Table 3 Categories of 4074 reported protocol deviations
Deviation categories | N (%a) | Requirement not met (may overlap and be ≤ or ≥ N) |
|---|---|---|
Omitted data or procedures | 2815 (69.10%) | Omissions were reported on laboratory tests (n = 1902), physical examinations (n = 1604), eligibility (n = 338), adverse event reporting (n = 119), questionnaires (n = 226), counselling (n = 63), full visits (n = 57), non-study product treatment provision (n = 36), study product (n = 13), and regulatory approvals (n = 9). |
Error in data or procedures | 545 (13.38%) | There were errors on physical examination (n = 263), laboratory tests (n = 103), adverse events and reactogenicity (n = 52), study product (n = 51), participant unique identifier assignment, eligibility, enrolment and randomisation (n = 29), delays with implementing updated study materials (n = 15), the full visit (n = 8), reimbursement (n = 4), non-study treatment (n = 3), questionnaires (n = 3), and counselling (n = 1). The major errors with physical examinations were related to thermometry methodology inconsistent with protocol. Regarding laboratory tests, a prominent issue was not following requirements to take some samples before certain events. |
Use of materials not approved by ethics committee | 406 (9.97%) | Study materials distributed to participants before ethics committee approval were study brochures (n = 403) and participant diary cards (n = 3). |
Consent errors | 172 (4.22%) | Consent errors were implementing the incorrect consent form version (n = 112), documentation issues (n = 39), delays in implementing an approved consent form (n = 18) and 2 could not be coded. Documentation issues included form completion that did not meet good documentation practice standards. It also included consent process documentation that was insufficiently detailed, e.g. the distribution of copies, or whether a participant understood explanations of their incorrect responses on their assessments of understanding consent. |
Needless data collection or procedures | 130 (3.19%) | Unnecessary data collection or procedures were laboratory samples (n = 113), full visits (n = 7), questionnaires (n = 4), study product (n = 1), treatment (n = 1), and unblinding (n = 1). |
Not categorised | 6 (0.15%) | Insufficient information for categorization (n = 6). |