Jump to content

Double-blind study: Blinding to be safeguarded for monitor and data management?



  • Eve Kortleven
    03/04/2014 09:36
    GCP as well as ISO14155 state that blinding, if any, should be safeguarded/maintained during/throughout data entry and processing. In case of double-blind study with medical device, what would be the meaning of 'data entry and processing': only site staff assessing the patients, or also sponsor's monitoring and data management team monitoring and processing the data? According to some, in the latter case, you would have a triple-blind study. Is that assumption correct? In my experience with pharma trials, data management and monitor were also blinded to the treatment in case of double-blind trials, so I would expect the same for medical device trials...?
  • Medical Device Committee
    21/07/2014 08:03
    Data entry and processing could mean either at the site (eCRF) or in-house with the sponsor. It is sometimes harder to double blind a device study – generally not done if invasive however if truly double blind then there are multiple ways to manage this: 1. For an ophthalmic dissolvable steroid implant there was one investigator at each site (unblinded) who performed the procedure and initial follow up after a week or so then another investigator at the same site (blinded) who performed all subsequent follow up visits and assessments. Separation of site staff in this way can maintain the blind. 2. Use of two monitors for a single site, one blinded, the other not (to perform drug accountability where it revealed treatment assignment). Generally, once data has been collected, it is easy to maintain the blind. Again, in-house there is a separate person/team responsible for the randomization schedule – distinct from the team doing the data processing/analysis. Controlled access to the unblinding codes. Strict separation of staff and roles is acceptable. Double blind means the patient and the clinician (and possibly others including the monitor) are blinded to treatment. ISO 14155 states: 3.4 blinding/masking: procedure in which one or more parties to the clinical investigation are kept unaware of the treatment assignment(s) NOTE Single blinding usually refers to the subject(s) being unaware of the treatment assignment(s). Double blinding usually refers to the subject(s), investigator(s), monitor and, in some cases, centralized assessors being unaware of the treatment assignment(s). During the data processing stage i.e collecting of data from the visits (data collection), completing the CRFs (data entry) and data processing (reviewing entered data and raising queries and cleaning data) the site team and the monitor is not aware of the treatment allocation unless there has been an unblinding episode of a single patient in the case of a medical emergency for instance Even under these circumstances there may still be members of the team (data management) that may remain blind. This is still referred to as double blind. Control procedures for unblinding for final (or interim) analysis purposes should be in place. The case should also be the same for medical devices. Triple-blind studies do take place. In such cases the treatment or intervention is unknown to (a) the research participant, (b) the individual(s) who administer the treatment or intervention, and (c) the individual(s) who assess the outcomes. So you could have an "assessment committee" who is simply given data for groups A and B. A triple-blind study has the theoretical advantage of allowing the monitoring committee to evaluate the response variable results more objectively. This assumes that appraisal of efficacy and harm, as well as requests for special analyses, may be biased if the details about which group a patient is in is known. However, if the assessment committee also has a safety monitoring role, then such a design may be counterproductive as it may not be possible to proactively monitor trends.
Reply to Thread