If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Medicine, Division of Neurology and the Djavad Mowafaghian Centre for Brain Health, University of British Columbia, Vancouver, British Columbia, Canada
Department of Medicine, Division of Neurology and the Djavad Mowafaghian Centre for Brain Health, University of British Columbia, Vancouver, British Columbia, Canada
Trends in MS incidence rates over time have not been examined on a global basis.
•
Our systematic review included 65 regional estimates across 24 countries.
•
By absolute number of studies, the predominant pattern was increasing MS incidence.
•
No pattern was seen with consistent case definitions/high population coverage.
•
There were few eligible studies to assess change in incidence in many regions.
Abstract
Background
The incidence of multiple sclerosis (MS) has reportedly increased over time; however, change in MS incidence has not been rigorously assessed globally.
Methods
We followed the guidelines for the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement. Two independent reviewers systematically searched Scopus, PubMed and Web of Science for peer-reviewed publications in English from 1 January 1985 to 24 September 2020 reporting MS incidence for at least two contiguous five-year periods with clearly-defined case ascertainment. The outcome was change in MS incidence rate according to geographical region.
Results
We identified 64 papers providing 65 regional estimates (including three paediatric-onset MS) across 24 countries covering ∼3% of the world's population (in 2000/1 or closest available total population for the entire country), with quality (adapted Newcastle-Ottawa Scale) ranging from sufficient to good. Studies were mainly from Italy (n=14 including San Marino), Norway (n=10) or Canada (n=9), with no studies in the Africa or South-East Asia regions. Of the 62 whole-of-population estimates, MS incidence rates: significantly increased in 38 (61%), significantly decreased in 13 (21%) and remained stable in 11 (18%). In the paediatric-onset studies, MS incidence was stable in two (67%) and increased in one (33%). Many estimates derived from only selected (often small) regions of a country. For 42 (68%) of the whole-of-population estimates (and two of the paediatric-onset estimates) a consistent case definition or diagnostic criteria over the entire study period was explicitly reported. Across the n=9 whole-of-population estimates based on a consistent case definition for the duration of the study period, and including a substantial proportion of the population of a country (≥one-third), incidence rates were stable in n=3, increased in n=3 and decreased in n=3. Studies using a consistent case definition covered ∼2.7% of the global population; incidence rates were stable in 0.9% of the global population, decreased in studies covering 1%, and increased in those covering 0.8% of the global population.
Conclusion
The studies reporting change in MS incidence rate over time were limited by world region and the proportion of the global population covered. Although by number of studies, the predominant pattern was increasing MS incidence, in studies where a consistent case definition was used across the duration of the study and with high population coverage, no predominant pattern of MS incidence was evident.
Multiple sclerosis (MS) is an inflammatory and neurodegenerative disorder of the central nervous system, affecting around 2.8 million people worldwide (
The prevalence of MS appears to have increased over time, likely a combined result of earlier detection and diagnosis, improved disease management and life expectancy, and improvements in data quality and accuracy (
), changes in incidence rates over time, globally, have not been examined. We performed a systematic review of the peer-reviewed literature from 1985 to 2020 to examine trends over time in MS incidence rates, globally and within geographic regions.
2. Methods
2.1 Search strategy and selection criteria
The review protocol was prospectively registered on PROSPERO (CRD42020152035) and we followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (
). We searched three databases (Scopus, PubMed and Web of Science) using the terms ‘multiple sclerosis, MS, and incidence’ and the reference lists of recent meta-analyses of MS incidence. Full text articles published in English from 1 January, 1985 to 24 September, 2020 were potentially eligible. The study start (1985) broadly corresponded with the use of MRI in clinical practice to aid MS diagnosis (in high-income countries) (
Study inclusion criteria were: MS incidence reported for at least ten years of observation (including at least two contiguous five-year periods); a defined geographical region covered, with a well-defined underlying study population (by geography and time period); clearly stated data source(s) and case ascertainment method(s), with cases identified using internationally recognised MS diagnostic criteria (e.g., Schumacher, Poser, or McDonald) or a validated algorithm applied to hospital or physician-based medical records or health administrative data. Studies were excluded if: case ascertainment was clearly limited in scope (e.g., from only one of several possible clinics or hospitals in a region and/or the geographic region a clinic/hospital served was poorly defined); incidence estimates were likely to be unstable due to the underlying population being small and/or with limited follow-up (defined as <100,000 person-years); the source population was not representative of the geographic region for the time period (e.g., military cohorts); or, case identification was by self-report.
2.3 Identification of studies
Two researchers (JL and HSN) independently screened titles, followed by abstracts and full texts. Duplicates were removed and disagreements were resolved by consensus. Articles reporting unique results were included.
2.4. Data extraction and quality assessment
For each included study, two authors (JL and HSN) extracted the following data: the primary author; year of publication; study location; the case definition or diagnostic criteria and data source(s); MS case characteristics (number, sex, race, age at symptom onset, or if unavailable, MS diagnosis), total population number; measurement period; crude MS incidence rates; and sex and/or age standardised MS incidence rates (noting the standard population used, where available). Where incidence rates were reported graphically only, GetData Graph Digitizer version 2.26.0.20 was used to extract the relevant data. Where study authors reported more than five-year blocks of incidence, data for all of the most recent five-year blocks were extracted (e.g., if eighteen years, the last fifteen were extracted). Each study's findings were reported by country or region (using the six World Health Organization regions, as used in the Atlas of MS 3rd Edition) (
Wells G., Shea B., O'Connell D., Peterson J., Welch V., Losos M., et al. (2020) The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analysis [10 October 2020]. Available from: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp.
) was used to assess study quality and risk of bias, based on five categories: diagnostic criteria or case ascertainment; representativeness of the study sample; definition of the study location; adequacy of data reporting across the study years; and incidence data report method (unadjusted/crude and adjusted/age and/or sex-specific). Each category was graded from good to insufficient quality (A to C; in the Supplement). Any study scoring grade C (‘insufficient’) in any single category was excluded.
) to estimate the average annual change in incidence within each study, with 95% confidence intervals and p-values, and graphed the available data points (and trend line) to allow the reader to visualise the trends (provided in the Supplementary data). A p-value of <0.05 was considered to denote a statistically significant change in incidence over time. Where data were provided only as the mean for a period, we assigned that value to each year in the period, and used those data in the regression analysis. We used these results to provide a qualitative summary for each study of the incidence trend over time.
3. Results
We identified 65 regional estimates (including three for paediatric-onset MS (
Pediatric-onset multiple sclerosis and other acquired demyelinating syndromes of the central nervous system in Denmark during 1977-2015: a nationwide population-based incidence study.
)) from 24 countries spanning four continents (Table 1 and PRISMA study flow chart provided in the Supplement). The combined total underlying population was ∼180 million people, representing ∼3% of the world's population in 2000 (
United Nations (2017). Department of Economic and Social Affairs, Population Division. World Population Prospects: The 2017 Revision, Volume I: Comprehensive Tables (ST/ESA/SER.A/399).
) to summarise findings, the eligible studies in this systematic review covered (by population, in the year 2000): 3% of the Americas Region, ∼15% of the Europe Region, ∼4% of the Eastern Mediterranean Region and ∼1% of the Western Pacific Region. No eligible studies in the Africa and South-East Asia Regions were identified. Over half the MS incidence estimates (n=33) came from three countries: Italy (n=15, including two regional estimates in Mount Etna and San Marino) (
Development and validation of an administrative data algorithm to estimate the disease burden and epidemiology of multiple sclerosis in Ontario, Canada.
Absolute annual change is the coefficient of the linear trend line across the data points provided for each year or 5-year period; standardised incidence rates used where provided, otherwise, crude incidence rates.
Qualitative summary based on the regression coefficient, 95% confidence interval for the absolute annual change, and p-value. IPMSSG = International pediatric MS Study Group.
Crude
Standardised
Americas Region
Al-Sakran 201819 Saskatchewan, Canada
‘Marrie algorithm’: ≥3 hospital, physician, or MS DMD claims (prescriptions filled) (
Age- and sex-standardised to the Canadian population, Census 2001.
(CCDSS)
Incidence decreased CCDSS (2004–2013)
Hader 200721 Saskatoon, Saskatchewan, Canada
Allison and Millar, 1954; Schumacher; Poser
Regional hospital records, family physicians, neurologists, nursing homes, Home Care Program, MS Society and a provincial MS Treatment Database (est. 1997)
1990–1999
9.20 (7.90–10.70)
NR
-0.12 (-0.16, -0.07) p<0.001
Incidence decreased (1990–2004)
2000–2004
7.90 (6.20–9.90)
NR
Kingwell 201522 British Columbia, Canada
≥ 7 MS claims (ICD-9/10 codes 340/G35) for people who were resident in British Columbia for >3 years following the first MS or demyelinating disease claim; ≥ 3 MS claims for those with ≤ 3 years of residency (akin to the Marrie algorithm)
Health administrative data, including hospital, physician and prescriptions filled
Age- and sex-standardised to the Canadian population, Census 2001.
Marrie 201323 Nova Scotia, Canada
≥ 7 MS claims (ICD-9/10 codes 340/G35) for people with >3 years of administrative data; ≥ 3 MS claims for those with ≤ 3 years of data (akin to the Marrie algorithm)
Health administrative data including hospital, physician claims and MS clinic data
Age- and sex-standardised to the Canadian population, Census 1996.
Cabre 200928 French West Indies (Martinique and Guadeloupe)
McDonald 2005
Inflammatory CNS diseases registry (1992–2007) based on: hospital records, open-care neurologists, health records, health insurance organisations, MS patient associations
July1992-June1997
0.93 (0.76–1.20)
NR
0.07 (0.05, 0.08) p<0.001
Incidence increased (1992–2006)
July1997-June2002
1.20 (1.01–1.39)
NR
July2002-June2007
1.67 (1.46–1.88)
NR
Gracia 200929 Republic of Panama
Poser; McDonald 2001 (used from 2003)
Clinical files from the National Public Health System and private practice
1991–1995
0.20
NR
0.02 (0.01, 0.35) p=0.006
Incidence increased
1996–2000
0.34
NR
(1991–2005)
2001–2005
0.42
NR
Europe Region
Joensen 201130 Faroe Islands
Poser (applied from 1943–1986); Poser, McDonald and Thompson (used 1986–2007)
Hospital records, government disability registry, private neurologist
1993–2002
4.90 (2.90–7.00)
NR
-0.22 (-0.32, -0.13) p<0.001
Incidence decreased (1993–2007)
2003–2007
2.40 (2.10–3.30)
NR
Koch-Henriksen 201831 Denmark
Allison and Millar, 1954 (for MS onset before 1994); Poser (for MS onset 1994–2004); McDonald (for MS onset 2005-end); (initially all ages, then restricted to: 10–64 years)
Age-standardised to the European Standard Population (1976).
Celius 200136 Oslo, Norway
Poser
Oslo MS-Registry including hospital records, patient registries, private neurologists, MS rehabilitation and nursing homes, patient societies, disability pension records from the National Health Insurance Administration
1987–1991
7.20 (5.00–10.20)
NR
0.23 (0.12, 0.33) p=0.001
Incidence increased (1987–1996)
1992–1996
8.70 (6.30–11.90)
NR
Dahl 200437 Nord-Trøndelag County, Norway
Poser
Hospitals, outpatient clinic and the sole private neurologist records
Data from previously published studies (1950–1991), 2 unpublished master's theses (presented in 2009 and 2011), subsequent years: hospital records and outpatient clinics
Age-standardised to the European Standard Population (1976).
(1990–1999)
Grassivaro 201957 Province of Padua (Padova), Northeast Italy
McDonald 2001, 2005
Neurological centres, drug-dispensing records, outpatient records of neurorehabilitation services, chronic care services, local branch of the Italian Society for MS, archives of the National Pension Institute and National Health Insurance
1990–1999
4.10
NR
0.12 (0.10, 0.13) p<0.001
Incidence increased (1990–2015)
2000–2009
5.50
NR
2011–2015
6.50
NR
Nicoletti 201158 City of Catania, Sicily, Italy
Poser
Neurological departments, Italian MS Association and clinical records
Age-standardised to the Italian population, Census 2001.
Ranzato 200364 Province of Padova (Padua), Northeast Italy
Poser
Hospital records, health records, Association if Invaliding Disease of Padova, National Pension Institute and National Health Insurance Scheme, Italian MS Society
1985–1989
2.60 (2.08–3.12)
NR
0.14 (0.01, 0.19) p<0.001
Incidence increased (1985–1999)
1990–1994
3.90 (3.19–4.61)
NR
1995–1999
4.20 (3.74–4.66)
NR
Salemi 200065 Bagheria City, Palermo, Sicily, Italy
Poser
Hospital records, out-patient neurology clinic, GPs & neurologists were ‘informed’, rehabilitation archives, public & private MRI centres, Office for Handicapped People, social workers, television channel
1985–1989
3.50 (1.50–6.90)
NR
0.27 (0.15, 0.40) p=0.001
Incidence increased (1985–1994)
1990–1994
5.30 (2.70–9.20)
NR
Solaro 201566 Province of Genoa, Italy
Poser
Hospital records, regionally serving neurologists, health records, disability pension records, Italian MS Society
1998–2002
6.82
NR
-0.05 (-0.34, 0.25) p=0.72
Incidence stable
2003–2007
6.26
NR
(1998–2007)
Granieri 200867 Republic of San Marino
Poser
Hospital records (private & public), telephone surveys of neurologists, ophthalmologists, GPs, pharmacists, social workers
1990–1994
10.80 (6.00–19.00)
NR
-0.32 (-0.49, -0.15) p=0.001
Incidence decreased (1990–2005)
1995–1999
6.20 (2.80–12.60)
NR
2000–2005
6.80 (3.20–11.40)
NR
1999–2003
1.7 (0.00–3.40)
NR
Valadkeviciene 201868 Lithuania
ICD-10 code G35: newly diagnosed MS in ambulatory service
Compulsory Health Insurance System including ambulatory and stationary visits
Age-standardised to the WHO World Standard Population 2000–2025.
(2006–2015)
Paediatric-only studies
Marrie 201818 Ontario, Canada
Marrie algorithm ≥3 hospital or physician claims for MS (ICD-9/10 codes 340/G35) CCDSS algorithm ≥1 hospital or ≥5 physician claims (ICD-9/10 codes 340/G35) within 2 years; (restricted to: 0–18 years)
Province-wide health administrative including hospital and physician claims
Kuwait National MS Registry (hospital and MS clinic databases)
1994–1998
0.66
NR
0.07 (0.03, 0.10) p=0.001
Incidence increased
1999–2003
1.29
NR
(1994–2013)
2004–2008
1.45
NR
2009–2013
1.77
NR
CCDSS=Canadian Chronic Disease Surveillance System. NR=not reported. CNS=central nervous system. DMD=disease modifying drug (used in the treatment of MS). ICD=International Classification of Diseases. BOT-MS = Buskerud (Vestre Viken Hospital Trust in Drammen), Telemark (Telemark Hospital Trust in Skien), Oslo University Hospital. ESP=European Standard Population. WHO=World Health Organization. GP=general practitioner.
# Absolute annual change is the coefficient of the linear trend line across the data points provided for each year or 5-year period; standardised incidence rates used where provided, otherwise, crude incidence rates.
Qualitative summary based on the regression coefficient, 95% confidence interval for the absolute annual change, and p-value. IPMSSG = International pediatric MS Study Group.
a Age- and sex-standardised to the Canadian population, Census 2001.
b Age- and sex-standardised to the Ontario population, Canadian Census 2001.
c Age- and sex-standardised to the Ontario population, Census 1991.
d Age- and sex-standardised to the Canadian population, Census 1996.
e Age-standardised to the European Standard Population (2013).
f Age-standardised to the European Standard Population (1976).
g Age-standardised to 2006–7 European population (EUROSTAT; December 3, 2007).
h Total population of the South-Western Sardinia Population Cocco et al., 2011.
i Age-standardised to the Italian population, Census 2001.
j Age- and sex-standardised to the Italian population, National Census 2011.
k Age-standardised to the WHO World Standard Population 2000–2025.
l Age-standardised to the Australian population, Census 2011.
m Age- and sex standardised to the Canadian population, Census 2006.
Estimates of the number of reported incident MS cases (and the periods of study) varied greatly, from a low of 20 (1985–1994) in Bagheria City, Italy (
). Less than half of the estimates were in the form of age and sex-standardised incidence rates (n=26, 42%), using a variety of standard populations (Table 1).
Table 2Incident multiple sclerosis over time: average annual number of incident cases over total time period and approximate proportion of total country population covered by the regional study (year 2000/1 or closest available total population for the entire country).
Broad study location (number of studies); total general underlying population, plus calendar year; percentage of country's population covered by the study
Regional study location [population]
Total number incident MS cases/ total time period in years (study calendar years and other relevant detail)
Average annual number of cases
Approximate proportion of total country population covered by regional study
Development and validation of an administrative data algorithm to estimate the disease burden and epidemiology of multiple sclerosis in Ontario, Canada.
Total population at mid-year by region in United Nations (United Nations, 2017)
Martinique and Guadeloupe population 2000=812,000 Approximate proportion of total (unique) country population covered by regional studies in the Caribbean=2%
estimated from the average incidence for the period from the relevant paper and the average population for the period from the relevant national statistics organization.
estimated from the average incidence for the period from the relevant paper and the average population for the period from the relevant national statistics organization.
estimated from the average incidence for the period from the relevant paper and the average population for the period from the relevant national statistics organization.
8.0%
France (1) Total France population 2000=59,608,000
estimated from the average incidence for the period from the relevant paper and the average population for the period from the relevant national statistics organization.
5.2%
Western Greece, Greece, Prefectures of Aitolokarnania, Achaia and Ilia (
estimated from the average incidence for the period from the relevant paper and the average population for the period from the relevant national statistics organization.
estimated from the average incidence for the period from the relevant paper and the average population for the period from the relevant national statistics organization.
estimated from the average incidence for the period from the relevant paper and the average population for the period from the relevant national statistics organization.
estimated from the average incidence for the period from the relevant paper and the average population for the period from the relevant national statistics organization.
16.2%
Western Pacific Region
Australia (1) Total Australia population 2000=19,066,000
Pediatric-onset multiple sclerosis and other acquired demyelinating syndromes of the central nervous system in Denmark during 1977-2015: a nationwide population-based incidence study.
Pediatric-onset multiple sclerosis and other acquired demyelinating syndromes of the central nervous system in Denmark during 1977-2015: a nationwide population-based incidence study.
Pediatric-onset multiple sclerosis and other acquired demyelinating syndromes of the central nervous system in Denmark during 1977-2015: a nationwide population-based incidence study.
(denominator=year 2000/1, or closet available total population for the entire country)
^ estimated from the average incidence for the period from the relevant paper and the average population for the period from the relevant national statistics organization.
a Total population at mid-year by region in United Nations (
United Nations (2017). Department of Economic and Social Affairs, Population Division. World Population Prospects: The 2017 Revision, Volume I: Comprehensive Tables (ST/ESA/SER.A/399).
) the remainder had an approximate coverage of 0.04% to 38% of the total country population (Table 2). Case ascertainment source(s) varied from single (e.g., regional MS Society records (
)). When clinical diagnostic criteria were used, these were largely reflective of the era; Poser (1983) criteria were most common, followed by McDonald (2001–2017), Allison and Millar (1954) and Schumacher (1965). Other approaches included validated algorithms, typically based on administrative data from hospital and/or physician claims, using International Classification of Diseases (ICD) diagnostic codes to identify MS cases (
Development and validation of an administrative data algorithm to estimate the disease burden and epidemiology of multiple sclerosis in Ontario, Canada.
According to the results from the linear regression analysis, over time and across all whole-of-population estimates (n=62), MS incidence rates: significantly increased (n=38), significantly decreased (n=13), or remained stable (n=11).
Of the 62 regional estimates, 42 were explicitly reported to be based on a consistent case definition or diagnostic criteria to identify MS cases over the entire study period covered. These estimates covering ∼2.7% of the global population.; 57% (n=24/42) showed increasing incidence rates (covering 0.8% of the global population), 21% (n=9/42) stable incidence rates (0.9% of the global population), and 21% (n=9/42) decreasing incidence rates (1.0% of the global population). Nine of these studies each covered a substantial proportion (≥one-third) of the country's population (and in total, 1.9% of the global population). Three studies showed stable incidence rates (
Development and validation of an administrative data algorithm to estimate the disease burden and epidemiology of multiple sclerosis in Ontario, Canada.
Thus, for the whole-of-population estimates, from studies using a consistent case definition over time, and covering a substantial proportion of the country's population, there was no predominant pattern of trend in MS incidence over time. In studies covering a smaller proportion of the country's population or not using a consistent case definition across the whole study period, the predominant trend was increasing incidence over time.
A greater proportion of studies undertaken wholly prior to the mid-point of our period of interest (2003), and using a consistent case definition across the study periods, showed increasing incidence (23 estimates; 69% increasing, 9% stable, 22% decreasing) compared to those with contiguous study periods wholly after 2003 (7 estimates; 29% increasing, 42% stable, 29% decreasing). Furthermore, when grouping these studies by latitude band (<30, 30–50, >50 degrees), stable or decreasing incidence was more common in the highest latitude band (n=19 estimates; 53% increasing, 16% stable, 32% decreasing) compared to the middle latitude band (n=22 estimates; 64% increasing, 23% stable, 14% decreasing), with too few studies in the lowest latitude band (n=2; 1 increasing, 1 stable) to allow assessment.
3.1.1 Italy
Twelve of the regional estimates in this region derived from studies explicitly reporting consistent methods to identify cases over time (ten Poser criteria and two McDonald criteria). Of these, incidence rates significantly increased in eight (1985–2018) (
Studies covered approximately 7% of Italy's population, ranging from 0.04% to 1.9% for individual regional estimates (Fig. 1; Table 2). Of the 15 regional estimates in Italy and San Marino, MS incidence significantly increased in ten (including both geographical locations in Mount Etna) (
Of the eight studies in Canada, seven explicitly used consistent methods to identify cases across the course of the study (six used algorithms and one Poser criteria); the MS incidence rates were stable in three studies (1999–2013) (
Development and validation of an administrative data algorithm to estimate the disease burden and epidemiology of multiple sclerosis in Ontario, Canada.
Studies covered approximately 69% of Canada's population (Fig. 2; Table 2), with individual estimates for 0.6% to 38% of the population. Three estimates showed stable incidence (1999–2013), (
Development and validation of an administrative data algorithm to estimate the disease burden and epidemiology of multiple sclerosis in Ontario, Canada.
Five studies in Norway explicitly used consistent methods (Poser criteria) to identify cases over time. Of these studies, MS incidence significantly increased in four studies (1987–2006) (
). Studies covered ∼56% of Norway's population (Fig. 3; Table 2), with individual estimates covering 2.8% to 11%. In Norway, MS incidence rates increased over time in eight counties (from 1985–2018) (
) covering ∼41% of the population, but significantly decreased in two southern counties (from 1988–2012), covering ∼15% of the total Norwegian population (
Of the seven studies in this region, three explicitly used a consistent case definition throughout the study; the incidence rate significantly decreased (1986–2000) in two studies (
), covering 45% of Finland's population. The two Swedish studies each covered ∼3% of Sweden's population and reported similar numbers of incident cases (from 15 to 18 cases per year; totalling 181 (Bostrom et al., 2009) and 201 (Svenningsson et al., 2015) cases). However, the MS incidence rate significantly decreased in Western Sweden (Värmland County; 1991–2000) (
In both of the identified studies from this region, the MS incidence rate significantly increased over three consecutive 5-year periods. However, the study in Martinique and Guadeloupe covered only ∼2% of the total Caribbean population (
There were six single studies for (or within) a specific country for this region, covering from 3% to 100% of that country's population (table 2). In these studies the MS incidence rate was: stable in Bavaria, Germany (2006–2015) (
). For the two other studies covering a larger population size, the MS incidence rates significantly increased in Crete (1985–2004; 5.2% of Greece's population) (
). Two studies in Spain covered ∼0.5% of the total population; the MS incidence rates were stable in Central-Eastern Spain (Bajo Aragon, Teruel, 1988–2002; 42 incident cases), (
) while a smaller study in South East Wales (Cardiff unitary authority, covering <1% of the UK's population) showed a significant increase in the MS incidence rate (1988–2007) (
Qualitative summary based on the regression coefficient, 95% confidence interval for the absolute annual change, and p-value. IPMSSG = International pediatric MS Study Group.
00443-6&id=gr1.jpg){kind=link}
00443-6&id=gr2.jpg){kind=link}
00443-6&id=gr3.jpg){kind=link}