Risk of prostate cancer among Swedish-born and foreign-born men in Sweden, 1961–2004

Authors

  • Omid Beiki,

    Corresponding author
    1. Department of Public Health Sciences, Division of Social Medicine/Epidemiology, Karolinska Institutet, Stockholm, Sweden
    2. Kermanshah University of Medical Sciences, Kermanshah, Iran
    • Karolinska Institutet, Department of Public Health Sciences, Division of Social Medicine/Epidemiology, Norrbacka, Karolinska University Hospital, SE-171 76, Stockholm, Sweden
    Search for more papers by this author
    • Fax: +46-8-737-3840.

  • Anders Ekbom,

    1. Clinical Epidemiology Unit, Department of Medicine, Karolinska Institutet, Stockholm, Sweden
    Search for more papers by this author
  • Peter Allebeck,

    1. Department of Public Health Sciences, Division of Social Medicine/Epidemiology, Karolinska Institutet, Stockholm, Sweden
    2. Stockholm Centre for Public Health, Stockholm, Sweden
    Search for more papers by this author
  • Tahereh Moradi

    1. Department of Public Health Sciences, Division of Social Medicine/Epidemiology, Karolinska Institutet, Stockholm, Sweden
    2. Stockholm Centre for Public Health, Stockholm, Sweden
    Search for more papers by this author

Abstract

To elucidate the importance of environmental and genetic factors in prostate cancer etiology, we compared the risk of prostate cancer among foreign-born men to that of Swedish-born men in Sweden and to that in the country of origin. We estimated rate ratios (RRs) with 95% confidence intervals (CIs) adjusted for age, calendar period of year and education using Poisson regression in a cohort of 3.8 million men aged 45 years and older between 1961 and 2004. During the 45 years of follow-up, 8,244 and 187,675 cases of prostate cancer occurred among foreign-born and Swedish-born men, respectively. Overall, foreign-born men had a significantly 40% decreased risk of prostate cancer compared to Swedish-born men (RR = 0.62, 95% CI = 0.61–0.63). Men born in Middle Africa and in the Caribbean had an increased risk (RR = 1.89, 95% CI = 0.95–3.78 and RR = 1.24, 95% CI = 0.71–2.19, respectively). The overall risk in both strata of duration of residence or age at immigration was lower among immigrants compared to Swedish-born men. After additional adjustment for birthplace and age at immigration, although the risk remained lower among immigrants compared to Swedish-born, but it was increased among immigrants who stayed 35 years and longer compared to those who stayed shorter (RR = 1.33, 95% CI = 1.21–1.46). Both environmental and genetic factors seem to be involved in the etiology of prostate cancer. Duration of residence was an important factor affecting the risk among immigrants. Studies focusing on the etiology of prostate cancer specifically in African immigrants and their descendants and increasing preventive and diagnostic activities on old immigrants are recommended. © 2008 Wiley-Liss, Inc.

Prostate cancer is the most common cancer diagnosed among men in most western countries.1 About 75% of patients are diagnosed after 65 years.2

The incidence of prostate cancer varies considerably between countries and between different ethnic groups. The highest incidence rates have been observed in North America especially among African-American, Australia/New Zealand and Scandinavia, whereas the lowest rates have been observed in Asia, particularly in India, China and Japan.3 However, there is a general increase in incidence in most countries, with low as well as high incidence rates.4–6

Studies on migrants that have focused either on comparing the risks in country of origin and host country, or on studying risk in relation to age at immigration and duration of residence are approaches to explore the importance of environmental and genetic factors in the etiology of most cancers.7 Results from previous migrant studies on prostate cancer have been inconsistent. Several studies have shown that the incidence rate of prostate cancer in migrants is on a level between the incidence in the countries of origin and the incidence in the new country of residence,8 whereas others have shown that the rates among migrants are similar to natives in the host country.9

In Sweden, the high proportion of foreign-born residents, 12.9% in 2006,10 and the existence of an established system of population-based registers containing both demographic and medical information, provide a unique opportunity to conduct migrant studies. The aim of this study was to compare the risk of prostate cancer among foreign-born men to that of Swedish-born men, and to compare the incidence rate of prostate cancer among foreign-born men in Sweden to that in their country of origin.

Material and methods

Study cohort

The study cohort consisted of 2 groups, the foreign-born and the Swedish-born. The foreign-born group consisted of 372,663 men aged 45 years and older, and the Swedish-born group consisted of 3,385,160 same age men, whose both parents were born in Sweden. All cohort members were free of prostate cancer at the start of follow-up. The follow-up started in January 1, 1961, or birth date or first immigration date, whichever occurred last. Information about the place of birth was not available for 1,617 individuals and thus excluded from the study cohort.

Data on highest acquired level of education were obtained from the National Population and Housing Census, and also from the Longitudinal integration database for health insurance and labour market studies 1990–2003.11 Data on the birthplace of parents of the cohort members were obtained through linkage to the Multigeneration register.12 This register provides parental information on all Swedish inhabitants born after 1931, and who were alive in 1960.12

Follow-up

All Swedish residents can be identified by their 10-digit national registration number.13 The linkages have been completed by Statistics Sweden and Centre for Epidemiology at the National Board of Health and Welfare by using this unique identifier. To ensure confidentiality, the national registration numbers were replaced by serial numbers through Statistics Sweden. The study was approved by the Regional Board of The Ethical Committee, Stockholm.

The cohort members were followed until diagnosis of cancer (ICD-7 code: 177 Malignant Neoplasm of Prostate), first emigration, death, or end of follow-up (December 31, 2004), whichever occurred first. Follow-up was achieved through linkage with the following databases: (i) The Cancer Register containing data on all cases of cancer since 1958. The completeness of cancer registration and the percentage of cytologically or histologically verified cases is considered to be close to 100%14, 15; (ii) The Total Population Register11, 16 and (iii) The Cause of Death Register, which contains information on date of death and underlying cause of death since 1952. Present completeness of this register is estimated to be 99%.11, 16

Classification of the country of birth

We classified foreign-born individuals by 6 continents, which were further subdivided into 19 world regions, as defined by the United Nations Population Division:

  • Africa (Eastern, Middle, Northern, Southern and Western Africa)

  • Asia (Eastern, South-Central, South-Eastern and Western Asia)

  • Europe (Eastern, Northern, Southern and Western Europe)

  • Latin America (Caribbean, Central America and South America)

  • Northern America

  • Oceania (Australia/New Zealand, Melanesia, Micronesia/Polynesia)

We report information for all continents and regions, except for subgroups of Oceania because of the limited number of prostate cancer cases. We also report individual countries having 10 or more cases of prostate cancer.

Statistical methods

We analyzed data in a grouped form, with foreign-born men as the main explanatory variable. The analysis was performed in 2 stages.

We first compared the risk of prostate cancer between foreign-born and Swedish-born men. We used Poisson models estimated by the maximum likelihood method and rate ratios (RRs) were computed with 95% confidence intervals (CIs). Adjustments were made for attained age (age at follow-up) in 5-year intervals (45–49; 50–54; 55–59… 115–119; 120 years or more), calendar period of follow-up (3 categories: 1961–75; 1976–90; 1991–2004), and number of years of education as an indicator of socio-economic position (SEP; 0–9; 10–12; 13 years or more). In addition, we performed the analysis stratified by attained age (45–54; 55–64; 65–74; 75–84; 85 years or more). Furthermore, we additionally adjusted the RRs based on occupational SEP obtained from censuses 1980, 1985 and 1990 classified as manual workers, nonmanual workers and employers, group with missing information on SEP and others (including pensioners, students, part-time workers and employees without information on occupation). In an attempt to study the possible influence of lifestyle and environmental exposures, we stratified the immigrants by their age at immigration (<40 and ≥40 years) and duration of residence in Sweden (<35 and ≥35 years). This analysis was limited to only foreign-born men with a known date of immigration. We repeated this stratified analysis adjusting for birth country in addition to attained age, calendar period of follow up and years of education. These analyses were also repeated with additional adjustment for age at immigration and duration of residence when appropriate. Because a cancer diagnosis shortly after immigration may indicate selection of prevalent cases, we repeated the main analysis after exclusion of subjects with diagnoses occurring within 1 and 2 years after immigration to Sweden. The results did not differ and thus are not presented.

Secondly, to compare cancer rates among immigrants in Sweden with that from countries of origin, in our cohort we calculated age-standardized rates (ASRs) between 1963 and 2002 by 4-year intervals (1963–1967; 1968–1972; 1973–1977; 1978–1982; 1983–1987; 1988–1992; 1993–1997; 1998–2002) using the world population as standard.1 To prevent high variability in ASRs, we selected only countries with at least 5 cases in each 5 consecutive periods. We also obtained standardized rates, age-adjusted to the world population, for prostate cancer during these years in the countries of origin. We retrieved these rates from volumes II–IX of Cancer Incidence in 5 Continents from the IARC.1, 3 The criteria used to select countries were the availability of population-based data from 5 consecutive volumes of IARC publications. Where ASR for the entire population of the country was not reported, we pooled data from subpopulations of that country. A final selection of 9 countries including Denmark, Estonia, Finland, Germany, Latvia, Norway, Poland, UK and USA was made.

The Joinpoint regression models were used to find the best-fit line for trends of prostate cancer incidence rates.17, 18 This analysis involves fitting a series of joined straight lines on a log scale to the trends in the age-adjusted rates. Line segments were joined at points called joinpoints. Each joinpoint denotes a statistically significant change in trend. Once the line segments are established, the estimated annual percent change was used to describe and test the statistical significance of the trends. The null hypothesis in this analysis is that the trend in incidence rates is neither increasing nor decreasing.

Results

During the 45 years of follow up, 8,244 cases of prostate cancer were diagnosed among 372,663 foreign-born men older than 45 years, contributing with 6.3 million person years at risk. During the same period, 187,675 cases of prostate cancer were found among 3.4 million Swedish-born same age men, corresponding to more than 63 million person years of follow up (Tables I and II). Around 80% of the foreign-born men emigrated from European countries; mostly from Northern Europe, and mainly from Finland, Denmark and Norway.

Table I. Population Size, Age at Exit, Age at Immigration1 and Duration of Residence1 Among Foreign-Born and Swedish-Born Men 45 Years and Older by Birth Region/Country, 1961–2004
Birth region/countryNo. of menAge at exitAge at immigrationDuration of residence
Mean (±SD)Mean (±SD)Mean (±SD)
  • 1

    Only among immigrants with known date of immigration.

Foreign-born men372,66363.1 (±12.0)32.0 (±14.5)27.8 (±15.2)
Africa10,59855.1 (±8.6)33.0 (±10.2)21.0 (±9.8)
 Eastern Africa4,34153.8 (±7.9)35.7 (±10.7)17.7 (±8.7)
 Middle Africa32154.1 (±9.0)34.9 (±11.1)17.3 (±11.3)
 Northern Africa4,18456.2 (±8.8)30.2 (±9.0)24.7 (±9.5)
  Morocco1,44356.7 (±8.3)28.7 (±8.3)26.7 (±9.2)
 Southern Africa43861.6 (±12.5)31.1 (±13.7)25.6 (±13.6)
 Western Africa1,31453.7 (±6.9)32.4 (±7.9)20.6 (±8.6)
Asia46,01856.4 (±10.4)38.1 (±13.1)17.6 (±9.0)
 Eastern Asia3,07259.8 (±12.0)40.6 (±13.3)17.6 (±11.5)
  China1,97161.2 (±12.9)42.7 (±13.2)16.5 (±10.3)
 South-Central Asia15,79855.3 (±9.7)37.4 (±12.2)17.5 (±7.5)
  India1,69259.2 (±10.7)33.4 (±12.5)24.0 (±10.6)
  Iran11,56955.0 (±9.7)38.0 (±12.2)16.8 (±6.1)
 South-Eastern Asia2,72456.1 (±9.8)35.6 (±11.4)19.9 (±8.4)
 Western Asia24,42456.6 (±10.5)38.7 (±13.6)17.4 (±9.5)
  Iraq9,26254.3 (±8.7)42.4 (±11.6)11.8 (±6.8)
  Lebanon2,77754.6 (±9.1)36.7 (±11.4)17.7 (±7.1)
  Syria2,80156.4 (±10.0)39.0 (±12.3)17.1 (±7.2)
  Turkey7,94759.9 (±12.2)35.2 (±16.0)23.7 (±9.3)
Europe295,90364.6 (±11.8)30.1 (±14.6)31.3 (±15.5)
 Eastern Europe33,49665.2 (±12.5)33.4 (±14.0)28.9 (±14.5)
  Bulgaria1,01060.7 (±11.8)37.9 (±13.2)21.4 (±11.8)
  Czechoslovakia4,29666.8 (±12.0)29.7 (±11.9)34.4 (±13.8)
  Hungary8,43965.3 (±10.9)27.0 (±11.2)36.6 (±12.0)
  Poland10,87163.5 (±12.9)34.4 (±13.5)26.5 (±12.5)
  Romania3,22361.7 (±12.2)41.2 (±13.5)19.2 (±9.9)
  Soviet Union4,87571.4 (±12.5)34.8 (±16.3)31.7 (±19.1)
 Northern Europe175,07265.0 (±11.8)27.7 (±14.3)34.0 (±15.1)
  Denmark29,92867.6 (±12.4)30.8 (±14.8)33.5 (±17.3)
  Estonia10,58473.8 (±10.6)23.9 (±14.1)49.7 (±14.6)
  Finland99,97862.9 (±10.8)25.0 (±13.0)35.9 (±12.9)
  Iceland1,13057.8 (±10.7)37.0 (±11.9)19.1 (±12.5)
  Latvia1,94374.7 (±11.1)25.6 (±14.1)47.9 (±14.8)
  Lithuania24072.1 (±12.3)36.5 (±15.6)32.4 (±21.9)
  Norway24,98767.5 (±12.4)35.1 (±17.0)27.7 (±18.6)
  UK6,03159.5 (±11.5)34.5 (±12.3)21.9 (±12.9)
 Southern Europe52,99661.2 (±10.6)35.3 (±14.3)23.9 (±13.7)
  Bosnia10,20158.2 (±10.3)47.8 (±11.7)10.3 (±4.9)
  Croatia1,51359.6 (±8.6)30.1 (±12.2)29.2 (±12.9)
  Greece7,43061.8 (±9.9)30.2 (±11.2)28.6 (±10.2)
  Italy6,18367.9 (±12.0)27.2 (±11.8)36.3 (±13.6)
  Portugal1,13860.9 (±10.1)29.8 (±10.7)28.6 (±9.6)
  Spain3,07462.7 (±10.1)29.2 (±10.4)30.1 (±11.1)
  Yugoslavia22,36160.6 (±10.0)32.6 (±13.1)26.7 (±12.9)
 Western Europe34,33967.4 (±11.7)27.0 (±13.8)36.6 (±15.6)
  Austria4,92766.8 (±11.1)22.6 (±11.1)41.1 (±11.1)
  Belgium43465.8 (±13.2)33.3 (±16.0)26.1 (±17.4)
  France1,91063.1 (±12.4)30.1 (±12.5)28.1 (±15.0)
  Germany23,08868.1 (±11.7)26.9 (±14.3)37.4 (±15.7)
  Netherlands2,52365.0 (±12.1)31.4 (±13.0)30.3 (±17.2)
  Switzerland1,43967.1 (±11.1)27.9 (±12.5)36.0 (±15.4)
Latin America10,75156.6 (±9.7)36.2 (±12.0)19.8 (±8.5)
 Caribbean43857.9 (±10.6)37.8 (±11.0)18.0 (±11.4)
 Central America67155.4 (±9.9)35.8 (±11.9)18.5 (±9.1)
 Southern America9,64256.7 (±9.6)36.2 (±12.0)19.9 (±8.3)
  Argentina75358.9 (±10.7)33.0 (±13.0)23.7 (±11.4)
  Chile5,69356.5 (±9.5)37.2 (±11.4)19.2 (±6.6)
Northern America8,88567.4 (±13.8)34.8 (±16.5)22.4 (±17.0)
  Canada69961.6 (±12.7)29.5 (±16.9)25.3 (±18.4)
  USA8,17367.9 (±13.8)35.4 (±16.4)22.1 (±16.8)
Oceania50855.8 (±10.6)35.1 (±13.2)17.8 (±13.6)
Table II. Rate Ratio (RR)1 and 95% Confidence Interval (CI) of Prostate Cancer Among Foreign-Born Men 45 Years and Older in Sweden by Birth Region/Country, 1961–2004
Birth region/countryCasesPYRSRR2 (95% CI)RR3 (95% CI)
  • 1

    Swedish-born men are the reference group. PYRS, Person-years.

  • 2

    Adjusted for age and calendar year of follow-up.

  • 3

    Adjusted for age, calendar year of follow-up and education.

Foreign-born men8,2446,299,0260.57 (0.56–0.58)0.62 (0.61–0.63)
Africa78105,5540.58 (0.47–0.73)0.63 (0.50–0.78)
 Eastern Africa1738,0280.40 (0.25–0.64)0.45 (0.28–0.72)
 Middle Africa82,9112.00 (1.00–4.00)1.89 (0.95–3.78)
 Northern Africa3446,5410.54 (0.39–0.76)0.59 (0.42–0.82)
  Morocco1216,8670.56 (0.32–0.98)0.64 (0.36–1.13)
 Southern Africa96,6210.60 (0.31–1.16)0.59 (0.31–1.14)
 Western Africa1011,4531.01 (0.55–1.88)1.03 (0.55–1.92)
Asia319514,1690.36 (0.32–0.40)0.41 (0.37–0.46)
 Eastern Asia3044,0100.32 (0.22–0.45)0.34 (0.24–0.48)
  China2430,5570.32 (0.22–0.48)0.35 (0.23–0.52)
 South-Central Asia117161,1520.46 (0.38–0.55)0.51 (0.43–0.61)
  India2623,1430.60 (0.41–0.88)0.62 (0.42–0.91)
  Iran81114,9610.44 (0.36–0.55)0.50 (0.40–0.62)
 South-Eastern Asia1530,0740.31 (0.19–0.51)0.33 (0.20–0.55)
 Western Asia157278,9330.32 (0.27–0.37)0.37 (0.32–0.44)
  Iraq4585,5490.39 (0.29–0.52)0.48 (0.36–0.65)
  Lebanon1226,5890.32 (0.18–0.56)0.37 (0.21–0.65)
  Syria2031,7280.38 (0.24–0.58)0.46 (0.30–0.71)
  Turkey66114,0850.26 (0.20–0.33)0.30 (0.23–0.38)
Europe7,3465,390,4250.57 (0.56–0.59)0.62 (0.61–0.64)
 Eastern Europe903615,5050.58 (0.54–0.62)0.58 (0.54–0.62)
  Bulgaria1015,3590.31 (0.17–0.57)0.32 (0.17–0.60)
  Czechoslovakia15286,3490.67 (0.57–0.79)0.66 (0.56–0.77)
  Hungary214164,5050.56 (0.49–0.64)0.55 (0.48–0.63)
  Poland262186,2730.57 (0.50–0.64)0.57 (0.51–0.64)
  Romania5352,1310.44 (0.34–0.58)0.46 (0.35–0.61)
  Soviet Union202100,2780.64 (0.56–0.74)0.65 (0.57–0.75)
 Northern Europe4,8303,240,1050.62 (0.60–0.64)0.67 (0.65–0.69)
  Denmark951626,6400.55 (0.52–0.59)0.60 (0.56–0.64)
  Estonia701242,9150.90 (0.83–0.96)0.89 (0.82–0.96)
  Finland2,0301,712,1450.57 (0.55–0.60)0.63 (0.60–0.65)
  Iceland1014,2720.38 (0.21–0.71)0.41 (0.22–0.75)
  Latvia13144,9890.85 (0.72–1.01)0.84 (0.71–1.00)
  Lithuania185,3211.02 (0.64–1.62)1.00 (0.63–1.59)
  Norway882508,1080.63 (0.59–0.68)0.68 (0.63–0.72)
  UK10482,9860.63 (0.52–0.76)0.66 (0.55–0.80)
 Southern Europe495847,0020.29 (0.27–0.32)0.33 (0.30–0.36)
  Bosnia75133,7690.31 (0.25–0.39)0.40 (0.31–0.50)
  Croatia2022,1490.59 (0.38–0.91)0.61 (0.40–0.95)
  Greece45122,9370.19 (0.14–0.26)0.22 (0.16–0.29)
  Italy103135,7430.28 (0.23–0.33)0.31 (0.25–0.37)
  Portugal1917,9200.56 (0.36–0.87)0.60 (0.38–0.94)
  Spain3453,6170.31 (0.22–0.43)0.35 (0.25–0.49)
  Yugoslavia183344,2560.29 (0.25–0.34)0.31 (0.27–0.36)
 Western Europe1,118687,8120.62 (0.59–0.66)0.64 (0.61–0.68)
  Austria12098,9890.49 (0.41–0.59)0.50 (0.41–0.59)
  Belgium138,3110.59 (0.34–1.01)0.61 (0.36–1.06)
  France5232,4880.67 (0.51–0.88)0.69 (0.53–0.91)
  Germany821469,0720.66 (0.62–0.71)0.67 (0.63–0.72)
  Netherlands6948,1300.57 (0.45–0.73)0.59 (0.47–0.75)
  Switzerland4330,4340.54 (0.40–0.73)0.61 (0.45–0.82)
Latin America112123,3780.59 (0.49–0.71)0.63 (0.52–0.76)
 Caribbean125,5871.20 (0.68–2.12)1.24 (0.71–2.19)
 Central America76,8050.65 (0.31–1.36)0.70 (0.34–1.48)
 Southern America93110,9860.55 (0.45–0.67)0.58 (0.48–0.72)
  Argentina1310,2090.69 (0.40–1.19)0.72 (0.42–1.24)
  Chile5265,0410.53 (0.41–0.70)0.57 (0.44–0.75)
Northern America382160,3400.79 (0.72–0.88)0.81 (0.73–0.90)
  Canada2111,0840.78 (0.51–1.19)0.80 (0.52–1.23)
  USA361149,1510.79 (0.72–0.88)0.81 (0.73–0.90)
Oceania75,1600.79 (0.37–1.65)0.86 (0.41–1.81)

Age at exit was on average lower among foreign-born men (63.1, SD ± 12) compared to Swedish-born men (69.5, SD ± 12.9). The exception was among immigrants from Latvia (74.7, SD ± 11.1), Estonia (73.8, SD ± 10.6), Lithuania (72.1, SD ± 12.3) and Soviet Union (71.4, SD ± 12.5) whose age at exit was higher than Swedish-born men (Table I).

Comparison of overall risk and by attained age

In general, foreign-born men had a 40% decreased risk of prostate cancer compared to Swedish-born men (RR = 0.62, 95% CI = 0.61–0.63; Table II). Analyses by continents showed a statistically significantly reduced risk among immigrants from all continents, except for Oceania where the reduced risk was not significant (RR = 0.86, 95% CI = 0.41–1.81).

At the level of World regions, immigrants from all regions had a lower risk of prostate cancer compared to Swedish-born men. The exceptions were men from Western Africa who had similar risk as Swedish-born men (RR = 1.03, 95% CI = 0.55–1.92), men born in the Caribbean who had 20% nonsignificant elevated risk (RR = 1.24, 95% CI = 0.71–2.19) and men born in Middle Africa with a borderline 90% increased risk of prostate cancer (RR = 1.89, 95% CI = 0.95–3.78).

At the country level, immigrants from all 39 out of 198 countries of origin that had more than 10 cases of prostate cancer showed a decreased risk, except for Lithuania which had risk levels similar to Swedish-born men (Table II). The reduced risk was not statistically significant for immigrants from Morocco, Latvia, Belgium, Argentina and Canada. In addition, the risk among immigrants from Democratic Republic of the Congo who constituted the majority of immigrants from Middle Africa was about 3 times higher than men born in Sweden (6 cases, RR = 2.84, 95% CI = 1.27–6.31; data not shown in the tables).

We did not find any variation in risk of prostate cancer by attained age (Fig. 1). Overall, the risk was reduced by 35–50% in all age strata. Adjustment for education did not change the risk of prostate cancer.

Figure 1.

Rate ratio (RR) and 95% confidence interval (CI) of prostate cancer adjusted for age, calendar year of follow-up and education among foreign-born men 45 years and older by attained age, 1961–2004. Swedish-born men are the reference group. Data from Oceania has been deleted due to high variability.

Because the analyses by age at immigration and duration of residence were confined to a subcohort with known date of immigration available for only 65% of immigrants, we repeated the above analyses in this subcohort. The results were similar to the results from the complete cohort (Table III).

Table III. Rate Ratio (RR)1 and 95% Confidence Interval (CI) of Prostate Cancer Among Foreign-Born Men 45 and Older with Known Date at Immigration in Sweden, 1961–2004
Birth region/countryCasesPYRSRR2 (95% CI)RR3 (95% CI)
  • 1

    Swedish-born men are the reference group. PYRS = Person-years.

  • 2

    Adjusted for age and calendar year of follow-up.

  • 3

    Adjusted for age, calendar year of follow-up and education.

Foreign-born men4,1203,112,7130.64 (0.62–0.67)0.66 (0.64–0.68)
Africa5375,4690.65 (0.50–0.85)0.68 (0.52–0.89)
 Eastern Africa1028,3010.33 (0.18–0.62)0.36 (0.20–0.68)
 Middle Africa51,6223.34 (1.39–8.02)3.13 (1.30–7.53)
 Northern Africa2533,6370.64 (0.43–0.95)0.66 (0.45–0.98)
  Morocco1112,0760.80 (0.44–1.44)0.87 (0.48–1.58)
 Southern Africa52,7001.19 (0.50–2.86)1.10 (0.46–2.65)
 Western Africa89,2091.20 (0.60–2.39)1.18 (0.59–2.36)
Asia266338,2350.43 (0.38–0.49)0.48 (0.43–0.55)
 Eastern Asia1424,9640.25 (0.15–0.42)0.27 (0.16–0.45)
  China1016,7060.23 (0.12–0.42)0.25 (0.14–0.47)
 South-Central Asia101112,0860.54 (0.45–0.66)0.59 (0.49–0.72)
  India1514,7670.60 (0.36–0.99)0.60 (0.36–1.00)
  Iran7781,2710.54 (0.43–0.68)0.60 (0.48–0.75)
 South-Eastern Asia1222,4750.32 (0.18–0.57)0.35 (0.20–0.61)
 Western Asia139178,7100.41 (0.35–0.49)0.47 (0.40–0.56)
  Iraq4349,8840.57 (0.42–0.77)0.68 (0.50–0.91)
  Lebanon1120,2740.35 (0.20–0.64)0.41 (0.22–0.73)
  Syria2022,4290.48 (0.31–0.75)0.58 (0.37–0.89)
  Turkey5572,2290.33 (0.25–0.43)0.37 (0.28–0.48)
Europe3,6552,574,6000.67 (0.64–0.69)0.68 (0.65–0.70)
 Eastern Europe518349,9590.61 (0.56–0.67)0.60 (0.55–0.66)
  Bulgaria89,9610.40 (0.20–0.79)0.40 (0.20–0.80)
  Czechoslovakia9552,6560.76 (0.62–0.92)0.71 (0.58–0.87)
  Hungary145105,8890.63 (0.54–0.74)0.60 (0.51–0.71)
  Poland165111,6920.61 (0.52–0.71)0.61 (0.52–0.71)
  Romania4332,9680.53 (0.39–0.71)0.55 (0.41–0.75)
  Soviet Union5432,8200.53 (0.41–0.69)0.55 (0.42–0.71)
 Northern Europe2,2611,459,0320.73 (0.70–0.76)0.74 (0.71–0.78)
  Denmark459260,9610.68 (0.62–0.75)0.70 (0.63–0.76)
  Estonia12244,2990.82 (0.69–0.98)0.80 (0.67–0.96)
  Finland1,279932,2060.73 (0.69–0.77)0.75 (0.71–0.79)
  Iceland87,3040.68 (0.34–1.37)0.66 (0.33–1.31)
  Latvia1910,8750.51 (0.33–0.81)0.50 (0.32–0.78)
  Lithuania21,7600.32 (0.08–1.28)0.31 (0.08–1.23)
  Norway312155,6190.79 (0.71–0.88)0.80 (0.72–0.90)
  UK5844,3340.85 (0.66–1.10)0.84 (0.65–1.09)
 Southern Europe387465,7540.45 (0.40–0.49)0.48 (0.43–0.53)
  Bosnia6867,4480.41 (0.32–0.52)0.52 (0.41–0.66)
  Croatia1718,5020.60 (0.37–0.96)0.62 (0.38–0.99)
  Greece2960,8780.31 (0.21–0.44)0.33 (0.23–0.47)
  Italy6848,4020.63 (0.50–0.80)0.63 (0.50–0.80)
  Portugal1610,8760.96 (0.59–1.56)0.99 (0.61–1.62)
  Spain2425,0400.58 (0.39–0.86)0.58 (0.39–0.87)
  Yugoslavia150222,0590.38 (0.33–0.45)0.40 (0.34–0.47)
 Western Europe489299,8550.73 (0.67–0.80)0.70 (0.64–0.76)
  Austria7246,7090.73 (0.58–0.92)0.69 (0.55–0.87)
  Belgium12,6020.18 (0.03–1.26)0.17 (0.02–1.22)
  France2113,9120.82 (0.53–1.26)0.79 (0.51–1.20)
  Germany336204,0220.73 (0.65–0.81)0.69 (0.62–0.77)
  Netherlands3921,6290.79 (0.58–1.08)0.75 (0.55–1.03)
  Switzerland2010,8920.75 (0.49–1.17)0.72 (0.46–1.12)
Latin America8890,4060.64 (0.52–0.78)0.67 (0.54–0.83)
 Caribbean83,3531.53 (0.77–3.07)1.52 (0.76–3.05)
 Central America54,3490.75 (0.31–1.79)0.80 (0.33–1.92)
 Southern America7582,7040.59 (0.47–0.74)0.63 (0.50–0.78)
  Argentina76,0710.75 (0.36–1.57)0.74 (0.35–1.55)
  Chile4651,7050.56 (0.42–0.75)0.60 (0.45–0.80)
Northern America5331,7060.85 (0.65–1.11)0.84 (0.64–1.10)
  Canada52,8360.98 (0.41–2.36)0.96 (0.40–2.31)
  USA4828,7910.84 (0.63–1.11)0.83 (0.62–1.10)
Oceania52,2961.86 (0.77–4.47)1.71 (0.71–4.11)

Effect of age at immigration or duration of residence

Overall, the risk adjusted for attained age, calendar period of year and education was lower in both strata of age at immigration compared to the Swedish-born men (Table IV). The risk among men born in Middle Africa who were older than 40 years at immigration was about 8 times higher than that for Swedes (RR = 7.70, 95% CI = 3.21–18.51). In contrast, men born in the Caribbean were at increased risk if they were younger than 40 years at immigration (RR = 2.53, 95% CI = 1.14–5.63). At country level, men born in Canada who immigrated at ages older than 40 had 3 times higher risk than that for Swedes (RR = 3.27, 95% CI = 1.05–10.14).

Table IV. Rate Ratio (RR)12 and 95% Confidence Interval (CI) of Prostate Cancer Among Foreign-Born Men 45 Years and Older by Age at Immigration and Country of Birth, 1961–2004
Birth region/country<40≥40
RR (95% CI)RR (95% CI)
  • 1

    Swedish-born men are the reference group.

  • 2

    Adjusted for age, calendar period of follow up and education.

Foreign-born men0.71 (0.69–0.74)0.54 (0.50–0.57)
Africa0.75 (0.54–1.05)0.57 (0.36–0.90)
 Eastern Africa0.53 (0.22–1.27)0.28 (0.12–0.67)
 Middle AfricaN/A7.70 (3.21–18.51)
 Northern Africa0.69 (0.44–1.09)0.59 (0.26–1.30)
  Morocco0.97 (0.52–1.80)0.44 (0.06–3.12)
 Southern Africa1.17 (0.44–3.13)0.89 (0.13–6.34)
 Western Africa1.36 (0.65–2.86)0.61 (0.09–4.33)
Asia0.52 (0.41–0.67)0.47 (0.41–0.54)
 Eastern Asia0.16 (0.04–0.63)0.31 (0.17–0.54)
  China0.17 (0.02–1.22)0.27 (0.14–0.51)
 South-Central Asia1.00 (0.70–1.41)0.50 (0.39–0.63)
  India0.75 (0.40–1.39)0.43 (0.18–1.04)
  Iran1.33 (0.83–2.15)0.52 (0.41–0.67)
 South-Eastern Asia0.29 (0.09–0.90)0.37 (0.19–0.72)
 Western Asia0.39 (0.27–0.58)0.50 (0.41–0.60)
  Iraq1.05 (0.50–2.21)0.63 (0.46–0.88)
  Lebanon0.23 (0.03–1.62)0.44 (0.24–0.82)
  Syria0.22 (0.03–1.58)0.63 (0.40–0.98)
  Turkey0.31 (0.18–0.54)0.39 (0.29–0.53)
Europe0.71 (0.69–0.74)0.54 (0.50–0.59)
 Eastern Europe0.65 (0.58–0.72)0.52 (0.45–0.61)
  Bulgaria0.64 (0.29–1.43)0.19 (0.05–0.75)
  Czechoslovakia0.75 (0.60–0.94)0.60 (0.39–0.93)
  Hungary0.63 (0.53–0.75)0.44 (0.28–0.71)
  Poland0.66 (0.54–0.80)0.55 (0.44–0.69)
  Romania0.63 (0.37–1.06)0.52 (0.36–0.75)
  Soviet Union0.56 (0.41–0.76)0.51 (0.30–0.87)
 Northern Europe0.77 (0.73–0.80)0.63 (0.57–0.70)
  Denmark0.70 (0.63–0.77)0.68 (0.54–0.85)
  Estonia0.85 (0.70–1.01)0.41 (0.18–0.91)
  Finland0.77 (0.73–0.82)0.63 (0.55–0.73)
  Iceland0.67 (0.28–1.61)0.64 (0.20–1.97)
  Latvia0.52 (0.33–0.83)0.26 (0.04–1.81)
  Lithuania0.19 (0.03–1.35)0.82 (0.12–5.82)
  Norway0.84 (0.74–0.95)0.67 (0.51–0.86)
  UK1.00 (0.74–1.35)0.56 (0.33–0.94)
 Southern Europe0.50 (0.44–0.57)0.44 (0.37–0.52)
  Bosnia0.51 (0.16–1.58)0.52 (0.40–0.66)
  Croatia0.71 (0.43–1.15)0.20 (0.03–1.44)
  Greece0.38 (0.25–0.56)0.20 (0.08–0.48)
  Italy0.58 (0.45–0.75)1.09 (0.62–1.92)
  Portugal1.16 (0.69–1.96)0.49 (0.12–1.96)
  Spain0.63 (0.41–0.97)0.37 (0.12–1.16)
  Yugoslavia0.42 (0.34–0.50)0.37 (0.28–0.50)
 Western Europe0.72 (0.66–0.79)0.48 (0.35–0.67)
  Austria0.72 (0.57–0.91)0.36 (0.12–1.12)
  Belgium0.23 (0.03–1.63)N/A
  France0.85 (0.54–1.34)0.45 (0.11–1.78)
  Germany0.72 (0.64–0.80)0.51 (0.35–0.76)
  Netherlands0.79 (0.57–1.10)0.54 (0.20–1.43)
  Switzerland0.75 (0.48–1.17)0.42 (0.06–2.97)
Latin America0.87 (0.61–1.23)0.59 (0.46–0.77)
 Caribbean2.53 (1.14–5.63)0.70 (0.17–2.78)
 Central America1.45 (0.47–4.49)0.48 (0.12–1.91)
 Southern America0.71 (0.47–1.07)0.59 (0.45–0.78)
  Argentina0.70 (0.22–2.16)0.77 (0.29–2.06)
  Chile0.75 (0.43–1.32)0.56 (0.40–0.78)
Northern America0.69 (0.48–1.00)1.09 (0.74–1.62)
  Canada0.47 (0.12–1.86)3.27 (1.05–10.14)
  USA0.72 (0.49–1.06)1.00 (0.66–1.53)
Oceania1.44 (0.46–4.45)2.41 (0.60–9.62)

After additional adjustment for birth country, we observed a significantly increased risk among immigrants who immigrated at ages younger than 40 compared to those who immigrated at ages 40 or older (RR = 1.24, 95% CI = 1.12–1.36; results not presented in the table). Inclusion of duration of residence into the model attenuated the risk, and it was the same regardless of age at immigration (overall RR = 1.04, 95% CI = 0.93–1.16).

The results were similar at country level and it was increased among most immigrant groups who immigrated at ages younger than 40 compared to those who immigrated at ages 40 or older. The exception was among Danish immigrants in whom the risk was not different between both strata of age at immigration. Those who immigrated at ages less than 40 years had a risk similar to those who immigrated at ages 40 or older (RR = 1.07, 95% CI = 0.79–1.44; results not presented in the table).

The overall risk adjusted for attained age, calendar period of year and education was significantly lower than that of Swedish-born men in both strata of duration of residence (Table V). At the level of regions, immigrants from Middle Africa had 4 times higher risk compared to Swedish-born men in the first 35 years of residence in Sweden (RR = 3.88, 95% CI = 1.61–9.31). Further stratification into narrower categories showed that this increased risk was highest during the first 5 years of residence among immigrants from Middle Africa (RR = 8.49, 95% CI = 2.12–33.94).

Table V. Rate Ratio (RR)12 and 95% Confidence Interval (CI) of Prostate Cancer Among Foreign-Born Men 45 Years and Older by Duration of Residence and Country of Birth, 1961–2004
Birth region/country<35≥35
RR (95% CI)RR (95% CI)
  • 1

    Swedish-born men are the reference group.

  • 2

    Adjusted for age, calendar period of follow up and education.

Foreign-born men0.53 (0.51–0.56)0.78 (0.75–0.81)
Africa0.61 (0.44–0.83)1.01 (0.60–1.71)
 Eastern Africa0.35 (0.18–0.66)0.68 (0.10–4.81)
 Middle Africa3.88 (1.61–9.31)N/A
 Northern Africa0.54 (0.32–0.89)1.02 (0.55–1.90)
  Morocco0.65 (0.29–1.45)1.48 (0.62–3.57)
 Southern Africa0.70 (0.17–2.79)1.83 (0.59–5.68)
 Western Africa1.30 (0.65–2.61)N/A
Asia0.48 (0.43–0.55)0.51 (0.33–0.80)
 Eastern Asia0.28 (0.16–0.49)0.18 (0.02–1.25)
  China0.25 (0.13–0.48)0.30 (0.04–2.11)
 South-Central Asia0.58 (0.47–0.71)0.87 (0.41–1.82)
  India0.45 (0.24–0.87)1.22 (0.55–2.71)
  Iran0.61 (0.48–0.76)0.53 (0.07–3.75)
 South-Eastern Asia0.30 (0.16–0.57)0.70 (0.23–2.16)
 Western Asia0.48 (0.40–0.57)0.42 (0.21–0.84)
  Iraq0.66 (0.48–0.89)1.90 (0.47–7.59)
  Lebanon0.42 (0.23–0.75)N/A
  Syria0.58 (0.37–0.90)N/A
  Turkey0.38 (0.29–0.50)0.29 (0.11–0.76)
Europe0.53 (0.51–0.57)0.78 (0.75–0.81)
 Eastern Europe0.55 (0.48–0.62)0.67 (0.59–0.75)
  Bulgaria0.19 (0.06–0.58)1.24 (0.52–2.98)
  Czechoslovakia0.73 (0.55–0.98)0.69 (0.52–0.91)
  Hungary0.47 (0.35–0.65)0.66 (0.55–0.80)
  Poland0.57 (0.47–0.70)0.67 (0.52–0.85)
  Romania0.53 (0.38–0.74)0.71 (0.37–1.36)
  Soviet Union0.47 (0.29–0.76)0.59 (0.43–0.81)
 Northern Europe0.61 (0.57–0.66)0.82 (0.78–0.86)
  Denmark0.52 (0.43–0.63)0.77 (0.69–0.85)
  Estonia0.54 (0.29–1.00)0.84 (0.70–1.01)
  Finland0.62 (0.57–0.68)0.83 (0.78–0.89)
  Iceland0.34 (0.11–1.05)1.52 (0.63–3.64)
  LatviaN/A0.58 (0.37–0.91)
  Lithuania0.74 (0.10–5.27)0.20 (0.03–1.39)
  Norway0.71 (0.59–0.87)0.85 (0.74–0.97)
  UK0.66 (0.47–0.94)1.22 (0.83–1.77)
 Southern Europe0.39 (0.34–0.45)0.68 (0.58–0.79)
  Bosnia0.52 (0.41–0.66)0.48 (0.07–3.39)
  Croatia0.36 (0.16–0.79)1.02 (0.57–1.85)
  Greece0.26 (0.16–0.42)0.51 (0.29–0.89)
  Italy0.42 (0.25–0.71)0.73 (0.56–0.95)
  Portugal0.76 (0.40–1.47)1.61 (0.77–3.37)
  Spain0.44 (0.24–0.83)0.75 (0.44–1.26)
  Yugoslavia0.33 (0.26–0.40)0.59 (0.46–0.75)
 Western Europe0.51 (0.41–0.63)0.76 (0.69–0.84)
  Austria0.39 (0.20–0.76)0.77 (0.60–0.99)
  BelgiumN/A0.32 (0.04–2.24)
  France0.92 (0.50–1.71)0.69 (0.38–1.25)
  Germany0.52 (0.41–0.67)0.75 (0.67–0.84)
  Netherlands0.38 (0.17–0.84)0.92 (0.66–1.30)
  Switzerland0.49 (0.16–1.51)0.79 (0.49–1.26)
Latin America0.63 (0.51–0.79)1.53 (0.80–2.95)
 Caribbean1.49 (0.71–3.13)1.86 (0.26–13.21)
 Central America0.55 (0.18–1.69)2.77 (0.69–11.07)
 Southern America0.60 (0.47–0.76)1.30 (0.58–2.90)
  Argentina0.78 (0.35–1.73)0.58 (0.08–4.14)
  Chile0.61 (0.45–0.81)N/A
Northern America0.88 (0.62–1.25)0.79 (0.52–1.20)
  Canada0.99 (0.25–3.97)0.94 (0.30–2.92)
  USA0.87 (0.61–1.26)0.77 (0.49–1.20)
Oceania1.79 (0.58–5.56)1.62 (0.41–6.48)

At country level, the risk was significantly higher among immigrants from Denmark, Austria, Croatia, Finland, Germany, Norway, Poland and United Kingdom (UK) who stayed longer than 35 years compared to those who stayed shorter.

After additional adjustment for birth country, the overall risk was significantly higher among foreign-born men who stayed longer than 35 years compared to those who stayed shorter (RR = 1.36, 95% CI = 1.25–1.47). Inclusion of age at immigration into the model did not affect the risk (overall RR = 1.33, 95% CI = 1.21–1.46).

Cancer rates in immigrants compared to their countries of origin

Figure 2 shows the incidence rates among foreign-born and Swedish-born men in Sweden and among men residing in the countries of origin in a selection of 9 countries including Denmark, Estonia, Finland, Germany, Latvia, Norway, Poland, UK and USA. The incidence rates generally increased over time across populations, with the fastest rise occurring in men residing in USA, Finland, Germany and Swedish-born men in Sweden.

Figure 2.

Age-standardized rate (ASR) trends of prostate cancer among foreign-born and Swedish-born men 45 years and older in Swedish study (Swe.) and among men 45 years and older in countries of origin from “Cancer Incidence in Five Continents” databases (CI5). *DOR = Duration of residence. §Selected countries including Denmark, Estonia, Finland, Germany, Norway, Latvia, Poland, UK, USA.

The rate of prostate cancer among all immigrants as one group, as well as the rates among immigrants for the selected countries was lower compared to Swedish-born men. Except for immigrants from Estonia and USA, the country-specific trends were similar to that of the trend among immigrants as one group. Furthermore, immigrants from Denmark, Germany and UK had a pattern similar to that of residents of their countries of origin. Immigrants from Estonia and USA had a more similar pattern to Swedish-born men. In addition, after a longer duration of residence (≥35 years) the rates of prostate cancer among immigrants showed a convergence to the rate of Swedish-born men (Fig. 2).

Figure 3 shows the results of the joinpoint regression analysis of prostate cancer incidence rates. Prostate cancer incidence among Swedish-born men increased by 2.0% per year for the period of 1964–1995 (p < 0.0001). The increase in rates was about 8.0% per year for the period of 1996–2004 (p < 0.0001). Rates among immigrants from Estonia also increased by 1.6% (p = 0.006) and 9.0% (p = 0.0002) per year for the periods of 1961–1993 and 1994–2004, respectively. Incidence rates were relatively stable for 1961 through 1995 among immigrants from Denmark, Germany and Finland, whereas their recent rates (1996–2004) increased by 8.9% (p = 0.0002), 8.8% (p = 0.003), and 9.8% (p < 0.0001) per year, respectively. We did not find any joinpoint for rates among migrants from UK, USA, Norway and Poland. While rates were stable among immigrants from UK for the period of 1968–2004, it increased by 1.4% (p = 0.002), 1.8% (p = 0.0009) and 4.4% (p < 0.0001) per year for the same time period among immigrants from Norway, USA and Poland, respectively.

Figure 3.

Fitted lines from Joinpoint regression for Age-standardized rate (ASR) trends of prostate cancer among foreign-born and Swedish-born men 45 years and older.

Discussion

In this population-based cohort study, we found that foreign-born men had an overall lower risk of prostate cancer compared to Swedish-born men. In general, the risk was not affected by education and occupation, and it did not vary by attained age. Duration of residence was an effect modifier for prostate cancer risk among the immigrants.

The major strength of our study is the long follow up of a large cohort including all Swedish-born and foreign-born men aged 45 years and older in Sweden. Our study covered all immigrants in Sweden originating from around 200 countries. The completeness and reliability of data in Swedish registers, and the uniform health care system, are other strengths of our study.14

The observed differences in risk of prostate cancer among foreign-born compared to Swedish-born men may be related to a number of factors including the following: (i) diagnosis by prostate-specific antigen (PSA), the most important factor that affects the reported incidence of prostate cancer19, 20; (ii) access and quality of health care and different diagnostic procedures including digital rectal examination (DRE) and transurethral resection of the prostate21; (iii) genetic susceptibility and (iv) exposure to environmental and lifestyle risk factors.

Despite its high morbidity, the etiology of prostate cancer remains largely unknown. The only established risk factors are age, family history and race.22, 23 Many risk factors such as diet, obesity, physical activity, occupation and androgens have been suggested; but their role in etiology of prostate cancer is unclear.22, 23

A well-known epidemiological observation about prostate cancer is the large difference in incidence rates between different countries, and between groups of different ethnicity.4, 24 African-Americans have the highest rates in the world, followed by Caucasian-Americans. Populations of the Caribbean, Scandinavia, Western Europe, Canada and Oceania have the next highest incidence rates of prostate cancer. Rates within Europe vary considerably and are highest in Western Europe, and lowest in Eastern European countries. Asia has the lowest rates of prostate cancer in the world; however, there are obvious differences in incidence rates within Asia. Japan and the Philippines have much higher rates than India, China and Pakistan.1, 3 Prostate cancer incidence data from Africa are sparse; however, several reports have suggested that prostate cancer has become the number one cancer among men in Nigeria and Uganda.25, 26

The risk of prostate cancer among first-generation immigrants was modified by duration of residence. However, the risk remained statistically lower among these immigrants compared to Swedish-born men in both strata of duration of residence. This might suggest that either genetic factors are more influential than environmental factors in the susceptibility to prostate cancer, or more likely that it takes more than one generation for environmental factors to act.

We found the highest risk of prostate cancer among immigrants from Middle Africa, within their first 5 years of residence in Sweden, and among those who immigrated after an age of 40. Our findings are in agreement with the consistently observed high risk of prostate cancer among African immigrants in the USA and the UK that remained high even after adjustment for a number of known and suspected risk factors.27–29 Although immigrants from Middle Africa may not be representative of their counterparts in Africa, in the absence of a viable cancer register in Africa, it can be concluded that they reflect the high risk of prostate cancer in their countries of origin. On the other hand, immigrants from the Caribbean who have immigrated in ages younger than 40 years showed higher risk of prostate cancer. Previous studies have shown high risk of prostate cancer among immigrants from the Caribbean in the UK.29 Interpretation is related to the real risk in country of origin. If one accepts high rates in the Caribbean, as some studies have shown earlier,30 then better access to health care and diagnostic PSA test in Sweden could explain the increased risk observed in our study. In general, the observed high risk of prostate cancer among immigrants from Middle Africa and the Caribbean that have common ancestors in Africa, may suggest a strong influence of genetics or, less likely, shared lifestyle risk factors in the etiology of prostate cancer. The data is also suggestive of higher rates of prostate cancer in Africa and the Caribbean than that presented in available international data indicating the possible unreported cases in these countries.2

The observed lower risk of prostate cancer among immigrants from Asia compared to Swedish-born men is similar to other studies on Asian migrants in the USA, the UK and Australia.28, 31, 32 We found a tendency of convergence of the risk among Asian immigrants who immigrated before 40 years of age or among those who stayed 35 years and longer toward the risk among Swedish-born men. We did not have enough power for the more detailed analysis within Asia. Existence of unknown risk factors associated with western lifestyle in this group, especially diet may partly explain this convergence.22, 33

The incidence of prostate cancer is more affected by screening than any other cancer, because it is a slow-growing tumor with a long latency period. In addition, the prevalence of microscopic latent tumors has been shown to be quite high and similar among the elderly in most populations.34 The observed significant increase in incidence among immigrants from Estonia before 1994, among Swedish-born men before 1995, and among immigrants from Norway, Poland and USA for the period of 1968–2004 could be due to better access to health care with more frequent routine DRE, or more frequent use of transurethral resection of the prostate, increased awareness of prostate-related symptoms or change in prevalence of risk factors, and thus a real increase in the number of patients with prostate cancer.5, 19, 35, 36 A more rapid increase in incidence of prostate cancer, which started around 1995 among immigrants from Denmark, Estonia, Germany and Finland, and for the same period among Swedish-born men, can be attributed to the introduction of diagnostic PSA test in Sweden around 1990.37 These 2 different scenarios in trends of prostate cancer give us some clues about disparities in diagnostic activities and in access to health care system among different migrant groups in Sweden. Previous studies in USA showed differences in the frequency of DRE and diagnostic PSA tests among different ethnic groups, even after controlling for demographic factors and access to these diagnostic tools.38, 39 Information about disparity in diagnostic activities and screening behaviour by country of birth is not available in Sweden. Such differences may help to explain part of the observed lower risk of prostate cancer among some immigrant groups compared with Swedish-born men.

Migrants from Denmark had similar trends of incidence to their counterparts in Denmark. They also had a sharp increase after 1995, suggesting an increasing level of diagnostic activity due to the PSA test that was introduced around 1990. This is in agreement with results from previous studies in Nordic countries that showed an increase in incidence in Denmark before and after introduction of PSA test.5 We also found that the risk of prostate cancer among immigrants from Denmark remained lower compared to Swedish-born men in all strata of attained age, age at immigration and duration of residence. This might be explained by genetic factors or persistent lifestyle risk factors in the etiology and progression of prostate cancer within this population; however, the effect of diagnostic density should also be taken into account.

Migrants from Estonia had a trend in incidence of prostate cancer more similar to the Swedish-born men than to their counterparts in Estonia. The increase occurred in 2 periods, a slow increase before and a sharp increase after 1994. Among all immigrants, men from Estonia were youngest at immigration with the longest duration of residence in Sweden. Nearly all immigrants from Estonia arrived in Sweden at the same time, in 1944 during World War II, and are known as the most integrated immigrants into the Swedish society.40 We found a higher risk among Estonians who immigrated at ages younger than 40 compared to those who immigrated at ages 40 or older. We also found higher risk among those who stayed longer than 35 years in Sweden compared to those who stayed shorter. These observations indicate the importance of modifiable environmental factors in the etiology of prostate cancer among immigrants from Estonia.

Higher SEP has been shown to be associated with increased incidence of prostate cancer.41, 42 These differences in SEP have often been explained by differences in diagnostic activity, and are at least partly due to greater use of diagnostic PSA test among higher educated men. However, in our study adjustment for education did not change the risk. Although, education is one of the best indicators of SEP, there might be a residual confounding regarding social position. Diagnostic activity may also vary by occupation due to available occupational health services. The overall risk was not affected by additional adjustment for occupational SEP. However, a high proportion of cohort members had missing or unclassified information on occupational SEP.

It has been reported that 25,000–50,000 non-Nordic immigrants have left the country without this being registered in 1997.43 This over coverage, i.e., registers including more people than they are meant to cover due to not reporting emigration, is a concern in register-based studies. Because of high-quality health care system, cancer patients might be less likely to move from Sweden. However, it is likely that the very sick and very old might prefer to die in their birth country. If emigration not reported then an underestimation of both cancer risk and mortality will occur. In that case over coverage would be less probable regarding cancer. Over coverage does not apply to immigrants from Nordic countries as they have coordinated their population registers. Selective return migration may also have influenced our results, especially for migrants from the USA and several European countries such as Scandinavian countries.44 These immigrants are more likely to return to their countries of origin in case of cancer or after retirement. It is less probable for immigrants from other countries; however, such as Iran, Iraq and Poland to return to their home countries, mostly due to better health care and welfare benefits in Sweden and due to unstable political and economical situations in their birth countries.44

Lack of information on individual risk factors, the clinical stage and histological grade, and true incidence of prostate cancer in most countries of origin are weaknesses of our study. Because immigrants are not representative of the total population in the country of origin, caution must be taken when comparing incidence rates among immigrants with those of the country of origin. For example, the risk of prostate cancer may vary across regions within one country. However, such information is not available in the Swedish registry.

In summary, being an immigrant in Sweden seems to be a major determinant for prostate cancer risk. Differences found in the risk of prostate cancer between Swedish-born and foreign-born men are not explained merely by differences in diagnostic activities or access to health care. The roles of environmental and genetic factors are also important. A genetic susceptibility of Africans and their descendants to prostate cancer calls for more studies to clarify the complex interplay of genetic and lifestyle factors in prostate cancer progression.

Ancillary