Development of the home cooking enviRonment and equipment inventory observation form (Home-CookERITM): an assessment of content validity, face validity, and inter-rater agreement

Introduction: Quantifying Home Cooking EnviRonments has applications in nutrition epidemiology, health promotion, and nutrition interventions. This study aimed to develop a tool to quantify household cooking environments and establish its content validity, face validity, and inter-rater agreement. Methods: The Home Cooking EnviRonment and equipment Inventory observation form (Home-CookERI™) was developed as a 24-question (91-item) online survey. Items included domestic spaces and resources for storage, disposal, preparation, and cooking of food or non-alcoholic beverages. Home-CookERITM was piloted to assess content validity, face validity, and usability with six Australian experts (i.e., dietitians, nutrition researchers, chefs, a food technology teacher, and a kitchen designer) and 13 laypersons. Pilot participants provided feedback in a 10 min telephone interview. Home-CookERI™ was modified to an 89-item survey in line with the pilot findings. Inter-rater agreement was examined between two trained raters in 33 unique Australian households. Raters were required to observe each item before recording a response. Home occupants were instructed to only assist with locating items if asked. Raters were blinded to each other’s responses. Inter-rater agreement was calculated by Cohen’s Kappa coefficient (κ) for each item. To optimize κ, similar items were grouped together reducing the number of items to 81. Results: Home-CookERITM had excellent content and face validity with responding participants; all 24 questions were both clear and relevant (X² (1, <i>n</i> = 19; 19.0, <i>p</i> = 0.392)). Inter-rater agreement for the modified 81-item Home-CookERI™ was almost-perfect to perfect for 46% of kitchen items (<i>n</i> = 37 items, <i>κ</i> = 0.81–1), moderate to substantial for 28% (<i>n</i> = 23, <i>κ</i> = 0.51–0.8), slight to fair for 15% (<i>n</i> = 12, <i>κ</i> = 0.01–0.5), and chance or worse for 11% of items (<i>n</i> = 9, <i>κ</i> ≤ 0.0). Home-CookERITM was further optimized by reduction to a 77-item version, which is now available to researchers. Conclusion: Home-CookERI™ is a comprehensive tool for quantifying Australian household cooking environments. It has excellent face and content validity and moderate to perfect inter-rater agreement for almost three-quarters of included kitchen items. To expand Home-CookERI™ applications, a home occupant self-completion version is planned for validation.