Explore the vast range of titles available.

The associations of blood pressure with the different manifestations of incident cardiovascular disease in a contemporary population have not been compared. In this study, we aimed to analyse the associations of blood pressure with 12 different presentations of cardiovascular disease. We used linked electronic health records from 1997 to 2010 in the CALIBER (CArdiovascular research using LInked Bespoke studies and Electronic health Records) programme to assemble a cohort of 1·25 million patients, 30 years of age or older and initially free from cardiovascular disease, a fifth of whom received blood pressure-lowering treatments. We studied the heterogeneity in the age-specific associations of clinically measured blood pressure with 12 acute and chronic cardiovascular diseases, and estimated the lifetime risks (up to 95 years of age) and cardiovascular disease-free life-years lost adjusted for other risk factors at index ages 30, 60, and 80 years. This study is registered at ClinicalTrials.gov, number NCT01164371. During 5·2 years median follow-up, we recorded 83 098 initial cardiovascular disease presentations. In each age group, the lowest risk for cardiovascular disease was in people with systolic blood pressure of 90–114 mm Hg and diastolic blood pressure of 60–74 mm Hg, with no evidence of a J-shaped increased risk at lower blood pressures. The effect of high blood pressure varied by cardiovascular disease endpoint, from strongly positive to no effect. Associations with high systolic blood pressure were strongest for intracerebral haemorrhage (hazard ratio 1·44 [95% CI 1·32–1·58]), subarachnoid haemorrhage (1·43 [1·25–1·63]), and stable angina (1·41 [1·36–1·46]), and weakest for abdominal aortic aneurysm (1·08 [1·00–1·17]). Compared with diastolic blood pressure, raised systolic blood pressure had a greater effect on angina, myocardial infarction, and peripheral arterial disease, whereas raised diastolic blood pressure had a greater effect on abdominal aortic aneurysm than did raised systolic pressure. Pulse pressure associations were inverse for abdominal aortic aneurysm (HR per 10 mm Hg 0·91 [95% CI 0·86–0·98]) and strongest for peripheral arterial disease (1·23 [1·20–1·27]). People with hypertension (blood pressure ≥140/90 mm Hg or those receiving blood pressure-lowering drugs) had a lifetime risk of overall cardiovascular disease at 30 years of age of 63·3% (95% CI 62·9–63·8) compared with 46·1% (45·5–46·8) for those with normal blood pressure, and developed cardiovascular disease 5·0 years earlier (95% CI 4·8–5·2). Stable and unstable angina accounted for most (43%) of the cardiovascular disease-free years of life lost associated with hypertension from index age 30 years, whereas heart failure and stable angina accounted for the largest proportion (19% each) of years of life lost from index age 80 years. The widely held assumptions that blood pressure has strong associations with the occurrence of all cardiovascular diseases across a wide age range, and that diastolic and systolic associations are concordant, are not supported by the findings of this high-resolution study. Despite modern treatments, the lifetime burden of hypertension is substantial. These findings emphasise the need for new blood pressure-lowering strategies, and will help to inform the design of randomised trials to assess them. Medical Research Council, National Institute for Health Research, and Wellcome Trust.

Background Multiple imputation is a commonly used method for handling incomplete covariates as it can provide valid inference when data are missing at random. This depends on being able to correctly specify the parametric model used to impute missing values, which may be difficult in many realistic settings. Imputation by predictive mean matching (PMM) borrows an observed value from a donor with a similar predictive mean; imputation by local residual draws (LRD) instead borrows the donor’s residual. Both methods relax some assumptions of parametric imputation, promising greater robustness when the imputation model is misspecified. Methods We review development of PMM and LRD and outline the various forms available, and aim to clarify some choices about how and when they should be used. We compare performance to fully parametric imputation in simulation studies, first when the imputation model is correctly specified and then when it is misspecified. Results In using PMM or LRD we strongly caution against using a single donor, the default value in some implementations, and instead advocate sampling from a pool of around 10 donors. We also clarify which matching metric is best. Among the current MI software there are several poor implementations. Conclusions PMM and LRD may have a role for imputing covariates (i) which are not strongly associated with outcome, and (ii) when the imputation model is thought to be slightly but not grossly misspecified. Researchers should spend efforts on specifying the imputation model correctly, rather than expecting predictive mean matching or local residual draws to do the work.

Fetal growth restriction is a major determinant of adverse perinatal outcome. Screening procedures for fetal growth restriction need to identify small babies and then differentiate between those that are healthy and those that are pathologically small. We sought to determine the diagnostic effectiveness of universal ultrasonic fetal biometry in the third trimester as a screening test for small-for-gestational-age (SGA) infants, and whether the risk of morbidity associated with being small differed in the presence or absence of ultrasonic markers of fetal growth restriction. The Pregnancy Outcome Prediction (POP) study was a prospective cohort study of nulliparous women with a viable singleton pregnancy at the time of the dating ultrasound scan. Women participating had clinically indicated ultrasonography in the third trimester as per routine clinical care and these results were reported as usual (selective ultrasonography). Additionally, all participants had research ultrasonography, including fetal biometry at 28 and 36 weeks' gestational age. These results were not made available to participants or treating clinicians (universal ultrasonography). We regarded SGA as a birthweight of less than the 10th percentile for gestational age and screen positive for SGA an ultrasonographic estimated fetal weight of less than the 10th percentile for gestational age. Markers of fetal growth restriction included biometric ratios, utero-placental Doppler, and fetal growth velocity. We assessed outcomes for consenting participants who attended research scans and had a livebirth at the Rosie Hospital (Cambridge, UK) after the 28 weeks' research scan. Between Jan 14, 2008, and July 31, 2012, 4512 women provided written informed consent of whom 3977 (88%) were eligible for analysis. Sensitivity for detection of SGA infants was 20% (95% CI 15–24; 69 of 352 fetuses) for selective ultrasonography and 57% (51–62; 199 of 352 fetuses) for universal ultrasonography (relative sensitivity 2·9, 95% CI 2·4–3·5, p<0·0001). Of the 3977 fetuses, 562 (14·1%) were identified by universal ultrasonography with an estimated fetal weight of less than the 10th percentile and were at an increased risk of neonatal morbidity (relative risk [RR] 1·60, 95% CI 1·22–2·09, p=0·0012). However, estimated fetal weight of less than the 10th percentile was only associated with the risk of neonatal morbidity (pinteraction=0·005) if the fetal abdominal circumference growth velocity was in the lowest decile (RR 3·9, 95% CI 1·9–8·1, p=0·0001). 172 (4%) of 3977 pregnancies had both an estimated fetal weight of less than the 10th percentile and abdominal circumference growth velocity in the lowest decile, and had a relative risk of delivering an SGA infant with neonatal morbidity of 17·6 (9·2–34·0, p<0·0001). Screening of nulliparous women with universal third trimester fetal biometry roughly tripled detection of SGA infants. Combined analysis of fetal biometry and fetal growth velocity identified a subset of SGA fetuses that were at increased risk of neonatal morbidity. National Institute for Health Research, Medical Research Council, Sands, and GE Healthcare.

There is an urgent need for mental health promotion in nonclinical settings. Mindfulness-based programmes (MBPs) are being widely implemented to reduce stress, but a comprehensive evidence synthesis is lacking. We reviewed trials to assess whether MBPs promote mental health relative to no intervention or comparator interventions. Following a detailed preregistered protocol (PROSPERO CRD42018105213) developed with public and professional stakeholders, 13 databases were searched to August 2020 for randomised controlled trials (RCTs) examining in-person, expert-defined MBPs in nonclinical settings. Two researchers independently selected, extracted, and appraised trials using the Cochrane Risk-of-Bias Tool 2.0. Primary outcomes were psychometrically validated anxiety, depression, psychological distress, and mental well-being questionnaires at 1 to 6 months after programme completion. Multiple testing was performed using p < 0.0125 (Bonferroni) for statistical significance. Secondary outcomes, meta-regression and sensitivity analyses were prespecified. Pairwise random-effects multivariate meta-analyses and prediction intervals (PIs) were calculated. A total of 11,605 participants in 136 trials were included (29 countries, 77% women, age range 18 to 73 years). Compared with no intervention, in most but not all scenarios MBPs improved average anxiety (8 trials; standardised mean difference (SMD) = -0.56; 95% confidence interval (CI) -0.80 to -0.33; p-value < 0.001; 95% PI -1.19 to 0.06), depression (14 trials; SMD = -0.53; 95% CI -0.72 to -0.34; p-value < 0.001; 95% PI -1.14 to 0.07), distress (27 trials; SMD = -0.45; 95% CI -0.58 to -0.31; p-value < 0.001; 95% PI -1.04 to 0.14), and well-being (9 trials; SMD = 0.33; 95% CI 0.11 to 0.54; p-value = 0.003; 95% PI -0.29 to 0.94). Compared with nonspecific active control conditions, in most but not all scenarios MBPs improved average depression (6 trials; SMD = -0.46; 95% CI -0.81 to -0.10; p-value = 0.012, 95% PI -1.57 to 0.66), with no statistically significant evidence for improving anxiety or distress and no reliable data on well-being. Compared with specific active control conditions, there is no statistically significant evidence of MBPs' superiority. Only effects on distress remained when higher-risk trials were excluded. USA-based trials reported smaller effects. MBPs targeted at higher-risk populations had larger effects than universal MBPs. The main limitation of this review is that confidence according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach is moderate to very low, mainly due to inconsistency and high risk of bias in many trials. Compared with taking no action, MBPs of the included studies promote mental health in nonclinical settings, but given the heterogeneity between studies, the findings do not support generalisation of MBP effects across every setting. MBPs may have specific effects on some common mental health symptoms. Other preventative interventions may be equally effective. Implementation of MBPs in nonclinical settings should be partnered with thorough research to confirm findings and learn which settings are most likely to benefit.

Missing covariate data commonly occur in epidemiological and clinical research, and are often dealt with using multiple imputation. Imputation of partially observed covariates is complicated if the substantive model is non-linear (e.g. Cox proportional hazards model), or contains non-linear (e.g. squared) or interaction terms, and standard software implementations of multiple imputation may impute covariates from models that are incompatible with such substantive models. We show how imputation by fully conditional specification, a popular approach for performing multiple imputation, can be modified so that covariates are imputed from models which are compatible with the substantive model. We investigate through simulation the performance of this proposal, and compare it with existing approaches. Simulation results suggest our proposal gives consistent estimates for a range of common substantive models, including models which contain non-linear covariate effects or interactions, provided data are missing at random and the assumed imputation models are correctly specified and mutually compatible. Stata software implementing the approach is freely available.

Background Multiple imputation is frequently used to deal with missing data in healthcare research. Although it is known that the outcome should be included in the imputation model when imputing missing covariate values, it is not known whether it should be imputed. Similarly no clear recommendations exist on: the utility of incorporating a secondary outcome, if available, in the imputation model; the level of protection offered when data are missing not-at-random; the implications of the dataset size and missingness levels. Methods We used realistic assumptions to generate thousands of datasets across a broad spectrum of contexts: three mechanisms of missingness (completely at random; at random; not at random); varying extents of missingness (20–80% missing data); and different sample sizes (1,000 or 10,000 cases). For each context we quantified the performance of a complete case analysis and seven multiple imputation methods which deleted cases with missing outcome before imputation, after imputation or not at all; included or did not include the outcome in the imputation models; and included or did not include a secondary outcome in the imputation models. Methods were compared on mean absolute error, bias, coverage and power over 1,000 datasets for each scenario. Results Overall, there was very little to separate multiple imputation methods which included the outcome in the imputation model. Even when missingness was quite extensive, all multiple imputation approaches performed well. Incorporating a secondary outcome, moderately correlated with the outcome of interest, made very little difference. The dataset size and the extent of missingness affected performance, as expected. Multiple imputation methods protected less well against missingness not at random, but did offer some protection. Conclusions As long as the outcome is included in the imputation model, there are very small performance differences between the possible multiple imputation approaches: no outcome imputation, imputation or imputation and deletion. All informative covariates, even with very high levels of missingness, should be included in the multiple imputation model. Multiple imputation offers some protection against a simple missing not at random mechanism.

Organisations such as the National Institute for Health and Care Excellence require the synthesis of evidence from existing studies to inform their decisions—for example, about the best available treatments with respect to multiple efficacy and safety outcomes. However, relevant studies may not provide direct evidence about all the treatments or outcomes of interest. Multivariate and network meta-analysis methods provide a framework to address this, using correlated or indirect evidence from such studies alongside any direct evidence. In this article, the authors describe the key concepts and assumptions of these methods, outline how correlated and indirect evidence arises, and illustrate the contribution of such evidence in real clinical examples involving multiple outcomes and multiple treatments

Background Risk prediction models are routinely used to assist in clinical decision making. A small sample size for model development can compromise model performance when the model is applied to new patients. For binary outcomes, the calibration slope (CS) and the mean absolute prediction error (MAPE) are two key measures on which sample size calculations for the development of risk models have been based. CS quantifies the degree of model overfitting while MAPE assesses the accuracy of individual predictions. Methods Recently, two formulae were proposed to calculate the sample size required, given anticipated features of the development data such as the outcome prevalence and c-statistic, to ensure that the expectation of the CS and MAPE (over repeated samples) in models fitted using MLE will meet prespecified target values. In this article, we use a simulation study to evaluate the performance of these formulae. Results We found that both formulae work reasonably well when the anticipated model strength is not too high (c-statistic < 0.8), regardless of the outcome prevalence. However, for higher model strengths the CS formula underestimates the sample size substantially. For example, for c-statistic = 0.85 and 0.9, the sample size needed to be increased by at least 50% and 100%, respectively, to meet the target expected CS. On the other hand, the MAPE formula tends to overestimate the sample size for high model strengths. These conclusions were more pronounced for higher prevalence than for lower prevalence. Similar results were drawn when the outcome was time to event with censoring. Given these findings, we propose a simulation-based approach, implemented in the new R package ‘samplesizedev’, to correctly estimate the sample size even for high model strengths. The software can also calculate the variability in CS and MAPE, thus allowing for assessment of model stability. Conclusions The calibration and MAPE formulae suggest sample sizes that are generally appropriate for use when the model strength is not too high. However, they tend to be biased for higher model strengths, which are not uncommon in clinical risk prediction studies. On those occasions, our proposed adjustments to the sample size calculations will be relevant.

Background It has long been advised to account for baseline covariates in the analysis of confirmatory randomised trials, with the main statistical justifications being that this increases power and, when a randomisation scheme balanced covariates, permits a valid estimate of experimental error. There are various methods available to account for covariates but it is not clear how to choose among them. Methods Taking the perspective of writing a statistical analysis plan, we consider how to choose between the three most promising broad approaches: direct adjustment, standardisation and inverse-probability-of-treatment weighting. Results The three approaches are similar in being asymptotically efficient, in losing efficiency with mis-specified covariate functions and in handling designed balance. If a marginal estimand is targeted (for example, a risk difference or survival difference), then direct adjustment should be avoided because it involves fitting non-standard models that are subject to convergence issues. Convergence is most likely with IPTW. Robust standard errors used by IPTW are anti-conservative at small sample sizes. All approaches can use similar methods to handle missing covariate data. With missing outcome data, each method has its own way to estimate a treatment effect in the all-randomised population. We illustrate some issues in a reanalysis of GetTested , a randomised trial designed to assess the effectiveness of an electonic sexually transmitted infection testing and results service. Conclusions No single approach is always best: the choice will depend on the trial context. We encourage trialists to consider all three methods more routinely.

Two approaches commonly used to deal with missing data are multiple imputation (MI) and inverse‐probability weighting (IPW). IPW is also used to adjust for unequal sampling fractions. MI is generally more efficient than IPW but more complex. Whereas IPW requires only a model for the probability that an individual has complete data (a univariate outcome), MI needs a model for the joint distribution of the missing data (a multivariate outcome) given the observed data. Inadequacies in either model may lead to important bias if large amounts of data are missing. A third approach combines MI and IPW to give a doubly robust estimator. A fourth approach (IPW/MI) combines MI and IPW but, unlike doubly robust methods, imputes only isolated missing values and uses weights to account for remaining larger blocks of unimputed missing data, such as would arise, e.g., in a cohort study subject to sample attrition, and/or unequal sampling fractions. In this article, we examine the performance, in terms of bias and efficiency, of IPW/MI relative to MI and IPW alone and investigate whether the Rubin’s rules variance estimator is valid for IPW/MI. We prove that the Rubin’s rules variance estimator is valid for IPW/MI for linear regression with an imputed outcome, we present simulations supporting the use of this variance estimator in more general settings, and we demonstrate that IPW/MI can have advantages over alternatives. IPW/MI is applied to data from the National Child Development Study.