The foreword states that communicating information "accurately, without causing misunderstanding" is increasingly important. The word order matters. "Without causing misunderstanding" comes before "accurately." That sequence is deliberate. Not deceiving is a necessary condition — but the foreword signals from the outset that it is not sufficient.
The imperative not to mislead connects directly to the first of the three structural pillars: blocking "factually accurate yet misleading" expression. This design principle drives the majority of the specific rules in later chapters — behind every requirement about axis scales, emphasis, and text placement, this principle is at work.
01Why "not deceiving" is not enough
Prohibiting falsehood is where all regulation begins. But it does not, by itself, achieve accurate communication. Using true facts as raw material while controlling how they are arranged, cropped, or weighted can produce in the reader an understanding that diverges sharply from the overall picture.
Consider a concrete case. A primary endpoint failed to reach statistical significance, but an exploratory subgroup analysis produced a significant result, and that result is placed prominently. Every number is accurate. But the reader concludes that solid efficacy evidence exists. No deception has occurred. Misunderstanding has.
The same applies to presenting only the relative risk reduction in large type while relegating the absolute risk reduction to a footnote. Both figures accurately reflect the trial. The presentation nonetheless engineers the impression that "this drug works." Accurate facts can be used to induce incorrect judgment — and that is why "not causing misunderstanding" is a demand one tier higher than "not deceiving."
02What it means to cause misunderstanding
Misunderstanding is hard to define because it depends on the reader's interpretation. In the context of the Creation Guide, however, it can be understood as follows: a state in which a reasonable healthcare professional, having reviewed the material, arrives at an understanding inconsistent with the actual weight and limitations of the underlying evidence.
Defined this way, the problem reaches beyond "what is written" to "what is absent." If unfavorable safety data are omitted, the reader may interpret their absence as "no particular concerns." If confidence intervals are not shown, the width of uncertainty around a point estimate goes unnoticed. Omission and asymmetric placement distort the reader's understanding just as surely as inaccurate text.
Three routes through which misunderstanding arises
Misunderstanding enters through three main channels. Selection — presenting only favorable results. Layout — giving favorable information large treatment and unfavorable information small treatment. Context stripping — excising a portion from the whole. None of these is fabrication. But each moves the reader's understanding away from reality in the same way that a lie would.
03Why this is a higher standard than accuracy alone
The obligation to "communicate accurately" imposes a higher standard than the obligation to "deliver correct information." Delivering correct information while the reader forms an incorrect understanding still falls short of compliance — that is the foreword's position.
The practical context makes this demand rational. Clinicians are busy. There is rarely time to read a material from cover to cover. Impressions formed at a glance, numbers that catch the eye, the summary block — these are the inputs from which judgments are quickly assembled. Material authors cannot assume careful reading. They must design for what a distracted, skimming reader takes away.
"There are no falsehoods" is not a defense. What is required is whether a correct understanding forms in the reader's mind. Regardless of intent, expression that leads to misunderstanding can constitute a violation.
In closing
The imperative not to mislead is not an extension of "not deceiving" — it stands in the foreword as an independent obligation. Once the reality is acknowledged that true facts can be used to manipulate impressions, regulation must extend beyond "what may be written" to "what the reader will actually understand." Most of the rules governing format, layout, and expression in later chapters are translations of this imperative into specific, enforceable requirements.
Case studies — the "it isn't written, so…" mindset, and how each lands
"We didn't lie" is not a defense against the foreword's standard. What is asked is whether the reader ends up with an accurate picture. Knowing that misimpression travels through selection, placement, and omission of context — and then deliberately working those channels — leads somewhere. The four levels below show where.
GRAY 1 — Straying from the spirit (untouched by the wording, but against the foreword's spirit)
The scheme: "The primary endpoint didn't reach significance, but we present the full trial data accurately. Placing a favorable subgroup analysis — which was pre-specified in the analysis plan — at the front of the figure panel contains nothing false."
Verdict: Leading with the subgroup result without first stating the primary endpoint's outcome structurally produces the impression that there is an established efficacy base, even though the numbers are accurate. The placement itself generates the misreading. The correct approach is to present the primary endpoint result and its statistical status first, with the subgroup analysis explicitly labelled as supplementary.
GRAY 2 — Exploiting the gap (deliberately working a loophole to invite misreading)
The scheme: "Both the relative risk reduction (RRR) and the absolute risk reduction (ARR) are in the material. The RRR is displayed large at the centre of the figure, the ARR in small text alongside. Both numbers are there — the obligation is met."
Verdict: The presence of information and its actual transmission are separate questions. The physical size asymmetry produces the impression that the product is highly effective, while the ARR is effectively invisible. "Not misleading" requires not just placing information but placing it so it registers. RRR and ARR should appear at near-equal prominence, accompanied by the baseline risk of the study population.
GRAY 3 — Violating higher norms (arguably "not in the text," yet almost certainly out under the PMD Act, Fair Advertising Standards, and the JPMA Code)
The scheme: "Nowhere in the Guidelines does it say confidence intervals are mandatory. We report the point estimates accurately. Omitting confidence intervals is not a violation of any written rule."
Verdict: The Sales Information Provision Guidelines and the Fair Advertising Standards broadly prohibit information provision that distorts the judgments of healthcare professionals; omitting confidence intervals to conceal the uncertainty around a point estimate falls within that prohibition. "Not explicitly required by the Guidelines" is not a license. Confidence intervals must accompany all primary efficacy metrics.
BLACK — Breaching the explicit text (a plain violation of the provision itself)
The scheme: "We're just quoting from the clinical trial used for regulatory approval. The adverse-event comparison with the comparator shown in the figure is documented in the trial report — it's a fact, not an unpublished claim absent from the electronic PI."
Verdict: Displaying in a material an adverse-event comparison with a comparator drug that does not appear in the electronic PI constitutes a safety claim outside the approved scope, which conflicts with PMD Act Article 66's prohibition on exaggerated advertising. The Guidelines require that safety content remain within the bounds of the electronic PI — departing from that is a plain textual violation. Safety comparisons must be confined to what the electronic PI states.
Improper intent by implication — three scopes of "insinuating" without stating it
The Preamble places "do not mislead" above "do not lie" as an independent obligation. All three examples below involve no numerical falsehood; the deception operates through formatting priority, the ordering of evidence, and the narrative arc constructed by the document's chapter structure.
Implication 1 — Local: carried in a single word or sentence
How it is built: "Present the relative risk reduction (RRR) without parentheses and tuck the absolute risk reduction (ARR) inside parentheses immediately after: e.g., 'risk decreased by 40% (absolute difference 1.2 percentage points).'"
Intent that can be read: The typographic difference of a single pair of parentheses reorders the informational hierarchy. Readers process the large unparenthesised figure (RRR) as the primary measure of effect; the small ARR retreats to background status without being suppressed.
Verdict: Deliberately depressing the visual salience of ARR through parenthetical placement distorts the reader's judgment. The "do not mislead" obligation covers the legibility design of information, not just its presence. ARR and RRR must be presented with equivalent typographic weight.
Implication 2 — Adjacent: read across neighboring clauses
How it is built: "State 'a statistically significant difference was observed on the primary endpoint' immediately followed by 'a consistent trend was confirmed across subgroup analyses,' with no indication that the second finding is exploratory."
Intent that can be read: Adjacency prompts readers to read the two sentences as cumulative confirmation: primary endpoint achieved and further corroborated by subgroups. The exploratory nature of the subgroup analysis is never stated, so the weaker evidence type is absorbed into the stronger.
Verdict: Placing a confirmatory result and an exploratory result side by side, without differentiating their evidentiary status, is a proximity implication that inflates the persuasive weight of the exploratory finding. Exploratory analyses must be labelled as such and kept visually separate from the primary endpoint conclusion.
Implication 3 — Whole-context: arising from the document as a whole
How it is built: "Structure the document as: (1) disease burden (emotional tone), (2) unmet need, (3) efficacy of the drug (bulk of pages), (4) safety (table only, last page). Open Chapter 3 with a clinical photograph and patient-centred language before presenting data."
Intent that can be read: The emotional arc built in Chapters 1 and 2 amplifies the efficacy data in Chapter 3, then treats safety as a dry bureaucratic afterthought. The document as a whole functions as a persuasive narrative rather than a balanced information transfer.
Verdict: A chapter structure designed to emotionally prime readers before presenting efficacy data, and to minimise engagement with safety data, violates the "do not mislead" duty at the level of document architecture. Each sentence may be factual; the whole creates a false impression. Safety must receive equivalent space and comparable rhetorical weight to efficacy throughout.