Objective:

To create sex-specific reference curves (RCs) and optimal curves (OCs) for hemoglobin levels in children, assess differences between RCs and OCs, and develop a web-based tool for clinical assessments.

Approach:

• Study Design: Cross-sectional study using data from healthy children in Toronto, Ontario, collected during regular health supervision visits.
• Data Collection: Blood samples were analyzed for hemoglobin levels, with exclusion criteria for various health conditions and language barriers.
• Statistical Analysis: Quantile regression was used to model hemoglobin as a continuous function of age, estimating sex-specific RCs and OCs.

Key Findings:

• Reference curves (RCs) and optimal curves (OCs) provide a more nuanced understanding of hemoglobin levels compared to traditional reference intervals.
• The World Health Organization's (WHO) fixed hemoglobin thresholds may not adequately reflect age-related variability in pediatric populations.
• The study cohort was ethnically diverse, enhancing the generalizability of the findings.

Interpretation:

Age- and sex-specific curves may help clinicians interpret pediatric hemoglobin levels in the context of expected variation across childhood. Comparing reference curves with optimal curves may also help distinguish typical population distributions from levels associated with optimal health. These curves offer an alternative to applying fixed WHO thresholds across ages, although their use should be considered in light of the study’s population and exclusions.

Limitations:

• Exclusion of children with certain health conditions may limit the applicability of findings to the broader pediatric population.
• Fixed thresholds from WHO do not account for variability in hemoglobin levels among younger children.

Conclusion:

The authors concluded that age- and sex-specific reference and optimal curves can support a more nuanced assessment of hemoglobin levels in children than fixed thresholds. They developed a web-based tool to help clinicians apply these curves in practice, while noting that the findings should be interpreted in the context of the study population and exclusions.