Acceptability and Feasibility of a Family-Inclusive, Community-Based Lifestyle and Mindfulness Program for Latinx Adolescents at Risk for Type 2 Diabetes: A Two-Arm Pilot and Feasibility Randomized Controlled Trial

Abstract

Adolescent-onset type 2 diabetes (T2D) disproportionally affects Latinx communities. Prevention is, therefore, critical. Unfortunately, preventative T2D interventions remain limited and do not address key mental health concerns that disparately affect youth of Latinx backgrounds. We evaluated the acceptability and piloted the feasibility of a community-based, family-inclusive lifestyle + mental health program (Health Without Barriers [HWB]) compared to a lifestyle program alone (Healthy Living Program) among Latinx adolescents and their families. We conducted a randomized controlled pilot and feasibility trial with follow-up assessments at 6 weeks and 6 months. Enrolled families (N = 18) participated in either HWB or Healthy Living Program. Both programs involved 12 sessions of lifestyle programming for 6 weeks. Only HWB had a mental health component targeting stress and depression symptoms via mindfulness practices. Study assessments included quantitative, qualitative, and clinical data. Primary outcomes were program acceptability and feasibility. Exploratory outcomes included within-group patterns of change in behavioral and metabolic/physiologic indices. Acceptability was high for parents and mixed for adolescents, particularly with regard to the mental health component of HWB. Feasibility was strong for program implementation and mixed for long-term retention. Exploratory within-group patterns of change varied by program. Findings demonstrate preliminary feasibility of implementing an integrative lifestyle + mental health program among Latinx families, with very positive acceptability among parents and mixed acceptability among youth. Future adaptations to improve acceptability of mental health programming and access to more resources to strengthen retention rates are critical next steps in integrative approaches for T2D prevention among Latinx youth.

Type 2 diabetes (T2D) is a serious public health concern associated with high risk of health complications including earlier mortality (Asif, 2014). Commonly considered a disease of adulthood, the prevalence of youth-onset T2D is increasing rapidly (Wagenknecht et al., 2023). Unfortunately, youth-onset T2D appears to be associated with a more aggressive disease course as compared to adult-onset T2D (Nadeau et al., 2016). One of the primary risk factors for developing T2D in childhood is obesity (Kuczmarski et al., 2000), which disproportionally affects Latinx youth (Ogden et al., 2018). Latinx youth in the United States not only demonstrate higher prevalence of obesity compared to White youth (Ogden et al., 2018) but also higher levels of insulin resistance (Diaz-Thomas et al., 2023), a key physiological antecedent to T2D (Kelsey & Zeitler, 2016). Partially driven by disparities in these established risk factors, youth-onset T2D affects those from Latinx communities at significantly higher rates (Diaz-Thomas et al., 2023). About 20% of the U.S. population identified as “Hispanic or Latino” in the 2020 U.S. Census, representing a 23% increase over the last decade (U.S. Census Bureau, 2020). This growing representation of Latinx communities in the United States forecasts that T2D health disparities will persist and continue to affect Latinx youth at increasingly higher rates. Unfortunately, effective treatment and preventative interventions for T2D in childhood remain limited (Hannon & Arslanian, 2015), particularly for those from historically underserved backgrounds (McCurley et al., 2017).

Standard of care and clinical practice guidelines for T2D prevention center on lifestyle behavioral modifications, including changes in diet and physical activity to reduce excess weight and consequently decrease clinical precursors (e.g., insulin resistance) associated with T2D risk (Hampl et al., 2023). Extant behavioral approaches that target T2D prevention in adolescents, however, have shown mixed results, particularly among adolescents from minoritized identities (McCurley et al., 2017). While there are many complex and multilayered contributing factors driving these disparities (e.g., lack of cultural relevance in interventions, sociostructural obstacles, underrepresentation in clinical trials), unaddressed mental health concerns may be one modifiable barrier that negatively impacts treatment engagement and responsiveness among Latinx youth.

Latinx youth are at increased risk of experiencing internalizing symptoms (e.g., depression, stress, anxiety) compared to youth from other ethnic and racial backgrounds (Anderson & Mayes, 2010; J. S. Brown et al., 2007; Mikolajczyk et al., 2007; Varela et al., 2008). Mental health concerns have been identified as a key barrier to engaging in many of the lifestyle behaviors that are core to T2D prevention efforts (Alberga et al., 2019; Bruggink et al., 2019; N. Sanchez et al., 2022). Negative emotions, for example, have been described by Latinx adolescents and their families as an important contributor to overeating patterns, suggesting that obstacles to mental health in this group can challenge engagement in dietary recommendations for T2D prevention (N. Sanchez et al., 2022). Further, difficulties with mental health may affect health outcomes not only via behavioral pathways but also through their influence on key physiologic and metabolic pathways that contribute to T2D risk. Symptoms of depression, for example, have been associated with both higher insulin resistance and, also, worsening insulin resistance and T2D onset over time (Joseph & Golden, 2017; Shomaker & Goodman, 2015; Shomaker et al., 2011; Suglia et al., 2016). Together, extant evidence demonstrates the bio-behavioral influence that depression symptoms have on T2D risk. Unfortunately, mental health is not typically addressed in lifestyle and T2D prevention interventions, highlighting the urgent need to build upon existing programs to address the multilayered factors that contribute to T2D risk in Latinx youth.

One such program specifically developed for Latinx youth with overweight/obesity is the Healthy Living Program (HeLP) or La Vida Saludable (Haemer et al., 2023). HeLP is a family-inclusive lifestyle intervention that meets the 2023 American Academy of Pediatrics clinical practice guidelines for intensive health behavior and lifestyle treatment (Hampl et al., 2023) and was designed around the concept of familism, the value placed on strong family ties and the interconnected of family members (Hernández & Bámaca-Colbert, 2016). The original curriculum of HeLP, however, did not include a focus on adolescent mental health. To address this gap, our team developed Health Without Barriers (HWB), or Salud Sin Barreras, a modified version of HeLP that retains most of its content and adds an adolescent mental health component to its programming (see Methods for details). HWB was developed with the intent to offer distinctive benefits for Latinx adolescents at risk for T2D, as we anticipate that targeting mental health in combination with lifestyle will lead to greater behavioral (e.g., more physical activity) and physiological (e.g., lower insulin resistance) benefits compared to a lifestyle intervention alone. Yet, before a comparative efficacy trial can be completed to test this hypothesis, acceptability of HWB needs to be assessed, and the feasibility of conducting a randomized trial comparing a lifestyle + mental health program versus a lifestyle program alone must be piloted.

Thus, the primary aims of this randomized controlled pilot and feasibility trial were (a) to evaluate the acceptability of a lifestyle + mental health program (HWB) relative to a lifestyle program alone (HeLP) among Latinx adolescents and their families and (b) to pilot the preliminary feasibility of conducting a two-arm randomized controlled trial comparing HWB versus HeLP in Latinx communities of the Western United States. We anticipated that acceptability ratings for HWB would be noninferior to acceptability ratings for HeLP and that running a randomized pilot trial with two arms would be feasible. An additional exploratory aim of this pilot was to describe preliminary within-group (HWB or HeLP) changes from baseline to 6-weeks and from baseline to 6-months posttreatment in psychosocial functioning and clinical/physiological variables.

Method

Trial Design

This study was a two parallel-group randomized controlled pilot and feasibility trial (NCT03527641) conducted in a metropolitan area of Northern Colorado from May 2018 to March 2019. The allocation ratio was 1:1, with no stratification or block randomization. Study procedures were approved by the Institutional Review Board of Colorado State University.

Participant Eligibility, Recruitment, and Screening

Eligible participants included males and females who (a) self-identified as Latinx; (b) were 12–15 years; (c) had overweight/obesity (body mass index [BMI] ≥ 85th percentile for age and sex); (d) had a family history of T2D, prediabetes, or gestational diabetes; (e) were in good general health as assessed by a pediatrician; and (f) had the ability to speak in English to participate in adolescent-group sessions. Parent sessions were facilitated in English and Spanish to support participation for monolingual Spanish-speaking parents. Exclusionary criteria were (a) parent-reported adolescent major psychiatric disorder (i.e., pervasive developmental disorder, bipolar, conduct, anorexia, bulimia, panic, obsessive/compulsive, psychosis, schizophrenia, and substance use disorders), (b) parent-reported adolescent major medical problem, (c) self-reported medication use impacting weight or mood if started <3 months before screening, and (d) self-reported pregnancy in females.

For recruitment efforts, the research team partnered with a local federally qualified health center (FQHC) serving Latinx families and identified potential participants through three main strategies selected with input from a previously developed community advisory board involved in guiding recruitment and retention efforts (N. Sanchez et al., 2022): (a) direct referrals from promotoras (i.e., community health workers), (b) bilingual/bicultural research staff calls to patients identified by health care staff via chart review from community partner FQHC, and (c) recruitment by bilingual/bicultural research staff at Latinx community events. Phone screenings were used to determine initial eligibility. Potentially eligible families were invited to participate in two additional in-person screening visits. One screening took place at a local community organization, where staff completed the informed consent/assent process, confirmed eligibility based on BMI ≥ 85th percentile and T2D family history, and facilitated completion of adolescent and parent surveys. The next screening/baseline appointment occurred at a partner FQHC where a physician completed a physical exam and a phlebotomist performed a fasted-state blood draw.

Randomization

There were two sets of dates on which the program was offered (i.e., Monday/Wednesday or Tuesday/Thursday). Only one group condition was available on each of those dates. Study randomization was conducted by a lead research coordinator and involved randomizing study condition to one of the two schedules offered. Families were then asked to identify which of the two schedules they could attend without awareness of the condition.

Interventions

Lifestyle Only Program

HeLP is a family-inclusive (i.e., all siblings, parents, and other caregivers are welcome to attend), culturally responsive, and community-based lifestyle intervention developed through community-engaged partnerships with Latinx communities for Latinx youth with obesity/overweight living in low-income, urban areas of the Western United States (Haemer et al., 2023). HeLP was developed with a particular attention to preschool and elementary age children and meets the 2023 American Academy of Pediatrics clinical practice guidelines for intensive health behavior and lifestyle treatment (Hampl et al., 2023). Initial evaluations of HeLP suggest the program is feasible, cost-effective, and has shown preliminary signals for decreasing obesity (Cardel et al., 2018; Cason-Wilkerson et al., 2022; Haemer et al., 2022). HeLP is also the subject of an ongoing registered (NCT05041855) randomized controlled trial (Haemer et al., 2023).

During the HeLP program, adolescents and their families participate in a 6-week, 12-session lifestyle intervention consisting of (a) a weekly 2-hr multidisciplinary session and (b) a weekly 2-hr cooking and nutrition session led by staff trained in the Cooking Matters curriculum (Supplemental Figure 1). Cooking Matters is a national campaign that seeks to promote a healthier lifestyle and works with communities to address hunger and/or food insecurity. Its curriculum was developed for families living on a low income and/or facing food insecurity and was delivered by bilingual facilitators, including opportunities to consume affordable and culturally appropriate recipes as a family (Shrodes et al., 2023; A. Williams et al., 2023).

During the six multidisciplinary sessions, adolescents gathered without parents and participated in group physical fitness activities (e.g., moderate-to-vigorous physical activity [MVPA], resistance training, team sports) delivered by a certified athletic trainer. Afterward, adolescents participated in a 1-hr adolescent-oriented health education program delivered by a trained graduate student and a Latinx bilingual/bicultural research assistant. The health education program was based on the middle and high-school “Hey Durham” manualized curriculum (Bravender, 2005). Parents received a 1-hr psychoeducation session that covered various health topics, parenting guidance, and goal setting around health goals. Parent sessions were cofacilitated by a Latinx bilingual/bicultural promotora and a trained Latinx bilingual/bicultural graduate student.

Lifestyle and Mental Health Program

HWB is a modified version of HeLP that adds a mental health component and specifically focuses on adolescents instead of preschool and elementary children. HWB addresses adolescent stress and depression symptoms via empirically supported mindfulness practices (Dunning et al., 2019). Paralleling U.S. national trends, community needs assessments of adolescent mental health concerns have been mounting throughout Colorado since before and escalating even further after the pandemic (Colorado Department of Public Health and Environment, 2023). Within this context, decisions on creation of HWB were informed by a 9-month community translation process with key stakeholders (Nease et al., 2020; Norman et al., 2013), in combination with focus groups conducted with Latinx local-area families (N. Sanchez et al., 2022). Adolescents in HWB received six 1-hr mindfulness-based intervention group sessions of equivalent length and in place of health education. Sessions were delivered by the graduate student and Latinx research assistant who also delivered HeLP. The mindfulness program was derived from “Learning to BREATHE (L2B),” a socioemotional and developmentally appropriate learning curriculum that is based on mindfulness-based stress reduction and seeks to teach adolescents stress management and mindfulness skills (Broderick, 2021). L2B has demonstrated acceptability for adolescents and has been feasibly delivered among teenagers at risk for excess weight gain and T2D (Shomaker et al., 2019), as well as among ethnically diverse and marginalized youth and families (Tobin et al., 2021). Among Latinx adolescents specifically, empirical evidence supports the efficacy of mindfulness-based programs for reducing depression symptoms and stress (Clarke et al., 2022; Edwards et al., 2014; B. Sanchez et al., 2023). In L2B, activities include body scans, breath awareness, mindful eating, guided meditation, and mindful movement. All adolescent mindfulness group sessions were recorded and reviewed by a licensed clinical psychologist trained in the L2B program. There were three primary modifications to the L2B curriculum included in HWB. First, a brief explanation about the potential benefit of mindfulness for health and reductions in T2D risk was added to Session 1. Second, vignettes were modified to enhance relevance to Latinx adolescents. Lastly, although the focus was to integrate mindfulness-based training into the adolescent curriculum, parents also received a 30-min version of L2B mindfulness training to learn stress management strategies and ways to support adolescent mindfulness practice.

Facilitator Training and Supervision

Facilitator trainings for both interventions were provided through directed readings, discussion, and observation of other facilitators leading group sessions. The certified physical trainer and the facilitator leading the mindfulness portion of HWB both had expertise in these areas. A licensed clinical psychologist provided regular supervision.

Data Collection and Settings

Baseline, posttreatment, and 6-month follow-up assessments were conducted at a local family center or FQHC largely serving Latinx families. HeLP and HWB were delivered from June 2018 to October 2018 at a local recreation center and community organization serving Latinx families. Families received transportation and phone/text reminders support to enhance attendance. Participants were compensated for their participation (up to $200).

Sample Size

The original proposed sample size for this study was estimated at N = 40 based upon empirical recommendations for pilot and feasibility trials expected to have small-to-medium effects on outcomes (Whitehead et al., 2016). The final sample size was N = 18 (HWB n = 11, HeLP n = 7). Recruitment details are shown in the CONSORT flow diagram (Supplemental Figure 2).

Measures

Prior to data collection, families were pre-screened and asked to report on T2D family history, adolescent age, sex, height, and weight to estimate BMI percentile. All questionnaires were available in English and Spanish. When a measure was not available in Spanish from the developers, a professional certified translator was hired to translate the measure.

Baseline Sociocontextual and Clinical Characteristics

Participants reported on various sociocultural and environmental characteristics to contextualize results, including (a) race and ethnicity; (b) acculturation, measured with the four-item Brief Acculturation Scale for Hispanics (Norris et al., 1996); (c) perceived discrimination, assessed with the 10-item Everyday Discrimination Scale (D. R. Williams et al., 1997); and (d) home environmental chaos, measured with the 15-item Confusion, Hubbub, and Order Scale (Matheny et al., 1995).

Caregivers also completed a brief questionnaire to report on adolescent medical and family history. Additional clinical characteristics, including BMI (kg/m²), BMI z score for age and sex, and BMI percentile, were collected as part of study screening procedures. Body measurements were collected in the morning after an overnight fast by a medical assistant. A stadiometer was used to measure height, and a calibrated scale was used to measure weight.

Primary Outcomes

Acceptability.

Program acceptability was assessed using an adapted version of the Treatment Acceptability Questionnaire (Hunsley, 1992). Items are rated on a 5-point Likert scale ranging from 1 to 5 (not at all to definitely/extremely), with higher scores demonstrating greater acceptability. Participants were also given the opportunity to provide open-ended feedback about the program. Acceptability was operationalized as ratings of “4–very” or “5–extremely.” Thus, high acceptability ratings (i.e., 4 and 5 scores) were combined into a “high acceptability” category, and low acceptability ratings (i.e., 1–3 scores) were recoded into a “low acceptability” category. Program acceptability benchmarks were set at ≥80%.

Feasibility.

Feasibility of implementation was operationalized as (a) a median session attendance of ≥80% and (b) a proportion of facilitators trained and retained ≥80%. The benchmark for retention (i.e., the proportion of participants who completed study procedures through postassessment) was ≥70%. Lastly, the feasibility of data collection was operationalized as <5% of missing data for surveys and <20% of device wear.

Exploratory Outcomes

Depression Symptoms.

The 20-item Center for Epidemiologic Studies-Depression Scale (CES-D) was used to assess depression symptoms. The CES-D is a validated measure with adequate psychometric qualities in Latinx adolescents (Skriner & Chu, 2014; Stockings et al., 2015). This scale measured how often over the past week adolescents and their caregivers experienced depression symptoms. The total score ranges from 0 to 60, with higher scores reflecting more depressive symptoms. Scores > 20 are indicative of potentially clinically elevated depression. Scale consistency reliability was α = .93 for adolescents and α = .73 for caregivers.

Stress.

To measure stress, two surveys were used. The Perceived Stress Scale is a 14-item measure assessing the degree to which various situations over the past month are appraised as stressful (Kupst et al., 2015). Total scores range from 0 to 40, with higher values reflecting greater levels of perceived stress. Internal consistency reliability was α = .80 for adolescents and α = .62 for parents. The Patient-Reported Outcomes Measurement Information System Psychological Stress Experiences Short Form is an eight-item measure assessing psychological stress (Bevans et al., 2018). The total score for this measure ranges from 8 to 40, which is then transformed into T scores. Higher T scores represent greater psychological stress. Only adolescents completed this measure (α = .96).

Dispositional Mindfulness.

The 15-item Mindful Attention Awareness Scale (MAAS) was used to assess adolescents’ propensity for present-moment attention (K. W. Brown & Ryan, 2003). The MAAS total score ranges from 15 to 90, and higher scores reflect greater mindfulness. Internal consistency reliability for the MAAS in this study was α = .95.

Physical Activity and Sedentary Behavior.

Adolescent physical activity and sedentary behavior were measured by lightweight accelerometer ActiGraph GT3X worn on the nondominant wrist. Adolescents wore the accelerometer 24 hr a day, for 7 consecutive days except for immersive water-related activities. Actigraph data yielded three total values for the 7-day wear period, including (a) total number of average minutes per day spent in sedentary activity over the 7 days, (b) average total number of daily minutes spent in MVPA, and (c) average percentage of daily time spent in MVPA.

Insulin Resistance and Metabolic Health Outcomes.

Following an overnight fast, a blood draw was conducted for adolescents at the health center. Laboratory values were evaluated by a physician at the participating community health center to assess fasting glucose, fasting insulin, hemoglobin A1c, alanine transaminase/aspartate aminotransferase, high-density lipoprotein/low-density lipoprotein, total cholesterol, and triglycerides. To measure insulin resistance, the homeostasis model assessment of insulin resistance was used and calculated as (fasting insulin [μ U/ML] × fasting glucose [mmol/L])/22 (Matthews et al., 1985).

Salivary Cortisol.

The change in cortisol levels before and after sessions was used as a marker of stress. Immediately before and after HeLP or HWB, adolescents provided a salivary sample with an oral sublingual swab held under the tongue for 120 s. Salivary samples were stored at −20C and assayed at the University of Trier. Samples were centrifuged for 10 min and cortisol levels established by the time-resolved fluorescence immunoassay method.

Data Analytic Plan

Given the pilot feasibility nature of the study, all analyses are descriptive in nature. Summary statistics were calculated to evaluate acceptability and feasibility. Group medians and interquartile ranges were presented at each time point, including baseline, 6-week, and 6-month postintervention to describe pre-to-post within-condition changes within HWB and HeLP intervention groups. For cortisol data, before and after session values are presented. Missing data by group and visit were reported for each measure. Participants with recorded visits but incomplete data were included in the analysis due to the smaller sample size of the study and descriptive nature of the analyses. No imputation methods were used given the pilot nature of the sample. Qualitative data were collected with the adapted Treatment Acceptability Questionnaire.

Results

A total of N = 18 families consented to participate in the study. Enrollment rate was n = 9–10 per month of eligible, randomized families, but recruitment had to be discontinued prematurely due to limited resources available to cover facilitator fees, research coordination support, and supplies beyond the 1-year pilot grant period. Eleven adolescents attended HWB, and seven adolescents attended HeLP. Seven of the adolescent–caregiver dyads had additional family members attend the program, all of whom were siblings (n = 11). Baseline sociodemographic and clinical characteristics are presented in Table 1. Mean age and sex distribution were comparable across HWB and HeLP. Across conditions, adolescents experienced similar levels of everyday discrimination, and about two thirds had high levels of acculturation. In general, more than a third of adolescents reported moderately elevated depression symptoms (CES-D > 20) at baseline.

Table 1.

Baseline Demographic and Clinical Characteristics

Variable name	Health Without Barriers Salud Sin Barreras n = 11 M ± SD or % (n)	Healthy Living Program La Vida Saludable n=7 M ± SD or % (n)
Age, years	13.59 ± 1.04	13.68 ± 1.21
Female	54% (6)	71% (5)
Race/Ethnicity
White	36% (4)	0% (0)
Black or African American	9% (1)	0% (0)
American Indian or Alaska Native	18% (2)	0% (0)
Hispanic, Latino or Spanish	100% (11)	100% (7)
Mexican/Mexican American	91% (10)	86% (6)
Missing/chose not to report	9% (1)	14% (1)
BMI, kg/m2	32.66 ± 6.23	31.32 ± 3.63
BMI, z score	2.27 ± 0.54	2.10 ± 0.33
BMI, percentile	97.65 ± 2.32	97.79 ± 1.31
Obesity, BMI ≥ 95th percentile	82% (9)	100% (7)
Fasting insulin, mIU/L	25.45 ± 13.25	22.47 ± 10.47
Fasting glucose, mg/dL	90.55 ± 6.22	89.71 ± 7.76
Insulin resistance, HOMAR-IR	5.51 ± 2.74	4.90 ± 2.09
HbA1c, %	5.08 ± 0.18	5.09 ± 0.18
No. First-/second-degree family member(s) with diabetes	3.00 ± 1.48	3.43 ± 1.72
% with first-degree family member(s) with diabetes	46% (5)	43% (3)
Elevated depression symptoms (CES-D > 20)	36% (4)	43% (3)
Acculturationa	16.00 ± 4.11	13.71 ± 2.50
High (average score > 3)	64%	71%
Low (average score ≤ 3)	36%	29%
Everyday discriminationb	41.18 ± 10.50	42.71 ± 9.74
Home chaosc	11.67 ± 5.57	6.86 ± 3.53

Note. BMI = body mass index; HOMAR-IR = homeostasis model assessment of insulin resistance; HbA1c = hemoglobin A1c; CES-D = Center for Epidemiologic Studies-Depression Scale.

^aAcculturation was assessed with the Brief Acculturation Scale for Hispanics. Averaged total scores range from 1 to 5 and are used to calculate a dichotomous level of acculturation.

^bEveryday discrimination was assessed using the Everyday Discrimination Scale. Total scores range from 10 to 54, with higher scores indicating lower levels of everyday discrimination.

^cHome environment was assessed using the confusion, hubbub, and order scale. Total scores range from 0 to 15. Higher scores indicate higher levels of environmental chaos in the home.

Acceptability

In both HWB and HeLP, 75% and 100% of the adolescents, respectively, reported that the program would help other adolescents (Table 2). The majority (≥88%) of adolescents in both conditions perceived facilitators as “helpful/supportive” and liked the physical fitness component of the program. All adolescents in the HWB group and 80% in the HeLP group thought their mood improved because of the program, and 75%–80% in each condition believed their stress improved.

Table 2.

Primary Study Outcomes: Caregiver and Adolescent Intervention Acceptability Ratings

Acceptability domain	Health Without Barriers Salud Sin Barreras		Healthy Living Program La Vida Saludable
	Caregiver (%)	Adolescent (%)	Caregiver (%)	Adolescent (%)
Participant comfort	100	38	100	80
Leaders supportive	100	88	100	100
Health improved	100	63	100	80
Mood improved	100	100	100	80
Stress improved	71	75	80	80
Recommend to other parents	100		80
Help other teens like me		75		100
Liked recipes	100	42	100	80
Liked exercise		88		100
Liked mindfulness	100	42

Note. % indicates the percentage of parents and teens who rated acceptability of the program on a 1–5 scale (5 best) as very high (4) or extremely high (5).

The level of comfort teens experienced with HWB, however, was low (38%) as compared to HeLP (80%), and the acceptability of mindfulness activities was also low (42%) in HWB. Further, perceived improvements in overall health were lower in HWB (63%) compared to HeLP (80%). Adolescents in HWB reported low likability ratings of the recipes provided (42%), as compared to adolescents in HeLP (80%), with both groups getting the same recipes. With regard to program delivery location, most adolescents (78%; n = 7) who answered this question endorsed liking where the program was held. No adolescent shared open-ended written feedback.

Parents’ acceptability ratings for both interventions were high, with 100% of parents reporting that they felt “very” or “extremely” comfortable during the program, perceived leaders as “helpful,” and believed their health and mood “improved.” Most parents in both interventions reported they would recommend the program, and 100% reported liking the recipes provided. Among parents in HWB, 100% reported liking mindfulness. Every parent (100%; n = 12) who answered the question about program location endorsed liking where the program was held. All parents who shared written feedback (n = 8) had positive comments about the interventions.

Feasibility

Program Implementation and Retention

Regarding feasibility of program implementation, attendance for HWB (n = 11) was high (≥10 of 12 sessions) for n = 7 families and low (≥3 of 12 sessions) for n = 2 families. Two of the enrolled families withdrew before the program started and completed no sessions after randomization. The median session attendance was 10 of 12 sessions (83%). Using the intent-to-treat principle, retention was 72% and 63% at the 6-week and 6-month follow-up, respectively.

For HeLP (n = 7), attendance was high (≥10 of 12 sessions) for n = 5 families, and moderate (≥8 of 12 sessions) for n = 1. One of the enrolled families completed only the first session after randomization. The median session attendance was 11 of 12 sessions (92%). Accounting for all randomized families, retention was 75% and 71% at the 6-week and 6-month follow-up, respectively. There was 100% retention for facilitators in HWB and HeLP.

Data Collection

With regard to feasibility of data collection, protocol compliance and assessment completion data are presented in Table 3. For exploratory outcomes, there was minimal missing data (<1%). For follow-up data collection, 73% (6-week) and 64% (6-month) of participants in HWB and 71% (6-week) and 71% (6-month) of participants in HeLP completed at least some follow-up assessments. Adolescent device wear was 60% in HWB and 71% in HeLP, after accounting for randomized participants, including those who dropped out.

Table 3.

Psychosocial and Behavioral Exploratory Outcomes by Intervention Group Across Study Time Points

Measure	Health Without Barriers Salud Sin Barr eras						Healthy Living Program La Vida Saludable
	Baseline		6-weeks		6-months		Baseline		6-weeks		6-months
	n	Score	n	Score	n	Score	n	Score	n	Score	n	Score
Adolescent psychosocial and behavioral outcomes
Depression symptoms	11	14 [9.5, 19]	8	8.5 [4.5, 9]	7	11 [6, 12]	7	9 [7.24]	5	9 [8, 16]	5	13 [7, 16]
Perceived stress	11	28.0 [25.5, 32.0]	8	28.5 [26.5, 29.0]	6	24.5 [18.8, 28.8]	7	28.0 [20.5, 29.0]		27.0 [22.0, 28.0]	5	28.0 [19.0, 32.0]
Psychological stress	9	50.3 [45.2, 55.0]	8	52.2 [48.5, 54.3]	5	51.6 [50.3, 55.0]	7	50.3 [43.0, 57.9]	4	48.9 [37.0, 55.0]	5	57.1 [55.0, 57.1]
Mindfulness	11	64.0 [53.0, 69.0]	8	69.0 [57.5, 70.5]	5	71.0 [61.0, 77.0]	7	59.0 [49.0, 79.5]	5	76.0 [65.0, 84.0]	5	81.0 [53.0, 81.0]
Sedentary minutes/day	10	1680 [1390, 2210]	6	1630 [1620, 1730]	6	1410 [1260, 2160]	7	1660 [1370, 2100]	5	1760 [1620, 2280]	4	1700 [1610, 1840]
Total MVPA	10	941 [842, 1130]	6	855 [674, 1090]	6	456 [338, 642]	7	857 [760, 1300]	5	671 [662, 1090]	4	556 [371, 723]
% time spent in MVPA	10	15.4 [9.27, 16.8]	6	14.8 [10.4, 17.2]	6	8.33 [4.43, 11.1]	7	14.2 [11.0, 21.2]	5	14.8 [8.41, 17.2]	5	10.9 [8.22, 12.8]
Caregiver psychosocial and behavioral outcomes
Depression symptoms	11	12 [9, 17]	8	8.50 [4.5, 12.5]	7	11 [10.36, 15.5]	7	14 [9.50, 16]	5	5 [0, 8]	5	3 [2, 9]
Perceived stress	6	15 [13, 22]	5	15 [11.8, 17.5]	4	14.5 [14, 18.8]	6	21.0 [20, 22]	4	18.0 [6.00, 18.0]	5	19.0 [16.0, 19.0]

Note. Statistics in bold indicate median scores and CI in brackets at 6-weeks and 6-months postintervention that were improved compared to values at baseline. MVPA = moderateto-vigorous physical activity; % time = percentage of time spent in MVPA; CI = confidence interval.

Description of Within-Group Pre-to-Post Changes in Outcomes

Psychosocial and Behavioral Outcomes

In the HWB program, adolescents reported reductions in depression symptoms, perceived stress, and showed less daily sedentary time via accelerometry from baseline to follow-up, while mindfulness increased during the same period. No within-group changes in psychological stress or total time spent in MVPA were reported in HWB. Caregivers in this program also experienced decreases in depression symptoms and perceived stress from baseline to follow-up (Table 3).

Adolescents in HeLP experienced decreases in perceived stress and psychological stress at 6-weeks posttreatment that were not sustained at the 6-month follow-up. They also had an increase in mindfulness at the 6-week and 6-month follow-ups. No changes were observed in depression symptoms, sedentary time, or total time spent in MVPA. Caregivers in HeLP reported reductions in depression symptoms and stress at both 6-week and 6-month follow-ups (Table 3).

Metabolic and Physiological Outcomes

Generally, there was a trend toward decreased insulin resistance in both HWB and HeLP (Table 4). A similar trend was present for various secondary metabolic outcomes. Table 5 shows before and after session cortisol levels for adolescents in both groups. Overall, there was a pattern of decreased cortisol levels across conditions, with adolescents in HeLP experiencing drops in cortisol after every session.

Table 4.

Clinical and Metabolic Exploratory Outcomes by Intervention Group Across All Study Time Points

	Health Without Barriers Salud Sin Barr eras						Healthy Living Program La Vida Saludable
	Baseline		6-weeks		6-months		Baseline		6-weeks		6-months
Measure	n	Score	n	Score	n	Score	n	Score	n	Score	n	Score
Clinical and metabolic outcomes
HOMAR-IR	11	4.82 [3.85, 6.28]	8	3.64 [2.70, 4.68]	7	4.97 [2.36, 6.11]	7	5.13 [3.80, 5.51]	5	4.53 [3.91, 5.54]	5	3.73 [3.46, 5.06]
Insulin, mlU/L	11	22.0 [18.2, 28.8]	8	17.8 [12.1, 23.7]	7	21.4 [11.4, 28.7]	7	21.0 [16.7, 26.7]	5	21.1 [17.2, 22.0]	5	16.4 [14.9, 25.0]
Glucose, mg/dL	11	92.0 [88.5, 94.5]	8	85.0 [81.5, 91.5]	7	88.0 [86.5, 94.0]	7	92.0 [85.0, 95.0]	5	90.0 [87.0, 92.0]	5	92.0 [85.0, 94.0]
HbA1c, %	11	5.10 [5.00, 5.30]	8	5.00 [4.90, 5.15]	7	5.10 [5.00, 5.25]	7	5.10 [5.00, 5.25]	5	5.00 [5.00, 5.20]	5	5.20 [5.20, 5.30]
ALT, U/L	11	17.0 [14.0, 43.0]	8	17.0 [13.8, 22.5]	7	16.0 [12.5, 19.5]	7	25.0 [18.5, 29.0]	5	23.0 [15.0, 23.0]	5	19.0 [14.0, 31.0]
AST, U/L	11	19.0 [16.0, 27.5]	8	17.5 [14.8, 21.0]	7	15.0 [13.5, 20.5]	7	19.0 [17.0, 22.5]	5	19.0 [16.0, 27.5]	5	17.0 [14.0, 20.0]
Triglycerides, mg/dL	11	115 [95.5, 168]	7	136 [93.5, 184]	7	80.0 [63.5, 145]	7	135 [93.5, 146]	5	111 [93.0, 148]	5	108 [89.0, 153]
Cholesterol, mg/dL	11	158 [149, 166]	7	148 [141, 161]	7	136 [122, 160]	7	167 [122, 185]	5	139 [121, 172]	5	141 [129, 164]
HDL, mg/dL	11	43.0 [37.0, 49.5]	7	46.0 [37.5, 55.0]	7	45.0 [39.0, 55.5]	7	38.0 [32.0, 42.0]	5	37.0 [33.0, 38.0]	5	37.0 [37.0, 38.0]
LDL, mg/dL	11	95.0 [79.0, 99.5]	7	76.0 [68.0, 109]	7	79.0 [72.0, 103]	7	102 [64.0, 118]	5	76.0 [68.0, 109]	5	79.0 [72.0, 103]

Note. Statistics in bold indicate median scores at 6-weeks and 6-months postintervention that were improved compared to values at baseline. HOMAR-IR = homeostasis model assessment of insulin resistance, with higher values indicating greater insulin resistance; HbA1c = hemoglobin A1c; ALT = alanine transaminase; AST = aspartate aminotransferase; HDL = high-density lipoprotein; LDL = low-density lipoprotein.

Table 5.

Salivary Cortisol Exploratory Outcomes Before and After Each Multidisciplinary Session

	Health Without Barriers Salud Sin Barreras n = 8		Healthy Living Program La Vida Saludable n = 5
Session number	Before	After	Before	After
Session 1
Median [Q1, Q3]	2.29 [1.15, 2.57]	1.40 [0.866, 1.69]	2.29 [1.86, 5.38]	1.15 [0.855, 1.61]
Missing	2 (25.0%)	2 (25.0%)	0 (0%)	0 (0%)
Session 2
Median [Q1, Q3]	2.43 [2.24, 2.50]	1.27 [0.788, 1.72]	1.77 [1.48, 1.94]	0.685 [0.565, 0.720]
Missing	3 (37.5%)	2 (25.0%)	2 (40.0%)	2 (40.0%)
Session 3
Median [Q1, Q3]	2.05 [1.32, 2.64]	1.67 [0.960, 2.53]	2.79 [2.43, 3.47]	1.26 [1.15, 1.58]
Missing	1 (12.5%)	1 (12.5%)	0 (0%)	0 (0%)
Session 4
Median [Q1, Q3]	1.90 [1.33, 2.29]	1.38 [0.578, 2.10]	2.38 [2.09, 2.69]	1.54 [1.47, 3.39]
Missing	1 (12.5%)	1 (12.5%)	0 (0%)	0 (0%)
Session 5
Median [Q1, Q3]	1.68 [1.22, 4.15]	2.09 [0.710, 2.97]	2.50 [1.83, 2.71]	1.23 [0.915, 1.53]
Missing	2 (25.0%)	2 (25.0%)	0 (0%)	0 (0%)
Session 6
Median [Q1, Q3]	2.04 [1.41, 3.29]	2.15 [1.08, 5.01]	2.33 [1.72, 2.98]	1.11 [0.936, 1.23]
Missing	1 (12.5%)	1 (12.5%)	1 (20.0%)	1 (20.0%)

Note. Statistics in bold indicate median scores before the session that improved compared to values after the session.

Discussion

Latinx youth are disproportionally affected by adolescent-onset T2D, psychological stress, and associated risk factors (Diaz-Thomas et al., 2023; Goran et al., 2002; Morrison et al., 2010). Youth in our sample had a strong family history of diabetes, very elevated insulin resistance, and a very high percentage of them endorsed elevated depression symptoms(HWB = 36%; HeLP = 43%), with rates comparable to recent reports of global prevalence rates (34%) of elevated depression symptoms among adolescents (Shorey et al., 2022). Together, these clinical characteristics highlight the strong need for T2D preventative interventions that target both lifestyle and mental health among Latinx youth.

Overall, adolescent program acceptability for HWB was mixed compared to HeLP. On one hand, the majority of adolescents in HWB and HeLP perceived group leaders as supportive and experienced benefits in their mood and stress levels, lending some support to the acceptability of mindfulness-based interventions among adolescents with diverse ethnic backgrounds, high weight, and at risk for T2D (Shomaker et al., 2017, 2019). On the other hand, only 63% of adolescents in HWB reported the program had a positive impact on overall health compared to 80% in HeLP. Adolescents in HWB also endorsed low levels of comfort participating in the program (38% endorsed high comfort), whereas the majority (80%) endorsed high comfort participating in HeLP. Lower acceptability ratings in HWB may stem from the potential discomfort generated by mindfulness activities, given that less than half of adolescents in HWB (42%) reported mindfulness activities met the benchmark for “acceptable.”

Prior work on community-based mindfulness interventions among Latinx youth shows that adolescents may feel uncomfortable or “weird” practicing mindfulness, particularly in group settings (Tobin et al., 2021). Thus, although some degree of discomfort is not unexpected, future tailoring efforts should focus on increasing adolescents’ comfort and acceptability. It is possible that discomfort may partially stem from a missed opportunity to incorporate Latinx values such as personalismo, which emphasizes close interpersonal relationships and may drive comfort levels in group-based interventions. Creating more opportunities throughout the program for social connection (e.g., rapport-building activities) may better align with Latinx cultural values and help teenagers feel more comfortable with and connected to one another. Additionally, consideration of cultural factors that may pose a perceived conflict with mindfulness practices (e.g., faith-based traditions and values) may offer the potential to mitigating some of the discomfort experienced by adolescents (Palitsky & Kaplan, 2021). Engagement and comfort with the program also may have been partially shaped by intersectional factors (e.g., gender, socioeconomic status, sexual orientation), which are yet to be explored in the context of T2D prevention efforts. Importantly, future work remains to be done to evaluate the extent to which acceptability ratings relate to clinical outcomes and whether comfort with mindfulness activities is a prerequisite for creating lasting therapeutic change. It is worth noting that despite mixed ratings on acceptability, 75% of adolescents in HWB believed the program would help other adolescents.

Interestingly, HWB acceptability ratings among caregivers were very high, paralleling the high acceptability ratings for HeLP. All caregivers in both programs felt comfortable participating in the interventions and reported positive benefits with regard to their own health and their mood. All parents in HWB reported liking the mindfulness activities, which were facilitated in English and Spanish by a Latinx promotora, suggesting that this approach to promote mental health was well received. These findings align with prior research on the acceptability and feasibility of community-based mindfulness programs demonstrating that despite initial, potential skepticism, caregivers of Latinx youth often tend to find mindfulness practices enjoyable and acceptable, especially when there is a sense of connection with other caregivers and instructors are perceived as supportive (Tobin et al., 2021). More broadly, these findings are consistent with literature reviews on mindfulness-based interventions in the Latinx community showing moderate-to-high acceptability ratings in this group (Cotter & Jones, 2020).

Study start-up took longer than anticipated, and as a result, enrollment initiation was delayed several months. A 1-year pilot timeline and budget, with restricted carryforward of funds, restricted the research team’s capacity to extend the enrollment period and cover research staffing needs. In a future trial, accounting for a realistic start-up period (6–9 months) will be paramount.

The feasibility of program implementation was strong. Intervention attendance for both interventions was high, with median session attendance exceeding 80% across conditions, indicating that most families who started the program attended most sessions. A very small number (n = 2) of participants withdrew from the study after randomization prior to starting the program, suggesting that once participants enrolled in the study, they were likely to stay involved. The strong feasibility of program implementation for both interventions likely was driven by a combination of factors, including strong logistical support surrounding transportation for families, consistent phone and text reminders from research staff prior to session, and high acceptability of program location, which was selected based on recommendations from the community advisory board.

Short-term (6-weeks) retention of adolescents and caregivers was above the set benchmark of >70% for HWB and HeLP. Long-term (6-months) retention exceeded this benchmark for HeLP, but not for HWB, likely because two (of 11) adolescents withdrew after randomization but prior to program initiation. These rates are generally consistent with retention for medical (e.g., blood drawing) and psychosocial assessments in other pilot and feasibility studies with racially and ethnically diverse youth at risk for T2D (McCurley et al., 2017) and highlight retention as a challenging aspect of conducting randomized controlled trials for T2D prevention. Adolescent compliance for device wear was also a challenge, although comparable or slightly better than compliance rates in other studies with at-risk adolescents (Scott et al., 2017). The limited number of staff available to assist with device-related difficulties likely drove lower compliance rates.

Regarding metabolic outcomes, participants in HWB had decreased insulin resistance at the 6-week postintervention assessment relative to their baseline, although reductions were not sustained at the 6-month follow-up. Overall, there was a pattern of short-term, long-term, or sustained across the range of metabolic markers assessed for HWB. A similar picture emerged for youth in HeLP, with decreased insulin resistance at the 6-week and 6-month follow-up assessments, and a general pattern of short-term, long-term, or sustained reductions across most metabolic markers. For salivary cortisol levels, there was a trend toward postsession cortisol reductions across the majority of sessions for HWB and HeLP. Combined, these descriptive results provide preliminary data showing that metabolic outcomes generally changed in the desired direction throughout the intervention and at postintervention across conditions.

Consistent with prior literature, we observed reductions in adolescent depression symptoms at both the 6-week and 6-month follow-up assessment for those in HWB but not HeLP (Cotter & Jones, 2020), suggesting that the mental health component added to HWB may be of value in addressing depressive symptoms despite adolescents’ low acceptability ratings of mindfulness activities. Further, youth in HWB, but not HeLP, spent less time engaged in sedentary activities, possibly due to improvements in mood, as depression has been linked to physical inactivity (Redig et al., 2022). Together, descriptive results from this study add preliminary support to the body of literature demonstrating the potential benefits of mindfulness training in mitigating T2D risk (Shomaker et al., 2017). A less clear pattern emerged for perceived and psychological stress despite reductions in salivary cortisol following most sessions. It is possible that some of the immediate physiological benefits of the programs on stress reduction are not sustained, possibly reflecting larger structural and contextual stressors not directly addressed in the study (McCord et al., 2019). Acculturation stress, for example, may be a particularly salient but unaddressed factor contributing to sustained psychological stress among Latinx youth and their caregivers. Future studies are needed to evaluate how acculturation challenges contribute to psychological stress and how it could be addressed in the context of a mindfulness-based stress reduction intervention. Interestingly, parents enrolled in both HWB and HeLP reported baseline to follow-up reduction in depression symptoms. It is possible that besides mindfulness skills, the opportunities for social connection during the program, the perceived support from facilitators, and the opportunity to receive this program in Spanish or English by Latinx facilitators had a particularly positive influence on caregivers’ mental health.

Several study limitations should be noted. Despite a high recruitment rate, budgetary obstacles related to this pilot-level, multilayered intervention challenged our ability to (a) prolong the recruitment phase to reach the target sample size and (b) provide additional incentives and research support to bolster retention and follow-up. Additionally, although teenagers were offered the opportunity to provide feedback about their experiences with the program, we did not obtain any open-ended feedback from teenagers. Leveraging focus groups or posttreatment interviews in the future would facilitate a more nuanced understanding of adolescents’ experiences with mindfulness activities. Another limitation includes the lack of intervention fidelity checks. Although facilitators were trained and supervised by professional/licensed staff, we did not complete standardized fidelity checks.

Despite these limitations, this pilot study has some notable strengths, including (a) the addition of amental health component to a lifestyle intervention specifically developed for Latinx youth, (b) the use of bicultural staff who spoke Spanish and facilitated participation of caregivers whose primary language was not English, and (c) the strong preexisting relationships with various local community health centers, which facilitated recruitment. The present study also demonstrates the preliminary feasibility of implementing an integrative lifestyle and mental health program among Latinx families and highlights the need to explore future modifications to improve the acceptability of mindfulness-based approaches among Latinx youth at risk for T2D. The promising preliminary signals for various outcomes in the lifestyle and mental health intervention also underscore the need to further evaluate adjunctive mindfulness-based interventions in T2D prevention efforts among Latinx youth and families.

Resumen

La diabetes tipo 2 (DT2) de inicio en la adolescencia afecta desproporcionadamente a las comunidades latinas. Por lo tanto, la prevención es fundamental. Desafortunadamente, las intervenciones preventivas de la diabetes tipo 2 siguen siendo limitadas y no abordan los problemas de salud mental principales que afectan de manera desigual a los jóvenes de origen latino. Evaluamos la aceptabilidad y pusimos a prueba la viabilidad de un programa de estilo de vida + salud mental (HWB) comunitario e inclusivo para la familia, en comparación con un programa de solo estilo de vida (HeLP) entre adolescentes latinos y sus familias. Realizamos un ensayo de viabilidad piloto aleatorio controlado con evaluaciones de seguimiento a las 6 semanas y a los 6 meses. Familias inscritas (N = 18) participaron en el programa HWB o en el programa HeLP. Ambos programas incluyeron 12 sesiones de programación del estilo de vida durante 6 semanas. Sólo el programa HWB tenía un componente de salud mental que se enfocaba en los síntomas de estrés y depresión a través de prácticas de conciencia plena. Las evaluaciones de los estudios incluyeron datos cuantitativos, cualitativos y clínicos. Los resultados primarios fueron la aceptabilidad y la viabilidad de los programas. Los resultados exploratorios incluyeron patrones de cambio intragrupales en los índices conductuales y metabólicos/fisiológicos. La aceptabilidad fue alta para los padres y mixta para los adolescentes, particularmente con respecto al componente de salud mental del programa HWB. La viabilidad fue alta con respecto a la implementación de los programas y mixta con respecto a la retención de largo plazo. Los patrones de cambio exploratorios intragrupales variaron según el programa. Los resultados demuestran la viabilidad preliminar de implementar un programa integrador de estilo de vida + salud mental entre familias latinas, con una aceptabilidad muy positiva entre los padres y una aceptabilidad mixta entre los jóvenes. Las adaptaciones futuras para mejorar la aceptabilidad de los programas con salud mental y el acceso a más recursos para fortalecer las tasas de retención son los próximos pasos fundamen-tales en los métodos integradores para la prevención de la diabetes tipo 2 entre los jóvenes latinos.

Supplementary Material

Supplemental materials: https://doi.org/10.1037/lat0000279.supp

Public Significance Statement.

Implementation of a community-based, culturally tailored, and family-inclusive lifestyle + mental health program for the prevention of type 2 diabetes in Latinx youth was found to be feasible and generally acceptable, with some components of the program requiring further adaptations. Preliminary findings further showed that the observed behavioral and metabolic changes following the intervention were generally in the desired direction.