Development and reliability of a questionnaire to assess the dietary knowledge of Italian adult and elderly population: the QCAI (Questionario delle Conoscenze Alimentari Italiano)

Background: Reliable tools to quantify nutrition knowledge are crucial for planning education and prevention. We developed the Questionario delle Conoscenze Alimentari Italiano (QCAI) for Italian adults and assessed its reliability. Methods: A convenience sample of 165 participants (18-89 years; 70.9% women) completed a self-administered, 69-item questionnaire organized into four content sections (experts' recommendations; nutrient content of foods; healthy choices; meal frequency and labels) plus demographics. Participants completed the questionnaire at baseline (T0) and again after seven days (T1) under identical conditions. Internal consistency was evaluated with Cronbach's alpha (and KR-20 for dichotomous items). Temporal stability was assessed with intraclass correlation coefficients (ICC(2,1)) for section and total scores, Cohen's kappa (κ) for items, and Bland-Altman analysis. Results: Total-scale internal consistency was acceptable (Cronbach's α=0.757 for available items), but domain-level α values were heterogeneous and in some sections low (α: 0.307-0.638), indicating limited coherence of some subscales. Test-retest stability for the total score was moderate (ICC(2,1)=0.62), and 64.29% of items showed at least moderate agreement (κ≥0.40). Age-stratified analyses showed no linear trend in the total score across groups (18-39, 40-59, ≥60; Ptrend=0.26). Conclusions: QCAI is a low-cost instrument that shows moderate temporal stability and acceptable reliability for the total score in Italian adults. However, the low internal consistency observed in some domains limits the interpretability of section scores and highlights the need for item refinement and formal validation (content, construct, and criterion validity) against established measures before routine clinical or public-health use.