TECH uses team-based care similar to TEAM, but in-person care is substituted with telemedicine (i.e., video visits and consultations) and digital behavioral health tools.

IBD-IBS Symptom Severity was measured using the Irritable Bowel Syndrome Symptom Severity Scale (IBS-SSS) to more accurately capture IBD symptom severity for many individuals with IBD, particularly those with inactive IBD but persistent functional GI symptoms (e.g., IBD-IBS). The IBS-SSS scores pain severity, pain frequency, abdominal bloating, bowel satisfaction and interference with life on a 0-100 scale with a total score of 500. A 50-point or more reduction in this score is considered clinically meaningful. Total scores range from 0 to 500 with higher scores indicating more severe symptoms. Answers to all questions are summed to achieve the total score. Subjects can be categorized as having mild (75-175), moderate (175-300), or severe (>300) IBS. Only total scores are reported.

Health care utilization was captured through EHR data to measure categorical (yes or no) instances of ED visits and inpatient hospitalizations at both 6- and 12-months post-enrollment.

Patient engagement was measured using the validated IBD Self-Efficacy Scale (IBD-SES) which assesses one's confidence level in managing stress and emotions, managing medical care, managing symptoms and disease, and maintaining remission. The IBD-SES measures self-management activities across four subscales (Stress and Emotions; Medical Care; Symptoms and Disease; Remission), with higher scores indicating greater confidence in self-management. Responses for each question are summed to produce one overall and four subscale scores. The total score range is between 29 and 290, with 29 indicating the lowest level of self-efficacy and 290 indicating the highest. Each item is scored 1-10, with 1 being assigned for responses of "Not confident at all" and a score of 10 assigned for responses of "Totally confident." Only total scores are reported.

Quality of life was evaluated using the Short Quality of Life in Inflammatory Bowel Disease Questionnaire (SIBDQ), a validated tool which measures physical, social, and emotional health-related quality of life (QOL) for patients with Crohn's disease and ulcerative colitis. On a 7-point scale, lower scores correspond with a lower perceived QOL, while higher scores represent a higher perceived QOL. Responses for each question are summed for a total score between 10 and 70, with 10 reflecting poor QOL and 70 reflecting high QOL. Each item is scored 1-7, with 1 being assigned for responses indicating the highest severity of symptoms, and 7 being assigned for those reflecting the lowest severity of symptoms. Only total scores are reported.

Total of all reporting groups

Ethnicity self-reported by participants at baseline.

Race self-reported by participants at baseline.

IBD diagnosis confirmed by physician at baseline.

Incidence of steroid prescription at baseline was extracted from the electronic medical record.

Incidence of opioid prescription at baseline was extracted from the electronic medical record.

Incidence of advanced IBD therapy prescription at baseline was extracted from the electronic medical record. Advanced therapies include: Infliximab/Remicade, Adalimumab/Humira, Certolizumab/Cimzia, Golimumab/Simponi, Vedolizumab/Entyvio, Natalizumab/Tysabri, Ustekinumab/Stelara, Tofacitinib/Xeljanz, and Ozanimod/Zyposia.

Incidence of any one of 14 investigator-defined select comorbidities at baseline. Comorbidities of interest include: AIDS; autonomic disorder; chronic kidney disease; chronic pain syndrome; connective tissue disease; COVID-19; diabetes (Type 1 or 2); irritable bowel syndrome; liver disease; peptic ulcer disease; primary sclerosing cholangitis; rheumatologic disease; short gut syndrome; and systemic autoimmune conditions.

Age of participants at baseline.

Behavioral health symptom severity was measured utilizing the Patient Health Questionnaire Anxiety-Depression Scale (PHQ-ADS) - which combines the PHQ-8 and GAD-7 scales - as a composite measure of depression and anxiety. The Patient Health Questionnaire 8-item depression scale (PHQ-8) and 7-item Generalized Anxiety Disorder scale (GAD-7) are among the best validated and most commonly used depression and anxiety measures, respectively. The PHQ-ADS is the sum of the PHQ-8 and GAD-7, where a decrease in scores indicates a decrease in BH symptom severity; a decrease of 3-4 points indicates MCID. Lower scores indicate lower levels of behavioral health symptom severity; higher scores indicate higher symptom severity. The minimum score is 0; the maximum score is 45. Only the total PHQ-ADS is reported.

Functional impairment was measured using the validated Short Form 12 Health Survey Version 2 (SF-12v2) which includes 12 items from the Short-Form 36 Health Survey and yields a physical and mental composite score as well as 8 subscale values: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. Responses are tallied via a norm-based scoring algorithm provided by the assessment licensing company. Lower scores correspond with poorer health-related functioning; higher scores indicate better health-related functioning. The minimum score is 18.5; the maximum score is 142.6. Only the total composite score (the sum of both the physical and mental scales) is reported.

IBD-IBS Symptom Severity was measured using the Irritable Bowel Syndrome Symptom Severity Scale (IBS-SSS) to more accurately capture IBD symptom severity for many individuals with IBD, particularly those with inactive IBD but persistent functional GI symptoms (e.g., IBD-IBS). The IBS-SSS scores pain severity, pain frequency, abdominal bloating, bowel satisfaction and interference with life on a 0-100 scale with a total score of 500. A 50-point or more reduction in this score is considered clinically meaningful. Total scores range from 0 to 500 with higher scores indicating more severe symptoms. Answers to all questions are summed to achieve the total score. Subjects can be categorized as having mild (75-175), moderate (175-300), or severe (>300) IBS. Only total scores are reported.

Health care utilization was captured through EHR data to measure categorical (yes or no) instances of ED visits and inpatient hospitalizations at both 6- and 12-months post-enrollment.

Patient engagement was measured using the validated IBD Self-Efficacy Scale (IBD-SES) which assesses one's confidence level in managing stress and emotions, managing medical care, managing symptoms and disease, and maintaining remission. The IBD-SES measures self-management activities across four subscales (Stress and Emotions; Medical Care; Symptoms and Disease; Remission), with higher scores indicating greater confidence in self-management. Responses for each question are summed to produce one overall and four subscale scores. The total score range is between 29 and 290, with 29 indicating the lowest level of self-efficacy and 290 indicating the highest. Each item is scored 1-10, with 1 being assigned for responses of "Not confident at all" and a score of 10 assigned for responses of "Totally confident." Only total scores are reported.

Quality of life was evaluated using the Short Quality of Life in Inflammatory Bowel Disease Questionnaire (SIBDQ), a validated tool which measures physical, social, and emotional health-related quality of life (QOL) for patients with Crohn's disease and ulcerative colitis. On a 7-point scale, lower scores correspond with a lower perceived QOL, while higher scores represent a higher perceived QOL. Responses for each question are summed for a total score between 10 and 70, with 10 reflecting poor QOL and 70 reflecting high QOL. Each item is scored 1-7, with 1 being assigned for responses indicating the highest severity of symptoms, and 7 being assigned for those reflecting the lowest severity of symptoms. Only total scores are reported.

TECH is a low-human touch approach leveraging behavioral digital tools - primarily RxWell, a digital cognitive behavioral therapy mobile application with embedded health coaches - to deliver behavioral health care at the patient's convenience.

Aim 2a, moderation by disease activity. Null hypothesis: Disease activity does not moderate the trajectory of behavioral health score by treatment group.

Threshold for statistical significance was 0.05.

Considering a range of effect sizes for B1, B2, and B3, we estimated the outcome Y as Y=B1*X+B2*M+B3*X*M+ε, where ε~N(0,1). Using Monte Carlo methods, we averaged across scenarios to determine the typical effect of B3 that can be detected with our current and projected sample sizes after 25% attrition. With our projected sample sizes we expect 0.80 power to detect a significant interaction when the magnitude of B3 is at least 0.50.

Aim 2b, moderation by age. Null hypothesis: Age does not moderate the trajectory of behavioral health score by treatment group.

The Chi-squared statistic for the group*time*age interaction term is 1.37 with two degrees of freedom.

Threshold for statistical significance was 0.05.

Considering a range of effect sizes for B1, B2, and B3, we estimated the outcome Y as Y=B1*X+B2*M+B3*X*M+ε, where ε~N(0,1). Using Monte Carlo methods, we averaged across scenarios to determine the typical effect of B3 that can be detected with our current and projected sample sizes after 25% attrition. With our projected sample sizes we expect 0.80 power to detect a significant interaction when the magnitude of B3 is at least 0.50.

Aim 2b, moderation by diagnosis. Null hypothesis: Diagnosis does not moderate the trajectory of behavioral health score by treatment group.

The Chi-squared statistic for the group*time*diagnosis interaction term is 5.16 with two degrees of freedom.

Threshold for statistical significance was 0.05.

Considering a range of effect sizes for B1, B2, and B3, we estimated the outcome Y as Y=B1*X+B2*M+B3*X*M+ε, where ε~N(0,1). Using Monte Carlo methods, we averaged across scenarios to determine the typical effect of B3 that can be detected with our current and projected sample sizes after 25% attrition. With our projected sample sizes we expect 0.80 power to detect a significant interaction when the magnitude of B3 is at least 0.50.

Aim 2b, moderation by site. Null hypothesis: Site does not moderate the trajectory of behavioral health score by treatment group.

The Chi-squared statistic for the group*time*site interaction term is 0.11 with two degrees of freedom.

Threshold for statistical significance was 0.05.

Considering a range of effect sizes for B1, B2, and B3, we estimated the outcome Y as Y=B1*X+B2*M+B3*X*M+ε, where ε~N(0,1). Using Monte Carlo methods, we averaged across scenarios to determine the typical effect of B3 that can be detected with our current and projected sample sizes after 25% attrition. With our projected sample sizes we expect 0.80 power to detect a significant interaction when the magnitude of B3 is at least 0.50.

Aim 2b, moderation by eHealth literacy. Null hypothesis: eHealth literacy does not moderate the trajectory of behavioral health score by treatment group.

The Chi-squared statistic for the group*time*eHealth literacy interaction term is 3.23 with two degrees of freedom.

Threshold for statistical significance was 0.05

Considering a range of effect sizes for B1, B2, and B3, we estimated the outcome Y as Y=B1*X+B2*M+B3*X*M+ε, where ε~N(0,1). Using Monte Carlo methods, we averaged across scenarios to determine the typical effect of B3 that can be detected with our current and projected sample sizes after 25% attrition. With our projected sample sizes we expect 0.80 power to detect a significant interaction when the magnitude of B3 is at least 0.50.

Null hypothesis: The trajectory of ED visits does not differ by treatment group.

For ED visits, the Chi-squared statistic for the group*time interaction term is 0.82 with two degrees of freedom.

The threshold for statistical significance was .05.

We calculated statistical power for the target sample size of 675 (~337 in each of TEAM and TECH arms). After adjusting for the multiple testing in Aim 1 (2 tests) and considering 25% attrition, a sample size of 505 (~252 per treatment group) will provide over 99.9% power for the hypothesis test for Aim 1.