Youth with anorexia nervosa (AN) and atypical anorexia nervosa (AAN; i.e., meeting all criteria for AN but not underweight at diagnosis) are at risk for significant medical complications (American Psychiatric Association, 2013; Arcelus et al., 2011; Mehler et al., 2022). It is a widespread misconception that the presenting weight of an individual with AN or AAN is a proxy for how “sick” they are (Garber et al., 2019; Whitelaw et al., 2018). However, recent research comparing characteristics of youth with AN and AAN have shown commensurate or more severe complications among youth with AAN, despite higher weight at presentation (Sawyer et al., 2016; Swenne, 2016; Walsh et al., 2023; Whitelaw et al., 2014). Instead, for youth with either AN or AAN, percent of total body weight lost, coupled with the rapidity of weight loss—not necessarily the degree of malnutrition—predict the occurrence and severity of medical complications, including vital sign abnormalities, refeeding hypophosphatemia, and longer hospital stays among medically hospitalized patients (Assalone et al., 2022; Garber et al., 2019; Kennedy et al., 2017; Meierer et al., 2019; Swenne, 2016; Whitelaw et al., 2018).

Illness severity in AN/AAN may also be shaped by one's premorbid weight, or one's weight status prior to developing an eating disorder (ED). Notably, a significant percentage of youth with AN/AAN have body mass indexes [BMI] characterized as overweight/obesity (BMI percentile ≥85th percentile for age and sex per the United States [U.S.] Centers for Disease Control [CDC]) prior to onset of their ED (Lebow et al., 2015; Matthews et al., 2023). Compared to youth with premorbid normal weights, youth with AN/AAN and premorbid overweight/obesity are more often encouraged to lose weight and receive positive reinforcement for dieting (Lin et al., 2023). This subset of patients more often meets criteria for AAN due to presenting at categorically “normal” weights (Matthews et al., 2022; Sawyer et al., 2016), yet can also meet AN criteria if their weight loss is severe enough. In fact, youth with AN/AAN and premorbid overweight/obesity have often lost more weight and at a faster rate than patients without a history of overweight/obesity (Kennedy et al., 2017; Matthews, Kramer and Mitan, 2022, Matthews, Gordon, Kramer, Lin and Ziv, 2023; Meierer et al., 2019). Therefore, it is critical to better understand whether premorbid weight status presents unique risk factors in AN/AAN.

It is conceivable that youth with AN/AAN and premorbid overweight/obesity present for treatment with more severe medical complications, given the likelihood of having lost more weight at a faster rate than youth with premorbid normal weights (Kennedy et al., 2017; Matthews, Kramer and Mitan, 2022, Matthews, Gordon, Kramer, Lin and Ziv, 2023; Meierer et al., 2019). Whereas research has clarified the association between presenting weight status (i.e., AN/AAN) and medical complications in youth (Assalone et al., 2022; Garber et al., 2019; Kennedy et al., 2017; Meierer et al., 2019; Sawyer et al., 2016; Swenne, 2016; Walsh et al., 2023; Whitelaw et al., 2018), the relationship between premorbid weight status and medical sequalae is poorly understood. It is important to determine whether premorbid weight status places youth with AN/AAN at differential risk, as findings could direct and improve care for youth with these life-threatening illnesses. This study aimed to examine the association between premorbid weight status and medical complications (i.e., lab and vital sign abnormalities) while controlling for presenting weight status in youth with AN/AAN hospitalized for medical stabilization.