Explore the vast range of titles available.

Inference about the causal structure that induces correlations between two traits can be achieved by combining genetic associations with a mediation-based approach, as is done in the causal inference test (CIT). However, we show that measurement error in the phenotypes can lead to the CIT inferring the wrong causal direction, and that increasing sample sizes has the adverse effect of increasing confidence in the wrong answer. This problem is likely to be general to other mediation-based approaches. Here we introduce an extension to Mendelian randomisation, a method that uses genetic associations in an instrumentation framework, that enables inference of the causal direction between traits, with some advantages. First, it can be performed using only summary level data from genome-wide association studies; second, it is less susceptible to bias in the presence of measurement error or unmeasured confounding. We apply the method to infer the causal direction between DNA methylation and gene expression levels. Our results demonstrate that, in general, DNA methylation is more likely to be the causal factor, but this result is highly susceptible to bias induced by systematic differences in measurement error between the platforms, and by horizontal pleiotropy. We emphasise that, where possible, implementing MR and appropriate sensitivity analyses alongside other approaches such as CIT is important to triangulate reliable conclusions about causality.

Missing data are ubiquitous in medical research. Although there is increasing guidance on how to handle missing data, practice is changing slowly and misapprehensions abound, particularly in observational research. Importantly, the lack of transparency around methodological decisions is threatening the validity and reproducibility of modern research. We present a practical framework for handling and reporting the analysis of incomplete data in observational studies, which we illustrate using a case study from the Avon Longitudinal Study of Parents and Children. The framework consists of three steps: 1) Develop an analysis plan specifying the analysis model and how missing data are going to be addressed. An important consideration is whether a complete records’ analysis is likely to be valid, whether multiple imputation or an alternative approach is likely to offer benefits and whether a sensitivity analysis regarding the missingness mechanism is required; 2) Examine the data, checking the methods outlined in the analysis plan are appropriate, and conduct the preplanned analysis; and 3) Report the results, including a description of the missing data, details on how the missing data were addressed, and the results from all analyses, interpreted in light of the missing data and the clinical relevance. This framework seeks to support researchers in thinking systematically about missing data and transparently reporting the potential effect on the study results, therefore increasing the confidence in and reproducibility of research findings. •Missing data are ubiquitous in medical research.•Guidance is available, but missing data are still often not handled appropriately.•We present a framework for handling and reporting analyses of incomplete data.•This framework encourages researchers to think systematically about missing data.•Adoption of this framework will increase the reproducibility of research findings.•This article provides a much needed framework for handling and reporting the analysis of incomplete data in observational studies.•The framework puts a strong emphasis on preplanning the statistical analysis and encourages transparency when reporting the results of a study.•Adoption of this framework will increase the confidence in and reproducibility of research findings.

The past decade has been proclaimed as a hugely successful era of gene discovery through the high yields of many genome-wide association studies (GWAS). However, much of the perceived benefit of such discoveries lies in the promise that the identification of genes that influence disease would directly translate into the identification of potential therapeutic targets, but this has yet to be realized at a level reflecting expectation. One reason for this, we suggest, is that GWAS, to date, have generally not focused on phenotypes that directly relate to the progression of disease and thus speak to disease treatment.

Background Natural or quasi experiments are appealing for public health research because they enable the evaluation of events or interventions that are difficult or impossible to manipulate experimentally, such as many policy and health system reforms. However, there remains ambiguity in the literature about their definition and how they differ from randomized controlled experiments and from other observational designs. We conceptualise natural experiments in the context of public health evaluations and align the study design to the Target Trial Framework. Methods A literature search was conducted, and key methodological papers were used to develop this work. Peer-reviewed papers were supplemented by grey literature. Results Natural experiment studies (NES) combine features of experiments and non-experiments. They differ from planned experiments, such as randomized controlled trials, in that exposure allocation is not controlled by researchers. They differ from other observational designs in that they evaluate the impact of events or process that leads to differences in exposure. As a result they are, in theory, less susceptible to bias than other observational study designs. Importantly, causal inference relies heavily on the assumption that exposure allocation can be considered ‘as-if randomized’. The target trial framework provides a systematic basis for evaluating this assumption and the other design elements that underpin the causal claims that can be made from NES. Conclusions NES should be considered a type of study design rather than a set of tools for analyses of non-randomized interventions. Alignment of NES to the Target Trial framework will clarify the strength of evidence underpinning claims about the effectiveness of public health interventions.

Numerous observational studies have attempted to identify risk factors for infection with SARS-CoV-2 and COVID-19 disease outcomes. Studies have used datasets sampled from patients admitted to hospital, people tested for active infection, or people who volunteered to participate. Here, we highlight the challenge of interpreting observational evidence from such non-representative samples. Collider bias can induce associations between two or more variables which affect the likelihood of an individual being sampled, distorting associations between these variables in the sample. Analysing UK Biobank data, compared to the wider cohort the participants tested for COVID-19 were highly selected for a range of genetic, behavioural, cardiovascular, demographic, and anthropometric traits. We discuss the mechanisms inducing these problems, and approaches that could help mitigate them. While collider bias should be explored in existing studies, the optimal way to mitigate the problem is to use appropriate sampling strategies at the study design stage. Many published studies of the current SARS-CoV-2 pandemic have analysed data from non-representative samples from populations. Here, using UK BioBank samples, Gibran Hemani and colleagues discuss the potential for such studies to suffer from collider bias, and provide suggestions for optimising study design to account for this.

The COVID-19 pandemic and mitigation measures are likely to have a marked effect on mental health. It is important to use longitudinal data to improve inferences. To quantify the prevalence of depression, anxiety and mental well-being before and during the COVID-19 pandemic. Also, to identify groups at risk of depression and/or anxiety during the pandemic. Data were from the Avon Longitudinal Study of Parents and Children (ALSPAC) index generation (n = 2850, mean age 28 years) and parent generation (n = 3720, mean age 59 years), and Generation Scotland (n = 4233, mean age 59 years). Depression was measured with the Short Mood and Feelings Questionnaire in ALSPAC and the Patient Health Questionnaire-9 in Generation Scotland. Anxiety and mental well-being were measured with the Generalised Anxiety Disorder Assessment-7 and the Short Warwick Edinburgh Mental Wellbeing Scale. Depression during the pandemic was similar to pre-pandemic levels in the ALSPAC index generation, but those experiencing anxiety had almost doubled, at 24% (95% CI 23-26%) compared with a pre-pandemic level of 13% (95% CI 12-14%). In both studies, anxiety and depression during the pandemic was greater in younger members, women, those with pre-existing mental/physical health conditions and individuals in socioeconomic adversity, even when controlling for pre-pandemic anxiety and depression. These results provide evidence for increased anxiety in young people that is coincident with the pandemic. Specific groups are at elevated risk of depression and anxiety during the COVID-19 pandemic. This is important for planning current mental health provisions and for long-term impact beyond this pandemic.

Arterial ischaemic stroke is an important cause of acquired brain injury in children. Few prospective population-based studies of childhood arterial ischaemic stroke have been undertaken. We aimed to investigate the epidemiology and clinical features of childhood arterial ischaemic stroke in a population-based cohort. Children aged 29 days to less than 16 years with radiologically confirmed arterial ischaemic stroke occurring over a 1-year period (July 1, 2008, to June 30, 2009) residing in southern England (population denominator 5·99 million children) were eligible for inclusion. Cases were identified using several sources (paediatric neurologists and trainees, the British Paediatric Neurology Surveillance Unit, paediatricians, radiologists, physiotherapists, neurosurgeons, parents, and the Paediatric Intensive Care Audit Network). Cases were confirmed by personal examination of cases and case notes. Details of presenting features, risk factors, and investigations for risk factors were recorded by analysis of case notes. Capture–recapture analysis was used to estimate completeness of ascertainment. We identified 96 cases of arterial ischaemic stroke. The crude incidence of childhood arterial ischaemic stroke was 1·60 per 100 000 per year (95% CI 1·30–1·96). Capture–recapture analysis suggested that case ascertainment was 89% (95% CI 77–97) complete. The incidence of arterial ischaemic stroke was highest in children aged under 1 year (4·14 per 100 000 per year, 95% CI 2·36–6·72). There was no difference in the risk of arterial ischaemic stroke between sexes (crude incidence 1·60 per 100 000 per year [95% CI 1·18–2·12] for boys and 1·61 per 100 000 per year [1·18–2·14] for girls). Asian (relative risk 2·14, 95% CI 1·11–3·85; p=0·017) and black (2·28, 1·00–4·60; p=0·034) children were at higher risk of arterial ischaemic stroke than were white children. 82 (85%) children had focal features (most commonly hemiparesis) at presentation. Seizures were more common in younger children (≤1 year) and headache was more common in older children (>5 years; p<0·0001). At least one risk factor for childhood arterial ischaemic stroke was identified in 80 (83%) cases. Age and racial group, but not sex, affected the risk of arterial ischaemic stroke in children. Investigation of such differences might provide causative insights. The Stroke Association, UK.

Mendelian Randomisation (MR) is a powerful tool in epidemiology that can be used to estimate the causal effect of an exposure on an outcome in the presence of unobserved confounding, by utilising genetic variants as instrumental variables (IVs) for the exposure. The effect estimates obtained from MR studies are often interpreted as the lifetime effect of the exposure in question. However, the causal effects of some exposures are thought to vary throughout an individual’s lifetime with periods during which an exposure has a greater effect on a particular outcome. Multivariable MR (MVMR) is an extension of MR that allows for multiple, potentially highly related, exposures to be included in an MR estimation. MVMR estimates the direct effect of each exposure on the outcome conditional on all the other exposures included in the estimation. We explore the use of MVMR to estimate the direct effect of a single exposure at different time points in an individual’s lifetime on an outcome. We use simulations to illustrate the interpretation of the results from such analyses and the key assumptions required. We show that causal effects at different time periods can be estimated through MVMR when the association between the genetic variants used as instruments and the exposure measured at those time periods varies. However, this estimation will not necessarily identify exact time periods over which an exposure has the most effect on the outcome. Prior knowledge regarding the biological basis of exposure trajectories can help interpretation. We illustrate the method through estimation of the causal effects of childhood and adult BMI on C-Reactive protein and smoking behaviour.

Large studies such as UK Biobank are increasingly used for GWAS and Mendelian randomization (MR) studies. However, selection into and dropout from studies may bias genetic and phenotypic associations. We examine genetic factors affecting participation in four optional components in up to 451,306 UK Biobank participants. We used GWAS to identify genetic variants associated with participation, MR to estimate effects of phenotypes on participation, and genetic correlations to compare participation bias across different studies. 32 variants were associated with participation in one of the optional components ( P  < 6 × 10 −9 ), including loci with links to intelligence and Alzheimer’s disease. Genetic correlations demonstrated that participation bias was common across studies. MR showed that longer educational duration, older menarche and taller stature increased participation, whilst higher levels of adiposity, dyslipidaemia, neuroticism, Alzheimer’s and schizophrenia reduced participation. Our effect estimates can be used for sensitivity analysis to account for selective participation biases in genetic or non-genetic analyses. Large BioBank studies are commonly used in GWAS, but may be biased by factors affecting participation and dropout. Here the authors show that some of the factors affecting participation may have underlying genetic components.

Patients with opioid dependency prescribed opioid agonist treatment (OAT) may also be prescribed sedative drugs. This may increase mortality risk but may also increase treatment duration, with overall benefit. We hypothesised that prescription of benzodiazepines in patients receiving OAT would increase risk of mortality overall, irrespective of any increased treatment duration. Data on 12,118 patients aged 15-64 years prescribed OAT between 1998 and 2014 were extracted from the Clinical Practice Research Datalink. Data from the Office for National Statistics on whether patients had died and, if so, their cause of death were available for 7,016 of these patients. We identified episodes of prescription of benzodiazepines, z-drugs, and gabapentinoids and used linear regression and Cox proportional hazards models to assess the associations of co-prescription (prescribed during OAT and up to 12 months post-treatment) and concurrent prescription (prescribed during OAT) with treatment duration and mortality. We examined all-cause mortality (ACM), drug-related poisoning (DRP) mortality, and mortality not attributable to DRP (non-DRP). Models included potential confounding factors. In 36,126 person-years of follow-up there were 657 deaths and 29,540 OAT episodes, of which 42% involved benzodiazepine co-prescription and 29% concurrent prescription (for z-drugs these respective proportions were 20% and 11%, and for gabapentinoids 8% and 5%). Concurrent prescription of benzodiazepines was associated with increased duration of methadone treatment (adjusted mean duration of treatment episode 466 days [95% CI 450 to 483] compared to 286 days [95% CI 275 to 297]). Benzodiazepine co-prescription was associated with increased risk of DRP (adjusted HR 2.96 [95% CI 1.97 to 4.43], p < 0.001), with evidence of a dose-response effect, but showed little evidence of an association with non-DRP (adjusted HR 0.91 [95% CI 0.66 to 1.25], p = 0.549). Co-prescription of z-drugs showed evidence of an association with increased risk of DRP (adjusted HR 2.75 [95% CI 1.57 to 4.83], p < 0.001) but little evidence of an association with non-DRP (adjusted HR 0.79 [95% CI 0.49 to 1.28], p = 0.342). There was no evidence of an association of gabapentinoid co-prescription with DRP (HR 1.54 [95% CI 0.60 to 3.98], p = 0.373) but evidence of an association with increased non-DRP (HR 1.83 [95% CI 1.28 to 2.62], p = 0.001). Concurrent benzodiazepine prescription also increased mortality risk after consideration of duration of OAT (adjusted HR for DRP with benzodiazepine concurrent prescription 3.34 [95% CI 2.14 to 5.20], p < 0.001). The main limitation of this study is the possibility that unmeasured confounding factors led to an association between benzodiazepine prescription and DRP that is not causal. In this study, co-prescription of benzodiazepine was specifically associated with increased risk of DRP in opioid-dependent individuals. Co-prescription of z-drugs and gabapentinoids was also associated with increased mortality risk; however, for z-drugs there was no evidence for a dose-response effect on DRP, and for gabapentinoids the increased mortality risk was not specific to DRP. Concurrent prescription of benzodiazepine was associated with longer treatment but still increased risk of death overall. Clinicians should be cautious about prescribing benzodiazepines to opioid-dependent individuals.