Cancer Causes Control DOI 10.1007/s10552-013-0217-x
ORIGINAL PAPER
Cancer incidence and mortality of Surinamese migrants in the Netherlands: in-between Surinamese and Dutch levels? Gracie¨lle Williams • Dennis R. A. Mans • Joop Garssen • Otto Visser • Danie¨lle Kramer Anton E. Kunst
•
Received: 21 September 2012 / Accepted: 20 April 2013 Ó Springer Science+Business Media Dordrecht 2013
Abstract Purpose It has been suggested that the cancer risk of migrants from low-income to high-income countries will converge toward the levels of their host country. However, comparisons with country of origin are mostly lacking. We compared cancer incidence and mortality rates of Surinamese migrants in the Netherlands to both native Dutch and Surinamese levels. Methods Data covering the period 1995–2008 were obtained from Surinamese and Dutch national cancer registries and national cause-of-death registries. Cancer incidence was studied for 21 types of cancer and cancer mortality for nine types. We calculated age-standardized incidence/mortality ratios (SIR/SMR) for the Surinamese migrants and for Suriname, using the native Dutch population as reference. Results Significantly lower overall cancer incidence (SIR = 0.77, 95 % CI = 0.69–0.84) and mortality rates (SMR = 0.63, 95 % CI = 0.55–0.72) were found for Surinamese migrants compared to native Dutch. Generally, cancer risk was lower for most cancers (e.g., cancer of the
G. Williams D. Kramer A. E. Kunst (&) Department of Public Health, Academic Medical Centre (AMC), University of Amsterdam, Amsterdam, The Netherlands e-mail:
[email protected] D. R. A. Mans Department of Pharmacology, Faculty of Medical Sciences, Anton de Kom University, Paramaribo, Suriname J. Garssen Statistics Netherlands, The Hague, The Netherlands O. Visser Department of Registration and Research, Comprehensive Cancer Centre Amsterdam (IKA), Amsterdam, The Netherlands
breast, colon and rectum, lung), but higher for other cancers (e.g., cancer of the uterine cervix, liver). For most cancers, cancer risk of the Surinamese migrants was inbetween Surinamese and native Dutch levels. Importantly, for many cancers, migrants’ incidence and mortality rates had not closely approached native Dutch rates. For skin cancer, incidence levels for Surinamese migrants were lower than both Surinamese and native Dutch levels. Conclusions The results suggest that cancer incidence and mortality rates of Surinamese migrants generally converge from Surinamese toward Dutch levels, though not for all cancer types. Overall, Surinamese migrants still had a much more favorable cancer profile than the native Dutch population. Keywords Incidence Mortality Cancer types Migrants Suriname
Introduction Cancer incidence and mortality rates vary greatly among countries [1]. For many cancers, incidence and mortality rates are relatively low in South Asia, while they are relatively high in most western countries. Migrant studies may offer insights into the causes behind these differences [2, 3]. Comparisons of cancer risk among migrants, the country of origin, and the host population may offer insights into the relative importance of genetic makeup, early-life exposures, and later-life environment exposures in the etiology of cancer [2, 3]. Migrant studies focusing on migrants from low- and middle-income countries have found the overall cancer incidence and mortality for migrants to be lower compared to the host population [2, 4–11]. However, differences were
123
Cancer Causes Control
noted among different types of cancer. Compared to the host population, migrants from these countries were found to have lower risk of cancers that are assumed to be related to a western lifestyle (e.g., breast, lung, and colorectal cancer) but higher risk of cancers potentially related to infections (e.g., cervical and liver cancer) [2, 9, 10, 12, 13]. The results of several studies suggest that cancer rates of migrants from low- and middle-income countries were converging from the levels of the home population toward the levels of the host population [2, 4–11]. However, this trend was not observed for all cancers. Moreover, the extent of convergence varied greatly according to cancer type and migrant group. In these studies, the extent of convergence was mostly explored by assessing the relationship with length of stay in the host country or with age at the time of immigration. Fewer studies made comparisons to the cancer risk in low- or middle-income countries of origin [5, 6]. In the Netherlands, the proportion of migrants from lowand middle-income countries currently amounts to 11 % [14, 15]. Dutch studies have found a lower overall cancer risk of the migrants from low- or middle-income countries compared to the native Dutch population [4, 10]. Within most migrant groups, overall cancer mortality was highest in second-generation migrants, those who migrated at younger age, and those with longer duration of residence [4]. This suggests that, over time, cancer mortality in migrants converges toward the levels of the native Dutch population. Unfortunately, comparisons with the countries of origin are lacking. The most consistent pattern of convergence was found for migrants from the Republic of Suriname, a South American country that used to be a Dutch colony [4]. In the decade after independence in 1975, many Surinamese migrated to the Netherlands. Migration rates have been lower since the 1990s and especially since 2005 [14]. The main Surinamese immigrant groups, the Hindustani and Creole, have similarities with South Asian and Caribbean migrant populations in other countries such as England [16]. Migrant studies in England, Australia, and the United States of America have suggested a convergence of overall cancer risk of South Asian migrants, while no consistent patterns were found for the Caribbean migrants [17–21]. The aim of our study is to assess to what extent cancer incidence and mortality rates of Surinamese migrants in the Netherlands were in-between the rates of the Suriname population and those of the native Dutch population. We will extend previous Dutch work by comparing cancer rates of Surinamese migrants not only to the native Dutch population but also to their country of origin. Moreover, we will simultaneously study incidence rates and mortality rates, which have been studied apart from each other in most migrant studies so far [6, 10, 14, 18, 22, 23].
123
Methods Our study populations included the Surinamese migrant population residing in the Netherlands (i.e., Surinamese migrants), the native Surinamese population residing in Suriname (i.e., Suriname), and the native Dutch population residing in the Netherlands (i.e., native Dutch population). For each group, we collected data on cancer incidence and mortality, and the corresponding data on population at risk. Population data The number of person years at risk corresponding to the mortality and incidence data was estimated from data on number of residents by age, sex, and calendar year. For the Surinamese migrants and ethnic Dutch, these population data were obtained from Statistics Netherlands. In the Netherlands, accurate estimates of mid-year population were derived from the continuous population registry. For Suriname, population data were obtained from the General Bureau of Statistics, Ministry of Planning And Developmental Cooperation of Suriname. In Suriname, estimates of the mid-year population were derived, through extrapolation, from data of the seventh general population and housing census Suriname of 2004. Cancer incidence data Numbers of new cases of cancer by age and sex among the Surinamese migrants and among the native Dutch population were obtained from the Netherlands Cancer Registry (NCR) for years 1995–2004. The NCR is a national, population-based cancer registry with complete national coverage since 1989. The NCR contains complete, detailed, and validated data on each cancer patient [24]. Numbers of new cases of cancer by age and sex in Suriname were obtained from the Surinamese pathologybased cancer registry for years 2000–2004. This is a national registry maintained by the Pathologic Anatomy Laboratory of the Academic Hospital Paramaribo. The registry is based on information from hospitals; information from death certificates is not used to identify or diagnose cases. The registry has a complete national coverage since 1980, although with under-registration of new cancer cases at a rate of 10–15 % [25]. For this study, we included data on 21 different cancer types (see Table 1). We excluded cancer types for which pathology was confirmed in \80 % of all cases in the Surinamese cancer registry (i.e., cancer of the liver, gall bladder and bile ducts, and pancreas) or because of other registration problems (i.e., lung cancer and cancer of the mesothelioma). Furthermore, in Fig. 1 below, we excluded cancers for which the number of cases was too small for
Cancer Causes Control Table 1 List of cancers included in the incidence and mortality analyses, with their abbreviations, and ICD 10 codes Description
ICD 10 codesa
Incidence
Mortality
Cancer of the oral cavity
C01–C06
x
Cancer of the pharynx
C09–C14
x
Cancer of the esophagus Cancer of buccal cavity, pharynx and esophagus
C15 C00–C15
x
Cancer of stomach
C16
x
x
Cancer of colon and/or rectum
C18–C21
x
x
Cancer mortality data x
Cancer of liver
C22.0, C22.1
Cancer of the larynx
C32
x
Cancer of larynx, trachea, bronchus and lung
C30–C34; C39
Cancer of bone and soft tissues
C40–C41; C46–C49
x
Cancer of the skin, Melanoma
C43, C44
x
Cancer of breast
C50
x
x
Cancer of uterine cervix Cancer of uterine corpus
C53 C54, C55
x x
x
Cancer of the ovary
C56
x
Cancer of the prostate
C61
x
Cancer of the kidney
C64
x
Other cancers of the urinary system (renal pelvis, urethra, urinary bladder and urinary tract, other/not specified)
C65–C68
x
x x
Cancer of the brain
C71
x
Cancer of the thyroid gland
C73
x
Hodgkin’s disease and leukaemiab
C81, C91– C95
Hodgkin’s lymphomab
C81
x
Mature b cell and t cell and precursor cell tumorsb
C82–C88; C90–C91; C95–C96
x
Myeloid tissueb
C92
x
Other neoplasm’s
C00–D48
x
Unknown cancers
C80
x
In the Dutch cancer registry data, Surinamese migrants were classified according to country of birth of the subject. Thus, only first-generation migrants were classified as being a migrant. All age groups were included in the incidence data.
x
x
a
Cancer codes as specified by the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD10), by the World Health Organization (WHO)
b
These types of cancers may be categorized as cancer of hematopoietic and lymphoid tissue
comparative analyses. More specifically, we excluded cancers for which the observed and expected number of cases was lower than 30 among either the Surinamese migrants or in Suriname (i.e., cancers of the oral cavity, larynx, and thyroid gland, as well as Hodgkin lymphoma and myeloid malignancies).
Number of deaths from cancer by age and sex among the Surinamese migrants and the native Dutch population were obtained from the Dutch cause-of-death registry for years 1995–2005. For Suriname, we obtained data on cancer deaths by age and sex for years 2000–2008 from the Bureau for Public Health Care, Ministry of Public Health of Suriname. While the Dutch cause-of-death registry is complete, the under-registration in the Surinamese registry was estimated at 10–15 % between years 2000 and 2008 [26]. All deaths were coded according to International Classification of Diseases (ICD), 10th revision [27]. For this study, we studied the nine cancers that were the largest causes of death among Surinamese migrants (Table 1). In the Dutch mortality data, a subject was considered to be a Surinamese migrant if he/she, or at least one of his/her parents was born in Suriname. This rule, which follows the standard definition of Statistics Netherlands [24], implies that both first- and second-generation migrants were classified as migrant. Individuals of 80 years and older were excluded from the mortality analyses, because only a small part of Surinamese migrants were over 80 years old, and because the cause-of-death registration of these older migrants may be compromised. Statistical methods Because of limitations to the available data, Surinamese migrants and the Surinamese population could not always be compared in a direct way. Instead, we chose to compare these groups indirectly, by using the Dutch population as a common reference population. For this, we calculated standardized incidence/mortality ratios (SIR/SMR) with the native Dutch population as the reference group. We calculated two series of ratios aimed to compare the native Dutch population with, respectively, Suriname and the Surinamese migrants. SIR and SMR are ratios of the observed and the expected number of cases. The former represents the number of cases observed for, respectively, the Surinamese migrants and Suriname. The latter represents the number of cases expected for these two populations in the hypothetical case that they would have the age–sex-specific incidence/mortality rates of the native Dutch. We calculated 95 % confidence intervals (CI) for the SIR and SMR using the exact method proposed by Ulm [28].
123
Cancer Causes Control
To take into account under-registration of cases in the Surinamese cancer and mortality registries, we decreased the expected values for Suriname with 15 %. The same correction was made for both the SIR and the SMR. Results are presented for men and women combined. For most types of cancers, the number of observed cases among Surinamese migrants was too small for genderstratified analyses. The exception was lung cancer mortality, for which the number of cases were sufficiently large to present results for men and women separately.
Results Cancer incidence The overall cancer incidence rate of Surinamese migrants (SIR = 0.77, 95 % CI = 0.69–0.84) and Suriname (SIR = 0.72, 95 % CI = 0.61–0.83) was significantly lower than that of the native Dutch population (Table 2). Surinamese migrants had significantly lower incidence rates of cancers of oral cavity, esophagus, colon and rectum, pancreas, larynx, skin including melanoma, breast, ovary, urinary system including kidney, brain, myeloid tissue, and the rest group of ‘other neoplasms.’ Surinamese migrants had significantly higher incidence rates of cancers of the liver, uterine cervix and corpus, prostate, and thyroid gland. Figure 1 compares the SIR of the Surinamese migrants with the SIR of Suriname. Minor cancer types are excluded from this presentation (see ‘Methods.’) For most cancers,
incidence rates of the Surinamese migrants were inbetween those of Suriname and those of the Netherlands, suggesting a convergence from Surinamese to Dutch levels. Results were suggestive of a downward convergence (i.e., from higher Surinamese levels toward lower Dutch levels) for cancers of uterine cervix and corpus, bone and soft tissues, and pharynx. Results were suggestive of an upward convergence (i.e., from lower Surinamese levels to higher Dutch levels) for overall cancer, cancers of colon and rectum, breast, kidney, brain, esophagus, as well as mature B cell and T cell tumors, and the group of ‘other neoplasms.’ Incidence rates of the Surinamese migrants were relatively close to Dutch levels for colorectal cancer, breast cancer, uterine cervix cancer, and mature B cell and T cell tumors. For some cancer types, incidence rates of the Surinamese migrants were not in-between the Surinamese and Dutch levels, suggesting a divergence of incidence rates. While cancers of the urinary system and skin cancer had lower incidence rates in Suriname than in the Netherlands, their incidence was even lower in Surinamese migrants. Conversely, prostate cancer and stomach cancer had higher incidence rates in Suriname than in the Netherlands and even higher incidence rates in Surinamese migrants. Cancer mortality The overall cancer mortality rate of Surinamese migrants (SMR = 0.63, 95 % CI = 0.55–0.72) and Suriname (SMR = 0.56, 95 % CI = 0.50–0.63) was significantly
Fig. 1 SIR (95 % CI) for Surinamese migrants and Suriname, by type or location of cancer
123
Cancer Causes Control Table 2 Observed and expected cancer incidence rates and the SIR (95 % CI) for Surinamese migrants and Suriname, stratified by type or location of cancer Type or location of cancer
Surinamese migrants versus native Dutch
Suriname versus native Dutch
Observed
Observed
Expected
SIR (95 % CI)
Expected
SIR (95 % CI)
Oral cavity
64
84.4
0.76 (0.58–0.96)
17
28.7
0.59 (0.35–0.93)
Pharynx
52
59.1
0.88 (0.66–1.14)
32
22
1.45 (0.99–2.03)
Esophagus Stomach
45 220
140.9 191.9
0.32 (0.23–0.42) 1.15 (1.00–1.31)
10 64
43.2 63.7
0.23 (0.11–0.41) 1.00 (0.77–1.27)
Colon, rectum
541
730.3
0.74 (0.68–0.81)
153
300.1
0.51 (0.43–0.60)
Larynx
27
62.6
0.43 (0.28–0.62)
19
23.5
0.81 (0.49–1.23)
Bone, soft tissue
64
65.6
0.98 (0.75–1.24)
47
34.8
1.35 (0.99–1.78)
Skin, melanoma
78
461.6
0.17 (0.13–0.21)
142
356.5
0.40 (0.34–0.47)
Breast
726
1,016.7
0.71 (0.66–0.77)
230
425.5
0.54 (0.47–0.61)
Cervix
153
81.6
1.88 (1.59–2.19)
248
29
8.56 (7.52–9.67)
Uterus
144
120.1
1.20 (1.01–1.41)
90
52
1.73 (1.39–2.12) 0.62 (0.40–0.90)
Ovary
83
134.3
0.62 (0.49–0.76)
25
40.4
Prostate
511
415.7
1.23 (1.12–1.34)
280
248.4
1.13 (1.00–1.27)
Kidney
80
134.4
0.60 (0.47–0.74)
18
54.7
0.33 (0.20–0.51)
Other urinary tract
56
202.6
0.28 (0.21–0.36)
31
73.8
0.42 (0.29–0.59)
Brain
54
135.2
0.40 (0.30–0.52)
7
43.9
0.16 (0.06–0.31)
Thyroid gland
53
33.1
1.60 (1.20–2.08)
25
18
1.39 (0.90–2.02)
35 367
30.5 342.3
1.15 (0.80–1.58) 1.07 (0.97–1.19)
14 69
24.2 154.1
0.58 (0.32–0.95) 0.45 (0.35–0.56)
Myeloid tissue
88
110.3
0.80 (0.64–0.98)
3
21.7
0.14 (0.03–0.37)
Other cancers
115
200.4
0.57 (0.47–0.69)
73
136.5
0.53 (0.42–0.67)
184
264.5
0.70 (0.60–0.80)
52
80.5
0.65 (0.48–0.84)
4,065
5,308.5
0.77 (0.69–0.84)
1,700
2,364.7
0.72 (0.61–0.83)
Hodgkin’s lymphoma Mature cell tumors
Unknown Total
lower than that of the native Dutch population (Table 3). Compared to the native Dutch population, Surinamese migrants had significantly lower mortality rates for cancers of the buccal cavity, colon and rectum, lung, breast, and for ‘other neoplasms.’ Surinamese migrants had significantly higher mortality rates for cancers of the liver and the uterine cervix. Figure 2 compares the SMR of the Surinamese migrants with the SMR of Suriname. For half of the cancers, mortality rates of the Surinamese migrants were in-between those of the Surinamese and native Dutch, suggesting a convergence from Surinamese to Dutch levels. For cancers of the liver and uterine cervix, results were suggestive of a downward convergence from higher Surinamese levels to the lower Dutch levels. For cancers of colon and rectum, buccal cavity, and lung, results were suggestive of an upward convergence from lower Surinamese levels to the higher Dutch levels. For other cancer types, mortality rates of the Surinamese migrants were not in-between Surinamese and Dutch levels, suggesting a divergence of mortality rates. For some of the cancers with lower mortality rates in Suriname than in
the Netherlands (Hodgkin’s disease and leukemia, breast cancer, the groups of ‘other neoplasms’), Surinamese migrants had mortality rates that were even lower than the Surinamese levels. Striking differences in cancer incidence and mortality rates were found for breast cancer: While breast cancer incidence rates of Surinamese migrants were close to the relatively high Dutch levels, mortality rates were lower even as compared to Suriname. Such a contrast between incidence and mortality rates was not observed for other cancer types.
Discussion Surinamese migrants had significantly lower overall cancer incidence and mortality rates than the native Dutch population. Cancer risk was lower for cancers that might be lifestyle related (e.g., breast, colorectal, and lung cancer) and higher for cancers that might be infection related (e.g., cancer of the uterine cervix and liver). For the majority of cancers, incidence or mortality rates of the Surinamese
123
Cancer Causes Control Table 3 Observed and expected cancer mortality rates and the SMR (95 % CI) for Surinamese migrants and Suriname, stratified by type or location of cancer Type or location of cancer
Buccal cavity, pharynx, esophagus
Surinamese migrants versus native Dutch
Suriname versus native Dutch
Observed
Observed
Expected
SMR (95 % CI)
Expected
SMR (95 % CI)
72
143.89
0.50 (0.39–0.63)
64
221.81
0.34 (0.26–0.43)
Stomach
113
101.8
1.11 (0.91–1.33)
140
168.29
0.98 (0.82–1.15)
Colon, rectum Liver
195 100
266.86 31.58
0.73 (0.63–0.84) 3.17 (2.58–3.83)
180 169
433.81 51.66
0.49 (0.42–0.56) 3.85 (3.29–4.46)
Lung (both sexes)
323
690.86
0.47 (0.42–0.52)
319
1,146.72
0.33 (0.29–0.36)
Breast
177
332.61
0.53 (0.46–0.61)
202
422.03
0.56 (0.49–0.64)
Cervix
42
25.3
1.66 (1.20–2.22)
238
29.67
9.44 (8.28–10.69)
Hodgkin’s lymphoma, leukemia Other cancers Total
69
82.84
0.83 (0.65–1.05)
96
125.2
766
1,088.14
0.70 (0.55–0.87)
1,343
1,735.1
0.90 (0.73–1.10) 0.91 (0.86–0.96)
2,180
3,454.74
0.63 (0.55–0.72)
3,070
5,481.01
0.68 (0.66–0.70)
Fig. 2 SMR (95 % CI) for Surinamese migrants and Suriname, by type or location of cancer
migrants were in-between Surinamese and Dutch levels. The results suggested an upward convergence (i.e., from lower Surinamese to higher Dutch levels) for most cancers, but a downward convergence (i.e., from higher Surinamese to lower Dutch levels) for a few other cancers, notably cancer of uterine cervix. For skin cancer, the data suggested a divergence: Rates of migrants were not only below Dutch levels but even below the relatively low Surinamese levels. Evaluation of data limitations Data available to this study had some potential limitations. First, the study period covered by the Suriname data differed from that of the Dutch data, both for incidence (2000–2004 vs. 1995–2004) and for mortality (2000–2008 vs. 1995–2005). The main reason for this discrepancy is, that the Surinamese incidence and mortality data had to be close to the 2004 population census, while the Dutch data
123
had to cover at least 10 years in order to include sufficient cancer cases among Surinamese migrants. The central years of these study periods differ by 3–5 years. Such a difference may have affected our comparisons to a slight extent only in cases where secular changes are strong. The data sources in Suriname are estimated to underregister the incidence and mortality of cancer by 10–15 % [25, 26]. We corrected for this degree of under-registration in our statistical analysis, by adjusting the expected values in SMR and SIR estimations. However, this procedure did not take into account potential differences in degree of under-registration between cancer types. Especially for mature B cell and T cell tumors, we expect under-registration in Suriname to be much higher than 15 %. This would explain why we found the incidence of these malignancies in Suriname to be much lower than that in both Surinamese migrants and the native Dutch population.
Cancer Causes Control
The population data for Suriname were available for only one point in time, the 2004 census. These population counts had to be extrapolated in order to calculate the denominator of Surinamese cancer incidence and mortality rates. Moreover, Surinamese population data were estimated to have an undercount of 3.16 % [29]. Population undercounting might have caused an overestimation of cancer risk of Suriname, or it may have mitigated some of the underestimation due to under-registration of cancer incidence and mortality cases. When interpreting the results, two selection effects have to be considered. First, migrants could be healthier individuals than their fellow countrymen who did not migrate [3]. If so, this so-called healthy migrant effect might have contributed to our finding that cancer incidence and mortality rates among Surinamese migrants are generally higher than in Suriname [10, 21]. This effect may be compounded by the so-called unhealthy remigration effect, according to which migrants have the tendency to return to their home country when they experience serious disease [4, 30]. However, remigration of cancer patients would only have affected our results if these remigrants would be administratively removed from the Dutch population registry. This is a relatively rare event among Surinamese in the Netherlands [31]. Interpretation of key results Surinamese migrants were found to have lower overall cancer rates than the native Dutch population. These results are consistent with prior research on Surinamese migrants in the Netherlands [4, 10] and Caribbean and South Asian migrants in England, Australia, and the United States of America [17–21]. Also in agreement with prior studies [2, 9, 10, 12, 13] is our finding of higher incidence and mortality rates among Surinamese migrants than the native Dutch population for cancers that may be related to infection exposure, and lower rates for cancers that may be related to a western lifestyle. For the majority of cancers, incidence and mortality rates of the Surinamese migrants were in-between Surinamese and Dutch levels. This observation is consistent with the hypothesis of a convergence of cancer risk from country of origin to the current host country. A similar pattern suggesting convergence was observed in prior Dutch studies and in foreign studies on South Asian migrants, but not in foreign studies on Caribbean migrants [2, 4–11, 17–21]. For cancer of the uterine cervix, rates of Surinamese migrants were in-between the high levels in Suriname and the much lower Dutch levels, although the latter had not yet been approached. This pattern might be related to the lower prevalence of human papilloma virus infection in the Netherlands [25, 32–34].
Risk of colorectal and lung cancer were suggested to converge from relatively low Surinamese levels toward higher Dutch levels. Prior migrant studies suggested that this pattern may be due to the adaption of a western lifestyle, which is characterized by higher levels of smoking, unhealthier diets, and less physical activity compared to the countries of origin [4]. In the Netherlands, the prevalence of migrant Surinamese smokers has been found to be twice as high as in Suriname [35]. Results for breast cancer incidence suggested a convergence from lower Surinamese levels toward higher Dutch levels. Breast cancer risk has been related, among others, to low parity, older age at first pregnancy, and use of oral contraceptives [4, 34, 36]. Compared to their homeland, Surinamese migrants in the Netherlands may be more exposed to these reproductive risk factors [36]. Furthermore, Surinamese migrants in the Netherlands may benefit from nationwide proactive programmes for breast cancer screening [37]. Their participation to these programmes may have resulted in earlier detection and treatment of their disease, which may explain why breast cancer incidence levels were increased, while mortality levels were decreased, as compared to Suriname. Skin cancer and melanoma incidence among migrants was lower compared to both Surinamese and Dutch levels. High levels of sunlight exposure and a pale skin color have been found to be important risk factors for cancer of skin and melanoma [38]. In these respects, Surinamese migrants may benefit from a double protection: while their African and/or Asian roots provide them with a darker skin, the Dutch temperate climate limits sunlight exposure. Risks of cancers of the ovary and the urinary system (except kidney) were found to be close to Suriname levels and lower than the Dutch levels. The persistently low levels of ovary cancer are not in line with the convergence observed for breast cancer, suggesting that reproductive factors [4, 34, 36] are not the entire explanation. Women in Suriname might have a lower risk of ovary cancer due to higher levels of vitamin D resulting from higher exposure to sunlight [39, 40]. Migrants might still benefit from these high vitamin D levels. Of cancers of the urinary system, bladder cancer is the most common cancer. Bladder cancer incidence among Surinamese migrants was found to be lower than that in the native Dutch population [33]. Bladder cancer is linked to chronic infection with Schistosoma hematobium [14, 33], which is rare in Suriname [41]. Other factors could play an additional role, such as smoking, but we could not obtain data to compare smoking prevalence in Suriname with that in Surinamese migrants. Risks of cancers of the liver and the prostate were also found to be close to Suriname levels, but then at higher levels than the Netherlands. Persistently high levels of liver cancer mortality might be due to long-lasting effects of
123
Cancer Causes Control
viral infections during early childhood. Of all liver cancer cases worldwide, 80 % is attributable to persistent viral infection, mainly with hepatitis B and C viruses [42]. Studies from Australia and other countries found the prevalence of hepatitis B among Asian migrants to be as high as in their countries of origin [17, 43]. Persistently high levels of prostate cancer might in part be due to genetic factors. About 40 % of the Surinamese population has its roots in West Africa, similarly to African-Americans [10]. African-Americans have been found to have higher prostate cancer incidence rates than white Americans [10]. Conclusion Results of our study partially support the hypothesis of a convergence of cancer risk from Surinamese toward Dutch levels. Continued convergence in forthcoming years would imply a decrease in migrants’ risk of some cancers (such as cancer of the uterine cervix) and an increasing risk of some other cancers (such as colorectal and breast cancer). However, for many cancers, recent cancer risk of migrants was much closer to Surinamese than to Dutch levels. The results suggest that the relatively favorable cancer profile of Surinamese migrants, even though subject to change, may persist for some years or even decades [4]. Further comparisons to countries of origin, not only for Surinamese but also for other migrant populations, are necessary to predict future trends in cancer risk among migrants [2]. Acknowledgments We would like to acknowledge drs. W. Punwasi of the Bureau for Public Health Care (Suriname), and the General Bureau of Statistics (Suriname) for preparing and providing tabulation data on mortality and populations numbers of Suriname. Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Jemal A, Center MM, DeSantis C (2010) Global patterns of cancer incidence and mortality rates and trends. Cancer Epidemiol Biomark Prev 19:1893–1907 2. Arnold M, Razum O, Coebergh JW (2010) Cancer risk diversity in non-western migrants to Europe: an overview of the literature. Eur J Cancer 46:2647–2659 3. Parkin DM, Khlat M (1996) Studies of cancer in migrants: rational and methodology. Eur J Cancer 32:761–771 4. Stirbu I, Kunst AE, Vlems FA, Visser O, Bos V, Deville W, Nijhuis HGJ, Coebergh JW (2006) Cancer mortality rates among first and second generation migrants in The Netherlands: convergence toward the rates of the native Dutch population. Int J Cancer 119:2665–2672 5. Fang J, Madhavan S, Alderman MH (1996) Cancer mortality of Chinese in New York City 1988–1992. Int J Epidemiol 25: 907–912
123
6. Shimizu H, Ross RK, Bernstein L (1991) Cancers of the breast and prostate among Japanese and white immigrants in Los Angeles Country. Br J Cancer 63:963–966 7. Harding S, Rosatob M, Teyhan A (2009) Trends in cancer mortality among migrants in England and Wales, 1979–2003. Eur J Cancer 45:2168–2179 8. Spallek J, Arnold M, Hentschel S, Razum O (2009) Cancer incidence rate ratios of Turkish immigrants in Hamburg, Germany: a registry based study. Cancer Epidemiol 33:413–418 9. Harding S (2004) Mortality of migrants from the Caribbean to England and Wales: effect of duration of residence. Int J Epidemiol 33:382–386 10. Visser O, van Leeuwen FE (2007) Cancer risk in first generation migrants in North-Holland/Flevoland, The Netherlands, 1995–2004. Eur J Cancer 43:901–908 11. Wild S, Mckeigue P (1997) Cross sectional analysis of mortality by country of birth in England and Wales, 1970–92. Br Med J 314:705–710 12. McCredie M (1998) What have we learned from the studies of migrants. Cancer Causes Control 9:1–2 13. McMichael AJ, Giles GG (1998) Cancer in migrants to Australia: extending the descriptive epidemiological data. Cancer Res 48:751–756 14. Nicolaas H, Wobma E, Ooijevaar J (2010) Demografie van (nietwesterse) allochtonen in Nederland. In: van der Vliet R, Ooijevaar J, Boerdam A (eds) Jaarrapport integratie 2010. Statistics Netherlands, The Hague 15. Bindraban NR, Stronks K, Klazinga NS (2003) Cardiovasculaire risicofactoren bij Surinamers in Nederland: een literatuuroverzicht. Ned Tijdschr Geneeskd 147:1591–1594 16. Agyemang C, Stronks K, Tromp N, Bhopal R, Zaninotto P, Unwin N, Nazroo J, Kunst AE (2010) A cross-national comparative study of smoking prevalence and cessation between English and Dutch South Asian and African origin populations; the role of national context. Nicotine Tob Res 12:557–566 17. Grulich AE, McCredie M, Coates M (1995) Cancer incidence in Asian migrants tot New South Wales, Australia. Br J Cancer 71:400–408 18. Thomas DB, Karagas MR (1987) Cancer in first and second generation Americans. Cancer Res 47:5771–5776 19. Marmot MG, Adelstein AM, Bulusu L (1984). Lessons from the study of immigrant mortality. The Lancet 1455–1457 20. Shai D (1991) Cancer mortality in Cuba and among the Cuban-Born in the United States: 1979–1981. Public Health Rep 106:68–73 21. McCredie M, Williams S, Coates M (1999) Cancer Mortality in East and Southeast Asian migrants to New South Wales, Australia, 1975–1995. Int J Cancer 83:179–185 22. Zeeb H, Razum O, Blettner M, Stegmaier C (2002) Transition in cancer patterns among Turks residing in Germany. Eur J Cancer 38:705–711 23. Winkler V, Ott JJ, Holleczek B, Stegmaier C, Becher H (2009) Cancer profile of migrants from the former soviet union in Germany: incidence and mortality. Cancer Causes Control 20:1873–1879 24. Van der Sanden GA, Coebergh JWW, Schouten LJ, Visser O, van Leeuwen FE (1995) Cancer incidence in The Netherlands in 1989 and 1990. First results of the nationwide Netherlands Cancer Registry. Eur J Cancer 31:1822–1829 25. Mans DRA, Rijkaard E, Dollart J et al (2008) Differences between urban and rural areas of the Republic of Suriname in the ethnic and age distribution of cancer—a retrospective study from 1980 through 2004. Open Epidemiol J 1:30–35 26. Punwasi W (2009) Doodsoorzaken in Suriname 2007. Public Health Agency, Paramaribo 27. World Health Organization (2010) International statistical classification of diseases and related health problems, 10th revision. World Health Organization, Geneva
Cancer Causes Control 28. Ulm K (1990) A simple method to calculate the confidence interval of a standardized mortality ratio (SMR). Am J Epidemiol 131:373–375 29. General Bureau of Statistics, Census Office (2006) Census 2004 coverage evaluation. General Bureau of Statistics, Census Office 30. Uitenbroek DG, Verhoeff AP (2002) Life expectancy and mortality differences between migrant groups living in Amsterdam, The Netherlands. Soc Sci Med 54:1379–1388 31. Bos V, Kunst AE, Keij-Deerenberg IM, Garssen J, Mackenbach JP (2004) Ethnic inequalities in age-and cause-specific mortality in The Netherlands. Int J Epidemiol 33:1112–1119 32. Stirbu I, Kunst AE, Bos V, Mackenbach JP (2006) Differences in avoidable mortality between migrants and the native Dutch in The Netherlands. BMC Public Health 27:78 33. Parkin DM, Bray F, Ferlay J, Pisani P (2005) Global cancer statistics, 2002. CA Cancer J Clin 55:74–108 34. Kamangar F, Dores GM, Anderson WF (2006) Patterns of cancer incidence, mortality and prevalence across five continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol 24:2137–2150 35. Nierkens V, van der Vloeg MV, van Eer MY, Stronks K (2011) How do psychosocial determinants in migrant women in the Netherlands differ from these among their countersparts in their country of irigin? A cross-sectional study. BMC Public Health 11:397
¨ ry FG, van 36. Visser O, van der Kooy K, van Peppen AM, O Leeuwen FE (2004) Breast cancer risk among first-generation migrants in The Netherlands. Br J Cancer 90:2135–2137 37. Fracheboud J, Otto SJ, Van Dijck JA, Broeders MJ, Verbeek AL, De Koning HJ, National Evaluation Team for Breast Cancer Screening (NETB) (2004) Decreased rates of advanced breast cancer due to mammography screening in The Netherlands. Br J Cancer 91:861–867 38. Diepgen TL, Mahler V (2002) The epidemiology of skin cancer. Br J Dermatol 146:1–6 39. Lefkowitz ES, Garland CF (1994) Sunlight, vitamin D, and ovarian cancer mortality rates in US women. Int J Epidemiol 23:1133–1136 40. Lappe JM, Travers-Gustafson D, Davies KM, Recker RR, Heaney RP (2007) Vitamin D and calcium supplementation reduces cancer risk: results of a randomized trial. Am J Clin Nutr 85:1586–1591 41. International Association for Medical Assistance to Travellers (IAMAT) (2010) World Schistosomiasis Risk Chart. IAMAT, Toronto 42. Bosch X, Ribes J, Dı´az M, Cle´ries R (2004) Primary liver cancer: worldwide incidence and trends. Gastroenterology 127:5–16 43. Van Dijk MG (2008) Thesis on Hepatitis B in Suriname. Prince Leopold institute of tropical medicine, Antwerp
123
