This article was submitted to various
medical journals in 1994. Although it was never published, I believe
the conclusions are still relevant today.
Depression in Myalgic Encephalomyelitis and Chronic Fatigue States; a re-examination of the data.
Authors
Ellen M. Goudsmit
Chartered Health Psychologist. London, UK
Renee J.E. Faas.
Psychologist in private practice.
Haarlem. The Netherlands.
Many patients with myalgic encephalomyelitis and chronic fatigue syndrome suffer from emotional lability and psychological distress. However, efforts to determine the prevalence of clinically significant depression have been complicated by the marked differences between the populations studied. An analysis of the recent research indicates that where the selection of subjects is based on little more than the presence of unexplained chronic fatigue, the estimates of moderate and severe depression tend to be much higher than in the more strictly defined samples. Indeed, the rates for patients with M.E./CFS are not very different from those reported in medically ill populations, suggesting that the role of depression in the perpetuation of the former may have been overestimated.
A number of authors have suggested that the rate of affective disorders in myalgic encephalomyelitis (M.E.) and chronic fatigue syndrome (CFS) is significantly higher than that documented in comparable 'medical' conditions (David 1991, Manu et al 1993, Walford et al. 1993). Moreover, it has been proposed that depression plays a major role in perpetuating the symptoms of postviral and chronic fatigue syndrome (Butler et al. 1990, David 1991).
To assess whether this view is supported by recent research, we examined the existing literature on M.E. and CFS, noting the estimated prevalence of both marked depression and major depressive disorder. We then separated the data into two groups. The first consisted of the studies where all or the majority of the patients fulfilled the diagnostic criteria for M.E., post-viral fatigue syndrome or CFS (e.g. Holmes et al. 1988, Ho-Yen et al 1991, Lloyd et al. 1990). The second group was restricted to reports which had adopted less specific criteria for CFS (e.g. Sharpe et al. 1991). To avoid confusion, the latter will henceforth be referred to as chronic fatigue states (CF). Data was also collected from 10 of the most recent studies involving patients suffering from specific 'medical' diseases.
We limited ourselves to depression because this is claimed to be the most common of the psychiatric disorders in M.E. and CFS (David 1991, Manu et al 1993). All the estimates were based on standardised rating scales and/or structured and semi-structured psychiatric interviews e.g. the Present State Examination and Diagnostic Interview Schedule.
Where authors have published several studies giving different rates, the most often quoted figure was used. Likewise, where authors gave more than one estimate, the original text was checked to see which estimate the researchers regarded as the most significant, or if there was no indication, we selected the higher figure of two (Peterson et al 1991) or the middle of three (Millon et al 1989). In cases where the criteria for psychiatric diagnoses had been corrected to exclude somatic CFS symptoms, these were used (Katon et al. 1991) in preference to uncorrected figures.
Tables 1 and 2 give the rates of depression found in the most recent studies on M.E. and chronic fatigue states respectively, while Table 3 summarises the rates found in other conditions.
As the analysis shows, the average rate of depression in patients with M.E. is lower than those of people who fulfil the broader criteria for unexplained chronic fatigue, but similar to those of people suffering from 'medical' illnesses. Even if one removes the lowest figure from the M.E. category, and the highest one from the CF group, the difference between patients with M.E. and chronic fatigue states remains.
The results show that it is worth differentiating the research on M.E. from that on more general chronic fatigue states. We have chosen not to refer to the latter as chronic fatigue syndrome since this term covers a number of disorders, not just M.E. According to the criteria formulated by Sharpe et al. (1991), the majority of patients with M.E. should be listed under post-infectious fatigue syndrome (PIFS). In practice however, few British researchers distinguish between M.E. and other conditions characterised by unexplained chronic fatigue. Thus regrettably, one can not assume that the terms CFS and M.E. are synonymous (Macintyre and Hume 1993).
While it is clear that the majority of patients with chronic fatigue states suffer from emotional distress, and that the latter often signifies the presence of a psychiatric disorder (Manu et al. 1988 David 1991), it is important not to overlook the influence of other psychological and social stressors, for instance, the strain of looking after very young children (Popay 1992).
All such cases may present with chronic fatigue, myalgia, headaches and a host of other symptoms also seen in post-viral syndromes. However, there are a number of differences between these patients and those with M.E. (Faas 1992, Hickie et al 1990, Macintyre and Hume 1993). Two of the most important are the close relationship between minimal exertion and the onset of symptoms, and the marked fluctuations in the severity of the complaints.
Another distinguishing feature is the response to exercise. Recent research supports the observation that patients with M.E. do not respond as well to graded exercise and cognitive therapy as do people with more general chronic fatigue states (Lloyd et al 1993, Butler et al. 1991). Indeed, the effect of graded exercise regimes is now used by some clinicians to separate the two groups (Lescrauwaet, personal communication).
Where there is sound evidence of treatable affective disorders, we agree with colleagues that these must not be ignored, or dismissed as an "understandable" reaction to ill-health (Rodin et al 1991). However, in our experience, the majority of patients with M.E. suffer from less severe mood disturbances than those seen in psychiatric populations (see also Jenkins 1991, Yeomans and Conway 1991) and most will require little more than support and counselling.
The finding that the rate of clinical and marked depression in these patients is generally no greater than that found in medically ill populations supports the view that depressive disorders do not play a major role in the perpetuation of M.E. However, since the present analysis involved only 10 studies relating to physical illness, we also compared our results with the data from a more extensive survey (Rodin et al. 1991).
Selecting all the studies on neurological disorders (N=22) and cancer (N=17) which provided estimates for either current major depressive episode or marked depression, we found the means to be 33% and 25.3% respectively. These not only exceed the mean for patients with M.E. but also the rates reported for 'disabled' control groups in CFS research (Katon et al 1991, Wood et al 1991).
Another limitation of our research is the fact that certain studies were omitted from the analysis and it may be argued that this alone could be responsible for the low rates of depression. However, the majority of the excluded studies involved heterogeneous groups and therefore would not have affected the data relating to M.E. For instance, Kroenke et al (1988) investigated people who reported fatigue to be "a major problem". We excluded the data from Taerk et al (1987) because their criteria did not allow us to determine the proportion of patients with M.E. or CF and we omitted Kruesi et al (1989) because they only provided estimates for lifetime prevalence of depression (46.4%).
The findings of this analysis suggest that the prevalence of moderate depression and major depressive disorder in M.E. and strictly defined CFS is no higher than that documented in many medical disorders. However, where researchers have used less rigorous definitions focusing on unexplained chronic fatigue without signs of encephalopathy or continued infection, the rates of depression appear to be much higher.
We conclude that since there is little evidence to indicate that clinical depression is particularly common in M.E., the role of psychiatric illness in the aetiology of this disorder may have been overestimated.
Finally, we believe that until we know more about the mechanisms underlying M.E. and related conditions such as fibromyalgia, it may be more fruitful to continue to distinguish between these disorders, and not to label them all as chronic fatigue syndrome.
Depression % |
Measure used |
Authors |
|
0 |
HAD |
Ho-Yen et al. 92 |
|
13 |
Interview |
Yeomans and Conway 91 |
|
14 |
BDI |
Peterson et al. 91 |
|
20 |
POMS |
Millon et al. 89 |
|
22 |
Interview |
Riccio et al. 92 |
|
45 |
Interview |
Hickie et al. 90 |
Depression % |
Measure used |
Authors |
|
10.2 |
Interview |
Katon et al. 91 |
|
23.5 |
Interview |
Wood et al. 91 |
|
36 |
Interview |
Manu et al. 88 |
|
42 |
Interview |
Gold et al. 90 |
|
47 |
Interview |
Wessely and Powell 89 |
|
52 |
BDI |
Blakely et al. 91 |
|
56* |
HAD (A+D) |
Sharpe et al. 92 |
Depression % |
Measure |
Group |
Authors |
|
3.2 |
Interview |
R.A. |
Katon et al. 91 |
|
11 |
Interview |
S.L.E. |
Hay et al. 92 |
|
12.5 |
Interview |
Muscle Dis. |
Wood et al. 91 |
|
19 |
Interview |
M.I. |
Forrester et al 92 |
|
25 |
HAD |
Rectal cancer |
MacDonald 1988 |
|
30 |
BDI |
Diabetes** |
Leedom et al. 91 |
|
38 |
BDI |
Chronic pain |
Blakely et al. 91 |
|
40 |
Various |
Parkinson's |
Cummings 92 |
|
40 |
HAD |
Breast cancer |
Greer et al. 92 |
|
40 |
BDI |
M.S. |
Minden et al. 87 |
Ellen M. Goudsmit ©
Chartered Health Psychologist. London, UK
Renee J.E. Faas. ©
Psychologist in private practice.
Haarlem. The Netherlands.
All rights reserved. This article
may not be reproduced without
permission from the authors. See the full
copyright
notice.
Be sure to see the many other valuable articles at our Main M.E. Home Page
