Next Article in Journal
Diabetes Differentially Alters Glial Cells in Different Brain Regions
Previous Article in Journal
Exploring the Role of AbaComplex in Managing Dysglycemia: Insights from a Randomized, Three-Arm, Placebo-Controlled Trial
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Racial and Ethnic Disparities in Type 2 Diabetes Complications and In-Hospital Mortality in the United States: A Retrospective Cohort Study

by
Lainy A. Burress
1 and
John M. Clements
2,*
1
Department of Family Medicine, Michigan State University College of Human Medicine, B103 Clinical Center 788 Service Road, East Lansing, MI 48824, USA
2
Charles Stewart Mott Department of Public Health, Michigan State University College of Human Medicine, 130 East 2nd Street, Suite 202, Flint, MI 48502, USA
*
Author to whom correspondence should be addressed.
Diabetology 2025, 6(3), 15; https://doi.org/10.3390/diabetology6030015
Submission received: 10 December 2024 / Revised: 31 January 2025 / Accepted: 24 February 2025 / Published: 27 February 2025

Abstract

:
Objectives: To describe the association of race with type 2 diabetes complications and determine if differences in rates of complications exist between racial/ethnic groups of adult type 2 diabetes patients in the United States. Additionally, we model the odds of in-hospital patient mortality across racial/ethnic groups. Methods: A retrospective cohort study was conducted using data from the 2018 National Inpatient Sample of Healthcare Cost and Utilization Project, including 97,314 unweighted and 486,500 weighted adults with type 2 diabetes. Chi-square analysis was used to determine the association of race with diabetes complications, along with z-tests to determine the differences in complication rates of 11 different complications between racial/ethnic groups and binary logistic regression to model in-hospital mortality. Results: Our analysis revealed significant racial/ethnic disparities in both complication rates and odds of in-hospital mortality. Whites had the lowest rate of complications overall, except for arthropathy/oral complications (18.8%) and foot/skin ulcers (18.2%), while Black/African Americans had the highest rates of hyperosmolarity (7.3%), ketoacidosis (21.2%), neurological complications (8.9%), and hyperglycemia (13.4%). Asian/Pacific Islanders had the highest rates of hypoglycemia (17.6%) as well as kidney (7.2%) and ophthalmic (0.3%) complications, and Hispanics the highest rates of circulatory complications (19.0%). Hispanic ethnicity was associated with 10.6% reduced odds of in-hospital mortality, and Asian/Pacific Islanders and Other races had increased odds of mortality by 25.2% and 27.0%, respectively. Notably, neurological (OR = 0.278, 95% CI: 0.111, 0.702) complications and hyperglycemia (OR = 0.304, 95% CI: 0.124, 0.749) were associated with a reduction in mortality odds by 62.2% and 69.6%, possibly reflecting the study’s focus on in-hospital rather than all-cause or 30-day mortality. Conclusions: We demonstrated disparities in both rates of type 2 diabetes complications and odds of mortality between different racial/ethnic groups. These results lay groundwork for future research into the root causes of these disparities and highlight the importance of targeting interventions and equitable case for those most at risk.

1. Introduction

Approximately 38 million people in the United States have diabetes, most commonly type 2 (95%), with 7.6 million people having undiagnosed diabetes [1]. As of 2022, diabetes was the eighth leading cause of death in the United States [1]. The prevalence of type 2 diabetes is highest in adults over 65 and disproportionately affects minorities [2,3]. Members of Black and Hispanic race/ethnic groups also face a disproportionate prevalence of diabetic health complications and increased odds of diabetes-related mortality compared to Whites [3,4,5,6]. Data from the National Inpatient Sample (NIS) of the Healthcare Cost and Utilization Project (HCUP) includes several important complications for diagnostic coding and billing including kidney, ophthalmic, neurological, oral, and circulatory complications as well as hyperosmolarity, ketoacidosis, arthropathy, skin and foot issues, and hypo- and hyperglycemia [7]. This study focuses on these specific type 2 diabetes complications, particularly highlighting the disproportionate burden experienced by different racial/ethnic groups.
A few studies describe the associations between race and diabetes complications. For instance, Black and Hispanic adults with diabetes tend to have a higher burden of microvascular complications including nephropathy/albuminuria and retinopathy, as well as macrovascular complications such as cardiovascular and cerebrovascular disease [4]. While diabetes rates declined from 1990 to 2010, Blacks were still more likely to suffer from hyperglycemic death, amputation, end-stage renal disease (ESRD), and stroke than Whites [8]. Additionally, Blacks, Hispanics, and Native Americans with diabetic foot infections had an increased risk of major amputations (1.4, 1.3, and 1.5 odds ratios, respectively) in comparison to Whites, and similar outcomes were seen for hospital stays with Blacks and Hispanics admitted for diabetic foot infections having longer hospital stays (9.2 and 8.6 days, respectively) compared to Whites (8.1 days) [9]. Blacks and Hispanics have also been more commonly undertreated, receiving lower-quality treatment for foot infections [9]. These populations also have higher risk and incidence rates of ESRD compared to Whites [6,10], despite Blacks and Hispanics having lower cardiovascular events and equal access to care [10]. Cardiovascular mortality declined in adults with self-reported diabetes from 1988 to 2015; however, disparities by race and ethnicity continued to exist [3].
Blacks diagnosed with diabetes also have increased odds of hypertension compared to Whites, and these disparities exist for Blacks with diabetes regardless of their income and education level [11]. Race is also associated with poor quality of care for diabetic patients [7]. Other studies focusing on socioeconomic status, social determinants of health (SDOH), and diabetes explain how diabetic health outcomes can be multifactorial and influenced by many variables including income, education, and location [12]. As is common in diabetes literature, these studies combine type 1 and type 2 diabetes outcomes rather than evaluating type 2 diabetes independently. Without sufficient research into disparities in rates of type 2 diabetic complications, minority populations are at a higher risk of developing complications and receiving undertreatment, as well as having higher odds of mortality related to the condition [13].
This study uses data from the National Inpatient Sample (NIS) of the Healthcare Cost and Utilization Project (HCUP) [7] to examine differences in the prevalence of complications including kidney, ophthalmic, neurological, oral, circulatory, hyperosmolarity, ketoacidosis, arthropathy, skin and foot issues, and hypo- and hyperglycemia. We also explore the association of race/ethnicity and complications on in-hospital mortality for a cohort of adults with type 2 diabetes while controlling for age, sex, insurance status, patient location (rural), and income. The aims of this research are to determine if race/ethnicity is associated with these complications as well as the differences in rates of complications between these groups. Further, this study determines if race/ethnicity groups and diabetes-related complications increase the odds of in-hospital mortality. In this way, this research helps close the gap in research on disparities in the context of type 2 diabetes complications.

2. Materials and Methods

This study is an IRB-exempt, retrospective cohort study using discharge data from the 2018 National Inpatient Sample (NIS), Healthcare Cost and Utilization Project (HCUP), and Agency for Healthcare Research and Quality [7]. Created to support the HCUP, the NIS represents all short-term, non-Federal hospitals in the United States [7]. Data from the NIS cover more than 97% of the national population, including over seven million hospital stays [7]. Patient information is collected regardless of payer, making the NIS the largest database of this type in the United States [7]. The 2018 HCUP database includes 7,105,498 discharge cases collected from 1 January 2018 to 31 December 2018. Discharge records include up to 40 International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10) diagnosis codes [14]. The first diagnosis code is generally considered the primary diagnosis for that hospital stay. For this study, the primary diagnosis code was used to create groups based on diabetes complications. Discharge cases for patients with a primary diagnosis of type 2 diabetes (ICD-10: E11.XXX) were included for a total of 97,975 discharges. Because E11 is the broad diagnosis code for type 2 diabetes and ICD-10 guidelines indicate that subcodes should be used instead, five cases with a diagnosis code of E11 and no subcodes were excluded. In addition, 660 discharges with an age less than 18 were excluded to examine solely the adult population. After these exclusions, the final unweighted sample size is 97,310. The use of NIS data requires a weighting procedure to make reliable national estimates [15]. The final sample size for this study is 486,550, reflecting the use of weighting procedures.
To determine the association of race with diabetes complications, as well as differences in rates of complications between groups based on race/ethnicity, the primary independent variable is “Type of Diabetes Complication”. This categorical variable was created by selecting cases with a primary diagnosis code of E11.XXX, which classified the main type 2 diabetes code (E11.XXX) into 11 categories coded as: 0 = type 2 diabetes w/out complications (E11.9), used as the reference for modeling mortality, 1 = type 2 diabetes w/hyperosmolarity (E11.0X), 2 = type 2 diabetes w/ketoacidosis (E11.1X), 3 = type 2 diabetes w/kidney complications (E11.2X), 4 = type 2 diabetes w/ophthalmic complications (E11.3X), 5 = type 2 diabetes w/neurological complications (E11.4X), 6 = type 2 diabetes w/circulatory complications (E11.5X), 7 = type 2 diabetes w/other complications, diabetic arthropathy, oral complications (E11.61X, 11.638, 11.69, 11.8), 8 = type 2 diabetes w/skin complications—foot/skin ulcers (E11.62X), 9 = type 2 diabetes w/hypoglycemia (E11.64X), and 10 = type 2 diabetes w/hyperglycemia (E11.65X).
The main dependent variable to test for odds of mortality in this study is in-hospital mortality (alive on discharge = 0 or died in the hospital = 1). Race, along with Type of Diabetes Complications, are the primary independent variables to model in-hospital mortality. Race is coded as White = 1 (reference group), Black = 2, Hispanic = 3, Asian/Pacific Islander = 4, Native American = 5, and Other = 6. Other independent variables collected for analyses include the following: patient age (in years at admission), sex (male = 0—reference group, or female = 1), insurance status (Medicare = 1, Medicaid = 2, private insurance = 3—reference group), patient location, based on Ingram and Franco [16] (metropolitan = 0—reference group, and non-metropolitan (or rural) = 1), income (median income for zip code categorized into quartiles from lowest to highest: USD < 45,999 = 1—reference group, USD 46,000–58,999 = 2, USD 59,000–78,999 = 3, and USD > 79,000 = 4).
SPSS version 27 [17] was used for statistical analyses, beginning with descriptive statistics for each variable. A chi-square analysis was conducted to determine the overall association between race and different type 2 diabetes complications. Differences in the rates of complications were determined using a z-test for differences in proportions. Finally, in-hospital mortality was modeled using a binary logistic regression analysis to determine the association of race, complications, and other independent variables on the odds of mortality. Statistical significance was determined using an alpha value of 0.05 in combination with confidence intervals for odds ratios where appropriate.

3. Results

Table 1 includes descriptive statistics for the sample, reflecting the use of weighting procedures that allow national estimates for all results. The mean age is 60.28, with the majority being White (53.8%) and male (59.1%). The most common complications are ketoacidosis (17.2%), circulatory (16.5%), arthropathy/oral (16.2%), and skin and foot complications (15.3%). Approximately 0.7% (n = 3600) of discharges experienced in-hospital mortality.

3.1. Complications and Race

Table 2 shows a crosstabulation of race and diabetes complications with differences in the proportions of each complication between each race/ethnicity category. Figure 1 also provides these results. First, there is a significant association between race and type 2 diabetes complications (chi-square = 16,617, df = 50, p < 0.001), indicating that race and complications are dependently related. Z-tests for differences in proportions indicate that for every type 2 diabetes complication, significant differences exist between the rate of each complication for differing race categories. For each complication category, significant differences in proportions of each complication between race/ethnic groups are designated by different alphabetical superscripts within each complication subcategory. Insignificant differences in proportions are designated using the same alphabetic letter superscript.
There are several noticeable differences in complication rates by race. For instance, Whites had among the lowest complication rates compared to all other races for each complication except arthropathy and oral complications (18.8%, second only to Native Americans at 19.2%) and foot/skin ulcers (18.2%). Conversely, Black/African Americans had the highest rates of many complications including hyperosmolarity (7.3%), ketoacidosis (21.2%), neurological complications (8.9%), and hyperglycemia (13.4%), but had the lowest rate of circulatory complications (14.8%). Asian/Pacific Islanders had the highest rates of kidney complications (7.2%), ophthalmic complications (0.3%), and hypoglycemia (17.6%), aside from having the lowest rate of skin and foot complications (8.7%). Hispanics had the highest rates of circulatory complications (19.0%), the second highest rate of ophthalmic complications (0.2%), and neurological complications (7.1%) and did not have the lowest rate of any complications. Native Americans had the highest rate of arthropathy and oral complications (19.2%) and the second highest rate of many other complications including hyperosmolarity (5.0%), ketoacidosis (17.7%), and skin and foot complications (17.4%). Generally, the Other race categories were commonly found in third or fourth rank for rates of complications but had the lowest rates of ketoacidosis (16.5%, matching the rate for Hispanics).

3.2. In-Hospital Mortality

Table 3 includes the results of the binary logistic regression analysis to determine if race and diabetes complications are associated with increased odds of in-hospital mortality. Asian/Pacific Islanders (OR = 1.252, 95% CI: 1.019–1.537, p = 0.032) and Other race groups (OR = 1.270, 95% CI: 1.049–1.537, p = 0.014) had 25.2% and 27.0% increased odds of experiencing in-hospital mortality compared to Whites. However, Hispanics (OR = 0.894, 95% CI: 0.802–0.997, p = 0.044) had 10.6% decreased odds of experiencing hospital mortality compared to Whites. Finally, there was no significant difference in the odds of in-hospital mortality between Whites and Black/African Americans (OR = 0.958, 95% CI: 0.875–1.048) or between Whites and Native Americans (OR = 1.219, 95% CI: 0.868–1.711).
Some type 2 diabetes complications also influenced mortality, but not in an expected direction. In fact, neurological complications and hyperglycemia were protective against mortality. Neurological complications (OR = 0.278, 95% CI: 0.111–0.702, p = 0.007) reduced the odds of mortality by approximately 72.2% compared to diabetes without complications and hyperglycemia (OR = 0.304, 95% CI: 0.124–0.749, p = 0.010) reduced the odds of mortality by 69.6% compared to diabetes without complications. None of the other diabetes complications were significant predictors of in-hospital mortality.
Finally, several control variables were associated with in-hospital mortality. For every year increase in age (OR = 1.039, 95% CI: 1.036–1.043, p < 0.001), the odds of in-hospital mortality increased by 3.9%. Compared to patients with private insurance, patients with Medicare (OR = 1.126, 95% CI: 1.005–1.260, p = 0.040) had 12.6% increased odds of in-hospital mortality. Income appeared to be one of the largest drivers of in-hospital mortality, with higher incomes protective against mortality. The highest income quartile (OR = 0.707, 95% CI: 0.628–0.796, p < 0.001) was associated with 29.3% decreased odds of in-hospital mortality compared to the lowest income quartile. Income Quartile 3 (OR = 0.873, 95% CI: 0.792–0.962, p = 0.006) and Quartile 2 (OR = 0.854, 95% CI: 0.781–0.933, p < 0.001) were also associated with 12.7% and 14.6% decreased odds of mortality compared to the lowest quartile income.

4. Discussion

Using data from the 2018 NIS [7], this study describes the association between race and type 2 diabetes complications, the difference in prevalence of type 2 diabetes complications between races, and the odds of mortality associated with race and type 2 diabetes complications. Our analyses indicated significant associations between race and type 2 diabetes complications and revealed significant differences in the proportions of several complications among race/ethnicity groups. Here, Whites had the lowest rates of all complications compared to non-White races (aside from arthropathy and oral complications), and Blacks had the highest rates for most complications, followed closely by Asian/Pacific Islanders and then Hispanics and Native Americans. These results align with other studies that report Blacks and Hispanics have an increased prevalence of diabetes complications compared to Whites [18,19] These results do not consider the severity of diabetes complications, which may be important considering research by Tan et al. [9] that reported Blacks, Hispanics, and Native Americans admitted to the hospital for diabetic foot infections had longer hospital stays and increased risk of major amputations compared to Whites. Therefore, while Whites were found to have higher rates of skin and foot complications in this study, they may very well have shorter hospital stays and be less likely to have amputations [9]. These results also do not consider the number of diabetes complications on a specific discharge record. However, subsequent analyses indicate each record included approximately two complications, and while there are statistically significant differences between some race/ethnicity groups (e.g., Hispanics had a greater number of complications than Whites, Blacks, and Asian Pacific Islanders: 2.23 versus 2.13, 1.97, and 2.09, respectively), these small differences are not likely clinically significant, as evidenced by the lower odds of in-hospital mortality in the Hispanic group, as an example.
Our results modeling the odds of in-hospital mortality partially supported the notion that race/ethnicity and type 2 diabetes complications would influence mortality. Asian/Pacific Islanders and Other races had significantly increased odds of mortality compared to Whites, however, Black/African Americans and Native Americans had no significant difference in odds of mortality compared to Whites, and Hispanics had a decreased odds of mortality by 10.6% compared to Whites. Our finding that the Hispanic race is protective against mortality does not align with other research [20,21,22] reporting that Blacks and Hispanics had higher odds of diabetes-related mortality compared to Whites. However, the data for this study only assessed whether patients were alive upon discharge or died at the hospital and did not collect follow-up information after discharge, which could cause differences in the odds of mortality outcomes between this study and others that consider mortality occurring within 30 days after a hospital stay.
Regarding type 2 diabetes complications, while hyperosmolarity, ketoacidosis, hypoglycemia, kidney complications, and circulatory complications were associated with increased odds of mortality, these were not significant associations, and, in fact, some complications were associated with decreased odds of mortality including neurological complications and hyperglycemia. Hyperglycemia as a protective variable against mortality contradicts other literature [23,24] and may have resulted from a high prevalence of hyperglycemia in non-critically ill patients as well as the ability to treat hyperglycemia through insulin [25]. A more likely explanation may be related to the length of stay in the hospital, as well as how mortality is measured in the NIS. Further analyses of these data indicate that the length of stay (LOS) in the hospital was shorter for cases with hyperglycemia (3.21 days) than the length of stay for most other complications which ranged from approximately four (4) days to over eight (8) days. Because the mortality variable in the NIS only measures in-hospital mortality, not all-cause or 30-day mortality, even if cases with hyperglycemia died after their hospital stay, it would not be captured in NIS data. A longer hospital stay experienced by patients with other complications increases the chances of observing a death in the hospital just because they were in the hospital for a longer time.
Neurological complications may have acted protectively against the odds of mortality in this sample due to large variations in symptoms, slow development of the condition, and hyperglycemia as a cause of neurological complications; however, this needs to be further explored [26]. In addition, patients with neurological complications, similar to patients with hyperglycemia, had a shorter LOS (4.26) than most other complications, possibly reducing the odds of observing in-hospital mortality solely because they were no longer in the hospital. In addition, additional analyses about where patients were discharged revealed that patients with neurological complications were at increased odds of being discharged to a skilled nursing or inpatient care facility (OR = 1.41, 95% CI: 1.13–1.75) or to home healthcare (OR = 2.02, 95% CI: 1.59–2.56) compared to routine discharges with no restrictions or additional treatments. Deaths in these settings would not be captured in the NIS. Hypotheses related to the results for hyperglycemia and neurological complications could be better tested using prospective, longitudinal studies with longer-term follow-up of these patients in different settings.
It is also possible that the NIS HCUP type 2 diabetes mortality data do not include all diabetes-related deaths. Rodriguez et al. [21] stated that there is a national underrepresentation of type 2 diabetes in mortality cause of death and that the use of a “multiple-cause-of-death-analysis” would allow for more accurate mortality data related to diabetes. Diabetes is also underreported by adult populations [19]. Additionally, studies addressing associations between race and health are complex, as many independent variables interact [12,22]. In this study, low-income and Medicaid insurance coverage was significantly associated with higher odds of mortality. Thus, race, income, insurance, and other independent variables may interact to collectively create differences and disparities in health outcomes for people with type 2 diabetes.
Results for other variables including age, insurance status, income, and type of complication, suggest an important contribution to the odds of mortality. Thus, race was not the only predictor of increased odds of mortality; however, where race was significant, it proved to be a stronger influence on increased odds of mortality than any other variable, except for the highest income quartile. That said, it is important to understand the ways in which multiple independent variables may interact to increase the odds of mortality for underserved populations. For example, specific complications vary in their health impact, and higher income is often associated with better insurance, care, and access to healthcare [12,13].
Diabetic health outcomes are strongly associated with race; however, socioeconomic factors like income, healthcare access, and education are strong influences on outcomes [12]. These variables are rarely mutually exclusive, with Blacks and Hispanics commonly experiencing lower income and educational achievement and poorer healthcare access compared to Whites [13]. Race is not so much a biological construct as it is a social one, and race may be considered as a proxy measure of socioeconomic status, education, housing, and/or transportation. We were only able to use income as a measure of socioeconomic status. These experiences, as well as racism inherent in institutions, rather than race alone, may be more heavily associated with increased prevalence, morbidity, and mortality in people with diabetes. Culture also varies by race and may play an important role in health beliefs and affected health outcomes [27]. Future research should attempt to include additional reliable and validated SDOH measures.
A strength of this study includes the large sample size (N = 97,310 unweighted and N = 486,550 weighted) from a national database including 7,105,495 hospital discharges representing 97% of national hospital discharges [7] allowing for an examination of multiple type 2 diabetes morbidities and in-patient mortality, as well as many independent variables such as race, age, sex, and income and insurance status. Results from this study could be used to advocate for better screening of type 2 diabetes complications as well as routine treatment. These diagnosis and treatment efforts should target populations most in need, and results from this study can help identify who that population is (i.e., non-White people in the lowest income quartile). This study also expands on the limited health research of American Asian/Pacific Islanders and Native Americans with type 2 diabetes. Prior studies focus primarily on larger populations such as Black and Hispanic groups, failing to observe other non-White races. There is a lack of literature on diabetes health outcomes for smaller race groups, and this affects the ability to make conclusive statements that could lead to positive changes in healthcare for these populations [28].
This study’s cross-sectional design limits its ability to establish causal relationships between diabetes complications and in-hospital mortality. Future research should incorporate longitudinal studies to better assess these connections. Other limitations in this study lie primarily in data collection regarding race. Patients cannot choose multiple race categories to describe themselves; therefore, there is no way to collect accurate data on biracial patients in this dataset. Furthermore, there is no information on the “Other” race category and who that may include. This is important because the other race category experiences 27% increased odds of mortality compared to Whites. The Hispanic ethnicity race category is also quite vague given that this population is not monolithic and has many different origins of people such as of Cuba, Mexican, Puerto Rican, and South and Central American descent [29]. That said, no race category is truly monolithic, and the same issue could exist for all races.
Finally, we also recognize that there are some diabetes complications that may be considered acute versus chronic [30]. In this study, we included all complications together to examine differences among races in the prevalence of complications including kidney, ophthalmic, neurological, oral, circulatory, hyperosmolarity, ketoacidosis, arthropathy, skin and foot issues, and hypo- and hyperglycemia. We also explored the association of race/ethnicity and complications on in-hospital mortality, while controlling for age, sex, insurance status, patient location (rural), and income, and using as a reference group those with diabetes but without any of these complications. Our intent was to study the group and these complications in their entirety, which we have reported here. However, chronic and acute conditions by their very nature may have differential influences on the course of disease. Because of this, we conducted supplemental analyses grouping acute complications (type 2 diabetes w/out complications (as a reference category), hyperosmolarity, ketoacidosis, hypoglycemia, and hyperglycemia) and chronic complications (type 2 diabetes w/out complications (as a reference category), kidney, ophthalmic, neurological, circulatory, other/arthropathy/oral, and foot/skin ulcers) into separate variables and ran separate unadjusted and adjusted analyses for each complication group. Results related to the complications were essentially similar for adjusted and unadjusted regressions. When conducting regressions on mortality using only the chronic complications, we found that none of these were significantly associated with in-hospital mortality. Of the control variables, only age (OR = 1.04, 95% CI: 1.03–1.05) and the highest income quartile (OR = 0.586, 95% CI: 0.409, 0.839) were significantly associated with in-hospital mortality in this model. We also ran regressions using only acute complications and found that ketoacidosis (OR = 3.51, 95% CI: 2.04, 6.03), hypoglycemia (OR = 4.42, 95% CI: 2.78, 7.05), and hyperglycemia (OR = 4.29, 95% CI: 2.70, 6.80) significantly increase the odds of in-hospital mortality. Of the control variables, only age (OR = 1.04, 95% CI: 1.03–1.05) was significantly associated with in-hospital mortality in this model. These results suggest that acute complications, rather than chronic ones, may have a stronger immediate impact on in-hospital mortality. Chronic conditions may progress more gradually, potentially reducing their immediate mortality risk. Future research should consider examining the timing of complications, the role of treatment interventions, and the interaction between acute and chronic conditions to better understand the differential impacts on mortality. Future studies, especially larger ones, should explore the timing and treatment of complications and the interaction between acute and chronic conditions and observe mortality over a longer time frame to better understand their effects.

5. Conclusions

We found that some categories based on race/ethnicity are significantly associated with differences in morbidity and odds of mortality for adult type 2 diabetes patients in the United States. Most commonly, Black and Asian/Pacific Islander race categories have the highest rates of morbidity, with Hispanic and Native American groups also disproportionately affected compared to Whites. In terms of mortality, this study found that Asian/Pacific Islander and Other race categories had the highest odds of mortality, with odds increased by 25 and 27%, respectively, compared to Whites. Other variables were also important in predicting mortality, including income and insurance status. However, analyzing mortality using only chronic or acute complications negated the effects of race and most other control variables, while some acute conditions increased the odds of mortality by as much as 600%.

Author Contributions

Conceptualization, L.A.B. and J.M.C.; Methodology, L.A.B. and J.M.C.; Formal Analysis, L.A.B. and J.M.C.; Data Curation, J.M.C.; Writing—Original Draft Preparation, L.A.B.; Writing—Review and Editing, L.A.B. and J.M.C.; Supervision, J.M.C.; Funding Acquisition, J.M.C. All authors have read and agreed to the published version of the manuscript.

Funding

Data acquisition was funded by a faculty development grant from the Charles Stewart Mott Department of Public Health to John M. Clements.

Institutional Review Board Statement

Not applicable.

Data Availability Statement

The datasets presented in this article are not readily available because HCUP specifically prohibits data sharing as part of the Data Use Agreement. Requests to access HCUP datasets should be directed to HCUP at the Agency for Healthcare Research and Quality.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Centers for Disease Control and Prevention. National Diabetes Statistics Report. Centers for Disease Control and Prevention. 2022. Available online: https://www.cdc.gov/diabetes/data/statistics-report/index.html (accessed on 23 February 2023).
  2. Centers for Disease Control and Prevention. Diabetes Report Card 2021. Centers for Disease Control and Prevention. 19 July 2022. Available online: https://www.cdc.gov/diabetes/library/reports/reportcard.html#:~:text=American%20Indian%20or%20Alaska%20Native,and%207.4%25%2C%20respectively (accessed on 23 February 2023).
  3. Cheng, Y.J.; Kanaya, A.M.; Araneta, M.R.; Saydah, S.H.; Kahn, H.S.; Gregg, E.W.; Fujimoto, W.Y.; Imperatore, G. Prevalence of diabetes by race and ethnicity in the United States, 2011–2016. JAMA 2019, 322, 2389–2398. [Google Scholar] [CrossRef] [PubMed]
  4. Haw, S.J.; Shah, M.; Turbow, S.; Egeolu, M.; Umpierrez, G. Diabetes complications in racial and ethnic minority populations in the USA. Curr. Diabetes Rep. 2021, 21, 1–8. [Google Scholar] [CrossRef] [PubMed]
  5. Peek, M.E.; Cargill, A.; Huang, E.S. Diabetes health disparities. Med. Care Res. Rev. 2007, 64, 101S–156S. [Google Scholar] [CrossRef] [PubMed]
  6. Spanakis, E.K.; Golden, S.H. Race/ethnic difference in diabetes and diabetic complications. Curr. Diabetes Rep. 2013, 13, 814–823. [Google Scholar] [CrossRef] [PubMed]
  7. HCUP National Inpatient Sample (NIS). Healthcare Cost and Utilization Project (HCUP). 2018. Agency for Healthcare Research and Quality, Rockville, MD. Available online: www.hcup-us.ahrq.gov/nisoverview.jsp (accessed on 23 February 2023).
  8. Gregg, E.W.; Li, Y.; Wang, J.; Burrows, N.R.; Ali, M.K.; Rolka, D.; Williams, D.E.; Geiss, L. Changes in diabetes-related complications in the United States, 1990–2010. N. Eng. J. Med. 2014, 370, 1514–1523. [Google Scholar] [CrossRef]
  9. Tan, T.-W.; Shih, C.-D.; Concha-Moore, K.C.; Diri, M.M.; Hu, B.; Marrero, D.; Zhou, W.; Armstrong, D.G. Disparities in outcomes of patients admitted with diabetic foot infections. PLoS ONE 2019, 14, e0211481. [Google Scholar] [CrossRef]
  10. Lewis, E.F.; Claggett, B.; Parfrey, P.S.; Burdmann, E.A.; McMurray, J.J.V.; Solomon, S.D.; Levey, A.S.; Ivanovich, P.; Eckardt, K.-U.; Kewalramani, R.; et al. Race and ethnicity influences on cardiovascular and renal events in patients with diabetes mellitus. Am. Heart J. 2015, 170, 322–329. [Google Scholar] [CrossRef]
  11. Bell, C.N.; Thorpe, R.J.; Bowie, J.V.; LaVeist, T.A. Race disparities in cardiovascular disease risk factors within socioeconomic status strata. Ann. Epidemiol. 2018, 28, 147–152. [Google Scholar] [CrossRef]
  12. Hill-Briggs, F.; Adler, N.E.; Berkowitz, S.A.; Chin, M.H.; Gary-Webb, T.L.; Navas-Acien, A.; Thornton, P.L.; Haire-Joshu, D. Social Determinants of Health and diabetes: A scientific review. Diabetes Care 2021, 44, 258–279. [Google Scholar] [CrossRef]
  13. Canedo, J.R.; Miller, S.T.; Schlundt, D.; Fadden, M.K.; Sanderson, M. Racial/ethnic disparities in diabetes quality of care: The role of healthcare access and socioeconomic status. J. Racial Ethn. Health Disparities 2017, 5, 7–14. [Google Scholar] [CrossRef]
  14. ICD10 Data. 2023 ICD-10-CM Codes E11*: Type 2 Diabetes Mellitus. ICD10 Data. 2023. Available online: https://www.icd10data.com/ICD10CM/Codes/E00-E89/E08-E13/E11- (accessed on 14 January 2023).
  15. Houchens, R.; Ross, D.; Elixhauser, A. Final Report on Calculating National Inpatient Sample (NIS) Variances for Data Years 2012 and Later. U.S.; HCUP Methods Series Report; Agency for Healthcare Research and Quality: Rockville, MD, USA, 2015. Available online: https://www.hcup-us.ahrq.gov/reports/methods/methods.jsp (accessed on 14 January 2023).
  16. Ingram, D.D.; Franco, S.J. NCHS urban-rural classification scheme for counties. Vital Health Stat. 2012, 154, 1–65. [Google Scholar] [PubMed]
  17. IBM Corp. IBM SPSS Statistics for Windows, Version 27.0, Computer Software; IBM Corp.: Armonk, NY, USA, 2020.
  18. Osborn, C.Y.; de Groot, M.; Wagner, J.A. Racial and ethnic disparities in diabetes complications in the Northeastern United States: The role of socioeconomic status. J. Natl. Med. Assoc. 2013, 105, 51–58. [Google Scholar] [CrossRef] [PubMed]
  19. Black, S.A. Diabetes, diversity, and disparity: What do we do with the evidence? Am. J. Public Health 2002, 92, 543–548. [Google Scholar] [CrossRef] [PubMed]
  20. Lanting, L.C.; Joung, I.M.A.; Mackenbach, J.P.; Lamberts, S.W.J.; Bootsma, A.H. Ethnic differences in mortality, end-stage complications, and quality of care among diabetic patients. Diabetes Care 2005, 28, 2280–2288. [Google Scholar] [CrossRef] [PubMed]
  21. Rodriguez, F.; Blum, M.R.; Falasinnu, T.; Hastings, K.G.; Hu, J.; Cullen, M.R.; Palaniappan, L.P. Diabetes-attributable mortality in the United States from 2003 to 2016 using a multiple-cause-of-death approach. Diabetes Res. Clin. Pract. 2019, 148, 169–178. [Google Scholar] [CrossRef]
  22. Rosenstock, S.; Whitman, S.; West, J.F.; Balkin, M. Racial disparities in diabetes mortality in the 50 most populous US cities. J. Urban Health 2014, 91, 873–885. [Google Scholar] [CrossRef]
  23. Falciglia, M.; Freyberg, R.W.; Almenoff, P.L.; D’Alessio, D.A.; Render, M.L. Hyperglycemia–related mortality in critically ill patients varies with admission diagnosis. Crit. Care Med. 2009, 37, 3001–3009. [Google Scholar] [CrossRef]
  24. Leite, S.A.O.; Locatelli, S.B.; Niece, S.P.; Oliveira, A.R.F.; Tockus, D.; Tosin, T. Impact of hyperglycemia on morbidity and mortality, length of hospitalization and rates of re-hospitalization in a general hospital setting in Brazil. Diabetol. Metab. Syndr. 2010, 2, 49. [Google Scholar] [CrossRef]
  25. Dhatariya, K.; Corsino, L.; Umpierrez, G.E. Management of Diabetes and Hyperglycemia in Hospitalized Patients. National Library of Medicine. 30 December 2020. Available online: https://www.ncbi.nlm.nih.gov/books/NBK279093/ (accessed on 29 January 2023).
  26. Johns Hopkins Medicine. Diabetic Neuropathy. Johns Hopkins Medicine. Available online: https://www.hopkinsmedicine.org/health/conditions-and-diseases/diabetes/diabetic-neuropathy-nerve-problems (accessed on 29 January 2023).
  27. Rosal, M.C.; Ockene, I.S.; Restrepo, A.; White, M.J.; Borg, A.; Olendzki, B.; Scavron, J.; Candib, L.; Welch, G.; Reed, G. Randomized trial of a literacy-sensitive, culturally tailored diabetes self-management intervention for low-income Latinos. Diabetes Care 2011, 34, 838–844. [Google Scholar] [CrossRef]
  28. King, G.L.; McNeely, M.J.; Thorpe, L.E.; Mau, M.L.M.; Ko, J.; Liu, L.L.; Sun, A.; Hsu, W.C.; Chow, E.A. Understanding and addressing unique needs of diabetes in Asian Americans, Native Hawaiians, and Pacific Islanders. Diabetes Care 2012, 35, 1181–1188. [Google Scholar] [CrossRef]
  29. U.S. Census Bureau. About the Hispanic Population and Its Origin. Census.gov. 15 April 2022. Available online: https://www.census.gov/topics/population/hispanic-origin/about.html#:~:text=About%20Hispanic%20Origin&text=OMB%20defines%20%22Hispanic%20or%20Latino,or%20origin%20regardless%20of%20race (accessed on 20 January 2023).
  30. Farmaki, P.; Damaskos, C.; Garmpis, N.; Garmpi, A.; Savvanis, S.; Diamantis, E. Complications of the type 2 diabetes mellitus. Curr. Cardiol. Rev. 2020, 16, 249–251. [Google Scholar] [CrossRef]
Figure 1. Prevalence (%) of type 2 diabetes complications by race/ethnicity. * API = Asian Pacific Islander; NA = Native American.
Figure 1. Prevalence (%) of type 2 diabetes complications by race/ethnicity. * API = Asian Pacific Islander; NA = Native American.
Diabetology 06 00015 g001
Table 1. Characteristics of cohort, N = 486,500.
Table 1. Characteristics of cohort, N = 486,500.
Characteristicn (%)
Age—mean (SD)60.28
Sex
 Male287,325 (59.1)
 Female199,210 (40.9)
Insurance Status
 Medicare254,800 (57.9)
 Medicaid92,775 (21.1)
 Private Insurance92,655 (21.0)
Race
 White256,860 (53.8)
 Black/African American117,840 (24.7)
 Hispanic74,295 (15.6)
 Asian/Pacific Islander9350 (2.0)
 Native American5270 (1.1)
 Other13,610 (2.9)
Patient Location
 Metropolitan407,595 (84.4)
 Non-Metropolitan75,610 (15.6)
Income
 Lowest Quartile (USD <45,999)179,880 (37.7)
 Quartile 2 (USD 46,000–58,999)132,725 (27.8)
 Quartile 3 (USD 59,000–78,999)99,265 (20.8)
 Highest Quartile (USD >79,000)64,720 (13.6)
Diabetes Type 2 Complications
 No complications (E11.9)980 (0.20)
 Hyperosmolarity (E11.0X)24,180 (5.0)
 Ketoacidosis (E11.1X)83,545 (17.2)
 Kidney complications (E11.2X)10,325 (2.1)
 Ophthalmic complications (E11.3X)595 (0.12)
 Neurological complications (E11.4X)31,780 (6.5)
 Circulatory complications (E11.5X)80,465 (16.5)
 Arthropathy, oral complications (E11.61X, E11.63X)78,815 (16.2)
 Skin complications—foot/skin ulcers (E11.62X)74,410 (15.3)
 Hypoglycemia (E11.64X)46,520 (9.6)
 Hyperglycemia (E11.65X).54,935 (11.3)
Mortality
 Alive on Discharge482,7690 (99.3)
 Died3600 (0.74)
Table 2. Chi-square test and z-tests for differences in proportions *.
Table 2. Chi-square test and z-tests for differences in proportions *.
n (%) of Diabetes Complications by Race/Ethnicity Group
Type 2 Diabetes Complication CategoryWhiteBlackHispanicAsian/Pacific
Islander
Native
American
Other
No complications (E11.9)530 (0.2) a190 (0.2) a145 (0.2) a25 (0.3) a,b25 (0.5) b25 (0.2) a
Hyperosmolarity (E11.0X)10,965 (4.3) a8570 (7.3) b2940 (4.0) c430 (4.6) a265 (5.0) a605 (4.4) a,c
Ketoacidosis (E11.1X)40,360 (15.7) a24,940 (21.2) b12,295 (16.5) c1225 (13.1) d935 (17.7) c2245 (16.5) a,c
Kidney complications (E11.2X)3890 (1.5) a2425 (2.1) b2455 (3.3) c675 (7.2) d215 (4.1) e410 (3.0) c
Ophthalmic complications (E11.3X)285 (0.1) a115 (0.1) a150 (0.2) b25 (0.3) b##
Neurological complications (E11.4X)13,775 (5.4) a10,485 (8.9) b5260 (7.1) c560 (6.0) a220 (4.2) d955 (7.0) c
Circulatory complications (E11.5X)42,250 (16.4) a17,485 (14.8) b14,095 (19.0) c1760 (18.8) c,d880 (16.7) a,e2380 (17.5) d,e
Arthropathy, oral complications (E11.61X, E11.63X)48,375 (18.8) a13,045 (11.1) b11,945 (16.1) c1160 (12.4) d1010 (19.2) a1875 (13.8) e
Skin complications—foot/skin ulcers (E11.62X)46,695 (18.2) a11,860 (10.1) b10,655 (14.3) c815 (8.7) d915 (17.4) a1950 (14.3) c
Hypoglycemia (E11.64X)23,265 (9.1) a12,880 (10.9) b6045 (8.1) c1645 (17.6) d325 (6.2) e1465 (10.8) b
Hyperglycemia (E11.65X).26,470 (10.3) a,b15,845 (13.4) c8310 (11.2) d1030 (11.0) b,d480 (9.1) a1695 (12.5) e
Chi-square = 16,617.778, df = 50, p < 0.001
* For each complication category, insignificant differences in proportions are designated using the same alphabetic letter superscript. Significant (p < 0.05) differences in group proportions between race/ethnic groups are designated by different alphabetic superscripts within each complication subcategory. # HCUP data use agreements prohibit reporting group sizes less than or equal to 10 to reduce the likelihood of identifying individual discharge records.
Table 3. Binary logistic regression model predicting in-hospital mortality.
Table 3. Binary logistic regression model predicting in-hospital mortality.
CharacteristicOdds Ratio (95% CI) *,
p-Value
Age1.039 (1.036, 1.043), <0.001
Female Sex (Reference: Male)0.972 (0.905, 1.043), 0.427
Insurance Status (Reference: Private Insurance)
 Medicare1.126 (1.005, 1.260), 0.040
 Medicaid1.117 (0.972, 1.284), 0.119
Race (Reference: White)
 Black/African American0.958 (0.875, 1.048), 0.348
 Hispanic0.894 (0.802, 0.997), 0.044
 Asian/Pacific Islander1.252 (1.019, 1.537), 0.032
 Native American1.219 (0.868, 1.711), 0.253
 Other1.270 (1.049, 1.537), 0.014
non-Metropolitan Location (Reference: Metropolitan)1.029 (0.932, 1.136), 0.577
Income (Reference: Lowest Quartile, USD <45,999)
 Quartile 2 (USD 46,000–58,999)0.854 (0.781, 0.933), <0.001
 Quartile 3 (USD 59,000–78,999)0.873 (0.792, 0.962), 0.006
 Highest Quartile (USD >79,000)0.707 (0.628, 0.796), <0.001
Type of Diabetes Complications (Reference: Diabetes w/out complications)
 Hyperosmolarity (E11.0X)1.075 (0.439, 2.630), 0.874
 Ketoacidosis (E11.1X)1.395 (0.575, 3.380), 0.462
 Kidney complications (E11.2X)1.320 (0.534, 3.263), 0.548
 Ophthalmic complications (E11.3X) #---
 Neurological complications (E11.4X)0.278 (0.111, 0.702), 0.007
 Circulatory complications (E11.5X)1.843 (0.762, 4.458), 0.175
 Arthropathy, oral complications (E11.61X, E11.63X)0.537 (0.221, 1.307), 0.171
 Skin complications—foot/skin ulcers (E11.62X)0.515 (0.211, 1.254), 0.144
 Hypoglycemia (E11.64X)1.268 (0.523, 3.075), 0.599
 Hyperglycemia (E11.65X)0.304 (0.124, 0.749), 0.010
# Subjects with this complication were not included in the analysis due to the small sample size for this category. * CI = confidence interval.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Burress, L.A.; Clements, J.M. Racial and Ethnic Disparities in Type 2 Diabetes Complications and In-Hospital Mortality in the United States: A Retrospective Cohort Study. Diabetology 2025, 6, 15. https://doi.org/10.3390/diabetology6030015

AMA Style

Burress LA, Clements JM. Racial and Ethnic Disparities in Type 2 Diabetes Complications and In-Hospital Mortality in the United States: A Retrospective Cohort Study. Diabetology. 2025; 6(3):15. https://doi.org/10.3390/diabetology6030015

Chicago/Turabian Style

Burress, Lainy A., and John M. Clements. 2025. "Racial and Ethnic Disparities in Type 2 Diabetes Complications and In-Hospital Mortality in the United States: A Retrospective Cohort Study" Diabetology 6, no. 3: 15. https://doi.org/10.3390/diabetology6030015

APA Style

Burress, L. A., & Clements, J. M. (2025). Racial and Ethnic Disparities in Type 2 Diabetes Complications and In-Hospital Mortality in the United States: A Retrospective Cohort Study. Diabetology, 6(3), 15. https://doi.org/10.3390/diabetology6030015

Article Metrics

Back to TopTop