Longitudinal trajectories of behavior change in a national sample of patients seeking eating‐disorder treatment

Abstract

Objective: Rapid response to treatment, indicated by substantial decreases in eatingdisorder (ED) symptoms within the first 4-6 weeks of treatment, is the most reliable predictor of treatment outcomes for EDs. However, there is limited research evaluating short-term longitudinal trajectories of ED symptoms during treatment. Thus, it is difficult to know which aspects of ED psychopathology are slow or fast to change. The purpose of this study was to elucidate three-month trajectories of ED psychopathology during treatment and test whether ED diagnosis influenced the direction and rate of change.Method: Participants were Recovery Record users seeking treatment for an ED (N = 4,568; 86.8% female). Participants completed the Eating Pathology Symptoms Inventory once per month for 3 months.Results: Latent growth curve models indicated that ED diagnosis influenced the rate of ED behavior change. Anorexia nervosa was associated with faster reductions in cognitive restraint, excessive exercise, restricting, yet slower reductions in body dissatisfaction, and binge eating. Bulimia nervosa was associated with faster reductions in binge eating, cognitive restraint, excessive exercise, and purging. Binge-eating disorder was associated with faster reductions in body dissatisfaction and binge eating, yet slower reductions in restricting. Conclusions:Our results have implications for future research by providing initial information about the direction and rate of ED change over the course of treatment.If clinicians and researchers know which ED symptoms are slow to change, on average, across diagnostic groups, treatment protocols could be adjusted to target slow changing symptoms more quickly, and therefore improve ED treatment outcomes.

Patient use of a self‐monitoring app during eating disorder treatment: Naturalistic longitudinal cohort study

“…Unfortunately, we were unable to explore the association between app activity and ED symptoms, as the EDE‐Q data were omitted due to missing data. Yet, as the link between patients’ long‐term app engagement and their symptom development is seemingly important to explore, we recommend that future studies collect longitudinal data on patient app engagement as well as changes in ED symptoms and severity similarly to the longitudinal study on RR (Chapa et al., 2019). As we were unsuccessful collecting these data using self‐report measures, future studies may consider using clinician‐collected data when possible, such as the EDE (Fairburn et al, 2008).…”

“…Thus, although concerning, high drop‐out rates are expected in ED treatment, but little is known about patients’ long‐term engagement in apps applied as part of ED treatment. A recent longitudinal study explored the changes in ED symptoms in RR users during three months (Chapa et al., 2019). The study found major significant improvements on most ED symptomatology when assessed by the Eating Pathology Symptoms Inventory (Forbush et al., 2013) measuring body dissatisfaction, binge‐eating, cognitive restraint, purging, restricting, and excessive exercising (Chapa et al., 2019).…”

“…A recent longitudinal study explored the changes in ED symptoms in RR users during three months (Chapa et al., 2019). The study found major significant improvements on most ED symptomatology when assessed by the Eating Pathology Symptoms Inventory (Forbush et al., 2013) measuring body dissatisfaction, binge‐eating, cognitive restraint, purging, restricting, and excessive exercising (Chapa et al., 2019). Still, the study did not report on the participants’ app engagement, such as the change in the number of completed logs over time.…”

Eating‐disorder psychopathology and driven exercise change models: A latent change score analysis

“…The final sample ( N = 4,568; 86.8% female) included treatment‐seeking RR users who completed 3 months of EPSI assessments. Past research demonstrated that this sub‐sample was older by 2 years and had a longer illness duration by 1 year compared to treatment‐seeking RR users who completed only one or two EPSI assessments (Chapa et al, 2020). This sub‐sample was comparable to the full sample of all treatment‐seeking RR users (Chapa et al, 2020) in ED psychopathology levels at baseline.…”

“…Past research demonstrated that this sub‐sample was older by 2 years and had a longer illness duration by 1 year compared to treatment‐seeking RR users who completed only one or two EPSI assessments (Chapa et al, 2020). This sub‐sample was comparable to the full sample of all treatment‐seeking RR users (Chapa et al, 2020) in ED psychopathology levels at baseline. A wide range of ages (11–78 years old) were represented with a mean ( SD ) age of 29.40 ( 11.83 ) years.…”