The prevalence of chronic kidney disease (CKD) in adults ≥18 years of age with type 1 diabetes in the U.S. was determined using National Health and Nutrition Examination Survey (NHANES) data.
A modified treatment-based algorithm applying a subset of NHANES diabetes questionnaires was used. The number of respondents with CKD and type 1 diabetes was weighted (extrapolated) to the U.S. population.
Based on data between 2015 and 2018, type 1 diabetes was identified in 47 out of 19,225 adults with evaluable kidney function data. CKD was present in 20 out of 47 people identified with type 1 diabetes. The weighted estimate of CKD in type 1 diabetes was 21.5%, corresponding to 258,196 (95% CI 71,189–445,203) people in the U.S.
Applying a conservative approach in our study indicates that CKD is common in adults with type 1 diabetes in the U.S.
Introduction
Chronic kidney disease (CKD) is a complication that has been historically ascribed to 30–40% of individuals with type 1 diabetes worldwide (1). The prevalence of individuals with CKD due to type 1 diabetes continues to increase globally, rising by 21.7% between 2007 and 2017, and was ∼3.2 million individuals in 2017 (2). These individuals are at increased risk of cardiovascular complications, and ∼25–30% will progress to kidney failure (3). Although kidney protective treatment such as angiotensin-converting enzyme inhibitors and angiotensin receptor blockers can reduce the risk of kidney failure, a significant disease burden remains (4,5).
The accurate surveillance of CKD and type 1 diabetes is especially important for understanding disease burden and pathogenesis (6,7). The identification of at-risk subgroups and potential risk factors can help promote appropriate health care delivery and choice of treatment course (6,8). However, direct estimates for the prevalence of CKD in people with type 1 diabetes are difficult because of challenges in differentiating and defining types of diabetes, as well as ascertainment methodologies (6). Surveillance using large administrative databases and health records may help improve the current understanding of diabetic kidney disease burden; however, accuracy can limit such strategies (6).
The National Health and Nutrition Examination Survey (NHANES) is a program of studies that combines interviews and physical examinations to assess the health and nutritional status of people in the U.S.; this database is representative of the U.S. population and is used for the extrapolation to the whole population (9). Using NHANES data from 2015 to 2018, this study sought to estimate the prevalence of CKD in adults with type 1 diabetes in the U.S.
Research Design and Methods
This cross-sectional study aimed to estimate the prevalence of CKD and type 1 diabetes in the U.S. using NHANES, a complex, multistage probability sample of the U.S. population that collects health and nutritional status information through interviews and examinations. The NHANES sampling weights account for oversampling of specific populations, noncoverage, and nonresponse; therefore, this weighting (extrapolation) may be used to provide national estimates representative of the overall U.S. population (9). SAS procedures PROC SURVEYFREQ and PROC SURVEYMEANS (10) were used to estimate weighted frequencies, percentages, standard errors, and 95% CIs. For the standard error estimation, the Taylor series linearization method, which accounts for the NHANES complex sampling design, was used. All weighted calculations were based on the multistage stratified design of the survey.
NHANES data cycles taken from 2015 to 2016 and 2017 to 2018 were used to identify adults ≥18 years of age with CKD and type 1 diabetes. This survey does not collect data on the specific subtypes of diabetes. Therefore, adults with CKD and type 1 diabetes were identified using a modified version of a published treatment-based algorithm: by applying a subset of diabetes questionnaires from NHANES (11) derived from recommendations made by the American Diabetes Association. The algorithm criteria require insulin treatment initiation within the first year of diabetes diagnosis and insulin as the only prescribed antihyperglycemic drug; the age limit at disease onset was not defined for the identification of adults with type 1 diabetes (11). The modified algorithm uses a threshold of an estimated glomerular filtration rate (eGFR) of ≤60 mL/min/1.73 m2 or a urinary albumin-to-creatinine ratio (UACR) of ≥30 mg/g for the identification of those with CKD (12). The CKD Epidemiology Collaboration equation was used to calculate eGFR (13). UACR was available as a laboratory value.
Results
A total of 19,225 adults were recorded in NHANES. Overall, 1,647 respondents were identified as having diabetes, of whom 54 were identified with type 1 diabetes (Fig. 2). Of the 54 respondents, 47 had evaluable eGFR and UACR information. In these adults with type 1 diabetes and evaluable eGFR and UACR, 20 were identified with CKD.
Weighting the number of identified respondents with type 1 diabetes in NHANES to the overall U.S. population corresponded to an estimate of 1,281,913 (95% CI 758,202–1,805,624) adults. In addition, CKD with type 1 diabetes were estimated to affect 258,196 (95% CI 71,189–445,203) adults in the U.S. (Fig. 2). The weighted U.S. prevalence of adults with CKD and type 1 diabetes corresponded to 21.5% (95% CI 10.4–25.8) (Fig. 3).
This study also evaluated baseline demographics and disease characteristics of the weighted population. The baseline demographics and disease characteristics were estimated based on the NHANES sample weights and estimation procedures (9). The median age at interview was 65 years (range: 63–76 years) among 258,196 adults with CKD and type 1 diabetes (Table 1). In addition, non-Hispanic White adults accounted for the highest proportion of respondents with CKD and type 1 diabetes, followed by non-Hispanic Black adults (60% and 18%, respectively) (Table 1). Across adults with CKD and type 1 diabetes, the mean recorded eGFR was 57 mL/min/1.73 m2, and the median UACR was 89 mg/g (Table 1).
Conclusions
CKD is highly prevalent in adults with type 1 diabetes and is associated with a risk of adverse cardiovascular and kidney outcomes (1). Although current standard of care generally consists of glycemic control and renin-angiotensin system blockers, a moderate risk of kidney failure is still reported in adults with type 1 diabetes (4,14). This therefore necessitates the development of novel kidney protective treatment strategies and increased surveillance of people with type 1 diabetes (1,15,16). A Study to Learn How Well the Study Treatment Finerenone Works and How Safe it is in People With Long-term Decrease in the Kidneys' Ability to Work Properly (Chronic Kidney Disease) Together With Type 1 Diabetes (FINE-ONE, NCT05901831), A Pilot Study of Fenofibrate to Prevent Kidney Function Loss in Type 1 Diabetes (PERL-FENO, NCT04929379), and Ambrisentan Sotagliflozin and Prevention of Renal Injury; a Randomized Evaluation (ASPIRE, NCT06072326) are trials investigating pharmacologic treatments in this setting (1,16–18).
Our study indicates that CKD was common in U.S. adults with type 1 diabetes. The algorithm identified a U.S. prevalence of CKD in type 1 diabetes of 21.5%, which has been similarly recorded in previously published European analyses (19,20). An observational study in 1,166 adults aged ≥18 years with type 1 diabetes identified CKD in 18% of participants in the Republic of Ireland (19). In addition, 8.4% of 25,762 adults aged 18–49 years old, and 22.1% of 5,121 adults aged 50–59 years were reported to have moderate to very high-stage CKD in the Swedish National Diabetes Register (20).
Race/ethnic group is associated with CKD incidence among adults with diabetes (21). Our study identified that a high proportion of adults with CKD and type 1 diabetes were non-Hispanic White followed by Black ethnicities. This is consistent with a published analysis that observed an increased risk of kidney function loss in adults of African Caribbean ethnicities with type 1 diabetes (22). Our results with respect to race/ethnicity should be treated with caution, as non-Hispanic Black individuals were oversampled in NHANES. Applying weights adjusted the distribution to reflect that of the U.S. population.
Despite the similarities identified with previously published analyses, our study is limited by the small data set used in the analysis and the exclusion of individuals <18 years of age. Respondents <18 years of age were excluded from our analysis because diabetes-related kidney disease is rare in children. In 2019, diabetes-related causes only accounted for 0.6% of cases in children with kidney disease in the U.S. (23). In the U.S., the incidence of type 1 diabetes in people <20 years of age is 22 per 100,000 (6). In addition to the small data set, eGFR and UACR values were not available for some respondents identified to have type 1 diabetes in the NHANES database. The results of the weighted population are also based on estimates and rely on certain assumptions. Furthermore, our study used a restrictive definition of type 1 diabetes; individuals using oral antihyperglycemic therapies were excluded from the analyses, which may have also led to an underestimate (11). Since the occurrence of CKD in people with type 1 diabetes is high, our results may underestimate the prevalence of CKD in adults with type 1 diabetes in the U.S. (1). This study provides a U.S. national estimate of CKD in type 1 diabetes, which has not been previously reported in the literature. Future studies enrolling larger cohorts should therefore help to provide accurate estimates on the national prevalence of CKD in type 1 diabetes in the U.S.
Article Information
Acknowledgments. The authors thank Ryan Fiano and Augustina Ogbonnaya from AmerisourceBergen (Conshohocken, PA) for the statistical analysis of the data; this assistance was funded by Bayer AG. Medical writing support was provided by Hussain Merchant, MSc, and editorial support was provided by Melissa Ward, BA, both of Orion (a division of Prime, London, U.K.), supported by Bayer AG according to Good Publication Practice guidelines (24).
Funding. This study was funded by Bayer AG.
Duality of Interest. P.R. reports personal fees from Bayer, research support and personal fees from AstraZeneca and Novo Nordisk, and personal fees from Astellas, Boehringer Ingelheim, Eli Lilly, MSD, Gilead, and Sanofi. All fees were given to Steno Diabetes Center Copenhagen. P.-H.G. reports receiving lecture honoraria from Astellas, AstraZeneca, Bayer, Boehringer Ingelheim, Eli Lilly, Elo Water, Medscape, MSD, Mundipharma, Nestlè, Novo Nordisk, PeerVoice, Sanofi, and SCIARC, and being an advisory board member of Astellas, AstraZeneca, Bayer, Boehringer Ingelheim, Eli Lilly, Medscape, MSD, Mundipharma, Nestlé, Novo Nordisk, PeerVoice, Sanofi, and SCIARC. R.S. is an employee of Bayer AG, USA, and/or may hold shares in the company. R.L. is an employee of Bayer AG, Germany. K.R.T. reports consultancy fees from AstraZeneca, Boehringer Ingelheim, Eli Lilly, Novo Nordisk, and Travere, grant support from Bayer and Travere, and speaker fees from AstraZeneca, Eli Lilly, and Novo Nordisk. No other potential conflicts of interest relevant to this article were reported.
Author Contributions. All authors contributed to the conception and design of the work, or the acquisition, analysis, or interpretation of data for the work. All authors drafted the work or revised it critically for important intellectual content. All authors provided final approval of the version to be submitted and published in the journal and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. R.L. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Prior Presentation. This study was presented as an abstract (A-2505) and poster (1460-P) at the 83rd Scientific Sessions of the American Diabetes Association, San Diego, CA, 23–26 June 2023.
Handling Editors. The journal editors responsible for overseeing the review of the manuscript were John B. Buse and Casey M. Rebholz.
