This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Formative Research, is properly cited. The complete bibliographic information, a link to the original publication on https://formative.jmir.org, as well as this copyright and license information must be included.
Depression has a profound impact on population health. Although using web-based mental health programs to prevent depression has been found to be effective in decreasing depression incidence, there are obstacles preventing their use, as reflected by the low rates of use and adherence.
The aims of the study are to understand the barriers to using web-based mental health programs for the prevention of depression and the possible dangers or concerns regarding the use of such programs.
BroMatters and HardHat were two randomized controlled trials (RCTs) that evaluated the effectiveness of e–mental health programs for preventing workplace depression. In the BroMatters RCT, only working men who were at high risk of having a major depressive episode were included. The participants were assigned to either the control group or 1 of 2 intervention groups. The control participants had access to the general depression information on the BroMatters website. Intervention group 1 had access to BroMatters and BroHealth—the depression prevention program. Intervention group 2 had access to BroMatters and BroHealth along with weekly access to a qualified coach through telephone calls. The HardHat trial targeted both men and women at high risk of having a major depressive episode. The participants in the intervention group were given access to the HardHat depression prevention program (which included a web-based coach), whereas HardHat access was only granted to the control group once the study was completed. This qualitative study recruited male participants from the intervention groups of the two RCTs. A total of 2 groups of participants were recruited from the BroMatters study (after a baseline interview: n=41; 1 month after the RCT: n=20; 61/744, 8.2%), and 1 group was recruited from the HardHat RCT 1 month after the initial quantitative interview (9/103, 8.7%). Semistructured interviews were performed with the participants (70/847, 8.3%) and analyzed using content analysis.
There were both personal and program-level barriers to program use. The three personal barriers included time, stress level, and the perception of depression prevention. Content, functionality, and dangers were the program-level barriers to the use of web-based mental health programs. Large amounts of text and functionality issues within the programs decreased participants’ engagement. The dangers associated with web-based mental health programs included privacy breaches and inadequate help for severe symptoms.
There are personal and program-level barriers to the use of web-based mental health programs. The stigmatization of help seeking for depression symptoms affects the time spent on the program, as does the public perception of depression. Certain barriers may be mitigated by program updates, whereas others may require a complete shift in the perception of depression prevention.
Major depressive episodes (MDEs) have a considerable impact on human health, with an annual prevalence of 4.7% in Canada [
Web-based mental health programs can play an important role in the prevention of depression by increasing accessibility, confidentiality, and sustainability [
The demographic characteristics of age and sex seem to influence the use of web-based depression prevention resources. Web-based mental health programs are also less likely to be perceived as helpful by older populations [
Although studies to date offer some insight into this topic, there are limited studies on the aversion to using web-based programs and examining other concerns or dangers that limit their use. The aims of this study are to (1) understand the barriers to using web-based mental health programs for the prevention of depression and (2) explore any possible dangers or concerns regarding the use of such programs.
This qualitative study was embedded in two RCTs that each examined a different web-based program to prevent mental illness (BroHealth and HardHat, as described below). This project was approved by the research ethics board of the Royal Mental Health Centre in Ottawa, Canada.
BroHealth is a web-based program that was evaluated in the BroMatters RCT. It aims to reduce the risk of depression among working men at high risk of having an MDE. Details of the BroMatters RCT can be found in a previous publication [
The participants in the BroMatters study were recruited using random digit dialing across Canada. These participants were working men at high risk of having an MDE. Their risk was calculated using a multivariable risk prediction (MVRP) algorithm, which is used to estimate the chances of developing depression in the next 4 years [
HardHat is an enhancement of BroHealth, targeting both men and women in the workplace. It encompasses enhanced visual design, depression information, self-assessment tools, and nine work-focused CBT and problem-solving therapy (PST) sessions designed by a psychiatrist with expertise in CBT and PST. Each session includes a 5-minute video recorded by professional voice actors in both English and French, in addition to in-class and/or homework assignments. Users are required to submit assignments for review and approval by a coach before moving on to the next session. Before finalization, HardHat was pilot-tested among 12 potential users from the community and revised based on feedback.
The target population of the HardHat RCT was working men and women who were at high risk of having an MDE but were not currently experiencing one. The risk of MDE was calculated using the same sex-specific MVRP algorithms [
Originally, the qualitative interviews were only planned to be conducted after the completion of the BroMatters RCT; however, because of low use throughout the duration of both RCTs, an additional round of interviews was conducted to investigate the reasons for the low use. The participants for this qualitative study were randomly selected from the intervention groups (ie, those given access to BroHealth or HardHat). A total of 3 different groups were recruited from the 3 different sample sets. Group 1 consisted of participants from the BroMatters RCT who had little or no use of BroHealth (maximum of one log-in). The interviews were conducted 1 month after the RCT began. Participants were randomly selected and interviewed until 41 participants were reached and code saturation was reached within these interviews. Group 2 was recruited after the BroMatters RCT was completed. Overall, 20 of these participants were randomly selected from either of the intervention groups and interviewed, and code saturation was reached within these interviews. The total population size for groups 1 and 2 was 744 individuals; thus, our use of 61 participants for qualitative data collection represents 8.2% (61/744) of the total population of participants in the BroMatters source study. Group 3 was recruited from the HardHat RCT 1 month after the study began; this study had a total of 103 participants to sample from. The participants were randomly selected if they had a maximum of one log-in, and they were interviewed until code saturation was reached. To maintain the homogeneity of the study sample for this analysis, we included nine interviews conducted with male participants from the HardHat trial, representing 8.7% (9/103) of the HardHat population.
The semistructured interview guides were designed by the team members; different interview guides were used for each group, but the questions included similar topics. The questions focused on the lack of program use, motivations, and program perception. The qualitative interview guides are included in
The principle of data saturation was used to determine the sample size for this qualitative study. Data saturation allows for a complete picture to be formed about participant perceptions on the topic. For this study, data saturation was defined as
Inductive content analysis was conducted using NVivo version 12 (QSR International) [
A total of 70 participants completed the semistructured telephone interviews—61 and 9 participants from the BroMatters and HardHat RCTs, respectively. The demographic characteristics of each group are presented in
A total of 6 barriers to using web-based mental health programs were described. These six barriers were categorized as personal-level barriers that revolved around the personality of the user and program-level barriers that were specific to the web-based program itself. The personal-level barriers included lack of time, level of stress, and disbelief in prevention, whereas the program-level barriers included content complexity and redundancy, program functionality, and perceived dangers (
Demographics for the participants in the different participant groups (N=70).
Demographics | All participants (N=70) | BroMatters low use (n=41) | BroMatters after RCTa (n=20) | HardHat, (n=9) | |||||
Age (years), mean (range) | 40.6 (20-67) | 39.5 (20-63) | 42.7 (27-66) | 43.0 (22-67) | |||||
|
|||||||||
|
British Columbia | 12 (17) | 7 (17) | 3 (15) | 2 (22) | ||||
|
Alberta | 8 (11) | 5 (12) | 3 (15) | 0 (0) | ||||
|
Saskatchewan | 4 (6) | 3 (7) | 1 (5) | 0 (0) | ||||
|
Manitoba | 1 (1) | 1 (2) | 0 (0) | 0 (0) | ||||
|
Ontario | 35 (50) | 21 (51) | 7 (35) | 7 (78) | ||||
|
Quebec | 5 (7) | 1 (2) | 4 (20) | 0 (0) | ||||
|
New Brunswick | 1 (1) | 0 (0) | 1 (5) | 0 (0) | ||||
|
Nova Scotia | 4 (6) | 3 (7) | 1 (5) | 0 (0) | ||||
Depression risk score (%), mean (range) | 23.1 (7-84) | 19.2 (7-84) | 20.4 (7-84) | 32.3 (22-67) | |||||
Number of log-ins, mean (range) | —b | 1.2 (0-11) | 2.3 (1-8) | — | |||||
|
|||||||||
|
Complete | — | — | — | 2 (22) | ||||
|
Partially | — | — | — | 3 (33) | ||||
|
Not started | — | — | — | 4 (44) |
aRCT: randomized controlled trial.
bNot available.
Lack of time
Life is busy.
People do not prioritize stress prevention.
Level of stress
Too little stress, and people do not think they need it
Too much stress, and people cannot find the time and motivation
Disbelief in prevention
Stigma
People do not focus on depression.
Content complexity and redundancy
Text heavy
People do not like activities in program
Redundant information
Program functionality
Disorganized flow
Broken links
Internet as a medium
Perceived dangers
Privacy
Using it as a treatment instead of seeking help
Most participants reported a lack of time or an inability to prioritize use as the largest barrier to program use. For instance, one participant said the following:
...it’s in my list of things to get to but life is like super busy, I have a young kid and a newish job and so it’s there, but I haven’t gotten to it.
Prioritizing other aspects of their lives often caused the participants to forget about the program entirely. A participant commented as follows:
...I keep forgetting. It’s in the pile of things to do and I keep forgetting to go back and look at it again.
The level of participant stress, both high and low, emerged as a barrier to program use. Some participants neglected to use the web-based programs because they believed that they did not need to use it:
I didn’t really need anything because work’s going pretty well.
Others managed their stress well using other support, including therapy, medication, and self-management strategies. For instance, one participant mentioned not using the program:
...because I currently have a psychologist...and I’ve been meeting with that doctor on a regular basis.
The participants may also have been disinclined to use the program because they were struggling to manage their mental health; hence, they felt overwhelmed and unable to avail themselves of the web-based program. For instance, one participant said the following:
...over the past month...my mental state of mind has deteriorated. I just got no determination or willpower to do anything.
These participants were often overwhelmed by the content when they tried to use the programs. One participant explained as follows:
I was already pretty stressed out and a little overwhelmed and had high levels of anxiety as a result, and so when I got to the website and saw the information that there was, I had to like pull back because it was too much.
In multiple cases, participants were skeptical as to whether anyone would use the program as a preventive measure before developing symptoms of depression:
I don’t think people are going to use the resource before they have depression...people don’t turn to something until they have it.
Program use as a preventive measure was also hindered by the stigma associated with getting mental health help. One participant stated:
There’s some stigma attached to the whole idea of not being mentally well, so I think it’s something that people like to keep private.
Content that is difficult to digest or redundant when considering the information available elsewhere is a barrier to program use. Some participants felt that the programs were too text heavy, thus deterring participation and causing them to lose interest in the program. One participant remarked the following:
I think what it comes down to is the amount of reading. If there is a lot of reading, it’s just kind of a put off.
Some participants felt that the modules were too long to complete in one sitting, and they neglected to either start the modules or return to complete them. In addition, some participants found the program content to be redundant when considering other resources that they had access to elsewhere, leading to the limited use of the programs. One participant explained this as follows:
...after a half a dozen times I found that the content was redundant with what I had already gotten from my healthcare provider.
For some participants, the types of activities did not align well with their preferences. For example, one participant said the following:
I’m not the type of person to divide a problem up into six component pieces.... My personality is a type where I don’t care about how many different parts of the problem there are.
Another topic discussed by the participants was how the program functioned. Functionality issues such as broken links and disorganized flow discouraged the use of the programs. When the participants had difficulty using the program, they felt discouraged and were less likely to return to it. The internet itself was also perceived to be problematic as a medium for the delivery of mental health prevention resources. Some participants did not like using the internet at all or did not use it often enough. Many participants also had an aversion to using the internet as a resource for mental health prevention. For instance, one participant said the following: “When you’re stressed out or something like that, last thing you want to do is go on your phone and tell your phone how stressed you are.” [BroMatters post-RCT; age: 47 years].
Overall, very few participants reported concerns about personal dangers or negative impacts of web-based mental health programs. A few of the participants voiced concerns about privacy. For example, one participant described how navigating contemporary times that are pervaded by telephone and email scams leaves them skeptical about the legitimacy of certain websites. For instance, a participant commented as follows: “...there’s so many traps for an old man to fall into” (BroMatters low use; age: 63 years). Furthermore, he provided additional context: “...99% of these emails and phone calls and people at my door are scammers wanting money” (BroMatters low use; age: 63 years). In addition, the programs’ focus on personal mental health left some participants concerned about privacy:
I’m a very private person and so the fact that I opened up as much as I did in the first call kind of concerned me, and so I’m hesitant just to spill my guts when I don’t really know who it is I’m talking to.
The other danger that the participants conveyed was the possibility that individuals with severe symptoms may use the program at the expense of specialized treatment and thus may not receive the care that they need. One participant explained this as follows:
...if somebody would try to use this to replace seeking other help or talking to someone else about their issues.
Most categories were consistent across the interview groups and programs. However, the participants who were interviewed only 1 month after the RCT started (both BroMatters and HardHat) reported
This study found that barriers to program use exist at both personal and program levels. Personal-level barriers affected initial use of the program. Many participants did not prioritize their mental health, citing that time constraints or a busy lifestyle did not allow them the time to use the program. This led many participants to believe that depression cannot be prevented. Although all participants in the study had a high risk of MDE, some did not believe that they had a stressful life and therefore did not believe that they needed to use the program. The barrier of continued use was discussed when the program itself was up for discussion. Content had a large effect on continued use. When content was perceived as redundant or not applicable to the participants, they discontinued use. Similarly, if program use caused frustration because of technical problems or the participants’ dislike of the internet as a medium, they were less likely to use the programs. Some participants were also concerned about the possibility of privacy breaches resulting from program use.
The results of our study are consistent with previous research in terms of personal and program-level barriers (lack of time, stress, content and medium, and privacy concerns) [
The design and development of BroHealth and HardHat were informed by a large national survey and several rounds of usability testing. Nevertheless, the use of these programs was not optimal. This qualitative study uncovered several barriers to web-based program use and has significant implications for mental health. Lack of time and perceived stress are the main personal barriers. It should be noted that the RCTs evaluating these two web-based programs were conducted from 2017 to 2019. The reported barriers reflect how potential users perceived web-based mental health programs during that time. Such barriers may be different in the context of COVID-19 with the closure of in-person health services [
Many participants believed that depression cannot be prevented. This mistaken view leads to a reluctance to use programs aimed at prevention, including web-based mental health programs. This finding is echoed in previous studies [
This study includes some limitations. First, some participants were interviewed a number of months after having used the program. As such, for some, recalling the program details was difficult. In addition, because demographic data were not collected, the results of this study cannot be extrapolated to larger populations. Furthermore, the barriers described in this study were specific to the BroHealth and HardHat programs; therefore, the barriers in this study, such as functionality, reminders, and content, cannot be generalized to other programs.
Although the use of web-based mental health programs to decrease the prevalence and incidence of depression has been shown quantitatively to be effective, there are barriers to their use. Barriers such as time and the perception of depression prevention need to be changed at the population level. Increasing the perceived importance and priority of depression prevention is likely to mitigate these barriers; therefore, research into these areas is imperative. Having easy-to-use programs with minimal text may improve engagement with web-based mental health programs, especially among those who may have had difficulty with previous attempts to use such programs.
Qualitative interview guides.
cognitive behavioral therapy
Green Shield Canada
major depressive episode
multivariable risk prediction
personal informatics
problem-solving therapy
randomized controlled trial
This study was supported by a grant from the Movember Foundation. The funder played no role in the study design and interpretation of the results.
All authors contributed to the study design, data interpretation, and the review and final approval of the manuscript.
None declared.