The Guide divides clinical trial findings into two tiers: those that constitute the primary evidence for an approved indication, and those that represent incidental, accompanying observations. The second tier carries the label reference information, and the Guide requires that it be recorded with its status explicitly distinguished. The classification is not a housekeeping matter. Where something is placed determines the ceiling on how forcefully it may be communicated — that is the core of the discipline the Guide calls isolation of reference information.
The three pillars of the foreword — alignment with the package insert, accurate communication without misleading, and structural verifiability — extend to the question of which findings are brought forward and which are held back. Isolating reference information is a visible, on-the-page safeguard against secondary findings appearing to be of the same standing as the principal approved indication.
01Secondary results are recorded separately as reference information
When a treatment conducted within the scope of an approved indication produces results that are incidental to the primary objective, those results must be clearly distinguished from the primary evidence of efficacy and recorded as reference information. And regardless of that framing, expression that could cause the reader to misunderstand the approved indication must be avoided.
Secondary findings were not pre-specified as primary endpoints in the trial design. They were observed exploratorily, not confirmed through a hypothesis-testing analysis. Even numbers drawn from the same clinical trial carry fundamentally different evidential weight depending on whether they come from the primary, confirmatory analysis or from a secondary observation. The approved indication rests on the confirmatory analysis of the primary endpoint. Secondary results can only be used in the context of supplementing that approval — no more.
Why the distinction in writing is necessary
Without an explicit distinction, the reader receives primary evidence and secondary findings as equivalent. When a physician makes a prescribing decision, knowing which figure was confirmatory and which was an exploratory observation is directly relevant to interpreting the evidence correctly. Failing to mark the distinction on the page allows the material as a whole to leave an inflated impression of the approved indication. The explicit "reference information" label is the brake on that impression.
02Daily activity and QoL are treated as reference information as a rule
Assessments of activities of daily living (ADL) and quality of life (QoL) are directly relevant to patients' experience of illness, and carry real weight for clinicians. Yet the Guide designates them reference information as a rule.
The reason lies in the nature of the measurement. ADL and QoL assessments reflect the patient's subjective experience and living environment; compared with direct pathophysiological indicators of disease, they carry a wider range of variability and are more susceptible to measurement bias. The principal basis of efficacy in the approval review is generally the core disease symptoms or objective clinical endpoints. Improvement in ADL or QoL may be expected as a downstream consequence, but it is distinct from the approved indication itself.
Detailed rule: when the assessment scale is standardized
The detailed rules carve out one qualification. When the definition of the assessment scale or score is clear and widely established, there are cases in which it does not fall under reference information. Standardized scales — internationally validated functional assessment instruments, for example — have definitions and measurement methods shared across the clinical community, leaving little room for arbitrary interpretation. In such cases the reference-information classification may not apply.
That qualification does not, however, grant an unconditional licence to make stronger claims. Even with a standardized scale, the requirement to avoid expression that could mislead about the approved indication remains in force. Only the evidential standing of the scale changes; the duty to prevent misunderstanding persists.
03Pharmacological actions with unclear relevance to the indication are treated the same way
A pharmacological action whose relevance to the approved indication is not sufficiently established is likewise treated as reference information, in the same way as secondary clinical results.
Descriptions of pharmacological action sit within the material as explanations of the mechanism that underlies the approved indication. Not every pharmacological change observed in a study connects directly to that approved indication. Placing findings of unclear relevance alongside the primary pharmacological data creates the risk of implying that the drug has benefits beyond what was approved. The Guide assigns such findings to the reference-information category and requires that they be presented as knowledge whose relationship to the approved indication has not been established.
The "reference information" frame is not a device for erasing facts. It is a device for accurately conveying where a finding stands in the hierarchy of evidence. Recording is permitted — but the tier of evidence must be made explicit before it is recorded. This is what the foreword's principle of accurate communication without misleading looks like when applied to the architecture of information.
04Reference information must not be recorded in the features column
A finding classified as reference information must not be placed in the material's features/characteristics column.
The features column is where the principal points of a drug's efficacy and safety are described. Information placed there is naturally received by the reader as representing the drug's central value. Writing reference information — secondary findings, pharmacological actions of unclear relevance — in that column would misrepresent the evidential tier. The on-page separation is the structural mechanism for preventing that misreading.
The prohibition on placing content in the features column and the requirement to record it explicitly as reference information are two sides of the same rule. The first defines where something must not go; the second defines where and how it may go. Together they inscribe the evidential standing of secondary findings into the layout of the page.
Isolating reference information is a way of expressing in the structure of a material the fact that "this finding exists, but its evidential weight differs from that of the primary indication." Secondary clinical results, ADL and QoL findings (except standardized scales meeting the criteria), and pharmacological actions of unclear relevance to the indication — all are framed as reference information and kept out of the features column. This tier-marking runs through the three pillars of the foreword. The package insert is the authoritative original; the summary merely supplements it. The supplement must not create an impression that exceeds the original. Making the strength of information visible on the page is the final safeguard of accurate, unmisleading communication.