Analyses were conducted to determine the effects of each of the four chatbot components on secondary outcomes. This analysis in particular explores the effects of the motivational interviewing component (MI) on participant willingness to seek psychotherapy for concerns their disordered weight and shape behaviors and thoughts since engagement with the intervention. Linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

Gender, age, and race were included as auxiliary variables to improve imputation quality. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. For sensitivity analysis, models were run again using full information maximum likelihood to handle missing data.

Analyses were conducted to determine the effects of each of the four chatbot components on willingness to seek psychotherapy. This analysis in particular explores the effects of the psychoeducation component (PE) on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

Analyses were conducted to determine the effects of each of the four chatbot components on willingness to seek psychotherapy. This analysis in particular explores the effects of the personalized recommendation component (PR) on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

Analyses were conducted to determine the effects of each of the four chatbot components on willingness to seek psychotherapy. This analysis in particular explores the effects of the repeated administration component (RA) on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis provides results on the 2 week time point post Intervention and contributes to understanding of the effect of time since chatbot engagement on willingness to seek psychotherapy, regardless of the combination of the four chatbot components participants were assigned (ie., component turned off or on).

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis provides results on the 6 week time point post Intervention and contributes to understanding of the effect of time since chatbot engagement on willingness to seek psychotherapy, regardless of the combination of the four chatbot components participants were assigned (ie., component turned off or on).

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis provides results on the 14 week time point post Intervention and contributes to understanding of the effect of time since chatbot engagement on willingness to seek psychotherapy, regardless of the combination of the four chatbot components participants were assigned (ie., component turned off or on).

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the interaction effects of 2-week time since engagement with the intervention and the motivational interviewing (MI) component on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns. since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the interaction effects of 6-week time since engagement with the intervention and the motivational interviewing (MI) component on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns. since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the interaction effects of 14-week time since engagement with the intervention and the motivational interviewing (MI) component on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns. since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the interaction effects of 2-week time since engagement with the intervention and the psychoeducation (PE) component on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns. since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the interaction effects of 6-week time since engagement with the intervention and the psychoeducation (PE) component on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns. since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the interaction effects of 14-week time since engagement with the intervention and the psychoeducation (PE) component on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns. since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the interaction effects of 2-week time since engagement with the intervention and the personalized recommendation (PR) component on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns. since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the interaction effects of 6-week time since engagement with the intervention and the personalized recommendation (PR) component on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns. since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the interaction effects of 14-week time since engagement with the intervention and the personalized recommendation (PR) component on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns. since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the interaction effects of 2-week time since engagement with the intervention and the repeated administration (RA) component on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns. since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the interaction effects of 6-week time since engagement with the intervention and the repeated administration (RA) component on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns. since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the interaction effects of 14-week time since engagement with the intervention and the repeated administration (RA) component on participant willingness to seek psychotherapy for concerns related to their eating, weight, and/or shape concerns. since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

We defined sufficiently large main effects as those with p values ≤ .05 or with 95% confidence intervals that do not include 0.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

This analysis explores the time effect of attitudinal changes over the 14 week course of the study, regardless of the combination of components participants were assigned (ie., component turned off or on). Results will help determine whether mean participant attitudes toward change for concerns related to their eating, weight, and/or shape concerns varied based on the time elapsed since engagement with the intervention (ie., do attitudinal changes post-inntervention endure over time?).

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We computed both unstandardized (B) and standardized (ß) coefficients to indicate effect sizes. As suggested by Fey et al. (2023), we interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

Analyses were conducted to determine the interaction effects of time since intervention and each of the four chatbot components. This analysis explores the interaction effects of time and the motivational interviewing (MI) component on participant attitudes toward change for concerns related to their eating, weight, and/or shape concerns since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We computed both unstandardized (B) and standardized (ß) coefficients to indicate effect sizes. As suggested by Fey et al. (2023), we interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.

Analyses were conducted to determine the interaction effects of time since intervention and each of the four chatbot components. This analysis explores the interaction effects of time and the personalized recommendation (PR) component on participant attitudes toward change for concerns related to their eating, weight, and/or shape concerns since engagement with the intervention.

The linear mixed effects models incorporated a random intercept to accommodate the correlation of repeated measures within participants.

We computed both unstandardized (B) and standardized (ß) coefficients to indicate effect sizes. As suggested by Fey et al. (2023), we interpreted absolute ß values below 0.2 as small, between 0.2 and 0.5 as medium, and greater than 0.5 as large.

To improve the quality of the imputations, gender, age, and race were included as auxiliary variables. We generated ten imputed datasets with the predictive mean matching method and pooled coefficients, degrees of freedom, and standard errors with Rubin's (1987) rules. As a sensitivity analysis, we also ran the same models using full information maximum likelihood to handle missing data.