Stress is a major public health concern. It can cause serious psychological and physical problems such as fatigue, sleeping problems, coronary diseases, depression, and even related mortality [1-4]. Hence, proper stress management for employees is particularly important; work-related stress affects approximately 28% of all workers [5]. Stress has also been reported to be a major factor in up to 80% of all work-related injuries and 40% of workplace turnovers [6]. Data from the American Institute of Stress show that job stress in the United States accounts for over US $300 billion annually as a result of increased absenteeism, employee turnover, diminished productivity, medical, legal, and insurance expenses, and worker compensation [7].

Traditional face-to-face or in-person stress reduction interventions for employees are proven to be effective, with small-to-moderate effect sizes [8]. However, a major barrier to managing stress is limited access due to time and location constraints, as it may be difficult for employees to find the time to see a therapist during work hours on weekdays [9]. The success of nonpsychopharmacological psychiatric treatment is related to treatment adherence, which is linked to barriers like time and distance. A useful way to increase access, and thereby increase treatment adherence, is by incorporating telemedicine-based methods. Employees can easily access videoconferencing-based treatments at their convenience, ie, during break hours in the workplace or after office hours at home [10].

There is a bulk of evidence suggesting that videoconferencing- based telepsychology is no less effective than in-person treatment in a variety of psychiatric disorders [11], including depression [12], panic disorders [13], obsessive compulsive disorder [14], posttraumatic stress disorder [15], and eating disorders [16]. A recent meta-analysis of 26 randomized controlled trials demonstrated noninferiority of remote psychiatric counseling, with respect to both assessment and treatment, compared with in-person counseling [17]. To the best of our knowledge, to date, there has been no study evaluating the effectiveness of videoconferencing in reducing stress and enhancing resilience in the workplace.

An important issue in videoconferencing is equipment. In most previous studies, participants were instructed to use the equipment provided at remote videoconference sites [11]. However, traveling to these sites is no different from traveling to see a counselor; it would still be a barrier to treatment for those whose workplace or home is far from the videoconference site. To further increase accessibility to treatment, we used mobile-based videoconferencing equipment. Access to treatment by smartphones could be particularly beneficial in South Korea, since almost all Korean employees own a smartphone (94% of population), making it one of the countries with the highest smartphone ownership worldwide [18].

The purpose of this study was to compare the effects of mobile videoconference-based intervention on stress reduction and resilience enhancement with that of in-person and self-care methods in Korean employees. We hypothesized that the effectiveness of the mobile videoconference condition would be superior to the control condition (self-care condition), but comparable to the in-person condition.

Of the 98 individuals screened for this study, 17 did not meet the inclusion criteria; 81 individuals were thus enrolled and randomly allocated to 1 of the 3 conditions. Among them, 4 in the mobile videoconference condition and 1 in the in-person condition dropped out after randomization but before treatment initiation (mobile videoconference condition: 3 had trouble installing the app on their smartphone and 1 refused participation due to difficulty in scheduling appointments; in-person condition: 1 needed psychiatric treatment due to aggravation of psychiatric symptoms). As a result, 21 participants were allocated to the mobile videoconference condition, 27 to the in-person condition, and 28 to the self-care condition; all 81 subjects completed the pretreatment assessment. After the start of treatment, 3 participants in the mobile videoconference condition and 1 participant in the self-care condition dropped out (mobile videoconferencing condition: 2 dropped out due to their personal schedules, 1 complained of unstable Wi-Fi connection; self-care condition: 1 dropped out of because of personal matters, but refused to give a detailed explanation). A total of 18 individuals in the videoconference condition, 27 in the in-person condition, and 27 in the self-care condition completed all 4 sessions of the intervention and underwent the posttreatment and 1-month follow-up assessment (Figure 1). The drop-out rates after treatment engagement were 14% (3/21), 0%, and 3% (1/27) for the mobile videoconferencing, in-person, and self-care conditions, respectively; this distribution was not statistically significant (P=.09 by Fisher exact test).

The demographic and clinical characteristics of participants who completed the assessments at all three time points are presented in Table 1. The mean age of participants in the mobile videoconference condition and in-person condition was higher than that of the self-care condition (P<.001), and there was a significant difference in marital status among the conditions (P=0.03); hence, age and marital status were included as covariates in the main analyses. There were no significant differences in gender, length of work, education status, or baseline scores of the clinical variables. When classifying the occupations of the participants according to the International Standard Classification of Occupations 08, 5 were managers, 26 were professionals, 6 were technicians and associate professionals, 12 were clerical support workers, 19 were service and sales workers, 1 was a plant and machine operator, and 3 had elementary occupations. Overall, 42% (30/72) were hospital employees.

The effects of condition, time, and condition × time for all clinical variables are shown in Multimedia Appendix 2; Figures 2-5 depict changes in the PSS, KELS, BRS, and AIS scores across time. The interaction between time and condition was significant for 4 clinical variables (PSS: F_3.40=3.1, P=.03; BRS: F_3.47=3.9, P=.008; KELS: F_3.45=2.8, P=.03; AIS: F_3.04=4.5, P=.005). There were significant main effects for condition on PSS (F₂=8.7, P<.001) and KOS (F₂=11.6, P<.001).

At posttreatment, the mobile videoconferencing condition showed a greater decrease in KELS scores at posttreatment, but this was not significant at follow-up (Multimedia Appendix 2). There were no significant differences in any clinical variable between the mobile videoconferencing condition and in-person condition at follow-up. The mobile videoconferencing condition showed a greater decrease in KELS scores compared with the self-care condition at posttreatment, but this was without significance at follow-up. The mobile videoconferencing condition showed a greater increase in BRS scores at follow-up than the self-care condition. The in-person condition showed a greater decrease in PSS, KOSS, BRS, and AIS scores at posttreatment than the self-care condition, but only KOSS, BRS, and AIS scores were significant at follow-up.

Regarding the 4 questions about therapeutic alliance, there were no differences in the scores between the videoconferencing condition and in-person condition: question 1: average score 4.7 (SD 0.6) versus 4.7 (SD 0.4), P=.58; question 2: average score 4.7 (SD 0.6) versus 4.9 (SD 0.3), P=.32; question 3: average score 4.6 (SD 0.5) versus 4.9 (SD 0.3), P=.06; question 4: average score 4.7 (SD 0.6) vs 4.7 (SD 0.6), P=.98.

To the best of our knowledge, this is the first study investigating the effectiveness of a mobile videoconferencing-based intervention for stress reduction and resilience enhancement in employees. Videoconferencing was delivered using a smartphone app, thus combining the merits of both videoconferencing and mobile devices. Overall, there were significantly differential effects across time according to treatment condition on perceived stress, resilience, emotional labor, and sleep. Moreover, there were no significant differences between the mobile videoconferencing and in-person conditions at follow-up, suggesting that the benefits of mobile videoconferencing therapy were comparable to those of in-person therapy.

At follow-up, both the mobile videoconferencing and in-person conditions had significantly enhanced resilience compared with the self-care condition, which was the primary goal of the intervention. Resilience enhancement has been recognized as an important part of stress reduction. Resilience refers to the process that allows individuals to adapt positively despite stress or trauma [47]. Resilience-based interventions emphasize strengths within individuals and community members to persevere and recover from environmental, physical, or emotional stress [48]. Previous research regarding resilience suggests approaches that build on strengths rather than problem-focused strategies [47]. Southwick and Charney suggested 5 essential components to resilience interventions: (1) emotional regulation training to recognize and manage reactivity and impulsivity; (2) cognitive behavioral approaches to reframe thought processes and increase positive emotion; (3) physical health information on exercise, nutrition, sleep, and relaxation to increase protective behavior; (4) social support to build connections with family, peers, and mentors to increase protective factors; and (5) neurobiological components, such as mindfulness-based stress reduction, to increase the ability to manage stress [49]. Our intervention contained most of these components, leading to successful resilience enhancement.

Most studies on videoconferencing-based treatments have focused on addressing rural populations. A few studies have investigated the effectiveness of videoconferencing in homebound populations, such as disabled adults with dementia or older adults with depression [50,51]. This study suggests that telepsychology methods can also be useful to young employees living in metropolitan areas. Most adults in their 30s and 40s are smartphone savvy, which can allow them to access treatment at their convenience as long as they have a Wi-Fi or LTE (Long-Term Evolution) connection [10]. Videoconferencing via smartphones is also cost-effective as most people already own smartphones and they will be able to save travel costs [52].

To date, a variety of effective stress management programs for employees has been proposed and made available [53]. A recent meta-analysis found that while cognitive behavioral therapy interventions yielded the largest effect sizes, relaxation and meditation techniques were the most popular [54]. A SMART-3RP program that included both of these components has been found to improve resilience and minimize perceived stress in palliative care clinicians, medical interpreters, and resident physicians [44]. The intervention duration for the original SMART-3RP program was 8 weeks; however, this study showed that the intervention was effective only after 4 sessions. Moreover, this intervention was effective and also had a low drop-out rate (videoconferencing condition 14.2%, in-person condition 0% after treatment engagement) compared with previous studies. For comparison, a recent meta-analysis found that the mean completion of workplace psychological treatments was 45%, with a range of 3%-95% [55]. With respect to the 0% drop-out rate in the in-person condition, almost 40% of participants were employees of the hospitals in which the study took place; hence, high accessibility may have contributed to the low drop-out rate.

Although there was no statistical significance at follow-up, the post hoc analysis revealed that the mobile videoconferencing condition had a greater effect on emotional labor than both the in-person and self-care conditions at posttreatment. Emotional labor is a unique type of stress experienced in employees. Brotheridge and Grandey proposed that there may be two sources to job-related stress: emotional demands of the work environment and employees’ ability to control their emotions. This indicates that workers with emotionally demanding jobs and a low capacity for emotional control likely experience the greatest job-related stress [29]. Of the participants who completed the study, 19 (26.4%) were nurses. Given that emotional labor is key to making patients feel safe and comfortable [56], many nurses experience emotional labor, resulting in higher job stress, poorer health, greater self-alienation, and increased frequency of depressive mood [57]. The results of this study suggest that mobile videoconference interventions may be effective in reducing emotional labor in employees, at least in the short-term.

A major concern of videoconference treatment is the quality of therapeutic alliance between patient and therapist [58]. Therapeutic alliance has been defined as collaborative effort by the therapist and patient to facilitate healing [59]. In this study, there were no significant differences in therapeutic alliance ratings between the conditions. Our findings demonstrate that stress intervention via mobile videoconferencing does not compromise therapeutic alliance, which is in line with previous telepsychology research results [60].

There are a few notable limitations to this study. The length of intervention and follow-up interval were relatively short. Thus, this study provides no information on the long-term effect of mobile videoconference interventions. Participants were mostly female and all were Korean, limiting the generalizability of the findings to the male gender and other ethnic groups. We did not exclude participants with depression, insomnia, or anxiety disorders, making the study population clinically heterogeneous; as the sample size was not sufficient for subgroup analyses according to the presence or absence of psychiatric diagnosis, we were unable to evaluate whether psychiatric diagnoses influenced the effect of mobile videoconference treatment. Moreover, we did not measure if the intervention improved any workplace variables, such as work performance, absenteeism, and turnover rate. There is also the possibility of a selection bias caused by the recruitment of highly motivated participants. Lastly, participants and therapists were not blinded to their treatment conditions, which may have caused an expectation bias.

In conclusion, this study demonstrates that videoconferencing- based stress reduction interventions can be effective in employees. Further studies with larger sample sizes and longer follow-up intervals may be helpful for determining the long-term effect of this intervention.