Assessing the performance of physician's prescribing preference as an instrumental variable in comparative effectiveness research with moderate and small sample sizes: a simulation study.

Aim: This simulation study is to assess the utility of physician's prescribing preference (PPP) as an instrumental variable for moderate and smaller sample sizes. Materials & methods: We designed a simulation study to imitate a comparative effectiveness research under different sample sizes. We compare the performance of instrumental variable (IV) and non-IV approaches using two-stage least squares (2SLS) and ordinary least squares (OLS) methods, respectively. Further, we test the performance of different forms of proxies for PPP as an IV. Results: The percent bias of 2SLS is around approximately 20%, while the percent bias of OLS is close to 60%. The sample size is not associated with the level of bias for the PPP IV approach. Conclusion: Irrespective of sample size, the PPP IV approach leads to less biased estimates of treatment effectiveness than OLS adjusting for known confounding only. Particularly for smaller sample sizes, we recommend constructing PPP from long prescribing histories to improve statistical power.

Using Claims Data to Predict Pre-Operative BMI Among Bariatric Surgery Patients: Development of the BMI Before Bariatric Surgery Scoring System (B3S3).

Background: Lack of body mass index (BMI) measurements limits the utility of claims data for bariatric surgery research, but pre-operative BMI may be imputed due to existence of weight-related diagnosis codes and BMI-related reimbursement requirements. We used a machine learning pipeline to create a claims-based scoring system to predict pre-operative BMI, as documented in the electronic health record (EHR), among patients undergoing a new bariatric surgery. Methods: Using the Optum Labs Data Warehouse, containing linked de-identified claims and EHR data for commercial or Medicare Advantage enrollees, we identified adults undergoing a new bariatric surgery between January 2011 and June 2018 with a BMI measurement in linked EHR data ≤30 days before the index surgery (n=3226). We constructed predictors from claims data and applied a machine learning pipeline to create a scoring system for pre-operative BMI, the B3S3. We evaluated the B3S3 and a simple linear regression model (benchmark) in test patients whose index surgery occurred concurrent (2011-2017) or prospective (2018) to the training data. Results: The machine learning pipeline yielded a final scoring system that included weight-related diagnosis codes, age, and number of days hospitalized and distinct drugs dispensed in the past 6 months. In concurrent test data, the B3S3 had excellent performance (R2 0.862, 95% confidence interval [CI] 0.815-0.898) and calibration. The benchmark algorithm had good performance (R2 0.750, 95% CI 0.686-0.799) and calibration but both aspects were inferior to the B3S3. Findings in prospective test data were similar. Conclusion: The B3S3 is an accessible tool that researchers can use with claims data to obtain granular and accurate predicted values of pre-operative BMI, which may enhance confounding control and investigation of effect modification by baseline obesity levels in bariatric surgery studies utilizing claims data.

Pre-operative BMI is an important potential confounder in comparative effectiveness studies of bariatric surgeries.Claims data lack clinical measurements, but insurance reimbursement requirements for bariatric surgery often result in pre-operative BMI being coded in claims data.We used a machine learning pipeline to create a model, the B3S3, to predict pre-operative BMI, as documented in the EHR, among bariatric surgery patients based on the presence of certain weight-related diagnosis codes and other patient characteristics derived from claims data.Researchers can easily use the B3S3 with claims data to obtain granular and accurate predicted values of pre-operative BMI among bariatric surgery patients.

Behavioral carry-over effect and power consideration in crossover trials.

A crossover trial is an efficient trial design when there is no carry-over effect. To reduce the impact of the biological carry-over effect, a washout period is often designed. However, the carry-over effect remains an outstanding concern when a washout period is unethical or cannot sufficiently diminish the impact of the carry-over effect. The latter can occur in comparative effectiveness research, where the carry-over effect is often non-biological but behavioral. In this paper, we investigate the crossover design under a potential outcomes framework with and without the carry-over effect. We find that when the carry-over effect exists and satisfies a sign condition, the basic estimator underestimates the treatment effect, which does not inflate the type I error of one-sided tests but negatively impacts the power. This leads to a power trade-off between the crossover design and the parallel-group design, and we derive the condition under which the crossover design does not lead to type I error inflation and is still more powerful than the parallel-group design. We also develop covariate adjustment methods for crossover trials. We evaluate the performance of cross-over design and covariate adjustment using data from the MTN-034/REACH study.

Comparative Effectiveness of Placental Allografts in the Treatment of Diabetic Lower Extremity Ulcers and Venous Leg Ulcers in U.S. Medicare Beneficiaries: A Retrospective Observational Cohort Study Using Real-World Evidence.

Objective: To compare the effectiveness of cellular tissue products (CTP) versus standard care in U.S. Medicare beneficiaries with diabetic lower extremity ulcers (DLEUs) or venous leg ulcers (VLUs). Approach: We performed a retrospective cohort study using real-world evidence from U.S. Medicare claims for DLEUs or VLUs between 2016 and 2020. There were three cohorts evaluated: viable cryopreserved placental membrane (vCPM) or viable lyopreserved placental membrane (vLPM); other CTP; and standard care. Claims were collapsed into episodes of care. Univariate and bivariate statistics were used to examine the frequency distribution of demographics and clinical variables. Multivariable zero-inflated binomial regressions were used to evaluate mortality and recurrence trends. Logistic regression compared three adverse outcomes (AOs): amputation; 1-year mortality; and wound recurrence. Results: There were 333,362 DLEU episodes among 261,101 beneficiaries, and 122,012 VLU episodes among 80,415 beneficiaries. DLEU treatment with vLPM was associated with reduced 1-year mortality (-26%), reduced recurrence (-91%), and reduced AOs (-71%). VLU treatment with vCPM or vLPM was associated with reduced 1-year mortality (-23%), reduced recurrence (-80%), and 66.77% reduction in AOs. These allografts were also associated with a 49% and 73% reduced risk of recurrence in DLEU and VLU, respectively, compared with other CTPs. Finally, vCPM or vLPM were associated with noninferior prevention of AOs related to amputation, mortality, and recurrence (95% CI: 0.69-1.14). Conclusions: DLEUs and VLUs treated with vCPM and vLPM allografts are associated with lowered 1-year mortality, wound recurrence, and AOs in DLEUs and VLUs compared with standard care. Decision makers weighing coverage of placental allografts should consider these added short- and long-term clinical benefits relative to costly management and high mortality of Medicare's most frequent wounds.

Comparative Effectiveness of Second-line Antihyperglycemic Agents for Cardiovascular Outcomes: A Large-scale, Multinational, Federated Analysis of the LEGEND-T2DM Study.

Background: SGLT2 inhibitors (SGLT2is) and GLP-1 receptor agonists (GLP1-RAs) reduce major adverse cardiovascular events (MACE) in patients with type 2 diabetes mellitus (T2DM). However, their effectiveness relative to each other and other second-line antihyperglycemic agents is unknown, without any major ongoing head-to-head trials. Methods: Across the LEGEND-T2DM network, we included ten federated international data sources, spanning 1992-2021. We identified 1,492,855 patients with T2DM and established cardiovascular disease (CVD) on metformin monotherapy who initiated one of four second-line agents (SGLT2is, GLP1-RAs, dipeptidyl peptidase 4 inhibitor [DPP4is], sulfonylureas [SUs]). We used large-scale propensity score models to conduct an active comparator, target trial emulation for pairwise comparisons. After evaluating empirical equipoise and population generalizability, we fit on-treatment Cox proportional hazard models for 3-point MACE (myocardial infarction, stroke, death) and 4-point MACE (3-point MACE + heart failure hospitalization) risk, and combined hazard ratio (HR) estimates in a random-effects meta-analysis. Findings: Across cohorts, 16·4%, 8·3%, 27·7%, and 47·6% of individuals with T2DM initiated SGLT2is, GLP1-RAs, DPP4is, and SUs, respectively. Over 5·2 million patient-years of follow-up and 489 million patient-days of time at-risk, there were 25,982 3-point MACE and 41,447 4-point MACE events. SGLT2is and GLP1-RAs were associated with a lower risk for 3-point MACE compared with DPP4is (HR 0·89 [95% CI, 0·79-1·00] and 0·83 [0·70-0·98]), and SUs (HR 0·76 [0·65-0·89] and 0·71 [0·59-0·86]). DPP4is were associated with a lower 3-point MACE risk versus SUs (HR 0·87 [0·79-0·95]). The pattern was consistent for 4-point MACE for the comparisons above. There were no significant differences between SGLT2is and GLP1-RAs for 3-point or 4-point MACE (HR 1·06 [0·96-1·17] and 1·05 [0·97-1·13]). Interpretation: In patients with T2DM and established CVD, we found comparable cardiovascular risk reduction with SGLT2is and GLP1-RAs, with both agents more effective than DPP4is, which in turn were more effective than SUs. These findings suggest that the use of GLP1-RAs and SGLT2is should be prioritized as second-line agents in those with established CVD. Funding: National Institutes of Health, United States Department of Veterans Affairs.

Automated generation of comparator patients in the electronic medical record.

Background: Well-designed randomized trials provide high-quality clinical evidence but are not always feasible or ethical. In their absence, the electronic medical record (EMR) presents a platform to conduct comparative effectiveness research, central to the emerging academic learning health system (aLHS) model. A barrier to realizing this vision is the lack of a process to efficiently generate a reference comparison group for each patient. Objective: To test a multi-step process for the selection of comparators in the EMR. Materials and Methods: We conducted a mixed-methods study within a large aLHS in North Carolina. We (1) created a list of 35 candidate variables; (2) surveyed 270 researchers to assess the importance of candidate variables; and (3) built consensus rankings around survey-identified variables (ie, importance scores >7) across two panels of 7-8 clinical research experts. Prioritized algorithm inputs were collected from the EMR and applied using a greedy matching technique. Feasibility was measured as the percentage of patients with 100 matched comparators and performance was measured via computational time and Euclidean distance. Results: Nine variables were selected: age, sex, race, ethnicity, body mass index, insurance status, smoking status, Charlson Comorbidity Index, and neighborhood percentage in poverty. The final process successfully generated 100 matched comparators for each of 1.8 million candidate patients, executed in less than 100 min for the majority of strata, and had average Euclidean distance 0.043. Conclusion: EMR-derived matching is feasible to implement across a diverse patient population and can provide a reproducible, efficient source of comparator data for observational studies, with additional testing in clinical research applications needed.

Percutaneous electrical nerve field stimulation for adolescents with irritable bowel syndrome: Cost-benefit and cost-minimization analysis.

Abdominal pain drives significant cost for adolescents with irritable bowel syndrome (IBS). We performed an economic analysis to estimate cost-savings for patients' families and healthcare insurance, and health outcomes, based on abdominal pain improvement with percutaneous electrical nerve field stimulation (PENFS) with IB-Stim® (Neuraxis). We constructed a Markov model with a 1-year time horizon comparing outcomes and costs with PENFS versus usual care without PENFS. Clinical outcomes were derived from a sham-controlled double-blind trial of PENFS for adolescents with IBS. Costs/work-productivity impact for parents were derived from appropriate observational cohorts. PENFS was associated with 18 added healthy days over 1 year of follow-up, increased annual parental wages of $5,802 due to fewer missed work days to care for the child, and $4744 in cost-savings to insurance. Percutaneous electrical field nerve stimulation for adolescents with IBS appears to yield significant cost-savings to patients' families and insurance.

Integrating Evidence to Guide Use of Biologics and Small Molecules for Inflammatory Bowel Diseases.

Advances in science have led to the development of multiple biologics and small molecules for the treatment of inflammatory bowel diseases (IBDs). This growth in advanced medical therapies has been accompanied by an increase in methodological innovation to study and compare therapies. Guidelines provide an evidence-based approach to integrating therapies into routine practice, but they are often unable to provide timely recommendations as new therapies come to market, and they have limited incorporation of real-world evidence when making recommendations. This limits the scope and usability of guidelines, and a gap remains in defining how best to position and integrate advanced medical therapies for IBD. In this review, we provide a framework for clinicians and researchers to understand key differences in sources of evidence, how different methodologies are applied to study the comparative effectiveness of advanced medical therapies in IBD, and considerations for how these sources of evidence can be used to better integrate current guideline recommendations. Over time, we anticipate this framework will allow for a transition to living guidelines and/or practice recommendations.

Effectiveness of Biologic Agents Among Hispanic Patients With Metastatic Colorectal Cancer.

BACKGROUND: Randomized clinical trials have defined the survival advantage with the addition of biologic drugs to chemotherapy in patients with metastatic colorectal cancer (mCRC). Under representation of Hispanics contributes to poorly defined outcomes in this group. We aim to determine whether the real-world benefit of biologics extends to Hispanics using a comparative effectiveness research approach. METHODS: This retrospective cohort study included all treatment centers contributing to SEER registry with available claims in the SEER-Medicare linked database (2001-2011) and 2 hospitals (2004-2016) catering to minorities. Metastatic CRC patients were classified as receiving chemotherapy or biochemotherapy (CT plus biologics; if initiated within 3 months of chemotherapy). The primary outcome was overall survival (OS) among the Hispanic patients calculated from time of administration of first dose of chemotherapy to death or last follow-up. A weighted Cox regression model was used to assess differences in survival. RESULTS: We identified 182 Hispanic patients with mCRC from the Patient Entitlement and Diagnosis Summary (PEDSF) file (n = 101) and hospital database (n = 81). Overall, 52% were women and 72% received biologics. The median OS was 11.3 and 17.0 months in chemotherapy and biochemotherapy group, respectively. Biochemotherapy offered a survival benefit compared with chemotherapy alone, with an average hazard rate reduction of 39% (95% CI 6%-60%, p = .0236) using inverse probability of treatment weighting (IPTW) based analysis. CONCLUSION: In this cohort of Hispanic patients with mCRC, biochemotherapy was associated with longer survival. Clinicians may offer biochemotherapy therapy to all patients regardless of race/ethnicity to maximize clinical benefit.

Controls, comparator arms, and designs for critical care comparative effectiveness research: It's complicated.

BACKGROUND: Comparative effectiveness research is meant to determine which commonly employed medical interventions are most beneficial, least harmful, and/or most costly in a real-world setting. While the objectives for comparative effectiveness research are clear, the field has failed to develop either a uniform definition of comparative effectiveness research or an appropriate set of recommendations to provide standards for the design of critical care comparative effectiveness research trials, spurring controversy in recent years. The insertion of non-representative control and/or comparator arm subjects into critical care comparative effectiveness research trials can threaten trial subjects' safety. Nonetheless, the broader scientific community does not always appreciate the importance of defining and maintaining critical care practices during a trial, especially when vulnerable, critically ill populations are studied. Consequently, critical care comparative effectiveness research trials sometimes lack properly constructed control or active comparator arms altogether and/or suffer from the inclusion of "unusual critical care" that may adversely affect groups enrolled in one or more arms. This oversight has led to critical care comparative effectiveness research trial designs that impair informed consent, confound interpretation of trial results, and increase the risk of harm for trial participants. METHODS/EXAMPLES: We propose a novel approach to performing critical care comparative effectiveness research trials that mandates the documentation of critical care practices prior to trial initiation. We also classify the most common types of critical care comparative effectiveness research trials, as well as the most frequent errors in trial design. We present examples of these design flaws drawn from past and recently published trials as well as examples of trials that avoided those errors. Finally, we summarize strategies employed successfully in well-designed trials, in hopes of suggesting a comprehensive standard for the field. CONCLUSION: Flawed critical care comparative effectiveness research trial designs can lead to unsound trial conclusions, compromise informed consent, and increase risks to research subjects, undermining the major goal of comparative effectiveness research: to inform current practice. Well-constructed control and comparator arms comprise indispensable elements of critical care comparative effectiveness research trials, key to improving the trials' safety and to generating trial results likely to improve patient outcomes in clinical practice.

ICD-10 diagnosis codes in electronic health records do not adequately capture fracture complexity for proximal humerus fractures.

BACKGROUND: The ability to do comparative effectiveness research (CER) for proximal humerus fractures (PHF) using data in electronic health record (EHR) systems and administrative claims databases was enhanced by the 10th revision of the International Classification of Diseases (ICD-10), which expanded the diagnosis codes for PHF to describe fracture complexity including displacement and the number of fracture parts. However, these expanded codes only enhance secondary use of data for research if the codes selected and recorded correctly reflect the fracture complexity. The objective of this project was to assess the accuracy of ICD-10 diagnosis codes documented during routine clinical practice for secondary use of EHR data. METHODS: A sample of patients with PHFs treated by orthopedic providers across a large, regional health care system between January 1, 2016, and December 31, 2018, were retrospectively identified from the EHR. Four fellowship-trained orthopedic surgeons reviewed patient radiographs and recorded the Neer Classification characteristics of displacement, number of parts, and fracture location(s). The fracture characteristics were then reviewed by a trained coder, and the most clinically appropriate ICD-10 diagnosis code based on the number of fracture parts was assigned. We assessed congruence between ICD-10 codes documented in the EHR and radiograph-validated codes, and assessed sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for EHR-documented ICD-10 codes. RESULTS: There were 761 patients with unilateral, closed PHF who met study inclusion criteria. On average, patients were 67 years of age and 77% were female. Based on radiograph review, 37% were 1-part fractures, 42% were 2-part, 11% were 3-part, and 10% were 4-part fractures. Of the EHR diagnosis codes recorded during clinical practice, 59% were "unspecified" fracture diagnosis codes that did not identify the number of fracture parts. Examination of fracture codes revealed PPV was highest for 1-part (PPV = 0.66, 95% confidence interval [CI] 0.60-0.72) and 4-part fractures (PPV = 0.67, 95% CI 0.13-1.00). CONCLUSIONS: Current diagnosis coding practices do not adequately capture the fracture complexity needed to conduct subgroup analysis for PHF. Conclusions drawn from population studies or large databases using ICD-10 codes for PHF classification should be interpreted within this limitation. Future studies are warranted to improve diagnostic coding to support large observational studies using EHR and administrative claims data.

Comparative effectiveness of sotrovimab and molnupiravir for preventing severe COVID-19 outcomes in patients on kidney replacement therapy: observational study using the OpenSAFELY-UKRR and SRR databases.

Background: Due to limited inclusion of patients on kidney replacement therapy (KRT) in clinical trials, the effectiveness of coronavirus disease 2019 (COVID-19) therapies in this population remains unclear. We sought to address this by comparing the effectiveness of sotrovimab against molnupiravir, two commonly used treatments for non-hospitalised KRT patients with COVID-19 in the UK. Methods: With the approval of National Health Service England, we used routine clinical data from 24 million patients in England within the OpenSAFELY-TPP platform linked to the UK Renal Registry (UKRR) to identify patients on KRT. A Cox proportional hazards model was used to estimate hazard ratios (HRs) of sotrovimab versus molnupiravir with regards to COVID-19-related hospitalisations or deaths in the subsequent 28 days. We also conducted a complementary analysis using data from the Scottish Renal Registry (SRR). Results: Among the 2367 kidney patients treated with sotrovimab (n = 1852) or molnupiravir (n = 515) between 16 December 2021 and 1 August 2022 in England, 38 cases (1.6%) of COVID-19-related hospitalisations/deaths were observed. Sotrovimab was associated with substantially lower outcome risk than molnupiravir {adjusted HR 0.35 [95% confidence interval (CI) 0.17-0.71]; P = .004}, with results remaining robust in multiple sensitivity analyses. In the SRR cohort, sotrovimab showed a trend toward lower outcome risk than molnupiravir [HR 0.39 (95% CI 0.13-1.21); P = .106]. In both datasets, sotrovimab had no evidence of an association with other hospitalisation/death compared with molnupiravir (HRs ranged from 0.73 to 1.29; P > .05). Conclusions: In routine care of non-hospitalised patients with COVID-19 on KRT, sotrovimab was associated with a lower risk of severe COVID-19 outcomes compared with molnupiravir during Omicron waves.

Benefits and Harms of Standard Versus Reduced-Dose Direct Oral Anticoagulant Therapy for Older Adults With Multiple Morbidities and Atrial Fibrillation.

Background Dose reduction of direct oral anticoagulant (DOAC) medications is inconsistently applied to older adults with multiple morbidities, potentially due to perceived harms and unknown benefits of standard dosing. Methods and Results Using 2013 to 2017 US Medicare claims linked to Minimum Data Set records, we conducted a retrospective cohort study. We identified DOAC initiators (apixaban, dabigatran, rivaroxaban) aged ≥65 years with nonvalvular atrial fibrillation residing in a nursing home. We estimated inverse-probability of treatment weights for DOAC dose using propensity scores. We examined safety (hospitalization for major bleeding) and effectiveness outcomes (all-cause mortality, thrombosis [myocardial infarction, stroke, systemic embolism, venous thromboembolism]). We estimated hazard ratios (HRs) and 95% CIs using cause-specific hazard-regression models. Of 21 878 DOAC initiators, 48% received reduced dosing. The mean age of residents was 82.0 years, 66% were female, and 31% had moderate/severe cognitive impairment. After estimating inverse-probability of treatment weights, standard dosing was associated with a higher rate of bleeding (HR, 1.18 [95% CI, 1.03-1.37]; 9.4 versus 8.0 events per 100 person-years). Standard-dose therapy was associated with the highest rates of bleeding among those aged >80 years (9.1 versus 6.7 events per 100 person-years) and with a body mass index <30 kg/m2 (9.4 versus 7.4 events per 100 person-years). There was no association of dosing with mortality (HR, 0.99 [95% CI, 0.96-1.06]) or thrombotic events (HR, 1.16 [95% CI, 0.96-1.41]). Conclusions In this nationwide study of nursing home residents with nonvalvular atrial fibrillation, we found a higher rate of bleeding and little difference in effectiveness of standard versus reduced-dose DOAC treatment. Our results support the use of reduced-dose DOACs for many older adults with multiple morbidities.

Comparative Effects of Bivalent, Quadrivalent, and Nonavalent Human Papillomavirus Vaccines in The Prevention of Genotype-Specific Infection: A Systematic Review and Network Meta-Analysis.

BACKGROUND: Human papillomavirus (HPV) infection is a major global disease burden and the main cause of cervical cancer. Certain HPV genotypes, with are the most common etiologic pathogens and cause a significant disease burden, are being targeted for vaccine development. However, few studies have focused on the comparative effectiveness of the bivalent HPV (2v-HPV), quadrivalent HPV (4v-HPV), and nonavalent HPV (9v-HPV) vaccines against HPV strain-specific infection. This study investigated the comparative effects of these vaccines against genotype-specific infection. MATERIALS AND METHODS: We conducted a pairwise and network meta-analysis of published randomized clinical trials of HPV vaccines according to sex and HPV infection status for nine HPV genotypes (HPV 6/11/16/18/31/33/45/52/58). RESULTS: Overall, 10 randomized controlled trials (12 articles) were included in this study. In the network meta-analysis, no statistically significant differences were observed in the prevention of carcinogenic HPV strains (16/18/31/33/45/52/58) between the 2v-HPV and 4v-HPV vaccines in female HPV infection-naïve populations. However, the 9v-HPV vaccine showed a significantly superior effect compared with 2v-HPV and 4v-HPV vaccines in preventing HPV 31/33/45/52/58 infections. Although 2v-HPV and 4v-HPV vaccines provided some cross-protection against HPV 31/33/45/52/58 infections, the effect was significant only on HPV 31 infection. For HPV 16 and 18, neither statistically significant nor small differences were found in the prevention of HPV infection among the 2v-HPV, 4v-HPV, and 9v-HPV vaccines. CONCLUSION: Our study complements previous understanding of how the effect of HPV vaccines differs according to the HPV genotype. This is important because HPV genotype prevalence varies among countries. We advocate for continued efforts in vaccinating against HPV, while public health agencies should consider the difference in the vaccine effect and HPV genotype prevalence when implementing HPV vaccination in public vaccination programs.

Remdesivir Is Associated With Reduced Mortality in COVID-19 Patients Requiring Supplemental Oxygen Including Invasive Mechanical Ventilation Across SARS-CoV-2 Variants.

Background: This comparative effectiveness study investigated the effect of remdesivir on in-hospital mortality among patients hospitalized for coronavirus disease 2019 (COVID-19) requiring supplemental oxygen including low-flow oxygen (LFO), high-flow oxygen/noninvasive ventilation (HFO/NIV), or invasive mechanical ventilation/extracorporeal membrane oxygenation (IMV/ECMO) across variant of concern (VOC) periods. Methods: Patients hospitalized for COVID-19 between December 2020 and April 2022 and administered remdesivir upon admission were 1:1 propensity score matched to patients not administered remdesivir during their COVID-19 hospitalization. Analyses were stratified by supplemental oxygen requirement upon admission and VOC period. Cox proportional hazards models were used to derive adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs) for 14- and 28-day mortality. Results: Patients treated with remdesivir (67 582 LFO, 34 857 HFO/NIV, and 4164 IMV/ECMO) were matched to non-remdesivir patients. Unadjusted mortality rates were significantly lower for remdesivir-treated patients at 14 days (LFO: 6.4% vs. 8.8%; HFO/NIV: 16.8% vs. 19.4%; IMV/ECMO: 27.8% vs. 35.3%) and 28 days (LFO: 9.8% vs. 12.3%; HFO/NIV: 25.8% vs. 28.3%; IMV/ECMO: 41.4% vs. 50.6%). After adjustment, remdesivir treatment was associated with a statistically significant reduction in in-hospital mortality at 14 days (LFO: aHR, 0.72; 95% CI, 0.66-0.79; HFO/NIV: aHR, 0.83; 95% CI, 0.77-0.89; IMV/ECMO: aHR, 0.73; 95% CI, 0.65-0.82) and 28 days (LFO: aHR, 0.79; 95% CI, 0.73-0.85; HFO/NIV: aHR, 0.88; 95% CI, 0.82-0.93; IMV/ECMO: aHR, 0.74; 95% CI, 0.67-0.82) compared with non-remdesivir treatment. Lower risk of mortality among remdesivir-treated patients was observed across VOC periods. Conclusions: Remdesivir treatment is associated with significantly reduced mortality among patients hospitalized for COVID-19 requiring supplemental oxygen upon admission, including those requiring HFO/NIV or IMV/ECMO with severe or critical disease, across VOC periods.

Long-term Prostate Cancer-specific Mortality After Prostatectomy, Brachytherapy, External Beam Radiation Therapy, Hormonal Therapy, or Monitoring for Localized Prostate Cancer.

BACKGROUND: The optimal treatment of localized prostate cancer (PCa) remains controversial. OBJECTIVE: To compare long-term survival among men who underwent radical prostatectomy (RP), brachytherapy (BT), external beam radiation therapy (EBRT), primary androgen deprivation therapy (PADT), or monitoring (active surveillance [AS]/watchful waiting [WW]) for PCa. DESIGN, SETTING, AND PARTICIPANTS: This is a cohort study with long-term follow-up from the multicenter, prospective, largely community-based Cancer of the Prostate Strategic Urologic Research Endeavor (CaPSURE) registry. Men with biopsy-proven, clinical T1-3aN0M0, localized PCa were consecutively accrued within 6 mo of diagnosis and had clinical risk data and at least 12 mo of follow-up after diagnosis available. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: PCa risk was assessed, and multivariable analyses were performed to compare PCa-specific mortality (PCSM) and all-cause mortality by primary treatment, with extensive adjustment for age and case mix using the Cancer of the Prostate Risk Assessment (CAPRA) score and a well-validated nomogram. RESULTS AND LIMITATIONS: Among 11 864 men, 6227 (53%) underwent RP, 1645 (14%) received BT, 1462 (12%) received EBRT, 1510 (13%) received PADT, and 1020 (9%) were managed with AS/WW. At a median of 9.4 yr (interquartile range 5.8-13.7) after treatment, 764 men had died from PCa. After adjusting for CAPRA score, the hazard ratios for PCSM with RP as the reference were 1.57 (95% confidence interval [CI] 1.24-1.98; p < 0.001) for BT, 1.55 (95% CI 1.26-1.91; p < 0.001) for EBRT, 2.36 (95% CI 1.94-2.87; p < 0.001) for PADT, and 1.76 (95% CI 1.30-2.40; p < 0.001) for AS/WW. In models for long-term outcomes, PCSM differences were negligible for low-risk disease and increased progressively with risk. Limitations include the evolution of diagnostic and therapeutic strategies for PCa over time. In this nonrandomized study, the possibility of residual confounding remains salient. CONCLUSIONS: In a large, prospective cohort of men with localized PCa, after adjustment for age and comorbidity, PCSM was lower after local therapy for those with higher-risk disease, and in particular after RP. Confirmation of these results via long-term follow-up of ongoing trials is awaited. PATIENT SUMMARY: We evaluated different treatment options for localized prostate cancer in a large group of patients who were treated mostly in nonacademic medical centers. Results from nonrandomized trials should be interpret with caution, but even after careful risk adjustment, survival rates for men with higher-risk cancer appeared to be highest for patients whose first treatment was surgery rather than radiotherapy, hormones, or monitoring.

Effectiveness of an artificial intelligence clinical assistant decision support system to improve the incidence of hospital-associated venous thromboembolism: a prospective, randomised controlled study.

BACKGROUND: Thromboprophylaxis has been determined to be safe, effective and cost-effective for hospitalised patients at venous thromboembolism (VTE) risk. However, Chinese medical institutions have not yet fully used or improperly used thromboprophylaxis. The effectiveness of information technology applied to thromboprophylaxis in hospitalised patients has been proved in many retrospective studies, lacking of prospective research evidence. METHODS: All hospitalised patients aged >18 years not discharged within 24 hours from 1 September 2020 to 31 May 2021 were prospectively enrolled. Patients were randomly assigned to the control (9890 patients) or intervention group (9895 patients). The control group implemented conventional VTE prevention programmes; the intervention group implemented an Artificial Intelligence Clinical Assistant Decision Support System (AI-CDSS) on the basis of conventional prevention. Intergroup demographics, disease status, hospital length of stay (LOS), VTE risk assessment and VTE prophylaxis were compared using the χ2 test, Fisher's exact test, t-test or Wilcoxon rank-sum test. Univariate and multivariate logistic regressions were used to explore the risk factor of VTE. RESULTS: The control and intervention groups had similar baseline characteristics. The mean age was 58.32±15.41 years, and mean LOS was 7.82±7.07 days. In total, 5027 (25.40%) and 2707 (13.67%) patients were assessed as having intermediate-to-high VTE risk and high bleeding risk, respectively. The incidence of hospital-associated VTE (HA-VTE) was 0.38%, of which 86.84% had deep vein thrombosis. Compared with the control group, the incidence of HA-VTE decreased by 46.00%, mechanical prophylaxis rate increased by 24.00% and intensity of drug use increased by 9.72% in the intervention group. However, AI-CDSS use did not increase the number of clinical diagnostic tests, prophylaxis rate or appropriate prophylaxis rate. CONCLUSIONS: Thromboprophylaxis is inadequate in hospitalised patients with VTE risk. The role of AI-CDSS in VTE risk management is unknown and needs further in-depth study. TRIAL REGISTRATION NUMBER: ChiCTR2000035452.

Secondary Primary Cancer Risk After Radiation Therapy in Rectal Cancer: A Population-Based Cohort Study With Propensity Score Matching.

BACKGROUND: It remains unclear whether radiation therapy (RT) has an impact on the development of secondary primary cancer (SC) in rectal cancer (RC) patients, especially within the true pelvis. AIM: To examine the incidence of SC in a population-based cohort of RC after surgical treatment with or without radiation therapy (RT, NRT). PATIENTS AND METHODS: The epidemiological cohort consisting of 13,919 RC patients with primary M0 stage diagnosed between 1998 and 2019 was collected from cancer registry data of Upper Bavaria. Competing risk analyses were conducted regarding the development of SC on 11 687 first malignancies, stratified by RT/NRT. A propensity score (PS) was generated by logistic regression modeling of RT to repeat competing risk analyses on a PS-matched cohort. RESULTS: The median age (interquartile range) of the epidemiological cohort was 68.9 years (60.4-76.7). About 60.8%, were men, 38.7% had UICC III, 35.8% of tumors were localized lower than 8 cm, 41.3% underwent RT. Only 17.1% of patients older than 80 years at diagnosis received RT. In general, RT patients were 5 years younger than NRT patients (65.9 years [58.0-73.0] vs. 71.3 years [62.4-79.2], P < .0001). The 20-year cumulative incidence of SC was 16.5% in RT and 17.4% in NRT patients (P = .2298). Men with RT had a lower risk of prostate cancer (HR = 0.55, 95%CI [0.34-0.91], P = .0168). In the PS-matched cohort, RT patients had a significantly higher risk of bladder cancer during follow-up (10-year cumulative incidence of 1.1% vs. 0.6% in NRT). The direction of the RT effects in men and women and different tumor sites may cancel each other. CONCLUSION: A protective effect of RT in rectal cancer patients on developing prostate SC by half is reproduced. Further analyses studying the long-term SC risks of RT should essentially focus on stratification by sex, and focus on more recent data.

Addressing missing data in the estimation of time-varying treatments in comparative effectiveness research.

Comparative effectiveness research is often concerned with evaluating treatment strategies sustained over time, that is, time-varying treatments. Inverse probability weighting (IPW) is often used to address the time-varying confounding by re-weighting the sample according to the probability of treatment receipt at each time point. IPW can also be used to address any missing data by re-weighting individuals according to the probability of observing the data. The combination of these two distinct sets of weights may lead to inefficient estimates of treatment effects due to potentially highly variable total weights. Alternatively, multiple imputation (MI) can be used to address the missing data by replacing each missing observation with a set of plausible values drawn from the posterior predictive distribution of the missing data given the observed data. Recent studies have compared IPW and MI for addressing the missing data in the evaluation of time-varying treatments, but they focused on missing confounders and monotone missing data patterns. This article assesses the relative advantages of MI and IPW to address missing data in both outcomes and confounders measured over time, and across monotone and non-monotone missing data settings. Through a comprehensive simulation study, we find that MI consistently provided low bias and more precise estimates compared to IPW across a wide range of scenarios. We illustrate the implications of method choice in an evaluation of biologic drugs for patients with severe rheumatoid arthritis, using the US National Databank for Rheumatic Diseases, in which 25% of participants had missing health outcomes or time-varying confounders.