Combining Wearable Technology and Telehealth Counseling for Rehabilitation After Lumbar Spine Surgery: Feasibility and Acceptability of a Physical Activity Intervention

Abstract

Objective

The purpose of this study was to examine the feasibility and acceptability of a wearable device and telehealth counseling physical activity intervention early after lumbar spine surgery.

Methods

Sixteen patients were randomized to an 8-session physical activity intervention or to usual postoperative care after surgery. The intervention included a wearable device (ie, Fitbit) and telehealth counseling by a licensed physical therapist. The feasibility of study procedures was assessed through recruitment, randomization, retention, and participation rates. Acceptability was assessed through a satisfaction survey and median within-participant change in objective physical activity (steps per day and time spent in moderate-to-vigorous physical activity [MVPA]) and patient-reported outcomes.

Results

Of 64 participants who were eligible, recruitment and randomization rates were 41 and 62%, respectively. Retention for objective physical activity and patient-reported outcomes was 94 and 100%, respectively, at 6-month follow-up. Seven (88%) participants in the intervention group completed all telehealth sessions, and 6 (75%) met step goals over the 8 sessions. All participants in the intervention group found the wearable device and telehealth counseling to be helpful and reported it much or somewhat more important than other postoperative services. Median within-participant change for steps per day improved from baseline (preoperative) to 6 months after surgery for both the intervention (1070) and usual care (679) groups, while MVPA only improved for the intervention group (2.2. minutes per day). Improvements in back and leg pain and disability were noted for both groups. No adverse events were reported in the study.

Conclusion

Combining wearable technology and telehealth counseling is a feasible approach to promote the physical activity during the early postoperative period after spine surgery. Future randomized controlled trials are needed to investigate the efficacy of leveraging wearables and telehealth during postoperative rehabilitation.

Impact

This study has implications for the clinical dissemination of physical activity strategies in the rehabilitation setting.

Introduction

Studies have shown that up to 80% of patients remain physically inactive after spine surgery.^1–4 Individuals who are physically inactive are at higher risk of disability, diabetes, and cardiovascular diseases^5–8 and consequently all-cause mortality.⁹ Physical activity, specifically walking, is an important determinant of patient-reported outcomes after spine surgery.^10–12 Master et al¹¹ found that patients who walked at least 3500 steps per day during the early postoperative period had 3.75 times higher odds of achieving “best outcome” at 1 year. Early assessment and management of physical inactivity have the potential to improve outcomes after spine surgery.

Wearable technology, such as smart watches, allows tracking of physical activity during the postoperative period. Gilmore et al¹³ demonstrated that activity monitors are valid for assessing steps immediately after spine surgery. Additionally, these tools can serve as an important intervention component in rehabilitation settings for self-monitoring and patient-centered goal setting.^14,15 A recent systematic review found that combining wearable technology and step goals may be effective in increasing postoperative physical activity in patients undergoing joint replacement surgery.¹⁶ To date, intervention studies focused on increasing physical activity through walking have not been conducted in patients following spine surgery. Qualitative work has found that although patients acknowledge physical activity is important, they are resistant to increasing activity due to pain, spine-related symptoms, and fear of reinjury, especially during the early postoperative period.^17,18 Feasibility studies are needed to determine whether specific intervention components, such as wearable trackers and behavioral counseling, are acceptable prior to larger-scale studies in this surgical patient population.

Based on our published work that increased steps per day at 6 weeks after surgery was associated with improved long-term outcomes,¹¹ we developed an early postoperative physical activity intervention that included wearable technology and telehealth physical activity counseling by a licensed physical therapist. Intervention components were informed by prior trials in surgical and nonsurgical patient populations with musculoskeletal pain^16,19,20 and a Brief Action Planning framework that has been found to be effective for physical activity behavior change.^20,21 Since qualitative work^17,18 has identified specific challenges with increasing physical activity, the primary aim of this study was to examine the feasibility and acceptability of our telehealth physical activity intervention in patients after lumbar spine surgery. Feasibility was assessed through recruitment, randomization, retention, and intervention participation rates,²² and acceptability was assessed through a posttreatment satisfaction survey and median within-participant change in objective physical activity and patient-reported outcomes. Benchmarks for feasibility based on prior spine surgery studies^23–25 were established a priori. These benchmarks included: ≥50% for recruitment of participants who were eligible, >85% randomization of participants who consented, >90% for retention, and ≥ 75% for adherence. We hypothesized that participants would report being satisfied with the wearable device and telehealth counseling physical activity intervention, and within-participant improvements in outcomes would be larger for the intervention group when compared descriptively to a usual care group.

Methods

Study Design and Participants

This was a single-center feasibility randomized controlled trial (NCT04591249) in 16 patients undergoing spine surgery for a lumbar degenerative condition. The initial sample size was 30 participants; however, this was reduced due to the impact of the COVID-19 pandemic. Patients were recruited prior to surgery, completed a preoperative (baseline) assessment, and were randomized at 2 weeks after surgery to the physical activity intervention (n = 8) or usual care (n = 8). Follow-up assessments for objective physical activity and patient-reported measures occurred for both groups posttreatment (3 months after surgery) and 6 months after surgery.

The study population included English-speaking adults aged 18 years or older who were undergoing surgical treatment using laminectomy with or without fusion for a lumbar degenerative condition. Patients having surgery for spinal deformity as the primary indication or surgery secondary to pseudarthrosis, trauma, infection, or tumor were excluded. Additional exclusion criteria included prior history of lumbar spine surgery, presence of back and/or lower extremity pain <3 months, history of neurological disorder resulting in moderate to severe movement dysfunction, and those who are unable to provide stable address and access to a telephone indicating the inability to participate in the intervention.

Participants were recruited through both in-person and remote study procedures. The preoperative clinic schedule was reviewed daily, and the surgery schedule was reviewed weekly. Patients were approached and consented either in-clinic or over the telephone with a standardized electronic consent process.

Procedures

Demographic and clinical characteristics were collected from a self-report questionnaire and the medical record prior to surgery. Outcome assessment occurred at baseline and at 3 and 6 months after surgery. Physical activity was objectively assessed using a research grade triaxial accelerometer (Actigraph GT3x). Patient-reported outcomes were assessed using validated measures. REDCap (Research Electronic Data Capture) was used for administering the web-based questionnaires and storing study data.^26,27 Outcome assessors who collected physical activity data were blinded to group assignment.

Randomization occurred electronically by research personnel 2 weeks after surgery during an in-person postoperative clinic visit. Patients were frequency-matched in a 1:1 ratio in blocks of assignments stratified by type of surgery (ie, fusion or no fusion). Block sizes of 2 and 4 were determined randomly with the patient as the unit of randomization.

Physical Activity Intervention (8 Weekly Sessions)

The overall structure of the physical activity intervention is summarized using the Template for Intervention Description and Replication (see Supplementary Appendix 1).^28,29 The physical activity intervention was based on a Brief Action Planning framework,²¹ which includes motivational interviewing and SMART goal setting. At the 2-week postoperative clinic visit, participants received a Fitbit Inspire HR and a folder that included an introduction to their therapist and the program, a daily step goal tracking sheet, and instructions on how to use Fitbit and download and use Zoom (a HIPPA compliant web-based interface) for the telehealth counseling sessions. Study personnel reviewed the Fitbit and Zoom instructions with the patient, setup the Fitbit, and downloaded Zoom software for those patients requesting help. Patients who were not able to come back to the clinic received the Fitbit and folder through the mail, and a video conference call was used to review material and the Fitbit and Zoom setup process. Session 1 with the study physical therapist occurred over Zoom within 1 week of patients receiving the Fitbit. During the session, the physical therapist confirmed if participants were comfortable using the Fitbit and helped participants through motivational intervention techniques to set a weekly walking goal. Patients were asked to write down their weekly walking goal and daily progress on the tracking form provided in the folder. Patients accessed their walking data from the Fitabase system. At each subsequent weekly session, the physical therapist asked participants whether they met their weekly step goal, reviewed participants’ physical activity using Fitabase, and objectively checked whether they met their weekly step goal. Physical activity progression each week occurred when participants met their walking goal for at least 4 out of 7 days. If a participant did not achieve their walking goal, the physical therapist discussed barriers and helped participants identify potential solutions. Walking goal progression was documented each week by the physical therapist in REDCap.

One physical therapist (S.W.V), with 15-years delivering remote behavioral interventions involving motivational interviewing, provided the intervention. The physical therapist completed 3 hours of in-person training on implementing study procedures, creating SMART goals, and using a Fitbit and the Fitabase system to monitor steps. To ensure accurate use of the Fitbit and real-time monitoring, a mock trial of Fitbit data collection with the physical therapist and study personnel was completed prior to the start of the feasibility trial.

Fidelity of the intervention was assessed through a checklist that was completed by the study physical therapist after each session. The checklist included the length of the telehealth session, whether or not participants met their step goal, and whether a new step goal and course of action was initiated. A study investigator (H.M.) monitored 100% of the Zoom recordings for the first 2 patients and then performed a random audit of 20% of the remaining sessions.

Usual Care

Participants randomized to usual care received postoperative care as determined by their treating surgeon. This included lifting restrictions, advice to stay active, and oral analgesics as needed. Postoperative physical therapist referral was at the discretion of the surgeon and was initiated, if needed, after the intervention phase of the trial.

Feasibility Outcomes

Feasibility was assessed through recruitment (proportion of eligible participants who consented to participate), randomization (proportion of participants who consented who were randomized to intervention or usual care group), retention (number of participants who completed the objective and patient-reported outcome measures at 6-month follow-up), and participation rates.

Adherence to the intervention was assessed through completion of weekly telehealth calls with the physical therapist and weekly step goal attainment (examined via Fitabase).

Acceptability Outcomes

Acceptability was measured through a posttreatment satisfaction survey that was completed by participants in the physical activity intervention group. The survey consisted of 5 questions related to helpfulness, likelihood to recommend, benefit to effort, impact on walking, and comparison to other services. Responses to the questions related to helpfulness and likelihood to recommend were recorded on 0 to 10 Likert scales, with 0 being not helpful/unlikely to 10 being very helpful/most likely. The other 3 questions had discrete answer choices. Participants also completed open-ended questions related to their experience with the intervention. Adverse events were assessed by study personnel at follow-up time points through review of the electronic medical record and weekly by the study physical therapist during the telehealth sessions. Participants were informed they could contact study personnel anytime during the course of the study to report adverse events.

A research-grade triaxial accelerometer (Actigraph GT3x) was used to objectively assess physical activity, which was defined as steps per day and time in moderate-to-vigorous physical activity (MVPA) per day. The Actigraph GT3X validly quantifies physical activity in free-living conditions.^30,31 The accelerometer was distributed via mail at all assessment time-points. Participants were instructed to wear the accelerometer on their right hip for a minimum of 10 hours each day over a 7-day period. Once returned, a publicly available accelerometer algorithm provided by the National Cancer Institute (NCI) was used to process accelerometer data (https://epi.grants.cancer.gov/nhanes_pam/).³² Nonwear time was defined as ≥90 minutes of consecutive activity of <100 counts per minute and detected using a validated algorithm from adults with knee osteoarthritis.³³ Valid wear day was defined as at least 10 hours of wear time; total number of steps and time spent in MVPA were summed over a valid wear day. The thresholds used by NCI³² on a minute-by-minute basis were used to categorize activity counts into time spent in ≥2020 counts per minute, which was used to classify MVPA. To reliably estimate free-living physical activity, at least 3 valid wear days were needed.³⁴

Patient-reported outcomes included validated measures for disability and pain. The 10-item Oswestry Disability Index (ODI) assessed the effects of disability on various aspects of daily life.³⁵ A total score ranges from 0 to 100 with higher scores representing higher disability. The 11-point Numeric Rating Scale (NRS) assessed back and leg pain intensity using a single-item. The NRS is scored from 0 to 10, with 0 representing no pain and 10 representing worst pain imaginable.^36,37

Data Analysis

Descriptive statistics including medians, 25th and 75th percentile, means, standard deviations, and frequencies were used to describe participant characteristics and to summarize measures of feasibility and acceptability. Due to the small sample size, median values were used to describe characteristics and measures within the manuscript text. Participant open-ended feedback about the physical activity intervention was assessed, and selected quotes about the intervention were extracted. Formal qualitative analyses were not performed.

Accelerometer and patient-reported outcomes were summarized by group and time point. Within-participant change in outcomes at 3 and 6 months was computed and summarized by group. Consistent with recommendations for pilot studies,^38,39 formal statistical testing for between-group efficacy was not performed. SAS 9.3 (SAS Institute Inc, Cary, NC) was used to process Actigraph data, and R (R Foundation for Statistical Computing, Vienna, Austria) was used for the descriptive analyses.

Role of the Funding Source

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

Results

Participant Characteristics

The median age for the sample was 65.0 years, and most participants were White (94%), married (88%), and had some college experience (75%) (Tab. 1). Characteristics were mostly similar between the 2 groups, except the physical activity intervention group, which, compared to usual care, was older (71 vs 61 years) and had fewer individuals who were currently working (38 vs 75%).

Table 1.

Baseline Demographic and Clinical Characteristics of Randomized Sample (n = 16)^a

Characteristic	All Sample (n = 16)	Physical Activity (n = 8)	Usual Care (n = 8)
Age, y  Median (25th–75th percentile)  Mean (SD)	65.0 (56.8–73.0) 64.2 (13.2)	71.5 (54.0–76.8) 65.4 (15.7)	61.0 (57.5–68.3) 63.0 (11.0)
Female, n (%)	8 (50)	4 (50)	4 (50)
White, n (%)	15 (94)	7 (88)	8 (100)
Some college or more, N (%)	12 (75)	6 (75)	6 (75)
Married, n (%)	14 (88)	6 (75)	8 (100)
BMI in kg/m2  Median (25th–75th percentile)  Mean (SD)	29.0 (26.7–30.4) 29.5 (5.7)	29.4 (27.3–30.3) 28.5 (5.1)	29.0 (26.7–32.6) 30.5 (6.5)
Working status, n (%)  Currently working  Retired  Elected not to work	9 (56) 6 (38) 1 (6)	3 (38) 4 (50) 1 (13)	6 (75) 2 (25) 0 (0)
One or more comorbidities, n (%)	13 (81)	6 (75)	7 (88)
Fusion surgery, n (%)	12 (75)	6 (75)	6 (75)

^a BMI = body mass index

Feasibility

Over a 5-month period, 451 patients undergoing lumbar surgery for a degenerative condition were screened for eligibility, and 64 were eligible (Figure). The most common reasons for ineligibility were having a prior lumbar spine surgery (n = 191) and canceling spine surgery (n = 45) (Supplementary Appendix 2). The recruitment rate was 41% (26 consented/64 eligible), and randomization rate was 62% (16 randomized/26 consented). Of the 10 participants who were not randomized, 4 participants cancelled surgery, and 6 participants had missing or invalid baseline Actigraph data prior to randomization. Fifteen (94%) participants had 3 or more valid days of Actigraph data, and all participants (100%) completed the patient-reported outcomes measures at 6-month follow-up for high retention rates.

Figure.

Consolidated Standards of Reporting Trials (CONSORT) Flowchart.

Seven participants randomized to the physical activity intervention completed all 8 calls with the physical therapist (88% adherence to intervention). The median duration of the first telehealth session was 19.6 minutes. The median duration of subsequent telehealth sessions ranged from 5.2 to 13.7 minutes. Intervention participants showed an overall increase in their steps per day goals over the course of the 8-session intervention. The median step goal was 4000 steps per day at week 1 and 7500 steps per day at week 8 (Tab. 2). Six participants (75% adherence to step goals) either partially (1–3 out of 7 days) or fully (≥4 out of 7 days) met the physical therapist’s recommended weekly steps goal for the 8 intervention sessions.

Table 2.

Steps Per Day Goals During Physical Activity Intervention

Participant	Week 1	Week 2	Week 3	Week 4	Week 5	Week 6	Week 7	Week 8
1	2000	2000	2500	3000	3500	4000	4500	5000
2	10,000	10,000	10,000	10,000	10,000	12,000	10,000	10,000
3	5500	5700	5850	6200	6500	6800	8500	8000
4	1225	1584	2200	2662	2662	2800	2800	2800
5	3000	3200	3520	4000	4000	4400	–	4000
6	5000	7000	8000	8500	9000	9000	9500	10,000
7	7000	8000	8000	8000	8500	9000	9000	9000
8	2000	6000	6500	6500	6500	6500	7000	7000
Median (25th–75th percentile)	4000 (2000–5875)	5850 (2900–7250)	6175 (3265–8000)	6350 (3750–8125)	6500 (3875 8625)	6650 (4300–9000)	8500 (5750–9250)	7500 (4750–9250)
Mean (SD)	4465.6 (3005.1)	5435.5 (2973.3)	5821.3 (2850.0)	6107.8 (2688.3)	6332.8 (2732.3)	6812.5 (3080.1)	7328.6 (2727.5)	6975.0 (2768.1)

A review of the physical therapist checklists found that 100% of the components were delivered, which was consistent with a review of the Zoom recordings indicating treatment fidelity.

Acceptability

All intervention participants completed the posttreatment satisfaction survey. Participants perceived the intervention as being useful for their recovery and would recommend the intervention to others (Tab. 3). Six (75%) participants reported that the benefits they received far or somewhat outweighed the effort put forth. Seven (88%) participants noted that their physical activity increased a meaningful amount. All participants said that the intervention was much more (75%) or somewhat more (25%) important than other postoperative services.

Table 3.

Physical Activity Intervention Acceptability

Acceptability Measure	Median (25th–75th percentile) or n (%)	Mean (SD)
Helpfulness of intervention to recovery: 0–10	9.5 (9.0–10.0)	9.1 (1.4)
Likelihood to recommend intervention: 0–10	10.0 (9.0–10.0)	9.5 (0.8)
Benefit to effort The benefits far outweighed the effort The benefits somewhat outweighed the effort The benefits equaled the effort I put into it The effort far outweighed the benefits	5 (63%) 1 (12%) 1 (12%) 1 (12%)	– – – –
Impact on walking  My walking increased a meaningful amount  There was some increase in my walking, but not enough to be meaningful  There was no change in my walking  There was some decrease in walking, but not enough to be meaningful  My walking decreased a meaningful amount	7 (88%) 1 (12%) 0 (0%) 0 (0%) 0 (0%)	– – – – –
Importance compared to other services  Much more important  Somewhat more important  As important  Somewhat less important  Much less important	6 (75%) 2 (25%) 0 (0%) 0 (0%) 0 (0%)	– – – – –

When participants were asked to describe the program and provide their experiences, participants highlighted the benefit of working with a physical therapist to set goals, usefulness of the Fitbit, and described an increase in their motivation to walk because of the intervention (Tab. 4). When asked for recommendations to improve the program, some participants noted issues with the Fitbit (“I would have liked to have known that the Fitbit doesn’t work right or it won’t count my steps in certain positions such as leaning my arm on a grocery cart or putting my hand in my pocket while I’m walking” and “The font on the FitBit is so small that it can be difficult to read”).

Table 4.

Participant Feedback About Physical Activity Intervention

Topic	Participant Feedback
Description of program	“The walking program was my motivator to get up and get going. I also would tell them that I overcame my fears of hurting my back with the walking program because I had someone to talk to and encourage me.” “I would describe the walking program as a way to get your life back to better than before surgery. For me, I hadn’t been able to walk that much for many years!” “I ended up putting the focus on my physical therapist and told anyone who asked that my PT required me to meet my step goals. I did not frame the program as if it were optional—I’d tell everyone my PT and Doctor require this of me, so let’s go walk!” “It helps you get stronger and gain confidence in yourself. It also motivates you to get moving even when you don’t feel like it!!” “The Fitbit keeps up with your steps and alerts you when you need to walk.” “The more effort you put into the step count you reap more benefits period!”
Recommendations for improving program	“I would have liked to have known that the Fitbit doesn’t work right or it won’t count my steps in certain positions such as leaning my arm on a grocery cart or putting my hand in my pocket while I’m walking. Also, I wish I would of had the walking program longer with the therapist she was a big encouragement to me.” “What would be cool is if in between weekly calls, you could get little encouragements or messages to help keep focused on your goals.” “Maybe pair the program with a shoe store to help patients get comfortable and affordable shoes. Maybe pair it with a podiatrist as well to help get quality inserts.” “The font on the FitBit is so small that it can be difficult to read.” “It seems to me the program was well thought out and I can’t think of anything that would make it work better, the main thing would stress how important it is to invest as much effort in order to get the best results.”
Additional thoughts on program	“I’m grateful I had the opportunity to be in the walking program and privileged to have met the people who helped me on my journey to healing.” “[Therapist] was amazing every week! She helped set realistic goals, understanding the challenges and benefits of walking. Being able to track daily progress and accountability was extremely beneficial.” “Weekly coaching sessions are very encouraging.” “I liked using the Fitbit more than I thought I would. It was very useful and easy to use. The recovery coach was helpful in answering questions that arose.” “Being compliant is key to recovery. Having the weekly visits with the PT to monitor what you [are] doing [and] how you’re doing it, asking suggestions and helping set goals is encouraging.” “There were a few days it took a little extra effort to get motivated but the thought always came to mind I had excellent support from my coach….a solid help.”

Steps per day, time spent in MVPA, and patient-reported outcome measures for each group and time point are summarized in Table 5. The median within-participant change for steps per day and MVPA at 3 and 6 months in the physical activity group was 1010 and 1070 steps per day and 5.9 and 2.2 minutes per day in MVPA, respectively. In contrast, the median within-participant change in the usual care group was −58 and 679 steps per day and 3.4 and 0.1 minutes/day in MVPA, respectively. For the patient-reported-measures, improvements in the ODI and NRS were observed for both groups. At 3 and 6 months, the median within-participant difference in ODI score for the physical activity group was −26 and −24, in NRS back pain score was −3 and −2.5, and in NRS leg pain score was −4 and −3. The median within-participant difference for the usual care group at these same time points was −22 and −23 for the ODI, −3 and −3.5 for NRS back pain, and − 4.5 and −5 for NRS leg pain. No adverse events were reported in the study.

Table 5.

Acceptability Outcome Assessment^a

	Physical Activity			Usual Care
Outcome	Preop	3 mo	6 mo	Preop	3 mo	6 mo
Actigraph
Steps per day  Median (25th–75th percentile)  Mean (SD)	2883 (1533–5032) 3189.6 (2062.0)	4079 (1724–6206) 4133.0 (2602.9)	4944 (1871–6150) 4457.0 (2697.5)	4599 (3079–6911) 5100.6 (2676.7)	4600 (2915–6336) 4884.0 (2405.4)	6113 (3695–7762) 5708.0 (2183.7)
Time in MVPA per day  Median (25th–75th percentile)  Mean (SD)	2.3 (1.8–8.1) 5.6 (6.4)	8.1 (0.9–11.3) 8.1 (8.7)	12.1 (2.9–12.1) 11.8 (8.9)	9.1 (2.5–10.9) 8.1 (6.7)	7.9 (3.4–21.3) 15.2 (18.0)	8.7 (4.1–15.8) 17.3 (24.2)
Patient-reported outcomes
Disability: ODI  Median (25th–75th percentile)  Mean (SD)	32.0 (20.0–40.0) 31.5 (16.2)	13.3 (8.3–16.0) 13.4 (10.0)	7.0 (1.5–13.5) 10.8 (13.4)	25.0 (23.0–36.5) 32.3 (16.3)	3.0 (1.5–10.0) 5.0 (4.9)	3.0 (1.5–4.0) 5.8 (9.9)
Back pain intensity: NRS  Median (25th–75th percentile)  Mean (SD)	5.0 (4.0–7.3) 5.6 (1.8)	1.0 (0.5–3.0) 1.9 (1.9)	2.0 (0.8–3.3) 2.4 (2.3)	5.0 (3.5–5.8) 5.0 (2.5)	1.0 (0.8–2.3) 1.5 (1.4)	2.0 (0.8–2.0) 1.6 (1.3)
Leg pain intensity: NRS  Median (25th–75th percentile)  Mean (SD)	5.0 (3.5–8.0) 5.0 (3.0)	1.0 (0.0–2.0) 1.3 (1.6)	1.5 (0.0–2.5) 1.9 (2.2)	6.5 (5.8–7.3) 6.5 (1.6)	1.5 (0.8–2.3) 2.0 (2.3)	1.0 (0.8–1.3) 1.4 (1.6)

^a MVPA = moderate-to-vigorous physical activity; NRS = Numerical Rating Scale; ODI = Oswestry Disability Index.

Discussion

We examined the feasibility and acceptability of a wearable and telehealth counseling physical activity intervention for patients after lumbar spine surgery. Our results showed that the study procedures and combining wearable technology and telehealth physical therapist support are feasible to implement, and the physical activity intervention is acceptable to patients. Retention and participation rates were above a priori feasibility metrics; however, recruitment and randomization were below the a priori feasibility metrics. High retention rates for the study, as well as high participation and satisfaction with the intervention, support future clinical trials of a telehealth physical activity intervention. In addition, within-participant improvements noted in objective physical activity and patient-reported disability and pain suggest that the intervention may potentially benefit patients after spine surgery. However, further work is needed to determine whether the telehealth physical activity intervention can provide improvement over usual care in this surgical population.

The current study highlighted several feasibility issues relating to Actigraph data collection that can inform future rehabilitation research in surgical population. Six participants (23%) were not randomized due to missing or invalid Actigraph data. A short time interval between consent and surgery and issues with timely receipt of the Actigraphs through the mail led to retention issues. Recommendations for future work include a wide window for preoperative Actigraph data collection (~1 month) prior to surgery and remote procedures for confirmation of valid wear time. Cloud-based data capture technology, such as CentrePoint (https://actigraphcorp.com/centrepoint/), has the potential for remote verification of Actigraph valid wear days when transit delays occur. Other forms of remote data monitoring through rings or wrist-worn trackers may also be acceptable; however, validation for these other types of wearable technology is needed in the spine surgery population. Finally, the use of FedEx or priority mail to reduce mailing delays appears to be needed for both preoperative and early postoperative time points.

The within-participant assessment of steps per day showed a larger median increase in steps per day in the intervention group at 3 and 6 months. However, the small sample size and lack of efficacy testing in this study limit conclusions regarding steps per day and MVPA results. Additionally, there were large preoperative differences noted in both steps and MVPA per day, despite randomization. Although a larger sample size may address this baseline imbalance, future trials may want to consider randomization procedures that stratify by physical activity level. It is important to note that both groups engaged in more than 3500 steps per day after surgery, a threshold that is known to be associated with achieving best outcome and lower opioid use 12 months after spine surgery.¹¹ Since participants in the usual care group exceeded this threshold by a considerable amount prior to surgery, it is not surprising that a high level of activity continued after surgery. Future trials and clinical dissemination of physical activity interventions may want to include screening criteria for physical activity to target those at highest risk for poor outcomes after surgery. Previous studies have shown that conducting physical activity screening using a single-item physical activity measure⁴⁰ or national physical activity guidelines⁴¹ may be a strategy to target patients who are physically inactive.

Two physical activity outcomes were chosen for this feasibility trial. At 6 months after spine surgery, the descriptive comparison between the 2 groups was not consistent across these outcomes. Median within-participant change for steps per day appeared to improve for both groups; however, the change for MVPA only improved for the intervention group. The intervention group was engaged in at least 10 minutes of MVPA per day, a threshold which is known to be associated with lower risk of disability in patients with lower extremity joint symptoms.⁵ Although the initial focus of the intervention content was on increasing steps per day, counseling on increasing the intensity of daily walking occurred for those patients already at maximum levels (ie, 8000–10,000 steps per day). It may be that categorizing physical activity by intensity level is a more sensitive outcome metric for a patient population with wide variability in baseline physical activity. Further work may be needed to determine the most appropriate primary outcome definition for surgical patient populations, especially if future trials include a higher-risk group that requires more focus on increasing steps than intensity.

There were other notable observations in this study, which have the potential to direct future research. First, the eligibility rate for this pilot work was low (14%), and the recruitment rate of 41% was below our feasibility benchmark. Prior published work in this patient population by our group has reported higher rates.^23–25 Results may be due to the timing of recruitment, which occurred during the COVID-19 pandemic; surgery cancellation rates increased across the medical center, while elective surgeries decreased. In addition, the reliance on remote study procedures during this time and an increase in patient stress and anxiety related to the pandemic may have contributed to a lower recruitment rate. Second, this pilot study measured outcomes (physical activity and patient-reported outcome measures) at both 3 and 6 months postoperatively. For some individuals, the 3-month assessment overlapped with the intervention phase due to delays in starting the intervention. Future work may want to consider a postintervention assessment or only using 6 months as a short-term follow-up timepoint. Third, although all participants progressed through the telehealth physical activity intervention and typically met step count goals, it is unclear whether this was reflective of an outcome change in physical activity as measured by the Actigraph. The median steps per day goal at the 8th session of the intervention were 7500 steps per day. However, the physical activity group showed median step counts of 4079 and 4944 steps per day at the 3- and 6-month postoperative time points, respectively. Since the 3-month assessment either overlapped or was close to the end of the intervention, future research may be needed to better understand this discrepancy between Fitbit and Actigraph data. One potential reason for the discrepancy may be attributed to the placement of the wearable technology⁴² (ie, Fitbit was worn on the wrist vs Actigraph was worn on the hip). With regard to the 6-month data, this may indicate a need for ongoing encouragement to monitor and meet step count goals following the conclusion of the 8-session intervention phase. One potential solution would be “booster sessions” or text messages (as suggested by a participant during their exit interview—“What would be cool is if in between weekly calls, you could get little encouragements or messages to help keep focused on your goals”). So, although the results of this study indicate that the wearable and telehealth counseling physical activity intervention is feasible to implement and acceptable to participants, further consideration of how to measure physical activity and how to ensure lasting effects of the intervention is necessary in future research.

Limitations

The limitations of this study include the small sample size. We had planned for 30 participants based on our primary aim of feasibility and acceptability and anticipated recruitment rate; however, the COVID-19 pandemic limited enrollment to 16. Prior work has recommended larger sample sizes for pilot work,^38,43 and a notable impact of the small sample size was the baseline imbalance of demographics and outcomes between the 2 groups. For example, most participants in the usual care group were younger, currently employed, and more physically active prior to surgery. Also, participants enrolled in the study displayed lower ODI scores than expected for patients scheduled for spine surgery, which impacts the generalizability of our findings.

Conclusion

A physical activity intervention incorporating wearable technology and telehealth counseling from a physical therapist appears to be feasible and acceptable for patients recovering from lumbar spine surgery. Recommendations include additional randomized controlled trials to assess the efficacy of leveraging wearables and telehealth through early rehabilitation programs for improving postoperative spine outcomes. This line of work has implications for the clinical dissemination of physical activity strategies in the rehabilitation setting.

Author Contributions

Hiral Master (Conceptualization [equal], Funding acquisition [lead], Methodology [equal], Project administration [equal], Supervision [equal], Writing—original draft [lead], Writing—review & editing [equal]), Rogelio A. Coronado (Conceptualization [supporting], Investigation [supporting], Methodology [supporting], Supervision [supporting], Writing—original draft [supporting], Writing—review & editing[ supporting]), Sarah Whitaker (Data curation [equal], Methodology [supporting], Writing—review & editing [equal]), Shannon Block (Data curation [supporting], Methodology [supporting], Project administration [equal], Supervision [supporting], Writing—review & editing [equal]), Susan W. Vanston (Conceptualization [supporting], Data curation [supporting], Methodology [supporting], Writing—review & editing [equal]), Jacquelyn S. Pennings (Conceptualization [supporting], Formal analysis [lead], Methodology [equal], Software [equal], Validation [equal], Writing—review & editing [equal]), Rishabh Gupta (Data curation [supporting], Investigation [supporting], Resources [supporting], Writing—review & editing [equal]), Payton Robinette (Data curation [supporting], Project administration [supporting], Resources [supporting], Supervision [supporting], Writing—review & editing–[supporting]), Byron Stephens (Conceptualization [supporting], Data curation [supporting], Investigation [supporting], Supervision [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), Amir Abtahi (Data curation [supporting], Investigation [supporting], Supervision [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), Jacob Schwarz (Data curation [supporting], Investigation [supporting], Supervision [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), and Kristin R. Archer (Conceptualization [equal], Funding acquisition [supporting], Investigation [equal], Methodology [equal], Project administration [equal], Resources [equal], Supervision [equal], Visualization [equal], Writing—original draft [equal], Writing—review & editing [lead])

Funding

This study was funded by research grants from the Center for Musculoskeletal Research at Vanderbilt University Medical Center, the Academy of Orthopaedic Physical Therapy, and by a CTSA award (UL1TR000445) from the National Center for Advancing Translational Sciences. R. Coronado was also supported by a Vanderbilt Clinical and Translational Research Scholars award (KL2TR002245) during manuscript development.

Ethics Approval

Ethical approval was obtained by the Institutional Review Board at the participating site.

Clinical Trial Registration

This trial was registered with ClinicalTrials.gov (NCT04591249).

Data Availability

The dataset used and analyzed for the current study is available by reasonable request and at the discretion of the corresponding author.

Disclosures

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

K. Archer reports personal fees from NeuroSpinal Innovation, Inc and Spine outside the submitted work. J. Pennings reports personal fees from 3Spine and Steamboat Orthopaedic and Spine Institute.