Within-Trial Decisions

2. Code-breaking (unblinding)

In most pharmaceutical companies and contract research organisations (CRO) there is a procedure for unblinding an individual participant’s treatment allocation while the trial is still ongoing. This happens if there is a medical emergency or serious medical condition during a clinical study when clinicians (investigators) cannot manage treatment without knowledge of the participant’s study treatment.

To provide the documentation of the unblinding process is a regulatory requirement for study sponsors. This is usually done in a standard operating procedure (SOP) and there may be a guidance document or process flow to support the SOP.

In some organisations there is a call-centre or automated system to manage the process on behalf of the study sponsor. In many organisations the study staff who may receive a call from a healthcare provider, particularly out-of-hours, will be trained to access the system, identify the treatment allocation and complete the required documentation of the case.

There may be situations where special product-specific circumstances exist, such as a known risk for overdose or a potentially harmful interaction with other medicines. These cases require the availability of an on-call doctor. The process would then be modified to ensure that a procedure is in place to ensure that an on-call doctor discusses the request for unblinding, prior to its execution. This is to protect the patient from any serious impact of removing a treatment that requires dose titration, or from a potential treatment that may react badly with the study medicine that has been taken.

Trial sites are provided with specific training on the unblinding procedure and ‘patient cards’ (or equivalent) are provided to the patient to enable emergency unblinding by anyone who may happen to see the patient such as clinicians not involved with the trial in emergency rooms. The patient cards carry information about the clinical trial in which the patient is taking part, including clinical trial name and number, the principal investigator’s and institutions contact details, and the number to call for emergency unblinding. The patient is instructed to offer the card to any health worker other than those related to the trial in which they are participating.

Once the patient blind is broken, many patients will be required to discontinue the trial because of the possible bias. If there are too many cases where the treatment allocation has been disclosed during the study, the statistical integrity of the trial may be jeopardised. As additional safeguard, the trial-related clinical staff are often requested to discuss the need for knowing the patient’s treatment in case they have to treat an emergency with the medical monitor/on-call doctor, prior to making the call to the automated system or helpdesk to break the blind.