The Stress in America™ survey 2015 (American Psychological Association, 2016) revealed that 24% of the adult population experienced high stress-levels and that 34% perceived increased stress during the last year (vs. 16% who reported decreased stress). These are alarming results given the fact that a high stress-level is a risk factor for several metabolic and mental disorders (e.g., Cohen et al., 2007; Lupien et al., 2009; Provencal and Binder, 2015), more specifically, for example, for cardiovascular disease (e.g., Dimsdale, 2008; Steptoe and Kivimäki, 2012; Booth et al., 2015) and depression (e.g., van Praag, 2004; Wang, 2005; Vinkers et al., 2014).

These results underline the relevance of stress management/coping. Folkman and Lazarus (1980, 1985) distinguished problem-focused (addressing the source of the stress) from emotion-focused (tackling the stress-induced emotions) coping. Another distinction is dysfunctional vs. functional coping. A dysfunctional coping strategy leads to problems even though the stress-level may be reduced, whereas a functional coping strategy has no harmful effects. The functionality of certain coping strategies also depends on how extensively they are applied: They might be functional in the short-term, but dysfunctional in the long-term (e.g., repression).

Behavioral and mental disengagement (Carver et al., 1989) or self-harm (e.g., Chapman et al., 2006; Kleindienst et al., 2008) and related drug abuse (e.g., Brady and Sonne, 1999; Sinha, 2001) as well as unhealthy eating behavior (e.g., Greeno and Wing, 1994; Torres and Nowson, 2007) are examples of dysfunctional coping. Functional coping includes, for example, exercising mindfulness (e.g., Chiesa and Serretti, 2009; Khoury et al., 2015), practicing relaxation techniques (e.g., Carlson and Hoyle, 1993; Ernst and Kanji, 2000; Chiesa and Serretti, 2009), taking part in psychoeducation (PE) (e.g., Donker et al., 2009; Van Daele et al., 2012), or being physically active (e.g., Penedo and Dahn, 2005; Warburton et al., 2006).

Many of these functional coping strategies make use of bodily processes. For example, tensing and relaxing the muscles of different body parts is essential in progressive muscle relaxation, and several mindfulness-based interventions (e.g., body-scan, breathing meditation, yoga) rely on observing and appreciating body sensations (see Burg et al., 2017). Another intervention with a body focus is “Functional Relaxation” (FR; Fuchs, 2013). FR originally belongs to the psychodynamically based body-oriented psychotherapy methods most frequently used in Germany, Austria, and Switzerland. Minute movements of small joints, hardly noticeable to observers, are performed during relaxed expiration, accompanied by observing differences of body feelings triggered by these movements. Thereby, FR focuses on the discovery of proprioception by exploring the perceived differences of body sensations and the finding of one's own rhythm, aiming at rebalancing the autonomous nervous system triggered by small movements of the joints. In contrast to exercise-based methods such as, progressive muscle relaxation, and in accordance with the theoretical foundation in psychodynamic psychotherapy, the resulting sensations are described either nonverbally or explicitly during the therapeutic interaction.

Up to now, studies on FR solely rely on clinical samples and the results support the efficacy of FR for the following disorders: Somatoform heart disorder (Lahmann et al., 2008a), asthma (Loew et al., 1996, 2001; Lahmann et al., 2009, 2010a), irritable bowel syndrome (Lahmann et al., 2010b), chronic headache (Loew et al., 2000), and dental anxiety (Lahmann et al., 2008b).

Although these studies have shown that FR can reduce clinically relevant problems, it remains unclear whether FR is an effective intervention for stress in non-clinical samples. Reducing stress in non-clinical samples is important as this might prevent clinically relevant problems. Therefore, this randomized controlled trial was conducted, which compared the combination of FR and psychoeducation (PE) with PE alone in individuals with elevated stress-levels. The hypotheses were that the condition offering FR + PE is superior to a condition with PE alone at the end of the intervention as well as at 6-month follow-up to reduce stress-levels (primary outcome) as well as depression and somatization (secondary outcomes).

The Ethics Committee of the Technical University of Munich approved the trial and the applied materials and methods. The study was conducted in accordance with the approved guidelines. Written informed consent was obtained from all subjects in accordance with the Declaration of Helsinki.

The materials and the data are available upon request from the first author.

The current randomized controlled trial compared FR + PE with PE alone. FR + PE was superior to PE in reducing stress-levels at the end of the intervention as well as at 6-month follow-up. Moreover, FR + PE was more beneficial than PE to reduce somatization as well as depression.

The effect sizes for PE to reduce stress-levels as measured with the PSQ (end of intervention: d = 0.57; 6-months follow-up: d = 0.67) fall into the range of the corresponding effect sizes (end of intervention: d = −0.03 to 0.89; follow-up: d = −0.10 to 0.78) reported in a meta-analysis on PE for stress-reduction (Van Daele et al., 2012). The effect sizes for PE were, however, higher than the overall effects found in the cited meta-analysis (end of intervention: d = 0.27; follow-up: d = 0.20). For FR + PE, the effect sizes to reduce stress-levels were large. And large effect sizes have also been shown for mindfulness-based stress-reduction on measures of stress-levels in non-clinical samples (Chiesa and Serretti, 2009; Khoury et al., 2015). Future trials could directly compare FR with mindfulness-based interventions to reveal whether one of these approaches is more beneficial for stress. With regard to depression, FR + PE produced large effects, whereas Khoury et al. (2015) reported that mindfulness-based stress reduction (MBSR) has a moderate effect on measures of depressive symptoms in healthy people. Yet, pre-treatment differences in depressive symptoms might also explain why FR + PE produced large improvements of depression, whereas MBSR resulted in moderate improvements. Our sample had depression scores on the PHQ-9 (Table 3: estimate for FR + PE = 10.60; estimate for PE = 10.03) typical for mild and moderate depression (Kroenke et al., 2001), and the studies on MBSR might have analyzed data from samples with lower depression scores.

A shortcoming of the current study is that no structured or standardized clinical interviews were applied to have information whether and how many of the stressed participants had a depressive or any other mental disorder. Another limitation is the fact that only self-reports were used as outcome measures. Besides subjective stress responses, physiological parameters (e.g., salivary cortisol, heart rate, electrodermal activity) should be considered important complementary data sources to evaluate stress-reduction interventions. Moreover, the use of newer technologies (e.g., smartphones) might be fruitful in future research to measure outcome variables not only at pre-, post-, and follow-up assessments but also as ecological momentary assessments on a daily life basis (e.g., Trull and Ebner-Priemer, 2013; Adams et al., 2017). Moreover, PE might not be the best control condition since psychoeducational conditions “may reduce the likelihood of detecting a true effect in the intervention arms of the trial. Therefore, alternatives to psychoeducational intervention as control groups (for example, attention placebo) are recommended, in order to avoid bias in study outcomes.” (Donker et al., 2009, p. 8).

Despite these limitations the study at hand has also several strengths. The representativeness of the results is increased by the fact that several FR therapists took part in the study. Another strength (with regard to internal validity) is that each therapist treated FR + PE participants as well as PE participants. If the therapists had been different between the conditions, this could have negatively affected the internal validity (potential differences in experience, therapeutic skills,…). Nevertheless, it is possible that there were differences between therapists within each condition. “That is, even though the therapists are delivering the same specific ingredients, some therapists will do so more skillfully and therefore achieve better outcomes than other therapists delivering the same treatment” (Wampold, 2015, p. 274). A meta-analysis reported that therapist effects amount to d = 0.35 in clinical trials (Baldwin and Imel, 2013). Yet, the variance component (for the primary outcome) associated with the therapists was below 1% and statistically insignificant in the current study. A further strength regarding the internal validity is that manuals were used to standardize the delivery of the interventions across the therapists. Although the use of manuals, the same therapists in both conditions, and the randomized design increase the internal validity of the result, the following threats to the internal validity have to be kept in mind, too.

First, the time interval between the pre-treatment assessment and the post-treatment assessment was shorter for PE (approx. 6 weeks) than for FR + PE (10–12 weeks), so that spontaneous remission of elevated stress-levels might have occurred more probably for participants of the FR + PE condition. Yet, this potential confounder is relevant for the short-term treatment effects only, since the time interval from pre-treatment to follow-up (6-months) was the same for both conditions.

Second, the participants of the FR + PE condition received a higher dose (10 × 90 min) than participants of the PE condition (2 × 90 min). Therefore, it is possible that the different dose contributed to the differences between FR + PE and PE in the outcome variables. However, PE shorter in duration has been shown to be more beneficial than PE with longer durations to reduce stress-levels (Van Daele et al., 2012) and, therefore, matching the length might lead to results even more in favor of FR + PE. Moreover, an extended PE of 10 sessions might contain a lot of redundancy and could lead to more drop-outs.

Third, the conditions also differed in supervision: While therapists were supervised for participants of the FR + PE condition, no supervision was provided in the context of the study for participants of the PE condition. Thus, therapist competence and adherence to the treatment manual, for example, might have been higher in FR + PE than in PE. However, inconsistent findings exist whether supervision leads to better outcomes (Watkins, 2011) and a meta-analysis showed that adherence (r = 0.02) as well as competence (r = 0.07) contribute only marginally to the outcome (Webb et al., 2010).

Future trials with larger samples are needed that take these threats to the internal validity and the other shortcomings into account in order to replicate that FR is beneficial for stress. These further studies should also use structured or standardized clinical interviews to investigate whether FR prevents clinical disorders through stress-reduction.

In summary, the study at hand provides first evidence that the combination of FR and PE is more suited than PE alone for stress. FR might be a potent component of stress interventions. Adding FR to such interventions might better help prevent clinically relevant disorders such as depression or somatization.

CL was the primary investigator of this study, drafted and revised the manuscript; MG realized the follow-up assessment and revised the manuscript; HS contributed to the statistical analyses and revised the manuscript; AD contributed the study design and revised the manuscript; CP contributed to the study design and revised the manuscript; TP performed the statistical analyses and drafted the manuscript.