Abstract
Aims
To identify the prevalence and mortality of type 2 diabetes in Asian Americans (Asians) vs. non-Hispanic whites (Whites).
Methods
We analyzed a nationally representative sample of 237,354 U.S. adults aged ≥30 years using National Health Interview Survey data from 2000–2014 to estimate the prevalence and trends of type 2 diabetes. Additionally, 144,638 Asians and Whites represented in surveys from 2000–2009 were included in the mortality analysis with follow-up to 2011.
Results
Type 2 diabetes was higher in Asians than Whites (7.0–11.2 vs. 5.6–8.3%) and increased over time. Prevalence rates increased from 8.1 (2000–2002) to 9.6% (2012–2014) in Asians and from 6.0 (2000–2002) to 7.9% (2012–2014) in Whites (both P<0.05). The age-standardized mortality rates were 72.7 and 138.8 per 1,000 person-years in Asians and Whites with diabetes, respectively, and 58.1 and 77.8 per 1,000 person-years, respectively, in those without diabetes. Among Asians and Whites with diabetes, hazard ratios for total and CVD mortality were 0.7 (95% CI: 0.5–0.9) and 0.3 (95% CI: 0.1–0.6), respectively, with no difference in cancer mortality. Asians and Whites without diabetes exhibited no differences in total or cause-specific mortality.
Conclusions
Type 2 diabetes was more prevalent in Asians, with a significant upward trend since 2000, but overall mortality was lower in Asians than Whites with diabetes. Asians are susceptible to type 2 diabetes; thus, prevention programs are still needed.
Keywords: type 2 diabetes, mortality, Asian Americans
1. Introduction
Asian Americans comprise 5% of the U.S. population (15 million people)1 and have diverse cultures, languages, and health issues.2–4 Although previous studies have shown that Asian Americans have an increased risk of type 2 diabetes mellitus compared to non-Hispanic whites (henceforth “Whites”) after accounting for their lower prevalence of obesity,5, 6 few studies have examined mortality among Asian Americans (henceforth “Asians”) with and without diabetes. We therefore analyzed data from the National Health Interview Survey (NHIS) and linked these data to mortality follow-up data to identify the prevalence and trends of type 2 diabetes and estimate total and cause-specific mortality in Asian vs. White adults with and without diabetes.
2. Subjects, materials and methods
2.1. Data source
The National Health Interview Survey (NHIS) provides data on the health status, health care access, and health behavior of the non-institutionalized civilian population in the U.S. using a multistage probability sampling design. The data are collected by trained interviewers with a computer-assisted personal interviewing (CAPI) program and are based on self-reports from the respondents. Details regarding the study design and procedure are available elsewhere.7
2.2. Study participants
To analyze prevalence and trends, we first identified a sample of 453,480 NHIS adult (18 years and older) participants who were interviewed between 2000 and 2014. We subsequently excluded individuals with any of the following criteria (individuals could be excluded for more than one reason): extreme outliers for BMI, defined as the upper and lower 0.5% of BMI (n=23,214); missing data for age at the onset of diabetes (n=2,208); assumed to have type 1 diabetes (age at the onset of diagnosis less than 25 years old) (n=3,757); or aged less than 30 years old (n=82,779). Our final sample consisted of 15,881 Asians and 221,473 Whites with complete data on race/ethnicity, diabetes status, age at diabetes onset, age, sex, educational attainment, nation of birth, smoking status, alcohol consumption status, leisure-time physical activity, height, and weight.
For mortality analyses, pooled NHIS data from 2000 to 2009 were used linked with National Death Index (NDI) data and follow-up data until December 31, 2011.8 We excluded subjects with extremely outlying BMI (n=15,820), with missing data for age at the onset of diabetes (n=859), identified as having type 1 diabetes (n=1,647), aged less than 30 years (n=52,917), or with missing data on mortality status (n=10,234). The final sample for the mortality analysis consisted of Asians (n=7,822) and Whites (n=136,816) with complete data.
2.3. Measurements
Self-reported race/ethnicity was categorized using a combination of variables, including “race coded to single race group” and “Hispanic origin” in 2000–2005 and the predefined “race/ethnicity recode” in 2006–2014. Asian subgroups consisted of Chinese, Filipino, Asian Indian and other Asian individuals (i.e., Korean, Japanese, Vietnamese, and other Asian subgroups). Those classified as American Indian, Alaska Native, and multiple race without any primary race selection were included only in the larger category of “All Asians” and not included in any of the Asian subgroups.
Individuals were classified as having diabetes if they provided a positive response to any of the following three questions: (1) “Have you ever been told by a doctor or health professional that you have diabetes or sugar diabetes?”; (2) “Are you now taking insulin?”; or (3) “Are you now taking diabetic pills to lower your blood sugar? These are sometimes called oral agents or oral hypoglycemic agents”. Those who reported being told they had “prediabetes”, “impaired fasting glucose”, “impaired glucose tolerance”, “borderline diabetes” or “high blood sugar” and women who reported having had diabetes only during pregnancy (gestational diabetes) were classified as non-diabetes in our analysis.
Total and cause-specific mortality data were obtained from the NDI following the 10th revision of the International Statistical Classification of Diseases, Injuries and Causes of Death (ICD-10) guidelines, including cardiovascular disease (CVD)- (I00-I09, I11, I13, I20-I51) and cancer-related mortality (C00-C97).8
Other variables of interest were age, sex, educational attainment, nation of birth, smoking and alcohol consumption status, leisure-time physical activity, and BMI (weight in kilograms divided by height in meters squared). Educational attainment was dichotomized as college graduate vs. high school graduate. Nation of birth was identified as U.S.-born vs. foreign-born, i.e., whether participants were born in the U.S., including the 50 states or the District of Columbia. The participants were also dichotomized according to smoking and alcohol consumption status as current smokers vs. current non-smokers and current alcohol drinkers vs. current alcohol abstainers, respectively. Leisure-time physical activity was categorized as active vs. inactive according to the recommendation of the American Heart Association (AHA) for general heart health (at least 25 minutes of vigorous aerobic activity 3 times per week or at least 30 minutes of light-to-moderate aerobic activity 5 times per week).9 BMI was categorized as underweight or normal weight, overweight, and obese (BMI<25; 25≤BMI<25; 25≤BMI<30; and BMI≥30 kg/m2, respectively) according to NHLBI guidelines for Whites. For Asians, we applied the WHO Asian standard (underweight or normal weight, BMI<23; overweight, 23≤BMI<27.5; obese, BMI≥27.5 kg/m2).10
2.4. Statistical analysis
To examine trends in the prevalence of type 2 diabetes among Asians vs. Whites, we pooled 15 years of data from 2000 to 2014 and created new design variables incorporating stratum, primary sampling unit (PSU), and sampling weight.7 This approach accounted for complex sampling designs and weights and was limited to Asians and Whites using the STATA “subpop” command for correct variance estimation.
Baseline characteristics of Asians vs. Whites were compared using ANOVA for continuous variables (e.g., age and BMI) and χ2 tests for categorical variables (e.g., age group, sex, educational attainment, nation of birth, current smoking and alcohol consumption status, leisure-time physical activity, and overweight/obesity).
To examine differences in the prevalence of type 2 diabetes in Asians vs. Whites, we calculated age- and sex-adjusted prevalence proportions using the 2000 U.S. standard population and displayed them as Lowess-smoothed lines. The mean differences of age- and sex-adjusted type 2 diabetes prevalence between the earlier sample (2000–2002) and the later sample (2012–2014) as well as overall trends in both races were examined using a linear regression model. In addition, age- and sex-adjusted BMI between Asians and Whites was compared over five consecutive three-year periods (2000–2002; 2003–2005; 2006–2008; 2009–2011; and 2012–2014).
We first calculated age-standardized death rates (per 1,000 person-years) in Asians and Whites using the 2000 U.S. standard population stratified by diabetes status. To compare Asian vs. White individuals with and without diabetes, Cox proportional hazard regression models were constructed to estimate hazard ratios (HR) and their 95% confidence intervals (CI) across cause-specific mortalities, such as total, CVD-related, and cancer-related deaths with adjustments for age (timescale), sex, obesity status, educational attainment, nation of birth, current smoking and alcohol consumption status and leisure-time physical activity.
Follow-up time was calculated from the age at enrollment (quarter-year) to either the age of death (quarter-year) or the age at the end of the study (December 31, 2011), if alive, depending on which event occurred first. The proportional hazards assumption was satisfied.
All tests of statistical significance were two-tailed with alpha levels of 0.05. Analyses were performed using STATA 12 (StataCorp LP, College Station, TX, USA).
3. Results
Table 1 shows the characteristics of the 237,354 study participants who were Asian or White stratified by diabetes status. Among both diabetes and non-diabetes participants, Asians and the Asian subgroups were younger, were less likely to be U.S.-born, had higher educational attainment and were less likely to be current smokers or drinkers than Whites (all P<0.05). Among adults with diabetes, Asians and the Asian subgroups were more likely to meet the AHA leisure-time physical activity recommendation9 than Whites (32.5% and 28.6–48.7% vs. 27.9%; P<0.05). However, among non-diabetes participants, Asians and the Asian subgroups were less likely to meet the physical activity recommendation than Whites (30.1% and 28.7–30.8% vs. 34.2%; P<0.05). Asians and the Asian subgroups had a lower BMI than Whites (with diabetes: 27.0 kg/m2 and 25.0–27.9 kg/m2 vs. 31.3 kg/m2; without diabetes: 24.6 kg/m2 and 23.3–25.3 kg/m2 vs. 27.1 kg/m2), were less likely to be obese (with diabetes: 38.5% and 21.4–45.7% vs. 52.9%; without diabetes:18.3% and 9.1–25.0% vs. 24.1%) and were more likely to be overweight (with diabetes: 40.1% and 37.5–45.1% vs. 32.1%; without diabetes: 43.6% and 40.0–50.7% vs. 37.7%) based on the WHO Asian standard (overweight: 23≤BMI<27.5 kg/m2, obese: BMI≥27.5 kg/m2; all P<0.05).
Table 1.
Characteristics of adult Asians, Asian subgroups, and Whites in the U.S. based on the 2000–2014 NHISa
| Diabetes |
||||||
|---|---|---|---|---|---|---|
| All Asiansb | Chinese | Filipinos | sian Indians | Other Asians | Whites | |
|
| ||||||
| N | 1,383 | 183 | 389 | 235 | 458 | 18,522 |
| Age (years) | ||||||
| Mean (SE) | 59.7 (0.4) | 64.4 (1.1) | 60.2 (0.7) | 56.5 (0.9) | 60.4 (0.7) | 63.1 (0.1) |
| ≥65 (%) | 36.2 | 48.1 | 39.9 | 29.4 | 36.1 | 46.2 |
| Female (%) | 46.3 | 41.7 | 49.8 | 35.9 | 53.8 | 45.7 |
| Education (%)c | 59.9 | 60.2 | 72.4 | 69.5 | 49.3 | 46.6 |
| Smoking (%)d | 11.9 | 7.2 | 10.8 | 9.4 | 11.5 | 16.4 |
| Alcohol consumption (%)e | 35.8 | 43.4 | 38.6 | 36.6 | 31.3 | 45.2 |
| U.S.-born (%)f | 22.4 | 16.2 | 22.8 | 0.3 | 20.0 | 95.5 |
| Physically active (%)g | 32.5 | 48.7 | 31.3 | 28.6 | 31.2 | 27.9 |
| BMI (kg/m2) | ||||||
| Mean (SE) | 27.0 (0.2) | 25.0 (0.4) | 27.9 (0.4) | 26.6 (0.3) | 26.2 (0.3) | 31.3 (0.1) |
| Overweight (%)h | 40.1 | 41.7 | 37.5 | 45.1 | 44.7 | 32.1 |
| Obese (%)i | 38.5 | 21.4 | 45.7 | 34.6 | 31.3 | 52.9 |
| N | 14,498 | 3,033 | 2,985 | 2,329 | 5,370 | 202,951 |
| Age (years) | ||||||
| Mean (SE) | 47.9 (0.2) | 48.9 (0.3) | 49.3 (0.3) | 43.8 (0.3) | 48.6 (0.3) | 52.6 (0.1) |
| ≥65 (%) | 13.1 | 14.7 | 15.3 | 6.5 | 14.6 | 22.0 |
| Female (%) | 52.2 | 53.8 | 57.0 | 44.8 | 52.5 | 51.2 |
| Education (%)c | 70.0 | 70.9 | 75.7 | 82.1 | 63.3 | 61.2 |
| Smoking (%)d | 11.6 | 7.5 | 12.0 | 5.9 | 13.4 | 20.7 |
| Alcohol consumption (%)e | 47.6 | 46.3 | 53.7 | 40.8 | 46.9 | 68.5 |
| U.S. born (%)f | 20.7 | 15.6 | 25.7 | 3.3 | 19.1 | 94.9 |
| Physically active (%)g | 30.1 | 30.1 | 30.1 | 28.7 | 30.8 | 34.2 |
| BMI (kg/m2) | ||||||
| Mean (SE) | 24.6 (0.05) | 23.3 (0.1) | 25.3 (0.1) | 25.0 (0.1) | 24.0 (0.1) | 27.1 (0.02) |
| Overweight (%)h | 43.6 | 40.0 | 46.9 | 50.7 | 41.9 | 37.7 |
| Obese (%)i | 18.3 | 9.1 | 25.0 | 19.1 | 15.1 | 24.1 |
All characteristics of Asians are significantly different from those of Whites (P<0.05) regardless of diabetes status, except the proportion of females among diabetic patients.
Data are presented as weighted means or proportions.
All Asians include American Indian, Alaska Native, and individuals of multiple races without a primary race in addition to Asian subgroups, such as Chinese, Filipino, Asian Indian, and other Asian individuals.
Some college or college graduate.
Current smoking.
Current alcohol consumption.
Born in the U.S. (50 states or the District of Columbia), excluding U.S. territories.
Percentage that met the recommended leisure-time physical activities (25 minutes of vigorous activity three times per week or 30 minutes of light-to-moderate activity five times per week).
23≤BMI<27.5 kg/m2 for Asians and 25≤BMI<30 kg/m2 for Whites.
BMI≥27.5 kg/m2 for Asians and BMI≥30 kg/m2 for Whites.
Figure 1 displays the trends in the age- and sex-adjusted prevalence of type 2 diabetes among Asians and Whites over a study period of 15 years. Type 2 diabetes was more prevalent in Asians than in Whites throughout the study period. In Asians, the prevalence of type 2 diabetes increased from 8.1% in 2000–2002 to 9.6% in 2012–2014, while in Whites, its prevalence increased from 6.0% in 2000–2002 to 7.9% in 2012–2014 (P=0.04 for the trend in Asians, P<0.0001 for the trend in Whites). In the 2012–2014 survey cycle, the prevalence of type 2 diabetes in Chinese, Filipino, Asian Indian, and other Asian individuals was 5.3%, 12.8%, 11.3%, and 9.4%, respectively. Age- and sex-adjusted mean BMI was lower in Asians than in Whites over five consecutive three-year periods (24.5–24.9 kg/m2 vs. 26.8–27.9 kg/m2), and the difference in BMI between Asians and Whites increased over time (data not shown).
Figure 1.
The two lines represent Lowess-smoothed annual age- and sex-standardized prevalence of type 2 diabetes in 2000–2014 with adjustments for sampling weight. The solid line represents all Asians (A), and the dotted line represents Whites (W). The box-plots reveal age- and sex-standardized BMI in each period (2000–2002, 2003–2005, 2006–2008, 2009–2011, and 2012–2014) and are drawn at the midpoint of each period.
The characteristics of the 144,638 participants included in the mortality analysis were similar to those of the participants included in the analysis of type 2 diabetes prevalence (data not shown). During the 12-year follow-up period, 15,627 participants died (401 Asians and 15,226 Whites). The age-standardized mortality rates were higher in individuals with diabetes than in individuals without diabetes. The overall mortality rates were 134.9 and 77.3 per 1,000 person-years for diabetes and non-diabetes, respectively. CVD-related mortality was 27.0 and 14.1, and all-cancer-related mortality was 27.8 and 21.1 per 1,000 person-years for adults with and without diabetes, respectively (Appendix 1). Regardless of diabetes status, Asians had lower age-standardized mortality rates that Whites regardless of the cause of death. Among the participants with diabetes, the age-standardized overall mortality rates were 72.7 and 138.8 per 1,000 person-years for Asians and Whites, respectively. Among the non-diabetes participants, the overall mortality rates were 58.1 and 77.8 per 1,000 person-years for Asians and Whites, respectively (Table 2).
Table 2.
Adjusted hazard ratios (HR) for all-cause, cardiovascular disease (CVD)-related and all-cancer-related mortality in Asians vs. Whites stratified by diabetes status based on the 2000–2009 NHIS and a follow-up survey in 2011
| N | Total Deaths | Age-Standardized Mortality Rate (per 1,000 person-years) | HR |
|||
|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | ||||
|
| ||||||
| All-Cause | ||||||
| Diabetes | ||||||
| Asians | 636 | 76 | 72.7 | 0.6 (0.4–0.7) | 0.6 (0.4–0.7) | 0.7 (0.5–0.9) |
| Whites | 10,457 | 2,657 | 138.8 | 1 | 1 | 1 |
| Non-Diabetes | ||||||
| Asians | 7,186 | 325 | 58.1 | 0.9 (0.7–1.0) | 0.9 (0.7–1.0) | 1.0 (0.8–1.1) |
| Whites | 126,359 | 12,569 | 77.8 | 1 | 1 | 1 |
|
| ||||||
| CVD | ||||||
| Diabetes | ||||||
| Asians | 636 | 10 | 7.0 | 0.3 (0.1–0.5) | 0.3 (0.1–0.5) | 0.3 (0.1–0.6) |
| Whites | 10,457 | 598 | 28.0 | 1 | 1 | 1 |
| Non-Diabetes | ||||||
| Asians | 7,186 | 51 | 8.5 | 0.7 (0.5–0.9) | 0.7 (0.5–0.9) | 0.8 (0.5–1.1) |
| Whites | 126,359 | 2,412 | 14.2 | 1 | 1 | 1 |
|
| ||||||
| All-Cancer | ||||||
| Diabetes | ||||||
| Asians | 636 | 21 | 25.5 | 0.8 (0.5–1.3) | 0.8 (0.5–1.3) | 0.9 (0.5–1.5) |
| Whites | 10,457 | 514 | 28.0 | 1 | 1 | 1 |
| Non-Diabetes | ||||||
| Asians | 7,186 | 84 | 15.5 | 0.9 (0.7–1.1) | 0.8 (0.7–1.1) | 0.9 (0.7–1.2) |
| Whites | 126,359 | 3,297 | 21.2 | 1 | 1 | 1 |
Numerical data are HRs (95% CI). Model 1: Adjusted for age (continuous) and sex. Model 2: Model 1 + obesity categories (BMI≥27.5 kg/m2 for Asians and BMI≥30 kg/m2 for Whites). Model 3: Model 2 + education, U.S. born, current smoking and alcohol consumption status, and leisure-time physical activity.
After adjusting for age and sex, overall mortality was 80% higher in adults with diabetes than in those without diabetes (HR=1.8 [95% CI: 1.7–1.8]). CVD-related mortality was two times higher (HR=2.0 [95% CI: 1.8–2.3]) and all-cancer-related mortality was 40% higher in adults with diabetes than in those without diabetes (HR=1.4 [95% CI: 1.2–1.5]). This association persisted in the fully adjusted model (HRCVD=1.9 [95% CI: 1.7–2.1], HRTotal=1.7 [95% CI: 1.6–1.8]) (Appendix 1).
In adults with diabetes, after adjusting for age and sex, overall mortality was 40% lower in Asians than in Whites (HR=0.6 [95% CI: 0.4–0.7]) and CVD-related mortality was 70% lower in Asians than in Whites (HR=0.3 [95% CI: 0.1–0.5]). The associations persisted in the obesity-adjusted model (HRCVD=0.3 [95% CI: 0.1–0.5], HRTotal=0.6 [95% CI: 0.4–0.7]) as well as the fully adjusted model (HRCVD=0.3 [95% CI: 0.1–0.6], HRTotal=0.7 [95% CI: 0.5–0.9]) (Table 2). No significant differences in total or CVD-related mortality were observed between Asians and Whites without diabetes. For all-cancer-related mortality, there was no difference between the two racial groups among adults with or without diabetes.
4. Discussion
These data supported the following two major conclusions. First, type 2 diabetes was more prevalent in Asians than in Whites between 2000 and 2014. Asians are at least 20% more likely to have type 2 diabetes than their non-Asian counterparts. The prevalence of type 2 diabetes among Filipinos and Asian Indians was significantly higher than that among Whites, while Chinese individuals had similar prevalence rates as Whites. Furthermore, the prevalence rates in both Asians and Whites increased significantly during the study period. Second, among adults with diabetes, Asians exhibited lower overall mortality and CVD-related mortality than Whites, but there was no significant difference in cancer-related mortality. Among adults without diabetes, there was no significant difference in mortality between Asians and Whites.
Our findings were consistent with those of previous studies on Asian Americans. According to the U.S. national survey, Asians are 10% more likely to be diagnosed with diabetes than Whites,11, 12 but are 30% less likely to have heart disease and 50% less likely to die from heart disease.11, 13 In the UK Prospective Diabetes Study, whose participants were newly diagnosed with type 2 diabetes and were followed for 18 years, Asian Indians had an 11% lower all-cause mortality (HR=0.89 [95% CI: 0.80–0.97]) and a 57% lower peripheral vascular disease mortality (HR=0.43 [95% CI: 0.23–0.82]) than White Caucasians.14 In U.S. Hispanics, patterns similar to Asians were reported. Hispanics had a 33% higher self-reported prevalence of diabetes than Whites but approximately 25% lower total and cause-specific mortalities.15 This phenomenon has been called the “Hispanic paradox”, which describes the lower overall mortality rate and longer life expectancy among Hispanics than among White counterparts despite their higher prevalence of diabetes and lower socio-economic status and educational attainment, which are risk factors for CVD.15–17
There are several possible explanations for the disparities in the risk of developing diabetes and mortality between Asian and White adults with diabetes. Asians are known to be more genetically vulnerable to developing type 2 diabetes18, 19 and have lower muscle mass at any given level of BMI than Whites.20 Physical inactivity, an established risk factor for type 2 diabetes, would be one explanation for Asians having a higher risk of developing type 2 diabetes than Whites.21 In addition, migration effects, which were proposed as an explanation for the “Hispanic paradox”,16, 17 could also serve as a protective factor against mortality risks among Asian adults with diabetes. The migration effect is composed of two hypotheses: the “healthy migrant effect” and the “salmon bias effect”. The “healthy migrant effect” states that people who successfully migrated to the U.S. are healthier than those residing in their home countries as well as the general U.S. population. Our results were consistent with previous findings and showed that all Asians were younger, had higher education, were less likely to be current smokers and alcohol drinkers, were more physically active and were less obese than their White counterparts (Table 1). In addition, Asians with diabetes had significantly lower CVD and total mortality rates (Table 2). This phenomenon may occur because U.S. immigrants are not randomly selected from their original countries based on physical and psychological health.22, 23 Nationally representative data for the U.S. indicate that foreign-born individuals have better health than U.S.-born non-Hispanic whites and that recent Hispanic immigrants have better health than those who have resided in the U.S. for a longer period of time.22 The “salmon bias effect” describes the tendency of some foreign-born Hispanic subgroups to return to their home countries when they are older or very sick, as these individuals prefer to die in their home country; therefore, the mortality rate may be reduced among foreign-born subgroups, as foreign deaths are not included in U.S. mortality statistics. Indeed, the returning immigrants tend to be in poorer health and have a higher risk of mortality than immigrants who stay in the U.S.; therefore, the “return migration effect” may artificially lower the mortality rates of immigrants in the US.22 In addition, a study of health and internal migration in China supported these two hypotheses in Asians, showing that 1) healthier people migrate and move away from the home, and 2) immigrants with worse health conditions tend to return or move closer to their home communities.24 Further studies are required to explore the possible physiological and behavioral protective factors for mortality in Asians, especially Asian patients with diabetes, and the theories regarding overall immigrant health conditions.
The strengths of our study include a nationally representative sample with a sufficient sample size for Asians, which allowed us to explore trends in the prevalence of type 2 diabetes in Asians and Whites and the mortality risk ratios between the two races over a period of ten years. However, there are limitations in our study. First, the NHIS is based on self-reported data; thus, misclassifications of diabetes or BMI were possible. Nonetheless, there is no evidence that the degree of underestimation in BMI systematically differed in Asians vs. Whites25 due to using self-reported height and weight to calculate BMI. Second, there may be residual confounding factors due to the lack of information on other diabetes risk factors in the NHIS, such as dietary intake and family history of diabetes. Finally, the mortality benefits in Asians might be overestimated due to the “salmon bias effect”, in which the lower mortality rate observed among Asians is due to their tendency to migrate back to their original countries when they are older and sick.24
In conclusion, this study demonstrates that Asians are vulnerable to type 2 diabetes and that the prevalence of type 2 diabetes in this group has exhibited an upward trend in the U.S since 2000. Thus, prevention efforts for at-risk Asians are still needed. Physical inactivity, a recognized modifiable risk factor for developing type 2 diabetes, may be of particular interest from a public health perspective. In addition, among adults with diabetes, Asians have lower total and CVD-related mortality risks than Whites. Asian patients with diabetes should continue to pay attention to diabetes control in order to reduce the risk of complications.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
HCY was supported in part by a grant from NIDDK (P30DK079637).
Appendix 1. Adjusted hazard ratios (HR) for all-cause, cardiovascular disease (CVD)-related and all-cancer-related mortality in Diabetes vs. non-Diabetes based on the 2000–2009 NHIS and follow-up survey in 2011
| N | Total Deaths | Age-Standardized Mortality Rate (per 1,000 person-years) | HR |
|||
|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | ||||
|
| ||||||
| All-Cause | ||||||
| Diabetes | 11,093 | 2,733 | 134.9 | 1.8 (1.7–1.8) | 1.8 (1.7–1.8) | 1.7 (1.6–1.8) |
| Non-Diabetes | 133,545 | 12,894 | 77.3 | 1 | 1 | 1 |
|
| ||||||
| CVD | ||||||
| Diabetes | 11,093 | 698 | 27.0 | 2.0 (1.8–2.3) | 2.0 (1.8–2.2) | 1.9 (1.7–2.1) |
| Non-Diabetes | 133,545 | 2,463 | 14.1 | 1 | 1 | 1 |
|
| ||||||
| All-Cancer | ||||||
| Diabetes | 11,093 | 535 | 27.8 | 1.4 (1.2–1.5) | 1.4 (1.3–1.5) | 1.4 (1.2–1.5) |
| Non-Diabetes | 133,545 | 3,381 | 21.1 | 1 | 1 | 1 |
Numerical data are HRs (95% CI). Model 1: Adjusted for age (continuous) and sex. Model 2: Model 1 + obesity categories (BMI≥27.5 kg/m2 for Asians and BMI≥30 kg/m2 for Whites). Model 3: Model 2 + education, U.S. born, current smoking and alcohol consumption status, and leisure-time physical activity.
Footnotes
Declaration of Interest
None
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCES
- 1.Hoeffel EM, Rastogi S, Kim MO, Shahid H. The Asian population: 2010. Census 2010 Briefs. Washington, DC. U.S. Department of Commerce, U.S. Census Bureau, 2012 [Google Scholar]
- 2.Reeves TJ, Bennett CE. We the people: Asians in the United States. Census 2000 Special Reports. Washington, DC. U.S. Department of Commerce, U.S. Census Bureau, 2004 [Google Scholar]
- 3.Chen J, Vargas-Bustamante A, Ortega AN. Health care expenditures among Asian American subgroups. Med Care Res Rev. 2013;70: 310–329. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Barnes PM, Adams PF, Powell-Griner E. Health characteristics of the Asian adult population: United States, 2004–2006. Adv Data. 2008: 1–22. [PubMed] [Google Scholar]
- 5.Lee JW, Brancati FL, Yeh HC. Trends in the prevalence of type 2 diabetes in Asians versus whites: results from the United States National Health Interview Survey, 1997–2008. Diabetes Care. 2011;34: 353–357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.McNeely MJ, Fujimoto WY. Epidemiology of diabetes in Asian North Americans. In: Ekoé JM, Rewers M, Williams R, Zimmet P, eds. The epidemiology of diabetes mellitus. Wiley-Blackwell, West Sussex; 2008:323–337. [Google Scholar]
- 7.National Center for Health Statistics. Survey Description, National Health Interview Survey, 2014. Hyattsville, Maryland. 2015 [Google Scholar]
- 8.National Center for Health Statistics. Office of Analysis and Epidemiology, Public-use Linked Mortality File, 2015. Hyattsville, Maryland. http://www.cdc.gov/nchs/data_access/data_linkage/mortality.htm. [Accessed August, 2016]. [Google Scholar]
- 9.The American Heart Association recommendations for physical activity in adults. http://www.heart.org/HEARTORG/HealthyLiving/PhysicalActivity/FitnessBasics/American-Heart-Association-Recommendations-for-Physical-Activity-in-Adults_UCM_307976_Article.jsp. [Accessed September, 2016].
- 10.WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet. 2004;363: 157–163. [DOI] [PubMed] [Google Scholar]
- 11.Blackwell DL, Lucas JW, Clarke TC. Summary health statistics for U.S. adults: national health interview survey, 2012. Vital Health Stat 10. 2014: 1–161. [PubMed] [Google Scholar]
- 12.Centers for Disease C, Prevention. Health status of Cambodians and Vietnamese--selected communities, United States, 2001–2002. MMWR Morb Mortal Wkly Rep. 2004;53: 760–765. [PMC free article] [PubMed] [Google Scholar]
- 13.Xu J, Murphy SL, Kochanek KD, Bastian BA. Deaths: Final Data for 2013. Natl Vital Stat Rep. 2016;64: 1–119. [PubMed] [Google Scholar]
- 14.Davis TM, Coleman RL, Holman RR, Group U. Ethnicity and long-term vascular outcomes in Type 2 diabetes: a prospective observational study (UKPDS 83). Diabet Med. 2014;31: 200–207. [DOI] [PubMed] [Google Scholar]
- 15.Dominguez K, Penman-Aguilar A, Chang MH, et al. Vital signs: leading causes of death, prevalence of diseases and risk factors, and use of health services among Hispanics in the United States - 2009–2013. MMWR Morb Mortal Wkly Rep. 2015;64: 469–478. [PMC free article] [PubMed] [Google Scholar]
- 16.Medina-Inojosa J, Jean N, Cortes-Bergoderi M, Lopez-Jimenez F. The Hispanic paradox in cardiovascular disease and total mortality. Prog Cardiovasc Dis. 2014;57: 286–292. [DOI] [PubMed] [Google Scholar]
- 17.Cortes-Bergoderi M, Goel K, Murad MH, et al. Cardiovascular mortality in Hispanics compared to non-Hispanic whites: a systematic review and meta-analysis of the Hispanic paradox. Eur J Intern Med. 2013;24: 791–799. [DOI] [PubMed] [Google Scholar]
- 18.Chan JC, Malik V, Jia W, et al. Diabetes in Asia: epidemiology, risk factors, and pathophysiology. JAMA. 2009;301: 2129–2140. [DOI] [PubMed] [Google Scholar]
- 19.Yoon KH, Lee JH, Kim JW, et al. Epidemic obesity and type 2 diabetes in Asia. Lancet. 2006;368: 1681–1688. [DOI] [PubMed] [Google Scholar]
- 20.Park YW, Allison DB, Heymsfield SB, Gallagher D. Larger amounts of visceral adipose tissue in Asian Americans. Obes Res. 2001;9: 381–387. [DOI] [PubMed] [Google Scholar]
- 21.Lee DC, Sui X, Church TS, Lee IM, Blair SN. Associations of cardiorespiratory fitness and obesity with risks of impaired fasting glucose and type 2 diabetes in men. Diabetes Care. 2009;32: 257–262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Abraido-Lanza AF, Dohrenwend BP, Ng-Mak DS, Turner JB. The Latino mortality paradox: a test of the “salmon bias” and healthy migrant hypotheses. Am J Public Health. 1999;89: 1543–1548. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Sorlie PD, Backlund E, Johnson NJ, Rogot E. Mortality by Hispanic status in the United States. JAMA. 1993;270: 2464–2468. [PubMed] [Google Scholar]
- 24.Lu Y, Qin L. Healthy migrant and salmon bias hypotheses: a study of health and internal migration in China. Soc Sci Med. 2014;102: 41–48. [DOI] [PubMed] [Google Scholar]
- 25.Rothman KJ. BMI-related errors in the measurement of obesity. Int J Obes (Lond). 2008;32 Suppl 3: S56–59. [DOI] [PubMed] [Google Scholar]