Validity of the Mediterranean Diet and Culinary Index (MediCul) for Online Assessment of Adherence to the ‘Traditional’ Diet and Aspects of Cuisine in Older Adults

Abstract

The Mediterranean diet is associated with multiple health benefits. Yet, no tool has been specifically developed to assess adherence to the ‘traditional’ Mediterranean diet and cuisine within a Western cohort, and validated for online use. We tested the reliability and validity of online administration of the Mediterranean Diet and Culinary Index (MediCul) among middle-aged and older adults. Participants were recruited in January⁻March 2017 from the 45 and Up Study, completing MediCul twice. Test-retest reliability was assessed using the paired t-test, intra-class correlation coefficient (ICC) and Bland-Altman plot. Validity was tested against a three-day food record (FR)-derived MediCul score using Bland-Altman and nutrient trends across the MediCul score tertiles. Participants (n = 84; 60% female; 65.4 years (SD = 5.9)), were overweight (BMI 26.1; SD = 4.0) with 1.7 (SD = 1.5) chronic illnesses/conditions. Sequential MediCul tool scores were 56.1/100.0 and 56.8/100.0, respectively (t = −1.019; p = 0.311). Reliability via ICC (ICC = 0.86, 95% CI: 0.789, 0.910, p < 0.0001) and Bland-Altman was good. In Bland-Altman validity analyses, the tool over-reported FR MediCul score by 5.6 points with no systematic bias ((y = 8.7 − 0.06*x) (95% CI: −0.278, 0.158, p = 0.584)). Nutrient trends were identified for MediCul consistent with expected Mediterranean patterns. Online MediCul administration demonstrated good reliability and moderate validity for assessing adherence to a ‘traditional’ Mediterranean pattern among older Australians.

Keywords

• validity
• reliability
• repeatability
• dietary assessment
• index tool
• score
• Mediterranean diet
• Mediterranean dietary pattern
• traditional