One-Year Results of a Community-BasedTranslation of the Diabetes PreventionProgramHealthy-Living Partnerships to Prevent Diabetes (HELP PD) Project

JEFFREY A. KATULA, PHD1

MARA Z. VITOLINS, DRPH2

ERICA L. ROSENBERGER, MS2

CAROLINE S. BLACKWELL, BS2

TIMOTHY M. MORGAN, PHD3

MICHAEL S. LAWLOR, PHD4

DAVID C. GOFF JR., MD, PHD2

OBJECTIVE—Although the Diabetes Prevention Program (DPP) and the Finnish DiabetesPrevention Study (FDPS) demonstrated that weight loss from lifestyle change reduces type 2diabetes incidence in patients with prediabetes, the translation into community settings has beendifficult. The objective of this study is to report the first-year results of a community-basedtranslation of the DPP lifestyle weight loss (LWL) intervention on fasting glucose, insulin re-sistance, and adiposity.

RESEARCH DESIGN AND METHODS—We randomly assigned 301 overweight andobese volunteers (BMI 25–40 kg/m2) with fasting blood glucose values between 95 and 125mg/dL to a group-based translation of the DPP LWL intervention administered through a di-abetes education program (DEP) and delivered by community health workers (CHWs) or to anenhanced usual-care condition. CHWs were volunteers with well-controlled type 2 diabetes. Atotal of 42.5% of participants were male, mean age was 57.9 years, 26% were of a race/ethnicityother than white, and 80% reported having an education beyond high school. The primaryoutcome is mean fasting glucose over 12 months of follow-up, adjusting for baseline glucose.

RESULTS—Compared with usual-care participants, LWL intervention participants experi-enced significantly greater decreases in blood glucose (24.3 vs.20.4 mg/dL; P, 0.001), insulin(26.5 vs.22.7 mU/mL; P, 0.001), homeostasis model assessment of insulin resistance (21.9vs. 20.8; P , 0.001), weight (27.1 vs. 21.4 kg; P , 0.001), BMI (22.1 vs. 20.3 kg/m2; P ,0.001), and waist circumference (25.9 vs. 20.8 cm; P , 0.001).

CONCLUSIONS—This translation of the DPP intervention conducted in community set-tings, administered through a DEP, and delivered by CHWs holds great promise for the pre-vention of diabetes by significantly decreasing glucose, insulin, and adiposity.

Diabetes Care 34:1451–1457, 2011

The Diabetes Prevention Program(DPP) and the Finnish Diabetes Pre-vention Study (FDPS) demonstrated

that the incidence of type 2 diabetes couldbe reduced by almost 60% in patients withprediabetes through weight loss resultingfrom changes in diet and physical activity

(1,2). Despite these promising findings,the prevalence of type 2 diabetes contin-ues to increase (3). Furthermore, althoughdiabetes mortality has declined 8.3% inthe last decade, diabetes-related compli-cations continue to increase, resulting inrising disease burden (4).

Several recent translations (5–13) ofthe DPP and FDPS have demonstrated en-couraging effects across diverse settings, in-cluding primary care settings (6,9,11–13),cardiac rehabilitation programs (10),churches (8), YMCAs (5), and health carefacilities (7). The personnel who imple-mented the intervention included nurses(6,11,12), registered dietitians and ex-ercise physiologists (7,10), health careprofessionals (9,13), volunteer medicalpersonnel (8), and YMCA trainers (5).Evaluating these translated interventionsis challenging because of the differencesin outcome measures, sample sizes, andstudy designs. Although these interven-tions typically produced ~6% weightloss, the studies tended to have small sam-ple sizes, often did not include compari-son conditions, and lacked randomizedallocation to treatment conditions. Notranslational study to date has reportedsignificant reductions in blood glucose inindividuals with prediabetes. Therefore,additional translational research is neededthat uses innovative, cost-effective systemsto deliver effective lifestyle interventionstargeting patients at risk for diabetes. Nu-merous studies have tested the use of com-munity health workers (CHWs) in themanagement of diabetes (14), but no stud-ies to date have tested the use of CHWs inimplementing lifestyle interventions de-signed to prevent diabetes.

This report presents the first-year re-sults of the Healthy-Living Partnershipsto Prevent Diabetes (HELP PD) Project onglucose, insulin resistance, and adiposity(15). The HELP PD project was designedto translate the methods of the DPP intothe community via key modifications toenhance feasibility and dissemination: thedelivery of a group-based lifestyle weightloss (LWL) intervention via a partnershipbetween an existing community-based dia-betes education program (DEP) andCHWs.

RESEARCH DESIGN ANDMETHODS—The design,methods (15),recruitment procedures, and participantcharacteristics (16) have been described

c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c

From the 1Department of Health and Exercise Science, Wake Forest University, Winston-Salem, NorthCarolina; the 2Department of Epidemiology and Prevention, Wake Forest School of Medicine, Winston-Salem, North Carolina; the 3Department of Biostatistics, Wake Forest School of Medicine, Winston-Salem,North Carolina; and the 4Department of Economics, Wake Forest University, Winston-Salem, NorthCarolina.

Corresponding author: Jeffrey A. Katula, katulaj@wfu.edu.Received 9 November 2010 and accepted 12 April 2011.DOI: 10.2337/dc10-2115. Clinical trial reg. no. NCT00631345, clinicaltrials.gov.This article contains Supplementary Data online at http://care.diabetesjournals.org/lookup/suppl/doi:10.

2337/dc10-2115/-/DC1.© 2011 by the American Diabetes Association. Readers may use this article as long as the work is properly

cited, the use is educational and not for profit, and thework is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ for details.

care.diabetesjournals.org DIABETES CARE, VOLUME 34, JULY 2011 1451

C l i n i c a l C a r e / E d u c a t i o n / N u t r i t i o n / P s y c h o s o c i a l R e s e a r c hO R I G I N A L A R T I C L E

elsewhere. In brief, the HELP PD projectis a randomized controlled trial that teststhe effect of a community-based transla-tion of the DPP LWL intervention versusa comparison condition of enhanced usualcare on fasting blood glucose, other met-abolic outcomes, adiposity, psychosocialvariables, health-related quality of life,and health care costs and use. Data werecollected at baseline and every 6 monthsup to 24 months of follow-up. The insti-tutional review board of Wake ForestBaptist Medical Center approved thestudy, and all participants providedsigned informed consent.

The eligibility criteria targeted a sam-ple at risk for diabetes representative ofthe local community. Recruitment (16)was accomplished primarily throughmass mailings to targeted ZIP codes. In-terested individuals contacted a studytelephone number and were providedbasic information about participation, in-cluding additional steps in the screeningand randomization process and the timecommitment and expectations associatedwith participation before completing atelephone-screening questionnaire. Par-ticipants were required to have evidenceof prediabetes on two occasions, with aconfirmatory fasting glucose between 95and 125 mg/dL, and to have a BMI$25.0kg/m2 and#39.9 kg/m2 (17). Candidateswere screened for other comorbid condi-tions that would make physical activityunsafe or limit participation in the study.These conditions included recent historyof an acute cardiovascular disease event,clinical history of type 2 diabetes, uncon-trolled hypertension, cancer or other con-ditions limiting life expectancy, chronicuse of medicines known to influence glu-cose metabolism (i.e., corticosteroids),and major psychiatric or cognitive prob-lems, including depression. The PhysicalActivity Readiness Questionnaire wasadministered to identify people with con-traindications to exercise, who were re-quired to obtain a medical clearancefrom their physician prior to randomiza-tion. Participation in a supervised pro-gram for weight loss or another researchstudy that would interfere with HELP PDalso was an exclusion criterion.

Outcome measuresFasting blood glucose, insulin, and an-thropometry were assessed at baseline andrandomization and at the 6- and 12-monthvisits by trained study staff. Phlebotomywas performed after at least an 8-h fast,in accordance with American Diabetes

Association guidelines (18). All biochem-ical measurements were performed in acentral laboratory by technicians maskedto the intervention assignment.

Fasting glucose, the primary outcomemeasure, was measured using a timed–end point method supplied by BeckmanCoulter for the Synchron LX Analyzer, amethod accepted by the Centers for Dis-ease Control and Prevention. Within-runcoefficients of variation for this methodwere#3.9% and total coefficients of var-iation were #6.45%.

Insulin was assayed using the para-magnetic particle chemiluminescent im-munoassay for the Access ImmunoassaySystems (Beckman Coulter). There is,0.3% cross-reactivity with human pro-insulin and no detectable cross-reactivitywith human C-peptide. Low- and high-level human serum quality-control sam-ples were run during each 24-h timeperiod. The overall within-assay variabil-ity was 3.9%, and the between-assay var-iability was 5.5%.

Insulin resistance was examined us-ing the homeostasis model assessment ofinsulin resistance (HOMA-IR) index(HOMA IR = [{fasting insulin 3 fastingglucose}/22.5]) (19). HOMA-IR is a bettermeasure of insulin resistance than is fast-ing insulin alone and is highly correlatedwith other, more complex and invasivemeasures of insulin resistance from thefrequently sampled intravenous glucosetolerance test and the euglycemic clamp(20,19).

Anthropometric measurements weretaken with participants wearing light-weight clothing and without shoes. Mea-surements were taken twice during eachexam, and means were used in analyses.Seca Accu-Hite wall-mounted stadio-meters were used to measure height tothe nearest 0.5 cm. Weight was measuredusing a Cardinal Detecto digital scale (758C Series) to the nearest 0.1 kg. BMI wascalculated as weight in kilograms dividedby the square of height in meters. Waistcircumference was assessed, to the nearest0.1 cm, using a Gulick II 150-cm anthro-pometric tape with the participant in arecumbent position (21).

Community-based implementationOur translation of the DPP involvedconducting the LWL intervention incommunity-based sites via a local DEPand CHWs. Study investigators and staffconducted study administration, but theregistered dietitians employed by the DEPmanaged the day-to-day operations of the

intervention as well as the training andmonitoring of the CHWs. CHWs werecommunity members with type 2 diabe-tes, well-controlled HbA1c, and a historyof healthy eating and physical activity.CHWs were recruited through our DEPby the study investigators and registereddieticians. CHWs were responsible forconducting the intervention group ses-sions, managing participants, and en-tering data on each participant’s bodyweights obtained at group sessions.CHWs were compensated $100 per weekduring the first 6 months for weekly ses-sions and $200 per month during the sec-ond 6 months for monthly sessions. CHWtraining consisted of a 36-h program con-ducted over the course of 6–9 weeks andinvolved experiential learning, didactic in-struction, peer mentoring, and observa-tion. Ten CHWs were trained in twogroups of five; one group before recruit-ment started and another group 4 monthsinto recruitment.

LWL interventionThe LWL intervention targeted decreasedcaloric intake (goal of 1,200–1,800 kcalper day) and increased caloric expendi-ture through moderate physical activity(goal $180 min per week). The primaryobjective was to produce a total weightloss of 5–7% during the first 6 months oftreatment. During the second 6 months,participants were encouraged to continueto meet or maintain their weight lossgoals as long as their BMI did not fall be-low 20 kg/m2. This approach was consis-tent with the recommendations of theAmerican Diabetes Association, the NorthAmerican Association for the Study ofObesity, and the American Society forClinical Nutrition (22).

Participants met weekly for CHW-ledgroup sessions during the first 6 months.Fourteen different groups of 8–12 partic-ipants were conducted at various com-munity sites (e.g., parks and recreationcenters). Participants also received threepersonalized consultations with a regis-tered dietician (during months 1, 3, and6). During months 7–12, participants re-ceived two scheduled contacts with theCHW each month, one group sessionand one telephone contact. Interventioncontent standardization was supportedby a DVD series developed by the researchteam to cover nutrition and physical ac-tivity basics, energy balance, healthy eat-ing, goal setting, and problem solving. Wealso included presentations from localcommunity experts (e.g., the YMCA, local

1452 DIABETES CARE, VOLUME 34, JULY 2011 care.diabetesjournals.org

HELP PD: 1-year results

grocery stores, and specialty athletic foot-wear stores).

Enhanced usual careThe usual-care condition was designed toexceed the usual care provided to patientswith prediabetes and to enhance reten-tion. Usual care consisted of two individ-ual sessions with a nutritionist during thefirst 3 months involving healthy eatingand physical activity education to supportweight loss. Usual-care participants alsoreceived a monthly newsletter with topicsrelated to healthy lifestyles and informa-tion about community resources.

PowerThe HELP PD project targeted the re-cruitment of 300 participants. The DPPfound a 4 mg/dL difference in fasting glu-cose at 2 years. The correlation betweenglucose values at baseline and follow-upwas estimated to be between 0.45 and0.6 in two studies, with larger ranges inbaseline glucose than found in our study.Therefore, we assumed a more modestcorrelation of r = 0.2. The DPP reporteda cross-sectional SD of 8.3 mg/dL (1). Theestimated SD for a follow-up measureafter adjusting for the baseline value is8.3 3 the square root of (1 – 0.22) =8.132. This value of the adjusted SD wasused in the NQuery program to deter-mine sample size in a two-group trial.The trial was designed to have at least80–90% power to detect differences infasting glucose of 3–3.5 mg/dL. The re-quired number of evaluable subjects pergroup to have 80% power to detect a dif-ference of 3 mg/dL is 117. The requirednumber of evaluable subjects per groupto have 90% power to detect a differenceof 3.5 mg/dL is 115. Allowing for up to80% loss to follow-up over 2 years, thetotal required number of subjects neededto be randomly assigned to meet thesetwo goals is 294 and 288, respectively.With a total sample size of 300 subjectsrandomly assigned, we estimate that thetrial has 81% power to detect a 3 mg/dLdifference between groups and 91%power to detect a 3.5 mg/dL difference.

Data analysisAlthough the study was originally de-signed with 24 months of follow-up, theanalyses reported here are based on thefirst 12 months. To compare our resultswith those of other diabetes preventiontranslational studies, the vast majority ofwhich report 12-month outcomes, the in-vestigators received approval to unblind

the first-year results and amend the pro-tocol to conduct independent analysesof results in the first and second yearsof follow-up rather than the originally

planned approach of analyzing the 2 yearstogether.

Our randomization procedure in-volved a permuted block design with

Figure 1—Flow of participants from screening to completion of the final follow-up assessment.

care.diabetesjournals.org DIABETES CARE, VOLUME 34, JULY 2011 1453

Katula and Associates

varying block size to reduce the timerandomly assigned individuals had towait for intervention groups to developand ensure a degree of balance. Arith-metic means and SDs are presented forcontinuous variables. For baseline com-parisons between the LWL interventionand usual-care groups, t tests were usedfor continuous variables and the Fisherexact test was used for categorical vari-ables. The primary hypothesis for thisreport involved comparing the main ef-fect of the LWL intervention versus theusual-care control on fasting glucose. Weused general linear models for repeated-measures ANCOVA (SAS PROC MIXED)to compare the main effect of the interven-tion on the 6- and 12-month values mea-sured during the 1-year follow-up period.For each outcome measure, the baselinevalue for that outcome was used as a co-variate. The test for intervention effect isbased on the least square means of theestimated main effect of intervention(estimate of the average of the 6- and12-month means). We tested for groupand visit interactions and none were sig-nificant. We used the intention-to-treatapproach and included all postrandom-ized values according to the group towhich they were assigned. We performedsecondary analyses, making a reasonableexception to this rule by deleting obser-vations at visits where the subject was tak-ing hypoglycemic medication. Inferences

for comparisons were tested at a 5% two-sided level of significance. An unstructuredvariance model was used for intrasubjectlongitudinal covariance.

RESULTS—The participant screening,enrollment, and follow-up experience isdepicted in Fig. 1. The 301 participantsincluded a relatively well-educated, bi-racial group of middle-aged and oldermen and women with an average BMI of32.8 kg/m2 (Table 1). Over the first yearof the intervention, participants attended67.7% of all potential group interventionsessions. Including make-up visits (eitherin person or via telephone), participantscompleted 79.4% of all potential ses-sions. During the first 6 months, 72.4%of planned sessions were attended and11.0% were made up with a CHW (com-bined attendance = 83.4%). In months7–12, 49.2% of planned sessions were at-tended and 13.9% were made up (com-bined attendance = 63.1%). Informationwas available regarding the primary out-come, fasting glucose, for 280 (93.4%)participants at month 6 and 273 (90.2%)participants at month 12, with no sub-stantive difference in data completenessbetween groups.

Table 2 displays changes in measuresof adiposity, fasting glucose, and insulinresistance by treatment group. Relative tousual-care participants, LWL interventionparticipants lost a net of 6.0% of their

body weight and 5.0 cm in waist circum-ference (P , 0.001 for both compari-sons). Similar results were observed forabsolute weight loss and BMI.

On the basis of the adjusted averageof the 6- and 12-month means, fastingglucose decreased by 4.3 mg/dL in theLWL intervention participants versus adecrease of 0.4 mg/dL in the usual-careparticipants (P , 0.001). In secondaryanalyses that examined 6- and 12-monthassessments separately, fasting glucosehad decreased by 4.0 mg/dL in the LWLintervention versus an increase of 1.0mg/dL in the usual-care intervention(P , 0.001) at 6 months. At 12 months,fasting glucose had decreased by 4.6mg/dL in the LWL intervention versus adecrease of 1.8 mg/dL in the usual-careintervention (P = 0.02). Relative to usual-care participants, LWL intervention par-ticipants also demonstrated substantialdecreases in fasting insulin and HOMA-IR (P , 0.001 for both comparisons).

Supplementary Table A1 containsdata on adverse events, serious adverseevents, and the number of new cases ofdiabetes by treatment group. No differ-ences were found in these events betweentreatment groups.

CONCLUSIONS—These results indi-cate that the HELP PD project was suc-cessful in translating the DPP LWLintervention into a community-based ap-proach that induced significant reductionsin blood glucose, insulin, HOMA-IR, bodyweight, waist circumference, and BMI over12 months in overweight and obese pa-tients with prediabetes. To our knowledge,this study is the largest to date to success-fully translate the DPP into the communityand is the only DPP translational study todocument significant changes in fastingblood glucose, insulin, and insulin resis-tance using a randomized controlled de-sign.

The glucose-lowering effect of theHELP PD intervention compares favor-ably with the effects documented in theDPP and FDPS (24 and 25 mg/dL, re-spectively, albeit over longer follow-upperiods [DPP follow-up = 2.8 years andFDPS follow-up = 3.2 years]). Our futureanalyses will examine the sustainability ofthe HELP PD effect over 24 months. Theeffect of the HELP PD intervention on ad-iposity also compares favorably to otherstudies. A review (23) of nine randomizedcontrolled trials examining weight lossinterventions in individuals with pre-diabetes reported a pooled estimate of

Table 1—Baseline sample characteristics

Variable LWL intervention Usual care Total P*

n 151 150 301Sex 1.00Male 64 (42.4) 64 (42.7) 128 (42.5)Female 87 (57.6) 86 (57.3) 173 (57.5)

Race 0.43African American 39 (25.8) 35 (23.3) 74 (24.6)White 111 (73.5) 111 (74.0) 222 (73.8)Other/refused 1 (0.7) 4 (2.7) 5 (1.6)

Age (years) 57.3 6 10.1 58.5 6 9.0 57.9 6 9.5 0.28Educational attainment 0.97High school or less 29 (19.2) 32 (21.3) 61 (20.3)Associate degree or other 49 (32.5) 47 (31.3) 96 (31.9)Bachelor’s degree 37 (24.5) 37 (24.7) 74 (24.6)Beyond Bachelor’s degree 36 (23.8) 34 (22.7) 70 (23.3)

Weight (kg) 94.4 6 14.7 93.0 6 16.2 93.7 6 15.5 0.44BMI (kg/m2) 32.8 6 3.9 32.6 6 4.1 32.7 6 4.0 0.54Glucose (mg/dL) 105.4 6 12.5 105.7 6 10.0 105.5 6 11.3 0.79Insulin (mU/mL) 16.7 6 9.7 16.7 6 10.0 16.7 6 9.8 0.95HOMA-IR 4.4 6 3.0 4.4 6 2.9 4.4 6 2.9 0.99Data are n (%) or means6 SD. *t Tests were used for continuous variables, and the Fisher exact test was usedfor categorical variables.

1454 DIABETES CARE, VOLUME 34, JULY 2011 care.diabetesjournals.org

HELP PD: 1-year results

22.8 kg of weight loss (23.3% of ini-tial body weight) calculated from fourstudies with 1 year of follow-up. TheDPP (1) and FDPS (2) reported weightloss of 25.5 kg (24.9%) and 24.2 kg(24.7%), respectively. In addition, theDPP documented a weight loss of26.0 kg at 12 months of follow-up. Par-ticipants in the HELP PD LWL interven-tion lost an average of 27.1 kg (7.3%) at12 months. However, trials consistentlyshow that approximately one-third ofthe weight lost during the first 6 monthsof behavioral weight loss interventions istypically regained by 1 year, and weightreturns to baseline in 3–5 years (24). Ourfuture analyses will determine whetherweight loss is sustained over 24 monthsof follow-up.

Previous diabetes prevention transla-tional studies have reported 12-monthweight losses ranging from 20.45 kg (8)to 25.7 kg (5), but other than Boltriet al. (8), none have reported significantchanges in fasting blood glucose. The

DEPLOY (Diabetes Education and Pre-vention with a Lifestyle Intervention Of-fered at the YMCA) study (5) delivered agroup-based translation of the DPP LWLintervention via YMCAs to 77 participantswith elevated glucose and reported sig-nificant decreases in weight (25.7 kg)and BMI (26.7%) but no differencesbetween groups in changes in cardiome-tabolic outcomes (e.g., HbA1c). Likewise,the Weight Loss through Living Well(WiLLoW) study (11), using a non-randomized controlled cohort design,delivered a group-based DPP LWL in-tervention to overweight patients viaprimary care practices and reported a12-month weight change of –5.7 kg butdid not report changes in cardiometabolicoutcomes. Kramer et al (9) assessed theimpact of a group-based DPP LWL inter-vention in patients at risk for diabetes (n =42) delivered via primary care practicesusing a one-group design and reportedsignificant 12-month changes in weight(24.2 kg [24.5%]), waist circumference

(27.1 cm [26.8%]), and BMI (21.6[24.8%]), but changes in glucose (21.5mg/dL [21.4%]) were not significant.At a national level, Saaristo et al. (13) im-plemented the FDPS lifestyle interven-tion in 2,798 individuals with elevatedrisk for diabetes in 400 primary care set-tings and reported a mean of 21.2 kg ofweight loss. Although this study did notreport changes in fasting glucose, the re-sults indicate that the reduction in inci-dence of diabetes (assessed using an oralglucose tolerance test) was strongly re-lated to weight loss (relative risk of diabe-tes was 0.31 in the group who lost 5%weight, 0.72 in the group who lost 2.5–4.9% weight, and 1.10 in the group whogained 2.5% compared with the groupwho maintained weight). The results pre-sented here suggest that the HELP PDproject approach may be a more powerfultranslation of the DPP than previouslypublished approaches.

Fasting insulin and HOMA-IR, twomeasures of insulin resistance, responded

Table 2—Measures of adiposity and metabolic outcomes of HELP PD at 6 and 12 months by treatment group

Variable Baseline 6 months 12 monthsAdjusted means overthe follow-up period* P

Weight (kg)Control group 92.67 6 1.37 91.55 6 1.38 90.93 6 1.37 92.12 6 0.29Intervention group 94.41 6 1.24 87.14 6 1.22 87.44 6 1.28 86.39 6 0.30Difference 25.73 6 0.42 ,0.001

BMI (kg/m2)Control group 32.42 6 0.35 31.96 6 0.34 31.95 6 0.36 32.15 6 0.10Intervention group 32.81 6 0.34 30.38 6 0.35 30.52 6 0.36 30.25 6 0.10Difference 21.90 6 0.14 ,0.001

Waist (cm)Control group 104.22 6 0.91 103.37 6 0.90 103.45 6 0.89 103.70 6 0.27Intervention group 104.83 6 0.79 98.67 6 0.84 99.22 6 0.90 98.65 6 0.27Difference 25.05 6 0.38 ,0.001

Glucose (mg/dL)Control group 105.92 6 0.83 106.97 6 0.77 104.16 6 0.99 105.36 6 0.53Intervention group 105.58 6 1.07 101.69 6 0.81 101.11 6 0.84 101.60 6 0.54Difference 23.76 6 0.76 ,0.001

Insulin (mU/mL)Control group 16.92 6 0.84 15.54 6 1.07 12.81 6 0.64 14.09 6 0.41Intervention group 16.73 6 0.82 10.73 6 0.61 9.76 6 0.46 10.34 6 0.41Difference 23.75 6 0.58 ,0.001

HOMA-IRControl group 4.50 6 0.24 4.18 6 0.30 3.33 6 0.18 3.73 6 0.12Intervention group 4.48 6 0.26 2.77 6 0.18 2.48 6 0.13 2.65 6 0.12Difference 21.08 6 0.17 ,0.001

Percentage weight lossControl group 0.00 6 0.00 21.18 6 0.29 21.33 6 0.39 21.25 6 0.31Intervention group 0.00 6 0.00 27.52 6 0.48 27.21 6 0.57 27.37 6 0.31Difference 26.11 6 0.44 ,0.001

Data are means 6 SE. *Least square means from a repeated-measures ANCOVA using the baseline value as a covariate; P values represent the between-groupcomparison of the average of the 6- and 12-month means.

care.diabetesjournals.org DIABETES CARE, VOLUME 34, JULY 2011 1455

Katula and Associates

favorably to the intervention. Insulin re-sistance is a key link between obesity andthe risk of both diabetes and cardiovas-cular disease (25). The beneficial effectson insulin resistance provide support forthe hypothesis that the intervention mayhave beneficial effects on the risk for car-diovascular disease beyond the effects onblood glucose.

The HELP PD project made severalkey modifications to the DPP interventionto create a model that can be translatedfor use in any community with a DEP. Theintervention was delivered in a groupformat by CHWs in community-basedsettings and was overseen by registereddieticians employed by a local DEP,thereby minimizing the contributions ofresearch resources and maximizing theresponsibilities of community-based staff.The results of the current study indicatethat this model is effective at inducingmeaningful metabolic changes in individ-uals at high risk for diabetes.

Although the .3,000 American Dia-betes Association–recognized DEPs inthe U.S. are well positioned to implementthis intervention, several limitations exist.First, the HELP PD project was conductedin only one community located in thesoutheastern U.S. It is unknown whetherthis approach can be effectively dissemi-nated to other communities. In addition,a training program must be developed toprepare personnel. Finally, reimburse-ment policies must be developed tosupport the cost of program delivery.Economic analysis of the HELP PD pro-gram is underway and may help informpolicy development. Despite these limita-tions, the results of the HELP PD programindicate that empowering communitymembers through partnerships with ex-isting DEPs may effectively translate dia-betes prevention efforts and ultimatelyalter the course of the obesity and diabetesepidemics.

Acknowledgments—This study was fundedby a grant from the National Institute of Di-abetes and Digestive and Kidney Diseases(R18-DK-69901). The funding source had norole in the design and conduct of the study;collection, management, analysis, and inter-pretation of the data; and preparation, review,or approval of the manuscript.D.C.G. is a consultant for a trial testing a

glucose-lowering medication funded by Merckand a DSMBmember for a trial testing a glucose-loweringmedication funded by Takida. No otherpotential conflicts of interest relevant to thisarticle were reported.

J.A.K. researched data and wrote the man-uscript. M.Z.V., E.L.R., and C.S.B. researcheddata and reviewed and edited the manuscript.T.M.M. conducted data analyses, researcheddata, and wrote the manuscript. M.S.L. re-searched data and reviewed and edited themanuscript. D.C.G. researched data and wrotethe manuscript.The authors are grateful for the con-

tributions of Mark A. Espeland, PhD (studydesign and data analysis), Wake Forest Uni-versity Baptist Medical Center; W. Jack Rejeski,PhD (study design), Wake Forest University;Melicia C. Whitt-Glover, PhD (study design),Wake Forest University Baptist Medical Center;Jorge C. Escandon, MD (study design), WakeForest University Baptist Medical Center;Carolyn F. Pedley, MD (study design), WakeForest University Baptist Medical Center; KaraA. Foster (recruitment anddata collection),WakeForest University Baptist Medical Center; EileenSearson (recruitment), Wake Forest UniversityBaptist Medical Center; Terry Tembreull (datacollection), Wake Forest University BaptistMedical Center; Wesley Roberson (data anal-ysis), Wake Forest University Baptist MedicalCenter; Scott Isom (data analysis),Wake ForestUniversity Baptist Medical Center; and RhondaSpach (study administration), Wake ForestUniversity Baptist Medical Center.

References1. Knowler WC, Barrett-Connor E, Fowler

SE, et al.; Diabetes Prevention ProgramResearch Group. Reduction in the inci-dence of type 2 diabetes with lifestyle in-tervention or metformin. N Engl J Med2002;346:393–403

2. Tuomilehto J, Lindström J, Eriksson JG,et al.; Finnish Diabetes Prevention StudyGroup. Prevention of type 2 diabetes mel-litus by changes in lifestyle among subjectswith impaired glucose tolerance. N Engl JMed 2001;344:1343–1350

3. Cowie CC, Rust KF, Byrd-Holt DD, et al.Prevalence of diabetes and impaired fast-ing glucose in adults in the U.S. pop-ulation: National Health And NutritionExamination Survey 1999-2002. DiabetesCare 2006;29:1263–1268

4. Sloan FA, Bethel MA, Ruiz D, Shea AH,Feinglos MN. The growing burden ofdiabetes mellitus in the US elderly pop-ulation. Arch Intern Med 2008;168:192–199

5. Ackermann RT, Finch EA, Brizendine E,Zhou H, Marrero DG. Translating the Di-abetes Prevention Program into the com-munity: the DEPLOY Pilot Study. Am J PrevMed 2008;35:357–363

6. Absetz P, Valve R, Oldenburg B, et al.Type 2 diabetes prevention in the “realworld”: one-year results of the GOAL Im-plementation Trial. Diabetes Care 2007;30:2465–2470

7. Amundson HA, Butcher MK, Gohdes D,et al. Translating the diabetes prevention

program into practice in the generalcommunity: findings from the MontanaCardiovascular Disease and Diabetes Pre-vention Program. Diabetes Educ 2009;35:204–209

8. Boltri JM, Davis-Smith YM, Seale JP,Shellenberger S, Okosun IS, CorneliusME. Diabetes prevention in a faith-basedsetting: results of translational research.J Public Health Manag Pract 2008;14:29–32

9. Kramer MK, Kriska AM, Venditti EM,et al. Translating the Diabetes PreventionProgram: a comprehensive model for pre-vention training and program delivery. AmJ Prev Med 2009;37:505–511

10. McBride PE, Einerson JA, Grant H, et al.Putting the Diabetes Prevention Programinto practice: a program for weight lossand cardiovascular risk reduction for pa-tients with metabolic syndrome or type 2diabetes mellitus. J Nutr Health Aging2008;12:745S–749S

11. McTigue KM, Conroy MB, Bigi L, MurphyC, McNeil M. Weight loss through livingwell: translating an effective lifestyle in-tervention into clinical practice. DiabetesEduc 2009;35:199–204, 208

12. Whittemore R, Melkus G, Wagner J,Dziura J, Northrup V, GreyM. Translatingthe diabetes prevention program to pri-mary care: a pilot study. Nurs Res 2009;58:2–12

13. Saaristo T, Moilanen L, Korpi-Hyövälti E,et al. Lifestyle intervention for preven-tion of type 2 diabetes in primary healthcare: 1-year follow-up of the FinnishNational Diabetes Prevention Program(FIN-D2D). Diabetes Care 2010;33:2146–2151

14. Norris SL, Chowdhury FM, Van Le K,et al. Effectiveness of community healthworkers in the care of persons with di-abetes. Diabet Med 2006;23:544–556

15. Katula JA, Vitolins MZ, Rosenberger EL,et al. Healthy Living Partnerships to Pre-vent Diabetes (HELP PD): design andmethods. Contemp Clin Trials 2010;31:71–81

16. Blackwell CS, Foster KA, Isom S, et al.Healthy Living Partnerships to PreventDiabetes: recruitment and baseline char-acteristics. Contemp Clin Trials 2011;32:40–49

17. Mechanick JI, Kushner RF, Sugerman HJ,et al. American Association of ClinicalEndocrinologists, The Obesity Society, andAmerican Society for Metabolic & BariatricSurgery Medical guidelines for clinicalpractice for the perioperative nutritional,metabolic, and nonsurgical support of thebariatric surgery patient. Surg Obes RelatDis 2008;14(Suppl. 1):1–83

18. American Diabetes Association. Diagnosisand classification of diabetes mellitus. Di-abetes Care 2004;27(Suppl. 1):S5–S10

19. Matthews DR, Hosker JP, Rudenski AS,Naylor BA, Treacher DF, Turner RC.

1456 DIABETES CARE, VOLUME 34, JULY 2011 care.diabetesjournals.org

HELP PD: 1-year results

Homeostasis model assessment: insulinresistance and beta-cell function fromfasting plasma glucose and insulin con-centrations in man. Diabetologia 1985;28:412–419

20. Katsuki A, Sumida Y, Gabazza EC, et al.Homeostasis model assessment is a reli-able indicator of insulin resistance duringfollow-up of patients with type 2 diabetes.Diabetes Care 2001;24:362–365

21. Williamson DF, Kahn H, Worthman C,Burnette J, Russell C. Precision of recumbent

anthropometry. Am J Hum Biol 1993;5:159–167

22. Klein S, Sheard NF, Pi-Sunyer X, et al.Weight management through lifestylemodification for the prevention andmanagement of type 2 diabetes: ratio-nale and strategies: a statement of theAmerican Diabetes Association, the NorthAmerican Association for the Study ofObesity, and the American Society forClinical Nutrition. Diabetes Care 2004;27:2067–2073

23. Norris SL, Zhang X, Avenell A, et al. Long-term effectiveness of weight-loss interven-tions in adults with pre-diabetes: a review.Am J Prev Med 2005;28:126–139

24. Wing RR, Tate DF, Gorin AA, Raynor HA,Fava JL. A self-regulation program formaintenance of weight loss. N Engl J Med2006;355:1563–1571

25. Kahn SE, Hull RL, Utzschneider KM.Mechanisms linking obesity to insulin re-sistance and type 2 diabetes. Nature 2006;444:840–846

care.diabetesjournals.org DIABETES CARE, VOLUME 34, JULY 2011 1457

Katula and Associates