IMPORTANCE Celebrity social media posts engage millions of young followers daily, but the nutritional quality of foods and beverages in such posts, sponsored and unsponsored, is unknown. OBJECTIVE To quantify the nutritional quality of foods and beverages depicted in social media accounts of highly followed celebrities and assess whether nutritional quality is associated with post sponsorship, celebrity profession or gender, and followers' likes and comments. DESIGN, SETTING, AND PARTICIPANTS This cross-sectional study analyzed the content of foodand beverage-containing posts from Instagram (a photo- and video-sharing social media platform) accounts of 181 highly followed athletes, actors, actresses, television personalities, and music artists. Data were collected from May 2019 to March 2020. MAIN OUTCOMES AND MEASURES The nutritional quality of foods and beverages posted in celebrity social media accounts was rated using the Nutrient Profile Index (NPI) based on the sugar, sodium, energy, saturated fat, fiber, protein, and fruit and/or vegetable content per 100-g sample (a score of 0 indicated least healthy and 100, healthiest); foods with scores less than 64 and beverages with scores less than 70 were rated as "less healthy." Secondary outcomes were whether the nutritional quality of foods and beverages in social media posts was associated with post sponsorship, celebrity profession or gender, and followers' likes and comments. Mixed-effects regression models were used to estimate how outcomes differed across fixed effects. RESULTS The sample included social media accounts of 181 celebrities (66 actors, actresses, and television personalities [36.5%]; 64 music artists [35.4%]; and 51 athletes [28.2%]). A total of 102 celebrities (56.4%) were male, and the median age was 32 years (range, 17-73 years). Among 3065 social media posts containing 5180 total foods and beverages (2467 foods [47.6%]; 2713 beverages [52.4%]), snacks and sweets (920 [37.3%] of the foods) and alcoholic beverages (1375 [50.7%] of the beverages) were most common. Overall, 158 celebrity social media accounts (87.3%) earned a less healthy overall food nutrition score and 162 (89.5%) earned a less healthy overall beverage nutrition score, which would be unhealthy enough to fail legal youth advertising limits in the UK. For foods, social media posts with healthier nutrition scores were associated with significantly fewer likes (b, -0.003; 95% CI, -0.006 to 0.000; P=.04) and comments (b, -0.006; 95% CI, -0.009 to -0.003; P<.001) from followers. For beverages, nutrition scores were not significantly associated with likes (b, -0.010; 95% CI, -0.025 to 0.005; P=.18) or comments (b, -0.003; 95% CI, -0.022 to 0.016; P=.73). Only 147 food- or beverage-containing posts (4.8%) were sponsored by food- or beverage-relevant companies. Beverages in sponsored posts contained more than twice as much alcohol as those in nonsponsored posts (10.8 g [95% CI, 9.3 g to 12.3 g] per 100 g of beverage vs 5.3 g [95% CI, 4.7 g to 5.9 g] per 100 g of beverage). CONCLUSIONS AND RELEVANCE In this cross-sectional study, most highly followed celebrity social media accounts depicted an unhealthy profile of foods and beverages, primarily in nonsponsored posts. These results suggest that influential depictions of unhealthy food and beverage consumption on social media may be a sociocultural problem that extends beyond advertisements and sponsorships, reinforcing unhealthy consumption norms.
