Walking Exercise Sustainability Through Telehealth for Veterans With Lower-Limb Amputation: A Study Protocol

Abstract

Objective

This randomized controlled superiority trial will determine if an 18-month telehealth walking exercise self-management program produces clinically meaningful changes in walking exercise sustainability compared to attention-control education for veterans living with lower-limb amputation.

Methods

Seventy-eight participants with lower-limb amputation (traumatic or nontraumatic) aged 50 to 89 years will be enrolled. Two groups will complete 6 one-on-one intervention sessions, and 6 group sessions over an 18-month intervention period. The experimental arm will receive a self-management program focusing on increasing walking exercise and the control group will receive attention-control education specific to healthy aging. Daily walking step count (primary outcome) will be continuously monitored using an accelerometer over the 18-month study period. Secondary outcomes are designed to assess potential translation of the walking exercise intervention into conventional amputation care across the Veteran Affairs Amputation System of Care. These secondary outcomes include measures of intervention reach, efficacy, likelihood of clinical adoption, potential for clinical implementation, and ability of participants to maintain long-term exercise behavior.

Impact

The unique rehabilitation paradigm used in this study addresses the problem of chronic sedentary lifestyles following lower-limb amputation through a telehealth home-based walking exercise self-management model. The approach includes 18 months of exercise support from clinicians and peers. Trial results will provide rehabilitation knowledge necessary for implementing clinical translation of self-management interventions to sustain walking exercise for veterans living with lower-limb amputation, resulting in a healthier lifestyle.

Introduction

Severe peripheral artery disease, diabetes mellitus, and trauma are the most common reasons for lower-limb amputation (LLA) in veterans.^1,2 Veterans living with LLA have poor physical health outcomes,³ high medical service utilization,⁴ and high self-reported disability.¹ In the years following LLA, sedentary lifestyle is prevalent and frequently compounded by chronic comorbidities leading to further debility.^1,5

One way to reduce disability and improve functional independence for veterans with LLA is through sustained life-long walking exercise.⁶ However, 50% of people with LLA achieve routine walking activity that supports independent community participation,⁷ highlighting a critical need to improve walking exercise participation after LLA.⁸ Routine exercise can mitigate common postamputation sequelae, including diabetes, osteoporosis, cardiovascular disease, and falls.^9,10 Starting an exercise routine after LLA is a common adjunct to prosthetic training¹¹; however, long-term adherence is rarely achieved, which may explain poor long-term physical function outcomes.⁷ For example, up to 65% of individuals with LLA do not perform prescribed home exercises between physical therapist visits.¹² Following discontinuation of physical therapy, only 33% of adults with chronic health conditions report full adherence to an exercise protocol.¹³ Evidence supporting long-term exercise adherence remains limited because the majority of exercise trials report short-term outcomes, not exercise sustainability.¹⁴ Therefore, novel interventions targeting long-term exercise self-management are needed to best understand exercise sustainability.

Delivering clinician-supported physical activity behavior change and promoting self-management via telehealth within a person’s home holds promise for disrupting the cycle of sedentary behaviors and improving health outcomes.¹⁵ Telehealth offers many of the same benefits as in-person consultation while reducing socioeconomic barriers to accessing health care for veterans.¹⁶ Current Department of Veterans Affairs (VA) clinical practice guidelines for rehabilitation after LLA support assessment of psychosocial and behavioral health throughout rehabilitation,¹⁷ which can be delivered via telehealth. Furthermore, peer social support is recommended as part of the psychological adjustment process during rehabilitation after LLA. A similar exercise self-management approach for people with type 2 diabetes successfully increased physical activity for 2 years and was sustained for 3 years.¹⁸

The objective of this study is to provide a telehealth walking exercise sustainability intervention to veterans living with LLA. This study has 2 primary aims: (1) Determine if the telehealth walking exercise self-management program produces clinically meaningful exercise sustainability compared to attention-control education after LLA and (2) evaluate potential for large-scale clinical translation of the self-management program using RE-AIM (reach, effectiveness, adoption, implementation, and maintenance) framework assessments.¹⁹

Methods

This 2-arm randomized controlled trial implements a superiority trial design to determine the potential of veterans sustaining walking exercise over 18 months. The trial is designed to determine potential for a future large-scale clinical trial. The approach encompasses (1) sustainability of walking exercise through self-management; (2) optimizing telehealth intervention delivery; (3) remote monitoring and tailored messaging to eliminate access barriers; (4) peer support; and (5) leveraging established clinical resources.

Participants

Seventy-eight veterans (aged 50–89 years) with LLA will be recruited following completion of acute postamputation prosthetic training and randomized after baseline testing to the exercise self-management (EXP) or attention-control education (CTL) group. Randomization will be stratified by age (<63 and ≥ 63) and amputation etiology (traumatic and nontraumatic) using the Research Electronic Data Capture (REDCap)²⁰ randomization module. As a result only the interventionists will be unblinded to group allocation. Eligibility criteria are presented in Table 1.

Table 1.

Eligibility Criteria^a

Inclusion Criteria	Exclusion Criteria
Veterans between 50 and 89 y old. Unilateral or bilateral LLA (trans-metatarsal to hip disarticulation; traumatic or nontraumatic etiology). Ability to walk 2 min without seated rest while using the prosthesis and assistive device if needed. Living without assistance for basic activities of daily living.	Congenital or cancer-related amputation. Unstable heart condition (including unstable angina, uncontrolled cardiac dysrhythmia, acute myocarditis, and acute pericarditis). Acute systemic infection. Prisoner. Mild cognitive impairment. Active cancer treatment. Discretion of principal investigator to exclude patients who are determined to be unsafe or inappropriate to participate.

^a LLA = lower-limb amputation.

Recruitment

Participants will be recruited from the Rocky Mountain Regional VA Medical Center. The VA provides lifelong care for veterans with LLA, including annual multidisciplinary team clinic visits with the VA Regional Amputation Clinic (RAC) Team. Potential participants who express interest will complete a brief eligibility screen. Among the 78 participants to be enrolled, 62 participants are expected to complete final testing (Fig. 1). Consent and authorization forms will be emailed to each participant and reviewed at a prearranged time, approximately 1 month prior to their annual RAC clinic. Following consent and authorization, participants will be mailed an accelerometer-based activity monitor and instructions for use.

Figure 1.

Anticipated Consolidated Standards of Reporting Trials (CONSORT) flow diagram. CTL = attention-control group. EXP = experimental group.

Intervention Overview

Intervention for both groups will last 18 months and consist of (1) 2 annual multidisciplinary team sessions, (2) 6 individual telehealth intervention sessions (decreasing in frequency over time), and (3) 6 peer group telehealth sessions (every 3 months) for a total of 14 total sessions across 18 months. EXP and CTL groups will have separate primary interventionists to minimize risk of intervention contamination. Outcomes testing will occur at baseline, 6 months (M6), and 18 Months (M18) following enrollment. Study testers will have undergone training and reliability testing to administer testing. Outcome testing will occur remotely. All scheduled participant interactions are detailed in Figure 2.

Figure 2.

Testing and intervention timeline. CTL = attention-control education group; EXP = exercise self-management group.

Annual Multidisciplinary Team Sessions

Annual multidisciplinary team telehealth sessions consist of reviewing the veteran’s medical history, assessment of prosthetic fit and function, visualization of residual limb(s), physical evaluation, and care planning. Members of the amputation care team include physiatrist, surgeon, primary care physician, podiatrist, prosthetist, physical therapist, occupational therapist, nurse practitioner, psychologist, social worker, and nurse (Fig. 2). Study interventionists will attend annual appointments and share the participant’s physical function test results for both EXP and CTL groups. Participants in the EXP group will additionally share their exercise goals and intervention plan. The RAC team will review the current status of each participant as an opportunity to collectively identify ways to support goal achievement and optimize outcomes for each participant.

Experimental Intervention

The EXP intervention will leverage behavior-change techniques, including exercise self-management training using a wrist-worn Fitbit and tailored messaging with the VA-approved “Annie App” behavior-change techniques that promote exercise self-management (Tab. 2). The key behavior-change techniques are based on Social Cognitive Theory, Control Theory, Operant Conditioning, and Two-Minds Theory.^21,22 The Fitbit is a consumer-based activity tracker with an easy-to-use touch-screen interface that provides users the ability to create, self-monitor, and track personal walking activity goals. The Annie App is an automated VA messaging system that has demonstrated success in other veteran populations²³ and will prompt participants to consider behavior-change techniques and facilitate exercise self-management.

Table 2.

Walking Exercise Sustainability Training: Intervention Outline^a

Behavior-Change Technique	Individual Session	Peer Group Whole Health Session
Self-monitoring of behavior	Fitbit for step monitoring and rate of perceived exertion for intensity monitoring	Veterans will be encouraged throughout to: (1) Create 2 action plans, 1 involving walking exercise, and 1 in any of the related Whole Health domains: moving the body, surroundings, personal development, food and drink, recharge, family/friends/co-workers, spirit and soul, and power of the mind. (2) Use self-monitoring and graded task assessment to modify their own action plans and support others’ action plans. (3) Group problem solving to help restructure the physical and social environment and encourage prompts and cues to maximize Whole Health through individualization of self-management and community (peer and professional) support.
Action planning	Weekly action planning based on participant step count for identified individual targets and secondary activity action plan of participant’s choice
Graded tasks	Connection to self-monitoring and noted differences with higher or lower step count days
Restructuring environments	Discussion of personal barriers and facilitators to exercise and discussion of environmental components that can change
Problem solving	Discussion of reducing personal barriers and leveraging personal facilitators to increase walking exercise
Prompts and cues	Identification of cues that will be useful in self-management of walking exercise moving forward and long term

^a Sessions progress from intervention led (semi-passive) to participant and peer led (active).

Individual intervention sessions will track participant health care visits and falls. In telehealth intervention session 1, participants will meet with the interventionist to discuss behavior-change techniques, with a focus on walking exercise self-monitoring (Fig. 2). The participant will also be introduced to action planning and will create an action plan with exercise goals guided by the Patient-Specific Functional Scale (PSFS), a clinimetrically sound tool used to quantify activity limitations and aid in rehabilitation goal setting.²⁴ The 5 remaining individual sessions will focus on behavior-change techniques from an established Behavior Change Technique Taxonomy²² to promote exercise self-management and patient-centered communication.²⁵

Each 30-minute session will concentrate on one behavior-change technique and address others as warranted for individualized walking exercise action plans. Feedback on step counts provided by the Fitbit will be used for creating and self-monitoring personal walking activity goals during each EXP session. Exercise goals beyond walking will also be tracked using the PSFS and participant self-report of goal attainment. Throughout the intervention, the focus will progress from the interventionist guiding the discussion to the participant leading the sessions (Tab. 2).²⁶

Control Intervention

The CTL intervention will incorporate interventionist-delivered education sessions on general health topics (eg, falls, wound care, sock ply and skin care, pain education, mental health, and well-being, sleep) and general health education via Annie App prompts to match the timing and duration of the EXP group. Attention-control sessions will track health care visits and falls. The education sessions and Annie App prompts are based on topics and educational materials consistent with recommendations from the Center for Disease Control and the VA Whole Health Program.²⁷ Exercise will not be discouraged in the CTL group; rather, it will not be discussed in the research intervention. Any exercise-related questions during research sessions will be referred to a VA RAC outpatient physical therapist.

Peer Group Sessions

The 1-hour peer group sessions will be offered every 2–3 months. The group sessions for the EXP and CTL groups will be different. Groups will target 5–10 participants who are veterans based on recommendations for ideal group intervention size.²⁸ Groups will be guided by a physical therapist and trained peer mentors. The EXP group will focus on peer support of the participants in attaining sustained exercise. The CTL group peer sessions will not include any exercise-related self-management topics or techniques and will focus on preselected discussion topics such as fall risk education, pain education, etc.

Outcome Measures

Walking Exercise

Walking exercise will be recorded continuously across the 18-month study period, using a waist-mounted tri-axial accelerometer (ActiGraph GT9X Link).²⁹ The ActiGraph waist-mounted device allows for accurate tracking of daily walking activity and can accurately capture free-living daily walking patterns such as cadence and walking bouts. The ActiGraph is well-validated in the LLA³⁰ population and is used extensively for assessing daily walking activity.^31,32 The primary outcome will be the total daily step counts. Secondary activity outcomes will be total number of daily stepping bouts over a 10-minute duration and total number of minutes per day at a stepping cadence >40 steps per minute, both of which will also be quantified continuously over the 18-month study duration.

Clinical Translation Assessment

The secondary aim is to evaluate large-scale clinical translation potential of the intervention with the RE-AIM framework (Tab. 3).¹⁹ Intervention reach potential will be assessed as the percent of eligible veterans who receive the intervention. Fifty percent enrollment of all potential participants will be considered a positive local reach. Reasons for non-enrollment (eg, decline, unable to contact) will be noted to optimize approaches to maximize reach in future trials. We will use a brief participant satisfaction survey at the end of intervention (M18) using the Knowledge, Attitudes, and Beliefs Questionnaire³³ to also assess aspects of reach from the participant’s perspective, including reasons for enrollment.

Table 3.

RE-AIM Framework and Associated Outcome Measures^a

RE-AIM Component	Outcome Measure
Reach	Percent of eligible participants enrolled Satisfaction (50% enrollment considered a positive local reach)
Efficacy	Exercise duration and intensity (ActiGraph) Self-efficacy for exercise (SEE Scale)34 Socket satisfaction (CLASS)35 Physical function (PLUS-M,36 2-Minute Step Test,37 30-Second Chair Stand Test37) Social participation (PROMIS-Social)38 Disability (WHODAS 2.0)39
Adoption	Semi-structured mini focus groups with stakeholders: veteran and VA clinicians
Implementation	Intervention session log sheets Intervention fidelity checklist Adverse events
Maintenance	Percent of participants engaging in sustained walking exercise (at least 10 minutes for 3 d/wk) at study primary endpoint

^a All measures will be assessed at 3 discrete test points (M0, M6, and M18). CLASS = Comprehensive Lower Limb Amputation Socket Survey; PLUS-M = Prosthetic Limb Users Survey of Mobility; PROMIS-Social = Patient-Reported Outcomes Measurement Information System—Ability To Participate In Social Roles And Activities—Short Form 8a; RE-AIM = reach, effectiveness, adoption, implementation, and maintenance framework assessments; SEE Scale = Self-Efficacy For Exercise Scale; VA = Veterans Affairs; WHODAS 2.0 = World Health Organization Disability Assessment Schedule 2.0.

Efficacy outcomes beyond physical activity include exercise duration, exercise intensity, exercise self-efficacy, socket satisfaction, physical function, social participation, and disability (Tab. 3). The Self-Efficacy for Exercise Scale will assess exercise self-efficacy which has been validated for older adults.³⁴ The Comprehensive Lower Limb Amputee Socket Survey (CLASS) will be included as a multidimensional assessment measure of socket satisfaction.³⁵ Physical function will be assessed by the Prosthetic Limb Users Survey of Mobility (PLUS-M), 2-Minute Step Test, and 30-Second Chair Stand Test. The PLUS-M is a valid and reliable self-report measure of mobility for individuals with LLA.³⁶ The 2-Minute Step Test is a valid community and home-based test of endurance for older adults that measures total steps in place performed in 2 minutes.³⁷ The 30-Second Chair Rise Test is valid in older adults for measuring lower body strength and muscular endurance and is scored as the number of full stands completed in 30 seconds.³⁷ Social participation will be measured with the Patient-Reported Outcomes Measurement Information System (PROMIS) Ability to Participate in Social Roles and Activities—Short Form 8a, a validated questionnaire to measure social functioning in a variety of populations.³⁸ Lastly, we will utilize the WHODAS 2.0, a comprehensive measure of disability.³⁹

Adoption potential for the intervention will be understood through stakeholder engagement and input using semi-structured focus groups. Stakeholders will consist of RAC team members who are part of the annual multidisciplinary team telehealth sessions and veterans participating in the intervention. RAC stakeholders will participate in 30-minute focus groups up to 4 times throughout study. Veteran stakeholders will provide targeted input on the factors that influence intervention adoption currently and in the future. Typically, a range of codes is established with 4 focus groups, and meaning saturation is reached with 5 focus groups.⁴⁰ Focus groups with 4 to 6 people are recommended to provide in-depth insight, promote increased interaction, and allow for effective moderation⁴¹; therefore, up to 30 participants who are veterans will be recruited.

Implementation potential of the intervention will be evaluated based on whether EXP group interventionists deliver the intervention as planned.⁴² Data will be collected using a session log sheet in which interventionists will report protocol fidelity. We will consider >90% intervention fidelity to be a positive result for implementation potential. Interventionist delivery of the walking exercise behavioral intervention will also be assessed via supervision meetings and an intervention fidelity checklist completed upon review of 5% of randomly selected telehealth sessions.

Finally, intervention maintenance will be measured based on participant enactment measures and retention of behavior change.⁴² We will assess walking exercise of all participants (ActiGraph and self-report) at the primary endpoint (M18) to determine the number of participants continuing with intentional walking exercise (at least 10 minutes purposeful walking [cadence ≥40 steps per minute] 3 days per week). Specific walking bouts will be isolated from the ActiGraph data to quantify bouts of at least 10 minutes of walking. We will consider >75% participation in walking exercise at M18 to be positive maintenance. Additionally, we will assess participants’ self-management progression as reported on intervention session records over the course of EXP group participation.

Data Analysis

Sample size was estimated using pilot and published change scores,¹⁵ variability, and clinically meaningful difference data in average daily step count. We will test the null hypothesis that the EXP group has a mean change in activity of less than 1000 daily steps over the CTL group. With 31 individuals per group (total of 62 completers at M18), based on a 1-sided t-test, we will be able to detect a standardized effect size of 0.83 with 90% power (α = 0.05). Assuming approximately 20% attrition, we will randomize 78 participants to graduate 62 (31 per group) at the primary endpoint (M18). Analyses will be performed according to the principle of intention-to-treat (including all randomized participants). Unless otherwise specified, performed hypothesis tests will be 2-sided (α = .05), with 95% CI or P values reported. We will employ linear mixed models (LMMs) to incorporate longitudinal data structures.

The primary analysis will use appropriate contrasts from LMM and the primary hypothesis will be tested from the CIs of the difference between the 2 groups. Specifically, if the 95% CI of the estimated expected difference between the 2 groups in the change in step counts at 18 months from baseline does not contain 1000, we will reject the null hypothesis. If there is evidence of normality assumption violations, we will use appropriate transformations or link functions (eg, logit link for dichotomized measures). Goodness-of-fit statistics and model fitting diagnostics will be used to assess for influential points, outliers, and heteroscedasticity.

A secondary analysis of efficacy will explore differences over time in the number of stepping bouts >10-minutes per day, and time per day in stepping cadences >40 steps per minute. Statistical parametric mapping (SPM) will be used to assess the differences in these outcome measures between groups across the continuous duration of the study period.^43,44 SPM allows statistical examination of continuous data across an entire time series.⁴⁵ SPM will be used to analyze stepping activity (stepping bouts and cadence), using established Matlab software code that is accessible online.⁴³ Other secondary outcomes will be reported descriptively including the RE-AIM outcomes.

Role of the Funding Source

The funding source had no role in the study's design, conduct, and reporting.

Ethics

Ethical approval was obtained from the Colorado Multiple Institutional Review Board (#22–20,702) and is registered with ClinicalTrials.gov (NCT05412550). All participants will provide informed consent and authorization. Personal information will be collected electronically or via telephone and stored in a secure password-protected REDCap database. A deidentified database will be publicly available through the trial registry. Results will be reported via publication.

Discussion

Potential Impact of the Study

This study targets walking exercise sustainability through an innovative long-term telehealth self-management approach. We will determine if the exercise self-management program is superior to an attention-control group in sustaining walking exercise. In addition, the superiority trial design will allow early intervention adjustments so that a future multisite trial protocol will optimize intervention reach, efficacy, adoption, implementation, and maintenance. In the case of a positive superiority result (EXP group walking activity > CTL group), we will design the future trial with any warranted intervention modifications based on the results of the current proposed trial. In the case of a negative superiority result (EXP group walking activity = CTL group), the investigative team will refocus intervention efforts based on the knowledge gained to determine novel ways for better achieving veteran exercise sustainability after LLA.

Alternative Approaches & Anticipated Limitations

Based on the number of veterans seen annually at the VA Rocky Mountain Regional Medical Center, we expect to meet recruiting goals with veterans. Nevertheless, collaborators at other Denver regional medical centers will be approached to serve as additional referral sources as needed for recruiting veterans outside of the VA system. Based on previous studies with activity monitoring, we anticipate minimal noncompliance with activity monitor wear.³⁰ However, analyses using a likelihood-based longitudinal models will be completed in the presence of any missing activity data to give the least biased estimate of the treatment effect. Lastly, the sample size of the trial cannot account for all potential covariates (eg, age, socioeconomic, psychosocial status), we will examine changes in key characteristics descriptively to identify any outcomes that have a signal change. This examination of descriptive variables will inform potential changes to protocol and outcome plans in the future stages of this research.

Based on the complex health conditions individuals with LLA experience, participants may have periods during the 18-month study participation in which they cannot walk as prescribed. We will address this potential problem in 2 ways. First, participants in the intervention group will select secondary, nonwalking activity goals other than walking. By including both walking activity and nonwalking activity goals, the intervention allows for altered foci of participant support even during health-related pauses in walking ability. Second, the study will track any pauses in walking exercise by use of continuous daily activity monitoring, while continuing to have participants engage in the 1:1 and group intervention sessions. These pauses will be accounted for in the SPM analysis. These approaches simulate clinical situations in which patients have setbacks that require alterations in rehabilitation intervention. Potential bias for the intervention satisfaction results is a limitation because the interventionists are members of the study team. Generalizability of our implementation results may be limited to the VA health care system. Specific behavior change techniques used in the intervention will not be distinguishable when making inferences on overall intervention effect. Influence of specific techniques will need to be assessed in future studies designed to analyze them.

Author Contributions

Shawn L. Hanlon (Investigation [supporting], Project administration [supporting], Writing—original draft [lead], Writing—review & editing [lead]), Laura A. Swink (Conceptualization [supporting], Methodology [supporting], Project administration [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), Rachael Brink Akay (Investigation [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), Thomas T. Fields (Writing—original draft [supporting], Writing—review & editing [supporting]), Paul F. Cook (Conceptualization [equal], Data curation [supporting], Funding acquisition [supporting], Methodology [lead], Supervision [lead], Writing—original draft [supporting], Writing—review & editing [supporting]), Brecca M.M. Gaffney (Conceptualization [supporting], Data curation [supporting], Formal analysis [supporting], Funding acquisition [supporting], Investigation [supporting], Methodology [lead], Writing—original draft [supporting], Writing—review & editing [supporting], Elizabeth Juarez-Colunga (Conceptualization [supporting], Data curation [lead], Formal analysis [lead], Funding acquisition [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), and Cory L. Christiansen (Conceptualization [lead], Formal analysis [supporting], Funding acquisition [lead], Investigation [lead], Methodology [lead], Project administration [lead], Resources [lead], Writing—original draft [supporting], Writing—review & editing [supporting])

Ethics Approval

Ethical approval has been obtained from the Colorado Multiple Institutional Review Board (#22-20,702).

Funding

This study was funded by the Veterans Affairs Eastern Colorado Health Care System VA RR&D (I01RX003917) and NIH/NCATS Colorado CTSA Grant Number UL1 TR002535.

Clinical Trial Registration

This study is registered with ClinicalTrials.gov (NCT05412550).

Data Availability

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

Disclosures

The authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported no conflicts of interest.

The contents of this manuscript do not necessarily represent the views of the VA or the United States Government.