What exactly are decentralised and pragmatic elements?
Annex 2 defines decentralised elements as any activities related to a clinical trial carried out outside the investigator’s location – at a participant’s home, at a local healthcare facility, using mobile medical units, or remotely, e.g. via video links or digital health technologies (Digital Health Technologies, DHT), such as mobile apps, wearable devices or sensors.
Pragmatic elements, in turn, are solutions that incorporate aspects of usual clinical practice into the trial design; these may include eligibility criteria reflecting the real-world patient population, as well as simplified data collection limited to information that is genuinely necessary, often fed by automatic export from electronic health records (EHRs).
The third pillar of the document is real-world data (RWD). RWD comprises information on patients’ health status sourced from outside the clinical trial – from EHRs, disease registries or claims databases – and can be used in an interventional trial either as primary data or as a secondary use of data originally collected for other purposes.
Regardless of the approach, the foundation remains the principle of a proportionate, risk-based approach and the concept of “quality by design” (QbD), already familiar from the General Principles and Annex 1: the methodology must be fit for purpose, and the quantity and quality of the data generated must be sufficient to enable appropriate decisions to be made.
What does this mean for the investigator and the ethics committee?
Annex 2 does not create a new, parallel system of ethical review – it extends the existing obligations of the researcher and the bioethics committee (IRB/IEC) to cover issues specific to the new methodologies.
The researcher must provide the IRB/IEC with information enabling it to assess the validity of the approaches used, whilst the process of obtaining informed consent itself becomes more flexible.
Consent may be obtained remotely, provided that the researcher verifies the participant’s identity – for example, by checking an identification document during a video call – and the verification method and data privacy safeguards are specified in advance.
Importantly, the document requires that an alternative be maintained: if the characteristics of the study population (e.g. lack of familiarity with computer systems) so require, the participant should be able to choose a traditional, paper-based form of consent or an in-person consent process.
Information materials must also clearly specify what participant data will be collected and who will have access to it – particularly when procedures take place in the participant’s home or via a DHT.
Management of the investigational product outside the research centre
One of the practical areas that Annex 2 regulates in detail is the logistics of the investigational product in a decentralised trial.
Dispatch to a participant – whether by the trial site or directly by the sponsor, where local regulations permit – requires documented procedures covering:
- protection of the participant’s privacy and medical status,
- confirmation of receipt by the intended recipient (the participant or their carer),
- processes for the receipt, storage, administration, return or disposal of the product,
- a mechanism to maintain blinding, if the trial is blinded,
- availability of technical support and instructional materials for the participant.
Regardless of who physically organises transport, the investigator remains responsible for the safe and appropriate use of the investigational product in participants under their care – which Annex 2 sets out as a principle taking precedence over logistical arrangements.
Real-world data: when is it fit for the purpose of the trial?
The sponsor is responsible for assessing whether RWD is fit for the purpose of the trial – which the document defines through two criteria: reliability (covering accuracy, completeness, origin and traceability of the data) and relevance, i.e. the availability of key data elements needed to answer a specific research question.
In practice, this means that several variables must be assessed simultaneously: variations in formats and terminology across sources; the lack of standardisation regarding the timing of data collection in relation to the protocol schedule; the comparability of the RWD-based control group with the study group; and the risk of missing data resulting, for example, from a patient switching healthcare systems.
The greater the criticality of RWD for the study outcome – for example, when it supports key efficacy or safety endpoints – the greater the sponsor’s access to source data should be, rather than merely to data aggregated at the individual level.
For endpoints of lower criticality, such as exploratory objectives, an assessment at the system and process level may be sufficient.
Where the entity controlling the RWD (e.g. a healthcare facility, insurer, registry administrator) restricts the sponsor’s access to source records, the document recommends consulting the relevant regulatory authority – which in itself indicates how seriously the ICH takes the issue of the verifiability of such data.
Safety and sponsor oversight in a decentralised model
As the number of data sources increases – remote visits, DHT, EHR data, face-to-face visits – the complexity of monitoring participants’ safety grows.
Annex 2 requires that information on adverse events reaches the investigator in a manner that enables a timely therapeutic decision to be made, regardless of the source from which it originates.
The sponsor must ensure that data from the DHT is provided to the investigator in a relevant, comprehensible and manageable format – so as to enable an effective review of the participant’s health status.
The sponsor’s oversight of service providers (e.g. home nursing or logistics companies) must be proportionate to the risk, but the investigator retains the final decision-making authority as to whether a particular service provider is suitable to carry out the activities falling within their scope of responsibility.
What should the sponsor do now?
Annex 2 is entering the practical implementation phase, and the Regulatory Affairs and Clinical Operations departments should already:
- review current CIP protocols in terms of the description of decentralised, pragmatic or RWD methodologies and the justification for their use,
- review informed consent procedures – in particular, the availability of a paper-based or face-to-face alternative for participants without digital experience,
- assess existing agreements with entities controlling RWD sources to ensure that the sponsor and regulatory authorities have access to source data,
- update the safety management plan to reflect the variety of channels used to collect data on adverse events.
Link to the document: Annex 2 ICH GCP