- Email: [email protected]
Cancer incidence in European migrants to New South Wales
Annals of Oncology 1: 219-225, 1990. © 1990 Kluwer Academic Publishers. Printed in the Netherlands.
Original article Cancer incidence in European migrants to New South Wales M. McCredie, M.S. Coates & J.M. Ford NSW Central Cancer Registry, NSW Cancer Council, North Ryde, New South Wales, Australia
Summary. The incidence of cancer in migrants to New South Wales (NSW) from Italy, Greece, Yugoslavia, Germany, the Netherlands, Poland and USSR has been compared with that in the Australian-born population using data from the NSW Central Cancer Registry for 1972-84. The indirectly age-standardized incidence ratios (SIR) in all seven countries were low for melanoma of skin and high for gastric cancer. Cancers of the colon, oesophagus and lip also tended to have low SIRs. Migrants from Italy, Greece and Yugoslavia had significantly less cancer at all sites than the native-bom Australians mainly due to low SIRs for cancers of colon, lung (except Yugoslavian-born men), prostate and, in men, 'head and neck' (excluding nasopharynx). Cancers of breast and testis were relatively less common in migrants from Italy and Yugoslavia. SIRs were high for cancers of bladder (in Italian-born men), liver (in Greek- and Yugoslavian-born men) and nasopharynx (in Greek-born men and Italian-bom men and women). Amongst migrants from the four more northerly European countries, ovarian cancer was relatively more common in women from Germany and Poland as was bladder cancer in men, but not women, from Germany and the Netherlands. Cancers which had significantly increased SIRs in one migrant group only were lung (Dutch-bom men), cervix uteri and body of uterus (German-born women), gallbladder and bile ducts (Polish-bom women), thyroid (Italian-bom women), connective and other soft tissue (Russian-bom men) and brain (Greek-bom men and women computed together). Lymphomas were relatively less common in men bom in Yugoslavia. Key words: cancer, relative incidence, migrant, Europe, New South Wales
Introduction Studies of Southern European migrants in Australia as a whole (mortality) [1-5] and in the State of South Australia (incidence) [5-7] have shown higher rates of stomach cancer but lower rates of melanoma of skin and cancers of the colon, and in women, of lung and breast, relative to the Australian-bom population. While cancer mortality of migrants to Australia from individual countries within continental Europe has been described [1-5] cancer incidence has not Migrants from Southern Europe comprise almost one quarter of the population of overseas-bom residents of New South Wales (NSW) and migrants from other countries within continental Europe, a smaller but substantial proportion. The large and representative database at the NSW Central Cancer Registry makes it possible to analyse cancer incidence by individual migrant groups within these regions. Thus, in this descriptive study for the period 1972-84, cancer incidence of migrants to NSW from Italy, Greece, Yugoslavia, Germany, the Netherlands, Poland and the USSR has been compared with that of Australian-bom residents. Methods The NSW Central Cancer Registry receives statutory notifications from hospitals and radiotherapy departments,
as well as pathology reports and death certificates, for all cases of invasive cancer which occur in NSW as described previously [8]. Carcinoma in situ and benign tumours (e.g. papillomas) are not registered, nor are skin cancers other than melanoma. During the period 1972-84, 183,232 new cases were reported. The following analysis was carried out on 177,167 cases (97% of the total) for whom country of birth was known. Insufficient data are available at the Registry to include duration of residence in Australia as a factor in the analysis. Seven countries from continental Europe met the criterion for this study, namely, more than 1000 new cases of cancer amongst their migrants living in NSW occurring in 1972-84. Age-standardized incidence ratios were calculated by the indirect method [9] using, as the reference population, Australian-bom residents of NSW at the mid-point of the period 1972-84. Populations stratified by sex, 5-year age group and country of birth were interpolated from unpublished data from the 1976 and 1981 censuses obtained from the Australian Bureau of Statistics. The population for the USSR included people from Ukraine but not those from the three Baltic states. Confidence limits were calculated assuming that the observed cases followed a Poisson distribution and, because of the number of comparisons, the level of significance was set at 1 %. Average annual cancer incidence rates, directly age-standardized to the 'world' population [10], were calculated for
220 the migrant groups so that comparisons could be made with the incidence rates for NSW and their countries of birth published in 'Cancer Incidence in Five Continents' [11].
the bladder (in men from Italy), liver (in men from Greece and Yugoslavia), nasopharynx (in men from Greece and both sexes from Italy), thyroid (in Italian-bom women) and brain (in Greek-bom men and women, combined SIR=154 with a 99% confidence interval of 105-217).
Results Germany, the Netherlands, Poland and the USSR During 1972-84, Italian, Greek, Yugoslav, German, Dutch, Polish and Russian (including Ukrainian) migrants comprised 5.6% of the NSW population and contributed 5.7% of the cancers diagnosed in NSW residents (Table 1). Populations of the first five migrant groups were somewhat younger than native-bom Australians but, by contrast, the Polish- and Russian-bom populations were considerably older. Table 1. Numbers of persons in New South Wales at the mid-point of 1972-94 and of cancers diagnosed during 1972-84 according to country of birth. Country of birth
Persons
Persons aged 2 65 yr
Cancers
Comparison with rates in countries of birth
Italy
78315(1.6%)
8.8%
2.843(1.6%)
Greece
49.479(1.0%)
6.0%
1.384(0.8%)
Yugoslavia
58.217(1.2%)
3.9%
1.270(0.7%)
Germany
34.686 (0.7%)
6.6%
1.300(0.7%)
Netherlands
24.595 (0.5%)
8.1%
1.040(0.6%)
Poland
17.386 (0.4%)
16.4%
1280(0.7%)
USSR
10.443 (0.2%)
29.8%
1.006(0.6%)
9.1%
135.692(76.6%)
Australia
3.942.079 (80.2%)
Total
4.916.926(100%)
By contrast, in migrants from the four countries from northwestern and eastern European, cancer at all sites combined was not significantly less common than in the Australianbom. The only cancer to show a significant trend was melanoma, low both in men and women. Cancers which were relatively more common were stomach and ovary (the occasional SIR not reaching significance), bladder in men, but not women, from Germany and the Netherlands, lung (Dutch-bom men), cervix uteri and body of uterus (German-bom women), gallbladder and bile ducts (Polishbom women) and connective and other soft tissue (Russianbom men).
177.167(100%)
Migrants from all seven countries had high age-standardized incidence ratios (SIR) for gastric cancer and low SIRs for melanoma of skin and for cancer of colon (Polishbom men were the exception in respect of colon cancer and the occasional SIR did not reach significance; Tables 2 and 3). Cancers of the lip, 'head and neck' (excluding nasopharynx), oesophagus and rectum and lymphomas tended to be less common in migrants than in the Australian-bom but the low SIRs generally did not reach significance. Italy, Greece and Yugoslavia Migrants from the three southern European countries had significantly less cancer at all sites combined than the native-bom Australians (Tables 2 and 3). In addition to the cancer pattern described above, significantly low relative incidence rates were found for cancers of the lung (except for men bom in Yugoslavia), prostate, breast and testis (except in those bom in Greece) and, in men, cancers of 'head and neck' (excluding nasopharynx) and lip. On the other hand, the SIRs were significantly high for cancers of
Directly age-standardized incidence rates in migrants to NSW were compared with those in their homeland for all countries except Greece and the USSR, for which there are no published rates [11] (Table 4). Rates for cancers of the stomach, lung and melanoma of skin more nearly reflected those of the migrants' countries of birth than of NSW. This was true also for breast cancer in the Southern European migrants but, in migrants from Germany, the Netherlands and Poland, the rates were closer to those of NSW. On the whole, rates of colon and prostate cancer tended to fall between the rates of NSW and their homelands with the exception of colon cancer in women bom in Italy and Germany and men from the Netherlands, in whom the rates were lower than either the country of origin or adoption. For migrants from Germany and Poland, the rates of ovarian cancer were clearly higher than in NSW or their homelands. This was true also for bladder cancer in men from all five countries for which the comparisons could be made and in women from Poland.
Discussion Two distinct patterns of cancer incidence were seen in the European migrants, one in those bom in the southern countries of Italy, Greece and Yugoslavia and the other, less uniform, in migrants from Germany, the Netherlands, Poland and the USSR. Common to all seven countries, however, were the high rates of stomach cancer and low rates of melanoma of skin and, except for Polish-bom men, colon cancer relative to the Australian-bom residents of NSW. A lesser exposure to the sun, particularly at a young age [12], than that of native-bom Australians may explain the low SIRs of melanoma and of Up in the fair-skinned north-western and eastern Europeans. The darker skin types of the southern European migrants may also contribute to
Table 2. Age-standardized cancer incidence ratios during 1972-84 for male migrants to New South Wales from Italy, Greece, Yugoslavia, Germany, the Netherlands, Poland and USSR1 Site (ICD-9 2 ) Up (140)
Italy 29(12-57)
Greece 35(11-82)
Yugoslavia 17 (3-54)
Germany 28(5-87)
Netherlands 29(5-90)
Poland
USSR
42 (12-102)
64(16-167)
Hi-nrl Xr ru-rt^ nCUU OL IKUL
(141-146, 148,149, 160) Nasopharynx (147) Oesophagus (ISO) Stomach (151) Colon (153) Rectum (154) Liver (155) Gallbladder4 (156) Pancreas (157) Larynx(161) Trachea, bronchus & lung (162) Connective3 (171) Melanoma (172) Prostate (185) Testis(186) Bladder (188) Kidney (189) Brain (191) Thyroid (193) Lymphomas (200-202) Leukaemias (204-208) All cancers (140-208)
68(48-94) 324 (144-625) 58(30-101) 152(121-188) 55 (42-70) 84 (65-108) 188(83-358) 114(52-213) 117(84-159) 70(43-107)
40(19-74) 378 (122-885) 41(11-108) 149 (104-206) 50 (33-73) 71 (45-106) 347 (134-713) 175 (65-376) 78 (40-135) 87(43-154)
27(11-57) 136 (15^99) 67 (23-148) 243(181-319) 57(38-82) 90(59-130) 353(139-736) 148(47-344) 131 (78-204) 102 (54-175)
38(14-81) 74(0-571) 72 (21-177) 161 (104-237) 89 (60-126) 95 (58-146) 138(15-499) 106(18-331) 142 (79-237) 93(40-182)
55 (25-103) 0(0-408) 10(0-71) 115(69-180) 51(30-80) 72(41-116) 42(0-310) 99(17-315) 117(62-201) 97(45-183)
63(35-103) 61(0-464) 74 (30-148) 190(140-253) 105 (78-137) 109(76-151) 162(42-423) 131(42-305) 121 (73-186) 62(27-118)
76(35-142) 228(11-1030) 60(15-157) 237 (170-321) 81(53-117) 84 (49-134) 231(49-643) 26(0-195) 80(35-153) 66(21-153)
85 (74-98) 147 (80-247) 14(8-23) 65 (52-80) 26 (7-53) 133(107-163) 95 (65-135) 93 (59-137) 116(42-246) 86(63-114) 126(91-168)
79 (63-98) 120 (44-258) 19 (10-35) 71 (51-96) 55(20-118) 118(82-163) 76 (39-134) 156(96-239) 126(33-333) 62 (36-100) 101 (59-162)
100(81-122) 66(17-174) 23 (12-39) 64(43-91) 36(11-84) 121 (84-169) 82(42-143) 100 (55-166) 78(13-247) 63(37-99) 81 (44-136)
111(88-139) 92 (20-253) 40(22-66) 87 (59-122) 103 (44-201) 170(118-237) 79 (34-156) 110(52-203) 32 (0-240) 90 (52-145) 92 (45-167)
154(128-184) 82 (14-257) 38(20-66) 78(52-111) 77(22-189) 172(120-237) 81 (36-155) 116(55-214) 150(25-466) 84 (46-140) 80(35-153)
89 (73-107) 177 (61-392) 25(11^*7) 71 (52-95) 124(42-354) 100(68-141) 121(72-190) 102(48-189) 289(85-714) 73(41-119) 103 (57-170)
93(72-117) 304(104-666) 30(10-68) 71(50-98) 55 (0-413) 120(78-176) 63(22-140) 76(19-198) 231(26-844) 87(42-158) 142(76-241)
81(76-86)
77(70-85)
82(75-90)
95 (85-105)
98(88-108)
93(85-101)
94(84-104)
'Standardized Incidence ratio for Australian-bom = 100. Nlnth Revision of the International Classification of Diseases. •'Oral cavity, pharynx and nasal cavity excluding nasopharynx. ^Gallbladder and extrahepao'c bile ducts. ^Connective and other soft tissue. 2
S3
to
Table 3. Age-standardized cancer incidence ratios during 1972-84 for female migrants to New South Wales rrom Italy, Greece, Yugoslavia, Germany, the Netherlands, Poland and USSR1 Site (ICD-92) Up (140) ncttu ot IKVH
(141-146,148,149,160) Nasopharynx (147) Oesophagus(150) Stomach (151) Colon (153) Rectum (154) Liver (155) Gallbladder4 (156) Pancreas (157) Larynx(161) Trachea, bronchus & lung (162) Connective5 (171) Melanoma (172) Breast (174) Cervix uteri (180) Body of uterus (182) Ovary (183) Bladder(188) Kidney(189) Brain (191) Thyroid (193) Lymphomas (200-202) Leukaemias (204-208) All cancers (140-208)
Italy 17 (0-124)
Greece
Germany
Netherlands
Poland
USSR
0(0-166)
70 (4-320)
0(0-189)
0(0-279)
0(0-279)
0(0-252)
66 (43-97) 45 (0-338) 150(76-265) 101(59-161) 60(3-281)
82(28-182) 518(56-1830) 0(0-90) 266(176-383) 55(34-82) 61 (31-108) 0(0-441) 154(53-339) 128 (62-232) 0(0-331)
69 (20-168) 181(1-1238) 39(2-182) 225 (137-349) 48 (28-76) 76(39-131) 92 (0-675) 222(86^56) 107 (44-216) 70(0-531)
119(51-234) 0(0-1060) 50(6-183) 237(151-355) 51 (30-78) 80(44-135) 298 (34-1098) 89(19-253) 125 (59-230) 0(0-353)
14 (0-104) 0(0-1358) 24 (0-177) 120(51-236) 75(45-116) 61(25-123) 0(0-757) 52 (3-238) 135(55-271) 88 (0-675)
81(23-200) 637(34-3091) 20(0-149) 192(106-318) 74(46-113) 100(54-170) 0(0-757) 291 (123-567) 97(35-208) 406(90-1179)
24(1-110) 622 (34-3091) 101 (26-265) 224(141-335) 70 (45-105) 105(60-171) 0(0-757) 229 (97-^(48) 161 (85-277) 84(0-«19)
65(42-95) 47(8-150) 14(7-24) 85(73-97) 86(63-116) 87(60-122) 82 (55-116) 79(43-130) 54(25-101) 111 (62-182) 200(127-298) 91 (61-130) 97 (56-155)
45 (20-86) 116(30-301) 14(6-28) 100(83-119) 99 (68-139) 80(45-132) 77 (43-126) 111 (52-205) 46 (12-120) 149 (76-263) 121 (57-223) 82(45-136) 128(66-222)
42 (17-85) 161 (51-372) 24(12-42) 60 (47-76) 95 (64-135) 90(51-146) 85(48-139) 103 (44-202) 75 (26-167) 87(34-180) 93(38-186) 62(30-112) 78 (32-158)
137(89-202) 176 (51^128) 31 (16-55) 98(80-118) 156(111-213) 152(101-219) 182(124-257) 122 (59-222) 124(58-229) 74 (24-174) 147 (65-283) 91 (48-152) 81 (32-169)
81 (40-144) 115(13-422) 36(17-68) 112(89-139) 94(53-153) 79(38-144) 146 (86-232) 72(21-177) 154(68-295) 95 (27-232) 32(2-150) 69 (28-140) 90(29-210)
62(28-116) 313(93-779) 25 (8-58) 82 (62-105) 127(74-203) 104(58-172) 166(102-254) 110(45-220) 69 (20-170) 147(54-315) 158(46-389) 72 (29-144) 79 (23-194)
134 (82-208) 141 (16-523) 44(17-92) 77(56-101) 133 (72-223) 100(53-168) 136(76-224) 112(50-215) 80(26-186) 132(43-310) 202 (58^89) 99 (48-179) 152 (72-280)
78(72-84)
83(75-92)
71 (63-80)
105(95-116)
91 (80-103)
95 (83-107)
103(91-116)
51 (21-102) 480(108-1415)
8(0-56) 171 (121-235)
48(35-66)
'Standardized incidence ratio for Australian-bom = 100. Ninth Revision of the International Classification of Diseases. 3 Oral cavity, pharynx and nasal cavity excluding nasopharynx. ^Gallbladder and extrahepatic bile ducts. 'Connective and other soft tissue.
2
Yugoslavia
223 their lower incidences of this cancer. In gastric cancer, also believed to have its origins early in life [13], the rates in the migrants were close to those in their countries of birth. Despite mortality from stomach cancer in Greece, Yugoslavia and the Netherlands being relatively low compared with other European countries [14], the raised SIRs for their migrants were not significantly different from those for the other countries considered in this study. The cancer patterns of the southern European migrants to NSW in 1972-84 accord with those found in similar migrants to Australia in another period, 1962-71 [3] and to another State, South Australia, during an overlapping decade (1977-86) [7]. Despite higher rates for stomach cancer, the incidence for cancer at all sites combined was low relative to that for native-bom Australians due mainly to lower rates for melanoma of skin and cancers of the colon, prostate, and to some extent, breast and lung. Dietary differences between Southern European countries and Australia, and the changes made by migrants towards the more 'affluent' eating habits prevalent here, described comprehensively by McMichael et al. [2, 5-7, 15, 16] may be sufficient to explain the lower relative rates of cancers of the colon, prostate and breast. With respect to breast cancer, Greek-bom women were an exception in that their incidence, and mortality [3], were not significantly lower than those of Australian-bom women. That the point estimate for the incidence ratio (100) was higher than the earlier mortality ratio (74) may be due in part to the later time period covered by the incidence study resulting in a longer average duration of residence in Australia of the cases. With time, breast cancer mortality in Italian and Yugoslav migrants has been shown to converge on the Australian rates [3]. Examination of the incidence data by individual country of birth, rather than for Southern Europe as a region, has allowed differences between the countries to become evident The low relative rate of lymphoma in Yugoslav-bom men was due to non-Hodgkin's lymphoma, SIR=51 (25-90), rather than to Hodgkin's disease, SIR=100 (41—201). While the incidence of all lymphomas in Slovenia is approximately half that in NSW [11], it is not known in which areas of Yugoslavia our migrants were bom. That thyroid cancer was more common in Italianbom women is supported by the earlier mortality data [3]. In Italy, this cancer is concentrated in areas of iodine deficiency; these include Calabria and parts of Sicily [17], two provinces of birth of large numbers of Italian migrants to Australia. However, the high SIR for brain cancer in Greek migrants is a new finding yet to be confirmed independently. High SIRs for nasopharyngeal cancer were found in men bom in Greece and migrants of both sexes from Italy. Although nasopharyngeal cancer is not notably common in these countries [11, 18] and high mortality ratios were not found in their migrants [3], published incidence rates in men for some nearby Mediterranean or Middle Eastern countries were high for non-Chinese populations. These were Malta (3.5 per 100,000) [19], Israel - bom in Africa or Asia (3.3) [11] and Kuwait - Kuwaitis (2.1) and non-
Kuwaitis (2.9) [11]. The SIR for nasopharyngeal cancer in Maltese-bom men in NSW was 847 with a 99% confidence interval of 286-2064 [20]. In respect of Polish migrants, incidence data in NSW in 1972-84 were similar to those for mortality in Australia during 1962-66 [1] published for cancers of all sites, stomach, lung and breast The majority of directly age-standardized rates fell between those of Poland and NSW, an exception in each period being the rates for all sites in women, which were higher man for either the country of birth or adoption. That the SIR for cancers at all sites in Polish-bom women was lower, while the age-standardized rate was higher, than for the Australian-bom can be explained by the older age structure of the Polish-born migrants. The high relative rates of two cancers associated with socioeconomic class accord with the higher social status of migrants from north-western and eastern European countries, as measured by die proportion with post-school qualifications [21]. These are ovary (in women from Germany and Poland) and gallbladder and extrahepatic bile ducts (in women from Poland; in women from USSR, the SIR of 229 was significant at the 95% level (confidence interval; 121-390) but not the 99% level). For biliary cancer, it is the high incidence rates recorded in Warsaw and Cracow rather than the lower rates of rural Nowy Sacz which are reflected. An east-west gradient is apparent in Poland for this cancer widi higher mortality rates in the west [20]. The excesses of cancers of the cervix uteri and body of uterus in German-bom women reflect the age-standardized rates published for two regions of Germany which are 50% higher than for NSW [11]. Widiout knowledge of the frequency of cancer of connective and other soft tissue in the USSR one cannot draw any conclusion about the high SIR in male Russian and Ukrainian migrants. While it may be an isolated finding, the SIR for women bom in neighbouring Poland of 313 was significant at the 95% level (confidence interval 128-656) but not at the 99% level. Directly age-standardized incidence rates were available for Slovenia in Yugoslavia, for three Italian provinces (rates in Ragusa being lower dian in the two northern provinces of Varese and Parma), for Hamburg and Saarland in Germany, for Eindhoven in the Netherlands and, in Poland, for Warsaw City, Cracow City and Nowy Sacz, a rural area south-east of Cracow. As the Australian Department of Immigration only records country of birth, no data are available as to the particular regions from which the migrant groups came, although it is known drat the majority of the Italian migrants are from the south. The pattern of two important smoking-related cancers was different Lung cancer was less common in migrants from southern Europe dian in native-bom Australians and more common in men bom in the Netherlands (previous mortality rates were high in men from Germany, Poland and the Netherlands [3]). On the other hand, bladder cancer in men was notable in that rates were generally higher amongst migrants than in either NSW or their homelands. Except in the case of the Dutch-bom men this is unlikely to
224 Table 4. Average annual standardized incidence rates (per 100.000) for various cancers in New South Wales (NSW), Italy, Yugoslavia, Germany, the Netherlands and Poland, 1978-82, 1 and migrants to NSW from those countries, 1972-84. NSW Site (ICD-9) 2
Italy3
M
F
Italian-born migrants to NSW M F
229.3
170.9
Period
All sites (140-208) 8
1978-82 1972-84
272.8 269.2*
207.2 206.5 9
Stomach (151)
1978-82 1972-84
1X9 12.0
5.9 5.6
18.0
9.1
1978-82 1972-84
23.7 24.3
20.1 20.5
20.2
9.8
1978-82 1972-84
53.4 52.2
11.3 9.9
43.7
6.8
Melanoma (172) 1978-82 1972-84
17.1 19.4
16.1 18.2
2.7
3.4
Colon (153) Lung(162)
Breast (174) Ovary (183) Prostate (185) Bladder (188)
144.6-2043
2.1-3.5
240.6
160.6
24.9
11.6
16.6
10.0
54.7
52
8.7
4.4
8.4-19.9 10.4-16.6 2.8-7.5 1.8-3.7
8.9 8.7
7.2
M
F
235.0
159.1
34.9
15.1
8.7
7.8
57.7
6.7
2.4
2.7 37.7
333 9.5
7.9-11.7 9.7 18.7
17.9-203 21.8
22.1 5.0 4.6
Yugoslavia4
46.7-59.6
1978-82 1972-84
17.1 15.0
193.2-334 5
32.9-80.5
45.6
1978-82 1972-84
F
12.1-20.8
53.1 53.1
33.8 33.9
M
19.8-44.0
1978-82 1972-84
1978-82 1972-84
Yugoslavian-born migrants to NSW M F
11.7-273 20.0
0.5^.0
93
1.6
3.7
be due to a unduly high prevalence of smoking among migrants (no data are available for migrant groups in NSW). Infections or bladder calculi are an improbable explanation as the ratio of squamous cell to transitional cell carcinomas was no higher in the migrants than in the Australian-bom. The occurrence of high rates in bladder rather than lung cancer suggests instead possible exposure to occupational risk factors. A substantial proportion of new arrivals in Australia, regardless of their educational level, may have obtained work in factories or hazardous industries. As the median duration of residence in Australia for migrants from these six countries was 20-29 years in 1986 [21], bladder carcinogenesis due to exposure since migration to chemicals acting as initiators or promoters is conceivable. Exceptions from the usual trend amongst migrants of a gradual convergence of many of their cancer rates towards those of the host country, such as these high rates for bladder cancer in men, point to possible environmental hazards which warrant investigation.
References 1. Staszewski J, McCall MO, Stenhouse NS. Cancer mortality in 1962-66 among Polish migrants to Australia. Br J Cancer 1971; 25: 599-610. 2. McMichael AJ, McCall MG, Hamhome JM et aJ. Patterns of gastrointestinal cancer in European migrants to Australia: the role of dietary change. Im J Cancer 1980; 25:431-7. 3. Armstrong BK, Woodings TL, Stenhouse NS et al. Mortality from cancer in migrants to Australia, 1962-1971. Perth: NHMRC Research Unit in Epidemiology and Preventive Medicine, The
University of Western Australia, 1983. 4. Young C. Selection and survival. Immigrant mortality in Australia. Canberra: Department of Immigration and Ethnic Affairs, 1986. 5. McMichael AJ, Giles GG. Cancer in migrants to Australia: extending the descriptive epidemiological data. Cancer Res 1988; 48: 751-6. 6. McMichael AJ, Bonett A. Cancer profiles of British and SouthernEuropean migrants. Exploring South Australia's Cancer Registry data. MedJAust 1981; 1:229-3Z 7. McMichael AJ, Bonett A, Roder D. Cancer incidence among migrant populations in South Australia. Med J Aust 1989; 150: 417-20. 8. McCredie M, Coates MS, Ford JM. The changing incidence of cancers in adults in New South Wales. Int J Cancer 1988; 42: 667-71. 9. Armitage P, Berry G. Statistical methods in medical research. Blackwells: Oxford, 1987:403-5. 10. Doll R. Comparison between registries. Age-standardized rates. In: Waterhouse J, Muir C, Correa P et al., eds. Cancer incidence in five continents. Vol. III. IARC Sci Publ No 18, IARC: Lyon, 1976: 453-9. 11. Muir C, Waterhouse J, Mack T et al., eds. Cancer incidence in five continents, Vol. V. IARC Sci Publ No 88, IARC: Lyon, 1987. 12. Holman CDJ, Armstrong BK, Heenan PJ et al. The causes of malignant melanoma: results from the West Australian Lions Melanoma research project In: Gallagher RP ed. Epidemiology of malignant melanoma. Berlin: Springer-Verlag, 1986. 13. Nomura A. Stomach. In: Schottenfeld D, Franmeni JF, eds. Cancer epidemiology and prevention. Philadelphia: WB Sannders, 1982; 194-207. 14. World Health Organization. World Health Statistics Annual. Geneva: WHO, 1980,1981, 1982. 15. McMichael AJ, Potter JD, Hetzel BS. Tune trends in colo-rectal cancer mortality inrelationto food and alcohol consumption: United States, United Kingdom, Australia and New Zealand. Int J Epidemiol 1979; 8: 295-303. 16. McMichael AJ. Diet, nutrition and disease. Med J Aujt 1985; 142: 121-4.
225 Table 4. (Continued)
German born migrants to NSW M F
281.4
217.1
19.2
13.5
22.8
10.1
60.0
12.9
8.5
5.1
Germany3
M
F
247.7,2963
213.0,213.3
23.6, 23.7 15.6, 21.1 65.1, 7X7 4.0,
3.7
Netherlands6
Dutch-bom migrants to NSW M F
275.6
191.7
16.3
7.1
14.7
15.4
85.7
85
7.6
10.4
11.7, 111
4.6,
4.2
4.0
1X8,
3
3 6 7 8 9
10.9
25.1
15.1
47.7
7.6
3.1
4.1
19.7 5.8 5.9
F
206.3-242.6
1245-1823
23.1-43.7
8.9-17.0
4.6-10.5
4.8-8.7
47.0-73.1
4.8-123
1.2-2^
2.1-3.8 18.4-39.6 105-12.0
11.6 17.4 10.3-13.8
283 3.5,
M
48.0
25.9
26.1 14.8, 20.9
4
23.5
12.6 26.5, 28.7
1
223.1
9.5
71.6
9.1
31.3 5.9
251.5
60.4
16.0
2
204.0
94.4
53.5
25.3
290.6
20.6
7.1
58.0, 56.8
F
20.7
15.5, 17.2 11.1,
M
Poland7
Polish-bom migrants to NSW M F
3.8
19.8 25.9
3.7
9.8-105
3.0 16.0
1.9-2.1
62
Directly standardized to the 'World' population (source: ref 11). Ninth Revision of the International Classification of Diseases. Range for provinces of Parma (1978-82), Ragusa (1981-82) and Varese (1978-81). Slovenia (1978-81). Hamburg (1978-79), Saarland (1978-82). Eindhoven (1978-82). Range for Warsaw City (198O-S2), Cracow City (1978-81) and Nowy Sacz Rural Area (1978-81). Excludes 173 (other skin cancers). Rates for Australian-bom only.
17. Cislaghi C, Decarli A, La Vecchia C et al. Italian atlas of cancer mortality. In: Boyle P, Muir CS, Grundmann E, eds. Recent results cancer res. Vol. 114. Cancer mapping. Heidelberg: Springer-Verlag, 1989:143-53. 18. Shanmugaratnam K. Nasopharynx. In: Schottenfeld D, Fraumeni JF, eds. Cancer epidemiology and prevention. Philadelphia: WB Saundens, 1982: 194-207. 19. Waterhouse J, Muir C, Shanmugaratnam K et al., eds. Cancer incidence in five continents, Vol. IV. IARC Sci Pub! No 42, IARC: Lyon, 1982. 20. McCredie M, Coates MS. Cancer incidence in migrants to New South Wales, 1972 to 1984. Sydney: NSW Cancer Council, 1989. 21. Australian Bureau of Statistics. Overseas-bom Australians-a statistical profile. Cat No 4112.0. Canberra: ABS, 1989. 22. Zatonski W, Tyczynski J, Becker N. Geographical distribution of cancer in Poland. Ire Boyle P, Muir CS, Grundmann E, eds. Recent results cancer res. Vol. 114. Cancer Mapping. Heidelberg: SpringerVerlag, 1989: 176-95.
Received 9 October 1989; accepted 16 November 1989. Correspondence to: Dr. Margaret McCredie NSW Central Cancer Registry NSW Cancer Council P.O. Box 380 North Ryde NSW 2113, Australia
Original article Cancer incidence in European migrants to New South Wales M. McCredie, M.S. Coates & J.M. Ford NSW Central Cancer Registry, NSW Cancer Council, North Ryde, New South Wales, Australia
Summary. The incidence of cancer in migrants to New South Wales (NSW) from Italy, Greece, Yugoslavia, Germany, the Netherlands, Poland and USSR has been compared with that in the Australian-born population using data from the NSW Central Cancer Registry for 1972-84. The indirectly age-standardized incidence ratios (SIR) in all seven countries were low for melanoma of skin and high for gastric cancer. Cancers of the colon, oesophagus and lip also tended to have low SIRs. Migrants from Italy, Greece and Yugoslavia had significantly less cancer at all sites than the native-bom Australians mainly due to low SIRs for cancers of colon, lung (except Yugoslavian-born men), prostate and, in men, 'head and neck' (excluding nasopharynx). Cancers of breast and testis were relatively less common in migrants from Italy and Yugoslavia. SIRs were high for cancers of bladder (in Italian-born men), liver (in Greek- and Yugoslavian-born men) and nasopharynx (in Greek-born men and Italian-bom men and women). Amongst migrants from the four more northerly European countries, ovarian cancer was relatively more common in women from Germany and Poland as was bladder cancer in men, but not women, from Germany and the Netherlands. Cancers which had significantly increased SIRs in one migrant group only were lung (Dutch-bom men), cervix uteri and body of uterus (German-born women), gallbladder and bile ducts (Polish-bom women), thyroid (Italian-bom women), connective and other soft tissue (Russian-bom men) and brain (Greek-bom men and women computed together). Lymphomas were relatively less common in men bom in Yugoslavia. Key words: cancer, relative incidence, migrant, Europe, New South Wales
Introduction Studies of Southern European migrants in Australia as a whole (mortality) [1-5] and in the State of South Australia (incidence) [5-7] have shown higher rates of stomach cancer but lower rates of melanoma of skin and cancers of the colon, and in women, of lung and breast, relative to the Australian-bom population. While cancer mortality of migrants to Australia from individual countries within continental Europe has been described [1-5] cancer incidence has not Migrants from Southern Europe comprise almost one quarter of the population of overseas-bom residents of New South Wales (NSW) and migrants from other countries within continental Europe, a smaller but substantial proportion. The large and representative database at the NSW Central Cancer Registry makes it possible to analyse cancer incidence by individual migrant groups within these regions. Thus, in this descriptive study for the period 1972-84, cancer incidence of migrants to NSW from Italy, Greece, Yugoslavia, Germany, the Netherlands, Poland and the USSR has been compared with that of Australian-bom residents. Methods The NSW Central Cancer Registry receives statutory notifications from hospitals and radiotherapy departments,
as well as pathology reports and death certificates, for all cases of invasive cancer which occur in NSW as described previously [8]. Carcinoma in situ and benign tumours (e.g. papillomas) are not registered, nor are skin cancers other than melanoma. During the period 1972-84, 183,232 new cases were reported. The following analysis was carried out on 177,167 cases (97% of the total) for whom country of birth was known. Insufficient data are available at the Registry to include duration of residence in Australia as a factor in the analysis. Seven countries from continental Europe met the criterion for this study, namely, more than 1000 new cases of cancer amongst their migrants living in NSW occurring in 1972-84. Age-standardized incidence ratios were calculated by the indirect method [9] using, as the reference population, Australian-bom residents of NSW at the mid-point of the period 1972-84. Populations stratified by sex, 5-year age group and country of birth were interpolated from unpublished data from the 1976 and 1981 censuses obtained from the Australian Bureau of Statistics. The population for the USSR included people from Ukraine but not those from the three Baltic states. Confidence limits were calculated assuming that the observed cases followed a Poisson distribution and, because of the number of comparisons, the level of significance was set at 1 %. Average annual cancer incidence rates, directly age-standardized to the 'world' population [10], were calculated for
220 the migrant groups so that comparisons could be made with the incidence rates for NSW and their countries of birth published in 'Cancer Incidence in Five Continents' [11].
the bladder (in men from Italy), liver (in men from Greece and Yugoslavia), nasopharynx (in men from Greece and both sexes from Italy), thyroid (in Italian-bom women) and brain (in Greek-bom men and women, combined SIR=154 with a 99% confidence interval of 105-217).
Results Germany, the Netherlands, Poland and the USSR During 1972-84, Italian, Greek, Yugoslav, German, Dutch, Polish and Russian (including Ukrainian) migrants comprised 5.6% of the NSW population and contributed 5.7% of the cancers diagnosed in NSW residents (Table 1). Populations of the first five migrant groups were somewhat younger than native-bom Australians but, by contrast, the Polish- and Russian-bom populations were considerably older. Table 1. Numbers of persons in New South Wales at the mid-point of 1972-94 and of cancers diagnosed during 1972-84 according to country of birth. Country of birth
Persons
Persons aged 2 65 yr
Cancers
Comparison with rates in countries of birth
Italy
78315(1.6%)
8.8%
2.843(1.6%)
Greece
49.479(1.0%)
6.0%
1.384(0.8%)
Yugoslavia
58.217(1.2%)
3.9%
1.270(0.7%)
Germany
34.686 (0.7%)
6.6%
1.300(0.7%)
Netherlands
24.595 (0.5%)
8.1%
1.040(0.6%)
Poland
17.386 (0.4%)
16.4%
1280(0.7%)
USSR
10.443 (0.2%)
29.8%
1.006(0.6%)
9.1%
135.692(76.6%)
Australia
3.942.079 (80.2%)
Total
4.916.926(100%)
By contrast, in migrants from the four countries from northwestern and eastern European, cancer at all sites combined was not significantly less common than in the Australianbom. The only cancer to show a significant trend was melanoma, low both in men and women. Cancers which were relatively more common were stomach and ovary (the occasional SIR not reaching significance), bladder in men, but not women, from Germany and the Netherlands, lung (Dutch-bom men), cervix uteri and body of uterus (German-bom women), gallbladder and bile ducts (Polishbom women) and connective and other soft tissue (Russianbom men).
177.167(100%)
Migrants from all seven countries had high age-standardized incidence ratios (SIR) for gastric cancer and low SIRs for melanoma of skin and for cancer of colon (Polishbom men were the exception in respect of colon cancer and the occasional SIR did not reach significance; Tables 2 and 3). Cancers of the lip, 'head and neck' (excluding nasopharynx), oesophagus and rectum and lymphomas tended to be less common in migrants than in the Australian-bom but the low SIRs generally did not reach significance. Italy, Greece and Yugoslavia Migrants from the three southern European countries had significantly less cancer at all sites combined than the native-bom Australians (Tables 2 and 3). In addition to the cancer pattern described above, significantly low relative incidence rates were found for cancers of the lung (except for men bom in Yugoslavia), prostate, breast and testis (except in those bom in Greece) and, in men, cancers of 'head and neck' (excluding nasopharynx) and lip. On the other hand, the SIRs were significantly high for cancers of
Directly age-standardized incidence rates in migrants to NSW were compared with those in their homeland for all countries except Greece and the USSR, for which there are no published rates [11] (Table 4). Rates for cancers of the stomach, lung and melanoma of skin more nearly reflected those of the migrants' countries of birth than of NSW. This was true also for breast cancer in the Southern European migrants but, in migrants from Germany, the Netherlands and Poland, the rates were closer to those of NSW. On the whole, rates of colon and prostate cancer tended to fall between the rates of NSW and their homelands with the exception of colon cancer in women bom in Italy and Germany and men from the Netherlands, in whom the rates were lower than either the country of origin or adoption. For migrants from Germany and Poland, the rates of ovarian cancer were clearly higher than in NSW or their homelands. This was true also for bladder cancer in men from all five countries for which the comparisons could be made and in women from Poland.
Discussion Two distinct patterns of cancer incidence were seen in the European migrants, one in those bom in the southern countries of Italy, Greece and Yugoslavia and the other, less uniform, in migrants from Germany, the Netherlands, Poland and the USSR. Common to all seven countries, however, were the high rates of stomach cancer and low rates of melanoma of skin and, except for Polish-bom men, colon cancer relative to the Australian-bom residents of NSW. A lesser exposure to the sun, particularly at a young age [12], than that of native-bom Australians may explain the low SIRs of melanoma and of Up in the fair-skinned north-western and eastern Europeans. The darker skin types of the southern European migrants may also contribute to
Table 2. Age-standardized cancer incidence ratios during 1972-84 for male migrants to New South Wales from Italy, Greece, Yugoslavia, Germany, the Netherlands, Poland and USSR1 Site (ICD-9 2 ) Up (140)
Italy 29(12-57)
Greece 35(11-82)
Yugoslavia 17 (3-54)
Germany 28(5-87)
Netherlands 29(5-90)
Poland
USSR
42 (12-102)
64(16-167)
Hi-nrl Xr ru-rt^ nCUU OL IKUL
(141-146, 148,149, 160) Nasopharynx (147) Oesophagus (ISO) Stomach (151) Colon (153) Rectum (154) Liver (155) Gallbladder4 (156) Pancreas (157) Larynx(161) Trachea, bronchus & lung (162) Connective3 (171) Melanoma (172) Prostate (185) Testis(186) Bladder (188) Kidney (189) Brain (191) Thyroid (193) Lymphomas (200-202) Leukaemias (204-208) All cancers (140-208)
68(48-94) 324 (144-625) 58(30-101) 152(121-188) 55 (42-70) 84 (65-108) 188(83-358) 114(52-213) 117(84-159) 70(43-107)
40(19-74) 378 (122-885) 41(11-108) 149 (104-206) 50 (33-73) 71 (45-106) 347 (134-713) 175 (65-376) 78 (40-135) 87(43-154)
27(11-57) 136 (15^99) 67 (23-148) 243(181-319) 57(38-82) 90(59-130) 353(139-736) 148(47-344) 131 (78-204) 102 (54-175)
38(14-81) 74(0-571) 72 (21-177) 161 (104-237) 89 (60-126) 95 (58-146) 138(15-499) 106(18-331) 142 (79-237) 93(40-182)
55 (25-103) 0(0-408) 10(0-71) 115(69-180) 51(30-80) 72(41-116) 42(0-310) 99(17-315) 117(62-201) 97(45-183)
63(35-103) 61(0-464) 74 (30-148) 190(140-253) 105 (78-137) 109(76-151) 162(42-423) 131(42-305) 121 (73-186) 62(27-118)
76(35-142) 228(11-1030) 60(15-157) 237 (170-321) 81(53-117) 84 (49-134) 231(49-643) 26(0-195) 80(35-153) 66(21-153)
85 (74-98) 147 (80-247) 14(8-23) 65 (52-80) 26 (7-53) 133(107-163) 95 (65-135) 93 (59-137) 116(42-246) 86(63-114) 126(91-168)
79 (63-98) 120 (44-258) 19 (10-35) 71 (51-96) 55(20-118) 118(82-163) 76 (39-134) 156(96-239) 126(33-333) 62 (36-100) 101 (59-162)
100(81-122) 66(17-174) 23 (12-39) 64(43-91) 36(11-84) 121 (84-169) 82(42-143) 100 (55-166) 78(13-247) 63(37-99) 81 (44-136)
111(88-139) 92 (20-253) 40(22-66) 87 (59-122) 103 (44-201) 170(118-237) 79 (34-156) 110(52-203) 32 (0-240) 90 (52-145) 92 (45-167)
154(128-184) 82 (14-257) 38(20-66) 78(52-111) 77(22-189) 172(120-237) 81 (36-155) 116(55-214) 150(25-466) 84 (46-140) 80(35-153)
89 (73-107) 177 (61-392) 25(11^*7) 71 (52-95) 124(42-354) 100(68-141) 121(72-190) 102(48-189) 289(85-714) 73(41-119) 103 (57-170)
93(72-117) 304(104-666) 30(10-68) 71(50-98) 55 (0-413) 120(78-176) 63(22-140) 76(19-198) 231(26-844) 87(42-158) 142(76-241)
81(76-86)
77(70-85)
82(75-90)
95 (85-105)
98(88-108)
93(85-101)
94(84-104)
'Standardized Incidence ratio for Australian-bom = 100. Nlnth Revision of the International Classification of Diseases. •'Oral cavity, pharynx and nasal cavity excluding nasopharynx. ^Gallbladder and extrahepao'c bile ducts. ^Connective and other soft tissue. 2
S3
to
Table 3. Age-standardized cancer incidence ratios during 1972-84 for female migrants to New South Wales rrom Italy, Greece, Yugoslavia, Germany, the Netherlands, Poland and USSR1 Site (ICD-92) Up (140) ncttu ot IKVH
(141-146,148,149,160) Nasopharynx (147) Oesophagus(150) Stomach (151) Colon (153) Rectum (154) Liver (155) Gallbladder4 (156) Pancreas (157) Larynx(161) Trachea, bronchus & lung (162) Connective5 (171) Melanoma (172) Breast (174) Cervix uteri (180) Body of uterus (182) Ovary (183) Bladder(188) Kidney(189) Brain (191) Thyroid (193) Lymphomas (200-202) Leukaemias (204-208) All cancers (140-208)
Italy 17 (0-124)
Greece
Germany
Netherlands
Poland
USSR
0(0-166)
70 (4-320)
0(0-189)
0(0-279)
0(0-279)
0(0-252)
66 (43-97) 45 (0-338) 150(76-265) 101(59-161) 60(3-281)
82(28-182) 518(56-1830) 0(0-90) 266(176-383) 55(34-82) 61 (31-108) 0(0-441) 154(53-339) 128 (62-232) 0(0-331)
69 (20-168) 181(1-1238) 39(2-182) 225 (137-349) 48 (28-76) 76(39-131) 92 (0-675) 222(86^56) 107 (44-216) 70(0-531)
119(51-234) 0(0-1060) 50(6-183) 237(151-355) 51 (30-78) 80(44-135) 298 (34-1098) 89(19-253) 125 (59-230) 0(0-353)
14 (0-104) 0(0-1358) 24 (0-177) 120(51-236) 75(45-116) 61(25-123) 0(0-757) 52 (3-238) 135(55-271) 88 (0-675)
81(23-200) 637(34-3091) 20(0-149) 192(106-318) 74(46-113) 100(54-170) 0(0-757) 291 (123-567) 97(35-208) 406(90-1179)
24(1-110) 622 (34-3091) 101 (26-265) 224(141-335) 70 (45-105) 105(60-171) 0(0-757) 229 (97-^(48) 161 (85-277) 84(0-«19)
65(42-95) 47(8-150) 14(7-24) 85(73-97) 86(63-116) 87(60-122) 82 (55-116) 79(43-130) 54(25-101) 111 (62-182) 200(127-298) 91 (61-130) 97 (56-155)
45 (20-86) 116(30-301) 14(6-28) 100(83-119) 99 (68-139) 80(45-132) 77 (43-126) 111 (52-205) 46 (12-120) 149 (76-263) 121 (57-223) 82(45-136) 128(66-222)
42 (17-85) 161 (51-372) 24(12-42) 60 (47-76) 95 (64-135) 90(51-146) 85(48-139) 103 (44-202) 75 (26-167) 87(34-180) 93(38-186) 62(30-112) 78 (32-158)
137(89-202) 176 (51^128) 31 (16-55) 98(80-118) 156(111-213) 152(101-219) 182(124-257) 122 (59-222) 124(58-229) 74 (24-174) 147 (65-283) 91 (48-152) 81 (32-169)
81 (40-144) 115(13-422) 36(17-68) 112(89-139) 94(53-153) 79(38-144) 146 (86-232) 72(21-177) 154(68-295) 95 (27-232) 32(2-150) 69 (28-140) 90(29-210)
62(28-116) 313(93-779) 25 (8-58) 82 (62-105) 127(74-203) 104(58-172) 166(102-254) 110(45-220) 69 (20-170) 147(54-315) 158(46-389) 72 (29-144) 79 (23-194)
134 (82-208) 141 (16-523) 44(17-92) 77(56-101) 133 (72-223) 100(53-168) 136(76-224) 112(50-215) 80(26-186) 132(43-310) 202 (58^89) 99 (48-179) 152 (72-280)
78(72-84)
83(75-92)
71 (63-80)
105(95-116)
91 (80-103)
95 (83-107)
103(91-116)
51 (21-102) 480(108-1415)
8(0-56) 171 (121-235)
48(35-66)
'Standardized incidence ratio for Australian-bom = 100. Ninth Revision of the International Classification of Diseases. 3 Oral cavity, pharynx and nasal cavity excluding nasopharynx. ^Gallbladder and extrahepatic bile ducts. 'Connective and other soft tissue.
2
Yugoslavia
223 their lower incidences of this cancer. In gastric cancer, also believed to have its origins early in life [13], the rates in the migrants were close to those in their countries of birth. Despite mortality from stomach cancer in Greece, Yugoslavia and the Netherlands being relatively low compared with other European countries [14], the raised SIRs for their migrants were not significantly different from those for the other countries considered in this study. The cancer patterns of the southern European migrants to NSW in 1972-84 accord with those found in similar migrants to Australia in another period, 1962-71 [3] and to another State, South Australia, during an overlapping decade (1977-86) [7]. Despite higher rates for stomach cancer, the incidence for cancer at all sites combined was low relative to that for native-bom Australians due mainly to lower rates for melanoma of skin and cancers of the colon, prostate, and to some extent, breast and lung. Dietary differences between Southern European countries and Australia, and the changes made by migrants towards the more 'affluent' eating habits prevalent here, described comprehensively by McMichael et al. [2, 5-7, 15, 16] may be sufficient to explain the lower relative rates of cancers of the colon, prostate and breast. With respect to breast cancer, Greek-bom women were an exception in that their incidence, and mortality [3], were not significantly lower than those of Australian-bom women. That the point estimate for the incidence ratio (100) was higher than the earlier mortality ratio (74) may be due in part to the later time period covered by the incidence study resulting in a longer average duration of residence in Australia of the cases. With time, breast cancer mortality in Italian and Yugoslav migrants has been shown to converge on the Australian rates [3]. Examination of the incidence data by individual country of birth, rather than for Southern Europe as a region, has allowed differences between the countries to become evident The low relative rate of lymphoma in Yugoslav-bom men was due to non-Hodgkin's lymphoma, SIR=51 (25-90), rather than to Hodgkin's disease, SIR=100 (41—201). While the incidence of all lymphomas in Slovenia is approximately half that in NSW [11], it is not known in which areas of Yugoslavia our migrants were bom. That thyroid cancer was more common in Italianbom women is supported by the earlier mortality data [3]. In Italy, this cancer is concentrated in areas of iodine deficiency; these include Calabria and parts of Sicily [17], two provinces of birth of large numbers of Italian migrants to Australia. However, the high SIR for brain cancer in Greek migrants is a new finding yet to be confirmed independently. High SIRs for nasopharyngeal cancer were found in men bom in Greece and migrants of both sexes from Italy. Although nasopharyngeal cancer is not notably common in these countries [11, 18] and high mortality ratios were not found in their migrants [3], published incidence rates in men for some nearby Mediterranean or Middle Eastern countries were high for non-Chinese populations. These were Malta (3.5 per 100,000) [19], Israel - bom in Africa or Asia (3.3) [11] and Kuwait - Kuwaitis (2.1) and non-
Kuwaitis (2.9) [11]. The SIR for nasopharyngeal cancer in Maltese-bom men in NSW was 847 with a 99% confidence interval of 286-2064 [20]. In respect of Polish migrants, incidence data in NSW in 1972-84 were similar to those for mortality in Australia during 1962-66 [1] published for cancers of all sites, stomach, lung and breast The majority of directly age-standardized rates fell between those of Poland and NSW, an exception in each period being the rates for all sites in women, which were higher man for either the country of birth or adoption. That the SIR for cancers at all sites in Polish-bom women was lower, while the age-standardized rate was higher, than for the Australian-bom can be explained by the older age structure of the Polish-born migrants. The high relative rates of two cancers associated with socioeconomic class accord with the higher social status of migrants from north-western and eastern European countries, as measured by die proportion with post-school qualifications [21]. These are ovary (in women from Germany and Poland) and gallbladder and extrahepatic bile ducts (in women from Poland; in women from USSR, the SIR of 229 was significant at the 95% level (confidence interval; 121-390) but not the 99% level). For biliary cancer, it is the high incidence rates recorded in Warsaw and Cracow rather than the lower rates of rural Nowy Sacz which are reflected. An east-west gradient is apparent in Poland for this cancer widi higher mortality rates in the west [20]. The excesses of cancers of the cervix uteri and body of uterus in German-bom women reflect the age-standardized rates published for two regions of Germany which are 50% higher than for NSW [11]. Widiout knowledge of the frequency of cancer of connective and other soft tissue in the USSR one cannot draw any conclusion about the high SIR in male Russian and Ukrainian migrants. While it may be an isolated finding, the SIR for women bom in neighbouring Poland of 313 was significant at the 95% level (confidence interval 128-656) but not at the 99% level. Directly age-standardized incidence rates were available for Slovenia in Yugoslavia, for three Italian provinces (rates in Ragusa being lower dian in the two northern provinces of Varese and Parma), for Hamburg and Saarland in Germany, for Eindhoven in the Netherlands and, in Poland, for Warsaw City, Cracow City and Nowy Sacz, a rural area south-east of Cracow. As the Australian Department of Immigration only records country of birth, no data are available as to the particular regions from which the migrant groups came, although it is known drat the majority of the Italian migrants are from the south. The pattern of two important smoking-related cancers was different Lung cancer was less common in migrants from southern Europe dian in native-bom Australians and more common in men bom in the Netherlands (previous mortality rates were high in men from Germany, Poland and the Netherlands [3]). On the other hand, bladder cancer in men was notable in that rates were generally higher amongst migrants than in either NSW or their homelands. Except in the case of the Dutch-bom men this is unlikely to
224 Table 4. Average annual standardized incidence rates (per 100.000) for various cancers in New South Wales (NSW), Italy, Yugoslavia, Germany, the Netherlands and Poland, 1978-82, 1 and migrants to NSW from those countries, 1972-84. NSW Site (ICD-9) 2
Italy3
M
F
Italian-born migrants to NSW M F
229.3
170.9
Period
All sites (140-208) 8
1978-82 1972-84
272.8 269.2*
207.2 206.5 9
Stomach (151)
1978-82 1972-84
1X9 12.0
5.9 5.6
18.0
9.1
1978-82 1972-84
23.7 24.3
20.1 20.5
20.2
9.8
1978-82 1972-84
53.4 52.2
11.3 9.9
43.7
6.8
Melanoma (172) 1978-82 1972-84
17.1 19.4
16.1 18.2
2.7
3.4
Colon (153) Lung(162)
Breast (174) Ovary (183) Prostate (185) Bladder (188)
144.6-2043
2.1-3.5
240.6
160.6
24.9
11.6
16.6
10.0
54.7
52
8.7
4.4
8.4-19.9 10.4-16.6 2.8-7.5 1.8-3.7
8.9 8.7
7.2
M
F
235.0
159.1
34.9
15.1
8.7
7.8
57.7
6.7
2.4
2.7 37.7
333 9.5
7.9-11.7 9.7 18.7
17.9-203 21.8
22.1 5.0 4.6
Yugoslavia4
46.7-59.6
1978-82 1972-84
17.1 15.0
193.2-334 5
32.9-80.5
45.6
1978-82 1972-84
F
12.1-20.8
53.1 53.1
33.8 33.9
M
19.8-44.0
1978-82 1972-84
1978-82 1972-84
Yugoslavian-born migrants to NSW M F
11.7-273 20.0
0.5^.0
93
1.6
3.7
be due to a unduly high prevalence of smoking among migrants (no data are available for migrant groups in NSW). Infections or bladder calculi are an improbable explanation as the ratio of squamous cell to transitional cell carcinomas was no higher in the migrants than in the Australian-bom. The occurrence of high rates in bladder rather than lung cancer suggests instead possible exposure to occupational risk factors. A substantial proportion of new arrivals in Australia, regardless of their educational level, may have obtained work in factories or hazardous industries. As the median duration of residence in Australia for migrants from these six countries was 20-29 years in 1986 [21], bladder carcinogenesis due to exposure since migration to chemicals acting as initiators or promoters is conceivable. Exceptions from the usual trend amongst migrants of a gradual convergence of many of their cancer rates towards those of the host country, such as these high rates for bladder cancer in men, point to possible environmental hazards which warrant investigation.
References 1. Staszewski J, McCall MO, Stenhouse NS. Cancer mortality in 1962-66 among Polish migrants to Australia. Br J Cancer 1971; 25: 599-610. 2. McMichael AJ, McCall MG, Hamhome JM et aJ. Patterns of gastrointestinal cancer in European migrants to Australia: the role of dietary change. Im J Cancer 1980; 25:431-7. 3. Armstrong BK, Woodings TL, Stenhouse NS et al. Mortality from cancer in migrants to Australia, 1962-1971. Perth: NHMRC Research Unit in Epidemiology and Preventive Medicine, The
University of Western Australia, 1983. 4. Young C. Selection and survival. Immigrant mortality in Australia. Canberra: Department of Immigration and Ethnic Affairs, 1986. 5. McMichael AJ, Giles GG. Cancer in migrants to Australia: extending the descriptive epidemiological data. Cancer Res 1988; 48: 751-6. 6. McMichael AJ, Bonett A. Cancer profiles of British and SouthernEuropean migrants. Exploring South Australia's Cancer Registry data. MedJAust 1981; 1:229-3Z 7. McMichael AJ, Bonett A, Roder D. Cancer incidence among migrant populations in South Australia. Med J Aust 1989; 150: 417-20. 8. McCredie M, Coates MS, Ford JM. The changing incidence of cancers in adults in New South Wales. Int J Cancer 1988; 42: 667-71. 9. Armitage P, Berry G. Statistical methods in medical research. Blackwells: Oxford, 1987:403-5. 10. Doll R. Comparison between registries. Age-standardized rates. In: Waterhouse J, Muir C, Correa P et al., eds. Cancer incidence in five continents. Vol. III. IARC Sci Publ No 18, IARC: Lyon, 1976: 453-9. 11. Muir C, Waterhouse J, Mack T et al., eds. Cancer incidence in five continents, Vol. V. IARC Sci Publ No 88, IARC: Lyon, 1987. 12. Holman CDJ, Armstrong BK, Heenan PJ et al. The causes of malignant melanoma: results from the West Australian Lions Melanoma research project In: Gallagher RP ed. Epidemiology of malignant melanoma. Berlin: Springer-Verlag, 1986. 13. Nomura A. Stomach. In: Schottenfeld D, Franmeni JF, eds. Cancer epidemiology and prevention. Philadelphia: WB Sannders, 1982; 194-207. 14. World Health Organization. World Health Statistics Annual. Geneva: WHO, 1980,1981, 1982. 15. McMichael AJ, Potter JD, Hetzel BS. Tune trends in colo-rectal cancer mortality inrelationto food and alcohol consumption: United States, United Kingdom, Australia and New Zealand. Int J Epidemiol 1979; 8: 295-303. 16. McMichael AJ. Diet, nutrition and disease. Med J Aujt 1985; 142: 121-4.
225 Table 4. (Continued)
German born migrants to NSW M F
281.4
217.1
19.2
13.5
22.8
10.1
60.0
12.9
8.5
5.1
Germany3
M
F
247.7,2963
213.0,213.3
23.6, 23.7 15.6, 21.1 65.1, 7X7 4.0,
3.7
Netherlands6
Dutch-bom migrants to NSW M F
275.6
191.7
16.3
7.1
14.7
15.4
85.7
85
7.6
10.4
11.7, 111
4.6,
4.2
4.0
1X8,
3
3 6 7 8 9
10.9
25.1
15.1
47.7
7.6
3.1
4.1
19.7 5.8 5.9
F
206.3-242.6
1245-1823
23.1-43.7
8.9-17.0
4.6-10.5
4.8-8.7
47.0-73.1
4.8-123
1.2-2^
2.1-3.8 18.4-39.6 105-12.0
11.6 17.4 10.3-13.8
283 3.5,
M
48.0
25.9
26.1 14.8, 20.9
4
23.5
12.6 26.5, 28.7
1
223.1
9.5
71.6
9.1
31.3 5.9
251.5
60.4
16.0
2
204.0
94.4
53.5
25.3
290.6
20.6
7.1
58.0, 56.8
F
20.7
15.5, 17.2 11.1,
M
Poland7
Polish-bom migrants to NSW M F
3.8
19.8 25.9
3.7
9.8-105
3.0 16.0
1.9-2.1
62
Directly standardized to the 'World' population (source: ref 11). Ninth Revision of the International Classification of Diseases. Range for provinces of Parma (1978-82), Ragusa (1981-82) and Varese (1978-81). Slovenia (1978-81). Hamburg (1978-79), Saarland (1978-82). Eindhoven (1978-82). Range for Warsaw City (198O-S2), Cracow City (1978-81) and Nowy Sacz Rural Area (1978-81). Excludes 173 (other skin cancers). Rates for Australian-bom only.
17. Cislaghi C, Decarli A, La Vecchia C et al. Italian atlas of cancer mortality. In: Boyle P, Muir CS, Grundmann E, eds. Recent results cancer res. Vol. 114. Cancer mapping. Heidelberg: Springer-Verlag, 1989:143-53. 18. Shanmugaratnam K. Nasopharynx. In: Schottenfeld D, Fraumeni JF, eds. Cancer epidemiology and prevention. Philadelphia: WB Saundens, 1982: 194-207. 19. Waterhouse J, Muir C, Shanmugaratnam K et al., eds. Cancer incidence in five continents, Vol. IV. IARC Sci Pub! No 42, IARC: Lyon, 1982. 20. McCredie M, Coates MS. Cancer incidence in migrants to New South Wales, 1972 to 1984. Sydney: NSW Cancer Council, 1989. 21. Australian Bureau of Statistics. Overseas-bom Australians-a statistical profile. Cat No 4112.0. Canberra: ABS, 1989. 22. Zatonski W, Tyczynski J, Becker N. Geographical distribution of cancer in Poland. Ire Boyle P, Muir CS, Grundmann E, eds. Recent results cancer res. Vol. 114. Cancer Mapping. Heidelberg: SpringerVerlag, 1989: 176-95.
Received 9 October 1989; accepted 16 November 1989. Correspondence to: Dr. Margaret McCredie NSW Central Cancer Registry NSW Cancer Council P.O. Box 380 North Ryde NSW 2113, Australia