ESPE Yearbook of Paediatric Endocrinology

Pediatrics. 2020 Sep;146(3):e20200265. doi: 10.1542/peds.2020-0265. PMID: 32778539.

In brief: This cross-sectional analysis of a nationally representative US sample showed that 25.5% of US children and adolescents (10.6 million in 2016) are eligible for screening for diabetes according to American Diabetes Association (ADA) criteria. The eligibility criteria showed low sensitivity and specificity to detect hyperglycaemia defined by HbA1c ≥5.7% (55.5% and 76.3%, respectively) or fasting plasma glucose (FPG) ≥100 mg/dl (35.8% and 77.1%, respectively). Confirmed undiagnosed diabetes was rare (<0.5% of youth). Hence, the ADA screening approach selects a very large number of children and adolescents, of whom only tiny proportion test positive for diabetes or prediabetes, and a substantial number of children with diabetes but non-eligible for screening are evidently missed.

Comment: Clinical practice guidelines are considered one of the most influential and effective tools for the promotion of evidence-based medicine. The traditional approach to developing clinical guidelines has been labelled GOBSAT (good old boys sitting around a table). While substantial international effort and high-quality evidence are invested in the development of guidelines, considerably less effort has been invested in updating guidelines (1). Hopefully, recent findings such as these will alter the strength of the body of evidence.

In 2000, the ADA and the American Academy of Pediatrics recommended screening for T2DM in high-risk youth aged 10 years and older (or after the onset of puberty). Risk was considered high if overweight plus at least two of the following: non-white race, family history of T2DM (first- or second-degree relatives), maternal gestational diabetes and signs of insulin resistance (including: acanthosis nigricans, hypertension, dyslipidemia, polycystic ovarian syndrome and small-for-gestational-age). This recommendation was expanded in 2018 to include all youth with overweight plus one or more of these risk factors. This study quantifies the implications of this change: one-quarter of all US children and adolescents are eligible for screening for diabetes by the 2018 criteria, compared only 10% (3.6 million in 2016) according to the pre-2018 criteria.

Unconfirmed undiagnosed diabetes, defined by a single HbA1c of 6.5%, was observed in 0.3% (95% confidence interval [CI]: 0.1%–0.5%) of youth who were eligible for screening, and in 0.1% of those ineligible. A single elevated FPG value did not enable precise prediction of diabetes. Diagnosed diabetes was seen in 0.5% (95% CI: 0.4%–0.7%) of youth (0.2 million), equating to 85% of the total confirmed persons with diabetes. Similarly, the prevalence of prediabetes varied significantly depending on the definition. In conclusion, current screening criteria are not sensitive or specific and may miss many youth with diabetes. This study is important as it shows that clinical recommendations must be critically appraised.

Reference: 1. Kredo T, Bernhardsson S, Machingaidze S, Young T, Louw Q, Ochodo E, Grimmer K. Guide to clinical practice guidelines: the current state of play, International Journal for Quality in Health Care, Volume 28, Issue 1, February 2016, Pages 122–128, https://doi.org/10.1093/intqhc/mzv115.