. No significant difference was found in technical success, postoperative length of stay, length of intensive care unit stay, postoperative bowel ischemia, and spinal cord ischemia. After adjustment for baseline covariates, octogenarians were more likely to suffer from a postoperative complication (odds ratio [OR]: 1.16; [95% confidence interval (CI): 0.98-1.37], P < .001) and be discharged to a rehabilitation center (OR: 1.60; [95% CI: 1.27-2.00], P < .001) or nursing home (OR: 2.23; [95% CI: 1.64-3.01], P < .001). Five-year survival was lower in octogenarians (83% vs 71%, hazard ratio [HR]: 1.70; [95% CI: 1.46-2.0], P < .0001). Multivariate Cox proportional hazard analysis demonstrated that age was associated with increased all-cause mortality (HR: 1.72, [95% CI: 1.39-2.12], P < .001) and aortic-specific mortality (HR: 1.92, [95% CI: 1.04-3.68], P = .038). Crawford extent II aortic disease was associated with an increase in all-cause mortality (HR 1.49; [95% CI: 1.01-2.19], P < .001), aortic-specific mortality (HR: 5.05; [95% CI: 1.35-18.9], P = .016), and aortic-specific reintervention (HR: 1.91; [95% CI: 1.24-2.93], P = .003). Functional dependence was associated with increased all-cause mortality (HR: 2.90; [95% CI: 1.87-4.51], P < .001) and aortic-specific mortality (HR: 4.93; [95% CI: 1.69-14.4], P = .004). ConclusionsOur findings suggest that octogenarians do have a mildly increased mortality rate and rate of adverse events after F/BEVAR procedures. Despite this, when adjusted for other risk factors, age is on par with other medical comorbidities and therefore should not be a strict exclusion criterion for F/BEVAR procedures, rather considered in the global context of patient’s aortic anatomy, health, and functional status.

Association between SARS-CoV-2 infection and select symptoms and conditions 31 to 150 days after testing among children and adults

BackgroundAn increasing number of studies have described new and persistent symptoms and conditions as potential post-acute sequelae of SARS-CoV-2 infection (PASC). However, it remains unclear whether certain symptoms or conditions occur more frequently among persons with SARS-CoV-2 infection compared with those never infected with SARS-CoV-2. We compared the occurrence of specific COVID-associated symptoms and conditions as potential PASC 31- to 150-day following a SARS-CoV-2 test among adults and children with positive and negative test results.MethodsWe conducted a retrospective cohort study using electronic health record (EHR) data from 43 PCORnet sites participating in a national COVID-19 surveillance program. This study included 3,091,580 adults (316,249 SARS-CoV-2 positive; 2,775,331 negative) and 675,643 children (62,131 positive; 613,512 negative) who had a SARS-CoV-2 laboratory test during March 1, 2020–May 31, 2021 documented in their EHR. We used logistic regression to calculate the odds of having a symptom and Cox models to calculate the risk of having a newly diagnosed condition associated with a SARS-CoV-2 positive test.ResultsAfter adjustment for baseline covariates, hospitalized adults and children with a positive test had increased odds of being diagnosed with ≥ 1 symptom (adults: adjusted odds ratio[aOR], 1.17[95% CI, 1.11–1.23]; children: aOR, 1.18[95% CI, 1.08–1.28]) or shortness of breath (adults: aOR, 1.50[95% CI, 1.38–1.63]; children: aOR, 1.40[95% CI, 1.15–1.70]) 31–150 days following a SARS-CoV-2 test compared with hospitalized individuals with a negative test. Hospitalized adults with a positive test also had increased odds of being diagnosed with ≥ 3 symptoms or fatigue compared with those testing negative. The risks of being newly diagnosed with type 1 or type 2 diabetes (adjusted hazard ratio[aHR], 1.25[95% CI, 1.17–1.33]), hematologic disorders (aHR, 1.19[95% CI, 1.11–1.28]), or respiratory disease (aHR, 1.44[95% CI, 1.30–1.60]) were higher among hospitalized adults with a positive test compared with those with a negative test. Non-hospitalized adults with a positive test also had higher odds or increased risk of being diagnosed with certain symptoms or conditions.ConclusionsPatients with SARS-CoV-2 infection, especially those who were hospitalized, were at higher risk of being diagnosed with certain symptoms and conditions after acute infection.

Adaptive selection of the optimal strategy to improve precision and power in randomized trials.

Benkeser etal. demonstrate how adjustment for baseline covariates in randomized trials can meaningfully improve precision for a variety of outcome types. Their findings build on a long history, starting in 1932 with R.A. Fisher and including more recent endorsements by the U.S. Food and Drug Administration and the European Medicines Agency. Here, we address an important practical consideration: how to select the adjustment approach-which variables and in which form-to maximize precision, while maintaining Type-I error control. Balzer etal. previously proposed Adaptive Pre-specification within TMLE to flexibly and automatically select, from a prespecified set, the approach that maximizes empirical efficiency in small trials (N< 40). To avoid overfitting with few randomized units, selection was previously limited to working generalized linear models, adjusting for a single covariate. Now, we tailor Adaptive Pre-specification to trials with many randomized units. Using V-fold cross-validation and the estimated influence curve-squared as the loss function, we select from an expanded set of candidates, including modern machine learning methods adjusting for multiple covariates. As assessed in simulations exploring a variety of data-generating processes, our approach maintains Type-I error control (under the null) and offers substantial gains in precision-equivalent to 20%-43% reductions in sample size for the same statistical power. When applied to real data from ACTG Study 175, we also see meaningful efficiency improvements overall and within subgroups.

Comparison of baseline covariate adjustment methods for restricted mean survival time

The restricted mean survival time provides a straightforward clinical measure that dispenses with the need for proportional hazards assumptions. We focus on two strategies to directly model the survival time and adjust covariates. Firstly, pseudo-survival time is calculated for each subject using a leave-one-out approach, followed by a model analysis that adjusts for covariates using all pseudo-values. This method is used to reflect information of censored subjects in the model analysis. The second approach adjusts for covariates for those subjects with observed time-to-event while incorporating censored subjects using inverse probability of censoring weighting (IPCW). This paper evaluates these methods' power to detect group differences through computer simulations. We find the interpretation of pseudo-values challenging with the pseudo-survival time method and confirm that pseudo-survival times deviate from actual data in a primary biliary cholangitis clinical trial, mainly due to extensive censoring. Simulations reveal that the IPCW method is more robust, unaffected by the balance of censors, whereas pseudo-survival time is influenced by this balance. The IPCW method retains a nominal significance level for the type-1 error rate, even amidst group differences concerning censor incidence rates and covariates. Our study concludes that IPCW and pseudo-survival time methods differ significantly in handling censored data, impacting parameter estimations. Our findings suggest that the IPCW method provides more robust results than pseudo-survival time and is recommended, even when censor probabilities vary between treatment groups. However, pseudo-survival time remains a suitable choice when censoring probabilities are balanced.

Prognostic Impact of Repeated NT-proBNP Measurements in Patients With Heart Failure With Reduced Ejection Fraction

BackgroundAlthough clinical studies have demonstrated the association between a single N-terminal pro–B-type natriuretic peptide (NT-proBNP) measurement and clinical outcomes in chronic heart failure, the biomarker is frequently measured serially in clinical practice. ObjectivesThe aim of this study was to determine the added prognostic value of repeated NT-proBNP measurements compared with single measurements alone for chronic heart failure patients. MethodsIn the GUIDE-IT (Guiding Evidence Based Therapy Using Biomarker Intensified Treatment in Heart Failure) study, 894 study participants with chronic heart failure with reduced ejection fraction were enrolled at 45 outpatient sites in the United States and Canada. Repeated NT-proBNP levels were measured over a 2-year study period. Associations between repeated NT-proBNP measurements and trial endpoints were assessed using a joint longitudinal and survival model. ResultsAfter adjustment for baseline covariates, each doubling of the baseline NT-proBNP level was associated with a HR of 1.17 (95% CI: 1.08-1.28; P = 0.0003) for the primary trial endpoint of cardiovascular death or heart failure hospitalization. Serial measurements increased the adjusted HR for the primary trial endpoint to 1.66 (95% CI: 1.50-1.84; P < 0.0001), and a similar increased risk was observed across secondary trial endpoints. In joint modeling, an increase in NT-proBNP occurred weeks before the onset of adjudicated events. ConclusionsRepeated NT-proBNP measurements are a strong predictor of outcomes in heart failure with reduced ejection fraction with an increase in concentration occurring well before event onset. These results may support routine NT-proBNP monitoring to assist in clinical decision making. (Guiding Evidence Based Therapy Using Biomarker Intensified Treatment in Heart Failure [GUIDE-IT]; NCT01685840)

Beta-Blockers, Digoxin, or Both Following an Incident Diagnosis of Atrial Fibrillation: A Prospective Cohort Study.

Atrial fibrillation is one of the most common arrhythmias, but the optimal drug choice for a rate control strategy remains uncertain. A retrospective cohort claims database study of patients with an incident hospital discharge diagnosis of atrial fibrillation between 2011 and 2015. The exposure variables were a discharge prescription for beta-blockers, digoxin, or both. The primary outcome was a composite of total in-hospital mortality or a repeat cardiovascular (CV) hospitalization. Baseline confounding was controlled with propensity score inverse probability weighting using a entropy balancing algorithm and the prespecified estimand was the average treatment effect among the treated. Treatment effects for the weighted samples were calculated from a Cox proportional hazards model. A total of 12,723 patients were discharged on beta-blockers alone, 406 on digoxin alone, and 1499 discharged on combined beta-blocker and digoxin therapy with a median follow-up time of 356 days. After baseline covariate adjustment, the digoxin alone (hazard ratio [HR], 1.24; 95% confidence interval [CI], 0.85-1.81) and the combined group (HR, 1.09; 95% CI, 0.90-1.31) were not associated with increased risk for the composite endpoint compared with the beta- blocker-alone group. These results were robust to sensitivity analyses. Patients hospitalized for incident atrial fibrillation and discharged on digoxin alone or the combination of digoxin and a beta-blocker were not associated with an increase in the composite outcome of recurrent CV hospitalizations and death compared with those discharged on isolated beta-blocker therapy. However, additional studies are required to refine the precision of these estimates.

Longitudinal plasmode algorithms to evaluate statistical methods in realistic scenarios: an illustration applied to occupational epidemiology

IntroductionPlasmode simulations are a type of simulations that use real data to determine the synthetic data-generating equations. Such simulations thus allow evaluating statistical methods under realistic conditions. As far as we know, no plasmode algorithm has been proposed for simulating longitudinal data. In this paper, we propose a longitudinal plasmode framework to generate realistic data with both a time-varying exposure and time-varying covariates. This work was motivated by the objective of comparing different methods for estimating the causal effect of a cumulative exposure to psychosocial stressors at work over time.MethodsWe developed two longitudinal plasmode algorithms: a parametric and a nonparametric algorithms. Data from the PROspective Québec (PROQ) Study on Work and Health were used as an input to generate data with the proposed plasmode algorithms. We evaluated the performance of multiple estimators of the parameters of marginal structural models (MSMs): inverse probability of treatment weighting, g-computation and targeted maximum likelihood estimation. These estimators were also compared to standard regression approaches with either adjustment for baseline covariates only or with adjustment for both baseline and time-varying covariates.ResultsStandard regression methods were susceptible to yield biased estimates with confidence intervals having coverage probability lower than their nominal level. The bias was much lower and coverage of confidence intervals was much closer to the nominal level when considering MSMs. Among MSM estimators, g-computation overall produced the best results relative to bias, root mean squared error and coverage of confidence intervals. No method produced unbiased estimates with adequate coverage for all parameters in the more realistic nonparametric plasmode simulation.ConclusionThe proposed longitudinal plasmode algorithms can be important methodological tools for evaluating and comparing analytical methods in realistic simulation scenarios. To facilitate the use of these algorithms, we provide R functions on GitHub. We also recommend using MSMs when estimating the effect of cumulative exposure to psychosocial stressors at work.

Clinical and genomic characterisation of early-onset pancreatic cancer

BackgroundThe incidence of early-onset pancreatic cancer (EOPC) has risen dramatically in recent years. We aimed to characterise the clinical and genomic features of EOPC and evaluate their therapeutic implications. MethodsWe performed a comparative, single-centre, retrospective analysis of clinical, germline, and genomic features in EOPC (≤50 years) patients and compared them with a control group of average-onset pancreatic cancer patients (AOPC, ≥70 years). Key molecular findings were compared with an external, publicly available cohort. ResultsWe reviewed 336 patients who met all inclusion criteria (EOPC N = 139, AOPC N = 197). EOPC was associated with smoking status, lower prevalence of diabetes, better performance status, higher CA19.9 levels, and higher albumin levels at diagnosis. After adjustment for baseline covariates, we observed no differences in overall survival (OS). Age was associated with an increase in the incidence of KRASMUT both in our cohort and the validation cohort. EOPC were enriched in potentially actionable alterations according to ESCAT tiers I–IIIA when compared with AOPC in discovery and validation cohorts (19% versus 14% and 14% versus 8%, respectively). In the first-line metastatic setting, EOPC had a longer progression-free survival (hazard ratio [HR] 0.61, 95% confidence interval (CI) 0.43–0.87) and OS (HR 0.65, 95% CI 0.45–0.95), although there were no differences in response rate. After adjusting for the number of treatment lines, EOPC patients who did receive targeted therapies exhibited longer OS compared with EOPC who did not (HR 0.34, 95% CI 0.12–0.93). ConclusionsEOPC patients have improved outcomes in the metastatic setting when compared to AOPC and are enriched for targetable alterations that open opportunities for precision oncology-based approaches.

Covariate Adjustment in Chronic Pain Trials: An Oft-Missed Opportunity

Self-reported pain intensity, frequently used as an outcome in randomized clinical trials (RCTs) of chronic pain, is often highly variable and could be associated with multiple baseline factors. Thus, the assay sensitivity of pain trials (ie, the ability of the trial to detect a true treatment effect) could be improved by including prespecified baseline factors in the primary statistical model. The objective of this focus article was to characterize the baseline factors included in statistical analyses of chronic pain RCTs. Seventy-three RCTs published between 2016 and 2021 that investigated interventions for chronic pain were included. The majority of trials identified a single primary analysis (72.6%; n = 53). Of these, 60.4% (n = 32) included one or more covariates in the primary statistical model, most commonly baseline value of the primary outcome, study site, sex, and age. Only one of the trials reported information regarding associations between covariates and outcomes (ie, information that could inform prioritization of covariates for prespecification in future analyses). These findings demonstrate inconsistent use of covariates in the statistical models in chronic pain clinical trials. Prespecified adjustments for baseline covariates that could increase precision and assay sensitivity should be considered in future clinical trials of chronic pain treatments. PerspectiveThis review demonstrates inconsistent inclusion and potential underutilization of covariate adjustment in analyses of chronic pain RCTs. This article highlights areas for possible improvement in design and reporting related to covariate adjustment to improve efficiency in future RCTs.

Clinical and genomic characterization of early-onset pancreatic cancer.

4018 Background: Pancreatic cancer is a leading cause of cancer-related death worldwide. Early-onset tumours (≤ 50 years) have risen alarmingly in recent years. However, the clinical and genomic particularities of early-onset pancreatic cancer (EOPC) remain poorly defined. We aimed to characterize EOPC and study its implications for treatment and prognosis. Methods: We performed a retrospective analysis of EOPC patients and a control group of patients ≥ 70 years (defined as average-onset pancreatic cancer, AOPC) followed at a tertiary cancer centre from 2010 to 2022. We collected baseline patient characteristics, tumor molecular profiling, germline genetic alterations, survival and treatment outcomes. We used a targeted gene panel to identify somatic genomic events and classified them according to the ESMO scale for clinical actionability of molecular targets (ESCAT). Key molecular findings were validated in an external cohort. We used a propensity score weighting method and multivariate Cox regression analysis to adjust for covariates. Results: We reviewed 824 patients, 336 of whom met all inclusion criteria (EOPC N = 139, AOPC N = 197). EOPC was associated with smoking status (current, 15.9 vs 7.9%, p = 0.03), lower prevalence of diabetes (3.7 vs 39.6%, p &lt; 0.01), better performance status (ECOG 0, 42.9 vs 19.3%, p &lt; 0.01), higher CA19.9 levels (median 574 vs 207.6UI/L, p = 0.07) and higher albumin levels (median 4.2 vs 3.9g/L, p &lt; 0.01). EOPC showed a non-significant higher prevalence of germline alterations (22.4 vs 14.5%, p = 0.18). After adjustment for baseline covariates, we observed no differences in survival (HR 0.94, 95% CI 0.63-1.4). We found a lower prevalence of KRASMUT tumors in EOPC when compared with AOPC (83.1 vs 91.1%, p = 0.12) and validated this finding in an independent cohort of 803 patients (82.9 vs 94.5%, p &lt; 0.01). Notably, EOPC were enriched in potentially actionable alterations when compared with AOPC, both in our cohort (19.1 vs 14.4%) and the validation cohort (14.4 vs 7.9%, p &lt; 0.01). Moreover, 294 patients in our cohort (EOPC N = 130; AOPC N = 164) were diagnosed with or eventually presented metastasis. EOPC more frequently received first-line 5FU-based chemotherapy (44.1 vs 14.7%, p &lt; 0.01) and received a greater number of treatment lines (median 2 vs 1, p &lt; 0.01). EOPC had a longer progression-free survival on first-line chemotherapy when compared with AOPC (HR 0.61, 95% CI 0.43-0.87, p &lt; 0.01) although there were no differences in response rate. Six EOPC patients received matched targeted therapies and 5 patients with AOPC. After adjusting for the number of treatment lines received, EOPC treated with targeted therapies exhibited longer OS compared with EOPC who did not (HR 0.34 95% CI 0.12-0.93, p = 0.04), whereas this trend was not observed in AOPC (HR 0.82 95% CI 0.32-2.11, p = 0.68). Conclusions: EOPC patients harbor unique clinical and molecular features and may particularly benefit from precision-based oncology approaches.

Complex intervention to promote human papillomavirus (HPV) vaccine uptake in school settings: A cluster-randomized trial

Using a cluster-randomized trial design, we aimed to evaluate a complex intervention to increase uptake of human papillomavirus (HPV) vaccination in schools. The study was undertaken in high schools in Western Australia and South Australia between 2013 and 2015 with adolescents aged 12–13 years. Interventions included education, shared decision-making, and logistical strategies. The main outcome was school vaccine uptake. Secondary outcomes included consent forms returned and mean time to vaccinate 50 students. We hypothesised that a complex intervention would increase 3-dose HPV vaccine uptake. We recruited 40 schools (21 intervention, 19 control) with 6, 967 adolescents. There was no difference between intervention and control (3-dose mean 75.7% and 78.9%, respectively). Following adjustment for baseline covariates, absolute differences in coverage in favour of the intervention group were: dose 1, 0.8% (95% CI, −1.4,3.0); dose 2, 0.2% (95% CI, −2.7, 3.1); dose 3, 0.5% (95% CI, −2.6, 3.7). The percentage of returned consent forms in intervention schools (91.4%) was higher than in control schools (difference: 6%, 95% CI, 1.4, 10.7). There was a shorter mean time to vaccinate 50 students at dose 3. The difference for dose 3 was 110 min (95% CI, 42, 177); for dose 2, 90 min (95% CI, −15, 196); and dose 1, 28 min (95% CI, −71, 127). Logs revealed the inconsistent implementation of logistical strategies. The intervention had no impact on uptake. Inadequate resourcing for logistical strategies and advisory board reluctance toward strategies with potential financial implications impacted the implementation of logistical components. Trial registrationAustralian and New Zealand Clinical Trials Registry, ACTRN12614000404628, 14.04.2014.The study protocol was published in 2015 before data collection was finalised (Skinner et al., 2015). The HPV.edu study groupWe would like to acknowledge the contributions to this study by members of the HPV.edu Study Group, including: Professor Annette Braunack-Mayer: Australian Centre for Health Engagement, Evidence and Values, School of Health and Society, Faculty of Arts, Social Sciences and Humanities, University of Wollongong, NSW, Australia; Dr. Joanne Collins: Women's and Children's Health Network and School of Medicine and Robinson Research Institute, University of Adelaide, SA, Australia; Associate Professor Spring Cooper: School of Public Health, City University of New York (CUNY), New York, NY, USA; Heidi Hutton: Telethon Kids Institute, University of Western Australia, WA, Australia; Jane Jones: Telethon Kids Institute, University of Western Australia, WA, Australia; Dr. Adriana Parrella: Women's and Children's Health Network and School of Medicine and Robinson Research Institute, University of Adelaide, SA, Australia; and South Australian Health and Medical Research Institute (SAHMRI), Adelaide, Australia; Associate Professor David G. Regan: The Kirby Institute for Infection and Immunity in Society, Faculty of Medicine, UNSW Sydney, NSW, Australia; Professor Peter Richmond: Perth Children's Hospital, Child and Adolescent Health Service, Western Australia, Wesfarmers Centre of Vaccines and Infectious Diseases, Telethon Kids Institute, WA, Australia, and School of Medicine, University of Western Australia, Perth, WA, Australia; Dr. Tanya Stoney: Telethon Kids Institute, University of Western Australia, WA, Australia. Contact for the HPV.edu study group: Cristyn.Davies@sydney.edu.au or Rachel.Skinner@sydney.edu.au.

PO-01-031 TO CHECK OR NOT TO CHECK: LACK OF FIRST PASS ISOLATION MAY IDENTIFY PATIENTS WHO WOULD BENEFIT FROM PROVOCATIVE PHARMACOLOGIC CHALLENGE DURING PULMONARY VEIN ISOLATION

Durable pulmonary vein isolation (PVI) is the cornerstone of successful ablation of atrial fibrillation. Adenosine and isoproterenol are two commonly used pharmacologic agents for assessing conduction recovery during PVI. However, the use of a provocative pharmacologic challenge (PPC) is highly operator dependent, and the effect on outcomes after PVI using contemporary methods is unclear. Additionally, first pass isolation (FPI) has recently gained recognition as a marker for durable PVI. We sought to explore the utility of a PPC depending on the achievement of FPI. To assess the utility of a PPC with either adenosine or isoproterenol in patients with or without FPI during PVI. The Real-world Experience of Catheter Ablation for the Treatment of Symptomatic Paroxysmal and PsAF Using Novel Contact Force Technologies registry (Real-AF) is a prospective, observational, multicenter registry of patients undergoing radiofrequency (RF) ablation. A total of 910 patients with paroxysmal atrial fibrillation undergoing de novo ablation were included in this study and were retrospectively analyzed. Patients were divided into four groups based on whether FPI was achieved in both pulmonary veins and whether a PPC was performed. Atrial arrhythmia recurrence at one year was reported. A total of 590 patients (65.5%) had FPI and 313 (34.5%) did not. A PPC was performed in 488 (82.7%) of the FPI group and 258 (82.2%) of the no FPI group. Baseline covariates were compared using an analysis of variance (ANOVA), shown in Table 1. Left atrial diameter, CHADS2VAsc score, left atrial volume, procedure time, and total RF ablation time were significantly different among the four groups. Adjusted Kaplan-Meier curves are shown in Figure 1. Using a Cox-proportional hazards adjustment for baseline covariates, PPC was not associated with atrial arrhythmia free survival in the FPI group (p=0.145) (Figure 1A) but was associated with atrial arrhythmia survival in the no FPI group (p=0.025) (Figure 1B). In patients without FPI, the use of PPC resulted in similar outcomes as the FPI group (Figure 1C). In patients undergoing PVI who have FPI, performing a PPC does not improve atrial arrhythmia free survival at one year. However, in patients without FPI, a PPC is associated with improved atrial arrhythmia free survival at one year, leading to similar outcomes to patients who achieved FPI. This may have important implications for intraprocedural decision-making and procedural efficiency during PVI.Tabled 1Table 1. Baseline Demographics and Procedural Characteristics (PO-01-031).FPI and PPC (n=488)FPI and No PPC (n=102)No FPI and PPC (n=258)No FPI and No PPC (n=55)p-value (ANOVA)Ejection Fraction (%)57.4 ±7.558.3±8.557.7±6.457.7±6.40.337Left atrial diameter (cm)3.9±0.74.0±0.63.9±0.54.2±0.60.042Age (years)64.9±11.666.6±10.163.6±10.965.5±8.80.117Hypertension336 (68%)65 (63%)179 (69%)30 (54%)0.101Diabetes Mellitus69 (14%)20 (19%)41 (16%)10 (18%)0.508Congestive Heart Failure29 (6%)9 (9%)18 (7%)8 (14%)0.114Sex (Female)245 (50%)42 (42%)108 (42%)21 (38%)0.076CHADS2Vasc Score2.6±1.52.5±1.42.3±1.42.0±1.60.025Left Atrial Volume (cm3)84.6±26.188.4±28.985.2±22.1103.2±30.6<0.001Left Atrial Low Voltage Area (% of surface area)2.6±7.74.7±9.02.5±7.74.3±6.70.062Procedure time (minutes)74±35102±4094±42117±44<0.001Total Radiofrequency Ablation time (minutes)18±826±1125±1128±11<0.001 Open table in a new tab

Associations of alternative cannabis product use and poly-use with subsequent illicit drug use initiation during adolescence.

Specific cannabis products may differentially increase risk of initiating non-cannabis illicit drug use during adolescence. To determine whether ever- and poly-use of smoked, vaporized, edible, concentrate, or blunt cannabis products are associated with subsequent initiation of non-cannabis illicit drug use. High school students from Los Angeles completed in-classroom surveys. The analytic sample (N = 2163; 53.9% female; 43.5% Hispanic/Latino;baselineM age = 17.1years) included students who reported never using illicit drugs at baseline (spring, 11th grade) and provided data at follow-up (fall and spring, 12th grade). Logistic regression models assessed associations between use of smoked, vaporized, edible, concentrate, and blunt cannabis at baseline (yes/no for each product) and any non-cannabis illicit drug use initiation-including cocaine, methamphetamine, psychedelics, ecstasy, heroin, prescription opioids, or benzodiazepines-at follow-up. Among those who never usednon-cannabis illicit drugs at baseline, ever cannabis use varied by cannabis product (smoked = 25.8%, edible = 17.5%, vaporized = 8.4%, concentrates = 3.9%, and blunts = 18.2%) and patterns of use (single product use = 8.2% and poly-productuse = 21.8%). After adjustment for baseline covariates, odds of illicit drug use at follow-up were largest for baseline ever users of concentrates (aOR [95% CI] = 5.74[3.16-10.43]), followed by vaporized (aOR [95% CI] = 3.11 [2.41-4.01]), edibles (aOR [95% CI] = 3.43 [2.32-5.08]), blunts (aOR [95% CI] = 2.66[1.60-4.41]), and smoked (aOR [95% CI] = 2.57 [1.64-4.02]) cannabis. Ever use of a single product (aOR [95% CI] = 2.34 [1.26-4.34]) or 2 + products (aOR [95% CI] = 3.82 [2.73-5.35]) were also associated with greater odds of illicit drug initiation. For each of five different cannabis products, cannabis use was associated with greater odds of subsequent illicit drug use initiation, especially for cannabis concentrate and poly-product use.

Sex differences in risk factors for end-stage kidney disease and death in type 2 diabetes: A retrospective cohort study.

This study investigated the sex differences in the risk of end-stage kidney disease (ESKD) and mortality, as well as the effect modification of sex on associated factors in patients with type 2 diabetes. This multicenter observational cohort study included 4328 patients with type 2 diabetes. Hazard ratios (HRs) with 95% confidence intervals (CIs) of sex for ESKD and death were estimated using Cox proportional regression with adjustment for baseline covariates. For assessing risk modification, HRs and incidence rates for ESKD and death were compared between sexes across patient characteristics using Cox proportional and Poisson regression models. During a median follow-up of 7 years, 276 patients (70% men) developed ESKD, and 241 patients (68% men) died. Men had higher risks of ESKD (HR 1.34; 95% CI 1.02-1.75; p=.034) and death (HR 1.64; 95% CI 1.24-2.16; p=.001) versus women after adjusting for multiple covariates. Among patients with microalbuminuria, men had a substantially higher risk of ESKD versus women, compared to those with normo- and macroalbuminuria (p for interaction .04). Incidence rates were also increased in men versus women with albuminuria of around 300 mg/g. No differences were detected in the association of sex and death across baseline patient subgroups. In type 2 diabetes, men had an increased risk of ESKD and death versus women. Moderately increased albuminuria was strongly associated with sex difference in developing ESKD.

A comparison of covariate adjustment approaches under model misspecification in individually randomized trials

Adjustment for baseline covariates in randomized trials has been shown to lead to gains in power and can protect against chance imbalances in covariates. For continuous covariates, there is a risk that the the form of the relationship between the covariate and outcome is misspecified when taking an adjusted approach. Using a simulation study focusing on individually randomized trials with small sample sizes, we explore whether a range of adjustment methods are robust to misspecification, either in the covariate–outcome relationship or through an omitted covariate–treatment interaction. Specifically, we aim to identify potential settings where G-computation, inverse probability of treatment weighting (IPTW), augmented inverse probability of treatment weighting (AIPTW) and targeted maximum likelihood estimation (TMLE) offer improvement over the commonly used analysis of covariance (ANCOVA). Our simulations show that all adjustment methods are generally robust to model misspecification if adjusting for a few covariates, sample size is 100 or larger, and there are no covariate–treatment interactions. When there is a non-linear interaction of treatment with a skewed covariate and sample size is small, all adjustment methods can suffer from bias; however, methods that allow for interactions (such as G-computation with interaction and IPTW) show improved results compared to ANCOVA. When there are a high number of covariates to adjust for, ANCOVA retains good properties while other methods suffer from under- or over-coverage. An outstanding issue for G-computation, IPTW and AIPTW in small samples is that standard errors are underestimated; they should be used with caution without the availability of small-sample corrections, development of which is needed. These findings are relevant for covariate adjustment in interim analyses of larger trials.