Frequency of Substance Use in Adolescents With Eating Disorders and its Association With Eating Disorder Medical Severity

Abstract

Objective

The developmental characteristics of adolescence make them more susceptible to eating disorders (ED) and substance use (SU). We aimed to compare the prevalence of SU in adolescents with ED and controls, and among ED subgroups; and to examine the relationship between SU and the medical severity of EDs.

Methods

We enrolled 510 adolescents with ED and 520 age- and sex-matched controls. Current SU (≥1 occasion in the past 30 days) was obtained from psychosocial interviews. Medical severity parameters were examined for anorexia nervosa (AN)/atypical AN (AAN) and bulimia nervosa (BN).

Results

Tobacco use was higher in the ED group than in the control group (p = 0.002). Adolescents with BN reported more tobacco and alcohol use than AN and AAN (p < 0.001). There was no SU in adolescents with avoidant/restrictive food intake disorder (ARFID). Smoking was associated with higher medical severity in BN (p = 0.047), but not in AN/AAN.

Conclusions

We found that SU is more common in ED than healthy controls and in binging-purging subgroups than restrictive subgroups in ED. In addition, smoking was associated with medical severity in BN. The absence of SU in ARFID and subgroup differences underscore the need for longitudinal studies, to clarify the future impacts.

Plain language summary

Research comparing the prevalence of substance use (SU) in adolescents with eating disorder (ED) across ED subgroups is rare; and there is also a lack of studies examining relationship between SU and ED medical severity. Our paper emphasizes that while the prevalence of SU in adolescents with ED, particularly those with a binging-purging component, is more frequent than healthy adolescents, restrictive EDs, such as avoidant/restrictive food intake disorder (ARFID) were associated with lower observed rates. We show that smoking is associated with disease severity in adolescents with bulimia nervosa (BN). Our findings highlight the need for more comprehensive investigation of SU in adolescents with ARFID, as well as in other ED subtypes, along with a more precise definition of the relationship between SU and the medical severity of EDs.