Tsai Sarah, Patton Susana, DeLurgio Stephen, Williams David D, Dileepan Kavitha, Karmazin Alexander, Storm Melinda, Clements Mark
Division of Pediatric Endocrinology, Children's Mercy Kansas City, Kansas City, MO, USA.
University of Missouri-Kansas City School of Medicine, Kansas City, MO, USA.
Eur Endocrinol. 2019 Apr;15(1):1-5. doi: 10.17925/EE.2019.15.1.1. Epub 2019 Apr 12.
The purpose of this study was to develop a measure of type 1 diabetes mellitus (T1DM) knowledge that is aimed at youth and is based on contemporary management standards. An 88-item test was derived from the American Association of Diabetes Educators 7 Self-Care Behaviors. A multidisciplinary team selected the best 49 items which were piloted in a sample of 119 youths (59 males, aged 12-18, having a mean ± standard deviation glycated haemoglobin (A1C) of 9.9%±1.80 (84.7±19.7 mmol/mol). A minimum absolute point-biserial correlation coefficient of 0.250 was used to choose 49 items from the original 88 questions. Categorical principal component analysis was then used to identify the best factor analytical model that consisted of five factors composed of 19 items. These five factors explained 57% of item variances. Factors were associated with the latent variables: advanced problem-solving, hypoglycaemia prevention and management, taking insulin/medication administration, daily management and healthy active living. A new T1D knowledge test for youth was refined from 88 to 49 questions based on expert opinion and empirical test construction. The instrument was then refined to 19 items based on exploratory factor analysis. Future goals are to validate this factor model with another cohort and confirm concurrent validity based on youth's glycated haemoglobin and adherence behaviours. Our new T1DM knowledge measure initially appears valid and promising as a new clinical and research tool.
本研究的目的是开发一种针对青少年的1型糖尿病（T1DM）知识测量方法，该方法基于当代管理标准。一个88项的测试源自美国糖尿病教育者协会的7种自我护理行为。一个多学科团队从这88项中选出了最佳的49项，并在119名青少年样本中进行了试点（59名男性，年龄在12 - 18岁之间，糖化血红蛋白（A1C）的平均值±标准差为9.9%±1.80（84.7±19.7 mmol/mol））。使用最小绝对点二列相关系数0.250从最初的88个问题中选择49项。然后使用分类主成分分析来确定最佳的因子分析模型，该模型由五个因子组成，包含19个项目。这五个因子解释了57%的项目方差。这些因子与潜在变量相关：高级问题解决、低血糖预防与管理、胰岛素/药物服用、日常管理以及健康积极的生活方式。基于专家意见和实证测试构建，针对青少年的新的T1D知识测试从88个问题精简到了49个问题。然后基于探索性因子分析将该工具进一步精简为19个项目。未来的目标是用另一个队列验证这个因子模型，并基于青少年的糖化血红蛋白和依从行为确认同时效度。我们新的T1DM知识测量方法初步看来是有效的，并且作为一种新的临床和研究工具很有前景。
