When asked, many people can recite the broad principles of healthy eating^{(Reference Bisogni, Jastran and Seligson1)}. However, the degree to which these principles must be followed for someone to perceive their diet as ‘healthy’ may be influenced by their perceived health and weight status^{(Reference Sørensen, Matthiessen and Holm2)}. Moreover, for some, a diet can be deemed healthy because it provides satisfaction^{(Reference Bisogni, Jastran and Seligson1)}. The aim of this study was to compare the influence of self- perceived health (SPH), self-perceived weight (SPW) and satisfaction with food-related life (SWFL) on perceived healthy eating behaviours (PHEB) and measured diet quality. This study used data from the National Adult Nutrition Survey collected between 2008–2010 (n = 1219, male/female, 599/620).

Diet was assessed using a 4-day semi-weighed food diary. Diet quality was assessed using the Nutrient-Rich Food Index (NRF9.3). PHEB was assessed using an average of 6-items taken from previously validated tools (e.g., “Healthy eating is something I do frequently”). Satisfaction with food- related life (SWFL) was measured using 4-items taken from the SWFL scale. All items were measured on a 7-point Likert scale, ranging from 1 = strongly disagree to 7 = strongly agree. SPH and SPW were assessed using single items and five responses were possible for each. Separate hierarchical multiple linear regressions were used to explore the influence of SPH, SPW, SWFL and sociodemographic factors on PHEB and NRF9.3. The NRF9.3 model found that 23.3% of the variation could be explained by PHEB. Adding SPH, SPW and SWFL had no effect on the predictive strength of the model. The final model, including PHEB, SPH, SPW, SWFL, BMI and sociodemographics predicted a further 26.1% (49.4% in total) of the variation in NRF9.3. In addition to PHEB, being female, older age and having a lower BMI were significantly associated with higher NRF9.3. The model predicting PHEB found that the 23.3% of the variation explained by NRF9.3 could be increased by 8.5% (31.8% in total) when SPH and SPW were added. Adding SWFL explained a further 12% of the variation in PHEB (43.8% in total). However, the addition of BMI and sociodemographics only increased the prediction by a further 0.5% (44.3% in total). Comparing the models, we found that SPH, SPW and SWFL greatly increased the predictive strength of the model for PHEB but not NRF9.3. Whereas sex, age and BMI greatly enhanced the predictive strength for NRF9.3 but not PHEB. These differences suggest people may use their perceived health characteristics and overall dietary satisfaction to guide how they judge the healthiness of their diet. In addition, it suggests that women, older adults, and people with a lower BMI may assess the healthiness of their diets more closely to dietary recommendations. Further research is needed to understand how perceived health characteristics and dietary satisfaction influence people's interpretations of healthy eating.