Cognitive behavioral group therapy for patients with physical diseases and comorbid depressive or adjustment disorders on a waiting list for individual therapy: results from a randomized controlled trial

Imagine that you are diagnosed with a chronic disease and realize at some point that it is too difficult to cope on your own. You contact a psychotherapist and he tells you that you have to wait 6 months until you can begin your therapy. You would probably feel rather left alone. This is what happens to many patients in Germany. According to a survey based on data of 8620 outpatient psychotherapists, waiting times in Germany are on average 12.5 weeks for an initial consultation and 23.4 weeks for the first therapy session [1]. McCarthy, McGee and O’Boyle [2] found that most patients (64%) were dissatisfied with these waiting times. Huckert, Hank and Krampen [3] examined changes in psychopathological symptoms from registration for therapy to initial appointment (on average 6 months) and found that the symptoms of the majority of patients (68%) remain unchanged while waiting. However, studies found a significant correlation between waiting time and non-attendance at initial appointment [4, 5]. Hoyer, Helbig and Wittchen [6] found that 41% of patients under current depression treatment had already thought about starting the therapy at an earlier point in their lifetime and showed that long waiting time was one of the three most commonly-mentioned reasons for not even searching for a therapist at all. Thus, the problematic waiting situation increases the rate of late or completely untreated mental disorders, which in turn leads to higher direct and indirect costs [7]. From an ethical, economic and therapeutic perspective, it seems unwise to leave these patients untreated for such a long time.

The provision of low-threshold interventions during waiting time is one option to improve this problem. Psycho-educative information materials [8], bibliotherapy [9, 10], motivational interviewing training [11], group therapy [12] and web-based self-help [13] have already been used as waiting list interventions.

None of these interventions were addressed to patients with physical diseases and comorbid depressive or adjustment disorders. Group interventions seem to be a promising treatment option for the waiting time of these patients, because group-specific factors such as the universality of burden, altruism or interpersonal learning are associated with benefits in coping [14]. To date, there is only one study examining a waiting list group for patients (no special diagnostic indication) to talk about their problems. However, given that this study did not examine effects on mental outcomes and had an attendance rate of only 14% [12], there is a need for a study developing and evaluating a waiting list group intervention for patients with physical diseases and comorbid depressive or adjustment disorders.

The evidence for cognitive behavioral group therapy (CBGT) in depression treatment has been demonstrated in several meta-analyses [15‐17], although for patients with physical diseases the body of evidence is less extensive. Many studies in this field have investigated the effects of CBGT on quality of life in all physically ill patients and did not target patients with a mental comorbidity. There is some research on the efficacy of CBGT for depressed outpatients who suffer from the same physical disease. Kelly et al. [18] compared eight sessions CBGT to a waiting list control group in depressed outpatients with HIV and found a medium post-treatment effect (d = 0.54¹) on depression. Evans and Connis [19] compared eight sessions CBGT to a waiting list control group in depressed outpatients with cancer and also found a medium post-treatment effect (d = 0.55¹) on depression. Heckman et al. [20] compared 12 sessions CBGT to a waiting list control group in depressed outpatients with HIV and found a small post-treatment effect (d = 0.34), whereas Chesney, Chambers, Taylor, Johnson and Folkman [21] compared 10 sessions CBGT to a waiting list control group in depressed outpatients with HIV and found no significant difference in depression (d = 0.16¹) at post-treatment. To our knowledge, there is only one study investigating CBGT in patients with various physical diseases: Schuster, Rueddel, Keck and Schwarting [22] examined the efficacy of CBGT in inpatients with coronary or orthopedic diseases and comorbid depression, anxiety or somatoform disorders and demonstrated that compared to treatment as usual the group with an additional six sessions CBGT reported significantly lower depression scores at post-treatment (no results provided to calculate effect sizes) and follow-up (strong effect; d = 0.89). In summary, existing group interventions for patients with physical diseases and comorbid depression – of which the efficacy has been proven in a randomized controlled trial (RCT) – used either a closed group format with consecutive session contents [18‐20] or were designed for an inpatient setting [22]. Therefore, they cannot simply be adopted for an outpatient waiting list group intervention which should be able to continuously include new patients from the waiting list. Our intervention fills this gap.

To overcome waiting times of six months or more for individual therapy in our outpatient clinic for mental disorders in patients with physical diseases, we wanted to establish a waiting list group intervention as a first step of treatment. We developed STEpS, a half-open eight-session CBGT for patients with various physical diseases and comorbid depressive or adjustment disorders waiting for individual therapy [23] and conducted a RCT to examine its efficacy. The primary outcome of the study was depression; secondary outcomes were global psychological distress and health-related quality of life. We hypothesized that the STEpS group would be superior in these outcomes compared to a waiting list control group (CG) at the end of treatment as well as at 2-month follow-up, although the benefits would probably be smaller.

We conducted a two-armed RCT to compare a waiting list group intervention (STEpS) with a waiting list control group (CG). The study took place at the outpatient clinic of the Institute of Psychology at the Albert-Ludwigs-University of Freiburg, which is specialized in the treatment of mental disorders in patients with chronic physical diseases. Measurement points were taken at baseline (before initial consultation, T₁), post-treatment (T₂) and 2-month follow-up (T₃). The study protocol was approved by the ethics committee of the Albert-Ludwigs-University of Freiburg [23].

The aim of this RCT was to evaluate the efficacy of STEpS, an eight-session CBGT for outpatients with various physical diseases and comorbid depressive or adjustment disorders on a waiting list for individual therapy. As hypothesized, patients receiving STEpS showed significantly lower depression scores at post-treatment compared to the CG. The effect size achieved was small (d = 0.37). As there is no other study examining CBGT in a comparable sample of depressed outpatients with various physical diseases, we have to compare our findings with the results of studies that examined CBGT either in depressed outpatients who suffered from the same physical disease (HIV or cancer) or in inpatients with various physical diseases and mental comorbidity. Studies on the efficacy of CBGT in depressed outpatients who suffered from the same physical disease have revealed heterogeneous effects on depression (d = 0.16 [21], d = 0.34 [20], d = 0.54 [18], d = 0.55 [19]). Different effect sizes across these studies may be partly influenced by different strategies of handling missing data (using a completer analysis [18, 19] vs. imputing missing values by last observation carried forward [20]). Our findings strengthen the position that CBGT is efficacious in reducing depression in these patients. Compared to Schuster et al. [22] – who compared an additional six sessions CBGT to treatment as usual in inpatients with different physical diseases and mental comorbidity and revealed a strong effect on depression at 6-month follow-up – our effect on depression was considerably smaller. However, their strong effect size should probably be interpreted with caution, for two reasons: first, their inpatient CBGT explicitly focused on encouraging patients to start subsequent individual outpatient psychotherapies and thus it seems very likely that the follow-up effect also reflects improvements through these subsequent therapies; and second, the results were based on a completer sample with unknown size and thus might overestimate the effect. Comparisons with studies on other waiting list interventions cannot be drawn because these studies did not report effects on mental outcomes [7].

Effects on secondary outcomes were mixed. STEpS resulted in higher health-related quality of life scores at post-treatment. The effect on self-reported mental health was small (d = 0.47) and the effect on self-reported physical health was medium (d = 0.70). However, STEpS had no effect on global psychological distress. This may be explained by the severe burden due to the multimorbidity of the sample. Patients reported at least a depressive or an adjustment disorder, although more than half had further mental diagnoses. Taking this into account, it is unsurprising that our intervention – which mainly focused on dealing with depressive symptoms – did not lead to reduced global psychological distress.

The benefits of the STEpS group in depression and mental health did not remain stable. Eight weeks after the end of the treatment, there were no significant differences between the groups. This is in line with findings from the above-mentioned studies examining CBGT in patients who suffered from the same physical diagnosis, which also failed to reveal long-term effects (at 3-month, 4-month or 6-month follow-up) on depression [18‐20]. One reason for the symptom deterioration could be that the support of the group terminated and the individual therapy did not start right away, whereby patients may have felt left alone again. Receiving something that seems helpful and then having it taken away may have triggered this adverse effect. Indeed, when considering the multimorbidity of the sample, it again becomes quite understandable why eight sessions CBGT did not lead to follow-up effects. To prevent this rebound in depression and mental health after group termination, it could be promising to reduce the interval between the last group session and the beginning of the individual therapy by a) prolonging the group therapy to about 12 sessions, b) adding booster sessions or c) increasing the intervals between the eight sessions. Future research should clarify whether this could lead to better follow-up effects. It seems a little counterintuitive that STEpS had the largest and most stable effect on self-reported physical health. One hypothesis for this is that the exchange with other patients suffering from different physical diseases and seeing one’s own health in relation to the poor health of others could have resulted in a re-evaluation of one’s own health status. Another hypothesis could be that the behavioral activation intervention led to new positive experiences that changed the evaluation of their health status.

About half the patients in our sample had previous psychotherapy. Further analyses showed that IG patients with or without previous treatment experience did not significantly differ in depression at post-treatment (B = −1.61, SE = 0.73, CI = −1.59 – 1.27, T = −0.22, p = .825). Thus, the group intervention even seemed to be helpful for patients with previous treatment experience.

Overall, a large spectrum of depressive disorders is represented in our study sample. Only severe depressive episodes were underrepresented, although this in turn is representative for the German outpatient setting [35]. Moreover, the study sample had a large spectrum of physical diseases and mental comorbidities and were allowed to have previous psychotherapy experience or medication. All of these points enhance the external validity of our study and contribute to the comparability of the study sample to the routine patient care [36].

The fact that 57 of 97 (59%) of the self-referred patients were interested in participating in the study may indicate the strong need for waiting list interventions. Compared to the 14% who made use of the waiting list group in the study of Stone and Klein [12], this rate emphasizes the attractiveness of our intervention. The STEpS group sessions were highly accepted. This can be concluded from the attendance rate of 76% and the aforementioned predominantly external reasons for non-completion. The rate of patients who attended seven or eight out of eight sessions (58%) also supports acceptance. The attendance rate of the subgroup of self-referred patients was lower (65%), which might indicate a lower attendance rate under routine conditions. It must also be noted that 42% of the target group refused study participation. One of the main reasons for refusal seemed to be the group setting. Thus, it could be hypothesized that the implementation of group therapy as waiting list intervention in standard routine may not serve the needs of all patients.

This study contributes to the limited evidence on the efficacy of CBGT in patients with physical diseases and comorbid depressive or adjustment disorders. It is the first RCT to examine the efficacy of a waiting list group intervention, as well as the first RCT to examine the efficacy of CBGT in outpatients with various physical diseases and comorbid depressive or adjustment disorders. The waiting list group intervention proved effective in improving depression and health-related quality of life in the short term, but not at 2-month follow-up. Even though the intervention revealed only small to medium short-term effects on these outcomes, its implementation as a waiting list intervention could help multimorbid patients to handle long waiting periods for individual therapy. Therefore, the presented waiting list intervention contributes to a main problem in outpatient mental care in Germany. Future research should focus on methods to obtain better follow-up effects.