Table 3 Needed a Clearer Denominator for ELM Score Interpretation
A clarification in Journal of Hepatology has tightened the interpretation of the ELM score performance tables, focusing on how percentages in Table 3 relate to the underlying patient counts. The correction centers on a deceptively simple but clinically important point: percentages only mean something when the denominator is explicit. In diagnostic work, especially when a score is designed to place patients into rule-out and rule-in zones, a mislabeled percentage can distort how readers judge the test’s practical utility at the bedside.
The note addresses concerns raised by Dr. Toshikuni regarding the reported percentages in Table 3 and their relationship to the number of patients correctly allocated to each zone. The original wording did not make the denominator sufficiently explicit, leaving room for confusion about what the percentages actually described. In a score-based diagnostic framework, that omission matters because performance tables are often used to translate statistical outputs into clinical triage decisions. A reader deciding whether a patient belongs in a rule-out zone or a rule-in zone needs to know whether the percentage reflects all participants, only those with analyzable data, or another subset entirely.
The clarification also acknowledges two numerical reporting errors in the ELM column, both of which were corrected in the revised proofs. No new study cohort, endpoint, or validation experiment is introduced in this note. The purpose is narrower and more exacting: to repair the table so that the published performance metrics align with the intended counts and can be read without ambiguity.
In diagnostic hepatology, that kind of correction is not cosmetic. Score tables are often the most immediate source clinicians consult when deciding how much confidence to place in a noninvasive tool. If the percentages are detached from their denominators, even briefly, the apparent strength of the score can be overstated or misunderstood. A rule-out zone is only useful if it can be trusted to identify patients who truly fall below a threshold of concern; a rule-in zone only has value if the positive classification reflects the intended clinical population.
One sentence carries the entire practical weight of the correction: percentages are not self-explanatory when the patient count is unclear.
The revised proofs now provide the intended numerical framing, and the clarification makes explicit that the percentages in Table 3 were meant to describe patients correctly allocated to the rule-out and rule-in zones. That distinction is central to any score intended for pathway management. In liver disease assessment, where diagnostic thresholds can shape referral, surveillance, biopsy decisions, or additional noninvasive testing, small differences in table presentation can ripple outward into how clinicians perceive the score’s reliability.
The note does not alter the broader clinical question behind the ELM score. It does, however, reinforce the need for precise reporting in performance tables, where one mislabeled denominator can create a false impression of accuracy or misclassification risk. Readers evaluating a diagnostic instrument depend on exact arithmetic as much as on statistical design. When the table language is tightened, the clinical message becomes easier to interpret: the score’s performance should be read through the lens of the correct patient counts, not through percentages detached from their source population.
A second sentence is equally direct: the corrected ELM column now reflects the revised proofs rather than the initially printed values.
For clinicians, the value of this clarification lies in protecting downstream interpretation. Diagnostic scores are only as trustworthy as the way their results are reported, and performance tables are often the final bridge between analytic work and clinical use. In a journal setting, especially in hepatology where risk stratification tools are scrutinized for their ability to separate low-risk from high-risk patients, even modest reporting imprecision can create avoidable uncertainty.
This correction therefore serves as a reminder that table construction is part of the evidence itself. When percentages correspond clearly to patient counts, the diagnostic meaning of the rule-out and rule-in zones can be judged on its own terms. When they do not, readers are left to infer a denominator that should have been stated outright. The revised wording closes that gap and restores the intended interpretation of the ELM score performance tables.