High incidence and prevalence of systemic lupus erythematosus in Norway

Abstract

Our aim was to identify all patients with systemic lupus erythematosus (SLE) within the city of Oslo from 1999–2008 and to estimate the incidence and prevalence of SLE according to age, sex and ethnicity. Adults (16 years and over) with SLE were identified from five different sources. Only patients fulfilling four or more of the updated 1997 American College of Rheumatology (ACR) criteria were included. The incidence was stable during the nine year study period, with a mean annual incidence rate of 3.0 per 100,000 at risk (95% confidence interval (CI) 2.4–3.5). Females exhibited a bimodal pattern in age specific incidence with the first peak at 16–29 years of age and the second at 50–59 years of age. The overall prevalence was 51.8 per 100,000 population (95% CI 45.2–58.4), with 91.0 (95% CI 78.8–103.1) for females and 10.7 (95% CI 6.4–15.0) for males. The prevalence proportions for European descendants were similar to those for Asians but statistically significant lower than those for individuals adopted from non-European countries. The findings indicate a higher prevalence in Norwegians compared to Caucasians in Denmark and England. The higher prevalence of SLE in foreign adopted individuals warrants further examination.

Keywords

Systemic lupus erythematosus epidemiology prevalence incidence

Introduction

The incidence of several autoimmune diseases typically exhibits geographic association with higher latitudes.^1–3 Whether or not this is true regarding systemic lupus erythematosus (SLE) is unclear. Prevalence and incidence figures of SLE show, however, considerable variation in different regions, and also between ethnic groups within a region.⁴

Studies undertaken in the United Kingdom and North America indicate that some ethnic groups, particularly Afro-Americans and Asians, are more susceptible to develop SLE than other ethnic groups.^5–12 Since an increased disease susceptibility has been shown in first generation immigrants, a genetic basis for the high risk is a likely explanation.¹³ On the other hand, this high prevalence has not been confirmed in studies undertaken in Asia and Africa,^14,15 which might suggest the involvement of either an environmental protective factor in the country they left or an aggravating factor present in the United Kingdom or North America that affects the Afro-American and Asian populations.

Studies of SLE patients of the same ethnicity who have grown up in different societies and studies on different ethnicities in the same environment may give us a better understanding of the complex interaction between genetic and environmental factors. There are no data on ethnic differences in prevalence and incidence from Scandinavian countries. Prevalence figures from the Scandinavian studies have earlier been considered to be quite accurate,⁴ ranging from 28.3 to 68 per 100,000.^16–18

The aim of the study was to describe annual incidence over a time period and prevalence figures of SLE in Oslo, the largest city of Norway, and to compare these figures with other studies in Northern Europe. Since Oslo has reliable census figures and a relatively large group of immigrants, the prevalence and incidence figures in separate ethnic groups have also been assessed.

Materials and methods

Study area

Oslo is the capital and the largest city in Norway with a population of 580,000, of whom 20% are immigrants of non-European origin (the majority from Pakistan, Somalia, Sri-Lanka and Iraq). Immigration from Asia started in 1970. Thus, the majority of immigrants are first generation and younger than the ethnic Norwegian population. Immigrants from Africa came later. International adoption started in 1970 and the majority of adopted children came from Asia.

Every newcomer who is planning to stay in Norway for more than three months is registered in a governmental database (the National Population Register) with a unique 11 digit number. All inhabitants are obliged to inform the population register within eight days of moving. All deaths are reported by doctors who are required to complete a death certificate and registered in the Norwegian Cause of Death Registry.

At the time of the study, there were six hospitals in Oslo. Two hospitals had a rheumatology service. Three hospitals had a nephrology service and one of these dealt with kidney transplantation exclusively. Since 2004 the rheumatology service for all SLE patients has been assigned to the Rheumatology Department at Oslo University Hospital, Rikshospitalet.

In Norway, most specialists are hospital-based, and at the time of the study, there were two rheumatologists in private practice in Oslo. The health care system in Norway is almost fully publicly funded with an all-encompassing insurance for all inhabitants.

Identification and validation of patients with systemic lupus erythematosus

Five independent sources were used to identify patients with diagnosed SLE in the study area. All hospitals in Oslo conducted a computerized search within all their departments for patients with SLE in the 10^th revision of the International Classification of Diseases (ICD-10) diagnosis code registry. We received lists containing inpatients and outpatients visits from 1 January 1999–1 January 2008 from all six hospitals. In addition, patients were identified from the SLE cohort at Diakonhjemmet Hospital established in 1995.¹⁹ Furthermore, the Norwegian systemic connective tissue disease and vasculitis registry (NOSVAR) at Oslo University Hospital (OUS) Rikshospitalet examined their database for patients with SLE and rheumatologists in private practice were asked to send data of their SLE patients. Finally, a list of patients registered with SLE as the cause of death, or underlying cause of death, was obtained from the Cause of Death registry. The results of the search for patients are displayed in Figure 1.

Only adult (≥16 years) patients living in Oslo at the time of incidence or prevalence registration were included. Two researchers (IMG, KL) evaluated and reached consensus on SLE diagnosis based on the evidence of multisystem disease, immunological abnormalities and the absence of a better diagnosis.²⁰ The updated American College of Rheumatology (ACR) classification criteria²¹ were subsequently used for classification, and only patients fulfilling four or more criteria were included in the study. The study was approved by the Regional Committee for Medical Research Ethics and the National Data Inspectorate.

Population base

Population figures in Oslo according to year, age and country background were retrieved from Statistics Norway.²² The adult population on 1 January 2008 was 459,535, including 347,826 ethnic Norwegians, 42,197 people of European descent other than Norwegian and 48,560 Asians. Statistics Norway did not have exact population figures for foreign adopted residents in Oslo as they consider them to be ethnic Norwegian and thereby register the adoptive child’s country of birth together with the adoptive parents’ country of birth. Thus 2191 adult residents (885 males and 1306 females) had been born in Asia, Africa or South/Central-America of Norwegian parents, or had been adopted as children from these countries. This merged group was used as a proxy for the non-European adopted population in Oslo.

Data collection

We used the ACR criteria defined according to the guidelines,²¹ and the year of the first ACR criterion (symptom debut), the year the patient received their diagnosis by a doctor, parents country of birth and year of immigration were collected from medical records. A questionnaire was mailed out to all patients to confirm the patient and the parent’s country of birth. Data on the year of immigration were missing in five of 35 cases and the year of first symptom in three of 116 incident cases. Demographic data was collected from The National Population Register and hospital records. For comparison with newer prevalence data from England,¹⁰ data on biopsy-proven lupus-nephritis was collected from medical records. Immunological tests were assessed as described elsewhere.²³

Definition of ethnicity

To match the population based figures from the Statistics Norway,²⁴ ethnicity was defined as the parents’ country of birth. If one parent was born in Norway, the ethnicity was defined as Norwegian. European descent was defined as all those except Asians, Africans and South- and Central Americans. Turkey was added to Asia. Russia was included in Europe. We defined the ethnicity of the adopted population according to the biological parents’ country of birth.

Statistical analysis

The annual incidence rate and point prevalence are reported per 100,000 at risk. Ninety-five percent confidence intervals (CIs) for rates were calculated with the approximation formula for standard error. Age standardization was done by the direct method of standardization²⁵ using the population of European descent in Oslo as the standard population. Differences in prevalence between groups were regarded as statistically significant when the confidence intervals did not overlap.

Results

The results of the search for patients are displayed in Figure 1.

Of the nine cases not assessed, one medical record was not found and eight patients had insufficient information for a definite diagnosis. All incident cases and all (except eight cases) prevalent cases in 2008 were found in the hospital database.

Incidence of SLE

There were 116 adult patients diagnosed with SLE during the nine year study period, giving a mean annual incidence rate of 3.0 per 100,000 at risk (95% CI 2.4–3.5). Of these, 101 patients were females (5.0 (95% CI 4.0–6.0)) and 15 were males (0.8 (95% CI 0.4–1.2)).

Incidence in ethnic subgroups

Eighty-eight patients were of European descent, of whom 79 Norwegians (incidence 2.6/year (95% CI 2.0–3.2)) and nine immigrants (incidence 2.7/year (95% CI 0.9–4.3)). Of 17 incident Asians (4.0/year (95% CI 1.9–6.0)), only two patients were born in Norway and only one had symptoms of SLE before immigration. The mean number of years in Norway before the onset of symptoms was eight years (range: 12–29). There were five incident cases of foreign adopted SLE patients, all under 39 years of age, giving a crude incidence of 33.6 (95% CI 4.2–63.0) and age-adjusted incidence of 25.4 (95% CI–0.2-50.9). No analysis of Africans (3) and South-Americans (3) was possible due to the low number of patients. Numbers of incident cases and annual incidence rates for the largest ethnic groups are shown in Table 1.

Table 1

Ethnic- and age-specific mean annual incidence of systemic lupus erythematosus (SLE) per 100,000 at risk, 1999–2008

Age group	All*		European		Asian + Turkish
	n	rate	n	Rate	n	Rate
16–19	8	4.6	6	4.6	0
20–29	35	4.5	21	3.3	9	9.2
30–39	26	2.9	19	2.5	4	3.9
40–49	14	2.2	11	2.0	2	2.8
50–59	21	3.8	19	3.8	2	5.5
60–69	6	1.7	6	1.9	0
70–79	5	1.8	5	1.9	0
80–89	1	0.5	1	0.5	0
90>	0		0		0
ALL	116	3.0	88	2.6	17	4.7
Age standardized**						4.0

ALL: crude rate.

All: European descendants, Asian+Turkish, adopted, African, South/Central American.

Age standardized to European descendant population in Oslo.

Age distribution in females

Females exhibited a bimodal pattern in distribution of age specific incidence (Figure 2). The peak incidence of 8.1 (95% CI 2.1–14) in females occurred between 16 and 19 years of age and reached a new peak (5.9 (95% CI 3.0–8.7)) between 50 and 59 years of age. The bimodal pattern was present for European descendants and Asian females separately.

Females of 16–39 years of age had a statistically significant higher incidence compared to females older than 60 years of age. Age distribution had a similar pattern in females when we used the first SLE symptom to determine incident cases.

Age distribution in males

Males had their peak incidence of 1.82 (95% CI 0.2–3.4) between 50 and 59 years of age (Figure 2). Asian and adopted males had peak incidence at a younger age (Data not shown). The age curve for incidence in European males flattened out with a small peak at 40–49 years of age when we used first symptom to determine the numerator. (Data not shown).

Prevalence

A total of 238 SLE patients fulfilling four or more criteria were residents in the study area on 1 January 2008. Of these, 196 (82%) were European descendants (181 Norwegians), 28 (12%) were Asians, three (1%) were Africans, four (2%) were from South- Central America, seven (3%) were adopted from Asia and Africa as children (of whom six were from Asia and one was from Africa). There were 214 females and 24 males. The median age at diagnosis was 32 years (range 9–83 years), 35 years for Norwegians and 26 years for Asians. The median interval between the first symptom and diagnosis was one year.

The overall prevalence per 100,000 January 1 2008 was 52.8 (95% CI 45.2–58.4), with 91.0 (95% CI 78.8–103.1) for females and 10.7 (95% CI 6.4–15.0) for males. The peak prevalence for female patients occurred between the ages of 60–69 years (153.2 (95% CI 103.2–203.2)) while males had the peak prevalence between the ages of 45–66 years (19.3 (95% CI 8.8–29.8)).

Ethnic groups

The prevalence proportions were 52.0 (95% CI 44.7–59.9) in the ethnic Norwegian population, 35.5 (95% CI 17.6–53.5) in the European descent immigrants, 59.7 (95% CI 38.0–81.5) in age-adjusted Asian population and 307.0 (95% CI 75.4–538.7) in age-adjusted foreign adopted population. The female/male ratio was 7.9 (95% CI 4.9–13.5) in Norwegians, 13.2 (95% CI 3.3–114.8) in Asians, and 1.7 (95% CI 0.2–12.2) in foreign adopted SLE patients. Ethnic- and age-specific prevalence rates for the largest ethnic groups are shown in Table 2.

Table 2

Ethnic- and age-specific prevalence of systemic lupus erythematosus (SLE) per 100,000 at risk, 1 January 2008

Age Group	All*		European		Asian + Turkish
Age Group	n	rate	n	rate	n	rate
16–19	4	18.1	3	18.0	1	25.5
20–29	33	35.0	20	26.6	7	51.9
30–39	52	48.2	41	46.2	9	70.6
40–49	54	69.6	33	52.4	8	80.8
50–59	49	79.0	46	83.8	3	57.6
60–69	40	87.2	38	88.7	0
70–79	11	41.9	11	43.5	0
80–89	2	10.2	2	10.4	0
90>	2	48.2	2	48.7	0
ALL	238	51.8	196	50.3	28	57.7
Age standardized**						59.7

ALL: crude rate.

All: European descendants, Asians+Turkish, adopted, African, South/Central American.

Age standardized to European descendant population in Oslo.

Prevalence of biopsy proven lupus nephritis

Of the 238 SLE patients alive on January 1 2008, 62 patients had lupus nephritis of whom 50 patients were biopsy-proven. Prevalence rates of biopsy-proven lupus nephritis were 9.2 (95% CI 6.2–12.2) in the ethnic European descent population and 18.5 (95% CI 6.4–30.6) in the Asian population

Hospital registry

Of the 575 people given the SLE ICD-10 code in a hospital in the time period 1999–2008, 448 people were alive and living in the capture area on January 1 2008. This gives a prevalence of people given the ICD-10 code as 1.8 times higher than the prevalence of SLE by criteria.

Temporality

Annual incidence rates and prevalence rates for the European descent population in the nine year study period are shown in Figure 3. Among European descendants, the incidence and prevalence appeared stable over time. Among Asians there appeared to be a trend of increasing age adjusted incidence from 4.4 in the first period to 6.3 in the second period.

Discussion

In the present report we present a relatively high, but stable incidence and prevalence of SLE from 1999 to 2008 among European descendants in Oslo.

This is the first study in Scandinavia to report the incidence and prevalence of SLE in separate ethnic groups. The biggest strength of our study is our comprehensive search for patients, incidence studied over several years and the reliable consensus figures of Norway.

The major limitation to our study is the likelihood of missing data, especially milder cases. Still, we believe that by collecting data from several sources we have managed to keep missing information for each ACR criteria below 10%. Incidence and prevalence studies in Sweden, Denmark and Northern Norway were conducted twice in the same area.^{16–18,26–28} They are therefore more likely to detect milder cases of SLE due to increased awareness of SLE. Additionally, they are less likely to have losses to follow-up, which might explain some of the increase in prevalence from the first to second prevalence study in these areas. Based upon our earlier study from a minor area of Oslo, where 8 of 93 (9%) patients were lost to follow-up in the hospital registry, we estimated a similar figure for the remaining patients in the present study. This would increase the prevalence from 51.8 to 54.4 per 100,000. The study of immigrants is hampered by the heterogenic culture and genetic background of the few Asian immigrants and the highly skewed age distribution.

The prevalence and incidence of SLE in Norwegians in the present study (52.0 and 2.6, respectively) are similar to the prevalence and incidence figures of populations in Norway’s Arctic region (44.9-–64.1 and 2.6–3.0, respectively).^16,26 The finding also corresponds to the prevalence in Southern Sweden (36.3–68)^17,28 although a higher incidence was found (4.5).

However, the prevalence of SLE in our study differs markedly from that found in the studies from Denmark and England.^8–12,18,27 The estimated prevalence and incidence in European descendants in our study is about twice as high as in Denmark, and even higher than the prevalence among Caucasians in England. Similar differences between the countries were found comparing SLE prevalence as defined by criteria,^8,9,12,18,27 ICD-10 code from hospitals¹¹ or prevalence of biopsy proven lupus nephritis.¹⁰ Therefore, we do not believe that the method of ascertainment of data can explain these differences. Although genetic differences cannot be discounted, it is unlikely that genetics would explain all of the differences in prevalence. The Scandinavian countries are genetically homogenous, and have close genetic ties to the United Kingdom.^29,30 Thus, in explaining the divergences, environmental factors should also be taken into consideration. Considering known environmental hypotheses in the development of SLE,³¹ the differing patterns of alcohol consumption are perhaps the most striking difference between these nations. Whereas Norway and Sweden have a tradition of weekend binge drinking, England and Denmark have a higher total alcohol consumption although more frequently and with small amounts.³² It is believed that moderate alcohol consumption has an anti-inflammatory effect while heavy consumption augments inflammation.^33,34 Binge drinking might therefore increase inflammation and, likewise, SLE incidence.

We found an annual incidence of 4 per 100,000 Asian immigrants at risk, which is 1.5 times higher than in European descents in Oslo. Unfortunately, there are few studies reporting incidence in Asians and only three from Asia (incidence from 0.9–3.1).¹⁴ Studies of incidence in Asian immigrants have been conducted in England but only calculated over a one year period and the incidence varied from 4.1–15.2.^8,9 Despite a probable healthy migrant effect in Oslo, the prevalence amongst Asians in Oslo is slightly above the prevalence range of most Asian countries (30–50)¹⁴ which may be due to methodological reasons (not capturing all cases in Asia) or due to increased early mortality in the individuals at risk. Lower incidence and prevalence in Asian countries may also be explained by an aggravating environmental factor in Oslo, for example, lower vitamin D level due to less sunlight,³⁵ or a protective factor in the country of origin, for example, a protective role of infections.³⁶ Due to the low number of cases in the present study, these explanations remain speculative only. Consequently, further studies of prevalence and incidence in Asians are warranted, particularly those that measure incidence over a period of time.

Interestingly, although there is a strikingly higher prevalence in European descendants in Norway compared to England, the prevalence in Asian immigrants in Oslo (57.7) is only slightly higher than Asian immigrants in England (40–50.4).^8–10,12 This may indicate that the causative factor does not affect Asian immigrants as much as European descendants in Norway and England. Possible explanations being genetic factors, environmental factors in early life or a social determined environmental factor not commonly adopted by Asian immigrants. A possible environmental factor is unlikely to be geographically determined as this would have the same effect across ethnicities.

The foreign adopted cases were grouped separately because we noticed a very high prevalence compared to the other groups (307.0). Even though this prevalence was statistically higher than the prevalence in European descents, the findings might be random because of the low number of cases. However, this is a new finding and it is tempting to speculate whether possible early environmental causes are responsible for the high occurrence.

In the present study, females of both European descent and Asian origin exhibit a bimodal pattern in age specific incidence (Figure 2). The first peak (16–29 years) coincides with an age in which many women use contraceptive pills in the capture area³⁷ and the second peak (50–59 years of age) with frequent use of menopausal hormone therapy.^38,39 This finding is in line with the findings of a large prospective cohort study, showing an increased risk of SLE with use of oral contraceptives and postmenopausal use of hormones.⁴⁰

Although using age at diagnosis to calculate incidence showed an increase in incidence for males in the age group 50–59 years, this peak disappeared when we used age of first symptom to determine incidence. We find age of first symptom more reliable as age of diagnosis is more dependent on other factors, such as when an individual tells their doctor about the symptoms and when the doctor thinks of SLE as a possible diagnosis. Thus, it appears that males have a constant incidence across the age groups, indicating that a possible environmental factor does not vary across ages.

In summary, this is the first study in Scandinavia to present prevalence and incidence by ethnic group and the first study to present incidence in Asian immigrants over a time period.

The findings may suggest the presence of an aggravating environmental factor in Norway which is not prevalent in Denmark or England, and is less likely to affect Asian immigrants as much as European descendants. On the other hand, there might also be a protective environmental factor in Denmark and England that may not be common in Norway. The finding of a very high incidence and prevalence in the foreign adopted population was interesting, but must be confirmed by other studies, involving more cases than the present study.

Footnotes

Funding

This work was supported by grants from the Oslo Sanitetsforenings Legacy and the Vivi Irene Hansens Legacy.

Conflicts of interest

None declared.

References

Vieira

Hart

Webster

. Association between residences in U.S. northern latitudes and rheumatoid arthritis: A spatial analysis of the Nurses’ Health Study. Environ Health Perspect 2010; 118: 957–961.

Geographic patterns of childhood insulin-dependent diabetes mellitus. Diabetes Epidemiology Research International Group. Diabetes 1988; 37: 1113–1119.

Simpson

Jr. Blizzard

Otahal

Van DM

Taylor

. Latitude is significantly associated with the prevalence of multiple sclerosis: a meta-analysis. J Neurol Neurosurg Psychiatry 2011; 82: 1132–1141.

Borchers

Naguwa

Shoenfeld

Gershwin

. The geoepidemiology of systemic lupus erythematosus. Autoimmun Rev 2010; 9: A277–A287.

McCarty

Manzi

Medsger

Jr. Ramsey-Goldman

LaPorte

Kwoh

. Incidence of systemic lupus erythematosus. Race and gender differences. Arthritis Rheum 1995; 38: 1260–1270.

Siegel

Lee

. The epidemiology of systemic lupus erythematosus. Semin Arthritis Rheum 1973; 3: 1–54.

Michet

Jr. McKenna

Elveback

Kaslow

Kurland

. Epidemiology of systemic lupus erythematosus and other connective tissue diseases in Rochester, Minnesota, 1950 through 1979. Mayo Clin Proc 1985; 60: 105–113.

Hopkinson

Doherty

Powell

. Clinical features and race-specific incidence/prevalence rates of systemic lupus erythematosus in a geographically complete cohort of patients. Ann Rheum Dis 1994; 53: 675–680.

Johnson

Gordon

Palmer

Bacon

. The prevalence and incidence of systemic lupus erythematosus in Birmingham, England. Relationship to ethnicity and country of birth. Arthritis Rheum 1995; 38: 551–558.

10.

Patel

Clarke

Bruce

Symmons

. The prevalence and incidence of biopsy-proven lupus nephritis in the UK: Evidence of an ethnic gradient. Arthritis Rheum 2006; 54: 2963–2969.

11.

Eaton

Pedersen

Atladottir

Gregory

Rose

Mortensen

. The prevalence of 30 ICD-10 autoimmune diseases in Denmark. Immunol Res 2010; 47: 228–231.

12.

Samanta

Roy

Feehally

Symmons

. The prevalence of diagnosed systemic lupus erythematosus in whites and Indian Asian immigrants in Leicester city, UK. Br J Rheumatol 1992; 31: 679–682.

13.

Molokhia

McKeigue

Cuadrado

Hughes

. Systemic lupus erythematosus in migrants from west Africa compared with Afro-Caribbean people in the UK. Lancet 2001; 357: 1414–1415.

14.

Osio-Salido

Manapat-Reyes

. Epidemiology of systemic lupus erythematosus in Asia. Lupus 2010; 19: 1365–1373.

15.

Bae

Fraser

Liang

. The epidemiology of systemic lupus erythematosus in populations of African ancestry: a critical review of the "prevalence gradient hypothesis". Arthritis Rheum 1998; 41: 2091–2099.

16.

Eilertsen

Becker-Merok

Nossent

. The influence of the 1997 updated classification criteria for systemic lupus erythematosus: epidemiology, disease presentation, and patient management. J Rheumatol 2009; 36: 552–559.

17.

Stahl-Hallengren

Jonsen

Nived

Sturfelt

. Incidence studies of systemic lupus erythematosus in Southern Sweden: increasing age, decreasing frequency of renal manifestations and good prognosis. J Rheumatol 2000; 27: 685–691.

18.

Laustrup

Voss

Green

Junker

. Occurrence of systemic lupus erythematosus in a Danish community: an 8-year prospective study. Scand J Rheumatol 2009; 38: 128–132.

19.

Gilboe

Husby

. Application of the 1982 revised criteria for the classification of systemic lupus erythematosus on a cohort of 346 Norwegian patients with connective tissue disease. Scand J Rheumatol 1999; 28: 81–87.

20.

21.

Hochberg

. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1997; 40: 1725–1725.

22.

23.

Lerang

Gilboe

Gran

. Differences between rheumatologists and other internists regarding diagnosis and treatment of systemic lupus erythematosus. Rheumatology (Oxford) 2012; 51: 663–669.

24.

25.

Hill

. Bradford Hill's principles of medical statistics, 12th ed. London: Edward Arnold; 1991, pp. 177–177.

26.

Nossent

. Systemic lupus erythematosus in the Arctic region of Norway. J Rheumatol 2001; 28: 539–546.

27.

Voss

Green

Junker

. Systemic lupus erythematosus in Denmark: clinical and epidemiological characterization of a county-based cohort. Scand J Rheumatol 1998; 27: 98–105.

28.

Nived

Sturfelt

Wollheim

. Systemic lupus erythematosus in an adult population in southern Sweden: incidence, prevalence and validity of ARA revised classification criteria. Br J Rheumatol 1985; 24: 147–154.

29.

30.

Novembre

Johnson

Bryc

. Genes mirror geography within Europe. Nature 2008; 456: 98–101.

31.

Wang

Pan

. Moderate alcohol drinking might be protective for systemic lupus erythematosus: a systematic review and meta-analysis. Clin Rheumatol 2008; 27: 1557–1563.

32.

Sieri

Agudo

Kesse

. Patterns of alcohol consumption in 10 European countries participating in the European Prospective Investigation into Cancer and Nutrition (EPIC) project. Public Health Nutr 2002; 5: 1287–1296.

33.

Imhof

Froehlich

Brenner

Boeing

Pepys

Koenig

. Effect of alcohol consumption on systemic markers of inflammation. Lancet 2001; 357: 763–767.

34.

Solomon

Costenbader

Keenan

Chibnik

Karlson

. Alcohol consumption and markers of inflammation in women with preclinical rheumatoid arthritis. Arthritis Rheum 2010; 62: 3554–3559.

35.

Gilkeson

James

Kamen

. The United States to Africa lupus prevalence gradient revisited. Lupus 2011; 20: 1095–1103.

36.

Okada

Kuhn

Feillet

Bach

. The 'hygiene hypothesis' for autoimmune and allergic diseases: an update. Clin Exp Immunol 2010; 160: 1–9.

37.

Skjeldestad

. [Use of contraceptives in Norway in 2005]. Tidsskr Nor Laegeforen 2007; 127: 2803–2805.

38.

Hofvind

Sakshaug

Ursin

Graff-Iversen

. Breast cancer incidence trends in Norway-explained by hormone therapy or mammographic screening?. Int J Cancer 2012; 130: 2930–2938.

39.

Lee

Ross

Pike

. An overview of menopausal oestrogen-progestin hormone therapy and breast cancer risk. Br J Cancer 2005; 92: 2049–2058.

40.

Costenbader

Feskanich

Stampfer

Karlson

. Reproductive and menopausal factors and risk of systemic lupus erythematosus in women. Arthritis Rheum 2007; 56: 1251–1262.