Prevalence of diabetes, metabolic syndrome, and cardiovascular risk
factors in US Asian Indians: results from a national study
Ranjita Misraa,⁎, Thakor Patelb, Purushotham Kothac, Annaswamy Rajid, Om Gandae,
MaryAnn Banerjif, Viral Shahf, Kris Vijayg, Sundar Mudaliarc,
Dinakar Iyerh, Ashok Balasubramanyamh
aTexas A&M University, College Station, TX, USA
bUSUHS, Bethesda, MD, USA
cUC San Diego, San Diego, CA, USA
dBrigham and Women's Hospital, Harvard Medical School, Boston, MA, USA
eJoslin Diabetes Center, Harvard Medical School, Boston, MA, USA
fSUNY Downstate Medical Center, Brooklyn, NY, USA
gScottsdale Heart Institute, Scottsdale, AZ, USA
hBaylor College of Medicine, Houston, TX, USA
Received 22 July 2008; received in revised form 23 November 2008; accepted 6 January 2009
Background: Although studies of immigrant Asian Indians in other countries show high rates of diabetes (DM), metabolic syndrome
(MetS), and cardiovascular disease (CVD), no randomized, population-based studies of this rapidly growing ethnic group exist in the US.
Methods: The sample comprised 1038 randomly selected Asian Indian immigrants, aged 18 years and older at seven US sites. Prevalence of
diabetes and MetS (age-adjusted and sex-adjusted means) was estimated and ANOVA was used to calculate gender and group differences
(normoglycemia/impaired fasting glucose/diabetes) for CVD risk factors. Results: The mean age was 48.2 years. The majority of
respondents were male, married, educated, and with some form of health insurance. Prevalence of diabetes was 17.4%, and 33% of the
respondents had prediabetes. Cardiovascular risk factors, especially high levels of triglycerides, total cholesterol, LDL cholesterol,
homocysteine, and C-reactive protein, and low levels of HDL cholesterol, were also prevalent; elevated lipoprotein(a) was not observed. The
age-adjusted prevalence of MetS was 26.9% by the original NCEP/ATP III criteria, 32.7% by the modified NCEP/ATP III criteria, and 38.2%
by the IDF criteria. The MetS rates for women, but not for men, increased with age using all three criteria. There was a progressive worsening
of all metabolic parameters as individuals progressed from normal to IFG to diabetes. Conclusion: The prevalence rates of diabetes and MetS
among US Asian Indians are higher than reported in earlier, nonrandomized, smaller surveys. These data provide a firm basis for future
mechanistic and interventional studies.
Published by Elsevier Inc.
Keywords: Diabetes prevalence; Asian Indian; CVD risk factors; Metabolic syndrome
Migrant Asian Indians are reported to have high rates of
diabetes (DM), metabolic syndrome (MetS), cardiovascular
disease (CVD), and related complications in the US,
Canada, and UK (Anand et al., 2000; Chandie Shaw
et al., 2002; Enas et al., 1996; Hughes et al., 1997;
McKeigue et al., 1993, 1992; Mohanty et al., 2005; Omar
et al., 1985; Ramaiya et al., 1995; Samanta et al., 1991).
Insulin resistance is highly prevalent in Asian Indian
migrants, despite low rates of obesity (McKeigue, 1996;
McKeigue et al., 1992; Whincup et al., 2002). In the United
Journal of Diabetes and Its Complications 24 (2010) 145–153
E-mail address: email@example.com (R. Misra).
1056-8727/09/$ – see front matter. Published by Elsevier Inc.
States, Asian Indians have the highest ethnic-specific
prevalence of CVD, with age-specific mortality two to
three times higher than Caucasians (Enas & Senthilkumar,
2001; Enas et al., 1996; Wild et al., 1995). Traditional risk
factors such as hypertension, obesity, and hypercholester-
olemia do not account completely for these high rates.
Prevalence of DM and related conditions among US Asian
Indians was assessed by Venkataraman et al. (2004) using a
faith-based sample in Atlanta, GA. The overall prevalence
of DM was 18.3% (22.5% in men and 13.6% in women).
Mohanty et al. (2005) compared 555 Asian Indians to
87,846 non-Hispanic whites in the NHIS dataset from 1997
to 2000 and reported that the former had significantly
higher odds of having diabetes. However, they also
reported lower CHD and hypertension rates, in contrast to
prior studies that showed much higher age-standardized
CVD rates and related mortality in this ethnic group. Data
from national surveys are limited due to small sample sizes
or aggregation of ethnic data into a heterogeneous group of
“Asian Americans” or “Asian and Pacific Islanders.”
Population-based national studies on prevalence and risk
factors for DM and CVD among US Asian Indians are
Asian Indians, the third largest and fastest growing US
Asian subgroup, are heterogeneous, with numerous
languages, religions, racial types, social habits, cultural
practices, and diets. Despite a perception that they have
high socio-economic status and good access to health care
(Gupta, 2000), US Asian Indians have marked variations
in educational attainment, income, and wealth, and a
significant number lack education and job skills (Rangas-
wamy, 1995). Recent immigrant cohorts comprise both
highly educated professionals and individuals who lack
education and job skills. The latter are mostly family
members of earlier immigrants (Rangaswamy, 1995). This
heterogeneity makes it imperative to use large, randomized
samples to determine disease prevalence and risk factors.
Previous studies of Asian Indian health have employed
hospital-based or convenience samples (Abate et al., 1995,
1996, 2004; Banerji et al., 1999; Enas et al., 1996; Raji,
Seely et al., 2001; Raji, Williams et al., 2001; Venkatara-
man et al., 2004). The Diabetes among Indian Americans
(DIA) study is the first to develop and utilize a large,
randomly selected, nationwide cohort of US Asian Indians
to determine the prevalence of DM, MetS, and CVD
2. Research design and methods
Asian Indian adults were randomly selected at seven US
urban sites—Houston, TX; Phoenix, AZ; Washington, DC;
Boston, MA; San Diego, CA; Edison, NJ; and Parsippany,
NJ. There is no directory/sampling frame available for
Asian Indians in the US; hence directories were created at
each site through compilations of several sources: (1) city-
wide telephone directories with a search for the 100 most
common Asian Indian last names, (2) directories of ethnic
associations (e.g., Gujarati Association, Telugu Associa-
tion), (3) faith-based associations (e.g., temple directories),
and (4) professional associations. These sources enabled
recruitment of a diverse cross section of subjects. Given
the lack of public access to a more complete database of
Asian Indians in the United States, this approach to
achieve comprehensive representation and random sam-
pling is highly rigorous. Subjects (n=5000) were selected
from this database (n=43,120) using computerized random
numbers and were invited to participate by letters and
follow-up telephone calls; 30% of the letters were returned
due to incomplete address or no forwarding address, and
10% of the phone numbers were disconnected. Inclusion
criteria were having both parents from India, age ≥18
years, and residence in the US for at least 1 year.
Exclusion criteria included self-reported pregnancy status
and family members' report of significant mental or
2.2. Data collection
Telephone interviews were completed by trained,
multilingual Asian Indian staff. Fasting blood tests and
anthropometry were completed at a local community
center, clinic, or hospital. Blood samples were drawn
after a 10-h fast, coded, immediately centrifuged to
separate plasma or serum, and shipped on ice to three
core laboratories for biochemical analysis—Atherotech
Laboratory (Birmingham, AL, USA), Diabetes Diagnostic
Laboratory (Columbia, MO, USA), and the Translational
Metabolism Unit, Baylor College of Medicine (Houston,
TX, USA). The response rate was 37% (n=1838) for the
phone interview; of these 56% (n=1038) completed the
fasting blood work. For respondents who were unwilling
to participate in the study, interviewers asked the reasons
for nonparticipation as well as demographic information
(age, gender, educational level, number of years lived in
the US, smoking status, and family history of diabetes and
heart disease) for comparison of respondents to non-
respondents. Nonparticipants did not differ in educational
level, family history of diabetes and CVD, or smoking
status, but were significantly older than participants. The
overall response rate was higher than in published health
surveys of Asians or Asian Indians (Lien & Van Nuys,
2004; Misra et al., 2000; Misra & Vadaparampil, 2004;
Yagalla et al., 1996).
Demographic information included age, gender, marital
status, education, income, health insurance, place of birth,
and duration of US residency. Participants were asked
146 R. Misra et al. / Journal of Diabetes and Its Complications 24 (2010) 145–153
Cholesterol in Adults (Adult Treatment Panel III) final report (2002).
Circulation, 106(25), 3143−3421.
Venkataraman, R., Nanda, N. C., Baweja, G., Parikh, N., & Bhatia, V.
(2004). Prevalence of diabetes mellitus and related conditions in Asian
Indians living in the United States. American Journal of Cardiology, 94,
Whincup, P. H., Gilg, J. A., et al. (2002). Early evidence of ethnic
differences in cardiovascular risk: cross sectional comparison of British
South Asian and white children. BMJ, 324(7338), 635.
Wild, S. H., Laws, A., et al. (1995). Mortality from coronary heart disease
and stroke for six ethnic groups in California, 1985 to 1990. Annals of
Epidemiology, 5(6), 432−439.
World Health Organization Western Pacific Region, International Associa-
tion for the Study of Obesity, & International Obesity Task Force.
(2000). The Asia-Pacific Perspective: Redefining Obesity and Its
Treatment. Melbourne, Australia.
Wu, T., Dorn, J. P., et al. (2002). Associations of serum C-reactive protein
with fasting insulin, glucose, and glycosylated hemoglobin: the Third
National Health and Nutrition Examination Survey, 1988–1994. Amer-
ican Journal of Epidemiology, 155(1), 65−71.
Yagalla, M. V., Hoerr, S. L., et al. (1996). Relationship of diet, abdominal
obesity, and physical activity to plasma lipoprotein levels in Asian
Indian physicians residing in the United States. Journal of American
Dietetic Association, 96(3), 257−261.
153R. Misra et al. / Journal of Diabetes and Its Complications 24 (2010) 145–153