Cancer mortality by country of birth and cancer type in Sweden: A 25‐year registry‐based cohort study

Abstract Numerous studies have reported lower overall cancer mortality rates among immigrants compared to native populations. However, limited information exists regarding cancer mortality among immigrants based on specific birth countries and cancer types. We used population‐based registries and followed 10 million individuals aged 20 years or older in Sweden between 1992 and 2016. The Cox proportional hazard model was used to explore the disparities in cancer mortality by country of birth and cancer type, stratified by gender. Age‐standardized mortality rates were also computed using the world standard population. Hazard ratio (HR) of all‐site cancer was slightly lower among immigrants (males: HRm = 0.97: 95% confidence interval: 0.95, 0.98; females: HRf = 0.93: 0.91, 0.94) than Swedish‐born population. However, the immigrants showed higher mortality for infection‐related cancers, including liver (HRf = 1.10: 1.01, 1.19; HRm = 1.10: 1.02, 1.17), stomach (HRf = 1.39: 1.31, 1.49; HRm = 1.33: 1.26, 1.41) cancers, and tobacco‐related cancers, including lung (HRm = 1.44: 1.40, 1.49), and laryngeal cancers (HRm = 1.47: 1.24, 1.75). The HR of mesothelioma was also significantly higher in immigrants (HRf = 1.44: 1.10, 1.90). Mortality from lung cancer was specifically higher in men from Nordic (HRm = 1.41: 1.27, 1.55) and non‐Nordic Europe (HRm = 1.49: 1.43, 1.55) countries and lower in Asian (HRm = 0.78: 0.66, 0.93) and South American men (HRm = 0.70: 0.57, 0.87). In conclusion, there are large variations in cancer mortality by country of birth, and cancer type and require regular surveillance. Our detailed analyses lead to some novel findings such as excess mortality rate of mesothelioma and laryngeal cancers in Immigrants in Sweden. A targeted cancer prevention program among immigrants in Sweden is needed.


| INTRODUCTION
Cancer poses a major global public health problem, accounting for nearly 10 million deaths in 2020, 1 and it is the second most common cause of death in the Swedish population, with nearly 24,000 deaths in 2021. 2 Sweden's growing and aging immigrant population, now constituting approximately 20% of the total population, underscores the importance of prioritizing cancer research and health care for immigrants. 3esearch in the field of migration and health has found that most migrant groups live longer than the nonimmigrant population despite occupying lower socioeconomic positions, a phenomenon known as the "healthy immigrant paradox". 4[7][8][9][10][11][12] The risk of dying due to malignancies is higher in males, as compared to females, for a majority of cancer types 13 Socioeconomic status also plays a significant role in cancer mortality disparities. 7Immigrants with a lower socioeconomic position (SEP) may have limited access to preventive healthcare services, early cancer detection, and high-quality cancer treatment. 14Health risk behaviors (e.g., smoking and alcohol consumption) and environmental risks (poor housing, air pollution, workrelated health hazards) related to a higher risk of cancer mortality might also be higher among immigrants with low SEP. 15ost of studies on cancer among immigrants have primarily focused on the incidence and mortality disparities in all cancer sites 7,12 or in more common cancer sites. 8,10,11,16In addition, they have examined the disparities across immigrants by nativity (immigrants vs. nonimmigrants) or broad groups of immigrants according to their region of birth. 11,16,17As the immigrant population grows in size and the person-time and age of the immigrants increase, it would be possible to investigate the disparities for rare cancers and at the country of origin.Moreover, the composition of the immigrant population is evolving, marked by distinct reasons for immigration and diverse country origins among more recent immigrants compared to those in previous decades.This shift makes it challenging to generalize previous findings to this new cohort of immigrants.Consequently, there is a noticeable gap in comprehensive studies that explore cancer mortality by country of birth, as highlighted by a recent systematic review on all-cause mortality rates in the Nordic countries. 18his study took advantage of the Swedish high-quality national registries to examine disparities in cancer mortality by country of birth, gender, and specific cancer type, while also considering confounding by socio-demographic factors.Such comprehensive knowledge is pivotal for the formation of health policies and preventive action aiming to reduce the cancer burden in immigrants.

| Data sources
This study utilized registry data linkage from three sources, including (1) The Cause of Death Register provided data on the date of death, and underlying and contributing causes of death using ICD-9 codes from 1992 to 1996 and ICD-10 codes from 1997 onwards. 19The quality of the cause of death register is generally considered as high and being a valuable resource for public health research and policy; 20 (2) The Longitudinal Database of Health, Insurance, and Labor Market Studies (LISA) maintained by Statistics Sweden compiles annual data covering the Swedish population aged ≥16 years registered on December 31 each year since 1990 (since 2010 individuals aged 15 years included).We used this registry to obtain information on socioeconomic factors including education, disposable income, and civil status; 21 (3) The Total Population Registry is also maintained by Statistics Sweden and provides information on the entire population's dates of birth, death, sex, in and out-migration, and country of birth. 22These datasets were merged using a unique 10-digit personal identity number, which was then replaced by a serial number to ensure anonymity.The study did not require informed consent from participants and was approved by the Swedish Ethical Review Authority (decision 1: 2017/716-31 and decision 2: 2023-02550-02).mortality rate of mesothelioma and laryngeal cancers in Immigrants in Sweden.
A targeted cancer prevention program among immigrants in Sweden is needed.

K E Y W O R D S
cancer, immigrants, mortality, registered-based, Sweden

| Cohort
The study participants were enrolled using an opencohort design, for all individuals who had been residing in Sweden since January 1, 1992, and aged 20 years or older.Participants were followed up until they emigrated, died, or until the end of the study period, that is, on December 31, 2016, whichever came first.Individuals with incomplete data on the country of birth and those who had moved out of Sweden before 1992 were excluded from the study.Figure 1 shows the selection of the study cohort and the number of excluded participants with their respective origins.

| Variables (outcomes and covariates)
The person-years, measured in days, was determined by calculating the duration between the start and exit of the follow-up.For immigrants, the follow-up began 2 years after their arrival in Sweden.This adjustment was made to reduce the likelihood of including those immigrants who arrived in Sweden nearly at the endof-life stage with cancer.This commencement date was either January 01, 1992, for those arriving in Sweden before January 01, 1990, or 2 years after for those who arrived in Sweden after January 01, 1990.For individuals born in Sweden, the follow-up commenced on January 01, 1992.Immigrants were stratified into the largest immigrant groups in Sweden to ensure a sufficiently large analytical sample size: Nordic (Finland, Norway, Denmark), Non-Nordic Europe (Turkey, Germany, Austria, Netherlands, France, former Yugoslavia, Bosnia-Hercegovina, Italy, Spain, Greece, Croatia, Poland, Hungary, former Czechoslovakia, Romania, former Soviet Union, Russia, Estonia, UK and North Ireland), North America (USA), South America (Chile), Africa (Ethiopia, Somalia), Middle East (Iraq, Syria, Lebanon, Iran), and Asia (India, Vietnam, Thailand, and China).
The mortality rates were estimated for all cancers combined and separately for 16 cancer types in both sexes (lung, colorectal, pancreas, liver, stomach, brain and CNS, kidney, hematological (leukemia, Hodgkin lymphoma, non-Hodgkin lymphoma, and multiple myeloma), head and neck (oropharyngeal, hypopharyngeal, nasopharyngeal, lip, and oral), bladder, esophageal, melanoma, gall bladder, thyroid, mesothelioma, and laryngeal), four cancers specific to females (cervical, ovarian, corpus uteri, vulva, and vaginal), and prostate cancer in males.In the case of breast cancer, the analysis was restricted to women because of the small number of events among men.
To account for potential confounding factors for disparities of cancer mortality rate ratios between immigrants and non-immigrants, covariates including attained age, education (grouped into primary education, secondary education, post-secondary education or tertiary education, and unknown/missing), disposable income (categorized into five groups based on quintiles), marital status (single, married, divorced, or widowed), and calendar period (with 5 years interval) were utilized for adjustment.All socioeconomic characteristics were observed at the exit date of the study.

| Statistical methods
We calculated the age-standardized mortality rate (males: ASMR m , females: ASMR f ) per 100,000 personyears using the world standard population for all sites combined and 22 specific cancer types.ASMRs for all cancer sites combined are reported in this publication, and the results for specific cancer types are available in Tables S1 and S2.
The Cox proportional hazard model, using age as the underlying time scale, was employed to determine the relative risk of mortality, represented as hazard ratios (males: HR m , females: HR f ) and corresponding 95% confidence intervals (CI), in immigrant populations by country of birth.In all estimates, the reference group for comparison was individuals born in Sweden.The HRs of cancer mortality by country of birth were estimated for three aggregated levels of immigrants-all immigrants together, immigrants grouped based on region and specific country of birth.HRs were only reported for countries with more than at least five deaths reported during the follow-up.We used forest plots to visualize the HRs on a log scale for the top 9 common cancer types in both genders.However, results for all other cancer types are provided as Data S1.Two Cox regression models were constructed to evaluate the HRs by cancer type for all immigrant groups.Model 1 adjusted only for age, while Model 2 adjusted for all covariates-age, income, education, marital status, and calendar year.A pvalue of <0.05 was used to declare statistical significance.
All analyses were stratified by gender and performed using Stata software (Stata Ver.17, Stata Corp, College Station, Texas 77,845 USA).

| RESULTS
The study cohort included 10,011,841 individuals, of which 18% were foreign-born individuals and 50.2% were female (Table 1).A higher proportion of immigrants were either married or divorced, had completed tertiary education, and were younger at entry of the follow-up compared to Swedish-born individuals.A higher percentage of the immigrants had very low disposable income compared with non-immigrants.Across the calendar periods, the proportion of cancer deaths is relatively constant in Swedish-born individuals whereas on average increased by 0.4% in immigrants annually (Table 1).

| All-cancermortality rates
A total of 532,873 cancer deaths were reported in Sweden during the study period, from which 52,168 deaths were registered for the immigrant population.The ASMRs for all cancer types combined were 81.8 and 113.7 per 100,000 person-time in female and male immigrants, respectively.The rates were lower in females but higher in male immigrants than the Swedish-born population (ASMR f = 86.4;ASMR m = 106.4per 100,000).
Compared to Swedish-born individuals, HRs were somewhat lower both in immigrant women (HR f = 0.93, 95% CI: 0.91, 0.94) and men (HR m = 0.97: 0.95, 0.98), and also among immigrants from non-Nordic European countries, North America, South America, Africa, Middle East, and Asia.However, male immigrants from the Nordic countries had 11% higher overall cancer mortality rates compared to the Swedish-born population (HR m = 1.11: 1.09, 1.13).The highest age-standardized cancer mortality rate was observed among immigrants from the Nordic region (ASMR f = 91.3;ASMR m = 125.)followed by non-Nordic Europe (ASMR f = 80.5; ASMR m = 114.4).The lowest rate was observed in immigrants from the Middle East (ASMR f = 52.3;ASMRm = 79.1).Stratified estimates by calendar period show that mortality rates are generally been on a downward trend over the past decades for both immigrant groups and Sweden-born.By age groups, Nordic immigrants have shown the highest rates of cancer deaths, mainly in the mid and older age (Figures S1 and S2).
At the country level, we found that immigrants from Denmark (HR f = 1.17: 1.12, 1.23; HR m = 1.15: 1.10, 1.19), Norway (HR f = 1.04: 1.00, 1.08; HR m = 1.04: 1.00, 1.09), and male immigrants from Finland (HR m = 1.11: 1.09, 1.14) had significantly higher risk of all-site cancer deaths compared to the non-immigrants in Sweden.Except for Bosnian-Hercegovina and Estonia female (HR f = 1.17: 1.09, 1.25; HR f = 1.08: 1.00, 1.16, respectively) and male immigrants from former Yugoslavia (HR m = 1.06: 1.01, 1.11), Bosnia-Hercegovina (HR m = 1.13: 1.06, 1.21), and T A B L E 1 -Characteristics of participants in the study by gender and origin for all-site cancer mortality, Sweden, 1992  Estonia (HR m = 1.13: 1.06, 1.21) who had significantly higher risk of death due to all cancer types combined, most of other immigrants had a significantly lower risk of all-site cancer mortality compared to Swedishborn individuals, with the Middle East and Asia group being in the lowest rank (Table 2).The highest ASMR was observed among male immigrants from Bosnia-Hercegovina (ASMR = 142.7).

| Stomach cancer
The mortality rate of stomach cancer was significantly higher among immigrant men and women compared to Swedish-born individuals (HR f = 1.39: 1.30, 1.49; HR m = 1.33: 1.26, 1.41) (Table 3).At the country of birth level, a significantly elevated risk of stomach cancer morality was observed among male and female immigrants from the Nordic countries particularly from Finland, and some non-Nordic European countries, mainly countries in Central and South East Europe.The highest risk was observed for Russians (HR f = 3.19: 1.76, 5.79; HR m = 2.91: 1.31, 6.50) followed by Estonians (HR m = 1.98: 1.55, 2.53; HR f = 1.93: 1.44, 2.59).However, we observed a significantly lower mortality rate of stomach cancer in male immigrants from the United States (HR m = 0.35: 0.17, 0.73) (Figures 2 and 3).

| Liver cancer
We observed a 10% elevated mortality rate of liver cancer both in immigrant men (HR m = 1.10: 1.02, 1.17) and women (HR m = 1.10: 1.01, 1.19) compared to their Swedish-born counterparts (

| Esophageal cancer
Compared to Swedish-born individuals, the mortality rate of esophageal cancer was significantly lowered among male immigrants (HR m = 0.76: 0.69, 0.83) (Table 3).However, the risk of death due to this cancer was higher in female immigrants from Finland (HR f = 1.23: 1.02, 1.49), Somalia (HR f = 5.18: 2.29, 11.72), as well as male immigrants from Denmark (HR m = 1.33: 1.06, 1.67), and the UK and Northern  S2).

| Head and neck cancers
The risk of mortality stemming from lip, oral cavity, and pharyngeal cancers combined was significantly lower among male immigrants (HR m = 0.84: 0.76, 0.94) than Swedes (Table 3), whereas there was significantly higher risk of death among immigrants from specific countries/regions including Finland (HR m = 1.21: cancers was inconclusive, although those from Somalia (HR f = 5.40: 2.52, 11.55) and Austria (HR f = 2.67: 1.20, 5.95) showed a significantly higher risk (Table S2).

| Prostate cancer
Immigrant men exhibited a 30% lower risk for prostate cancer mortality (HR = 0.70: 0.68, 0.73) in comparison with men born in Sweden.This lower mortality rate in the immigrant group appeared consistent for most country-specific estimates, ranging from a 10% lower risk in immigrants from Finland (HR = 0.90: 0.84, 0.96) to a substantial 78% lower risk among Turkish men (HR = 0.22: 0.15, 0.33) (Figure 2).

| Brain cancer
The risk of brain and CNS cancer mortality was significantly lower among immigrant women (HR f = 0.82: 0.76, 0.89) and men (HR m = 0.74: 0.68, 0.79) compared to nonimmigrants (Table 3).The results were also consistent when analyzed at region or country levels, the risk being lowest among immigrants from South America (HR f = 0.33: 0.17, 0.63 and HR m = 0.31: 0.18, 0.55).However, male immigrants from France exhibited a significantly higher risk of death (HR m = 2.43: 1.44, 4.10) compared to the Swedishborn individuals (Figure 2; Table S2).

| Mesothelioma cancer
Compared to the non-immigrants, the mortality rate of mesothelioma cancer was significantly higher in female immigrants (HR f = 1.44: 1.10, 1.90) (Table 3).At the region or country level, male immigrants from the Nordic (HR m = 1.26: 1.03, 1.53) and female immigrants from Denmark (HR m = 2.96: 1.47, 5.98) exhibited a significantly higher mortality rate.Turkish men (HR f = 11.92:6.16, 23.06) and women (HR m = 2.34: 1.35, 4.06) had an exceptionally higher risk of death for mesothelioma cancer (Tables S1 and S2).

| Hematological cancer
There were no significant differences on mortality rate of hematological malignancies among female immigrants compared to Sweden-born women (Table 3), except for immigrants from Bosnian (HR = 134: 1.02, 1.76), and those from Africa (HR = 1.47: 1.07, 2.02) who exhibited a significantly higher mortality rate.In men, we found a 5% lower mortality rate in all immigrants compared to the Swedish-born population (HR m = 0.95: 0.90, 0.99), and among Asian immigrants (HR m = 0.74: 0.57, 0.97) at the region level.An exception was Danish male immigrants who had a significantly higher mortality rate (HR m = 1.17: 1.02, 1.35) than non-immigrants (Figures 2 and 3).

| DISCUSSION
This study investigated disparities in cancer mortality by country of birth and cancer type in Sweden while also considering potential gender differences.The study highlights that immigrants generally experience a somewhat lower overall cancer mortality rate compared to Swedish-born.
This finding aligns with previous research, both from Sweden 6,8,9 and internationally, [23][24][25] which has consistently suggested that immigrants experience a reduced risk of all-cause cancer-related mortality.This is also in line with the "healthy immigrant paradox" suggesting that immigrants constitute a selection of younger and healthier individuals from their country of origin. 26By performing a more detailed analysis of specific cancer types by country of birth, we were able to show an unambiguous and more complex pattern.Some immigrant groups face an increased risk of mortality from specific cancers, most notably lung, liver, laryngeal, and stomach cancers; and the risk was lower for colorectal, breast, and prostate cancers in most immigrant groups, when compared to nonimmigrants.][29][30] The etiology and risk factors vary greatly by cancer type.For infection-related cancers like liver, stomach, and cervical cancer, pre-migration factors play a pivotal role in explaining the disparities in cancer mortality between immigrants and non-immigrants. 31Many immigrant groups encounter barriers to healthcare access in their home countries (such as vaccine availability or prevention of specific infections), which can result in delayed detection and treatment of infections that contribute to these cancers.This is particularly evident in the context of chronic hepatitis B virus (HBV) infections, which is the main risk factor for liver cancer in many regions of the world.HBV infection is more prevalent among immigrants due to lower vaccination and treatment rates in their countries of origin. 32In 2016, the European Centre for Disease Prevention and Control (ECDC) report showed that immigrants account for an estimated 25% of the HBV cases in the EU/EEA. 33A previous study in Sweden also shows that a significant proportion of hepatitis B carriers are immigrants from non-Western countries, including Asian and African nations. 34Chronic HBV infections are relatively rare among native Swedes, 35 and this contrast in the prevalence of HBV infections could explain the observed disparities in liver cancer mortality.Furthermore, due to limited healthcare availability and standards in many immigrant nations, chronic liver diseases like HBV infection can progress asymptomatically over an extended period. 36his can lead to advanced-stage liver cancer at the time of diagnosis which cannot be treated anymore.
Similarly, the prevalence of Helicobacter pylori (H.pylori) infection, a well-known risk factor for stomach cancer, is notably high (ranges from 60% to 90%) in low-resource countries, 37 and it persists at significant rates in some European regions, particularly Eastern and Southeastern Europe despite improvements in food preservation and storage. 38This high infection rate may contribute to the observed disparities in stomach cancer mortality between non-Nordic European immigrant groups and non-immigrants in Sweden.Furthermore, the age-standardized incidence and mortality rates of stomach cancer are higher in central-east and southern Europe than in the northern European countries. 39The delayed detection and treatment of H. pylori infections, combined with other lifestyle factors like diet, smoking, and excessive alcohol consumption, 39 may further exacerbate the risk of stomach cancer mortality among immigrants.However, some findings may require further investigation.For example, despite high H. pylori infection rates in Africa, the mortality rate of stomach cancer appears lower among African immigrants in Sweden.This phenomenon, often referred to as the "African enigma", 40 suggests that factors beyond H. pylori infection may be at play in this specific case.Additionally, the higher mortality rate of stomach cancer in the Nordic regions may not be explained by H. pylori infection, as low infection rates are reported in these countries.Lifestyle factors, such as smoking, as well as body weight which is associated with cardia and stomach cancer may explain this difference.
Estimating the mortality of laryngeal cancer was limited to few immigrant groups due to rarity of the disease.However, the available data suggested that certain immigrant groups, particularly male immigrants from Nordic and specific European countries, exhibited significantly higher mortality risk compared to non-immigrants.Likewise, a heightened risk of lung cancer mortality was observed among similar immigrant groups from Nordic and several European countries, as well as male immigrants from Middle Eastern countries.The results are consistent with prior research conducted in Sweden. 8The primary factors contributing to the increased mortality rates of both laryngeal and lung cancer appear to be cigarette smoking, which accounts for about 90% of worldwide mortality for laryngeal, 41 and more than 80% in the United States and France and 40% in sub-Sahara Africa for lung cancer. 42The higher prevalence of smoking in several non-Nordic European countries is possibly influenced by historical factors such as wartime and post-war periods, along with other lifestyle factors, and may explain the elevated mortality risk of this cancer among European immigrant groups compared to non-immigrants in Sweden where the prevalence of cigarette smoking is notably lower than most European countries. 43he study also revealed that certain immigrant groups, such as those from Africa, exhibited notably higher mortality rate of esophageal and head and neck cancers, compared to individuals born in Sweden.Incidence of esophageal and pharyngeal cancers has previously been reported as significantly higher among immigrant groups in Sweden. 44Given that these types of cancers are predominately associated with smoking, 45 the discrepancy might be attributed, in part, to the relatively low prevalence of tobacco smoking within the Swedish population.Moreover, early diagnosis and treatment can significantly impact the outcomes (progression and metastasis) of esophageal and head and neck cancers, and thus, differences in healthcare access before migration appear to play a crucial role and could contribute to disparities in mortality rates between immigrant groups and Swedish-born individuals.Lifestyle factors like alcohol consumption and unhealthy diet, infections (HBV and Epstein Barr virus), and low socioeconomic status could also further contribute to explaining these disparities.
The risk of mesothelioma mortality also exhibited a notably elevated risk among immigrants, particularly those originating from the Nordic countries and Turkey.Mesothelioma is a relatively rare cancer, primarily caused by asbestos exposure. 46To the best of our knowledge, no previous report is available on the incidence or mortality rates of Mesothelioma among immigrants in European countries.Some studies conducted in Sweden in the 1980s and studied the risk of Mesothelioma in immigrant cohorts with a history of exposure to occupational hazards in their home country.These studies reported a higher incidence of mesothelioma cancer among Turkish 9,47 and Danish 9 immigrant men in Sweden.A recent study using the Global Cancer Observatory database revealed that the highest incidence rate of mesothelioma cancer was observed in North Europe, possibly due to better diagnosis and more complete registration. 48Sweden banned the use of asbestos in new construction projects in the early 1980s; however, given its long latency period, often ranging from 20 to 50 years between asbestos exposure and the development of the disease, immigrants who moved to Sweden in the aftermath of World War II to work in industries such as construction, shipbuilding, mining, and insulation manufacturing may have been exposed to asbestos.This fact complemented by their pre-migration history of asbestos exposure might be the reason for excess mortality risk among these groups.It is, however, important to approach these findings with caution due to the relatively low number of reported deaths.
With some exceptions, such as the increased risk of cervical, pancreas, and colorectal cancers among some immigrant groups, immigrants typically experience a lower mortality rate for most other cancer types including breast, prostate, hematological, gynecological, melanoma, gall bladder, kidney, and thyroid cancers.This pattern has been confirmed in many other studies for breast, prostate, 8,9 and gynecological 9 cancers.As immigrants often maintain their cultural practices, dietary habits, and lifestyle choices with them when they move to a new country, these can influence the risk of mortality from these cancers.The "healthy migrant effect" may also further contribute to this trend.For example, compared to dietary habits in Western countries like Sweden, diets in Africa and Asia are characterized by their reduced consumption of processed foods and higher intake of fiber. 49his dietary contrast may be a contributing factor to the observed lower rates of colorectal cancer mortality among immigrants from these regions.Genetic factors, as some immigrant groups may have genetic predispositions that are associated with lower cancer incidence and mortality, 50 might contribute further to the observed lower risk of mortality, for example, in the case of breast and cervical cancers.However, the observed lower cancer mortality rates among immigrants are not a universal trend and can depend on a multitude of factors including adapting to a new way of life in the host country-the phenomenon called acculturation, and thus, in the long run, mortality advantage for certain cancer types may diminish.Therefore, an ongoing monitoring of this population segment regarding mortality risk disparities by cancer types is essential.These findings will be instrumental in enhancing our understanding of cancer epidemiology among immigrants in Sweden and developing targeted interventions and policies that aim at reducing the cancer burden in this segment of the population within the national cancer control program.Given that the observed mortality rates may be closely tied to the incidence of some cancers for immigrants in their home countries, it is also imperative to boost cancer screening initiatives and tobacco control measures among immigrants.Establishing a comprehensive surveillance system to enhance screening and early detection of cancer in newcomers would be a necessary step to consider.Moreover, the changing demographics in composition and aging migrant population in Sweden-in which cancer incidence and mortality are prevalent, also necessitate ongoing monitoring, and further studies to explore cancer mortality disparities by behavioral and socio-cultural and economic risk factors, considering age and calendar period effects, as well as the economic implications within the immigrant population are warranted.

| STRENGTH AND LIMITATIONS
The study used high-quality total population data with a long-term follow-up and provided comprehensive evidence on cancer mortality rates among distinct groups of immigrants in Sweden on various site-specific cancer types.Another strength is that the estimates that were adjusted for socio-demographic factors.Income and education status were strong contributing factors for all-cause mortality disparities between immigrants and Swedish-born individuals in a previous Swedish study. 7There are also limitations in this study that need to be considered.First, immigrants returning to their country of birth when they become seriously ill might cause bias and leading to underestimates of cancer mortality in immigrant groups, a phenomenon called salmon bias.Nevertheless, it is essential to consider opposing viewpoints also, as some scholars argue that since access to and quality of cancer health care in many immigrants' home nations are not advanced, immigrants may be less inclined to move back to their home countries.This is evidenced in a Danish study among immigrants, which showed a higher disease severity was associated with fewer emigrations. 51In our study, loss to follow-up due to emigration was highest among immigrants from France (~44%), while the lowest was from Syria (~3%) followed by Eritrea (~4%).For the majority of immigrant groups, loss to follow-up due to emigration was between 10% and 20%.Second, exclusively sociodemographic factors are used as confounders.However, it is important to note that estimates on cancer mortality rates are also determined by other factors such as health behaviors (e.g., smoking, diet, alcohol, and physical activity), medical-related factors, genetic risk factors, and environmental hazards.Thus, a comprehensive analysis that takes into account behavioral, cultural, socioeconomic, and biological factors, as well as the impact of acculturation or post-migration lifestyle changes, and barriers to healthcare access could potentially offer a relatively better understanding of cancer disparities in the immigrant population in Sweden.In addition, the study exclusively relied on cancer mortality data due to data limitations.Cancer mortality data alone might not provide a comprehensive understanding of cancer disparities.It is essential to consider factors such as incidence and survival ratesfor example, reporting mortality-to-incidence rate rations.

| CONCLUSIONS
This study reveals that while immigrant groups exhibited somewhat lower mortality rates for overall and some specific cancer types, certain groups face a higher risk of mortality, particularly for cancers linked to infections and tobacco use, as compared with non-immigrants.Policies and prevention actions should especially focus on cancers related to infections and tobacco to reduce cancer mortality in immigrants.

F I G U R E 1
Final study population and exclusion criteria, Sweden, 1992 to 2016.α The Longitudinal Database of Health, Insurance, and Labor Market Studies.

Table 3
Mortality risk by cancer type and gender in migrant groups, compared to Sweden-born individuals, Sweden, 1992 to 2016.