Prognosis for myalgic encephalomyelitis and chronic fatigue syndrome

Prognosis refers to the expected course of a disease and the prospect of recovery. The prognosis for myalgic encephalomyelitis and chronic fatigue syndrome (ME and CFS) is considered to be poor with only a small minority (a median estimate of 5%) returning to pre-morbid levels of functioning. The majority of patients remains significantly impaired. A substantial improvement however is noted in an estimated 40% of patients and the prognosis in adolescents is considered to be better than in adults.

Due to heterogeneity in methodology, such as case definition, length of follow-up and assessment of recovery, the estimated prognosis of ME/CFS is variable and uncertain. According to ME expert Charles Shepherd, “there is a wide spectrum of severity, variability and degrees of improvement and deterioration in ME/CFS – and so prognosis can very difficult to predict with any certainty in an individual case."

Follow-up studies
In one of the first extensive studies on the prognosis of ME/CFS, Peterson et al. studied 65 patients who fulfilled the Holmes criteria after a one-year follow up. Though 40% reported improvement, none of the patients considered themselves recovered. Similar results were reported by an Australian team in 1994. They followed up on 103 patients for a mean of 3.2 years. Although the majority reported improvement, less than 6% had completely recovered. Two years later, a Dutch research team from Nijmegen published results for their large sample of 246 ME/CFS patients. After a 1.5 year follow-up only 3% said they had recovered and only 17% reported improvement. One of the first reports on prognosis using the Fukuda criteria was published by Saltzstein et al. Because they studied ME/CFS patients in primary care, recovery rates were generally higher, around 20%. In 2017, Collin & Crawley studied the prognosis of 418 ME/CFS patients who had been treated in a British ME/CFS centre and diagnosed according to the NICE criteria. Two and a half years later less than 6% reported they no longer suffered from ME/CFS and 30% claimed their health had deteriorated. All of the aforementioned studies used different diagnostic criteria – the Holmes, Australian, Oxford, Fukuda and NICE criteria respectively – but reported similar poor outcomes, with only a tiny minority of ME/CFS reaching full recovery.

Community-based studies
The poor prognosis observed in tertiary and primary care was confirmed by the epidemiological study in Chicago. Of the 32 ME/CFS patients identified in the late 1990s, 24 were evaluated after a period of approximately 10 years. Of these, two thirds continued to have ME/CFS, 21% had developed exclusionary illnesses, 8% was now labeled as suffering of idiopathic chronic fatigue and only 4% had remitted.

Work-related outcomes and objective measures
Although research has generally used self-reported improvement to estimate prognosis, some studies included more objective measurements such as work-related outcomes. These have generally reported less improvement. Vercoulen et al. and Hill et al. for example found no significant change in employment status. Another study from the research team of Benjamin Natelson found that unemployment rates increased after 3.5 year follow-up. A Danish study followed up on 33 patients for an extensive period of 5 years. Though work disability was already high at baseline (77%), this increased even further to 91%. A trend towards a decrease in employment status was also reported by Collin & Crawley, as more patients reported a deterioration than an improvement in hours spent on work, education and social activities (see table below).

Other studies have shown that ME/CFS patients do experience improvement on objective measures. Tiersky et al. (2002) for example noted better results on neurocognitive testing after a period of 3.5 years. Rowe et al. reported a significant improvement in range of motion (ROM) in 53 adolescents with ME/CFS after a two year follow-up.

The CDC's optimistic prognosis
Despite heterogeneity in diagnostic criteria and patient sample, most research is in agreement on the poor prognosis in ME/CFS (see table below). Research by the Centers for Disease Control and Prevention (CDC) however has controversially reported much higher recovery rates. In 1999, for example, the CDC surveillance system reported a “cumulative probability of recovery” from ME/CFS of 31.4% during the first 5 years of illness and 48.1% during the first 10 years of illness. Data published in 2003 from the epidemiological research in Wichita, Kansas indicated even higher recovery rates. At the three year follow-up almost 80% no longer fulfilled the diagnostic criteria for ME/CFS, though a large minority (23.1%) were diagnosed with other diseases such as sleep disorders. More than half of the studied ME/CFS patients had experienced partial to total remission during follow-up.

However, it should be noted that ME/CFS recovery definitions are inherently based on opinions rather than scientific measures, and influenced by factors like the diagnostic criteria used, etiological perceptions of ME/CFS, age and lifestyle status of patients, and misdiagnosis. The CDC information webpage has been criticized by patient organizations for painting the prognosis of ME/CFS in a positive light and overemphasizing natural recovery of the disease. ME Action, for example, wrote:"“While presenting a more positive outlook may decrease physician discomfort in dealing with a chronic and debilitating disease like ME, it is disingenuous to imply that patients will return to good health when so few do. Moreover, this can delay patients from getting the social support they need to cope with a lifelong illness.”"

Outbreaks
Longitudinal studies on ME outbreaks show that in its epidemic form, the disease tends to have a better prognosis. This might be because less strict diagnostic criteria were used in these studies or because patients were followed-up much earlier in their disease process (bypassing the normal six month waiting period). Levine et al. provided follow-up data from the outbreaks in the Nevada-California region (Incline Village, Truckee, Yerington and Placerville) during the 1980s. Twenty-nine patients were revaluated after a period of three years. Almost all subjects were able to return to pre-illness activity and the authors concluded that the overall prognosis was generally favorable."“Among the patients we observed were a marathon runner, a marathon bicyclist, avid skiers, mountain climbers, and other individuals pursuing an unusually physically active life-style prior to their acute illness. All of these patients attempted to continue their activities and most of them, after periods ranging from 6 to 18 months, eventually were able to resume their previous activities.”"Another follow-up study however, concluded that only an estimated 41% of patients during these outbreaks would fulfill the Holmes criteria of ME/CFS. The recovery rate for these ME/CFS patients was only 15%, much lower than the 30% for all patients of the Lake Tahoe outbreak.

A 10 year follow-up of the outbreak at West Otago, New Zealand also reported a good prognosis. Though some needed to adapt their lifestyle to prevent relapses, most patients returned to premorbid activity.

Follow-up data is also available for the outbreak in Akureyri, Iceland. Sigurdsson & Gudmundsson reported in 1955 that 13% of the patients he evaluated considered themselves completely recovered. In 1988, Byron Hyde reported about 10 Icelandic ME patients in the Lancet, 40 years after the initial outbreak. Two of them (20%) indicated a “total physical and intellectual recovery”.

Pediatric
Prognosis for adolescents with ME/CFS is generally considered to be more favorable than in adults. Gill et al. reported that 4.5 years after an initial evaluation almost a quarter of adolescent ME/CFS patients reached “near to complete improvement”. The diagnosis in this study however was made retrospectively (by looking at old charts and test results instead of a clinician’s assessment). A Dutch team followed up on 54 adolescents with ME/CFS. After an average of 2.2 years almost half of the sample said they've had almost completely recovered. David Bell followed up on 35 ME/CFS patients who fell ill during the Lyndonville outbreak in the 1980s. Prognosis was favorable: 37% said they had "recovered completely" and felt "entirely well.” A further 42.9% agreed with the statement “I have never recovered completely but feel pretty well." Other studies (Rangel et al., Sulheim et al.,, Rimes et al. ) have reported similar high improvement/recovery rates in pediatric ME/CFS, though these studies did not use generally accepted diagnostic criteria of ME/CFS. In the Bulletin of the IACFS/ME Katherine Rowe claimed to have data for 256 young ME/CFS patients 12 years after initial evaluation. Eighty-eight percent reported recovery, though this information has not been reported in a peer-review publication. In its 2015 report (p.183, 213), the National Academy of Medicine noted a lack of well-designed longitudinal studies on the prognosis of ME/CFS in children and adolescents.

Dr. Bell notes that if an adolescent spent three months in bed due to ME/CFS they will still be ill at age 35 even if their symptoms were mild in their adult years. Becoming increasingly ill with activity and symptom severity is expected 15-20 years later. "One study of young adults followed for fifteen years demonstrated clear improvement in activity, but not illness resolution8. The same group of patients continued to do relatively well for a further five to ten years and then became worse in both activity limitation and symptom severity12. It is rare for an adolescent to become completely free of the disease13."Children are diagnosed with ME/CFS at three months of illness under the Systemic Exertion Intolerance Disease (SEID) and Canadian Consensus Criteria (CCC) and myalgic encephalomyelitis (ME) is diagnosed immediately under the International Consensus Criteria (ICC).

Predictors of outcome
Several studies have looked at factors that influence the prognosis of ME/CFS, though results have generally been conflicting.

Age
Although some studies have found the age of adult ME/CFS patients to be associated with poorer outcomes (with older people having a worse prognosis), most studies did not find this correlation (see table below).

Matthews & Komaroff were able to follow-up on 92 ME/CFS patients for a long period of nine years. The authors noted that “physical function tended to improve for many patients over time, despite the fact that they were aging.” This suggests age is not a predictor of poor outcomes in adult ME/CFS patients. Kidd et al. segmented ME/CFS patients into two groups: those younger and those older than 55. A comparison showed few meaningful differences. But when limited to those patients who had been sick for a long time, younger patients did have more autonomic and immune-related symptoms.

Illness duration
Although many physicians believe that patients would benefit from earlier diagnosis and treatment, evidence for this is currently lacking. Studies have generally been inconsistent regarding the duration of illness as a predictor of outcome. (see table below).

The CDC information website has been criticized for suggesting early intervention leads to better outcomes. In a detailed critique Jaime Seltzer from #MEAction wrote: "“Anyone who has gone undiagnosed for years or even decades understands that an accurate diagnosis and useful management strategies — such as avoiding overactivity and providing symptomatic relief — can improve quality of life. But there is no evidence that early intervention affects the likelihood of full recovery. When the CDC has been so cautious regarding what it recommends as treatment, this assertion, which is insupportable with current evidence, stands out.”"Friedberg et al. compared ME/CFS patients with a median length of illness of 18 years compared to a group of ME/CFS-patients who had been ill for less than three years. Few differences existed between the two groups; although those with long illness duration had more cognitive difficulties (they also reported more sexual problems). In a study by [Santamarina-Perez et al. (2011), no association was found between illness duration and neuropsychological test results. Hornig et al. did find a distinct an increase in cytokine expression in the plasma of ME/CFS patients with illness duration less than three years, an increase that was not present in patients with longer illness duration.

Illness severity
Many studies have found that illness severity at baseline is associated poorer outcomes over time. Disagreement exists however on which measures of illness best predict a worse prognosis. While Pheley et al. (1999) found an association with fatigue, Ray et al. (1997) reported the broader concept of somatic symptoms to be associated with poor outcomes. In a Norwegian study that followed ME/CFS patients who got sick after mononucleosis, joint pain was an important predictor of prognosis.

Psychiatric comorbidity
Though some studies have reported it, no convincing evidence exists that a comorbid psychiatric diagnosis worsens the prognosis of ME/CFS. One well-controlled study (Tiersky et al. 2002) even found an inverse relationship; patients with a concurrent psychiatric history showed more improvement over time. The authors attributed this to varied presentations of symptoms in patients meeting the Fukuda criteria, which allows some patients with an underlying psychiatric etiology to be diagnosed with ME/CFS.

Infectious onset
A Japanese study indicated that the prognosis of postinfectious ME/CFS patients is better than in ME/CFS patients without an infectious onset. Though the sample size was small, results were in line with other studies. A Norwegian team for example followed up on ME/CFS patients who had developed ME/CFS after infectious mononucleosis. After 6.5 years a large increase in employment status was seen.

Sudden/gradual onset
DeLuca et al. (1997) compared patients with a sudden and gradual onset and found relatively few differences, except that the gradual onset group had more psychiatric comorbidity. Two studies (Reyes et al. 1999 ; Hill et al. 1999 ) found that the type of illness onset was not a significant predictor of prognosis in ME/CFS.

Comorbid fibromyalgia
Ciccone et al. reported that ME/CFS patients with comorbid fibromyalgia have a worse prognosis than ME/CFS patients without fibromyalgia. Other studies had already shown that patients with both diseases are more physically impaired than patients who only have ME/CFS.

Being diagnosed with CFS
In the community-study in Wichita, Kansas, researchers were able to compare the prognosis of ME/CFS patients who were diagnosed versus those than weren’t. Gijs Bleijenberg summarized the results as follows: "“Remission was also not associated with the report of being ever diagnosed as CFS. Although the number of CFS diagnosed subjects was very low, it may mean that the diagnosis of CFS cannot be seen as unfavourable for remission. This is contrary the idea of some family doctors who are reluctant to diagnose their patients as having CFS, as they are afraid that this will facilitate an unfavourable course of the condition.'"Bell et al. (2001) were able to dispel another myth, namely the idea that ME/CFS symptoms persist because of encouragement by sympathetic physicians who might reinforce illness beliefs. In their 13-year follow-up study, the longest to date in ME/CFS, clinical empathy was not a prognostic factor.

ME worse prognosis than CFS?
Hamilton et al. (2005) were able to probe the data of 18,000 patients who had been diagnosed by primary care physicians with either CFS, ME, fibromyalgia or postviral fatigue syndrome (PVFS). To estimate prognosis they looked at the percentage of patients who retained the label of any of these "fatiguing illnesses" at follow-up. Patients with ME were found to have a worse prognosis than patients with CFS, fibromyalgia or PVFS.

Sex
Most studies have not found gender to be a significant predictor of prognosis (more references needed). Matthews & Komaroff (2007) did find women to have consistent higher physical function than men.

‘Recovered’ but with lingering symptoms
In one of the most influential studies on the prognosis in ME/CFS, David Bell followed-up on the adolescents who fell ill during the Lyndonville outbreak in the 1980s. After a period of 13 years, more than a third of the sample considered themselves recovered. Yet when researchers looked at the symptoms these ‘recovered’ patients reported, they seemed to be much more disabled than healthy controls. As Bell explains: "“Some of the young adults rated their health as ‘good’, while the amount of activity they could perform was minimal. They had become so used to their performance level, they accepted it as their new normal. Unfortunately, many of them, while they described their health as good, they were unable to work full time or carry on other duties.”"This suggested that ‘recovered’ ME/CFS patients did not reach their premorbid health, but have simply adjusted to ongoing somatic symptoms. Similar findings were reported by Reyes et al., who noted that “all patients, including those who reported recovery, stated that they continued to have some CFS-defining symptoms”. Sankey et al. noted that the improvement patients reported “contrasts greatly with the numbers of symptoms that they were still experiencing when specifically asked. […] although 21 (75%) regarded themselves as fully recovered, they described symptom persistence of fatigue and headache, depression and sleep disturbance.”

Another aspect that might have led to an overestimation of recovery in ME/CFS is the fluctuating nature of the disease. Anderson et al. (2004) for example noted: "“Almost all our patients reported symptom fluctuation, including daily, weekly or monthly periods. This, together with our other findings, suggests that symptoms often improve or worsen without altering the overall severity of the illness.”"This is a confirmed by a study where ME/CFS symptoms were recovered at three time points in time. Hill et al. found that among patients who reported improvement at time two, about half were worse again at time three.

Finally Adamowicz et al. (2014) noted that ME/CFS studies often use a limited assessment of recovery, for example by simple asking patients if they recovered or by looking at only one measure to assess improvement. This might also have led to an overestimation of recovery in ME/CFS, as patients can improve on one measure and worsen on another. This was seen in a Norwegian study that used multiple measures of recovery. While the total sleep and rest time decreased, work disability increased. While attention during reading worsened, difficulty in driving a car improved. Adamowicz et al. pleaded for a multidimensional assessment of recovery that captures a broad-based return to health. Twisk argued that objective measures such as a cardiopulmonary exercise test, a tilt table test or neurocognitive tests, should be used to assess recovery. Qualitative studies with physician experts and patients with ME/CFS show support for a multidimensional assessment of recovery, which would incorporate measurements of daily functioning, symptomatology, quality of life, and physical functioning.

Disease onset
The disease ME/CFS often occurs when a person does not recover from a flu-like illness, EBV, mononucleosis, HHV, Q fever, another virus or other infections. Patients must meet one or more diagnostic criteria that define either myalgic encephalomyelitis (ME), chronic fatigue syndrome (CFS) or ME/CFS. "Occasionally, ME/CFS has been triggered by environmental toxins, the receipt of an immunizing injection, or surviving a major physical trauma."

Fluctuation
For most ME/CFS patients, their health is never the same and they experience numerous symptoms and disease severity which fluctuate day to day and though the years. In multiple sclerosis the clinical course of the disease can divided into different subtypes as some patients experience a progressive decline, while others a more relapsing course with ups and downs. Stroothoff et al. tried to make a similar subdivision in ME/CFS. Their analysis showed that the majority of ME/CFS patients describe their illness as fluctuating, with a minority (15.9%) said they were constantly getting worse. Howes & Goudsmit had previously reported on a subtype of ME that progressively worsens over time.

Fennells 4 stages
Others have subdivided the course of ME/CFS into different stages. In the 1990s, Patricia Fennell for example suggested four distinct stages. The first phase is about the process of getting sick, giving up hobbies or work and looking for an explanation of symptoms. In the second phase, the diagnosis has provided a kind of relief, while the disease usually stabilizes. In the third phase, patients understand that they will not recover quickly and that their disease is chronic. At this stage patients go through a mourning process as they realize that their old self is gone and they have to adapt to a new identity. Although the illness is stable, they still experience relapses. In the fourth phase, the patients reintegrate with their new identity; they set new goals and value a new way of living. Some experience improvement. Jason et al. tested this model in 65 ME/CFS patients. Though analysis showed a three-factor solution that corresponded with Fennell's stages, almost none of the patients were in the fourth stage. This was probably because participants were selected in a clinic and were still actively seeking health care. The analysis also showed that patients experiencing the first phase, the crisis phase, were sick for a longer time than patients in other stages. This suggests Fennell's subdivisions should not be seen as a linear development in time, but rather the progression patients make in dealing with illness and disability.

Bell’s 3 stages
Based on the experience of adolescents he followed up for many years, David Bell subdivides the course of ME/CFS into three stages. At the first stage, there is an acute illness where EBV is likely responsible for most adolescents but there is no standard viral illness. The initial virus likely fades away in a week and the patient feels better and is ready for regular activities. Bell adds "in a person with a gradual onset of symptoms, this stage would be different." The second stage "occurs when the symptoms do not disappear, but persist for months to years in roughly the same severity. There is little variation day-to-day, and this time is very confusing. Many medical providers are contacted, most giving conflicting opinions varying from psychological disorders to possible exotic infections, but the tests come back normal." In the third stage patients will have mild, gradual improvements over years and patients adjust their lives to the symptoms. Patients may look well and recovery is thought to have taken place, but they still have to adapt their lifestyle to avoid relapses and symptom exacerbation.

Learn more

 * Lecture by Dr. David Bell, April 16, 2011
 * Prognosis of ME/CFS – by David S. Bell, MD