The effectiveness and cost-effectiveness of cognitive behavioral therapy administered in group or via the internet for patients with irritable bowel syndrome

Abstract

Background: Irritable Bowel Syndrome (IBS) is a benign but debilitating disorder. It is the most commonly diagnosed gastrointestinal condition and affects between 7-21% of the world’s population. It is characterized by abdominal pain and altered bowel habits. Patients with refractory IBS tend to develop behavioral avoidance that may severely affect life quality and lead to aggravated symptoms. Psychological treatment has emerged as a potent way of treating patients with IBS. Professor Brjánn Ljótsson at Karolinska Institutet has developed a cognitive behavioral treatment protocol specific for IBS. One of the core components in the treatment is exposure to feared stimuli such as specific foods or situations. Four different RCTs have proven the efficacy of the treatment. In randomized controlled trials the goal is to find a potential treatment effect and the setting is normally adjusted to eliminate confounding factors (i.e by careful selection of patients and by utilizing specially trained staff that is supervised during the treatment phase and by providing many examinations or questionnaires to patients). Clinical routine care may often differ from the context of a randomized controlled trial, and it is therefore important to examine the effectiveness of a potentially effective treatment under real world circumstances. When a new and potentially more effective treatment is developed health care managers must decide whether it should be implemented in routine care or not, and one of the important questions concerns the costs for the new treatment in comparison to already implemented methods. Knowledge regarding cost-effectiveness is therefore needed.

Aims: The general aim of the present thesis is to examine the real-world effectiveness of the exposure-based cognitive behavioral treatment (ECBT) our research group has developed and to evaluate its cost-effectiveness to provide managers in health care with a better foundation for good decisions. Specific aims of this theses were to: a) evaluate the cost-effectiveness from a societal perspective of the exposure component in the ECBT when delivered over the internet (study I), b) evaluate the real-world treatment effectiveness of the ECBT when delivered face-to-face in a group format and to investigate possible predictors of treatment outcome (study II), and to evaluate the real-world effectiveness of the ECBT when delivered over the internet and to replicate an earlier finding that behavioral change predicts symptom-reduction and thus a potential working mechanism of the treatment (study III).

Methods: Study I randomized 309 self-referred patients to internet-delivered CBT (ICBT) with or without exposure (WE) and examined the costs of the intervention as well as potential cost reductions for society and related them to treatment effects. Study II used data from the electronical medical record at a gastroenterological clinic and compared symptom burden before and after patents went through a group based ECBT face-to-face. The study included 129 patients with refractory IBS. Baseline data was used to try to create a prediction model for treatment outcome. Study III included 309 consecutively recruited patients at the Internet Psychiatry Unit who received a 12-week internet-based ECBT for IBS. The effectiveness of the treatment was calculated at the end of treatment and six months after treatment started. In the study a random intercept cross-lagged panel model was used to investigate the relationship between behavioral change and symptom reduction over time.

Results: In both effectiveness studies (studies II and III) the patients were significantly improved after treatment and the calculated effect sizes indicated that the treatment effect is large. In study II, 52.7% of the patients had a symptomatic reduction measured with GSRS-IBS of 30% or more which is considered a clinically significant improvement. There was not enough support for the construction of a prediction model based on pre-treatment characteristics that would predict symptomatic improvement. The cost-effectiveness study (study I) showed that the exposure component makes the treatment approximately 20% more expensive, but that the money invested in exposure pays back with almost 6 times the investment in saved expenses from a societal perspective during the six months following the treatment. There is an 84% chance that the ECBT will be more cost-effective than ICBT-WE if society is willing to pay no more for the ECBT than for the ECBT-WE. Finally, in study III, we also found support for the hypothesis that the symptom reduction patients experience after ECBT is related to behavior change during the treatment, prior to the outcome.

Conclusion: ECBT is highly effective, also under real world circumstances, both when delivered face-to-face and over the internet. The exposure component in the treatment is related to behavior change that in turn predicts treatment outcome, thus pointing to a possible chain of events partly explaining the working mechanism of the treatment. Lastly, we conclude that it is very cost-effective from a societal perspective to include exposure in the treatment since the extra costs are weighed out many times by saved expenses down the road.