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 Neurology, Telemark Hospital Trust, P.b. 2900 Kjørbekk, 3710 SKIEN, NorwayInstitute of Health and Society, University of Oslo, P.b. 1072 Blindern, 0316 OSLO, Norway
Department of Neurology, Vestre Viken Hospital Trust, P.b. 800, 3004 Drammen, NorwayInstitute of Clinical Medicine, University of Oslo, P.b. 1072 Blindern, 0316 OSLO Norway
Institute of Clinical Medicine, University of Oslo, P.b. 1072 Blindern, 0316 OSLO NorwayDepartment of Neurology, Oslo University Hospital, Ullevål, P.b. 4956 Nydalen, 0424 OSLO, Norway
Department of Research, Telemark Hospital Trust, Skien, NorwayDepartment of Pharmaceutical Bioscience, University of Oslo, Boks 1072 Blindern, 0316 OSLO Norway.
Institute of Clinical Medicine, University of Oslo, P.b. 1072 Blindern, 0316 OSLO NorwayDepartment of Neurology, Oslo University Hospital, Ullevål, P.b. 4956 Nydalen, 0424 OSLO, Norway
“Prevalence of multiple sclerosis in rural and urban districts in Telemark County, Norway”
•
The paper present a complete cohort of patients with a confirmed diagnosis of Multiple Sclerosis (MS) in Telemark County, Norway, from 1999 to 2019.
•
Between 1999 and 2018, the yearly incidence of MS increased from 8.4/105 to 14.4/105.
•
The prevalence of MS in Telemark County is among the highest ever reported in Norway, 260.6/105 as of January 1st 2019.
•
The prevalence are even higher in rural areas of Telemark, in 2019 the prevalence rates were 250.4/105 in urban and 316.2 /105 in rural areas.
Abstract
Objective
To explore the trends in prevalence and incidence of multiple sclerosis (MS) in Telemark, Norway (latitude 58.7-60.3˚N), over the past two decades, with focus on differences between rural and urban areas.
Methods
Data from all patients with a confirmed diagnosis of MS in Telemark since 1993 were prospectively recorded and collected in a retrospective chart review. Prevalence estimates on January 1st 1999, 2009 and 2019, and incidence rates at five-year intervals between 1999 and 2018 were calculated and all results were adjusted to the European Standard Population. The study population was divided into urban and rural residency using a Norwegian governmental index.
Results
We registered 579 patients with MS in Telemark between 1999 and 2019. The adjusted prevalence estimates for January 1st 1999, 2009 and 2019 were 105.8/105, 177.1/105 and 260.6/105, respectively. In 2019, the prevalence estimates were 250.4/105 in urban and 316.2 /105 in rural areas. Between 1999 and 2018, the yearly incidence increased from 8.4/105 to 14.4/105.
Conclusions
The prevalence of MS in Telemark is among the highest ever reported in Norway, consistent with an increasing incidence in the county over the past twenty years. The even higher prevalence in the rural areas is unlikely to be explained by possible risk factors like latitude, exposure to sunlight and diet. Further studies on differences between urban and rural areas are required to reveal possible new risk factors.
Multiple sclerosis (MS) is an inflammatory disease with neurodegeneration. Onset is mainly in young adulthood with impact on function, employment, income and quality of life (
). Globally, there are an estimated 2.2-2.3 million people living with MS, and Europe is a region with high prevalence, estimated at 127/100 000 (105) in 2016 (
Collaborators GBDMS. Global, regional, and national burden of multiple sclerosis 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016.
). They claim that parts of Telemark are high-incidence areas for MS, and postulate that there is an association with farming, dairying and low seafood consumption in inland areas. The incidence and prevalence of MS in Telemark have not been systematically investigated before, but a nationwide study from Norway in 2012, estimated the prevalence in Telemark to be 194/105 (
There has been some focus on the variations in prevalence between rural and urban areas worldwide. A recently published study from Bavaria, Germany, describes a higher incidence and prevalence in urban than in rural areas (
). However, studies on environmental factors in early childhood have shown a significantly increased risk of developing MS among inhabitants in rural areas (
The aim of this study was to explore the trends in prevalence and incidence of MS in Telemark over the past two decades, particularly focusing on differences between rural and urban areas.
2. Material and methods
2.1 Geographical setting
Telemark county is located in the southeastern part of Norway, at latitude 58.7-60.3˚N, with a total area of 15 296 km2 (Fig. 1a). The county extends from the coastline of Skagerrak to the Hardanger Plateau, approximately 1 200 meters above sea level. The main city is Skien, where the county's only neurological department is located. Telemark and Skien had a population of 173 318 and 54 645 respectively as of January 1st 2019. Telemark consists of 18 municipalities with a wide variation in population density, topography and culture, comprising both smaller cities and rural areas, and the distance to specialist health services varies greatly.
Fig. 1a) Map of Norway with Telemark county marked in grey. b) Details of Telemark county, municipality by color according to centrality index
The Norwegian government has developed an index characterizing the different municipalities by how centrally they are located. The index comprises information on service functions and work places a resident can reach within 90 minutes. Added up, each municipality receives an index from 1 to 6, where 1 denotes the most central areas (
). In Telemark, the different municipalities have indices ranging from 3 to 6. For the comparison of different areas, we have considered an index of 3 as an urban area whereas indices 5 and 6 are grouped together as rural areas. Fig. 1b shows the different municipalities of Telemark, labelled by the centrality index.
2.2 Data collection and study population
This study is a part of the ongoing BOT-MS project, which is a database consisting of all patients registered with a confirmed MS diagnosis at the two regional hospitals in the counties Buskerud (Vestre Viken Hospital Trust in Drammen) and Telemark (Telemark Hospital Trust in Skien). The BOT database also includes the majority of the MS patients registered at Oslo University Hospital (OUS). The regional ethics committee of South East Norway and the Data Protection Officer at OUS have approved the project. All individuals registered in the electronic patient records with the ICD-10 code G35 (MS) between 1999 and 2019 and patients who fulfilled the diagnostic criteria for definite or probable MS (
) were included. An additional search for the ICD-9 code 340 (MS) between 1993 and 1998 was performed and patients with a verified diagnosis of MS were included. We registered all patients by their unique personal identification number and noted the year of change in status (deceased, migrated to or from the county). The year of the first symptom suggestive of MS was defined as the year of onset. This information, as well as year of diagnosis and subtype of MS, were derived from the medical record review. We classified subtypes of MS as progressive-onset or relapse-onset, the latter including those initially registered with a clinically isolated syndrome (CIS) that was later verified as definite MS, as well as those with secondary progressive MS at the time of diagnosis.
2.3 Prevalence and incidence
Prevalence was calculated based on population data for Telemark on January 1st 1999, 2009 and 2019. The prevalence was defined as the total number of MS patients residing in Telemark per 105 inhabitants in the county at each date. Prevalence according to the centrality index was calculated based on population data for each municipality.
The crude annual incidence was defined as the number of patients diagnosed with definite MS or CIS later converting to definite MS per year when residing in Telemark per 105 inhabitants. We calculated mean yearly incidence at five-year intervals between 1999 and 2019, using the average population at risk during the corresponding five-year interval. Population data stratified by age and sex was obtained from Statistics Norway. For the calculation of age standardized incidence and prevalence, we used the new European Standard Population as reference population (
We used IBM SPSS Statistics for Windows, version 21 (IBM Corp., Armonk, N.Y., USA) for the main statistical analysis, including two-sample independent t-test to compare characteristics at the first and last prevalence dates. 95 % confidence intervals (CI) for prevalence were calculated manually from the formula p ± 1,96 x SD, where SD is the standard deviation, given by the formula , p being the crude prevalence and n the number of persons participating. We used the mid-P exact test (
) to compare the prevalence in rural versus urban areas of Telemark, using OpenEpi.com.
3. Results
3.1 Demographics
Table 2 shows the demographic characteristics of the population on the three prevalence dates. The percentage of females with MS increased from 1999 to 2009 and remained stable from 2009 to 2019. The mean age at onset increased over the two decades, from 32.5 years in 1999, to 36.0 years in 2019. The increase in age at onset was significant for the whole group, as well as for both sexes separately. Accordingly, the study cohort had a significantly higher age in 2019 (53.8 years) than in 1999 (50.5 years) (p=0.009). There was an equivalent significant increase in mean age in the female cohort separately (p=0.009), but not for males. The mean time from onset to diagnosis decreased between 1999 and 2019, from 6.0 to 5.0 years respectively, but the reduction was not significant. The proportion of patients with a relapsing disease at diagnosis increased from 84.7% in 1999 to 90.9% in 2019, with a corresponding trend for each sex separately.
Table 2MS population demographics at the prevalence dates 01.01.1999, 01.01.2009 and 01.01.2019
A total of 625 patients were identified by the ICD-10 code G35, and 32 patients were identified by the ICD-9 code 340. Based on information from the electronic patient record, we excluded 74 patients as they did not fulfill the diagnostic criteria or were miscoded, and 9 patients as deceased prior to the first prevalence date of 01.01.1999. Through the BOT-collaboration, we included five patients diagnosed and treated in Buskerud, while residing in Telemark. Finally, 579 patients with MS, residing in Telemark at any time during the time-period 1999-2018 were included in the calculations. Table 3 shows the changes in the MS population in Telemark during the twenty-year period.
Table 3Changes in MS population in Telemark 1999-2019
The crude prevalence on 01.01.1999 was 97.3/105, on 01.01.2009, it was 176.1/105, and on 01.01.2019, it was 259.6/105. Table 2 shows the prevalence calculations for all three prevalence dates, including 95 % confidence intervals (CI) for the estimates. After adjusting to the European standard population, the prevalences were 105.8/105, 177.7/105, and 260.6/105 respectively. We also calculated the prevalence with adjustment according to the 1976 European standard population, finding a lower prevalence for 1999 and 2009, but the exact same prevalence for 2019 (data not shown).
The age-adjusted prevalence increased for all age groups over the two decades as shown in Fig. 2. The highest age-adjusted prevalence observed was for females aged 60-69 years on prevalence date 01.01.2019, with a prevalence of 683/105, as shown in Fig. 3.
Fig. 2Age-adjusted prevalence of MS in Telemark with 95% confidence interval, 1999 - 2009 2019.
Comparing the prevalence in the most rural (centrality indices 5 and 6) with the most urban areas (centrality index 3) of Telemark showed a significantly higher prevalence in rural areas. There was a significantly higher prevalence of MS among females in rural areas compared to females in urban areas, while no such difference was seen for males. The finding of a prevalence for females living in areas with centrality index 4 (suburban) of 354.6/105, indicating a gradual decrease towards more urban areas, reinforced this sex-specific pattern. There were no significant differences in mean age for the whole study population, nor for females residing in rural versus urban areas. Data for the last prevalence date are shown in Table 4.
Table 42019 Prevalence of MS in urban (Centrality index 3), suburban (Centrality index 4) and rural (Centrality indices 5 and 6) areas, Telemark, by sex and total. See map in Fig. 1 for index areas. C.I. = confidence interval
Prevalence date 01.01.2019
Male
Female
Total
Centrality index 3 (Urban areas)
Number of cases (% of total)
97 (37.3%)
163 (62.7%)
260 (100%)
Mean age MS patient at prevalence date (95%C.I.)
53.2 (50.6-55.8)
53.9 (51.76-56.1)
53.6 (51.9-55.3)
Population at risk
52 197
52 761
104 958
Prevalence/100 000 (95% C.I.)
185.8 (148.9-222.8)
308.9 (261.6-356.3)
247.7 (217.6-277.8)
Age-adjusted prevalence/100 000 (95% C.I.)
189.8 (152.5-227.2)
308.3 (261.0-355.6)
250.4 (220.2-280.7)
Centrality index 4 (Suburban areas)
Number of cases (% of total)
33 (30.6%)
75 (69.4%)
108 (100%)
Mean age MS pat at prev. date (95 % C.I.)
59.4 (55.2-63.6)
52.6 (49.3-55.9)
54.7 (52.0-57.4)
Population at risk
21 667
21 211
42 878
Prevalence/100 000(95% C.I.)
152.3 (100.4-204.2)
353.6 (273.7-433.5)
251.9 (204.4-299.3)
Age-adjusted prevalence/100 000 (95% C.I.)
155.3 (102.9-207.7)
354.6 (274.6-434.6)
252.3 (204.8-299.8)
Centrality indices 5&6 (Rural areas)
Number of cases (% of total)
20 (23.5%)
62 (72.9%)
82 (100%)
Mean age MS pat at prev. date (95 % C.I.)
51.8 (46.7-56.9)
53.4 (50.2-56.6)
53.0 (50.3-55.7)
Population at risk
12 875
12 607
25 482
Prevalence/100 000(95% C.I.)
155.3 (87.3-223.4)
491.8 (369.7-613.9)
321.8 (252.3-391.3)
Age-adjusted prevalence/100 000 (95% C.I)
146.0 (80.0-211.9)
493.5 (371.2-615.8)
316.2 (247.3-385.1)
p-value for comparison prevalence in rural (indices 5&6) vs urban (index 3)
The crude number of persons in Telemark diagnosed with definite MS or CIS later converted to definite MS in the period 1999-2018 varies between 11 and 27 per year (Fig. 4), with an overall increasing trend. Table 5 shows the crude incidence rates at five-year intervals, and age-adjusted incidence rates using the 2013 European standard population as a reference. Table 6 shows the age-adjusted incidence per year at five-year intervals, per sex.
Fig. 4Number of new cases diagnosed per year in Telemark, 1999-2018
The yearly incidence rate increased, although not significantly, from 8.2/105 to 13.9/105 from the first five-year interval to the last. Both sexes analyzed separately show the same trend, with an increase from 11.0/105 to 17.6/105 in females and from 5.4/105 to 10.2/105 in males. There is a dip in incidence from the second to third five-year intervals for the total group and for the females, which is due to low numbers and the large variation in new cases from one year to the next. When adjusted to the 2013 European standard population, the incidences were higher for all time-intervals for the female subgroup, whereas the adjustment only led to minor changes in the male subgroup and in the total population. We also calculated the adjustment according to the 1976 European standard (data not shown), which gave an even higher incidence for all time-intervals for females, but a lower incidence for males in the last time-interval. However, for the population as a whole, the differences between the two versions of European standards are minor.
4. Discussion
The prevalence of MS in Norway is among the highest worldwide, and studies from many Norwegian counties consistently report individually high rates. No systematic MS prevalence report from Telemark county has previously been published, and the present study confirms a prevalence of MS that has increased remarkably over the past 20 years, culminating in January 2019 with one of the highest MS prevalences ever published from Norway. Unlike previous studies, which have mainly pointed to a tendency towards increasing incidence of MS in urban versus rural areas, we report a clear trend towards higher prevalence of MS in the most rural areas, with a gradual decrease in more urban areas.
The prevalence estimate from Telemark was 105.8/105 at the first time-point, which is lower than roughly simultaneous calculations from other parts of Norway. In January 1995 the prevalence estimate from Oslo was 120.4 /105, and even higher when only native Norwegians were considered (136/105) (
) corresponded with our finding of 177.8/105 in Telemark in 2009. The most recent national study estimated the MS prevalence for Telemark at 194/105 as of January 1st 2012 (
), which is the latest reported prevalence from Norway until our finding of a prevalence in Telemark of 260/105 in 2019. It is, however, difficult to compare different areas of Norway, with their differences in availability of neurological services and changes in diagnostic criteria (
), especially based on historical data. Despite the possibility for underestimation at the first time point (01.01.1999), the significant increase from the first five-year period (1999-2004) to the next, and throughout the whole study period, is clear.
Prevalence estimates can increase with repeated surveys from the same area for several reasons (
). The Telemark Hospital Trust has the only neurological department in the county, and there are no private neurologists treating MS in Telemark. A team consisting of MS neurologists and nurses organizes the MS care in Telemark, and the team keeps track of all the MS-patients with regular controls. The Telemark Hospital Trust implemented electronic patient records in 1993, thus making searches for diagnoses for historical data easy and precise. We used both ICD-9 and ICD-10 diagnosis of MS as search criteria in this study, and we believe there are few missed cases. Through the research collaboration with the neighboring county of Buskerud and the capital Oslo, we have only identified five patients who were followed up by other hospitals while residing in Telemark over a period of 20 years. Through clinical collaboration with MS neurologists from the other counties in our region, and an evaluation of data from the Norwegian prescription registry, we have not been able to identify other MS patients from Telemark being followed up outside of the county. This confirms the impression of the completeness of our cohort.
The numbers of newly diagnosed MS patients per year is small, and a variation from one year to another is to be expected because of natural fluctuations, but the increase from 2017 to 2018 is most likely related to implementation of the latest revision of the McDonald diagnostic criteria (
). However, the incidence rates for five-year periods in Telemark have shown a clear increase over the past twenty years.
The incidence and prevalence of MS are dependent on the population's age distribution, and adjustment of rates by a hypothetical standard population is common in more recent studies. We have adjusted all our findings to the European Standard Population to be able to compare our data with findings from other countries and regions. We would like to highlight the fact that there are two versions of the standard population: 1976 and 2013. The latter takes into account the growing age of the population (
). In our data, this yielded different results for the first two prevalence calculations of 1999 and 2009, but no differences for the last prevalence date of 2019. There is reason to believe that the Norwegian population was not in accordance with the previous standard, and published adjusted Norwegian prevalence and incidence estimates from the first decade of the millennium using the old European standard may thus be underestimated.
In contrast to most previous studies, we have demonstrated an uneven geographical distribution in terms of rural aggregation of MS in Telemark. These differences are unlikely to be explained by an association of the prevalence of MS with latitude (
). In Telemark, there is a relatively small range of latitude (58.7-60.3˚N) and the UV radiation is considered similar throughout the area, although it is interesting to note that one of the largest rural municipalities, Tinn (see Fig. 1), is surrounded by high mountains, and its inhabitants are not exposed to sunlight for half the year.
The composition of various ethnicities may influence the prevalence. In a previous study, non- western immigrants to Norway had lower crude and adjusted prevalence estimates compared to the total population (
). According to Statistics Norway, the proportion of the population with non-Western background is 6.4 % in the urban areas and 4.1 % in the rural areas of Telemark, and this can only in part explain the higher rural prevalence of MS.
Smoking is a known risk factor for MS on the individual level (
). According to Statistics Norway, the proportion of Norwegians who smoke regularly has decreased from 32 % in 1999 to 12 % in 2018, but this is not reflected in the observed increase in incidence and prevalence estimates of MS. There are, however, well-documented differences in several lifestyle factors according to residency in Norway (
), like findings of 15 % daily smokers in the most rural areas, versus 11 % daily smokers in urban areas (Statistics Norway, 2015). The level of individual education may influence the development of diseases. One Norwegian study showed an inverse relationship between higher education and MS risk (
). Statistics Norway confirms a higher education level among residents in urban versus rural areas of Norway. Dietary patterns have been discussed regarding differences in the prevalence of MS with, traditionally, a higher intake of fat in the inland farming areas, and higher consumption of fish in coastal areas (
). Our experience, however, is that these differences are almost non-existent today. This statement is confirmed by the survey on living conditions performed by Statistics Norway, showing no significant difference in intake of fish/seafood, nor milk products between areas of residence. We would therefore argue that diet alone cannot explain the observed differences between rural and urban areas.
Due to a low sample size, we have not been able to report incidence related to urban and rural areas, which is a shortcoming in this study. Another limitation is the lack of a bigger city in the county (centrality indices 1 or 2). Our findings should be further investigated in a larger cohort, in order to be able to calculate incidence. The overall results should also be adjusted for lifestyle habits and other socioeconomic factors.
The proportion of patients with progressive MS at diagnosis has varied between studies, most likely mainly due to different definitions and classifications (
). There are also differences in the proportions of patients with a primary progressive disease course in Norwegian studies, with 22.3% in Oslo in 1995 (
). These national reports show a time-trend of a decreasing proportion of primary progressive disease, and correspond to our findings in Telemark of 15.3 % primary progressive disease in 1999 and 9.1 % in 2019. This development is predictable, and is most likely due to several factors, including an increased focus on anamnestic reports of earlier episodes of relapsing symptoms. This secures the relapsing diagnosis, which is a prerequisite for disease modifying treatments. The mean age of onset and the mean age of the prevalent population increases over two decades in Telemark. These findings are in accordance with some Norwegian studies (
). A flattening of the increase during the last ten-year period, as we found, may indicate that this is largely due to historically undiagnosed cases among females.
In conclusion, this study from Telemark shows one of the highest reported prevalences of MS in Norway, consistent with an increasing incidence in the county during the last twenty years. We also found an even higher prevalence of MS in the rural areas of the county, which partly confirms the findings of Swank from 1952 that claimed parts of Telemark were particularly high incidence areas. The results need to be further investigated in order to ascertain factors, other than latitude and sunlight, explaining the geographical differences in the prevalence of MS. An understanding of the distribution of MS is important to allow for better planning of health services, which may in turn bring us closer to an understanding of the disease susceptibility, and even development of further strategies for prevention of the disease.
Author contributions for paper
Prevalence of multiple sclerosis in rural and urban districts in Telemark County, Norway
Data statement,
Prevalence of multiple sclerosis in rural and urban districts in Telemark County, Norway
Due to the sensitive nature of the variables registered and the questions asked in this study, survey respondents were assured raw data would remain confidential and would not be shared.
A limited version of the data can be released upon reasonable request to the corresponding author.
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:
Acknowledgments
We would like to thank all the patients that were included in this study as well as Dr. Frøydis Moan Dalene and Dr. Tore Jørgen Mørland, both of them neurologists at Telemark Hospital Trust.
Funding
HØF has received research funding from Telemark Hospital Trust to perform this work.
p < 0.05,
30427-2&id=gr1.jpg){kind=link}
30427-2&id=gr2.jpg){kind=link}
30427-2&id=gr3.jpg){kind=link}
30427-2&id=gr4.jpg){kind=link}