Trends in Prevalence of Type 1 and Type 2 Diabetes in Children and Adolescents in the US, 2001-2017

Key Points

Question  Has the prevalence of type 1 and type 2 diabetes changed from 2001 to 2017 among those younger than 20 years in the US?

Findings  In this observational, cross-sectional, multicenter study of a mean of 3.47 million youths for each prevalence year from 6 areas in the US, the estimated prevalence of type 1 diabetes among those 19 years or younger increased significantly, from 1.48 per 1000 youths to 2.15 per 1000 youths, and the estimated prevalence of type 2 diabetes among those aged 10 to 19 years increased from 0.34 per 1000 youths to 0.67 per 1000 youths.

Meaning  In 6 areas of the US from 2001 to 2017, the estimated prevalence of diabetes among children and adolescents increased significantly for both type 1 and type 2 diabetes.

Importance  Changes in the prevalence of youth-onset diabetes have previously been observed.

Objective  To estimate changes in prevalence of type 1 and type 2 diabetes in youths in the US from 2001 to 2017.

Design, Setting, and Participants  In this cross-sectional observational study, individuals younger than 20 years with physician-diagnosed diabetes were enumerated from 6 areas in the US (4 geographic areas, 1 health plan, and select American Indian reservations) for 2001, 2009, and 2017.

Exposures  Calendar year.

Main Outcomes and Measures  Estimated prevalence of physician-diagnosed type 1 and type 2 diabetes overall and by race and ethnicity, age, and sex.

Results  Among youths 19 years or younger, 4958 of 3.35 million had type 1 diabetes in 2001, 6672 of 3.46 million had type 1 diabetes in 2009, and 7759 of 3.61 million had type 1 diabetes in 2017; among those aged 10 to 19 years, 588 of 1.73 million had type 2 diabetes in 2001, 814 of 1.85 million had type 2 diabetes in 2009, and 1230 of 1.85 million had type 2 diabetes in 2017. The estimated type 1 diabetes prevalence per 1000 youths for those 19 years or younger increased significantly from 1.48 (95% CI, 1.44-1.52) in 2001 to 1.93 (95% CI, 1.88-1.98) in 2009 to 2.15 (95% CI, 2.10-2.20) in 2017, an absolute increase of 0.67 per 1000 youths (95%, CI, 0.64-0.70) and a 45.1% (95% CI, 40.0%-50.4%) relative increase over 16 years. The greatest absolute increases were observed among non-Hispanic White (0.93 per 1000 youths [95% CI, 0.88-0.98]) and non-Hispanic Black (0.89 per 1000 youths [95% CI, 0.88-0.98]) youths. The estimated type 2 diabetes prevalence per 1000 youths aged 10 to 19 years increased significantly from 0.34 (95% CI, 0.31-0.37) in 2001 to 0.46 (95% CI, 0.43-0.49) in 2009 to 0.67 (95% CI, 0.63-0.70) in 2017, an absolute increase of 0.32 per 1000 youths (95% CI, 0.30-0.35) and a 95.3% (95% CI, 77.0%-115.4%) relative increase over 16 years. The greatest absolute increases were observed among non-Hispanic Black (0.85 per 1000 youths [95% CI, 0.74-0.97]) and Hispanic (0.57 per 1000 youths [95% CI, 0.51-0.64]) youths.

Conclusions and Relevance  In 6 areas of the US from 2001 to 2017, the estimated prevalence of diabetes among children and adolescents increased for both type 1 and type 2 diabetes.

Youth-onset type 1 and type 2 diabetes are serious chronic health conditions. Individuals with youth-onset diabetes are at risk for early complications, comorbidities, and excess mortality, particularly those who develop type 2 diabetes and those from racial and ethnic minority groups.^1,2 The SEARCH for Diabetes in Youth Study reported increases in the prevalence of type 1 diabetes and type 2 diabetes from 2001 to 2009³ and estimated that there were 187 246 US youths 19 years or younger with diabetes in 2009.⁴ In an article published in 2020, data from the SEARCH study demonstrated increases in the incidence of type 1 and type 2 diabetes in the US from 2002 through 2015.⁵ Although non-Hispanic White (hereafter referred to as White) youths (aged 19 years or younger) have the highest type 1 diabetes incidence, larger relative percent increases in incidence were observed among non-Hispanic Black (hereafter referred to as Black) and Hispanic youths from 2002 through 2015 and among Asian or Pacific Islander youths from 2011 through 2015 compared with White youths. American Indian and Black youths aged 10 to 19 years had the highest type 2 diabetes incidence. Significant increases in type 2 diabetes incidence were observed among American Indian, Asian or Pacific Islander, Black, and Hispanic youths.⁵

To monitor trends in youth-onset diabetes prevalence, data for 2001, 2009, and 2017 were analyzed to establish whether there were differences in any observed trends by diabetes type, race and ethnicity, age, and sex. Elucidating differences in diabetes prevalence trends by diabetes type and demographic characteristics is essential to describe the burden of disease and to estimate current and future resource needs.

The methods,⁶ incidence rates (2002-2015),^5,7and prevalence estimates (2001 and 2009)^3,4,8 for the SEARCH study have been previously published. Changes in estimated prevalence between 2001, 2009, and 2017, the 3 years for which prevalence was assessed in this study, were described in the current study. Case ascertainment was conducted under a Health Insurance Portability and Accountability Act Privacy Rule waiver of consent approved by the institutional review board(s) with jurisdiction for each of the clinical centers and the data coordinating center.

Data were collected from clinical centers located in 6 areas: California (Kaiser Permanente Southern California members in 7 counties, including Los Angeles), Colorado (residents of 14 counties, including Denver), Ohio (residents of 8 counties, including Cincinnati), South Carolina (residents of 4 counties, including Columbia), Washington State (residents of 5 counties, including Seattle), and (coordinated by the Colorado center) Indian Health Services users in select areas of Arizona and New Mexico.

Methods of type 1 diabetes and type 2 diabetes case ascertainment and prevalence estimation were the same in the 3 periods and included an additional 22-month window of ascertainment after the end of each prevalence year. Eligible individuals included those younger than 20 years at the end of the prevalent years (ie, December 31, 2001; December 31, 2009; and December 31, 2017) with physician-diagnosed diabetes who were residing in the geographic study areas, Kaiser Permanente Southern California members, or Indian Health Services enrollees during the prevalence year. Active-duty military personnel and individuals who were institutionalized were ineligible for inclusion.

Active surveillance networks for the 5 clinical centers were composed of pediatric and adult endocrinologists and other clinicians, hospitals, and health plans in the study areas. Potential youths with diabetes identified using International Classification of Diseases, Ninth Revision or International Classification of Diseases, Tenth Revision diagnosis codes, based on prescription medications used to manage diabetes, or from laboratory values indicative of diabetes were validated by verifying the diagnosis of diabetes and date of diagnosis in the clinical notes of the medical record, direct physician confirmation, or presence of these case reports in clinical diabetes databases that required physician verification for inclusion.

Race and ethnicity were examined in this study because of their association with diabetes prevalence.^3,4 Race and ethnicity were based primarily on self-reported information collected on a short survey administered to parents or guardians of children younger than 18 years and adults 18 years or older using the 2000 US census race and ethnicity categories, with medical records serving as the secondary source. When both sources were missing, race and ethnicity were imputed based on the racial and ethnic distribution of the residents of their census track in the prevalent year. Clinician-reported diabetes type was obtained from medical records via manual review. Sex and date of birth were obtained primarily from self-report, with medical records serving as the secondary source.

The denominators for the geographically based sites were the 2001, 2009, and 2017 bridged-race intercensal population estimates (vintage 2018) by race and ethnicity, age, and sex.⁹ Counts of active-duty military personnel were removed. For Kaiser Permanente Southern California, addresses of members were geocoded to the census block level and race and ethnicity–specific proportions of the population in the census blocks were applied to these counts to estimate the racial and ethnic composition of the membership by sex and age in 1-year increments for each prevalence year.¹⁰ Users of the Indian Health Services for the 3 years preceding the prevalence years were counted to define the American Indian study population in the select areas. Race and ethnicity categories included Hispanic ethnicity (regardless of race); race-bridging methods⁹ were then applied to multiracial youths (those who were not Hispanic and self-reported as 2 or more races) to determine the probability of belonging to each of the 4 racial groups. The 5 race and ethnicity categories were American Indian, Asian or Pacific Islander, Black, Hispanic (regardless of race), and White. All race categories included only individuals who were not Hispanic.

The primary outcome was the estimated prevalence of type 1 diabetes and type 2 diabetes per 1000 youths and accompanying 95% CIs in 2001, 2009, and 2017 overall (total sample) and individually stratified by race and ethnicity, age category, and sex. The secondary outcomes were the estimated prevalence of type 1 diabetes and type 2 diabetes simultaneously stratified by race and ethnicity, age category, and sex.

Physician-assigned diabetes type was compared with an etiologic assessment of diabetes type. Etiologic type 1 diabetes was defined as the presence of at least 1 autoantibody (glutamic acid decarboxylase, insulinoma-associated, or zinc transporter 8 autoantibody) or being insulin sensitive in the absence of diabetes autoantibodies, and type 2 diabetes was defined by insulin resistance in the absence of diabetes autoantibodies.^11-14 In-person study visits, including blood collection, were conducted for some prevalent cases in 2001 but were not conducted for prevalent cases in 2009 or 2017. Therefore, antibody test results and insulin sensitivity/resistance assessed for individuals with incident diabetes in 2002, 2008, and 2016 who completed an in-person study visit were used as a proxy to estimate the concordance between physician-assigned diabetes type and etiologic type for each of the 3 adjacent prevalent years.

Estimated prevalence was expressed as diabetes cases per 1000 youths with 95% CIs. Both the numerators (number of cases) and denominators (total population) were aggregated across the 6 areas from 5 clinical centers prior to calculating the estimated prevalence. Type 1 diabetes prevalence was reported in 4 age groups and type 2 diabetes prevalence was reported in 2 age groups, because the number of children 9 years or younger with type 2 diabetes was insufficient to calculate prevalence. Differences between prevalence estimates were calculated for 8-year intervals (2001-2009 and 2009-2017) and for the 16-year period (2001-2017) using a 2-sided skew-corrected inverted-score test assuming a binomial distribution.¹⁵ An unadjusted Poisson model using the prevalence years as a categorical variable was used to test whether 2009 prevalence was the midpoint of 2001 and 2017 prevalence. Annual percent change was calculated for each interval using the parameter estimates (β) from a continuous measure of time included in Poisson models estimating prevalence. Assuming linear trends and a significance level of 5%, initial power analyses conducted before the 2017 prevalence data were observed suggested that there was 90% power to detect changes in prevalence of 0.04 per 1000 youths for type 1 diabetes and 0.28 per 1000 youths for type 2 diabetes. In the exploratory analysis of the joint associations of race and ethnicity, age, and sex with diabetes prevalence (secondary outcome), P values with 2 degrees of freedom were used to test for significant differences among the 3 prevalent years.

Comparisons between etiologic type based on presence of diabetes autoantibodies and insulin sensitivity and clinician type for 3 three incident years were done using Fisher exact tests. The number and percent of cases in agreement with etiologic type were reported.

All tests were 2-sided, P < .05 was considered significant, and 95% CIs were reported for all prevalence estimates and differences in these estimates. All analyses were done using SAS, version 9.4 (SAS Institute).

Among youths (mean of 3.47 million youths for each prevalence year from 6 areas in the US) 19 years or younger, 4958 of a total population of 3.35 million had type 1 diabetes in 2001, 6672 of 3.46 million had type 1 diabetes in 2009, and 7759 of 3.61 million had type 1 diabetes in 2017 (Table 1). Among youths aged 10 to 19 years, 588 of 1.73 million had type 2 diabetes in 2001, 814 of 1.78 million had type 2 diabetes in 2009, and 1230 of 1.85 million had type 2 diabetes in 2017. Children 9 years or younger with type 2 diabetes were excluded from the analysis because of the small sample sizes (n = 5 in 2001; n = 18 in 2009; n = 8 in 2017). Race and ethnicity data were not available from self-report or medical records for 4.5% of those with type 1 or type 2 diabetes in 2001, 3.0% of those with type 1 or type 2 diabetes in 2009, and 2.1% those with type 1 or type 2 diabetes in 2017.

The estimated type 1 diabetes prevalence per 1000 youths for those 19 years or younger increased significantly from 1.48 (95% CI, 1.44-1.52) in 2001 to 1.93 (95% CI, 1.88-1.98) in 2009 to 2.15 (95% CI, 2.10-2.20) in 2017 (Table 1), which represented a significant absolute increase of 0.67 per 1000 youths (95%, CI, 0.64-0.70) and a 45.1% (95% CI, 40.0%-50.4%) relative increase over 16 years (Table 2). Overall, the annual percent change was 3.4% (95% CI, 2.9%-3.8%) in the first interval (2001-2009) and 1.4% (95% CI, 1.0%-1.8%) in the second interval (2009-2017); the change was not linear (P < .001). In the first interval, statistically significant increases were observed within each age, race and ethnicity, and sex subgroup, except for those 4 years or younger and American Indian youths. In the second interval, significant increases were observed among those aged 10 to 14 years and those aged 15 to 19 years in each race and ethnicity and sex subgroup. The largest absolute increases in estimated prevalence by race and ethnicity from 2001 to 2017 were among White (0.93 per 1000 [95% CI, 0.88-0.98]) and Black (0.89 per 1000 [95% CI, 0.80-0.99]) youths. Estimated prevalence increases were significantly greater in the first compared with the second interval for both sexes; the annual percent change was 3.4% (95% CI, 2.7%-4.1%) in the first interval compared with 1.1% (95% CI, 0.5%-1.7%) in the second interval for females (P < .001) and 3.3% (95% CI, 2.7%-4.0%) in the first interval compared with 1.6% (95% CI, 1.0%-2.2%) in the second for males (P = .02). The estimated prevalence increases from 2001 to 2017 were not significantly different by sex (0.71 per 1000 [95% CI, 0.66-0.75] for males vs 0.63 per 1000 [95% CI, 0.58-0.67] for females; P = .32).

The estimated prevalence of type 2 diabetes per 1000 youths for those aged 10 to 19 years increased significantly from 0.34 (95% CI, 0.31-0.37) in 2001 to 0.46 (95% CI, 0.43-0.49) in 2009 to 0.67 (95% CI, 0.63-0.70) in 2017 (Table 1), which represented a significant absolute increase of 0.32 per 1000 youths (95% CI, 0.30-0.35) and a 95.3% (95% CI, 77.0%-115.4%) relative increase over 16 years (Table 2). The annual percent change was 3.7% (95% CI, 2.4%-5.1%) in the first interval and 4.8% (95% CI, 3.7%-6.0%) in the second interval; the change was linear (P = .35). In the first interval, statistically significant increases in the estimated prevalence of type 2 diabetes were observed among White and Hispanic youths, within both age groups, and for both sexes. In the second interval, significant increases in estimated type 2 diabetes prevalence were observed for all subgroups by age, sex, and race and ethnicity, except for White youths.

The greatest absolute increase in the estimated prevalence of type 2 diabetes from 2001 to 2017 was among Black (0.85 per 1000 youths; 95% CI, 0.74 to 0.97) and Hispanic (0.57 per 1000 youths; 95% CI, 0.51 to 0.64) youths. Increases were not linear; Hispanic youths had a significantly greater increase in the first interval compared with the second interval (annual percent change, 7.3% [95% CI, 4.7%-10.0%] vs 3.2% [95% CI, 1.4%-5.0%]; P = .03), while Black youths had no significant increase in the first interval and a significant increase in the second interval (annual percent change, 1.2% [95% CI, −1.3% to 3.8%] vs 7.1% [95% CI, 4.8%-9.4%]; P = .007). Significant linear increases were observed in those aged 10 to 14 years, with a significant overall increase in estimated prevalence of 0.13 per 1000 youths (95% CI, 0.11-0.16) from 2001 to 2017. For those aged 15 to 19 years, the increase in estimated prevalence of type 2 diabetes was significantly greater in the second interval (0.36 per 1000 youths [95% CI, 0.32-0.41]) than in the first interval (0.14 per 1000 youths [95% CI, 0.10-0.18]) (P = .04).

Estimated prevalence and 95% CIs over the 3 periods stratified simultaneously by race and ethnicity, age, sex are shown for type 1 diabetes in Figure 1 and for type 2 diabetes in Figure 2, with all prevalence estimates and 95% CIs shown in eTable 1 and eTable 2 in the Supplement. There was no significant difference in estimated type 1 diabetes prevalence among those aged 4 years or younger across all 3 periods. Significant differences in estimated prevalence between study years were observed among those aged 5 to 9 years for Hispanic females and White males and females and among those aged 10 to 14 years for Asian or Pacific Islander, Black, Hispanic, and White males and American Indian females (Figure 1; eTable 1 in the Supplement).

Significant differences in estimated type 2 diabetes prevalence were observed among Black females and males and Hispanic females aged 10 to 14 years, but not among others in this age group (Figure 2; eTable 2 in the Supplement). Among those aged 15 to 19 years, significant differences were observed for females in all race and ethnicity groups, except White females, and for Black, Hispanic, and White males.

There were no significant differences between etiologic type and physician’s diagnosis of type 1 diabetes overall or by age, sex, or race and ethnicity subgroup or for type 2 diabetes overall (Table 3). However, among White participants with physician-diagnosed type 2 diabetes, the percentage of participants who met criteria for etiologic type in 2002 (50.0%) was significantly lower than the other 2 incident years (85.7% [P = .03] for 2009 and 86.2% [P = .02] for 2017). No significant differences were observed for other race and ethnicity groups. Significant differences were observed among those aged 10 to 14 years in 2002 vs 2008 (P = .03) and among males in 2002 vs 2016 (P = .03). No significant differences were observed for those aged 15 to 19 years or for females.

In 6 regions in the US, significant increases in the estimated prevalence of type 1 diabetes and type 2 diabetes were observed from 2001 to 2017. Although the percentage increase in prevalence was greater for type 2 diabetes, the absolute prevalence increase was greater for type 1 diabetes, which remains more common than type 2 diabetes in youth. The absolute increases in the estimated prevalence of type 1 diabetes were greatest among Black and White youths. Type 2 diabetes remained more common among racial and ethnic minority youths, with the absolute increases in estimated type 2 diabetes prevalence being greatest among Black youth and Hispanic youth.

Estimates of type 1 diabetes prevalence based on US MarketScan Multi-State data for those younger than 18 years were 1.29 per 1000 Medicaid recipients and 1.48 per 1000 commercially insured youths for 2002, 2.04 per 1000 Medicaid recipients and 2.10 per 1000 commercially insured youths for 2009, and 2.34 per 1000 Medicaid recipients and 2.32 per 1000 commercially insured youths for 2016.^16,17 These estimates were similar to the findings from the current study. Moreover, the US MarketScan data and findings from the current study highlight that the highest estimated prevalence of type 1 diabetes was among White youths, with the next highest estimates among Black youths. A study from the Netherlands reported a significant increase in prevalence of type 1 diabetes from approximately 1.0 per 1000 youths younger than 20 years to 1.5 per 1000 youths younger than 20 years from 1998 to 2011, with a significant annual percent change of 3.8% during this period,¹⁸ which is similar to the 3.4% annual percent change in the comparable period (2001-2009) for the current study.

In contrast with the present study, which yielded type 2 diabetes prevalence estimates ranging from 0.34 to 0.67 per 1000 youths for the 3 years for which prevalence was assessed and a doubling of prevalence in 16 years, a study conducted in China among those aged 7 to 17 years yielded type 2 diabetes prevalence estimates of 2.4 per 1000 youths in 2002 and 5.2 per 1000 youths in 2012.¹⁹ The US MarketScan multistate data yielded prevalence per 1000 youths of 0.70 in 2002, 2.21 in 2009, and 2.12 in 2016, which represents an absolute increase of 1.42 per 1000 youths, or 200%, in 14 years.¹⁶ The type 2 diabetes prevalence in the latter study was highest in the oldest age group (12-17 years), among Black participants, and among female participants, which mirrors the findings from the current study. A Canadian study reported an increase in the prevalence of type 2 diabetes among First Nations youths younger than 20 years from 1.7 per 1000 youths in 1995 to 5.7 per 1000 youths in 2015, an absolute increase of 4.0 per 1000 youths, or 235%, over 20 years.²⁰ In these 3 studies, the type 2 diabetes prevalence estimates were higher than the current study, but the prevalence approximately doubled over the observation period in each study. Reports of type 2 diabetes prevalence in individuals younger than 20 years in a German study ranged from 0.23 to 0.24 per 1000 youths from 2004 to 2016,²¹ while prevalence of type 2 diabetes in 2012 to 2013 in the UK for those younger than 16 years (0.03 per 1000 youths) were both much lower than in the current study, although race and ethnicity variation in prevalence was observed in the UK study.²²

Increases in prevalence can be driven by increased incidence, declining mortality, or both. The SEARCH study recently reported significant increases in the incidence of type 1 and type 2 diabetes from 2002 to 2015, which is likely a driver of increasing prevalence in the US.⁵ Steeper increases in age- and sex-adjusted incidence of type 1 diabetes from 2002 to 2015 were seen among Black and Hispanic individuals than among White individuals, which likely contributed to the narrowing in the prevalence gap by race and ethnicity. Although the etiology of type 1 diabetes is unknown, environmental factors (infectious and mucosal exposures in the first 2 years of life) interacting with type 1 diabetes susceptibility genes appear to play a role.^23-29 Increases in type 2 diabetes incidence is a likely driver of increasing prevalence of type 2 diabetes. Changes in anthropometric risk factors appear to play a significant role. Specifically, childhood obesity increased from 13.9% in 1999-2000 to 18.5% in 2015-2016.³⁰ Black and Mexican American teenagers experienced the greatest increase in prevalence of obesity/severe obesity from 1999 to 2018, which may contribute to race and ethnicity differences.³¹ Other contributing factors may include increases in exposure to maternal obesity and diabetes (gestational and type 2 diabetes)^32,33 and exposure to environmental chemicals.^34,35

Increasing awareness of type 2 diabetes in youth may have led to different screening practices that may also have contributed to the increases. In the current study, the diagnosis and classification of diabetes type were both based on physician report as part of clinical care. Although a change in diagnostic criteria to include hemoglobin A_1c of at least 6.5% in the diagnostic criteria for adults was recommended by the American Diabetes Association in 2010,³⁶ data for adolescents suggest that this recommendation is controversial. Expert pediatric groups do not recommend using hemoglobin A_1c alone to diagnosis diabetes.^37-39 Thus, it is unlikely that the introduction of hemoglobin A_1c into the diagnostic criteria in adults had an influence on the prevalence in 2017.

Considering the entire 16-year period, the differences in estimated type 1 diabetes prevalence from 2001 to 2017 equated to 0.72 to 1.70 additional cases per 1000 Black, Hispanic, and White individuals. Prevalence differences in the second interval (2009-2017) exceeded 0.5 additional cases per 1000 individuals among Black, Hispanic, and White females and males and American Indian and Asian or Pacific Islander females. Translating the difference in estimated type 2 diabetes prevalence from 2001 to 2017 into additional cases indicated that Black and Hispanic females aged 15 to 19 years experienced an estimated increase of more than 1 additional case per 1000 females.

This study has several limitations. First, only individuals with diagnosed diabetes were included, which missed those with undiagnosed type 2 diabetes; however, the number of individuals missed for this reason was likely small.⁴⁰ Undiagnosed cases were less likely among those with type 1 diabetes due to symptom severity at onset. Second, there were relatively small numbers of individuals in some race and ethnicity groups, particularly American Indian and Asian or Pacific Islander, reducing precision of these prevalence estimates and the ability to identify trends, if present, for these groups. Third, the correspondence between physician-assigned type 2 diabetes and etiologic type was lower than for type 1 diabetes and lower in 2002 than in subsequent years, particularly for White individuals. Fourth, because 2017 was the last year of prevalent case ascertainment in this study, it is uncertain whether the findings from 2017 reflect the prevalence in 2021. Fifth, although the data in this study were collected from 6 geographic regions of the US and the populations were racially and ethnically diverse, the 2020 census data were not available for comparison with the 2017 sample and the comparisons for 2001 with the 2000 census and 2009 with the 2010 census showed underrepresentation of Black youths and overrepresentation of Asian and Pacific Islander youths.²

In 6 areas of the US from 2001 to 2017, the estimated prevalence of diabetes among children and adolescents increased for both type 1 diabetes and type 2 diabetes.

Corresponding Author: Jean M. Lawrence, ScD, MPH, MSSA, Division of Diabetes, Endocrinology & Metabolic Diseases, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, 6707 Democracy Blvd, Room 6049, Bethesda, MD 20892 (jean.lawrence@nih.gov).

Accepted for Publication: June 25, 2021.

Correction: This article was corrected on October 5, 2021, to fix an error in Figure 2 in which the x-axis was mislabeled as the prevalence of type 1 diabetes, rather than the prevalence of type 2 diabetes. The x-axis is now labeled correctly.

Author Contributions: Dr Lawrence and Mr Isom had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Lawrence, Divers, Saydah, Imperatore, Pihoker, Mayer-Davis, Dabalea, Pettitt, Liese.

Acquisition, analysis, or interpretation of data: Lawrence, Divers, Isom, Imperatore, Pihoker, Marcovina, Hamman, Dolan, Dabalea, Liese.

Drafting of the manuscript: Lawrence, Divers, Isom, Liese.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Divers, Isom, Imperatore, Dabalea.

Obtained funding: Lawrence, Divers, Saydah, Pihoker, Hamman, Dolan, Dabalea, Liese.

Administrative, technical, or material support: Saydah, Pihoker, Marcovina, Dabalea.

Supervision: Lawrence, Saydah, Dolan, Dabalea, Liese.

Conflict of Interest Disclosures: None reported.

Funding/Support: The authors wish to acknowledge the involvement of the Kaiser Permanente Southern California’s Marilyn Owsley Clinical Research Center (funded by Kaiser Foundation Health Plan and supported in part by the Southern California Permanente Medical Group); the South Carolina Clinical & Translational Research Institute at the Medical University of South Carolina, National Institutes of Health (NIH)/National Center for Advancing Translational Sciences (NCATS) grant UL1 TR000062 and UL1 Tr001450; Seattle Children's Hospital and the University of Washington, NIH/NCATS grant UL1 TR00423; University of Colorado Pediatric Clinical and Translational Research Center, NIH/NCATS grant UL1 TR000154; the Barbara Davis Center at the University of Colorado at Denver (DERC NIH grant P30 DK57516); the University of Cincinnati, NIH/NCATS grant UL1 TR000077 and UL1 TR001425; and the Children with Medical Handicaps program managed by the Ohio Department of Health. SEARCH 4: The Population Based Registry of Diabetes in Youth Study (1U18DP006131, U18DP006133, U18DP006134, U18DP006136, U18DP006138, U18DP006139) is funded by the Centers for Disease Control and Prevention (DP-15-002) and by the NIH/National Institute of Diabetes and Digestive and Kidney Diseases. The SEARCH for Diabetes in Youth Cohort Study (1R01DK127208-01, 1UC4DK108173) is funded by the NIH/National Institute of Diabetes and Digestive and Kidney Diseases. SEARCH 1-3: SEARCH for Diabetes in Youth was funded by the Centers for Disease Control and Prevention (00097, DP-05-069, and DP-10-001) and supported by the National Institute of Diabetes and Digestive and Kidney Diseases. Kaiser Permanente Southern California (U48/CCU919219, U01 DP000246, and U18DP002714), University of Colorado Denver (U48/CCU819241-3, U01 DP000247, and U18DP000247-06A1), Cincinnati's Children's Hospital Medical Center (U48/CCU519239, U01 DP000248, and 1U18DP002709), University of North Carolina at Chapel Hill (U48/CCU419249, U01 DP000254, and U18DP002708), Seattle Children's Hospital (U58/CCU019235-4, U01 DP000244, and U18DP002710-01] and Wake Forest University School of Medicine (U48/CCU919219, U01 DP000250, and 200-2010-35171).

Role of the Funder/Sponsor: The project scientists (G.I. and S.S.) for the Centers for Disease Control and Prevention were involved in the design and conduct of the study; the interpretation of the data; the review and approval of the manuscript; and the decision to submit and where to submit the manuscript for publication. The project scientist from the National Institute of Diabetes and Digestive and Kidney Diseases was not involved with this manuscript. J.M.L. was a principal investigator for this study through the end of the funding period before joining the National Institute of Diabetes and Digestive and Kidney Diseases. The manuscript was submitted for clearance by the Centers for Disease Control and Prevention and the National Institute of Diabetes and Digestive and Kidney Diseases prior to submission to JAMA, which is required for all manuscripts with authors from these agencies.

SEARCH for Diabetes in Youth Study Group: The writing group acknowledges the contributions of the following individuals to the SEARCH for Diabetes in Youth Study (the principal investigator [PI] is listed first followed by other individuals on the study team in alphabetical order for each site): California: (PI) Jean M. Lawrence, ScD, MPH, MSSA (now with the National Institute of Diabetes and Digestive and Kidney Diseases); Peggy Hung, MPH; Corinna Koebnick, PhD, MSc; Xia Li, MS; Eva Lustigova, MPH; Kristi Reynolds, PhD, MPH (Department of Research & Evaluation, Kaiser Permanente Southern California, Pasadena California), and David J. Pettitt, MD (Santa Barbara, California). The Carolinas: (PI) Elizabeth J. Mayer-Davis, PhD; Amy Mottl, MD, MPH; Joan Thomas MS, RD (University of North Carolina, Chapel Hill); Malaka Jackson, MD; Lisa Knight, MD; Angela D. Liese, PhD, MPH; Christine Turley, MD (University of South Carolina); Deborah Bowlby, MD (Medical University of South Carolina); James Amrhein, MD; Elaine Apperson, MD; Bryce Nelson, MD (Greenville Health System and Eau Claire Cooperative Health Center). Colorado: (PI) Dana Dabelea, MD, PhD; Anna Bellatorre, PhD; Tessa Crume, PhD, MSPH; Richard F. Hamman, MD, DrPH; Katherine A. Sauder, PhD; Allison Shapiro, PhD, MPH; Lisa Testaverde, MS (LEAD Center in the Department of Epidemiology, Colorado School of Public Health, University of Colorado Denver); Georgeanna J. Klingensmith, MD; David Maahs, MD; Marian J. Rewers, MD, PhD; Paul Wadwa, MD (Barbara Davis Center for Childhood Diabetes); Stephen Daniels, MD, PhD; Michael G. Kahn, MD, PhD; Greta Wilkening, PsyD (Department of Pediatrics and Children’s Hospital); Clifford A. Bloch, MD (Pediatric Endocrine Associates); Jeffrey Powell, MD, MPH (Shiprock Service Unit, Navajo Area Indian Health Service); Kathy Love-Osborne, MD (Denver Health and Hospital Authority); Diana C. Hu, MD (Pediatrics Department, Tuba City Regional Health Care Center, Tuba City, AZ). Ohio: (PI) Lawrence M. Dolan, MD; Amy S. Shah, MD, MS; Debra A. Standiford, MSN, CNP; Elaine M. Urbina, MD, MS (Cincinnati Children’s Hospital Medical Center, Department of Pediatrics, University of Cincinnati). Washington: (PI) Catherine Pihoker, MD; Irl Hirsch, MD; Grace Kim, MD; Faisal Malik, MD, MSHS; Lina Merjaneh, MD; Alissa Roberts, MD; Craig Taplin, MD; Joyce Yi-Frazier, PhD (University of Washington); Natalie Beauregard, BA; Cordelia Franklin, BS; Carlo Gangan, BA; Sue Kearns, RN; Mary Klingsheim, RN; Beth Loots, MPH, MSW; Michael Pascual, BA (Seattle Children’s Hospital); Carla Greenbaum, MD (Benaroya Research Institute). Central Laboratory: (PI) Santica M. Marcovina, PhD, ScD (PI); Alan Chait, MD; Noemie Clouet-Foraison, PhD; Jessica Harting; Greg Strylewicz, PhD (University of Washington Northwest Lipid Metabolism and Diabetes Research Laboratories). Coordinating Center: (PI) Ralph D’Agostino, Jr, PhD; (PI) Elizabeth T. Jensen, MPH, PhD; Lynne E. Wagenknecht, DrPH; Ramon Casanova, PhD; Jasmin Divers, PhD; Maureen T. Goldstein, BA; Leora Henkin, MPH, MEd; Scott Isom, MS; Kristin Lenoir, MPH; June Pierce, AB; Beth Reboussin, PhD; Joseph Rigdon, PhD; Andrew Michael South, MD, MS; Jeanette Stafford, MS; Cynthia Suerken, MS; Brian Wells, MD, PhD; Carrie Williams, MA, CCRP (Wake Forest School of Medicine); Centers for Disease Control and Prevention: Giuseppina Imperatore, MD, PhD; Sharon H. Saydah, PhD; National Institute of Diabetes and Digestive and Kidney Diseases: Barbara Linder, MD.

Disclaimer: This study includes data provided by the Ohio Department of Health, which should not be considered an endorsement of this study or its conclusions. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention or the National Institute of Diabetes and Digestive and Kidney Diseases.

Meeting Presentation: A subset of these data were presented at the American Diabetes Association’s 80th Scientific Sessions; June 12-16, 2020.

Additional Contributions: The SEARCH for Diabetes in Youth Study is indebted to the many participants and their families and clinicians, whose participation made this study possible.