Efficacy of Telehealth for Movement-Evoked Pain in People With Chronic Achilles Tendinopathy: A Noninferiority Analysis

Abstract

Objective

The purpose of this study was to compare the efficacy of physical therapy delivered via an all telehealth or hybrid format with an all in-person format on movement-evoked pain for individuals with chronic Achilles tendinopathy (AT).

Methods

Sixty-six individuals with chronic AT participated (age, 43.4 [SD = 15.4] years; 56% female; body mass index, 29.9 [SD = 7.7] kg/m²). Participants completed all in-person visits from the initiation of recruitment in September 2019 to March 16, 2020 (in-person group). From March 17 to July 15, 2020, participants completed all telehealth visits (telehealth group). From July 16, 2020, to enrollment completion in December 2020, participants could complete visits all in-person, all telehealth, or a combination of in-person and telehealth (hybrid group) based on their preference. A physical therapist provided 6 to 7 visits, including an exercise program and patient education. Noninferiority analyses of the telehealth and hybrid groups compared with the in-person group were completed for the primary outcome of movement-evoked pain during single-limb heel raises.

Results

All groups demonstrated decreases in movement-evoked pain beyond the minimal clinically important difference from baseline to 8 weeks (2 out of 10 on a numeric pain rating scale). Lower bounds of the 95% CIs for mean differences between groups did not surpass the preestablished noninferiority margin (2 out of 10) for movement-evoked pain in both the telehealth and hybrid groups (telehealth vs in-person: 0.45 [−1.1 to 2.0]; hybrid vs in-person: 0.48 [−1.0 to 1.9]).

Conclusion

Individuals with chronic AT who completed a tendon-loading program with patient education through a telehealth or hybrid format had no worse outcomes for pain than those who received the same intervention through in-person visits.

Impact

Physical therapist–directed patient care delivered via telehealth may enhance accessibility to best practice AT rehabilitation, including exercise and education. Use of telehealth technology may also provide an opportunity to prioritize patient preference for physical therapy visit format.

Lay Summary

If you are a patient with chronic AT, physical therapist–directed patient care delivered via telehealth may improve your accessibility to best practice AT rehabilitation, including exercise and education. Use of telehealth technology may also prioritize your preferences regarding the format of the physical therapy visit.

Achilles, Noninferiority, Telehealth, Tendinopathy

Introduction

The coronavirus disease 2019 (SARS-CoV-2) pandemic substantially disrupted accessibility to in-person health care. With stay-at-home restrictions and quarantine orders, face-to-face visits were replaced with telehealth appointments (eg, the use of electronic technology, including video and/or audio) to remotely evaluate and follow up with patients. Telehealth initially provided a steep learning curve for health care providers and patients requiring new processes for arranging appointments and challenges to establish rapport. Telehealth also prevented many traditional clinical assessments and provision of treatment, leaving health care providers uncertain as to the overall effectiveness and quality of the care provided. However, the physical therapy profession was well suited to transition to the use of telehealth during the SARS-CoV-2 pandemic. Modifications to traditional care led to innovative methods in delivering physical therapy services,¹ including increasing the focus on patient education and empowerment and interventions that emphasized functional activities in real-world settings,² warranting further assessment of how providers can incorporate technology into clinical practice moving forward.

Although both patients and providers became more familiar with using telehealth following pandemic restrictions, there is uncertainty regarding its permanent role in clinical practice as social distancing restrictions ease and in-person health care visits have resumed. Instead of completely abandoning telehealth as the pandemic subsides, recent discussion has centered on its use with specific patient populations,³ including those living in rural and remote communities and those who prefer receiving instruction on self-management and education with exercise rather than hands-on interventions.³ However, the long-term effectiveness of telehealth compared with in-person physical therapy for common musculoskeletal diagnoses, including lower extremity tendinopathies, remains unclear.

Randomized controlled trials and reviews conducted prior to the SARS-CoV-2 pandemic have examined the effectiveness of using telehealth for rehabilitation of conditions such as rheumatic diseases,⁴ chronic and general musculoskeletal pain,^5,6 and postoperative orthopedic care.^7–10 Compared with face-to-face visits or standard care, these studies showed encouraging results with respect to the use of telehealth on outcomes related to pain, function, and psychological factors associated with chronic musculoskeletal pain, including pain catastrophizing and self-efficacy.^11,12 More recently, 1 large urban health care system reported high rates of patient-reported satisfaction and willingness to complete telehealth visits during the beginning of the SARS-CoV-2 pandemic.¹³ Despite these promising studies, a recent review highlighted that previous systematic reviews provide only a weak level of evidence supporting the use of telehealth and that there is an urgent need for research that captures use of current telehealth technology, includes a wider range of diagnoses, and uses appropriate noninferiority analyses.^14,15 Moreover, additional information on the effectiveness of telehealth not only for pain and function but also for key associated outcomes, including psychological factors and therapeutic alliance, is needed to enhance the confidence of clinicians in the use of telehealth and to inform clinical decision-making about which patients are appropriate for telehealth.¹⁶

The treatment of Achilles tendinopathy (AT) is well-suited for telehealth, with best practice rehabilitation strategies that include tendon-loading exercise and education.^17,18 The exercise progression for AT can be implemented in a home setting without equipment or hands-on feedback, and the activities can be incorporated into daily activities. Thus, telehealth technology may provide an appropriate platform to prescribe tendon-loading exercises as well as patient education on tendinopathy rehabilitation and activity modification to facilitate return to function. Yet there is currently limited evidence on the effectiveness of using telehealth for lower extremity tendinopathy, with only 1 qualitative study on a gym-based exercise program for AT that was monitored via telehealth.¹⁹ Determining the efficacy of physical therapist–directed care including exercise and education for AT on outcomes achieved via in-person, telehealth, or a hybrid approach (combination of telehealth and in-person visits) for individuals with chronic AT may influence patient preference on physical therapy visit format, ultimately reducing the burden of attending in-person physical therapy appointments and potentially enhancing patient engagement. Therefore, we aimed to compare the efficacy of an exercise and education program delivered via all in-person compared with all telehealth or to a hybrid format on movement-evoked pain for individuals with chronic AT. We hypothesized that participants who completed the exercise and education program via telehealth or hybrid approach would demonstrate improvements in movement-evoked pain that are no worse than those who completed all in-person visits. To test this, we utilized a noninferiority analysis. A secondary aim of this analysis was to determine the effect that the mode of delivery had on outcomes related to function, psychological variables, therapeutic alliance, and participant-perceived changes in AT symptoms.

Methods

This was a secondary analysis of a randomized, blinded, placebo-controlled trial for individuals with chronic AT: Tendinopathy Education on the Achilles.¹⁸ Details specific to the aims of this secondary analysis are provided below, with a more comprehensive description of the methods provided in the protocol paper.²⁰ All participants completed the same progressive tendon-loading exercise program and received patient education on AT.

Participants

Participants with chronic AT were recruited from the University of Iowa and local community through mass emails and referrals from the Department of Orthopaedics and Rehabilitation from September 2019 to December 2020. Potential participants completed an online screening for eligibility and were evaluated by an experienced physical therapist to confirm the diagnosis of AT and their eligibility. The inclusion criteria were as follows: primary source of pain localized to Achilles tendon insertion or midportion; localized pain ≥3 of 10 in the Achilles tendon (midportion, insertion, unilateral, or bilateral) during walking, heel raises, or hopping; and pain that increased ≥1 point on an 11-point scale with increasing load. The exclusion criteria were as follows: younger than 18 years old; inability to read and write in English; Achilles tendon pain for <3 months; history of Achilles tendon rupture verified by surgical or conservative management; treatment for AT other than exercise (eg, injection, surgery) in the preceding 3 months; systemic inflammatory condition (eg, rheumatoid arthritis, ankylosing spondylitis); endocrine disorder (eg, type 1 or 2 diabetes); connective tissue disorder (eg, Marfan syndrome); cardiovascular condition that might be exacerbated by a 90-second submersion of a hand in cold water (Raynaud syndrome, cold-contact urticaria); cardiovascular condition that prevents participation in an exercise program; history of taking fluoroquinolones in the preceding 3 months; foot and ankle pain owing to other causes, such as posterior impingement, bursitis, paratendonitis, sural nerve injury, ankle osteoarthritis, and radicular or referred symptoms (pain, altered sensation, weakness, and altered reflexes) from the lumbar spine into the lower extremities; and Four Square Step Test score >15 seconds (in-person fall risk assessment). The study was approved by the University of Iowa Institutional Review Board and registered on www.clinicaltrials.gov (NCT04059146) and Open Science Framework (https://osf.io, JF2XU). All participants provided informed consent prior to participating.

Participant Groups and Mode of Delivery

Participants were divided into 3 groups based on mode of delivery for the exercise and education program: in-person, telehealth, or hybrid. All in-person visits were completed from September 2019 to March 16, 2020, when a telehealth format was adopted due to university restrictions on in-person human research. Participants in the telehealth group completed all visits via the Zoom platform and only engaged the study staff remotely. Telehealth visits were completed in a one-on-one format with the treating physical therapist using a laptop computer with a camera. Participants were permitted to utilize a variety of devices (eg, cell phone, tablet, laptop). To help participants prepare for telehealth visits, they were emailed a handout with information to assist with logging into Zoom, hardware requirements, and ensuring a strong internet connection (available at https://doi.org/10.25820/data.006188). On July 15, 2020, restrictions on human research were lifted, and participants were offered the option to complete their visits either in-person or via telehealth based on their preference. The hybrid group consisted of participants who enrolled in the study prior to March 16, 2020, and transitioned into telehealth visits to maintain participation and participants who chose to complete at least 1 visit via telehealth following the resumption of in-person human research (Fig. 1). Participants were not eligible for telehealth visits if they reported symptoms that required consistent assessment of vital signs, including uncontrolled hypertension and/or inconsistent use of hypertension medications. Prior to the initiation of the exercise protocol, vital signs were screened for all participants who completed in-person visits according to the American College of Sports Medicine Guidelines for Exercise Testing.²¹ With the transition to telehealth, these stopping criteria (eg, uncontrolled hypertension) were converted to verbal screening questions to determine if medical clearance was needed prior to participation.

Figure 1

Participant visit format throughout the duration of enrollment. All participants completed 8 total visits (2 evaluations and 6 follow-up visits). A single bin includes all visits for an individual participant, and the visit format is represented by color. Beginning in March 2020, in-person research was paused, and the study shifted to an all-telehealth format. In July 2020, in-person research was reinitiated, and visits were completed based on the participant’s preference for either in-person or telehealth.

Intervention

Participants completed a baseline evaluation visit at 0 weeks, 6 to 7 one-on-one follow-up visits with a physical therapist over the following 8 weeks, and 2 additional evaluations at the 8- and 12-week time points. To maximize consistency in provision of the intervention, videos and handouts were provided to participants about the exercise and educational programs (available at https://doi.org/10.25820/data.006166). The exercise log and educational homework were both completed via online surveys. The digital format of the intervention facilitated the transition to telehealth, where all materials could be sent to participants in advance via email and reviewed together during visits using a shared screen. All participants received the same tendon-loading exercise program that consisted of 3 phases, including isometric heel raises, concentric-eccentric heel raises, and a functional spring phase (Suppl. Fig. 1). The progressive tendon-loading program required only the use of body weight and could be completed at home. Participants were also encouraged to continue their personal exercise or recreational activities as tolerated.²² Progression of the tendon-loading program was based on functional criteria, participant’s pain rating, participant preference, and the physical therapist’s clinical judgment. Modifications to the loading program were implemented at each follow-up visit, including alterations to exercise type, sets, and repetitions. After the primary endpoint (8 weeks), participants were encouraged to continue with their home exercise program (HEP). Participants received 1 of 2 standardized patient education programs that were delivered during follow-up visits over the same 8-week period where they received education about either pain science related to AT symptoms (pain science education) or the pathoanatomy of AT (pathoanatomical education). The treating physical therapist completed 1 additional follow-up via phone or email at 10 weeks to assess the participant’s HEP and make any necessary adjustments to the loading and dosage. To facilitate consistent engagement with their HEP, participant preference was incorporated into the decision-making process on the rate of progression within the exercise protocol. The patient education program included free response questions about the participant’s pain experience, which provided the therapist opportunities for active listening. A single physical therapist provided care throughout the duration of the study for all modes of delivery. The treating licensed physical therapist had 7 years of clinical experience with additional training as an orthopedic clinical specialist and as a fellow of the American Academy of Orthopedic Manual Physical Therapy.

Outcomes

The baseline and 8-week (primary endpoint) evaluation visits were completed in-person or via Zoom. At 12 weeks (secondary endpoint), participants completed surveys via an automated survey link to REDCap. Performance-based outcomes were collected at baseline and 8 weeks according to the instructions given by the outcomes assessor. Survey-based outcomes were completed at baseline, 8 weeks, and 12 weeks.

AT Symptoms

The primary outcome was movement-evoked pain during heel raises. Movement-evoked pain during heel raises is commonly utilized to evaluate the severity of AT and inform clinical decision-making on the prescription of tendon-loading exercise dosage and frequency. Movement-evoked pain is characterized as pain resulting from a specified physical activity.^23,24 Participants reported pain in their Achilles tendon on their most painful side during single-limb heel raises using an 11-point numeric pain rating scale (NRS), where 0 indicates “no pain” and 10 indicates “worst pain imaginable.”²⁵ The NRS has moderate to high test–retest reliability (r = 0.67–0.96) and high convergent validity (r = 0.79–0.95).²⁵ A decrease of 2 points on the NRS is the minimal clinically importance difference (MCID) in musculoskeletal conditions.²⁶ The Victorian Institute of Sports Assessment–Achilles (VISA-A) questionnaire is a reliable and valid measurement for Achilles tendon symptom severity.²⁷ Scores on the VISA-A range from 0 to 100, where a lower score indicates more severe AT symptoms.

Function

Self-reported function was assessed using the Patient-Reported Outcomes Measurement Information System (PROMIS) 2.0 for Physical Function (PROMIS-PF) with the computer adaptive testing function. The PROMIS-PF has excellent reliability (r = 0.96) and good construct and convergent validity (r = 0.792) with foot and ankle diagnoses.^28,29 Participants also completed the maximum number of heel raises on the most painful side. Participants were instructed to “raise your heel up and down as high as possible, as many times as you can. Please try to keep your knee straight and your trunk upright. You can use fingertip balance as needed.” The activity was discontinued when the participant was unable to continue with good technique (knee bending, trunk flexion) or due to muscle fatigue.

Psychological Variables

Kinesiophobia is defined as an excessive, irrational, and debilitating fear to carry out a physical movement due to feeling vulnerable to painful injury or reinjury.³⁰ Kinesiophobia was assessed using the Tampa Scale for Kinesiophobia 17-item (TSK-17) questionnaire.³¹ Scores on the TSK-17 range from 17 to 68, with a higher score indicating greater kinesiophobia.³¹ Pain catastrophizing is defined as magnification, rumination, and a sense of helplessness due to pain and was assessed using the Pain Catastrophizing Scale 13-item (PCS-13) questionnaire.³² Scores on the PCS-13 range from 0 to 52, with higher scores indicating great pain catastrophizing. PROMIS domains for anxiety, self-efficacy for managing symptoms, and depression were also assessed. All PROMIS measures have a general population T-score of 50 and SD of 10. Therefore, a patient with an anxiety T-score of 40 is 1 SD below the US general population mean.

Treatment Fidelity and Therapeutic Alliance

Treatment fidelity was assessed using the average duration of visit length (minutes) and by documenting the percentage of prescribed HEP completed by participants between visits to assess for adherence to the treatment plan. Therapeutic alliance was assessed using 2 questions from the Working Alliance Inventory (WAI).³³ The WAI is scored on a 7-point scale, where 0 = “never” and 7 = “always.” Questions included from the WAI were “I was confident in my physical therapist’s ability to help me” and “My physical therapist and I were working towards mutually agreed upon goals.” Participants provided responses to the WAI questions at the 8-week time point.

Perceived Change

Patient-perceived change in symptoms was assessed using the Global Rating of Change (GROC). The GROC is a 15-point scale anchored at −7 or “a very great deal worse” to +7 or “a very great deal better,” with 0 representing “about the same.” The MCID for the GROC is +4 points.³⁴ Participants were asked to rate their anticipated improvement prior to their initial evaluation and rated their perceived improvement at the 8- and 12-week time points.

Data Analysis

We chose a noninferiority statistical approach to compare the effect of different modes of delivering exercise and education for AT on movement-evoked pain. A noninferiority approach can demonstrate that a novel treatment is no worse than standard care by a prespecified amount (noninferiority margin), with the premise that the novel treatment offers an additional advantage, including availability of resources, cost, and patient preference.^35,36 The normality of continuous variables was assessed using the Kolmogorov–Smirnov test and by examining the quantile-quantile plot. Baseline participant demographics and outcomes for the 3 groups were compared using 1-way analysis of variance for continuous variables and the χ² test for categorical variables. A noninferiority margin was established using the MCID for movement-evoked pain for chronic musculoskeletal pain conditions (2 out of 10 on the NRS).²⁶ Noninferiority of the telehealth and hybrid groups for movement-evoked pain was established if the lower bound of the 95% CI of the absolute mean difference between groups (telehealth—in-person or hybrid—in-person) did not surpass the preestablished noninferiority margin for movement-evoked pain. Two independent-samples t test comparisons between groups were performed separately (telehealth vs in-person and hybrid vs in-person). A noninferiority approach was also utilized for secondary outcomes related to function and psychological variables. The upper (or lower, depending on the direction of change from baseline) bound of the 95% CI of the absolute mean difference between groups was examined in relation to the noninferiority margin (MCID) of each outcome (eg, a lower score at 8 weeks for kinesiophobia demonstrates improvement, and an elevated score for self-efficacy indicates improvement). Depending on the normality assumption, 1-way analysis of variance or the nonparametric alternative of Kruskal-Wallis tests and χ² tests was used to examine for differences between groups in treatment fidelity, therapeutic alliance, and the GROC. Secondary to the 3 groups not being concurrently enrolled, we also evaluated for the potential confounding effect of time during the SARS-CoV-2 pandemic on baseline values of the primary and secondary outcomes using 1-way analysis of variance between 3 time cohorts: September 2019 to March 15, 2020 (cohort 1); March 16 to July 15, 2020 (cohort 2); and July 16 to December 16, 2020 (cohort 3).

Role of the Funding Source

The funders played no role in the design, conduct, or reporting of this study.

Results

Baseline Demographics and Outcomes

A total of 66 individuals were enrolled, with 25 participants completing all in-person visits, 20 participants completing all telehealth visits, and 21 participants completing visits via the hybrid format (Fig. 2). Participants in the hybrid group completed 81 in-person visits (48%) and 87 telehealth visits (52%). No significant differences in demographics or baseline values for movement-evoked pain, function, or psychological variables were found between groups (Tabs. 1 and 2).

Figure 2

CONSORT flow chart of study participants and mode of physical therapy delivery through the clinical trial. AT = Achilles tendinopathy; m = months.

AT Symptoms

All groups demonstrated decreases in movement-evoked pain with heel raises that surpassed the MCID for pain by 8 weeks (NRS, from 0 to 10; in-person: 3.0 [95% CI = 1.9 to 4.2] [P < .001]; telehealth: 3.5 [95% CI = 2.4 to 4.5] [P < .001]; hybrid: 3.5 [95% CI = 2.6 to 4.4] [P < .001]) (Tab. 2). The absolute difference for changes in movement-evoked pain from baseline to 8 weeks between the telehealth and in-person groups was 0.45 (NRS, from 0 to 10; 95% CI = −1.1 to 2.0), and the absolute difference between the hybrid and in-person groups was 0.48 (NRS, from 0 to 10; 95% CI = −1.0 to 1.9). Because the lower bounds of the 95% CIs did not surpass the preestablished noninferiority margin (2 out of 10) for movement-evoked pain in both the telehealth and hybrid groups (Tab. 3; Fig. 3A), both telehealth and hybrid group outcomes were noninferior to the in-person group. At the primary endpoint (8 weeks), all groups demonstrated improvements in the VISA-A beyond the MCID of 10³⁷ (in-person: 14.9 [95% CI = 7.6 to 22.3]; telehealth: 14.6 [95% CI = 6.4 to 22.8]; hybrid: 20.7 [95% CI = 11.0 to 30.3]) (Tab. 2). Noninferiority for improvement on the VISA-A was observed in the hybrid group compared with the in-person group at the 8-week time point but not at the 12-week time point (Figs. 3C and 3D). In contrast, noninferiority on the VISA-A was not demonstrated in the telehealth group compared with the in-person group at any time point (Figs. 3C and 3D).

Function

Noninferiority was demonstrated in both the telehealth and hybrid groups at 8 and 12 weeks for the PROMIS-PF (Figs. 3G and 3H). All groups demonstrated increases in the number of single-leg heel raises performed from baseline to 8 weeks (in-person: 6.8 [95% CI = 2.8 to 10.7]; telehealth: 4.8 [95% CI = 2.4 to 7.1]; hybrid: 3.4 [95% CI = −1.9 to 8.7]) (Tab. 2). Both the telehealth and hybrid groups demonstrated noninferiority compared with the in-person group for the maximum number of heel raises performed at 8 weeks (Fig. 3B).

Psychological Variables

All groups demonstrated improvement on the TSK-17, PCS-13, PROMIS–self-efficacy, and PROMIS-anxiety questionnaires. Improvement on the TSK-17 in all groups approached or surpassed the MCID of 6 points at 8 weeks³⁸ (in-person: 7.3 [95% CI = 5.0 to 9.7]; telehealth: 6.6 [95% CI = 4.1 to 9.0]; hybrid: 5.5 [95% CI = 3.0 to 7.9]) (Tab. 2). Noninferiority was established in the telehealth and hybrid groups on the TSK-17, PCS-13, and PROMIS-anxiety assessments at 8 and 12 weeks (Figs. 3E and 3F). The telehealth and hybrid groups did not demonstrate noninferiority on the PROMIS–self-efficacy assessment at 8 and 12 weeks compared with the in-person group. Only the hybrid group demonstrated noninferiority on the PROMIS-depression assessment at 8 and 12 weeks; the telehealth group did not demonstrate noninferiority at either time point.

Treatment Fidelity, Therapeutic Alliance, and Perceived Change

There were no significant differences in the average length of physical therapy visits between groups (in-person: 37.5 [SD = 6.9] minutes; telehealth: 37.4 [SD = 7.9] minutes; hybrid: 38.9 [SD = 7.7] minutes; P = .78) or the average percentage of HEPs that were completed (in-person: 97.5 [83.3 to 100.0]; telehealth: 95.2 [89.4 to 100.0]; hybrid: 100.0 [86.4 to 100.0]; P = .62).

No significant differences in therapeutic alliance were demonstrated between groups for either WAI question (Suppl. Tab. 1). There were also no significant differences between groups on anticipated GROC (in-person: 5.0 [3.5–6.0]; telehealth: 4.5 [3.3–5.0]; hybrid: 5.0 [4.0–6.0]; P = .26), GROC at 8 weeks (in-person: 4.0 [3.0–6.8]; telehealth: 5.5 [5.0–6.0]; hybrid: 6.0 [4.0–6.5]; P = .20), or GROC at 12 weeks (in-person: 5.0 [3.8–7.0]; telehealth: 6.0 [5.0–7.0]; hybrid: 6.0 [5.0–7.0]; P = .15) (Suppl. Tab 1).

Potential Confounding Effect of Time

There were no significant differences between time cohorts for the primary or secondary outcomes, except for the TSK-17 (Suppl. Tab. 2). A significant difference between time cohorts was found for TSK-17 values at baseline (cohort 1: 37.4 [SD = 5.1]; cohort 2: 39.8 [SD = 5.5]; cohort 3: 35.3 [SD = 5.9]; P = .04). Post hoc analysis demonstrated that at baseline, cohort 2 (39.8 [SD = 5.5]) had a higher TSK-17 than cohort 3 (35.3 [SD = 5.9]) (P = .03). No differences were found between time cohorts for the TSK-17 at other time points (8 weeks: P = .14; 12 weeks: P = .22).

Table 1

Baseline Demographics Presented by Group (Mode of Delivery of Physical Therapy)^a

Characteristic	In-Person	Telehealth	Hybrid	P
Sample size	25	20	21
Group randomization
PAE	13	8	12	.64b
PSE	12	12	9	.12c
Age, mean (SD), y	46.0 (14.4)	44.25 (15.0)	39.3 (17.0)	.46
Sex				.49
Male	48.0 (12)	50.0 (10)	33.3 (7)
Female	52.0 (13)	50.0 (10)	66.7 (14)
AT type				.58
Insertional	64.0 (16)	55.0 (11)	47.6 (10)
Midportion	36.0 (9)	45 (9)	52.4 (11)
BMI, median (IQR)	31.3 (25.1– 37.8)	28.2 (26.6–33.8)	26.3 (21.6– 29.6)	.10
Pain duration, median (IQR), mo	12.0 (6.0– 24.0)	15.0 (6.0–36.0)	24.0 (14.5– 48.0)	.07
Stiffness duration, median (IQR), minutes	30.0 (20.0–47.0)	21.5 (16.3–35.0)	24.0 (13.0–50.0)	.65
Race				.87
White	88.0 (22)	85.0 (17)	81.0 (17)
Black/African American	8.0 (2)		9.5 (2)
Asian		5.0 (1)
American Indian/Alaska Native	4.0 (1)		4.8 (1)
Hispanic/Latino
Other		10.0 (2)	4.8 (1)
Do not know/not sure
Ethnicity: “Do you consider yourself Hispanic, Latino, or of Spanish origin?”	0.0 (0)	10.0 (2)	4.8 (1)	.28
Education				.23
8th grade or less	0 (0)
Some high school	0 (0)
High school diploma/GED	4.0 (1)		4.8 (1)
Some college	16.0 (4)	10.0 (2)	28.6 (6)
Vocational/technical/associate’s certificate/degree	4.0 (1)
Bachelor’s degree	40.0 (10)	30.0 (6)	38.1 (8)
Advanced/professional degree (eg, MA, PhD, MD)	36.0 (9)	60.0 (12)	28.6 (6)
Sought provider earlier	76.0 (19)	75.0 (15)	95.2 (20)	.16
No. of providers sought earlier, median (IQR)	2.0 (0.5–3.0)	1.5 (0.3–2.0)	2.0 (1.0– 3.0)	.30
Providers sought
Physical therapist	60.0 (15)	45.0 (9)	61.9 (13)
Primary care physician	36.0 (9)	30.0 (6)	57.1 (12)
Orthopedic surgeon	12.0 (3)	15.0 (3)	19.0 (4)
Sports medicine physician	20.0 (5)	10.0 (2)	33.3 (7)
Podiatrist	16.0 (4)	30.0 (6)	23.8 (5)
Acupuncturist	4.0 (1)	0.0 (0)	4.8 (1)
Massage therapist	8.0 (2)	5.0 (1)	9.5 (2)
Chiropractor	12.0 (3)	10.0 (2)	23.8 (5)
Other	4.0 (1)	5.0 (1)	4.8 (1)
Tried treatment earlier	92.0 (23)	100.0 (20)	100.0 (21)	.19
No. of treatments tried earlier, median (IQR)	4.0 (3.0–6.0)	5.5 (4.0–6.0)	6.0 (4.0–7.0)	.13
Treatments tried
Strengthening exercises	56.0 (14)	65.0 (13)	76.2 (16)
Stretching	76.0 (19)	90.0 (18)	100.0 (21)
Manual therapy	20.0 (5)	35.0 (7)	23.8 (5)
Taping	24.0 (6)	50.0 (10)	14.3 (3)
Heel lifts	28.9 (7)	45.0 (9)	66.7 (14)
Foot orthotics	24.0 (6)	40.0 (8)	42.9 (9)
Iontophoresis	0.0 (0)	0.0 (0)	0.0 (0)
Night splints	12.0 (3)	10.0 (2)	19.0 (4)
Laser therapy	4.0 (1)	0 (0)	4.8 (1)
Alterations in running technique	4.0 (1)	15.0 (3)	9.5 (2)
Massage/soft-tissue techniques	32.0 (8)	55.0 (11)	57.1 (12)
Instrument-assisted soft-tissue mobilization	12.0 (3)	5.0 (1)	14.3 (3)
Dry needling	4.0 (1)	5.0 (1)	9.5 (2)
Pain science education	0.0 (0)	0.0 (0)	4.8 (1)
Extracorporeal shock wave therapy	4.0 (1)	5.0 (1)	4.8 (1)
Oral pain medication (eg, NSAIDs)	60.0 (15)	40.0 (8)	57.1 (12)
Ice	56.0 (14)	65.0 (13)	76.2 (16)
Injections (PRP, sclerosing therapy, prolotherapy, corticosteroids)	4.0 (1)	5.0 (1)	9.5 (2)
Surgery	0.0 (0)	0.0 (0)	0.0 (0)
Treatments effective earlier	68.0 (17)	80.0 (16)	90.5 (19)	.18
No. of treatments effective earlier, median (IQR)	2.0 (0.0– 3.5)	2.5 (1.0–3.0)	3.0 (2.0–3.0)	.59
Treatments effective
Strengthening exercises	32.0 (8)	35.0 (7)	38.1 (8)
Stretching	48.0 (12)	50.0 (10)	81.0 (17)
Manual therapy	12.0 (3)	0.0 (0)	9.5 (2)
Taping	4.0 (1)	25.0 (5)	4.8 (1)
Heel lifts	4.2 (1)	15.0 (3)	9.5 (2)
Foot orthotics	12.5 (3)	10.0 (2)	19.0 (4)
Iontophoresis	0.0 (0)	0.0 (0)	0.0 (0)
Night splints	4.0 (1)	5.0 (1)	9.5 (2)
Laser therapy	0.0 (0)	0.0 (0)	0.0 (0)
Alterations in running technique	4.0 (1)	0.0 (0)	4.8 (1)
Massage/soft-tissue techniques	8.0 (2)	25.0 (5)	23.8 (5)
Instrument-assisted soft-tissue mobilization	8.0 (2)	0.0 (0)	0.0 (0)
Dry needling	4.0 (1)	5.0 (1)	4.8 (1)
Pain science education	0.0 (0)	0.0 (0)	0.0 (0)
Extracorporeal shock wave therapy	4.0 (1)	5.0 (1)	0.0 (0)
Oral pain medication (eg, NSAIDs)	24.0 (6)	25.0 (5)	23.8 (5)
Ice	20.0 (5)	40.0 (8)	38.1 (8)
Injections (PRP, sclerosing therapy, prolotherapy, corticosteroids)	4.0 (1)	0.0 (0)	4.8 (1)
Surgery	4.0 (1)	0.0 (0)	0.0 (0)

^a

Data are presented as percentage (number) of participants unless otherwise indicated. AT = Achilles tendinopathy; BMI = body mass index; IQR = interquartile range; NSAIDs = non-steroidal anti-inflammatory drugs; PAE = pathoanatomical education; PSE = pain science education; PRP = platelet-rich plasma.

^b

χ² test comparison of virtual and in-person data.

^c

χ² test for hybrid and in-person data.

Table 2

Outcomes at Baseline and 8 and 12 Weeks, Presented by Group (Mode of Delivery of Physical Therapy)^a

Outcome and Time Point	Mean (SD) for:			P
	In-Person	Telehealth	Hybrid
Movement-evoked pain, measured with NRS, scored from 0 to 10
Baseline	4.7 (2.0)	5.4 (2.4)	4.9 (2.0)	.55
8 wk	1.7 (2.0)	2.0 (1.7)	1.4 (1.3)	.61
Maximum no. of heel raises
Baseline	17.6 (12.9)	15.4 (6.0)	16.6 (12.5)	.80
8 wk	24.5 (14.2)	20.1 (6.2)	20.0 (7.7)	.24
VISA-A, scored from 0 to 100
Baseline	40.8 (16.5)	48.3 (16.5)	44.9 (17.4)	.33
8 wk	55.0 (18.3)	62.9 (13.3)	65.6 (19.4)	.11
12 wk	63.3 (21.8)	68.9 (19.0)	66.5 (19.3)	.66
TSK-17, scored from 17 to 68
Baseline	36.8 (5.3)	39.8 (5.5)	35.9 (5.8)	.07
8 wk	29.5 (6.4)	33.3 (6.6)	30.4 (6.2)	.15
12 wk	29.9 (6.8)	32.5 (6.8)	30.8 (7.0)	.47
PCS-13, scored from 0 to 52
Baseline	10.1 (6.0)	12.1 (7.5)	11.0 (7.5)	.64
8 wk	5.3 (4.3)	7.9 (8.2)	4.3 (3.6)	.11
12 wk	4.4 (4.1)	8.3 (9.6)	5.2 (6.9)	.20
PROMIS-PF, scored from 20 to 80
Baseline	47.8 (5.9)	49.9 (7.2)	47.5 (7.3)	.47
8 wk	50.2 (6.8)	51.3 (5.8)	50.2 (8.6)	.84
12 wk	50.3 (7.1)	52.4 (6.2)	53.0 (9.0)	.44
PROMIS–self-efficacy, scored from 20 to 80
Baseline	46.5 (6.4)	46.9 (7.1)	46.7 (6.1)	.98
8 wk	52.8 (9.6)	52.9 (6.9)	51.3 (7.9)	.77
12 wk	52.6 (10.6)	54.6 (8.8)	53.5 (8.4)	.79
PROMIS-anxiety, scored from 20 to 80
Baseline	52.4 (10.2)	55.1 (8.6)	56.0 (8.0)	.38
8 wk	50.3 (9.7)	53.2 (8.7)	52.6 (9.2)	.56
12 wk	50.9 (9.1)	52.2 (9.6)	51.8 (10.0)	.89
PROMIS-depression, scored from 20 to 80
Baseline	47.1 (7.1)	46.2 (8.4)	50.6 (8.0)	.16
8 wk	45.0 (7.5)	47.4 (9.1)	50.1 (8.2)	.13
12 wk	45.8 (8.2)	48.1 (7.5)	50.4 (7.0)	.14

^a

NRS = numeric pain rating scale; PCS-13 = Pain Catastrophizing Scale 13-item questionnaire; PF = physical function; PROMIS = Patient-Reported Outcomes Measurement Information System; TSK-17 = Tampa Scale of Kinesiophobia 17-item questionnaire; VISA-A = Victorian Institute of Sport Assessment–Achilles.

Table 3

Comparisons of Absolute Mean Differences and 95% CIs of Telehealth vs In-Person Groups and Hybrid vs In-Person Groups at Each Time Point^a

Domain	Variable	Change Score From Baseline to 8 wk, Mean (SD)	Change Score From Baseline to 12 wk, Mean (SD)	8-wk Pooled Difference (95% CI)	12-wk Pooled Difference (95% CI)
AT symptoms	Pain with heel raises, scored from 0 to 10
	In-person	−3.0 (2.7)	N/A	N/A	N/A
	Telehealth	−3.5 (2.3)	N/A	0.5 (−1.1 to 2.0)	N/A
	Hybrid	−3.5 (1.9)	N/A	0.5 (−1.0 to 1.9)	N/A
	VISA-A
	In-person	14.9 (17.4)	23.8 (19.1)	N/A	N/A
	Telehealth	14.6 (17.5)	20.6 (19.1)	−0.3 (−11.0 to 10.4)	−3.2 (−15.1 to 8.8)
	Hybrid	20.7 (21.2)	21.6 (22.1)	5.8 (−5.8 to 17.4)	−2.2 (−14.9 to 10.5)
Function	PROMIS-PF
	In-person	2.4 (4.6)	2.4 (4.5)	N/A	N/A
	Telehealth	1.4 (4.5)	2.5 (6.0)	−0.9 (−3.7 to 1.9)	0.09 (−3.1 to 3.3)
	Hybrid	2.7 (5.2)	5.5 (6.4)	0.3 (−2.6 to 3.3)	3.1 (−0.3 to 6.5)
	No. of heel raises
	In-person	6.8 (9.2)	N/A	N/A	N/A
	Telehealth	4.8 (5.0)	N/A	−0.2 (−1.4 to 0.9)	N/A
	Hybrid	3.4 (11.6)	N/A	0.3 (−0.7 to 1.3)	N/A
Psychological	TSK-17
	In-person	−7.3 (5.6)	−6.7 (6.3)	N/A	N/A
	Telehealth	−6.6 (5.1)	−7.4 (6.0)	0.8 (−2.5 to 4.1)	−0.6 (−4.5 to 3.2)
	Hybrid	−5.5 (5.4)	−5.1 (6.1)	1.9 (−1.5 to 5.2)	1.6 (−2.3 to 5.4)
	PCS-13
	In-person	−4.5 (4.5)	−5.3 (6.2)	N/A	N/A
	Telehealth	−4.2 (6.7)	−3.9 (6.1)	0.3 (−3.1 to 3.8)	1.5 (−2.4 to 5.3)
	Hybrid	−6.7 (6.1)	−5.7 (9.4)	−2.1 (−5.4 to 1.1)	−0.4 (−5.3 to 4.5)
	PROMIS–self-efficacy
	In-person	6.4 (10.7)	6.6 (11.6)	N/A	N/A
	Telehealth	5.9 (7.4)	7.7 (12.6)	−0.4 (−6.1 to 5.3)	1.1 (−6.5 to 8.6)
	Hybrid	4.6 (9.1)	6.9 (7.6)	−1.8 (−7.8 to 4.3)	0.3 (−5.8 to 6.3)
	PROMIS-anxiety
	In-person	−1.1 (5.8)	−1.4 (6.4)	N/A	N/A
	Telehealth	−1.9 (5.4)	−2.9 (6.2)	−0.8 (−4.3 to 2.6)	−1.6 (−5.5 to 2.4)
	Hybrid	−3.4 (7.0)	−4.2 (6.5)	−2.3 (−6.1 to 1.6)	−2.8 (−6.8 to 1.2)
	PROMIS-depression
	In-person	−1.8 (5.8)	−1.7 (6.1)	N/A	N/A
	Telehealth	1.2 (4.7)	1.9 (5.6)	2.9 (−0.3 to 6.2)	3.6 (−0.1 to 7.3)
	Hybrid	−0.5 (3.5)	−0.2 (4.7)	1.3 (−1.7 to 4.2)	1.5 (−1.9 to 4.8)

^a

AT = Achilles tendinopathy; N/A = not applicable; PCS-13 = Pain Catastrophizing Scale 13-item questionnaire; PF = physical function; PROMIS = Patient-Reported Outcomes Measurement Information System; TSK-17 = Tampa Scale of Kinesiophobia 17-item questionnaire; VISA-A = Victorian Institute of Sport Assessment–Achilles.

Individual Participant Data Sharing Plan

Study outcomes and descriptive metadata were deposited in the University of Iowa open-access institutional repository, Iowa Research Online at https://doi.org/10.25820/data.006188.

Discussion

The primary objective of this secondary analysis was to examine the effect of delivering care through different methods (eg, in-person, telehealth, or hybrid) on movement-evoked pain in individuals with chronic AT pain. Our data supported our hypothesis that individuals who participated in an exercise and education program through either telehealth or hybrid visits had improvements in movement-evoked pain that were no worse at 8 weeks after completing treatment relative to those who received in-person care. Additionally, for many of the outcomes related to function and psychological variables (kinesiophobia, pain catastrophizing, and anxiety), noninferiority analyses demonstrated similar effects in the telehealth and hybrid groups compared with the in-person group. We also found similar levels of patient participation in the HEPs between groups. Therapeutic alliance was reported at high levels in all 3 groups with no significant differences, and a continued increase in participant’s perceived improvement at each time point was reported with no significant differences between all modes of physical therapy delivery.

Telehealth and Hybrid Formats Have No Worse Pain Outcomes Than In-Person Visits

All 3 groups in our study demonstrated decreased Achilles tendon pain during single-leg heel raises from baseline to 8 weeks that surpassed the MCID for musculoskeletal pain with no significant differences between groups.²⁶ Consistent with our findings, in a noninferiority study with individuals who completed a 6-week exercise and education program via either telehealth or in-person visits following a total knee arthroplasty, Russell et al.¹⁰ reported no differences between groups in knee pain outcomes. Similarly, individuals following a subacromial decompression completing rehabilitation either through telehealth or in-person demonstrated improvements in shoulder pain, with no significant differences between groups at 12 weeks.³⁹ A recent meta-analysis found that both telehealth and in-person formats decreased pain over time and that 1 format is not more favorable than the other when addressing musculoskeletal pain.¹⁵ The current study on chronic AT adds to this growing body of literature indicating that telehealth provides an effective alternative to in-person care for the management of musculoskeletal pain.

Positive Psychosocial Effects of Physical Therapy via All Modes of Delivery

Pain-related psychological factors, including kinesiophobia and self-efficacy for managing symptoms, may impede on participation in a tendon-loading exercise program in those with chronic AT.⁴⁰ All groups initially demonstrated baseline scores on the TSK-17 around or exceeding the threshold of ≥37 for elevated kinesiophobia^31,38 and decreased below the threshold by 8 weeks and were maintained through the 12-week time point. Moreover, all 3 groups in our study demonstrated improvements in the PROMIS self-efficacy assessment at 8 and 12 weeks. A previous randomized controlled trial completed prior to the pandemic demonstrated improved self-efficacy in individuals with fibromyalgia with delivery of cognitive behavior therapy through in-person and telehealth formats.¹¹ Similarly, another study demonstrated that cognitive behavioral therapy delivered through an internet-based self-management program improved self-efficacy in those with osteoarthritis, rheumatoid arthritis, and fibromyalgia.¹² Telehealth, while minimizing clinician reliance on hands-on assessment and treatment, provides a communication medium with opportunities to focus more on a biopsychosocial approach to identify patient beliefs through active listening and one-on-one, individualized health care visits.

Figure 3

Upper- and lower-bound 95% CIs in relation to prespecified noninferiority margin of telehealth and hybrid groups for movement-evoked pain at 8 weeks (a), heel raises (b), Victorian Institute of Sport Assessment–Achilles (VISA-A) at 8 weeks (c), VISA-A at 12 weeks (d), 17-item Tampa Scale for Kinesiophobia (TSK-17) at 8 weeks (e), TSK-17 at 12 weeks (f), Patient-Reported Outcomes Measurement Information System (PROMIS) 2.0 for Physical Function (PROMIS-PF) at 8 weeks (g), and PROMIS-PF at 12 weeks (h).

Strong Therapeutic Alliance and Perceived Improvement for All Modes of Delivery

In our study, participants developed high therapeutic alliance with their treating physical therapist, regardless of the mode of delivery. A strong therapeutic alliance has been reported to be linked with improvements in chronic musculoskeletal pain for those completing physical therapy.⁴¹ Factors that contribute to developing a strong therapeutic alliance include a trusting relationship between the physical therapist and patient and the ability to individualize treatment plans so that patient values are incorporated and barriers minimized. In our study, each participant met one-on-one with a physical therapist every week and had a plan of care catered to their needs. Physical therapy services provided by telehealth are often conducted in a one-on-one format, thus demonstrating a potential benefit of virtual visits, which could further contribute to reducing chronic musculoskeletal pain conditions.

Another systematic review reported a significant positive association between therapeutic alliance and patients’ perceived effect of treatment.⁴² In our study, participants from all 3 groups reported perceived improvement that surpassed the MCID for the GROC at 8 and 12 weeks. There was also no significant difference in anticipated improvement between groups when participants were being enrolled in the study. Physical therapy services offered through telehealth appointments may cater to a unique patient population with a desire to complete their health care using a virtual format; as such, clinic owners could emphasize marketing strategies to those seeking this format for care.³ Beginning on March 17, 2020, our recruitment strategies highlighted the use of an all-telehealth format and indeed may have accessed the unique population willing to complete physical therapy remotely, ultimately influencing their anticipated improvement and overall outcomes.

Strengths and Limitations

This progressive tendon-loading program for individuals with chronic AT improved movement-evoked pain, function, and psychological variables and is feasible to deliver through different formats (in-person, telehealth, or hybrid). Additionally, these findings are representative of a wide spectrum of patients seeking care for AT, including individuals with both midportion and/or insertional AT and a wide range of body mass indexes, activity levels, and ages. Although the current study did not exclude anyone from participation based on ability and willingness to participate in telehealth, individuals who volunteered to participate in telehealth or hybrid treatment options may differ from individuals who participated in-person. Another limitation was our telehealth group exclusion criteria of individuals with cardiovascular medical conditions, including uncontrolled hypertension, where telehealth precluded our ability to directly assess vitals and ensure participant safety prior to initiation of an exercise program. Rehabilitation delivered via telehealth may not be an ideal format for those who require consistent monitoring of vital signs or screening for medical red flags that would necessitate referral to other health care providers. Future research that identifies participant characteristics of those who select and respond positively to physical therapist–directed care via telehealth may highlight unique patient features to assist clinician decision-making and clinic scheduling practices for in-person vs telehealth visits.

The feasibility of establishing high therapeutic alliance using telehealth is supported by this study, with all 3 groups reporting high therapeutic alliance on 2 questions from the WAI. Yet, a caution is that use of individual questions from the WAI has not been validated, and a more comprehensive assessment using the full version of a validated therapeutic alliance questionnaire may reveal components of therapeutic alliance that differ between treatment delivery formats. Another limitation of this study is the potential for being underpowered for the secondary outcomes and endpoints. Specifically, the study results were inconclusive for the VISA-A, PROMIS–self-efficacy, and PROMIS-depression, where the lower limit of 95% CI extended beyond the noninferiority margin. A larger sample size may provide a more robust representation of the impact of varying modes of physical therapist–directed care for chronic musculoskeletal pain diagnoses. Although the original study was powered for superiority testing for the effect of an education program on of the primary outcome of movement-evoked pain, we explored noninferiority of movement-evoked pain in the secondary manuscript. Finally, the 3 groups in our analysis were not enrolled simultaneously, resulting in a potential cohort effect of time impacting our results. Although a statistically significant difference in baseline TSK-17 values was found, a 3.9-point difference between groups at baseline may not be clinically meaningful; no differences were found for the TSK-17 or any other outcome measure between time cohorts at any other time point.

Conclusion

Individuals with chronic AT who completed an exercise and education program through telehealth or hybrid formats have no worse outcomes than those that receive the same intervention through in-person visits. Using telehealth for AT rehabilitation provides an opportunity to prioritize patient preferences on the format of physical therapy visits and increases accessibility of best practice education and exercise in the management of AT.

Author Contributions

Concept/idea/research design: A.A. Post, E.K. Rio, K.A. Sluka, G.L. Moseley, M.M. Hall, C. de Cesar Netto, J.M. Wilken, E.O. Bayman, R.L. Chimenti

Writing: A.A. Post, E.K. Rio, K.A. Sluka, M.M. Hall, C. de Cesar Netto, J.M. Wilken, E.O. Bayman, R.L. Chimenti

Data collection: A.A. Post, J. Danielson, R.L. Chimenti

Data analysis: A.A. Post, E.O. Bayman, R.L. Chimenti

Project management: A.A. Post, R.L. Chimenti

Fund procurement: R.L. Chimenti, K.A. Sluka, G.L Moseley, A.A Post

Providing participants: M.M. Hall, C. de Cesar Netto, R.L. Chimenti

Providing facilities/equipment: R.L. Chimenti

Consultation (including review of manuscript before submitting): A.A. Post, E.K. Rio, K.A. Sluka, G.L. Moseley, E.O. Bayman, M.M. Hall, C. de Cesar Netto, J. Danielson, J.M. Wilken, R.L. Chimenti

Funding

This study was funded by grants from the National Institute of Arthritis Musculoskeletal and Skin Disease (R00 AR071517) and by the Collaborative Research Grant from the International Association for the Study of Pain. Research reported in this publication was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health (under award nos. UL1TR002537, including use of REDCap, and UL1TR002537). This research was funded in part by a Promotion of Doctoral Studies (PODS) I and II scholarship from the Foundation for Physical Therapy Research. G.L. Moseley was supported by a Leadership Investigator Grant from the National Health and Medical Research Council of Australia (ID 1178444).

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or other sponsors.

Ethics Approval

The study was approved by the Institutional Review Board at the University of Iowa.

Clinical Trial Registration

This trial was registered on www.clinicaltrials.gov (NCT04059146) and Open Science Framework (https://osf.io, JF2XU).

Disclosures and Presentations

All authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported the following: R.L. Chimenti receives software from Siemens Medical Solutions USA Inc. Professional, and scientific societies have reimbursed her for time and travel costs related to presentation of research on pain and pain education at scientific conferences. G.L. Moseley receives royalties for key resources used for pain science education (Explain Pain, Explain Pain Handbook: Protectometer, Explain Pain Supercharged; NOIgroup Publications, Adelaide, Australia) and speaker fees for talks on contemporary pain science education and consults to various organizations on pain science education and management. G.L. Moseley also has received support from Reality Health; ConnectHealth UK; Kaiser Permanente; AIA Australia; Workers’ Compensation boards; and professional sporting organizations in Australia, Europe, South America, and North America. Professional and scientific bodies have reimbursed him for travel costs related to presentation of research on pain and pain education at scientific conferences/symposia. He has received speaker fees for lectures on pain and rehabilitation. M.M. Hall receives consulting fees from Tenex Health and royalties from UpToDate Inc and holds stock in Sonex Health. C. de Cesar Netto is a paid consultant for Ossio, Stryker, and Medartis; is a paid consultant for and receives stock options to CurveBeam and Tayco Brace; a paid consultant or receives royalties from Paragon, Zimmer-Biomet, Nextremity, and Artelon; and is the editor-in-chief of Foot and Ankle Clinics. K.A. Sluka serves as a consultant for Pfizer Consumer Health and Novartis Consumer Healthcare/GSK Consumer Health care and receives royalties from IASP Press. The other authors declare no conflicts of interest for this study.

A portion of this study was presented at the 19th International Association for the Study of Pain; September 19–23, 2022; Toronto, Ontario, Canada.