Abstract
We report two studies examining family interest in, and the feasibility of, technology-based treatment of childhood anxiety disorders and obsessive compulsive disorder (OCD). In the first study, 33 parents of children (51.5% boys; age 5–18 years) seen for anxiety or OCD completed a survey regarding access to technology, experience with the mental health system, and interest in technology-based tools. The majority of parents (76%) expressed high interest in electronic treatment options and preferred to use technology to facilitate, rather than replace, face-to-face treatment. In the second study, 10 youth (40% boys; age 6–17 years) and a parent completed an interactive online prototype to deliver psychoeducation for childhood anxiety disorders and OCD. User data supported the ability of the application to cover the breadth of symptom presentations, with most families able to identify multiple feared stimuli as well as related beliefs, avoidance behaviors, and potential exposures exercises. Only two entries, one fear stimuli and one exposure item, were made via free text. In addition, user experience was positive, with 80% of respondents rating the prototype highly. Overall, the results of both studies support and inform further development of electronic treatment tools for children with anxiety and OCD.
Anxiety disorders are among the most common childhood mental health diagnoses, with lifetime prevalence rates ranging from 15% to 20% (Beesdo, Knappe, & Pine, 2009). Anxiety disorders that develop in childhood generally follow a persistent course and lead to additional psychological concerns that interfere with social, emotional, and academic development (Ezpeleta, Keeler, Erkanli, Costello, & Angold, 2001). Within the United States, anxiety disorders are estimated to cost $42.3 billion annually (Greenberg et al., 1999). Most of these costs come from reduced productivity, absenteeism, greater use of nonpsychiatric medical and emergency care, and suicide (Lepine, 2002). The use of exposure therapy (ie, gradually facing one’s fears) is widely accepted as the most effective treatment for anxiety disorders (Kendall et al., 2005). Although exposure therapy is an effective, efficient, and well-tolerated treatment, few anxious children receive this intervention (Whiteside, Deacon, Benito, & Stewart, 2016; Whiteside et al., 2016). Only approximately one-third of children with anxiety disorders receive any treatment at all, whereas approximately 40% to 50% of children with depression and 50% to 80% of children with attention deficit hyperactivity disorder receive treatment (Merikangas et al., 2010). Even when children receive help, the treatment is rarely consistent with best practices, because it typically is much shorter and rarely includes exposure therapy (Whiteside et al., 2016, 2016, 2016). The failure to provide evidence-based therapy places children at risk for continued and exacerbated suffering, as well as imposing considerable financial costs on society.
Unfortunately, many barriers limit access to high-quality, therapist-delivered treatment. Although exposure therapy is supported by empirical evidence, many therapists lack the training to use exposure well and are also reluctant to increase their use of this technique (Gunter & Whittal, 2010; Meyer, Farrell, Kemp, Blakey, & Deacon, 2014; Whiteside et al., 2016, 2016). Specifically, surveys of clinicians suggest that approximately 5% of therapists provide exposure therapy, and reviews of medical records indicate that fewer than 25% of children receiving treatment outside of anxiety specialty clinics are ever introduced to exposure therapy (Whiteside et al., 2016, 2016). Rather than providing exposure therapy, most therapists use other techniques, such as relaxation, that are less effective (Whiteside et al., 2016, 2016). To date, efforts to increase therapists’ adherence to best practices have had limited success (Beidas & Kendall, 2010; Herschell, Kolko, Baumann, & Davis, 2010). Additionally, even if all therapists embraced the use of best practices, the number of available providers are insufficient to address the psychiatric needs of all patients (Kazdin & Blase, 2011).
Given these barriers to increasing access to traditional evidence-based therapy, attention has been given to developing technologies to deliver treatment directly to families through electronic self-help programs or to guide clinician delivery of evidence-based treatment. Internet-delivered treatments for childhood anxiety tend to be effective and acceptable to patients (Ebert et al., 2015; Rooksby, Elouafkaoui, Humphris, Clarkson, & Freeman, 2015). Likewise, computer-assisted therapy programs to guide therapist delivery of exposure for childhood anxiety disorders have been well-tolerated and successful (Khanna & Kendall, 2010). As such, technology has the potential to increase access to treatment via self-help and increase therapist use of exposure. However, these programs tend to be accessed via computers, rather than through mobile devices, which does not facilitate their use by patients during real-world, anxiety-provoking situations, and use of exclusively computer-based programs limits dissemination to underserved populations for whom a smartphone may be the only means of accessing the internet (Pew Research Center, 2018). Moreover, currently existing mobile applications are either consistent with evidence-based principles, but lack direct empirical support or, more commonly, are devoid of evidence-based components (Bakker, Kazantzis, Rickwood, & Rickard, 2016; Kertz, MacLaren Kelly, Stevens, Schrock, & Danitz, 2017). As such, there is a need to develop and validate new technologies for increasing the availability of high quality treatment.
Anxiety Coach, an iOS application for mobile devices, was developed to increase access to exposure-based treatment (Whiteside, 2016). To be consistent with evidence-based principles, this application focuses primarily on exposure through an extensive library of activities to help users build an individualized exposure plan as well as the ability to record real-time anxiety levels during exposure to real-life situations. The usefulness of Anxiety Coach as an adjunct to therapy has been illustrated in a case series (Whiteside et al., 2016). However, the initial version of the application had two primary limitations. First, the psychoeducation material introducing families to anxiety and its treatment was text-based and static. Second, the limited functionality associated with small screen size hindered its use with a therapist. As such, an integrated system that completed initial education and treatment planning online via computer, with ongoing treatment (exposure) completed via mobile device, could potentially maximize the advantages of each medium and facilitate use as a self-help tool and therapist guide.
Specifically, a video-based and interactive psychoeducational component available through a computer, could prove more accessible, thorough, and engaging then the current mobile device-based application. Treatment planning through fear ladder building might also be more manageable on a computer screen where the ladder and the activity library could be viewed simultaneously. All information entered through the computer-based application would be accessible through the mobile application, allowing the family to review and edit the education and fear ladders as needed. Exposures could be recorded through the mobile application to allow immersion in feared environments, while those that don’t require being in a particular setting could be recorded through the computer. Progress monitoring or symptom ratings and completion of exposures, could be monitored through tables or graphs that are best viewed on a larger computer screen. Such a program could be used entirely as self-help, as a guide to increase therapist use of exposure (i.e., computer in session, mobile device between sessions), or as a minimal contact hybrid.
The current study presents two components of the initial development of a multiplatform version of Anxiety Coach. These projects represent initial proof-of-concept explorations that were conducted before committing to resource-intensive content and software development. Study 1 involves a survey of parents to gauge interest in technology-based self-help tools for children with anxiety and obsessive-compulsive disorder (OCD). Study 2 involves the testing of a prototype of an interactive psychoeducational component with patients and their parents. All activities involved in the two studies were approved by the institutional review board.
Study 1: Parental interest in technology-based tools
Overview
An important tenet in successful product design is to understand the needs and wants of the intended audience. To this end, Study 1 surveyed parents of children presenting for evaluation and/or treatment of anxiety disorders or OCD. Parents were asked about access to technology, previous experience with the mental health system, and interest in technology-based treatment tools. We hypothesized that 1) most families would have access to technology, but children would not have their own devices; 2) parents would search online and consult their pediatrician when accessing treatment; 3) parents would report difficulties accessing quality care; and 4) families, especially those traveling from a distance, would be interested in technology that integrated mobile and online access and reduced the need to attend in-person treatment appointments.
Methods
Participants
Forty parents were eligible; each represented 1 of 40 patients evaluated and/or treated through a specialty pediatric anxiety disorders clinic at a Midwestern medical center. The survey was completed by 33 parents (response rate, 82.5%). The sample of children was 51.5% male (n = 17) and predominately white (n = 31 [93.9%]), and patient age ranged from 5 to 18 years (mean [SD] age, 12.5 [3.6] years). Per the parent’s report, 25 children (75.8%) had an anxiety diagnosis, 15 (45.5%) had an OCD diagnosis, 9 (27.3%) had both OCD and anxiety diagnoses, and 2 (6.1%) received neither diagnosis. Parents were well educated, with at least some postsecondary education (n = 32 [97.0%]), and most were married (n = 25 [80.6%]). Just over half of respondents (n = 18/32 [56.3%]) lived greater than 60 minutes’ travel distance from the specialty clinic.
Procedure
After the child’s initial assessment, parents were invited by the evaluating clinician to complete the survey. Written informed consent was deemed unnecessary; instead, parents were provided a verbal and written description of the study that emphasized the voluntary nature of participation. To protect anonymity, parents were instructed to return the survey to the clinic front desk when completed.
Materials
The survey did not include identifiable information. It consisted of three sections. Section 1 gathered basic demographic information, distance traveled to the clinic, and access to technology. Section 2 assessed experience accessing mental health care in their home area. The questions covered: 1) who prompted the decision to seek help; 2) actions taken to find help, 3) treatment resources accessed; 4) satisfaction with local resources; and 5) knowledge of online treatment tools. Section 3 assessed interest in technology-based treatment resources. The questions covered: 1) interest in different approaches to use the app (for self-help, with a primary care provider, with a therapist, as a method to find a therapist); 2) interest in content (assessment, psychoeducation, treatment exercises); 3) interest in technology (website, mobile application, combined website and mobile, videos online, text online, traditional books or paper handouts); and 4) importance of endorsements (pediatrician, family or friends, trusted institution, professional organization). The level of satisfaction with local resources (addressed in section 2) and satisfaction, interest, or importance of technology-based treatment (addressed in section 3) were rated on a scale from 0 (not at all), to 6 (very). Survey questions are summarized in Tables 1 and 2, and the complete survey is available upon request.
Table 1.
Access to technology and mental health care.
| Survey Itema | Total (N = 33)b | Local Residents (n = 14) | Nonlocal (> 60 miles) Residents (n = 18) |
|---|---|---|---|
| Access to devices, No. (%) | |||
| A computer with internet access at home | 31 (93.9) | 12 (85.7) | 18 (100) |
| A smartphone that can access apps | 31 (93.9) | 13 (92.9) | 17 (94.4) |
| Wireless internet access at home | 31 (93.9) | 13 (92.9) | 18 (100) |
| Tablet | 27 (81.8) | 11 (78.6) | 16 (88.9) |
| Mobile device YOUR CHILD carries | 23 (69.7) | 6 (42.9) | 16 (88.9) |
| Influence decision to seek help, No. (%) | |||
| We as parents decided there was a problem | 28 (84.8) | 12 (85.7) | 15 (83.3) |
| Pediatrician recommended treatment | 9 (27.3) | 3 (21.4) | 6 (33.3) |
| School personnel recommended treatment | 4 (12.1) | 3 (21.4) | 1 (5.6) |
| Another specialist recommended treatment | 6 (18.2) | 3 (21.4) | 2 (11.1) |
| Actions taken when looking for help, No. (%) | |||
| Talk to my pediatrician or family doctor | 25 (75.8) | 10 (71.4) | 14 (77.8) |
| Search online for help | 17 (51.5) | 7 (50.0) | 9 (50.0) |
| Talk to friends and family | 10 (30.3) | 4 (28.6) | 6 (33.3) |
| Talk to school personnel | 9 (27.3) | 5 (35.7) | 4 (22.2) |
| Get a list from insurance company | 4 (12.1) | 0 (0) | 4 (22.2) |
| Local treatment resources, No. (%) | |||
| My child’s pediatrician | 17 (51.5) | 4 (28.6) | 12 (66.7) |
| Therapist SPECIALIZING in anxiety or OCD | 16 (48.5) | 9 (64.3) | 7 (38.9) |
| General therapist | 16 (48.5) | 4 (28.6) | 11 (61.1) |
| Child psychiatrist | 14 (42.4) | 6 (42.9) | 7 (38.9) |
| Satisfaction with local resources mean (SD)c | 3.06 (2.0) | 4.18 (1.5) | 3.06 (2.3) |
| Aware of websites or apps to treat anxiety or OCD, No. (%) | 4 (12.1) | 1 (7.1) | 3/17 (17.6) |
Items are abbreviated. Full survey is available upon request.
One family did not answer distance questions; their answers were excluded from the subgroup analysis.
Items were scored on a scale of 0–6 (0 = not at all; 6 = very).
Table 2.
Interest in technologya.
| Survey Itemb | Total, Mean (SD) | Local Residents, Mean (SD) | Nonlocal (> 60 miles) Residents, Mean (SD) |
|---|---|---|---|
| Interest in approaches to using technology | |||
| Use with mental health therapist | 4.25 (2.1) | 3.50 (2.5) | 4.82 (1.6) |
| Provides names of local doctors | 4.09 (2.2) | 3.71 (2.3) | 4.47 (2.1) |
| Use on your own | 3.61 (2.4) | 2.79 (2.2) | 4.22 (2.2) |
| Use with pediatrician or family doctor | 3.42 (2.1) | 2.86 (2.3) | 3.83 (1.9) |
| Interest in components of care | |||
| Information about treatment | 4.88 (1.6) | 4.71 (1.6) | 5.11 (1.6) |
| Program to complete treatment exercises | 4.76 (1.7) | 4.29 (1.7) | 5.17 (1.6) |
| Diagnostic test or questionnaire | 4.67 (1.8) | 4.43 (1.9) | 5.00 (1.7) |
| Interest in platform | |||
| Integrated computer and mobile device program | 4.55 (1.9) | 4.00 (2.1) | 5.06 (1.7) |
| Applications accessed via mobile device | 4.52 (1.8) | 4.07 (1.9) | 4.94 (1.7) |
| Websites accessed via computer | 4.39 (1.8) | 3.79 (2.0) | 4.94 (1.6) |
| Interest in format | |||
| Written information online | 4.55 (1.5) | 4.50 (1.7) | 4.67 (1.5) |
| Paper books or handouts | 4.27 (1.7) | 3.93 (1.7) | 4.61 (1.7) |
| Instructional videos online | 4.24 (1.7) | 3.71 (1.9) | 4.72 (1.5) |
| Importance of endorsements | |||
| Trusted institution (developed) | 5.03 (1.5) | 4.43 (1.9) | 5.44 (0.8) |
| Professional organization (recommended) | 4.33 (1.7) | 4.00 (1.9) | 4.50 (1.4) |
| Pediatrician or family doctor (recommended) | 4.16 (1.7) | 3.77 (2.0) | 4.33 (1.5) |
| Friends or family (recommended) | 3.16 (1.7) | 2.77 (2.0) | 3.61 (1.2) |
Items are abbreviated. Full survey is available upon request.
Items were scored on a scale of 0–6 (0 = not at all; 6 = very).
Planned analyses
The analysis of survey data was primarily descriptive, with frequencies and mean scores. To examine the preferences of families with less access to mental health resources, differences in responses between families living 60 miles or more and those living within 60 miles were compared through χ2 analyses and t tests. We predicted that parents that traveled from a distance to seek treatment, would be less satisfied with their local treatment options and thus, more interested in technological solutions. We used repeated measures analysis of variance (RMANOVA) to compare parent interest in different approaches (e.g., web vs mobile applications). Given the exploratory nature of the study, corrections were not made for multiple comparisons.
Results
Access to technology
Parent responses to the survey questions about access to technology are presented in Table 1. Parents reported almost universal access to technology in the home (i.e., computers, internet, wireless internet access, smartphones), with the majority (69.7%) reporting that their child had a mobile device. The responses from parents traveling from a distance endorsed similar, although slightly greater, access to technology at home than local parents, with only 1 item (child-specific mobile device) differing significantly between groups (local vs distance, 42.9% vs 88.9%; χ2 = 7.77; P = .005). In addition, all children older than 12 years (n = 16) had their own device, whereas fewer of those 12 years and younger had devices (n = 7 [41.2%]).
Treatment access
Parent responses to the survey questions about treatment access are presented in Table 1. Most parents described making the decision to seek treatment on their own, as opposed to following recommendations from a professional; this response did not differ by distance from the clinic. After making the decision to seek help, parents were most likely to report consulting with a pediatrician, followed by searching online, with less than half reporting consulting with their insurance company, friends or family, or school personnel. This pattern was similar between distance groups (no significant differences). Nearly all families (n = 31 [93.9%]) were able to identify at least 1 local resource, often a mental health resource (n = 26 [78.8%]). Parents traveling a greater distance to the clinic more commonly reported finding help from a pediatrician (66.7% vs 28.6%;χ2 = 4.57; P = .03). Parents reported being only somewhat satisfied with local services (m = 3.48 on a scale from 0 [not at all] to 6 [very]), with 35.3% providing a positive rating (5 or 6), 44.7% providing a moderate rating (2–4), and 20.6% providing a low rating (0 or 1). Parents who traveled at least 60 miles reported lower satisfaction with local services, but the difference was not statistically significant (3.06 vs 4.18; t (26) = 1.43; P = .16). Most parents were unaware of online resources or mobile applications to assist with treatment of childhood anxiety and OCD. Three of the 4 who indicated awareness of resources mentioned Anxiety Coach.
Interest in technology
Parent responses to the survey questions about interest in technology are presented in Table 2. Overall, parents expressed interest in technological tools, with 25 (75.8%) rating their interest as a 5 or 6 (6 being the highest possible rating) for at least 1 approach. The omnibus RMANOVA was significant (F (3,27) = 3.03; P = .047) and indicated significant differences in the level of interest, depending on the proposed approach (F (3,87) = 3.15; P = .03). Posthoc pairwise comparisons indicated that parents were more interested in using technology with a therapist than as a solitary self-help approach or with primary care, and they were more interested in obtaining assistance in finding specialists than in using it with primary care (P < .05 for all). Parents who lived more than an hour away from the clinic generally expressed more interest in all approaches compared with local parents (m = 4.29 vs 3.21), but the difference was not significant (F (1,29) = 2.83; P = .10). Among the parents living far from the clinic, 16 (88.9%) rated their interest in at least 1 approach as a 5 or 6. Parents expressed interest in technology to assist with each of 3 components of mental health care content (ie, assessment, psychoeducation, treatment), with no significant differences among content components or groups based on distance from the clinic.
Parents reported similar levels of interest in different delivery platforms and formats (e.g., computer online vs mobile application, or videos vs reading material). RMANOVAs were conducted separately for each technological platform (computer, mobile, integrated) and format (video, reading, paper). We observed no significant differences (P > .05 for all) among options or between distance groups on interest in various platforms (computer, mobile, integrated). Local parents preferred text-based online information over videos or paper materials, but the omnibus test (F (2,29) = 3.24; P = .05), the interaction (F (2,29) = 3.27; P = .05), and the main effect (F (2,60) = 2.05; P = .14) did not show significant differences. Parents reported different levels of importance for various endorsements (omnibus RMANOVA: F (3,27) = 10.65; P < .001; main effect: F (3,87) = 10.91; P < .001). Specifically, parents rated a technology being offered by a trusted institution as being more important than recommendations from a pediatrician, family or friends, or a professional organization (P < .05 for all). In addition, parents rated recommendations from family or friends as being less important than the other endorsements (P < .05 for all).
Discussion
Study 1 was conducted to gauge parent interest, readiness, and preferences for technology-based treatment of childhood anxiety disorders and OCD. Interest in technology-based solutions was positive, although parents preferred to use technology to facilitate working with a therapist, rather than to alleviate the need for face-to-face therapy sessions. Parents reported similar levels of interest in different delivery platforms (i.e., computer, mobile, integrated). The results were largely consistent with our hypotheses, with some notable exceptions. For example, the high frequency of youth (including younger children) possessing their own devices was unexpected, as was the degree to which parents reported making the decision to seek help on their own, absent impetus from health or school professionals, and the relatively low reported use of the internet compared with consulting a pediatrician when seeking help. The low rates of school involvement are consistent with parents’ low level of confidence in the school’s ability to meet children’s mental health needs (CS Mott Children’s Hospital, 2017). Parents also reported more success accessing mental health resources than we expected, although rates of satisfaction were predictably low.
Study 2: Prototype of an interactive psychoeducational component
Overview
Study 2 was conducted to examine the usability of an interactive psychoeducation component for a technology-based self-help treatment tool for childhood anxiety disorders and OCD. Children and parents interacted with the computer-based prototype and rated its usability. In addition, data from the patients’ entries were collected to examine the feasibility of accurately applying the cognitive behavioral conceptualization of anxiety and OCD to their symptoms. We hypothesized that 1) families would rate the prototype positively; and 2) most patients would be able to select items from the preset options in prototype (rather than entering them via free text) to capture their symptoms.
Methods
Participants
Ten children (40% male) ranging in age from 6 to 17 years (mean [SD] age, 13.6 [3.3] years) and their parents were recruited. All families had completed a treatment orientation session covering 1) psychoeducation about anxiety and OCD and its treatment; and 2) treatment planning (i.e., building an exposure hierarchy). The patients were at different stages of treatment; 9 had completed at least 1 exposure session and some were close to terminating treatment.
Procedure
Study participants were recruited from a child anxiety and OCD specialty clinic. Families were approached by a treating clinician after a therapy session and given a written description of the study. Those who expressed interest were contacted by the study coordinator to schedule a study appointment. During the appointment, the coordinator provided a written description of the study and reviewed it verbally, emphasizing the voluntary nature of participation. Oral consent was provided by the family. Patients and their parents then interacted with the computer-based prototype application; activities consisted of watching instructional videos and completing interactive worksheets. At the end of the session, the patient and parent answered questions about the helpfulness and user-friendliness of the prototype. No identifiable information was collected during the study. Families were provided $25 remuneration.
Materials
The prototype application consisted of a psychoeducation module that presented the cognitive behavioral model of anxiety and OCD and described its treatment through videos and interactive worksheets. The content was based on a program that has been developed and refined clinically in individual and mixed diagnostic (i.e., anxiety and OCD) group treatment, as well as examined empirically (Whiteside et al., 2015; Whiteside, Dammann, Tiede, Biggs, & Hillson Jensen, 2017). The adaptation of the content for the current prototype was reviewed and approved by three content experts for adherence to evidence-based treatment of anxiety and OCD. The module began with a video overview of the cognitive behavioral model explaining the role of avoidance in maintaining symptoms and the need for exposure to reduce those symptoms. This introduction was followed by 4 shorter videos highlighting each component of the model: 1) stimuli that set off the anxiety or OCD cycle, 2) expectations in response to the stimuli that lead to distress, 3) methods used to avoid feared items, and 4) how to replace avoidance with exposure.
An interactive worksheet followed each video and prompted the family to identify how each component related to the child’s symptoms; they could select items from expandable lists or add free-text entries. Specifically, after viewing a video about the different types of stimuli that set off anxiety or OCD, the family could select the stimuli that set off the child’s symptoms from a list beginning with 3 domains (ie, external, mental, bodily stimuli) that expanded to 13 general categories (eg, social situations, germs, worries, intrusive thoughts, panic symptoms, nausea or vomiting), which in turn expanded to 80 specific items (eg, talking to kids, germs from other people, worries about health and safety, thoughts of hurting someone, heart pounding, throwing up from eating). The 3 subsequent interactive worksheets (ie, expectations, avoidance, and exposure) provided options based on the general category of anxiety or OCD stimuli selected by the family on the first worksheet. A final video summarized how the child’s symptoms, as entered into the program, fit the cognitive behavioral model. For example, a family might identify social situations as a stimulus, fears of being laughed at as an expectation, resistance to joining activities as avoidance, and starting conversations as an exposure.
The study coordinator collected basic demographic information (ie, child age, sex, and phase in treatment). The prototype recorded all the responses entered in the interactive worksheets by the patient and parent regarding the child’s symptoms. Finally, the patients and parents separately completed a questionnaire designed to assess usability. Two items were from the modified System Usability Scale (Brooke, 2013): 1) “This product was easy to use” was rated on a 5-point scale (1 = strongly disagree to 5 = strongly agree); and 2) “Overall, I would rate the user friendliness of this product as _____” (1 = worst imaginable to 7 = best imaginable). Three additional questions asked parents and children to rate 1) the ability of the product to capture the child’s symptoms, 2) ease of finding answers that fit the child symptoms, and 3) degree the product helped them understand the child’s symptoms. Answers were based on a 5-point scale (1 = strongly disagree to 5 = strongly agree).
Results
Use data
Table 3 presents the mean number of entries and the mean number of entries per selected general fear category. The families selected a mean (SD) of 20.1 (13.9) feared items (range, 1–39) from a mean (SD) of 5.5 (3.0) categories (range, 1–10). Except for the 1 patient who selected a single fear, the remaining patients selected at least 5 fears from at least 2 categories. For each fear category selected, the families were able to identify an average of 2 to 3 beliefs, avoidance behaviors, and exposures exercises. Free-text entry for symptoms was rarely used, with only 1 feared item and 1 exposure item entered.
Table 3.
Prototype usability.
| Total Entries | |||
|---|---|---|---|
| Selected Item | Range | Mean (SD) | Average Entries Per Category, Mean (SD) |
| Fear categories | 1–10 | 5.50 (3.0) | … |
| Feared items | 1–39 | 20.10 (13.9) | 3.20 (1.2) |
| Beliefs | 4–29 | 14.60 (9.3) | 2.62 (0.8) |
| Feelings | 6–42 | 17.80 (12.9) | 3.44 (1.7) |
| Avoidance | 4–28 | 14.10 (8.4) | 2.86 (1.3) |
| Exposures | 3–29 | 12.60 (9.4) | 2.33 (1.1) |
User experience data
Table 4 presents average user-experience ratings. Overall, the user experience of the application was rated highly, with 8 of 10 patients and parents rating the prototype as “excellent” or “best imaginable” (the top 2 categories). The ability of the prototype to capture the child’s experience, ease of finding applicable items, help in understanding the child’s experience, and ease of use in general were all rated on average as “agree” to “strongly agree.” Overall, of the 100 user experience ratings (5 items completed by 10 children and 10 parents), 88 indicated the best 2 response options.
Table 4.
User experience rating of prototype application.
| Parent | Child | |
|---|---|---|
| Itema | Mean (SD) | Mean (SD) |
| User friendliness | 5.80 (0.8) | 6.10 (1.0) |
| Capture experience | 4.60 (0.5) | 4.20 (0.6) |
| Easy to apply | 4.40 (1.0) | 4.30 (0.8) |
| Help to understand | 4.10 (0.9) | 4.00 (1.1) |
| Easy to use | 5.00 (0.8) | 4.70 (0.5) |
| Totalb | 4.53 (0.5) | 4.30 (0.6) |
User friendliness was rated from 1–7; all other scales were rated from 1–5. Higher ratings indicate more satisfaction.
Average of items rated 1–5 (ie, does not include user friendliness).
Discussion
Study 2 was conducted to explore the feasibility, adequacy of content, and usability of a basic prototype application for delivering and individualizing psychoeducation to children with anxiety and OCD. Use data suggest that the prototype effectively captured patients’ reported anxiety and OCD symptoms. Overall, these data suggest that families were able to individualize the cognitive behavioral conceptualization of anxiety and OCD to their child’s experience. Moreover, the design of the module appears to be user friendly. Lower user experience ratings were infrequent and may have reflected methodologic issues. For example, some patients were close to the end of treatment and the module may not have provided new information for them (e.g., low rating on “help to understand”).
General discussion
The goals of the current study were to examine patient interest in and feasibility of a technological solution to providing evidence-based treatment for childhood anxiety disorders and OCD that accurately reflects the needs of families. Overall, the results of the two studies supported the initial design of Anxiety Coach expansion and underscored the potential for using technology to deliver treatment for childhood anxiety and OCD.
The results of both studies support the hypothesis underlying the development of a computer-based interface to complement the mobile Anxiety Coach application. Results suggest that electronic online treatment could meet an identified need, given parents’ relatively low satisfaction with local resources and difficulty with identifying a specialist in their area. In addition, the ubiquity of device ownership, openness to using technology in different settings, and interest across different media for access (mobile device, computer, both) suggest that families are able and willing to access the proposed integrated platform design. Families were open to participating in all components of care (assessment, education, intervention) online, supporting efforts to augment the currently existing treatment components with enhanced psychoeducation and assessment. Moreover, the families’ interaction with the prototype supported the feasibility of individualizing the cognitive behavioral conceptualization of anxiety to a patient’s specific symptoms. Not only did patients and parents positively rate the ability of the prototype to capture and explain the patient’s symptoms, their entries into the prototype suggested that the content was sufficient to cover the families’ presenting concerns. As such, families were able to use the program to apply key cognitive behavioral concepts to their child’s unique experience.
More broadly, the study has implications for the development of technology-based treatment programs. Device ownership, openness to various delivery platforms, and interest in various treatment components suggest that families would likely be willing to consider a range of approaches to delivering mental health care electronically. However, the low awareness of existing programs suggests that marketing and dissemination plans must accompany initial development and validation. Ideally, efforts to understand the needs and interests of the intended audience should be assessed and understood early in the development process, as we did in the current investigation. Given the frequency with which families report consulting their child’s physician when seeking help, pediatricians and family doctors may be an important access point for educating families about existing self-help programs. This approach is consistent with previous work identifying health providers as a high-potential avenue for dissemination (Brouwer et al., 2008). Another consideration when developing electronically-delivered treatments is finding the appropriate level of in-person support. Families were most interested in technology that would facilitate treatment with a therapist. This desire is consistent with the theory that supportive accountability improves adherence and the effects of electronic treatment (Mohr, Cuijpers, & Lehman, 2011), however it does not increase the ability to provide care beyond the available human resources. As such, electronic treatment tools are most likely to be used and have a positive effect if they facilitate some type of personal contact where it would otherwise not be available (i.e., distance treatment) or improve the efficacy of in-person sessions (i.e., guide the implementation of exposure therapy; Khanna & Kendall, 2010).
The studies described here have a number of important limitations. Primarily, both samples were selected to provide convenient and time-efficient data collection in a setting where direct clinical contact could address potential negative effects of technological malfunctions or misunderstanding of clinical content. As such, the samples are unrepresentative of the target population in multiple ways. To begin with, to limit the length and intrusiveness of the survey, minimal demographic and diagnostic information was collected. However, based on data available from Study 1, both samples likely resembled the specialty clinic’s patient population, which tends to be white, educated, and affluent. Moreover, the sample had sought and participated in treatment, and thus may have needs, preferences, and barriers that differ substantially from families that have not made contact with traditional mental healthcare. Finally, the families in the sample had a presumably positive relationship with the exposure therapy and the clinic that developed the application, which also could affect their responses. All of these biases limit the generalizability of the results. As such, the needs, resources, and interests of diverse populations, as well as the cultural acceptability of the prototype to different populations, need to be examined. Similarly, given the goal of a self-help online treatment tool to extend care to those without access to traditional care, the current findings from patients within a mental health clinic should be confirmed in families who have not connected with specialty care. Finally, although the samples were sufficient to demonstrate initial support for the hypotheses underlying the application’s design and the prototype’s feasibility, larger samples are needed to support generalizability of the findings.
Despite these limitations, the current results support the continued development of an Internet-based program to complement the existing mobile Anxiety Coach iOS application. Specifically, the success of the psychoeducation prototype suggests that it could be integrated with the treatment component of the existing mobile application. If successful, this program would allow patients to identify their feared stimuli, specify the outcomes they fear will occur, identify a general approach to replacing avoidance with exposure, select and order specific exposure tasks, and record their progress as they complete those tasks. It will also be important to continue to receive input from intended users, including those from underserved populations, throughout the development and refinement of additional psychoeducational components, treatment tools, and assessment measures. Development of an internet-based version of the application will facilitate delivery of more interactive psychoeducational material and provide a medium that can be used collaboratively with a therapist.
In summary, the present investigation assessed the needs and interests of potential users and examined the usability and functionality of an early prototype to guide the development of a self-help electronic tool for childhood anxiety disorders and OCD. Overall, the results were positive and supportive and inform continued development. In addition, the collected data suggest potential avenues for increasing the likelihood that other similar projects can be successful.
Funding
These projects were supported by a grant from the National Institute on Disability Independent Living, and Rehabilitation Research (90BI0035-01-00).
Role of the funding source
The funder had no role in study design, in the collection, analysis and interpretation of data, in the writing of the report, or in the decision to submit the article for publication.
Footnotes
Disclosure statement
Dr. Whiteside and Mayo Clinic receive royalties from Mayo Clinic Anxiety Coach. This research has been reviewed by the Mayo Clinic Conflict of Interest Review Board and is being conducted in compliance with Mayo Clinic conflict of interest policies.
References
- Bakker D, Kazantzis N, Rickwood D, & Rickard N (2016). Mental health smartphone apps: Review and evidence-based recommendations for future developments. JMIR Ment Health, 3(1), e7. doi: 10.2196/mental.4984 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Beesdo K, Knappe S, & Pine DS (2009). Anxiety and anxiety disorders in children and adolescents: Developmental issues and implications for DSM-V. Psychiatric Clinics of North America, 32(3), 483–524. doi: 10.1016/j.psc.2009.06.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Beidas RS, & Kendall PC (2010). Training therapists in evidence-based practice: A critical review of studies from a systems-contextual perspective. Clinical Psychology, 17(1), 1–30. doi: 10.1111/j.1468-2850.2009.01187.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Brooke J (2013). SUS: A Retrospective. Journal of Usability Studies, 8(2), 29–40. [Google Scholar]
- Brouwer W, Oenema A, Crutzen R, De Nooijer J, De Vries NK, & Brug J (2008). An exploration of factors related to dissemination of and exposure to internet-delivered behavior change interventions aimed at adults: A Delphi study approach. Journal of Medical Internet Research, 10(2), e10. doi: 10.2196/jmir.956 [DOI] [PMC free article] [PubMed] [Google Scholar]
- CS Mott Children’s Hospital. (2017). Handling child health needs during the school day. Mott Poll Report, 30(2), 2. Retrieved from https://mottpoll.org/sites/default/files/documents/091817_SchoolHealth.pdf. [Google Scholar]
- Ebert DD, Zarski AC, Christensen H, Stikkelbroek Y, Cuijpers P, Berking M, & Riper H (2015). Internet and computer-based cognitive behavioral therapy for anxiety and depression in youth: A meta-analysis of randomized controlled outcome trials. PloS One, 10(3), e0119895. doi: 10.1371/journal.pone.0119895 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ezpeleta L, Keeler G, Erkanli A, Costello EJ, & Angold A (2001). Epidemiology of psychiatric disability in childhood and adolescence. Journal of Child Psychology and Psychiatry and Allied Disciplines, 42(7), 901–914. doi: 10.1111/jcpp.2001.42.issue-7 [DOI] [PubMed] [Google Scholar]
- Greenberg PE, Sisitsky T, Kessler RC, Finkelstein SN, Berndt ER, Davidson JR, … Fyer AJ (1999). The economic burden of anxiety disorders in the 1990s. Journal of Clinical Psychiatry, 60(7), 427–435. [DOI] [PubMed] [Google Scholar]
- Gunter RW, & Whittal ML (2010). Dissemination of cognitive-behavioral treatments for anxiety disorders: Overcoming barriers and improving patient access. Clinical Psychology Review, 30(2), 194–202. doi: 10.1016/j.cpr.2009.11.001 [DOI] [PubMed] [Google Scholar]
- Herschell AD, Kolko DJ, Baumann BL, & Davis AC (2010). The role of therapist training in the implementation of psychosocial treatments: A review and critique with recommendations. Clinical Psychology Review, 30(4), 448–466. doi: 10.1016/j.cpr.2010.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kazdin AE, & Blase SL (2011). Rebooting psychotherapy research and practice to reduce the burden of mental illness. Perspectives on Psychological Science, 6(1), 21–37. doi: 10.1177/1745691610393527 [DOI] [PubMed] [Google Scholar]
- Kendall PC, Robin JA, Hedtke KA, Suveg C, Flannery-Schroeder E, & Gosch E (2005). Considering CBT with anxious youth? Think exposures. Cognitive and Behavioral Practice, 12(1), 136–148. doi: 10.1016/S1077-7229(05)80048-3 [DOI] [Google Scholar]
- Kertz SJ, MacLaren Kelly J, Stevens KT, Schrock M, &Danitz SB (2017). A review of free iPhone applications designed to target anxiety and worry. Journal of Techology in Behavioral Science, 2, 61–70. [Google Scholar]
- Khanna MS, & Kendall PC (2010). Computer-assisted cognitive behavioral therapy for child anxiety: Results of a randomized clinical trial. Journal of Consulting and Clinical Psychology, 78(5), 737–745. doi: 10.1037/a0019739 [DOI] [PubMed] [Google Scholar]
- Lepine JP (2002). The epidemiology of anxiety disorders: Prevalence and societal costs. Journal of Clinical Psychiatry, 63(Suppl 14), 4–8. [PubMed] [Google Scholar]
- Merikangas KR, He JP, Brody D, Fisher PW,Bourdon K, & Koretz DS (2010). Prevalence and treatment of mental disorders among US children in the 2001–2004 NHANES. Pediatrics, 125(1), 75–81. doi: 10.1542/peds.2008-2598 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Meyer JM, Farrell NR, Kemp JJ, Blakey SM, & Deacon BJ (2014). Why do clinicians exclude anxious clients from exposure therapy? Behaviour Research and Therapy, 54, 49–53. doi: 10.1016/j.brat.2014.01.004 [DOI] [PubMed] [Google Scholar]
- Mohr DC, Cuijpers P, & Lehman K (2011). Supportive accountability: A model for providing human support to enhance adherence to eHealth interventions. Journal of Medical Internet Research, 13(1), e30. doi: 10.2196/jmir.1602 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pew Research Center. (2018, Feb 5). Mobile fact sheet. Retrieved from http://www.pewinternet.org/fact-sheet/mobile/
- Rooksby M, Elouafkaoui P, Humphris G, Clarkson J, & Freeman R (2015). Internet-assisted delivery of cognitive behavioural therapy (CBT) for childhood anxiety: Systematic review and meta-analysis. Journal of Anxiety Disorders, 29, 83–92. doi: 10.1016/j.janxdis.2014.11.006 [DOI] [PubMed] [Google Scholar]
- Whiteside SP (2016). Mobile device-based applications for childhood anxiety disorders. Journal of Child and Adolescent Psychopharmacology, 26(3), 246–251. doi: 10.1089/cap.2015.0010 [DOI] [PubMed] [Google Scholar]
- Whiteside SP, Ale CM, Young B, Olsen MW, Biggs BK, Gregg MS, … Homan K (2016). The length of child anxiety treatment in a regional health system. Child Psychiatry and Human Development, 47(6), 985–992. doi: 10.1007/s10578-016-0628-5 [DOI] [PubMed] [Google Scholar]
- Whiteside SP, Deacon BJ, Benito K, & Stewart E (2016). Factors associated with practitioners’ use of exposure therapy for childhood anxiety disorders. Journal of Anxiety Disorders, 40, 29–36. doi: 10.1016/j.janxdis.2016.04.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Whiteside SPH, Ale CM, Young B, Dammann J, Tiede MS, & Biggs BK (2015). The feasibility of improving CBT for childhood anxiety disorders through a dismantling study. Behaviour Research and Therapy, 73, 83–89. doi: 10.1016/j.brat.2015.07.011 [DOI] [PubMed] [Google Scholar]
- Whiteside SPH, Dammann JE, Tiede MS, Biggs BK, & Hillson Jensen A (2017). Increasing availability of exposure therapy through intensive group treatment for childhood anxiety and OCD. Behav Modif, 145445517730831. doi: 10.1177/0145445517730831 [DOI] [PubMed] [Google Scholar]
- Whiteside SPH, Sattler A, Ale CM, Young B, Hillson Jensen A, Gregg MS, & Geske JR (2016). The use of exposure therapy for child anxiety disorders in a medical center. Professional Psychology, Research and Practice, 47(3), 206–214. doi: 10.1037/pro0000077 [DOI] [Google Scholar]