The impact of internet-based cognitive
behavioral therapy and depressive symptoms on
self-care behavior in patients with heart failure:
A secondary analysis of a randomised controlled
trial
Peter Johansson, Tiny Jaarsma, Gerhard Andersson and Johan Lundgren
The self-archived postprint version of this journal article is available at Linköping University Institutional Repository (DiVA):
http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-173656
N.B.: When citing this work, cite the original publication.
Johansson, P., Jaarsma, T., Andersson, G., Lundgren, J., (2019), The impact of internet-based cognitive behavioral therapy and depressive symptoms on self-care behavior in patients with heart failure: A secondary analysis of a randomised controlled trial, International Journal of Nursing
Studies, , 103454. https://doi.org/10.1016/j.ijnurstu.2019.103454
Original publication available at:
https://doi.org/10.1016/j.ijnurstu.2019.103454 Copyright: Elsevier
1
The impact of internet based cognitive behavioural therapy and depressive
symptoms on self-care behaviour in patients with heart failure. A secondary
analysis of a randomised controlled trial.
2
Contribution of the paper
What is already known about the topic
• In patients with heart failure, depressive symptoms are common and can act as a barrier to performing self-care behaviours. This may also be a possible explanation for why depressive symptoms in heart failure are associated with an impaired quality of life and a poorer prognosis in heart failure.
• There are studies reporting that psychological interventions have possible positive short-term impact on self-care in heart failure. However, most of these studies did not include heart failure patients who also have elevated levels of depressive symptoms. • There is a lack of knowledge as to whether internet-based cognitive behaviour therapy
targeting depressive symptoms in patients with heart failure can improve self-care and the different aspects autonomy-based self-care, provider-based self-care and
consulting behaviour.
What this paper adds
• Internet-based CBT targeting depressive symptoms in patients with heart failure was not superior to an on-line discussion group, as regards improving self-care.
• Internet-based CBT may improve provider-based self-care behaviours such as taking heart failure medications.
• An improvement in depressive symptoms was associated with an improvement in autonomy-based self-care behaviours such as exercise and daily weighing.
4 Abstract
Background: Patients with chronic heart failure (HF) may require treatment of depressive symptoms to improve self-care behaviour. Objectives: To investigate the impact of internet-based cognitive behavioural therapy (CBT) on self-care behaviour in heart failure patients, and to study the association between changes in depressive symptoms and changes in self-care behaviour. Design: A secondary analysis of data collected in a pilot randomized
controlled study. Setting: 50 heart failure patients with depressive symptoms were recruited from four hospitals in southeastern Sweden. Methods: Patients were randomized to nine weeks of internet-based CBT (n=25) or to an active control group participating in an online discussion forum (n=25). A week two and three those in the internet-based CBT group worked with psychoeducation about heart failure and depression, emphasizing heart failure self-care.During the same weeks those in the on-line discussion forum specifically discussed heart failure self-care. Patient Health Questionnaire-9 was used to measure depressive
symptoms at baseline and at the nine-week follow-up. The European Heart Failure Self-care Behaviour Scale-9 was used to measure self-care behaviour (i.e., the summary score and the subscales autonomous based, provider based and consulting behaviour) at baseline, and at the three-week and nine-week follow-ups. Results: No significant differences were found in self-care between the patients in the internet-based CBT and the patients in the online discussion group at the three- and nine-week follow-ups. Within-group analysis of the changes in the European Heart Failure Self-care Behaviour Scale showed that from baseline to week three, the summary score increased significantly for the online discussion group (p=0.04), but not for the internet-based CBT group (p=0.15). At the nine-week follow-up, these scores had decreased. Similarly, consulting behaviour improved at week three for the online discussion group (p=0.04), but not for the internet-based CBT group (p=0.22). Analysis of provider-based adherence at the nine-week follow-up compared to baseline showed that the score in the
5
internet-based CBT group had increased (p=0.05) whereas it had decreased in in the on-line discussion group. Improvement in symptoms of depression was significantly associated with improvement in autonomy-based self-care (r=0.34, p=0.03).
Conclusion: Improvement in depressive symptoms was associated with improved
autonomous-based self-care. ICBT for depression in HF may benefit aspects of self-care that are vital to improve symptoms and prognosis.
6 1. BACKGROUND
Depressive symptoms are common in patients with heart failure (HF), and studies suggest a significantly increased risk of hospital readmissions and mortality (Gathright et al., 2017, Ghosh et al., 2016) and impairment of quality of life (Muller-Tasch et al., 2007) for these patients.
Poorer performance of self-care may be one underlying mechanism behind the negative effects of depressive symptoms in heart failure (Sedlar et al., 2017). For example, studies have reported that heart failure patients with depressive symptoms have more difficulty adhering to prescribed medications (Goldstein et al., 2017) and a greater tendency to delay seeking healthcare despite worsening symptoms (Johansson et al., 2011). This suggests that interventions targeting depressive symptoms in heart failure patients may improve self-care behaviour. A meta-analysis, including 25 RCTs, reported a possible positive short-term impact of psychological interventions on self-care in heart failure (Jiang et al., 2018).
However one problem was that only one study by Freedland et al. (Freedland et al., 2015) in the meta-analysis included a sample who were diagnosed with depression. In that study, six months of face-to-face cognitive behavioural therapy (CBT), compared to enhanced standard care, improved depression but surprisingly did not improve self-care. But self-care has more than one facet. For example self-care as measured by the 9-item European Heart Failure Self-care Behaviour Scale-9 (EHFScBS) (Jaarsma et al., 2009) has been reported to consist of three different facets of heart failure self-care (Vellone et al., 2014): autonomy-based adherence (three items: I weigh myself every day, I limit the amount of fluids, I exercise regularly), provider-based adherence (two items: I eat a low-salt diet, I take my medication as prescribed), and consulting behaviour (four items: If shortness of breath increases I contact my doctor or nurse, If my legs/feet are more swollen I contact my doctor or nurse, If I gain weight more than two kg in seven days I contact my doctor or nurse, If I experience fatigue I
7
contact my doctor or nurse). Thus, instead of analysing self-care as one facet it can be of interest to explore if CBT may have a different impact on the various facets of self-care, such as autonomy-based adherence, provider-based adherence and consulting behaviour, in heart failure patients with elevated levels of depressive symptom.
Regrading CBT, one problem is low access due to a lack of psychologists. Provision of CBT over the internet (i.e. internet-based CBT) has been proven effective in patients with
depression only (Andrews et al., 2018, Karyotaki et al., 2018). However, there is a lack of studies that have examined whether internet-based CBT in heart failure aimed at improving depression can also improve self-care and different facets of self-care.
Therefore, the objectives of this explorative study including HF patients with elevated levels of depressive symptoms were: (I) to investigate the impact of Internet-based Cognitive Behaviour Therapy (ICBT) on different facets self-care behaviour and (II) to study the association between changes in depressive symptoms and different facets of self-care behaviour.
2. METHODS
2.1 Study design and population
A secondary analysis of data collected in a randomized controlled trial (RCT) aimed to evaluate the impact of a nine-week internet-based CBT programme on depressive symptoms in patients with heart failure. The design of the RCT study has been reported in the main publication (Lundgren et al., 2016).In brief, a total of 50 heart failure patients with at least mild depressive symptoms (i.e. Patient Health Questionnaire-9 score >5) were recruited from four hospitals in southeastern Sweden. They were randomized to either a nine-week internet-based CBT programme (n=25) or to an active control group participating in a nine-week online discussion group (n=25). The regional ethical review board of Linköping Sweden
8
approved the study (ref. no. 2011/166-31). The primary study is registered at clinical trials.gov (NCT01681771).
2.2 Procedures
The internet-based CBT programme has been described previously (Lundgren et al., 2015, Lundgren et al., 2016). In brief, it consists of seven modules: (1) Introduction; (2) Living with HF; (3) Depressive symptoms and heart failure; (4) Behaviour activation – enabling change; (5) Behaviour activation – implementing change; (6) Problem-solving – a tool for dealing with problems, and (7) Consummation. The second and third modules (i.e. weeks 2 and 3) included psychoeducation and homework assignments emphasizing heart failure self-care such as adherence to medical treatment, symptom monitoring and physical activity. In the online discussion forum, discussion was performed in writing and new topics were presented each week over the study period. At weeks 2 and 3, the topics related to different aspects of self-care. Thus, at weeks 2 and 3, heart failure self-care was addressed, but the internet-based CBT group and the on-line discussion group approached it differently.
2.3 Measurements
2.3.1 Depressive symptoms
The Patient Health Questionnaire-9 (Spitzer et al., 1999) was used to measure depressive symptom at baseline and at the nine-week follow-up. The Patient Health Questionnaire-9 consists of nine items to be answered on a four-point scale, and summed to a total score in the range of 0-27, where higher numbers represent a higher level of depressive symptoms
(Kroenke et al., 2001). The Patient Health Questionnaire-9 has been found to be valid and reliable in patients with heart failure (Hammash et al., 2013).
9 2.3.2 Heart failure self-care behaviour
The 9-item European Heart Failure Self-care Behaviour Scale (Jaarsma et al., 2009) was used to measure heart failure self-care behaviour and the three sub-scales autonomy-based as described by Vellone at al. (Vellone et al., 2014) adherence; provider-based adherence; consulting behaviour at baseline, three weeks and at the nine-week follow-up. The items are rated on a five-point scale between 1 (I completely agree) and 5 (I completely disagree). The score was standardized to 0-100, and reversed so that a higher score indicated better self-care (Vellone et al., 2014).
2.4 Statistical methods
Descriptive data are presented as percentages or mean ± SD. A Student’s t-test was used for continuous variables and the Chi-square test was used for discrete variables. An evaluation of differences between groups regarding the summary score of the European Heart Failure Self-care Behaviour Scale and the three subscales scores (i.e. autonomy-based, provider-based and consulting behaviour) at the three and nine-week follow-up was performed with Analysis of Covariance (ANCOVA). A paired Student’s t-test was used to analyze if the changes in European Heart Failure Self-care Behaviour Scale from baseline to three weeks, and from three weeks to follow-up at nine weeks differed significantly within the internet-based CBT group and the on-line discussion group. To explore possible associations between changes in depressive symptoms and changes of European Heart Failure Self-care Behaviour Scale summary score, as well as in the subscales of autonomy-based and provider-based adherence and consulting behavior we did not separate data from the two groups. Since the change scores were skewed, analyses were performed with Spearman rank correlations. A positive correlation between the change in Patient Health Questionnaire-9 and the change in the
10
European Heart Failure Self-care Behaviour Scale indicates that an improvement in
depressive symptoms correlates with an improved self-care. P-values < 0.05 were considered significant. All data were analyzed using standard software (IBM, SPSS version 24).
3. RESULTS 3.1 Population
The population has been described in detail in the main publication (Lundgren et al., 2016). In brief, the mean age of the study population was 63 years (SD 12.8), 59% were male and 4% were smokers. About three-quarters of the population were in New York Heart Association Class II (40%) and III (36 %). Approximately 90 % took beta-blockers and angiotensin-converting enzyme inhibitors or angiotensin receptor blockers. The mean score in the Patient Health Questionnaire-9 before the start of the study did not vary between the groups (internet-based CBT group 11.8 compared to online discussion group 11.2).
3.2 Internet based Cognitive Behavioural Therapy and heart failure self-care
No significant differences were found in self-care between the patients in the internet-based CBT and the patients in the online discussion group at the three- and nine-week follow-ups (Table 1). As can be seen in Table 1, self-care behaviours increased from baseline to week three in both groups. From week three to the end of the study at week nine, a slightly different pattern was seen. The mean scores of the patients in the on-line discussion group decreased in all four scales, whereas in the internet-based CBT group the scores remained stable or
11
Within-group analysis (Table 1) of the changes in the European Heart Failure Self-care Behaviour Scale shows that from baseline to week three, the summary score increased significantly for the online discussion group (p=0.04), but not for the internet-based CBT group (p=0.15). At the nine-week follow-up, these scores had decreased. Similarly, consulting behaviour improved at week three for the online discussion group (p=0.04), but not for the internet-based CBT group (p=0.22). At the nine-week follow-up, these scores had also decreased to approximately the level found at baseline. For autonomy-based adherence, only minor increases were found at three weeks. However, at nine weeks, the mean score in the on-line discussion group had decreased to baseon-line level, whereas the score of the internet-based CBT group was sustained. The analysis of provider-based adherence revealed that the score in the internet-based CBT group increased significantly from baseline to the nine-week follow-up (p=0.05) whereas in the on-line discussion grofollow-up, the score decreased to baseline level.
3.3 Changes in symptoms of depression and changes in heart failure self-care A significant and moderate correlation between improvement in symptoms of depression
(mean change score 0.9 SD 4.0) and improvement in autonomy-based adherence (mean change score 1.2 SD 22.4) was found (r=0.34, p=0.03). The correlation between the
improvement in depressive symptoms and the summary score of European Heart Failure Self-care Behaviour Scale was of the same magnitude (mean change score 2.8 SD 14.2), but not significant (r=0.28, p=0.07). For subscales consulting behaviour (mean change score 1.8 SD 19.7), and provider-based adherence (mean change score 7.1 SD 18.3), the correlations were weaker and not significant (r=0.18, p=0.27 and r=0.07, p=0.66). Correlational analysis in each group did not reveal any significant associations.
12
4. Discussion
In this study we compared the effectiveness of internet-based CBT compared to an online discussion group in the improvement of self-care behaviour in heart failure patients with depressive symptoms. We could not find any beneficial effect of internet-based CBT on self-care in depressed heart failure patients. Freedland et al. (Freedland et al., 2015) also reported no improvements from CBT for self-care in heart failure patients. An explanation may be that CBT interventions may not impact all facets of self-care. We therefore undertook further analyses to explore whether internet-based CBT could have different impact on the three subscales in the European Heart Failure Self-care Behaviour Scale, but no beneficial effects could be found. Another possible explanation for the lack of effects on self-care, is that in our study and that of Freedland et al (Freedland et al., 2015), poor self-care was not a criterion for eligibility, which limits our possibilities of detecting improvements in self-care.
The within-group analyses showed that online-discussion group had a significant
improvement in the summary score at week 3, but not at week 9. This may be the result of increased attention by way of participation in the study, thus indicating a possible digital-placebo effect (Torous and Firth, 2016). On the other hand, the score for provider-based adherence was significantly different than baseline (p=0.05) in the internet-based CBT group at the nine-week follow-up. It is possible that the internet-based CBT programme was helpful in improving provider-based adherence since content in the psychoeducative heart failure module and the homework assignments were intended to increase the understanding of, and the benefit of, medication and exercise. Furthermore, in the behavioural activation module of
13
the internet-based CBT programme, the participants could, in a structured way, plan and implement self-care behaviours in their daily lives, such as taking medications.
We found an association between improvement in depressive symptoms and improvement in autonomy-based adherence (exercise, weighing, fluid restriction). In this analysis we pooled the data from the two groups to study how change in depression was related to changes in self-care, regardless of group assignment. We realize that from this analysis, one cannot determine whether internet-based CBT has a direct effect on self-care, or if it affects self-care indirectly by improving depression. However, in our pilot RCT study, only patients in the internet-based CBT group had a significant decrease in depressive symptoms (Lundgren et al., 2016). In our internet-based CBT programme, exercise (i.e. an aspect of autonomy-based adherence) was emphasized as a good self-care behaviour. This may indicate that
improvements in depressive symptoms may be associated with improvements in self-care behaviours that are more endorsed by patients themselves.
These results illustrated to us the difficulty of finding the optimal combination of programmes in the management of depressive symptoms in heart failure patients. It is possible that
internet-based CBT on top of exercise may be a more effective option than exercise or CBT alone. A previous study by Gary et al. (Gary et al., 2010) showed that CBT combined with home-based exercise had the best effect in reducing depression, increasing physical function, and improving health-related quality of life. These data also suggest a new hypothesis
regarding the optimal components of interventions to improve self-care. Maintaining autonomy, physical well-being and symptom relief are reported as important motivational goals for heart failure patients (Jaarsma et al., 2017). These goals may be more associated with intrinsic motivations, i.e. the patient performs self-care activities due to internal
autonomy-14
based or provider-based self-care behaviours such as exercise, choosing healthy food, increasing control by daily weighing, and experiencing fewer symptoms by taking HF medications. In our internet-based CBT programme, participants established their own personal goals (i.e. intrinsic goals). It is therefore hypothesized that an internet-based CBT programme could have served as an aid to achieving self-care. This hypothesis requires testing in additional research that explores the effects of CBT or internet-based CBT on self-care, and on different facets of self-care behaviours and depressive symptoms in heart failure patients.
A major limitation is the fact that this was a secondary-analysis of data collected in a pilot-RCT. Thus, the study is underpowered to detect associations and interactions between internet-based CBT, self-care behaviours and depressive symptoms. Therefore, the results from this study should be interpreted with caution. Another limitation is that the analysis of self-care behaviours was self-reported. It could have been valuable if objective data regarding self-care behaviours such as physical activity were also collected. Since this study is one of the first internet-based CBT studies performed with heart failure patients with elevated depressive symptoms, we believe that the results of this study, despite its limitations, are of interest.
In conclusion, this study found improvement in depressive symptoms that were associated with improved autonomous-based self-care adherence. More studies that evaluate whether internet-based CBT for depression in heart failure has beneficial effects on self-care are needed.
15
This study was funded by grants from the Swedish Heart and Lung Association (grant number E08/14), the Medical Research Council of Southeast Sweden (grant number FORSS-470121) and the Region of Östergötland (grant number LIO-470271). The funding source had no influence on the design, procedure, analysis or interpretation of the results in this study.
References
Andrews, G., Basu, A., Cuijpers, P., Craske, M.G., McEvoy, P., English, C.L., Newby, J.M., 2018. Computer therapy for the anxiety and depression disorders is effective, acceptable and practical health care: An updated meta-analysis. J Anxiety Disord 55, 70-78.
Freedland, K.E., Carney, R.M., Rich, M.W., Steinmeyer, B.C., Rubin, E.H., 2015. Cognitive Behavior Therapy for Depression and Self-Care in Heart Failure Patients: A Randomized Clinical Trial. JAMA Intern Med 175 (11), 1773-1782.
Gary, R.A., Dunbar, S.B., Higgins, M.K., Musselman, D.L., Smith, A.L., 2010. Combined exercise and cognitive behavioral therapy improves outcomes in patients with heart failure. J Psychosom Res 69 (2), 119-131.
Gathright, E.C., Goldstein, C.M., Josephson, R.A., Hughes, J.W., 2017. Depression increases the risk of mortality in patients with heart failure: A meta-analysis. J Psychosom Res 94, 82-89.
Ghosh, R.K., Ball, S., Prasad, V., Gupta, A., 2016. Depression in heart failure: Intricate relationship, pathophysiology and most updated evidence of interventions from recent clinical studies. Int J Cardiol 224, 170-177.
Goldstein, C.M., Gathright, E.C., Gunstad, J., M, A.D., Redle, J.D., Josephson, R., Moore, S.M., Hughes, J.W., 2017. Depressive symptoms moderate the relationship between medication regimen complexity and objectively measured medication adherence in adults with heart failure. J Behav Med 40 (4), 602-611.
Hammash, M.H., Hall, L.A., Lennie, T.A., Heo, S., Chung, M.L., Lee, K.S., Moser, D.K., 2013. Psychometrics of the PHQ-9 as a measure of depressive symptoms in patients with heart failure. Eur J Cardiovasc Nurs 12 (5), 446-453.
Jaarsma, T., Arestedt, K.F., Martensson, J., Dracup, K., Stromberg, A., 2009. The European Heart Failure Self-care Behaviour scale revised into a nine-item scale (EHFScB-9): a reliable and valid international instrument. Eur J Heart Fail 11 (1), 99-105.
Jaarsma, T., Cameron, J., Riegel, B., Stromberg, A., 2017. Factors Related to Self-Care in Heart Failure Patients According to the Middle-Range Theory of Self-Care of Chronic Illness: a Literature Update. Curr Heart Fail Rep 14 (2), 71-77.
Jiang, Y., Shorey, S., Seah, B., Chan, W.X., Tam, W.W.S., Wang, W., 2018. The effectiveness of psychological interventions on self-care, psychological and health outcomes in patients with chronic heart failure-A systematic review and meta-analysis. Int J Nurs Stud 78, 16-25. Johansson, P., Nieuwenhuis, M., Lesman-Leegte, I., van Veldhuisen, D.J., Jaarsma, T., 2011.
Depression and the delay between symptom onset and hospitalization in heart failure patients. Eur J Heart Fail 13 (2), 214-219.
Karyotaki, E., Ebert, D.D., Donkin, L., Riper, H., Twisk, J., Burger, S., Rozental, A., Lange, A., Williams, A.D., Zarski, A.C., Geraedts, A., van Straten, A., Kleiboer, A., Meyer, B., Unlu Ince, B.B., Buntrock, C., Lehr, D., Snoek, F.J., Andrews, G., Andersson, G., Choi, I., Ruwaard, J., Klein, J.P., Newby, J.M., Schroder, J., Laferton, J.A.C., Van Bastelaar, K., Imamura, K., Vernmark, K., Boss, L., Sheeber, L.B., Kivi, M., Berking, M., Titov, N., Carlbring, P., Johansson, R., Kenter, R., Perini, S., Moritz, S., Nobis, S., Berger, T., Kaldo, V., Forsell, Y., Lindefors, N., Kraepelien, M.,
16
Bjorkelund, C., Kawakami, N., Cuijpers, P., 2018. Do guided internet-based interventions result in clinically relevant changes for patients with depression? An individual participant data meta-analysis. Clin Psychol Rev 63, 80-92.
Kroenke, K., Spitzer, R.L., Williams, J.B., 2001. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 16 (9), 606-613.
Lundgren, J., Andersson, G., Dahlstrom, O., Jaarsma, T., Kohler, A.K., Johansson, P., 2015. Internet-based cognitive behavior therapy for patients with heart failure and depressive symptoms: A proof of concept study. Patient Educ Couns 98 (8), 935-942.
Lundgren, J.G., Dahlstrom, O., Andersson, G., Jaarsma, T., Karner Kohler, A., Johansson, P., 2016. The Effect of Guided Web-Based Cognitive Behavioral Therapy on Patients With Depressive Symptoms and Heart Failure: A Pilot Randomized Controlled Trial. J Med Internet Res 18 (8), e194.
Muller-Tasch, T., Peters-Klimm, F., Schellberg, D., Holzapfel, N., Barth, A., Junger, J., Szecsenyi, J., Herzog, W., 2007. Depression is a major determinant of quality of life in patients with chronic systolic heart failure in general practice. J Card Fail 13 (10), 818-824.
Sedlar, N., Lainscak, M., Martensson, J., Stromberg, A., Jaarsma, T., Farkas, J., 2017. Factors related to self-care behaviours in heart failure: A systematic review of European Heart Failure Self-Care Behaviour Scale studies. Eur J Cardiovasc Nurs 16 (4), 272-282.
Spitzer, R.L., Kroenke, K., Williams, J.B.W., Group, a.t.P.H.Q.P.C.S., 1999. Validation and Utility of a Self-report Version of PRIME-MD. JAMA: The Journal of the American Medical Association 282 (18), 1737-1744.
Torous, J., Firth, J., 2016. The digital placebo effect: mobile mental health meets clinical psychiatry. Lancet Psychiatry 3 (2), 100-102.
Vellone, E., Jaarsma, T., Stromberg, A., Fida, R., Arestedt, K., Rocco, G., Cocchieri, A., Alvaro, R., 2014. The European Heart Failure Self-care Behaviour Scale: new insights into factorial structure, reliability, precision and scoring procedure. Patient Educ Couns 94 (1), 97-102.
Table 1
Table 1. Mean values at the three measurement points and analysis of changes between the internet-based CBT and the discussion groups regarding the summary score autonomy-based adherence provider-based adherence and consulting behaviors) of the European Heart-Failure Self-Care Behavior Scale. The table also provide an analysis of changes within the internet-based CBT group and the discussion group.
Summary Score
Mean (SD) Autonomy-based adherence Mean (SD) Provider-based adherence Mean (SD) Consulting-behavior Mean (SD)
Baseline W3 W9 Baseline W3 W9 Baseline W3 W9 Baseline W3 W9
ICBT-group 38.9 (17) 47.4 (21) 42.3 (17) 45.6 (19) 47.6 (26) 47.6 (31) 68.4 (25) 76.2 (23) 78.5(20) 19 (26) 33 (30) 20 (20) Discussion-group 40.5 (21) 48.1 (18) 41.8 (18) 42.5 (23) 45.8 (26) 42.2 (20) 67.4 (22) 75 (15) 69.4 (20) 25.7 (31) 36.4 (27) 27.7 (34) Between-group analysis ANCOVAA F(1,41)=0.20, p=0.88 F(1,41)=0.36, p=0.85 F (1,41)=0.69, p=0.79 F(1,37)=0.15, p=0.70 Between-group analysis ANCOVAB F(1,37)=0.08, p=0.78 F(1,37)=0.15, p=0.69 F (1,37)=2.2, p=0.14 F(1,37)=0.28, p=0.60 Within-group analysis ICBT-group p=0.151 p=0.352 p=0.711 p=0.712 p=0.201 p=0.052 p=0.221 p=0.922 Within-group analysis Discussion -group p=0.041 p=0.422 p=0.421 p=0.932 p=0.061 p=0.192 p=0.041 p=0.432
Footnote:
ANCOVAA difference between the groups a week 3
ANCOVAB difference between the groups at week 9
1 p-value for the difference between the score at baseline and the score at week 3 2 p-value for the difference between the score at baseline score and the score at week 9
