BACKGROUND/RATIONALE:
Depression in individuals with diabetes is common and complicates their diabetes care, and is related to negative outcomes and increased health care expenditures. There is some evidence that both pharmacologic (specifically the new generation antidepressants) and non-pharmacologic treatment for major depression improves glycemic control, and good glycemic control in individuals with diabetes is related to lower health care utilization and costs. However, it has not been established in naturalistic settings, among individuals with diabetes, whether depression treatment actually improves process and intermediate outcomes of care, and decreases utilization and costs over time.

OBJECTIVE(S):
Our major hypothesis was that guideline-consistent treatment for depression will improve diabetes-related treatments among VHA patients with diabetes and depression (DM/D). Our specific aims were: 1) to analyze variations in patterns of guideline-consistent treatment; 2) to evaluate the impact of guideline-consistent depression treatment on patient adherence to diabetes care treatments and on intermediate outcomes of diabetes care; and 3) using longitudinal data, to examine the effect of depression on health care utilization and expenditures, and evaluate the impact of guideline-consistent depression treatment on health care utilization and expenditures.

METHODS:
We used both cross-sectional and longitudinal designs to examine the relationship between depression care and diabetes care (FY1999-2003). The study used repeated measurements within each episode-of-care to support and explain the study findings. Multivariate longitudinal regression techniques were utilized to examine variation in guideline-consistent depression treatment. Declining effect models were used to analyze the association between guideline-consistent depression treatment and intermediate outcomes and health care expenditures.

FINDINGS/RESULTS:
An important finding from our study was that a majority of veterans with diabetes and major depressive disorders (60%) received guideline-consistent antidepressants for depression. Further, such treatment for incident depression was associated with a reduction in expenditures (mainly inpatient expenditures) compared to those without any antidepressant treatment for incident depression, indicating the importance of treating both incident and prevalent depression. No differences in VHA expenditures between those with incident and prevalent depression were observed when controlling for patient characteristics and other factors. Quantile regression estimates revealed that depressive disorders are not associated with additional expenditures at every point of the expenditures distribution. When looking at overall expenditures between those with and without depression and better control of diabetes outcomes (measured as glycemic control), there were excess expenditures associated with depression (especially incident depression) and mortality was high for those with both incident and prevalent depression.
Among women veterans with diabetes, different screening algorithms revealed that a substantial proportion of individuals (16%) were diagnosed with non-major depression (Shen et al., 2008) that was clinically significant; and a 5% overall rate existed for prevalent depression vs. 9% for newly diagnosed depression in a one year period (Shen et al., 2009), emphasizing the need to focus on both non-major and incident depression as well.

IMPACT:
Our findings highlight the need for the use of a comprehensive algorithm to detect diagnosed depression, both incident and prevalent. Excess mortality associated with diagnosed depression suggests a need exists for routine screenings and interventions for depression. Lack of a relationship between glycemic control and depression suggests that the negative effects of depression may not be through its effect on diabetes outcomes. The excess costs associated with depression, and documented net cost savings associated with treating depression in patients with DM/D, suggest the need for aggressive treatment for depression. These findings also will help in making informed decisions about prioritization of limited resources. The study will advance research methods by applying a common unit of cost schedule applicable to both VA and non-VA sources of care (VA reasonable charges) that would reflect differences across patterns of care rather than changes in production efficiencies across the sources for that care. Further, our findings suggest that efforts are needed to monitor non-major depression and promote guideline-consistent treatment and pharmaco-therapy for major depression among veterans with diabetes, especially women. Guideline-consistent antidepressant treatment has the potential to improve clinical outcomes and produce cost savings.

---

External Links for this Project

---

PUBLICATIONS:

Journal Articles

1. Banerjea R, Sambamoorthi U, Smelson D, Pogach LM. Chronic illness with complexities: mental illness and substance use among Veteran clinic users with diabetes. The American journal of drug and alcohol abuse. 2007 Jan 1; 33(6):807-21. [view]
2. Meduru P, Helmer D, Rajan M, Tseng CL, Pogach L, Sambamoorthi U. Chronic illness with complexity: implications for performance measurement of optimal glycemic control. Journal of general internal medicine. 2007 Dec 1; 22 Suppl 3:408-18. [view]
3. Banerjea R, Sambamoorthi U, Smelson D, Pogach LM. Expenditures in mental illness and substance use disorders among veteran clinic users with diabetes. The journal of behavioral health services & research. 2008 Jul 1; 35(3):290-303. [view]
4. Tiwari A, Rajan M, Miller D, Pogach L, Olfson M, Sambamoorthi U. Guideline-consistent antidepressant treatment patterns among veterans with diabetes and major depressive disorder. Psychiatric services (Washington, D.C.). 2008 Oct 1; 59(10):1139-47. [view]

Conference Presentations

1. Tiwari A, Kashner M, Olfson M, Rajan M, Pogach LM, Sambamoorthi U. Cost Savings Associated with Guideline-consistent Treatment for Depression in Diabetes. Paper presented at: VA HSR&D National Meeting; 2007 Feb 21; Arlington, VA. [view]
2. Shen C, Kashner M, Olfson M, Rajan M, Pogach LM, Sambamoorthi U. Effects of Incident and Prevalent Depressive Disorders on Healthcare Costs among Veterans with Diabetes. Poster session presented at: AcademyHealth Annual Research Meeting; 2008 Jun 8; Washington, DC. [view]
3. Banerjea R, Sambamoorthi U, Smelson D, Pogach LM. Expenditures in Chronic Illness with Complexities: Mental Health and Substance Abuse among Veterans with Diabetes. Poster session presented at: AcademyHealth Annual Research Meeting; 2007 Jun 3; Orlando, FL. [view]
4. Tiwari A, Rajan M, Miller DR, Pogach LM, Olfson M, Sambamoorthi U. Guideline-Consistent Anti-depressant Treatment Patterns Among Veterans with Diabetes and Depression. Paper presented at: AcademyHealth Annual Research Meeting; 2006 Jun 1; Seattle, WA. [view]
5. Banerjea R, Pogach LM, Smelson DA, Sambamoorthi U. Multi-morbidity in Physical Illness and Co-occurring Mental Illness and Substance Use Disorders: Who are at Risk? Poster session presented at: American Public Health Association Annual Meeting and Exposition; 2009 Nov 7; Philadelphia, PA. [view]
6. Shen C, Banerjea R, Findley P, Pogach LM, Sambamoorthi U. Prevalent and Incident Depression and Mortality among Veterans with Diabetes. Poster session presented at: AcademyHealth Annual Research Meeting; 2009 Jun 29; Chicago, IL. [view]
7. Banerjea R, Pogach LM, Sambamoorthi U. Type of Mental Illness and Quality of Diabetes Care among Veterans. Poster session presented at: AcademyHealth Annual Research Meeting; 2008 Jun 8; Washington, DC. [view]
8. Sambamoorthi U, Shen C, Kashner M, Olfson M, Banerjea R. VHA Expenditures and Incident and Prevalent Depressive Disorders among Veteran Health Administration Clinic Users with Diabetes. Paper presented at: VA HSR&D National Meeting; 2009 Feb 13; Baltimore, MD. [view]

DRA: Mental, Cognitive and Behavioral Disorders, Health Systems Science
DRE: Epidemiology
Keywords: Cost, Research measure, Research method
MeSH Terms: none