How should intercurrent events be dealt with in clinical trials or clinical investigations?

In order for the results of clinical trials to be compared with existing treatments, they must be prepared and reported transparently and comprehensibly for doctors, authorities and associations with regard to the investigated treatment effect. Since many studies have deficiencies in this regard, the so-called estimate concept was introduced.

A Estimand is a precise description of the treatment effect that reflects the clinical question of the study.

that Estimands concept offers a structured method to determine which treatment effect should be investigated in a study. An estimand consists of five characteristics: the treatment itself, the group of study participants, the size to be measured, the summary of results at group level, and how to deal with events that could occur during the study and influence the results. These characteristics are defined in interdisciplinary exchange during study planning based on the clinical question.

The term intercurrent events (English intercurrent events, ICEs) defines events that occur after the start of study-related treatment and influence either the interpretation or existence of measurements associated with the clinical question. This includes, for example, starting a new therapy, changing or discontinuing treatment, or taking emergency medication. In its scientific guideline documents on statistical principles for clinical trials of medicinal products (CPMP/ICH/363/96 and EMA/CHMP/ICH/436221/2017), the EMA suggests 5 different approaches to deal with these events.

This procedure can also be applied mutatis mutandis to clinical trials involving medical devices. The strategy preferred and expected by the EMA for handling intercurrent events is the strategy of Treatment guidelines (English treatment policies).

The addendum suggests a further three strategies to deal with these events: “hypothetical,” “composite,” and “during treatment.” It is crucial to understand the differences between the strategies and to formulate exactly which are used when creating the estimate.

For intercurrent events, the Treatment guideline defined as a preferred specific strategy. The treatment guideline requires that the data of all study participants be included in the main analysis, regardless of whether intercurrent events have occurred or not. This approach aims to evaluate the effectiveness of the studied intervention under real conditions.

The procedure is then as follows:

1. Data collection: All data will continue to be collected regardless of the occurrence of ICEs.
2. Data analysis: The data is included in the main analysis as if the ICE had not occurred.
3. Interpretation: The effectiveness of the intervention is assessed under real conditions, including potential ICEs.

This structured approach is intended to ensure that the clinical question is clearly defined, that the appropriate study methods can be used to answer this question and that it is therefore possible to interpret the study results and assess their reliability. Even though the estimand concept is requested for clinical trials for medicinal products by EMA, this approch is now also required for clinical investigations for medical devices.

This knowledge is relevant for manufacturers of digital health applications (DiGA) and for statisticians who support DIGA manufacturers in preparing the statistical analysis plan or advise them, as the German Federal Institute for Drugs and Medical Devices (Bundesinstitut für Arzneimittel und Medizinprodukte, BfArM) expects this approach.

(as of July 2022)

May we also support you? Feel free to arrange a first non-binding appointment here.