Development and validation of a multidimensional instrument to assess complementary feeding practices among infants and young children aged 6–23 months

Abstract

The complementary feeding period, from 6 to 23 months of age, is a critical window for healthy growth and malnutrition prevention, making accurate measurement of caregiver practices essential. However, existing instruments have primarily focused on dietary intake or feeding style and often neglected broader aspects of the feeding process, such as the feeding environment and hygiene practices, despite their documented importance for child nutrition. This study aimed to develop and validate a multidimensional instrument for assessing complementary feeding practices among infants and young children. Domains were informed by the best-practice complementary feeding behaviors framework, a literature review, and a two-round Delphi process with 31 experts. The draft questionnaire was pretested and administered in a validation survey of 556 caregiver–child dyads with children aged 6–23 months, recruited through multistage sampling in rural Sichuan, China. Item screening was conducted using both classical test theory and item response theory. Exploratory factor analysis identified the latent structure, and confirmatory factor analysis evaluated structural validity. Reliability was assessed using Cronbach’s α and split-half reliability coefficients. Indicator weights were derived by combining Delphi-based subjective weights with entropy-based objective weights. The two Delphi rounds demonstrated high expert authority coefficients (0.89 and 0.90) and improved consensus (Kendall’s $$\:W$$ from 0.10 to 0.19, P  < 0.001). The finalized instrument comprised four domains: dietary quality, feeding style, feeding environment, and feeding hygiene, and seven factors: appropriate complementary feeding (7 items), avoidance of unhealthy food and beverage consumption (3 items), avoidance of controlling feeding practices (9 items), restriction of unhealthy food and beverage provision (3 items), guided feeding practices (3 items), eating environment (3 items), and hand hygiene (3 items). Confirmatory factor analysis provided evidence of acceptable structural validity, with adequate model fit indices (GFI = 0.920, CFI = 0.903, TLI = 0.917, RMSEA = 0.038, SRMR = 0.046), and internal consistency was acceptable (overall Cronbach’s α = 0.78; split-half reliability coefficient = 0.77). Weight determination analysis indicated that dietary quality had the highest domain weight (0.319), followed by feeding style (0.272), feeding hygiene (0.240), and feeding environment (0.169). This newly developed instrument provides a reliable and comprehensive assessment of complementary feeding practices, extending beyond dietary intake to encompass feeding style, environment, and hygiene. It may support more precise assessment and inform targeted interventions, although further validation in diverse populations and settings is warranted.