Appendix

Data Sources

• 2017–March 2020 National Health and Nutrition Examination Survey (NHANES), National Center for Health Statistics (NCHS), Centers for Disease Control and Prevention.
• 2018–2019 National Health Interview Survey (NHIS), National Center for Health Statistics, Centers for Disease Control and Prevention.
• Annual Estimates of the Resident Population by Sex, Single Year of Age, Race, and Hispanic Origin for the United States: April 1, 2010 to July 1, 2019, Population Division, US Census Bureau.

Methods

The total number of people with diabetes is the sum of the number of those aged 18 years or older with diagnosed or undiagnosed diabetes and the number of those younger than age 18 years with diagnosed diabetes. Undiagnosed diabetes for children and adolescents younger than age 18 years was not assessed due to insufficient sample size for reliable estimates. The 2017–March 2020 NHANES was used to calculate the percentage of adults aged 18 years or older with diagnosed and undiagnosed diabetes (see next section for detail). The 2018–2019 NHIS was used to calculate the percentage of children and adolescents younger than 18 years with diagnosed diabetes. These percentages were then applied to the corresponding July 1, 2019 US resident population estimates from the US Census Bureau to derive the total number of people with diabetes.

NHANES data collected from 2019 to March 2020 were combined with data from the 2017–2018 cycle to form a nationally representative sample of NHANES 2017–March 2020 pre-pandemic data. NCHS stopped all NHANES data collection for the 2019–2020 cycle after March 2020 due to the COVID-19 pandemic. Applying 2017–March 2020 NHANES estimates to the 2019 US resident population estimates has limitations. This methodology assumes that the prevalence of diabetes in 2019 was the same as it was in earlier years (2017–2018) and that the prevalence of diabetes in the resident population was identical to those in the civilian, noninstitutionalized population (from NHANES). Deviations from these assumptions may result in overestimated or underestimated numbers and rates.

Data Sources

• 2017–March 2020 National Health and Nutrition Examination Survey (NHANES), National Center for Health Statistics (NCHS), Centers for Disease Control and Prevention.
• Annual Estimates of the Resident Population by Sex, Single Year of Age, Race, and Hispanic Origin for the United States: April 1, 2010 to July 1, 2019, Population Division, US Census Bureau.

Methods

The percentage of adults aged 18 years or older with diabetes (diagnosed or undiagnosed) was estimated using 2017–March 2020 NHANES data. People who self-reported being told by a doctor or health professional that they had diabetes (other than during pregnancy) were classified as having diagnosed diabetes. Those not reporting a history of diagnosed diabetes but who had either a fasting plasma glucose greater than or equal to 126 mg/dl or an A1C level greater than or equal to 6.5% were classified as having undiagnosed diabetes. For consistency with earlier estimates, fasting glucose values were adjusted using backward regression equations provided by NCHS. People with missing values for either fasting glucose or A1C and pregnant women were excluded. People with diagnosed diabetes from the interviewed sample were combined with people with undiagnosed diabetes from the fasting plasma glucose subsample. Appropriate sampling weights were used so that estimates were representative of the total US adult population.

To estimate the number of adults with diagnosed and undiagnosed diabetes, the age-, sex-, race/ethnicity-specific percentages from three-way cross-tabulations were applied to the corresponding July 1, 2019 US resident population estimates from the US Census Bureau. These subgroup-specific numbers of adults were summed to obtain the estimated number of adults with diagnosed and undiagnosed diabetes for the following: total population; age groups 18–44, 45–64, and 65 years or older; men, women; non-Hispanic White, non-Hispanic Black, non-Hispanic Asian, and Hispanic adults. Age-adjusted percentages of diagnosed and undiagnosed diabetes were calculated among adults aged 18 years or older by sex, race/ethnicity, and education level by the direct method to the 2000 US Census standard population, using age groups 18–44, 45–64, and 65 years or older.

References

National Health and Nutrition Examination Survey. 2005–2006 Data Documentation, Codebook, and Frequencies https://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/GLU_D.htm.

National Health and Nutrition Examination Survey. 2007–2008 Data Documentation, Codebook, and Frequencies https://wwwn.cdc.gov/Nchs/Nhanes/2007-2008/GLU_E.htm.

National Health and Nutrition Examination Survey. 2017-March 2020 Data Documentation, Codebook, and Frequencies https://wwwn.cdc.gov/Nchs/Nhanes/2017-2018/P_GLU.htm.

Data Source

• 2001–March 2020 National Health and Nutrition Examination Surveys (NHANES), National Center for Health Statistics, Centers for Disease Control and Prevention.

Methods

Percentages of diagnosed, undiagnosed, and total diabetes using overlapping 4-year survey periods during 2001–2020 were calculated among adults aged 18 years or older and age-adjusted by the direct method to the 2000 US Census standard population, using age groups 18–44, 45–64, and 65 years or older. Joinpoint regression was used to analyze varying trends in non-overlapping 2-year estimates. This analysis used the age-adjusted estimates and permutation tests to identify a maximum of three points where linear trends in prevalence changed significantly in either direction or magnitude.

Reference

National Cancer Institute. Joinpoint Trend Analysis Software https://surveillance.cancer.gov/joinpoint/

Data Sources

• 2017–March 2020 National Health and Nutrition Examination Survey (NHANES), National Center for Health Statistics, Centers for Disease Control and Prevention.
• 2018–2019 National Health Interview Survey (NHIS), National Center for Health Statistics, Centers for Disease Control and Prevention.
• Annual Estimates of the Resident Population by Sex, Single Year of Age, Race, and Hispanic Origin for the United States: April 1, 2010 to July 1, 2019, Population Division, US Census Bureau.

Methods

The percentage of people with diagnosed diabetes was obtained from 2018–2019 NHIS data and 2017–March 2020 NHANES data. The percentage of people aged <20 years with diagnosed diabetes was obtained from NHIS based on information reported by a knowledgeable adult family member residing in the household. The percentage of people aged 20 years or older with diagnosed diabetes was obtained from self-reported data in the 2017–March 2020 NHANES. The combined estimate of diagnosed diabetes for all ages was applied to the July 1, 2019 US resident population from the US Census Bureau to derive the number of people with diagnosed diabetes for all ages and for children and adolescents younger than age 20 years.

Validated methods to distinguish between types of diabetes in surveys are not available. The percentage of adults aged 20 years or older with diagnosed diabetes who self-reported type 1 diabetes plus current insulin use and the percentage of adults aged 20 years or older with diagnosed diabetes who started using insulin within a year of their diagnosis were estimated from 2019 NHIS data. To estimate the number of adults aged 20 years or older with type 1 diabetes, these percentages were then applied to the derived number of adults aged 20 years or older with diagnosed diabetes. To estimate the number of youths with type 1 diabetes, the percentage of youth aged less than 20 years with type 1 diabetes (86.3%) calculated using prevalence data from the SEARCH for Diabetes in Youth Study was applied to the derived number of youth aged less than 20 years with diagnosed diabetes.

Reference

Lawrence JM, Divers J, Ison S, Saydah S, Imperatore G, Pihoker C, Marcovina SM, Mayer-Davis EJ, Hamman RF, Dolan L, Dabelea D, Pettitt DJ, Liese AD, SEARCH for Diabetes in Youth Study Group. Trends in Prevalence of Type 1 and Type 2 Diabetes in Children and Adolescents in the US, 2001-2017. JAMA. 2021 Aug 24;326(8):717-727.

Data Sources

• 2018–2019 National Health Interview Survey (NHIS), National Center for Health Statistics (NCHS), Centers for Disease Control and Prevention.
• National Data Warehouse (NDW), Indian Health Service (IHS).

Methods

With the exception of American Indian/Alaska Native (AI/AN) people, who are not well-represented in national surveys because of small population size, diagnosed diabetes by race-ethnicity, education level, and income, overall and by sex, were calculated using 2018–2019 NHIS self-reported data. Two years of data were averaged to provide more statistically reliable estimates. Adults aged 18 years or older who self-reported being told by a doctor or health professional that they had diabetes were classified as having diagnosed diabetes.

Prevalence of diagnosed diabetes among AI/AN people was calculated using fiscal year 2019 data from the IHS NDW. This data system includes patient registration and encounter data that are received from IHS facilities, tribally operated programs, and urban and contract health systems. These health care facilities serve about 2.6 million AI/AN people who belong to 574 federally recognized tribes in 37 states. Data for active patients (i.e., those with at least one encounter during the preceding 3 years) aged 18 years or older were used to calculate these estimates. Diabetes cases among these patients were identified based on encounter data using International Classification of Diseases and Related Health Problems, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis codes starting with 250 and ICD-10-CM (Tenth Revision) diagnosis codes starting with E10, E11, or E13. Patients were considered to have diagnosed diabetes if they had at least two encounters with one of these diagnosis codes reported during fiscal year 2019. Comparison of diabetes prevalence estimates calculated from NHIS and IHS NDW data should be interpreted with caution due to differences in the data sources and methods used to define diabetes.

Education level was classified in 3 categories: less than high school, high school graduate, or more than high school. Income level was based on the ratio of family income to the federal poverty level and classified in 4 categories: <100% federal poverty level (i.e., below poverty), 100–299% federal poverty level, 300–499% federal poverty level, and ≥500% federal poverty level. To handle missing income data in NHIS, NCHS provides files of multiply imputed income values. Five imputation datasets were used to calculate prevalence estimates by income level.

Percentages for all subgroups were age-adjusted, using age groups 18−44, 45−64, and 65 years or older, by the direct method to the 2000 US Census standard population.

Reference

National Center for Health Statistics. Multiple Imputation of Family Income and Personal Earnings in the National Health Interview Survey: Methods and Examples. 2019. https://www.cdc.gov/nchs/data/nhis/tecdoc18.pdf [PDF – 541KB].

Data Source

• 2017–2018 National Health Interview Survey (NHIS), National Center for Health Statistics (NCHS), Centers for Disease Control and Prevention.

Methods

Diagnosed diabetes among Hispanic and Asian subgroups were calculated using 2017–2018 NHIS self-reported data. Beginning in 2019 NHIS, data used to identify survey respondents in these subgroups were no longer publicly available. Two years of data were averaged to provide more statistically reliable estimates. Adults aged 18 years or older who self-reported being told by a doctor or health professional that they had diabetes were classified as having diagnosed diabetes. Estimates of diagnosed diabetes for Native Hawaiians and Other Pacific Islanders were not included because of small sample size.

Percentages for all subgroups were age-adjusted, using age groups 18−44, 45−64, and 65 years or older, by the direct method to the 2000 US Census standard population.

Data Sources

• 2003–2019 Behavioral Risk Factor Surveillance System (BRFSS), National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention.
• Annual Estimates of the Resident Population for Selected Age Groups by Sex for the United States, States, Counties, and Puerto Rico Commonwealth and Municipios: April 1, 2010 to July 1, 2019, Population Division, US Census Bureau.
• United States Diabetes Surveillance System (USDSS), National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention.

Methods

Year-specific, county-level estimates and maps of prevalence and incidence of diagnosed diabetes are available from the USDSS website (https://gis.cdc.gov/grasp/diabetes/DiabetesAtlas.html). Data from the BRFSS and the US Census Bureau’s Population Estimates Program were used to estimate county-level prevalence and incidence of diagnosed diabetes among adults aged 20 years or older. Three years of data were used to improve the precision of the year-specific incidence estimates. For example, 2004 estimates used BRFSS survey data for 2003, 2004, and 2005. Two years of data were used for 2019 estimates, as 2020 data were not available at the time of analysis. County-level estimates for over 3,100 counties or county equivalents (e.g., parish, borough, municipality) in the 50 US states, Puerto Rico, and the District of Columbia were based on indirect model-dependent estimates using Bayesian multilevel modeling techniques. This model-dependent approach uses a statistical model that “borrows strength” in making an estimate for one county from BRFSS data collected in other counties. For incidence, multilevel binomial regression models with random effects of demographic variables (age groups 20–44, 45–64, and 65 years or older; race/ethnicity; and sex) at the county level were developed. County-level prevalence was based on design-assisted model-based estimates using the power prior log-weights (PLOW) technique developed by Xie et al. Estimates were age-adjusted to the 2000 US Census standard population using age groups 20–44, 45–64, and 65 years or older.

References

Rao JNK. Small Area Estimation. Hoboken, New Jersey: John Wiley & Sons, Inc.; 2003.

Xie H, Barker LE, Rolka DB. Incorporating design weights and historical data into model-based small-area estimation. J Data Sci. 2020;18(1):115–131.

Barker LE, Thompson TJ, Kirtland KA, Boyle JP, Geiss LS, McCauley MM, Albright AL. Bayesian small area estimates of diabetes incidence by United States county, 2009. J Data Sci. 2013;11:249–269.

Data Sources

• 2018–2019 National Health Interview Survey (NHIS), National Center for Health Statistics, Centers for Disease Control and Prevention.
• 2017–March 2020 National Health and Nutrition Examination Survey (NHANES), National Center for Health Statistics, Centers for Disease Control and Prevention.
• Annual Estimates of the Resident Population by Sex, Single Year of Age, Race, and Hispanic Origin for the United States: April 1, 2010 to July 1, 2019, Population Division, US Census Bureau.

Methods

The rate of new cases of diabetes was calculated using 2019 NHIS data on respondents’ age at diagnosis and age at interview. Two-year averages of 2018–2019 NHIS were used to improve the precision of race/ ethnicity- and education level-specific estimates. Adults who reported being diagnosed with diabetes were asked at what age they were diagnosed. The number of years since diagnosis was calculated by subtracting the person’s age at diagnosis from the person’s current age. Adults who had a value of zero were identified as having been diagnosed with diabetes within the last year. In addition, half of the adults who had a value of one were classified as having been diagnosed within the last year. To calculate the rate, the numerator included the number of adults who were diagnosed with diabetes within the last year. The denominator was the estimate of the adult population, excluding those who had been diagnosed for more than 1 year and those who were categorized on the NHIS as “refused” or “don’t know” or who had missing values on the diabetes status question.

To estimate the number of new cases of diabetes for adults in 2019, the age-, sex-, race/ethnicity-specific rates of new cases from three-way cross-tabulations of 2018–2019 NHIS data were applied to the corresponding July 1, 2019 US resident population estimates from the US Census Bureau after excluding the number of adults who had been diagnosed with diabetes for more than 1 year, estimated from NHANES. These subgroup-specific numbers of adults were summed to obtain the estimated number of adults with newly diagnosed diabetes for the following groups: total population; age groups 18–44, 45–64, and 65 years or older; men, women; non-Hispanic White, non-Hispanic Black, non-Hispanic Asian, and Hispanic adults. Age-adjusted incidence of diagnosed diabetes was calculated among adults aged 18 years or older by sex, race/ethnicity and education level by the direct method to the 2000 US Census standard population, using age groups 18–44, 45–64, and 65 years or older.

Data Source

• 2000–2019 National Health Interview Survey (NHIS), National Center for Health Statistics, Centers for Disease Control and Prevention.

Methods

The same method used to calculate 2019 incidence was also applied to 2000–2019 NHIS data. For every year from 2000 to 2019, age-adjusted incidence of diagnosed diabetes was calculated among adults aged 18 years or older by the direct method to the 2000 US Census standard population, using age groups 18–44, 45–64, and 65 years or older. Joinpoint regression was used to analyze varying trends in annual age-adjusted incidence estimates. This analysis used the age-adjusted estimates and permutation tests to identify a maximum of three points where linear trends in incidence changed significantly in either direction or magnitude.

Reference

National Cancer Institute. Joinpoint Trend Analysis Software https://surveillance.cancer.gov/joinpoint/

Data Source

• 2002–2015 SEARCH for Diabetes in Youth Study, SEARCH Study Group.

Methods

We reported data from the SEARCH for Diabetes in Youth Study, a population-based registry of incident diabetes among youth less than age 20 years from five clinical sites in the United States. Diabetes type was based on physician diagnosis. Estimates of incidence for type 1 diabetes were included for all patients less than 20 years. Estimates for incidence of type 2 diabetes were only included for youth age 10 to 19 years since there are too few cases among patients less than 10 years for reliable estimates. Patients with all other types of diabetes were excluded. Race/ethnicity was based on self-report, from medical records, or from geocoding for youth with missing data. Annual denominators included civilian youths who were younger than 20 years of age on December 31 of the index year and who were civilian residents of the geographic study areas and summed across all five centers. Annual incidence rates, by type, were calculated as the number of valid, registered patients divided by the number of persons in the surveillance networks over the same period across the five centers. Rates for each index year were presented as two-year moving averages and were expressed per 100,000 youths, overall, and according to race/ethnicity for 2003 (2002–2003) through 2015 (2014–2015). The 95% confidence intervals for the annual unadjusted rates were calculated with the use of the skew-corrected inverted-score test, assuming a binomial distribution. To estimate the number of youths in the United States with type 1 or type 2 diabetes, the incidence rates from the SEARCH study were applied to the total US population for the four racial and ethnic groups for the years of interest. Data for American Indian (AI) youth who participated in the SEARCH study were not shown, since rates cannot be generalized to all AI youth in the United States. Trends in incidence were modeled separately for type 1 and type 2 diabetes with the number of diagnosed cases in each year as the outcome, the corresponding denominator as an offset, and the incidence year as the main predictor. The models were assumed to have a negative binomial distribution with a logarithmic link and used a generalized autoregressive moving average to account for serial correlation. To determine whether incidence trends were constant over the 2002–2015 period, a knot in index year 2011 was selected based on the minimum Akaike information criteria to allow for comparison of incidence trends between the 2002–2010 and 2011–2015 periods.

References

Hamman RF et al.; SEARCH for Diabetes in Youth Study Group. The SEARCH for Diabetes in Youth study: rationale, findings, and future directions. Diabetes Care. 2014 Dec;37(12):3336–44.

Divers J, Mayer-Davis EJ, Lawrence JM, et al. Trends in Incidence of Type 1 and Type 2 Diabetes Among Youths— Selected Counties and Indian Reservations, United States, 2002–2015. MMWR Morb Mortal Wkly Rep. 2020 Feb 14;69(6):161–165.

Benjamin MA, Rigby RA, Stasinopoulos DM. Generalized autoregressive moving average models. J Am Stat Assoc. 2003 Mar;98(461):214–23.

Data Sources

• 2017–March 2020 National Health and Nutrition Examination Survey (NHANES), National Center for Health Statistics (NCHS), Centers for Disease Control and Prevention.
• Annual Estimates of the Resident Population by Single Year of Age and Sex for the United States: April 1, 2010 to July 1, 2019, Population Division, US Census Bureau.

The percentage of adults aged 18 years or older with prediabetes was estimated using 2017–March 2020 NHANES data. People without diabetes were classified as having prediabetes if they had fasting plasma glucose values of 100 to 125 mg/dL or A1C values of 5.7% to 6.4%. In addition to this definition, prediabetes estimates according to other definitions were calculated, using 2-hour fasting glucose from an oral glucose tolerance test or more stringent glycemic criteria.

For consistency with earlier estimates, fasting glucose values were adjusted using backward regression equations provided by NCHS. People with missing values for either fasting glucose or A1C and pregnant women were excluded. To estimate the number of adults with prediabetes, the age-, sex-, race/ethnicity specific percentages from three-way cross-tabulations were applied to the corresponding July 1, 2019 US resident population estimates from the US Census Bureau. These subgroup-specific numbers of adults were summed to obtain the estimated number of adults with prediabetes for the following: total population; age groups 18–44, 45–64, and 65 years or older; men, women; non-Hispanic White, non-Hispanic Black, non-Hispanic Asian, and Hispanic adults.

Among those who tested positive for prediabetes, awareness was defined as (1) answered “yes” to the question, “Have you ever been told by a doctor or other health professional that you have any of the following: prediabetes, impaired fasting glucose, impaired glucose tolerance, borderline diabetes or that your blood sugar is higher than normal but not high enough to be called diabetes or sugar diabetes?” or (2) reported having prediabetes or borderline diabetes when asked whether they had diabetes.

Age-adjusted prevalence of prediabetes was calculated among adults aged 18 years or older by sex, race/ethnicity and education level by the direct method to the 2000 US Census standard population, using age groups 18–44, 45–64, and 65 years or older. Joinpoint regression was used to analyze varying trends in annual age-adjusted prevalence estimates. This analysis used the age-adjusted estimates and permutation tests to identify a maximum of three points where linear trends in prediabetes prevalence changed significantly in either direction or magnitude.

References

American Diabetes Association. Classification and diagnosis of diabetes. Diabetes Care. 2019 Jan; 42 (Supplement 1): S13-S28.

National Health and Nutrition Examination Survey. 2015-2016 Data Documentation, Codebook, and Frequencies https://wwwn.cdc.gov/Nchs/Nhanes/2015-2016/GLU_I.htm.

National Cancer Institute. Joinpoint Trend Analysis Software https://surveillance.cancer.gov/joinpoint/.

Data Source

• 2015–2018 National Health and Nutrition Examination Survey (NHANES), National Center for Health Statistics, Centers for Disease Control and Prevention.

Methods

The percentages of adults aged 18 years or older with diagnosed diabetes who had selected risk factors were estimated using the following definitions:

Smoking

Current cigarette smoking status was based on self-report. Tobacco use was based on self-reported current cigarette smoking or serum cotinine level >10 ng/mL. Former cigarette smoker was based on both 1) no current tobacco use and 2) a history of smoking at least 100 cigarettes in a lifetime.

Overweight and Obesity

Overweight and obesity were classified according to body mass index of 25.0–29.9 kg/m2 (overweight), 30.0–39.9 kg/m2 (obesity), or 40.0 kg/m2 or higher (extreme obesity) calculated from measured values of height and weight.

Physical Inactivity

Physical inactivity was based on self-report of less than 10 minutes per week of moderate or vigorous activity in each of the physical activity categories of work, leisure time, and transportation.

High Blood Pressure

High blood pressure was based on average measured systolic blood pressure of 140 mmHg or higher or the average diastolic blood pressure of 90 mmHg or higher or self-reported current use of prescription medication for high blood pressure.

High Cholesterol

High cholesterol was calculated as measured total cholesterol level minus HDL cholesterol level. A non-HDL cholesterol value of 130 mg/dL is roughly equivalent to an LDL level of 100 mg/dL, which indicates an increased risk of diabetes-related complications and eligibility for statin therapy.

A1C

A1C was classified based on measured glycated hemoglobin, with values higher than 9% indicating poor glycemic control.

Data Source

• 2015–2018 National Health and Nutrition Examination Survey (NHANES), National Center for Health Statistics, Centers for Disease Control and Prevention.

Methods

The percentages of adults aged 18 years or older with diagnosed diabetes who had selected protective factors were estimated using the following definitions:

Usual Source for Diabetes Care

Self-report of having at least one usual source of diabetes care, such as a doctor or other health care professional. Other health professionals did not include specialists, such as diabetes educators, dieticians, eye doctors, or foot doctors.

Physical Activity

Meeting the recommended physical activity goal was based on having at least 150 minutes per week of leisure-time physical activity.

Weight Management

Self-report of managing or losing weight to lower risk for developing certain diseases.

Statin Treatment

Statin therapy among adults aged 40–75 years with diagnosed diabetes was based on prescription information from a medication inventory.

A1C, Blood Pressure, Cholesterol, and Smoking (ABCs)

The ABCs definition was based on meeting all of the following criteria: A1C value <7.0%, blood pressure <140/90 mmHg, non-HDL cholesterol <130 mg/dL, and being a nonsmoker. A less stringent ABCs definition was based on meeting all of the following criteria: A1C value <8.0%, blood pressure <140/90 mmHg, non-HDL cholesterol <160 mg/dL, and being a nonsmoker.

References

American Diabetes Association. Standards of Medical Care in Diabetes—2021. Diabetes Care. 2021 Jan 1; 44 (Supplement 1).

AACE/ACE Guidelines for the Management of Dyslipidemia and Prevention of Cardiovascular Disease Writing Committee, Endocr Pract. 2017;23(Suppl 2).

Data Sources

• 2018 National Inpatient Sample (NIS), Agency for Healthcare Research and Quality.
• 2018 Nationwide Emergency Department Sample (NEDS), Agency for Healthcare Research and Quality.
• 2018 National Health Interview Survey (NHIS), National Center for Health Statistics, Centers for Disease Control and Prevention.
• 2015–2018 National Health and Nutrition Examination Survey (NHANES), National Center for Health Statistics, Centers for Disease Control and Prevention.
• 2020 United States Renal Data System (USRDS) Annual Report.
• 2019 Behavioral Risk Factor Surveillance System (BRFSS), National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention.

Methods

The number of emergency department (ED) visits for hypoglycemia (blood glucose <70 mg/dL) and hyperglycemic crisis in 2018 were calculated using NEDS. Hyperglycemic crisis includes diabetic ketoacidosis (DKA) and hyperglycemic hyperosmolar syndrome (HHS). DKA is characterized by hyperglycemia (blood glucose >250 mg/dL), metabolic acidosis, and increased blood ketone concentration. HHS is characterized by severe hyperglycemia (blood glucose >600 mg/dL), hyperosmolarity, and dehydration, without ketoacidosis. Discharge disposition was classified as the proportion of ED visits that were: treated and released; admitted to a hospital; transferred to another hospital; transferred to skilled nursing facility, intermediate care facility, or home with home healthcare; left against medical advice; died; unknown disposition but not admitted to a hospital. The number of hospitalizations for major cardiovascular diseases, lower-extremity amputation, hyperglycemic crisis and hypoglycemia in 2018 were calculated using NIS. Crude rates were calculated using the proportion of the population with diabetes from NHIS.

Prevalence of chronic kidney disease (CKD) stages 1–4 among US adults aged 18 years or older with diagnosed diabetes was calculated using laboratory data from the 2017–March 2020 NHANES and the 2009 and 2021 CKD Epidemiology Collaboration (CKD-EPI) equations for estimated glomerular filtration rate (eGFR). The 2009 CKD-EPI eGFR equation was based on serum creatinine, age, sex, and Black race; the updated 2021 equation was based on based on serum creatinine, age, and sex only. People with missing values for serum creatinine and pregnant women were excluded. CKD awareness was based on self-report among adults with eGFR levels 15–<60 mL/min/1.73 m² (i.e., moderate to severe CKD stages 3–4). Data on the numbers of people with end-stage kidney disease by primary cause (diabetes, high blood pressure, and glomerulonephritis) were obtained from the USRDS Annual Report Reference Tables. Prevalence of vision disability, defined as severe vision difficulty or blindness, was calculated using self-reported data from the 2019 BRFSS.

References

National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis. 2002 Feb;39(2 Suppl 1):S1–266.

Inker LA, Eneanya ND, Coresh J, Tighiouart H et al.; Chronic Kidney Disease Epidemiology Collaboration. New creatinine- and cystatin C-based equations to estimate GFR without race. N Engl J Med. 2021 Sep 23. doi: 10.1056/NEJMoa2102953. Online ahead of print.

Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, Kusek JW, Eggers P, Van Lente F, Greene T, Coresh J; CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration). A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009 May 5;150(9):604–12.

Johansen KL, Chertow GM, Foley RN, et al. US Renal Data System 2020 Annual Data Report: Epidemiology of Kidney Disease in the United States. Am J Kidney Dis. 2021;77(4)(suppl 1):Svii-Sviii, S1-S597.