Eosinophil thresholds are being applied to a count that varies by age and sex

Design

cross-sectional study of a national health-check dataset with multivariable regression and percentile curve construction

Population

over 680,000 Chinese health-check participants, 2010–2023

Primary outcome

distribution of blood eosinophil counts by age and sex

Effect

median 100 cells/µL (IQR 60–170); 120 (80–200) in men vs 90 (50–140) in women; U-shaped across age

Blood eosinophil count now guides inhaled corticosteroid decisions in COPD and biologic eligibility in asthma, using fixed thresholds applied to everyone. This cross-sectional study asked what the count actually looks like in a general population, using more than 680,000 Chinese health-check participants between 2010 and 2023, and constructed age- and sex-specific percentile curves.

The distribution is right-skewed, with a median of 100 cells per microlitre (interquartile range 60 to 170). Men sat consistently higher than women — 120 (80 to 200) against 90 (50 to 140). Across age the trajectory was U-shaped, highest under 20 and at 70 or above. In multivariable analysis, higher counts were associated with male sex, those age extremes, smoking and higher body mass index, as well as with asthma, COPD and several systemic comorbidities. Findings held in sensitivity analyses.

The practical implication is about what a threshold means rather than which threshold to use. A count of 300 cells per microlitre sits at a different percentile for a 75-year-old man who smokes than for a 40-year-old woman who does not, and the guideline cut-offs were derived largely from trial populations that were neither. This is the largest such dataset from an Asian population, which makes it more relevant to Indian practice than the reference ranges most laboratories print — though a Chinese health-check population is not an Indian one, and no Indian equivalent exists.

• Record smoking status alongside the eosinophil count — it shifts the distribution
• Interpret a borderline count in context of age and sex rather than against a single cut-off
• Repeat a threshold-straddling count before it decides a biologic or an inhaled steroid
• Note the count is higher in men, so the same number means different things in a mixed clinic
• Do not assume your laboratory's reference range is derived from a comparable population

Why it matters

Fixed eosinophil cut-offs are being applied to a measurement whose normal range moves with age, sex and smoking.

Don't overread it

This describes a Chinese health-check population — it does not establish what threshold should trigger treatment in anyone.

The statistics, in plain English

A right-skewed distribution means the median of 100 sits well below the mean, and percentile curves are the right way to read it rather than a mean and standard deviation. The male-female gap of 120 against 90 is a third higher at the median, which is substantial relative to the treatment thresholds in use. The associations with asthma and COPD were significant after Bonferroni correction, a conservative adjustment that makes false positives unlikely but tells you nothing about the size of the difference.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.