Country of Birth (Nativity) and Cancer Diagnosis: Findings From The National Health and Nutrition Examination Survey (NHANES) 2011-2018

Purpose: Cancer incidence in the US remains higher among certain groups, regions, and communities and there are variations based on nativity. Research has primarily focused on specic groups and types of cancer. This study expands on previous studies to explore the relationship between country of birth (nativity) and all cancer site incidences among US and foreign-born residents using a nationally representative sample. Methods: This is a cross-sectional study of (unweighted n= 22,554; weighted n =231,175,933) participants between the ages of 20 and 80 from the National Health and Nutrition Examination Survey (NHANES) 2011-2018. Using weighted logistic regressions, we analyzed the impact of nativity on self-reported cancer diagnosis controlling for routine care, smoking status, overweight, race/ethnicity, age, and gender. We ran a partial model, adjusting only for age as a covariate, and a full model with all other covariates. Results: In the partial and full models, our ndings indicate that US-born individuals were more likely to report a cancer diagnosis compared to their foreign-born counterparts (OR = 2.34, 95% CI [1.93; 2.84], p<0.01), and (OR=1. 39, 95 % CI [1.05; 1.84], p < 0.05), respectively. There was a signicant association between cancer diagnosis and routine care (OR=1.48, 95% [1.14; 1.93], p<0.01), overweight (OR=1.16, 95% CI [1.01; 1.34], p<0.05), and smoking status (OR=1.30, 95% CI [1.13; 1.49], p<0.01). Race/ethnicity, age and gender were also signicantly associated with cancer diagnosis. Conclusion: A variety of factors may reect lower cancer diagnosis in foreign-born individuals in the US other than a healthy immigrant advantage, including environmental factors. examine the relationship between self-reported lifetime cancer diagnosis and nativity. This study seeks to expand on previous research examining nativity and cancer diagnosis to explore the relationship between all cancer types among immigrant racial groups and US-born, using a large national dataset. We further examined the potential healthy immigrant advantage in terms of self-reported cancer diagnosis by specic racial groups. To our knowledge, the relationship between all types of self-reported cancer diagnoses and nativity among the general immigrant population, using a large population-based dataset, has not been examined. The ndings of this study may lead further investigations seeking the disparities in self-reported based on nativity. signicantly (OR a sensitivity the group as the reference group. were no signicant in the age as a continuous variable. examine compared to in the for longer periods and birthplace.

explored the relationship between nativity and cancer mortality with the general nding of lower mortality among foreign-born [8,[16][17][18][31][32]. While previous work addressed the heterogeneity among and within different immigrant populations, there remain shared socioeconomic factors among this population and the overall immigrant experience. Immigrants often must overcome linguistic and other barriers and are affected by immigration policies associated with their status that directly affect their access to care and, ultimately, their health [3].
The purpose of this study is to examine the relationship between self-reported lifetime cancer diagnosis and nativity. This study seeks to expand on previous research examining nativity and cancer diagnosis to explore the relationship between all cancer types among immigrant racial groups and US-born, using a large national dataset. We further examined the potential healthy immigrant advantage in terms of self-reported cancer diagnosis by speci c racial groups. To our knowledge, the relationship between all types of self-reported cancer diagnoses and nativity among the general immigrant population, using a large population-based dataset, has not been examined. The ndings of this study may lead to further investigations seeking to explain the disparities in self-reported cancer diagnosis based on nativity.

Data
To determine the association between nativity and self-reported cancer diagnosis, we used data from the National Health and Nutrition Examination Survey (NHANES). The survey was conducted by the National Center for Health Statistics (NCHS) at the CDC and included nationally representative information on both the health and nutrition status of the general US population, which was obtained through self-reported personal interviews and physical examination each year for children and adults. Data used included those released in two-year cycles from the demographic and questionnaire data les which included Medical Conditions, Hospital Utilization and Access to Care, and Smoking-Cigarette Use. This retrospective cross-sectional study included four interview cycles, 2011-2012, 2013-2014, 2015-2016, and 2017-2018. After applying an age criterion of ages 20 to 80, and deleting missing observations, our nal sample size was n = 22,554 (unweighted); n = 231,175,933 (weighted).

Variables
The dependent variable, self-reported cancer diagnosis, was obtained from the Medical Conditions le. That variable was extracted from responses to the question, "Have you ever been told by a doctor or other health professional that you had cancer or a malignancy of any kind?" and categorized as a dichotomous variable with "1 = yes" and "0 = no." The independent variable, country of birth (nativity), was identi ed from responses to the question, ''In what country were you born?'' Respondents who indicated being born in one of the 50 United States or Washington DC were classi ed as US-born. Those who indicated that they were born in other countries were classi ed as foreign-born. The variable was coded as "1 = US-born" and "0 = foreign-born." We controlled for routine care, overweight, smoking status, race/ethnicity, gender, and age. The variable routine care was obtained from the Hospital Utilization and Access to Care data le. The question "Is there a place that you usually go when you are sick or need advice?" was categorized as a dichotomous variable. Those who answered "yes," and indicated that they had one or more than one place, were combined and recoded as "yes," "1 = yes" and "0 = no." The variable overweight was obtained from the Medical Conditions data le. The question "Has a doctor or other health professional ever told you that you were overweight?" was categorized as a dichotomous variable with "1 = yes" and "0 = no." The smoking variable was obtained from the Smoking-Cigarette Use le. We combined the questions, "Have you smoked at least 100 cigarettes in your lifetime" and "Do you now smoke cigarettes?" for the smoking variable. Responses of "no" to both questions were combined and classi ed as "nonsmokers." All other responses were classi ed as smokers, current or past. We categorized this as a dichotomous variable "1 = yes" for smokers and "0 = no" for nonsmokers. Race/ethnicity was categorized as non-Hispanic White, non-Hispanic Black, Mexican American, other Hispanic, Asian, and other race/ethnicity. We created six dummy variables for each racial/ethnic category. We used non-Hispanic White, which was the largest group, as the reference category. Gender was classi ed as male and female. Age was used as a continuous variable.

Data Analysis
All analyses were performed using the complex samples module in IBM SPSS version 27 to adjust for the clustered hierarchical sample designs of NHANES, using cluster, stratum, and sample weights provided by the NCHS. Descriptive statistics were used to examine respondent characteristics. Weighted logistic regressions were performed to model the association between nativity and the dependent variable, self-reported cancer diagnosis. To determine the effect of age given the age differences between US and foreign-born individuals, we used two models. Model 1 adjusted for nativity and age only as a covariate.
In Model 2, we added covariates routine care, overweight, smoking status, race/ethnicity, gender, and age.
We further examined the association between nativity and cancer diagnosis among racial groups. The statistical signi cance was established at p < 0.05. Table 1 presents the descriptive characteristics of the sample. The weighted sample included 82% US-born and 18% foreign-born. Approximately 11% of the sample indicated that they had a cancer diagnosis. The percentage of US-born individuals who had a self-reported cancer diagnosis was more than twice that of foreign-born individuals (12% vs. 5%, p<0.001). The unweighted mean age of the sample was 49.75, median 50, and interquartile range 30. The weighted average age of the sample was 47.78. On average, foreign-born natives were younger than US-born. The weighted average age was 45.56 among foreign-born compared to 48.28 among US-born natives. More than one-third (64%) of the foreign-born natives were between the ages of 20 and 50, compared to (54%) US-born. A larger percentage of foreign-born was within the age group 51-64 versus US-born (36.3 vs. 26.3, p<0.001). Within age group 65-80, US-born accounted for a larger percentage (20.5% vs. 13.8%, p<0.001). The percentage of females was slightly higher than males (52% vs. 48%). There was almost an equal percentage of foreign-born and US-born males (49% vs. 48%) and an even split among foreign-born and US-born females (52% vs. 52%). Most of the weighted sample was non-Hispanic White (65%), followed by non-Hispanic Black (11%). Mexican Americans and other Hispanics accounted for 15%, with 9% and 6% for each group, respectively. US-born non-Hispanic Whites account for a higher percentage than foreign-born Whites (75% vs. 15%, p< 0.001). US-born non-Hispanic Blacks made up a larger percentage of the sample than foreign-born Blacks (13% vs. 7%, p<0.001). On the other hand, US-born Mexican Americans made up a smaller percentage than foreign-born Mexican Americans (5% vs. 26%, p<0.001). There was a similar pattern with US-born Other Hispanics vs foreign-born Other Hispanics (3% vs. 26. %, p < 0.001), US-born Asians vs. foreign-born Asians (1% vs. 26%, p<0.001) and those who were US-born of other races vs. foreign-born (2% vs. 3%, p<0.001).

Results
Most of the respondents received routine care in at least one location (84%). US-born individuals received routine care at a higher rate than foreignborn (85% vs.77%, p <0.001). Over one-third (36%) of the sample reported being told that they were overweight by their doctor and more US-born individuals reported being overweight compared to foreign-born (40% vs. 27%). Less than half of the sample were smokers who either smoked in the past or were current smokers (43%). However, US-born individuals were more likely to be smokers than foreign-born (46% vs. 31%, p = <0.001). Table 2 shows the multivariate logistic regression examining the association between nativity and cancer diagnosis after adjusting for routine care, being overweight, smoking status, race/ethnicity, gender, and age. In Model 1, we adjusted for age as a continuous variable using logistic regression. We found that nativity was significantly associated with a cancer diagnosis. US-born individuals were more likely to report a self-

Discussion
The ndings of this study reveal an association between nativity and self-reported cancer diagnosis for all racial groups combined. Compared to USborn natives, foreign-born individuals were less likely to have reported a cancer diagnosis. The ndings seem to validate the healthy immigrant paradox, which suggests that immigrants experience better health outcomes than US-born notwithstanding access to fewer resources [33][34], and show that it potentially extends to all racial/ethnic groups. Better health outcomes among foreign-born may also be attributed to the selective process that occurs during the immigration process. Only those who pass stringent health examinations are allowed entry in the US [35]. While those who enter the country without legal documents may bypass this health screening process, their health may be protected by cultural factors [3,14].
Additionally, the various pathways to entry which include, family relationships, having a particular skill set, or humanitarian protection may directly affect overall health status [19][20].
Mexican Americans and other Hispanics account for only 15 percent of the study sample, but almost half of the foreign-born in the sample. The Hispanic paradox of cultural and social advantages among this group that results in better health outcomes than non-Hispanic Whites despite lower socioeconomic status may also explain our ndings. It should be noted, however, that the protection may weaken over time with acculturation. The health of Hispanics is also said to be "protected" by cultural factors, including traditional diet and support of family [36] and others, lending support to our ndings.
As age increased, there was a higher likelihood of a cancer diagnosis, which is consistent with the development of cancer over time. The foreignborn group was slightly younger than the US-born group, and a higher percentage of US-born were in the 65 and older age category, which may explain our ndings of a lower odds of a self-reported cancer diagnosis among foreign-born individuals. The ndings on gender re ect reports on cancer incidence among gender which report men having higher cases on some site-speci c cancers than women in general and variations in gender-speci c cancer [3].
In the fully adjusted model, Mexican Americans, other Hispanics, non-Hispanic Blacks, and Asians were signi cantly less likely to have a selfreported cancer diagnosis compared to non-Hispanic Whites, which con rms other studies, particularly for speci c cancer types [24,27]. However, when strati ed by race, we found that US-born non-Hispanic Blacks were more likely to report a cancer diagnosis compared to foreign-born non-Hispanic Blacks. Hispanics and Asians account for the largest percentage of immigrants in the United States. Blacks account for a smaller percentage of immigrants and are more likely to be more recent immigrants who have maintained some protective social factors, such as particular dietary and behavioral patterns [20]. This may also be explained by the more stringent health screening in the selection process resulting in better health among this group.
Health insurance is a key factor in access to preventive screening, early diagnosis, and treatment of cancer [2]. In this study, we controlled for routine care, assuming those who have routine care or usual source of care would have health insurance and hence access to care. We found a signi cant association between self-reported cancer and routine care. Those who receive routine care were more likely to report a cancer diagnosis. This nding is not surprising as access and routine care result in diagnosis at the early, asymptomatic stage. Hence, the higher reported cases of cancer among individuals receiving routine care. Among racial groups, non-Hispanic Blacks and Mexican Americans who received routine care were signi cantly more likely to report having a cancer diagnosis which may simply suggest detection at screening.
It is important to note that although there may not be a cancer diagnosis, cancer may still be present as the frequency of screening differs among populations. Several studies report late-stage cancer diagnosis or disparity in screening among immigrants and those lacking health insurance or usual source of care [30,[37][38][39]. About a quarter of the foreign-born individuals in this study did not have access to routine care, and thus may have had missed opportunities to diagnose cancers that were asymptomatic or sub-clinical.
While approximately ve to ten percent of all cancers may be due in part to genetic defects, more cancers may be attributed to environmental or behavioral factors [40]. Smoking is a risk factor for lung, liver, and colorectal cancer. It is responsible for 80% of lung cancer, one of the most common types of cancer in the United States [3]. Our ndings of a signi cant association between smoking and cancer diagnosis are supported by studies pointing to environmental factors including behavior or lifestyle as the differentiating contributing risk factors to cancer. However, there are fewer smokers among foreign-born as we found in our study. This is validated by previous studies that have shown that US-born individuals are more likely to be smokers and at a greater risk for lung cancer [16,41]. Smoking, however, generally increases with acculturation diminishing the differences in health outcomes of new immigrants and US-born individuals. Individuals who are overweight are at a greater risk for many diseases, including cancer, particularly breast and colon cancer [3]. In our study, there was a small percentage of respondents who indicated being told that they were overweight. However, there was a signi cantly higher odds of a cancer diagnosis among those who reported being overweight. Our nding is supported by previous studies that show being overweight as a factor contributing to the increased risk of cancer [42][43][44][45].
Strengths of this study include its nationally representative nature of the NHANES data and the large sample size, yet the cross-sectional nature of the data does not allow for the observation of changes over time among the participants. Additionally, the self-reported nature of the data may result in a potential error with the data, such as recall bias or social desirability in survey responses. NHANES data do not provide information on speci c birthplace, but rather on whether individuals were born in the United States or another country. Therefore, in this study, Hispanics and Asians were presented as subgroups without capturing their heterogeneity and variations in health outcomes related to cancer [46][47]. Additionally, due to data limitations, we did not examine the length of time individuals had been in the United States, which may have had an impact on acculturation, socioeconomic status, and access to care. The scope of this study included self-reported cancer diagnosis or incidence and not prevalence, which is a function of disease incidence and survival time given the data. Additionally, the data did not include a speci c time frame for cancer diagnosis, and we were not able to capture the speci c types of cancer. Several studies have shown that immigrants who have resided in the United States for a longer period lose the positive immigrant advantage on health over time [48][49]. Future longitudinal studies may examine cancer diagnosis among recent immigrants compared to those who have resided in the US for longer periods and speci c birthplace.

Conclusions
This study nds that foreign-born individuals have a signi cantly lower likelihood of being diagnosed with all-site cancer compared to US-born and may re ect a healthy immigrant advantage. However, other factors may be driving these ndings, such as actual differences in environmental or behavioral exposures, lower rates of cancer screening, and a disparity in identifying cancers at the earliest and most curable stages. Future longitudinal studies are needed to further elucidate the factors behind the differences in cancer diagnoses based on nativity status.