A two-year follow-up study of chronic fatigue syndrome comorbid with psychiatric disorders
Specification of field: general topics in psychiatry and related fields.
Aims: Chronic fatigue syndrome patients often have comorbid psychiatric disorders such as major depressive disorders and anxiety disorders. However, the outcomes of chronic fatigue syndrome and the comorbid psychiatric disorders and the interactions between them are unknown. Therefore, a two-year prospective follow-up study was carried out on chronic fatigue syndrome patients with comorbid psychiatric disorders.
Methods: A total of 155 patients who met the Japanese case definition of chronic fatigue syndrome were enrolled in this study. Comorbid psychiatric disorders were diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders 4th edition criteria. Patients with comorbid psychiatric disorders received psychiatric treatment in addition to medical therapy for chronic fatigue syndrome. Seventy patients participated in a follow-up interview approximately 24 months later.
Results: Of the 70 patients with chronic fatigue syndrome, 33 patients were diagnosed as having comorbid psychiatric disorders including 18 major depressive disorders. Sixteen patients with psychiatric disorders and eight patients with major depressive disorders did not fulfill the criteria of any psychiatric disorders at the follow up. As for chronic fatigue syndrome, nine out of the 70 patients had recovered at the follow up. There is no significant influence of comorbid psychiatric disorders on the outcome of chronic fatigue syndrome.
Conclusions: Chronic fatigue syndrome patients have a relatively high prevalence of comorbid psychiatric disorders, especially major depressive disorders. The outcomes of chronic fatigue syndrome and psychiatric disorders are independent. Therefore treatment of comorbid psychiatric disorders is necessary in addition to the medical treatment given for chronic fatigue syndrome.
CHRONIC FATIGUE SYNDROME (CFS) is a controversial but significant issue in medical practice and is the focus of much research and clinical interest. CFS is defined by unexplained persistent and profound disabling fatigue and a combination of nonspecific accompanying symptoms that prominently feature self-reported impairments in concentration and short-term memory, sleep disturbance, and musculoskeletal pain.1 In 1988, Holmes published the first formal case definition in the USA primarily to standardize the patient population for research studies, and the United States Centers for Disease Control and Prevention (CDC) suggested the name ‘chronic fatigue syndrome’.2 Subsequently, an Australian case definition for CFS was reported in 1990,3 a British case definition in 1991,4 and a Japanese case definition in 1991 (Table 1).5 In 1994, a revised version of the CDC case definition was published by Fukuda and an international collaborative group.6
Table 1. Japanese case definition for CFS prepared by CFS research group of Japan and the Ministry of Health, Labour and Welfare in 1991
| 1||Chief complaint of at least 6 months of persistent or multiple relapsing and disabling fatigue|
| 2||Exclusion of medical conditions that may explain the prolonged fatigue as major criteria|
| Symptoms|| |
| 1||Mild fever or chills|
| 2||Sore throat|
| 3||Lymph node swelling|
| 4||Unknown muscle weakness|
| 6||Prolonged (>24 h) fatigue after exercise|
| 8||Migratory arthralgia|
| 9||Neuropsychological symptoms (more than one); sensitivity to light, temporary visual blind spots, forgetfulness, excessive irritability, confusion, difficulty thinking, inability to concentrate or depression|
| 10||Sleep disturbance|
| 11||Sudden onset of symptoms|
| Signs||(confirmed twice at least monthly intervals)|
| 1||Low-grade fever|
| 2||Nonexudative pharyngitis|
| 3||Lymph node swelling|
Estimates for the prevalence of CFS range from 0.007% to 2.8% in the general adult population and from 0.006% to 3.0% in primary care or general practice in the USA and the UK.7 In Japan, the prevalence of CFS was reported to be 1.5% in the general population in 1998.8 However, estimates of the prevalence of CFS vary depending on which definition is used, the type of population surveyed, and the study methods used.9
The etiology of CFS still remains obscure. It has been suggested that CFS is associated with central nervous system abnormalities, immune system abnormalities, endocrine abnormalities, and metabolic abnormalities. Many investigators have postulated that CFS is a condition of complex and multifactorial etiology.7 Therefore, CFS can be understood to be a special condition based on an abnormality of the psycho-neuro-endocrino-immunological system caused by psycho-social stress that has some genetic component. There are no confirmatory physical signs or characteristic laboratory abnormalities.
CFS often causes neuropsychological symptoms, such as irritability, loss of concentration, depression, anxiety, and sleep disturbance. In addition, patients with CFS have been reported to be more likely to have comorbid psychiatric diseases such as major depressive disorder (MDD), panic disorder, generalized anxiety disorder, somatization disorder, adjustment disorder, and personality disorder.7,10–13 Several studies have reported that 45–82% of patients with CFS fulfill the operational criteria for another psychiatric disorder.7,11–15 Among them, 27–73% of CFS patients are reported to have comorbid mood disorders.7,11–18 The prevalence rates for anxiety disorders in CFS are estimated to range from 20 to 53%11,12,14,15,17,18 including panic disorder which ranges from 11% to 25%,14,15,17,18 and generalized anxiety disorder which ranges from 8% to 30%.7,14,15,17–19 The prevalence rate for somatization disorder in CFS is estimated to range from 1 to 28%.7,11,14,15,18 However the association between CFS and psychiatric disorders remains obscure.
The prognosis of CFS is generally considered poor. Cairns reported in his review of 28 articles that the median full recovery rate was 5% (range 0–31%) and the median proportion of patients who improved during follow up was 39.5% (range 8–63%). Return to work rates at the follow up ranged from 8 to 30%.20 Older age, longer illness duration, fatigue severity, and a physical susceptibility to CFS tend to be risk factors for poorer prognosis.21
However, there are few studies on the outcome of patients with CFS and comorbid psychiatric disorders. The aim of this study is to clarify the mid-term outcome of CFS and comorbid psychiatric disorders, and the influence of CFS on the outcome of comorbid psychiatric disorders.
From May 2005 to August 2007, 260 patients visited the Osaka City University Hospital Fatigue Clinical Center with a chief complaint of chronic fatigue. The patients found and chose our clinic via newspapers, television, their family doctor's advice, and our university website. They were requested to bring a referral letter from their primary doctor prior to an examination at our clinic. Subjects received an extensive intake evaluation including a standardized physical examination, a questionnaire on past and current medical problems, and a battery of laboratory tests. After several psychological assessments, they had a psychiatric interview to determine whether they had any psychiatric disorders. As a result, 155 patients met the Japanese case definition of CFS criteria,5 all of whom gave their written informed consent after the study was fully explained to them.
A total of 155 patients with CFS were called to visit our hospital bimonthly (on average) and underwent supportive care for CFS including the prescription of Hochu-ekki-to (TJ-41) as Kampo (traditional Chinese) therapy as well as vitamin B12 and C.22 When the patients were diagnosed as having comorbid psychiatric disorders, they underwent standard psychiatric treatment, including a combination of supportive psychotherapy and medication, in addition to medical therapy for CFS at our clinic. Approximately two years later, we tried to conduct a follow up on the outcome of CFS and to diagnose any comorbid psychiatric disorders present.
Diagnosis and measurements
Diagnosis for CFS was made according to the Japanese case definition for CFS, which was prepared by the CFS research group of Japan and The Ministry of Health and Welfare (at that time, now The Ministry of Health, Labour, and Welfare) in 1991 on the basis of the CDC case definition 1988 (Table 1).5
The diagnosis of psychiatric disorders was made according to the Japanese version of the Structured Clinical Interview for DSM-IV Axis I disorders (SCID-I).23,24
To assess the severity of CFS and depressive symptoms, the following instruments were administered:
- 1Performance Status (PS) score: A notable feature of the Japanese case definition is that the severity of CFS is measured using self-reported ratings based on daily activities. It is represented by the PS score (Table 2). It indicates the degree of fatigue, and its scores range from zero to nine.25,26 The higher the score is, the greater the degree of fatigue. The level of fatigue for a diagnosis of CFS must be PS level 3 or above.
- 2Self-rating Depression Scale (SDS): This consists of 20 items describing depressive symptoms. Respondents describe how frequently they experience each symptom on a 4-point scale ranging from ‘none or a little of the time’ to ‘most or all of the time’. A lower score represents a more favorable psychological state.27
- 3The Hamilton Depression Rating Scale (HAM-D): This scale is a 21-question multiple choice questionnaire that clinicians use to rate the severity of a patient's depression.28 It was originally published in 1960 by Max Hamilton and is presently one of the most commonly used scales for rating depression in medical research. The higher the score is, the greater the degree of depression.
Table 2. PS scores for evaluating the severity of fatigue in CFS patients
|0||No complaints; able to carry on normal activity without fatigue|
|1||Able to carry on normal activity, but sometimes feels fatigue|
|2||Able to carry on normal activity or to do active work with effort; requires occasional rest|
|3||Several days a month, unable to carry on normal activity or to do active work; requires rest at home without work|
|4||Several days a week, unable to carry on normal activity or to do active work; requires rest at home without work|
|5||Unable to carry on normal activity or to do active work at all, although able to perform light tasks; requires rest at home without work for several days a week|
|6||Requires rest without work at home for over one-half of a week; able to do light tasks in good health|
|7||Unable to carry on normal activity or to do light tasks at all; able to care for self without assistance|
|8||Remains in bed for more than one-half of each day; able to care for self to some extent, but requires frequent assistance|
|9||Unable to care for self; must remain in bed with day-long assistance|
This study was approved by the institutional review board of Osaka City University Graduate School of Medicine.
All statistical analyses were performed using SPSS 11.5J for Windows (SPSS Japan, Tokyo, Japan). The χ2-test with Yates' continuity correction test and Fisher's exact test were used to evaluate categorical and non-parametric data. For continuous variables, the normality of the distribution of scores for all measures was examined. Correlations among the parameters were evaluated using Spearman's correlation coefficient. Statistical significance was set at P < 0.05 (two-tailed).
The demographic and clinical characteristics of patients with CFS at the baseline assessment
Of the 155 patients with CFS, 70 (45%) completed the follow-up assessment. However, 85 (55%) patients did not complete the follow-up assessment. Of the 85, 47 (30%) patients stopped visiting our hospital, 28 (18%) declined to participate in this research without expressing a reason, and 10 (7%) patients refused to participate in this research because they viewed it as too time consuming.
Table 3 shows the demographic and clinical characteristics of the 70 CFS patients. The mean age of the patients was 32.7 years. Forty-seven (67%) of them were female, and 30 (43%) patients were married. The average number of years of education was 13.6. As for their occupation, 33 (47%) patients had been employed, and 29 (42%) patients had not been employed. The mean duration of CFS was 54.0 months. The mean duration of comorbid psychiatric disorders was 13.9 months. The average length of the observation period from the baseline to the follow-up assessment was 22.5 months. The mean PS score was 5.2.
Table 3. Demographic and clinical characteristics of Patients with CFS at the baseline (N = 70)
|Age (years)||32.7 ± 9.09||16–53|
|Sex|| || |
|Marital status|| || |
|Education (years)||13.6 ± 1.89||9–16|
|Occupation|| || |
|Duration of CFS (months)||54 ± 61.8||4–295|
|Duration of psychiatric disorders (months) (n = 33)||13.9 ± 28.2||2–120|
|Observation period between baseline and follow up (months)||22.5 ± 8.7||9–34.7|
|PS score||5.2 ± 1.69||2–9|
Comorbid psychiatric disorders with CFS at the baseline assessment
Of the 70 CFS patients, 33 (47%) were diagnosed as having lifetime comorbid psychiatric disorders according to SCID-I (Table 4). Of these 33 patients, 20 (60%) had mood disorders including 18 (55%) patients with MDD, six (18%) patients had anxiety disorders including five (15%) patients with panic disorders without agoraphobia, one (3%) patient had an anxiety disorder not otherwise specified, and nine (22%) patients had other psychiatric disorders [including five (15%) patients with adjustment disorders and four (12%) patients with sleep disorders]. Two patients had both MDD and panic disorders without agoraphobia. Of the 33 patients, 26 (79%) had current comorbid psychiatric disorders.
Table 4. Outcome of comorbid psychiatric disorders with CFS patients
| Major depressive disorder||18||14||10|
| Depressive disorder, NOS||2||2||1|
| Panic disorder without agoraphobia||5||3||2|
| Anxiety disorder, NOS||1||1||1|
| Adjustment disorder||5||4||0|
| Sleep disorder||4||4||3|
|Number of patients with comorbid psychiatric disorders||Total||33†||28||17|
The outcome of CFS and comorbid psychiatric disorders at the follow-up assessment
In this study, we evaluated the outcome of CFS based on the PS score. Subjects were categorized as ‘good or poor outcome’ for CFS depending on whether their PS score was below 3 points at the follow up. This definition complies with the diagnostic criteria of the Japanese case definition of CFS. Approximately 2 years later, out of the initial 70 CFS patients, nine (13%) patients had a ‘good outcome’ and 61 (87%) patients had a ‘poor outcome’.
Table 4 shows the outcome of the comorbid psychiatric disorders. Seventeen out of 33 (52%) patients had current comorbid psychiatric disorders. Among them, 11 (65%) had mood disorders including 10 patients with MDD and one patient with depressive disorders not otherwise specified, three (18%) patients had anxiety disorders including two patients with panic disorders without agoraphobia and one patient with an anxiety disorder not otherwise specified, and three (18%) patients had other psychiatric disorders such as sleep disorders. Of the 17 CFS patients with comorbid psychiatric disorders, 15 (88%) patients still fulfilled the criteria of psychiatric disorders that had been diagnosed at the baseline, and two (12%) patients were in a state of recurrence of past psychiatric disorders. One was suffering from MDD, while the other had panic disorder without agoraphobia.
However, the remaining 16 (48%) patients who had had comorbid psychiatric disorders at the baseline did not fulfill the criteria for any psychiatric disorders at the follow up. They included eight patients who had suffered from MDD, one patient who had suffered from a depressive disorder not otherwise specified, one patient who had suffered from panic disorder, and six patients who had suffered from other psychiatric disorders such as adjustment disorders.
Among the 37 CFS patients who were free from comorbid psychiatric disorders at the baseline, five (14%) patients were diagnosed as having comorbid psychiatric disorders at the follow up. They included three (8%) patients with MDD, one (3%) patient with panic disorder without agoraphobia, one (3%) patient with dysthymic disorder, and one (3%) patient with undifferentiated somatoform disorder. The total number of disorders was six, but there was one patient who had MDD and panic disorder without agoraphobia. So the number of new-onset patients was five. Therefore, of the 70 CFS patients whom we observed approximately two years later, 38 (54%) had comorbid psychiatric disorders in total.
The correlation between the outcomes of CFS and comorbid psychiatric disorders
To evaluate the effects of the outcome of comorbid psychiatric disorders on the outcome of CFS, the 33 patients with comorbid psychiatric disorders were divided into two groups: the 16 patients with comorbid psychiatric disorders at the baseline who did not fulfill the criteria of any psychiatric disorders at the follow up and the 17 patients with comorbid psychiatric disorders at the baseline who fulfilled the criteria of a psychiatric disorder at the follow up (Table 5). The outcome of CFS was not related to that of the comorbid psychiatric disorders (Fisher's exact test; P = 1).
Table 5. The relationship between the outcomes of CFS and comorbid psychiatric disorders
|Outcome of CFS||Good||4||1|| * ||5||4|| ** |
|Poor||12||16|| ||28||28|| |
|Total|| ||16||17|| ||33||32|| |
Then, to evaluate whether the presence of comorbid psychiatric disorders influences the outcome of CFS, we divided the patients into two groups: CFS patients with a comorbid psychiatric disorder at the baseline (N = 33) and CFS patients without a comorbid psychiatric disorder during the two-year observation period (N = 32) (Table 5). The relationship between the presence of comorbid psychiatric disorders and the outcome of CFS during the two-year follow-up period was analyzed. No significant correlation between the presence of comorbid psychiatric disorders and the outcome of CFS was found using Fisher's exact test (P = 0.17).
Correlation of PS, SDS, and HAM-D score
Of the 33 CFS patients with comorbid psychiatric disorders, the mean PS score at the follow up was 4.7 [standard deviation (SD) 2.0], and the average improvement was 12% (SD 26.18). The mean SDS score at the follow up was 49.0 (SD 11.47), and the average improvement was 7.0% (SD 15.06). The mean HAM-D score at the follow up was 8.8 (SD10.37), and the average improvement was 23% (SD 97.55). No significant correlation was noted between the improvement rates of the PS score and those of the SDS score or between the improvement rates of the PS score and those of the HAM-D score (Spearman's correlation coefficient 0.143 P = 0.426; 0.246, P = 0.167).
Of the 18 CFS patients with comorbid MDD at the baseline, the mean PS score at the follow up was 5.4 (SD 1.69) and the average improvement was 3.6% (SD 21.57). The mean SDS score at the follow up was 55.9 (SD 8.72), and the average improvement was 5.0% (SD 12.01). The mean HAM-D score at the follow up was 12.6 (SD 12.52), and the average improvement was 14% (SD 100.82). There was no significant correlation between the improvement rates of the PS score and those of the SDS score or the improvement rates of the PS score and those of the HAM-D score (Spearman's correlation coefficient –0.128, P = 0.612; 0.014, P = 0.955).
Comorbidity of psychiatric disorders in patients with CFS
The prevalence of psychiatric disorders among patients with CFS ranges from 45% to 82%.7,11–15 MDD is the most frequent diagnosis and is present in 27 to 73%7,11–18 of patients, followed by anxiety disorder, which is diagnosed in 20 to 53% of patients, while somatoform disorder accompanies 1 to 28% of CFS cases.7,11,14,15,18 In this study, 47% of CFS patients had lifetime comorbidity of psychiatric disorders. MDD (26%) was the most common disorder, followed by panic disorders (15%) and adjustment disorders (15%). Our results were compatible with those of previous studies. This may be partly because the demographic and clinical features of our sample are generally consistent with those in previous reports,29,30 and partly because we have followed the DSM-IV assessment system, which is considered to be the most popular means of diagnosing psychiatric disorder in individuals with CFS.31 Among the 37 CFS patients without any comorbid psychiatric disorders at the baseline, five (14%) patients developed comorbid psychiatric disorders during the two year follow-up period. Therefore, the prevalence of comorbid psychiatric disorders in our sample increased from 33 (47%) to 38 (54%), which reinforced the finding that CFS is associated with a high prevalence of comorbid psychiatric disorders, especially MDD.
The relationship between the outcomes of CFS and comorbid psychiatric disorders
Longitudinal studies of CFS patients have shown that the risk factors for poorer prognosis tend to be older age, longer illness duration, fatigue severity, comorbid psychiatric disorders, and a physical susceptibility to CFS.21 With regards to the presence of comorbid psychiatric disorders, Russo et al.,32 Skapinakis et al.,33 and Wilson et al.34 thought that they predicted a poor prognosis for CFS, while Hill et al.35 reported that comorbid psychiatric disorder did not predict the outcome of CFS. We examined the relationship between the presence of comorbid psychiatric disorders and the outcome of CFS after a two-year follow-up period. Of the 33 CFS patients with comorbid psychiatric disorders, five (15%) patients showed a good outcome of CFS, and the remaining 28 (85%) patients had a poor outcome. No significant relationship was found between the presence of comorbid psychiatric disorders and the outcome of CFS. From the results of this prospective, longitudinal study, it is suggested that the presence of comorbid psychiatric disorders with CFS is not associated with a poor outcome of CFS. This result is consistent with that of Hill et al.35 Conversely, our result is opposite to that of Russo et al., Skapinakis et al., and Wilson et al. This discrepancy may be due to differences in the length of the follow-up period, using variable measurements for rating both the severity and the outcome of CFS, the absence of a standard definition of ‘remission or recovery’ from CFS, and partly due to differences in treatment programs for CFS. In terms of treatment programs for CFS patients in this study, Hochu-ekki-to (TJ-41), vitamin B12 and C were prescribed, however antidepressants were not prescribed. There is no report showing any effect of Hochu-ekki-to on MDD. On the other hand, antidepressants were not prescribed in our study, because they are reported to produce a range of side-effects or not to be effective for CFS when they are given to patients.36
The symptoms of CFS often overlap with MDD, and there are no physiological diagnostic markers for CFS or MDD, which makes it difficult to draw clear lines between these disorders. Some researchers have raised the possibility that CFS is an atypical manifestation of MDD,37,38 despite the fact that CFS patients generally do not experience anhedonia, guilt, or diminished energy.12,39,40 However, there have been few reports on the outcome of MDD in patients with CFS. In this study, no significant relationship between improvements in the PS and SDS or HAM-D scores was found among the CFS patients with comorbid psychiatric disorders including MDD. These results suggest that improvements in PS score do not influence the SDS score or HAM-D score. But these data have to be interpreted with caution. That is, the SDS and HAM-D include several psychosomatic questions that will influence the evaluation of patients with CFS. Therefore patients with CFS are liable to make a high score on these depression scales. In this respect, to make a more precise evaluation of the depressive symptoms of CFS, the Beck Depression Inventory and the Montgomery-Asberg Depression Rating Scale might have been more appropriate.41,42 The outcome of CFS is independent from that of comorbid psychiatric disorders, especially MDD, which may be an important milestone to clarify the pathophysiology of CFS and MDD.
Our study has several limitations and our results must be interpreted in the context of these methodological limitations. First, our findings, which are based on the clinical population at our clinic that were recruited via referral letters for chronic fatigue, should be interpreted with caution, because they may not generalize to other populations. Furthermore, our population size was relatively small. The number of patients who completed the two-year follow-up study was 70 of 155 (45%). This follow-up rate is relatively low compared with previous reports.43,44 We have to consider that some patients might be in too severe a condition to come to our clinic, which could influence the results of our study. Therefore, we will need to make all kinds of efforts to increase the number of patients in future studies including phone interviews and mail surveys. Second, the present study lacks MDD controls. Comparing the outcomes of MDD patients with CFS to those of MDD patients without CFS would be useful in the future. Despite these limitations, this study is meaningful in that it is the first study that shows that the outcomes of CFS and comorbid psychiatric disorders are not related, with 48% of the CFS patients with a psychiatric disorder and 44% of the CFS patients with MDD achieving improvement during the two-year follow up. These results underline the necessity to appropriately treat such comorbid psychiatric disorders in addition to the physical treatment provided for CFS.
This research was conducted with support from the 21st Century COE Program ‘Base to Overcome Fatigue,’ from the Ministry of Education, Culture, Sports, Science and Technology, the Japanese Government. The authors are grateful to Professor Yasuyoshi Watanabe and Professor Yoshiki Nishizawa for their warm words of encouragement. The authors would like to express our respect to Osaka University Graduate School of Medicine who launched the CFS study in the 1990s.