Verwerking Feedback en Proces:
1. De eerste twee dagen heb ik niet kunnen doen, omdat ik ziek was en omdat ik emotioneel nogal overstuur was van de feedback. Uiteindelijk mezelf toch een trap onder de kont gegeven en begonnen.
2. Ik begon gewoon bij het begin met het verwerken van de feedback, zo was mijn uitleg over sociale angststoornis en CBT te uitgebreid. Ook moest ik de adherence noemen als mogelijk gevolg van de toevoeging van assistentie.
3. Over de inleiding was ik redelijk te tevreden voor een eerste verbetering. Daarna moest ik de laatste alinea naar voren halen. Op zich was deze alinea makkelijker te plaatsen in mijn nieuwe structuur omdat ik daar ook al een beetje rekening houdt met adherence.
4. Toen besloot ik om die hele volgende alinea onder een kopje te zetten, namelijk het type contact en de invloed daarvan op adherence en effectiviteit. Daarmee was de eerste alinea van mijn eerste versie niet meer iCBT met exposure sessies maar iCBT met face-to-face contact assistentie. Dit zorgde er wel voor dat ik de non-face-to-face contact weer moest onderverdelen in nog twee subkopjes. Gelukkig kon ik in Starreveld (2nd ed., 2009) vinden hoe de opmaak er dan uit hoorde te zien.
5. Bij het schrijven van de tweede deelvraag kwam ik erachter dat ik wel soms twee soorten van begeleiding met elkaar vergelijk maar dan niet de drop-out rate per conditie heb uitgerekend. Dit moest dus nog wel gebeuren zodat ik ze kan vergelijken en wellicht conclusies kan trekken over de verschillende invloeden op adherence.
6. Na twee dagen rust, begon ik weer opnieuw met alles lezen. Om te zien of ik logisch schreef en ik snapte wat ik bedoelde zelf. Blijkbaar moest ik veel aan de discussie veranderen. En sommige overgangen waren heel raar.
7. Daarna heb ik het samen met iemand anders naast me door gelezen zodat hij feedback kon geven over wat vaag was. We zijn samen tot en met de Face-to-Face assistance.
8. Daarna heb ik de andere helft zelf nog een paar keer kritisch nagelezen en de referenties bij gewerkt.
Treating Social Anxiety Disorder with Internet-Based Cognitive Behavioral Therapy: the Influence of Assistance
Name: Rosanne Joosten Amount of Words: 5.352
Amount of Words (excluding Abstract): 5.204 Date: 22-06-2014
Abstract
A large amount of patients with social anxiety disorder (SAD) is not being treated. This could be the result of the unavailability of therapists, patients living in remote areas, or disorder-specific fear of social (therapeutic) contact. The currently new intervention internet-based cognitive behavioral treatment (iCBT) is a potential solution to these problems. The effectiveness of iCBT is already empirically been proven for multiple psychiatric disorders (Spek et al., 2007). But in some type of iCBT programs the therapist provides more assistance than in others. The influence of the amount of therapeutic contact on the effectiveness of, and adherence to iCBT in the treatment of SAD has not yet been discussed. This review focuses on the amount of contact and, if any, the type of contact. Research suggests that the effectiveness of iCBT is more effective when assisted. However, it seems that the adherence plays a part in this.
Index
1. Introduction
2. CAiCBT versus Unguided iCBT 2.1. Stepped-care assistance 3. CAiCBT and Types of Assistance
3.1. Face-To-Face Assistance 3.2. Non-Face-To-Face Assistance
3.2.1. Status of the communication partner. 3.2.2. Interactive contact. 3.2.3. Stepped-care assistance. 4. Discussion 4.1. Limitations 4.2. Future studies 5. References
Introduction
Social Anxiety Disorder (SAD) is a severe disorder with an unconscionably fear of specific or non-specific social situations (5th ed.; DSM-V; American Psychiatric Association, 2013). It is unfortunately a common disorder with a life-time prevalence of 7-15% in Western countries (Furmark, 2002). Clark and Wells (1995) conceived the cognitive behavioral model of social anxiety disorder to explain the disorder and its maintenance. According to this model, social anxiety is a result of problematic beliefs about one’s own social functioning and that of the social world, resulting in negative evaluations of social situations. People with SAD avoid social interactions, causing impossibility of falsification of negative assumptions, leading to the maintenance of SAD.
Recently, internet interventions are developed to treat psychiatric disorders (Barak et al., 2008; Spek et al., 2007). First, an advantage of therapy delivered through the internet is that less therapists are needed. Secondly, the distance between therapist and patient in unimportant for treatment (Berger et al., 2009). Thirdly, because the disorder-specific fear of SAD is social interaction (Newman et al., 2003) a treatment with less human contact could also work threshold lowering (Titov et al., 2009a). Internet-based interventions are proven to be effective in improving depressive and anxiety symptoms (Spek et al., 2007) suggesting that it is a potential ideal solution to the difficulties with treatment as usual.
The method from which internet-based treatment is derived is the treatment as usual cognitive behavioral therapy (CBT). With CBT maladaptive cognitions are restructured and avoidance is unlearned (Lydiard & Falsetti, 1995). An exposure element is used to achieve this (Lambert, 2013). Hereby patients are exposed to their feared situation and anxiety will decrease (Lambert, 2013). In internet-based CBT (iCBT), instead of a therapist-controlled exposure exercise, patients’ homework is to expose themselves to these feared situations (Andersson et al., 2006). Altogether, iCBT has been proven effective to treat patients with SAD (Tulbure, 2011).
There are different types of contact provided with iCBT program. For example, in some iCBT programs patients are assisted by a clinician (CAiCBT) while in others patients have no guidance. The assistance with iCBT could have two possible functions: to increase effectiveness of treatment and to increase adherence to treatment. The presence of a therapist with CAiCBT could be an in-session exposure element. This would mean that CAiCBT would lead to a better improvement of symptoms than unguided iCBT, hence increase effectiveness of treatment. The other possible function of the therapist’s presence is to be a motivational factor. This would mean patients would drop-out less, hence increase adherence to treatment. If motivating the patient is the only function of assistance, then there should be no difference found in improvements between CAiCBT and unguided iCBT. These two possible functions of assistance will be examined in this review.
In the first section CAiCBT is compared to unguided iCBT, with regards to adherence and effectiveness of treatment, by comparing improvements and drop-out rates between treatments. The second section will be on the different types of assistance with CAiCBT to see which ones cause improvements and or adherence. By making these comparisons the influence of assistance with iCBT on adherence to treatment and treatment effectiveness can be examined for patients with SAD.
CAiCBT versus Unguided iCBT
Before examining different types of assistance, the difference between CAiCBT and unguided iCBT regarding adherence and effectiveness has to be explored. If there is a difference in improvement, the assistance could be an in-session exposure. It would be unethical to prescribe unguided iCBT to patients with SAD, while knowing that effectiveness is greater in CAiCBT. But if there is a difference in drop-out rates, the assistance is a motivational factor. Then, it is also unethical to prescribe unguided iCBT, since adherence is more increased in CAiCBT. However, if it is found that there is no difference between CAiCBT and unguided iCBT in any of these domains, it would be recommendable to prescribe iCBT because more patients can be treated at once. By comparing CAiCBT with unguided iCBT, the preference of treatment for SAD with iCBT may be more empirical supported.
Titov and colleagues (2008c) randomly assigned participants to an unguided iCBT, CAiCBT, and a waiting-list control group. The assistance at CAiCBT was therapeutic contact via e-mail. Both iCBT groups had access to each an independent discussion forum. After treatment patients receiving CAiCBT were more improved on SAD measurements than patients who received unguided iCBT. Contrary to what previous studies have indicated (Spek et al., 2007), there was no difference between the patients on the waiting-list and patients receiving unguided iCBT. Suggesting the assistance with CAiCBT was an in-session exposure element. When results were further examined it was found that patients from the unguided iCBT group had a less lower completion of the program. In fact, the patients who did complete the program had improved much better. Indicating that the assistance with CAiCBT played a motivational role instead of an in-session exposure role.
Previous study seems to point out that the assistance in CAiCBT is an adherence stimulating factor instead of an in-session exposure factor in the treatment of SAD (Titov et al,. 2008c). To figure this issue out, there has to be a component that stimulates adherence but does not provide exposure to social interaction for both unguided iCBT and CAiCBT. One possible method in doing this is by automatic e-mails or short message services (SMS). To make a stronger distinction between the two types of contact, the assistance in CAiCBT may be increased. For example, the e-mails can be replaced by weekly phone-calls. In Titov and colleagues’ (2009b) study patients who were treated with CAiCBT received weekly phone-calls and access to a discussion forum. Patients who received
unguided iCBT received automatic e-mails and SMS’s and access to a discussion forum. There were patients on the waiting-list as well. When treatment had ended, patients from both treatment groups had similar improvements as patients receiving face-to-face CBT (McEvoy, 2007). Suggesting that the improvements were not due to the human contact in treatment. The adherence was greater with patients from the CAiCBT group compared to those from the unguided iCBT group. This points towards the suggestion that assistance was a stimulating adherence factor instead of an in-session exposure factor.
The addition of a discussion forum to both iCBTs could have any influence on the treatment, since contact through a discussion forum can also be an in-session exposure element or a motivational factor. Therefore, unguided iCBT may not be totally unguided in these previous studies. Furmark and colleagues (2009) provided patients with SAD either with CAiCBT or unguided iCBT, without access to a discussion forum. In the same study Furmark and colleagues (2009) did a trial where they compared unguided iCBT with and without augmentation of a discussion forum. This way it was explored if the addition of a discussion forum had any influence to treatment’s effectiveness or adherence of patients. The improvements were similar for both treatment groups. Thus the assistance provided in CAiCBT may not be in-session exposure as presumed. The drop-out rate was also similar. Furthermore in this study the assistance did not seem to have an influence on the treatment of SAD with iCBT.
Nordgreen and colleagues (2012) conducted a study with the same components. Patients also received e-mails when assigned to the CAiCBT group, were totally unguided in the unguided iCBT group, or were on the waiting-list. They found the same indifference between the improvements of patients, but the drop-out rate was much higher with the unguided iCBT patients than with the CAiCBT patients. This indicates that the assistance with CAiCBT is an adherence stimulating factor instead of an in-session exposure factor.
Stepped-care assistance
The presumption that social interaction in iCBT for SAD is an in-session exposure element seems to become less likely. However, the interaction in all studies was mandatory in CAiCBT and unavailable in unguided iCBT. There never was a contact method in between. That is, where contact is available and patients can choose themselves to make contact. This contact method is called stepped-care. There are a couple of reasons why mandatory assistance may have different effects than stepped-care assistance on adherence and effectiveness of treatment. Mandatory contact could be aversive, leading to a higher drop-out. Furthermore, if contact is mandatory, patients cannot avoid the exposure and this could lead to a higher effectiveness of treatment. Suggesting that when
contact is available instead of mandatory, the effectiveness can reduce and the adherence can increase.
Therefore the study of Berger and colleagues (2011) which compares the unguided iCBT, the CAiCBT, and the stepped-care iCBT when treating SAD is relevant when researching the influence of assistance. It was found that patients improved equally when contact was either absent, available, or mandatory present. The drop-out rate was somewhat lower with the unguided iCBT patients compared to the stepped-care iCBT and CAiCBT patients. However, there were no tests performed to prove its significance. 52% of the patients in the stepped-care iCBT contacted the therapist. This rate seems to state that not all patients need assistance to keep adherence to treatment. Overall, the results of this study suggest that assistance may not be an in-session exposure factor in the treatment of SAD with iCBT but a motivational factor for patients who need it.
In summary, contradictory results were found when unguided iCBT was compared with CAiCBT in the treatment of patients with SAD (Berger et al., 2011; Furmark et al., 2009; Nordgreen et al., 2012; Titov et al., 2008c, 2009b). In some studies CAiCBT was more effective than unguided iCBT, in others they caused the same improvements. What the reasons are for these inconsistencies is unclear. But when the drop-out rates were low the improvements were the same for patients receiving CAiCBT and unguided iCBT (Titov et al., 2008c). Indicating an adherence stimulating factor of assistance and not an in-session exposure element.
In this section there was not a distinction made between the types of assistance in CAiCBT. It is important to know if different types of assistance have different influences on adherence to treatment and effectiveness of treatment. When there is a difference, some types of assistance are more or less preferable to subjoin to treatment. That is why the types of assistance will be differentiated in the next section.
CAiCBT and Types of Assistance
The assistance of SAD patients with iCBT can be done in a variety of ways. To see which type of assistance is best for adherence to or effectiveness of treatment, different types must be distinguished. The first trivial separation is between face-to-face assistance and non-face-to-face assistance. For example, with iCBT the exposure exercises are ought to be done independently. These could be done wrong by patients, they hold on to safety behaviors or they do not perform the exposure exercises at all. Perhaps if therapist-controlled exposure sessions are added to iCBT the effectiveness increases. It is also possible that the therapist-controlled exposure sessions have influence on adherence to treatment. For patients can find the therapist-controlled exposure sessions to aversive and therefore withdrawal treatment. That is why a distinction must be made between face-to-face assistance and non-face-to-face assistance in iCBT for the treatment of SAD.
Face-To-Face Assistance
One way of face-to-face assistance is in the form of therapist-controlled exposure exercises. This way the patients perform the exercises properly which might increase the effectiveness of treatment. In the study that Andersson and colleagues (2006) conducted, participants with SAD either received CAiCBT with two therapist-controlled exposure sessions or they were on the waiting-list. On almost all social anxiety symptoms the patients were improved after treatment. However, when the improvements of patients in this study were compared with those from a meta-analysis, the addition of exposure sessions did not seem to be of a substantial value (Taylor, 1996). Tillfors and colleagues (2008) suggested two exposure sessions was too little to have a substantial influence on effectiveness. But even when they raised the amount of exposure sessions to five, there was no difference of effectiveness found between CAiCBT with or without therapist-controlled exposure sessions. Suggesting that this type of face-to-face contact has no influence on effectiveness of iCBT when treating patients with SAD.
In both studies the drop-out rates were low (6.25% and 5.26%) for patients receiving CAiCBT compared to the average drop-out rate of 19,7% in psychotherapy (Swift & Greenberg, 2012). Hence it seems this type of assistance has a good influence on adherence to treatment.
Research has shown that the addition of exposure sessions is not more effective in treating SAD than CAiCBT alone (Andersson et al., 2006; Tillfors et al., 2008). This seems to point out that CAiCBT is an effective stand-alone treatment for SAD. But there can only be concluded that the in-session exposure element is substantial vain if CAiCBT as stand-alone treatment is tested on its effectiveness when compared to a control condition. Nevertheless, the assistance did have a good influence on the adherence to treatment. By comparing CAiCBT with, for example, a waiting-list would help answering the question what type of therapeutic contact is necessary for the adherence to and effectiveness of the treatment of SAD.
Non-Face-To-Face Assistance
Most studies about CAiCBT are with non-face-to-face assistance, since the purpose of a self-help program like iCBT is to minimize therapeutic time spend per patient (Botella & García-Palacios, 1996). When the assistance is non-face the time is up to 13 times smaller than with face-to-face therapy (Andrews et al., 2011). However, it is not clear if this non-face-to-face-to-face-to-face assistance can be an in-session exposure element, since studies have shown that people with social anxiety feel more comfortable with a non-face-to-face interaction than with a face-to-face interaction (Pierce, 2009). This seems to indicate that this way of communicating has less of an in-session exposure element, for in an exposure situation there is an increase of anxiety and tendency to avoid. Since they feel more
comfortable with a non-face-to-face interaction this points to a less in-session exposure element. If the non-face-to-face contact can still influence the adherence to treatment will also be explored. Therefore there will be discussed if different types of non-face-to-face interaction are of any additive value to the effectiveness of or adherence to iCBT in the treatment of SAD.
Status of the communication partner. The first potential factor to be of influence on adherence to treatment and effectiveness of treatment is the status of the communication partner. Firstly, all sorts of assistance in CAiCBT can be done by a therapist as well as by a non-therapist. The assistance of patients with iCBT for example involve giving feedback on homework or progress, motivating the patient, giving explanation on the execution of the exercises, or express appraisal. These activities can be done by any person with knowledge of the program and its technicalities. Secondly, this separation between therapeutic and non-therapeutic contact is made because people are found to be much more likely to assume something is true if authorities say it is (“appeal to authority”, Mook, 2001, p5-7). If this is true for the non-face-to-face interaction with CAiCBT then the effectiveness and adherence will be greater for patients who are assisted by a therapist compared to those assisted by a non-therapist. To test if the status of the communication partner has influence on adherence or effectiveness of treatment, de distinction must be made between a therapist and a non-therapist giving the assistance.
First it will be examined if the therapeutic status increases effectiveness of treatment or adherence to treatment, because the operation of the appeal to authority has to be examined on its presence. The most direct way of therapeutic social interaction without face-to-face contact is through telephonic contact. In Carlbring and colleagues’ (2007) study participants with SAD received weekly therapeutic phone-calls besides the feedback via e-mail on their homework assignments of CAiCBT. After treatment participants receiving CAiCBT were more improved on social anxiety symptoms than participants on the waiting-list. Suggesting that therapeutic contact is a substantial addition to iCBT. The drop-out rate was as low as 6.66% for the participants receiving CAiCBT. But there cannot be drawn a valid conclusion out of this study about the effectiveness for there was no active control group, so there is not taken account for spontaneous recovery.
Titov and colleagues (2008a) did include a control group when examining the influence of therapeutic contact on treatment effectiveness and patient’s adherence. Therefore patients with SAD received either CAiCBT with e-mail contact and discussion forum contact, or discussion forum access without CAiCBT. The discussion forum access could represent non-therapeutic assistance. The way the groups are differentiated, the effect of non-therapeutic assistance has is taken into account. Compared to patients with only discussion forum access, patients improved on social anxiety symptoms with CAiCBT. This seems to suggest that the therapeutic contact had influence on the effectiveness of treatment. The drop-out rate was somewhat higher than of previous studies, 12% of
the participants did not complete the whole study. Compared to the average drop-out rate of 19,7% in psychotherapy (Swift & Greenberg, 2012) this is still low. However since in previous studies the contact was also with a therapist, the difference in drop-out rate between these two does not seem to be caused by the status of the communication partner but by something else. Other reasons for difference in adherence will be discussed later.
It cannot be concluded that the effectiveness is due to the therapeutic contact, since these are studies where the contact was always with a therapist (Carlbring et al., 2007; Titov et al., 2008). Maybe it was not that much the extra therapeutic support but more support in general that caused the increase in effectiveness or influence the adherence (Weiss et al., 2013) . Titov and colleagues (2009c) executed a study where participants received CAiCBT with non-therapeutic phone-calls. As the control condition they provided participants CAiCBT with access to a discussion forum. In this study the discussion forum was moderated by a clinician. By comparing a non-therapeutic assistance form with a therapeutic assistance form, the influence of status of the communication partner is better investigated. Participants from both interventions improved on social anxiety symptoms without any difference between the interventions. Indicating that it is not the status of the communication partner, but more support in general that influences effectiveness. The drop-out rate for the participants receiving non-therapeutic telephonic assistance was 7.00%. The drop-out rate for the participants receiving therapeutic discussion forum assistance was 8.00%. This seems to indicate that the status of the communication partner is of no influence on the adherence to treatment of patients with SAD and also not on the effectiveness of CAiCBT.
The available research showed that both therapeutic and non-therapeutic assistance cause improvements of social anxiety symptoms in patients with SAD receiving CAiCBT (Carlbring et al., 2007; Titov et al., 2008a; Titov et al., 2009c). Thus, the status of the communication partner does not seem to influence the effectiveness of treatment. The adherence to treatment did also vary much across studies with only therapeutic assistance (Carlbring et al., 2007; Titov et al., 2008a; Titov et al., 2009c). The reason for these differences will be discussed later. For now, the status of the communication partner does not seem to play a part in the effectiveness of treatment and there cannot be drawn a valid conclusion about the influence of the status of the communication partner on adherence to CAiCBT for the treatment of SAD either.
Interactive assistance. Another distinction that must be made is between the low and high degree of interactiveness of the assistance method. That is how long it takes to receive a reaction and respond to a message. Having direct and thus interactive contact with someone seems to be more frightening for patients with SAD than a less interactive contact (Pierce, 2009). For example, receiving response of an e-mail and sending a reaction to an e-mail takes much more time than interacting by phone. This makes e-mail a less interactive medium of assistance than telephonic
assistance. A high degree of interactiveness seems to be aversive, thus demotivating for adherence and stimulating for effectiveness. The results of the study of Carlbring and colleagues (2007) suggested that the less interactive form of e-mail is enough for improving social anxiety symptoms with CAiCBT. But there was also a phone-call added so this cannot be concluded, since this a more interactive form of assistance. In the previous study of Titov and colleagues (2008a) patients had access to a discussion forum plus they received the phone-calls. These types of assistance are both of a high degree of interactiveness. They concluded that the high degree of interactiveness of assistance had a good influence on effectiveness of treatment and medium influence on the adherence to treatment (Titov et al., 2008a). Furthermore, the distinction between interactive and less interactive must be made to conclude if the degree of interactivity is of any substantial influence on the effectiveness of and adherence to CAiCBT when treating SAD.
Titov and colleagues (2008b) executed a study where participants received assistance in the form of e-mail contact with iCBT or were on the waiting-list. That way they could investigate what the influence was of the less interactive contact method e-mail to iCBT. Results show that the treated participants were more improved on social anxiety symptoms compared to those on the waiting-list. The drop-out rate of the participants receiving CAiCBT was 11.63%. The results suggest that the less interactive contact method e-mail has a positive influence on the effectiveness of treatment and a medium influence on the adherence to treatment compared to previous studies (Andersson et al., 2006; Carlbring et al., 2007; Tillfors et al., 2008; Titov et al., 2009c).
This study is only examined the less interactive assistance method e-mail. In a more complex but more meaningful way the comparison between a less interactive and a more interactive form of assistance is further investigated. In the study of Carlbring and colleagues (2006) patients with SAD received CAiCBT with once a week an e-mail from the therapist. These patients were recruited from a waiting-list of another study (Carlbring et al., 2007). Directly after treatment patients receiving CAiCBT were more improved on SAD symptoms than patients on the waiting-list. There were no participants who drop-out of the study.
Then 30 months after treatment of the Carlbring and colleagues’ (2007) study when participants were measured again. For a more clearer image of participant recruitment and final comparison see Figure 2. The final comparison made is between the additional power of an once a weekly e-mail or an once a weekly e-mail and phone-call as augmentation to treatment (Carlbring et al., 2009). Here the interactive contact phoncall is put opposite to the less interactive contact e-mail. From results can be concluded that the degree of interactivity is not of influence on treatment outcome. The drop-out was higher with the participants not receiving the interactive contact method phone-calls, 6.66%, than those who did, 0%. Suggesting that the degree of interactivity has some
Study 3.
Carlbring et al., 2009: 30 months follow-up
influence on the adherence to iCBT for the treatment of SAD but not on the effectiveness of treatment.
Stepped-care assistance. Before there can be made a conclusion about the influence of the degree of interactiveness of assistance on adherence to treatment and effectiveness of treatment, stepped-care must be examined as well. With the stepped-care assistance of iCBT, the control over the assistance is owned by the patient. This indicates a higher degree of interactiveness with stepped-care iCBT than with CAiCBT. It is important to know if this degree of interactiveness is of influence on the adherence to and effectiveness of the treatment of SAD. In the study of Berger, Hohl and Caspar (2009), participants received either stepped-care assistance with iCBT or were on the waiting-list. In the iCBT program there was a reminder after every module that participants could contact the therapist via e-mail any time they wanted. On average patients contacted the therapist 5.5 times during the 10 weeks of the study, the minimal was 0 and the maximal was 16. This is on average less than once a week. The results show that participants did improve more when receiving stepped-care iCBT compared those on the waiting-list. The drop-out rate was 9.68%. This study indicated that when contact is not mandatory, patients can still improve. Suggesting that some patients improve and complete treatment when assistance with iCBT is available and others do not need this assistance for the same accomplishments.
Figure 2. Recruitement of Participants and Final Comparison.
CaICBT = Clinical Assisted Internet-Based Cognitive Behavioral Treatment Study 1. Carlbring et al., 2007 CaICBT + Phone-Calls + E-Mails Waiting-list (Study 2. Carlbring et al., 2006)
Discussion
The studies used in this review have shown contradictory results regarding the effectiveness of assistance with iCBT for the treatment of SAD. The proposed in-session exposure element did not seem to be present in most studies. That is, CAiCBT was only superior to unguided iCBT in one study (Titov et al., 2008c) and in most studies the improvements caused by treatment were the same for these two contact methods (Berger et al., 2011; Furmark et al., 2009; Nordgreen et al., 2012; Titov et al., 2009b). The reason for this different outcome seems to be in the adherence to treatment, since patients who completed this study improved equally (Titov et al., 2008c). In almost all studies was found that adherence to treatment was lower with patients receiving unguided iCBT. The pre- to post-treatment ratio and the drop-out rate of the studies are found in Table 1. for all discussed studies. The percentage of full completion varied from 77.19- 100% (M=91,23%) for CAiCBT and from 59.38-97.50%(M=80.17%) for unguided iCBT. There were two studies done with a stepped-care contact method, where the completion percentage was on average 89.61%. These rates seem to indicate that patients have more adherence to a CAiCBT program than to an unguided iCBT program and patients’ adherence to the stepped-care method of iCBT is somewhere in between.
When taking a closer look at the type of assistance, there was no substantial difference found between face-to-face contact and non-face-to-face contact in adherence or treatment effectiveness (Andersson et al., 2006; Berger et al., 2009; Carlbring et al., 2006, 2007, 2009;Tillfors et al., 2008; Titov et al., 2008a,b,2009c). Both contact methods caused improvements and, compared to the average 19,7% of drop-out of psychotherapy (Swift & Greenberg, 2012), increased adherence. Even when non-face-to-face contact is being divided into different types of assistance there was no substantial difference found between status of the communication partner (Carlbring et al., 2007; Titov et al., 2008a, 2009c) or degree of interactivity (Berger et al., 2009; Carlbring et al., 2006, 2007, 2009;Titov et al., 2008a,b,2009c). In conclusion the type or amount of contact does not seem to have an influence on the effectiveness of iCBT and a medium influence on the adherence to iCBT in the treatment of patients suffering from SAD.
Limitations
But there are some limitations to these conclusions. Firstly, although the mean percentage of completion of CAiCBT was high, it still varied much across studies. This can be due to a couple of differences. As is noticeable, research about iCBT for SAD is executed by a couple of different research teams. These research teams are from Sweden (Andersson et al., 2006; Tillfors et al., 2008; Carlbring et al., 2006, 2007, 2009; Furmark et al., 2009; Nordgreen et al., 2012), Australia (Titov et al., 2008a, 2008b, 2008c, 2009a, 2009b, 2009c) and Switzerland (Berger et al., 2009, 2011). When the results of each research team are compared, they are not that variable anymore. This could be due
to the fact that each research team have developed a slightly different iCBT program to treat SAD. These program can have different effects on the adherence to treatment. Since the research teams are from another country they have treated patients from different nationalities. One culture can react differently to a treatment than another culture (Eshun, Chang, & Owusu, 1998), possibly causing different outcome measurements as effectiveness and adherence for the different research teams. In these two ways, the results of the three research teams could vary and are therefore not completed legitimate to compare with each other.
Pre- to Post-Treatment Ratio and Drop-Out Rate per Discussed Study.
Secondly, it cannot validly be concluded that more assistance leads to better adherence because a drop-out of a study is not the same as a lack of adherence to treatment. According to Rizvi, Vogt, and Resick (2009), it is called treatment adherence when patients complete 75% of the modules/chapters. The only study where this criteria is used is in Nordgreen and colleagues’ (2012). From all other studies the drop-out rate is used instead of the adherence. The reason for dropping-out can be different than the reason of low adherence of the patient to treatment. A patient can for example forget to complete questionnaires, or they got disconnected from the internet causing them
Table 1.
Study
Pre- to Post-Treatment Ratio CAiCBT Unguided Stepped-care
Drop-Out Rate
Andersson et al. (2006) 30/32 - - 6.25%
Tillfors et al. (2008) 18/19 - - 5.26%
Carlbring et al. (2007) 28/30 - - 6.66%
Titov et al. (2009c) 76/85 - - 10.59%
Titov et al. (2008a) 44/50 - - 12.00%
Titov et al. (2008b) 38/43 - - 11.63% Carlbring et al. (2006) 26/26 - - 0% Carlbring et al. (2009) 57/60 - - 5.00% Berger et al. (2009) - - 28/31 9.68% Titov et al. (2008c) 30/32 27/31 - 6.25%; 12.90% Titov et al. (2009b) 72/84 75/84 - 19%; 32% Furmark et al. (2009) 39/40 39/40 - 2.50%; 2.50% Nordgreen et al. (2012) 149/149 57/96 - 0%; 41.62% Berger et al. (2011) 24/27 23/27 24/27 11.11%; 14.1%;11.1%
to drop-out of analysis. Thus there cannot be drawn valid conclusions about adherence and CAiCBT out these studies.
Future Studies
The only study that used the criteria of adherence to treatment from Rizvi and colleagues’ (2009) was that of Nordgreen and colleagues (2012). Interestingly, they also measured for reasons of adherence to treatment. Results showed that when patients were unguided, higher credibility ratings of the treatment were associated with higher adherence to treatment. Suggesting that if the program is more credible, or presented more credible, the program could be more suitable for an unguided format of iCBT. Since there was a higher drop-out rate in Titov and colleagues’ (2009b) study when proved unguided, this seems to indicate that their program is less credible-looking. However, there can be other reasons as well for low adherence. A study about internet-based therapy for anxiety disorders, have already found some predictors for pretreatment attrition and treatment withdrawal (AL-Asadi, Klein and, Meyer, 2014). However, the disorder-specific fear of SAD for social interaction could have an important and different influence on the adherence to treatment. Clearly, research is needed about the reasons for low adherence to iCBT in the treatment of SAD.
Furthermore, in these previous studies about treatment adherence there was no stepped-care condition of iCBT, so there cannot be drawn conclusions about this contact method and adherence. Plus to draw a more likely conclusion there has to be more than one study about patients adherence to iCBT treatment per contact method. There could be multiple reasons why patients keep adherence to the treatment. Perhaps there is a type of contact method more suitable for each patient. This is an area that still needs to be explored by future studies.
7. References
AL-Asadi, A.M., Klein, B., Meyer, D. (2014). Pretreatment Attrition and Formal Withdrawal During Treatment and Their Predictors: An Exploratory Study of the Anxiety Online Data. J Med
Internet Res 16(6):e152.
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.
Andersson, G., Carlbring, P., Holmström, A., Sparthan, E., Furmark, T., Nilsson-Ihrelt, E., …Ekselius, L. (2006). Internet-Based Self-Help with Therapist Feedback and In Vivo Group Exposure for Social Phobia: A Randomized Controlled Trial. Journal of Consulting and Clinical Psychology,
74, 677-685.
Andrews, G., Davies, M., & Titov, N. (2011). Effectiveness randomized controlled trial of face to face versus Internet cognitive behavior therapy for social phobia. Australian and New Zealand
Journal of Psychiatry, 45, 337-340.
Barak, A., Hen, L., Boniel-Nissim, M., & Shapira, N. (2008). A comprehensive review and a meta- analysis of the effectiveness of internet-based psychotherapeutic interventions. Journal of
Technology in Human Services, 26, 109-160.
Berger, T., Caspar, F., Richardson, R., Kneubügler, B., Sutter, D., & Andersson, G. (2011). Internet- based treatment of social phobia: A randomized controlled trial comparing unguided with two types of guided self-help. Behaviour Research and Therapy, 49, 158-169.
Berger, T., Hohl, E., & Caspar, F. (2009). Internet-Based Treatment for Social Phobia: A Randomized Controlled Trail. Journal of Clinical Psychology, 65, 1021-1035.
Botella, C., & García-Palacios, A. (1996). La estructuración de la terapia, la utilización de manuales de autoayuda y el costebeneficio terapéutico en el tratamiento del trastorno de pánico. Análisis
y Modificación de Conducta, 22, 115–136.
Carlbring, P., Bergman Nordgren, L., Furmark, T., & Andersson, G. (2009). Long-term outcome of Internet-delivered cognitive-behavioural therapy for social phobia: A 30-month follow-up.
Behaviour Research and Therapy, 47, 848-850.
Carlbring, P., Furmark, T., Steczkó, J., Ekselius, L., & Andersson, G. (2006). An open study of Internet- based bibliotherapy with minimal therapist contact via email for social phobia. Clinical
Psychologist, 10, 30-38.
Carlbring, P., Gunnarsdóttir, M., Hedensjö, L., Andersson, G., Ekselius, L., & Furmark, T. (2007). Treatment of social phobia: randomized trial of internet-delivered cognitive-behavioural therapy with telephone support. British Journal of Psychiatry, 190, 123-128.
Clark, D.M., & Wells, A. (1995). A cognitive model of social phobia. In R.G. Heimberg, M.R. Liebowitz, D.A. Hope, & F.R. Schneier (Eds.), Social phobia: Diagnoses, assessment, and treatment (pp. 69–93). New York: Guilford Press.
Eshun, S., Chang, E.C., & Owusu, V. (1998). Cultural and gender differences in responsesto depressive mood: a study of college students in Ghana and the U.S.A. Pergamon, 24, 581-583.
Furmark, T. (2002). Social phobia: overview of community surveys. Acta Psychiatrica Scandinavica,
105, 84-93.
Furmark, T., Carlbring, P., Hedman, E., Sonnenstein, A., Clevberger, P., Bohman, B., …Andersson, G. (2009). Guided and unguided self-help for social anxiety disorder: randomised controlled trial. The British Journal of Psychiatry, 195, 440-447.
Lambert, M.J. (2013). Handbook of Psychotherapy and Behavior Change. (6th ed.). New Jersey: John Wiley & Sons, Inc.
Lydiard, R.B., Falsetti, S.A., (1995). Treatment options for social phobia. Psychiatr. Ann. 25, 570–576. McEvoy P. (2007). Effectiveness of a cognitive behavioural group therapy for social phobia in a
community clinic: a benchmarking study. Behavioral Research Therapy, 45, 3030-3040. Mook, D.G. (2001). Psychological Research: The Ideas Behind The Methods. (1st ed.). New York: W.W.
Norton & Company, Inc.
Newman, M.G., Erickson, T., Przeworski, A., & Dzus, E. (2003). Self-help and minimal contact therapies for anxiety disorders: Is human contact necessary for therapeutic efficacy?
Journal of Clinical Psychology, 59, 259–274.
Nordgreen, T., Havik, O.E., Öst, L.G., Furmark, T., Carlbring, P., & Andersson, G. (2012). Outcome predictors in guided and unguided self-help for social anxiety disorder. Behaviour Research
and Therapy, 50, 13-21.
Pierce, T. (2009). Social anxiety and technology: Face-to-face communication versus technological communication among teens. Computers in Human Behavior, 25, 1367-1372.
Rizvi, S. L., Vogt, D. S., & Resick, P. A. (2009). Cognitive and affective predictors of treatment outcome in cognitive processing therapy and prolonged exposure for posttraumatic stress disorder.
Behaviour Research and Therapy, 47, 737-743.
Spek, V., Cuijpers, P., Nyklícek, I., Riper, H., Keyzer, J. and Pop, V. (2007). Internet-based cognitive behaviour therapy for symptoms of depression and anxiety: A meta-analysis. Psychological
Medicine: A Journal of Research in Psychiatry and the Allied Sciences, 37, 319–328.
Swift, J.K., & Greenberg, R.P. (2012). Premature Dicontinuation in Adult Psychotherapy: A Meta- Analysis. Journal of Consulting and Clinical Psychology, 80, 547-559.
Taylor, S. (1996). Meta-analysis of cognitive– behavioral treatments for social phobia. Journal of
Behavior Therapy and Experimental Psychiatry, 27, 1–9.
Tillfors, M., Carlbring, P., Furmark, T., Lewenhaupt, S., Spak, M., Eriksson, A., ...Andersson, G. (2008). Treating university students with social phobia and public speaking fears: internet delivered self-help with or without live group exposure sessions. Depression and Anxiety, 25, 708-717. Titov, N., Andrews, G., Johnston, L., Schwencke, G., & Choi, I. (2009a). Shyness programme: longer
term benefits, cost-effectiveness, and acceptability. Australian and New Zealand Journal of
Psychiatry, 43, 36-44.
Titov, N., Andrews, G., Choi, I., Schwencke, G., & Johnston, L. (2009b). Randomized controlled trial of web-based treatment of social phobia without clinician guidance. Australian and New
Zealand Journal of Psychiatry, 43, 913-919.
Titov, N., Andrews, G., Choi, I., Schwencke, G., & Mahoney, A. (2008c). Shyness 3: randomized controlled trial of guided versus unguided Internet-based CBT for social phobia. Australian
and New Zealand Journal of Psychiatry, 42, 1030-1040.
Titov, N., Andrews, G.,& Schwencke, G. (2008b). Shyness 2: treating social phobia online: replication and extension. Australian and New Zealand Journal of Psychiatry, 42, 595-605.
Titov, N., Andrews, G., Schwencke, G., Drobny, J., & Einstein, D. (2008a). Shyness 1: distance treatment of social phobia over the Internet. Australian and New Zealand Journal of
Psychiatry, 42, 585-594.
Titov, N., Andrews, G., Schwencke, G., Solley, K., Johnston, L., & Robinson, E. (2009c). An RCT comparing effect of two types of support on severity of symptoms for people completing Internet-based cognitive behaviour therapy for social phobia. Australian and New Zealand
Journal of Psychiatry, 43, 920-926.
Tulbure, B.T. (2011). The efficacy of Internet-supported intervention for social anxiety disorder: A brief meta-analytic review. Procedia – Social and Behavioral Sciences, 30, 552-557. Weiss, J.B., Berner, E.S., Johnson, K.B., Giuse, D.A., Murphy, B.A., & Lorenzi, N.M. (2013).
Recommendations or the design, implementation and evaluation of social support in online communities, networks, and groups. Journal of Biomedical Informatics, 46, 970-976.
Research Proposal Verwerking feedback:
1. Het moeilijkste was weer de sample size, ik wist dat dit en hoog aantal is, maar het kwam toch echt uit power analysis. Ik zou analysis kunnen uitvoeren over de samples van andere
onderzoeken en die proefpersonen vragenlijsten laten invullen. Maar als ik me houd aan het 75% voltooide criteria van Rizvi en collega’s, weet ik niet hoeveel proefpersonen ik daarvoor heb. Daarnaast zijn er maar weinig studies gedaan die stepped-care ook gebruiken als contact methode. Wellicht moet ik de vraag anders stellen, niet waarom adherence laag is, maar waarom adherence hoog is. Uiteindelijk wil ik zien welke factoren er samenhangen met
adherence, dus dat kan ook via die kant. Aangezien ik de analyse over de afvallers (die 19,7%) en de aanhouders (die 80,3%) doe, hoef ik eigenlijk geen rekening te houden met die percentages, ik wil gewoon zien of er via een regressieanalyse voorspellers zijn voor adherence. Beiden groepen mogen daarbij meedoen. Dan is 279 het hoogste getal waar ik mee moet werken voor de analyses om een effect size van 0.15 te krijgen.
2. Met het verbeteren van het verslag begon in echter gewoon weer vanaf het begin. Hopelijk kan ik het in een dag helemaal doorlezen en verbeteren en dan een andere dag nog een keer. Als ik het in een dag doe is de structuur iets mooier omdat ik dan niet vergeten ben hoe ik mijn betoog schreef.
3. Ik heb de methode sectie meer uitgebreid en sommige dingen meer moeten verklaren. Voor zover ik het nu zelf las, is het redelijk goed.
Treating Social Anxiety Disorder with Internet-Based Cognitive Behavioral Therapy: the Influence of the Amount of Assistance on Drop-Out Rates and its Reasons
Social anxiety disorder (SAD) is a disorder with a high life-time prevalence in western countries (7-15%; Furmark et al., 2002). The disorder is difficult to treat because of the disorder specific fear of social interaction (Newman et al., 2003). Internet-based cognitive behavioral therapy (iCBT) is a potential solution to this problem (Titov et al., 2009a). With iCBT there is less therapeutic contact then with the treatment as usual, cognitive behavioral therapy (CBT). Research has proven that iCBT is an effective intervention of treating psychological disorder (Spek et al., 2007). Programs of iCBT always consist education about the disorder and treatment, cognitive restructuring exercises, exposure exercises, and relapse prevention. Some programs include access to a discussion forum, weekly phone-calls, or e-mails automated or not. Some of these additions are used as assistance but they are time-consuming. There can be three contact methods distinguished in iCBT. First there is unguided iCBT, where there is no contact between patient and therapist. Second there is clinician assisted iCBT (CAiCBT), where patients are assisted by a therapist in one of the previously mentioned ways. Last there is stepped-care iCBT, where it is possible for patients to seek contact with the therapist but not mandatory. CAiCBT is proven to be most effective contact method in treating SAD with iCBT (Berger et al., 2011; Furmark et al., 2009; Nordgreen et al., 2012; Titov et al., 2008, 2009a,b).
However, the drop-out rates are variable between these contact methods of assistance with iCBT for SAD. The percentage of full completion varies from 77.19- 100% for CAiCBT and from 59.38-97.50% for unguided iCBT. Studies about the stepped-care contact method reported an average completion percentage of 89.61% (Andersson et al.,2006; Berger et al., 2009, 2011; Carlbring et al, 2006, 2007, 2009; Furmark et al., 2009; Nordgreen et al., 2012; Tillfors et al., 2008; Titov et al., 2008, 2009a, 2009b). This variance in adherence can result in the seemingly high effectiveness of CAiCBT. Because when focused on the results of the other studies, the participants who completed treatment with an unguided or stepped-care contact method of iCBT improved just as much (Berger et al., 2011; Furmark et al., 2009; Nordgreen et al., 2012; Titov et al., 2008, 2009a,b). Since unguided iCBT is much more beneficial (Titov et al., 2009a), it is important to find out what the reason is for the variability of adherence between and within these contact methods before concluding that CAiCBT is the most preferable intervention. Unfortunately, this variability of adherence to iCBT has only been studied once for SAD only between unguided iCBT and CAiCBT (Nordgreen et al., 2012).
Nordgreen and colleagues (2012) have found some predictors for adherence to treatment for patients undergoing iCBT and CAiCBT for SAD. However, they did not have a stepped-care iCBT group. There is only one research team which examined stepped-care iCBT for its effectiveness (Berger et al., 2009;2011). Since this contact method has been proven to be more cost-efficient than
CAiCBT(Berger et al., 2011), associated with higher adherence than unguided iCBT (Berger et al., 2009;2011) and as effective as both CAiCBT and unguided iCBT (Berger et al., 2009;2011) it is the potential most favorable contact method for SAD. This present study is a exploratory one for the predictors of adherence to stepped-care iCBT for SAD. There is also an open-ended question added for patients who drop-out. This way other not studied reasons for dropping-out can be explored. These are also exploratory and without hypothesis.
From previous research there are some hypotheses already proven (Nordgreen et al., 2012). These hypotheses will be tested again on their accuracy in this study. For results is expected that the lowest adherence will be in the unguided iCBT group (Titov et al., 2008, 2009b; Furmark et al., 2009; Nordgreen et al., 2012; Berger et al., 2011). The highest adherence is expected to be in the CAiCBT group (Titov et al., 2008, 2009b; Furmark et al., 2009; Nordgreen et al., 2012; Berger et al., 2011). The adherence of the stepped-care iCBT group is expected to be between these two rates (Titov et al., 2008, 2009b; Furmark et al., 2009; Nordgreen et al., 2012; Berger et al., 2011).
Reasons for adherence that will be examined are at least program’s credibility, pre-treatment symptom severity, and satisfaction with the treatment. Firstly, it is expected that main reason for low adherence in the unguided iCBT group is low credibility of the program (Nordgreen et al., 2012). The number of participants giving this reason will also be greater in this group compared to the other two treatment groups (Nordgreen et al., 2012). Furthermore, it is expected that high credibility is
associated with high adherence to the program (Nordgreen et al., 2012). Secondly, the main reason for low adherence in the CAiCBT group is expected to be associated with low treatment satisfaction (Nordgreen et al., 2012). Therefore it is also expected that high adherence is associated with high satisfaction (Nordgreen et al., 2012). Thirdly, there are no specific hypothesis about the reason for high or low adherence in the stepped-care iCBT group, for it is not yet examined. It is not expected that there is a correlation between pre-treatment symptom severity and adherence.
With regards to the effectiveness of treatment, it is expected that participants in all treatment groups improve on their social anxiety symptoms when drop-out is taking into account. These results are expected to maintain at follow-up. The improvements are expected to be greater than those of the waiting-list control group. There is not a difference between the three treatment groups expected.
Method
Participants
Participants will be recruited via the internet websites of psychological agencies. They will be screened with the Structured Clinical Interview for DSM–IV (SCID; First, Spitzer, Gibbon, & Williams, 1995; mentioned in Andersson et al., 2006) for their SAD diagnosis. They will also be measured on
suicidal thoughts, for it is not intended that participants are suicidal when participating in this study. This will be done with the Montgomery Åsberg Depression Rating Scale (MADRS-S; Svanborg & Åsberg, 1994; mentioned in Andersson et al., 2006). If participants have SAD diagnosed as most dominant disorder and score below 31 on the MADRS-S, they will be included in study. Via the program GPower 3.1 was found that there have to be 279 participants for an effect size of 0.15.
Material
Program. The iCBT program used is that of Andersson and colleagues (2006), because it has
been proven to be effective multiple times (Andersson et al., 2006, 2012; Carlbring et al., 2007; Furmark et al., 2009; Hedman et al., 2011). For a complete impression of the program, see those studies. In short, the program consists of nine modules, covering subjects as the cognitive model of SP, cognitive restructuring, exposure exercises, and attention training. Participants will have access to the program for ten weeks.
Questionnaires. Symptoms of social anxiety will be measured with the self-report version of
the Liebowitz Social Anxiety Scale (LSAS-SR; Baker, Heinrich, Kim, & Hofmann, 2002; Liebowitz, 1987) and Social Interaction Anxiety Scale (SIAS; Mattick & Clarke, 1998) and the SPSQ (Furmark et al., 1999). Treatment satisfaction is being measured with the Client Satisfaction Questionnaire (CSQ-8; Attkisson & Zwick, 1982). Credibility of the program is being measured with the
Credibility/expectancy scale (C-scale; Borkovec & Nau, 1972). This questionnaire also includes expectancy of improvements of patients, which can be another predictor of improvements for the explanatory part of this study. Adherence was as Rizvi, Vogt, and Resick (2009) defined as 75% completion of the 9 modules.
Procedure
The 279 participants will be randomly assigned with the website www.random.org to the CAiCBT, unguided iCBT, stepped-care iCBT and waiting-list control group. They will be screened for exclusion criteria and send the questionnaires via e-mail. Then the treatment starts for each of the treatment groups. The participants from the unguided iCBT group do not have any contact with a therapist during this study, when they do, they will be excluded from effectiveness analysis to test if the effectiveness is similar between contact methods when drop-out is taken into account.
Participants in the unguided iCBT group will not have the ability to contact a study-related therapist during the ten weeks of treatment for this could have effect to the adherence or effectiveness of the treatment. If they do, they are excluded from effectiveness analysis and send questionnaires about the reason for seeking a therapist. In the CAiCBT group a therapist will call the participants once a week, provide feedback on the homework via e-mail, and respond to e-mails within 3 days. In the
stepped-up iCBT participants have the possibility to contact a therapist by e-mail. After each exercise or module they are reminded to this fact. Every contact with another non-study related therapist, will be causing exclusion of effectiveness analysis. Reasons for seeking another therapist or lack of returning questionnaires will be asked via the.
Outcome Measurements and Data Analysis
Effectiveness
Before treatment participants will be measured on social anxiety symptoms with the SPSQ, SIAS, and the LSAS-SR. These questionnaires will also be administered at post-treatment and at one-year follow-up. To check for the effectiveness of treatment and the difference in effectiveness of treatment between the conditions, these results will be analyzed with a multivariate analysis of the variance (MANOVA).
Adherence
To check if the adherence correlates with treatment satisfaction the CSQ-8 will be
administered at treatment or at time of dropping-out. The C-scale will be administered at post-treatment or at time of dropping-out to check for correlations with adherence. The results of the questionnaires will be checked on dependence between conditions and reasons with a regression analysis. There will also be done an analysis of variance (ANOVA) over the number of drop-outs, these will be compared per treatment group.
References
Attkisson, C. C., & Zwick, R. (1982). The client satisfaction questionnaire. Psychometric properties and correlations with service utilization and psychotherapy outcome. Eval Program Plann, 5, 233- 237.
Andersson, G., Carlbring, P., Holmström, A., Sparthan, E., Furmark, T., Nilsson-Ihrelt, E., …Ekselius, L. (2006). Internet-Based Self-Help with Therapist Feedback and In Vivo Group Exposure for Social Phobia: A Randomized Controlled Trial. Journal of Consulting and Clinical Psychology,
74, 677-685.
Andersson, G., Paxling, B., Wiwe, M., Vernmark, K., Felix, C.B., Lundborg, L., …Carlbring, P. (2012). Therapeutic alliance in guided internet-delivered cognitive behavioural treatment of
depression, generalized anxiety disorder and social anxiety disorder. Behaviour Research and
Therapy, 50, 544-550.
self-report instrument: A preliminary psychometric analysis. Behaviour Research and
Therapy, 40, 701–715.
Berger, T., Caspar, F., Richardson, R., Kneubügler, B., Sutter, D., & Andersson, G. (2011). Internet- based treatment of social phobia: A randomized controlled trial comparing unguided with two types of guided self-help. Behaviour Research and Therapy, 49, 158-169.
Berger, T., Hohl, E., & Caspar, F. (2009). Internet-Based Treatment for Social Phobia: A Randomized Controlled Trail. Journal of Clinical Psychology, 65, 1021-1035.
Borkovec, T. D., & Nau, S. D. (1972). Credibility of analogue therapy rationales. Journal of Behavior
Therapy and Experimental Psychiatry, 3, 257-260.
Carlbring, P., Bergman Nordgren, L., Furmark, T., & Andersson, G. (2009). Long-term outcome of Internet-delivered cognitive-behavioural therapy for social phobia: A 30-month follow-up.
Behaviour Research and Therapy, 47, 848-850.
Carlbring, P., Furmark, T., Steczkó, J., Ekselius, L., & Andersson, G. (2006). An open study of Internet- based bibliotherapy with minimal therapist contact via email for social phobia. Clinical
Psychologist, 10, 30-38.
Carlbring, P., Gunnarsdóttir, M., Hedensjö, L., Andersson, G., Ekselius, L., & Furmark, T. (2007). Treatment of social phobia: randomized trial of internet-delivered cognitive-behavioural therapy with telephone support. British Journal of Psychiatry, 190, 123-128.
Furmark, T. (2002). Social phobia: overview of community surveys. Acta Psychiatrica Scandinavica,
105, 84-93.
Furmark, T., Carlbring, P., Hedman, E., Sonnenstein, A., Clevberger, P., Bohman, B., …Andersson, G. (2009). Guided and unguided self-help for social anxiety disorder: randomised controlled trial. The British Journal of Psychiatry, 195, 440-447.
Furmark, T., Tillfors, M., Everz, P. O., Marteinsdottir, I., Gefvert, O., & Fredrikson, M. (1999). Social phobia in the general population: Prevalence and sociodemographic profile. Social Psychiatry
and Psychiatric Epidemiology, 34, 416–424.
Hedman, E., Andersson, G., Ljótsson, B., Anderson, E., Rück, C., Mörtberg, E., & Lindefors, N. (2011). Internet-Based Cognitive Behavior Therapy vs. Cognitive Behavioral Group Therapy for Social Anxiety Disorder: A Randomized Controlled Non-inferiority Trial. PLoS ONE, 6, 1-10.
Liebowitz, M. R. (1987). Social phobia. Modern Problems of Pharmacopsychiatry, 22, 141–173. Mattick, R. P., & Clarke, J. C. (1998). Development and validation of measures of social phobia
scrutiny fear and social interaction anxiety. Behaviour Research and Therapy, 36, 455–470. Newman, M.G., Erickson, T., Przeworski, A., & Dzus, E. (2003). Self-help and minimal contact
therapies for anxiety disorders: Is human contact necessary for therapeutic efficacy?
Nordgreen, T., Havik, O.E., Öst, L.G., Furmark, T., Carlbring, P., & Andersson, G. (2012). Outcome predictors in guided and unguided self-help for social anxiety disorder. Behaviour Research
and Therapy, 50, 13-21.
Rizvi, S. L., Vogt, D. S., & Resick, P. A. (2009). Cognitive and affective predictors of treatment outcome in cognitive processing therapy and prolonged exposure for posttraumatic stress disorder.
Behaviour Research and Therapy, 47, 737-743.
Spek, V., Cuijpers, P., Nyklícek, I., Riper, H., Keyzer, J. and Pop, V. (2007). Internet-based cognitive behaviour therapy for symptoms of depression and anxiety: A meta-analysis. Psychological
Medicine: A Journal of Research in Psychiatry and the Allied Sciences, 37, 319–328.
Swift, J.K., & Greenberg, R.P. (2012). Premature Dicontinuation in Adult Psychotherapy: A Meta- Analysis. Journal of Consulting and Clinical Psychology, 80, 547-559.
Tillfors, M., Carlbring, P., Furmark, T., Lewenhaupt, S., Spak, M., Eriksson, A., ...Andersson, G. (2008). Treating university students with social phobia and public speaking fears: internet delivered self-help with or without live group exposure sessions. Depression and Anxiety, 25, 708-717. Titov, N., Andrews, G., Johnston, L., Schwencke, G., & Choi, I. (2009a). Shyness programme: longer
term benefits, cost-effectiveness, and acceptability. Australian and New Zealand Journal of
Psychiatry, 43, 36-44.
Titov, N., Andrews, G., Choi, I., Schwencke, G., & Johnston, L. (2009b). Randomized controlled trial of web-based treatment of social phobia without clinician guidance. Australian and New
Zealand Journal of Psychiatry, 43, 913-919.
Titov, N., Andrews, G., Schwencke, G., Drobny, J., & Einstein, D. (2008). Shyness 1: distance treatment of social phobia over the Internet. Australian and New Zealand Journal of
