A convergence model with a triangulated mixed methods approach was conducted to gain in-depth comprehension in the user engagement, acceptability, and usability of an mHealth intervention aimed at the reduction of sedentary behavior. This methodology allowed us to compare, corroborate, or relate quantitative data (ie, system usage and activity monitor data) and qualitative data (ie, interview data). Quantitative and qualitative data were analyzed separately, followed by an integrated interpretation of the results. Participants in the mixed methods study were recruited in Flanders, Belgium, from February to May 2019 using convenience sampling. Recruitment continued until data were saturated (ie, until no new themes emerged in additional interviews). Firstly, an advertisement was distributed via Facebook, and secondly, the advertisement was electronically sent to older adults who were included in a previous study by our research group and who had expressed interest in future studies. To be eligible for this study, participants needed to (1) be at least 60 years old, (2) be Dutch speaking, (3) be able to walk 100 meters without severe difficulties, and (4) have a smartphone. The study was registered at ClinicalTrials.gov (Identifier: NCT04003324) and was approved by the Committee of Medical Ethics of the Ghent University Hospital (Belgian registration number: 2019/0398). All participants provided written informed consent.

The study procedure is explained in Figure 1. Concretely, older adults who agreed to participate were contacted by phone to make an appointment for a first home visit. During this home visit, they received an information letter explaining the purpose of the study and an informed consent form. After signing the informed consent form, baseline measures were collected. Specifically, a structured interview was conducted to assess participants’ sociodemographic characteristics. Moreover, an accelerometer—activPAL (PAL Technologies)—was attached to the participants’ thighs to objectively measure their sedentary behavior. Participants were instructed to wear the accelerometer for 1 week and to fill out the accompanying diary. After 1 week, a researcher visited the participants again at their homes to collect the accelerometers. After baseline measurements, the self-monitoring–based mHealth intervention was introduced to the participants (see Self-Monitoring mHealth Intervention section). At the end of the intervention (ie, after 3 weeks), participants were asked to complete a semistructured interview that included questions on user engagement with the intervention and perceptions regarding usability and acceptability. At the end of this last home visit, participants were instructed to wear the accelerometer for another week (ie, postmeasurements). Participants were given a prestamped envelope and were asked to send the accelerometer back by postal mail.

The intervention consisted of general sedentary behavior information as well as visual and tactile feedback on participants’ sedentary behavior. General sedentary behavior information was provided to participants by means of a 10-minute presentation. The presentation was given by an expert in the field during the second home visit. Visual and tactile feedback were provided using a novel self-monitoring device—the Activator (PAL Technologies). The Activator has recently been validated by Gill et al [20]. The Activator is worn on the front of the thigh, either in a pants pocket or attached with an elastic band to clothing covering the upper thigh (eg, trousers, jeans, shorts, leggings, tights, or dresses), and provides visual and tactile feedback [21]. Visual feedback is presented through a smartphone app via Bluetooth connection. Both real-time feedback and a 7-day historical overview are presented based on participants’ sedentary time, upright time, and number of steps (see Figure 2). Visual feedback is constantly available and can be viewed whenever and as often as participants want. Tactile feedback is provided by means of a strong, but comfortable, vibration of the Activator device itself each time a participant is sitting for 30 uninterrupted minutes. If a participant remains sedentary, the vibration is repeated after another 30 minutes. Participants were able to turn the vibration function on and off.

Descriptive statistics were used to describe the baseline characteristics of the participants and to assess the extent of usage (ie, engagement). Qualitative data were thematically analyzed, using the NVivo 12 software package (QSR International), using the six-phase approach by Braun and Clarke [27] to gain insight into participants’ subjective experiences with the intervention (ie, engagement) and acceptability and usability of the intervention. More specifically, two researchers outside the project team—Charlotte Meersseman and Siel Mechelinck—read and reread the transcripts multiple times to become familiar with the data (phase 1). They independently coded the data line by line and defined an initial coding scheme using an inductive approach (phase 2). The coding schemes were then discussed with the first author (SC). By doing so, the triangulation technique was applied and the trustworthiness and validity of the findings were promoted. Based on the coding schemes, themes were searched (phase 3), reviewed (phase 4), and defined (phase 5). Subsequently, pen profiles (ie, diagrams of composite key emergent themes, frequency data, and verbatim quotes) were constructed based on the defined themes and results were written up (phase 6). This increasingly utilized technique is considered appropriate for presenting qualitative outcome data in a clear and useful manner [28]. Paired-samples t tests were performed to determine preliminary efficacy of the intervention on older adults’ sedentary time. All quantitative analyses were conducted in SPSS Statistics for Windows, version 25.0 (IBM Corp).

A total of 36 older adults expressed interest in participation. Out of these 36 participants, 2 (6%) of them could not be reached to make an appointment for the first home visit and 4 (11%) decided to withdraw from the study after receiving detailed study information. Reasons for withdrawal were health problems (2/36, 6%), lack of time (1/36, 3%), and death of a spouse (1/36, 3%). As such, 30 older adults completed the baseline measurements. Out of these 30 participants, 2 (7%) of them were excluded, as baseline data showed that they did not fulfill the inclusion criteria (ie, they were not able to walk 100 meters without severe difficulties). In addition, 2 (7%) participants dropped out during the intervention period due to health problems (1/30, 3%) and lack of motivation (1/30, 3%). Consequently, posttest data were collected from 26 out of 28 participants (93% retention).

Baseline characteristics of the participants are presented in Table 1. Just over half of the participants (15/28, 54%) were female, the average age was 65.0 years (SD 4.6), and the mean BMI was 25.4 kg/m² (SD 3.9). The majority of the participants were highly educated and were married or lived with a partner.

Qualitative data on user engagement were thematically analyzed and are presented in Figure 3. The main themes that emerged were positive and negative feelings about the intervention, preferences for the kind of feedback, and the pattern of use.

The participants mainly reported positive feelings, such as being motivated, surprised, and interested. Only a minority (3/28, 11%) indicated that they thought the intervention was not interesting and not helpful. There were mixed opinions on the preferred kind of feedback (ie, tactile vs visual). Some thought the vibrations were more useful, whereas others favored the visual information on the app. System usage data showed that the median number of days the self-monitoring device was worn by the participants was 20 out of 21 days (range 15-21). Half of the participants (16/28, 57%) reported that they accessed the app on a daily basis. This finding was confirmed by system usage data (see Multimedia Appendix 2), showing that 8 out of 28 participants (29%) consulted the app every day, while 5 participants (18%) consulted the app at least 80% of the days. Some participants reported that they consulted the visual feedback multiple times a day. Accordingly, system usage data showed that the median frequency of consulting the app ranged from 0 to 20 times a day (see Multimedia Appendix 2). Especially in the evening, and after doing physical activities, participants reported that they viewed the visual feedback. They indicated that the main reasons to access the app were out of curiosity, to go through their day, and to see the impact of certain physical activities. Participants also emphasized that they consulted the app more frequently in the beginning of the intervention period, compared to the end of the intervention period. This finding was also in line with the system usage data, which show that the median frequency of consulting the app ranged from 3 or 4 times a day in the beginning of the intervention to 1 or 2 times a day at the end of the intervention (see Figure 4).

Results of the thematic analysis on acceptability and usability of the intervention are presented in Figure 5. The main themes that were identified were the design and the ease of use, wearing preferences, problems and solutions, the focus, and the perceived relevance. The intervention was considered easy to use, and most participants described the design as clear. The only remark on the design were the colors of the behaviors. Participants frequently cited that it would be more logical if sedentary behavior (ie, the behavior that should be limited) were displayed in red and the number of steps (ie, the behavior that should increase) in green. Participants expressed mixed preferences regarding the way to wear the device. Some participants (11/28, 39%) preferred to use the elastic band, whereas others (13/28, 46%) preferred to wear it in their pockets. Frequent problems that older adults, especially women, experienced included small or loose pockets, loss of the device, and the imprint of the elastic band on their clothes after wearing it. Out of the 28 older adults, 5 (18%) indicated that they used a handkerchief to ensure that the device was fixed in their pocket and could not flip over. Despite the fact that the aim of the intervention was to reduce sedentary behavior, only 2 participants (7%) mainly focused on the sedentary behavior information. Although the majority of older adults rated the intervention as highly relevant, some older adults were not convinced about the relevance. The most important reasons for limited perceived relevance were (1) the fact that they do not spend a lot of time sitting, or at least think they are not sitting a lot, and thus have no need to reduce their sedentary time and (2) the fact that they often see no other option but sitting to perform certain tasks.

Participants commonly reported that the intervention changed their thinking (ie, they became more aware of their sedentary behavior) but not their actual sedentary behavior. The latter result was supported by quantitative data derived from the activity monitor (see Table 2). Sitting and standing time were very similar at pre- and postmeasurements. There was a small improvement in steps of around 400 per day. This improvement was not significant, probably because of the small sample size.

This study provides novel and in-depth insights into the potential of a self-monitoring–based mHealth intervention in older adults to reduce sedentary behavior. Overall, our results indicated that the intervention was generally well perceived by older adults, but preliminary analyses showed no reduction in sedentary time after the 3-week intervention period.

Previous research has shown that building sustained user engagement over time is challenging in mHealth interventions [29]. Low user engagement results in limited exposure to the intervention and, in turn, small or no intervention effects [30]. Therefore, gaining insight into the user engagement of mHealth interventions is crucial. Both objective usage data and subjective experiences showed that older adults were highly engaged with this study’s intervention. The participants generally expressed positive feelings, and the majority consulted the feedback frequently. They all agreed that the intervention made them more aware of their sedentary behavior, but the intervention did not result in a decrease in sedentary time.

The lack of a decrease in sedentary time is not entirely surprising given the following reasons. Firstly, the intervention period of 3 weeks was probably too short to actually change habitual behavior. Changing habits takes a long time [31] and, thus, it is likely that participants still need the cues to interrupt and/or reduce their sedentary behavior after the intervention has ended. Ending the cues might have resulted in relapse into their old and unhealthy habitual sedentary behavior [32]. Secondly, the intervention mainly targeted automatic processes underlying sedentary behavior by bringing the habitual behavior into conscious awareness. However, dual-process theories of motivation posit that both controlled and automatic processes regulate our sedentary behavior [33]. Thus, additional behavior change techniques (eg, goal setting, action planning, and coping planning) should be included in the intervention to affect the controlled processes and to actually achieve behavior change. Given that participants often mentioned that they saw no other options to reduce their sedentary behavior, it might be worth including concrete examples on how to reduce sedentary behavior. Thirdly, physical activity information (ie, number of steps) was also provided in the app, notwithstanding that the only aim of this intervention was to reduce sedentary behavior. The physical activity information could not be removed from the Activator app before the start of the study. Existing literature has indicated that participants of interventions targeting both sedentary behavior and physical activity simultaneously are more likely to focus on increasing physical activity due to (1) the clearer guidelines for physical activity (ie, 150 minutes of moderate- to vigorous-intensity physical activity a week) compared to sedentary behavior (ie, sit less), (2) the better-known negative health consequences of too little physical activity compared to too much sedentary behavior, and (3) the fact that physical inactivity is often still considered a synonym for sedentary behavior [34]. The latter was confirmed by the results of the semistructured interviews: participants often mentioned that they wanted to increase the number of steps in order to reduce their sedentary time. Objective physical activity data showed that the average daily number of steps increased by approximately 400, or 10%, over the 3-week intervention period. Although this increase was not significant, this indicates that Activator feedback is more likely to affect the number of steps than the sedentary time. This finding is in line with the results of previous Activator studies [21,35] and suggests that more efforts should be made to clarify the difference between sedentary behavior and physical inactivity and to emphasize the importance of standing and light-intensity physical activity.

Despite the fact that common aging-related barriers (eg, visual impairment, reduced working memory, limited motivation, and reduced mobility) can influence the use of mHealth in older adults [36], general perceptions on the acceptability and usability of this study’s intervention were positive. The app was easy to use and the design was clear. This is of great importance, as previous research has indicated that simplicity is one of the key principles for the design of mHealth interventions for older adults [37-39]. Although the Activator could be worn in different ways (ie, in the pants pockets or with an elastic band), the wearing of the device was often mentioned as challenging, especially among women. More research is therefore required to determine the ideal manner of attaching and wearing the Activator, especially when wearing pants without pockets or without deep pockets.

Strengths of this study include the innovativeness of the research. To our knowledge, this is the first study examining older adults’ experiences with an electronic self-monitoring device specifically developed to reduce sedentary behavior. Moreover, by collecting both qualitative and quantitative data, a comprehensive view was obtained on self-monitoring–based mHealth interventions to reduce older adults’ sedentary behavior. An important limitation of this study was the sampling method. Participants were not randomly selected and, therefore, selection bias may have occurred. The majority of the participants were highly educated, whereby generalization of the results to lower-educated groups might be limited. Moreover, no control group was included, as the main aim was to gain in-depth knowledge on participants’ perceptions with the intervention. Although data saturation was achieved in the qualitative analysis, the small sample size was only sufficient to get a first indication on effect sizes and was not meant to provide sufficient statistical power for the quantitative analysis. Finally, the intervention lasted only 3 weeks and, thus, no conclusions can be drawn on the long-term adherence to the intervention. Based on these limitations, future studies should endeavor to recruit a larger, more generalizable sample and should use a randomized controlled trial design to draw firm conclusions on the effectiveness of a self-monitoring tool to reduce older adults’ sedentary behavior. Furthermore, we believe that adding behavior change techniques to the mHealth intervention, ones that can affect the controlled processes underlying sedentary behavior, and extending the intervention duration might be recommended in future studies.

Results of this study suggest that the innovative self-monitoring–based mHealth intervention holds potential for the reduction of sedentary behavior in older adults. The intervention was considered interesting, helpful, and easy to use, and was able to increase awareness among older adults of their sedentary behavior. Despite the positive perceptions, no reductions in objective sedentary time were found in this study’s sample. Hence, the intervention was probably of insufficient intensity to reduce the sedentary behaviors of participants. In order to effectively achieve behavior change, a number of modifications to the intervention are suggested, such as the addition of behavior change techniques that target controlled processes underlying sedentary behavior.