10125652732795J Immigr Minor HealthJ Immigr Minor HealthJournal of immigrant and minority health / Center for Minority Public Health1557-19121557-192022481308404912910.1007/s10903-012-9617-3NIHMS591888ArticleMigration status in relation to clinical characteristics and barriers to care among youth with diabetes in the USJaacksLindsay M.1Oza-FrankReena2D’AgostinoRalphJr.3DolanLawrence M.4DabeleaDana5LawrenceJean M.6PihokerCatherine7O’ConnorM. Rebecca7LinderBarbara8ImperatoreGiuseppina9SeidMichael4NarayanK.M. Venkat10Mayer-DavisElizabeth J.11112Department of Nutrition, The University of North Carolina, Chapel Hill, NCResearch Institute at Nationwide Children’s Hospital and Department of Pediatrics, The Ohio State University, Columbus, OHDepartment of Biostatistics, Wake Forest University School of Medicine, Winston-Salem, NCCincinnati Children's Hospital Medical Center, Cincinnati, OHDepartment of Epidemiology, The University of Colorado, DenverDepartment of Research & Evaluation, Kaiser Permanente Southern California, Pasadena, CASeattle Children’s Hospital, Seattle, WANIDDK, National Institutes of Health, Bethesda, MDCenter for Disease Control and Prevention, Atlanta, GAHubert Department of Global Health, Emory University, Atlanta, GADepartment of Medicine, The University of North Carolina, Chapel Hill, NCCorresponding author: Elizabeth J. Mayer-Davis, 2211 McGavran-Greenberg Hall, 135 Dauer Drive, Campus Box 7461, Chapel Hill, NC 27599, 919-966-1991 (telephone), 919-843-2011 (fax), mayerdav@email.unc.edu29520141220120962014146949958

Migration status and the accompanying diversity in culture, foods and family norms, may be an important consideration for practitioners providing individualized care to treat and prevent complications among youth with diabetes. Approximately 20% of youth in the US have ≥ 1 foreign-born parent. However, the proportion and characteristics of youth with diabetes and ≥ 1 foreign-born parent have yet to be described. Study participants (n = 3,086) were from SEARCH for Diabetes in Youth, a prospective multi-center study in the US. Primary outcomes of interest included HbA1c, body mass index and barriers to care. Multivariable analyses were carried out using logistic regression and analysis of covariance. Approximately 17% of participants with type 1 diabetes (T1D) and 22% with type 2 diabetes (T2D) had ≥ 1 foreign-born parent. Youth with T1D and ≥ 1 foreign-born parent were less likely to have poor glycemic control [adjusted odds ratio (OR) (95% confidence interval): 0.70 (0.53, 0.94)]. Among youth with T2D, those with ≥ 1 foreign-born parent had lower odds of obesity [adjusted OR (95% CI): 0.35 (0.17, 0.70)]. This is the first study to estimate the proportion and characteristics of youth with diabetes exposed to migration in the US. Research into potential mechanisms underlying the observed protective effects is warranted.

diabetesyouthmigrationUSBACKGROUND

Diabetes is a unique chronic condition in that it requires rigorous and individualized ongoing treatment provided by a multidisciplinary healthcare team to maintain quality of life and prevent poor glycemia, cardiovascular disease and other complications. Providing effective individualized care relies on the consideration of individual-level factors, such as knowledge, attitudes and beliefs, as well as the socio-cultural environment in which a patient lives. The fact that poor glycemic control and cardiovascular disease risk factors are common among youth with diabetes in the US (1, 2) suggests that gaps remain in our understanding of the determinants of and prevention of these adverse health outcomes. Additional investigation of factors potentially associated with these clinical characteristics is needed in order to address these gaps. Research has supported associations of parental educational attainment, family structure and race/ethnicity (1) with glycemic control and it is plausible that other demographic factors, such as migration status, may also be important considerations for improving treatment and prevention efforts.

Migration to the US has been increasing since 1945 (3) with the current immigrant population estimated to be 38.5 million or 12% of the total population (4). According to the 2009 US Census, approximately 20% of youth (< 18 years) have at least one foreign-born parent (5). However, the proportion of youth with diabetes who have at least one foreign-born parent is unknown. Furthermore, the demographic, socioeconomic and clinical characteristics of this subpopulation in the US have yet to be described.

Current evidence suggests that the impact of migration status on glycemic control and self-management among youth with diabetes may differ in Europe and the US. Hsin et al. (6) found that among Hispanic youth with type 1 diabetes, those who migrated to the US more recently had better adherence measured using the Diabetes Self-Management Profile (7). In Italy, two cross-sectional studies reported that children of immigrants born in Italy have a younger age of type 1 diabetes diagnosis than children of immigrants born in their countries of origin or Italian children (8, 9). Children of immigrants (those born outside of Italy and those born in their country of origin, combined) were also more likely to have higher glycosylated hemoglobin A1c (A1C) than Italian patients (9). In France, a cross-sectional study of youth with type 1 diabetes found that youth with immigrant mothers had higher A1C and lower adherence to diet and insulin recommendations and that immigrant mothers had lower levels of diabetes knowledge relative to native French mothers (10).

SEARCH for Diabetes in Youth includes one of few cohorts with information on migration status concurrent with clinical characteristics and barriers to care among youth with diabetes in the US. These data provide a unique opportunity to explore the proportion and characteristics of youth with diabetes and at least one foreign-born parent. The objectives of this study were to: 1) determine the proportion of SEARCH participants with at least one foreign-born parent and the distribution of demographic and socioeconomic characteristics and 2) determine if having at least one foreign-born parent is associated with glycemic control, cardiovascular disease risk factors and barriers to care.

METHODSStudy Sample

SEARCH for Diabetes in Youth is a prospective cohort of youth with diabetes across six sites in the US: South Carolina, Ohio, Colorado, California, Washington and Hawaii. The study protocol was approved by Institutional Review Boards at each site and details of the methods have been published previously (11). Briefly, all prevalent cases of diabetes were identified in 2001 and incident cases were identified in subsequent years with estimated case ascertainment greater than 90% (12, 13). All ascertained cases were asked to complete a brief initial survey and after completion, were invited to a baseline visit during which anthropometric data and blood samples were collected. Participants in the 2002–2005 incident cohorts were also asked to attend periodic follow-up visits, on average, 12, 24 and 60 months after the baseline visit.

Initially, SEARCH did not query youth or parent migration status. In 2007, questions regarding migration were added to surveys that were administered to the 2001 prevalent cohort and the 2002–2005, and 2008 incident cohorts. Migration questions were consistent among the surveys and included whether the participant or either parent had migrated to the US, and if so, when they migrated and where they migrated from. SEARCH did not query the legal status of individuals that reported being born outside the US.

As a consequence of the SEARCH protocol, only baseline data were available for the 2001 prevalent and 2008 incident cohorts and follow-up visit data were additionally available for the 2002–2005 incident cohorts. In the event that data were available for more than one follow-up visit, the most recent visit was selected with the objective of increasing generalizability by including the most contemporary data and a diverse sample of disease durations. All outcome variables are therefore from a single visit.

A total of 3,191 youth diagnosed with type 1 or type 2 diabetes by a health care provider before the age of 20 years completed the survey containing the migration questions. Of these, 57 had missing data and 37 responded “Don’t know” to at least one of the three questions relating to migration, and thus were excluded from the present analysis. Additionally, 11 participants were born outside the US yet had US-born parents, possibly indicating that they were adopted. Due to potential differences in exposure, this subgroup of participants was excluded, leaving 3,086 participants in the final analysis.

Measures

Participants were categorized as exposed to migration if they had at least one parent that reported being born outside of the US. Self-reported country of origin was categorized into six macro geographical regions according to the United Nations’ 2008 Demographic Yearbook: Africa, Asia, Europe, Latin America and the Caribbean, Northern America, and Oceania (14). Parental duration of residency in the US was calculated as the period beginning with the self-reported year of migration to the US and ending with the year of the baseline visit or most recent follow-up visit. For participants with two foreign-born parents, the parental duration in the US was calculated as the mean of the two parents’ durations.

Participant age was calculated as the period beginning with date of birth and ending with the date of the baseline visit or most recent follow-up visit. Diabetes duration was calculated as the period from the date of diabetes diagnosis to the date of the baseline visit or most recent follow-up visit. Self-reported demographic and socioeconomic variables assessed included gender, race/ethnicity based on 2000 US Census classification (15) [non-Hispanic white, Hispanic (regardless of race), Asian/Pacific Islander, non-Hispanic black, Other (including Native American)], highest level of parental educational attainment (≤ high school, some college or a degree), family structure (two-parent household, single-parent household, other household structure), estimated total annual household income (<$24,999, $25,000–$49,999, $50,000–$74,999, ≥$75,000). Health insurance coverage was categorized as a binary variable (private, other) where other included participants reporting none (n = 32 for type 1 diabetes; n = 19 for type 2 diabetes), Medicaid/Medicare (n = 202 for type 1 diabetes; n = 46 for type 2 diabetes), and other (n = 20 for type 1 diabetes; n = 6 for type 2 diabetes). Proportions for health insurance coverage excluded the California and Hawaii SEARCH sites because these were healthcare plan sites and therefore all participants from these sites would be classified as having private insurance.

Barriers to care were assessed at follow-up visits via questionnaire items adapted from the Consumer Assessment of Healthcare Providers and Systems survey Supplemental Item Set for Children with Chronic Conditions (16). The seven barriers assessed included: not having a personal doctor (regular doctor), problems getting care (access to care), costs of care, problems getting medications/supplies, problems receiving care that takes into account personal and family context (contextual care), problems communicating with providers, and problems getting needed information. Barriers were identified as being absent or present based on operationalization schemes published previously (17).

Glycemic control was assessed using A1C measured in whole blood with automated non-porous ion-exchange high-performance liquid chromatography (Tosoh Bioscience, Montgomeryville, Pennsylvania) and was analyzed as a continuous outcome and as a binary outcome defined as A1C ≥ 9.5% (poor glycemic control) versus < 9.5 (1).

Cardiovascular disease risk factors assessed included body mass index (BMI, kg/m2), BMI z-score, obesity (BMI ≥ 95th percentile for age and sex), waist circumference (cm), systolic blood pressure (SBP, mmHg), diastolic blood pressure (DBP, mmHg), hypertension (SBP or DBP ≥ 95th percentile for age, gender, and height), low-density lipoprotein (LDL, mg/dL) cholesterol, high-density lipoprotein (HDL, mg/dL) cholesterol, triglycerides (mg/dL), and family history of diabetes (mother or father with diabetes). Methods used for anthropometric measures and clinical characteristics have been published previously (11). Quality of life was assessed using the Pediatric Quality of Life Inventory Generic Module (PedsQL) total score ranging from 0 to 100 with higher scores indicating better health-related quality of life. Frequency of self blood glucose monitoring (< 1 time per day, 1–2 times per day, 3 times per day, ≥ 4 times per day) was self-reported.

Analysis

All analyses were stratified by diabetes type. Chi-square tests were used to compare categorical outcomes and F tests (ANOVA) were used to compare continuous outcomes across migration status strata. A Fisher’s Exact Test was used to compare categorical outcomes when cell sizes were less than or equal to 5. Test assumptions, including homogeneity of variances, were met.

Race/ethnicity was assessed as a potential effect measure modifier using a Breslow-Day test for homogeneity of the stratum-specific odds ratios at an alpha level of 0.10 to account for small sample sizes and consequently lower power for stratified estimates. Race/ethnicity stratum-specific estimates could not be derived for participants with type 2 diabetes due to small cell sizes. In the absence of effect measure modification, race/ethnicity was treated as a potential confounder and a single effect estimate was presented. Additional potential confounders were identified using a directed acyclic graph (18): diabetes duration, SEARCH site, parental education, and household income. A change-in-estimate approach with a 10% change taken to be an indication of confounding was used to determine the final adjustment set for a given outcome.

Multiple logistic regression analysis for binary outcomes and analysis of covariance (ANCOVA; generalized linear models) for continuous outcomes were used to assess the effect of migration status on A1C, obesity, BMI z-score, and barriers to care after adjustment for confounders. All statistical analyses were conducted using SAS 9.2 (SAS Institute, Cary, North Carolina). Statistical significance was considered for P < 0.05.

RESULTS

Of the 3,086 youth included in the present analysis, 16.9% (n = 450) of those with type 1 diabetes and 22.0% (n = 92) of those with type 2 diabetes had at least one foreign-born parent. Due to limitations in sample size, participants with one (type 1 diabetes, n = 249; type 2 diabetes, n = 20) or both (type 1 diabetes, n = 201; type 2 diabetes, n = 72) parents born outside the US were combined. Within this group, 53 of the participants with type 1 diabetes and 24 of those with type 2 diabetes were themselves born outside the US.

Duration of parental residency in the US ranged from 4 to 54 years among participants with type 1 diabetes and 4 to 45 years among those with type 2 diabetes. The majority of parents born outside the US migrated from Latin America and the Caribbean (Table I). The proportion of males with type 2 diabetes was lower among participants with US-born parents: 36.7% compared to 48.9% among those with at least one foreign-born parent. Parents of participants with type 1 diabetes who migrated to the US were less likely to have some college or a degree than those born in the US and had lower household incomes. Participants with type 2 diabetes and at least one foreign-born parent were significantly more likely to live in a two-parent household and a trend towards lower parental education in this group approached significance.

Participants with type 1 diabetes and at least one foreign-born parent had a lower total PedsQL score, on average, than those with US-born parents (Table I). No statistically significant differences in frequency of blood glucose monitoring were observed between migration strata among participants with type 1 diabetes or participants with type 2 diabetes.

Information on barriers to care was available for the subsample with a follow-up visit (n = 1,496). In the unadjusted analysis, among participants with type 1 diabetes, having a regular doctor and receiving contextual care differed by parental migration status: participants with at least one foreign-born parent were more likely to experience these barriers (Table II). However, these differences were no longer statistically significant after adjustment for race/ethnicity and SEARCH site: adjusted odds ratios (OR) (95% confidence intervals (CI)) for not having a regular doctor: 1.10 (0.72, 1.67); not receiving contextual care: 1.14 (0.76, 1.72) (Table III). Among participants with type 2 diabetes, only access to care varied by parental migration status in the unadjusted analysis: those with US-born parents reported this barrier more frequently.

A total of 2,896 participants had glycemic control and cardiovascular disease risk factor data. Among participants with type 1 diabetes, there was no statistical evidence for heterogeneity across race/ethnicity strata (homogeneity test p = 0.62 for poor glycemic control and p = 0.13 for obesity). Race/ethnicity (non-Hispanic white versus all other) was therefore assessed as a potential confounder in all remaining multivariable models. Participants with type 1 diabetes and at least one foreign-born parent were significantly less likely to have poor glycemic control than participants with US-born parents [adjusted OR (95% CI): 0.70 (0.53, 0.94); Table IV]. Results based on A1C data analyzed continuously were similar [adjusted mean ± SD A1C among participants with US-born parents: 8.53 ± 0.04; participants with at least one foreign-born parent: 8.17 ± 0.10; p = 0.001].

Obesity was not significantly associated with migration status among participants with type 1 diabetes [adjusted OR (95% CI): 0.79 (0.55, 1.12); Table IV] and an association was also not found when BMI z-scores were analyzed continuously [adjusted mean ± SD BMI z-scores among participants with US-born parents: 0.62 ± 0.02; participants with at least one foreign-born parent: 0.55 ± 0.06; p = 0.3].

Glycemic control was not significantly associated with migration status among participants with type 2 diabetes [adjusted OR (95% CI): 0.65 (0.29, 1.45); Table IV] and results were again consistent when A1C was analyzed continuously [adjusted mean ± SD A1C among participants with US-born parents: 8.29 ± 0.19; participants with at least one foreign-born parent: 7.89 ± 0.44; p = 0.4].

Participants with type 2 diabetes that had at least one parent born outside the US were significantly less likely to be obese relative to participants with US-born parents [adjusted OR (95% CI): 0.35 (0.17, 0.70); Table IV]. The direction of this association was consistent when BMI z-scores were analyzed continuously, but was no longer statistically significant after adjustment [adjusted mean ± SD BMI z-scores among participants with US-born parents: 1.97 ± 0.08; participants with at least one foreign-born parent: 1.73 ± 0.18; p = 0.2].

The following sensitivity analyses were performed and resulted in similar effect estimates as the combined results above: 1) baseline data for all participants (compared to most-recent visit data), 2) stratifying participants into the following three exposure categories: US-born parents, one foreign-born parent, both parents foreign-born, and 3) stratifying participants into the following three exposure categories: US-born parents, at least one foreign-born parent with duration in the US ≥ median, at least one foreign-born parent with duration in the US < median (data not shown).

DISCUSSION

The present analysis is one of few studies to examine health-related aspects of youth with diabetes who are children of immigrants to the US. Approximately 17% of participants with type 1 diabetes and 22% with type 2 diabetes had at least one foreign-born parent; similar estimates to those reported for the general population of youth (< 18 years) in the US (5). This study contributes to the growing documentation of the “healthy immigrant effect” (19). Participants with type 1 diabetes and at least one foreign-born parent had better glycemic control than those with US-born parents. Participants with type 2 diabetes and at least one foreign-born parent were less likely to be obese than their counterparts with US-born parents. Systematic reviews suggest that these comparatively better health outcomes may deteriorate over time as immigrants acculturate and acquire their host country’s rates of disease risk factors and disease (20–22). Here we have identified a potential early intervention target associated with better glycemic control and weight status.

Youth with type 1 diabetes who had at least one foreign-born parent were less likely to have poor glycemic control and had a lower mean A1C than youth with US-born parents. A similar magnitude and direction of effect was also observed among youth with type 2 diabetes, though the association was not statistically significant. Results of adjusted barriers to care analysis among participants with type 1 diabetes suggested that children of immigrants in this sample did not have significantly higher odds of reporting the barriers of having a regular doctor, access and cost of care, getting medications and information, or communicating with practitioners, perhaps explaining why we did not observe detrimental effects. The apparent protective effect may stem from better adherence to clinical recommendations among Hispanic youth (6), who made up nearly 50% of participants with at least one foreign-born parent in this sample. Better adherence in this group may be the result of cultural values, such as respect of elders and more reverent treatment of healthcare workers, observed previously in both Hispanic and Asian populations (6, 23). Additionally, offspring of immigrants may have greater family involvement and support (24), which have been associated with improved adherence in youth with diabetes (25). In this sample, participants with at least one foreign-born parent were more likely to live in a two-parent household. Family structure, serving as a proxy for family support, may be a mediating pathway through which parent migration status improves glycemic control among youth with diabetes.

Among participants with type 2 diabetes, those with at least one parent born outside the US were less likely to be obese relative to their counterparts with US-born parents. This difference may be explained by the large proportion of immigrants from Asia, Latin America and the Caribbean in this sample. In groups migrating from these regions, type 2 diabetes has been shown to occur at a lower BMI than in other populations (20, 26), perhaps due to the interaction of a genetic predisposition and exposure to a Western diet (27, 28). Compared to genetically similar adults in their villages of origin in India, UK migrants were found to have higher total energy intakes, fat intakes and IGF-1 levels, independent of BMI (29). A clinical implication of this result may be the need for type 2 diabetes screening at lower BMIs in patients with foreign-born parents.

Additionally, though speculative, these results may indicate that in this sample of immigrant families there is greater adherence to traditional, healthier foods. Previous studies in Latino youth have revealed an association between lower levels of acculturation and lower dietary intakes of energy, fat, and saturated fat (30) and studies in Asian adolescents have reported higher levels of vegetable consumption and lower levels of soda consumption relative to white adolescents (31). Further research on family support, dietary intake, and acculturation is needed to improve our understanding of the observed protective effects.

There were several limitations to the present analysis. For participants with type 1 diabetes, our sample sizes within racial/ethnic strata were small, particularly for Asian/Pacific Islanders, thus limiting our power to detect differences across racial/ethnic strata. Furthermore, though sensitivity analyses did not indicate substantial differences between strata of participants with one versus both parents born outside the US, small sample sizes limited the interpretability of these results. There was therefore the potential for heterogeneity within the group of participants with at least one foreign-born parent. The total sample size for youth with type 2 diabetes was also small. However, to our knowledge, this is the first analysis to present these data on children of immigrants with type 2 diabetes in the US and therefore it provides insight into this unexplored subgroup. Over 37% of participants with type 1 diabetes and 55% of those with type 2 diabetes and at least one foreign-born parent came from the California SEARCH site. Although confounding by SEARCH site was assessed, this disproportion may have introduced selection bias in that California has one of the largest immigrant populations in the US (32) and living in immigrant enclaves has been associated with improved health behaviors (33). Although ascertainment for SEARCH is high (12, 13), selection bias in the form of “the healthy volunteer effect” may have influenced the results: participants with at least one foreign-born parent that participated in this study may reflect a healthier group of individuals exposed to migration relative to the general US population. Finally, the only measures of acculturation investigated in this analysis were generational status and duration of residence in the US. More sophisticated scales of acculturation, such as “A Short Acculturation Scale for Hispanics,” which has been adapted for use in other cultures (34), could provide further insight into additional aspects that we were not able to explore, such as linguistic acculturation and ethnicity of neighborhoods, schools, and close friends.

Diabetes is a unique condition in that it requires tailored and multidisciplinary ongoing care. Parent migration status may be an important factor for health care teams to consider when providing individualized care to youth with diabetes. The large proportion of participants with foreign-born parents reported here supports the need for further investigation of diabetes care in this understudied population.

ACKNOWLEDGEMENTS

The SEARCH for Diabetes in Youth Study is indebted to the many youth and their families, and their health care providers, whose participation made this study possible.

Grant Support: SEARCH for Diabetes in Youth is funded by the Centers for Disease Control and Prevention (PA numbers 00097, DP-05-069, and DP-10-001) and supported by the National Institute of Diabetes and Digestive and Kidney Diseases.

Site Contract Numbers: Kaiser Permanente Southern California (U48/CCU919219, U01 DP000246, and U18DP002714), University of Colorado Denver (U48/CCU819241-3, U01 DP000247, and U18DP000247-06A1), Kuakini Medical Center (U58CCU919256 and U01 DP000245), Children’s Hospital Medical Center (Cincinnati) (U48/CCU519239, U01 DP000248, and 1U18DP002709), University of North Carolina at Chapel Hill (U48/CCU419249, U01 DP000254, and U18DP002708-01), University of Washington School of Medicine (U58/CCU019235-4, U01 DP000244, and U18DP002710-01), Wake Forest University School of Medicine (U48/CCU919219, U01 DP000250, and 200-2010-35171).

The authors wish to acknowledge the involvement of General Clinical Research Centers (GCRC) at the South Carolina Clinical & Translational Research (SCTR) Institute, at the Medical University of South Carolina (NIH/NCRR Grant number UL1RR029882); Children’s Hospital and Regional Medical Center (Grant Number M01RR00037); Colorado Pediatric General Clinical Research Center (Grant Number M01 RR00069) and the Barbara Davis Center at the University of Colorado at Denver (DERC NIH P30 DK57516); and the Institutional Clinical and Translational Science Award (CTSA), NIH/NCRR at the University of Cincinnati (Grant Number 1UL1RR026314-01).

The contents of this paper are solely the responsibility of the authors and do not necessarily represent the official views of the Centers for Disease Control and Prevention and the National Institute of Diabetes and Digestive and Kidney Diseases.

CONFLICT OF INTEREST

LM Jaacks, R Oza-Frank, R D’Agostino, Jr., LM Dolan, D Dabelea, JM Lawrence, C Pihoker, MR O’Connor, B Linder, G Imperatore, M Seid, KMV Narayan, and EJ Mayer-Davis have no conflicts of interest to disclose.

REFERENCESPetittiDBKlingensmithGJBellRAAndrewsJSDabeleaDImperatoreGGlycemic control in youth with diabetes: The SEARCH for Diabetes in Youth StudyJ Pediatr2009155566867219643434Mayer-DavisEJMaBLawsonAD'AgostinoRBJrLieseADBellRACardiovascular disease risk factors in youth with type 1 and type 2 diabetes: Implications of a factor analysis of clusteringMetab Syndr Relat Disord200972899518847385MongerRUS legal permanent residents: 2009. Office of Immigration Statistics, US Department of Homeland SecurityAnnual Flow Report2010US Census BureauThe 2009 American Community Survey2010Washington DCUS Census BureauUS Census BureauCurrent Population Survey. 2010Annual Social and Economic Supplement2009HsinOLa GrecaAMValenzuelaJMoineCTDelamaterAAdherence and glycemic control among hispanic youth with type 1 diabetes: Role of family involvement and acculturationJournal of Pediatric Psychology201035215616619491214HarrisMAWysockiTSadlerMWilkinsonKHarveyLMBucklohLMValidation of a structured interview for the assessment of diabetes self-managementDiabetes Care20002391301130410977022CadarioFVercellottiATradaMZaffaroniMRapaAJafuscoDYounger age at diagnosis of type 1 diabetes mellitus in children of immigrated families born in ItalyJ Endocrinol Invest2004271091391815762037BaninPRimondiFDe TogniACantoniSChiariGIughettiLType 1 diabetes in children and adolescents of immigrated families in Emilia-Romagna (Italy)Acta Biomed2010811353920857851Tubiana-RufiNMoretLCzernichowPChwalowJRisk factors for poor glycemic control in diabetic children in FranceDiabetes Care19951811147914828722073SEARCH for Diabetes in Youth Study GroupSEARCH for Diabetes in Youth: A multicenter study of the prevalence, incidence and classification of diabetes mellitus in youthControl Clin Trials200425545847115465616SEARCH for Diabetes in Youth Study GroupLieseADD'AgostinoRBHammanRFKilgoPDLawrenceJMThe burden of diabetes mellitus among US youth: Prevalence estimates from the SEARCH for Diabetes in Youth StudyPediatrics200611841510151817015542SEARCH for Diabetes in Youth Study GroupDabeleaDBellRAD'AgostinoRBImperatoreGJohansenJMIncidence of diabetes in youth in the United StatesJAMA2007297242716272417595272United NationsDemographic Yearbook2008US Census BureauUnited States Census 20002000Washington, DCUS Department of CensusOMB no. 0607-0856.Consumer assessment of healthcare providers and systems (CAHPS) (tm) 2.0 questionnaires [Internet]1998Available from: http://www.ahrq.gov/qual/cahps/cahpques.htm.SeidMWaitzfelderBDabeleaDDolanLImperatoreGLinderBPrevalence of barriers to health care in the US youth with diabetes: The SEARCH for Diabetes in Youth StudyDiabetes200958Suppl. 1A271A272GreenlandSPearlJRobinsJMCausal diagrams for epidemiologic researchEpidemiology199910137489888278FloresGBrotanekJThe healthy immigrant effect: A greater understanding might help us improve the health of all childrenArch Pediatr Adolesc Med2005159329529715753277Oza-FrankRNarayanKMVOverweight and diabetes prevalence among US immigrantsAm J Public Health2010100466166819608956Oza-FrankRStephensonRNarayanKMDiabetes prevalence by length of residence among US immigrantsJ Immigr Minor Health20111311819688263SinghGKMillerBAHealth, life expectancy, and mortality patterns among immigrant populations in the United StatesCan J Public Health2004953114121Ingersoll-DaytonBSaengtienchaiCRespect for the elderly in Asia: Stability and changeInt J Aging Hum Dev199948211313010376957HernandezDJDemographic change and the life circumstances of immigrant familiesThe Future of Children2004141647GreyMBolandEAYuCSullivan-BolyaiSTamborlaneWVPersonal and family factors associated with quality of life in adolescents with diabetesDiabetes Care19982169099149614606CreatoreMIMoineddinRBoothGManuelDHDesMeulesMMcDermottSAge- and sex-related prevalence of diabetes mellitus among immigrants to Ontario, CanadaCMAJ2010182878178920403889FerreiraSRLerarioDDGimenoSGSanudoAFrancoLJJapanese-Brazilian Diabetes Study GroupObesity and central adiposity in japanese immigrants: Role of the western dietary patternJ Epidemiol200212643143812462278OstbyeTWelbyTJPriorIASalmondCEStokesYMType 2 (non-insulin-dependent) diabetes mellitus, migration and westernisation: The Tokelau Island Migrant StudyDiabetologia19893285855902777000HealdAHSharmaRAndersonSGVyasASiddalsKPatelJDietary intake and the insulin-like growth factor system: Effects of migration in two related populations in India and Britain with markedly different dietary intakePublic Health Nutr20058662062716236191MazurREMarquisGSJensenHHDiet and food insufficiency among Hispanic youths: Acculturation and socioeconomic factors in the Third National Health and Nutrition Examination SurveyAm J Clin Nutr20037861120112714668273AllenMLElliottMNMoralesLSDiamantALHambarsoomianKSchusterMAAdolescent participation in preventive health behaviors, physical activity, and nutrition: Differences across immigrant generations for Asians and Latinos compared with whitesAm J Public Health200797233734317138919State of California, Department of FinanceLegal immigration to California in 20022003Sacramento, CAOsypukTLDiez RouxAVHadleyCKandulaNRAre immigrant enclaves healthy places to live? The multi-ethnic study of atherosclerosisSoc Sci Med200969111012019427731CruzFADPadillaGVAgustinEOAdapting a measure of acculturation for cross-cultural researchJ Transcultural Nursing2000113191198

Demographic, socioeconomic and clinical characteristics of participants by diabetes type and parent migration status.a, b

Type 1 DiabetesType 2 Diabetes
US-bornParents(n = 2,217)≥1 Foreign-bornParent(n = 450)P-valuecUS-bornParents(n = 327)≥1 Foreign-bornParent(n = 92)P-valuec
DEMOGRAPHIC CHARACTERISTICS
Parent Duration in USd (years)N/A25.5 ± 11.2N/AN/A24.5 ± 10.8N/A
Parent Region of Origin
Latin America & CaribbeanN/A48.3 (213)N/AN/A70.0 (63)N/A
North America5.7 (25)0.0 (0)
Europe20.2 (89)1.1 (1)
Oceania1.6 (7)6.7 (6)
Asia19.3 (85)20.0 (18)
Africa5.0 (22)2.2 (2)
Age at Visite
2–9 years25.7 (535)28.0 (117)0.30.3 (1)3.7 (3)N/A
≥10 years74.3 (1,549)72.0 (301)99.7 (302)96.3 (79)
Age at Diagnosis (years)8.7 ± 4.48.3 ± 4.50.113.8 ± 2.714.3 ± 2.90.2
Male50.8 (1,126)48.2 (217)0.336.7 (120)48.9 (45)0.03
Race/Ethnicity
Non-Hispanic White72.1 (1,599)27.1 (122)<0.000121.1 (69)1.1 (1)<0.0001
Hispanic10.5 (232)49.6 (223)13.8 (45)69.6 (64)
Asian/Pacific Islander3.1 (68)15.3 (69)3.7 (12)27.2 (25)
Non-Hispanic Black13.3 (295)6.9 (31)45.0 (147)2.2 (2)
Other1.0 (23)1.1 (5)16.5 (54)0.0 (0)
SOCIOECONOMIC CHARACTERISTICS
Parent Highest Level of Education
≤ High School17.9 (391)30.0 (131)<0.000146.7 (146)58.8 (50)0.05
Some college or a degree82.1 (1,797)70.0 (306)53.4 (167)41.2 (35)
Family Structure
Two-parent household70.3 (1,544)72.3 (323)0.0937.7 (122)63.3 (57)<0.0001
Single-parent household26.6 (585)23.0 (103)51.2 (166)28.9 (26)
Other household structure3.1 (68)4.7 (21)11.1 (36)7.8 (7)
Estimated Total Annual Household Income
< $24,99913.9 (274)17.1 (66)0.00751.2 (132)43.6 (24)0.5
$25,000–$49,99920.8 (410)26.2 (101)24.8 (64)21.8 (12)
$50,000–$74,99919.7 (387)19.7 (76)12.8 (33)18.2 (10)
≥ $75,00045.6 (897)36.9 (142)11.2 (29)16.4 (9)
Type 1 DiabetesType 2 Diabetes
US-bornParents(n = 2,091)≥1 Foreign-bornParent(n = 420)P-valuecUS-bornParents(n = 303)≥1 Foreign-bornParent(n = 82)P-valuec
CLINCAL CHARACTERISTICS
PedsQL Score81.9 ± 12.479.3 ± 13.70.00176.9 ± 15.179.9 ± 10.50.2
Frequency of Glucose Monitoring
< 1 time per day2.0 (30)3.3 (10)0.526.6 (49)28.6 (12)0.7
1–2 times per day10.0 (149)9.9 (30)37.5 (69)42.9 (18)
3 times per day14.5 (216)15.9 (48)16.9 (31)16.7 (7)
≥ 4 times per day73.5 (1,094)70.9 (214)19.0 (35)11.9 (5)
Glycemic Control
A1C (%)8.47 ± 1.798.45 ± 1.780.98.37 ± 2.848.13 ± 2.700.5
A1C ≥ 9.5%22.7 (429)24.2 (93)0.533.9 (94)31.7 (25)0.7
Cardiovascular Disease Risk Factors
Family History of Diabetesf13.4 (199)14.5 (46)0.658.1 (125)46.0 (23)0.1
BMI (kg/m2)21.6 ± 5.821.7 ± 5.80.835.5 ± 8.532.5 ± 9.90.007
BMI-z0.59 ± 0.990.67 ± 0.970.11.96 ± 1.241.67 ± 1.060.06
Obeseg12.9 (260)15.3 (62)0.276.0 (219)59.8 (49)0.01
Waist Circumference (cm)75.8 ± 16.575.4 ± 15.40.7113.1 ± 21.2108.4 ± 26.50.1
Systolic Blood Pressure (mmHg)103.6 ± 11.7103.6 ± 12.30.96117.5 ± 12.5114.4 ± 11.70.05
Diastolic Blood Pressure (mmHg)66.1 ± 10.166.0 ± 10.20.873.2 ± 9.972.4 ± 10.10.5
Hypertensiveh5.2 (103)5.2 (21)0.9713.6 (39)9.9 (8)0.4
LDL Cholesterol (mg/dl)94.9 ± 28.895.0 ± 25.10.96105.3 ± 34.398.9 ± 31.40.1
HDL Cholesterol (mg/dl)56.8 ± 14.256.3 ± 14.00.542.1 ± 11.342.0 ± 12.80.9
Triglycerides (mg/dl)83.5 ± 90.383.8 ± 64.00.96178.6 ± 292.4197.7 ± 169.70.6

Sample sizes vary for clinical characteristics due to 190 participants in the 2008 incident cohort that completed the initial survey but did not attend the baseline visit. Unrounded percents sum to 100.

Values presented as mean ± SD or % (n).

Chi-square test for categorical outcomes; F test for continuous outcomes. Fisher’s Exact test for categorical outcomes with cell sizes less than or equal to 5.

Calculated using self-reported year of migration and the year of the baseline visit or most recent follow-up visit.

Calculated using self-reported date of birth and the date of the baseline visit or most recent follow-up visit.

Mother or father with diagnosed diabetes.

BMI ≥ 95th percentile for age and sex.

SBP or DBP ≥ 95th percentile for age, gender, and height.

Barriers to care by diabetes type and parent migration status.a, b

Type 1 DiabetesType 2 Diabetes
US-bornParents(n = 1,142)≥1 Foreign-bornParent(n = 183)P-valuecUS-bornParents(n = 133)≥1 Foreign-bornParent(n = 38)P-valuec
Insurance Coveraged
Private78.4 (793)69.5 (82)0.0339.3 (42)33.3 (3)0.7
Othere21.6 (218)30.5 (36)60.8 (65)66.7 (6)
Barriers to Care
Regular Doctor16.5 (188)23.1 (42)0.0340.2 (53)43.2 (16)0.7
Access to Care15.5 (177)16.9 (31)0.628.0 (37)10.8 (4)0.03
Cost of Care50.4 (572)48.1 (88)0.643.9 (58)29.7 (11)0.1
Medication33.4 (372)29.6 (50)0.332.1 (35)21.9 (7)0.3
Contextual Care24.2 (213)33.1 (49)0.0236.0 (18)60.0 (6)0.2
Communication50.8 (578)53.0 (97)0.653.0 (70)41.7 (15)0.2
Getting Information49.5 (275)56.8 (50)0.233.3 (10)42.9 (3)0.7

Sample sizes may vary slightly due to missing data. Unrounded percents sum to 100.

Values presented as % (n).

Chi-square test for categorical outcomes; F test for continuous outcomes. Fisher’s Exact test for categorical outcomes with cell sizes less than or equal to 5.

Proportions excluding California and Hawaii SEARCH sites because these sites were healthcare plan sites and therefore all participants from these sites would be coded as having private insurance.

Includes none (n = 32 for type 1 diabetes; n = 19 for type 2 diabetes), Medicaid/Medicare (n = 202 for type 1 diabetes; n = 46 for type 2 diabetes), and other (n = 20 for type 1 diabetes; n = 6 for type 2 diabetes).

Odds ratios (95% confidence intervals) for logistic regression models predicting barriers to care among participants with type 1 diabetes in the 2002–2005 incident cohorts. Participants with US-born parents were the referent group.

Regular DoctorAccess to CareCost of CareMedicationContextual CareCommunicationGetting Information
Unadjusted1.51 (1.04, 2.21)1.11 (0.73, 1.69)0.91 (0.67, 1.25)0.84 (0.59, 1.19)1.55 (1.07, 2.26)1.09 (0.80, 1.50)1.34 (0.85, 2.12)
Adjusted race/ethnicity and SEARCH sitea1.10 (0.72, 1.67)0.97 (0.62, 1.54)0.98 (0.69, 1.38)0.80 (0.55, 1.17)1.14 (0.76, 1.72)1.17 (0.83, 1.65)1.16 (0.71, 1.89)

Final adjustment set for barriers to care outcome in type 1 diabetes stratum given 10% change-in-estimate criteria.

Odds ratios (95% confidence intervals) for binary logistic regression models predicting poor glycemic control (A1C ≥ 9.5%) or obesity (BMI ≥ 95th percentile for age and sex) by diabetes type. Participants with US-born parents were the referent group.

Type 1 Diabetes(n = 2,511)Type 2 Diabetes(n = 385)
Poor Glycemic Control (A1C ≥ 9.5%)
Unadjusted1.08 (0.84, 1.40)0.90 (0.53, 1.54)
Adjusted race/ethnicitya0.73 (0.55, 0.96)0.77 (0.45, 1.33)
Adjusted diabetes durationb1.06 (0.82, 1.38)0.99 (0.57, 1.71)
Adjusted SEARCH sitec0.96 (0.74, 1.27)0.68 (0.36, 1.30)
Adjusted parental educationd1.01 (0.78, 1.32)0.94 (0.54, 1.62)
Adjusted household incomee1.10 (0.83, 1.46)1.16 (0.60, 2.24)
Adjusted race/ethnicity and SEARCH sitef0.70 (0.53, 0.94)
Adjusted race/ethnicity, SEARCH site, and household incomeg0.65 (0.29, 1.45)
Obesity (BMI ≥ 95th percentile for age and sex)
Unadjusted1.22 (0.90, 1.65)0.57 (0.34, 0.95)
Adjusted race/ethnicitya0.87 (0.63, 1.20)0.60 (0.36, 1.01)
Adjusted diabetes durationb1.23 (0.91, 1.65)0.48 (0.28, 0.81)
Adjusted SEARCH sitec1.16 (0.84, 1.60)0.47 (0.28, 0.80)
Adjusted parental educationd1.05 (0.77, 1.44)0.67 (0.36, 1.24)
Adjusted household incomee1.09 (0.78, 1.51)0.47 (0.25, 0.88)
Adjusted race/ethnicity, parental education, and household incomeh0.79 (0.55, 1.12)
Adjusted diabetes duration, parental education, and household incomei0.35 (0.17, 0.70)

Non-Hispanic white versus all other race/ethnicities. Coded as a binary variable.

Continuous.

South Carolina, Ohio, California, Colorado, Washington, Hawaii. Coded using indicator variables with Colorado as the referent.

Less than or equivalent to high school versus some college or degree. Coded as a binary variable.

Less than $24,999, $25,000–$49,999, $50,000–$74,999, ≥$75,000. Coded using indicator variables with ≥$75,000 as the referent.

Final adjustment set for outcome of poor glycemic control and A1C in type 1 diabetes stratum given 10% change-in-estimate criteria.

Final adjustment set for outcome of poor glycemic control and A1C in type 2 diabetes stratum given 10% change-in-estimate criteria.

Final adjustment set for outcome of obesity and BMI z-scores in type 1 diabetes stratum given 10% change-in-estimate criteria.

Final adjustment set for outcome of obesity and BMI z-scores in type 2 diabetes stratum given 10% change-in-estimate criteria.