Efficacy of short-term psychotherapy for multiple medically unexplained physical symptoms: A meta-analysis

Abstract

Multiple medically unexplained physical symptoms (MUPS) are considered to be difficult and costly to treat. The current meta-analysis therefore investigates the efficacy of short-term psychotherapy for MUPS. Based on a multiple-phase literature search, studies were selected according to a-priori defined inclusion criteria. The standardized mean gain was used as the effect size index. Separate data aggregation of between- and within-group contrasts was performed on the basis of a mixed effects model. Outcome variables were physical symptoms, disorder specific emotions, cognitions and behaviors, depressive symptoms, general psychopathology, functional impairment, and health care utilization. Based on 27 included studies, small between-group effect sizes (range: d₊ = 0.06–d₊ = 0.40) and small to large within-group effect sizes (range: d₊ = 0.36–d₊ = 0.80) were found for post-treatment and follow-up assessments for the different outcome variables. Significant moderator variables were identified as the type, mode, and setting of therapy, number of therapy sessions, profession of therapist, age and sex of patients, quality of diagnostic procedure, and the control of concomitant treatments. Implications of the results for clinical practice and future research are discussed.

Introduction

Individuals seeking medical care due to physical symptoms of unknown aetiology play an important role in our health care system. Several studies found disproportionately elevated rates of medical care utilization for these patients (e.g., Barsky, Orav, & Bates, 2005), as well as substantial psychosocial disabilities and high levels of psychological distress (e.g., De Waal, Arnold, Eekhof, & Van Hemert, 2004). The variability of diagnostic terms for these physical symptoms with no identifiable organic causes is confusing. In the literature, often the global term “medically unexplained physical symptoms (MUPS)” is used. In the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association, 1994) as well as in the International Classification of Diseases (ICD-10; World Health Organization, 1992) we find either very strict criteria for somatization disorder or low-threshold criteria for the residual category of undifferentiated somatization disorder. Less stringent criteria are required for “abridged somatization disorder” as introduced by Escobar, Rubio-Stipec, Canino, and Karno (1989; Somatic Symptom Index, SSI-4,6), “multisomatoform disorder” as proposed by Kroenke et al. (1997), and “polysymptomatic somatoform disorder” as suggested by Rief and Hiller (1999).

Independent of this variability in terminology, all diagnostic concepts have in common that they refer to patients suffering persistently from multiple physical symptoms with no identifiable organic pathology. The chronic nature of these symptoms often leads to a decrease in levels of activity in everyday life, working capacity and role functioning (De Waal et al., 2004). As a consequence, MUPS are usually related to high levels of psychological distress and high rates of comorbid psychiatric disorders. In a study by Barsky et al. (2005), 31% of the participants with somatization symptoms fulfilled criteria of a major depressive disorder and 19% had a panic disorder. Effective psychological management is needed for these patients in addition to medical examination and treatment of symptoms.

The management of MUPS and the motivation of somatizing patients for seeking psychological treatments do however pose particular challenges for clinicians. An important problem in treating this highly-stressed clinical population includes their usually somatically focused health beliefs. They often seek to find a “true” medical diagnosis and a “correct” medical treatment. In turn, this leads to an unnecessary dependency on medical care whilst the psychosocial problems are not addressed (Grol, 1986). Their physicians try to rule out treatable medical conditions, and when none are found, physicians shift to constraining the help-seeking behavior of their patients. This shift is often associated with a deterioration of the physician–patient relationship, as patients think that the physician has lost interest in them or views them as annoying (Kirmayer, 1994). Furthermore, if the clinicians refer their patients for psychological treatment, they often consider the act of referral as calling the reality of their physical symptoms into question (Kirmayer, 2000).

Taking all these aspects into account, it seems to be necessary to develop effective psychological treatments, tailored to meet the specific and partly conflicting demands of patients suffering from MUPS. Fortunately, the number of such approaches has increased over the past decade. Important developments can be identified in cognitive-behavior therapy in particular, where modern treatment concepts focus on restructuring patients' cognitions about their symptoms (Speckens, Spinhoven, van Hemert, Bolk, & Hawton, 1997) and the reattribution of physical symptoms to psychosocial problems (Goldberg et al., 1989, Morriss et al., 2006), or aim to enhance activity levels, reduce physiological arousal and to improve awareness and communication of emotions (Allen, Woolfolk, Escobar, Gara, & Hamer, 2006). Further elements of cognitive-behavior interventions for patients with MUPS are biofeedback (Nanke & Rief, 2003), relaxation techniques (Zaby, Heider, & Schröder, 2008), and problem-solving approaches (Wilkinson & Mynors-Wallis, 1994). In addition, there are some approaches adapting psychodynamic concepts for patients with somatization disorder. Uexküll and Köhle (1996) for example, emphasize the functionality of physical symptoms and a maladaptive emotional processing of bodily sensations. An interpersonal approach by Stuart and Noyes (2006) on the other hand, focuses on somatizing behavior as a form of interpersonal communication driven by insecure, anxious attachment, and uses specific interpersonal techniques aimed primarily at improving interpersonal functioning.

Despite the progress that has been made in the development of specific psychotherapies, there does not seem to be sufficient implementation in clinical practice. This phenomenon is reflected in the suboptimal procedures undergone by patients relating to high treatment costs, the chronic course of symptomatology (Lieb et al., 2002) and comorbidity of MUPS (De Waal et al., 2004). For this reason, it is important to develop treatment guidelines organizing the treatment procedures of MUPS efficiently. As a basis for treatment guidelines, the summary of existing evidence of the efficacy of the different treatment concepts is needed. Fortunately, linked to the increasing number of treatment approaches for MUPS over the past decade, the number of clinical trials has also increased. Nevertheless, an extensive literature search obtained only systematic reviews (summarized in Table 1) but no formal meta-analyses of psychological treatments for MUPS. Furthermore, several existing reviews focus on specific, monosymptomatic somatoform disorders such as irritable bowel syndrome, or other forms of somatoform disorders, for example hypochondriasis (see Table 1). We found six reviews addressing MUPS (Allen et al., 2002, Kroenke, 2007, Kroenke and Swindle, 2000, Looper and Kirmayer, 2002, Nezu et al., 2001, Sumathipala, 2007), however they are not up to date and also do not apply meta-analytical strategies. In addition, only a few of the reviews include studies with simple pre-post designs (see Table 1), even though, according to the differentiation between efficacy and effectiveness research, it is not only important to evaluate the efficacy of an intervention in a controlled clinical setting but also the applicability and feasibility of an intervention in real-world setting (Nathan, Stuart, & Dolan, 2000). The latter is often realized in less standardized and uncontrolled designs.

The current meta-analysis aims to give an overview of existing clinical trials of psychotherapy for chronic multiple MUPS and to summarize the data using meta-analytical strategies. Research literature was not limited specifically to short-term therapies, although we found that studies of long-term psychotherapy for MUPS do not yet exist. The effects of psychotherapy and their stability will be analyzed separately for different outcome variables as well as both within- and between-group contrasts. Furthermore, possible moderators of the efficacy of psychological treatment approaches for MUPS will be evaluated.