Search

Before embarking on an individual participant data meta-analysis (IPDMA) project, researchers and funders need assurance it is worth their time and cost. This should include consideration of how many studies are promising their IPD and, given the characteristics of these studies, the power of an IPDMA including them. Here, we show how to estimate the power of a planned IPDMA of randomized trials to examine treatment-covariate interactions at the participant level (ie, treatment effect modifiers). We focus on a binary outcome with binary or continuous covariates, and propose a three-step approach, which assumes the true interaction size is common to all trials. In step one, the user must specify a minimally important interaction size and, for each trial separately (eg, as obtained from trial publications), the following aggregate data: the number of participants and events in control and treatment groups, the mean and SD for each continuous covariate, and the proportion of participants in each category for each binary covariate. This allows the variance of the interaction estimate to be calculated for each trial, using an analytic solution for Fisher's information matrix from a logistic regression model. Step 2 calculates the variance of the summary interaction estimate from the planned IPDMA (equal to the inverse of the sum of the inverse trial variances from step 1), and step 3 calculates the corresponding power based on a two-sided Wald test. Stata and R code are provided, and two examples given for illustration. Extension to allow for between-study heterogeneity is also considered.

Precision medicine research often searches for treatment-covariate interactions, which refers to when a treatment effect (eg, measured as a mean difference, odds ratio, hazard ratio) changes across values of a participant-level covariate (eg, age, gender, biomarker). Single trials do not usually have sufficient power to detect genuine treatment-covariate interactions, which motivate the sharing of individual participant data (IPD) from multiple trials for meta-analysis. Here, we provide statistical recommendations for conducting and planning an IPD meta-analysis of randomized trials to examine treatment-covariate interactions. For conduct, two-stage and one-stage statistical models are described, and we recommend: (i) interactions should be estimated directly, and not by calculating differences in meta-analysis results for subgroups; (ii) interaction estimates should be based solely on within-study information; (iii) continuous covariates and outcomes should be analyzed on their continuous scale; (iv) nonlinear relationships should be examined for continuous covariates, using a multivariate meta-analysis of the trend (eg, using restricted cubic spline functions); and (v) translation of interactions into clinical practice is nontrivial, requiring individualized treatment effect prediction. For planning, we describe first why the decision to initiate an IPD meta-analysis project should not be based on between-study heterogeneity in the overall treatment effect; and second, how to calculate the power of a potential IPD meta-analysis project in advance of IPD collection, conditional on characteristics (eg, number of participants, standard deviation of covariates) of the trials (potentially) promising their IPD. Real IPD meta-analysis projects are used for illustration throughout. ispartof: STATISTICS IN MEDICINE vol:39 issue:15 pages:2115-2137 ispartof: location:England status: published

Previous articles in Statistics in Medicine describe how to calculate the sample size required for external validation of prediction models with continuous and binary outcomes. The minimum sample size criteria aim to ensure precise estimation of key measures of a model's predictive performance, including measures of calibration, discrimination, and net benefit. Here, we extend the sample size guidance to prediction models with a time-to-event (survival) outcome, to cover external validation in datasets containing censoring. A simulation-based framework is proposed, which calculates the sample size required to target a particular confidence interval width for the calibration slope measuring the agreement between predicted risks (from the model) and observed risks (derived using pseudo-observations to account for censoring) on the log cumulative hazard scale. Precise estimation of calibration curves, discrimination, and net-benefit can also be checked in this framework. The process requires assumptions about the validation population in terms of the (i) distribution of the model's linear predictor and (ii) event and censoring distributions. Existing information can inform this; in particular, the linear predictor distribution can be approximated using the C-index or Royston's D statistic from the model development article, together with the overall event risk. We demonstrate how the approach can be used to calculate the sample size required to validate a prediction model for recurrent venous thromboembolism. Ideally the sample size should ensure precise calibration across the entire range of predicted risks, but must at least ensure adequate precision in regions important for clinical decision-making. Stata and R code are provided.

In prediction model research, external validation is needed to examine an existing model's performance using data independent to that for model development. Current external validation studies often suffer from small sample sizes and consequently imprecise predictive performance estimates. To address this, we propose how to determine the minimum sample size needed for a new external validation study of a prediction model for a binary outcome. Our calculations aim to precisely estimate calibration (Observed/Expected and calibration slope), discrimination (C-statistic), and clinical utility (net benefit). For each measure, we propose closed-form and iterative solutions for calculating the minimum sample size required. These require specifying: (i) target SEs (confidence interval widths) for each estimate of interest, (ii) the anticipated outcome event proportion in the validation population, (iii) the prediction model's anticipated (mis)calibration and variance of linear predictor values in the validation population, and (iv) potential risk thresholds for clinical decision-making. The calculations can also be used to inform whether the sample size of an existing (already collected) dataset is adequate for external validation. We illustrate our proposal for external validation of a prediction model for mechanical heart valve failure with an expected outcome event proportion of 0.018. Calculations suggest at least 9835 participants (177 events) are required to precisely estimate the calibration and discrimination measures, with this number driven by the calibration slope criterion, which we anticipate will often be the case. Also, 6443 participants (116 events) are required to precisely estimate net benefit at a risk threshold of 8%. Software code is provided.

One-stage individual participant data meta-analysis models should account for within-trial clustering, but it is currently debated how to do this. For continuous outcomes modeled using a linear regression framework, two competing approaches are a stratified intercept or a random intercept. The stratified approach involves estimating a separate intercept term for each trial, whereas the random intercept approach assumes that trial intercepts are drawn from a normal distribution. Here, through an extensive simulation study for continuous outcomes, we evaluate the impact of using the stratified and random intercept approaches on statistical properties of the summary treatment effect estimate. Further aims are to compare (i) competing estimation options for the one-stage models, including maximum likelihood and restricted maximum likelihood, and (ii) competing options for deriving confidence intervals (CI) for the summary treatment effect, including the standard normal-based 95% CI, and more conservative approaches of Kenward-Roger and Satterthwaite, which inflate CIs to account for uncertainty in variance estimates. The findings reveal that, for an individual participant data meta-analysis of randomized trials with a 1:1 treatment:control allocation ratio and heterogeneity in the treatment effect, (i) bias and coverage of the summary treatment effect estimate are very similar when using stratified or random intercept models with restricted maximum likelihood, and thus either approach could be taken in practice, (ii) CIs are generally best derived using either a Kenward-Roger or Satterthwaite correction, although occasionally overly conservative, and (iii) if maximum likelihood is required, a random intercept performs better than a stratified intercept model. An illustrative example is provided.

Many meta-analysis models contain multiple parameters, for example due to multiple outcomes, multiple treatments or multiple regression coefficients. In particular, meta-regression models may contain multiple study-level covariates, and one-stage individual participant data meta-analysis models may contain multiple patient-level covariates and interactions. Here, we propose how to derive percentage study weights for such situations, in order to reveal the (otherwise hidden) contribution of each study toward the parameter estimates of interest. We assume that studies are independent, and utilise a decomposition of Fisher’s information matrix to decompose the total variance matrix of parameter estimates into study-specific contributions, from which percentage weights are derived. This approach generalises how percentage weights are calculated in a traditional, single parameter meta-analysis model. Application is made to one- and two-stage individual participant data meta-analyses, meta-regression and network (multivariate) meta-analysis of multiple treatments. These reveal percentage study weights toward clinically important estimates, such as summary treatment effects and treatment-covariate interactions, and are especially useful when some studies are potential outliers or at high risk of bias. We also derive percentage study weights toward methodologically interesting measures, such as the magnitude of ecological bias (difference between within-study and across-study associations) and the amount of inconsistency (difference between direct and indirect evidence in a network meta-analysis).

A one-stage individual participant data (IPD) meta-analysis synthesises IPD from multiple studies using a general or generalised linear mixed model. This produces summary results (e.g. about treatment effect) in a single step, whilst accounting for clustering of participants within studies (via a stratified study intercept, or random study intercepts) and between-study heterogeneity (via random treatment effects). We use simulation to evaluate the performance of restricted maximum likelihood (REML) and maximum likelihood (ML) estimation of one-stage IPD meta-analysis models for synthesising randomised trials with continuous or binary outcomes. Three key findings are identified. Firstly, for ML or REML estimation of stratified intercept or random intercepts models, a t-distribution based approach generally improves coverage of confidence intervals for the summary treatment effect, compared to a z-based approach. Secondly, when using ML estimation of a one-stage model with a stratified intercept, the treatment variable should be coded using ‘study-specific centering’ (i.e. 1/0 minus the study-specific proportion of participants in the treatment group), as this reduces the bias in the between-study variance estimate (compared to 1/0 and other coding options). Thirdly, REML estimation reduces downward bias in between-study variance estimates compared to ML estimation, and does not depend on the treatment variable coding; for binary outcomes, this requires REML estimation of the pseudo-likelihood, although this may not be stable in some situations (e.g. when data are sparse). Two applied examples are used to illustrate the findings.

Introduction For tests reporting continuous results, primary studies usually provide test performance at multiple but often different thresholds. This creates missing data when performing a meta-analysis at each threshold. A standard meta-analysis (no imputation [NI]) ignores such missing data. A single imputation (SI) approach was recently proposed to recover missing threshold results. Here, we propose a new method that performs multiple imputation of the missing threshold results using discrete combinations (MIDC). Methods The new MIDC method imputes missing threshold results by randomly selecting from the set of all possible discrete combinations which lie between the results for 2 known bounding thresholds. Imputed and observed results are then synthesised at each threshold. This is repeated multiple times, and the multiple pooled results at each threshold are combined using Rubin's rules to give final estimates. We compared the NI, SI, and MIDC approaches via simulation. Results Both imputation methods outperform the NI method in simulations. There was generally little difference in the SI and MIDC methods, but the latter was noticeably better in terms of estimating the between-study variances and generally gave better coverage, due to slightly larger standard errors of pooled estimates. Given selective reporting of thresholds, the imputation methods also reduced bias in the summary receiver operating characteristic curve. Simulations demonstrate the imputation methods rely on an equal threshold spacing assumption. A real example is presented. Conclusions The SI and, in particular, MIDC methods can be used to examine the impact of missing threshold results in meta-analysis of test accuracy studies.

OBJECTIVES: Our aim was to improve meta-analysis methods for summarizing a prediction model's performance when individual participant data are available from multiple studies for external validation. STUDY DESIGN AND SETTING: We suggest multivariate meta-analysis for jointly synthesizing calibration and discrimination performance, while accounting for their correlation. The approach estimates a prediction model's average performance, the heterogeneity in performance across populations, and the probability of "good" performance in new populations. This allows different implementation strategies (e.g., recalibration) to be compared. Application is made to a diagnostic model for deep vein thrombosis (DVT) and a prognostic model for breast cancer mortality. RESULTS: In both examples, multivariate meta-analysis reveals that calibration performance is excellent on average but highly heterogeneous across populations unless the model's intercept (baseline hazard) is recalibrated. For the cancer model, the probability of "good" performance (defined by C statistic ≥0.7 and calibration slope between 0.9 and 1.1) in a new population was 0.67 with recalibration but 0.22 without recalibration. For the DVT model, even with recalibration, there was only a 0.03 probability of "good" performance. CONCLUSION: Multivariate meta-analysis can be used to externally validate a prediction model's calibration and discrimination performance across multiple populations and to evaluate different implementation strategies.

Background Researchers and funders should consider the statistical power of planned Individual Participant Data (IPD) meta-analysis projects, as they are often time-consuming and costly. We propose simulation-based power calculations utilising a two-stage framework, and illustrate the approach for a planned IPD meta-analysis of randomised trials with continuous outcomes where the aim is to identify treatment-covariate interactions. Methods The simulation approach has four steps: (i) specify an underlying (data generating) statistical model for trials in the IPD meta-analysis; (ii) use readily available information (e.g. from publications) and prior knowledge (e.g. number of studies promising IPD) to specify model parameter values (e.g. control group mean, intervention effect, treatment-covariate interaction); (iii) simulate an IPD meta-analysis dataset of a particular size from the model, and apply a two-stage IPD meta-analysis to obtain the summary estimate of interest (e.g. interaction effect) and its associated p-value; (iv) repeat the previous step (e.g. thousands of times), then estimate the power to detect a genuine effect by the proportion of summary estimates with a significant p-value. Results In a planned IPD meta-analysis of lifestyle interventions to reduce weight gain in pregnancy, 14 trials (1183 patients) promised their IPD to examine a treatment-BMI interaction (i.e. whether baseline BMI modifies intervention effect on weight gain). Using our simulation-based approach, a two-stage IPD meta-analysis has

If individual participant data are available from multiple studies or clusters, then a prediction model can be externally validated multiple times. This allows the model’s discrimination and calibration performance to be examined across different settings. Random-effects meta-analysis can then be used to quantify overall (average) performance and heterogeneity in performance. This typically assumes a normal distribution of ‘true’ performance across studies. We conducted a simulation study to examine this normality assumption for various performance measures relating to a logistic regression prediction model. We simulated data across multiple studies with varying degrees of variability in baseline risk or predictor effects and then evaluated the shape of the between-study distribution in the C-statistic, calibration slope, calibration-in-the-large, and E/O statistic, and possible transformations thereof. We found that a normal between-study distribution was usually reasonable for the calibration slope and calibration-in-the-large; however, the distributions of the C-statistic and E/O were often skewed across studies, particularly in settings with large variability in the predictor effects. Normality was vastly improved when using the logit transformation for the C-statistic and the log transformation for E/O, and therefore we recommend these scales to be used for meta-analysis. An illustrated example is given using a random-effects meta-analysis of the performance of QRISK2 across 25 general practices.

Access to big datasets from e-health records and individual participant data (IPD) meta-analysis is signalling a new advent of external validation studies for clinical prediction models. In this article, the authors illustrate novel opportunities for external validation in big, combined datasets, while drawing attention to methodological challenges and reporting issues.