Department of Nutrition, School of Health Sciences, University of Brasilia (UnB), Campus Darcy Ribeiro, Asa Norte, Brasilia DF 70910-900, Brazil.
Faculty of Medicine, University of Brasilia (UnB), Campus Darcy Ribeiro, Asa Norte, Brasilia DF 70910-900, Brazil.
Nutrients. 2018 Sep 10;10(9):1274. doi: 10.3390/nu10091274.
Gluten cross-contamination in gluten-free food may jeopardize treatment of celiac patients. Considering the deficit of appropriate instruments to enable the implementation of safe production practices for gluten-free food, this study aimed to evaluate the application of a check-list elaborated for gluten cross-contamination prevention in food services. The instrument was applied in 60 Brazilian food services. Interobserver reproducibility and internal consistency of the check-list were tested. A score classification was created for establishments according to the food contamination risk assessment. Subsequent to the application and statistical analysis, the original instrument was reduced to a 30-item check-list. In the reproducibility analysis, none of the 30 items showed significant divergence among the evaluators ( > 0.05 in the Cochran Q test). The 30-item version of the check-list presented Kuder⁻Richardson Formula 20 (KR-20) = 0.771, indicating good internal consistency. The proposed classification score is obtained by adding 1 point for each item with an "adequate" response; therefore, the final score may vary between 0 and 30 points. Establishments with up to 15 points exhibit risk of gluten contamination, while establishments with a score above 16 points exhibit low risk of contamination. The check-list displayed good reproducibility and internal consistency, suggesting that it could be a useful gluten contamination control instrument in food services.
无麸质食品中的麸质交叉污染可能危及乳糜泻患者的治疗。鉴于缺乏适当的工具来实施无麸质食品的安全生产实践，本研究旨在评估用于食品服务中预防麸质交叉污染的检查表的应用。该工具应用于 60 家巴西食品服务机构。测试了检查表的观察者间再现性和内部一致性。根据食品污染风险评估，为各机构创建了评分分类。在应用和统计分析之后，原始工具被简化为 30 项检查表。在再现性分析中，30 项检查表中的任何一项在评估者之间都没有明显差异（Cochran Q 检验> 0.05）。检查表的 30 项版本的 Kuder⁻Richardson 公式 20（KR-20）= 0.771，表明内部一致性良好。所提出的分类评分是通过为每个“适当”回答的项目加 1 分获得的；因此，最终分数可能在 0 到 30 分之间变化。得分在 15 分以下的机构存在麸质污染的风险，而得分在 16 分以上的机构则存在低污染风险。检查表显示出良好的再现性和内部一致性，表明它可能是食品服务中控制麸质污染的有用工具。
