OBJECTIVE: To describe the association between height, demographics, and treatment in youths with type 1 diabetes participating in an international network for pediatric diabetes centers (SWEET).

METHODS: Data were collected from 55 centers with documented patients' height. All subjects below 20 years of age, diabetes duration >1 year, and without celiac disease were included. World Health Organization growth charts were used to calculate height and body mass index z-scores. Multiple hierarchic regression models adjusting for known confounders were applied.

RESULTS: Data on 22 941 subjects (51.8% male) were analyzed with a median and interquartile range for age 14.8 years (11.2, 17.6), diabetes duration 5.6 years (3.1, 8.9), and height z-score 0.34 (-0.37, 1.03). Children were taller in the youngest age groups: adjusted height z-scores of 0.31 (±0.06) and 0.39 (±0.06), respectively; with shorter diabetes duration (<2 years: 0.36 [±0.06]; 2-<5 years: 0.34 [±0.06]; ≥5 years: 0.21 [±0.06]) and if they were pump users: 0.35 ± 0.05 vs 0.25 ± 0.05 (>three injections/day and 0.19 ± 0.06 [0-3 injections daily]), respectively. High hemoglobin A1c (HbA1c) and low to normal weight were associated with a lower height z-score. Trends were identical in all models except for gender. No gender differences were found except in the final height model where females exhibited higher z-score than males.

CONCLUSION: For youths treated at centers offering modern diabetes management, major growth disturbances are virtually eliminated. For children with a young age at onset, high HbA1c, injections, and/or non-intensive diabetes, treatment still requires attention in order to attain normal growth.