BACKGROUND: Assessing pediatric patients for insulin resistance is one way to identify those who are at a high risk of developing type 2 diabetes mellitus. The homoeostasis model assessment (HOMA) is a measure of insulin resistance based on fasting blood glucose and insulin levels. Although this measure is widely used in research, cutoff values for pediatric populations have not been established. OBJECTIVE: To assess the validity of HOMA cutoff values used in pediatric studies published in peer-reviewed journals. METHODS: Studies published from January 2010 to December 2015 were identified through MEDLINE. Initial screening of abstracts was done to select studies that were conducted in pediatric populations and used HOMA to assess insulin resistance. Subsequent full-text review narrowed the list to only those studies that used a specific HOMA score to diagnose insulin resistance. Each study was classified as using a predetermined fixed HOMA cutoff value or a cutoff that was a percentile specific to that population. For studies that used a predetermined cutoff value, the references cited to provide evidence in support of that cutoff were evaluated. RESULTS: In the 298 articles analyzed, 51 different HOMA cutoff values were used to classify patients as having insulin resistance. Two hundred fifty-five studies (85.6%) used a predetermined fixed cutoff value, but only 72 (28.2%) of those studies provided a reference that supported its use. One hundred ten studies (43%) that used a fixed cutoff either cited a study that did not mention HOMA or provided no reference at all. Tracing of citation history indicated that the most commonly used cutoff values were ultimately based on studies that did not validate their use for defining insulin resistance. CONCLUSION: Little evidence exists to support HOMA cutoff values commonly used to define insulin resistance in pediatric studies. These findings highlight the importance of validating study design elements when training medical students and novice investigators. Using available data to generate population ranges for HOMA would improve its clinical utility.
BACKGROUND: Assessing pediatric patients for insulin resistance is one way to identify those who are at a high risk of developing type 2 diabetes mellitus. The homoeostasis model assessment (HOMA) is a measure of insulin resistance based on fasting blood glucose and insulin levels. Although this measure is widely used in research, cutoff values for pediatric populations have not been established. OBJECTIVE: To assess the validity of HOMA cutoff values used in pediatric studies published in peer-reviewed journals. METHODS: Studies published from January 2010 to December 2015 were identified through MEDLINE. Initial screening of abstracts was done to select studies that were conducted in pediatric populations and used HOMA to assess insulin resistance. Subsequent full-text review narrowed the list to only those studies that used a specific HOMA score to diagnose insulin resistance. Each study was classified as using a predetermined fixed HOMA cutoff value or a cutoff that was a percentile specific to that population. For studies that used a predetermined cutoff value, the references cited to provide evidence in support of that cutoff were evaluated. RESULTS: In the 298 articles analyzed, 51 different HOMA cutoff values were used to classify patients as having insulin resistance. Two hundred fifty-five studies (85.6%) used a predetermined fixed cutoff value, but only 72 (28.2%) of those studies provided a reference that supported its use. One hundred ten studies (43%) that used a fixed cutoff either cited a study that did not mention HOMA or provided no reference at all. Tracing of citation history indicated that the most commonly used cutoff values were ultimately based on studies that did not validate their use for defining insulin resistance. CONCLUSION: Little evidence exists to support HOMA cutoff values commonly used to define insulin resistance in pediatric studies. These findings highlight the importance of validating study design elements when training medical students and novice investigators. Using available data to generate population ranges for HOMA would improve its clinical utility.
Authors: Marta Díaz; Judit Bassols; Abel López-Bermejo; Francis de Zegher; Lourdes Ibáñez Journal: Pediatr Diabetes Date: 2014-10-20 Impact factor: 4.866
Authors: Raquel M S Campos; Marco T de Mello; Lian Tock; Patrícia L Silva; Deborah C L Masquio; Aline de Piano; Priscila L Sanches; June Carnier; Flávia C Corgosinho; Denis Foschini; Sergio Tufik; Ana R Dâmaso Journal: J Strength Cond Res Date: 2014-03 Impact factor: 3.775
Authors: B García Cuartero; C García Lacalle; C Jiménez Lobo; A González Vergaz; C Calvo Rey; M J Alcázar Villar; E Díaz Martínez Journal: An Pediatr (Barc) Date: 2007-05 Impact factor: 1.500
Authors: Maria Izabel Siqueira de Andrade; Juliana Souza Oliveira; Vanessa Sá Leal; Poliana Coelho Cabral; Pedro Israel Cabral de Lira Journal: PLoS One Date: 2021-02-09 Impact factor: 3.240