Article Text
PDF
Statistics from Altmetric.com
What is already known on this topic?
Many patients who experience significant health anxiety and somatisation, and who also over-utilise medical services, are unresponsive to standard medical care.1 Cognitive behavioural therapy (CBT) has been shown to provide modest reductions in symptomatology and healthcare utilisation in these patients.1 Relaxation training (RT) also has been shown to benefit this population, but has been less widely studied.1
What does this paper add?
-
Given that somatising patients often fail to pursue mental health referrals, stepped-care approaches to treatment within primary care have been suggested. Stepped-care approaches provide a very brief intervention to all patients and a more intensive treatment to the subset of patients who are unresponsive to the initial brief intervention. Barsky and colleagues’ study extended the scope of previous research on treatment of somatising patients1 by examining the effectiveness of a stepped-care cognitive-behavioral approach administered by primary care clinicians in a primary care treatment setting.
-
This is the first direct comparison, conducted in primary care, of the clinical effectiveness of cognitive behavioural therapy and relaxation training for easing hypochondriacal beliefs and somatisation symptoms.
Limitations
-
Although the study examined a group of patients with high rates of healthcare utilisation, the patients were relatively high-functioning compared with those examined in earlier research.2 It is not clear that all the patients treated were either high in somatisation or hypochondriasis. Fewer than one-third of randomised participants met Diagnostic and Statistical Manual-IV criteria for hypochondriasis and their mean Whitely Index score and Somatic Symptom Inventory score were lower than those reported in previous trials.3 High-utilising patients who were neither somatising nor hypochondriacal would not be expected to benefit from the interventions examined.
-
The study design did not include a no-treatment or treatment-as-usual control group.
-
Many of the conclusions of the study are based on correlational data and a posteriori inferences, rather than confirmation of a priori hypotheses.
What next in research?
The outcome of primary interest for high-utilising patients would be change in healthcare utilisation. Much larger datasets are required to make meaningful comparisons of healthcare costs. Also, whether primary care clinicians can be trained to administer psychosocial treatments as effectively as specialists with more intensive training is yet to be examined.
Could these results change your practices and why?
This study bolsters the conclusions of other research indicating that brief psychosocial treatments administered in primary care can benefit patients who are high utilisers of medical services.4 The study serves as an important reminder of the potential benefits of relaxation training for somatising patients. Given that we employ similar methods to those studied in this research, the findings of this study would not alter our approach to treatment.
References
Commentary
ABSTRACT FROM: Barsky AJ, Ahern DK, Bauer MR, et al. A randomized trial of treatments for high-utilizing somatizing patients. J Gen Intern Med 2013;70:1396–404.
Participants Eighty-nine people selected from the highest 20% of healthcare utilisers in the prior year, identified through encounter claims data, and whose names were approved by their clinicians. Potential participants completed a telephone screening questionnaire to measure levels of medically unexplained symptoms and health-related anxiety. Those who scored above a predetermined cut-off were eligible to participate. Participants were excluded if they had more than one visit to the same subspecialty or more than one narcotic prescription during the previous year, a diagnosis of alcoholism or psychosis, symptom-contingent litigation or disability compensation or were under 18 or over 70 years of age. Less than one-third of participants met Diagnostic and Statistical Manual-IV (DSM-IV) criteria for hypochondriasis.
Setting Six primary care suites in two large, multispeciality community practices in the USA.
Intervention Cognitive behavioural therapy (CBT, n=59) versus relaxation therapy (RT, n=30). CBT consisted of two steps. The first step was of three, manualised 60 min counselling sessions focusing on symptom amplification, cognitive distortions and misunderstandings about symptoms, disease and medical care. Participants judged to have not improved after this step had a second step of six further, tightly manualised 60 min counselling sessions addressing dysfunctional beliefs and maladaptive illness behaviours, and teaching distraction techniques (including diaphragmatic breathing in half of one session). Relaxation therapy consisted of three tightly manualised 30 min sessions covering a range of relaxation techniques, including diaphragmatic breathing. Participants were given practice CDs and asked to practice for 20 min daily. Step 1 CBT and RT were delivered by practising clinicians based routinely at the clinic who had received three 60 min training sessions; step 2 CBT was delivered by psychologists.
Comparison CBT versus relaxation therapy (RT). There was no control arm.
Patient follow-up Seventy-eight per cent of CBT participants and 77% of RT participants completed 12-month follow-up. Twenty-four per cent of CBT participants received step 2.
Allocation Cluster randomisation by practice suite.
Blinding Unclear.
OUTCOMES
Hypochondriacal and somatisation symptoms at 12 months
Participants in both treatment arms showed significant reduction from baseline in hypochondriacal attitude (Whitely Index score, p value not reported), somatisation (Somatic Symptom Inventory, p<0.01 for both groups), health anxiety (Health Anxiety Inventory, p<0.01 for both groups), hypochondriacal cognitions (frequency and beliefs on the Medical Cognitions Questionnaire, p<0.01 for all but beliefs with RT (p=0.03)), role impairment (Functional Status Questionnaire, p<0.01 for CBT and p=0.05 for RT) and psychiatric symptoms (Hopkins Symptoms Checklist-90, p<0.01 for both groups) (see webextra table). There was no significant difference between the two treatment arms for any outcome, at any time-point.
Medical utilisation
The two treatment arms were combined to test the effect of treatment on healthcare utilisation. Ambulatory visits significantly decreased from an average 10 visits/year 1 year before the trial, to 8.8 visits in the year following the trial (p=0.036).
Footnotes
-
Competing interests None.