Adapting the “Resist Diabetes” Resistance Training Intervention for Veterans

Abstract

Purpose

The Resist Diabetes trial demonstrated that twice-per-week resistance training reduced prediabetes prevalence and improved strength among older adults with prediabetes. Our objective was to determine initial perceptions of patients and care providers in a Veterans Affairs Medical Center (VAMC) regarding Resist Diabetes (RD), and ultimately, inform adaptations to improve uptake of RD in the Veterans Health Administration.

Methods

A mixed-methods approach was utilized. Care providers (n=20) and veterans with prediabetes (n=12) were recruited to gauge perceptions of the RD program and identify barriers and facilitators to the program referral process and program implementation. Care provider perceptions of the acceptability, appropriateness and feasibility were determined using a validated survey. Open-ended questionnaires and interview guides, based upon the Consolidated Framework for Implementation Research, were utilized to determine major and minor themes within the provider and veteran responses. To identify the dissemination potential of RD, the availability of onsite fitness facilities at VAMC facilities nationally (n=159) was assessed.

Results

Providers rated (scaled 1–5; 1=completely disagree, 5=completely agree) the RD program as appealing (4.8+/−0.1), appropriate (4.8+/−0.0), and feasible (4.6+/−0.2). Providers reported that prediabetes/diabetes is a significant problem in the VAMC, and that different prevention programs will appeal to different types of VAMC patients. Patients (n=12; 58% female; aged 65+/−10yrs; BMI 34+/−6 kg/m²; HbA1c 5.7+/−1.8%) expressed interest in an exercise-focused diabetes prevention program and defined key barriers: travel, transportation, and time constraints. Among the responding national VAMC sites, 85% (97/114) reported having an onsite fitness facility.

Conclusion

Salem VAMC care providers and veteran patients demonstrated positive perceptions of the Resist Diabetes program. Program adaptations are needed to address barriers to patient participation including travel, transportation and time constraints.

Introduction

Evidence-based weight-loss and health behavior promotion interventions that prevent type 2 diabetes (T2D) are rarely taken to scale within intended delivery systems (1, 2). Alternative approaches for type 2 diabetes (T2D) prevention are needed to maximize intervention reach and effectiveness (3). The US Veterans Health Administration (VHA) system is our nation’s largest integrated health system. The prevalence of T2D among patients in the VHA is substantially higher than in the general US population, which may be attributed to veterans older age and higher prevalence of overweight/obesity (4–6). One in four veterans are being treated for T2D, which is the leading cause of blindness, amputations and renal disease within the veteran population (7). Diabetes prevention efforts within the VHA patient population may be uniquely challenging (8, 9), particularly regarding the feasibility and acceptability of lifestyle changes, and strategies to maximize intervention adherence is an area of significant research needed (10).

VHA providers are willing to refer to health behavior intervention programs and administrators support policy changes to prevent T2D (7, 10–12). Existing VA diabetes prevention programs, such as VA Diabetes Prevention Program (VA-DPP) and MOVE!, focus on weight loss (7–9) with little focus on exercise. Although these programs have comparable effectiveness for producing long-term weight loss, neither program has reduced HbA1c in veterans with prediabetes (7, 12) Importantly, VA providers report that the focus on weight loss was a barrier to referral (7, 12). Resistance training (RT) is recommended by the American Diabetes Association for T2D prevention (13, 14) and offers a more effective diabetes prevention approach (10). In addition, veterans report positive attitudes about exercise (15), strong preferences for exercising with gym equipment including those used for RT, and structured workouts (15, 16).

The Resist Diabetes trial (n=170) demonstrated that RT twice per week for 3 months in a supervised setting reduces prediabetes prevalence and improves strength among older (aged 50–69 years) adults with prediabetes. Further, it could be successfully translated and maintained without direct supervision for 12 months (17, 18). The Resist Diabetes Social Cognitive Theory (SCT)-based maintenance intervention emphasized self-regulation (e.g., goal setting, self-monitoring) as a key theory-based component for exercise adoption and maintenance (17–19). After 3 months of RT, 34% of program participants were no longer prediabetic; this prevalence of normoglycemia was maintained through month 15 (17). Muscular strength and body composition also improved (strength: +21% and +14% for chest and leg press, respectively; fat-free mass: +0.5%). However, the degree to which Resist Diabetes fits the mission, values, and resources of the VHA system and the patients it serves has yet to be explored. The objective was to determine perceptions of the Resist Diabetes (RD) program from patients and care providers in a Veterans Affairs Medical Center (VAMC) with particular attention to the fit within the system (e.g., are fitness facilities available) in order to address any adaptations needed before full scale dissemination and implementation.

Methods

Protocol Overview

This research utilized a convergent parallel mixed-methods design (20). Participants included veterans, care providers and administrators at the Salem VAMC, located in the community where RD was conducted. Care Providers, including physicians, nurses, behavioral health and administration staff completed a brief, validated survey (21) and an open-ended questionnaire. The brief survey and questionnaire took approximately 20 minutes to complete. A Waiver of Documentation of Informed Consent was obtained for the provider portion of the study from the Salem VA Institutional Review Board (IRB). An informational paragraph was provided on the open-ended provider questionnaire, to describe the research and the expectations of participation to providers.

Veteran participants were Salem VAMC patients, over the age of 18, who met the criteria for prediabetes (HbA1c 5.7–6.4%, fasting glucose 100–125mg/dl, or 2-hr oral glucose tolerance 140–199mg/dl (13, 14), were medically stable and able to function independently. Veteran participants were initially recruited to undergo two 45–60-minute study sessions at the Salem VAMC: one baseline assessment and one focus group session. However, due to a slow rate of enrollment and an insufficient number of participants enrolled in one time period, most participants completed all assessments in one study session which instead utilized an interview format with planned focus group questions. Veteran participants provided informed consent, HIPAA authorization and then completed questionnaires during the study session. Participant demographics, health history, anthropometrics, habitual physical activity and dietary questionnaires were collected in order to describe the participant population. The protocol and consent form were approved by the IRB at the Salem VAMC.

To identify the availability of onsite fitness facilities throughout the VHA system, which would be necessary to implement RD nationally, fitness facility information was collected by research staff (EV and JW) using an Internet search followed by phone calls to Veterans Affairs Healthcare Systems and Veterans Affairs Medical Centers across the nation.

Recruitment

Recruitment of providers occurred between March 2018 and January 2019, via brief presentations at three provider staff meetings or by informational emails from the VAMC Chief of Staff. Providers were eligible if they were physicians or other healthcare providers at the Salem VAMC. Providers were given a $7 meal voucher for participating in this investigation. This incentive option was informed by feedback from the VAMC research team members (12).

VA patients were enrolled between April 2018 and January 2019. Participants were identified and recruited by Salem VAMC care providers based on their medical history or electronic medical record (EMR) information. Recruitment cards (for patients) were distributed to providers at beginning of the staff meetings, during provider recruitment. Participants provided informed consent and HIPAA authorization to allow for EMR access (to verify prediabetes status) prior to the beginning of the study session. Participants received a $20 gift card to a local grocery chain for their participation at the end of the assessment session.

Quantitative Assessments: Providers

Providers completed a brief survey to assess the Acceptability of Intervention Measure (AIM), Intervention Appropriateness Measure (IAM), and Feasibility of Intervention Measure (FIM) for the RD program (21). This measure was developed by Weiner et al (21) based upon implementation outcomes defined by Proctor et al (22): Acceptability: perception of the intervention among stakeholders is agreeable, or satisfactory, Appropriateness: the perception of the fit of intervention into the practice setting or with consumers to fit a specific problem, Feasibility: the extent an intervention can be integrated and carried out within that given setting (22). These three implementation outcomes (Acceptability, Appropriateness and Feasibility) have shown to be leading indicators for implementation success (21). Each measure category (AIM, IAM, and FIM) included four items to measure the providers’ perceptions of the RD program. Each item was scored on a 1–5 scale (1=“Completely Disagree” to 5=“Completely Agree”). The brief and pragmatic qualities of the survey made it appropriate for providers of all levels in the Salem VAMC organization to complete in a short time period.

Qualitative Assessment: Providers

Providers completed an open-ended questionnaire developed by the research team to characterize perceptions of the RD program and prediabetes, barriers and facilitators to referral and program implementation within the Salem VAMC, communication and referral methods. All provider interactions were conducted by investigators who were not known to the providers (EV and BD).

Quantitative Assessments: Veteran Participants

Participants completed the following questionnaires to assess demographic and other descriptive characteristics. The Block Brief dietary screeners (23) assessed usual fruit, vegetable, fat and fiber intake, and took <10 minutes to administer. The brief Beverage Intake Questionnaire (BEVQ) (24, 25), provided information on habitual beverage intake. The BEVQ contains 15 items and takes 3–4 minutes to complete. Participants also completed a general Health History questionnaire.

Clinical measurements included height, body mass, BMI, hemoglobin A1C, fasting plasma glucose, and oral glucose tolerance test (OGTT). Height was measured to the nearest cm, with shoes (due to ease of session visit), on a scale-mounted stadiometer. Body mass was measured, with shoes, in clothing to nearest 0.1 lbs. using a clinical digital scale. BMI was calculated as weight (kg)/height (m²). HgbA1c, FPG and OGTT (if available) were obtained using the electronic medical record (EMR). The total time for participants to complete questionnaires, screeners and anthropometrics was approximately one hour.

Qualitative Assessments: Veteran Participants

A semi-structed interview guide was developed using the Consolidated Framework for Implementation Research (CFIR) (26). CFIR was created to consolidate existing implementation theories into one framework to identify factors that affect intervention implementation within VHA facilities and to provide consistent terminology within implementation science (26). CFIR includes 37 defined constructs, within five domains that influence implementation. These domains consist of: characteristics of the intervention (e.g. characteristics of the RD intervention, evidence strength, and quality), the outer setting (e.g. Salem VAMC patient needs and existing programs), inner setting (compatibility of RD with existing Salem VAMC programs and key stakeholders), process of implementation (e.g. receptiveness of stakeholders and leadership at the Salem VAMC), and characteristics of the individuals (e.g. knowledge and perceptions of Salem VAMC patients (26)). CFIR provides a framework to identify implementation barriers and facilitators, which informs intervention adaptations to develop the “best-fitting” version of the intervention for the organization. CFIR was also used to guide qualitative data coding and analysis (26).

Semi-structured interviews were conducted to characterize veteran’s perceptions of the RD program and the Salem VAMC, and barriers and facilitators to veteran referrals and program participation within the CFIR domains. Of the 12 participants enrolled, 11 participants completed interviews. Interviews were conducted by one investigator (EV), who recorded written responses to questions, and a second investigator (BD), who also recorded written responses and observations, and ensured all topics were covered.

VHA Fitness Facilities

One hundred and fifty-nine VHA facilities nationwide were identified via the U.S Department of Veterans Affairs “Location by State” website (27). This information was entered into an Excel spreadsheet, with the contact information for each facility. Research staff (EV and JW) contacted VAMC staff by phone (e.g. Employee Health, MOVE! Dietitians, Physical Therapists and Recreational Therapists) at each location to inquire about the availability of onsite fitness or gym facilities for veterans, employees or both. A phone messages were left if there was no response to the phone call.

Data Analysis

Block Brief Screeners (23), Health History and BEVQ (23–25) were reviewed for accuracy and completeness prior to analysis. Statistical analysis was conducted using SPSS statistical analysis software (SPSS v. 25 for Mac). Analyses included descriptive statistics (means, standard error, and frequencies). Data were expressed as mean ± SEM or %.

The provider open-ended questionnaires and participant interviews were analyzed using content analysis and the CFIR coding framework (26). Two investigators (BD and EV) entered provider and participant responses independently into Excel. Transcriptions were then compiled for comparison and content verification. One investigator (EV) tabulated individual responses across constructs by reviewing participant interview responses, written notes and provider questionnaire responses. Major and minor themes were identified within each construct. “Major themes” were defined as responses given by ≥50% of participants, and “minor themes” were defined as responses given by 25–49% of participants (28). Similar responses were grouped within each CFIR construct to detect the major and minor themes. These themes were organized within the CFIR framework domains and constructs and presented in Tables 4 and 5.

Results

Participants

Twenty providers were enrolled and consisted of clinical psychologists (PhD, PsyD) within the Mental Health Service at the Salem VAMC (65%), nursing staff (15%), and physicians (which included two administrators) (20%). All providers (n=20) completed the brief survey, but only 15 providers completed the longer open-ended questionnaire. Thirty-two veterans were referred to the study, and 12 of those individuals were enrolled. Reasons for not participating (remaining 20 patients) were as follows: 4 declined (2, study sessions conflict with work schedule; 2, residence is too far from VA medical center to participate), 2 were scheduled but did not show up or respond to rescheduling requests, 1 wanted to enroll but contacted us after data collection ended, 13 could not be reached via phone (i.e., did not respond or have active phone service).

The veteran participants enrolled (n=12) were predominately older adults, white, female, and overweight (Table 1). Of those enrolled, one veteran participant was transgender, and one veteran did not complete the interview component. Prediabetes diagnoses (i.e., ICD code) were only present in the EMR on 6 (50%) of the participants (Table 1).

Table 1.

Veteran participant demographic characteristics

Characteristic	Mean ± SEMa	n (%)
Gender
Male		5 (42.0)
Female		7 (58.0)
Age, y	65.0 ± 2.9
Race
White		9 (75.0)
African American		3 (25.0)
Height, in	68.5 ± 1.2
Weight, lbs.	222.3 ± 11.3
Body Mass Index (BMI), kg/m2	33.6 ± 1.9
Prediabetes Diagnosis in EMRb
Yes		6 (50.0)
No		6 (50.0)
HgbA1c, %	5.7 ± 0.5
FPG, mg/dLC	92.8 ± 13.7
Education
High School		2 (16.7)
College Degree		8 (66.7)
Master’s Degree		2 (16.7)
Occupation
Retired		7 (58.3)
Employed		2 (16.7)
No Response		3 (25.0)
Current Tobacco Use
Smoking		1 (8.3)
Non-Smoking		11 (91.7)
Alcohol Use
Alcohol Use		3 (75.0)
No Alcohol Use		9 (25.0)

^aSEM = standard error of mean

^bEMR = Electronic Medical Record

^cAvailable for 10 participants. No OGTT results were available in EMR.

Quantitative Results

Provider Perceptions of RD Acceptability, Appropriateness and Feasibility

Mean scores for the brief survey are provided in Table 2. Higher scores are associated with positive perceptions of the intervention in the context of that category. All providers scored the RD intervention between 3–5 (“Neutral” – “Completely Agree”) in all categories The lowest scores were in the Feasibility section.

Table 2.

Provider perceptions of Acceptability, Appropriateness and Feasibility of Resist Diabetes

	MD*/DO (n=3) Mean ± SEM	RN/LPN (n=3) Mean ± SEM	Other Providers** (n=13) Mean ± SEM
Acceptability of Intervention Measure (AIM)
Resist Diabetes VA meet my approval?	5.00 ± 0.00	5.00 ± 0.00	4.69 ± 0.48
Resist Diabetes VA is appealing to me?	4.75 ± 0.50	5.00 ± 0.00	4.69 ± 0.48
I like Resist Diabetes VA.	4.50 ± 1.00	5.00 ± 0.00	4.69 ± 0.48
I welcome Resist Diabetes VA.	5.00 ± 0.00	5.00 ± 0.00	4.84 ± 0.38
Intervention Appropriateness Measure (IAM)
Resist Diabetes VA seems fitting for Salem VAMC.	5.00 ± 0.00	5.00 ± 0.00	4.69 ± 0.48
Resist Diabetes VA seems suitable for Salem VAMC.	5.00 ± 0.00	5.00 ± 0.00	4.62 ± 0.51
Resist Diabetes VA seems applicable for Salem VAMC.	4.75 ± 0.50	5.00 ± 0.00	4.77 ± 0.44
Resist Diabetes VA seems like a good match for Salem VAMC.	5.00 ± 0.00	5.00 ± 0.00	4.69 ± 0.48
Feasibility of Intervention Measure (FIM)
Resist Diabetes VA seems implementable.	4.50 ± 1.00	5.00 ± 0.00	4.54 ± 0.52
Resist Diabetes VA seems possible.	4.75 ± 0.50	5.00 ± 0.00	4.77 ± 0.44
Resist Diabetes VA seems doable.	4.75 ± 0.50	5.00 ± 0.00	4.54 ± 0.52
Resist Diabetes VA seems easy to use	4.50 ± 1.00	4.33 ± 0.58	4.15 ± 0.80

^*MD included two Admin/MD staff members

^**Other providers included PsyD and PhD, one admin

^aScored on a scale 1–5.

Veteran’s Lifestyle Behaviors

In general, participants reported engaging in some regular physical activity (PA) but did not meet current PA recommendations (Table 3). However, more veterans were meeting (or exceeding) muscle strengthening exercise bout recommendations (~2.4 sessions per week). Mean reported sedentary time was ~ 7 hours per day. Participants reported meeting the recommended daily intake of fruits and vegetables, but inadequate amounts of dietary fiber (Table 3). Male and female veterans reported consuming 48+/−34 and 41+/−1 fl oz water per day, and 28+/−11 and 6+/−3 fl oz sugar-sweetened beverages per day, respectively.

Table 3.

Lifestyle characteristics of veteran participants.

Characteristic	Mean ± SEMa	Recommendedb
Moderate Exercise
Sessions per week	2.4 ± 0.7	150 minutes/week
Minutes per session	30.8 ± 8.9
Strenuous Exercise
Sessions per week	1.25 ± 0.6	75 mins/week
Minutes per session	4.17 ± 2.5
Muscle Strengthening Exercise
Sessions per week	2.4 ± 0.7	2–3× per week
Minutes per session	23.8 ± 7.4
Sedentary Time
Hours per week day	6.9 ± 0.8
Hours per weekend day	7.2 ± 1.3
Fruit and Vegetable Servings/day	5.2 ± 0.3	Adults 5 servings/day
Dietary Fiber (g)	7.8 ± 1.2	Adults 19yo+ 25–30g
Total Fat (g)	100.2 ± 8.0	Adults 19yo+ 20–35% total calories
Saturated Fat (g)	23.4 ± 2.0	Adults <10% total calories
Percent (%) Fat	36.0 ± 1.9	Adults 19yo+ 20–35% total calories
Dietary Cholesterol (mg)	219.4 ± 15.4	Adults 19yo+ <300mg

^aSEM = standard error of mean

^bReferences: 32, 33.

Qualitative Results

Provider and veteran responses including illustrative quotes, according to CFIR construct, are presented in Tables 4 and 5, respectively. The open-ended questionnaires and interviews revealed several major* and minor^† themes pertaining to perceptions of RD, barriers and facilitators to referral, the veteran population and barriers and facilitators to program implementation.

Table 4.

Provider questionnaire response themes for Resist Diabetes

CFIR Construct	Major* and minor† themes
Intervention Characteristics
Relative Advantage	Having multiple programs to target different patient demographics and preferences is preferred.* Would provide an additional option to the existing repertoire of lifestyle programs offered at Salem VAMC.* One-on-one structure of RD offers a more personalized approach. † P: This would be a nice addition and not competition. It adds another option.
Complexity	Completing the full program of 12 full body exercise or attending for 12-weeks may deter some veterans from finishing or participating.* Low motivation and unwillingness to experience discomfort. † P: I think a major thing is ‘life happens’ and you may have to learn that.
Cost	Motivated to integrate new programs as long as budget is minimally, if not affected. † P: Admin always wants new programs but won’t pony up the resources. Clinical staff [are] usually willing to pitch in.
Outer Setting
Needs & Resources of Those Served by the Organization	Majority of the veteran population will have travel and transportation as a barrier to participation*. Scheduling conflicts and time constraints will be existing medical or mental health appointments will likely affect participation and attendance.* The economic status of the veteran population will affect attendance and sustainability of the intervention if the veteran will be required to pay.* Fear of trying new interventions due to PTSD, gender and motivation may affect participation.† P: This region, due to poverty and culture (i.e. Appalachian), tends to have people who don’t believe these programs work for them”. P2: Travel is always a barrier.
Inner Setting
Networks and Communications	Providers prefer communication about new interventions at staff meetings or over email.* Primary Care staff (Physicians and nursing) are primary contacts for referrals and informing veterans of new programs.* Phone call and text messages are also effective methods for referring and communicating with veterans.* Email can be utilized for veteran communication. † Flyers provided by primary care, staff or visible in common areas to promote the intervention.* Veterans communicate with each other and spread news about new programs by word of mouth. † The Salem VAMC newsletter, social media and MyHealthe Vet mobile app could be utilized for informing veterans of new programs. † P: “Handouts and warm hand offs”
Implementation climate	Administration is receptive to new interventions.* Vets are used to referrals and would welcome the opportunity.* Veterans would be interested, but depending on their location, time and availability.* Veterans will be open if providers emphasize the effectiveness and benefits. † P: “I think overall veterans would be open to it, especially if encouraged by providers.
Relative Priority	High prevalence of T2D and prediabetes within the veteran population.* Veteran’s quality of life (sedentary, overweight/obese and comorbidities) are contributing to a high risk of T2D.* High risk due because of an aging population. † P: Yes! Diabetes is rampant in our patient population. It would be great to have more preventable interventions available.
Characteristics of Individuals
Knowledge and Belief	Yes, will refer patients to RD.* Encourages improvement in health. † P: Yes! It would be questionably ethical not to [refer patients].
Process
Engaging	Provider education at staff meetings and via email.* Reminders of intervention and inclusion criteria. † P: As long as inclusion criteria are widely shared, [identification] should be fine. P2: Only issue is effectively keeping it on the radar. P3: [Admin] is receptive if the steps are clear and efficient for referral.
Key Stakeholders	Primary care staff.*

^*Represents a “major theme”, defined as a response given by >50% of the participants

^†Represents a “minor theme”, defined as a response given by 25–49% of the participants

P = Participant illustrative quote

Table 5.

Veteran interview response themes for Resist Diabetes.

CFIR Construct	Major* and minor † themes
Intervention Characteristics
Relative Advantage	Have not participated in other lifestyle programs at the Salem VAMC.* Referred to the MOVE! program. † Gerofit participation. † Programs for prevention are important. †Excited to be able to train with a personal trainer.* Starting training initially one-on-one than transition to group format*. Group-based initially. † P: At Gerofit 3x/week and MOVE! once a month! P2: People pay a lot of money for that. A personal trainer like the movie stars!
Evidence Quality and Strength	Decreases the risk for getting T2D.* Reduced glucose, that huge. † Reduced HgbA1c. † Weight loss. † Sounds effective. †P: I don’t see them not having it, it would benefit veterans.
Complexity	Attending for 12-weeks may defer some veterans from finishing or participating.* Time period.* Conflicting appointments. † Don’t quite understand it all. † Weight lifting. † P: Schedule. That’s the only thing, otherwise I’d be here.
Outer Setting
Needs & Resources of Those Served by the Organization	Travel and distance from Salem VAMC*. Don’t know about available resources from the Salem VAMC to aid in program participation.* Scheduling conflicts, other appointments.* Fear of trying new interventions due to PTSD.†Limitations due to comorbidities or existing disabilities. † Need for childcare. † P: Travel mainly, I would have 3 hours of driving. P2: My handicap. Time with other appointments, some can be moved, some can’t.
Inner Setting
Networks and Communications	Find out about new programs through primary care physicians at visits.* Word of mouth from other veterans.* Phone calls, especially texts. Everyone has a phone* Email. † Social Media, maybe a Facebook group. † Occasionally flyers. † I read the newsletter. † MyHealthe Vet mobile app. † All something we are going to see. † P: Providers or I hear about it while I am here. P2: Only if you see you physician and they relay it to you. P3: Word of mouth or twice a year from my primary doctor.
Implementation climate	I think other veterans would be interested.* I don’t know about other veterans. † P: Maybe like half the patients. If it was explained why they should do it instead of being a full-blown diabetic. P2: Anybody who knows they have issues with diabetes, you’d be stupid not to participate... Exercise is ingrained in vets.
Relative Priority	High prevalence of T2D and prediabetes within our population.* Most of us are older, overweight and sedentary.* There’s no reason not to implement RD, they need it.* Prevention and education for T2D. † P: You see people missing their feet or legs, it was either war or diabetes. Most of the time it’s diabetes” P2: Prevention is better than a cure. This will prevent them from getting diabetes. P3: I don’t see not having it, it would benefit veterans.
Structural Characteristics	Yes, I know about the onsite fitness facility.* No, I didn’t know we even had an onsite gym. † You have to referred or in Gerofit to go to the gym. † P: Yes, but didn’t know before Gerofit. They had one program before though, but only for 6 weeks. Then no more.
Characteristics of Individuals
Knowledge and Belief on Innovation	Offers a way to resist diabetes and not become full blown diabetic.* Get people active, back in shape and doing something.* Sounds like a good program to participant in.* Overall health benefits. † P: So, I don’t get diabetes. When I get old, I want to be independent. Independence is a huge thing for me.
Other Personal Attributes: Knowledge of Prediabetes	I don’t know anything about prediabetes.* Prediabetes is when you have the characteristics leading up to T2D.* Not full blown T2D.* Informed by Primary Care Physician or staff.* Family history or knows someone with diabetes. † Don’t like it [prediabetes and T2D]. † Expected the diagnosis. † Frustrated, unhappy or depressed about the diagnosis. † Not aware of the diagnosis. † P: It brought back bad memories since my wife died of it [T2D]. P2: On your way to type 2 diabetes and all the medications.
Self-Efficacy	Yes, I want to be involved.* Depends on the exercises, but I can do it.* I could do the exercises.* Sharing the experience and doing it with others. † Learn the proper technique, then I can do it. † P: I need to do something, I sit around too much and eat too much. P: Losing weight and getting back in shape. Just sitting around, I don’t feel good. P2: I’m with somebody. I’m another guy.
Process
Engaging	Primary care referrals.* Email* Phone calls* Word of mouth from other veterans.* Flyers in common areas. †Newsletter. † P: Something I am going to see. P2: [With referrals] You have a have a basis of trust of where you heard it.
Key Stakeholders	Primary care staff.*

^*Represents a “major theme”, defined as a response given by >50% of the participants

^†Represents a “minor theme”, defined as a response given by 25–49% of the participants

P = Participant illustrative quote

Intervention Characteristics

Providers generally perceived that having multiple behavior change programs at the Salem VAMC could target different patient demographics and their varying preferences. For example, program offerings may include: MOVE! for weight management, Gerofit for older adults and Resist Diabetes for diabetes prevention. Therefore, RD could provide another option for Salem VAMC patients health care/improvement/prevention. Administration is open to offering new programs at the Salem VAMC, especially if budget is minimally impacted. The biggest concern for providers was the ability for veterans to complete the 12-week initial intervention period due to their medical conditions including PTSD.

The majority of veteran participants had not participated in other lifestyle programs at the Salem VAMC. Some had participated in or were referred to either MOVE! or Gerofit. Based upon a program description, participants were excited about the opportunity to reduce their risk of T2D through the Resist Diabetes program. A majority of the veterans were also enthusiastic about the possibility of resistance training with a personal trainer. Most participants stated a preference for beginning their training one-on-one with a personal trainer then transitioning to training with a group. However, the participants were open to group-based training as long as the group size was kept small (<10 participants).

Outer Setting, Needs and Resources of Those Served by the Organization

The biggest barrier for veterans based on provider response was travel and transportation to the Salem VAMC. Providers noted that a significant portion of veterans live long distances away from the Salem VAMC, and limited ability to travel and long driving time may influence program participation rates. Besides transportation, limited finances and time will adversely affect patient participant according to the providers. However, veterans may have existing appointments or commitments at Salem VAMC and these could be coordinated with program training sessions. If veterans are to be responsible for any program costs, this may deter participation.

Consistent with the provider results, participants perceived travel and distance to Salem VAMC as the major barrier to program participation. Participants noted they were not aware of any resources provided by the Salem VAMC that could aid in transportation or program involvement in general. Time and scheduling conflicts were also noted by the participations as a potential barrier to participation. Participants noted that existing disabilities and Post Traumatic Stress Disorder (PTSD) symptoms could limit program participation.

Inner Setting, Network and Communications

In terms of referring veterans to programs, the providers indicated that their preferred method of communication would come from primary care services. Phone calls and email can also be effective methods of referral. For communicating and educating providers, staff meetings and email were the preferred method of communication. Providers indicated that education on program and inclusion criteria at staff or department meetings would be most effective.

Providers generally believe that Salem VAMC administration is receptive to offering new programs. Although receptive, administration and providers, especially those in primary care, have overwhelming schedules and limited time. Despite these barriers, providers were receptive to offering and promoting new programs at the Salem VAMC. Providers also noted that veterans are used to receiving referrals on a regular basis, and welcome new programs. According to the Salem VAMC providers, there is a high prevalence of prediabetes and T2D within the veteran population, and a high prevalence of older, overweight/obese and sedentary veterans. As veterans often have other comorbidities that could contribute to a high risk of T2D development, Resist Diabetes was perceived to be a beneficial diabetes prevention program for this population.

Participants preferred finding out new programs through their Primary Care Physicians (PCP) at their appointments. Word of mouth from other veterans was also a preferred method of hearing about new programs. PCP and fellow veterans are trusted sources of information, which was important for the participants. In terms of technology-based communications, veterans preferred a phone call, and particularly favored text messages.

In general, participants believe that other veterans would be interested in and welcome the RD program. Participants noted a high prevalence of T2D and prediabetes within their population, consistent with their older age, overweight/obesity and a sedentary lifestyle. Diabetes prevention was important to the participants. The majority of the participants were aware of the Salem VAMC’s onsite fitness facility, and some participants enrolled in Gerofit already use this facility on a weekly basis.

Characteristics of Individuals

All providers had a positive view of the RD program, and all stated they would refer patients to RD. Based on the evidence basis for RD program effectiveness, providers perceived the program as encouraging to improve health.

A majority of the participants noted having a limited knowledge of prediabetes, and remarked that RD offered a way to resist T2D and “not become a full-blown diabetic”. Most participants were aware of their prediabetes diagnosis and had been informed of their diagnosis by their PCP, or they knew someone effected by the disease. In addition to preventing T2D, participants believed that RD allows participants to get active and “back in shape”. Most of the participants (all but one) wanted to be involved with RD when it is introduced to the Salem VAMC. One hesitation was not knowing what exercise were involved in the protocol, due to their past injuries.

VHA Fitness Facilities

Investigators identified 159 VAMC facilities nationwide. 45 (28%) of the 159 facilities contacted did not respond to phone calls. Among the 114 responders, over half (97/114, 61%) of the total facilities had a fitness or gym facility on their campus for either veterans or employees to use. Of these facilities, the majority of the fitness or gym facilities were for employees only, but the capabilities were there to offer access to veterans. A small portion of the responding VAMCs did not have any fitness or gym locations on their VAMC campus (17/114, 11%).

Discussion

These findings demonstrate that Salem VAMC providers and veteran participants had a high affinity for the RD program. Based on the Proctor et al (17) implementation outcome definitions for the brief survey (21), providers view the RD intervention as acceptable or satisfactory, appropriate for the Salem VAMC and its veteran population, and feasible to implement and integrate within the Salem VAMC. As positive perceptions of acceptability, appropriateness and feasibility are predictors of program uptake (21), these findings suggest that providers at the Salem VAMC may refer to and support the RD program at the VAMC. Veteran participants also demonstrated positive perceptions of the RD program, but identified key program adaptations needed for program implementation. Major barriers to RD participation, consistent in both provider questionnaires and participant interviews, included travel distance, transportation needs and time constraints (for both veterans and providers). RD program facilitators identified by both stakeholder groups suggest that primary care staff, specifically physicians, will be critical for successful program promotion and implementation. Both stakeholder groups also perceived the RD program as an effective and beneficial program that can aid in preventing T2D and improving veteran’s quality of life. Given that many national VAMC have an onsite fitness facility, if the RD program can demonstrate effectiveness in the VA setting, larger-scale program dissemination appears to be feasible from an infrastructure standpoint.

Within the VHA, MOVE! and VA-DPP are major translations of Diabetes Prevention Program (7–9). MOVE! specifically targeted weight management, while VA-DPP focused on diabetes prevention (7–9). MOVE! is a multidisciplinary program focused on patient centered goal setting and is delivered using small educational group sessions (10). VA-DPP used the existing MOVE! referral infrastructure, but targeted for T2D prevention in prediabetic veterans (7–9). VA-DPP utilized group-based, closed cohorts, goal setting and consistent program coaches. A comparison of these programs did not find significant differences in HgbA1c between the two programs, however VA-DPP had higher participation and weight loss at 6-months, and more VA-DPP participants completed more session than the MOVE! Participants (8). MOVE! and VA-DPP focus on lifestyle changes, including educating on increasing physical activity, however neither include in-session exercise (8). MOVE! is available at every VHA facility, but VA-DPP is only available at a few (8). Given veterans’ affinity for training either one-on-one with a trainer or in small groups (2–3 individuals) (15) which is consistent with the approach used in the original RD trial (17, 18), the availability of fitness centers in many VHA facilities, positive perceptions from both provider and veterans of the RD intervention and the need for multiple diabetes prevention program options within the VHA, the RD program could represent a promising new option within the VHA system. The RD program may be particularly appealing to those veterans searching for physical activity-based program diabetes prevention program, given veterans positive attitudes about structured exercise (15).

Strengths of this investigation include the mixed methods approach, the use of a validated survey (21) and CFIR implementation framework (26) to inform program adaptations needed to fit the VA system, the utilization of a research practice partnership team to develop and conduct this research (29, 30), the national audit of VA facility availability to address future dissemination efforts, and the inclusion of a diverse sample of providers and veteran patients. Based upon recruitment experiences, including a physician champion within the VHA Women’s Clinic (S.B.) was an effective method to increase the recruitment and participation of female participants. This is significant in that women are viewed as an underserved population within the VHA system. Provider participants were also included across multiple levels within the VAMC organization (i.e. nursing staff to administrators).

Limitations of this study included a small sample size, which could limit the generalizability and transferability of these results to the VAMC population as a whole. However, there was consistency between the provider and patient responses, and data saturation was achieved for responses related to program interest, benefits and barriers to participation. With regard to physical activity (PA) and dietary intake habits of veterans, self-reported methods are used which are often prone to bias. Previous research demonstrates that a small portion of veterans meet physical activity recommendations, and VHA-using veterans are less likely to meet physical activity recommendations than non-VHA-using veterans (31). In the present sample, several participants reported being involved in Gerofit (n=4), which is a weekly physical activity program for older veterans at the Salem VAMC. Future investigation could utilize objective methods to assess physical activity behaviors, such as accelerometry. The lack of an audio recording of veteran interviews limited our ability to transcribe answers to the semi-structured interview questions, although two interviewers were present to document interview fieldnotes and responses. The slow recruitment process precluded using the focus group format, which could have provided more insight to the veteran experience and culture at the Salem VAMC if larger samples of veterans were included, and response triangulation. Future investigations could also include more CFIR constructs, such as Culture which may play a large role within the VHA system.

Given the high prevalence of prediabetes and T2D in veterans, T2D prevention is a critical need in the VHA (7–9). However, the VHA is currently lacking an effective diabetes prevention program with a high rate of participation. Providers viewed the Resist Diabetes program as acceptable, appropriate and feasible for the Salem VAMC facility, although adaptations addressing travel, transportation, and scheduling needs should be addressed. These findings demonstrate that a Resist Diabetes: VA program could meet this need, and represent a lifestyle behavioral change program that is a good fit for veterans and the VAMC system. Furthermore, if the RD program is proven to be effective within the Salem VAMC it has the potential to be translated to other VHA facilities across the nation.