Methodologies and statistics

  • Email
  • Help

The use of appropriate statistical methods is important for planning, carrying out and evaluating the results of studies used in the development of medicines. These methods have an impact on the life-cycle of medicines at all stages from early development to after approval.

At the European Medicines Agency, discussions on scientific advice, paediatric investigation plans and marketing-authorisation applications often concern the choice of the most appropriate trial designs and statistical tools.

The Agency engages in dialogue with other regulators, the pharmaceutical industry, patient organisations and academia, to help foster a better understanding of the benefits and limitations of statistical methods in drug development.

New and innovative statistical approaches and methodologies can improve how clinical trials are designed, conducted, evaluated and interpreted. These include subgroup analysis, adaptive or flexible designs and modelling and simulation.

The Agency organises meetings, including conferences and training sessions, and publishes scientific guidelines to help industry, academia and regulatory scientists apply robust methodology.

Subgroup analysis

Subgroup analysis is the analysis of clinical trial results separately in subsets of the participants in a clinical trial. It is an important tool in the assessment of confirmatory trials in marketing authorisation applications.

Subgroup analyses are used to:

  • assess the internal consistency of a trial's results;
  • try to identify patient groups with the most favourable benefit-risk profile;
  • try to rescue trials that ‘fail’ based on the full analysis.

Subgroups may be pre-specified in the trial protocol, based on demographic, genomic or disease characteristics. Alternatively they may materialise during a trial, based on the researcher's need or desire to explore study results further.

Interpreting subgroup analyses is challenging. Therefore, the Agency's Committee for Medicinal Products for Human Use (CHMP) is preparing a guideline, setting out important methodological considerations and assessment strategies:

The Agency has hosted a meeting on subgroup analysis:

Adaptive (flexible) designs

The Agency has hosted a number of meetings on adaptive or flexible designs:

Modelling and simulation

Modelling and simulation are techniques that use mathematical models to understand and predict the outcomes of interventions. They are used widely in engineering and science, including the development of medicines, and can be used to improve the efficiency of the design and development of products.

Modelling and simulation methodology provides an opportunity to improve the efficiency of medicine development, as well as to facilitate the regulatory assessment of medicines. It can be applied in a range of areas, including early medicine development, dose finding, clinical pharmacology, studying special populations, and the optimisation and analysis of pivotal clinical trials.

The Agency has hosted a number of meetings on modelling and simulation:

How helpful is this page?

Average rating:

 Based on 31 ratings

Add your rating:

See all ratings
19 ratings
1 ratings
0 ratings
0 ratings
11 ratings

Tell us more

Related information