Diabetes Associated With Higher Health Care Utilization and Poor Outcomes After COPD-Related Hospitalizations

, , , , ,
The American Journal of Managed Care, September 2022, Volume 28, Issue 9

Patients with diabetes and chronic obstructive pulmonary disease (COPD) have worse outcomes when hospitalized and appear to be more vulnerable to respiratory and nonrespiratory complications after a COPD exacerbation, which highlights the need for targeted interventions in this population.

ABSTRACT

Objectives: Readmissions after hospitalizations for acute exacerbation of chronic obstructive pulmonary disease (COPD) have a high socioeconomic burden. Comorbidities such as diabetes increase the risk for hospital readmissions, but the impact of diabetes on hospital outcomes remains unknown. The aim of this study was to evaluate the effect of complicated or uncomplicated diabetes on outcomes and health care costs related to admissions and readmissions in patients 35 years and older with an index admission for COPD.

Study Design: This was a retrospective longitudinal data analysis. We analyzed data from the Healthcare Cost and Utilization Project (HCUP) Nationwide Readmissions Database.

Methods: We analyzed the 2012-2015 HCUP Nationwide Readmissions Database and used multivariable weighted regression analyses to adjust for confounding factors. Individuals with any chronic pulmonary disease other than COPD were excluded.

Results: Of 1,728,931 patients hospitalized for COPD, 522,020 (30.2%) had a diagnosis of diabetes. Risk of all-cause 30-day readmission was higher among patients with complicated diabetes (adjusted odds ratio [OR], 1.15; 95% CI, 1.11-1.18) and uncomplicated diabetes (adjusted OR, 1.10; 95% CI, 1.08-1.12) compared with patients without diabetes. Diabetes was associated with longer length of stay, higher rates of hospital complications during index hospitalizations and 30-day readmissions, and a higher health care cost. Although diabetes was not associated with higher hospital mortality, routine hospital discharges were less common and the need for home health care upon discharge was higher among those with diabetes.

Conclusions: Patients hospitalized for COPD and coexisting diabetes have worse clinical outcomes and higher 30-day readmissions compared with patients hospitalized for COPD without diabetes. Optimizing medical therapies and targeted interventions for both diseases is needed to alleviate disease burden to individuals and to society.

Am J Manag Care. 2022;28(9):e325-e332

_____

Takeaway Points

Chronic obstructive pulmonary disease (COPD) hospitalizations, including readmissions, have a high socioeconomic burden on patients and society. As a subgroup, patients with diabetes who are admitted for COPD exacerbations have longer hospital stays, higher costs, and worse outcomes. They are more likely to have serious breathing problems and higher rates of kidney failure, infections, confusion, and strokes. They are more likely to go to long-term or short-term care centers and to require assistance at home. Additionally, patients with diabetes who are admitted for COPD are more likely to come back to the hospital within 30 days for respiratory and nonrespiratory complications. Information in this article may help with policy and management of patients with diabetes and COPD.

  • Patients with diabetes and COPD should be recognized as being uniquely vulnerable to both respiratory issues and complications of other organs such as the kidneys and the brain.
  • There is a significant increase in cost per hospitalization and length of stay for this group.
  • Postacute care costs and resource utilization are also significantly higher, which is an area of future study.

_____

Chronic obstructive pulmonary disease (COPD) is a leading cause of mortality in the United States.1 US costs attributable to COPD were estimated at $32.1 billion in 2010 and were expected to reach $49 billion by 2020.2 Acute exacerbations of COPD (AECOPD) account for 70% of COPD-related health care costs.3 Comorbid conditions are associated with increased mortality, readmission, and health care utilization.3,4

Comorbidities are more prevalent in patients with COPD compared with age-matched nonsmokers or smokers without COPD.5 Diabetes was shown to be present in 10% of patients with COPD in a large multinational cohort.5 Diabetes has been independently associated with reduced lung function even in nonsmokers.6 Conversely, COPD has been implicated in the development of diabetes,7 possibly due to chronic systemic inflammation and insulin resistance.8

In 2014, the Hospital Readmissions Reduction Program (HRRP) was instituted to incentivize COPD care bundles to reduce readmissions and improve health care outcomes. HRRP links payment to the quality of hospital care and reduces payments to hospitals for excess readmissions.2 However, more than 8 years after the HRRP was introduced, there remains ambiguity regarding best practices to reduce COPD readmission.9 A 2019 American Thoracic Society workshop noted that COPD readmissions are more common in individuals with comorbidities.9 Interventions to reduce COPD readmissions must take into account comorbidities, as demonstrated by the results of a study of Medicare claims in which more than 70% of COPD-related readmissions were associated with conditions other than COPD.9,10 Moreover, although prior studies have identified diabetes as a predictor of COPD-related readmissions, the socioeconomic and health care burden of comorbid diabetes with COPD has not been determined.11-13

Given these findings, we hypothesized that diabetes is associated with an increased probability of poor outcomes and higher cost during the index COPD admission as well as COPD-related and all-cause rehospitalizations. Further, we hypothesized that the impact of diabetes on 30-day COPD readmission risk and hospitalization outcomes would demonstrate the need for targeted interventions that lower disease burden at both the individual and societal levels.

METHODS

Study Design

We analyzed the 2012-2015 Healthcare Cost and Utilization Project Nationwide Readmissions Database (HCUP-NRD). Sponsored by the Agency for Healthcare Research and Quality (AHRQ), the HCUP-NRD contains administrative data with encounter-level information from approximately 18 million discharges per year and estimates roughly 36 million discharges nationwide after weighting.14,15 Our methods have been previously described.16 The requisite data user agreement was signed with AHRQ.

Study Population

Our analysis included all adult individuals (≥ 35 years) with a principal diagnosis of chronic obstructive bronchitis (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] code 491.x and International Classification of Diseases, Tenth Revision, Clinical Modification [ICD-10-CM] code J44.x in the fourth quarter of 2015) or emphysema (ICD-9-CM code 492.x and ICD-10-CM code J43.x in the fourth quarter of 2015). We also included patients with a principal diagnosis of respiratory failure (HCUP Clinical Classification Software [CCS] code 13117) and secondary diagnosis of chronic obstructive bronchitis or emphysema. We limited our analyses to patients 35 years and older because the diagnosis of COPD is less likely in younger patients.18,19 We defined the index admission as the first admission during a particular year with these diagnoses. Thirty-day readmissions were included if they were linked to the index admission. We excluded individuals with any chronic pulmonary disease other than COPD (eAppendix Table 1 [eAppendix available at ajmc.com]) and those with dual diagnoses of COPD and asthma. To analyze 30-day readmissions, we excluded patients who died during the index admission or were discharged in December.

Measurements

HCUP-NRD includes information on patient demographics (age and sex), socioeconomic status (estimated household income, primary insurance type), medical conditions (ICD-9-CM diagnoses, medical comorbidities, and severity of illness codes at admission), and length of stay (LOS). The total admission cost is estimated from admission charges using HCUP Cost-to-Charge Ratio Files.20 Costs were then inflated to 2015 US$ using the gross domestic product (GDP) price index.21

Outcome Measures

Patients with diabetes who were hospitalized for COPD were compared with those without diabetes for the following outcome variables: hospital LOS, hospital admission cost, in-hospital mortality rate, acute complications during both index admission and readmissions, and 30-day all-cause and COPD-related rehospitalizations. Patients with diabetes were further divided into having diabetes without chronic complication (CCS code 49) or with chronic complications (CCS code 50). The acute complications during hospitalization included respiratory failure, sepsis, shock, acute kidney injury (AKI), delirium, encephalopathy, and stroke (eAppendix Table 2).

Statistical Analysis

Our statistical analysis methods have been previously described.16 We used analytic sample weights (discharge-level weight) to generate national estimates according to published HCUP-NRD guidelines.17,20 Data are presented as weighted median (IQR) for continuous variables and weighted percent (SE) for categorical variables. Patients with diabetes were compared with those without diabetes using t tests or analysis of variance for normally distributed continuous variables. The Wilcoxon rank sum test or the Kruskal-Wallis test were used when normality assumptions were not met. Categorical variables were compared using a χ2 test. The impact of diabetes on categorical outcomes was determined using weighted logistic regression and are presented as weighted odds ratios (ORs) with 95% CIs. LOS and cost were analyzed using linear regression with log transformation, and the exponential of the slope parameter and corresponding 95% CI are presented. All models were adjusted for demographics, socioeconomic features, comorbidities, and hospital characteristics. Additionally, models evaluating readmission cost and readmission hospital LOS were also adjusted for baseline cost and LOS during the index admission. Missing data were not imputed, given the low percentage of missing values (eAppendix Table 3), and therefore were excluded from the analysis. Analyses were done using the SURVEY package in R 4.0.0 (R Foundation for Statistical Computing). All tests were 2-tailed and performed at a significance level of 0.05.

RESULTS

Index Admission

Of the 1,728,931 patients hospitalized for COPD who met inclusion criteria, 522,020 (30.2%) also carried the diagnosis of diabetes, including 442,610 (25.6%) with uncomplicated diabetes and 79,401 (4.6%) with complicated diabetes (Table 1). The median (IQR) age was 69 (60-77) years, and 54.7% were female. Overall, the demographics and clinical characteristics varied considerably between patients with and without diabetes. Patients with diabetes were more likely to be female and had more chronic medical conditions, reflected by a higher number of diagnoses at discharge and higher diagnosis-related group severity and mortality risk scores (P < .001).

The hospital LOS and cost for the index admission were higher among patients with diabetes. Overall, patients admitted for COPD had a weighted median (IQR) LOS of 3 (2-4) days, with a weighted median (IQR) cost of $5343 ($3501-$8414). Weighted hospital LOS and cost were higher among patients with diabetes and highest among those with complicated diabetes (Table 2). Those differences remained significant after adjustment for patient demographics, socioeconomic status, and chronic medical comorbidities. Hospital LOS was 7% longer in patients with complicated diabetes than those without diabetes (adjusted exponential of slope, 1.07; 95% CI, 1.06-1.08). Similarly, complicated diabetes was associated with an 11% increase in hospital cost (adjusted exponential of slope, 1.11; 95% CI, 1.13-1.16) (Table 3).

For the index hospitalization, diabetes was associated with higher likelihood of hospital complications, although risk for mortality was not increased. Respiratory failure (18.1%), AKI (8.5%), and encephalopathy (0.9%) were the most common complications of the index hospitalization. They were more common among patients with uncomplicated diabetes and most common among those with complicated diabetes (Table 4). The in-hospital mortality rate for the entire cohort was 2% (Table 2). In contrast to the higher probability of hospital complications among those with diabetes, the adjusted probability of hospital mortality was 5.3% lower among patients with uncomplicated diabetes (adjusted OR, 0.947; 95% CI, 0.903-0.993) and 9.9% lower among those with complicated diabetes (adjusted OR, 0.901; 95% CI, 0.819-0.990) (Table 3). Routine hospital discharges were 3.2% less common among patients with uncomplicated diabetes and 10.6% less common among those with complicated diabetes (Table 2). In parallel, discharges to skilled nursing facilities and intermediate care facilities were significantly higher among patients with uncomplicated diabetes (12.5%) and those with complicated diabetes (15.1%) compared with patients without diabetes (11.3%) (P < .001). Similarly, there was a higher need for home health care upon discharge among patients with uncomplicated diabetes (18.1%) and those with complicated diabetes (21.5%) compared with those without diabetes (15.7%) (P < .001) (Table 2).

Hospital 30-Day Readmissions

Patients with an index admission of COPD were at significant risk for 30-day hospital readmission. Among the entire cohort of 1,728,931 patients, 272,282 (15.7%) were readmitted within 30 days of hospital discharge, including 77,444 (4.5%) readmitted for COPD (Table 2). The all-cause 30-day readmission rate was higher among patients with complicated diabetes (19.8%) than those with uncomplicated diabetes (17.3%) and patients without diabetes (14.9%) (P < .001). Even after adjustment for patient demographics, socioeconomic status, and chronic medical comorbidities, the likelihood of all-cause readmission was 15% higher in patients with complicated diabetes (adjusted OR, 1.15; 95% CI, 1.11-1.18) and 10% higher in those with uncomplicated diabetes (adjusted OR, 1.10; 95% CI, 1.08-1.12) compared with patients without diabetes. Interestingly, 30-day COPD-related readmissions were less likely to be associated with complicated diabetes (adjusted OR, 0.91; 95% CI, 0.85-0.98), but were not less likely to be associated with uncomplicated diabetes (adjusted OR, 0.99; 95% CI, 0.96-1.02), compared with those without diabetes (Table 2). The median (IQR) time to the first readmission from discharge was 12 (6-20) days for all-cause and 13 (6-21) days for COPD-related 30-day readmissions. Compared with the outcomes of the index admission, hospital LOS, hospital cost, and in-hospital mortality were significantly higher in both all-cause and COPD-related readmissions (P < .001) (Table 2).

Although readmission LOS was not significantly higher in patients with diabetes than in those without diabetes, readmission cost was on average $413 higher in patients with uncomplicated diabetes and $1080 in those with complicated diabetes compared with those without diabetes. Similar to findings from the index admission, diabetes was not associated with higher mortality. In fact, in-hospital mortality from all-cause 30-day readmission was 8.6% lower in patients with uncomplicated diabetes and 18.3% lower in those with complicated diabetes compared with those without diabetes (Table 2 and Table 3). Those differences continued to be significant after adjustment for patient demographics, socioeconomic status, and comorbidities (Table 3). In contrast, in-hospital mortality from 30-day COPD-related readmission was not influenced by diabetes (Table 2 and Table 3).

More severe outcomes were also demonstrated in patients with poor socioeconomic status. Multivariate analysis showed that having Medicaid as a payer is associated with higher risk of all-cause 30-day readmission (adjusted OR, 1.58; 95% CI, 1.52-1.64) and COPD-related 30-day readmission (adjusted OR, 1.59; 95% CI, 1.50-1.69) and higher hospital cost (adjusted OR, 1.01; 95% CI, 1.00-1.02) compared with having private insurance. Similarly, individuals in the lowest quartile of household income ($1-$41,999 for 2015) were also at a higher risk for all-cause 30-day readmission (adjusted OR, 1.06; 95% CI, 1.04-1.09) and COPD-related 30-day readmission (adjusted OR, 1.11; 95% CI, 1.06-1.16) compared with individuals in the highest quartile (> $68,000 for 2015).

Compared with the index admission, the rate of respiratory failure was significantly higher in COPD-related readmissions (20.7% vs 18.1%) but not in all-cause readmissions (0.3% lower). That risk was highest among those with complicated diabetes (adjusted OR, 1.31; 95% CI, 1.45-1.51) compared with those without diabetes. However, neither uncomplicated diabetes nor complicated diabetes were significantly associated with a higher likelihood of respiratory failure during all-cause readmission (P = .07) (Table 4). Compared with respiratory complications, diabetes was associated with higher rates of nonrespiratory complications during both 30-day all-cause and 30-day COPD-related readmission (Table 4). For example, the probability of AKI was significantly higher in patients with diabetes (both uncomplicated and complicated combined) during all-cause readmission (adjusted OR, 1.30; 95% CI, 1.25-1.36) and COPD-related readmissions (adjusted OR, 1.35; 95% CI, 1.22-1.50).

DISCUSSION

Our findings replicate those of previously published studies that demonstrate the significant role of comorbidities in COPD readmissions.9-11,22,23 Our study shows that after an index admission for COPD, the risk of non–COPD-related readmission was significantly increased, particularly among patients with multiple comorbidities. Diabetes has emerged as an independent predictor of COPD readmissions in multiple studies.11,13,24,25 Buhr et al22 reviewed discharges in a large national database between 2010 and 2016 and found the Elixhauser comorbidity index score (which includes diabetes) to be much higher among the 9.56% of patients who were readmitted within 30 days for non-COPD causes compared with COPD. Interestingly, patients who required higher postacute care services were also at higher risk for non–COPD-related readmissions. A large retrospective analysis using the HCUP Nationwide Inpatient Sample data from 2002 to 2014 showed that diabetes is associated with longer COPD-related hospital LOS and higher rate of respiratory failure, stroke, and AKI, but no change in mortality.26 Although these findings highlight the role of comorbid diabetes in worse outcomes for COPD, readmissions and socioeconomic consequences were not studied. Our analysis demonstrating increased LOS, complications, and health care costs in patients with comorbid diabetes and COPD can help design programs and interventions to target patients at highest risk and is an important area of future study.22,27

In our study, we found that diabetes is not only a risk factor for hospital readmission but also associated with increased rate of in-hospital complications and higher health care costs, including postacute care among patients hospitalized with AECOPD. None of the prior studies3,12,24,26 have delineated the health care costs associated with comorbid diabetes or distinguished between complicated vs uncomplicated diabetes in patients with AECOPD and the postacute care outcomes in the US health care system.

Our study also demonstrated that, considering all index admissions with COPD exacerbation, 30% of patients had diabetes, which is higher than in previous studies citing 10% prevalence of diabetes in patients with COPD.5 This observation would suggest that diabetes is more common than previously thought in those with severe disease requiring hospitalization for COPD exacerbation. In-hospital complications were higher in this group, as demonstrated by increased risk for respiratory failure, AKI, and encephalopathy, which likely corresponds to the increases in hospital cost and weighted LOS.

The increase in complications may result from several possibilities. Hyperglycemia associated with glucocorticoid use is more common among patients with diabetes and may itself be a contributing factor to several complications and worse AECOPD outcomes.28,29 Although a causal relationship between diabetes and COPD is not established, patients with COPD have higher levels of inflammatory markers, insulin resistance, and evidence for worse survival with comorbid diabetes.30,31 Further, the simultaneous elevation of inflammatory markers such as C-reactive protein, fibrinogen, and leukocyte count is associated with increased risk of an AECOPD even in those with mild disease.32 Chronic inflammation in COPD is linked to growth hormone resistance (decreased insulin-like growth factor 1/growth hormone ratio) and insulin resistance, which are associated with worse outcomes.33

The higher rate of respiratory failure seen in our study among patients admitted for COPD may be explained by the higher risks of hypercapnic respiratory failure in patients with diabetes.34,35 Diabetes is associated with increased sleep-disordered breathing, impaired lung function, and increased respiratory muscle fatigue.35-37

Our study also showed that diabetes was associated with higher rates of nonpulmonary hospital complications during both index admissions and 30-day readmissions. The increased inflammation seen in AECOPD, marked by such factors as elevated interleukin-6 and hyperglycemia, likely contributes to increased risk for AKI.38 Similarly, diabetes has also been associated with cognitive impairment in longitudinal studies, especially in older patients. Hyperglycemia, hypoglycemia, and vascular changes are thought to contribute to cognitive impairment, which, along with cerebrovascular complications of diabetes, may contribute to significant functional impairment and eventually poorer prognosis.39 Thus, diabetes may be key to a COPD phenotype with increased systemic inflammation, endothelial dysfunction, and worse outcomes.40

Discharge disposition to home was less likely in those with diabetes compared with those without diabetes. This has implications for patient outcomes and contributes to overall health care costs, given that long-term care facilities account for $136 billion per year in the United States. We found that nonroutine discharges to short-term hospital and short-term and intermediate care facilities were higher in patients with complicated diabetes and uncomplicated diabetes (16.9% and 13.9%, respectively) compared with 12.5% of COPD hospitalizations seen in Medicare data.2 We know from prior studies looking at COPD acute care outcomes that patients with low income are known to have higher hospitalization and readmission rates.41 Therefore, the prevalence of complicated diabetes and uncomplicated diabetes in groups with lower household incomes is another marker of increased risk that highlights the need for targeted interventions in this subpopulation.

Although inpatient mortality during both index COPD hospitalization and readmissions remained similar in all groups, this could be biased by the patient’s disposition upon discharge, given that outcomes of patients discharged to long-term care facilities are not captured in the HCUP-NRD, which is an important limitation of our study.

COPD exacerbations are clinical and socioeconomic events that have far-reaching consequences on patient health and functional capacity beyond the index hospitalization.42 Our study confirms that among patients hospitalized with a primary diagnosis of COPD, the presence of diabetes as a comorbidity is associated with worse clinical outcomes, increased risk for hospital complications, higher health care cost, and increased risk for all-cause and COPD-related readmissions. In addition to diabetes, our study shows that a nonfavorable socioeconomic status, reflected by low household income and Medicaid enrollment, is associated with significantly higher risk of 30-day all-cause and COPD-related readmission. As previously described,43 our study again demonstrates the importance of socioeconomic status on hospital outcomes and identifies a subgroup of patients who might benefit from targeted intervention to ensure appropriate follow-up post hospital discharge.

Limitations

Although our study utilizes a robust and previously validated methodology, we acknowledge that our study has several limitations. Observational studies are subject to bias introduced by measured and unmeasured confounding variables even after adjusting for multiple factors such as demographics, socioeconomic status, and comorbidities. Therefore, it is still plausible that the association of poor COPD-related outcomes and diabetes is caused by other confounding factors not adjusted for in this study. For example, information on time of onset of diabetes or disease duration was not available, so the impact of such important confounding factors on outcomes cannot be studied. Although we have provided insights into the pathophysiologic basis for increased complications among patients with diabetes, we cannot establish a causal relationship with an observational study. Furthermore, the HCUP-NRD database enabled us to track each unique patient during the same year. However, we were not able to track patients across the years; thus, some readmissions may be for the same patient. Hospital care is also not homogeneous across hospitals and can influence hospital LOS and cost. However, we adjusted for hospital characteristics, including the hospital, identification number, ownership, bed size, teaching status, and urban-rural designation, to account for these important confounders. Our study is also limited by the statistical method we used and the fact that NRD hospital identifiers (HOSP_NRD) vary from year to year. Although “survey” methodologies are recommended to produce national estimates by using weighted data, they do not account for patient clustering at the hospital level, where cost, hospital LOS, and quality of care vary among hospitals. Although we adjusted for hospital characteristics in the NRD, bias can be inevitably introduced by not accounting for patients’ clustering at hospital levels. Finally, although we used the GDP price index to adjust for inflation, this metric—compared with other commonly used metrics, such as the Consumer Price Index—could underestimate the cost difference between the groups.

CONCLUSIONS

COPD and diabetes are interlinked with progression through common pathological pathways and are associated with higher health care utilization and costs. The postdischarge period after an AECOPD is critical for such vulnerable patients with comorbid diabetes, and targeted interventions for this patient population are crucial.

Author Affiliations: Brooklyn Campus of the VA NY Harbor Healthcare System (PB), Brooklyn, NY; Lerner Research Institute and the Respiratory Institute, Cleveland Clinic (AA, UH, JGZ), Cleveland, OH; Center for Populations Health Research, Lerner Research Institute, Cleveland Clinic (RL), Cleveland, OH; SUNY Downstate Health Science Center (DD); Now with Albert Einstein Medical College/Jacobi Medical Center (DD), Bronx, NY.

Source of Funding: Supported by a grant from the National Institute of Health, National Heart, Lung, and Blood Institute: K08 HL133381 (PI: Dr Zein).

Author Disclosures: Dr Hatipoğlu is an associate editor for the COPD section of UpToDate. Dr Zein is a board member of the American Board of Internal Medicine and has grants pending and received from the National Institute of Health, National Heart, Lung, and Blood Institute. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (PB, DD, JGZ); acquisition of data (JGZ); analysis and interpretation of data (PB, RL, JGZ); drafting of the manuscript (PB, AA, RL, DD, UH, JGZ); critical revision of the manuscript for important intellectual content (PB, AA, RL, DD, UH, JGZ); statistical analysis (RL, JGZ); provision of patients or study materials (DD); obtaining funding (JGZ); supervision (RL, JGZ); supervision of statistical analysis (RL); data interpretation (AA, UH); and final approval of publication (JGZ).

Address Correspondence to:Joe G. Zein, MD, PhD, Respiratory Institute, Cleveland Clinic, 9500 Euclid Ave, Cleveland, OH 44106. Email: zeinj@ccf.org.

REFERENCES

1. Croft JB, Wheaton AG, Liu Y, et al. Urban-rural and state differences in COPD prevalence, Medicare hospitalizations, and mortality — United States, 2015. MMWR Morb Mortal Wkly Rep. 2018;67(7):205-211. doi:10.15585/mmwr.mm6707a1

2. Hospital Readmissions Reduction Program (HRRP). CMS. Updated December 1, 2021. Accessed March 17, 2021. https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/AcuteInpatientPPS/Readmissions-Reduction-Program

3. Press VG, Konetzka RT, White SR. Insights about the economic impact of COPD readmissions post implementation of the Hospital Readmission Reduction Program. Curr Opin Pulm Med. 2018;24(2):138-146. doi:10.1097/MCP.0000000000000454

4. Divo M, Cote C, de Torres JP, et al. Comorbidities and risk of mortality in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2012;186(2):155-161. doi:10.1164/rccm.201201-0034OC

5. Agusti A, Calverley PMA, Celli B, et al; Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE) Investigators. Characterisation of COPD heterogeneity in the ECLIPSE cohort. Respir Res. 2010;11(1):122. doi:10.1186/1465-9921-11-122

6. Kuziemski K, Słominski W, Jassem E. Impact of diabetes mellitus on functional exercise capacity and pulmonary functions in patients with diabetes and healthy persons. BMC Endocr Disord. 2019;19(1):2. doi:10.1186/s12902-018-0328-1

7. Lee CTC, Mao IC, Lin CH, Lin SH, Hsieh MC. Chronic obstructive pulmonary disease: a risk factor for type 2 diabetes: a nationwide population-based study. Eur J Clin Invest. 2013;43(11):1113-1119. doi:10.1111/eci.12147

8. Celli BR, Locantore N, Yates J, et al; ECLIPSE Investigators. Inflammatory biomarkers improve clinical prediction of mortality in chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2012;185(10):1065-1072. doi:10.1164/rccm.201110-1792OC

9. Press VG, Au DH, Bourbeau J, et al. Reducing chronic obstructive pulmonary disease hospital readmissions. an official American Thoracic Society workshop report. Ann Am Thorac Soc. 2019;16(2):161-170. doi:10.1513/AnnalsATS.201811-755WS

10. Shah T, Churpek MM, Coca Perraillon M, Konetzka RT. Understanding why patients with COPD get readmitted. Chest. 2015;147(5):1219-1226. doi:10.1378/chest.14-2181

11. Lau CS, Siracuse BL, Chamberlain RS. Readmission After COPD Exacerbation Scale: determining 30-day readmission risk for COPD patients. Int J Chron Obstruct Pulmon Dis. 2017;12:1891-1902. doi:10.2147/COPD.S136768

12. Njoku CM, Alqahtani JS, Wimmer BC, et al. Risk factors and associated outcomes of hospital readmission in COPD: a systematic review. Respir Med. 2020;173:105988. doi:10.1016/j.rmed.2020.105988

13. Crisafulli E, Torres A, Huerta A, et al. C-reactive protein at discharge, diabetes mellitus and ≥ 1 hospitalization during previous year predict early readmission in patients with acute exacerbation of chronic obstructive pulmonary disease. COPD. 2015;12(3):306-314. doi:10.3109/15412555.2014.933954

14. Overview of the Nationwide Readmissions Database (NRD): about the NRD. Agency for Healthcare Research and Quality. Updated October 18, 2021. Accessed February 19, 2021. https://www.hcup-us.ahrq.gov/nrdoverview.jsp#about

15. Healthcare Cost and Utilization Project user support. Agency for Healthcare Research and Quality. Updated February 10, 2022. Accessed February 11, 2021. https://www.hcup-us.ahrq.gov/overview.jsp

16. Zhang P, Lopez R, Attaway AH, et al. Diabetes mellitus is associated with worse outcome in patients hospitalized for asthma. J Allergy Clin Immunol Pract. 2021;9(4):1562-1569.e1. doi:10.1016/j.jaip.2020.10.054

17. Clinical Classifications Software (CCS) 2015. Agency for Healthcare Research and Quality. March 2016. Accessed January 13, 2021. https://www.hcup-us.ahrq.gov/toolssoftware/ccs/CCSUsersGuide.pdf

18. Titmarsh S, Poliziani M, Russell RE. “Breathing new life into chronic obstructive pulmonary disease (COPD)” – results from an online survey of UK patients. Int J Chron Obstruct Pulmon Dis. 2019;14:2799-2807. doi:10.2147/COPD.S222139

19. Attaway AA, Zein J, Hatipoğlu US. SARS-CoV-2 infection in the COPD population is associated with increased healthcare utilization: an analysis of Cleveland Clinic’s COVID-19 registry. EClinicalMedicine. 2020;26:100515. doi:10.1016/j.eclinm.2020.100515

20. Cost-to-Charge Ratio Files. Agency for Healthcare Research and Quality. Updated November 3, 2021. Accessed December 23, 2020. https://www.hcup-us.ahrq.gov/db/state/costtocharge.jsp

21. National income product accounts tables: section 1 domestic product and income: table 1.1.4 price indexes for gross domestic products. Bureau of Economic Analysis. Accessed January 22, 2021. https://apps.bea.gov/iTable/iTable.cfm?reqid=19&step=2#reqid=19&step=2&isuri=1&1921=survey

22. Buhr RG, Jackson NJ, Dubinett SM, Kominski GF, Mangione CM, Ong MK. Factors associated with differential readmission diagnoses following acute exacerbations of chronic obstructive pulmonary disease. J Hosp Med. 2020;15(4):219-227. doi:10.12788/jhm.3367

23. Press VG, Miller BJ. The Hospital Readmissions Reduction Program and COPD: more answers, more questions. J Hosp Med. 2020;15(4):252-253. doi:10.12788/jhm.3362

24. Roberts MH, Clerisme-Beaty E, Kozma CM, Paris A, Slaton T, Mapel DW. A retrospective analysis to identify predictors of COPD-related rehospitalization. BMC Pulm Med. 2016;16(1):68. doi:10.1186/s12890-016-0231-3

25. Barba R, de Casasola GG, Marco J, et al. Anemia in chronic obstructive pulmonary disease: a readmission prognosis factor. Curr Med Res Opin. 2012;28(4):617-622. doi:10.1185/03007995.2012.675318

26. Gunasekaran K, Murthi S, Elango K, et al. The impact of diabetes mellitus in patients with chronic obstructive pulmonary disease (COPD) hospitalization. J Clin Med. 2021;10(2):235. doi:10.3390/jcm10020235

27. Wadhera RK, Joynt Maddox KE, Wasfy JH, Haneuse S, Shen C, Yeh RW. Association of the Hospital Readmissions Reduction Program with mortality among Medicare beneficiaries hospitalized for heart failure, acute myocardial infarction, and pneumonia. JAMA. 2018;320(24):2542-2552. doi:10.1001/jama.2018.19232

28. Baker JM, Pace HA, Ladesich JB, Simon SD. Evaluation of the impact of corticosteroid dose on the incidence of hyperglycemia in hospitalized patients with an acute exacerbation of chronic obstructive pulmonary disease. Hosp Pharm. 2016;51(4):296-304. doi:10.1310/hpj5104-296

29. Benabdelghaffar H, Khmiss T, Fenniche S, et al. Hyperglycemia is associated with poor outcomes in patients with acute exacerbations of chronic obstructive pulmonary disease. Eur Respir J. 2013;42(suppl 57):P2164.

30. Machado FVC, Pitta F, Hernandes NA, Bertolini GL. Physiopathological relationship between chronic obstructive pulmonary disease and insulin resistance. Endocrine. 2018;61(1):17-22. doi:10.1007/s12020-018-1554-z

31. Miller J, Edwards LD, Agustí A, et al; Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE) Investigators. Comorbidity, systemic inflammation and outcomes in the ECLIPSE cohort. Respir Med. 2013;107(9):1376-1384. doi:10.1016/j.rmed.2013.05.001

32. Thomsen M, Ingebrigtsen TS, Marott JL, et al. Inflammatory biomarkers and exacerbations in chronic obstructive pulmonary disease. JAMA. 2013;309(22):2353-2361. doi:10.1001/jama.2013.5732

33. Barnes PJ. The cytokine network in chronic obstructive pulmonary disease. Am J Respir Cell Mol Biol. 2009;41(6):631-638. doi:10.1165/rcmb.2009-0220TR

34. Gupta K, Prasad A, Nagappa M, Wong J, Abrahamyan L, Chung FF. Risk factors for opioid-induced respiratory depression and failure to rescue: a review. Curr Opin Anaesthesiol. 2018;31(1):110-119. doi:10.1097/ACO.0000000000000541

35. Biswas N, Sangma MA. Factors related to noninvasive ventilation outcomes during an episode of hypercapnic respiratory failure in chronic obstructive pulmonary disease. Mymensingh Med J. 2019;28(3):605-619.

36. Lecube A, Simó R, Pallayova M, et al. Pulmonary function and sleep breathing: two new targets for type 2 diabetes care. Endocr Rev. 2017;38(6):550-573. doi:10.1210/er.2017-00173

37. Fujimaki S, Kuwabara T. Diabetes-induced dysfunction of mitochondria and stem cells in skeletal muscle and the nervous system. Int J Mol Sci. 2017;18(10):2147. doi:10.3390/ijms18102147

38. Yu SMW, Bonventre JV. Acute kidney injury and progression of diabetic kidney disease. Adv Chronic Kidney Dis. 2018;25(2):166-180. doi:10.1053/j.ackd.2017.12.005

39. Biessels GJ, Strachan MWJ, Visseren FLJ, Kappelle LJ, Whitmer RA. Dementia and cognitive decline in type 2 diabetes and prediabetic stages: towards targeted interventions. Lancet Diabetes Endocrinol. 2014;2(3):246-255. doi:10.1016/S2213-8587(13)70088-3

40. Barnes PJ, Celli BR. Systemic manifestations and comorbidities of COPD. Eur Respir J. 2009;33(5):1165-1185. doi:10.1183/09031936.00128008

41. Parekh TM, Cherrington AL, Bhatia S, et al. The association of low income and high stress with acute care use in COPD patients. Chronic Obstr Pulm Dis. 2020;7(2):107-117. doi:10.15326/jcopdf.7.2.2019.0165

42. Halpin DM, Miravitlles M, Metzdorf N, Celli B. Impact and prevention of severe exacerbations of COPD: a review of the evidence. Int J Chron Obstruct Pulmon Dis. 2017;12:2891-2908. doi:10.2147/COPD.S139470

43. Jiang X, Xiao H, Segal R, Mobley WC, Park H. Trends in readmission rates, hospital charges, and mortality for patients with chronic obstructive pulmonary disease (COPD) in Florida from 2009 to 2014. Clin Ther. 2018;40(4):613-626.e1. doi:10.1016/j.clinthera.2018.03.006