A web-based survey of technology use and participation was developed for dissemination in Israel on the basis of the National Alliance on Mental Illness (NAMI) mHealth survey [14]. The original survey was translated to Hebrew, adapted to the local context, and supplemented with additional items pertaining to recovery [14] and community participation [37]. The final version of the survey included four sections: (1) technology use, (2) recovery, (3) community participation, and (4) background and demographic characteristics.

The survey was distributed in Israel between February 2018 and May 2019 among adult individuals (over the age of 18 years) who identified as having SMI (the study group) and those without SMI (a comparison group), based on a screening question. SMI was defined on the basis of the Substance Abuse and Mental Health Services Administration criteria [42], which are acceptable for distribution of block grants to state governments, and the Israel National Insurance Institute criteria in Israel for disability support [43,44]. Accordingly, individuals with chronic conditions, such as schizophrenia, major depression, or another mental health disorder that results in serious disability are considered under the umbrella of SMI [45,46].

Before large-scale dissemination, we conducted a field-test of the final version of the survey with 4 individuals with SMI who commented on the questions and wording and further refinement. Survey participants were recruited via a snowball approach using Facebook groups and national email discussion lists (“Listservs”) of consumers and service providers as well as via nonprofit organizations and advocacy groups of and for people with SMI. We also provided an option of a paper-and-pencil version of the survey for individuals with SMI who were interested in participating via a face-to-face meeting with the research assistant.

The study was approved by both the University of Haifa Institutional Review Board and Israel Ministry of Health. SPSS (version 25, IBM Corp) was used to present descriptive statistics and calculate correlations and means differences. A P value of <.05 was set as the significance level.

The sample includes 381 respondents, of whom 199 (52%) reported having at least 1 SMI, with 16 (8%) having reported more than 1 SMI. The most common condition was schizophrenia (n=58, 29.1%), followed by affective disorders (bipolar disorder, n=16, 8%; depression, n=12, 6%; and anxiety n=5, 2.5%), personality disorders (n=7, 3.5%), and posttraumatic stress disorder (n=6, 3%). Most respondents with SMI (n=157, 78.9%) reported an additional nonmental health chronic condition, mostly diabetes (n=10, 6.4%) or an orthopedic condition (n=10, 6.4%), compared to those without SMI, who reported no such condition.

More women (n=246, 64.6%) than men were included in the survey, although the distribution was more balanced among those with SMI (Table 1). The most frequent age category was 18-34 years (n=146, 38.3%). While most respondents with SMI were single (n=129, 64.8%), the majority of respondents without SMI were married (n=103, 56.6%). While most of the respondents without SMI were employed (n=131, 72.0%), most respondents with SMI were employed only in supported employment programs (n=106, 53.3%). Lastly, while most respondents with SMI had at least a high school diploma (n=80, 40.2%), the majority of respondents without SMI had a higher education degree (n=136, 74.7%).

Overall, men reported having significantly greater access to technology (mean 3.58, SD 1.47) than did women (mean 3.01, SD 1.43; t₃₇₉=3.62; P<.001). Women reported significantly greater negative emotions (mean 2.06, SD 0.86) when using digital technologies compared to men (mean 1.79, SD 0.83; t₃₇₉=–2.95; P<.01). Compared to men, women also reported significantly more total days of community participation in terms of both amount (women: mean 42.02, SD 11.73; men: mean 37.67, SD 12.17; t₃₁₂=–3.09; P<.01) and extent of participation (women: mean 20.02, SD 10.11; men: mean 17.30, SD 10.74; t_262.3=–2.40; P<.05).

Respondents with SMI reported significantly greater access to technology (mean 3.63, SD 1.43) than those without SMI (mean 2.57, SD 1.38; t₃₇₉=–6.15; P<.001). Respondents with SMI also reported significantly greater negative (mean 2.05, SD 0.87) and positive emotions (mean 2.99, SD 0.96) when using digital technologies compared to those without SMI (negative emotions: mean 1.87, SD 0.80; t₃₇₉=–2.06; P<.01; positive: mean 2.59, SD 0.78; t_373.8=–4.43; P<.05). Finally, respondents with SMI reported significantly fewer days of participation in community activities (mean 37.6, SD 11.41) compared to those without SMI (mean 43.9, SD 11.92; t₃₁₂=4.79; P<.001).

Multiple linear regression analysis was conducted to predict recovery among the respondents with SMI (n=199) based on their access to digital devices, emotional experience with technology use (negative and positive), and the amount and extent of their community participation. A significant regression model was obtained (F_5,163=9.39; R²=0.224; P<.001). Greater experience of emotions while using technology, both negative (β=–.290; P<.001) and positive (β=.299; P<.001), and greater amount of community participation (β=.249; P<.01) significantly predicted recovery. Greater positive emotions and more days of community participation were positively related to recovery, while greater negative emotions were negatively related to recovery.

Regression analyses were conducted to assess the contribution of technology use (for illness management) and emotions experienced while using the technology to recovery among respondents with SMI (n=199). First, we assessed the direct association between technology use for illness management (also referred to “helpfulness of activities when using technological devices” in the NAMI study [14] and recovery; Figure 1, path c). Significant and positive β weights were obtained (β=.19; P<.01).

Second, 2 regression analyses were carried out with technology use for illness management as the independent variable; 1 model included negative emotions and the other included positive emotions as the dependent variables (Figure 1, path a). Significant and positive β weights were obtained for both negative (β=.15; P<.05) and positive emotions (β=.50; P<.001).

Third, 2 multiple regression analyses were conducted to predict recovery by positive and negative emotions and technology use for illness management, respectively (Figure 1, path b). A negative and significant β weight was obtained between negative emotions and recovery when controlling for technology use (β=–.17; P<.05). A positive and significant β weight was obtained between positive emotions and recovery when controlling for technology use (β=.36; P<.001). A positive and significant β weight was obtained when predicting recovery with technology use as the predictor and controlling for negative emotions (β=.22; P<.01). No such result was obtained when predicting recovery with technology use as the predictor and controlling for positive emotions (β=.01; P>.05) (Figure 1, path c). These patterns suggest partial mediation between technology use (for illness management) and recovery when negative emotions serve as a mediator, and complete mediation in the case of positive emotions under the 4 conditions of mediation [39].

This study is the first to explore and conceptualize digital community inclusion among individuals with SMI. Our study focused on digital participation of adults with SMI in Israel. We compared the mHealth and digital technology use (ie, virtual participation) of adults with and those without SMI and explored whether and how participation factors predict recovery among adults with SMI. The main findings included higher rates of digital participation among individuals with SMI compared to those without SMI. In addition, greater positive emotions were noted when digital participation in the community, digital technology use, and more days of face-to-face community participation were positively related to recovery among adults with SMI. Positive emotions during digital community participation mediated the relationship between technology use (for illness management) and recovery. The findings emphasize the important role of digital participation on community inclusion of adults with SMI.

Previous studies that compared digital technology use among individuals with and those without SMI have yielded contradictory findings. On one hand, lower rates of technology access and use were observed among individuals with SMI than among those from the general population [47,48]. On the other hand, some studies found that individuals with SMI used digital technologies at rates similar to those of the general population, and concluded that mental health problems may not be a barrier to technology use [14,49,50]. Although the aforementioned studies found similar or lower rates of technology use among individuals with SMI, our study reported a higher rate of technology use among individuals with SMI than in the general adult population. A plausible explanation for this finding is that the other findings of this study showed that adults with SMI reported significantly fewer days of participation in community activities than those without SMI. Another plausible explanation for this finding is the growing number of mental health and wellness apps available for individuals with SMI [51,52]. Furthermore, using digital technologies for mental health care has created a more accessible environment for people with SMI, thus enabling anonymous participation. In case they decide to disclose personal information, nonverbal and non–face-to-face communication may create a less stigmatic environment for interaction with others [8,53,54].

In addition to exploring mHealth and digital technology use among adults with SMI, this study examined whether digital community participation predicted recovery. The theoretical framework of community participation traditionally refers to face-to-face or actual participation and is defined as the empowered, self-determined choice and action among individuals to be active in valued roles in the communities of their choice [26]. According to this traditional framework, the term “community participation” includes 3 main types of participation: social (eg, attending a community event, entertaining family or friends at home, or visiting family or friends), productive (going to school to earn a degree or certificate, working for pay, and participating in volunteer activities), and leisure (going to a museum, theater or cultural event, going to a park or recreating center, and going to a restaurant) [41]. Engaging with others in the community may also reduce public stigma toward people with SMI, which in turn can contribute to their recovery [55-57].

However, it seems that the life domains of community participation have referred to the physical environment, while participation in the digital environment has been excluded, although the digital environment has become an integral part of our life [58,59] and even more so in the COVID-19 era [4]. Following the concept of recovery as a dynamic process [30], studies have indicated that social support plays a main role in an individual’s recovery [60,61]. Social support can be delivered not only through face-to-face interactions, as demonstrated in the traditional concept of community participation and from a distance through remote communication using digital technologies [62]. While some preliminary studies focused on the impact of social media use on face-to-face community participation among individuals with SMI [22,63,64], they have focused solely on social media and did not include mHealth and digital technologies for illness management.

Moreover, recent studies have not conceptually included digital participation as part of community participation. This study has addressed this concept and theoretical gaps by exploring the predictive factors to recovery while taking into account both digital and face-to-face, in-person participation. As reported by Hendryx et al [65], involvement in a wide range of activities, whether they are more or less social in nature, physically active, or occur inside or outside of home, was related to better recovery. Hence, using digital technologies for greater involvement in physical or digital activities, whether the activities are intended for spending time alone or for contacting others socially, can empower people to manage their recovery [66]. Our findings provide further support to this argument by showing that using technology for illness management predicted recovery.

Furthermore, emotional experience of technology use were found to mediate the relationship between technology use and recovery, while positive emotions completely mediated this relationship. Studies on human-computer interactions emphasize the important role of emotions in technology adoption [67,68]. Emotions, as a central component of attitude toward a referent, are a mental state of readiness for action, which promote behavioral activation [67,69]. Positive emotions are responsible for the user eventually trusting the technology and using it [70].

The study has several limitations. First, because the survey was conducted on the internet, sampling may be biased by recruiting adults who are likely to be more technologically savvy. However, we recruited adults with SMI not only through web-based groups but also in face-to-face meetings with individuals with SMI who were interested in participating, and they completed a paper-and-pencil version of the survey. This strategy enabled us to recruit a more heterogeneous sample in terms of access to technology. Second, the majority of survey respondents were young adults with only few over the age of 65 years, which could also explain the relatively high use of technology. Therefore, the results for older adults with SMI must be interpreted with caution, and future studies should target a subgroup of older adults with SMI.

Lastly, although we acknowledge that poverty plays a crucial role in access to technology [71,72], participants in our sample enjoy the social welfare benefits provided by the In Israel Ministry of Health and the Israel National Insurance Institute. This mental health support system provides financial and rehabilitation support, including housing, education, employment, and mental health care. Therefore, although often individuals with SMI experience poverty and lack of access, respondents with SMI in our sample enjoyed social welfare benefits (Table 1), which may explain their greater access to technology. Future studies should focus on evaluating the impact of poverty on digital participation and access to technology.

This study demonstrated the potential of digital community inclusion to recovery and well-being among individuals with SMI. Our findings indicate higher rates of access and use of mHealth and digital technologies among individuals with SMI than among the general population. Furthermore, our findings show that digital participation could promote recovery among adults with SMI. Accordingly, this study emphasizes the need to update and expand the definition and conceptualization of community participation, and include aspects of digital participation needs of individuals with SMI.

Our findings suggest that policy makers, service users, and researchers should use existing digital technologies and design novel mHealth interventions to support the recovery process of adults with SMI. In particular, the current COVID-19 crisis poses an opportunity for mental health care systems to adopt digital technologies for service provision. In this sense, this study, conducted before the current COVID-19 pandemic, contributes to the understating that digital participation of adults with SMI is valuable to their recovery not only in crisis but also in routine. Furthermore, it is important to support the participation of individuals with SMI in the virtual environment in a manner that facilitates a positive emotional experience. Positive emotional experience while using digital technologies is a key factor in their engagement in the web-based environment and consequently in their recovery.