On variance estimation of the inverse probability-of-treatment weighting estimator: A tutorial for different types of propensity score weights.

Propensity score methods, such as inverse probability-of-treatment weighting (IPTW), have been increasingly used for covariate balancing in both observational studies and randomized trials, allowing the control of both systematic and chance imbalances. Approaches using IPTW are based on two steps: (i) estimation of the individual propensity scores (PS), and (ii) estimation of the treatment effect by applying PS weights. Thus, a variance estimator that accounts for both steps is crucial for correct inference. Using a variance estimator which ignores the first step leads to overestimated variance when the estimand is the average treatment effect (ATE), and to under or overestimated estimates when targeting the average treatment effect on the treated (ATT). In this article, we emphasize the importance of using an IPTW variance estimator that correctly considers the uncertainty in PS estimation. We present a comprehensive tutorial to obtain unbiased variance estimates, by proposing and applying a unifying formula for different types of PS weights (ATE, ATT, matching and overlap weights). This can be derived either via the linearization approach or M-estimation. Extensive R code is provided along with the corresponding large-sample theory. We perform simulation studies to illustrate the behavior of the estimators under different treatment and outcome prevalences and demonstrate appropriate behavior of the analytical variance estimator. We also use a reproducible analysis of observational lung cancer data as an illustrative example, estimating the effect of receiving a PET-CT scan on the receipt of surgery.

Use and reporting of inverse-probability-of-treatment weighting for multicategory treatments in medical research: a systematic review.

OBJECTIVES: Causal inference methods for observational data represent an alternative to randomised controlled trials when they are not feasible or when real-world evidence is sought. Inverse-probability-of-treatment weighting (IPTW) is one of the most popular approaches to account for confounding in observational studies. In medical research, IPTW is mainly applied to estimate the causal effect of a binary treatment, even when the treatment has in fact multiple categories, despite the availability of IPTW estimators for multiple treatment categories. This raises questions about the appropriateness of the use of IPTW in this context. Therefore, we conducted a systematic review of medical publications reporting the use of IPTW in the presence of a multi-category treatment. Our objectives were to investigate the frequency of use and the implementation of these methods in practice, and to assess the quality of their reporting. STUDY DESIGN AND SETTING: Using Pubmed, Embase and Web of Science, we screened 5660 articles and retained 106 articles in the final analysis that were from 17 different medical areas. This systematic review is registered on PROSPERO (CRD42022352669). RESULTS: The number of treatment groups varied between 3 and 9, with a large majority of articles (90 [84.9%]) including 3 or 4 groups. The most commonly used method for estimating the weights was multinomial regression (51 [48.1%]) and generalized boosted models (48 [45.3%]). The covariates of the weight model were reported in 91 articles (85.9 %). Twenty-six articles (24.5 %) did not discuss the balance of covariates after weighting, and only 16 articles (15.1 %) referred to the assumptions needed to obtain correct inferences. CONCLUSION: The results of this systematic review illustrate that medical publications scarcely use IPTW methods for more than two treatment categories. Among the publications that did, the quality of reporting was suboptimal, in particular in regard to the assumptions and model building. IPTW for multi-category treatments could be applied more broadly in medical research, and the application of the proposed guidelines in this context will help researchers to report their results and to ensure reproducibility of their research.

Practical considerations for sample size calculation for cluster randomized trials.

Cluster randomized trials are an essential design in public health and medical research, when individual randomization is infeasible or undesirable for scientific or logistical reasons. However, the correlation among observations within clusters leads to a decrease in statistical power compared to an individually randomised trial with the same total sample size. This correlation - often quantified using the intra-cluster correlation coefficient - must be accounted for in the sample size calculation to ensure that the trial is adequately powered. In this paper, we first describe the principles of sample size calculation for parallel-arm CRTs, and explain how these calculations can be extended to CRTs with cross-over designs, with a baseline measurement and stepped-wedge designs. We introduce tools to guide researchers with their sample size calculation and discuss methods to inform the choice of the a priori estimate of the intra-cluster correlation coefficient for the calculation. We also include additional considerations with respect to anticipated attrition, a small number of clusters, and use of covariates in the randomisation process and in the analysis.

How should a cluster randomized trial be analyzed?

In cluster randomized trials, individuals from the same cluster tend to have more similar outcomes than individuals from different clusters. This correlation must be taken into account in the analysis of every cluster trial to avoid incorrect inferences. In this paper, we describe the principles guiding the analysis of cluster trials including how to correctly account for intra-cluster correlations as well as how to analyze more advanced designs such as stepped-wedge and cluster cross-over trials. We then describe how to handle specific issues such as small sample sizes and missing data.

Risk of death following chikungunya virus disease in the 100 Million Brazilian Cohort, 2015-18: a matched cohort study and self-controlled case series.

BACKGROUND: Chikungunya virus outbreaks have been associated with excess deaths at the ecological level. Previous studies have assessed the risk factors for severe versus mild chikungunya virus disease. However, the risk of death following chikungunya virus disease compared with the risk of death in individuals without the disease remains unexplored. We aimed to investigate the risk of death in the 2 years following chikungunya virus disease. METHODS: We used a population-based cohort study and a self-controlled case series to estimate mortality risks associated with chikungunya virus disease between Jan 1, 2015, and Dec 31, 2018, in Brazil. The dataset was created by linking national databases for social programmes, notifiable diseases, and mortality. For the matched cohort design, individuals with chikungunya virus disease recorded between Jan 1, 2015, and Dec 31, 2018, were considered as exposed and those who were arbovirus disease-free and alive during the study period were considered as unexposed. For the self-controlled case series, we included all deaths from individuals with a chikungunya virus disease record, and each individual acted as their own control according to different study periods relative to the date of disease. The primary outcome was all-cause natural mortality up to 728 days after onset of chikungunya virus disease symptoms, and secondary outcomes were cause-specific deaths, including ischaemic heart diseases, diabetes, and cerebrovascular diseases. FINDINGS: In the matched cohort study, we included 143 787 individuals with chikungunya virus disease who were matched, at the day of symptom onset, to unexposed individuals using sociodemographic factors. The incidence rate ratio (IRR) of death within 7 days of chikungunya symptom onset was 8·40 (95% CI 4·83-20·09) as compared with the unexposed group and decreased to 2·26 (1·50-3·77) at 57-84 days and 1·05 (0·82-1·35) at 85-168 days, with IRR close to 1 and wide CI in the subsequent periods. For the secondary outcomes, the IRR of deaths within 28 days after disease onset were: 1·80 (0·58-7·00) for cerebrovascular diseases, 3·75 (1·33-17·00) for diabetes, and 3·67 (1·25-14·00) for ischaemic heart disease, and there was no evidence of increased risk in the subsequent periods. For the self-controlled case series study, 1933 individuals died after having had chikungunya virus disease and were included in the analysis. The IRR of all-cause natural death within 7 days of symptom onset of chikungunya virus disease was 8·75 (7·18-10·66) and decreased to 1·59 (1·26-2·00) at 57-84 days and 1·09 (0·92-1·29) at 85-168 days. For the secondary outcomes, the IRRs of deaths within 28 days after disease onset were: 2·73 (1·50-4·96) for cerebrovascular diseases, 8·43 (5·00-14·21) for diabetes, and 2·38 (1·33-4·26) for ischaemic heart disease, and there was no evidence of increased risk at 85-168 days. INTERPRETATION: Chikungunya virus disease is associated with an increased risk of death for up to 84 days after symptom onset, including deaths from cerebrovascular diseases, ischaemic heart diseases, and diabetes. This study highlights the need for equitable access to approved vaccines and effective anti-chikungunya virus therapeutics and reinforces the importance of robust vector-control efforts to reduce viral transmission. FUNDING: Brazilian National Research Council (CNPq), Fundação de Amparo à Pesquisa do Estado da Bahia, Wellcome Trust, and UK Medical Research Council. TRANSLATION: For the Portuguese translation of the abstract see Supplementary Materials section.

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.

Association Between Fluoroquinolone Use and Hospitalization With Aortic Aneurysm or Aortic Dissection.

Importance: Fluoroquinolone use has been associated with increased hospitalization with aortic aneurysm or dissection in noninterventional studies, but the reason for this observed association is unclear. Objective: To determine the association between fluoroquinolone use and aortic aneurysm or dissection using multiple study designs and multiple databases to increase the robustness of findings. Design, Setting, and Participants: Cohort and case-crossover studies were conducted separately in 2 databases of UK primary care records. Clinical Practice Research Datalink Aurum and GOLD primary care records were linked to hospital admissions data. Adults with a systemic fluoroquinolone or cephalosporin prescription between April 1997 and December 2019 were included in the cohort study. Adults hospitalized with aortic aneurysm or dissection within the eligibility period were included in the case-crossover study. Individuals meeting inclusion criteria in the case-crossover study were matched 1:3 to control individuals on age, sex, index date, and clinical practice to adjust for calendar trends in prescribing. Data were analyzed from January to July 2022. Exposures: Systemic fluoroquinolone or comparator antibiotic. Main Outcomes and Measures: Hazard ratios (HRs) were estimated in the cohort study for the association between prescription of fluoroquinolones and hospitalization with aortic aneurysm or dissection using stabilized inverse probability of treatment-weighted Cox regression. Odds ratios (OR) were estimated in the case-crossover study for the association between systemic fluoroquinolone use and hospitalization with aortic aneurysm or dissection using a conditional logistic regression model. Estimates were pooled across databases using fixed-effects meta-analysis. Results: In the cohort study, we identified 3 134 121 adults in Aurum (mean [SD] age, 52.5 [20.3] years; 1 969 257 [62.8%] female) and 452 086 in GOLD (mean [SD] age, 53.9 [20.2] years; 286 502 [63.4%] female) who were prescribed fluoroquinolones or cephalosporins. In crude analyses, fluoroquinolone relative to cephalosporin use was associated with increased hospitalization with aortic aneurysm or dissection (pooled HR, 1.28; 95% CI, 1.13-1.44; P < .001) but after adjustment for potential confounders, this association disappeared (pooled adjusted HR, 1.03; 95% CI, 0.91-1.17; P = .65). In the case-crossover study, we identified 84 841 individuals hospitalized with aortic aneurysm or dissection in Aurum (mean [SD] age, 75.5 [10.9]; 23 551 [27.8%] female) and 10 357 in GOLD (mean [SD] age, 75.6 [10.5]; 2809 [27.1%] female). Relative to nonuse, fluoroquinolone use was associated with an increase in hospitalization with aortic aneurysm or dissection, but no association was found relative to other antibiotics (vs cephalosporin pooled OR, 1.05; 95% CI, 0.87-1.27; vs trimethoprim, 0.89; 95% CI, 0.75-1.06; vs co-amoxiclav, 0.98; 95% CI, 0.82-1.18). Conclusions and Relevance: The results in this study suggest that estimates of association of fluoroquinolones with aortic aneurysm or dissection may be affected by confounding. When such confounding is accounted for, no association was evident, providing reassurance on the safety of fluoroquinolones with respect to aortic aneurysm or dissection.

Associations between treatments, comorbidities and multidimensional aspects of quality of life among patients with advanced cancer in the Netherlands-a 2017-2020 multicentre cross-sectional study.

OBJECTIVE: To investigate associations between quality of life (QoL) and 1) immunotherapy and other cancer treatments received three months before QoL measurements, and 2) the comorbidities at the time of completion or in the year prior to QoL measurements, among patients with advanced cancer. METHODS: A cross-sectional study is conducted on patients with advanced cancer in the Netherlands. The data come from the baseline wave of the 2017-2020 eQuiPe study. Participants were surveyed via questionnaires (including EORTC QLQ-C30). Using multivariable linear and logistic regression models, we explored statistical associations between QoL components and immunotherapy and other cancer treatments as well as pre-existing comorbidities while adjusting for age, sex, socio-economic status. RESULTS: Of 1088 participants with median age 67 years, 51% were men. Immunotherapy was not associated with global QoL but was associated with reduced appetite loss (odds ratio (OR) = 0.6, 95%CI = [0.3,0.9]). Reduced global QoL was associated with chemotherapy (adjusted mean difference (ß) = - 4.7, 95% CI [- 8.5,- 0.8]), back pain (ß = - 7.4, 95% CI [- 11.0,- 3.8]), depression (ß = - 13.8, 95% CI [- 21.5,- 6.2]), thyroid diseases (ß = - 8.9, 95% CI [- 14.0,- 3.8]) and diabetes (ß = - 4.5, 95% CI [- 8.9,- 0.5]). Chemotherapy was associated with lower physical (OR = 2.4, 95% CI [1.5,3.9]) and role (OR = 1.8, 95% CI [1.2,2.7]) functioning, and higher pain (OR = 1.9, 95% CI [1.3,2.9]) and fatigue (OR = 1.6, 95% CI [1.1,2.4]). CONCLUSION: Our study identified associations between specific cancer treatments, lower QoL and more symptoms. Monitoring symptoms may improve QoL of patients with advanced cancer. Producing more evidence from real life data would help physicians in better identifying patients who require additional supportive care.

Effects of proactive vs fixed community health care delivery on child health and access to care: a cluster randomised trial secondary endpoint analysis.

Background: Professional community health workers (CHWs) can help achieve universal health coverage, although evidence gaps remain on how to optimise CHW service delivery. We conducted an unblinded, parallel, cluster randomised trial in rural Mali to determine whether proactive CHW delivery reduced mortality and improved access to health care among children under five years, compared to passive delivery. Here we report the secondary access endpoints. Methods: Beginning from 26-28 February 2017, 137 village-clusters were offered care by CHWs embedded in communities who were trained, paid, supervised, and integrated into a reinforced public-sector health system that did not charge user fees. Clusters were randomised (stratified on primary health centre catchment and distance) to care during CHWs during door-to-door home visits (intervention) or based at a fixed village site (control). We measured outcomes at baseline, 12-, 24-, and 36-month time points with surveys administered to all resident women aged 15-49 years. We used logistic regression with cluster-level random effects to estimate intention-to-treat and per-protocol effects over time on prompt (24-hour) treatment within the health sector. Results: Follow-up surveys between February 2018 and April 2020 generated 20 105 child-year observations. Across arms, prompt health sector treatment more than doubled compared to baseline. At 12 months, children in intervention clusters had 22% higher odds of receiving prompt health sector treatment than those in control (cluster-specific adjusted odds ratio (aOR) = 1.22; 95% confidence interval (CI) = 1.06, 1.41, P = 0.005), or 4.7 percentage points higher (adjusted risk difference (aRD) = 0.047; 95% CI = 0.014, 0.080). We found no evidence of an effect at 24 or 36 months. Conclusions: CHW-led health system redesign likely drove the 2-fold increase in rapid child access to care. In this context, proactive home visits further improved early access during the first year but waned afterwards. Registration: ClinicalTrials.gov NCT02694055.

Monitoring multidimensional aspects of quality of life after cancer immunotherapy: protocol for the international multicentre, observational QUALITOP cohort study.

INTRODUCTION: Immunotherapies, such as immune checkpoint inhibitors and chimeric antigen receptor T-cell therapy, have significantly improved the clinical outcomes of various malignancies. However, they also cause immune-related adverse events (irAEs) that can be challenging to predict, prevent and treat. Although they likely interact with health-related quality of life (HRQoL), most existing evidence on this topic has come from clinical trials with eligibility criteria that may not accurately reflect real-world settings. The QUALITOP project will study HRQoL in relation to irAEs and its determinants in a real-world study of patients treated with immunotherapy. METHODS AND ANALYSIS: This international, observational, multicentre study takes place in France, the Netherlands, Portugal and Spain. We aim to include about 1800 adult patients with cancer treated with immunotherapy in a specifically recruited prospective cohort, and to additionally obtain data from historical real-world databases (ie, databiobanks) and medical administrative registries (ie, national cancer registries) in which relevant data regarding other adult patients with cancer treated with immunotherapy has already been stored. In the prospective cohort, clinical health status, HRQoL and psychosocial well-being will be monitored until 18 months after treatment initiation through questionnaires (at baseline and 3, 6, 12 and 18 months thereafter), and by data extraction from electronic patient files. Using advanced statistical methods, including causal inference methods, artificial intelligence algorithms and simulation modelling, we will use data from the QUALITOP cohort to improve the understanding of the complex relationships among treatment regimens, patient characteristics, irAEs and HRQoL. ETHICS AND DISSEMINATION: All aspects of the QUALITOP project will be conducted in accordance with the Declaration of Helsinki and with ethical approval from a suitable local ethics committee, and all patients will provide signed informed consent. In addition to standard dissemination efforts in the scientific literature, the data and outcomes will contribute to a smart digital platform and medical data lake. These will (1) help increase knowledge about the impact of immunotherapy, (2) facilitate improved interactions between patients, clinicians and the general population and (3) contribute to personalised medicine. TRIAL REGISTRATION NUMBER: NCT05626764.

Immortal-time bias in older vs younger age groups: a simulation study with application to a population-based cohort of patients with colon cancer.

BACKGROUND: In observational studies, the risk of immortal-time bias (ITB) increases with the likelihood of early death, itself increasing with age. We investigated how age impacts the magnitude of ITB when estimating the effect of surgery on 1-year overall survival (OS) in patients with Stage IV colon cancer aged 50-74 and 75-84 in England. METHODS: Using simulations, we compared estimates from a time-fixed exposure model to three statistical methods addressing ITB: time-varying exposure, delayed entry and landmark methods. We then estimated the effect of surgery on OS using a population-based cohort of patients from the CORECT-R resource and conducted the analysis using the emulated target trial framework. RESULTS: In simulations, the magnitude of ITB was larger among older patients when their probability of early death increased or treatment was delayed. The bias was corrected using the methods addressing ITB. When applied to CORECT-R data, these methods yielded a smaller effect of surgery than the time-fixed exposure approach but effects were similar in both age groups. CONCLUSION: ITB must be addressed in all longitudinal studies, particularly, when investigating the effect of exposure on an outcome in different groups of people (e.g., age groups) with different distributions of exposure and outcomes.

Long-term oral antibiotic use in people with acne vulgaris in UK primary care: a drug utilization study.

BACKGROUND: The inappropriate use of antibiotics is understood to contribute to antimicrobial resistance. Oral antibiotics are regularly used to treat moderate-to-severe acne vulgaris. In practice, we do not know the typical length of oral antibiotic treatment courses for acne in routine primary care and what proportion of people receive more than one course of treatment following a new acne diagnosis. OBJECTIVES: To describe how oral antibiotics are prescribed for acne over time in UK primary care. METHODS: We conducted a descriptive longitudinal drug utilization study using routinely collected primary care data from the Clinical Practice Research Datalink GOLD (2004-2019). We included individuals (8-50 years) with a new acne diagnosis recorded between 1 January 2004 and 31 July 2019. RESULTS: We identified 217 410 people with a new acne diagnosis. The median age was 17 years [interquartile range (IQR) 15-25] and median follow-up was 4.3 years (IQR 1.9-7.6). Among people with a new acne diagnosis, 96 703 (44.5%) received 248 560 prescriptions for long-term oral antibiotics during a median follow-up of 5.3 years (IQR 2.8-8.5). The median number of continuous courses of antibiotic therapy (≥ 28 days) per person was four (IQR 2-6). The majority (n = 59 010, 61.0%) of first oral antibiotic prescriptions in those with a recorded acne diagnosis were between the ages of 12 and 18. Most (n = 71 544, 74.0%) first courses for oral antibiotics were for between 28 and 90 days. The median duration of the first course of treatment was 56 days (IQR 50-93 days) and 18 127 (18.7%) of prescriptions of ≥ 28 days were for < 6 weeks. Among people who received a first course of oral antibiotic for ≥ 28 days, 56 261 (58.2%) received a second course after a treatment gap of ≥ 28 days. The median time between first and second courses was 135 days (IQR 67-302). The cumulative duration of exposure to oral antibiotics during follow-up was 255 days (8.5 months). CONCLUSIONS: Further work is needed to understand the consequences of using antibiotics for shorter periods than recommended. Suboptimal treatment duration may result in reduced clinical effectiveness or repeated exposures, potentially contributing to antimicrobial resistance.

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.

Propensity score matching after multiple imputation when a confounder has missing data.

One of the main challenges when using observational data for causal inference is the presence of confounding. A classic approach to account for confounding is the use of propensity score techniques that provide consistent estimators of the causal treatment effect under four common identifiability assumptions for causal effects, including that of no unmeasured confounding. Propensity score matching is a very popular approach which, in its simplest form, involves matching each treated patient to an untreated patient with a similar estimated propensity score, that is, probability of receiving the treatment. The treatment effect can then be estimated by comparing treated and untreated patients within the matched dataset. When missing data arises, a popular approach is to apply multiple imputation to handle the missingness. The combination of propensity score matching and multiple imputation is increasingly applied in practice. However, in this article we demonstrate that combining multiple imputation and propensity score matching can lead to over-coverage of the confidence interval for the treatment effect estimate. We explore the cause of this over-coverage and we evaluate, in this context, the performance of a correction to Rubin's rules for multiple imputation proposed by finding that this correction removes the over-coverage.

Primary prevention of acute cardiovascular events by influenza vaccination: an observational study.

AIMS: Previous studies show a reduced incidence of first myocardial infarction and stroke 1-3 months after influenza vaccination, but it is unclear how underlying cardiovascular risk impacts the association. METHODS AND RESULTS: The study used linked Clinical Practice Research Datalink, Hospital Episode Statistics Admitted Patient Care and Office for National Statistics mortality data from England between 1 September 2008 and 31 August 2019. From the data, individuals aged 40-84 years with a first acute cardiovascular event and influenza vaccination occurring within 12 months of each September were selected. Using a self-controlled case series analysis, season-adjusted cardiovascular risk stratified incidence ratios (IRs) for cardiovascular events after vaccination compared with baseline time before and >120 days after vaccination were generated. 193 900 individuals with a first acute cardiovascular event and influenza vaccine were included. 105 539 had hypertension and 172 050 had a QRISK2 score ≥10%. In main analysis, acute cardiovascular event risk was reduced in the 15-28 days after vaccination [IR 0.72 (95% CI 0.70-0.74)] and, while the effect size tapered, remained reduced to 91-120 days after vaccination [0.83 (0.81-0.88)]. Reduced cardiovascular events were seen after vaccination among individuals of all age groups and with raised and low cardiovascular risk. CONCLUSIONS: Influenza vaccine may offer cardiovascular benefit among individuals at varying cardiovascular risk. Further studies are needed to characterize the populations who could derive the most cardiovascular benefits from vaccination.

Cluster randomised trials with a binary outcome and a small number of clusters: comparison of individual and cluster level analysis method.

BACKGROUND: Cluster randomised trials (CRTs) are often designed with a small number of clusters, but it is not clear which analysis methods are optimal when the outcome is binary. This simulation study aimed to determine (i) whether cluster-level analysis (CL), generalised linear mixed models (GLMM), and generalised estimating equations with sandwich variance (GEE) approaches maintain acceptable type-one error including the impact of non-normality of cluster effects and low prevalence, and if so (ii) which methods have the greatest power. We simulated CRTs with 8-30 clusters, altering the cluster-size, outcome prevalence, intracluster correlation coefficient, and cluster effect distribution. We analysed each dataset with weighted and unweighted CL; GLMM with adaptive quadrature and restricted pseudolikelihood; GEE with Kauermann-and-Carroll and Fay-and-Graubard sandwich variance using independent and exchangeable working correlation matrices. P-values were from a t-distribution with degrees of freedom (DoF) as clusters minus cluster-level parameters; GLMM pseudolikelihood also used Satterthwaite and Kenward-Roger DoF. RESULTS: Unweighted CL, GLMM pseudolikelihood, and Fay-and-Graubard GEE with independent or exchangeable working correlation matrix controlled type-one error in > 97% scenarios with clusters minus parameters DoF. Cluster-effect distribution and prevalence of outcome did not usually affect analysis method performance. GEE had the least power. With 20-30 clusters, GLMM had greater power than CL with varying cluster-size but similar power otherwise; with fewer clusters, GLMM had lower power with common cluster-size, similar power with medium variation, and greater power with large variation in cluster-size. CONCLUSION: We recommend that CRTs with ≤ 30 clusters and a binary outcome use an unweighted CL or restricted pseudolikelihood GLMM both with DoF clusters minus cluster-level parameters.

Causal and Associational Language in Observational Health Research: A Systematic Evaluation.

We estimated the degree to which language used in the high-profile medical/public health/epidemiology literature implied causality using language linking exposures to outcomes and action recommendations; examined disconnects between language and recommendations; identified the most common linking phrases; and estimated how strongly linking phrases imply causality. We searched for and screened 1,170 articles from 18 high-profile journals (65 per journal) published from 2010-2019. Based on written framing and systematic guidance, 3 reviewers rated the degree of causality implied in abstracts and full text for exposure/outcome linking language and action recommendations. Reviewers rated the causal implication of exposure/outcome linking language as none (no causal implication) in 13.8%, weak in 34.2%, moderate in 33.2%, and strong in 18.7% of abstracts. The implied causality of action recommendations was higher than the implied causality of linking sentences for 44.5% or commensurate for 40.3% of articles. The most common linking word in abstracts was "associate" (45.7%). Reviewers' ratings of linking word roots were highly heterogeneous; over half of reviewers rated "association" as having at least some causal implication. This research undercuts the assumption that avoiding "causal" words leads to clarity of interpretation in medical research.

Extracorporeal Membrane Oxygenation for Severe Acute Respiratory Distress Syndrome Associated with COVID-19: An Emulated Target Trial Analysis.

Rationale: Whether patients with coronavirus disease (COVID-19) may benefit from extracorporeal membrane oxygenation (ECMO) compared with conventional invasive mechanical ventilation (IMV) remains unknown. Objectives: To estimate the effect of ECMO on 90-day mortality versus IMV only. Methods: Among 4,244 critically ill adult patients with COVID-19 included in a multicenter cohort study, we emulated a target trial comparing the treatment strategies of initiating ECMO versus no ECMO within 7 days of IMV in patients with severe acute respiratory distress syndrome (PaO2/FiO2 < 80 or PaCO2 ⩾ 60 mm Hg). We controlled for confounding using a multivariable Cox model on the basis of predefined variables. Measurements and Main Results: A total of 1,235 patients met the full eligibility criteria for the emulated trial, among whom 164 patients initiated ECMO. The ECMO strategy had a higher survival probability on Day 7 from the onset of eligibility criteria (87% vs. 83%; risk difference, 4%; 95% confidence interval, 0-9%), which decreased during follow-up (survival on Day 90: 63% vs. 65%; risk difference, -2%; 95% confidence interval, -10 to 5%). However, ECMO was associated with higher survival when performed in high-volume ECMO centers or in regions where a specific ECMO network organization was set up to handle high demand and when initiated within the first 4 days of IMV and in patients who are profoundly hypoxemic. Conclusions: In an emulated trial on the basis of a nationwide COVID-19 cohort, we found differential survival over time of an ECMO compared with a no-ECMO strategy. However, ECMO was consistently associated with better outcomes when performed in high-volume centers and regions with ECMO capacities specifically organized to handle high demand.

Application of Inverse-Probability-of-Treatment Weighting to Estimate the Effect of Daytime Sleepiness in Patients with Obstructive Sleep Apnea.

Rationale: Continuous positive airway pressure (CPAP), the first line therapy for obstructive sleep apnea (OSA), is considered effective in reducing daytime sleepiness. Its efficacy relies on adequate adherence, often defined as >4 hours per night. However, this binary threshold may limit our understanding of the causal effect of CPAP adherence and daytime sleepiness, and a multilevel approach for CPAP adherence can be more appropriate. Objectives: In this study, we show how two causal inference methods can be applied on observational data for the estimation of the effect of different ranges of CPAP adherence on daytime sleepiness as measured by the Epworth Sleepiness Scale (ESS). Methods: Data were collected from a large prospective observational French cohort for patients with OSA. Four groups of CPAP adherence were considered (0-4, 4-6, 6-7, and 7-10 h per night). Multivariable regression, inverse-probability-of-treatment weighting (IPTW), and inverse propensity weighting with regression adjustment (IPW-RA) were used to assess the impact of CPAP adherence level on daytime sleepiness. Results: In this study, 9,244 patients with OSA treated by CPAP were included. The mean initial ESS score was 11 (±5.2), with a mean reduction of 4 points (±5.1). Overall, there was evidence of the causal effect of CPAP adherence on daytime sleepiness which was mainly observed between the lower CPAP adherence group (0-4 h) compared with the higher CPAP adherence group (7-10 h). There are no differences by considering higher level of CPAP adherence (>4 h). Conclusions: We showed that IPTW and IPW-RA can be easily implemented to answer questions regarding causal effects using observational data when randomized trials cannot be conducted. Both methods give a direct causal interpretation at the population level and allow the assessment of the appropriate consideration of measured confounders.

Effects of ACE inhibitors and angiotensin receptor blockers: protocol for a UK cohort study using routinely collected electronic health records with validation against the ONTARGET trial.

INTRODUCTION: Cardiovascular disease is a leading cause of death globally, responsible for nearly 18 million deaths worldwide in 2017. Medications to reduce the risk of cardiovascular events are prescribed based on evidence from clinical trials which explore treatment effects in an indicated sample of the general population. However, these results may not be fully generalisable because of trial eligibility criteria that generally restrict to younger patients with fewer comorbidities. Therefore, evidence of effectiveness of medications for groups underrepresented in clinical trials such as those aged ≥75 years, from ethnic minority backgrounds or with low kidney function may be limited.Using individual anonymised data from the Ongoing Telmisartan Alone and the Ramipril Global Endpoint Trial (ONTARGET) trial, in collaboration with the original trial investigators, we aim to investigate clinical trial replicability within a real-world setting in the area of cardiovascular disease. If the original trial results are replicable, we will estimate treatment effects and risk in groups underrepresented and excluded from the original clinical trial. METHODS AND ANALYSIS: We will develop a cohort analogous to the ONTARGET trial within the Clinical Practice Research Datalink between 1 January 2001 and 31 July 2019 using the trial eligibility criteria and propensity score matching. The primary outcome is a composite of cardiovascular death, non-fatal myocardial infarction, non-fatal stroke and hospitalisation for congestive heart failure. If results from the cohort study fall within pre-specified limits, we will expand the cohort to include under represented and excluded groups. ETHICS AND DISSEMINATION: Ethical approval has been granted by the London School of Hygiene & Tropical Medicine Ethics Committee (Ref: 22658). The study has been approved by the Independent Scientific Advisory Committee of the UK Medicines and Healthcare Products Regulatory Agency (protocol no. 20_012). Access to the individual patient data from the ONTARGET trial was obtained by the trial investigators. Findings will be submitted to peer-reviewed journals and presented at conferences.