• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Med Care. Author manuscript; available in PMC Jun 4, 2012.
Published in final edited form as:
PMCID: PMC3366691

The Impact of Comorbidity Type on Measures of Quality for Diabetes Care



Studies provide conflicting results about the impact of comorbid conditions on the quality of chronic illness care. We assessed the effect of comorbidity type (concordant, discordant, or both) on the receipt of guideline-recommended care among patients with diabetes.

Research Design

Patients were assigned to 1 of 4 condition groups: diabetes-concordant (hypertension, ischemic heart disease, hyperlipidemia), and/or -discordant (arthritis, depression, chronic obstructive pulmonary disease) conditions, or neither. We evaluated hemoglobin (Hb) A1c, blood pressure, and low-density lipoprotein cholesterol (LDL-C) readings at index and measured overall good quality of diabetes care, including a 6-month follow-up interval. We assessed the effect of condition group on overall good quality of care with logistic regression and generalized ordered logistic regression.


We assigned 35,872 patients to the diabetes comorbid condition groups, ranging from 2.0% in the discordant-only group to 58.0% in the concordant-only group. Patients with both types of conditions were more likely than those with no comorbidities to receive overall good quality for glycemic [odds ratio (OR), 2.13; 95% confidence interval (CI), 1.86-2.41], blood pressure (OR, 1.62; 95% CI, 1.40-1.84) and LDL-C (OR, 3.57; 95% CI, 3.08-4.05) control within 6 months of an index visit. They were also more likely to receive overall good quality for all 3 quality measures combined (OR, 2.17; 95% CI, 1.96-2.39).


Patients with the greatest clinical complexity were more likely than less complex patients to receive high quality diabetes care, suggesting that increased complexity does not necessarily predispose chronically ill patients to receiving poorer care. However, caution should be used in treating certain patient groups, such as the elderly, for whom adherence to multiple condition-specific guidelines may lack benefit or cause harm.

Keywords: diabetes mellitus, quality of health care, comorbidity


Although treatment of chronically ill patients consumes more than three-fourths of U.S. healthcare expenditures, evidence suggests that these patients receive just over half of the appropriate recommended care.1,2 This has implications particularly for patients with diabetes, as approximately 40% suffer from 3 or more comorbidities.1 Studies assessing the impact of comorbidities on quality of chronic illness care provide mixed results. One study found that discussion of unrelated conditions during a clinical encounter decreased the likelihood of addressing elevated blood pressure in diabetic patients.3 Similarly, patients with uncontrolled hypertension were less likely to receive antihypertensive therapy intensification if they had greater numbers of unrelated conditions.4 Conversely, emerging evidence suggests that the presence of multiple comorbidities may actually enhance quality for chronically ill patients. We recently reported that hypertensive patients with greater complexity were more likely to receive high quality care than those with fewer comorbidities.5 This and other studies6,7 highlight the need for additional research to elucidate the impact of comorbid conditions on the quality of chronic illness care.

Using a framework employed in prior research,5,8 we identified comorbidities that were concordant and discordant with diabetes. Diabetes-concordant comorbidities have similar pathogenesis and treatment strategies, while diabetes-discordant comorbidities are not related to diabetes development or management. We examined the impact of comorbidity type (concordant, discordant, or both) on 3 diabetes quality-of-care measures: glycemic, blood pressure, and lipid control. We also assessed the relationship between comorbidity type and likelihood of receiving overall good quality for all 3 measures.


Study population

We identified Veterans with diabetes receiving primary care at 7 Veterans Affairs (VA) facilities from July 2007 through June 2008. We classified patients as having diabetes if, in the study interval or 2 years prior, they had: 2 outpatient or 1 inpatient diagnosis code indicating diabetes;9 a filled prescription for diabetes medication; or at least 2 glucose readings ≥ 200 mg/dL. We excluded patients with limited life expectancy (diagnosis of metastatic cancer or hospice/palliative care use) and who died during the study or follow-up period.

Diabetes-concordant and -discordant conditions

Using diagnosis and procedure codes in the VA National Patient Care Database10 and Fee Basis files,11 we identified patients with diabetes-concordant (hypertension, ischemic heart disease, and hyperlipidemia) and diabetes-discordant [arthritis, chronic obstructive pulmonary disease (COPD), and depression] conditions within 2 years prior to their most recent primary care encounter during the study interval. We categorized patients into mutually exclusive diabetes condition groups: 1) no comorbid conditions; 2) discordant-only; 3) concordant-only; 4) both concordant and discordant conditions.5

Study outcomes

We used American Diabetes Association recommendations12 to assess hemoglobin A1c (HbA1c) (<7%), blood pressure (<130/80 mm Hg), and low-density lipoprotein cholesterol (LDL-C) (<100 mg/dL) control at index. Among uncontrolled patients, we examined a 6-month follow-up period to determine whether appropriate medication adjustments were made or the patient’s most recent follow-up reading was at goal (Figure 1).

Figure 1
Algorithm to Assess Appropriate Care for Patients with Diabetes

We obtained HbA1c, blood pressure, and LDL-C values from the VA Network data warehouse, which includes data from each hospital’s electronic medical records. We assessed control at index using the patient’s most recent readings. We used VA Decision Support System pharmacy data to identify current medications (prescriptions filled within 100 days prior to index)13 and those received during follow-up. We used the data warehouse to identify prescriptions filled outside the VA. We computed average daily medication dosages using the formula: (quantity of medication/days supplied) × numeric dosage. We calculated overall good quality for each measure by adding the number of patients controlled at index and those receiving appropriate follow-up care. We refer to attainment of either of these criteria as “overall good quality”

Statistical analyses

We assessed the proportion of patients that were controlled at index; received appropriate follow-up care; and received overall good quality by chronic condition group for the 3 measures. We used logistic regression to determine the impact of coexisting conditions on the likelihood of being controlled at index; receiving appropriate follow-up care; and receiving overall good quality. Additionally, we compared the likelihood of receiving overall good quality for 1 or more of the measures, using generalized ordered logistic regression. Models were adjusted for age, illness burden defined by Diagnostic Cost Group (DCG) Relative Risk Scores,14 and accounted for clustering of patients by facility (Table 1). Analyses were conducted using SAS v9.2 (SAS Institute Inc., Cary, North Carolina) and Stata 10 (StataCorp LP, College Station, TX).

Table 1
Definitions of study variables

Sensitivity analyses

We assessed the impact of a shorter follow-up interval (3 months); exclusion of patients with limited life expectancy; use of VA quality indicators [HbA1c (≤9%) and blood pressure (<140/90 mm Hg)]; and effect of VA primary and specialty care utilization, on the study outcomes. We identified 25 relevant specialty care clinics for the study conditions, and designated an encounter as specialty care if the patient had a condition and a visit to the relevant clinic (e.g., patient with COPD and a visit to pulmonary clinic) during the prior year. Additionally, because diabetes and depression commonly co-occur, we conducted univariate analyses to evaluate the influence of depression on study outcomes.

The Institutional Review Board at Baylor College of Medicine and the Michael E. DeBakey VA Research and Development Committee approved this study. The funding sources played no role in the study conduct or interpretation.


We identified 35,872 diabetic patients meeting inclusion criteria. Of these, 5.8% had no comorbidities, 58.0% had concordant-only, 2.0% had discordant-only, and 34.2% had both (Table 2). We excluded 2,337 patients who had limited life expectancy or died during follow-up.

Table 2
Characteristics of the Diabetes Cohort by Chronic Condition Group

For all measures, the proportion of patients controlled at index was lowest among those with no comorbidities (Table 3). The group with both concordant and discordant comorbidities had the highest proportion receiving appropriate follow-up care for glycemic (61.2%), blood pressure (66.5%), and LDL-C (55.8%) control. The no comorbidities group had the lowest proportion receiving appropriate follow-up care for glycemic and LDL-C control (50.0% and 25.8%, respectively) while the discordant-only group had the lowest proportion receiving appropriate follow-up for blood pressure control (44.0%). The total proportion of patients receiving overall good quality was 77.8% for glycemic, 80.4% for blood pressure, and 85.5% for LDL-C control. Similar to findings for control at index, patients with both types of conditions had the highest proportion of patients receiving overall good quality for each measure, while the no comorbidities group had the lowest.

Table 3
Glycemic, Blood Pressure, and LDL-C Control at Index, Appropriate Follow-up, and Overall Good Quality of Diabetes Care by Chronic Condition Group, Proportions and Adjusted ORs

Adjusting for age and illness burden, patients with both types of comorbidites were more likely than patients with no comorbidities to achieve glycemic and LDL-C control at index [odds ratio (OR), 2.15; 95% confidence interval (CI), 1.88-2.43; and 2.22; 1.94-2.49, respectively)], (Table 3). Patients with concordant-only comorbidities and those with both types were more likely than those with no comorbidities to receive appropriate follow-up care and overall good quality for each measure. Patients with discordant-only comorbidities were more likely than those with no comorbidities to receive overall good quality for glycemic (OR, 1.48; 95% CI, 1.13-1.83) and LDL-C (OR, 1.30; 95% CI, 1.06-1.59) control. When assessing attainment of guideline-recommended levels across the 3 quality-of-care measures simultaneously, patients with concordant-only and both types of comorbidities were more likely than those with no comorbidities to receive overall good quality (OR, 2.00; 95% CI, 1.82-2.20 and OR, 2.17; 95% CI, 1.96-2.39, respectively) (Table 4).

Table 4
Adjusted ORs of Overall Good Quality of Care for at least 1, 2, or 3 Diabetes Quality Indicators by Condition Group*

We found similar results for overall good quality using a 3-month follow-up period and when assessing less stringent glycemic (HbA1c </=9%) and blood pressure (<140/90 mm Hg) thresholds. Our findings that the most complex patients received higher levels of overall good quality for each measure persisted after adjusting for primary and specialty care visits. Although a higher proportion of patients excluded for limited life expectancy or death were in the both condition group, study outcomes were similar when including these patients in our analyses. Finally, univariate analyses showed that depression was not associated with decreased levels of overall good quality for any of the measures (data not shown).


We assessed quality for 3 dimensions of diabetes care: glycemic, blood pressure, and LDL-C control, and determined how this varied by condition group. We found that patients with both concordant and discordant conditions as well as those with concordant-only were more likely than those with no comorbidities to receive overall good quality for each dimension examined. These findings did not change when we controlled for prior visits, assessed a shorter follow-up period, or evaluated care using VA performance indicators for glycemic and blood pressure control. Patients with both types of comorbidities were also more likely than those with no comorbidities to receive overall good quality for all 3 dimensions of diabetes care combined.

This work extends prior studies by examining the correlation between different types of conditions and quality for 3 dimensions of diabetes care.15,16 As assessed by current standards, provision of high quality, comprehensive care in this population necessitates adherence to numerous clinical practice guidelines. Although our results show high levels of guideline-recommended care among the most complex diabetic patients, studies indicate that such care may not be equally beneficial across all chronically ill patients.17,18 For example, elderly patients with multiple comorbidities receiving care in accordance with multiple guidelines may have worse outcomes. This highlights the intricacies of managing such patients and the need to adapt current methods of quality assessment and provider reimbursement to address the challenges unique to this patient population.

Several factors may contribute to the higher quality found among more complex diabetic patients. First, providers may be more likely to treat coexisting conditions in complex patients to reduce the risk of adverse consequences. For example, they may aggressively target blood pressure and lipid control because their benefit in preventing or slowing the progression of coronary artery disease has been well documented.12 Additionally, these findings may result in part from system-level processes aimed at increasing evidence-based care within the VA.19

Medication adherence and healthcare utilization may also influence the relationship between diabetes complexity and quality of care. Patients with both types of comorbidities had more frequent primary and specialty care visits than other patients. This increased healthcare exposure may lead to stronger provider/patient relationships, which may increase adherence to treatment plans, including recommended visits and prescribed medications.20,21 More complex patients may also utilize ancillary services, such as case management, more often which may improve their overall health and increase adherence.20 However, we found that controlling for patient visits did not impact our findings.

Depression and diabetes frequently co-occur, yet differ in pathogenesis and treatment, so we felt this was an ideal discordant condition.8 While some studies have shown that depression is associated with poorer quality of diabetes care,22,23 others report that patients with comorbid mental illness, including depression, received similar quality of diabetes care to those without mental disorders.24,25 We examined the influence of depression on our models and did not find a significant impact.

Our study’s strengths include a large study population, evaluation of multiple dimensions of diabetes care, inclusion of a follow-up period, and use of a novel multinomial logistic approach to assess the impact of comorbidities on comprehensive diabetes care. We ascertained data from VA administrative sources which provide extensive inpatient and outpatient VA utilization data10 and obtained clinical variables from the data warehouse which includes data directly from VA electronic medical records. Nonetheless, there are limitations that should be considered when interpreting our findings. First, we conducted this study in the VA, which serves a predominantly male, elderly patient population; thus generalizability may be limited. Also, we identified common chronic conditions to classify patients into condition groups, but the conditions chosen may not reflect all existing comorbidities. Finally, although the patient distribution across the 4 condition groups is not balanced, each group is large enough for appropriate statistical modeling.

We found that diabetic patients with the highest level of complexity did not have lower levels of overall good quality compared with less complex patients. Further, complex patients were more likely to receive overall good quality of care across all 3 dimensions of care. This suggests that increased complexity does not necessarily predispose patients with diabetes to receiving poorer quality of care.


The research reported here was supported by the Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development Service, (PPO 09-316, PI LeChauncy D. Woodard, MD, MPH), (IIR 04-349, PI Laura A. Petersen, MD, MPH), National Institutes of Health (R01 HL079173-01, PI Laura A. Petersen, MD, MPH), and Houston VA HSR&D Center of Excellence HFP90-020 (PI Laura A. Petersen, MD, MPH). Dr. Petersen was a Robert Wood Johnson Foundation Generalist Physician Faculty Scholar (grant number 045444) and an American Heart Association Established Investigator Awardee (grant number 0540043N) at the time this work was conducted. Dr. Woodard is an assistant professor at the Michael E. DeBakey VA Medical Center Health Services Research and Development Center of Excellence, Houston, Texas. The funding sources played no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript. The views expressed are solely of the authors, and do not necessarily represent those of the VA.


This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. Wolff JL, Starfield B, Anderson G. Prevalence, expenditures, and complications of multiple chronic conditions in the elderly. Arch Intern Med. 2002;162:2269–2276. [PubMed]
2. McGlynn EA, Asch SM, Adams J, et al. The quality of health care delivered to adults in the United States. N Eng J Med. 2003;348:2635–2645. [PubMed]
3. Kerr EA, Zikmund-Fisher BJ, Klamerus ML, et al. The role of clinical uncertainty in treatment decisions for diabetic patients with uncontrolled blood pressure. Ann Intern Med. 2008;148:717–727. [PubMed]
4. Turner BJ, Hollenbeak CS, Weiner M, et al. Effect of unrelated comorbid conditions on hypertension management. Ann Intern Med. 2008;148:578–586. [PubMed]
5. Petersen LA, Woodard LD, Henderson LM, et al. Will hypertension performance measures used for pay-for-performance programs penalize those who care for medically complex patients? Circulation. 2009;119:2978–2985. [PMC free article] [PubMed]
6. Wong ND, Lopez V, Tang S, et al. Prevalence, treatment, and control of combined hypertension and hypercholesterolemia in the United States. Am J Cardiol. 2006;98:204–208. [PubMed]
7. Higashi T, Wenger NS, Adams JL, et al. Relationship between number of medical conditions and quality of care. N Engl J Med. 2007;356:2496–2504. [PubMed]
8. Piette JD, Kerr EA. The impact of comorbid chronic conditions on diabetes care. Diabetes Care. 2006;29:725–731. [PubMed]
9. Solberg LI, Engebretson KI, Sperl-Hillen JM, et al. Are claims data accurate enough to identify patients for performance measures or quality improvement? The case of diabetes, heart disease, and depression. Am J Med Qual. 2006;21:238–245. [PubMed]
10. Maynard C, Chapko MK. Data resources in the Department of Veterans Affairs. Diabetes Care. 2004;27:B22–B26. [PubMed]
11. Fee basis files [VA HERC website] VA Health Economics Resource Center; [Accessed November 16, 2010]. Available at: http://www.herc.research.va.gov/data/fb.asp.
12. American Diabetes Association Standards of medical care in diabetes – 2009. Diabetes Care. 2009;32:S6–S12. Executive Summary. [PMC free article] [PubMed]
13. Kerr EA, Smith DM, Hogan MM, et al. Building a better quality measure: are some patients with “poor quality” actually getting good care? Med Care. 2003;41:1173–1182. [PubMed]
14. Petersen LA, Pietz K, Woodard LD, et al. Comparison of the predictive validity of diagnosis-based risk adjusters for clinical outcomes. Med Care. 2005;43:61–67. [PubMed]
15. Saaddine JB, Cadwell B, Gregg EW, et al. Improvements in diabetes processes of care and intermediate outcomes: United States, 1988-2002. Ann Intern Med. 2006;144:465–474. [PubMed]
16. Baes S, Rosenthal MB. Patients with multiple chronic conditions do not receive lower quality of preventive care. J Gen Intern Med. 2008;23:1933–1939. [PMC free article] [PubMed]
17. Fried TR, Tinetti ME, Iannone L. [Accessed November 16, 2010];Primary care clinicians’ experiences with treatment decision making for older persons with multiple conditions. 2010 Sep 13; Arch Intern Med Online First. Available at: http://www.archinternmed.com.
18. Tinetti ME, Bogardus ST, Jr, Agostini JV. Potential pitfalls of disease-specific guidelines for patients with multiple conditions. N Engl J Med. 2004;351:2870–2874. [PubMed]
19. Craig TJ, Perlin JB, Fleming BB. Self-reported performance improvement strategies of highly successful Veterans Health Administration facilities. Am J Med Qual. 2007;22:438–444. [PubMed]
20. Osterberg L, Blaschke T. Adherence to Medication. N Engl J Med. 2005;353:487–497. [PubMed]
21. Kerse N, Buetow S, Mainous AG, et al. Physician-patient relationship and medication compliance: A primary care investigation. Ann Fam Med. 2004;2:455–461. [PMC free article] [PubMed]
22. Frayne SM, Halanych JH, Miller DR, et al. Disparities in diabetes care: Impact of mental illness. Arch Intern Med. 2005;165:2631–2638. [PubMed]
23. Gonzales JS, Safren SA, Cafliero E, et al. Depression, self-care, and medication adherence in type 2 diabetes: relationships across the full range of symptom severity. 2007;30:2222–2227. [PubMed]
24. Desai MM, Rosenheck RA, Druss BG, et al. Mental disorders and quality of diabetes care in the Veterans Health Administration. Am J Psychiatry. 2002;159:1584–1590. [PubMed]
25. Heckbert SR, Rutter CM, Oliver M, et al. Depression in relation to long-term control of glycemia, blood pressure, and lipids in patients with diabetes. J Gen Intern Med. 2010;25:524–529. [PMC free article] [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...