Usha Sambamoorthi, PhD, adjunct associate professor, Division of Health Systems and Policy, UMDNJ-School of Public Health, and director, Health Outcomes, Center for Healthcare Knowledge Management, NJ VA Healthcare System

I

email this print this Share this: digg this! del.icio.us facebook newsvine

t has been well documented that co-occurring, multiple chronic conditions are closely correlated with high healthcare expenditures. Nearly 65% of adults older than 65 have more than one condition and the per capita Medicare expenditures for those with multiple chronic conditions can be 66 times higher than for those without a chronic condition. It is also well established that psychiatric disorders, such as depression, tend to co-occur with general medical illnesses. For example, the rates of depressive disorder are elevated in a number of medical conditions, including infectious diseases such as HIV, various cardiac conditions and diabetes mellitus.

Depression as a co-morbidity with chronic medical conditions is often associated with a myriad of adverse health outcomes, poor quality of life, and increased healthcare use and expenditures. For example, depression increases early mortality during recovery from a myocardial infarction and results in an increased risk of heart failure among older persons with isolated systolic hypertension. Healthcare expenditures of individuals with depression are excessive in many chronic physical conditions like diabetes, heart disease, and hypertension.

Among individuals with co-morbid depression and medical illnesses, pharmaceutical treatment is one of the greatest challenges facing the healthcare system. Because most patients with co-morbid conditions use multiple medications, the potential for drug interactions and for variable impact of antidepressants may discourage their use. This has been seen in individuals with diabetes. Clinical trials have shown that older tricyclic antidepressants, such as nortriptyline, might worsen glycemic control. However, the use of new generation antidepressants, such as selective serotonin reuptake inhibitors (SSRIs), has been shown to result in better glycemic control among patients with diabetes.

While many studies have documented the negative impact of depression on healthcare use and expenditures among individuals with chronic conditions, the patterns of depression treatment and the effect of treated versus not-treated depression on these outcomes, are not well documented. In this context, for the last 15 years, I have been collaborating with colleagues across the country to understand the effect of co-morbid depression in various systems of healthcare, patterns of treatment for depression in individuals with chronic physical and mental illness, and how management of depression affects the management of chronic conditions using paid claims from Medicare, Medicaid, and more recently, with administrative data from the Veterans Health Administration (VHA). One of the exciting and challenging aspects of working with claims data is defining normative standards of care and pharmaceutical adherence measures that can be meaningfully developed using only the information available in claims data. Therefore, my research has been both substantive and methodological.

Working with my colleagues at Rutgers University, we first documented the high rates of affective disorders in the HIV Medicaid population and evaluated how the treatment for depression affects the treatment for HIV among Medicaid beneficiaries with HIV. According to editorial comments in the Journal of General Internal Medicine in 2000, our research project was considered to be one of the few to demonstrate that treatment for depression was accompanied by better quality of HIV care and reduced total healthcare expenditures. We also found that although individuals with depression were less likely to be adherent to antiretroviral medications, treatment for depression was associated with increased adherence to regimens of life-saving antiretroviral medications.

Continuing along the same lines of inquiry, we explored depression treatment patterns among individuals with diabetes, based on Medicaid data from four states (Alabama, Georgia, New Jersey, and Wisconsin). Although the use of SSRIs has been shown to be beneficial for glycemic control, we found that among individuals with diabetes, nearly a quarter of patients were still prescribed the older tricycle agents, suggesting poor quality of care for patients with diabetes and depression.

In addition, among individuals with diabetes in naturalistic settings, it is not clear whether depression treatment actually improves the process and intermediate outcomes of care, and decreases utilization of medical care and costs over time. Therefore, working with colleagues at the VHA, and using merged Medicare claims and VHA administrative data, we explored the patterns of use and cost-effectiveness of antidepressant treatment for depression. VHA is a national staff-model health maintenance organization (integrated healthcare delivery system) organized into 22 regions. Complete diagnostic and intermediate outcomes (laboratory results) from medical care encounters of veterans are captured through the electronic medical record system and are available to researchers for analysis.

Our analysis of merged Medicare and VHA administrative data indicated that only 60% of veterans with diabetes received guideline-consistent antidepressant treatment. Our findings suggested that competing clinical demands, defined as presence of multiple chronic physical and mental illness, may reduce the likelihood of antidepressant treatment among depressed veterans with diabetes. Our findings also substantiated that clinically modifiable factors (e.g., treatment by a mental health specialist, prescription of new generation antidepressants and good glycemic control prior to depression onset) could be targeted to improve rates of guideline-consistent antidepressant treatment.

Further exploration of the association between guideline-consistent depression treatment and healthcare expenditures among individuals with diabetes revealed that even after controlling for clustering and other independent variables, veterans with guideline-consistent depression treatment had reduced expenditures compared to those without any treatment for incident depression. Our findings suggest that aggressive efforts are needed to promote guideline-consistent antidepressant treatment for depression among those with chronic illnesses such as diabetes to improve clinical outcomes and capture cost savings. I have expanded this line of inquiry to include other cardiovascular conditions such as hypertension and heart disease and received funding from the VHA to analyze the healthcare of women veterans with chronic physical illness and depression, and compare their care to male veterans.

In summary, my collaborative research on the care of patients with chronic physical and mental illness, using large administrative databases, advances health services research and provides policy directives as these patients are typically excluded from clinical trials. Findings from the studies suggest that payors should incorporate depression treatment as an important aspect of chronic illness care to achieve better quality of care and cost savings.

Usha Sambamoorthi, PhD, is Director, Health Outcomes, Center for Healthcare Knowledge Management at the New Jersey VA Healthcare System; adjunct associate professor, Division of Health Systems and Policy, UMDNJ-School of Public Health; and adjunct professor, Department of Community Health and Preventive Medicine, Morehouse School of Medicine, Atlanta, GA.