Included participants were physical therapists who were affiliated with Chronisch ZorgNet and had a specialization in COPD treatment. Chronisch ZorgNet is a nationwide network that provides training for treating patients with chronic diseases according to the latest evidence (guidelines). The network consists of 3500 physical therapists working in primary, secondary, and tertiary care, of which 1200 physical therapists are specialized in lung physical therapy [19]. This network provided access to this target population. The survey was distributed by emails from Chronisch ZorgNet. All specialized lung physical therapists who were Chronisch ZorgNet members were eligible to participate in this survey. Physical therapists were excluded from this study if they treated less than 1 unique patient with COPD per week.

The survey (Multimedia Appendix 1) was drafted from documents based on international scientific literature: (1) the checklist “the use of measurement instruments in clinical practice” [20] based on the CFIR [17] (Q1), (2) the questionnaire “meaningful use of patient-specific goal setting within the clinical reasoning of physical therapists” [21] based on the TPB [18] (Q2), (3) the framework “experiences of patients with commercially available ATs” [22] (Q3), and (4) the KNGF guidelines for COPD [5]. Questions of the checklist Q1 and the questionnaire Q2 were adapted for the application of ATs instead of general measurement instruments. To gain insight into relevant barriers and facilitators, the topics of the framework Q3 were used. The survey was supplemented with questions based on the recommendations for the use of ATs in the COPD guidelines [5]. In total, 9 questions determined the eligibility of the participants (eg, sex, age, and work experience). Based on the question “I use ATs in people with COPD” (yes or no), participants answered either 59 questions if they used ATs or 33 questions if they did not.

Of the 59 questions for the AT users, 26 were based on the checklist “the use of measurement instruments in clinical practice”’ Q1 [20], 5 were based on questionnaire Q2 [21], 20 were based on the framework Q3 [22], and 8 were based on the COPD guidelines [5]. Of the 33 questions for the nonusers, 19 were based on Q1 [20], 5 were based on Q2 [21], and 9 were based on Q3 [22]. The questions were either answered on a 7-point Likert scale (0=totally disagree and 7=totally agree), on a multiple-answer scale, or were open comment questions. Only 4 questions were necessarily answered by yes or no (eg, ATs are provided to our patients: yes or no). Participants were able to review and change their answers before completing the survey.

To ensure the content validity of the combined questionnaires, the draft survey was sent to 5 experts for feedback on the usability and face validity of the survey: 2 experts in the field of measurement tools in health care, 1 expert in the field of ATs in health care, and 2 specialized physical therapists. The feedback required no changes to the survey and the final survey was constructed (Multimedia Appendix 1).

Data collection was performed in May 2021. The survey was completed via clickable link, and QR code in the email leading to the survey in QuestBack Essentials (2021). For each question, a digital page emerged; in total, there were between 33 and 59 pages excluding the start- and the end page and 7 questions regarding population characteristics. All questions were mandatory questions and needed to be completed before the survey automatically continued. One reminder was sent after 1 week. Participants were asked to answer the questions as accurate as possible, and it was explicitly addressed that the results were analyzed anonymously. Data were stored in a secured data repository of Zuyd University of Applied Sciences for 15 years, and only the participating researchers had access to these data.

Data analyses were performed using SPSS Statistics (version 23.0; IBM Crop). Only fully completed surveys were analyzed. Descriptive statistics of the participants’ characteristics were presented as a number (percentage) for the categorical variables sex, workplace setting, and whether they completed a recent COPD course. For the continuous variables, age, work experiences, and the number of treated unique patients with COPD per week were presented as a mean (SD). Answers to questions with a 7-point Likert scale were presented as mean (SD). Responses to open comment questions were treated as qualitative data and categorized into themes by using directed content analyses via inductive coding [23] and presented as a number (percentage) per theme. Open answers that did not fit any code were labeled as other. Responses were separately analyzed for users and nonusers of ATs. In case of any missing data, listwise deletion was applied.

No ethics approval was required since the research was not subject to the Medical Research Involving Human Subjects Act (Wet medisch-wetenschappelijk onderzoek met mensen). The research was part of continuous quality monitoring and improvement of daily COPD care within a national quality network (Chronisch ZorgNet). This study did not require written informed consent. Participating physical therapists were informed about the purpose of this survey prior to commencing the survey. Physical therapists voluntarily participated in this study. Study results were processed anonymously. Physical therapists were always able to end the survey without any consequences. There was no compensation for the physical therapists. The CHERRIES (Checklist for Reporting Results of Internet E-Surveys) was used (Checklist 1).

In total, 216 physical therapists completed the survey. Five physical therapists were excluded because they did not meet the inclusion criteria (treating less than 1 unique patient with COPD per week), leaving 211 responses for analysis. There were no missing data. The characteristics of the physical therapists are displayed in Table 1.

Results of the survey are presented per research questions.

Of the 211 participating physical therapists, 108 (51.2%) used an AT, whereas most of them (n=82, 75.9%) already used an AT before it was advised in the guidelines. Most physical therapists measured the number of steps (n=64, 55.2%) and used a standard app on a smartphone (n=97, 89.8%). Other parameters named were, for example, activities of daily living, distance, physical capacity, and heart rate. The 2 most common goals for the use of an AT were to inventory the physical activity level (n=94, 87%) and to stimulate the physical activity level of the patient (n=88, 81.5%). All results on how ATs were used in physical therapy in patients with COPD are presented in Table 2.

Physical therapists who used ATs because they believed that it is an added value to health care (n=46, 42.5%) and because the guidelines recommend to use ATs (n=44, 40.7%). Physical therapists considered the cutoff value of 5000 steps per day for patients with COPD to be a realistic measure of sufficient activity (mean 4.7, SD 1.5). The primary reason physical therapists did not use ATs both among users (n=58, 53.7%) and nonusers (n=35, 33.9%) was that patients themselves indicated that they did not want to use them. In the nonusers group, lack of knowledge (n=31, 30.9%) was the second important reason for nonuse. Other reasons mentioned were, for example, bad experiences, difficulty implementing ATs, and concerns about the clinometric properties of the ATs.

In most centers, there was no consensus on why they were going to use ATs (mean 3.4, SD 1.8). Users scored high on questions regarding whether ATs contributed to the assessment of physical activity levels (mean 5.8, SD 1.1), allocation of patients into profiles (mean 5.8, SD 1.1), evaluation of physical activity levels (mean 5.9, SD 0.9), and stimulation of physical activity levels (mean 5.9, SD 1.0). Results on why ATs were used in physical therapy in patients with COPD are presented in Table 3.

No remarkable differences exist between users and nonusers regarding the experienced barriers and facilitators. For both groups, a relevant barrier to using ATs was the costs (users: n=75, 69.4% vs nonusers: n=77, 74.7%) and the cognitive or communication skills of the patient (users: n=75, 69.4% vs nonusers: n=50, 48.5%). Other reasons that were mentioned were prioritizing the patient over scientific pursuits, privacy concerns, and AT fails to measure parameters that therapists find meaningful. Important facilitators in both groups for using ATs were the motivation of the patient (users: n=90, 83.3% vs nonusers: n=70, 67.9%) and the user-friendliness of the AT (users: n=64, 59.2% vs nonusers: n=57, 55.3%). Both users (mean 3.6, SD 1.8) and nonusers (mean 2.5, SD 1.6) indicated that they lack the knowledge to provide a recommendation of commercially available ATs to their patients. Likewise, both groups indicated insufficient education or coaching possibilities for the health care professionals on ATs within their center (users: mean 3.7, SD 1.9 and nonusers: mean 2.7, SD 1.8). Further barriers and facilitators are presented in Table 4.

This research aimed to assess how and why ATs are currently used in physical therapy in patients with COPD and to determine which barriers and facilitators are of relevance for optimal implementation of ATs during the clinical reasoning process of Dutch physical therapists.

It is known that the implementation of an innovation is often challenging [11]. However, of the 211 participating physical therapists, 108 (51.2%) used an AT, and most of them (n=82, 75.9%) already used an AT before it was advised in the guidelines. This relatively high percentage could be explained by the fact that physical therapists indicated that the most important reason for using ATs is the experienced added health care value. Another explanation could be that the included physical therapists were affiliated with Chronisch ZorgNet and might therefore be more sensitized to provide evidence-based care. Furthermore, it was notable that regardless whether a physical therapist was a user or nonuser, the barriers and facilitators for using ATs were roughly the same. Both users and nonusers indicated that the most important reason for not using ATs is because their patients do not want to use an AT. However, it is debatable whether this is an assumption of the physical therapist or the actual opinion of the patient, as a recent review describes that patients with COPD in general enjoyed using ATs and found using the ATs easy [24]. The second (nonusers) and third (users) reason for physical therapists to not use ATs was a general lack of knowledge. Physical therapists indicated that they did not know which brand of commercially available AT they could recommend to their patients, which is related to the experienced lack of knowledge regarding price, functions, and quality. In contradiction, knowledge regarding how to measure with ATs, how to interpret the data of the AT, and how to discuss the data of the AT with a patient was scored high by the physical therapists. A possible explanation could be that therapists lack different knowledge: the knowledge on how to measure with ATs, how to interpret the data of the AT, and how to discuss the data of the AT.

The general adaption rate of eHealth in physical therapy seems to be similar to the adaption rate of 52% found in this study [25,26]. However, the use of the ATs in this study was self-reported; therefore, the presence of biases in the reported use and perception of ATs might be present. Most physical therapists used an app on a smartphone as an AT, which links to the fact that availability and costs were important factors when selecting an AT. Most smartphones automatically measure step count, and apps are mostly free of use. However, the validity of apps regarding measuring physical activity is low, especially in older adults during low walking speed, as is the case in activities of daily living [27]. Most physical therapists considered the cutoff value of 5000 steps per day to be sufficiently active, used in the new national guidelines, to be realistic. This is valuable information, considering that this cutoff value in the revised guidelines is based on previous studies. Several studies mention a minimum of 7500‐10,000 steps per day in order to have a healthy active lifestyle [28-30]. However, it seems that a healthy person reaches only 5500‐6000 steps per day on average, and people with a chronic disease 3500‐5500 steps per day [29-31]. Consequently, in the national guidelines, a cutoff value of 5000 steps per day is used for people with COPD to be sufficiently active [32,33]. This study suggests that 5000 steps per day might be a minimal clinical important threshold in daily physical therapy practice.

Both users and nonusers indicated that the implementation of ATs was not prepared and planned for within their center. To successfully implement eHealth, internal and external facilitation, audit and feedback, management support, and training of clinicians are essential according to a recent systematic review [34] and are in line with the CFIR [17]. Wilde et al [35] similarly showed the physical therapists’ need for guidance and information to support the integration of ATs within the practice. A solution could be to provide education and training to management and physical therapists. Management should be trained to guide an implementation trajectory; however, the use of implementation support practitioners could also be an option [36]. Implementation support practitioners are professionals who support organizations, leaders, and staff in their implementation of evidence-informed practices and policies [37]. Current physical therapists and physical therapy students should be educated and trained. To optimize the adoption of ATs in clinical settings, educational interventions should be developed based on the best available evidence, clinical expertise, and patient values and circumstances [38].

However, it is important to train physical therapists’ and students’ general eHealth competencies in order to use eHealth in a broader spectrum than ATs only. It is suggested that education consists of 5 competency themes: information communication technology attitudes and skills, interpretation and analysis of eHealth data, support and guidance, communication skills, and privacy and confidentiality [39]. Current health care professionals are insufficiently trained in all of these competencies and therefore struggle to perform these skills in their daily clinical practice [40]. A better understanding and more digital competency could also lead to better experiences and better outcomes. On the other hand, a previous study showed that patients with chronic diseases (including COPD) experience ATs as technically complicated and wish to receive support by their physical therapist during the use within a diagnostic and/or therapeutic process [22]. Currently, several (digital) tools and websites are available to support health care professionals in the goal-oriented selection of eHealth [41,42].

Several limitations of this study need to be mentioned. First, this study addressed technology acceptance on a limited basis. The level of technology acceptance might influence whether ATs are used or not. Studies showed that if health care professionals are already experienced eHealth users, they report fewer implementation barriers and experience more advantages (eg, a more positive attitude toward eHealth) [43]. Second, participating physical therapists were affiliated with Chronisch ZorgNet, an organization in which physical therapists treating patients with chronic diseases are trained to work evidence-based. This might have biased the results since affiliated physical therapists might be more prone to adhere to recent guidelines and be motivated to implement innovations. On the other hand, all participating physical therapists were members of Chronisch ZorgNet, which includes almost all physical therapists who are specialized in lung physical therapy. Therefore, the study sample constitutes a representative sample for primary care. A strength of this study is that the survey was based on relevant international scientific literature and models in relation to implementation and behavioral change (CFIR and TPB), which enables generalization beyond the Dutch context [5,13-18,22,44].

Overall, these results show that ATs are not yet fully implemented in the Dutch general physical therapy practice in patients with COPD, as recommended by current evidence-based guidelines. Both users and nonusers indicated that the most important reason for not using ATs is because their patients do not want to or at least the health care professionals experienced it that way. However, it is debatable whether this is an assumption of the physical therapist or the actual opinion of the patient. Physical therapists need guidance for the successful implementation of ATs. This could be accomplished by providing training for current physical therapists, integrating ATs into the education of (future) physical therapists, and providing support during the implementation process of ATs for both the physical therapists and management.