Progressive Achilles Loading via Clinician e-Support (PACE) is a single-blind, noninferiority, 2-arm, parallel-group randomized controlled trial. Individuals with AT will be randomized to a single-visit, PT-initiated rehabilitation program or to a multivisit, PT-guided telehealth rehabilitation program. All participants will initiate care with a single synchronous in-person visit with a PT and have access to information on exercise and education materials via an online learning management system. Primary outcomes will be the numeric rating scale (NRS) for movement pain and the Victorian Institute of Sport Assessment–Achilles (VISA-A) score for disability.

We will enroll 160 individuals with AT across 2 sites: 1 military treatment facility and 1 civilian academic center. Outcomes will be assessed at 2, 4, 8, 12, 26, and 52 weeks, with the primary end point at week 4. Deviations to the protocol will be reported in real time on the Open Science Framework. This protocol is reported in accordance with the SPIRIT (Standard Protocol Items: Recommendations for Interventional Trials) guidelines [25].

Participants will be recruited and enrolled from 2 sites: the Carl R. Darnall Army Medical Center and the University of Iowa Hospitals & Clinics (UIHC). Active and reserve military personnel, veterans, and civilians will be eligible with recruitment carried out at site-specific clinics (family medicine, physical therapy, and sports medicine); on social media; through email requests; and through the distribution of brochures and flyers in community areas. A full list of the inclusion and exclusion criteria is provided in Textbox 1. To allow for the consideration of sex as a biological variable in the study findings, we will target at least 33% enrollment of female participants across both sites, with an anticipated higher recruitment rate of female participants at the UIHC to compensate for an anticipated lower rate of female recruitment at the CRDAMC (Carl R Darnall Army Medical Center, Fort Hood).

This study will be stopped prior to its completion if recruitment challenges significantly affect the ability to complete analyses of the primary outcomes. Therefore, we anticipate that it is more likely that we need to add and modify recruitment strategies rather than invoke this stopping rule.

We will follow a 3-stage screening process at the CRDAMC. First, a member of the study team will review new referrals to CRDAMC physical therapy for lower leg, ankle, and foot conditions. Second, study staff will review the past medical history of potentially eligible participants as part of the referral review process. If eligibility criteria are potentially met, a PT will contact the patient by phone to arrange a Health Insurance Portability and Accountability Act (HIPAA)–compliant video meeting to review the eligibility criteria in more detail and have the participant perform tendon-loading exercises to determine movement-evoked pain intensity localized to the Achilles tendon. The final stage of the screening process includes an in-person clinical examination to confirm the diagnosis of AT and a review of the eligibility criteria.

At the UIHC, we will use a similar 3-stage screening process. First, all recruited potential participants will complete a brief online eligibility survey. Second, if the criteria are potentially met, a PT will meet with them over a video meeting to review the eligibility criteria in more detail and have them perform tendon-loading exercises to determine movement pain intensity localized to the Achilles tendon. The final stage of the screening process includes an in-person clinical examination to confirm the diagnosis of AT and a review of the eligibility criteria.

Individuals who meet all eligibility criteria during the in-person visit will be randomized to either a single-visit or multivisit rehabilitation intervention. Randomization will occur at a 1:1 ratio with stratification at each site for each treatment arm using permuted random blocks. Stratification variables will be based on AT type (insertional or midportion) and symptom duration (acute <3 months or chronic >3 months), which will result in 4 strata per site. Variable block sizes of 2, 4, and 6 will be used so that randomization blocks cannot be guessed. To minimize selection bias, group assignments will only be revealed after eligibility is confirmed, all baseline measures have been collected, and the in-person treatment visit has been completed. The PTs enrolling participants will not have access to the allocation sequence, which will be concealed using a computer-based randomization system in REDCap (Research Electronic Data Capture). Unblinding of group allocation will occur in the case of serious adverse events or unanticipated problems related to the intervention, as decided by the safety officer (MH) in conjunction with the steering committee (RLC, JW, and MF).

Both arms of this trial will be using an online learning management system (TalentLMS; Epignosis) for education on the pathophysiology, self-management, and exercise progression for AT. This exercise and education content has been shown to be equally effective in reducing pain when delivered by telehealth compared to in-person treatment for AT [26,27]. For this study, the learning units (Table 1) are either completed independently at a self-paced rate (single-visit group) or reviewed during weekly sessions with a PT (multivisit group). The learning units include information about the home exercise program and provide guidance on progression. The exercise component has been previously described by Post et al [27] and is shown in Figure 1.

Participants in the single-visit, PT-initiated group will undergo a self-paced rehabilitation program initiated with 1 synchronous, in-person visit with a PT. At the first visit, participants will be instructed to complete the 6 successive online learning units over an 8-week period, with flexibility to choose the timing that works best for them.

Participants in the multivisit, PT-guided rehabilitation program will have 7 one-on-one telehealth visits over 8 weeks with a PT. These 30-minute visits will include exercise participation and education. Participants will be expected to complete 1 educational unit per week and will be given an exercise log to track exercises completed at home between treatment sessions. Instruction on exercise performance and progression will be provided by a PT throughout the plan of care. The PT will review the weekly educational content with the participant at each visit.

To maintain blinding of outcome assessment, baseline measures will be gathered in person prior to randomization. A research assistant blinded to the treatment groups will capture outcomes after randomization. Participants will complete 3 survey-only evaluation sessions at 2, 12, and 26 weeks. Participants will complete 3 video evaluation sessions, including surveys, at 4 weeks (primary end point), 8 weeks, and 52 weeks. The video evaluation sessions will be used to capture pain with tendon-loading activities (heel raises and hops) in real time, assess heel raise performance (number of heel raises), and screen for adverse events. A visualization of measure administration can be seen in Table 2.

The primary outcomes of this study are movement pain and disability (Table 3). Movement pain will be measured using the NRS. Assessing the participant’s worst pain during single-limb heel raises allows for comparison of outcomes with previous clinical trials for AT [26,28-30]. Disability will be measured using the VISA-A, which assesses disability specific to AT and allows comparison to other studies, as this is the most commonly used self-reported outcome for AT [31-34].

At baseline, we will capture demographics, military status, and tendon anterior-posterior diameter using ultrasound imaging [57]. In alignment with the core domain of tendon health, as defined by the International Scientific Tendinopathy Symposium Consensus [58], we will capture additional measures of pain and disability, along with measures of participation in life activities, physical function and capacity, psychological factors, patient rating of the condition, and quality of life over time (Table 3). Activity level will be captured using the Tegner scale [59], minutes per week spent doing moderate and vigorous physical activity, and questions about military readiness (eg, ability to complete a 2-mile run, wear military boots, and wear load-bearing equipment). Potential confounders that may influence the response to rehabilitation will also be captured, including the use of cointerventions or medications, self-reported exercise logs, and completion of educational units as captured in the learning management system. To assess treatment credibility, participants will complete a series of items evaluating the intervention’s impact on their perspective, understanding of self-management strategies, clarity of information, and appropriateness of the approach for managing AT, rated on a 5-point Likert scale (0=not at all, 1=a little bit, 2=somewhat, 3=quite a bit, and 4=very much).

To inform the feasibility of translating these rehabilitation strategies to a battlefield setting, we will use purposive sampling to invite participants at the CRDAMC who rotate to the field to participate in focus groups to discuss barriers and facilitators to participation in the rehabilitation programs. Participants will be invited to participate in the qualitative interviews until data saturation is reached. Semistructured interviews will be held via a videoconference platform that allows for HIPAA-compliant audio recording of the interviews. All audio recordings will be transcribed verbatim.

Exercise and education interventions are considered minimal risk by the University of Iowa Institutional Review Board (Biomedical IRB-01) and do not exceed the risk associated with standard clinical care; therefore, this study uses a safety officer (MH) rather than a data and safety monitoring board. The data and safety of participants will be monitored regularly by the intervention clinicians and outcome assessors, who will be contacting both the self-paced and PT-guided participants throughout the duration of study participation. All adverse events, serious adverse events, and unanticipated problems will be documented in REDCap by study team members and reviewed regularly by the safety officer. Any potential referrals for appropriate care for participants who report considerable pain or discomfort will be discussed with the safety officer and followed up in a timely manner.

The trial is powered to evaluate noninferiority between the two treatment groups at 4 weeks for 2 coprimary outcomes: (1) movement pain on a 0‐10 NRS, using a 1.1-point noninferiority margin, and (2) disability using the VISA-A, with an 8.2-point noninferiority margin. Both margins are consistent with previously published clinically meaningful thresholds [40]. Because the primary analysis will evaluate both outcomes using separate noninferiority tests, we applied a Bonferroni-adjusted, 2-sided α=.025 for each outcome. Power calculations assume comparison of mean group differences at 4 weeks using the same linear mixed-effects models planned for the primary analysis with adjustment for baseline values. A sample of 64 participants per group provides ≥80% power to conclude noninferiority for each outcome at α=.025. To account for an anticipated 20% dropout, we inflated the sample size to 80 participants per group.

Baseline characteristics will be presented as counts and percentages for categorical measures. Continuous measures will be reported as means and SDs or medians and IQRs, as appropriate. Summary statistics by group will be provided for the primary and secondary outcomes and will be compared using 2-sample t tests, Wilcoxon rank-sum tests, or Fisher exact tests, as appropriate. Our intention-to-treat analysis will use linear mixed-effects models to assess the noninferiority of the single-visit intervention compared to the multivisit intervention, adjusting for baseline values of the primary outcomes. Additional intention-to-treat analyses will include any covariates that differed between groups at baseline. Primary outcomes will be the 4-week changes in measures of pain and function, and models will account for site using a random effect. Point estimates, CIs, and P values will be obtained for between-group differences in 4-week mean change compared to the margin of noninferiority. Significance for testing pain and function will be assessed at an α level of .025. We plan to report all data disaggregated by sex, AT type, and AT duration.

Secondary analyses will use linear mixed-effects models to examine predictors of improvement in pain and disability separately at 4 weeks. The same analysis will be performed at 12, 26, and 52 weeks as secondary end points. Outcomes will be the changes measured at each follow-up time point, and the models will be controlled for the corresponding baseline values. Model selection will be conducted to identify the optimal predictor set for each change outcome over time. The pool of candidate predictors includes baseline demographic and ultrasound imaging variables, as well as immediate improvement (within 2 weeks) of pain-related psychological factors, knowledge on self-management, pain, and disability. Models for all additional predictor combinations up to order 3 will be fit, and the Akaike information criterion will be calculated to compare model fits with the same outcome. Point estimates, CIs, and P values will be reported for predictors in models selected by the Akaike information criterion for each outcome. Significance for model predictors will be assessed at an α level of .05. Categorization of continuous variables using clinically meaningful cut points, such as obesity categories for the continuous BMI variable, will also be explored.

Reasons for missing data will be recorded and compared qualitatively between participants who drop out and participants who complete the study. Participant characteristics collected prior to dropout will be compared to those of participants who complete the study. For the primary imputation strategy, if comparisons show that participant characteristics do not differ meaningfully between those with and without missing data, we will assume the data are missing at random and proceed accordingly. Under this assumption, we will use multiple imputation of the missing values for variables included in the linear mixed-effects models. Multiple imputation will be performed using SAS PROC MI and PROC MIANALYZE procedures in SAS software (version 9.4; SAS Institute Inc).

All variables in the primary analyses and, if feasible, other variables in the study database predictive of the missing values or influencing the cause of missing data will be included. For the sensitivity analyses addressing missing not at random, if assessment of reasons for missing data suggests that data may be systematically related to unobserved values (eg, inability to attend the next day assessment due to severe AT pain), sensitivity analysis will also be performed to assess the potential impact of including missing not at random data in the analysis. Methods for sensitivity analysis such as marginal delta adjustment, conditional delta adjustment, reference-based controlled imputation, and other pattern-mixture models will be considered. If needed, variables will be transformed to satisfy approximate normality before imputation and then retransformed to the original scale.

Data will be captured primarily via REDCap, a secure online data management system developed for building and managing online surveys and databases, and compliant with HIPAA and 21 Code of Federal Regulations Part 11. Electronic records collected will be retained for a minimum of 6 years after the completion of the study.

This study was approved by the University of Iowa Biomedical-01 Institutional Review Board (202405285) on September 19, 2024. During the screening process of this study, potentially eligible participants will be given the study information sheet and electronic consent form, with all questions answered and time provided for review. A copy of the electronically signed consent form will be obtained and kept within REDCap to document compliance, and a copy of the informed consent will be emailed to each participant. Participants will be compensated US $50 for completion of in-person consent and baseline measures, US $15 for each remote survey evaluation (up to US $45 total), US $50 for each remote video evaluation (up to US $150 total), and US $50 for focus group participation. All participant data will be deidentified and assigned a study-specific ID. Electronic records will be stored on University of Iowa network drives and will be accessible only to authorized personnel using password-protected computers.