A smartphone application for treating depressive symptoms: study protocol for a randomised controlled trial

A smartphone application for treating depressive symptoms: study protocol for a randomised... Background: Depression is a commonly occurring disorder linked to diminished role functioning and quality of life. The development of treatments that overcome barriers to accessing treatment remains an important area of clinical research as most people delay or do not receive treatment at an appropriate time. The workplace is an ideal setting to roll-out an intervention, particularly given the substantial psychological benefits associated with remaining in the workforce. Mobile health (mhealth) interventions utilising smartphone applications (apps) offer novel solutions to disseminating evidence based programs, however few apps have undergone rigorous testing. The present study aims to evaluate the effectiveness of a smartphone app designed to treat depressive symptoms in workers. Methods: The present study is a multicentre randomised controlled trial (RCT), comparing the effectiveness of the intervention to that of an attention control. The primary outcome measured will be reduced depressive symptoms at 3 months. Secondary outcomes such as wellbeing and work performance will also be measured. Employees from a range of industries will be recruited via a mixture of targeted social media advertising and Industry partners. Participants will be included if they present with likely current depression at baseline. Following baseline assessment (administered within the app), participants will be randomised to receive one of two versions of the Headgear application: 1) Intervention (a 30-day mental health intervention focusing on behavioural activation and mindfulness), or 2) attention control app (mood monitoring for 30 days). Participants will be blinded to their allocation. Analyses will be conducted within an intention to treat framework using mixed modelling. Discussion: The results of this trial will provide valuable information about the effectiveness of mhealth interventions in the treatment of depressive symptoms in a workplace context. Trial registration: The current trial is registered with the Australian and New Zealand Clinical Trials Registry (ACTRN12617000547347, Registration date: 19/04/2017). Keywords: Depression, Workplace, Mhealth, Treatment Background difficulties in all aspects of their life, including work, Depression is now recognised as one of the leading home, and social activities [5]. In response to the causes of disability worldwide [1]. As well as being rela- increased identification and recognition of depression tively common [2], depression can also be very debilitat- [6], there has been greater emphasis on the development ing [3], with core symptoms inclusive of; an absence of of innovative and effective psychological treatments. positive affect, persistent low mood, and low activity [4]. The treatment of depression, both pharmacologically Persons with severe depressive symptoms report serious and with non-pharmacological interventions, remains a critical area of ongoing clinical research. There is substan- tial evidence supporting the efficacy of psychotherapy in * Correspondence: m.deady@unsw.edu.au the treatment of both sub-clinical [7] and clinical levels of Black Dog Institute; Faculty of Medicine, UNSW, Sydney, Australia Full list of author information is available at the end of the article depression [8]. Whilst cognitive behaviour therapy (CBT) © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Deady et al. BMC Psychiatry (2018) 18:166 Page 2 of 9 remains the most empirically supported treatment [9], been trialled [20] and even fewer have been specifically other therapies such as behavioural activation therapy tailored to a workplace setting. (BAT) has been shown to be equally effective and also less The proposed HeadGear trial aims to evaluate the complex to administer [10]. Service use statistics indicate effectiveness of a new smartphone app in treating that despite the effectiveness of such treatments, few depressive symptoms in a workplace context. people readily access these services [11]. Overcoming identified barriers to accessing psychotherapy such as cost, Methods convenience and accessibility [12] may enhance people’s Design ability to access evidence based treatments. The aim of the current study will be achieved through a Despite experiencing depression many people still re- multicentre randomized controlled trial, with two parallel main as active participants in the workforce [13]. Ongoing arms. The trial will compare two smartphone app-based in- employment is associated with numerous individual bene- terventions: a novel intervention app designed for the treat- fits such as economic, psychosocial, and emotional well- ment of depression (Headgear) and an attention control being [14]. Given its immediacy and core role in the lives app (mood monitoring). Assessments will occur at baseline, of individuals, the workplace may be an avenue for provid- post-intervention (5-weeks), and at 3-month follow-up. ing psychological interventions to those who might not The study is registered with the Australian New Zealand otherwise access mental health services in a conventional Clinical Trials Registry (ACTRN12617000547347) and has manner [15]. Interventions implemented in the workplace ethical approval from the University of New South Wales have been found to be effective in improving both mental (UNSW) Human Research Ethics Committee (HC17021). health and occupational outcomes [16]. This highlights an Consent to participate in the trial will be obtained electron- added benefit of workplace interventions as medical inter- ically from all participants. The study will be conducted in ventions in isolation have not shown as positive an effect accordance with the Helsinki Declaration [25]and is com- on work-related outcomes when compared to workplace pliant with the CONSORT guidelines [26]. interventions [17]. To date, most workplace interventions face accessibility and scalability issues, thus limiting the Setting and participants opportunity for employees to access these services. This The study will recruit Australians who are currently issue appears to be particularly prevalent in workers employed, and will sample more selectively from a range employed in industries where roles are mobile and/or iso- of male-dominated industries [27]. In Australia, these lated; work hours are often intermittent or excessive; and industries include agriculture/forestry/fishing, utility ser- tasks may be repetitive and high-risk (e.g. such as con- vices (electricity, gas, water and waste), wholesale trade, struction, transport and mining). These industries tend to manufacturing, transport/postal/warehousing, mining, be male-dominated (males > 70% of workforce) [18]and and construction [28]. Emergency services and defence are associated with higher rates of mental health concerns also fit this definition, but were not considered unique [18]. Technological innovations, namely smartphones, industries by the ABS, though for this study they will provide an opportunity for individuals in such roles to also be considered as male-dominated. learn how to manage their wellbeing and seek further sup- Industry partner organisations will assist with recruit- port if required [19]. ment by promoting the study among specific groups or Mobile health (mhealth) technologies are increasingly their entire workforce. We will aim to recruit at least being recognised as an effective means through which 850 employed adults across Australia. Organisations that mental health interventions can be disseminated in the elected to participate will promote the study via their population [20, 21]. Such interventions overcome numer- respective health and wellbeing officers, along with email ous barriers associated with treatment seeking, including and newsletter advertisements. The study will also be stigma, cost, and accessibility [20]. Encouragingly, there is promoted via members of the research team presenting preliminary support that mhealth interventions can effect- at partner worksites. Social media advertising targeted at ively treat symptoms in adults with depression [22]. More- employed people will also be utilised to recruit individ- over, mhealth interventions offer the opportunity for an uals employed externally of partner organisations using autonomous, user-directed approach that motivates and an evidence-informed advertising approach [29]. Both offers personalised support by allowing the delivery of males and females will be recruited. content to be tailored to an individual’s interests and needs [23]. This may be particularly advantageous to Eligibility criteria workplace contexts where, individualised interventions Initial eligibility criteria are: having a valid telephone num- are associated with more consistent outcomes than organ- ber, ownership of an Apple/Android-operating smartphone, isational wide interventions [24]. However, of the many fluent in English and living in Australia. As this trial is mhealth interventions available on the market, few have focused on the treatment of depression, participants will Deady et al. BMC Psychiatry (2018) 18:166 Page 3 of 9 also be excluded from this trial (although still permitted to virtually identical look and ‘feel’ as the intervention use the app) if they do not have substantial levels of depres- version of Headgear and is accessed in the same man- sive symptoms at baseline, as indicated by either a PHQ-9 ner. However, there is no skill development and no score below 15, or the PHQ-9’s algorithm for a diagnosis component of behavioural activation or mindfulness for Major Depressive Disorder (MDD) not being met [30]. therapy. To control for the attentional component of the Participants were excluded if they were under 18 or not HeadGear application, the control condition will en- currently employed. No exclusion criteria were in place courage users to use the inbuilt mood monitor daily over regarding comorbidities or medication use. a 30-day period and users will also have access to the ‘risk calculator.’ Interventions Active intervention: HeadGear Procedure The intervention condition, HeadGear, is a smartphone All interested users will be directed to their respective application-based intervention utilising Behavioural Acti- app store (iTunes or Google Play) directly via a dedi- vation Therapy (BAT) and mindfulness-based therapies. cated website (headgear.org.au) where participants regis- Behavioural Action is a therapy based on learning theory ter and provide their phone number. Informed consent that reconnects people to an environment of positive will be sought digitally via both the website and the app reinforcement, incorporating elements of value driven ac- itself. This provides information around the study aims, tion and goal setting [31, 32]. When delivered face-to-face, risks and benefits, confidentiality and dissemination of BAT has been shown to perform as well as CBT for the results. After consent and app download participants treatment of depression [33] and preliminary evidence will undergo initial screening. Participants who meet the that this translates in mhealth form [34]. The other com- inclusion criteria will then be randomised to receive ei- ponent utilised in Headgear, Mindfulness, draws on medi- ther the full HeadGear app or the attention-matched tation practices to allow individuals to gain further insight control version of the app. Participants will be blinded into their emotional, physical, and/or cognitive experience to their allocation. All participants will be provided with to ultimately shape it [35]. Mindfulness has been identified appropriate referral information to health services and as a transtherapeutic process that targets transdiagnostic crisis lines and will be encouraged to seek help from mental processes [36], as evidenced by its effect in treating their GP (if this has not already occurred) while com- a range of psychopathologies including but not limited to pleting the trial. The flow of participants through the mood, anxiety and substance use disorders [37, 38]. study phases is shown in Fig.1. The therapeutic component of HeadGear encourages the user to complete one ‘challenge’ each day (5–10 min Random allocation per day), over 30 days. These ‘challenges’ incorporate a Randomisation will occur immediately following comple- variety of evidence-based BAT and mindfulness tech- tion of the baseline assessment using automated proce- niques and skills, including psychoeducational videos, dures integrated into the trial management software. The value-driven activity planning and goal-setting, practice algorithm for randomisation will consist of a block design, exercises, and techniques for developing coping and stratified by industry type, with a block size of 10. resilience (e.g., problem solving, improving sleep, mini- mising alcohol use, and/or assertiveness training). Other Assessment components of the Headgear intervention app include Administration of assessments mood monitoring, a skill ‘toolbox’ (progressively built as Assessments will be completed at baseline, post-intervention the skills are completed), and a technical service help- (5-weeks), and 3-month follow-up. Baseline assessment line. Steps have been taken to promote user motivation includes outcome measures pertaining to depression symp- and engagement by incorporating a ‘daily challenge’ tomatology (Patient Health Questionnaire (PHQ-9) [44]), framework where the user is ‘rewarded’ (through skill wellbeing (World Health Organisation-5 (WHO-5) tokens) upon completing each daily challenge. A chal- Well-Being Index [45, 46]), anxiety symptomatology (Gen- lenge framework has been shown to increase the general eral Anxiety Disorder-2 item (GAD-2) [47], resilience (Con- appeal of an app [39–41]. In addition to these elements, nor-Davidson Resilience Scale 10-item (CD-RISC10) [48]), the Headgear application has been designed using par- work performance and absenteeism (Health and Work ticipatory approaches and iterative human-computer Performance Questionnaire (HPQ) [49]). In addition to interaction design strategies [42, 43]. these measures, demographic and service use information will be collected. The application monitors usage data in- Attention-matched control cluding number of log-ins, frequency of use, time spent The attention-matched control condition is a smart- in-app, and activity completion rates. This data will be used phone application that will have the same name and a to examine program engagement. Deady et al. BMC Psychiatry (2018) 18:166 Page 4 of 9 Fig. 1 Flow of participants through the trial Post-intervention assessment will occur at 5 weeks’ receive up to three text messages and a call at each post-baseline, to allow users extra time to complete the follow-up assessment time point, linking them to an online 30-day program. Participants will be reminded to complete platform through which they can complete the assessment. the post-intervention assessment via text messages to On completion of each assessment, participants will be phone numbers provided in order to address for the possi- entered into a draw to win one of four $200 Visa gift cards. bility that users may delete the app during the trial. Partici- pants will complete an online questionnaire similar to the Specific measures used in online assessments baseline measure (see Table 1), and again at 3 months The PHQ-9 will be used to measure depression symptom- post-baseline (“3-month follow-up”). Participants will atology [50]. The PHQ-9 is a reliable and valid nine-item Table 1 Assessment measures Baseline Post-intervention 3-month follow-up Demographics × Patient Health Questionnaire-9 item (PHQ-9) [44]× × × General Anxiety Disorder-2 item (GAD-2) [47]× × × World Health Organisation-5 (WHO-5) Well-Being Index [45, 46]× × × Connor-Davidson Resilience Scale 10-item (CD-RISC10) [48]× × × Health and Work Performance Questionnaire (HPQ) [49]× × × Service utilisation and management items × × × Program feedback × Deady et al. BMC Psychiatry (2018) 18:166 Page 5 of 9 measure of depression severity over the past 2 weeks [44, where 0 indicates the worst possible quality of life while 51]. Each of the nine items of the PHQ-9 is scored as 0 a score of 25 represents the best possible quality of life. (not at all), 1 (several days), 2 (more than half the days), or A score ≤ 13 or an answer of 0 or 1 on any of the five 3 (nearly every day). Scores are summed to provide a total items shows poor wellbeing. WHO-5 is a psychometric- score (score range 0–27 with 0 indicating no depressive ally sound measure of well-being with high internal symptoms and 27 indicating all symptoms occurring consistency (Cronbach’s α = 0.84) and convergent associ- nearly daily). The criterion and construct validity of the ations with other measures of well-being [55]. PHQ-9 has previously been demonstrated, with 73% sensi- Work Performance will be measured using three items tivity and 98% specificity in detecting major depression (performance items A10, A11, A12) from the Health and compared to clinician-based assessment [28, 50]and, Work Performance Questionnaire (HPQ) [49]) and two regardless of diagnostic status, typically represents clinic- additional items pertaining to: i) sickness absence over ally significant depression [50]. The measure has excellent the past month (days absent more generally, and days internal consistency (Cronbach α > 0.85 in multiple sam- absent specifically for mental health reasons), and ii) ples) and test-retest reliability (α =0.84) [52]. weeklong sickness absence over the past 6-months Anxiety will be measured using the 2-item Generalised (weeks absent more generally, and weeks absent specific- Anxiety Disorder scale (GAD-2) [47]. The GAD-2 ally for mental health reasons). consists of the two core criteria for generalized anxiety Service use and management items comprised of seven disorder, which have been shown to be effective screen- items assessing lifetime and past month service use, ing items for panic, social anxiety, and post-traumatic along with current medication use. Participants were stress disorders [47]. The GAD-2 begins with the stem also asked about their abilities (perceived capability and question: “Over the last 2 weeks, how often have you effectiveness) to manage their mental fitness, and auton- been bothered by the following problems?” Response omy (choice and freedom) in management. These were options are “not at all”, “several days”, “more than half scored on a 7-point Likert scale from strongly disagree the days”, and “nearly every day”, scored as 0, 1, 2, and to strongly agree. 3, respectively (total ranging from 0 to 6). A total scale score ≥ 3 is suggested as a cut-off point between the nor- Safety protocol mal range and “probable anxiety” [47]. In any trial concerned with mental health, there is the po- Resilience will be measured by the 10 item tential for participants to experience psychological dis- Connor-Davidson Resilience Scale (CD-RISC-10) [53]. tress. Those who meet criteria for the trial will within the The CD-RISC-10 is a self-rated measure, with each app trigger the user to be directed to a “get support” page question rated on a 5-point scale from 0 (‘not true at (at each assessment point) and will suggest the participant all’)to4(‘true nearly all the time’). The CD-RISC-10 seek further help from these support services or their gen- has been shown to differentiate between individuals eral practitioner (GP). Additionally, an optional call-back who function well after adversity and those who do not service for individuals requiring further support or direc- and measures the core features of resilience and the tion is provided. This call-back will be conducted by a ability to tolerate experiences [53]. It is believed that team-member with psychology training, within 4 days, increased resilience may reduce rates of mental ill with the purpose to guide participants into necessary care health [54].The scale demonstrates high internal arrangements. If the team member still has concerns for consistency (Cronbach’s α = 0.89), construct validity, the participant’s safety, an accredited psychiatrist will con- and test-retest reliability in the general population and tact the participants within the next 24 working hours. in clinical settings [48]. Total scores range from 0 to 40 Participants will also receive an SMS with a range of sup- with higher scores corresponding to greater resilience. port service contacts, and another reminder to consult The scale has been shown to have good concurrent with their GP regarding their mental health. validity, with higher resilience on the scale associated with lower levels of perceived stress [48] Validity is high Study hypotheses and outcomes relative to other measures and reflects differentiation in We hypothesise that participants receiving the HeadGear resilience among diverse populations, showing that intervention will have reduced levels of depression symp- higher levels of resilience are consistent with lower tomatology at post-intervention and 3-month follow-up, levels of perceived stress vulnerability [48]. compared to participants in the attention-matched control Wellbeing will be assessed using the 5-item WHO condition. While the primary analyses will be conducted on Wellbeing Index (WHO-5) [45, 46]. Participants are the entire sample (to examine the intervention effect). We asked to self-report on the presence or absence of well- also predict the intervention effect to vary according to the being on a 6-point scale ranging from 5 (‘all of the time’) level of depression symptoms at baseline [56], with a greater to 0 (‘none of the time.’) Raw scores range from 0 to 25 effect amongst those with higher levels of symptomatology. Deady et al. BMC Psychiatry (2018) 18:166 Page 6 of 9 Secondly, it is hypothesized that—relative to the control symptom levels, level of functional improvement and re- group—HeadGear Intervention participants will have cruitment method at baseline will be explored using an lower rates of depressive disorder as detected by the PHQ analysis of covariance approach using baseline measures as algorithm at all follow-up time points. We also hypothe- a covariate and including a covariate by intervention arm sise that the intervention group will have reduced levels of interaction term in models. The effectiveness of the active anxiety symptomatology, and improved wellbeing, resili- intervention at clinically relevant levels of baseline covari- ence, and work performance, at all follow-up time points, ates will be assessed using planned comparisons while the relative to controls. lowest values of covariates associated with a significant benefit of the intervention will be established using a Primary outcome Johnson and Neyman [57]approach. The primary outcome measure of the study will be the All tests of treatment effects will be conducted using a level of depressive symptomatology (as measured by the two-sided alpha level of 0.05 and 95% confidence intervals. PHQ-9) at the 3-month follow up period. Sample size Secondary outcomes As a treatment, the size of the effect of the intervention A range of secondary outcomes will be considered, includ- is anticipated to be moderate. Meta-analysis of previous ing change in anxiety symptomatology (as measured by trials of internet and mobile based treatment of depres- the PHQ-2) at both 5-week and 3-month follow up. Other sion showed a large effect size of g = − 0.90 [22]; how- secondary outcomes include incident caseness of depres- ever unguided interventions typically show smaller effect sion at 5-week and 3-month follow-up (as measured by sizes. Power calculations were carried out using the R the PHQ-9 diagnostic algorithm), and the level of depres- package simR [58]. Power was set at 80%, alpha at 0.05, sion symptoms (PHQ-9) at 5-week follow up. Finally, 2-tailed tests, and a correlation of .50 between pre- and change in Wellbeing (as measured by the WHO-5) and post- intervention scores was assumed. Based on these change in occupational functioning (as measured by the calculations, a sample size of 266 per group was needed HPQ and sickness absence questions) at both 5-week and (total N = 532). A conservative dropout rate of 40% at 3-month follow-ups will be outcomes of interest. follow-up was estimated. An initial sample of 851 will therefore be recruited for the trial. Statistical analysis Analysis plan Dissemination Primary analyses will be undertaken on an intent-to-treat Results of this study will be disseminated for publication basis, including all participants as randomised, regardless in peer-reviewed journals and key findings presented at of treatment received or withdrawal from the study. Like- national and international conferences. lihood based methods (mixed-model repeated measures (MMRM)) methods will be favoured to analyse change in Discussion the primary outcome measure (PHQ-9). A priori planned This study is planned to be the largest randomised con- comparisons of change from baseline across the 3-month trolled trial of a smartphone intervention for depression. follow up period will be used to test the primary hypoth- By targeting the workers, this trial will provide valuable esis. An unstructured variance-covariance matrix will be evidence regarding the effectiveness of mhealth interven- used to accommodate relationships between observations tions in treating depressive symptoms in a workplace at different occasions. Variables found to be substantially context. Given the substantial impact that depression imbalanced between groups post randomisation will be has on the individual and the employer, if shown to be tentatively included in these models and retained if effective, this program would allow for a simple and eco- statistically significant and influential on outcomes. Simi- nomical means by which an organisations or govern- lar analyses of scaled secondary measures will assess ments could disseminate a tailored intervention for differential change due to intervention arm. Mathematical workers. Given the proliferation of untested smartphone transformation or categorisation of raw scores may be applications, the dissemination of evidence based prod- undertaken to meet distributional assumptions and ad- ucts into the workplace, and indeed the wider commu- dress any violation of assumptions attributable to outliers. nity, remains a pressing need. Baseline characteristics will be used to define subgroups Employees with significant depressive symptoms have that would be the targeted if the app were offered as treat- higher rates of absenteeism, presenteeism and job turnover ment. Group membership will be used for models to evalu- [59]. Remaining in the workforce is important as it offers ate moderation of effectiveness by adding appropriate structure, empowerment, financial security and protection interaction terms and undertaking planned comparisons. from the psychological impacts of unemployment [14]. The effect on outcome of level of baseline depressive Using the workplace as a means to dispense or promote an Deady et al. BMC Psychiatry (2018) 18:166 Page 7 of 9 intervention may not only be protective for the individual By using both methods, we are confident that participants concerned, but may also assist in overcoming the issue of without significant symptomatology will be excluded and individuals delaying or not receiving treatment [60]. From those with significant symptomatology will be included. an economic standpoint, it is has been established that, for The treatment of depression utilising evidence based the workplace, such an approach is more cost-effective for mhealth interventions remains an important area of clinical the organisation [61]: utilising smartphone technologies research. The Headgear trial will be the largest trial of a would improve upon this cost-effectiveness. It is also hoped smartphone application that seeks to offer an alternate or that dissemination via workplaces and social media will augmentation to traditional face-to-face therapy through help engage individuals who would not usually access help which working adults can manage their mental health and via the health care system. wellbeing. This trial will be unique in that it is advertising a This study will provide valuable evidence regarding the treatment via a workplace setting, and allowing for the effectiveness of mhealth tools in the treatment of depres- assessment of clinical and occupational outcomes. Finally, sive symptoms. Despite the proliferation of mental health the Headgear Trial will provide much needed information apps, there is scarce research on the effectiveness of such on the general effectiveness of evidence-based interventions apps. Indeed, in a systematic review of the literature on (BAT and mindfulness) delivered through smartphone smartphone interventions, only five mental health apps technology. were empirically tested and only one of these did not re- Funding quire the input from a mental health professional [20]. This study was developed in partnership with beyond blue with donations mHealth interventions offer advantages in that they in- from the Movember Foundation. RAC is funded by an Australian Research Council Future Fellowship FT140100824. SBH and MD are supported by crease user autonomy and anonymity, which may be im- funding from iCare and NSW Health. portant as stigma [62] and lack of knowledge of services [63] can impact upon help-seeking behaviour in the work- Availability of data and materials The anonymized datasets generated and analysed during the current study place. An intervention developed for the workplace also will not be publicly available due to legal and ethical restrictions. carries the benefit in that it could ameliorate some of the financial burden placed solely on the healthcare system. Dissemination The current trial has received ethics approval from the University of New The proposed trial does carry with it a number of limi- South Wales Human Research Ethics Committee (HC17021). Research tations. The use of a smartphone app as a delivery mo- findings will be disseminated via peer reviewed journals, conferences and dality does mean that the intervention is unguided and internal reports. that the user is responsible for managing their inter- Authors’ contributions action with the program. Thus, trial attrition and disen- All of the other authors contributed to reviews of the manuscript and the gagement are expected issues [64, 65]. It is worth original research design and the analysis of data. All authors read and approved the final manuscript. noting, however, that this has also been an issue for face-to-face psychotherapy trials [66]. The reason for Ethics approval and consent to participate drop out in both mhealth and face-to-face trials is often The proposed trial has received ethical approval from the University of New South Wales (UNSW) Human Research Ethics Committee (HC17021). multi-faceted, and whilst can be related to engagement Prior to study participation all participants viewed and consented to the with the program, it is rarely only due to dissatisfaction information that was provided in the Participant Information Sheet. [67].To account for potential drop out, two procedures Competing interests were put into place. Firstly, all follow-up communication All researchers have remained independent from the funders in the would occur via phone numbers to ensure that if the completion and submission of this work. All authors have no conflict of participant uninstalled the app before the follow-up interest or competing interests to declare, except that the intellectual property of the smartphone application is jointly owned. period, the participant could still be reached. Secondly, conservative drop-out estimation and the use of statis- Publisher’sNote tical methods robust to data missing at random, it is Springer Nature remains neutral with regard to jurisdictional claims in believed that this limitation will be minimised. published maps and institutional affiliations. Another limitation of the present trial is related to the reli- Author details ance on self-reported depressive symptoms, rather than a 1 2 Black Dog Institute; Faculty of Medicine, UNSW, Sydney, Australia. School of diagnosis of depression achieved through a structured diag- 3 Psychiatry, UNSW Sydney, Sydney, Australia. Central Clinical School, Brain nostic interview. This is a common issue faced by most and Mind Centre, Faculty of Medicine and Health, University of Sydney, Sydney, Australia. School of Electrical and Information Engineering, similar trials given the constraints around time and re- University of Sydney, Sydney, NSW 2006, Australia. School of Systems sources. To overcome this issue, we will use a well-validated Management and Leadership, Faculty of Engineering and IT, University of measure (PHQ-9) that contains two methods for classifying Technology Sydney, Sydney, Australia. Department of Mental Health and Suicide, Norwegian Institute of Public Health, Oslo, Norway. School of depression: 1) threshold total score above 14 (sensitivity = Psychology, UNSW Sydney, Sydney, Australia. MRC Cognition and Brain 67%; specificity = 95%) [68] or 2) meeting the depression 9 Sciences Unit, University of Cambridge, Cambridge, UK. Department of algorithm’s criteria (sensitivity = 0.53; specificity = 0.94) [69]. Community Medicine, University of Tromsø, Tromsø, Norway. Centre for Deady et al. BMC Psychiatry (2018) 18:166 Page 8 of 9 Work and Mental Health, Nordland Hospital Trust, Bodø, Norway. Centre for 28. Kroenke K, Spitzer RL. The PHQ-9: a new depression diagnostic and severity Research and Education in Forensic Psychiatry and Psychology, Haukeland measure. Psychiatr Ann. 2002;32(9):509–15. University Hospital, Bergen, Norway. 29. Choi I, et al. Using different Facebook advertisements to recruit men for an online mental health study: engagement and selection bias. Internet Received: 4 December 2017 Accepted: 17 May 2018 Interventions. 2017;8:27–34. 30. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–13. 31. Hopko DR, et al. Contemporary behavioral activation treatments for depression: References procedures, principles, and progress. Clin Psychol Rev. 2003;23(5):699–717. 1. Organization, W.H., Depression and other common mental disorders: global 32. Lejuez CW, et al. Ten year revision of the brief behavioral activation treatment health estimates. 2017. for depression: revised treatment manual. Behav Modif. 2011;35(2):111–61. 2. Slade T, et al. 2007 National Survey of mental health and wellbeing: 33. Richards DA, et al. Cost and outcome of Behavioural activation versus methods and key findings. Aust N Z J Psychiatry. 2009;43(7):594–605. cognitive Behavioural therapy for depression (COBRA): a randomised, 3. Harvey SB, et al. Depression and work performance: an ecological study controlled, non-inferiority trial. Lancet. 388(10047):871–80. using web-based screening. Occup Med (Lond). 2011;61(3):209–11. 34. Ly KH, et al. Behavioural activation versus mindfulness-based guided self- 4. American Psychiatric Association. Diagnostic and Statistical Manual of help treatment administered through a smartphone application: a Mental Disorders (DSM-5). 5th ed. Washington, DC: APA; 2013. randomised controlled trial. BMJ Open. 2014;4(1):e003440. 5. Kessler RC, et al. The epidemiology of major depressive disorder: results from the 35. Brown KW, Ryan RM, Creswell JD. Mindfulness: theoretical foundations and National Comorbidity Survey Replication (NCS-R). Jama. 2003;289(23):3095–105. evidence for its salutary effects. Psychol Inq. 2007;18(4):211–37. 6. Harvey SB, et al. Is the prevalence of mental illness increasing in Australia? 36. Greeson J, Garland EL, Black D. Mindfulness: a transtherapeutic approach for Evidence from national health surveys and administrative data, 2001–2014. transdiagnostic mental processes. Wiley Blackwell handbook of mindfulness. Med J Aust. 2017;206(11):490–3. 2014:531–62. 7. Cuijpers P, et al. Psychotherapy for subclinical depression: meta-analysis. Br J 37. Hofmann SG, et al. The effect of mindfulness-based therapy on anxiety and Psychiatry. 2014;205(4):268–74. depression: a meta-analytic review. J Consult Clin Psychol. 2010;78(2):169. 8. Cuijpers P, et al. Psychotherapy for depression in adults: a meta-analysis of 38. Zgierska A, et al. Mindfulness meditation for substance use disorders: a comparative outcome studies. J Consult Clin Psychol. 2008;76(6):909–22. systematic review. Subst Abus. 2009;30(4):266–94. 9. Butler AC, et al. The empirical status of cognitive-behavioral therapy: a 39. Titov N, et al. Improving adherence and clinical outcomes in self-guided review of meta-analyses. Clin Psychol Rev. 2006;26(1):17–31. internet treatment for anxiety and depression: randomised controlled trial. 10. Richards DA, et al. Cost and outcome of Behavioural activation versus PLoS One. 2013;8(7):e62873. cognitive Behavioural therapy for depression (COBRA): a randomised, 40. Matthews M, et al. Designing mobile applications to support mental health controlled, non-inferiority trial. Lancet. 2016;388(10047):871–80. interventions. In: Jo L, editor. Handbook of Research on User Interface 11. Olfson M, Marcus SC. National trends in outpatient psychotherapy. Am J Design and Evaluation for Mobile Technology. London: IGI, Information Psychiatr. 2010;167(12):1456–63. Science Reference; 2008. p. 635–56. 12. Mohr DC, et al. Barriers to psychotherapy among depressed and 41. Bakker D, et al. Mental health smartphone apps: review and evidence-based nondepressed primary care patients. Ann Behav Med. 2006;32(3):254–8. recommendations for future developments. JMIR Mental Health. 2016;3(1):e7. 13. Sanderson K, Andrews G. Common mental disorders in the workforce: 42. Deady M, et al. Designing smartphone mental health applications for recent findings from descriptive and social epidemiology. Can J Psychiatry. emergency service workers. Occup Med. 2017; 2006;51(2):63–75. 43. Calvo RA, Peters D. Positive computing: technology for wellbeing and 14. Modini M, et al. The mental health benefits of employment: results of a human potential. Cambridge: MIT Press; 2014. systematic meta-review. Australas Psychiatry. 2016;24(4):331–6. 44. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9. J Gen Intern Med. 2001; 15. Tan L, et al. Preventing the development of depression at work: a 16(9):606–13. systematic review and meta-analysis of universal interventions in the 45. Bech P. Measuring the dimensions of psychological general well-being by workplace. BMC Med. 2014;12(1):74. the WHO-5. QoL Newsletter. 2004;32:15–6. 16. Joyce S, et al. Workplace interventions for common mental disorders: a 46. Bech P, Gudex C, Staehr Johansen K. The WHO (ten) well-being index: systematic meta-review. Psychol Med. 2016;46(4):683–97. validation in diabetes. Psychother Psychosom. 1996;65:183–90. 17. Nieuwenhuijsen K, et al. Interventions to improve occupational health in 47. Kroenke K, et al. Anxiety disorders in primary care: prevalence, impairment, depressed people. Cochrane Database Syst Rev. 2008;2(2):CD006237. comorbidity, and detection. Ann Intern Med. 2007;146:317–25. https://doi.org/10.1002/14651858.CD006237.pub2. 48. Connor KM, Davidson JR. Development of a new resilience scale: the Connor- 18. Roche AM, et al. Men, work, and mental health: a systematic review of Davidson resilience scale (CD-RISC). Depress Anxiety. 2003;18(2):76–82. depression in male-dominated industries and occupations. Safety and 49. Kessler RC, et al. The world health organization health and work health at work. 2016;7(4):268–83. performance questionnaire (HPQ). J Occup Environ Med. 2003;45(2):156–74. 19. Deady M, et al. Designing smartphone mental health applications for 50. Spitzer RL, et al. Validation and utility of a self-report version of PRIME-MD - emergency service workers. Occup Med. 2017:kqx056. the PHQ primary care study. JAMA. 1999;282(18):1737–44. 20. Donker T, et al. Smartphones for smarter delivery of mental health 51. Spitzer RL, Kroenke K, Williams JB. Validation and utility of a self-report version programs: a systematic review. J Med Internet Res. 2013;15(11):e247. of PRIME-MD: the PHQ primary care study. Primary care evaluation of mental 21. Calvo RA, et al. Computing in mental health. In: Proceedings of the 2016 disorders. Patient health questionnaire. JAMA. 1999;282(18):1737–44. CHI Conference Extended Abstracts on Human Factors in Computing 52. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9 - validity of a brief Systems. San Jose: ACM; 2016. depression severity measure. J Gen Intern Med. 2001;16(9):606–13. 22. Josephine K, et al. Internet-and mobile-based depression interventions for 53. Campbell-Sills L, Stein MB. Psychometric analysis and refinement of the people with diagnosed depression: a systematic review and meta-analysis. J connor–Davidson resilience scale (CD-RISC): validation of a 10-item measure Affect Disord. 2017;223:28–40. of resilience. J Trauma Stress. 2007;20(6):1019–28. 23. Calvo R, et al. Positive computing: a new partnership between psychology, social sciences and technologists (editorial introduction), vol. 6; 2016. 54. Southwick SM, Vythilingam M, Charney DS. The psychobiology of 24. Reynolds S. Psychological well-being at work: is prevention better than depression and resilience to stress: implications for prevention and cure? J Psychosom Res. 1997;43(1):93–102. treatment. Annu Rev Clin Psychol. 2005;1:255–91. 25. Association, W.M. World medical association declaration of Helsinki: ethical 55. Krieger T, et al. Measuring depression with a well-being index: further principles for medical research involving human subjects. Jama. 2013; evidence for the validity of the WHO well-being index (WHO-5) as a 310(20):2191–4. measure of the severity of depression. J Affect Disord. 2014;156:240–4. 26. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines 56. Imamura KKNF, Matsuyama Yutaka TA, Akihito S, Rino U, Sonoko K, Kiyoto K. for reporting parallel group randomised trials. BMC Med. 2010;8(1):18. Effects of an internet-based cognitive behavioral therapy (iCBT) program in 27. Australian Bureau of Statistics. Australian and New Zealand standard manga format on improving subthreshold depressive symptoms among industrial classification. Canberra: Australian Bureau of Statistics; 2008. healthy workers: a randomized controlled trial. PLoS One. 2014;9(5):e97167. Deady et al. BMC Psychiatry (2018) 18:166 Page 9 of 9 57. Johnson PO, Neyman J. Tests of certain linear hypotheses and their application to some educational problems. Stat Res Memoir. 1936;1:57–93. 58. Green P, MacLeod CJ, SIMR. An R package for power analysis of generalized linear mixed models by simulation. Methods Ecol Evol. 2016;7(4):493–8. 59. Lerner D, Henke RM. What does research tell us about depression, job performance, and work productivity? J Occup Environ Med. 2008;50(4):401–10. 60. Wang PS, et al. Twelve-month use of mental health services in the United States: results from the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62(6):629–40. 61. Whiteford HA, et al. The work outcomes research cost-benefit (WORC) project: the return on investment for facilitating help seeking behaviour. Aust N Z J Psychiatry. 2005;39:A37. 62. Hanisch SE, et al. The effectiveness of interventions targeting the stigma of mental illness at the workplace: a systematic review. BMC Psychiatry. 2016;16(1):1. 63. French MT, et al. Factors that influence the use and perceptions of employee assistance programs at six worksites. J Occup Health Psychol. 1997;2(4):312. 64. Leykin Y, et al. Results from a trial of an unsupported internet intervention for depressive symptoms. Internet Interventions. 2014;1(4):175–81. 65. Donker T, et al. Internet-delivered interpersonal psychotherapy versus internet- delivered cognitive behavioral therapy for adults with depressive symptoms: randomized controlled noninferiority trial. J Med Internet Res. 2013;15:e82. 66. De Maat S, et al. Relative efficacy of psychotherapy and pharmacotherapy in the treatment of depression: a meta-analysis. Psychother Res. 2006;16(5):566–78. 67. Christensen H, et al. Free range users and one hit wonders: community users of an internet-based cognitive behaviour therapy program. Aust N Z J Psychiatry. 2006;40(1):59–62. 68. Manea L, Gilbody S, McMillan D. Optimal cut-off score for diagnosing depression with the patient health questionnaire (PHQ-9): a meta-analysis. Can Med Assoc J. 2012;184(3):E191–6. 69. Manea L, Gilbody S, McMillan D. A diagnostic meta-analysis of the patient health Questionnaire-9 (PHQ-9) algorithm scoring method as a screen for depression. Gen Hosp Psychiatry. 2015;37(1):67–75. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Psychiatry Springer Journals

A smartphone application for treating depressive symptoms: study protocol for a randomised controlled trial

Free
9 pages

Loading next page...
 
/lp/springer_journal/a-smartphone-application-for-treating-depressive-symptoms-study-qk593iv7Op
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Medicine & Public Health; Psychiatry; Psychotherapy
eISSN
1471-244X
D.O.I.
10.1186/s12888-018-1752-5
Publisher site
See Article on Publisher Site

Abstract

Background: Depression is a commonly occurring disorder linked to diminished role functioning and quality of life. The development of treatments that overcome barriers to accessing treatment remains an important area of clinical research as most people delay or do not receive treatment at an appropriate time. The workplace is an ideal setting to roll-out an intervention, particularly given the substantial psychological benefits associated with remaining in the workforce. Mobile health (mhealth) interventions utilising smartphone applications (apps) offer novel solutions to disseminating evidence based programs, however few apps have undergone rigorous testing. The present study aims to evaluate the effectiveness of a smartphone app designed to treat depressive symptoms in workers. Methods: The present study is a multicentre randomised controlled trial (RCT), comparing the effectiveness of the intervention to that of an attention control. The primary outcome measured will be reduced depressive symptoms at 3 months. Secondary outcomes such as wellbeing and work performance will also be measured. Employees from a range of industries will be recruited via a mixture of targeted social media advertising and Industry partners. Participants will be included if they present with likely current depression at baseline. Following baseline assessment (administered within the app), participants will be randomised to receive one of two versions of the Headgear application: 1) Intervention (a 30-day mental health intervention focusing on behavioural activation and mindfulness), or 2) attention control app (mood monitoring for 30 days). Participants will be blinded to their allocation. Analyses will be conducted within an intention to treat framework using mixed modelling. Discussion: The results of this trial will provide valuable information about the effectiveness of mhealth interventions in the treatment of depressive symptoms in a workplace context. Trial registration: The current trial is registered with the Australian and New Zealand Clinical Trials Registry (ACTRN12617000547347, Registration date: 19/04/2017). Keywords: Depression, Workplace, Mhealth, Treatment Background difficulties in all aspects of their life, including work, Depression is now recognised as one of the leading home, and social activities [5]. In response to the causes of disability worldwide [1]. As well as being rela- increased identification and recognition of depression tively common [2], depression can also be very debilitat- [6], there has been greater emphasis on the development ing [3], with core symptoms inclusive of; an absence of of innovative and effective psychological treatments. positive affect, persistent low mood, and low activity [4]. The treatment of depression, both pharmacologically Persons with severe depressive symptoms report serious and with non-pharmacological interventions, remains a critical area of ongoing clinical research. There is substan- tial evidence supporting the efficacy of psychotherapy in * Correspondence: m.deady@unsw.edu.au the treatment of both sub-clinical [7] and clinical levels of Black Dog Institute; Faculty of Medicine, UNSW, Sydney, Australia Full list of author information is available at the end of the article depression [8]. Whilst cognitive behaviour therapy (CBT) © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Deady et al. BMC Psychiatry (2018) 18:166 Page 2 of 9 remains the most empirically supported treatment [9], been trialled [20] and even fewer have been specifically other therapies such as behavioural activation therapy tailored to a workplace setting. (BAT) has been shown to be equally effective and also less The proposed HeadGear trial aims to evaluate the complex to administer [10]. Service use statistics indicate effectiveness of a new smartphone app in treating that despite the effectiveness of such treatments, few depressive symptoms in a workplace context. people readily access these services [11]. Overcoming identified barriers to accessing psychotherapy such as cost, Methods convenience and accessibility [12] may enhance people’s Design ability to access evidence based treatments. The aim of the current study will be achieved through a Despite experiencing depression many people still re- multicentre randomized controlled trial, with two parallel main as active participants in the workforce [13]. Ongoing arms. The trial will compare two smartphone app-based in- employment is associated with numerous individual bene- terventions: a novel intervention app designed for the treat- fits such as economic, psychosocial, and emotional well- ment of depression (Headgear) and an attention control being [14]. Given its immediacy and core role in the lives app (mood monitoring). Assessments will occur at baseline, of individuals, the workplace may be an avenue for provid- post-intervention (5-weeks), and at 3-month follow-up. ing psychological interventions to those who might not The study is registered with the Australian New Zealand otherwise access mental health services in a conventional Clinical Trials Registry (ACTRN12617000547347) and has manner [15]. Interventions implemented in the workplace ethical approval from the University of New South Wales have been found to be effective in improving both mental (UNSW) Human Research Ethics Committee (HC17021). health and occupational outcomes [16]. This highlights an Consent to participate in the trial will be obtained electron- added benefit of workplace interventions as medical inter- ically from all participants. The study will be conducted in ventions in isolation have not shown as positive an effect accordance with the Helsinki Declaration [25]and is com- on work-related outcomes when compared to workplace pliant with the CONSORT guidelines [26]. interventions [17]. To date, most workplace interventions face accessibility and scalability issues, thus limiting the Setting and participants opportunity for employees to access these services. This The study will recruit Australians who are currently issue appears to be particularly prevalent in workers employed, and will sample more selectively from a range employed in industries where roles are mobile and/or iso- of male-dominated industries [27]. In Australia, these lated; work hours are often intermittent or excessive; and industries include agriculture/forestry/fishing, utility ser- tasks may be repetitive and high-risk (e.g. such as con- vices (electricity, gas, water and waste), wholesale trade, struction, transport and mining). These industries tend to manufacturing, transport/postal/warehousing, mining, be male-dominated (males > 70% of workforce) [18]and and construction [28]. Emergency services and defence are associated with higher rates of mental health concerns also fit this definition, but were not considered unique [18]. Technological innovations, namely smartphones, industries by the ABS, though for this study they will provide an opportunity for individuals in such roles to also be considered as male-dominated. learn how to manage their wellbeing and seek further sup- Industry partner organisations will assist with recruit- port if required [19]. ment by promoting the study among specific groups or Mobile health (mhealth) technologies are increasingly their entire workforce. We will aim to recruit at least being recognised as an effective means through which 850 employed adults across Australia. Organisations that mental health interventions can be disseminated in the elected to participate will promote the study via their population [20, 21]. Such interventions overcome numer- respective health and wellbeing officers, along with email ous barriers associated with treatment seeking, including and newsletter advertisements. The study will also be stigma, cost, and accessibility [20]. Encouragingly, there is promoted via members of the research team presenting preliminary support that mhealth interventions can effect- at partner worksites. Social media advertising targeted at ively treat symptoms in adults with depression [22]. More- employed people will also be utilised to recruit individ- over, mhealth interventions offer the opportunity for an uals employed externally of partner organisations using autonomous, user-directed approach that motivates and an evidence-informed advertising approach [29]. Both offers personalised support by allowing the delivery of males and females will be recruited. content to be tailored to an individual’s interests and needs [23]. This may be particularly advantageous to Eligibility criteria workplace contexts where, individualised interventions Initial eligibility criteria are: having a valid telephone num- are associated with more consistent outcomes than organ- ber, ownership of an Apple/Android-operating smartphone, isational wide interventions [24]. However, of the many fluent in English and living in Australia. As this trial is mhealth interventions available on the market, few have focused on the treatment of depression, participants will Deady et al. BMC Psychiatry (2018) 18:166 Page 3 of 9 also be excluded from this trial (although still permitted to virtually identical look and ‘feel’ as the intervention use the app) if they do not have substantial levels of depres- version of Headgear and is accessed in the same man- sive symptoms at baseline, as indicated by either a PHQ-9 ner. However, there is no skill development and no score below 15, or the PHQ-9’s algorithm for a diagnosis component of behavioural activation or mindfulness for Major Depressive Disorder (MDD) not being met [30]. therapy. To control for the attentional component of the Participants were excluded if they were under 18 or not HeadGear application, the control condition will en- currently employed. No exclusion criteria were in place courage users to use the inbuilt mood monitor daily over regarding comorbidities or medication use. a 30-day period and users will also have access to the ‘risk calculator.’ Interventions Active intervention: HeadGear Procedure The intervention condition, HeadGear, is a smartphone All interested users will be directed to their respective application-based intervention utilising Behavioural Acti- app store (iTunes or Google Play) directly via a dedi- vation Therapy (BAT) and mindfulness-based therapies. cated website (headgear.org.au) where participants regis- Behavioural Action is a therapy based on learning theory ter and provide their phone number. Informed consent that reconnects people to an environment of positive will be sought digitally via both the website and the app reinforcement, incorporating elements of value driven ac- itself. This provides information around the study aims, tion and goal setting [31, 32]. When delivered face-to-face, risks and benefits, confidentiality and dissemination of BAT has been shown to perform as well as CBT for the results. After consent and app download participants treatment of depression [33] and preliminary evidence will undergo initial screening. Participants who meet the that this translates in mhealth form [34]. The other com- inclusion criteria will then be randomised to receive ei- ponent utilised in Headgear, Mindfulness, draws on medi- ther the full HeadGear app or the attention-matched tation practices to allow individuals to gain further insight control version of the app. Participants will be blinded into their emotional, physical, and/or cognitive experience to their allocation. All participants will be provided with to ultimately shape it [35]. Mindfulness has been identified appropriate referral information to health services and as a transtherapeutic process that targets transdiagnostic crisis lines and will be encouraged to seek help from mental processes [36], as evidenced by its effect in treating their GP (if this has not already occurred) while com- a range of psychopathologies including but not limited to pleting the trial. The flow of participants through the mood, anxiety and substance use disorders [37, 38]. study phases is shown in Fig.1. The therapeutic component of HeadGear encourages the user to complete one ‘challenge’ each day (5–10 min Random allocation per day), over 30 days. These ‘challenges’ incorporate a Randomisation will occur immediately following comple- variety of evidence-based BAT and mindfulness tech- tion of the baseline assessment using automated proce- niques and skills, including psychoeducational videos, dures integrated into the trial management software. The value-driven activity planning and goal-setting, practice algorithm for randomisation will consist of a block design, exercises, and techniques for developing coping and stratified by industry type, with a block size of 10. resilience (e.g., problem solving, improving sleep, mini- mising alcohol use, and/or assertiveness training). Other Assessment components of the Headgear intervention app include Administration of assessments mood monitoring, a skill ‘toolbox’ (progressively built as Assessments will be completed at baseline, post-intervention the skills are completed), and a technical service help- (5-weeks), and 3-month follow-up. Baseline assessment line. Steps have been taken to promote user motivation includes outcome measures pertaining to depression symp- and engagement by incorporating a ‘daily challenge’ tomatology (Patient Health Questionnaire (PHQ-9) [44]), framework where the user is ‘rewarded’ (through skill wellbeing (World Health Organisation-5 (WHO-5) tokens) upon completing each daily challenge. A chal- Well-Being Index [45, 46]), anxiety symptomatology (Gen- lenge framework has been shown to increase the general eral Anxiety Disorder-2 item (GAD-2) [47], resilience (Con- appeal of an app [39–41]. In addition to these elements, nor-Davidson Resilience Scale 10-item (CD-RISC10) [48]), the Headgear application has been designed using par- work performance and absenteeism (Health and Work ticipatory approaches and iterative human-computer Performance Questionnaire (HPQ) [49]). In addition to interaction design strategies [42, 43]. these measures, demographic and service use information will be collected. The application monitors usage data in- Attention-matched control cluding number of log-ins, frequency of use, time spent The attention-matched control condition is a smart- in-app, and activity completion rates. This data will be used phone application that will have the same name and a to examine program engagement. Deady et al. BMC Psychiatry (2018) 18:166 Page 4 of 9 Fig. 1 Flow of participants through the trial Post-intervention assessment will occur at 5 weeks’ receive up to three text messages and a call at each post-baseline, to allow users extra time to complete the follow-up assessment time point, linking them to an online 30-day program. Participants will be reminded to complete platform through which they can complete the assessment. the post-intervention assessment via text messages to On completion of each assessment, participants will be phone numbers provided in order to address for the possi- entered into a draw to win one of four $200 Visa gift cards. bility that users may delete the app during the trial. Partici- pants will complete an online questionnaire similar to the Specific measures used in online assessments baseline measure (see Table 1), and again at 3 months The PHQ-9 will be used to measure depression symptom- post-baseline (“3-month follow-up”). Participants will atology [50]. The PHQ-9 is a reliable and valid nine-item Table 1 Assessment measures Baseline Post-intervention 3-month follow-up Demographics × Patient Health Questionnaire-9 item (PHQ-9) [44]× × × General Anxiety Disorder-2 item (GAD-2) [47]× × × World Health Organisation-5 (WHO-5) Well-Being Index [45, 46]× × × Connor-Davidson Resilience Scale 10-item (CD-RISC10) [48]× × × Health and Work Performance Questionnaire (HPQ) [49]× × × Service utilisation and management items × × × Program feedback × Deady et al. BMC Psychiatry (2018) 18:166 Page 5 of 9 measure of depression severity over the past 2 weeks [44, where 0 indicates the worst possible quality of life while 51]. Each of the nine items of the PHQ-9 is scored as 0 a score of 25 represents the best possible quality of life. (not at all), 1 (several days), 2 (more than half the days), or A score ≤ 13 or an answer of 0 or 1 on any of the five 3 (nearly every day). Scores are summed to provide a total items shows poor wellbeing. WHO-5 is a psychometric- score (score range 0–27 with 0 indicating no depressive ally sound measure of well-being with high internal symptoms and 27 indicating all symptoms occurring consistency (Cronbach’s α = 0.84) and convergent associ- nearly daily). The criterion and construct validity of the ations with other measures of well-being [55]. PHQ-9 has previously been demonstrated, with 73% sensi- Work Performance will be measured using three items tivity and 98% specificity in detecting major depression (performance items A10, A11, A12) from the Health and compared to clinician-based assessment [28, 50]and, Work Performance Questionnaire (HPQ) [49]) and two regardless of diagnostic status, typically represents clinic- additional items pertaining to: i) sickness absence over ally significant depression [50]. The measure has excellent the past month (days absent more generally, and days internal consistency (Cronbach α > 0.85 in multiple sam- absent specifically for mental health reasons), and ii) ples) and test-retest reliability (α =0.84) [52]. weeklong sickness absence over the past 6-months Anxiety will be measured using the 2-item Generalised (weeks absent more generally, and weeks absent specific- Anxiety Disorder scale (GAD-2) [47]. The GAD-2 ally for mental health reasons). consists of the two core criteria for generalized anxiety Service use and management items comprised of seven disorder, which have been shown to be effective screen- items assessing lifetime and past month service use, ing items for panic, social anxiety, and post-traumatic along with current medication use. Participants were stress disorders [47]. The GAD-2 begins with the stem also asked about their abilities (perceived capability and question: “Over the last 2 weeks, how often have you effectiveness) to manage their mental fitness, and auton- been bothered by the following problems?” Response omy (choice and freedom) in management. These were options are “not at all”, “several days”, “more than half scored on a 7-point Likert scale from strongly disagree the days”, and “nearly every day”, scored as 0, 1, 2, and to strongly agree. 3, respectively (total ranging from 0 to 6). A total scale score ≥ 3 is suggested as a cut-off point between the nor- Safety protocol mal range and “probable anxiety” [47]. In any trial concerned with mental health, there is the po- Resilience will be measured by the 10 item tential for participants to experience psychological dis- Connor-Davidson Resilience Scale (CD-RISC-10) [53]. tress. Those who meet criteria for the trial will within the The CD-RISC-10 is a self-rated measure, with each app trigger the user to be directed to a “get support” page question rated on a 5-point scale from 0 (‘not true at (at each assessment point) and will suggest the participant all’)to4(‘true nearly all the time’). The CD-RISC-10 seek further help from these support services or their gen- has been shown to differentiate between individuals eral practitioner (GP). Additionally, an optional call-back who function well after adversity and those who do not service for individuals requiring further support or direc- and measures the core features of resilience and the tion is provided. This call-back will be conducted by a ability to tolerate experiences [53]. It is believed that team-member with psychology training, within 4 days, increased resilience may reduce rates of mental ill with the purpose to guide participants into necessary care health [54].The scale demonstrates high internal arrangements. If the team member still has concerns for consistency (Cronbach’s α = 0.89), construct validity, the participant’s safety, an accredited psychiatrist will con- and test-retest reliability in the general population and tact the participants within the next 24 working hours. in clinical settings [48]. Total scores range from 0 to 40 Participants will also receive an SMS with a range of sup- with higher scores corresponding to greater resilience. port service contacts, and another reminder to consult The scale has been shown to have good concurrent with their GP regarding their mental health. validity, with higher resilience on the scale associated with lower levels of perceived stress [48] Validity is high Study hypotheses and outcomes relative to other measures and reflects differentiation in We hypothesise that participants receiving the HeadGear resilience among diverse populations, showing that intervention will have reduced levels of depression symp- higher levels of resilience are consistent with lower tomatology at post-intervention and 3-month follow-up, levels of perceived stress vulnerability [48]. compared to participants in the attention-matched control Wellbeing will be assessed using the 5-item WHO condition. While the primary analyses will be conducted on Wellbeing Index (WHO-5) [45, 46]. Participants are the entire sample (to examine the intervention effect). We asked to self-report on the presence or absence of well- also predict the intervention effect to vary according to the being on a 6-point scale ranging from 5 (‘all of the time’) level of depression symptoms at baseline [56], with a greater to 0 (‘none of the time.’) Raw scores range from 0 to 25 effect amongst those with higher levels of symptomatology. Deady et al. BMC Psychiatry (2018) 18:166 Page 6 of 9 Secondly, it is hypothesized that—relative to the control symptom levels, level of functional improvement and re- group—HeadGear Intervention participants will have cruitment method at baseline will be explored using an lower rates of depressive disorder as detected by the PHQ analysis of covariance approach using baseline measures as algorithm at all follow-up time points. We also hypothe- a covariate and including a covariate by intervention arm sise that the intervention group will have reduced levels of interaction term in models. The effectiveness of the active anxiety symptomatology, and improved wellbeing, resili- intervention at clinically relevant levels of baseline covari- ence, and work performance, at all follow-up time points, ates will be assessed using planned comparisons while the relative to controls. lowest values of covariates associated with a significant benefit of the intervention will be established using a Primary outcome Johnson and Neyman [57]approach. The primary outcome measure of the study will be the All tests of treatment effects will be conducted using a level of depressive symptomatology (as measured by the two-sided alpha level of 0.05 and 95% confidence intervals. PHQ-9) at the 3-month follow up period. Sample size Secondary outcomes As a treatment, the size of the effect of the intervention A range of secondary outcomes will be considered, includ- is anticipated to be moderate. Meta-analysis of previous ing change in anxiety symptomatology (as measured by trials of internet and mobile based treatment of depres- the PHQ-2) at both 5-week and 3-month follow up. Other sion showed a large effect size of g = − 0.90 [22]; how- secondary outcomes include incident caseness of depres- ever unguided interventions typically show smaller effect sion at 5-week and 3-month follow-up (as measured by sizes. Power calculations were carried out using the R the PHQ-9 diagnostic algorithm), and the level of depres- package simR [58]. Power was set at 80%, alpha at 0.05, sion symptoms (PHQ-9) at 5-week follow up. Finally, 2-tailed tests, and a correlation of .50 between pre- and change in Wellbeing (as measured by the WHO-5) and post- intervention scores was assumed. Based on these change in occupational functioning (as measured by the calculations, a sample size of 266 per group was needed HPQ and sickness absence questions) at both 5-week and (total N = 532). A conservative dropout rate of 40% at 3-month follow-ups will be outcomes of interest. follow-up was estimated. An initial sample of 851 will therefore be recruited for the trial. Statistical analysis Analysis plan Dissemination Primary analyses will be undertaken on an intent-to-treat Results of this study will be disseminated for publication basis, including all participants as randomised, regardless in peer-reviewed journals and key findings presented at of treatment received or withdrawal from the study. Like- national and international conferences. lihood based methods (mixed-model repeated measures (MMRM)) methods will be favoured to analyse change in Discussion the primary outcome measure (PHQ-9). A priori planned This study is planned to be the largest randomised con- comparisons of change from baseline across the 3-month trolled trial of a smartphone intervention for depression. follow up period will be used to test the primary hypoth- By targeting the workers, this trial will provide valuable esis. An unstructured variance-covariance matrix will be evidence regarding the effectiveness of mhealth interven- used to accommodate relationships between observations tions in treating depressive symptoms in a workplace at different occasions. Variables found to be substantially context. Given the substantial impact that depression imbalanced between groups post randomisation will be has on the individual and the employer, if shown to be tentatively included in these models and retained if effective, this program would allow for a simple and eco- statistically significant and influential on outcomes. Simi- nomical means by which an organisations or govern- lar analyses of scaled secondary measures will assess ments could disseminate a tailored intervention for differential change due to intervention arm. Mathematical workers. Given the proliferation of untested smartphone transformation or categorisation of raw scores may be applications, the dissemination of evidence based prod- undertaken to meet distributional assumptions and ad- ucts into the workplace, and indeed the wider commu- dress any violation of assumptions attributable to outliers. nity, remains a pressing need. Baseline characteristics will be used to define subgroups Employees with significant depressive symptoms have that would be the targeted if the app were offered as treat- higher rates of absenteeism, presenteeism and job turnover ment. Group membership will be used for models to evalu- [59]. Remaining in the workforce is important as it offers ate moderation of effectiveness by adding appropriate structure, empowerment, financial security and protection interaction terms and undertaking planned comparisons. from the psychological impacts of unemployment [14]. The effect on outcome of level of baseline depressive Using the workplace as a means to dispense or promote an Deady et al. BMC Psychiatry (2018) 18:166 Page 7 of 9 intervention may not only be protective for the individual By using both methods, we are confident that participants concerned, but may also assist in overcoming the issue of without significant symptomatology will be excluded and individuals delaying or not receiving treatment [60]. From those with significant symptomatology will be included. an economic standpoint, it is has been established that, for The treatment of depression utilising evidence based the workplace, such an approach is more cost-effective for mhealth interventions remains an important area of clinical the organisation [61]: utilising smartphone technologies research. The Headgear trial will be the largest trial of a would improve upon this cost-effectiveness. It is also hoped smartphone application that seeks to offer an alternate or that dissemination via workplaces and social media will augmentation to traditional face-to-face therapy through help engage individuals who would not usually access help which working adults can manage their mental health and via the health care system. wellbeing. This trial will be unique in that it is advertising a This study will provide valuable evidence regarding the treatment via a workplace setting, and allowing for the effectiveness of mhealth tools in the treatment of depres- assessment of clinical and occupational outcomes. Finally, sive symptoms. Despite the proliferation of mental health the Headgear Trial will provide much needed information apps, there is scarce research on the effectiveness of such on the general effectiveness of evidence-based interventions apps. Indeed, in a systematic review of the literature on (BAT and mindfulness) delivered through smartphone smartphone interventions, only five mental health apps technology. were empirically tested and only one of these did not re- Funding quire the input from a mental health professional [20]. This study was developed in partnership with beyond blue with donations mHealth interventions offer advantages in that they in- from the Movember Foundation. RAC is funded by an Australian Research Council Future Fellowship FT140100824. SBH and MD are supported by crease user autonomy and anonymity, which may be im- funding from iCare and NSW Health. portant as stigma [62] and lack of knowledge of services [63] can impact upon help-seeking behaviour in the work- Availability of data and materials The anonymized datasets generated and analysed during the current study place. An intervention developed for the workplace also will not be publicly available due to legal and ethical restrictions. carries the benefit in that it could ameliorate some of the financial burden placed solely on the healthcare system. Dissemination The current trial has received ethics approval from the University of New The proposed trial does carry with it a number of limi- South Wales Human Research Ethics Committee (HC17021). Research tations. The use of a smartphone app as a delivery mo- findings will be disseminated via peer reviewed journals, conferences and dality does mean that the intervention is unguided and internal reports. that the user is responsible for managing their inter- Authors’ contributions action with the program. Thus, trial attrition and disen- All of the other authors contributed to reviews of the manuscript and the gagement are expected issues [64, 65]. It is worth original research design and the analysis of data. All authors read and approved the final manuscript. noting, however, that this has also been an issue for face-to-face psychotherapy trials [66]. The reason for Ethics approval and consent to participate drop out in both mhealth and face-to-face trials is often The proposed trial has received ethical approval from the University of New South Wales (UNSW) Human Research Ethics Committee (HC17021). multi-faceted, and whilst can be related to engagement Prior to study participation all participants viewed and consented to the with the program, it is rarely only due to dissatisfaction information that was provided in the Participant Information Sheet. [67].To account for potential drop out, two procedures Competing interests were put into place. Firstly, all follow-up communication All researchers have remained independent from the funders in the would occur via phone numbers to ensure that if the completion and submission of this work. All authors have no conflict of participant uninstalled the app before the follow-up interest or competing interests to declare, except that the intellectual property of the smartphone application is jointly owned. period, the participant could still be reached. Secondly, conservative drop-out estimation and the use of statis- Publisher’sNote tical methods robust to data missing at random, it is Springer Nature remains neutral with regard to jurisdictional claims in believed that this limitation will be minimised. published maps and institutional affiliations. Another limitation of the present trial is related to the reli- Author details ance on self-reported depressive symptoms, rather than a 1 2 Black Dog Institute; Faculty of Medicine, UNSW, Sydney, Australia. School of diagnosis of depression achieved through a structured diag- 3 Psychiatry, UNSW Sydney, Sydney, Australia. Central Clinical School, Brain nostic interview. This is a common issue faced by most and Mind Centre, Faculty of Medicine and Health, University of Sydney, Sydney, Australia. School of Electrical and Information Engineering, similar trials given the constraints around time and re- University of Sydney, Sydney, NSW 2006, Australia. School of Systems sources. To overcome this issue, we will use a well-validated Management and Leadership, Faculty of Engineering and IT, University of measure (PHQ-9) that contains two methods for classifying Technology Sydney, Sydney, Australia. Department of Mental Health and Suicide, Norwegian Institute of Public Health, Oslo, Norway. School of depression: 1) threshold total score above 14 (sensitivity = Psychology, UNSW Sydney, Sydney, Australia. MRC Cognition and Brain 67%; specificity = 95%) [68] or 2) meeting the depression 9 Sciences Unit, University of Cambridge, Cambridge, UK. Department of algorithm’s criteria (sensitivity = 0.53; specificity = 0.94) [69]. Community Medicine, University of Tromsø, Tromsø, Norway. Centre for Deady et al. BMC Psychiatry (2018) 18:166 Page 8 of 9 Work and Mental Health, Nordland Hospital Trust, Bodø, Norway. Centre for 28. Kroenke K, Spitzer RL. The PHQ-9: a new depression diagnostic and severity Research and Education in Forensic Psychiatry and Psychology, Haukeland measure. Psychiatr Ann. 2002;32(9):509–15. University Hospital, Bergen, Norway. 29. Choi I, et al. Using different Facebook advertisements to recruit men for an online mental health study: engagement and selection bias. Internet Received: 4 December 2017 Accepted: 17 May 2018 Interventions. 2017;8:27–34. 30. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–13. 31. Hopko DR, et al. Contemporary behavioral activation treatments for depression: References procedures, principles, and progress. Clin Psychol Rev. 2003;23(5):699–717. 1. Organization, W.H., Depression and other common mental disorders: global 32. Lejuez CW, et al. Ten year revision of the brief behavioral activation treatment health estimates. 2017. for depression: revised treatment manual. Behav Modif. 2011;35(2):111–61. 2. Slade T, et al. 2007 National Survey of mental health and wellbeing: 33. Richards DA, et al. Cost and outcome of Behavioural activation versus methods and key findings. Aust N Z J Psychiatry. 2009;43(7):594–605. cognitive Behavioural therapy for depression (COBRA): a randomised, 3. Harvey SB, et al. Depression and work performance: an ecological study controlled, non-inferiority trial. Lancet. 388(10047):871–80. using web-based screening. Occup Med (Lond). 2011;61(3):209–11. 34. Ly KH, et al. Behavioural activation versus mindfulness-based guided self- 4. American Psychiatric Association. Diagnostic and Statistical Manual of help treatment administered through a smartphone application: a Mental Disorders (DSM-5). 5th ed. Washington, DC: APA; 2013. randomised controlled trial. BMJ Open. 2014;4(1):e003440. 5. Kessler RC, et al. The epidemiology of major depressive disorder: results from the 35. Brown KW, Ryan RM, Creswell JD. Mindfulness: theoretical foundations and National Comorbidity Survey Replication (NCS-R). Jama. 2003;289(23):3095–105. evidence for its salutary effects. Psychol Inq. 2007;18(4):211–37. 6. Harvey SB, et al. Is the prevalence of mental illness increasing in Australia? 36. Greeson J, Garland EL, Black D. Mindfulness: a transtherapeutic approach for Evidence from national health surveys and administrative data, 2001–2014. transdiagnostic mental processes. Wiley Blackwell handbook of mindfulness. Med J Aust. 2017;206(11):490–3. 2014:531–62. 7. Cuijpers P, et al. Psychotherapy for subclinical depression: meta-analysis. Br J 37. Hofmann SG, et al. The effect of mindfulness-based therapy on anxiety and Psychiatry. 2014;205(4):268–74. depression: a meta-analytic review. J Consult Clin Psychol. 2010;78(2):169. 8. Cuijpers P, et al. Psychotherapy for depression in adults: a meta-analysis of 38. Zgierska A, et al. Mindfulness meditation for substance use disorders: a comparative outcome studies. J Consult Clin Psychol. 2008;76(6):909–22. systematic review. Subst Abus. 2009;30(4):266–94. 9. Butler AC, et al. The empirical status of cognitive-behavioral therapy: a 39. Titov N, et al. Improving adherence and clinical outcomes in self-guided review of meta-analyses. Clin Psychol Rev. 2006;26(1):17–31. internet treatment for anxiety and depression: randomised controlled trial. 10. Richards DA, et al. Cost and outcome of Behavioural activation versus PLoS One. 2013;8(7):e62873. cognitive Behavioural therapy for depression (COBRA): a randomised, 40. Matthews M, et al. Designing mobile applications to support mental health controlled, non-inferiority trial. Lancet. 2016;388(10047):871–80. interventions. In: Jo L, editor. Handbook of Research on User Interface 11. Olfson M, Marcus SC. National trends in outpatient psychotherapy. Am J Design and Evaluation for Mobile Technology. London: IGI, Information Psychiatr. 2010;167(12):1456–63. Science Reference; 2008. p. 635–56. 12. Mohr DC, et al. Barriers to psychotherapy among depressed and 41. Bakker D, et al. Mental health smartphone apps: review and evidence-based nondepressed primary care patients. Ann Behav Med. 2006;32(3):254–8. recommendations for future developments. JMIR Mental Health. 2016;3(1):e7. 13. Sanderson K, Andrews G. Common mental disorders in the workforce: 42. Deady M, et al. Designing smartphone mental health applications for recent findings from descriptive and social epidemiology. Can J Psychiatry. emergency service workers. Occup Med. 2017; 2006;51(2):63–75. 43. Calvo RA, Peters D. Positive computing: technology for wellbeing and 14. Modini M, et al. The mental health benefits of employment: results of a human potential. Cambridge: MIT Press; 2014. systematic meta-review. Australas Psychiatry. 2016;24(4):331–6. 44. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9. J Gen Intern Med. 2001; 15. Tan L, et al. Preventing the development of depression at work: a 16(9):606–13. systematic review and meta-analysis of universal interventions in the 45. Bech P. Measuring the dimensions of psychological general well-being by workplace. BMC Med. 2014;12(1):74. the WHO-5. QoL Newsletter. 2004;32:15–6. 16. Joyce S, et al. Workplace interventions for common mental disorders: a 46. Bech P, Gudex C, Staehr Johansen K. The WHO (ten) well-being index: systematic meta-review. Psychol Med. 2016;46(4):683–97. validation in diabetes. Psychother Psychosom. 1996;65:183–90. 17. Nieuwenhuijsen K, et al. Interventions to improve occupational health in 47. Kroenke K, et al. Anxiety disorders in primary care: prevalence, impairment, depressed people. Cochrane Database Syst Rev. 2008;2(2):CD006237. comorbidity, and detection. Ann Intern Med. 2007;146:317–25. https://doi.org/10.1002/14651858.CD006237.pub2. 48. Connor KM, Davidson JR. Development of a new resilience scale: the Connor- 18. Roche AM, et al. Men, work, and mental health: a systematic review of Davidson resilience scale (CD-RISC). Depress Anxiety. 2003;18(2):76–82. depression in male-dominated industries and occupations. Safety and 49. Kessler RC, et al. The world health organization health and work health at work. 2016;7(4):268–83. performance questionnaire (HPQ). J Occup Environ Med. 2003;45(2):156–74. 19. Deady M, et al. Designing smartphone mental health applications for 50. Spitzer RL, et al. Validation and utility of a self-report version of PRIME-MD - emergency service workers. Occup Med. 2017:kqx056. the PHQ primary care study. JAMA. 1999;282(18):1737–44. 20. Donker T, et al. Smartphones for smarter delivery of mental health 51. Spitzer RL, Kroenke K, Williams JB. Validation and utility of a self-report version programs: a systematic review. J Med Internet Res. 2013;15(11):e247. of PRIME-MD: the PHQ primary care study. Primary care evaluation of mental 21. Calvo RA, et al. Computing in mental health. In: Proceedings of the 2016 disorders. Patient health questionnaire. JAMA. 1999;282(18):1737–44. CHI Conference Extended Abstracts on Human Factors in Computing 52. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9 - validity of a brief Systems. San Jose: ACM; 2016. depression severity measure. J Gen Intern Med. 2001;16(9):606–13. 22. Josephine K, et al. Internet-and mobile-based depression interventions for 53. Campbell-Sills L, Stein MB. Psychometric analysis and refinement of the people with diagnosed depression: a systematic review and meta-analysis. J connor–Davidson resilience scale (CD-RISC): validation of a 10-item measure Affect Disord. 2017;223:28–40. of resilience. J Trauma Stress. 2007;20(6):1019–28. 23. Calvo R, et al. Positive computing: a new partnership between psychology, social sciences and technologists (editorial introduction), vol. 6; 2016. 54. Southwick SM, Vythilingam M, Charney DS. The psychobiology of 24. Reynolds S. Psychological well-being at work: is prevention better than depression and resilience to stress: implications for prevention and cure? J Psychosom Res. 1997;43(1):93–102. treatment. Annu Rev Clin Psychol. 2005;1:255–91. 25. Association, W.M. World medical association declaration of Helsinki: ethical 55. Krieger T, et al. Measuring depression with a well-being index: further principles for medical research involving human subjects. Jama. 2013; evidence for the validity of the WHO well-being index (WHO-5) as a 310(20):2191–4. measure of the severity of depression. J Affect Disord. 2014;156:240–4. 26. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines 56. Imamura KKNF, Matsuyama Yutaka TA, Akihito S, Rino U, Sonoko K, Kiyoto K. for reporting parallel group randomised trials. BMC Med. 2010;8(1):18. Effects of an internet-based cognitive behavioral therapy (iCBT) program in 27. Australian Bureau of Statistics. Australian and New Zealand standard manga format on improving subthreshold depressive symptoms among industrial classification. Canberra: Australian Bureau of Statistics; 2008. healthy workers: a randomized controlled trial. PLoS One. 2014;9(5):e97167. Deady et al. BMC Psychiatry (2018) 18:166 Page 9 of 9 57. Johnson PO, Neyman J. Tests of certain linear hypotheses and their application to some educational problems. Stat Res Memoir. 1936;1:57–93. 58. Green P, MacLeod CJ, SIMR. An R package for power analysis of generalized linear mixed models by simulation. Methods Ecol Evol. 2016;7(4):493–8. 59. Lerner D, Henke RM. What does research tell us about depression, job performance, and work productivity? J Occup Environ Med. 2008;50(4):401–10. 60. Wang PS, et al. Twelve-month use of mental health services in the United States: results from the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62(6):629–40. 61. Whiteford HA, et al. The work outcomes research cost-benefit (WORC) project: the return on investment for facilitating help seeking behaviour. Aust N Z J Psychiatry. 2005;39:A37. 62. Hanisch SE, et al. The effectiveness of interventions targeting the stigma of mental illness at the workplace: a systematic review. BMC Psychiatry. 2016;16(1):1. 63. French MT, et al. Factors that influence the use and perceptions of employee assistance programs at six worksites. J Occup Health Psychol. 1997;2(4):312. 64. Leykin Y, et al. Results from a trial of an unsupported internet intervention for depressive symptoms. Internet Interventions. 2014;1(4):175–81. 65. Donker T, et al. Internet-delivered interpersonal psychotherapy versus internet- delivered cognitive behavioral therapy for adults with depressive symptoms: randomized controlled noninferiority trial. J Med Internet Res. 2013;15:e82. 66. De Maat S, et al. Relative efficacy of psychotherapy and pharmacotherapy in the treatment of depression: a meta-analysis. Psychother Res. 2006;16(5):566–78. 67. Christensen H, et al. Free range users and one hit wonders: community users of an internet-based cognitive behaviour therapy program. Aust N Z J Psychiatry. 2006;40(1):59–62. 68. Manea L, Gilbody S, McMillan D. Optimal cut-off score for diagnosing depression with the patient health questionnaire (PHQ-9): a meta-analysis. Can Med Assoc J. 2012;184(3):E191–6. 69. Manea L, Gilbody S, McMillan D. A diagnostic meta-analysis of the patient health Questionnaire-9 (PHQ-9) algorithm scoring method as a screen for depression. Gen Hosp Psychiatry. 2015;37(1):67–75.

Journal

BMC PsychiatrySpringer Journals

Published: Jun 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off