روند رفتار درمانی شناختی برای سندرم خستگی مزمن: کدام تغییرات در تداوم شناخت و رفتار به کاهش خستگی مربوط می شود؟
کد مقاله | سال انتشار | تعداد صفحات مقاله انگلیسی |
---|---|---|
30111 | 2013 | 7 صفحه PDF |
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Psychosomatic Research, Volume 75, Issue 3, September 2013, Pages 235–241
چکیده انگلیسی
Objective Cognitive behaviour therapy (CBT) can significantly reduce fatigue in chronic fatigue syndrome (CFS), but little is known about the process of change taking place during CBT. Based on a recent treatment model (Wiborg et al. J Psych Res 2012), we examined how (changes in) cognitions and behaviour are related to the decrease in fatigue. Methods We included 183 patients meeting the US Centers for Disease Control criteria for CFS, aged 18 to 65 years, starting CBT. We measured fatigue and possible process variables before treatment; after 6, 12 and 18 weeks; and after treatment. Possible process variables were sense of control over fatigue, focusing on symptoms, self-reported physical functioning, perceived physical activity and objective (actigraphic) physical activity. We built multiple regression models, explaining levels of fatigue during therapy by (changes in) proposed process variables. Results We observed large individual variation in the patterns of change in fatigue and process variables during CBT for CFS. Increases in the sense of control over fatigue, perceived activity and self-reported physical functioning, and decreases in focusing on symptoms explained 20 to 46% of the variance in fatigue. An increase in objective activity was not a process variable. Conclusion A change in cognitive factors seems to be related to the decrease in fatigue during CBT for CFS. The pattern of change varies considerably between patients, but changes in process variables and fatigue occur mostly in the same period.
مقدمه انگلیسی
Chronic fatigue syndrome (CFS) is diagnosed when someone experiences medically unexplained, severe fatigue that has been present for at least six months and results in severe impairment in daily functioning. The fatigue is accompanied by at least four out of eight of the following symptoms: post-exertion malaise, unrefreshing sleep, short-term memory/concentration problems, muscle pain, joint pain without swelling or redness, tender lymph nodes in the neck or armpit, sore throat or headaches [1] and [2]. Several cognitive behavioural models of CFS exist, which try to explain the perpetuation of symptoms in CFS patients [3], [4], [5] and [6]. All start with a period of severe fatigue that many patients, rightfully or not, attribute to a physical illness. In order to recover, patients often rest and become less active. Most models assume that because of this inactivity, patients' physical condition will decline and they will become increasingly sensitive to fatigue. In one of these models, that of Vercoulen et al. [7], a low level of physical activity, a low sense of control over fatigue, and a high level of focusing on symptoms explain the perpetuation of fatigue and impairment. The low level of physical activity is related to somatic attributions, as patients may avoid physical activity when they attribute their symptoms to a somatic illness. Although there are more recent models that also include physiological and social factors [8], this is still the only model that simultaneously tested the influence of several perpetuating factors. Based on these models, several protocols for CBT were developed [9], [10], [11] and [12] that lead to a significant reduction of fatigue and impairment [13] and [14]. Although protocols differ, they all encompass a stepwise increase in physical activity and challenging of dysfunctional fatigue-related beliefs. Recently, Wiborg et al. developed a comprehensive treatment model of CBT for CFS [15], based on the model of Vercoulen et al. [7]. For their analyses they used data from previous randomised controlled trials (RCTs) testing the efficacy of CBT for CFS. Wiborg et al. first tested for each variable in the model of Vercoulen whether it was a mediator of CBT for CFS, i.e. whether it changed more in patients receiving CBT than in the control group and whether it indirectly explained (part of) the effect of treatment on symptom change. Two adaptations were made to the original variables: somatic attributions were not analysed, as previous research had shown that somatic attributions do not change during treatment [10], so they could never be a mediator of CBT; and perceived problems with activity were analysed, rather than objective activity assessed with actigraphy, as previous research had shown that objective activity or physical fitness do not mediate the effect of CBT for CFS [16] and other behavioural interventions for chronic fatigue [17] and [18]. In the final model, the decrease in fatigue is explained by an increased sense of control over fatigue, an increase in perceived activity, and improved physical functioning. There are three limitations to the model of Wiborg et al. First, the model relied only on measurements before and after therapy and did not study the process of change during therapy. We therefore do not know whether changes in mediating variables indeed precede changes in fatigue. This makes it impossible to draw causal conclusions. Furthermore, temporary changes, which may catalyse other change processes, will go unnoticed. For example, a permanent change in objective activity does not mediate the treatment effect, but a temporary increase in activity might lead to changes in cognitions, which then lead to a decrease in fatigue. Second, focusing on symptoms was measured with the subscale ‘somatic complaints’ of the Symptom Checklist 90. As the authors argued themselves, this questionnaire may have been unsuitable. Third, the model was tested in an RCT of a minimal intervention based on CBT for CFS [19]. Patients received a self-help booklet explaining the different components of CBT. In addition, they had regular email contact with a therapist. Tummers et al. showed that fatigue and impairment decrease significantly when patients follow additional CBT after such a minimal intervention [20]. So individual CBT may have a more complex mechanism of change, i.e. it may target more processes, than the minimal intervention Wiborg et al. tested. We built on the treatment model of Wiborg et al. by studying the change in different process variables during individual CBT. In Fig. 1 all proposed process variables are depicted. First of all, we selected the variables of the model of Wiborg et al., i.e. sense of control over fatigue, perceived activity and self-reported physical functioning. We added focusing on symptoms because this variable mediated the effect of CBT for CFS in an earlier RCT [21] and may have been measured inadequately in the study of Wiborg et al. These four variables had thus been proven to mediate the effect of CBT for CFS in formal mediation analyses of RCTs. Although objective activity was previously discarded as a mediator of CBT for CFS [16], we included this variable because a temporary change in objective activity might elicit other changes. Full-size image (20 K) Fig. 1. Potential process variables that we tested, based on the models of Wiborg [15] and Vercoulen [30].
نتیجه گیری انگلیسی
In Fig. 2 a flow chart depicting the selection process is presented. Between April 2008 and December 2010, 297 CFS patients meeting all inclusion criteria started CBT. Fifty-five refused to participate; the most frequent reason given was the perceived effort of having to do extra assessments during treatment. Ten patients decided not to start with therapy in the first session, before informed consent was given. Of the 232 patients who gave written informed consent, 192 completed at least two interim measurements during the study: Fourteen decided to stop therapy before 12 weeks, 14 no longer experienced severe fatigue (fatigue < 35) and therapist and patient agreed to stop, and 12 patients no longer want to participate in the study. At the end of treatment, 9 patients did not fill in the questionnaires. So data of 183 patients could be included in the analyses. They had on average 12.9 sessions (SD 3.6). In 49 of them one or more variables were missing, which were imputed using multiple imputation with 20 replications. Baseline characteristics are given in Table 1.