Multiple sclerosis in women: the sex ratio, pregnancy relapses, and prevalence gap

She was 29, two years into a relapsing-remitting diagnosis, and planning a first pregnancy when she asked the question every woman with multiple sclerosis eventually brings into clinic: will this get worse when I am pregnant, and how much worse after delivery? One neurologist had said pregnancy is protective. Another had warned about a postpartum "crash." Both were drawing from real data. Neither had put the numbers next to each other.

Multiple sclerosis is an immune-mediated disease of the central nervous system. It is not rare in women of reproductive age, and it is not evenly shared between the sexes. The female excess is large and measurable. In several high-quality population series it is still widening. That pattern sits inside a broader reality we covered in our review of autoimmune disease in women: most autoimmune conditions hit women harder, but MS is one of the few where incidence and clinical course differ by sex in opposite directions.

This article pulls the figures a reporter or clinician can cite with a primary source attached. You will find the female-to-male ratio and whether it is rising, age at onset, pregnancy effects on relapse rate, disability progression by sex, and U.S. and global prevalence estimates.

Period	Relapses per woman-year
Year before pregnancy	0.7
First trimester	0.5
Second trimester	0.6
Third trimester	0.2
First 3 months postpartum	1.2

Source: Confavreux et al., Pregnancy in Multiple Sclerosis (PRIMS) Group, New England Journal of Medicine, 1998 (254 women, 269 pregnancies).

Group	Prevalence per 100,000
Women	450.1
Men	159.7
Overall (higher 10-year estimate)	309.2

Source: Wallin et al., Neurology, 2019. Higher 10-year cumulated 2010 estimate; female-to-male ratio 2.8.

Group	Prevalence per 100,000
White	374.8
Black	298.4
Other non-Hispanic	197.7
Hispanic (any race)	161.2

Source: Hittle et al., JAMA Neurology, 2023. Overall female-to-male ratio 2.9 in the same analysis.

The sex ratio, and whether it is rising

Start with the present. In the National Multiple Sclerosis Society-supported U.S. prevalence study published in Neurology in 2019, Wallin and colleagues applied a validated algorithm to private and public insurance claims (plus military) and estimated 2010 adult prevalence at 450.1 per 100,000 in women and 159.7 per 100,000 in men, a female-to-male prevalence ratio of 2.8. A later analysis of the same claims infrastructure by Hittle and colleagues in JAMA Neurology (2023) identified 744,781 adults with MS: 76% female, 24% male, overall ratio 2.9.

Those are prevalence ratios (stock, not flow). Incidence ratios tell a sharper story about whether the gap is still moving. Orton and colleagues, writing in Lancet Neurology in 2006, used a Canadian dataset of 27,074 patients and calculated the female-to-male ratio by year of birth. The ratio had been increasing for at least 50 years and exceeded 3.2:1 among more recent birth years (p<0.0001; rank correlation r=0.84). The authors concluded the rise reflected a disproportionate increase in incidence among women, not earlier diagnosis in women alone.

A systematic review by Alonso and Hernán in Neurology in 2008 estimated that the female-to-male incidence ratio rose from 1.4 in 1955 to 2.3 in 2000, with overall incidence of 3.6 cases per 100,000 person-years in women and 2.0 in men. The pattern is not universal. Some Swedish analyses have not shown a clear secular rise. The best multi-decade series mostly point one way.

Denmark is the cleanest long-running national registry. Koch-Henriksen and colleagues reported in Neurology in 2018 that from 1950-1959 to 2000-2009, female incidence more than doubled (5.91 to 12.33 per 100,000 per year) while male incidence rose only 24% (4.52 to 6.08). Magyari's 2016 Danish Medical Journal thesis, citing Bentzen et al. (2010), puts the Danish ratio at 1.3:1 in 1950, 1.5:1 in 1977, and 2.02:1 in 1990. Holm and colleagues, updating the same registry through 2023 in Brain, found females were 58.7% of the Danish MS population in 1975, 65.7% in 2000, and 68.5% in 2023, with a female/male prevalence ratio of 2.2 in 2023.

So is the ratio still rising? In high-quality Northern European and Canadian data, yes, mainly because female incidence climbed while male incidence moved little. In the contemporary United States, the best national prevalence work lands near 2.8-2.9 women per man. That is not the 4:1 figure sometimes repeated in patient materials, and it is not a global constant. It is still a large, real female excess.

How common MS is in women

Absolute counts matter as much as ratios. Wallin et al. put the higher 2010 U.S. adult prevalence estimate at 309.2 per 100,000, or 727,344 people. Extrapolating observed growth in two of their datasets, they estimated 2017 prevalence between 851,749 and 913,925 adults. That is the figure behind the widely quoted "nearly 1 million Americans" line. Survival gains and multi-payer case-finding both pushed the total above older national estimates.

Hittle et al. (2023) added race and geography. White adults had the highest 2010 prevalence at 374.8 per 100,000, followed by Black adults at 298.4, other non-Hispanic groups at 197.7, and Hispanic adults at 161.2. Peak prevalence sat in the 45- to 64-year age band across groups. Prevalence rose with latitude: 11.7 additional cases per 100,000 per degree after direct adjustment. That gradient was strongest in women. The old "MS is a disease of Northern European white women" shorthand is incomplete. Black women carry a substantial U.S. share, and the north-south gradient persists after race adjustment.

Globally, Walton and colleagues summarized the third Atlas of MS in Multiple Sclerosis Journal in 2020: about 2.8 million people living with MS (35.9 per 100,000), pooled incidence 2.1 per 100,000 persons per year across 75 reporting countries, and a mean age at diagnosis of 32 years. Females were twice as likely as males to live with MS in that global compilation. That is a lower ratio than many high-latitude countries report, as expected when low-prevalence regions are averaged in.

Denmark's series runs longer than most. Holm et al. reported overall prevalence rising from 22.0 per 100,000 in 1950 to 295.0 in 2023 (female 402.1, male 186.7). Incidence climbed from about 3.5 per 100,000 until 1975 to 11.4 in 2000, then largely plateaued through 2022. Prevalence can keep rising after incidence stabilizes because people with MS are living longer. These numbers sit next to other female-predominant chronic conditions, including thyroid disorders and migraine in women.

Age at onset

NINDS states that MS symptoms usually begin between ages 20 and 40. The Atlas of MS puts the global mean age at diagnosis at 32 years. Those statements are not identical, because diagnosis often lags first symptoms. Both still place the disease in the years when people build careers and have children.

The age pattern is not static. Koch-Henriksen's Danish analysis found incidence rose most sharply among people with onset at age 50 or older: a 4.30-fold increase in women and a 2.72-fold increase in men from 1950-1959 to 2000-2009. Late-onset MS is still a minority of cases, but the female increase is not confined to classic early adulthood. Holm et al. put the mean age of the living Danish MS population at 54.2 years in 2023, up from a late-1970s plateau near 52.5.

In clinic that means two conversations: the 28-year-old with new sensory symptoms, and the 52-year-old with progressive gait trouble and a delayed workup. The sex ratio remains female-skewed across those ages; Hittle et al. found the highest female-to-male ratios in the 45-54 band (for example, 4.4 among Hispanic adults and 3.7 among Black adults).

Pregnancy and relapse rates

This is the section the patient in the opening vignette needed, with the numbers in order.

The Pregnancy in Multiple Sclerosis (PRIMS) study remains the foundational prospective series. Confavreux and colleagues followed 254 women through 269 pregnancies across 12 European countries (NEJM, 1998). Mean annualized relapse rate was 0.7 in the year before pregnancy, 0.5 in the first trimester, 0.6 in the second, and 0.2 in the third (P<0.001 versus pre-pregnancy for the third trimester). In the first three months after delivery the rate rose to 1.2 (P<0.001), then returned toward baseline. Disability on the Kurtzke scale worsened by 0.7 points over 33 months, without a clear acceleration confined to the postpartum window. Neither breastfeeding nor epidural analgesia showed an adverse effect on relapse or disability in that analysis.

The 2004 PRIMS two-year report by Vukusic and colleagues in Brain refined the risk picture. After the first postpartum trimester, annualized relapse rates did not differ significantly from the pre-pregnancy year. Despite the elevated early postpartum risk, 72% of women had no relapse in those first three months. Predictors of a postpartum relapse included higher pre-pregnancy relapse rate, relapse during pregnancy, and higher disability at pregnancy onset. Those predictors help with counseling. They are not precise enough for individual prediction alone.

Contemporary cohorts look different because disease-modifying therapy use has changed. Langer-Gould and colleagues (Neurology, 2020; Kaiser Permanente California, 466 pregnancies among 375 women, 2008-2016) found annualized relapse rates falling from 0.37 before pregnancy to 0.14-0.07 during pregnancy. In the postpartum year, 26.4% of women relapsed. Exclusive breastfeeding for at least two months was associated with lower postpartum relapse risk (adjusted hazard ratio 0.37). Most women with MS today can have children without an increased overall postpartum relapse burden relative to pre-pregnancy activity when disease is well controlled going into conception. Active disease before pregnancy still marks higher risk.

Pregnancy is usually quiet for inflammatory activity; the first postpartum months are the vulnerable window in classic untreated cohorts. For cycle tracking after delivery, our period calculator and ovulation calculator can help interpret irregular returns of menses while neurology plans therapy resumption. Postpartum mood risk also rises when a chronic illness flares; see our review of postpartum depression statistics.

Disability progression by sex

Here the sex pattern flips. Women are more likely to develop MS. Men who develop it often do worse on disability measures.

Ribbons and colleagues analyzed 15,826 patients from the international MSBase registry and reported in PLoS ONE in 2015 that among 14,453 relapse-onset patients, males progressed faster on the Expanded Disability Status Scale: 0.133 EDSS points per year versus 0.112 in females (P<0.001). Females had a reduced risk of conversion to secondary progressive MS (hazard ratio 0.77, 95% CI 0.67-0.90). In primary progressive MS (1,373 patients), both sexes worsened over time, but there was no significant sex effect on annualized EDSS change. The male disadvantage is clearest in relapse-onset disease, not as a universal rule across every phenotype.

Magyari and Koch-Henriksen's 2022 nationwide Danish analysis in the Journal of Neurology, Neurosurgery & Psychiatry quantified the tradeoff. Among 3,028 men and 6,619 women with relapsing MS onset since 1996 who received disease-modifying therapy, the weighted female-to-male relapse rate ratio was 1.16 (95% CI 1.10-1.22). Women had more inflammatory activity, and that sex difference in relapses disappeared after age 50. Annualized EDSS increase was 0.07 in men and 0.05 in women. With women as the reference, the hazard ratio for reaching EDSS 4 was 1.34 in men, and for EDSS 6 it was 1.43. Diagnostic delay did not differ significantly by sex.

Put plainly: women get more relapses, especially before midlife; men with relapse-onset disease reach hard disability milestones faster. Female sex is not protective once the diagnosis is established. Our review of disability and women's healthcare access covers how disability status itself changes access to care, including reproductive health. When someone asks whether MS is "worse in women," the accurate answer is: more common and often more inflammatory in women; often faster disability accumulation in men with relapse-onset disease.

What the data cannot yet explain

The sex ratio is not a mystery of underdiagnosis alone. Orton's Canadian analysis argued against simple diagnostic bias because time-to-diagnosis did not differ by sex. The speed of the rise points to environment (smoking patterns, adolescent obesity, vitamin D and sun exposure, later and fewer pregnancies, occupational change) interacting with sex-specific biology. No single exposure has been shown to account for the full shift, and good researchers say so.

Hormones are an incomplete answer. Pregnancy's third-trimester quiet period and the postpartum rebound fit hormonal immunomodulation, and Magyari's relapse data show the female excess in relapse activity fading after age 50. Hormones do not explain why male incidence stayed relatively flat while female incidence climbed, and they do not explain the male progression disadvantage. Genetic load, X-chromosome dosage, EBV timing, and lifestyle all remain on the table. The data are stronger on description than on mechanism.

Measurement has limits too. U.S. national prevalence rests on insurance claims algorithms, not a door-to-door census. Global Atlas figures depend on country reporting of uneven quality. Pregnancy studies from the 1990s describe a less treated generation than the one walking into clinic now. When I quote PRIMS next to Langer-Gould, I am comparing eras as well as sample sizes.

Research funding and trial enrollment still lag the female burden of autoimmune disease generally, a gap we document in women's underrepresentation in clinical trials. MS research has done better than many fields on including women, because the epidemiology forced the issue. The remaining hole is mechanistic: why the incidence rise is female-specific, and how to turn that into prevention.

For the patient who asked about pregnancy, the summary is short. Women carry most of the MS burden: about 2.8 times the male prevalence in the best U.S. study. The female excess has grown over decades in several countries. Pregnancy usually lowers relapse rates, especially in the third trimester. The first postpartum months are higher risk, though most women still do not relapse then. Disability progression is not gentler in women simply because incidence is higher.

Frequently asked questions about multiple sclerosis in women

What is the female-to-male ratio in multiple sclerosis?

In the United States, Wallin et al. (Neurology, 2019) estimated a 2010 prevalence ratio of 2.8 women per man (450.1 vs 159.7 per 100,000). Hittle et al. (JAMA Neurology, 2023) reported an overall ratio of 2.9 and found 76% of identified U.S. adult cases were female. Canadian birth-cohort data from Orton et al. (2006) put the ratio above 3.2:1 in recent birth years.

Is the MS sex ratio still rising?

In several high-quality series, yes. Alonso and Hernán's systematic review estimated the incidence ratio rose from 1.4 in 1955 to 2.3 in 2000. Danish registry work shows female incidence more than doubling from the 1950s to the 2000s while male incidence rose modestly, and the female share of prevalent cases in Denmark climbed from 58.7% in 1975 to 68.5% in 2023. Not every country shows the same trend.

What is the typical age at onset for MS in women?

NINDS reports that symptoms usually begin between ages 20 and 40. The Atlas of MS (2020) reported a global mean age at diagnosis of 32 years. Incidence has also risen among people with onset at 50 or older, especially women, in long-running Danish data. Classic young-adult onset is still typical, but not the whole picture.

How does pregnancy affect MS relapse rates?

In PRIMS (Confavreux et al., 1998), annualized relapse rate fell from 0.7 before pregnancy to 0.2 in the third trimester, then rose to 1.2 in the first three postpartum months. Vukusic et al. (2004) found 72% of women had no relapse in that early postpartum window. Contemporary Kaiser Permanente data (Langer-Gould et al., 2020) found lower pre-pregnancy activity overall and a 26.4% postpartum-year relapse proportion, with exclusive breastfeeding linked to lower risk (adjusted hazard ratio 0.37).

Do women with MS accumulate disability faster than men?

Usually the opposite. MSBase data (Ribbons et al., 2015) showed faster EDSS progression in men with relapse-onset MS (0.133 vs 0.112 points per year). Danish treated-cohort data (Magyari and Koch-Henriksen, 2022) found men had higher hazards of reaching EDSS 4 and EDSS 6 (HRs 1.34 and 1.43) even though women had higher relapse rates before age 50.

How many people in the United States live with MS?

Wallin et al. estimated 727,344 U.S. adults with MS in 2010 under their higher model, at a female-to-male prevalence ratio of 2.8 (450.1 per 100,000 women vs 159.7 per 100,000 men). Their 2017 extrapolation reached as high as 913,925 adults. Hittle et al. found 76% of identified cases in the underlying claims pools were female.