|Home | About | Journals | Submit | Contact Us | Français|
To examine whether video-based mHealth interventions are feasible, acceptable, understandable, and engaging to people with schizophrenia.
This study used a mixed methods design. Ten individuals with schizophrenia spectrum disorders were recruited for a month-long trial in which they used FOCUS-AV, a smartphone system that offers video and written intervention options. Participants completed post-trial measures and engaged in semi-structured interviews.
One participant dropped out. The remaining nine participants used intervention videos successfully. Participants responded to 67% of system-delivered prompts to engage FOCUS-AV, and 52% of FOCUS-AV use was initiated by the users. On average, participants used interventions six days a week, four times daily. Participants used video functions an average of 28 times. They chose video over written interventions on 67% of the times they used on-demand functions, but opted for written content 78% of the times they responded to pre-scheduled prompts. Clinician videos were rated as more personal, engaging, and helpful than written interventions. Video and written interventions were rated equally usable and understandable. Written interventions were rated as more favorable in letting users proceed at their own pace. Similarly to what is seen in live therapy, the communication style and demeanor of clinicians depicted in intervention videos reportedly affected participants' experience with treatment.
Video-based mHealth may be a feasible, usable, acceptable, and highly engaging method for flexible delivery of interventions to people with schizophrenia using mobile technology. Producing intervention videos is more time, labor, and cost intensive than generating written content, but participant feedback suggests there may be added value in this approach. Additional research will determine whether video-based mHealth interventions lead to better, faster, or more sustainable clinical gains.
Mobile Health (mHealth) programs that leverage mobile phones as instruments for illness monitoring and management are becoming increasingly popular. There is growing scientific evidence demonstrating that people with serious mental illnesses, including schizophrenia and bipolar disorder, can benefit from using mHealth interventions as adjunctive to psychosocial services (Ben-Zeev et al., 2014; Depp et al., 2015; Granholm et al., 2012; Pijnenborg et al., 2010). To date, mHealth interventions for people with serious mental illness have used reminders, suggestions, and illness management interventions in the form of text messages or screens with written content and static images. However, a large proportion of the U.S. population, including people with serious mental illness, now own smartphones that have the capacity to store, stream, and play video (Torous et al., 2014). These ubiquitous personal devices could potentially be used to support video-based mHealth interventions.
Smartphone-supported video interventions could be helpful in delivering illness management interventions to people with schizophrenia who experience illness-related difficulties in their day-to-day environments (Ben-Zeev et al., 2012). Videos depicting clinicians who “speak” directly to users via their mobile device could perhaps extend the therapeutic experience beyond the clinic setting and may be viewed as more personalized than text-based interventions. Videos may be more instrumental in modeling behavior (e.g. social skills training) than written messages or static images that cannot convey dynamic facial expressions and body language. Finally, video may be a particularly accessible medium for individuals who have limited education, literacy, or motivation as videos require less reading, scrolling, or “swiping” through digital pages of written material.
mHealth interventions that use video may be an important step in actualizing the vision of a digital “therapist in one's pocket” (NIMH, 2015). But this approach is not without potential pitfalls; people with schizophrenia are characterized by difficulties in information processing and reality testing (Brüne, 2005; Dudley et al., 2015; Garety & Freeman, 1999) which could impact their ability to fully distinguish between live interactions with a clinician (e.g., two-way video tele-psychiatry) and pre-prepared videos that are available on their personal mobile phone as part of an automated intervention. Some may be apprehensive about using video interventions outside of a clinic without on-site technical guidance and support.
Whether video-based mHealth interventions are feasible, acceptable, and engaging to people with schizophrenia is unclear. To answer these questions we conducted a proof-of-concept study in which participants were provided with a smartphone intervention system that offered a choice in intervention modality; either videos depicting clinicians who “speak” directly to the user or written clinical content. Participants used the intervention in their own environments over a month-long trial and provided comprehensive feedback on their experience.
The study was approved by the Committee for Protection of Human Subjects at Dartmouth College and conducted in collaboration with a community mental health agency in the Northeastern U.S. Clinical staff approached candidates with a chart diagnosis of schizophrenia or schizoaffective disorder to inquire whether they would be interested in learning about the study. Research staff described the study to referred individuals and invited them to attend a screening appointment at the community mental health center where their reading level was assessed and their ability to hear, see, and have sufficient dexterity to access content on a smartphone was evaluated using a demonstration device; To test dexterity and vision, participants were asked to unlock the device using the touchscreen and dial a number using the visual dial pad display. To test hearing, research staff launched a FOCUS-AV clinical assessment prompt and participants were asked to raise their hand after they heard the audio signal to respond. Participants who passed the screener underwent baseline assessment and received training on how to operate the smartphone and use all functions of the mHealth intervention.
Once participants demonstrated basic proficiency, they were provided with a smartphone with a data plan and the mHealth intervention installed and activated. Participants were instructed to use the smartphone and mHealth intervention over a month-long period and were encouraged to use the other smartphone functions. Research staff called participants to check-in and offer troubleshooting assistance weekly for the entire study period. At the end of the month participants returned the smartphone, underwent post-trial assessment, and received $80 compensation.
Participants were provided with a Samsung S5 Smartphone with FOCUS-AV (audio/video) mHealth system installed and headphones. FOCUS-AV is an adapted version of the FOCUS smartphone intervention that was shown to be usable, acceptable, and helpful in reducing psychotic and depressive symptom severity in people with schizophrenia (Ben-Zeev et al., 2014). Briefly, FOCUS was designed for people with schizophrenia and offers both pre-scheduled and on-demand illness management interventions targeting auditory hallucinations (i.e. voices), social functioning, medication use, mood problems, and sleep disturbances (Ben-Zeev et al., 2013). Interventions are structured as brief interactive modules in which users are asked to rate their clinical status daily using multiple choice options displayed on the touchscreen. Based on their self-ratings, FOCUS delivers an intervention in the form of screen sequences containing illness-management suggestions and support statements in written text and images (i.e., photos or cartoons). Participants' FOCUS use data (e.g., responses to prompts, interventions delivered) are stored and transmitted to a study server when the smartphone has internet connectivity. Once data are transmitted, clinicians or researchers can view updated summaries of participants' system use via secure online dashboard.
FOCUS-AV contains video adaptations for all the FOCUS content. With FOCUS-AV, users receive the same daily prompts to engage in assessments. After they provide a response, users are given the option to choose between a written intervention or video version. If they select the written intervention, users navigate at their own pace through a sequence of several screens by pressing a touchscreen button labeled “continue.” If they select video, a media player with a “play” touchscreen button appears on the screen. If a video is launched, it will run continuously unless users press the “pause” button. Videos can be replayed by pressing the “play” button again. Videos depict one of two clinicians trained in evidence-based interventions who speak directly to the camera and offer shortened versions of illness management strategies that would typically be administered in the context of live therapy (e.g., guided relaxation, cognitive restructuring). Several videos contain demonstrations played out by an actor and narrated by the clinicians (e.g. calling a friend for support, using post-it note reminders, using earphones to “drown out” voices). FOCUS-AV also offers an introduction video in which one of the clinicians introduces the system and encourages participants to use it in their day-to-day lives.
Ten individuals with schizophrenia or schizoaffective disorder were enrolled in the study. Participants had a mean age of 45.5 years (SD= 13.18), were 60% male and 90% Caucasian. They had an average 12th grade reading level (determined by the Wide Range Achievement Test-4th edition [Wilkinson & Robertson, 2006]). Participants reported an average of 1.8 (SD=1.17) psychiatric hospitalizations. At baseline, participants had on average mild to moderate psychotic symptom severity (Psychotic Symptom Rating Scale [Haddock et al., 1999] M=31.9, SD=10.67, range= 18-48) and mild symptoms of depression (Beck Depression Inventory-II [Beck et al., 1996] M=14.3, SD=10.50, range= 0-35). All but one participant (90%) owned a mobile phone of some kind and 33% indicated their device was a smartphone. Two participants (20%) had never used a smartphone before their participation in this study.
Quantitative data were collected on participants' objective use of the mHealth intervention, their evaluation of intervention usability, feasibility, acceptability and satisfaction, and their assessment of video vs. written content combined with static images. Demographic and clinical data were collected during the baseline assessment. We used a parallel convergent mixed methods design (Creswell & Plano Clark, 2007) with qualitative data from semi-structured interviews to complement our understanding of the quantitative findings regarding participants' intervention modality preferences, and to further explore their experiences with the video interventions. Mixed methods approaches in proof-of-concept research generate detail and context about the user's experience which are informative in intervention refinement and testing (Fetters, Curry, & Creswell, 2013).
The smartphone logs all FOCUS-AV activity. Feasibility was defined as the percent of participants who were able to use both system-initiated (i.e. in response to prompts) and participant-initiated (i.e. on-demand) videos independently and in their own environments for a minimum of three days after receiving the smartphone. Response Rate was defined as the proportion of clinical status assessments participants responded to from those sent. Video Selection was defined as the percent of incidences in which participants chose video (rather than written content) from the total number of interventions they received after responding to daily prompts. Video Initiation was defined as the total number of times participants self-initiated “on-demand” video interventions (not including videos tallied for the calculation of Response Rate).
Participants completed several measures during the post-trial assessment meeting. First, an 11-item measure in which they were asked to indicate their intervention modality preference: video, written content, or video and written content rated equally (all items and summary of responses appear in Table 1). Second, a 12-item measure examining usability, acceptability, and satisfaction with the FOCUS-AV video content (all items and summary of responses appear in Table 2). Finally, participants rated three statements pertaining to possible negative impact of the FOCUS-AV program as a whole (i.e., audio prompts, clinical status ratings, written interventions, and videos): “Using the program made me upset”, “I had concerns about my privacy using the program”, and “Using the program made me suspicious” (rated 1-“Strongly Disagree” to 5- “Strongly Agree”).
Following quantitative assessments, participants were asked a series of open-ended questions to learn more details about their experiences with the video and written interventions using a brief, targeted, and flexible semi-structured interview guide. A trained research interviewer conducted the 30-minute qualitative interviews and participants provided details about their experiences and preferences. The content of their responses was recorded and used for analyses.
Quantitative data analysis involved calculation of descriptive statistics including means and frequencies. Statistical analyses were performed using SPSS software, version 22.0. Qualitative data analysis involved an iterative team-based approach (Beebe, 2001). Three researchers (KA, RB, and GJ) independently reviewed written notes from the interviews. The study team identified key domains that would complement and expand our understanding of the quantitative findings regarding participants' preferences for intervention modality. The research team reviewed the key domains and selected qualitative quotes to ensure that they were representative of the data. Any disagreements, which were rare, were resolved through clarification and discussion.
Project staff screened eleven referred individuals. All candidates passed the screener and met with research staff at the community mental health center to undergo baseline assessment and receive training on how to operate the smartphone and mHealth intervention. One individual withdrew during the baseline assessment and explained that he did not have time to participate because “people were after him”. Ten individuals started the study but one unenrolled in the first day; he left the smartphone with his case manager with a note saying “it's not for me”. Of the nine participants who continued, one stopped using the phone mid-trial after receiving a system update notification from the service provider (AT&T) that made him anxious (unrelated to the FOCUS-AV system). One participant's device was stolen and was replaced by study staff after two days. Days in which participants did not use the smartphone (as a result of theft, forgetting to take the device with them, or because they elected not to engage in the intervention) were included in the calculations of the sample's average FOCUS-AV use (i.e., added as “0” in summation). All participants returned the study devices they had at the end of their participation.
All enrolled participants were able to use both system-initiated and participant-initiated videos for a minimum of three days. On average, participants used FOCUS-AV 5.9 days a week (week 1: average 6.4 days with an average of 5.8 times daily; week 4 average: 5.9 days with an average of 3.1 times daily). Throughout the month, participants interacted with FOCUS-AV on average 4.4 times daily. Fifty-two percent of FOCUS-AV use was initiated by participants (i.e., on-demand). The Response Rate (i.e., to prompts) was 66.8%. On average participants' Video Selection was 22.2% (i.e. when responding to pre-scheduled prompts, they opted for written content 77.7% of the time). However, when participants' self-initiated FOCUS-AV use (i.e. on-demand) they opted for videos 66.7% of the time. Their Video Initiation rate over the month-long trial was 28.3 times. All but one individual (who did not use interventions for “voices”), accessed on-demand content from all five treatment modules over the course of their participation: social functioning (on average 12.5 times), mood problems (on average 12.1 times), sleep disturbances (on average 8.8 times), voices (on average 6.9 times), and medication use (on average 4.5 times).
Participants rated their intervention modality preferences (see Table 1). Video was rated higher in most areas. On average, participants found videos to be more personal, engaging, and helpful in supporting their illness management. Video and written modalities were rated equally easy to use and understand, as well as similarly motivating and positive. Participants thought written interventions required less effort, and on average, were better at letting users go at their own pace.
Participants completed a measure evaluating the usability and acceptability of video-delivered content as well as their satisfaction with the system (see Table 2). Overall, they found the intervention videos to be very usable and acceptable. All study completers (100%) stated that they would be willing to play intervention videos in a private location. Approximately 78% stated that they would not feel comfortable playing videos in public places if others could hear the content. All participants indicated they would be willing to use intervention videos if they could use headphones.
Participants rated three statements pertaining to possible negative effects of the FOCUS-AV program as whole; approximately 22% reported feeling upset after using the program and 22% had some concern about their privacy. None of the participants endorsed feeling suspicious when using the system (0%).
The key domains and selected qualitative quotes are summarized in Table 3. Qualitative data were categorized within the following domains: Satisfaction with video interventions; Comparison between video and written interventions; and Suggestions for improving the video interventions. In-depth interviews revealed that participants thought the videos were a more personal mode of delivering the content. Participants commented on the exceptional clinical communication skills of the clinicians and noted that the clinical expert delivering the illness management suggestions gave credibility to the intervention content. While participants were highly satisfied with the video interventions, some participants reported that the written content was more appropriately paced and could be viewed more easily in public places (e.g., workplace) compared to the video-delivered intervention, which they thought would only be viewed in privacy (e.g., at home). Suggestions for improving the videos included extending their length, using a step-by-step method to present illness management strategies, and incorporating more interactive features.
To our knowledge, this proof-of-concept study is the first to report on the use of video-based smartphone interventions among people with schizophrenia. Overall, our sample found mHealth video interventions to be usable, understandable, and highly engaging. Participants rated video interventions as preferable to written mHealth interventions in most aspects. Participants were very satisfied and comfortable with the combination of video and written content used in the smartphone system. While specific events led to user dissatisfaction (i.e., operating system updates that were “pushed” by the mobile carrier, technical problems that caused prompting failures), these were unrelated to the use of video. Furthermore, participants with schizophrenia did not demonstrate or express any unique difficulties in understanding the nature, origin, or purpose of pre-programmed mHealth video interventions.
The interviews revealed that participants liked the more personal “feel” of videos. They commented on how the clinicians' expert delivery of clinical content added credibility to the illness management suggestions. Participants stated that the presenters' calm demeanor had a calming effect. It is noteworthy that both verbal and nonverbal behaviors depicted by clinicians in the videos appear to have had positive effects, similar to how communication style and body language affect live therapeutic interactions (Cruz & Pincus, 2002; McCabe & Priebe, 2008; Sherer & Rogers, 1980).
All study participants used on-demand intervention videos regularly (66.7% of all self-initiated use). However, when they responded to pre-scheduled prompts (which may have occurred when they were in public settings) users typically opted to access written content. Two thirds of participants indicated that written material better allowed them to go through intervention content at their own pace. Following the videos may require more cognitive resources which may be less available to participants when they are prompted, but more accessible when they self-initiate their mHealth use (i.e., they can prepare by sitting in a less distracting setting). Participants found mHealth video interventions to be acceptable, as long as they could watch them using headphones or in nonpublic situations. These findings suggest that people with schizophrenia may be apprehensive about using video interventions in public for fear of public stigmatization (Corrigan, 2004). Alternatively, they may be apprehensive about watching videos in public because it might disturb others and draw unwanted attention irrespective of the content or because it is more difficult to hear videos in noisier environments. Therefore, our study suggests that if mHealth video interventions are to be used with individuals with schizophrenia, providing users with written intervention options, as well as headphones, is recommended.
The study has several limitations. Our sample was small, limiting the generalizability of the findings. However, the number of participants in the current study is consistent with early proof-of-concept research designed to produce feasibility and tolerability outcomes (Bell & Weinstein, 2011; Farrell, Mahone, & Guilbaud 2004; Olincy et al., 2006). The consistency in our participants' feedback regarding ease of use, acceptability, and satisfaction, as well as their objective video intervention use patterns do increase confidence in some of our key findings. Given the small sample size and the objective to use the qualitative data to complement and expand on the quantitative data, we systematically categorized responses that appeared salient in relation to the research questions. A larger sample in future research will allow us to conduct more extensive qualitative data analysis. Participants in the study had an average 12th grade reading level, which is relatively high. Our sample was racially homogeneous, and so were the clinicians and actors depicted in intervention videos. How differences in variables related to education, culture, race, and ethnicity might impact the viability of the intervention in a more diverse sample of people with schizophrenia is unknown.
Previous research has shown that individuals with schizophrenia, as well as others with severe psychiatric disabilities, use social media platforms to post video testimonials as a method of connecting with others and strengthening their own recovery (Naslund et al., 2014). Computerized interventions for schizophrenia have already made good use of videos to relay psychoeducational information or depict training roleplays in supervised clinic settings (Gottlieb et al., 2013; Steinwachs et al., 2011). The study adds to these findings and suggests that video interventions may be a feasible, usable, acceptable, and highly engaging method for more flexible delivery of interventions to people with schizophrenia using mobile technology outside of the controlled clinic environment. Research involving larger and more diverse samples who use mHealth video interventions over longer timeframes will be helpful in evaluating how robust our proof-of-concept findings are. Finally, producing intervention videos is more time, labor, and cost intensive than generating written content. Participant feedback from this study suggests there may be added value in this approach. Systematic clinical research will be essential in evaluating whether mHealth video-interventions lead to better, faster, or more sustainable patient gains, warranting the additional investment.
The authors kindly thank Mr. Raymond Walker who assisted in conducting qualitative interviews and mHealth intervention set up and technical troubleshooting. This research was supported by the National Institute of Mental Health (Award R34MH100195, PI: Ben-Zeev) and Dartmouth SYNERGY Clinical and Translational Science Institute.