Internet-based treatment for Romanian adults with panic disorder: protocol of a randomized controlled trial comparing a Skype-guided with an unguided self-help intervention (the PAXPD study)

Hundreds of millions of people are affected by mental health disorders worldwide and failing to adequately address this issue brings significant suffering, disability and economic loss. The lifetime prevalence for mental health disorders in general is above 46 %, with the highest prevalence recorded for anxiety disorders, given that almost 1 out of 3 people (28.8 %) suffer from an anxiety disorder during their lifetime [1]. Every year over 38.2 % (164.7 million people) of the total EU population suffer from at least one mental disorder and the resulting economic costs rise at 3–4 % of EU Gross Domestic Product (GDP) [2].

One third of medical consultations made by general practitioners are in fact provided for mental health problems [3‐5]. A report released by WHO (World Health Organization) and WONCA (World Organization of Family Doctors) in 2008 stated that “up to 60 % of people attending primary care clinics have a diagnosable mental disorder” (page VII) [6]. There is an alarming underdetection of mental health problems – for example, detection rates of anxiety and mood disorders in general practice reach only 15 and 36 %, respectively [7, 8]. Moreover, collaboration between GPs and mental health services is either poor or non-existent.

Despite the potential to successfully treat mental disorders, only a small minority receive even the most basic treatment. Studies have shown that 60 % of those in need receive no treatment at all, and only 32.7 % of those who receive treatment get it at minimal quality standards [9]. Furthermore, there are numerous individual and systemic barriers to accessing mental health services [10], thus the treatment is provided only with a significant delay from the onset of the disorder (after 9–23 years for anxiety disorders and 6–8 years for mood disorders) [11]. In some countries, the situation is even worse. In Romania, where the current study is conducted, 76.4 % of the individuals with mental health disorders do not have access to any form of treatment; only about 11.5 % receive psychological or psychiatric help [12].

Computer and Internet-based cognitive behavioral treatments (ICBT) represent recent attempts to innovate the psychotherapeutic process, matching both the existent demands of the market, and the new attitudes of those who seek help in the digital era. Research on internet-based treatments has grown rapidly in the past decade, gathering more than 100 controlled studies that tested various intervention programs for anxiety disorders, mood disorders, or other psychiatric conditions [13‐15]. Overall, the outcomes of these studies have been positive and very convincing. Several studies reported no significant differences in treatment outcomes when comparing internet-based interventions to traditional face-to-face therapy [16‐21]. For a review see Andersson et al. [22]. Because half of all lifetime cases of mental disorders start by age 14 years and the median age of onset for anxiety is 11 years [1], it should be mentioned that ICBT shows promising results not just for adults, but also for treating anxiety and depression in youth [23, 24]. Anxiety disorders seem to be one of the most approachable and treatable types of mental disorder through online therapy [25‐31]. Panic disorder (PD) with or without agoraphobia is one of the conditions for which ICBT programs have been developed since the late 1990s and there are several empirical studies supporting their efficacy and effectiveness [13, 17, 19, 32‐42]. In Romania, only one internet-based treatment has been tested so far, the Swedish program Sofie [43]. The study yielded positive outcomes for the treatment of social anxiety disorder, but the program is currently not in use.

The most promising interventions designed and tested so far are guided self-help interventions. A guided ICBT intervention is a web-based self-help program combined with minimal, but regular therapist support. In guided programs, asynchronous communication (emails or text messages) is most commonly used and, on average, patient guidance does not take more than 10 min per week [44]. So far, research is very limited regarding web-based treatment programs that provide guidance through synchronous (real time) audio-video communication (e.g. video chat programs such as Skype). On the other hand, there is growing evidence that the efficacy of CBT delivered via videoconference is equivalent to face-to-face treatment for mood and anxiety disorders [45‐50]. This type of communication closely matches face-to-face therapy and enables access to important face-to-face cues such as verbal tone, facial expressions, and body language. Through synchronous guidance, the therapists may be able to offer more individually tailored feedback and support according to each patient’s personal needs. Furthermore, this type of contact may facilitate higher engagement, feelings of accountability and social support, and reduce the risk of misunderstandings, compared to emails and instant messaging [51]. All of the above could also facilitate the development of a better therapeutic alliance [52], a factor that has consistently proven to be a key predictor for therapeutic change in face-to-face therapy [53, 54]. So far, there is mixed evidence regarding the importance of the therapeutic alliance in guided ICBT [44]. Even if contact with the therapist is only minimal, patients seem to develop a strong alliance, but this makes no difference for the outcome, at least for self-help interventions with rather low-intensity guidance [44]. However, the therapeutic alliance in internet-based treatments that include video conferencing is yet to be investigated [55]. Of course, there are also disadvantages for using real-time audio-video communication. For instance, compared to asynchronous written exchanges, patients do not have the opportunity to revisit and reread the dialogues with their therapists [51]. The patients are still able to revisit and reflect upon the information enclosed in the treatment modules, but they cannot review their guidance sessions.

It is still not clear enough whether guided online treatments work better than unguided treatments [18, 31, 56‐59], however there is a strong tendency for the superiority of guided treatment, especially for depression [14, 60]. So far, the literature suggests a superiority of guided versus unguided self-help treatments in terms of efficacy, adherence to treatment and drop-out rates [14, 61‐63], but the conclusions are based on few studies and mainly limited to depression and social anxiety disorder, thus restricting the generalizability of the findings [64].

There is good quality evidence that ICBT treatments, especially guided interventions, are both efficacious and effective, either compared with waiting-list or with face to face interventions, but there is no intervention that is suitable for everyone [44]. In order to be able to make recommendations for a certain person, with certain characteristics, we need to know which variables can predict or mediate/ moderate the outcomes. Unfortunately, the studies conducted so far showed that there are few stable predictors across studies and diagnoses [65]. For example, treatment adherence was considered as a potential predictor variable, but there are inconsistencies, with some studies finding an association between adherence and clinical outcomes [66‐68] and others not [65, 69, 70]. Another example is the severity of the disorder at baseline. In one study, a high baseline severity predicted better outcomes [71], while another study showed the contrary, low levels of panic symptom severity predicting better outcomes [65]. Another interesting result concerns the role of comorbidity. A recent study showed that having a comorbid anxiety disorder was associated with a better treatment outcome among patients with panic disorder but not patients with social anxiety disorder [72], while others found no significant association [65, 73]. Considering these mixed results, we agree with Andersson who concluded “We do not know much about predictors of treatment outcome and the mechanisms involved in generating good outcome. Thus, more studies on moderators and mediators are needed” ([44], p.106).

The study design is reported in line with the CONSORT 2010 statement (Consolidated Standards of Reporting Trials) [77] and the SPIRIT 2013 Statement (Standard Protocol Items: Recommendations for Interventional Trials) [78].

The prevalence of mental health problems is very high and bears considerable suffering and burden worldwide. However, only a small percentage of the affected individuals reach out and receive treatment for their problems. There are many barriers at different levels, yet a viable solution to increase access to quality treatment resides in the possibilities provided by internet-based cognitive behavioral treatments. Although several ICBT programs have been tested with strong positive results and even integrated into regular health care, there are countries where validated internet-based interventions are yet to be developed. The present trial aims both to validate the overall efficacy of the first Romanian ICBT program for panic disorder, and to investigate aspects that are novel to ICBT, such as brief therapist guidance sessions via secured Skype.

There are still divergent results regarding the effect of therapist guidance in ICBT [18, 31, 56‐58], and the efficacy of Skype-based (or equivalent) therapy sessions in ICBT have not been previously investigated. Thus, a central focus of our trial is to test the additive effect of therapeutic guidance via secured Skype in self-help ICBT programs. Specifically, the current study design allows us to assess the effect of the same self-help intervention modules in two treatment conditions: therapist guided via Skype vs. unguided self-directed intervention.

There is a general consensus that internet-based therapy can be efficient for many psychiatric conditions, but not all the patients benefit from it [44]. Thus, there is an increasing interest to describe the characteristics of the patients who would benefit the most from this type of interventions. Taking into consideration this issue, additional exploratory analyses of potential predictors, moderators or mediators will be made. Patients’ treatment expectations, personality traits (e.g. personal autonomy), and usability of the online treatment program could prove to be key factors of treatment adherence and outcome, especially in the unguided self-help treatment group.

Several possible limitations should be noted. Firstly, the study sample size is large enough to reach our main research objectives, but will not permit us to perform detailed subgroup analyses. However, we expect to identify candidate predictors of treatment outcome that could be tested directly in future studies with larger samples. Secondly, the control group, for ethical reasons, will receive treatment in the second stage of the trial, thus the control group will not be available at follow-up. Furthermore, most of the recruitment will be made from the community, through advertisements in the media and on the study website, which could lead to a self-selected sample. Regarding the latter limitation, several findings from recent studies suggest that sampling bias may not affect the generalizability of internet-based trials. For example, Titov et al. [122] revealed that patients undergoing web-based treatments resemble national samples and present disorders as severe as those undergoing traditional psychotherapy. Another example is a recent study by Donkin et al. [123] on sampling bias in an internet-based clinical trial for depression. The authors revealed that out of 35 potentially biasing factors, only 4 were associated with consenting likelihood, accounting for limited variance in explaining participation, and none were associated with the primary outcomes. Future research will be required, nonetheless, to investigate whether the results from the PAXonline trial can be generalized to samples in regular mental health care settings.

In Romania, the need for easily accessible treatment programs for mental disorders is very high. 76.4 % of the individuals with mental health problems do not have access to any form of treatment, and only approximately 11.5 % receive psychological or psychiatric treatment [12]. Internet-based treatments can offer a viable alternative solution to the current situation, considering that internet access has increased rapidly over the past years in Romania (in 2004 only 6 % households had access to the internet, and presently the number has increased to 61 % [124]). However, although ICBT advantages are easily recognized, both patients and psychologists in Romania have a reserved attitude towards using internet-based psychological intervention [125]. By developing rigorous clinical trials on the efficiency of Romanian ICBT programs, we can increase the uptake of these solutions, and thus have a significant positive impact on the quality of life, both at an individual and national level.

Trial start date: May 22, 2014.

Currently recruiting (N_current = 99 as of December 15, 2015).