Fatigue and Multiple Sclerosis: A Global Look at How Common It Really Is

Table of Contents

• Key Points
• Understanding Multiple Sclerosis and Fatigue
• How the Research Was Conducted
• The Studies Behind the Numbers
• How Fatigue Was Measured
• Quality of the Included Studies
• The Big Picture: Global Fatigue Prevalence
• Who Is Most Affected by MS Fatigue?
• Why Do the Numbers Differ So Much?
• What This Means for Patients and Doctors
• Study Limitations
• Recommendations
• Frequently Asked Questions
• Source Information

Key Points

• About 59% of people with MS have clinically significant fatigue, based on 69 studies with 44,468 participants.
• The choice of fatigue scale and cut-off explains 46.4% of variation; with disability, age, MS duration, it explains 86.4%.
• Fatigue prevalence has declined each decade since 2000: 64.4%, 59.6%, and 51% in 2020–2023.
• Secondary progressive MS has the highest fatigue (74.4%), followed by primary progressive (64.3%) and relapsing-remitting (54.7%).
• No pharmacological treatments reliably relieve MS fatigue; physical activity, diet changes, and CBT may help.

Understanding Multiple Sclerosis and Fatigue

Multiple sclerosis (MS) is a chronic, inflammatory disease that damages the myelin sheath—the protective coating around nerve fibers—in the central nervous system (the brain and spinal cord). It is one of the most common causes of non-traumatic disability among young adults. The global burden of MS has been growing: an estimated 2.8 million people worldwide now live with MS, according to the latest Atlas of MS data. The highest rates are found in the WHO European Region (EUR) and the Region of the Americas (AMR), while the lowest rates are in the African Region (AFR) and the Western Pacific Region (WPR).

MS is categorized into three main disease phenotypes, based on disease activity, progression, and clinical course:

• Relapsing-remitting MS (RRMS): Approximately 85% of people with MS are initially diagnosed with this form, characterized by flare-ups followed by periods of recovery.
• Secondary progressive MS (SPMS): Many people with RRMS eventually develop this form, where disability steadily worsens with or without superimposed relapses.
• Primary progressive MS (PPMS): A less common form with continuous worsening from the start.

People with MS experience a wide range of symptoms that vary in severity, including spasticity (muscle stiffness), pain, fatigue, bladder and bowel problems, difficulty walking (gait impairments), mood disturbances, and sleep disorders.

Among all of these, fatigue is one of the most common and burdensome symptoms of MS. It can appear at any stage of the disease. The definition used by researchers is: "a significant lack of physical and/or mental energy, perceived by the individual or caregiver, that interferes with normal and desired activities."

MS-related fatigue is fundamentally different from ordinary tiredness. Healthy people recover from fatigue with rest or sleep, but MS-related fatigue is disabling and often cannot be relieved that way. Patients describe having to put an unduly high level of effort into daily tasks, which reduces their ability to perform everyday activities and to work.

Researchers divide MS-related fatigue into two categories:

• Primary fatigue: Considered specific to MS itself. It occurs without an obvious trigger and is a direct consequence of the underlying disease process.
• Secondary fatigue: Caused by other factors, including sleep disturbances, mood disorders (anxiety and depression), side effects of disease-modifying treatments (DMTs), and decreased physical activity.

The consequences are serious. Fatigue is one of the main drivers of low health-related quality of life and unemployment in people with MS. It is also a major predictor of claims for social benefits, including sick leave and disability pensions. That makes fatigue one of the most urgent clinical problems in the treatment and management of MS.

Currently, there are no convincing pharmacological (drug) treatments proven to reliably relieve MS-related fatigue. Fortunately, nonpharmacological approaches—including physical activity, dietary modification, and cognitive behavioral therapy (CBT)—can be beneficial for many patients.

This study was motivated by a real problem: previous research on fatigue prevalence has produced wildly inconsistent numbers. Individual studies have reported fatigue in anywhere from 28.4% to 88.2% of people with MS. An earlier systematic literature review covering 12 studies found prevalence rates ranging from 36.5% to 78%. This wide variation probably reflects differences in measurement tools (more than a dozen fatigue questionnaires exist) and differences in patient characteristics. The researchers conducted this systematic review and meta-analysis to establish a reliable global figure and identify exactly what causes the numbers to vary so much.

How the Research Was Conducted

The study was carefully designed according to international scientific standards. The research protocol was registered in the PROSPERO international prospective register of systematic reviews (registration number CRD42024499139), and the study was reported following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines—the globally accepted standard for this type of research.

The research team conducted a comprehensive computerized search of eight databases: PubMed, EMBASE, Cochrane Library, Web of Science, PsycINFO, CINAHL, and two Chinese databases (China National Knowledge Infrastructure [CNKI] and Wanfang). The search was completed on January 31, 2024, covering English- and Chinese-language studies published since 2000.

To be included in the analysis, studies had to meet all of the following criteria:

1. Be an observational study (cross-sectional, cohort, or case-control design)
2. Include people aged 17 years or older with an MS diagnosis, confirmed either by self-report or by a clinician using the Poser or McDonald diagnostic criteria
3. Be published as full text in a peer-reviewed journal
4. Use a validated fatigue measurement scale with a defined cut-off value indicating clinically significant fatigue

Studies were excluded if they involved only patients with clinically isolated syndrome (a single neurological episode that doesn't yet meet MS criteria) or only hospitalized inpatients; if fatigue was itself a requirement for entering the study; if the sample size was smaller than 100; if there wasn't enough information to calculate prevalence; or if they contained duplicated data.

The study selection process was rigorous. After removing duplicate records, two investigators independently screened all titles and abstracts. Whenever at least one reviewer thought an abstract might qualify, the full article was retrieved and evaluated by both reviewers. Disagreements were settled through consensus, or by a designated senior author if consensus couldn't be reached.

Data extraction was also done twice—independently by two pairs of reviewers—using standardized spreadsheets that were pilot-tested on 15 randomly selected articles. The extracted data covered study characteristics (design, sample size, country, diagnostic criteria), patient characteristics (age, sex, MS duration, disability scores, education level), and outcomes (fatigue cases, fatigue scale, and cut-off values). For longitudinal studies, baseline data were used.

The researchers assessed study quality with two established tools: the Agency for Healthcare Research and Quality (AHRQ) criteria for cross-sectional studies (scoring 11 items, with total scores of 0–3 labeled "low quality," 4–7 "moderate quality," and 8–11 "high quality"), and the Newcastle-Ottawa Scale (NOS) for case-control and cohort studies (scoring 0–9 stars, with 7–9 stars indicating low risk of bias, 5–6 moderate, and fewer than 4 high).

For the statistical analysis, the researchers used a random-effects model—the appropriate method when studies are expected to vary from each other—and applied a Freeman-Tukey transformation to correctly handle the non-normal distribution of prevalence data. Heterogeneity (how much results differ between studies) was measured with the χ²-based Cochrane Q statistic and the I² test. An I² value of 75% or higher indicates considerable heterogeneity. To find out what was driving the differences, the team ran subgroup analyses and meta-regression analyses on potential explanatory factors such as sample size, survey year, patient source, age, sex distribution, MS duration, disability score, fatigue scale, WHO region, and country income level. A minimum of three studies was required per subgroup. Statistical significance was set at p < 0.05.

The study did not formally evaluate publication bias, because established methods for this (such as funnel plot asymmetry tests) are known to be unreliable for prevalence studies. Instead, the researchers performed a sensitivity analysis—removing studies one by one to see whether any single study was skewing the results.

The Studies Behind the Numbers

The database search initially produced 7,258 records. After removing duplicates, 4,838 titles and abstracts were screened, and 446 full-text articles were assessed against the eligibility criteria. In the end, 69 studies met all criteria and were included in the final analysis.

Together, these 69 studies included 44,468 people with MS. The average study had 644 participants, but sample sizes varied enormously, ranging from 100 to 9,077 participants. Across studies, the mean age of participants ranged from 32.4 to 59.3 years. The mean Expanded Disability Status Scale (EDSS) score—a standard measure of MS-related disability that runs from 0 (no disability) to 10 (death from MS)—ranged from 1.9 to 6.5. The mean MS duration ranged from 4.1 to 22.2 years, and the proportion of female participants ranged from 55% to 86%.

Geographically, the studies spanned 27 countries across four WHO regions:

• Over half of the studies (n = 36) were conducted in Europe (EUR)
• There were no studies from Africa (AFR) or the Southeast Asia Region (SEAR)
• The United States contributed the most studies (n = 9), followed by Argentina (n = 5), and then the UK, Italy, and Australia (4 each)
• Saudi Arabia, the Netherlands, China, and Finland each contributed 3 studies
• Two studies were binational (one in the USA and Sweden, another in Turkey and Israel), and three studies were multinational

By income level, the vast majority of studies (n = 53, 77%) were performed in high-income countries, with the remainder in upper-middle-income countries. In terms of patient recruitment, almost half of the studies (n = 34) were population-based, 24 took place in MS outpatient clinics, and 4 were conducted in MS Medical Research Centres. Most studies (n = 64, 93%) used a cross-sectional design (collecting data at a single point in time), 4 used a case-control design, and 1 used a cohort design (following patients over time).

Among the studies that reported specific patient characteristics: 15 provided data on sex distribution, 15 reported fatigue prevalence broken down by MS phenotype, and 4 estimated prevalence by education level. When diagnostic criteria were reported (34 studies), two earlier studies used the older Poser criteria, while all later studies used the continuously updated McDonald criteria (versions 2001, 2005, 2010, and 2017).

How Fatigue Was Measured

The included studies used four different validated fatigue scales, and even within a single scale, different cut-off values were used to define "clinically significant" fatigue. This variation matters because it directly affects how many patients get counted as fatigued.

The most commonly used tool was the Fatigue Severity Scale (FSS), used in 47 studies (two-thirds of the total). But even here, cut-off values varied:

• 28 studies used a cut-off of 4 (mean score)
• 17 studies used a cut-off of 5 (mean score)
• 2 studies used cut-offs of 4.5 (mean score) or 28 (total score)

The Modified Fatigue Impact Scale (MFIS) was used in 16 studies (about one-fifth): 14 of these used a total score of 38 as the cut-off, while the other 2 used totals of 35.5 and 45. Four studies used the Fatigue Scale for Motor and Cognitive Functions (FSMC) with a total-score cut-off of 43. One study used the EMIF-SEP, a validated French version of the Fatigue Impact Scale, with a total-score cut-off of 55. One additional study used both the FSS and MFIS, with cut-offs of 4 (mean score) and 38 (total score), respectively.

This patchwork of tools and thresholds is not just a technical detail—it turned out to be the single most important factor explaining why different studies report such different prevalence rates.

Quality of the Included Studies

The quality of the evidence matters for how much we can trust the results. In the 4 case-control studies and the single cohort study, Newcastle-Ottawa Scale scores ranged from 4 to 6 stars, indicating moderate risk of bias on average.

For the 64 cross-sectional studies, the mean AHRQ quality score was 7.6 out of 11, with scores ranging from 5 to 10. Of these, 59% were rated "moderate quality", and 28 studies (41%) were rated "high quality." The main limitations flagged were: inadequate detail about quality-control methods for outcome measures, the use of non-blinded evaluators (who knew what the study was looking for), and unclear methods for handling missing data in the statistical analyses.

Reassuringly, when the researchers performed a sensitivity analysis—removing studies one at a time to see if any single study distorted the overall estimate—the results remained largely unchanged. This indicates that the findings were robust and reliable despite the varying quality of the underlying studies.

The Big Picture: Global Fatigue Prevalence

Across all 69 studies, the reported prevalence of fatigue ranged from a low of 28.4% to a high of 88.2%. When all the data were pooled together, the global prevalence of MS-related fatigue was 59.1% (95% confidence interval: 55.9% to 62.2%).

In plain language, this means that nearly 6 out of every 10 people with MS experience clinically significant fatigue. The 95% confidence interval indicates that if the analysis were repeated many times, the true global figure would fall between 55.9% and 62.2% in 95 out of 100 repetitions—giving us high confidence that the true number sits close to 59%.

There was significant heterogeneity among the studies (I² = 97.3%, p < 0.01), meaning that 97.3% of the variation between studies was due to real differences (rather than random chance), and the probability that these differences occurred by chance alone is less than 1%.

The five countries with the highest prevalence of MS-related fatigue were:

1. Austria: 80%
2. Norway: 79.5%
3. United Kingdom: 69.8%
4. Switzerland: 69.2%
5. Lithuania: 68.6%

The researchers also pooled results by demographic group. In the 15 studies that reported data by sex, fatigue prevalence was slightly higher in women (58% of 6,982 females) than in men (56.5% of 2,598 males). By MS phenotype, secondary progressive MS (SPMS) had the highest prevalence at 74.4% (10 studies), followed by primary progressive MS (PPMS) at 64.3% (8 studies), and relapsing-remitting MS (RRMS) at 54.7% (15 studies). Education also mattered: among the 4 studies reporting this, people with more than 12 years of education had a fatigue prevalence of 47.9%, compared with 64.3% among those with 12 or fewer years of education.

Who Is Most Affected by MS Fatigue?

To explore why prevalence rates varied so much between studies, the researchers regrouped the studies by study- and patient-level characteristics. The subgroup analysis produced clear patterns:

• Sample size: Studies with more than 1,000 participants reported higher fatigue prevalence (67%) than smaller studies (59.9%).
• Survey year: Fatigue prevalence has decreased every decade since 2000: 64.4% in 2000–2009, 59.6% in 2010–2019, and 51% in 2020–2023.
• Sex distribution: Prevalence rose with the proportion of female participants: 49.9% in studies with fewer than 60% women, 58.5% in studies with 60–80% women, and 64.6% in studies with 80% or more women.
• Age: Fatigue became more common with age—54.7% in studies with mean ages of 30–40 years, 58.6% in those aged 40–50, and 69.3% in those aged 50–60.
• Disability (EDSS score): People with greater disability (EDSS score greater than 4) had much higher fatigue prevalence (73.7%) than those with milder disability (EDSS ≤ 4).
• MS duration: Patients who had MS for more than 10 years had higher prevalence (63.4%) than those with a shorter disease duration (55.6%).
• Fatigue scale used: The FSMC produced the highest estimate (70.4%), while the MFIS with a cut-off of 38 produced the lowest (51%). Within the most common scale (FSS), a cut-off of 4 yielded a prevalence of 65.6%, versus 54.3% with a cut-off of 5.
• Patient source: Population-based studies reported 61.4%, outpatient studies 58.9%, MS Medical Research Centres 50.2%, and mixed populations 54.9%.
• WHO region: Europe had the highest estimate (61.2%), while the Western Pacific had the lowest (54.2%).
• Country income level: High-income countries reported significantly higher prevalence (59.9%) than upper-middle-income countries (53.4%).

Of the 10 characteristics examined, all but WHO region and patient source had a statistically significant impact on reported fatigue prevalence—meaning these factors are genuine sources of the differences seen between studies.

Why Do the Numbers Differ So Much?

Subgroup analyses help, but large variation still remained within each subgroup. To dig deeper, the researchers performed a meta-regression analysis that simultaneously evaluated multiple study and patient characteristics to see how much each factor explains the differences between studies.

The results were striking. Four factors were significantly correlated with the variation between studies, and together they explained 86.4% of the total between-study variance:

• Fatigue scale used: 46.4%—this was by far the largest source of heterogeneity
• EDSS score (disability level): 18.4%
• Age: 14.6%
• MS duration: 7.1%

In other words, the single most important reason that one study finds 30% fatigue while another finds 80% is simply the choice of fatigue questionnaire and its cut-off value. This is a critical insight: it

Frequently Asked Questions

What causes fatigue in multiple sclerosis?

Researchers divide MS-related fatigue into primary fatigue, caused by the disease itself, and secondary fatigue, caused by factors like sleep issues, mood disorders, medication side effects, and reduced physical activity. Primary fatigue occurs without an obvious trigger and stems directly from the underlying MS process.

Why do different studies report very different fatigue rates in MS?

The main reason is the choice of fatigue measurement scale and its cut-off value, which alone explained 46.4% of the variation between studies. Other factors include patient disability level, age, and MS duration. Together, these four factors accounted for 86.4% of the differences.

How was fatigue measured in the studies included in this meta-analysis?

The studies used four validated fatigue scales: the Fatigue Severity Scale (FSS), Modified Fatigue Impact Scale (MFIS), Fatigue Scale for Motor and Cognitive Functions (FSMC), and EMIF-SEP. Even within the same scale, different cut-off values were used, which affected prevalence estimates.

When should a person with multiple sclerosis and severe fatigue seek a second opinion?

Fatigue affects nearly 6 out of 10 people with MS worldwide. If your fatigue is disabling or unresponsive to physical activity, dietary changes, or cognitive behavioral therapy, a second opinion can help. Fatigue may be primary, from MS itself, or secondary to sleep issues, mood disorders, or medication side effects—so a thorough evaluation matters. There is no convincing drug treatment for MS fatigue, but nonpharmacological approaches can help. A specialist review can confirm your fatigue is properly measured and managed. Diagnostic Detectives Network provides independent expert second opinions.