Dataset Information

Pitfalls in the design and analysis of paediatric clinical trials: a case of a 'failed' multi-centre study, and potential solutions.

ABSTRACT: AIM: To increase awareness of possible pitfalls in the design and analysis of a multi-centre randomized clinical trial and to give an overview of alternative study designs and their consequences for power analyses in case of limited availability of trial participants. METHODS: Investigation of the assumptions in the power calculation and re-analysis of the original data of a 'failed' trial on the effect of dexamethasone on the duration of mechanical ventilation in young children with respiratory syncytial virus infection. Use of 'boundaries approach' is explored using the data from this trial. A comprehensive overview of the various modern solutions for the design of a subsequent trial in this field is given. RESULTS: Two frequent major deficiencies of trial design and data analysis are reviewed in depth, i.e. too optimistic assumptions for the sample size calculation and failure to adjust for centre effects. CONCLUSION: Critical review of trial assumptions and if necessary sample size recalculation based on an internal pilot by a data monitoring committee is recommended to maximize the probability of obtaining conclusive results.

SUBMITTER: van der Lee JH

PROVIDER: S-EPMC2659390 | biostudies-other | 2009 Feb

REPOSITORIES: biostudies-other

ACCESS DATA

Publications

Pitfalls in the design and analysis of paediatric clinical trials: a case of a 'failed' multi-centre study, and potential solutions.

van der Lee Johanna H JH Tanck Michael W T MW Wesseling Judit J Offringa Martin M

Acta paediatrica (Oslo, Norway : 1992) 20080924 2

<h4>Aim</h4>To increase awareness of possible pitfalls in the design and analysis of a multi-centre randomized clinical trial and to give an overview of alternative study designs and their consequences for power analyses in case of limited availability of trial participants.<h4>Methods</h4>Investigation of the assumptions in the power calculation and re-analysis of the original data of a 'failed' trial on the effect of dexamethasone on the duration of mechanical ventilation in young children wit ...[more]

PMID: 18826493

Similar Datasets

Project description:BackgroundMulti-centre randomized controlled clinical trials play an important role in modern evidence-based medicine. Advantages of collecting data from more than one site are numerous, including accelerated recruitment and increased generalisability of results. Mixed models can be applied to account for potential clustering in the data, in particular when many small centres contribute patients to the study. Previously proposed methods on sample size calculation for mixed models only considered balanced treatment allocations which is an unlikely outcome in practice if block randomisation with reasonable choices of block length is used.MethodsWe propose a sample size determination procedure for multi-centre trials comparing two treatment groups for a continuous outcome, modelling centre differences using random effects and allowing for arbitrary sample sizes. It is assumed that block randomisation with fixed block length is used at each study site for subject allocation. Simulations are used to assess operation characteristics such as power of the sample size approach. The proposed method is illustrated by an example in disease management systems.ResultsA sample size formula as well as a lower and upper boundary for the required overall sample size are given. We demonstrate the superiority of the new sample size formula over the conventional approach of ignoring the multi-centre structure and show the influence of parameters such as block length or centre heterogeneity. The application of the procedure on the example data shows that large blocks require larger sample sizes, if centre heterogeneity is present.ConclusionUnbalanced treatment allocation can result in substantial power loss when centre heterogeneity is present but not considered at the planning stage. When only few patients by centre will be recruited, one has to weigh the risk of imbalance between treatment groups due to large blocks and the risk of unblinding due to small blocks. The proposed approach should be considered when planning multi-centre trials.

Project description:BackgroundThe analysis of associations between accelerometer-derived physical activity (PA) intensities and cardiometabolic health is a major challenge due to multicollinearity between the explanatory variables. This challenge has facilitated the application of different analytic approaches within the field. The aim of the present study was to compare association patterns of PA intensities with cardiometabolic health in children obtained from multiple linear regression, compositional data analysis, and multivariate pattern analysis.MethodsA sample of 841 children (age 10.2 ± 0.3 years; BMI 18.0 ± 3.0; 50% boys) provided valid accelerometry and cardiometabolic health data. Accelerometry (ActiGraph GT3X+) data were characterized into traditional (four PA intensity variables) and more detailed categories (23 PA intensity variables covering the intensity spectrum; 0-99 to ≥10,000 counts per minute). Several indices of cardiometabolic health were used to create a composite cardiometabolic health score. Multiple linear regression and multivariate pattern analyses were used to analyze both raw and compositional data.ResultsBesides a consistent negative (favorable) association between vigorous PA and the cardiometabolic health measure using the traditional description of PA data, associations between PA intensities and cardiometabolic health differed substantially depending on the analytic approaches used. Multiple linear regression lead to instable and spurious associations, while compositional data analysis showed distorted association patterns. Multivariate pattern analysis appeared to handle the raw PA data correctly, leading to more plausible interpretations of the associations between PA intensities and cardiometabolic health.ConclusionsFuture studies should consider multivariate pattern analysis without any transformation of PA data when examining relationships between PA intensity patterns and health outcomes.Trial registrationThe study was registered in Clinicaltrials.gov 7th of April 2014 with identification number NCT02132494 .

Project description:BackgroundSnakebites are a public health problem in Nicaragua: it is a tropical developing country, venomous snakes are present and there are reports of snakebites treated both in the formal and informal health care system. We aimed to produce an incidence map using data reported by the health care system that would be used to allocate resources. However, this map may suffer from case detection bias and decisions based on this map will neglect snakebite victims who do not receive healthcare. To avoid this error, we try to identify where underreporting is likely based on available information.Method and findingsThe Nicaraguan municipalities are categorized by precipitation, altitude and geographical location into regions of assumed homogenous snake prevalence. Socio-economic and healthcare variables hypothesized to be related to underreporting of snakebites are aggregated into an index. The environmental region variable, the underreporting index and three demographic variables (rurality, sex and age distribution) are entered in a Poisson regression model of municipality-level snakebite incidence. In this model, the underreporting index is non-linearly associated with snakebite incidence, a finding we attribute to underreporting in the most deprived municipalities. The municipalities with the worst scoring on the underreporting index and those with combined low reported incidence and large rural population are identified as likely underreporting. 3,286 snakebite cases were reported in 2005-2009, corresponding to a 5-year incidence of 56 bites per 100,000 inhabitants (municipality range: 0-600 cases per 100,000 inhabitants).ConclusionsUsing publicly available data, we identified areas likely to be underreporting snakebites and highlighted these areas instead of leaving them "white" on the incidence map. The effects of the case detection bias on the distribution of resources against snakebites could decrease. Although not yet verified empirically, our study provides an example of how snake bite epidemiology may be investigated in similar settings worldwide at a low cost.

Dataset Information

Pitfalls in the design and analysis of paediatric clinical trials: a case of a 'failed' multi-centre study, and potential solutions.

Publications

Pitfalls in the design and analysis of paediatric clinical trials: a case of a 'failed' multi-centre study, and potential solutions.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets