European Journal of Neurology
Volume 8 Issue 5 Page 463 - September 2001
E. G. Celius & B. Vandvik
The Oslo Multiple Sclerosis (MS) Registry was established in 1990, and this is the first report on the prevalence and incidence of MS in the city of Oslo, Norway. The prevalence rate of definite MS on 1 January 1995 was 120.4/105. Inclusion of patients of native Norwegian ancestry only and exclusion of non-Norwegian immigrants yielded a prevalence rate of 136.0/105. A similar prevalence rate (136.5/105) was found when patients and immigrants from the other Nordic countries (Finland, Sweden, Denmark) were included. Segregation of the native Norwegian patients according to the counties where they were born showed no significant differences except for a disproportionate increase of patients born in the inland county of Oppland. A total of 794 cases were resident in Oslo at the time of a diagnosis of definite MS in the period 1972-99. The crude average annual incidence rate for each 5-year period, between 1972 and 1996, increased significantly from 3.7/105 in the 1972-76 to 8.7/105 in the 1992-96 period. The increase was more marked in relapsing-remitting (RR) than in primary progressive disease and in female cases.
A nation-wide survey of the prevalence of multiple sclerosis (MS) in Norway in 1948 (Swank et al., 1952) and a study based on mortality data from 1951 to 1965 (Westlund, 1970) suggested that the south-eastern part of Norway is the area with the highest MS prevalence in the country. Surveys of five out of a total of 19 counties (see Fig. 1) in Norway have since been reported. Studies of the county of Vestfold (on the west side of the Oslo Fjord) showed a prevalence rate of 61.6/105 in 1963 (Oftedal, 1965) and 86.4/105 in 1983 (Edland et al., 1996). In western Norway, the prevalence rate in Hordaland County was approximately 20/105 in 1960 (Presthus, 1960) and 59.8/105 in 1983 (Larsen et al., 1984a), and in Møre-Romsdal County it was 25.7/105 in 1961 (Presthus, 1966) and 75.4/105 in 1983 (Midgard et al., 1991). In northern Norway, the combined prevalence rate in Troms and Finmark counties was 20.6/105 in 1973 (DeGraaf, 1974), 31.5/105 in 1983 (Grønning and Mellgren, 1985) and 73.0/105 in 1993 (Grønlie et al., 2000). The prevalence rate in Gothenburg in Sweden (300 km south of Oslo, on the east coast of the Skagerrak) was 96/105 in 1988 (Svenningsson et al., 1990), and in Denmark 112/105 in 1990 (Koch-Henriksen, 1999).
In the Norwegian studies, an increase in incidence from 1953 to 1977 was concluded to have taken place in the county of Hordaland (Larsen et al., 1984b), whereas fluctuating incidence patterns were reported in Vestfold (Edland et al., 1996), Møre-Romsdal (Midgard et al., 1991) and Troms and Finmark (Grønning and Mellgren, 1985).
Oslo is at once both the capital and one of the 19 counties of Norway with 11.1% of the countrys population. The population of Oslo was 483 401 on January 1 1995 (Statistisk årbok, 1996). Since about 1970, the immigration of people of non-European (mainly African and Asian) ethnic origin has increased substantially, and the number of Asian and African immigrants on 1 January 1995 represented 9.4% of the total population of Oslo (Statistisk Årbok, 1996). The aim of this study was to determine the prevalence and incidence of definite MS in Oslo, and to study the possible variations in incidence over time.
Materials and methods
The Oslo MS-Registry was established at the Oslo City Hospitals Department of Neurology in 1990. Prospective registration of MS patients has since been systematically carried out. Incidence data prior to 1990 have been collected retrospectively. As of 1 January 2001, the registry comprises a total of 1274 definite MS patients with previous or present residence in Oslo since 1972. The primary data source was the patient registry of the Department of Neurology, which is the only neurological department serving Oslo. For historical reasons, some Oslo patients were referred to the Department of Neurology at the National Hospital in Oslo or to the local Oslo hospitals with neurological consultant services prior to 1995. Data from the patient registries of these hospitals were obtained. In addition, data were obtained from neurologists in private practice, MS rehabilitation and nursing homes, and patient societies. Records on all patients receiving a disability pension with a main or supplementary diagnosis of MS were obtained from the National Health Insurance Administration.
Only patients investigated and diagnosed by a neurologist or by a department of neurology were included. All hospital records were reviewed by the same neurologist (EGC) and reclassified according to the Poser criteria (Poser et al., 1983); only cases of definite MS according to the Poser criteria were included. MRI abnormalities were not included as paraclinical evidence for MS. A questionnaire was sent to all patients to obtain additional demographic data. Place of birth was also obtained from the Central Bureau of Statistics in Norway. Years of onset and diagnosis, ethnic origin, and home address at onset, at time of diagnosis and on the prevalence date were recorded for all patients. Additional recorded data included sex, onset symptom, disease type, presence or absence of oligoclonal bands (OB) in the cerebrospinal fluid (CSF) by agarose gel electrophoresis, and results of visual (VEP), somatosensory (SEP), and brainstem evoked potentials (BEP). Year of onset was defined as the year of the first symptom(s) probably related to the disease. When patients experienced more than one onset symptom, the symptom of greatest functional impact was defined as the onset symptom. The disease was classified as either relapsing-remitting (RR) including secondary progressive (SCP), or as primary chronic progressive (PCP) MS. PCP MS was defined as a progressive disease course without bout onset and without a history or recorded evidence of remissions. In 27 patients of the prevalence cohort, it was impossible retrospectively to distinguish between SCP and PCP disease.
The crude prevalence rate was calculated as the number of persons resident in Oslo with a diagnosis of definite MS per 105 inhabitants on 1 January 1995. Age adjusted prevalence rates were calculated based on population data from Oslo on the prevalence day (Statistisk årbok, 1996). Incidence calculations were based on data from 1972 onwards, as this was the first year of computerized diagnostic patient registries in Oslo. Data from 2000 have not been included, as they still are incomplete. The crude incidence rates were calculated as the number of persons resident in Oslo at the time of the diagnosis of definite MS per 105 inhabitants per year. Average annual incidence rates for each 5-year period between 1972 and 1996 were calculated based on the average annual population for the same periods. Population data for all incidence calculations were obtained from the Central Bureau of Statistics in Norway (Statistisk årbok, 1999).
The study was approved by the Regional Medical Research Ethics Committee. Relevant permissions from the Data Protection Registrar and the Public Health Department of the Ministry of Social Affairs were obtained.
Statistical calculations were performed using the Public Domain Software for Epidemiology and Disease Surveillance Epi-info Version 6.0 (Center for Disease Control, Epidemiology Program Office, Atlanta, GA, USA), and the Statistical package for the Social Sciences (SPSS for Windows, Version 9.0). When appropriate, P-values were corrected for number of comparisons (Pc).
A total of 582 patients with a diagnosis of definite MS were alive and resident in Oslo on the prevalence day, 1 January 1995, yielding a prevalence rate of 120.4/105 (95% CI: 110.9-130.7). Of these, 556 (95.5%) were of native Norwegian ancestry (one or both parents Norwegian) and 549 (94.3%) were born in Norway. Fourteen patients were immigrants from the other Nordic countries, seven from other European countries and two from the USA (both of European ancestry). Three patients were of Asian origin. No patients were of Lapp (Sami) origin. Inclusion of patients of native Norwegian ancestry only and exclusion of the non-Norwegian immigrants from the population of Oslo, yielded a prevalence rate of 136.0/105 (95% CI: 125.0-147.9). Inclusion of patients from the other Nordic countries and the Nordic immigrants (Finland, Sweden, Denmark) in the population of Oslo yielded a similar prevalence rate of 136.5/105 (95% CI: 125.4-148.0). The characteristics of the prevalence cohort are given in Table 1. The female : male ratio was 2.1 : 1. Age adjusted prevalence rates are given in Fig. 2. The mean age of the prevalence cohort was 50.1 (women 49.3, men 52.0) years. Segregation of the native Norwegian patients according to the counties where they were born, relative to the contribution from these counties to the population of Oslo (Table 2), showed a disproportionate increase of patients born in the inland county of Oppland, immediate north of Oslo (see Fig. 1) and a slight decrease of patients born in Oslo. However, neither of these differences showed statistical significance when corrected for number of comparisons.
The number of new cases diagnosed per year for the period 1972-99 is given in Fig. 3. Crude average annual incidence rates for 5-year intervals from 1972 to 1996 for the total cohort and for RR and PCP disease separately are shown in Table 3. The increase in incidence was statistically significant for the total cohort (P=0.003), and for both RR (P=0.005) and PCP (P=0.001) disease separately, whilst the difference in increase in incidence between RR and PCP disease was marginally significant (P=0.05). The mean age at onset was 32.8 years (95% CI: 32.1-33.5) and at diagnosis 38.3 years (95% CI: 37.5-39.1) in the incidence cohort.
As shown in Fig. 3, an increase in incidence seemed to emerge at about 1986. Therefore, the total incidence cohort was subdivided into two subgroups (1972-85 and 1986-99, respectively) in order to investigate any differences between the groups possibly explaining the increase in incidence. The characteristics of the two groups and the total incidence cohort are shown in Table 4, indicating a significant decrease in the proportion of abnormal evoked responses (SEP and BEP) and in the frequency of OB in the CSF between the first and the second period. In the first period, the female : male ratio was 1.7 : 1 and in the second 2.3 : 1, this difference was not statistically significant. Table 5 gives the mean age at diagnosis and the mean time from onset to diagnosis according to sex and disease type for the two periods, showing a slight but non-significant decrease in time from onset to diagnosis and in mean age at diagnosis for both sexes as well as for both disease types. The distribution of onset symptoms in the two subgroups and the total incidence cohort is given in Table 6.
Table 1 Characteristics of the prevalence-cohort on 1 January 1995
Table 2 Place of birth by counties in the prevalence cohort and in the population of Oslo, in percent
Table 3 Crude annual incidence rates for 5-year periods 1972-96, for all MS and for RR and PCP MS patients (95% CI)
Table 4 Characteristics of the total incidence cohort and of cohort subgroups 1972-85 and 1986-99
Table 5 Mean age at diagnosis and mean time from onset to diagnosis according to sex and disease type in the incidence cohort subgroups 1972-85 and 1986-99 (95% CI)
Table 6 Distribution of onset symptoms in the total incidence cohort and cohort subgroups 1972-85 and 1986-99
Figure 1 Map of Norway with the counties numbered according to numbers shown in Table 2.
Figure 2 Age adjusted prevalence rates.
Figure 3 Number of new cases diagnosed per year 1972-99.
The present prevalence rate of MS in Oslo of 120.4/105 is the highest so far reported in Norway. The prevalence rate of 136.0/105 in the native Norwegian population of Oslo is even higher. These prevalence rates in Oslo are significantly higher than the prevalence rates previously observed in Norway and in Gothenburg in Sweden, but not significantly different from the latest prevalence rate reported from Denmark. Studies of the Vestfold County (Edland et al., 1996), the city of Gothenburg in Sweden (Svenningsson et al., 1990), and Denmark (Koch-Henriksen, 1999) suggest that the region around the Oslo Fjord and Skagerrak is a high-risk area for MS in Scandinavia. This is supported by this study.
Some differences exist in both the diagnostic and the inclusion criteria used between the cited studies. Only definite cases according to the Poser criteria (Poser et al., 1983) were included in this study, whilst cases of probable MS were included in several of the other studies. This would underestimate rather than overestimate the prevalence rate compared with the other Scandinavian studies (Larsen et al., 1984a; Svenningsson et al., 1990; Midgard et al., 1991; Edland et al., 1996; Koch-Henriksen, 1999; Grønlie et al., 2000).
Only 50% of the native Norwegian inhabitants of Oslo are born in Oslo, the rest come from other parts of the country. With the exception of an increase in MS patients born in the inland county of Oppland north of Oslo (Fig. 1) from which no prevalence data are so far available, there were no major differences between the patients of different county origin (Table 2). From the evidence of uneven prevalence in the country, a lower proportion of Oslo MS-patients born in areas with lower prevalence such as western and northern Norway would be expected. This is not the case and might be explained by an environmental factor triggering the disease in the south-eastern part of Norway. Further analysis regarding age at move to Oslo would be of interest to answer this question.
The mean age, sex ratio, mean age at onset and diagnosis as well as the mean time from onset to diagnosis of the present prevalence cohort are in accordance with other studies from Norway (Larsen et al., 1984a; Midgard et al., 1991; Edland et al., 1996) as well as from Scotland (Rothwell and Charlton, 1998), England (Rice-Oxley et al., 1995; Robertson et al., 1995; Ford et al., 1998) and Switzerland (Groebke-Lorenz et al., 1992). The relative prevalence of RR and PCP disease is comparable with that reported in northern Norway (Grønlie et al., 2000), and Cambridgeshire, UK (Robertson et al., 1995). The proportion of PCP cases in the present study was significantly higher than that reported in the county of Møre-Romsdal (Midgard et al., 1991); whether this represents true differences between the populations studied or a different usage of criteria in distinguishing between SCP and PCP disease is unclear.
The crude average annual incidence rate of diagnosis for the period 1992-96 of 8.7/105 is the highest recorded in Norway till now. This incidence rate is comparable with that recently reported from the western part of Finland (Sumelahti et al., 2000). Higher incidence rates of 10-12/105 have been reported in Scotland (Rothwell and Charlton, 1998), but this study included cases with probable MS according to the Poser criteria. This may also explain why the median age at diagnosis in the Scottish study was 34 years, compared with 37 years in our study.
An increase in the incidence rates of MS has been reported in some studies (Larsen et al., 1984b; Wynn et al., 1989; Benedikz et al., 1994), whereas a fluctuating (Kinnunen, 1984; Wynn et al., 1990; Grønning et al., 1991; Midgard et al., 1991; Kurtzke et al., 1995; Edland et al., 1996; Sumelahti et al., 2000) or a decreasing incidence (Cook et al., 1985, 1988; Svenningsson et al., 1990) have been reported in others. In Denmark, a decrease in incidence in the period 1952-67 and a gradual increase since 1970 has been reported (Koch-Henriksen, 1999). This is comparable with the gradual increase seen in this study of Oslo since 1972.
All cases in our study were reviewed and classified by one neurologist, and MRI was not used in the diagnostic classification. An overestimation in the last decade (1986-99) caused by different usage of diagnostic criteria should therefore not contribute to the increase in incidence. Neither can the non-significant reduction in time from onset to diagnosis explain this increase. There has been an increase in foreign immigrants from areas with low MS prevalence to Oslo since 1970, and the second-generation immigrants are still too young to be at risk for developing MS. This results in an increase in the denominator in the incidence calculations. Hence the true increase in incidence in the population of Norwegian ancestry in Oslo might be even higher.
The increase in incidence rates in the present study was more marked for RR than PCP MS. This has also been reported in other studies from Norway (Grønning et al., 1991; Midgard et al., 1996). The difference is not explained by a change in time from onset to diagnosis, as the slightly more pronounced reduction in time from onset to diagnosis in the second (1986-99) compared with the first period (1972-85) for PCP cases compared with RR cases would reduce the difference in incidence rates. As observed by others (Kinnunen, 1984; Larsen et al., 1984b; Grønning and Mellgren, 1985; Wynn et al., 1990; Edland et al., 1996; Midgard et al., 1996; Koch-Henriksen, 1999), the increase in incidence was more marked in women. Both female sex and RR MS have in several studies been associated with a favourable course (Runmarker and Andersen, 1993; Brønnum-Hansen et al., 1994; Midgard et al., 1995; Weinshenker, 1996; Hawkins and McDonnell, 1999), and the increase in affected females and RR MS in our study could possibly reflect a trend towards more benign cases being diagnosed.
The decrease in the frequency of optic neuritis as onset symptom found in this study between the first (1972-85) and the second (1986-99) incidence subcohorts is similar to that reported in three other Norwegian studies (Larsen et al., 1984b; Grønning and Mellgren, 1985; Midgard et al., 1991). A possibly longer time from onset to diagnosis in patients with optic neuritis as onset symptom has been suggested as a possible explanation (Larsen et al., 1984b; Grønning and Mellgren, 1985; Midgard et al., 1991). A more probable explanation may be improved health care facilities and a general trend over the past three decades towards earlier physician and specialist consultation even for sensory disturbances and other less dramatic symptoms than visual impairment. Similarly, the decrease in frequency of abnormal evoked responses between the two periods may be due to the increased referral to such neurophysiological investigations in Oslo in the past decade, leading to such investigations being performed at an earlier stage of disease. An earlier referral to CSF studies may likewise be an explanation for the decreased frequency of OB in the CSF in the last period, as the agarose electrophoresis method used (Clausen et al., 1964) was the same throughout the study period.
In conclusion, the present prevalence data from Oslo support the earlier reports suggesting that the south-eastern part of Norway is a high-risk area for MS in this country. The increase in incidence from 1972 to 1996 is thought to reflect a true increase in disease incidence. The increase was more marked in RR than PCP MS and in females, possibly reflecting an increase in incidence of cases of benign MS.
This study was supported by grants from the Multiple Sclerosis Society of Norway and the Odd Fellow Research Fund for Multiple Sclerosis.
1 Benedikz J, Magnusson
H, Guthmundsson G (1994). Multiple sclerosis in Iceland, with observations
on the alleged epidemic in the Faroe Islands. Ann Neurol 36:S175-S179.
2 Brønnum-Hansen H, Koch-Henriksen N, Hyllested K (1994). Survival of patients with multiple sclerosis in Denmark: a nationwide, long-term epidemiologic survey. Neurology 44:1901-1907.
3 Clausen J, Matzke J, Gerhardt W (1964). Agar gel microelectrophoresis of proteins in the cerebrospinal fluid; normal and pathological findings. Acta Neurol Scand 40:49-56.
4 Cook SD, Cromarty JI, Tapp W, Poskanzer D, Walker JD, Dowling PC (1985). Declining incidence of multiple sclerosis in the Orkney Islands. Neurology 35:545-551.
5 Cook SD, MacDonald J, Tapp W, Poskanzer D, Dowling PC (1988). Multiple sclerosis in the Shetland Islands: an update. Acta Neurol Scand 77:148-151.
6 DeGraaf AS (1974). Multiple Sclerosis in northern Norway. Eur Neurol 11:281-295.
7 Edland A, Nyland H, Riise T, Larsen JP (1996). Epidemiology of multiple sclerosis in the county of Vestfold, eastern Norway: incidence and prevalence calculations. Acta Neurol Scand 93:104-109.
8 Ford HL, Gerry E, Airey CM, Vail A, Johnson MH, Williams DR (1998). The prevalence of multiple sclerosis in the Leeds Health Authority. J Neurol Neurosurg Psychiat 64:605-610.
9 Groebke-Lorenz W, Balint S, Frey UPet al. (1992). Multiple sclerosis in northwestern Switzerland. Schweiz Med Wochenschr 122:582-587.
10 Grønlie SA, Myrvoll E, Hansen G, Grønning M, Mellgren SI (2000). Multiple sclerosis in North Norway, and first appearance in an indigenous population. J Neurol 247:129-133.
11 Grønning M & Mellgren SI (1985). Multiple sclerosis in the two northernmost counties of Norway. Acta Neurol Scand 72:321-327.
12 Grønning M, Riise T, Kvale G, Nyland H, Larsen JP, Aarli JA (1991). Incidence of multiple sclerosis in Hordaland, western Norway: a fluctuating pattern. Neuroepidemiology 10:53-61.
13 Hawkins SA & McDonnell GV (1999). Benign multiple sclerosis? Clinical course, long term follow up, and assessment of prognostic factors. J Neurol Neurosurg Psychiat 67:148-152.
14 Kinnunen E (1984). Multiple sclerosis in Finland: evidence of increasing frequency and uneven geographic distribution. Neurology 34:457-461.
15 Koch-Henriksen N (1999). The Danish Multiple Sclerosis Registry: a 50-year follow-up. Multiple Sclerosis 5:293-296.
16 Kurtzke JF, Hyllested K, Heltberg A (1995). Multiple sclerosis in the Faroe Islands: transmission across four epidemics. Acta Neurol Scand 91:321-325.
17 Larsen JP, Aarli JA, Nyland H, Riise T (1984a). Western Norway, a high-risk area for multiple sclerosis: a prevalence/incidence study in the county of Hordaland. Neurology 34:1202-1207.
18 Larsen JP, Kvaale G, Riise T, Nyland H, Aarli JA (1984b). An increase in the incidence of multiple sclerosis in western Norway. Acta Neurol Scand 70:96-103.
19 Midgard R, Albrektsen G, Riise T, Kvale G, Nyland H (1995). Prognostic factors for survival in multiple sclerosis: a longitudinal, population based study in Møre and Romsdal, Norway. J Neurol Neurosurg Psychiat 58:417-421.
20 Midgard R, Riise T, Nyland H (1991). Epidemiologic trends in multiple sclerosis in Møre and Romsdal, Norway: a prevalence/incidence study in a stable population. Neurology 41:887-892.
21 Midgard R, Riise T, Svanes C, Kvale G, Nyland H (1996). Incidence of multiple sclerosis in Møre and Romsdal, Norway from 1950 to 1991. An age-period-cohort analysis. Brain 119:203-211.
22 Oftedal SI (1965). Multiple sclerosis in Vestfold county. Acta Neurol Scand Supplement 16:1-62.
23 Poser CM, Paty DW, Scheinberg Let al. (1983). New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol 13:227-231.
24 Presthus J (1960). Report on the multiple sclerosis investigations in West-Norway. Acta Psychiatr Neurol Scand 35:88-92.
25 Presthus J (1966). Multiple sclerosis in Møre og Romsdal County, Norway. Acta Neurol Scand 42:12-18.
26 Rice-Oxley M, Williams ES, Rees JE (1995). A prevalence survey of multiple sclerosis in Sussex. J Neurol Neurosurg Psychiat 58:27-30.
27 Robertson N, Deans J, Fraser M, Compston DA (1995). Multiple sclerosis in the north Cambridgeshire districts of East Anglia. J Neurol Neurosurg Psychiat 59:71-76.
28 Rothwell PM & Charlton D (1998). High incidence and prevalence of multiple sclerosis in south east Scotland: evidence of a genetic predisposition. J Neurol Neurosurg Psychiat 64:730-735.
29 Runmarker B & Andersen O (1993). Prognostic factors in a multiple sclerosis incidence cohort with twenty-five years of follow-up. Brain 116:117-134.
30 Statistisk årbok for Oslo (1996). Central Bureau of Statistics, Norway.
31 Statistisk årbok for Oslo (1999). Central Bureau of Statistics, Norway.
32 Sumelahti ML, Tienari PJ, Wikstrom J, Palo J, Hakama M (2000). Regional and temporal variation in the incidence of multiple sclerosis in Finland 1979-93. Neuroepidemiology 19:67-75.
33 Svenningsson A, Runmarker B, Lycke J, Andersen O (1990). Incidence of MS during two fifteen-year periods in the Gothenburg region of Sweden. Acta Neurol Scand 82:161-168.
34 Swank RL, Lerstad O, Strøm A, Backer J (1952). Multiple Sclerosis in rural Norway. New England J Med 246:721-728.
35 Weinshenker BG (1996). Epidemiology of multiple sclerosis. Neurol Clin 14:291-308.
36 Westlund K (1970). Distribution and mortality time trend of multiple sclerosis and some other diseases in Norway. Acta Neurol Scand 46:455-483.
37 Wynn DR, Rodriguez M, O'Fallon WM, Kurland LT (1989). Update on the epidemiology of multiple sclerosis. Mayo Clin Proc 64:808-817.
38 Wynn DR, Rodriguez M, O'Fallon WM, Kurland LT (1990). A reappraisal of the epidemiology of multiple sclerosis in Olmsted County, Minnesota. Neurology 40:780-786.