Quantifying association and disparities between diabetes complications and COVID-19 outcomes: A retrospective study using electronic health records

Abstract

Background

Despite established relationships between diabetic status and an increased risk for COVID-19 severe outcomes, there is a limited number of studies examining the relationships between diabetes complications and COVID-19-related risks. We use the Adapted Diabetes Complications Severity Index to define seven diabetes complications. We aim to understand the risk for COVID-19 infection, hospitalization, mortality, and longer length of stay of diabetes patients with complications.

Methods

We perform a retrospective case-control study using Electronic Health Records (EHRs) to measure differences in the risks for COVID-19 severe outcomes amongst those with diabetes complications. Using multiple logistic regression, we calculate adjusted odds ratios (OR) for COVID-19 infection, hospitalization, and in-hospital mortality of the case group (patients with diabetes complications) compared to a control group (patients without diabetes). We also calculate adjusted mean difference in length of stay between the case and control groups using multiple linear regression.

Results

Adjusting demographics and comorbidities, diabetes patients with renal complications have the highest odds for COVID-19 infection (OR = 1.85, 95% CI = [1.71, 1.99]) while those with metabolic complications have the highest odds for COVID-19 hospitalization (OR = 5.58, 95% CI = [3.54, 8.77]) and in-hospital mortality (OR = 2.41, 95% CI = [1.35, 4.31]). The adjusted mean difference (MD) of hospital length-of-stay for diabetes patients, especially those with cardiovascular (MD = 0.94, 95% CI = [0.17, 1.71]) or peripheral vascular (MD = 1.72, 95% CI = [0.84, 2.60]) complications, is significantly higher than non-diabetes patients. African American patients have higher odds for COVID-19 infection (OR = 1.79, 95% CI = [1.66, 1.92]) and hospitalization (OR = 1.62, 95% CI = [1.39, 1.90]) than White patients in the general diabetes population. However, White diabetes patients have higher odds for COVID-19 in-hospital mortality. Hispanic patients have higher odds for COVID-19 infection (OR = 2.86, 95% CI = [2.42, 3.38]) and shorter mean length of hospital stay than non-Hispanic patients in the general diabetes population. Although there is no significant difference in the odds for COVID-19 hospitalization and in-hospital mortality between Hispanic and non-Hispanic patients in the general diabetes population, Hispanic patients have higher odds for COVID-19 hospitalization (OR = 1.83, 95% CI = [1.16, 2.89]) and in-hospital mortality (OR = 3.69, 95% CI = [1.18, 11.50]) in the diabetes population with no complications.

Conclusions

The presence of diabetes complications increases the risks of COVID-19 infection, hospitalization, and worse health outcomes with respect to in-hospital mortality and longer hospital length of stay. We show the presence of health disparities in COVID-19 outcomes across demographic groups in our diabetes population. One such disparity is that African American and Hispanic diabetes patients have higher odds of COVID-19 infection than White and Non-Hispanic diabetes patients, respectively. Furthermore, Hispanic patients might have less access to the hospital care compared to non-Hispanic patients when longer hospitalizations are needed due to their diabetes complications. Finally, diabetes complications, which are generally associated with worse COVID-19 outcomes, might be predominantly determining the COVID-19 severity in those infected patients resulting in less demographic differences in COVID-19 hospitalization and in-hospital mortality.

Introduction

Coronavirus Disease 2019 (COVID-19) is an infectious disease caused by the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) virus. On March 20, 2020, following a total of 234,073 confirmed cases and 9,840 deaths globally [1], the World Health Organization (WHO) declared COVID-19 a pandemic. Demographic factors such as age, sex, race, and ethnicity are associated with the risk for COVID-19 infection, hospitalization, and death [2]. Furthermore, individuals with underlying health conditions have a higher risk for severe COVID-19 outcomes [3, 4]. The most common comorbidities among patients hospitalized with COVID-19 include hypertension, obesity and diabetes [5, 6]. In this study, we analyze the association between diabetes complications and COVID-19 outcomes. We are specifically interested in COVID-19 infection, hospitalization, hospital length of stay, and in-hospital mortality. Type 1 and Type 2 diabetes are associated with increased chance of hospitalization for COVID-19 and in-hospital mortality [7–10]. Cardiovascular conditions and hospitalizations, especially those related to heart failure, stroke, and myocardial infarction, are associated with increased COVID-19 related mortality in patients with diabetes [10]. Lim et al. [11] proposed several pathophysiological mechanisms leading to increased cardiovascular mortality after infection with SARS-CoV-2 in patients with diabetes such as increased levels of inflammatory mediators in the blood (e.g., lipopolysaccharide, cytokines, and metabolites), insulin resistance, and vascular endothelial damage. Zhou et al. [12] provided a synopsis of the reasons for increased severity of COVID-19 in patients with diabetes including aggravated inflammatory storm, immune system dysfunction, and diabetes-related comorbidities such as lung injury and renal damage. Guo et al. [13] investigated the pathophysiological mechanisms behind the deterioration of COVID-19 patients with diabetes, and found that they had higher ferritin and coagulation index compared to patients without diabetes, suggesting that COVID-19 patients with diabetes are more susceptible to an inflammatory storm. Furthermore, they observed higher odds of mortality in patients with diabetic complications. However, not all complications were represented in their study cohort. Additionally, having a history of microvascular diabetic complications (e.g., diabetic kidney disease, severe diabetic retinopathy, and history of diabetic foot ulcer) was associated with a reduced chance of discharge within 28 days of hospital admission for those infected with COVID-19 [14].The focus of this study was to quantify this association amongst those hospitalized and did not investigate the association of these complications with being hospitalized with COVID-19.

Diabetes may cause macrovascular and microvascular complications based on the injurious effects of hyperglycemia [15]. Hyperglycemia (high blood glucose) is considered a metabolic complication of diabetes [16, 17] and there are many studies that establish a relationship between hyperglycemia and severe COVID-19 outcomes, especially in patients with diabetes [18–23]. These studies focus on hyperglycemia alone and, despite being a common metabolic complication amongst diabetes patients, do not investigate a more general relationship between diabetes related metabolic complications and COVID-19 outcomes. Similar analysis has been done on the relationship between COVID-19 and other specific metabolic complications among the diabetes population including ketoacidosis [24–28] and hypoglycemia [29]. Associations between specific diabetic complications and COVID-19 extends to macrovascular complications including coronary artery disease [30], peripheral arterial disease [31], and stroke [32] as well as other microvascular complications such as diabetic nephropathy [33, 34], neuropathy [35], and retinopathy [34, 36].

Although there is an abundance of work investigating the relations between diabetes and COVID-19, previous research has focused on either the general diabetes population (i.e., not focused on complications) or patients with specific diabetic complications (e.g., hyperglycemia as a metabolic complication or peripheral arterial disease as a cardiovascular complication). There are currently no studies that compare the association of different diabetic complications with COVID-19 outcomes. Furthermore, there are no studies investigating demographic disparities in COVID-19 outcomes among patients with diabetic complications.

One of the objectives of this work is to understand the association of different microvascular (i.e., renal, neurological, ocular) and macrovascular (i.e., cardiovascular, peripheral vascular, cerebrovascular, metabolic) diabetic complications to COVID-19 outcomes. We specifically consider COVID-19 infection, hospitalization, in-hospital mortality and hospital length of stay outcomes. We also aim to investigate demographic disparities in COVID-19 outcomes among patients with diabetic complications. We accomplish these objectives through a retrospective cohort study analyzing the observations of the considered COVID-19 outcomes in patients with diabetic complications across seven different body systems. Our findings provide clinicians with actionable information when caring for patients that have diabetic complications by indicating which complications, if any, have a higher association with adverse COVID-19 outcomes.

Materials and methods

Data

This is a retrospective study that uses personal protected health information derived from electronic medical records at MedStar Health, a healthcare system that operates 10 hospitals in the Baltimore-Washington metropolitan area. The data was fully anonymized before being accessed by the research team. The study was approved by North Carolina State University’s Institutional Review Board (IRB No. 21006). The period of observation is between January 2019 to December 2020 including 1,389,115 unique patients from 6,265,821 encounters with patient level information. We consider 2019 only to identify the diabetic status of patients based on their medical history before the pandemic. Furthermore, by considering the first year of the COVID-19 pandemic when disease mitigation measures (e.g, antiviral drugs and vaccines) were not available and treatment strategies were not well understood, we focus on a period when diabetes patients were at a higher risk for adverse COVID-19 outcomes.

From this initial patient population, we exclude 6,008 patients who died in 2019. There are 156 patients (0.011%) with missing birthdate, 639 patients (0.046%) with missing sex, 169,276 patients (12.19%) with missing race, and 191,555 patients (13.79%) with missing ethnicity information. We exclude patients with missing birthdate and retain other patients with missing information in the analysis because excluding them all would significantly reduce the population size. We do not impute missing values for sex, race and ethnicity but instead create a missing category. After this exclusion, the final population includes 1,382,951 patients.

We split the population into a case group with 142,807 patients (10.33% of the patient population) and a control group with the rest of the patients. The patients in the case group are those who had at least one diabetes-related encounter during or before 2019 as indicated in their past medical history. A diabetes-related encounter is defined as an encounter with at least one of the ICD-10 codes associated with diabetes (E10—E14). Finally, our data includes 43,473 COVID-19-related encounters from 37,245 patients (2.69% of the patient population) with a positive COVID-19 test, and 8,431 of these encounters resulted in hospitalization of 6,988 patients throughout the observation period.

Observation generation.

The outcomes of interest in this work are COVID-19 infection, COVID-19 hospitalization, COVID-19 length of stay, and COVID-19 in-hospital mortality. A patient is said to have a COVID-19 infection if that patient had at least one COVID-19-related encounter during the observation period. If at least one of these COVID-19-related encounters is classified as an inpatient encounter, we say that the patient had a COVID-19 hospitalization. Thus, COVID-19 is not necessarily the primary or only cause for this hospitalization. For patients with a COVID-19 hospitalization, the COVID-19 length of stay is the duration of the COVID-19 hospitalization encounter or the maximum duration if the patient had more than one COVID-19-related hospitalization. COVID-19 in-hospital mortality is identified through the discharge disposition of the last COVID-19 hospitalization.

The covariates of interest are diabetes status and its complications. Diabetes complications are defined based on the adapted Diabetes Complications Severity Index (aDCSI) [16]. The aDCSI classifies diabetes complications into seven categories: cardiovascular, renal, ocular, peripheral vascular, cerebrovascular, neurological and metabolic complications. We use the translation of aDCSI to ICD-10 [17] to assign patients in the case group to the eight subcase groups, that is, a group of diabetes patients without any complications and seven groups of diabetes patients with different complications. For example, if a patient had at least one encounter with ICD-10 codes associated with cardiovascular complication as defined in aDCSI, then we include that patient in the subcase group of diabetes patients with cardiovascular complication. Note that a patient may be included in multiple complication groups. Fig 1 outlines the process of generating the case and control groups based on diabetes status and its complications.

[Figure omitted. See PDF.]

We consider age, sex, race, and ethnicity as confounding factors in our analysis. Figs 2 and 3 present the proportion of diabetes patients by complication among each race and ethnicity, respectively. We also consider a list of suspected underlying medical conditions associated with more severe COVID-19 illness, other than diabetes, as confounders related to the outcomes and covariates of interest. This includes obesity, hypertension, chronic kidney disease, chronic lower respiratory diseases, substance use disorder, heart disease, cancer, chronic liver disease, stroke, dementia, mood disorders, schizophrenia, rheumatoid arthritis, lupus, psoriasis, and disorders involving an immune mechanism [3, 4]. A patient is identified as having an underlying medical condition if they had at least one encounter during or before 2019 with ICD-10 codes associated with those conditions.

[Figure omitted. See PDF.]

[Figure omitted. See PDF.]

Statistical analysis

We calculate the adjusted odds ratio for COVID-19 infection, hospitalization, and in-hospital mortality of the case group compared to the control group. The odds ratio is an appropriate measure of risk in case-control studies under the “uncommon disease” assumption, i.e. when the probability of an outcome is less than 0.1 or 0.2 [37, 38]. We also calculate the adjusted mean difference in length of stay for COVID-19 hospitalization between diabetes and non-diabetes patients. From a clinical perspective, length of stay has been considered as a meaningful outcome and a potential indicator for the quality of patient care [39, 40].

Several factors, including a patient’s demographic characteristics and underlying medical conditions, influence the relationship between diabetic status and COVID-19 outcomes of interest. We use multiple logistic regression to calculate adjusted odds ratios (AOR) to consider the influence of confounding factors [41]. The multiple logistic regression model is formulated as:where Y is the outcome of interest, x is the diabetic status, z is a vector of demographics and underlying medical conditions, ϵ is the error term. Then the estimated AOR that associates diabetic status to the outcome of interest (e.g. COVID-19 infection) is exp (β).

To compute the adjusted mean difference (AMD) for length of stay, we use multiple linear regression [42]. The multiple linear regression model is formulated as follows:where Y is the COVID-19 hospital length of stay. The estimated difference in average length of stay between case and control groups adjusted for demographics and underlying medical conditions is given by β.

Results

We conduct a separate analysis for all patients with diabetes (case group), diabetes patients without complications (subcase group 1), and diabetes patients with complications (subcase groups 2–7). For each analysis, the control group includes patients without diabetes. We present the AOR for COVID-19 infection, hospitalization, in-hospital mortality and AMD of hospital length of stay along with the p-values. Table 1 presents the demographic characteristics of the study population. We perform Pearson’s Chi-Square test to identify differences between the demographic characteristics of the case and control groups that should be adjusted in the analysis. We observe statistically significant differences (p < 0.001) in Sex, Age, Race, and Ethnicity.

[Figure omitted. See PDF.]

COVID-19 infection and hospitalization

Table 2 reports the adjusted odds ratios of diabetes patients for COVID-19 infection and hospitalization. The odds ratios of diabetes patients for COVID-19 infection is significant and greater than one regardless of the complication type. The strongest effect of diabetic status on COVID-19 infection is observed in patients with renal complications (OR = 1.85, 95% CI = [1.71, 1.99]) followed by those with neurological complications (OR = 1.82, 95% CI = [1.67, 1.98]). The odds ratios of diabetes patients with cardiovascular, cerebrovascular and ocular complications for COVID-19 hospitalization is greater than one but they are not statistically significant. The strongest effect of diabetic status on COVID-19 hospitalization is observed in patients with metabolic complications (OR = 5.58, 95% CI = [3.54, 8.77]) followed by those with peripheral vascular complications (OR = 1.82, 95% CI = [1.46, 2.26]).

[Figure omitted. See PDF.]

COVID-19 hospitalization outcomes

Table 3 presents the adjusted mean difference in length of stay and the adjusted odds ratios for in-hospital mortality. The average hospital length of stay for diabetes patients, especially those with cardiovascular or peripheral vascular complications is significantly higher than non-diabetes patients. In general, the odds of COVID-19 in-hospital mortality for diabetes patients are higher than non-diabetes patients. The odds ratios are statistically significant for the case groups of diabetes patients with metabolic, neurological, or renal complications. The highest significant odds ratio of in-hospital mortality is observed for diabetes patients with metabolic complications (OR = 2.41, 95% CI = [1.35, 4.31]).

[Figure omitted. See PDF.]

Demographic disparities in COVID-19 outcomes

In this section we present the results of a comparison between diabetes patients with and without complications in different demographic groups. Table 4 presents the adjusted odds ratios for COVID-19 infection, hospitalization, in-hospital mortality and the mean differences of hospital length of stay between different demographic groups. This analysis is separately performed for all diabetes patients, diabetes patients with complications, and diabetes patients with no complications. We perform the disparity analysis between the case and control groups of a demographic factor by adjusting for other demographic factors and comorbidities. For example, to calculate the AOR for COVID-19 in-hospital mortality between seniors (65+) and younger adults (18–39), we adjust for sex, race, ethnicity and comorbidities.

[Figure omitted. See PDF.]

Among diabetes patients, the odds of COVID-19 infection (OR = 0.91, 95% CI = [0.85, 0.96]), hospitalization (OR = 0.73, 95% CI = [0.64, 0.83]), and in-hospital mortality (OR = 0.75, 95% CI = [0.60, 0.93]) for females are lower than males. The odds for COVID-19 in-hospital mortality of male diabetes patients with complications is also significantly higher than female patients (OR = 0.73, 95% CI = [0.57, 0.93]). The odds of COVID-19 infection for senior diabetes patients is lower than adult patients (ORs = 0.64–0.70, 95% CI = [0.55, 0.73]-[0.66, 0.75]). However, if these patients are infected with COVID-19, their odds of hospitalization are approximately twice that of adult patients. These results about senior and adult patients are still observed within the diabetes group with and with no complications. If senior (65 and older) diabetes patients are hospitalized, their average length of stay is 1.79 days longer than younger adult (18–39 years old) patients. Not surprisingly, senior diabetes patients have significantly higher odds of COVID-19 in-hospital mortality than younger (OR = 4.26, 95% CI = [1.79, 10.12]) and older adult (40–64 years old) patients (OR = 2.15, 95% ci = [1.66, 2.79]). In particular, the odds for COVID-19 in-hospital mortality of senior patients is significantly higher than younger adults patients within the diabetes group with no complications (OR = 15.01, 95% CI = [1.89, 119.21]).

African American diabetes patients have higher odds of COVID-19 infection (OR = 1.79, 95% CI = [1.66, 1.92]) and hospitalization (OR = 1.62, 95% CI = [1.39, 1.90]) than White diabetes patients. This significant higher odds for infection and hospitalization of African American patients pertains within the diabetes group with and with no complications. However, if these patients are hospitalized, their odds of in-hospital mortality is lower than White patients (OR = 0.76, 95% CI = [0.59, 0.99]). The statistical significance of this ratio persists in those with diabetic complications. Hispanic diabetes patients have higher odds of COVID-19 infection than non-Hispanic patients (OR = 2.86, 95% CI = [2.42, 3.38]). However, the mean length of stay for Hispanic patients is 2.46 (95% CI = [-4.56, -0.35]) days shorter than non-Hispanic patients, respectively. Specifically, this mean difference of COVID-19 hospital length of stay is significantly shorter in Hispanic than non-Hispanic patients within diabetes group with complications (OR = -4.82, 95% CI = [-7.98, -1.66]). Furthermore, the odds ratios for COVID-19 hospitalization (OR = 3.41, 95% CI = [2.75, 4.23]) and in-hospital mortality (OR = 3.69, 95% CI = [1.18, 11.50]) between Hispanic and non-hispanic patients with no diabetes complications are statistically significant.

The significance of the results in Table 4 can be affected by small sample sizes in the diabetes patient population with COVID-19 hospitalization. Therefore, we provide the sample sizes for these patients in the considered demographics groups in Table 5.

[Figure omitted. See PDF.]

Discussion

We first discuss our results regarding the association between diabetic complications and COVID-19 outcomes. This discussion is arranged by specific complication types. We then discuss demographic disparities among diabetes patients with complications and also among those without complications. In this discussion, diabetes patients with one or more complications are considered as a single group.

The association between diabetic complications and COVID-19 outcomes

We observe that diabetes patients have higher odds for COVID-19 infection compared to non-diabetes patients. As diabetes patients are at higher risk for severe COVID-19 outcomes [43–45], a possible explanation for this observation is that diabetes patients are more likely to seek care when they experience COVID-19-related symptoms. Furthermore, studies have shown that infectious diseases, respiratory tract infections in particular, occur more frequently and severely in diabetes patients [46–48]. In addition to studying the overall diabetes patient population, we have also analyzed subpopulations with specific complications. Our results show that among all diabetes patients, those with renal complications have the highest odds for COVID-19 infection compared to non-diabetes patients. Health systems should dedicate more resources to prevent the infection of diabetes patients with such complications in order to help alleviate the medical surge caused by an infectious disease outbreak.

Among patients infected with COVID-19, we find that diabetes patients have a higher risk of hospitalization compared to non-diabetes patients. Diabetes is one of the most common underlying medical conditions in hospitalized patients with COVID-19 [5, 49]. Furthermore, we observe diabetes patients with COVID-19 hospitalization are at higher risk of in-hospital mortality. This result is in line with several other studies that suggest 30 to 40 percent of all COVID-19 deaths have occurred among people with diabetes [50–52]. We observe that among all diabetes patients, those with metabolic complications have the highest risks of COVID-19 hospitalization and in-hospital mortality compared to non-diabetes patients. In aDCSI, only severe conditions, such as diabetes mellitus with ketoacidosis, diabetes mellitus with coma, and diabetes mellitus with hypoglycemic coma, are classified as metabolic complications. Therefore, regardless of their COVID-19 status, such patients have higher risk for hospitalization and mortality. In addition, studies have shown that COVID-19 infection increases the risk of diabetic ketoacidosis [24, 25, 53]. This indicates a bidirectional relationship between diabetes with metabolic complications and increased risk for COVID-19 hospitalization and mortality. Hence, diabetes patients with metabolic complications should be prioritized for intervention when diagnosed with COVID-19.

In general, diabetes patients have a longer COVID-19 hospital length of stay than non-diabetes patients. Among diabetes patients with at least one of the seven complications considered in this study, only those with cardiovascular or peripheral vascular complications had a significantly longer hospital length of stay than non-diabetes patients. One possible explanation for this result is that these patients require more diagnostic tests than non-diabetes patients. This result can provide hospital administrators information on how to schedule their staff based on their current diabetic patient population. For instance, if a hospital system has historically observed more diabetic patients with cardiovascular or peripheral vascular complications during specific periods throughout the year, they could plan to staff more providers during those times.

Demographic disparities among diabetes patients with complications

Male vs female.

In Table 4, male patients have higher odds for COVID-19 infection, hospitalization and in-hospital mortality than female patients in the general diabetes population. Although male patients with diabetes complications still have higher odds for COVID-19 in-hospital mortality, there is no statistically significant difference in COVID-19 infection and hospitalization between male and female patients. In general, males have higher risk for COVID-19 infection [54, 55], which can explain the significant differences observed in COVID-19 infection and hospitalization between male and female diabetes patients with no complications. However, when they have diabetes complications, male patients may be less frequently engaging in behaviors that generate high-risk for contracting COVID-19. This can explain the insignificant difference in the odds for COVID-19 infection and hospitalization between male and female patients with diabetes complications. Finally, the course of illness due to COVID-19 infection is known to be more severe in males leading to higher risks for worse outcomes and death [56]. Diabetic complications can exacerbate these risks [10], explaining our observation that male patients with diabetes complications have higher odds for COVID-19 in-hospital mortality than females. This result demonstrates the effect of diabetes complications on demographic disparities can vary with respect to COVID-19 outcome. More specifically, demographic disparities in one COVID-19 outcome might decrease while disparities in another outcome might increase at the same time in the presence of diabetic complications.

Senior vs younger adults.

In Table 4, senior adults have longer length of stay as well as higher odds for COVID-19 infection, hospitalization and in-hospital mortality compared to younger adults in the general diabetes population. The odds of seniors for COVID-19 in-hospital mortality remains significantly higher than younger adults in the diabetes group with no complications. This difference, however, is statistically less significant in the diabetes group with complications. One explanation for this result might be the small number of younger adults with diabetes complications and COVID-19 hospitalization in our data set (i.e., [53]). Furthermore, diabetes complications, which are associated with worse COVID-19 outcomes [10], might be predominantly determining the mortality risk of those patients resulting in less age specific differences. The hospital length of stay difference is no longer statistically significant between seniors and younger adults in both diabetes groups with and without complications. This result might again be due to the fact that our data include a small number of COVID-19 hospitalizations of the diabetic younger adults with complications (i.e., [53]) and without complications (i.e., [62]).

African American vs white.

We find that African American patients have higher odds for COVID-19 infection and hospitalization than White patients in the general diabetes population. However, whites have higher odds for COVID-19 in-hospital mortality than African Americans. The statistical significance of this difference persists in those with diabetic complications and disappears among the diabetic population with no complications. This significance change is potentially due to smaller sample size and lower mortality rate in the no complications group. Specifically, the total number of African American and white patients with no diabetes complications is approximately one third those with complications in our data set. Furthermore, the association between race and risk of COVID-19 mortality is inconsistent in the literature. Some studies considering patients in Georgia conclude that there is no difference in the risk of COVID-19 in-hospital mortality between African American and White patients [57, 58]. On the other hand, other studies found that African Americans have a higher risk of COVID-19 in-hospital mortality compared to White populations [59, 60] or vice versa [61, 62]. These diverse conclusions suggest that the racial disparities in COVID-19 mortality may manifest differently across hospital systems and patient characteristics.

The study population in Gupta et al [59] includes patients from a single hospital serving a majority Black population with high poverty rates, and the study population in Asch et al [60] includes Medicare beneficiaries across the U.S. Ogedegbe et al [61] and Buikema et al [62] do not focus on individuals with specific socio-economic or demographic characteristics, thus their study populations are more similar to ours. Ogedegbe et al [61] suggested that observing lower in-hospital COVID-19 mortality rates in African American patients despite well-established higher mortality rates of these patients in the general population could be related to out-of-hospital mortality because African American populations are more likely to be uninsured and underinsured than White populations and thus have poorer access to care. Buikema et al [62] suggested that the higher in-hospital COVID-19 mortality rates in African American patients underscores the need to promote earlier entry of racial/ethnic minorities into the healthcare system for COVID-19 preventive services and treatment.

Hispanic vs non-Hispanic.

In Table 4, the mean COVID-19 hospital length of stay of Hispanic diabetes patients is shorter than non-Hispanic diabetes patients. This result is in line with an analysis conducted by the Agency for Healthcare Research and Quality (AHRQ) in 2020 [63]. Interestingly, this disparity is magnified in the population with diabetes complications, whereas it is not statistically significant in the diabetes population without complications. A possible explanation for this result is that the Hispanic population in the US is more likely to be under-insured or uninsured [64]. Therefore, Hispanic patients have more limited access to the hospital care compared to non-Hispanic patients when longer hospitalizations are needed when diabetes complications exist. Although Hispanic patients with diabetes complications have higher odds for COVID-19 infection, there are no differences in the odds for COVID-19 hospitalization and in-hospital mortality between Hispanic and non-Hispanic patients with diabetes complications. Thus, diabetes complications, which are associated with worse COVID-19 outcomes [10], might be predominantly determining the COVID-19 severity in those infected patients resulting in less ethnic differences. There are significant differences in the odds for COVID-19 hospitalization and in-hospital mortality between Hispanic and non-Hispanic patients with no diabetes complications. Since the diabetes population with no complications is more similar to the general population, this result is driven by underlying health disparities between Hispanic and non-Hispanic populations [65–67]. Note that these differences in the odds for COVID-19 hospitalization and in-hospital mortality between Hispanic and non-Hispanic patients are not observed in the general diabetes population in our data.

Conclusion and limitations

This study contributes to the existing literature on understanding risk factors for COVID-19 related to underlying medical conditions and demographic characteristics. Specifically, we demonstrate that the presence of diabetes complications increases the risks of COVID-19 infection, hospitalization, and worse health outcomes with respect to in-hospital mortality and longer hospital length of stay. Furthermore, we show the presence of health disparities in COVID-19 outcomes across demographic groups in our diabetes population.

There are some limitations to our work that can be considered in future research. First, our data does not include a primary diagnosis code for each encounter. Thus, the primary reason for a COVID-19 infection or hospitalization encounter could be due to COVID-19 or other medical conditions which can obscure the definition of these two outcomes. Second, our results may not be generalizable to other patient populations since our dataset comes from a single hospital system. Future studies are needed in order to compare results from different hospital systems. Another limitation is that we did not account for time since diagnosis of diabetes in our analysis. We utilized a patients’ medical histories to identify diabetes diagnosis but were unable to include the time at which this diagnosis was first made. By collecting more data future research could investigate the impact of time since diagnosis of diabetes on different COVID-19 outcomes. Finally, we have applied logistic and linear regression to estimate the odds ratio and mean differences of COVID-19 outcomes among patients with different diabetic complications. Another interesting area of future research would be to measure the importance of these diabetic complications as independent variables in prediction models. For instance, future work could utilize binary indicators of diabetes complications as variables in a decision tree or random forest model. Then, mean impurity decrease or permutation feature importance could be used to evaluate the importance of diabetic complications for predicting COVID-19 outcomes. This analysis however should incorporate additional system and population level factors, e.g., ventilator availability and vaccine coverage.

Acknowledgments

The authors thank the team at the National Center for Human Factors in Healthcare at MedStar Health for their assistance in retrieving and interpretation of MedStar’s EHR data.

Citation: Paramita NLPSP, Agor JK, Mayorga ME, Ivy JS, Miller KE, Ozaltin OY (2023) Quantifying association and disparities between diabetes complications and COVID-19 outcomes: A retrospective study using electronic health records. PLoS ONE 18(9): e0286815. https://doi.org/10.1371/journal.pone.0286815

About the Authors:

Ni Luh Putu S. P. Paramita

Roles: Conceptualization, Formal analysis, Methodology, Visualization, Writing – original draft, Writing – review & editing

Affiliation: Operations Research Graduate Program, North Carolina State University, Raleigh, North Carolina, United States of America

Joseph K. Agor

Roles: Conceptualization, Investigation, Methodology, Supervision, Writing – original draft, Writing – review & editing

E-mail: [email protected]

Affiliation: School of Mechanical, Industrial, and Manufacturing Engineering, Oregon State University, Corvallis, Oregon, United States of America

ORICD: https://orcid.org/0000-0002-9380-4256

Maria E. Mayorga

Roles: Conceptualization, Methodology, Supervision, Writing – review & editing

Affiliation: Edward P. Fitts Department of Industrial and Systems Engineering, Raleigh, North Carolina, United States of America

Julie S. Ivy

Roles: Conceptualization, Methodology, Supervision, Writing – review & editing

Affiliation: Edward P. Fitts Department of Industrial and Systems Engineering, Raleigh, North Carolina, United States of America

Kristen E. Miller

Roles: Conceptualization, Project administration, Supervision, Writing – review & editing

Affiliation: National Center for Human Factor in Healthcare, MedStar Health, Washington, District of Columbia, United States of America

Osman Y. Ozaltin

Roles: Conceptualization, Investigation, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing

Affiliation: Edward P. Fitts Department of Industrial and Systems Engineering, Raleigh, North Carolina, United States of America

ORICD: https://orcid.org/0000-0002-0093-5645

References

1. Kerbl R, Zepp F. Coronavirus disease 2019. Monatsschr Kinderheilkd. 2021;169(4):308–11.

2. Jehi L, Ji X, Milinovich A, Erzurum S, Rubin BP, Gordon S, et al. Individualizing Risk Prediction for Positive Coronavirus Disease 2019 Testing: Results From 11,672 Patients. Chest [Internet]. 2020 Oct 1;158(4):1364–75. Available from: pmid:32533957

3. Williamson EJ, Walker AJ, Bhaskaran K, Bacon S, Bates C, Morton CE, et al. Factors associated with COVID-19-related death using OpenSAFELY. Nature [Internet]. 2020;584(7821):430–6. Available from: pmid:32640463

4. Taquet M, Geddes JR, Husain M, Luciano S, Harrison PJ. 6-month neurological and psychiatric outcomes in 236 379 survivors of COVID-19: a retrospective cohort study using electronic health records. The Lancet Psychiatry [Internet]. 2021 May 1;8(5):416–27. Available from: https://doi.org/10.1016/S2215-0366(21)00084-5

5. Richardson S, Hirsch JS, Narasimhan M, Crawford JM, McGinn T, Davidson KW, et al. Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the New York City Area. JAMA [Internet]. 2020;323(20):2052–9. Available from: pmid:32320003

6. Guan WJ, Liang WH, Zhao Y, Liang HR, Chen ZS, Li YM, et al. Comorbidity and its impact on 1590 patients with COVID-19 in China: a nationwide analysis. Eur Respir J. 2020 May;55(5).

7. Zhu L, She ZG, Cheng X, Qin JJ, Zhang XJ, Cai J, et al. Association of Blood Glucose Control and Outcomes in Patients with COVID-19 and Pre-existing Type 2 Diabetes. Cell Metab [Internet]. 2020 Jun 2;31(6):1068–1077.e3. Available from: pmid:32369736

8. Barron E, Bakhai C, Kar P, Weaver A, Bradley D, Ismail H, et al. Associations of type 1 and type 2 diabetes with COVID-19-related mortality in England: a whole-population study. Lancet Diabetes Endocrinol [Internet]. 2020 Oct 1;8(10):813–22. Available from: pmid:32798472

9. Gregory JM, Slaughter JC, Duffus SH, Smith TJ, LeStourgeon LM, Jaser SS, et al. COVID-19 Severity Is Tripled in the Diabetes Community: A Prospective Analysis of the Pandemic’s Impact in Type 1 and Type 2 Diabetes. Diabetes Care. 2021 Feb;44(2):526–32. pmid:33268335

10. Holman N, Knighton P, Kar P, O’Keefe J, Curley M, Weaver A, et al. Risk factors for COVID-19-related mortality in people with type 1 and type 2 diabetes in England: a population-based cohort study. Lancet Diabetes Endocrinol [Internet]. 2020 Oct 1;8(10):823–33. Available from: pmid:32798471

11. Lim S, Bae JH, Kwon HS, Nauck MA. COVID-19 and diabetes mellitus: from pathophysiology to clinical management. Nat Rev Endocrinol [Internet]. 2021;17(1):11–30. Available from: pmid:33188364

12. Zhou Y, Chi J, Lv W, Wang Y. Obesity and diabetes as high-risk factors for severe coronavirus disease 2019 (COVID-19). Diabetes Metab Res Rev. 2021 Feb;37(2):e3377. pmid:32588943

13. Guo W, Li M, Dong Y, Zhou H, Zhang Z, Tian C, et al. Diabetes is a risk factor for the progression and prognosis of COVID-19. Diabetes Metab Res Rev. 2020 Mar;36(7):e3319. pmid:32233013

14. Wargny M, Potier L, Gourdy P, Pichelin M, Amadou C, Benhamou PY, et al. Predictors of hospital discharge and mortality in patients with diabetes and COVID-19: updated results from the nationwide CORONADO study. Diabetologia [Internet]. 2021;64(4):778–94. Available from: pmid:33599800

15. Fowler MJ. Microvascular and Macrovascular Complications of Diabetes. Clin Diabetes [Internet]. 2008 Apr 1;26(2):77–82. Available from: https://doi.org/10.2337/diaclin.26.2.77

16. Chang HY, Weiner JP, Richards TM, Bleich SN, Segal JB. Validating the adapted Diabetes Complications Severity Index in claims data. Am J Manag Care. 2012 Nov;18(11):721–6. pmid:23198714

17. Wicke FS, Glushan A, Schubert I, Köster I, Lübeck R, Hammer M, et al. Performance of the adapted Diabetes Complications Severity Index translated to ICD-10. Am J Manag Care. 2019 Feb;25(2):e45–9. pmid:30763043

18. Unnikrishnan R, Misra A. Diabetes and COVID19: a bidirectional relationship. Nutr Diabetes. 2021 Jun;11(1):21. pmid:34168110

19. Sardu C, D’Onofrio N, Balestrieri ML, Barbieri M, Rizzo MR, Messina V, et al. Hyperglycaemia on admission to hospital and COVID-19. Diabetologia [Internet]. 2020;63(11):2486–7. Available from: pmid:32632527

20. Yan Y, Yang Y, Wang F, Ren H, Zhang S, Shi X, et al. Clinical characteristics and outcomes of patients with severe COVID-19 with diabetes. BMJ open diabetes Res care. 2020 Apr;8(1).

21. Gianchandani R, Esfandiari NH, Ang L, Iyengar J, Knotts S, Choksi P, et al. Managing Hyperglycemia in the COVID-19 Inflammatory Storm. Diabetes. 2020 Oct;69(10):2048–53. pmid:32778570

22. Ceriello A. Hyperglycemia and COVID-19: What was known and what is really new? Vol. 167, Diabetes research and clinical practice. Ireland; 2020. p. 108383.

23. Calvisi SL, Ramirez GA, Scavini M, Da Prat V, Di Lucca G, Laurenzi A, et al. Thromboembolism risk among patients with diabetes/stress hyperglycemia and COVID-19. Metabolism. 2021 Oct;123:154845. pmid:34364927

24. de Sá-Ferreira CO, da Costa CHM, Guimarães JCW, Sampaio NS, Silva L de ML, de Mascarenhas LP, et al. Diabetic ketoacidosis and COVID-19: what have we learned so far? Am J Physiol Metab [Internet]. 2021 Nov 15;322(1):E44–53. Available from: pmid:34779657

25. Reddy PK, Kuchay MS, Mehta Y, Mishra SK. Diabetic ketoacidosis precipitated by COVID-19: A report of two cases and review of literature. Diabetes Metab Syndr. 2020;14(5):1459–62. pmid:32771918

26. Rahmati M, Keshvari M, Mirnasuri S, Yon DK, Lee SW, Il Shin J, et al. The global impact of COVID-19 pandemic on the incidence of pediatric new-onset type 1 diabetes and ketoacidosis: A systematic review and meta-analysis. J Med Virol [Internet]. 2022 Nov 1;94(11):5112–27. Available from: pmid:35831242

27. Elgenidy A, Awad AK, Saad K, Atef M, El-Leithy HH, Obiedallah AA, et al. Incidence of diabetic ketoacidosis during COVID-19 pandemic: a meta-analysis of 124,597 children with diabetes. Pediatr Res [Internet]. 2022; Available from: https://doi.org/10.1038/s41390-022-02241-2

28. Pasquel FJ, Messler J, Booth R, Kubacka B, Mumpower A, Umpierrez G, et al. Characteristics of and Mortality Associated With Diabetic Ketoacidosis Among US Patients Hospitalized With or Without COVID-19. JAMA Netw Open [Internet]. 2021 Mar 10;4(3):e211091–e211091. Available from: pmid:33688962

29. Moin ASM, Al-Qaissi A, Sathyapalan T, Atkin SL, Butler AE. Type 2 Diabetes Coagulopathy Proteins May Conflict With Biomarkers Reflective of COVID-19 Severity [Internet]. Vol. 12, Frontiers in Endocrinology. 2021. Available from: https://www.frontiersin.org/articles/10.3389/fendo.2021.658304

30. Peterson E, Lo KB, DeJoy R, Salacup G, Pelayo J, Bhargav R, et al. The relationship between coronary artery disease and clinical outcomes in COVID-19: a single-center retrospective analysis. Coron Artery Dis [Internet]. 2021;32(5). Available from: https://journals.lww.com/coronary-artery/Fulltext/2021/08000/The_relationship_between_coronary_artery_disease.2.aspx pmid:32732512

31. Rastogi A, Dogra H, Jude EB. COVID-19 and peripheral arterial complications in people with diabetes and hypertension: A systematic review. Diabetes Metab Syndr. 2021;15(5):102204. pmid:34303918

32. Al-kuraishy HM, Al-Gareeb AI, Alblihed M, Cruz-Martins N, Batiha GES. COVID-19 and Risk of Acute Ischemic Stroke and Acute Lung Injury in Patients With Type II Diabetes Mellitus: The Anti-inflammatory Role of Metformin [Internet]. Vol. 8, Frontiers in Medicine. 2021. Available from: https://www.frontiersin.org/articles/10.3389/fmed.2021.644295

33. Mourad D, Azar NS, Azar ST. Diabetic Nephropathy and COVID-19: The Potential Role of Immune Actors. Int J Mol Sci. 2021 Jul;22(15). pmid:34360529

34. Basra R, Whyte M, Karalliedde J, Vas P. What is the impact of microvascular complications of diabetes on severe COVID-19? Microvasc Res [Internet]. 2022;140:104310. Available from: https://www.sciencedirect.com/science/article/pii/S0026286221001801 pmid:34979154

35. Tuan WJ, Lennon RP, Zhang A, Macherla A, Zgierska AE. Risks of Severe COVID-19 Outcomes Among Patients With Diabetic Polyneuropathy in the United States. J Public Heal Manag Pract [Internet]. 2022;28(6). Available from: https://journals.lww.com/jphmp/Fulltext/2022/11000/Risks_of_Severe_COVID_19_Outcomes_Among_Patients.13.aspx pmid:36037512

36. Corcillo A, Cohen S, Li A, Crane J, Kariyawasam D, Karalliedde J. Diabetic retinopathy is independently associated with increased risk of intubation: A single centre cohort study of patients with diabetes hospitalised with COVID-19. Diabetes Res Clin Pract [Internet]. 2021;171:108529. Available from: https://www.sciencedirect.com/science/article/pii/S0168822720307865 pmid:33152348

37. Greenland S. Model-based Estimation of Relative Risks and Other Epidemiologic Measures in Studies of Common Outcomes and in Case-Control Studies. Am J Epidemiol [Internet]. 2004 Aug 15;160(4):301–5. Available from: pmid:15286014

38. Cummings P. The Relative Merits of Risk Ratios and Odds Ratios. Arch Pediatr Adolesc Med [Internet]. 2009 May 4;163(5):438–45. Available from: pmid:19414690

39. O’Keefe GE, Jurkovich GJ, Maier R V. Defining excess resource utilization and identifying associated factors for trauma victims. J Trauma. 1999 Mar;46(3):473–8. pmid:10088854

40. Englert J, Davis KM, Koch KE. Using Clinical Practice Analysis to Improve Care. Jt Comm J Qual Improv [Internet]. 2001;27(6):291–301. Available from: https://www.sciencedirect.com/science/article/pii/S1070324101270254 pmid:11402776

41. Agresti A. Categorical Data Analysis. 3rd ed. Hoboken: John Wiley & Sons Inc; 2013.

42. Bland M. An Introduction to Medical Statistics. 4th ed. Oxford: Oxford University Press; 2015.

43. Fadini GP, Morieri ML, Boscari F, Fioretto P, Maran A, Busetto L, et al. Newly-diagnosed diabetes and admission hyperglycemia predict COVID-19 severity by aggravating respiratory deterioration. Diabetes Res Clin Pract. 2020 Oct;168:108374. pmid:32805345

44. Cardona-Hernandez R, Cherubini V, Iafusco D, Schiaffini R, Luo X, Maahs DM. Children and youth with diabetes are not at increased risk for hospitalization due to COVID-19. Pediatr Diabetes [Internet]. 2021;22(2):202–6. Available from: https://onlinelibrary.wiley.com/doi/abs/10.1111/pedi.13158 pmid:33205546

45. Caballero AE, Ceriello A, Misra A, Aschner P, McDonnell ME, Hassanein M, et al. COVID-19 in people living with diabetes: An international consensus. J Diabetes Complications. 2020 Sep 1;34(9). pmid:32651031

46. Muller LMAJ Gorter KJ, Hak E Goudzwaard WL, Schellevis FG Hoepelman AIM, et al. Increased risk of common infections in patients with type 1 and type 2 diabetes mellitus. Clin Infect Dis an Off Publ Infect Dis Soc Am. 2005 Aug;41(3):281–8. pmid:16007521

47. Peleg AY, Weerarathna T, McCarthy JS, Davis TME. Common infections in diabetes: pathogenesis, management and relationship to glycaemic control. Diabetes Metab Res Rev. 2007 Jan;23(1):3–13. pmid:16960917

48. Casqueiro J, Casqueiro J, Alves C. Infections in patients with diabetes mellitus: A review of pathogenesis. Indian J Endocrinol Metab [Internet]. 2012 Mar;16 Suppl 1(Suppl1):S27–36. Available from: https://pubmed.ncbi.nlm.nih.gov/22701840 pmid:22701840

49. Kompaniyets L, Pennington A, Goodman A, Rosenblum H, Belay B, Ko J, et al. Underlying Medical Conditions and Severe Illness Among 540,667 Adults Hospitalized With COVID-19, March 2020–March 2021. Prev Chronic Dis. 2021;18. pmid:34197283

50. Mahamat-Saleh Y, Fiolet T, Rebeaud ME, Mulot M, Guihur A, El Fatouhi D, et al. Diabetes, hypertension, body mass index, smoking and COVID-19-related mortality: a systematic review and meta-analysis of observational studies. BMJ Open. 2021 Oct;11(10):e052777. pmid:34697120

51. Sourij H, Aziz F, Bräuer A, Ciardi C, Clodi M, Fasching P, et al. COVID-19 fatality prediction in people with diabetes and prediabetes using a simple score upon hospital admission. Diabetes, Obes Metab [Internet]. 2021 Feb 1;23(2):589–98. Available from: https://doi.org/10.1111/dom.14256 pmid:33200501

52. Wortham J, Lee J, Althomsons S, Al. E. Characteristics of Persons Who Died with COVID-19—United States, February 12–May 18, 2020. MMWR Morb Mortal Wkly Rep 2020. (69):923–9.

53. Cuschieri S, Grech S. COVID-19 and diabetes: The why, the what and the how. J Diabetes Complications. 2020 Sep 1;34(9). pmid:32456846

54. Sharma G, Volgman AS, Michos ED. Sex Differences in Mortality From COVID-19 Pandemic: Are Men Vulnerable and Women Protected? Vol. 2, JACC. Case reports. Netherlands; 2020. p. 1407–10. pmid:32373791

55. Baker P, White A, Morgan R. Men’s health: COVID-19 pandemic highlights need for overdue policy action. Lancet (London, England). 2020 Jun;395(10241):1886–8. pmid:32563357

56. Jin JM, Bai P, He W, Wu F, Liu XF, Han DM, et al. Gender Differences in Patients With COVID-19: Focus on Severity and Mortality [Internet]. Vol. 8, Frontiers in Public Health. 2020. Available from: https://www.frontiersin.org/articles/10.3389/fpubh.2020.00152

57. Marmarchi F, Liu M, Rangaraju S, Auld SC, Creel-Bulos MC, Kempton CL, et al. Clinical Outcomes of Critically III Patients with COVID-19 by Race. J Racial Ethn Heal Disparities [Internet]. 2022;9(2):385–9. Available from: pmid:33469873

58. McCain JL, Wang X, Connell K, Morgan J. Assessing the impact of insurance type on COVID-19 mortality in black and white patients in the largest healthcare system in the state of georgia. J Natl Med Assoc [Internet]. 2022;114(2):218–26. Available from: https://www.sciencedirect.com/science/article/pii/S0027968422000049 pmid:35090747

59. Gupta R, Agrawal R, Bukhari Z, Jabbar A, Wang D, Diks J, et al. Higher comorbidities and early death in hospitalized African-American patients with COVID-19. BMC Infect Dis [Internet]. 2021;21(1):78. Available from: pmid:33461499

60. Asch DA, Islam MN, Sheils NE, Chen Y, Doshi JA, Buresh J, et al. Patient and Hospital Factors Associated With Differences in Mortality Rates Among Black and White US Medicare Beneficiaries Hospitalized With COVID-19 Infection. JAMA Netw Open [Internet]. 2021 Jun 17;4(6):e2112842–e2112842. Available from: pmid:34137829

61. Ogedegbe G, Ravenell J, Adhikari S, Butler M, Cook T, Francois F, et al. Assessment of Racial/Ethnic Disparities in Hospitalization and Mortality in Patients With COVID-19 in New York City. JAMA Netw Open [Internet]. 2020 Dec 4;3(12):e2026881–e2026881. Available from: pmid:33275153

62. Buikema AR, Buzinec P, Paudel ML, Andrade K, Johnson JC, Edmonds YM, et al. Racial and ethnic disparity in clinical outcomes among patients with confirmed COVID-19 infection in a large US electronic health record database. EClinicalMedicine. 2021 Sep;39:101075. pmid:34493997

63. Owens PL. COVID-19-Related Hospitalizations in Nine States, by Race/Ethnicity, 2020: Statistical Brief #272. In Rockville (MD); 2006.

64. Lee J, Hale N. Evidence and Implications of the Affordable Care Act for Racial/Ethnic Disparities in Diabetes Health During and Beyond the Pandemic. Popul Health Manag [Internet]. 2022 Apr 1;25(2):235–43. Available from: pmid:35442797

65. Adegunsoye A, Ventura IB, Liarski VM. Association of Black Race with Outcomes in COVID-19 Disease: A Retrospective Cohort Study. Ann Am Thorac Soc [Internet]. 2020 Jul 9;17(10):1336–9. Available from: pmid:32643398

66. Bandi S, Nevid MZ, Mahdavinia M. African American children are at higher risk of COVID-19 infection. Pediatr Allergy Immunol [Internet]. 2020 Oct 1;31(7):861–4. Available from: pmid:32469426

67. Martinez DA, Hinson JS, Klein EY, Irvin NA, Saheed M, Page KR, et al. SARS-CoV-2 Positivity Rate for Latinos in the Baltimore–Washington, DC Region. JAMA [Internet]. 2020 Jul 28;324(4):392–5. Available from: pmid:32556212