Behavioral activation-based guided self-help treatment administered through a smartphone application: study protocol for a randomized controlled trial

The aim of this study will be to test the effects of a smartphone-delivered behavioral activation treatment. The study is based on our previous work on guided Internet-treatment for depression [17], but in this study the treatment content will be delivered via smartphone, using a recently developed smartphone application. As waiting-list control groups are regarded as unsuitable to test the effects of treatments, we plan to have a non-specific smartphone application not dealing with depression or behavioral activation. This will probably be the first controlled study on smartphone delivered depression treatment and therefore we will begin by conducting an efficacy study [18], with participants recruited via mass media and the Internet. We expect the smartphone application to be superior to the control condition.

An 8-week smartphone-based behavioral activation intervention with minimal therapist contact has been developed by our group. The intervention is based on our previous experiences with Internet-based depression treatment [15, 16, 19‐23]. The smartphone-based intervention consists of a web-based psychoeducation, and a smartphone application. There is also a back-end system where the therapist can see reports from the patients or activities being reported. The concept of the intervention was built with the intention to target some of the aspects of current Internet-delivered interventions that could be improved. Although the Internet is effectively used for distributing psychological treatment programs to improve mental health [24‐26], including treatment of mild to moderate major depression [27], it has been argued that Internet-based treatment applications have limitations [28, 29]. One important limitation is the lack of assimilation into the user’s daily life, which is a crucial aspect of behavioral activation treatments for depression [12]. Even if paper versions of text material can be used and contact with the therapist can be handled via the Internet, these two options may not be practically feasible. Mobile phones, on the other hand, are often ubiquitous and are hence present when behaviors occur. Nowadays, access to the Internet can be facilitated by means of smartphones. Another limitation of Internet-based treatments for depression, delivered via computers, is that they tend to be text-based, often reaching close to 200 pages of reading [20]. Smartphone applications tend to be more user-friendly and condensed than regular web pages (for example the difference between a smartphone application for a newspaper and web pages for the same newspaper), which could be an attractive feature for persons with depression who easily feel fatigued and need encouragement to work through the program [30]. The latter aspect is important, as guidance appear to be crucial to prevent drop-out from guided self-help treatments [31], which was recently confirmed in a meta-analysis [15]. Finally, a challenge for modern depression treatments is to integrate information technology in the regular treatment [29]. In other words face-to-face treatments could benefit from using smartphones as an adjunct to the regular sessions, which in the case of behavioral activation treatments for depression could facilitate activity scheduling and homework, which are crucial elements of the treatment. It may also be possible to reduce the number of sessions.

A prototype of the smartphone application, together with a psychoeducation, has been tested in a pilot study (Ly KH, Dahl J, Andersson G: Development and initial evaluation of a smartphone application based on acceptance and commitment therapy, unpublished). This prototype was however not directed at depression. The results from the pilot study gave us useful information and we concluded that the intervention include some important features that could be further developed and tested. The results also showed that the smartphone application was used by some participants in situations where it was difficult to use a computer connected to the Internet, for example in the underground train, on the bus and train, and so on. More generally, this might carry implications that the platform of mobile phones does not only assimilate an intervention into a user’s daily life, but also increases the awareness of the intervention, and thus can increase adherence to the intervention. In light of these observations we believe it is feasible to proceed and test if behavioral activation can be presented in the format of a smartphone application.

In the attention control group, we will include brief online education and then recommend use of a smartphone application that is not directly aimed at depression (for example, ‘Effective meditation’). The duration of the control condition will also be 8 weeks. For ethical reasons we will give the participants in the control group access to the behavioral activation treatment following the 8 week treatment period.

Patients will be recruited via advertisement in newspapers, other media, and the Internet. Only persons who already use smartphones will be invited to participate.

We will include participants with an ongoing primary diagnosis of major depression of mild to moderate character. Diagnoses will be made over the telephone using the Mini-International Neuropsychiatric Interview (MINI) [32]. To administer clinical interviews via telephone has shown to give the same result as doing it face-to-face [33, 34]. Medication will be allowed with the requirement that the dose is stable. Consenting participants need to be fluent in Swedish and have access to the Internet as well as a smartphone.

Persons with suicidal intent and patients for whom the depression is directly linked to another major primary psychiatric disorder (for example, psychosis or bipolar disorder) will be excluded and referred to other treatment resources. In addition, people with a history of misuse of alcohol or drugs will be excluded.

Participation will require a written informed consent.

Participants can at any time withdraw from the treatment or the study without specifying the reason.

We will adhere to the Swedish Personal Data Act. No sensitive data will be saved on a computer in which the person providing data can be identified. All Internet and smartphone activities will be highly secure, with encrypted information as well as no contact outside of a closed contact system. If participants decline further participation, the data will be removed.

The primary outcome measures will be self-assessed depression using the Beck Depression Inventory (BDI-II) [35] and the Patient Health Questionnaire-9 (PHQ-9) [36]. PHQ-9 consists of nine items, each scored 0–3 and with a total score range from 0 to 27. PHQ-9 shows a sensitivity of 0.77 (0.71–0.84) and a specificity of 0.94 (0.90–0.97) [37]. BDI was constructed to assess the degree of depression in adolescents and adults, and consists of 21 questions [38]. The BDI has been extensively validated in various studies. A review of the BDI’s internal consistency estimates yielded a mean coefficient alpha of 0.86 for psychiatric patients and 0.81 for non-psychiatric subjects [39]. For psychiatric patients the test-retest reliability has been reported to range from 0.48 to 0.86, whereas the coefficients for non-psychiatric subjects ranged from 0.60 to 0.83 [39]. We will also administer the MINI [32]. This will be done via telephone to make participation possible for persons who live in all regions of Sweden. Moreover, an independent (blind) rating of clinical global impression-improvement scale (CGI) [40] will be performed. Hence, all patients will be interviewed again after the treatment period. All self-report measures will be administered via the Internet, which will facilitate 1-year and 2-year follow-ups. According to Thorndike et al. [41], self-report measures administrated via the Internet show good validity. We estimate a drop-out from completing the post-treatment measures of approximately 30%. This would lead to a reduced sample size of 84, which will maintain the estimated power. While the cross-over design will prohibit conclusions regarding specific long-term effects, the long-term effect of the new treatment format (that is smartphones) is not yet investigated, making the follow-up informative in the case both treatments are ineffective. Moreover, switching treatment will be optional and we will be able to investigate subgroups that only complete one of the treatments. The main outcome of interest in the study is if smartphone-enabled behavioral activation is superior to attention control (use of smartphone with no reference to depression or behavioral activation). We expect a moderate difference on the main outcome measures of depression. We also expect the number of responders (defined as a decrease according to Jacobson on the BDI-II [35] and PHQ-9 [36] and remission to lower levels of symptoms on the MINI-mood disorders section [32]) to be larger in the behavioral activation arm.

As secondary outcome we will include the Beck Anxiety Inventory (BAI) [42], the Quality of Life Inventory (QOLI) [43] and Acceptance and Action Questionnaire (AAQ-II) [44]. The BAI is a 21-item self-report inventory for measuring the severity of anxiety in psychiatric populations [42]. De Beurs et al.[45] showed that the BAI has high internal consistency in panic patients (Cronbach’s a = 0.92) and good test-retest reliability (0.83). The QOLI consists of 16 areas rated by the subject concerning importance and satisfaction. The instrument’s internal consistency is high, between a = 0.77 and 0.89, and the 1 month test-retest reliability lies between r = 0.80 and 0.91 [43]. The AAQ-II is a 10-item, 7-point Likert scale that assesses psychological flexibility - a term for the ability to accept unwanted internal experiences and engage in valued actions. The AAQ II has shown good reliability (test-retest reliability = 0.81–0.87) and validity [44].

Other measures will be included as potential predictors of outcome such as education, number of previous episodes, and so on. Finally, we will include the Trimbos and Institute of Medical Technology Assessment Cost Questionnaire for Psychiatry (TIC-P [46]), which is a questionnaire designed to measure health related costs. Cost-effectiveness studies also often involved a measure called the EQ-5D [47]. The study will follow the guidelines given in the CONSORT statement [48, 49], with the exception of double-blinding, which is not possible in psychotherapy studies. Other features such as intention-to-treat (mixed models approach [50]) and statistical power calculations will be adhered to. The theoretical rationale for smartphone supported behavioral activation is the observation that activity is related to depressive symptoms [8] and that behavioral change may be facilitated by the increased adherence provided by the smartphone.

An intention-to-treat design will be used when analyzing post-treatment data and data collected at 1-year follow-up. Between-group changes at post-treatment and at 1-year follow-up will be calculated using analyses of covariance (ANCOVAs), with pre-treatment scores as the covariate. Effect sizes between and within the two groups will be calculated with Cohen’s d computed with the pooled standard deviation.