We conducted a nonblinded RCT with an intervention group (video consultation) and a usual care group (face-to-face consultation), with 1:1 allocation.

Patients were recruited from a large teaching hospital (Rijnstate, Arnhem). To be eligible to participate, patients had to be older than 18 years, be diagnosed with moderate or severe OSA (AHI >15), require CPAP treatment, have no history of CPAP treatment, have access to a tablet or smartphone, and be proficient in the Dutch language. Exclusion criteria were having a psychiatric or cognitive disorder.

Prior to the study, a letter was sent to patients to confirm their appointments (eg, sleep study and consultation with the pulmonologist) including information about the study. During the first face-to-face consultation with the pulmonologist, patients received their treatment plan and information about the study (including information letter and informed consent form). This was followed by instruction about their CPAP treatment. After this consultation, the researcher provided patients with additional information about the study, and they were asked to sign the informed consent form. For reasons of clinical necessity, patients started treatment the same day.

After patients signed informed consent and completed the baseline questionnaire, they were randomized by the researcher to the intervention or usual care group using the software program Research Manager (Cloud9 Software) with block size of 10. The researcher informed the patients about their allocation, and the intervention group received additional information about the video consultation app (Facetalk, Qconferencing) [32]. All participants received a copy of the informed consent form, and a follow-up appointment was planned directly.

The video consultation app Facetalk [32] could be downloaded (for free) from Google Play [33] or the App Store [34]. The first video consultation with a nurse was planned for 1 week after the start of CPAP. Patients received an email with the date, time, and a link to start the video consultation in the app. Three focus points were discussed during the consultations: (1) adherence (>6 hours per night), (2) rest AHI <5 (or <10 if age over 70 years), and (3) (improvements in) symptoms. If these objectives were achieved after 1 week, a new consultation was planned for 3 weeks later (4 weeks after the start). If these objectives were not achieved, video consultations were planned for weekly (until 4 weeks after starting CPAP treatment). After 4 weeks, patients received a questionnaire. See Multimedia Appendix 1 for the study process.

The usual care group followed the same care process but with face-to-face consultation instead of video consultation. Patients received a confirmation letter with the day and time of their next consultation.

Data analysis was performed using SPSS (version 22.0, IBM Corp). Descriptive statistics were used to report the baseline characteristics, experiences, expectations, and satisfaction. Linear mixed models were used to analyze differences in CPAP use over time for the intervention and usual care group (interaction term: time × group). All available CPAP use data were used in the analysis, according to the intention-to-treat principle. Differences in adherence over time between groups was analyzed using generalized estimating equations. The relation between CPAP use and risk perception, outcome expectancies, and self-efficacy was analyzed with a linear regression. Normally distributed variables were reported as mean and standard deviation, and statistical differences were tested using an independent samples t test. Nonnormally distributed data were reported with medians and interquartile range (25th to 75th percentiles), and differences between groups were analyzed with Mann-Whitney U tests.

All participants signed a written informed consent form prior to inclusion in the study. The study was approved by the regional medical research ethics committee Commissie Mensgebonden Onderzoek Arnhem–Nijmegen and registered at Clinicaltrials.gov [NCT04563169].

Patients were included from January 2, 2019, until June 26, 2019. In total, 222 patients were screened for eligibility, and 50 patients did not meet the inclusion criteria: no tablet or smartphone (n=17), no proficiency in the Dutch language (n=10), AHI <15 (n=10), history of CPAP treatment (n=5), no OSA (n=4), psychiatric or cognitive disorder (n=3), and age <18 years (n=1). In total, 28 patients declined to participate, and 4 patients were not informed about the study for other reasons: 2 patients were not referred to the researcher due to logistical errors, 1 patient followed a different care process (there was no consultation with the pulmonologist that same day), and 1 patient had had CPAP for try out for a short period.

In total, 140 patients were randomized, and 70 patients were allocated to the intervention group and 70 patients to the usual care group. During the intervention period, 2 patients discontinued the intervention: 1 preferred face-to-face consultation, and 1 had no working device. Four patients stopped CPAP treatment during the intervention period (first 4 weeks). In total, 10 patients were lost to follow-up in the intervention group (n=9 stopped CPAP treatment and n=1 died) and 3 in the usual care group (n=3 stopped CPAP treatment). See Figure 1 for the CONSORT (Consolidated Standards of Reporting Trials) flow diagram.

Both groups had similar baseline characteristics (Table 1), only outcome expectancies (P=.048) and risk perception (P=.02) appeared to be significantly different between groups.

The use of video consultation does not lead to superior results on CPAP use compared with face-to-face consultation. A significant difference in change over time was found between groups for short-term (weeks 1 through 4) CPAP use (P-interaction=.008). However, the specific time points (week 1: P=.62; week 2: P=.15; week 3: P=.33, and week 4: P=.20) were not significantly different. See Multimedia Appendix 2 and Multimedia Appendix 3 for more detailed information on short-term CPAP use.

No significant difference in change over time for long-term CPAP use (week 4, week 12, and week 24) was found between groups (P-interaction=.68). CPAP use decreased for both groups in the long term (P=.008), but no significant difference was found between the intervention and usual care group (P=.09). See Table 2 and Figure 2 for change in CPAP use over time (week 4, week 12, and week 24).

The use of video consultation does not lead to superior results on CPAP adherence compared with face-to-face consultation. No significant difference was found between both groups for short-term (P=.95) and long-term (P=.12) CPAP adherence. Also, no significant difference in change over time between the intervention and usual care group was found for short-term (P-interaction=.17) and long-term (P-interaction=.51) CPAP adherence. See Multimedia Appendix 4 and Multimedia Appendix 5 for the short-term and long-term adherence rates per week.

No significant difference between groups was found for the SEMSA constructs: outcome expectancies (P=.64), self-efficacy (P=.41), and risk perception (P=.30). See Multimedia Appendix 6.

After 4 weeks, a relation was found between CPAP use and self-efficacy (P=.001), meaning that patients with higher levels of self-efficacy showed higher CPAP use. There was no relation between CPAP use and risk perception (P=.34) or outcome expectancies (P=.76). Also, the difference between the intervention and usual care group was not significant (P=.25).

Patients expressed positive expectations for the use of video consultation. After 4 weeks, 76% (48/63) indicated that video consultation had a positive effect on control over their treatment, and 75% (47/63) indicated that it positively affected the treatment itself. The majority (58/63, 92%) implied it did not cost them effort, 95% (60/63) reported that they had enough skills to use a tablet or smartphone and that they received enough support (53/63, 84%). Although, 64% (44/69) expected to be stimulated by people in their direct environment to use video consultation, only 25% (16/63) were actually stimulated. Almost all patients (60/63, 95%) intended to keep using video consultation. See Multimedia Appendix 7.

Patients in both groups were satisfied with the consultations. On average, the intervention group rated the consultations with an 8.5 and the usual care group with an 8.3 on a scale of 1 to 10 (1=not at all satisfied to 10=very satisfied). Patients indicated (intervention group versus usual care group) that health care professionals understood their problems (59/63, 94%, vs 58/68, 85%) and listened to them (60/63, 95%, vs 61/68, 90%). Almost all patients understood the content of the consultation (61/63, 97%, vs 62/68, 91%), could easily express their feelings (59/63, 94%, vs 62/68, 91%), and were satisfied with the information they received (58/63, 92%, vs 60/68, 88%). However, more patients with video consultation reported that they did not miss important information (56/63, 89%, vs 43/68, 63%). See Multimedia Appendix 8.

The majority (56/63, 89%) of the patients were very satisfied with video consultation, the quality of the video (50/63, 79%), and sound of the system (45/63, 71%). It also saved them time (61/63, 97%) and provided better access to health care professionals (43/63, 68%). Almost all patients felt safe about their privacy and confidentiality (61/63, 97%) and preferred a video consultation over a face-to-face consultation (51/63, 81%). According to almost half (28/63, 44%) the patients, face-to-face consultation can be replaced by video consultation. See Multimedia Appendix 9.

Nurses (n=3) rated the use of video consultation on average with a 7.3 (SD .57) on a scale of 1 to 10 (1=not at all satisfied to 10=very satisfied). They were all satisfied with privacy and confidentiality and quality of the sound and video and would recommend its use to colleagues and patients. Two nurses agreed that its use fits in their work process. However, only one nurse was completely satisfied with the information she could provide. They did not think that the use of video consultation helped them save time or work more efficiently.

The nurses reported that use of video consultation is not suitable for new patients, and they prefer to use it during follow-up:

The nurses also experienced some technical problems:

They also provided suggestions for improvement and described advantages of video consultations:

In this RCT, we evaluated the superiority of video consultation over face-to-face consultation for newly diagnosed OSA patients. For CPAP use, we found a significant difference in change over time between groups in the short term (P-interaction=.008). However, the specific time points (week 1: P=.62; week 2: P=.15; week 3: P=.33, and week 4: P=.20) were not significantly different. No significant difference in change over time was found for long-term CPAP use (P-interaction=.68). No significant difference in change over time between groups was found for short-term (P-interaction=.17) or long-term (P-interaction=.51) CPAP adherence. Self-efficacy appeared to have a statistically significant effect on CPAP use in both groups (P=.001) regardless of the intervention arm (P=.25). No significant difference between groups was found for outcome expectancies (P=.64), self-efficacy (P=.41), or risk perception (P=.30). The experiences with video consultation were very positive. Almost all patients (60/63, 95%) intended to keep using video consultation. Patients in both groups rated the consultations on average with an 8.4. All nurses (n=3) were satisfied with privacy and confidentiality aspects and quality of the sound and video. However, they expressed some recommendations for improvement (eg, to use video consultation only in follow-up).

Unfortunately, change over time was not evaluated in previous controlled studies [18,26,29], but this evaluation is as such a likely pattern. In our study, a significant difference in CPAP use between video consultation and face-to-face consultation was not found. Parikh et al [18] reported statistically equivalent CPAP use for new OSA patients (mean average use minutes per day 305.31 vs 340.55, P=.15). In a multicenter RCT, no statistically significant difference was found for CPAP use after 6 months (telemedicine mean use 4.4 [SD 2.0] hours per day vs face-to-face 4.2 [SD 2.0] hours per day, P=.83) and adherence (telemedicine 65% vs usual care 57% compliance, P=.33) [29]. Based on these findings, it appears that CPAP use is equivalent to using video consultation.

Where previous studies mainly focused on CPAP use, adherence, and satisfaction with video consultation [17,18,25,26,29], we additionally evaluated the combination of cognitive components (self-efficacy, outcome expectancies, and risk perception), experience with the technology (using the UTAUT model), and satisfaction of patients and nurses. This combination of outcomes has received little attention until now. Cognitive components are found to be increasingly important in predicting CPAP use [13,14,27]. Our results show that use of CPAP is higher in patients with high levels of self-efficacy (P=.001) regardless of the intervention arm (P=.25). In order to improve self-efficacy, it is necessary to positively influence patient perceptions. Patients may benefit from a self-management approach [27,41,42] with tailored education to change their perceptions about CPAP use and subsequently improve self-efficacy [43]. Lai et al [44] provided patients with additional education to enhance, for example, self-efficacy. This increased CPAP use compared with patients receiving usual care (P<.001). Stepnowsky et al [41] showed that a self-management program with information about OSA- and CPAP-related issues led to high self-efficacy scores (4.5 [SD 0.6]; scale 0 to 5) and CPAP adherence (5.5 [SD 2.3] mean hours per night). Because self-efficacy scores can be affected by the time that patients are treated, scores should be assessed regularly in order to be useful in clinical practice [14].

However, limited evidence was available about the effect of video consultation for newly diagnosed patients starting CPAP. Most previous RCTs were small, with sample sizes varying from 19 to 40 patients [17,25]. Only Isetta et al [29] evaluated CPAP compliance with a fully powered sample size. Although almost half of the patients (40%) in this study had insufficient digital skills, technology aspects were not evaluated [29]. In our study, 9% (20/222) were unable to participate because of lack of access to a mobile device or due to psychiatric or cognitive disorder. During the intervention, 2 patients (2/70, 3%) discontinued the video consultation intervention because of preference for face-to-face consultation or problems with their mobile device. The use of video consultation is evolving rapidly in clinical practice, but digital services are not applicable to all patients and digital health literacy remains a challenge [45]. This is especially due to lack of awareness or knowledge or unwillingness to change [46] and emphasizes the importance of personalized interventions rather than a one-size-fits-all approach.

The assessment of UTAUT components and self-efficacy can also be used to indicate technology use [47]. To our knowledge, no previous studies have identified technology acceptance for OSA patients using video consultation. Patients in our study had positive experiences with the use of video consultation and were satisfied with the video consultation system and consultations in general. Previous studies also reported high satisfaction scores [17,18,25,26], mostly regarding communication with a health care professional [18] and privacy and security factors [17]. Although most patients would recommend the use of video consultations to others, not all patients in our study are convinced that all visits can be replaced by video consultations. This is in line with findings from previous research [17].

The involvement of health care professionals is essential to achieve successful implementation of technology [48], but this is often not evaluated [17]. We found that nurses (n=3) preferred to start with a face-to-face consultation because education about the sleep mask and adjustments are often required during the first follow-up appointment with the nurse. The applicability of technology use may be dependent on the population [49], and for OSA patients, the use of video consultation in a blended care setting might therefore be beneficial. We found that the nurses were satisfied with video consultation and especially with the quality of the system, privacy and confidentiality. They would recommend it to colleagues and patients. Nurses also reported technical problems (eg, problems with Wi-Fi connections). Technological issues are often seen as a barrier [50], and it is important to take technical elements into account [48,51,52] during implementation. Another point for improvement is integration in existing health care processes (eg, planning). To achieve successful implementation, it can be beneficial to involve professionals during the implementation process itself [50].

Video consultation can be seen as a promising app to support OSA patients during treatment. Still, evidence was lacking and previous research was not strong enough in design or focused on a limited number of outcomes. With the evaluation of a broad range of outcomes affecting CPAP use and implementation of video consultation in clinical practice, this RCT adds value to current knowledge.

However, proper evaluation in this field is challenging because research often lags behind the rapid development of technology [53]. The use of pragmatic trials may be promising [54] to evaluate different elements of eHealth solutions in a hospital setting and can, for example, be used to get (more) rapid insights in relevant implementation outcomes such as feasibility, impact on an organization, and acceptance and adoption by health care professionals and patients. Future research should focus on blended care approaches in which self-efficacy especially receives greater emphasis. For organizations to be able to implement video consultation on a larger scale, integration in existing health care processes and technology acceptance by patients and professionals is necessary.

Several limitations should be considered. Risk perception and outcome expectancies were significantly different at baseline, despite randomization. For a limited number of patients (7/66, 11%, in the intervention group and 6/70, 9%, in the control group), video consultations or face-to-face consultations were replaced with a telephonic consultation due to technical problems in the intervention group and because patients in the control group could not come to the hospital. The protocol process were not strictly followed because patients failed to attend their scheduled appointment (no show, sick, on holiday) or there were organizational inaccuracies such as wrongly scheduled appointments. The percentage of patients that followed the process exactly as described (Multimedia Appendix 1) was higher in the intervention group (approximately half) than in the usual care group (approximately one-third). However, all patients received the intervention (type of consultation) they were allocated to except for the 2 patients who discontinued the intervention (Figure 1). Another limitation is that only 3 nurses were involved in the evaluation. Therefore, a firm conclusion on professional aspects cannot be drawn.

Support of OSA patients with video consultation does not lead to superior results on CPAP use and adherence compared with face-to-face consultation. The findings of this research show that a significant difference in change over time was found between groups for short-term CPAP use (but not on specific time points), but not for long-term CPAP use. Levels of self-efficacy were positively related to CPAP use in both groups. Patients were very satisfied with video consultation and reported positive experiences.

Therefore, the findings of this research suggest that self-efficacy is an important factor in improving CPAP use and that video consultation may be a feasible way to support patients starting CPAP. The integration in health care processes and tailoring video consultation use to patient and professional needs is essential to ensure successful use. A blended care setting, in which an initial video consultation is combined with face-to-face consults, may be beneficial. To our knowledge, this is the first RCT that examined the effects of video consultation on CPAP use over time for newly diagnosed OSA patients in combination with cognitive components and experience with technology use. Future research should focus on blended care approaches in which self-efficacy receives greater emphasis.