originalinvestigation | nutrition,obesity,andexercise

Original Investigation | Nutrition, Obesity, and Exercise

Effect of a Culturally Adapted Behavioral Interventionfor Latino Adults on Weight Loss Over 2 YearsA Randomized Clinical TrialLisa G. Rosas, PhD; Nan Lv, PhD; Lan Xiao, PhD; Megan A. Lewis, PhD; Elizabeth M. J. Venditti, PhD; Patricia Zavella, PhD; Kristen Azar, MSN/MPH; Jun Ma, MD, PhD

Abstract

IMPORTANCE Identifying effective weight loss interventions for Latino adults at risk of diabetes isof critical public health importance.

OBJECTIVE To determine whether a culturally adapted behavioral intervention for Latino adults wasmore effective than usual care for weight loss over 24 months.

DESIGN, SETTING, AND PARTICIPANTS In this randomized clinical trial, Latino adults with a bodymass index (BMI; calculated as weight in kilograms divided by height in meters squared) of 24 orgreater and a high risk for type 2 diabetes were recruited in primary care practices in the SanFrancisco, California, area, randomized to receive the Vida Sana intervention or usual care, andfollowed up for 24 months. The study was conducted from November 2015 to May 2019, and datawere analyzed from July 2019 to Septmeber 2020.

INTERVENTIONS The treatment group received Vida Sana, a culturally adapted lifestyleintervention that included a family-based orientation session and 22 group sessions over 12 months.Participants were encouraged to use a wearable activity tracker and mobile applications to track theirphysical activity and dietary intake. Participants received monthly email messages for an additional12 months. The control group received usual care.

MAIN OUTCOMES AND MEASURES The primary outcome was weight loss at 24 months.Secondary outcomes included weight loss at 12 months and achieving at least 5% weight loss at 12and 24 months. Associations of baseline characteristics and intervention adherence with weight lossoutcomes were also examined.

RESULTS Among 191 participants (mean [SD] age, 50.2 [12.2] years; 118 [61.8%] women; 107participants [57.2%] of Mexican origin; mean [SD] baseline BMI, 32.4 [5.7]) randomized, 92participants were randomized to the intervention and 99 participants were randomized to usualcare. Of these, 185 participants (96.9%) completed 24-month follow-up. Mean (SD) weight loss didnot differ significantly by group at 24 months (intervention: −1.1 [5.7] kg; control: −1.1 [7.1] kg; P = .93).However, mean (SD) weight loss was significantly greater in the intervention group (−2.6 [6.0] kg)than the control group (−0.3 [4.2] kg) at 12 months (mean difference, −2.1 [95% CI, −3.6 to −0.7] kg;P = .005). Intervention participants were more likely to achieve at least 5% weight loss than controlparticipants at 12 months (22 participants [25.9%] vs 9 participants [9.2%]; P = .003), andparticipants who achieved at least 5% weight loss attended more intervention sessions than thosewho did not (mean [SD], 16.6 [7.6] sessions vs 12.4 [7.5] sessions; P = .03).

(continued)

Key PointsQuestion Is a culturally adapted

behavioral lifestyle intervention using

technology more effective than usual

care for weight loss among Latino adults

in primary care?

Findings In this randomized clinical trial

of 191 Latino patients in primary care, a

culturally adapted behavioral lifestyle

intervention using technology, including

web-based self-monitoring and a

wearable activity monitor, was more

effective for weight loss over 12 months

but not 24 months.

Meaning These findings suggest that

the culturally adapted behavioral

intervention using technology was not

effective for long-term weight loss over

24 months, so research to optimize

intervention effectiveness over 24

months is needed.

+ Visual Abstract

+ Supplemental content

Author affiliations and article information arelisted at the end of this article.

Open Access. This is an open access article distributed under the terms of the CC-BY License.

JAMA Network Open. 2020;3(12):e2027744. doi:10.1001/jamanetworkopen.2020.27744 (Reprinted) December 18, 2020 1/16

Downloaded From: https://jamanetwork.com/ on 07/05/2022

https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamanetworkopen.2020.27744&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2020.27744


Abstract (continued)

CONCLUSIONS AND RELEVANCE These findings suggest that among Latino adults with highdiabetes risk, a culturally adapted behavioral lifestyle intervention was effective for weight loss over12 months but not 24 months.

TRIAL REGISTRATION ClinicalTrials.gov Identifier: NCT02459691

JAMA Network Open. 2020;3(12):e2027744. doi:10.1001/jamanetworkopen.2020.27744

Introduction

Latino adults in the United States experience high prevalence of overweight and obesity, with someestimates as high as 80%,1-3 and disparities in weight-related comorbidities, such as type 2 diabetes,compared with non-Hispanic White adults (12% vs 7%).4-7 Given that Latino people in the US are thelargest racial/ethnic minority group and account for the largest share of population growth,8

addressing overweight and obesity and related diabetes risk in this group is a critical publichealth need.9

The US Preventive Services Task Force recommends that primary care clinicians refer adultswith obesity (body mass index [BMI; calculated as weight in kilograms divided by height in meterssquared] �30), or those who are overweight (BMI �25) and have at least 1 additional cardiovasculardisease risk factor, such as prediabetes, to intensive behavioral weight loss interventions.10 TheDiabetes Prevention Program (DPP) set the criterion standard with an intensive behavioral lifestyleintervention that resulted in 6.8 kg weight loss over 12 months.11 Weight loss is the primary driver ofreduced diabetes risk.12

Latino adults have been underrepresented in studies of weight loss interventions, despite theirhigh risk.13 Cultural adaptation of evidence-based interventions, such as the DPP, is important toeffectively engage Latino adults owing to cultural differences in factors associated with risk andprotection.14 A systematic review of culturally adapted DPP interventions for Latino adultsdocumented weight loss at 12 months in 4 studies, ranging from 1.1 kg to 4.2 kg, showing thatclinically significant weight loss is possible but of smaller magnitude compared with the original DPPand other studies conducted with primarily non-Hispanic White adults.15-20 Only 1 study16

documented weight loss over 2 years, which was low (1 kg) and not statistically significantly differentthan weight loss in the control group. The interventions were primarily offered in home orcommunity settings. Primary care settings offer opportunities to leverage clinicians’ relationshipswith patients to identify and refer those who would benefit from behavioral weight lossinterventions.21

Technology-mediated weight loss strategies have emerged as an opportunity for improving theeffectiveness of weight loss interventions for minority populations.22,23 However, studies ofculturally adapted behavioral interventions focused on weight loss for Latino adults have notincorporated technology into the interventions. A large portion (80%) of Latino adults report accessto the internet, whether by mobile phone, tablet, or computer.24 Technology-mediated approachesmay increase intervention effectiveness by making the intervention more accessible, but theseapproaches need to be tested among Latino adults. Thus, the purpose of this randomized clinical trialwas to determine whether a culturally adapted behavioral intervention using technology was moreeffective than usual care for weight loss over 2 years among Latino primary care patients.

JAMA Network Open | Nutrition, Obesity, and Exercise Effect of a Culturally Adapted Behavioral Intervention for Latino Adults on Weight Loss



https://clinicaltrials.gov/ct2/show/NCT02459691


Methods

The institutional review board for Sutter Health, Northern California, approved this study. Allparticipants provided written informed consent. This study is reported following ConsolidatedStandards of Reporting Trials (CONSORT) reporting guideline for randomized clinical trials.

Research DesignThe Trial Protocol in Supplement 1 and results of the process to culturally adapt the intervention werepreviously published.25,26 Briefly, this was a translational trial in which Latino primary care patientswith a BMI of 24 or greater were randomized to receive usual care alone or with a culturally adaptedbehavioral lifestyle intervention and followed for 24 months with a primary outcome ofweight change.

Study ParticipantsParticipants were recruited from November 2015 to April 2017 in 3 cohorts with 1 each from 3different primary care sites within the Palo Alto Medical Foundation, a large community-basedmultispecialty group practice in Northern California. Adult (age �18 years) primary care patients whoself-reported Latino ethnicity and ability to speak Spanish (Spanish-only or bilingual) with a BMI 24or greater and prediabetes, a history of gestational diabetes, or 3 of 5 elements of the metabolicsyndrome27 but without type 1 or type 2 diabetes or cardiovascular disease were eligible toparticipate. Exclusions included significant psychiatric or medical comorbidities (eg, bipolar disorder,active cancer), pregnancy, or planned relocation during the follow-up period.

Study ProceduresActive primary care patients who were identified as meeting eligibility criteria in the electronic healthrecord (EHR) received a bilingual (Spanish and English) screening invitation via email or regular mail,if their primary care clinician approved. Patients self-screened online or via telephone with a bilingualstudy coordinator to assess eligibility criteria. Patients eligible after screening were invited tocomplete a baseline visit in which height and weight measurements were taken to confirm eligibility.Patients were not recruited on the basis of seeking or being recommended weight managementservices.

Randomization and BlindingEligible patients were randomized 1:1 to receive usual care or usual care plus a culturally adaptedbehavioral lifestyle intervention known as Vida Sana. All participants were provided with a wearableactivity tracker (Fitbit Zip) as an incentive. We used a covariate adaptive block randomizationmethod28 for block randomization to achieve balance between treatment groups according baselineclinic site, age, sex, BMI, waist circumference, and level of acculturation assessed by the ShortAcculturation Scale for Hispanics.29 Investigators, the data and safety monitoring board, outcomeassessors, and the data analyst were masked until after completion of the blind review of primaryoutcome data through 24 months.

InterventionThe intervention, which was delivered by a trained bilingual health coach in Spanish, is a culturaladaptation of the Group Lifestyle Balance curriculum30 derived from the original DPP lifestyleintervention.11 A bilingual health coach who did not have a specialized degree was trained by acertified master trainer for the Group Lifestyle Balance curriculum. The in-depth cultural adaptationincluded focus groups with Latino patients, key informant interviews with clinicians, and a structuredpretest with a Latino Patient Advisory Board.25 As a result of the adaptation, a family-wideorientation session was instituted to increase awareness about intervention goals and bestapproaches for providing positive social support, including structural, emotional, appraisal, and




http://www.equator-network.org/reporting-guidelines/consort/


informational support. Family members were also included in the session that focused on a healthyhome environment to promote desirable food and activity changes.

Intervention sessions were delivered in person for 1 year. Participants used a wearable activitytracker and mobile application to track their physical activity and the MyFitnessPal web or mobileapplication to track their dietary intake. The first 6 months, or core phase, included 16 group sessions(12 weekly sessions and then 4 bimonthly sessions). Based on social cognitive theory,31 the sessionsused behavioral strategies, such as self-monitoring, goal setting, stress management, and problemsolving, to achieve goals. The goals of the intervention were to achieve 7% weight loss and aminimum of 150 minutes per week of moderate-intensity physical activity. In addition, the healthcoach provided weekly individualized feedback to participants on their physical activity via theirfitness tracker application and diet via their diet tracking application. The postcore support phaseincluded an additional 6 monthly group sessions that focused on continued behavior change andother behavior maintenance strategies (eg, relapse control). A healthy meal was offered at eachin-person session to model healthy foods and to increase engagement and retention.25 Overall, thefirst 12 months of the intervention included approximately 23 hours of in-person time with the coach(orientation session plus 22 sessions) as well as feedback via the applications. In the second year,participants were sent monthly emails that reviewed the material from the first year and remindedparticipants to reach out to the coach for support. All intervention materials were provided inSpanish, with English available on request. Participants could elect to use the smartphoneapplications in their preferred language.

Usual CareParticipants in both treatment groups continued to receive usual care from their primary careclinicians. Primary care clinicians were not made aware of patients’ randomization assignment.Clinicians were neither encouraged nor prevented from offering weight management treatment topatients. Participants were not prevented from accessing weight management services from theirprimary care clinician or in the community. The health care system offered weight managementprograms, including bariatric surgery, group-based diabetes prevention programs, and mealreplacement.

Outcome MeasuresThis translational randomized clinical trial minimized patient burden and cost by relying on outcomemeasures that were easy to measure and required minimal time from participants and staff.32 Trainedbilingual study coordinators conducted in-person assessments at baseline, 12 months, and 24months. Assessments were completed in Spanish or English as preferred by the participant. Theprimary outcome was weight at 24 months, which was measured according to standardizedprotocols.33 Secondary outcomes included weight at 12 months, achieving 5% weight loss at 12 and24 months, as well as cardiometabolic risk factors, psychosocial well-being, and lifestyle behaviors at12 and 24 months. Achieving 5% baseline weight loss has been established as clinically meaningfulfor preventing diabetes.34 Cardiometabolic risk factors included blood pressure and waistcircumference, measured according to standardized protocols.35-37 Psychosocial well-beingmeasures included the Obesity-related Problems Scale (range, 0-100),38,39 which assesses obesity-specific quality of life, and the 9-item Patient Health Questionnaire (PHQ-9; range, 0-27), whichassesses depressive symptoms.40 The Obesity-related Problems Scale was translated into Spanishand found to be reliable and valid.39 The PHQ-9 has been tested in diverse populations, includingLatino adults, and found to be valid.40

Lifestyle behaviors included physical activity and diet. Physical activity was assessed using theStanford 7-Day Physical Activity Recall,41 and diet was assessed with a single multiple-pass 24-hourrecall using the Nutrition Data System for Research (NDSR).42,43 For physical activity, we calculatedtotal minutes and metabolic equivalent task (MET) minutes per week of leisure time physical activityby summing the unweighted and weighted physical activity minutes for moderate (weight: 4 MET




minutes), hard (weight: 6 MET minutes), and very hard (weight: 10 MET minutes) activities.44,45 Wealso calculated total energy expenditure, expressed as kilocalories per kilogram per day, by summingMET-minutes per day for each level of physical activity intensity and converting to kilocalories perkilogram per day using the conversion 1 MET-minute = 1 kcal/kg/h.44,45 We chose dietary indicatorsthat reflected intervention goals of increasing diet quality and fruit and vegetable consumption anddecreasing overall calories and fat. We calculated a Dietary Approach to Stop Hypertension (DASH)score and daily fruit and vegetable servings, daily calories (in kilocalories), and daily fat intake (ingrams). All diet variables were derived from the NDSR software. The DASH score was based on 9nutrient targets (ie, total fat, saturated fat, protein, cholesterol, fiber, magnesium, calcium, sodium,and potassium).46 For each nutrient target, participants were assigned a point if they achieved thetarget and half a point if they achieved an intermediate target (ie, half-way between the DASH targetand the population mean) and the DASH score was the sum of points for all 9 nutrients.47,48

Participation in outside weight management programs was assessed at study conclusion.Adverse events were monitored at each follow-up point and were also reported ad hoc byinterventionists.

Statistical AnalysisThree categories of prespecified analyses were performed: between-group differences in primaryand secondary outcomes, moderation analysis for primary outcome, and session attendance and itsassociation with weight change.

Analyses of between-group differences in primary and secondary outcomes included allparticipants with follow-up data at 12 or 24 months, and participants were analyzed based on thegroup to which they were assigned. Tests of group differences at 12 and 24 months in repeated-measures mixed-effects linear models were performed. The fixed effects of each model includedbaseline value of the outcome, randomization covariates, group (intervention or control), time (12 or24 months), and group-by-time interaction. In addition, we included Mexican origin as a fixed effectfor dietary outcomes, as dietary outcomes can vary among individuals from different countries oforigin. The random effects accounted for repeated measures with an unstructured covariance matrixand clustering of patients by primary care clinician. Adjusted differences in mean changes with 95%CIs and Cohen d for the primary outcome were obtained using model-based estimates. Analysesused all available data for each outcome, and missing data were handled directly through maximum-likelihood estimation via mixed modeling. Per the Trial Protocol in Supplement 1, in the case ofmissing study-measured weight, the closest EHR weight within 3 months of the due date of a missedstudy visit or self-reported weight (if no EHR weight available) was used.49 Sensitivity analyses wereconducted using study-measured weights only. A bootstrap resampling method was used to verifythat mixed-model–based results were not sensitive to violations of modeling assumptions.50

Because of the potential for type 1 error due to multiple comparisons, findings for analyses ofsecondary outcomes should be interpreted as exploratory.

Moderation analysis used the same mixed-effects linear model, as well as the main effect ofeach potential effect modifier and its interaction with group; the latter, if significant, resulted in arejection of the null hypothesis of no moderation. We then added the 3-way interaction of time,group, and moderator to generate time-specific moderation effects. Potential effect modifiersincluded sociodemographic factors (ie, age, sex, education, employment, occupation, marital status,household size, and income), food security (ie, household food security scale, which defined highfood security as a yes or no category), acculturation (assessed with the Short Acculturation scale forHispanics29), and health literacy (assessed with a short assessment of health literacy in Spanish andEnglish), all of which were prespecified.26

Among participants randomized to the intervention, we examined 4 indicators of interventionadherence and engagement and their correlation with weight loss: group session attendance,monitoring of weight at sessions, self-monitoring of diet in the diet tracking application, and self-monitoring of physical activity in the wearable activity tracker application. For group session





attendance, the correlation between the total number of attended group sessions and weightchanges from baseline to 12 and 24 months for the intervention group were plotted with the fittedregression lines. Monitoring of weight was determined based on whether the patient weighed in atthe session. Self-monitoring of diet and physical activity were determined based on whether thecoach recorded the patient’s self-monitoring data at the session. We categorized patients into 3groups of adherence based on if they had data in less than 50% of the sessions, 50% to less than75% of the sessions, or 75% or more of the sessions. We compared the mean weight changes frombaseline to 12 and 24 months in the 2 highest groups of adherence to the lowest using t tests.

As estimated, 93 participants would be needed per group to provide 80% power to detect a netbetween-group mean (SD) difference of 2.1 (4.6) kg at 24 months, assuming α = 5% (2-sided) and80% retention based on a previous trial in the same health care system.51 All analyses wereconducted using SAS Enterprise Guide version 7.1 (SAS Institute). Statistical significance was definedby 2-sided P < .05. Data were analyzed from July 2019 to September 2020.

Results

Study ParticipantsA total of 192 participants were randomized. After excluding one participant postrandomizationowing to a new cancer diagnosis that occurred before the patient was randomized but was notreported to the study until afterwards, 191 participants (mean [SD] age, 50.2 [12.2] years; 118 [61.8%]women; 107 participants [57.2%] of Mexican origin) were included, with 92 participants randomizedto receive the intervention and 99 participants randomized to receive usual care (Table 1). Primaryoutcome analyses included participants with follow-up data at 12 or 24 months, including 91participants in the intervention group and 99 participants in the usual care group (Figure 1).

Baseline demographic and clinical characteristics were similar between participants randomizedto the intervention and usual care, except the intervention group, compared with the usual caregroup, had fewer unemployed adults (13 participants [14.3%] vs 21 participants [21.6%]), smallerhousehold size (<2 members/household: 7 participants [7.7%] vs 4 participants [4.2%]), and lessconsumption of fruits and vegetables per day (mean [SD], 3.6 [2.5] servings/d vs 5.0 [4.4]servings/d). Most participants had at least some college experience (133 participants [71.1%]) andmiddle to upper incomes (90 participants [55.2%] with annual family incomes >$75 000) (Table 1).Participants’ mean (SD) BMI was 32.4 (5.7) overall and slightly higher among women compared withmen (33.0 [6.1] vs 31.4 [5.0]).

Primary and Secondary OutcomesThere was only 1 (0.5%) missing weight measurement. Mean (SD) weight loss at 24 months did notdiffer significantly among intervention participants compared with control participants (−1.1 [7.1] kg vs−1.1 [5.7] kg; adjusted mean difference, 0.1 [95% CI, −1.8 to 1.9] kg; P = .93) (Table 2). Sensitivityanalyses using only study-measured weights and bootstrap resampling method provided consistentresults (eTable 1 and eTable 2 in Supplement 2). However, mean (SD) weight loss at 12 months wassignificantly greater among intervention participants compared with control participants (−2.6 [6.0]kg vs−0.3 [4.2] kg; adjusted mean difference, −2.1 [95% CI, −3.6 to −0.7] kg; P = .005).

A significantly greater proportion of intervention participants achieved clinically significantweight loss (ie, �5% of baseline weight) compared with control participants at 12 months (22participants [25.9%] vs 9 participants [9.2%]; P = .003) but not 24 months (22 participants [24.2%]vs 15 participants [15.2%]; P = .10). There was no significant effect of the intervention compared withthe control on secondary outcomes of waist circumference, leisure time physical activity, total energyexpenditure, obesity-related problems, and health-related quality of life at 12 or 24 months (Table 2).We did not detect moderation by any of the prespecified variables.





Table 1. Demographic and Key Baseline Characteristics by Treatment Groupa

Characteristic

Mean (SD)

Overall (n = 191) Intervention (n = 92) Usual care (n = 99)Age, y 50.2 (12.2) 50.3 (12.5) 50.1 (12.0)

Female, No. (%) 118 (61.8) 57 (62) 61 (61.6)

Region of origin, No. (%)b

Mexico 107 (57.2) 52 (57.1) 55 (57.3)

Central America 32 (17.1) 17 (18.7) 15 (15.6)

South America 22 (11.8) 8 (8.8) 14 (14.6)

Puerto Rico, Cuba, other Spanish, or multi-origin 26 (13.9) 14 (15.4) 12 (12.5)

Education, No. (%)c

≤High school or GED 54 (28.9) 26 (28.9) 28 (28.9)

Some college 50 (26.7) 22 (24.4) 28 (28.9)

College graduate 41 (21.9) 21 (23.3) 20 (20.6)

Post college 42 (22.5) 21 (23.3) 21 (21.7)

Marital status, No. (%)d

Married or living with a partner 126 (67) 58 (63.7) 68 (70.1)

Single, separated, divorced, or widowed 62 (33) 33 (36.3) 29 (29.9)

Employment status, No. (%)d

Full-time 134 (71.3) 63 (69.2) 71 (73.2)

Part-time 20 (10.6) 15 (16.5) 5 (5.2)

Unemployed 34 (18.1) 13 (14.3) 21 (21.6)

Occupation industry, No. (%)e

Professional and business services 28 (18.8) 13 (16.7) 15 (21.1)

Educational and health services 43 (28.9) 20 (25.6) 23 (32.4)

Information, retail, transportation, construction, manufacturing,hospitality, or public administration

43 (28.8) 25 (32.1) 18 (25.4)

Other 35 (23.5) 20 (25.6) 15 (21.11)

Household size, No. (%)f

<2 11 (5.9) 7 (7.7) 4 (4.2)

2 28 (15.1) 7 (7.7) 21 (22.1)

≥3 147 (79) 77 (84.6) 70 (73.7)

Annual family income, $g

<75 000 73 (44.8) 39 (48.1) 34 (41.5)

75 000 to <125 000 40 (24.5) 18 (22.2) 22 (26.8)

≥125 000 50 (30.7) 24 (29.6) 26 (31.7)

Household food security, No. (%)d,h

High 158 (84) 80 (87.9) 78 (80.4)

Low 21 (11.2) 7 (7.7) 14 (14.4)

Very low 9 (4.8) 4 (4.4) 5 (5.2)

BMIa 32.4 (5.7) 32.4 (5.4) 32.4 (6.0)

Women 33.0 (6.1) 33.3 (6.0) 32.8 (6.1)

Men 31.4 (5.0) 30.8 (3.8) 31.9 (5.9)

Weight, kg 87.1 (19.1) 86.6 (17.2) 87.6 (20.8)

Women 82.3 (17.0) 83.2 (17.4) 81.5 (16.7)

Men 94.8 (19.9) 92.2 (15.5) 97.3 (23.2)

Waist circumference, cm 103.9 (14.4) 103.8 (13.7) 103.9 (15.0)

Women 103.8 (14.5) 104.3 (15.4) 103.5 (13.7)

Men 103.9 (14.3) 103.1 (10.6) 104.7 (17.1)

Short Acculturation Scale for Hispanics scorei 3.1 (0.8) 3.1 (0.7) 3.1 (0.8)

Short assessment of health literacy scored,j 16.6 (1.8) 16.6 (1.9) 16.6 (1.8)

Blood pressure, mm Hg

Systolic 121.8 (14.6) 122.5 (12.8) 121.3 (16.2)

Diastolic 74.1 (9.5) 74.8 (8.6) 73.5 (10.2)

(continued)




Intervention Adherence and Weight LossSession attendance was significantly positively correlated with weight loss at 12 months (ρ = −0.33[95% CI, −0.51 to −0.13]; P = .002) but not at 24 months (ρ = −0.21 [95% CI, −0.40 to <0.01]; P = .05)(Figure 2). Participants who achieved clinically significant weight loss at 12 months attended moresessions than those who did not (mean [SD], 16.6 [7.6] sessions vs 12.4 [7.5] sessions; P = .03).Greater monitoring of weight, diet, and physical activity were associated with greater weight loss(Table 3).

Table 1. Demographic and Key Baseline Characteristics by Treatment Groupa (continued)

Characteristic

Mean (SD)

Overall (n = 191) Intervention (n = 92) Usual care (n = 99)Leisure time moderate and vigorous physical activity total min/wkk 232 (241) 208 (191) 254 (280)

Leisure time physical activity, MET min/wkk,l 1126 (1249) 1000 (965) 1246 (1463)

Total energy expenditure, kcal/kg/dk,m 35.4 (4.2) 35.4 (4.2) 35.4 (4.1)

DASH scoren,o 2.5 (1.3) 2.4 (1.2) 2.6 (1.4)

Fruit and vegetable, servings/dn 4.3 (3.7) 3.6 (2.5) 5.0 (4.4)

Total calorie intake, kcal/dn 1734 (808) 1623 (657) 1839 (919)

Total fat, g/dn 71.4 (41.8) 69.7 (36.9) 73.1 (46.1)

Obesity-related problem scaled,p 40.2 (30.0) 44.0 (28.5) 36.8 (31.1)

PHQ-9 score d,q 5.4 (4.5) 6.0 (4.7) 4.8 (4.3)

T-scored,r

Sleep disturbance 51.6 (8.9) 51.6 (9.3) 51.5 (8.6)

Sleep impairment 51.2 (9.2) 51.6 (9.3) 50.7 (9.1)

Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); DASH, Dietary Approach to Stop Hypertension; GED, generaleducational development; MET, metabolic equivalent task; PHQ-9, Patient Health Questionnaire 9-item.a Prognostic factors for randomization were study site, age, sex, BMI, waist circumference, and Short Acculturation Scale for Hispanics.b Includes data for 91 individuals in the intervention group and 96 individuals in the usual care group.c Includes data for 90 individuals in the intervention group and 97 individuals in the usual care group.d Includes data for 91 individuals in the intervention group and 97 individuals in the usual care group.e Includes data for 71 individuals in the intervention group and 78 individuals in the usual care group.f Includes data for 91 individuals in the intervention group and 95 individuals in the usual care group.g Includes data for 81 individuals in the intervention group and 82 individuals in the usual care group.h Scores were calculated as the sum of affirmative responses to 6 questions in the module, then food security status was assigned as raw score 0 to 1, high or marginal food security

(raw score 1 may be considered marginal food security, but a large proportion of households that would be measured as having marginal food security using the household or adultscale will have a raw score of 0 on the 6-item scale); raw score 2 to 4, low food security; and raw score 5 to 6, very low food security.

i Measured using a 12-item questionnaire scaled from 1 to 5 for each item (1 = only Spanish, 2 = Spanish better than English, 3 = both equally, 4 = English better than Spanish, and5 = only English), and the score is calculated as mean of all items. A higher score indicates higher acculturation to US society.

j 18 test items to assess a Spanish-speaking adult’s ability to read and understand common medical terms in English or an English-speaking adult’s ability to read and understandcommon medical terms in Spanish. Each correct answer gets 1 point. The score is the sum of all items, ranging from 0 to 18.

k Includes data for 92 individuals in the intervention group and 97 individuals in the usual care group.l Physical activity levels were measured by the interview-administered Stanford 7-Day Physical Activity Recall. Leisure-time physical activity = non–work-related moderate activity

min/wk × 4 METs + hard activity min/wk × 6 METs + very hard activity min/wk × 10 METs. One MET is defined as the energy expenditure for sitting quietly.mThe Stanford 7-Day Physical Activity Recall data also provided estimates of total daily energy expenditures. Total energy expenditure = sleep hours × 1 MET + light activity

hours × 1.5 METs + moderate activity hours × 4 METs + hard activity hours × 6 METs + very hard activity hours × 10 METs.n Includes data for 76 individuals in the intervention group and 81 individuals in the usual care group.o DASH scores were calculated based on combining 9 nutrient targets (ie, total fat, saturated fat, protein, cholesterol, fiber, magnesium, calcium, sodium, and potassium). The

intermediate target of each nutrient was halfway between the DASH target and population mean (based on the National Health and Nutrition Examination Surveys 2007-2008,latest data available at the inception of this study). For a nutrient, participants reaching the DASH target were assigned 1 point, those reaching the intermediate target were assigneda half point, and those not meeting the intermediate target were given 0 points. The DASH score was the sum of points for all 9 nutrients.

p Calculated as mean scores of 8 questions, with a range from 0 (no at all) to 3 (extremely). Raw scores were multiplied by 100 then divided by 3. Higher score indicates more obesity-related psychosocial problems.

q Calculated as total scores of the 9 items with a range from 0 (no symptoms) to 27 (most severe symptoms). PHQ-9 score of 0 to 4 indicates minimal depression; 5 to 9, milddepression; 10 to 14, moderate depression; 15 to 19, moderately severe depression; and 20 to 27, severe depression.

r Measured using a population scale with a range from 25 (extreme disturbance or impairment) to 80 (none) converted from the raw total score of 8-item questionnaire. A score of50 represents the mean of the calibration sample, which was generally more enriched for chronic illness.




Outside Weight Management Participation and Adverse EventsThere were 19 participants in usual care who participated in various weight management programs,and 9 participants in the intervention who participated in other weight management programsoutside this study. Over the 24-month trial, there were 36 serious adverse events, 16 requiringhospitalization and 11 involving fractures or musculoskeletal injuries that required outpatient repairprocedures. Of these 36 serious adverse events, 4 (11.1%) were possibly related to the study (ie,hospital admissions following elective panniculectomy and abdominoplasty, torn meniscus, ankle

Figure 1. Participant Enrollment Flowchart

2871 Individuals identified inthe EHR as potentially eligible

1792 Sent invitation letters

617 Completed initial screening online or by telephone

202 Completed baseline visit

1080 Excluded193 Excluded by primary care clinician490 Not reviewed by primary

care clinician397 Letters not yet mailed

192 Randomized

92 Assigned to intervention

12-mo Follow-up

1 With self-reported weight

85 Patients with weight data74 With study measured weight10 With EHR-abstracted weight

24-mo Follow-up



99 Assigned to usual care

12-mo Follow-up



24-mo Follow-up



1175 Excluded362 Declined screening794 Pending initial screening19 Unreachable

415 Excluded338 Ineligible during initial screen74 Declined baseline visit3 Ineligible prior to baseline visit

10 Excluded8 Ineligible during baseline visit1 Protocol deviation1 Declined to complete

baseline questionnaire

10 Patients excluded owing to new cancerdiagnosis prior to randomization,detected after randomization

EHR indicates electronic health record.




Table 2. Primary and Secondary Outcomes

Outcome

Unadjusted change, mean (SD)Adjusted mean treatmentdifference (95% CI)a P valueaIntervention Usual care

Primary

Weight, kg

12 mo −2.6 (6.0) −0.3 (4.2) −2.1 (−3.6 to −0.7) .005

24 mo −1.1 (5.7) −1.1 (7.1) 0.1 (−1.8 to 1.9) .93

Secondary

BMI

12 mo −1.0 (2.3) −0.1 (1.6) −0.8 (−1.4 to −0.3) .004

24 mo −0.4 (2.2) −0.4 (2.7) 0.0 (−0.7 to 0.7) .99

5% weight loss, No. (%)b

12 mo 22 (25.9) 9 (9.2) 15.3 (5.1 to 26.0) .003

24 mo 22 (24.2) 15 (15.2) 8.7 (−2.0 to 20.2) .10

Waist circumference, cm

12 mo −4.3 (12.6) −1.9 (5.4) −2.3 (−5.1 to 0.5) .11

24 mo 0.3 (7.1) −0.1 (7.2) 0.5 (−1.6 to 2.7) .61

Leisure time moderate and vigorousphysical activity total min/wk

12 mo 35.8 (226.3) −36.9 (277.5) 33.3 (−27.5 to 94.1) .28

24 mo 6.4 (291.6) −74.1 (301) 46.4 (−24.9 to 117.7) .20

Leisure time physical activity METmin/wkb

12 mo 218.5 (1159) −231 (1411) 242.6 (−61.3 to 546.5) .12

24 mo −34.2 (1343) −386 (1541) 179 (−147.9 to 505.9) .28

Total energy expenditure,kcal/kg/d

12 mo 0.3 (5.0) −0.4 (3.9) 0.8 (−0.3 to 1.9) .18

24 mo 0.7 (5.6) −0.8 (4.3) 1.4 (−0.1 to 2.9) .06

DASH score

12 mo 0.4 (1.4) −0.2 (1.8) 0.2 (−0.2 to 0.7) .31

24 mo 0.3 (1.8) −0.2 (2.0) 0.1 (−0.5 to 0.6) .78

Fruit and vegetable,servings/d

12 mo −0.2 (3.1) −0.9 (4.3) 0 (−1.0 to 1.1) .95

24 mo 1.1 (4.0) −0.4 (5.2) 0.6 (−0.7 to 2.0) .36

Total calorie intake, kcal/d

12 mo −168 (760) −235 (910) −48.2 (−246.9 to 150.5) .63

24 mo −85.2 (671) −268 (703) 56.5 (−160 to 273.1) .61

Total fat, g/d

12 mo −15.0 (44.6) −10.2 (50.1) −6.5 (−17.4 to 4.5) .24

24 mo −8.0 (40.9) −8.7 (41.6) −0.6 (−12.9 to 11.6) .92

Obesity-related problem scale

12 mo −11.1 (28.0) −9.1 (23.2) 1.6 (−5.2 to 8.4) .65

24 mo −10.0 (23.8) −8.2 (26.0) 2.1 (−5 to 9.1) .56

EQ-5D, no symptoms, No. (%)

Mobilec

12 mo 57 (82.6) 68 (81.0) 1.8 (−11.0 to 10.2) .84

24 mo 66 (83.5) 74 (81.3) 1.1 (−8.7 to 9.2) .89

Carec

12 mo 64 (92.8) 78 (92.9) −0.7 (−0.8 to 0.2) .46

24 mo 75 (94.9) 82 (90.1) 0.4 (−0.3 to 0.6) .64

Activityc

12 mo 56 (81.2) 72 (85.7) −6.7 (−28.3 to 1.1) .22

24 mo 69 (87.3) 77 (84.6) 0.1 (−10.6 to 6.6) .98

(continued)




surgery, and finger fracture). There were 202 nonserious adverse events, with 64 (31.7%) involvingminor musculoskeletal injuries being the most common and 12 of 64 minor musculoskeletal injuriespossibly related to the study. Combined serious adverse events (14 events in the intervention groupvs 22 in the control group) and nonserious adverse events (101 events in the intervention group vs101 events in the control group) were comparably distributed across both groups. There wereno deaths.

Table 3. Percentage of Completions for the Key Intervention Components Associated With Weight Change From Baseline at 12 and 24 Months

Follow-up

Component sessions attended, %

Weight monitoring Diet monitoring Physical activity monitoring<50(n = 33)

50 to <75(n = 22)

≥75(n = 37)

<50(n = 70)

50 to <75(n = 9)

≥75(n = 13)

<50(n = 59)

50 to <75(n = 14)

≥75(n = 19)

Weight change from baseline,mean (SD) kg12 mo 0.1 (3.9) −2.2 (5.7) −4.7 (6.6)a −0.9 (4.5) −8.1 (9.2)a −6.4 (5.7)a −0.4 (4.5) −5.3 (5.5)a −6.5 (7.2)a

24 mo 0.2 (5.1) −1.3 (6.3) −2 (5.9) −0.2 (5.3) −4.5 (8)a −3.2 (5.1) 0.1 (5.4) −2.4 (4.7) −3.5 (6.6)a

a Indicates significant difference compared with <50% completion group.

Table 2. Primary and Secondary Outcomes (continued)

Outcome

Unadjusted change, mean (SD)Adjusted mean treatmentdifference (95% CI)a P valueaIntervention Usual care

Painc

12 mo 32 (46.4) 44 (52.4) −6.7 (−25.4 to 10.4) .45

24 mo 38 (48.1) 48 (52.8) −7.8 (−27.6 to 10.3) .39

Anxiousc

12 mo 41 (59.4) 47 (56.0) 12.2 (−8.8 to 33.5) .26

24 mo 49 (62) 52 (57.1) 10.0 (−13.1 to 31.9) .31

Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); EQ-5D,EuroQol 5-Dimension; DASH, Dietary Approach to Stop Hypertension; MET, metabolic equivalents.a Mixed-effects models accounting for the random effects of repeated measures and primary care physicians and adjusted

for baseline value of the outcome of interest, study site, age, sex, BMI, waist circumference, and Short Acculturation Scalefor Hispanics. For Nutrition Data System for Research outcomes (ie, DASH score, fruit and vegetable, total calorie intake,and total fat), the additional variable Mexican origin (yes or no) was adjusted.

b Includes data for 91 individuals in the intervention group and 99 individuals in the usual care group.c Includes data for 82 individuals in the intervention group and 92 individuals in the usual care group.

Figure 2. Correlation of Session Attendance and Weight Change

20

10

0

–10

–20

–300 2520

Wei

ght c

hang

e, k

g

Group sessions, No.15105




Discussion

In this randomized clinical trial, Latino primary care patients randomized to a culturally adapted,technologically mediated behavioral lifestyle intervention did not lose more weight at 24 monthsthan those randomized to usual care. However, participants in the intervention lost more weight andwere more likely to achieve clinically significant weight loss (ie, �5%) at 12 months than thoserandomized to usual care. There were no significant effects of the intervention on secondaryoutcomes at 12 or 24 months. Increased session attendance was significantly associated withachieving clinically significant weight loss at 12 months.

In the only other randomized clinical trial of a culturally adapted diabetes preventionintervention with 24 months of follow-up, to our knowledge, there was also no difference in weightloss between the intervention and usual care groups.16 Both studies provided minimal contact withparticipants during the second year. It is possible that participants need additional support in thesecond year to continue to lose weight or sustain weight loss, as has been done in previous trials,albeit not with primarily Latino adults.52-54

The range of weight loss observed over 12 months in this study and in other culturally adaptedinterventions (ie, 1.1 kg to 4.2 kg) is less than what was observed in the original DPP trial andbehavioral weight loss interventions in primary care.11,15-20,55 For instance, in a study of 243 primarycare patients (78% non-Hispanic White patients), those randomized to receive the Group LifestyleBalance lost 6.3 kg over 15 months, which was significantly greater than usual care.51 It is unknownwhether lower levels of weight loss are owing to the cultural adaptation or to barriers that Spanish-speaking Latino adults face in achieving weight loss.

As has been documented in prior studies,54,56-58 session attendance was positively associatedwith weight loss. Technology-mediated strategies, such as online videos or videoconferencesessions, could be incorporated to increase session attendance. In a study of 351 primarily AfricanAmerican primary care patients in North Carolina, patients received a study-specific smartphoneapplication and 18 telephone calls from a coach. More than two-thirds of participants randomized tothe intervention completed 80% or more of the coach telephone calls and lost 5.2 kg over12 months.22

LimitationsThis study has some limitations. There was insufficient power to detect whether moderation waspresent. Understanding possible heterogeneity of effects is critical for informing future strategies tooptimize effectiveness. Additionally, the findings of this study may not apply to other Latinopopulations. Latino adults in this study were recruited from a health system accessed primarily bypatients with employer-based health insurance and as a result had relatively high annual incomescompared with other similar studies with Latino adults. Thus, the findings of this study may not begeneralizable to Latino adults with different socioeconomic profiles. Additionally, this translationalstudy was not designed to examine mechanisms, such as change in diet and physical activity. As aresult, more rigorous measures of dietary intake and objective measures of physical activity werenot used.

Conclusions

This randomized clinical trial found that a technology-mediated culturally adapted behaviorallifestyle intervention was effective for weight loss over 12 months but not 24 months among adultLatino primary care patients at risk for diabetes. Identified factors associated with weight lossrevealed opportunities for future research to test strategies to optimize effectiveness of behaviorallifestyle interventions with Latino adults. Given that Latino individuals are disproportionatelyrepresented among adults in the US with obesity and diabetes, it is critical to pursue future researchaimed at increasing effectiveness of behavioral lifestyle interventions for chronic disease prevention.




ARTICLE INFORMATIONAccepted for Publication: October 4, 2020.

Published: December 18, 2020. doi:10.1001/jamanetworkopen.2020.27744

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Rosas LG et al.JAMA Network Open.

Corresponding Author: Jun Ma, MD, PhD, Vitoux Program on Aging and Prevention, Department of Medicine,University of Illinois at Chicago, 1747 W Roosevelt Rd, Room 586 (MC 275), Chicago, IL 60608([email protected]).

Author Affiliations: Department of Epidemiology and Population Health, Stanford University, Palo Alto, California(Rosas, Xiao); Vitoux Program on Aging and Prevention, Department of Medicine, University of Illinois at Chicago,Chicago (Lv, Ma); Center for Communications Science, RTI International, Seattle, Washington (Lewis); Departmentof Psychiatry, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania (Venditti); Department ofLatin American and Latino Studies, University of California, Santa Cruz (Zavella); Sutter Health Center for HealthSystems Research, Walnut Creek, California (Azar).

Author Contributions: Dr Ma had full access to all of the data in the study and takes responsibility for the integrityof the data and the accuracy of the data analysis.

Concept and design: Rosas, Lewis, Zavella, Ma.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Rosas, Lv, Xiao, Lewis, Azar, Ma.

Critical revision of the manuscript for important intellectual content: Lewis, Venditti, Zavella, Ma.

Statistical analysis: Rosas, Xiao.

Obtained funding: Lewis, Ma.

Administrative, technical, or material support: Rosas, Lv, Lewis, Venditti, Azar.

Supervision: Venditti, Azar, Ma.

Conflict of Interest Disclosures: Dr Lv reported receiving grants from Agency for Healthcare Research and Quality(AHRQ) during the conduct of the study. Dr Azar reported receiving grants from the National Institute of Diabetesand Digestive and Kidney Diseases, Patient-Centered Outcomes Research Institute, and Omada Health outside thesubmitted work. Dr Ma reported receiving personal fees from Health Mentor outside the submitted work. No otherdisclosures were reported.

Funding/Support: This work was supported by the Agency for Healthcare Research and Quality under award No.R01HS022702.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection,management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; anddecision to submit the manuscript for publication.

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the officialviews of the National Institutes of Health.

Data Sharing Statement: See Supplement 3.

Additional Contributions: The Data and Safety Monitoring Board (DSMB) members included William L. Haskell,PhD (Stanford University), Manisha Desai, PhD (Stanford University), Jeanette Aviles, MD (San Mateo MedicalCenter), and Armando Valdez, PhD (HealthPoint Communications). The DSMB reviewed the study protocol initiallyand data quality and safety monitoring reports semiannually thereafter and were compensated for their time.Elizabeth Jameiro, MD, Nancy Wittels, MS, Veronica Luna, BS, Andrea Blonstein, MBA, RD, and Alyssa Hernandez,BA (Sutter Health), and Mayra Amador, BA (County of San Mateo), contributed instrumentally to the conduct ofthe study as research team members and were paid for their work.

REFERENCES1. Flegal KM, Kruszon-Moran D, Carroll MD, Fryar CD, Ogden CL. Trends in obesity among adults in the unitedStates, 2005 to 2014. JAMA. 2016;315(21):2284-2291. doi:10.1001/jama.2016.6458

2. Summary Health Statistics: National Health Interview Survey: 2014. Table A-4: selected diseases and conditionsamong adults aged 18 and over, by selected characteristics: United States, 2014. Accessed May 22, 2019. https://ftp.cdc.gov/pub/Health_Statistics/NCHS/NHIS/SHS/2014_SHS_Table_A-4.pdf

3. Centers for Disease Control and Prevention. National Diabetes Statistics Report, 2017. Centers for DiseaseControl and Prevention; 2017.





https://jamanetwork.com/journals/jamanetworkopen/pages/instructions-for-authors#SecOpenAccess/?utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2020.27744

mailto:[email protected]


https://jama.jamanetwork.com/article.aspx?doi=10.1001/jama.2016.6458&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2020.27744

https://ftp.cdc.gov/pub/Health_Statistics/NCHS/NHIS/SHS/2014_SHS_Table_A-4.pdf

https://ftp.cdc.gov/pub/Health_Statistics/NCHS/NHIS/SHS/2014_SHS_Table_A-4.pdf

4. Avilés-Santa ML, Schneiderman N, Savage PJ, et al. Identifying probable diabetes mellitus among Hispanics/Latinos from four U.S. cities: findings from the Hispanic Community Health Study/Study of Latinos. Endocr Pract.2016;22(10):1151-1160. doi:10.4158/EP151144.OR

5. Schneiderman N, Llabre M, Cowie CC, et al. Prevalence of diabetes among Hispanics/Latinos from diversebackgrounds: the Hispanic Community Health Study/Study of Latinos (HCHS/SOL). Diabetes Care. 2014;37(8):2233-2239. doi:10.2337/dc13-2939

6. Al-Goblan AS, Al-Alfi MA, Khan MZ. Mechanism linking diabetes mellitus and obesity. Diabetes Metab SyndrObes. 2014;7:587-591. doi:10.2147/DMSO.S67400

7. Cowie CC, Rust KF, Ford ES, et al. Full accounting of diabetes and pre-diabetes in the U.S. population in 1988-1994 and 2005-2006. Diabetes Care. 2009;32(2):287-294. doi:10.2337/dc08-1296

8. Flores A. How the US Hispanic Population Is Changing. Fact Tank. September 18, 2017. Accessed November16,2020. https://www.pewresearch.org/fact-tank/2017/09/18/how-the-u-s-hispanic-population-is-changing/

9. Venditti EM. Behavioral lifestyle interventions for the primary prevention of type 2 diabetes and translation toHispanic/Latino communities in the United States and Mexico. Nutr Rev. 2017;75(suppl 1):85-93. doi:10.1093/nutrit/nuw041

10. Curry SJ, Krist AH, Owens DK, et al; US Preventive Services Task Force. Behavioral weight loss interventions toprevent obesity-related morbidity and mortality in adults: US Preventive Services Task Force recommendationstatement. JAMA. 2018;320(11):1163-1171. doi:10.1001/jama.2018.13022

11. Knowler WC, Barrett-Connor E, Fowler SE, et al; Diabetes Prevention Program Research Group. Reduction inthe incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002;346(6):393-403.doi:10.1056/NEJMoa012512

12. Hamman RF, Wing RR, Edelstein SL, et al. Effect of weight loss with lifestyle intervention on risk of diabetes.Diabetes Care. 2006;29(9):2102-2107. doi:10.2337/dc06-0560

13. Haughton CF, Silfee VJ, Wang ML, et al. Racial/ethnic representation in lifestyle weight loss intervention studiesin the United States: a systematic review. Prev Med Rep. 2018;9:131-137. doi:10.1016/j.pmedr.2018.01.012

14. Castro FG, Barrera M Jr, Holleran Steiker LK. Issues and challenges in the design of culturally adapted evidence-based interventions. Annu Rev Clin Psychol. 2010;6:213-239. doi:10.1146/annurev-clinpsy-033109-132032

15. McCurley JL, Gutierrez AP, Gallo LC. Diabetes prevention in U.S. Hispanic adults: a systematic review ofculturally tailored interventions. Am J Prev Med. 2017;52(4):519-529. doi:10.1016/j.amepre.2016.10.028

16. Rosas LG, Thiyagarajan S, Goldstein BA, et al. The effectiveness of two community-based weight loss strategiesamong obese, low-income US Latinos. J Acad Nutr Diet. 2015;115(4):537-550.e2. doi:10.1016/j.jand.2014.10.020

17. Ockene IS, Tellez TL, Rosal MC, et al. Outcomes of a Latino community-based intervention for the preventionof diabetes: the Lawrence Latino Diabetes Prevention Project. Am J Public Health. 2012;102(2):336-342. doi:10.2105/AJPH.2011.300357

18. Parikh P, Simon EP, Fei K, Looker H, Goytia C, Horowitz CR. Results of a pilot diabetes prevention interventionin East Harlem, New York City: Project HEED. Am J Public Health. 2010;100(suppl 1):S232-S239. doi:10.2105/AJPH.2009.170910

19. Van Name MA, Camp AW, Magenheimer EA, et al. Effective translation of an intensive lifestyle intervention forHispanic women with prediabetes in a community health center setting. Diabetes Care. 2016;39(4):525-531. doi:10.2337/dc15-1899

20. O’Brien MJ, Perez A, Scanlan AB, et al. PREVENT-DM comparative effectiveness trial of lifestyle interventionand metformin. Am J Prev Med. 2017;52(6):788-797. doi:10.1016/j.amepre.2017.01.008

21. Zamudio CD, Sanchez G, Altschuler A, Grant RW. Influence of language and culture in the primary care ofSpanish-speaking Latino adults with poorly controlled diabetes: a qualitative study. Ethn Dis. 2017;27(4):379-386.doi:10.18865/ed.27.4.379

22. Bennett GG, Steinberg D, Askew S, et al. Effectiveness of an app and provider counseling for obesity treatmentin primary care. Am J Prev Med. 2018;55(6):777-786. doi:10.1016/j.amepre.2018.07.005

23. Kozak AT, Buscemi J, Hawkins MA, et al. Technology-based interventions for weight management: currentrandomized controlled trial evidence and future directions. J Behav Med. 2017;40(1):99-111. doi:10.1007/s10865-016-9805-z

24. Brown A, López G, Lopez MH. Digital divide narrows for Latinos as more Spanish speakers and immigrants goonline. Published July 20, 2016. Accessed November 16, 2020. https://www.pewresearch.org/hispanic/2016/07/20/digital-divide-narrows-for-latinos-as-more-spanish-speakers-and-immigrants-go-online/




https://dx.doi.org/10.4158/EP151144.OR

https://dx.doi.org/10.2337/dc13-2939

https://dx.doi.org/10.2147/DMSO.S67400


https://www.pewresearch.org/fact-tank/2017/09/18/how-the-u-s-hispanic-population-is-changing/

https://dx.doi.org/10.1093/nutrit/nuw041

https://dx.doi.org/10.1093/nutrit/nuw041


https://dx.doi.org/10.1056/NEJMoa012512


https://dx.doi.org/10.1016/j.pmedr.2018.01.012

https://dx.doi.org/10.1146/annurev-clinpsy-033109-132032

https://dx.doi.org/10.1016/j.amepre.2016.10.028

https://dx.doi.org/10.1016/j.jand.2014.10.020

https://dx.doi.org/10.2105/AJPH.2011.300357






https://dx.doi.org/10.18865/ed.27.4.379


https://dx.doi.org/10.1007/s10865-016-9805-z

https://dx.doi.org/10.1007/s10865-016-9805-z

https://www.pewresearch.org/hispanic/2016/07/20/digital-divide-narrows-for-latinos-as-more-spanish-speakers-and-immigrants-go-online/

https://www.pewresearch.org/hispanic/2016/07/20/digital-divide-narrows-for-latinos-as-more-spanish-speakers-and-immigrants-go-online/

25. Rosas LG, Lv N, Lewis MA, et al. A Latino patient-centered, evidence-based approach to diabetes prevention.J Am Board Fam Med. 2018;31(3):364-374. doi:10.3122/jabfm.2018.03.170280

26. Rosas LG, Lv N, Xiao L, et al. Evaluation of a culturally-adapted lifestyle intervention to treat elevatedcardiometabolic risk of Latino adults in primary care (Vida Sana): a randomized controlled trial. Contemp ClinTrials. 2016;48:30-40. doi:10.1016/j.cct.2016.03.003

27. Grundy SM, Cleeman JI, Daniels SR, et al; American Heart Association; National Heart, Lung, and BloodInstitute. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart,Lung, and Blood Institute Scientific Statement. Circulation. 2005;112(17):2735-2752. doi:10.1161/CIRCULATIONAHA.105.169404

28. Xiao L, Lavori PW, Wilson SR, Ma J. Comparison of dynamic block randomization and minimization inrandomized trials: a simulation study. Clin Trials. 2011;8(1):59-69. doi:10.1177/1740774510391683

29. Marin G, Gamba RJ. A new measurement of acculturation for Hispanics: the Bidimensional Acculturation Scalefor Hispanics (BAS). Hisp J Behav Sci. 1996;18:297-316. doi:10.1177/07399863960183002

30. Kramer MK, Kriska AM, Venditti EM, et al. Translating the Diabetes Prevention Program: a comprehensivemodel for prevention training and program delivery. Am J Prev Med. 2009;37(6):505-511. doi:10.1016/j.amepre.2009.07.020

31. Bandura A. Social Foundations of Thought and Action: A Social Cognitive Theory. Prentice Hall; 1986.

32. Thorpe KE, Zwarenstein M, Oxman AD, et al. A pragmatic-explanatory continuum indicator summary(PRECIS): a tool to help trial designers. CMAJ. 2009;180(10):E47-E57. doi:10.1503/cmaj.090523

33. Ober C, Hoffjan S. Asthma genetics 2006: the long and winding road to gene discovery. Genes Immun. 2006;7(2):95-100. doi:10.1038/sj.gene.6364284

34. Thomas PR, ed. Committee to Develop Criteria for Evaluating the Outcomes of Approaches to Prevent and TreatObesity: Weighing the Options: Criteria for Evaluating Weight-Management Programs. National Academy Press;1995. doi:10.1037/10510-000

35. NHLBI Obesity Education Initiative Expert Panel on the Identification, Evaluation, and Treatment of Obesity inAdults. Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: TheEvidence Report. National Heart, Lung, and Blood Institute; 1998.

36. Pickering TG, Hall JE, Appel LJ, et al; Subcommittee of Professional and Public Education of the AmericanHeart Association Council on High Blood Pressure Research. Recommendations for blood pressure measurementin humans and experimental animals: part 1: blood pressure measurement in humans: a statement forprofessionals from the Subcommittee of Professional and Public Education of the American Heart AssociationCouncil on High Blood Pressure Research. Hypertension. 2005;45(1):142-161. doi:10.1161/01.HYP.0000150859.47929.8e

37. Measures from the PhenX Toolkit version 4.2. Accessed November 16, 2020. https://www.phenxtoolkit.org/

38. Karlsson J, Taft C, Sjöström L, Torgerson JS, Sullivan M. Psychosocial functioning in the obese before and afterweight reduction: construct validity and responsiveness of the Obesity-related Problems scale. Int J Obes RelatMetab Disord. 2003;27(5):617-630. doi:10.1038/sj.ijo.0802272

39. Bilbao A, Mar J, Mar B, Arrospide A, Martínez de Aragón G, Quintana JM. Validation of the Spanish translationof the questionnaire for the obesity-related problems scale. Obes Surg. 2009;19(10):1393-1400. doi:10.1007/s11695-009-9800-3

40. Huang FY, Chung H, Kroenke K, Delucchi KL, Spitzer RL. Using the Patient Health Questionnaire-9 to measuredepression among racially and ethnically diverse primary care patients. J Gen Intern Med. 2006;21(6):547-552.doi:10.1111/j.1525-1497.2006.00409.x

41. Blair SN, Haskell WL, Ho P, et al. Assessment of habitual physical activity by a seven-day recall in a communitysurvey and controlled experiments. Am J Epidemiol. 1985;122(5):794-804. doi:10.1093/oxfordjournals.aje.a114163

42. Conway JM, Ingwersen LA, Vinyard BT, Moshfegh AJ. Effectiveness of the US Department of Agriculture5-step multiple-pass method in assessing food intake in obese and nonobese women. Am J Clin Nutr. 2003;77(5):1171-1178. doi:10.1093/ajcn/77.5.1171

43. Conway JM, Ingwersen LA, Moshfegh AJ. Accuracy of dietary recall using the USDA five-step multiple-passmethod in men: an observational validation study. J Am Diet Assoc. 2004;104(4):595-603. doi:10.1016/j.jada.2004.01.007

44. Sallis JF, Haskell WL, Wood PD, et al. Physical activity assessment methodology in the Five-City Project. Am JEpidemiol. 1985;121(1):91-106. doi:10.1093/oxfordjournals.aje.a113987




https://dx.doi.org/10.3122/jabfm.2018.03.170280

https://dx.doi.org/10.1016/j.cct.2016.03.003

https://dx.doi.org/10.1161/CIRCULATIONAHA.105.169404

https://dx.doi.org/10.1161/CIRCULATIONAHA.105.169404

https://dx.doi.org/10.1177/1740774510391683

https://dx.doi.org/10.1177/07399863960183002



https://dx.doi.org/10.1503/cmaj.090523

https://dx.doi.org/10.1038/sj.gene.6364284

https://dx.doi.org/10.1037/10510-000

https://dx.doi.org/10.1161/01.HYP.0000150859.47929.8e

https://dx.doi.org/10.1161/01.HYP.0000150859.47929.8e

https://www.phenxtoolkit.org/

https://dx.doi.org/10.1038/sj.ijo.0802272

https://dx.doi.org/10.1007/s11695-009-9800-3

https://dx.doi.org/10.1007/s11695-009-9800-3

https://dx.doi.org/10.1111/j.1525-1497.2006.00409.x

https://dx.doi.org/10.1093/oxfordjournals.aje.a114163

https://dx.doi.org/10.1093/ajcn/77.5.1171

https://dx.doi.org/10.1016/j.jada.2004.01.007

https://dx.doi.org/10.1016/j.jada.2004.01.007

https://dx.doi.org/10.1093/oxfordjournals.aje.a113987

45. Wilson PW, Paffenbarger RS Jr, Morris JN, Havlik RJ. Assessment methods for physical activity and physicalfitness in population studies: report of a NHLBI workshop. Am Heart J. 1986;111(6):1177-1192. doi:10.1016/0002-8703(86)90022-0

46. Fung TT, Chiuve SE, McCullough ML, Rexrode KM, Logroscino G, Hu FB. Adherence to a DASH-style diet andrisk of coronary heart disease and stroke in women. Arch Intern Med. 2008;168(7):713-720. doi:10.1001/archinte.168.7.713

47. Sacks FM, Obarzanek E, Windhauser MM, et al. Rationale and design of the Dietary Approaches to StopHypertension trial (DASH): a multicenter controlled-feeding study of dietary patterns to lower blood pressure. AnnEpidemiol. 1995;5(2):108-118. doi:10.1016/1047-2797(94)00055-X

48. Sacks FM, Svetkey LP, Vollmer WM, et al; DASH-Sodium Collaborative Research Group. Effects on bloodpressure of reduced dietary sodium and the Dietary Approaches to Stop Hypertension (DASH) diet. N Engl J Med.2001;344(1):3-10. doi:10.1056/NEJM200101043440101

49. Xiao L, Lv N, Rosas LG, Au D, Ma J. Validation of clinic weights from electronic health records againststandardized weight measurements in weight loss trials. Obesity (Silver Spring). 2017;25(2):363-369. doi:10.1002/oby.21737

50. DiCiccio TJ, Efron B. Bootstrap confidence intervals. Statistic Sci. 1996;11(3):189-228. doi:10.1214/ss/1032280214

51. Ma J, Yank V, Xiao L, et al. Translating the Diabetes Prevention Program lifestyle intervention for weight lossinto primary care: a randomized trial. JAMA Intern Med. 2013;173(2):113-121. doi:10.1001/2013.jamainternmed.987

52. Diabetes Prevention Program (DPP) Research Group. The Diabetes Prevention Program (DPP): descriptionof lifestyle intervention. Diabetes Care. 2002;25(12):2165-2171. doi:10.2337/diacare.25.12.2165

53. Wadden TA, West DS, Delahanty L, et al; Look AHEAD Research Group. The Look AHEAD study: a descriptionof the lifestyle intervention and the evidence supporting it. Obesity (Silver Spring). 2006;14(5):737-752. doi:10.1038/oby.2006.84

54. Katzmarzyk PT, Martin CK, Newton RL Jr, et al. Weight loss in underserved patients—a cluster-randomizedtrial. N Engl J Med. 2020;383(10):909-918. doi:10.1056/NEJMoa2007448

55. Wadden TA, Butryn ML, Hong PS, Tsai AG. Behavioral treatment of obesity in patients encountered in primarycare settings: a systematic review. JAMA. 2014;312(17):1779-1791. doi:10.1001/jama.2014.14173

56. Wadden TA, West DS, Neiberg RH, et al; Look AHEAD Research Group. One-year weight losses in the LookAHEAD study: factors associated with success. Obesity (Silver Spring). 2009;17(4):713-722. doi:10.1038/oby.2008.637

57. Venditti EM, Bray GA, Carrion-Petersen ML, et al; Diabetes Prevention Program Research Group. First versusrepeat treatment with a lifestyle intervention program: attendance and weight loss outcomes. Int J Obes (Lond).2008;32(10):1537-1544. doi:10.1038/ijo.2008.134

58. Ely EK, Gruss SM, Luman ET, et al. A national effort to prevent type 2 diabetes: participant-level evaluation ofCDC’s National Diabetes Prevention Program. Diabetes Care. 2017;40(10):1331-1341. doi:10.2337/dc16-2099

SUPPLEMENT 1.Trial Protocol

SUPPLEMENT 2.eTable 1. Sensitivity Analysis of Primary Outcome Using Study-Measured WeightseTable 2. Sensitivity Analysis Using Bootstrap Results for the Treatment Effects on Weight Changes From Baselineat 12 and 24 Months

SUPPLEMENT 3.Data Sharing Statement




https://dx.doi.org/10.1016/0002-8703(86)90022-0

https://dx.doi.org/10.1016/0002-8703(86)90022-0

https://jama.jamanetwork.com/article.aspx?doi=10.1001/archinte.168.7.713&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2020.27744

https://jama.jamanetwork.com/article.aspx?doi=10.1001/archinte.168.7.713&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2020.27744

https://dx.doi.org/10.1016/1047-2797(94)00055-X

https://dx.doi.org/10.1056/NEJM200101043440101

https://dx.doi.org/10.1002/oby.21737

https://dx.doi.org/10.1002/oby.21737

https://dx.doi.org/10.1214/ss/1032280214

https://dx.doi.org/10.1214/ss/1032280214

https://jama.jamanetwork.com/article.aspx?doi=10.1001/2013.jamainternmed.987&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamanetworkopen.2020.27744

https://dx.doi.org/10.2337/diacare.25.12.2165

https://dx.doi.org/10.1038/oby.2006.84


https://dx.doi.org/10.1056/NEJMoa2007448




https://dx.doi.org/10.1038/ijo.2008.134


originalinvestigation | nutrition,obesity,andexercise

Documents