Abstract
This study assessed primary care providers’ (PCPs) perceptions of the usability of an electronic medical record tool to support physical activity counseling in primary care. Our analyses revealed that usability improved when the electronic medical record tool followed a PCP’s natural workflow and when the tool could assist in engaging in a discussion about physical activity. Poor usability was associated with the presence of large amounts of text on the screen and technological aspects that required additional learning. Overall, efficiency, workflow integration, and the inclusion of a care plan were vital in a physical activity counseling tool for a primary care setting.
Keywords
Currently, national physical activity guidelines in the United States and Canada recommend that adults (18–64 years) and older adults (65+ years) accumulate at least 150 minutes of moderate- to vigorous-intensity aerobic physical activity per week, plus strength training at least two times per week (American guidelines: https://health.gov/sites/default/files/2019-09/Physical_Activity_Guidelines_2nd_edition.pdf; Canadian guidelines: https://csepguidelines.ca/; Canadian Society for Exercise Physiology, 2020; Tremblay et al., 2011; U.S. Department of Health and Human Services, 2018). However, less than 30% of Canadian and American adults are meeting the minimum guidelines for physical activity (Colley et al., 2011; Gray et al., 2014; U.S. Department of Health and Human Services, 2018). High rates of physical inactivity are costly. As of 2018, physical inactivity resulted in approximately $117 billion in annual health care costs in the United States and $10.8 billion in Canada (Krueger et al., 2015; U.S. Department of Health and Human Services, 2018). With recent studies revealing no significant change in physical activity levels in Canada and the rest of the world, it is important that new strategies be tested and implemented to improve physical activity levels (Colley et al., 2017; Guthold et al., 2017). One possible avenue is through promotion of physical activity in primary care.
Dr. Robert Sallis (2009) has proposed and successfully implemented a “physical activity as a vital sign’ approach. It consists of two questions that a medical assistant will ask a patient on their arrival (Sallis, 2009). The first is, “On average, how often do you engage in at least moderate exercise?” (Sallis, 2009). The second question is, “On average, how many minutes per session?” (Sallis, 2009). Although Dr. Sallis’s approach has been successful at capturing the physical activity levels of 85% of individuals at Kaiser Permanente in Southern California, prompting a discussion on physical activity or changing physical activity behavior may require more probing than two simple questions (Sallis et al., 2016). Considering primary care providers’ (PCPs) lack of physical activity knowledge and their perceived ineffectiveness of physical activity counseling, PCPs need a tool that will assist them (Walsh et al., 1999).
Our research team developed a physical activity counseling tool for integration into electronic medical record (EMR). The tool builds on the physical activity vital sign approach by providing referral resources and by prompting patients to set physical activity–related goals (Clark et al., 2020). Throughout the design process, the researchers frequently collaborated with patients and PCPs to build a tool that would be thorough, yet easily adoptable by clinicians and well received by patients (Clark et al., 2020). Overall, PCPs desired that the physical activity counseling tool not be a simple screening tool, but rather one that would provide resources and advice that they can share with their patients when physical activity levels are not being met (Clark et al., 2020). The objective of this study was to gain insight into the usability of the EMR-integrated physical activity counseling tool to support the continued development of the tool and to further understand PCPs expectations of a physical activity counseling tool for primary care.
Method
Description of the Physical Activity Counseling Tool
The physical activity counseling tool was developed by a research team at the University of Waterloo, in collaboration with the eHealth Centre of Excellence in Waterloo, Ontario. The tool was created within the Telus PS Suite EMR platform and is currently compatible with the PS Suite EMR software. With further development, the physical activity counseling tool will be adapted to other EMR platforms.
The tool is composed of two sections; the Physical Activity Screening (PAS) questionnaire (Figure 1) and the care plan (Figure 2). The PAS questionnaire prompts the PCP to ask questions about weekly physical activity habits, which are benchmarked against physical activity guidelines (Figure 1). The care plan provides prompts and fillable and printable sections to support goal setting and the creation of a physical activity prescription (Figure 2). A typical interaction between a PCP using the physical activity counseling tool with a patient is provided in the sidebar.
An Example PCP-Patient Interaction
In a typical interaction between a PCP and a patient, the PCP begins by asking the first question located on the PAS questionnaire, “Are there any physical activities you enjoy doing?.” When a patient replies, “Yes,” the PAS questionnaire will prompt the PCP to determine the physical activity habits of their patient by using the list of activities shown and by inputting the number of days per week and the average time each activity is completed. Once weekly physical activity habits are documented, the PCP can click the “Calculate” button on the physical activity summary and their patient’s physical activity habits will be compared to the Canadian physical activity guideline recommendations. If Canadian physical activity guidelines are not being met, the PCP is prompted to ask, “How do you feel about setting a physical activity goal?.” If the patient agrees to set a physical activity goal, the link to the care plan will appear and the PCP can proceed to the care plan (Figure 2). By continuing to the care plan, the PCP is able to provide realistic next steps, disease-specific goals, physical activity resources, as well as other tips and advice that will encourage their patient to become more physically active. Since the care plan is designed to be modified and customized within an EMR platform, the PCP is provided the opportunity to print the care plan and provide the patient with a paper copy, detailing their next steps and their updated physical activity goals.

Screenshot of the Physical Activity Screening questionnaire.

Screenshot of the care plan.
Participants
We recruited primary care providers that had not previously seen or used the physical activity counseling tool. Individuals were eligible to participate if they (a) worked in a primary care environment (e.g., physician, nurse, allied health professional) in Ontario, (b) had experience using an EMR system, and (c) were able to communicate fluently in English. Purposeful sampling was used to represent diversity in health professions. Our target sample was five participants; a qualitative usability study requires three to five participants to retrieve an adequate amount of data, and five participants that are representative of the target population has been shown to capture a mean of 85% of usability issues (Faulkner, 2003; Kushniruk et al., 1997; Sonsteby & DeJonghe, 2012). The study was approved by the Office of Research Ethics at the University of Waterloo (ORE No. 40682), and participants provided written informed consent.
Testing Environment and Equipment
For this study, a minimalist portable lab approach was used (Rubin et al., 2008). The evaluations were completed in the primary care office of a participant or in the office of a research team member. The participants used a laptop that was compatible with the Telus Practice Solutions EMR software and contained a screen recording program (Open Broadcaster Software). Screen recordings were captured and analyzed for navigation difficulties that participants encountered while interacting with the tool. A handheld voice recorder was used to capture comments made by participants throughout the usability testing session. A Logitech webcam was used as a back-up screen recording and audio-recording device. Qualitative data analysis was completed using QSR International’s NVivo Version 12 software (QSR International, Melbourne).
Usability Testing Session Methodology
Each participant engaged in all three phases of the study design: think-aloud testing, a near-live scenario with a mock patient, and a semistructured interview. To begin the usability testing session, the moderator (member of the research team) of the study explained the usability testing methodology to the participant and then provided a brief description of the physical activity counseling tool. While thinking aloud, the participant was then asked to interact and learn the tool to the point that they would feel comfortable using the tool with a mock patient. Once the participant had familiarized themselves with the tool, they notified the moderator who then initiated a discussion regarding any questions or difficulties that the participant encountered while engaging with the tool. After the participant’s questions were addressed, the moderator provided the participant with a brief description of one of two mock patients. The mock patient was a member of the research team and was provided with a semistructured script to enact the clinical scenario. Both mock patient scenarios were scripted to not be meeting physical activity guidelines.
After the PCP completed the physical activity counseling tool with the mock patient, the moderator completed a brief semistructured interview with the PCP. A variation of flexible scripting was used during the interviews (Rubin et al., 2008). PCPs were then asked to complete a questionnaire that assessed demographic characteristics, which included age, sex, city of practice, profession, and years of practice. PCPs were also asked two questions related to their familiarity with EMR and their technological proficiency. Both responses were recorded by the participant on a 0- to 10 visual analogue scale.
Semistructured Interview Questions
Tell me what you liked about the physical activity counseling tool.
Tell me what you would change.
What aspects of the physical activity counseling tool were unclear?
How would you use the physical activity counseling tool in your practice?
What would prevent you from using it in your practice?
Is there anything you would like to ask me?
Data Analysis
The participant characteristics were described using the median and the range. The physical activity counseling tool was quantitatively analyzed according to its two components: the PAS questionnaire and the care plan. All audio-recordings were transcribed verbatim, coded and a content analysis were completed. One member of the research team systematically coded and analyzed the content of the transcripts. All screen recordings were analyzed for navigation difficulties. Coding categories (Table 1) were preset heuristics from the literature (Currie et al., 2010; Li et al., 2012; Richardson et al., 2017).
Predetermined Coding Categories (Derived From the Literature).
Once the participants’ comments were coded according to coding category, the comments within each coding category were organized with regard to the type of commentary that was being intended (i.e., positive, negative, or neutral attitude). All comments made with a neutral attitude were not included in the analysis, as they were not useful in determining strengths and weaknesses of the tool (Li et al., 2012). A content analysis of the positive and negative comments associated with each coding category was completed. Throughout the analysis, peer debriefing and audit trails were used to establish data trustworthiness.
Results
Five PCPs participated in the study: two physical therapists who work in an interdisciplinary primary care clinic, two family physicians, and one specialist who work in an interprofessional primary care clinic. The participants’ median age was 48 years (range: 38–57 years) and the median number of years of practice was 22 years (range: 5–30 years). The median for participants’ self-rated familiarity with EMR was 9/10 (range: 8–10), while the median for participants’ average self-rated technological proficiency was 8/10 (range: 6–9).
Time Requirement for Physical Activity Counseling Tool
The total time to complete the physical activity counseling tool was determined for each participant. The average time taken to complete both components of the tool with a mock patient was 9:31 minutes (±3:37). The average time required to complete the physical activity screening tool was 3:47 minutes (±1:26) and to complete the care plan was 5:44 minutes (±2:47). The fastest that a participant completed the physical activity counseling tool was 6:58 minutes, in which the screening tool was completed in 3:12 minutes and the care plan was completed in 3:46 minutes.
Features That Impeded the Usability of the Physical Activity Counseling Tool
A content analysis exposed issues related to each coding category. Two categories specifically, efficiency and visibility, had a greater number of associated negative comments/criticisms. Regarding efficiency, several participants noted that the care-planning portion of the tool was “too complex” and required too much thought and attention to read and complete given the time demands present in primary care environments. Respondents’ use of the word complexity was often associated with large amounts of text and options being present simultaneously. The problems surrounding “complexity” was especially prevalent in the care plan, as one general practitioner (GP) stated, “I’m kind of pressured because I’m trying to have a relationship communication with this patient, but I also have to read all this.” Another GP stated, “I don’t think a lot of the doctors are going to study it. They’re going to want to just have something that helps them.” For one participant, the number of options and large amount of text present in the care plan prevented them from fully completing the care plan tool with a mock patient.
With regard to visibility, issues were especially noticeable when a feature with poor visibility resulted in ineffective navigation. First, the navigation guide, shown in the top right-hand corner of Figure 2, was not noticed by 4/5 participants. The instructions were in a small font, and the guide was in a position of the form that is often not visible when viewed on a computer screen unless you scroll over, resulting in poor usability. Also, the pop-up associated with the “intensity cues” link often revealed itself in a location on the laptop that was rarely visible (appearing to the right of the field of view on a laptop with standard screen settings, shown on the right side of Figure 1). It was suggested that visibility for important text could be maximized by increasing font size, changing font colour, and by placing a border around the text.
Concerns about the content of the tool were raised. Two participants suggested having more diseases listed within the “Disease-Specific Goals” section within the care plan (Figure 2). Although the tool already has several common diseases and conditions such as diabetes and osteoporosis, additional diseases/conditions that were recommended included low back pain, chronic obstructive pulmonary disease, and Parkinson’s disease. A summary of major features that impeded the usability of the physical activity counseling tool is shown in Table 2.
Major Features That Impeded the Usability of the Physical Activity EMR Tool.
Note. EMR = electronic medical record; COPD = chronic obstructive pulmonary disease; PAS = physical activity screening.
Features That Facilitated the Usability of the Physical Activity Counseling Tool
Features that facilitated the usability of the tool were related to the coding categories content and workflow. With regard to content, the presence of simple and nonintimidating prompting questions assisted participants in completing the tool with a mock patient, despite having no previous exposure to it. For example, prompting questions were clearly visible in the PAS questionnaire and resulted in PCPs being able to determine the mock patient’s physical activity habits easily. But, when prompting questions were less visible, such as in the care plan, some participants had difficulty continuing the mock patient interaction, which was evident as one GP noted, “I’m not sure what I’m supposed to be reading or doing next. . . . At this point I’m really not sure what to recommend [to the mock patient].”
Features that allowed participants to document mock patients’ responses were essential in aligning with the workflow of PCPs, ultimately making documentation and charting convenient. These features included comment boxes, spaces for additional notes, and options to schedule a follow-up appointment. One participant stated, “I would use this tool in my interviewing process. It would be a really easy way to document. . . . A super easy way to document.” Additional major features that facilitated the usability of the physical activity counseling tool are summarized in Table 3.
Major Features That Facilitated The Usability of The Physical Activity EMR Tool.
Note. EMR = electronic medical record; PAS = physical activity screening; PCP = primary care provider.
PCPs Expectations of a Physical Activity Counseling Tool for Primary Care
Key insights that emerged related to PCPs expectations of an EMR-integrated physical activity counseling tool are summarized in Table 4.
PCPs’ Expectations of a Physical Activity Counseling Tool for Primary Care.
Note. PCP = primary care provider; EMR = electronic medical record; COPD = chronic obstructive pulmonary disease.
Discussion
Combining think-aloud testing, a near-live scenario, and a semistructured interview resulted in a thorough evaluation of the usability of the physical activity counseling tool. We identified areas requiring further development related to the efficiency, navigation, and visibility of features within the tool. Despite this being participants’ first exposure to the tool, PCPs could complete the tool in less than 10 minutes, suggesting that with changes to increase efficiency, and repeated exposure, the time required to use the tool could be condensed to fit within primary care practice. Overall, the comments made by PCPs suggest that further development of the physical activity counseling tool should address issues within all coding categories explored, but addressing those specific to efficiency and navigation may be especially important if the tool is to be integrated into primary care.
The time required to complete physical activity counseling, with or without an EMR tool, may be the biggest barrier preventing PCPs from integrating the tool into their daily practice. One barrier that was expressed in our study, and has been reported elsewhere, is the lack of extra time that PCPs have in their daily agendas (Clark et al., 2020; Viscomi et al., 2013). Currently, the physical activity counseling tool requires an average of 9:31 minutes for a first-time user, a number that will likely decrease with more familiarity. One GP in our study noted, “After using it two or three times I would be pretty quick with it. It would be like a 2-minute thing.” Researchers analyzing the usability of an EMR platform noted that the tasks within an EMR program need to be timesaving, or time-neutral at the very least (Rose et al., 2005).
To date, there has been little research involving usability testing on EMR tools that focus on physical activity counseling in primary care, although usability testing has been completed on EMR platforms and the results are relevant for EMR tools. One such study has shown that large amounts of text and a multitude of options can be overwhelming to a PCP (Rose et al., 2005). Ultimately, PCPs dislike when a lot of information is presented on a screen simultaneously as it interferes with their ability to complete tasks quickly (Rose et al., 2005). Similarly, PCPs in our study spoke negatively of having too much text present in the counseling portion of the tool at once, noting that it affected their ability to focus on the task at hand, communicating with a patient. As one GP stated, “I’m kind of pressured because I’m trying to have a relationship communication with a patient, and I’m thinking, okay wait a minute, I’ve got to read all this.”
In a usability study that assessed two complex decision support tools in an electronic health record, researchers noted that PCPs had a better understanding of the tool when clear instructions and simple language were present (Richardson et al., 2017). Furthermore, PCPs were more efficient in navigating the tool when next steps were clearly indicated (Richardson et al., 2017). Our study found that the presence of simple and nonintimidating prompting questions throughout the PAS questionnaire played an essential role in navigating the tool easily. When prompting questions were not available or were not visible (such as in the care plan tool), some PCPs had difficulty navigating the tool, and in one case, became completely disoriented, stating, “I’m not sure what I’m supposed to be reading or doing next.” Although we spent considerable effort in making the care planning portion of the tool comprehensive and easy to use, it is clear that we may need to sacrifice the amount of information presented at once, and instead include more prompts regarding counseling steps, to make it simpler. This is evident through a statement made by one of the PCPs, “Before I would see a patient in real-life now, I would spend some time going over that. It’s still not in my head.”
Overall, our insights from our usability analysis reveal that further evaluation of the tool is necessary to ensure that the physical activity counseling tool is properly integrated into primary care. Despite the need for further evaluation and development, the comments of the participants reflected a strong need for an EMR-integrated physical activity counseling tool in primary care, because as one of the GPs noted, “There’s [currently] nothing.”
We acknowledge that there are limitations of our study. As is the case with many usability studies in health care settings, the physical activity counseling tool was not evaluated in a real clinical environment, which was not realistic due to the tool still being in the preliminary stages of development. The PCPs’ perceptions of the tool may have changed if the usability testing would have been conducted in a realistic health care setting, as a result of factors such as time pressure. The participants in our study are also not likely to be a comprehensive, representative sample of the population of PCPs currently practicing in primary care in Southern Ontario. The participants were a convenience sample and were likely individuals interested in implementing physical activity counseling in primary care. Furthermore, two of our participants were physical therapists working in a primary care setting. Physical therapists may have a different user experience or needs when it comes to a physical activity EMR tool. However, the physical therapists we included worked in Family Health Teams, which are interdisciplinary primary care models, where physical therapists work directly in the primary care setting, and regularly use the EMR for documenting patient care. Thus, it was critical to get their feedback as primary users and arguably they may be more likely to use the tool on a day-to-day basis.
Conclusion
Using a unique make-up of think-aloud testing, near-live scenarios, and semistructured interviews, our study provided insight into the usability issues of an EMR-integrated physical activity counseling tool. Key findings revealed major features that impeded usability, which included the presence of large amounts of text and poor visibility of essential links that severely interfered with navigation of the EMR tool. A major feature that facilitated usability included the presence of simple and nonintimidating prompting questions to guide PCPs throughout a discussion regarding physical activity and thereby supporting PCP’s workflow. The findings of our study will support further development of our tool and provide insight into the expectations that PCPs have of a physical activity counseling tool for primary care.
Footnotes
The research was funded by the Strategy for Patient-Oriented Research and the Canadian Institutes of Health Research grant PEG-151773.
![]()
