Healthcare costs have increased significantly in the United States over the past 4 decades, essentially tripling every 20 years.1 Although the rate of growth in healthcare spending decreased in the 1990s with the onset of managed care, the rate of increase again accelerated during the past 12 years,2 at least in part, fueled by the costs of providing treatment to uninsured patients. Healthcare costs accounted for 16.2% of the gross domestic product (GDP) and exceeded $2.2 trillion dollars in 2007.
The economic burden of rising healthcare costs affects employers and government payers. Forty-six percent of the cost was public health expenditure, 42% was private insurance coverage, and 12% was through out-of-pocket spending, which includes nursing home, home health, dental services, durable medical products, and other professional services.3 The current economic crisis has provided additional impetus for the development and implementation of innovative interventions that can make healthcare more efficient and cost-effective.
Finding ways to manage the increasing cost of healthcare is critical and not only in the United States. The actuarial consulting firm, Watson Wyatt, reported in February 2008 that medical costs for employers were expected to continue to increase over the next 5 years through-out the world.4 Table 1 was prepared from its data of worldwide health care cost increases in 2006 and 2007, with projections for 2008.
Table 1. Worldwide Employer Medical Cost Increases
The Watson Wyatt report continued, “Rising medical costs have rapidly become a global issue that reaches far beyond the United States and other developed economies. … Many of the factors causing U.S. employers to experience significant increases in their healthcare costs — such as increased utilization, expensive medical technology and an aging population — are having comparable effects throughout the world.”4
It has been reported that 33% of private healthcare cost is due to inpatient services, with an additional 30% from physician and clinical services; whereas 37% of public healthcare cost is due to inpatient services, with 16% from physician and clinical services.5 With an aging population and a possible move toward expanded health coverage in the United States, it is unlikely the demand for care will diminish. It is also unlikely the cost of medical technology will significantly decrease. As a result, methods designed to optimize the effectiveness of medical interventions and to minimize complications are an essential component of a comprehensive cost-management strategy for healthcare.
One of the most widely replicated findings with respect to factors that worsen medical outcomes is the deleterious effect of comorbid psychiatric illness. Thomas et al reported an average 177% increase in medical health-care costs for Medicaid clients with a comorbid psychiatric diagnosis versus those clients with only a medical condition.6 Because major depressive disorder (MDD) is the most common psychiatric diagnosis in medical settings, it has received a great deal of research attention. MDD has been found to worsen outcomes in a variety of medical conditions, ranging from coronary artery disease7 to osteoporosis.8 The physiological mechanisms, or the “how,” for these phenomena is discussed throughout this issue. The current study uses records from commercially insured populations to illustrate and extend basic knowledge about the effect of comorbid psychiatric disorders on medical illness.
Health maintenance organization medical service records for 139,103 private sector clients between 2007 and 2008 were analyzed archivally. Data elements included age, gender, location of service (IP, ER, OP), date of service, and diagnoses (primary, secondary, and tertiary). Ethnicity and socioeconomic status were not available. Clients were grouped by gender and by five age groups: 0 to 9 years, 10 to 18 years, 19 to 39 years, 40 to 59 years, and 60 years and older.
International Classification of Disease (ICD)-9 diagnostic codes were grouped into 13 general medical categories (Cardiovascular, Dermatologic, Endocrine, ENT, Fractures/Injuries, Gastrointestinal [GI], Genitourinary/Gynecologic [GU/GYN], Hematologic, Infectious, Neoplasm, Neurologic, Obstetrics [OB], and Orthopedic);8 medical subgroups (Arthritis, Asthma, Coronary Artery Disease [CAD], Congestive Heart Failure [CHF], Chronic Obstructive Pulmonary Disorder [COPD], Diabetes, Hypertension [HTN], and Stroke); one general psychiatric category; and five psychiatric subgroups (Bipolar Disorder, Major Depressive Disorder, Personality Disorder, Schizophrenia, and Substance Abuse). Clients were considered to be positive for a medical diagnosis if they had a least one submitted claim with the appropriate ICD-9 code in a primary diagnostic field. A client was considered positive for a psychiatric diagnosis if they had a least one submitted claim with the appropriate psychiatric diagnosis in a primary, secondary, or tertiary diagnostic field. General groups and subgroups were not exclusive, thus a client with a diagnosis of diabetes would be positive for both a diagnosis of diabetes and for the general endocrine category. Diagnostic groupings are summarized in Table 2 (see page 1017).
Table 2. Diagnostic Groupings
For each medical diagnostic group, the annual average cost of care was determined for clients in each gender and age group. In each grouping, the target set was defined as all clients with a medical diagnosis, as well as a concomitant psychiatric diagnosis. The reference set was defined as all clients with the medical diagnosis without a concomitant psychiatric diagnosis. To offset the effect of inflationary increase in medical costs between 2007 and 2008, the factor increase in medical costs between each target group and control group was used for analysis. The average annual cost increase factor (AACIF) was calculated by using the average annual cost for clients with a psychiatric diagnosis, compared with the average annual cost for clients without a psychiatric diagnosis. (To simplify the results tables, the actual average annual costs have been converted to a relative cost unit [RCU], with each unit representing $2,000 above the least expensive cost, set as 1.)
Of the 139,103 private sector clients, 62,891 (45.2%) were male, and 76,212 (54.8%) were female, with an average age of 35.0 years for males and 36.0 years for females. Summary data also indicated that 14.6% of female clients and 12.9% of male clients had a psychiatric diagnosis in a primary, secondary, or tertiary diagnostic field.
Findings Related to Any Medical Diagnosis
In general, the presence of a concomitant psychiatric diagnosis produced a significant increase in average annual cost over clients without a psychiatric diagnosis. The average annual cost increase factor (AACIF) was 2.27 (t =−17.00, P < .001) for females and 2.28 (t=−11.92, P < .001) for males.
The increase was more pronounced when the psychiatric diagnosis was a substance abuse diagnoses, the most prevalent psychiatric subgroup. Analysis indicated that 3.4% of males had a concomitant substance abuse diagnosis and an AACIF of 3.05 (t =−12.05, P < .001). Concomitant substance abuse diagnoses occurred in 2.3% of females with an AACIF of 3.09 (t =−14.61, P < .001).
All five psychiatric subgroup diagnoses (Substance Abuse, Major Depressive Disorder, Bipolar Disorder, Schizophrenias, and Personality Disorders), had AACIFs from 2.11 to 3.51, with one group (females with a concomitant diagnosis of schizophrenia) having an AACIF of 8.95 (t = −17.43, P < .001). The only psychiatric subgroup without a significant AACIF was males with a concomitant schizophrenia diagnosis 2.25 (t=−1.23, P > .05). The results are summarized in Figure 1 (see page 1014).
Figure 1. Annual Average Cost Increase Factors (AACIF) by Psychiatric Concomitant Diagnoses.
Findings Related to General Medical Diagnoses
The 13 general medical diagnostic groups (Cardiovascular, Dermatologic, Endocrine, ENT, Fractures/Injuries, GI, GU/GYN, Hematologic, Infectious, OB, Orthopedic, Neoplasm, and Neurologic) with a concomitant psychiatric diagnosis had positive AACIFs ranging from 1.49 (t =−7.63, P > .001) for males, with a primary neoplasm diagnosis to 2.77 (t =−10.80, P <. 001) for females with an infectious disease diagnosis. The prevalence of concomitant psychiatric diagnoses ranged from 6.5% to 12.6%, and all comorbid groups differed significantly from the reference groups (medical diagnosis without concomitant psychiatric diagnosis). The results are summarized in Figure 2 (see page 1017).
Figure 2. AACIF for General Medical Conditions.
Findings Related to Specialized Medical Diagnostic Groups
The eight specialized medical diagnostic groups (Arthritis, Asthma, CAD, CHF, COPD, Diabetes, HTN, and Stroke) with concomitant psychiatric diagnoses also had positive AACIFs ranging from 1.19 (t =−2.34, P < .05) for males with a CHF to 3.80 (t = −2.60, P < .01) for females with a primary diagnosis of stroke. Prevalence of concomitant psychiatric diagnoses ranged from 5.8% to 12.5%. All but one group, females with CHF, differed significantly from the reference groups (medical diagnosis without concomitant psychiatric diagnosis). These results are summarized in Figure 3 (see page 1018).
Figure 3. AACIF for Specific Medical Conditions.
Findings Related to Age
The five age groups, with the exception of males younger than 19 years, had significant AACIFs for both genders, ranging from 1.94 (t = −6.84, P <. 001) for females between the 19 and 39 years to 3.29 (t = −4.94, P < .001) for females younger than 10 years. These results are summarized in Figure 4 (see page 1018).
Figure 4. AACIF by Age Groups.
A number of studies have reported increased medical costs for children with concomitant psychiatric disorders.9 Studies have reported similar effects in adults with specific psychiatric comorbid disorders.10 This study has demonstrated significant increases in medical cost of general and specific psychiatric concomitant diagnoses on 13 general primary medical diagnoses, as well as eight specific medical diagnostic groups. The increased cost, with few exceptions, has been significant for both male and female groups, ranging from 1.5 to 3 times the cost of patients without concomitant psychiatric diagnoses.
The study was restricted to health maintenance organization claims data for private sector medical services and is limited to data elements available. It also has the limitation of potential underdiagnosis of psychiatric conditions; however, the focus on diagnoses established by medical providers makes it better suited to design more effective interventions in the medical arena. The inclusion of public sector data, emergency room and inpatient data, additional demographics, and further analysis of age/medical diagnostic grouping relationships may provide more definitive guidelines for the development of effective interventions; however, the current study provides a compelling cost-of-care context for other articles in this issue.
Implications of these findings for clinical practice and program design are noteworthy, especially in an era of healthcare reform. First, the increased costs associated with comorbid medical and psychiatric conditions are clearly demonstrated by this and other studies in the literature. These costs are addressable through early screening for mental health and substance abuse conditions and comprehensive treatments plans designed to address the medical and psychiatric/substance abuse conditions present. Second, the significant prevalence of psychiatric conditions suggests that primary care and specialty medical practice should be designed to accommodate treatment of the whole person, meaning mind and body. Practical steps should be taken to incorporate psychiatric screening, mental health education, and streamlined referral and consultation procedures into clinic and institutional settings. Likewise, psychiatric/psychological practices should be alert to the likely presence of comorbid medical conditions and continuously monitor the well being and compliance with medical treatment of those undergoing behavioral health treatment.
Lastly, the complexities of managing comorbid medical and psychiatric practice, particularly in light of typical insurance regulated pre-authorization practices, restricted networks and benefit limitations, suggest new opportunities for primary care and specialty medical practice to collaborate with managed care companies. Medical practices and managed care companies should better leverage case management services and behavioral health specialty providers to ensure timely and effective intervention and case handling of comorbid patients. Doing so holds the promise of better clinical outcomes and reduced costs, which optimizes the medical economics of healthcare delivery.
- Congressional Budget Office Testimony of Peter R Orszag, Director on Growth in Health Care Costs, before the Committee on the Budget, U.S. Senate. January31, 2008.
- Tracking Health Care Costs, Trends Stabilize but Remain High in 2002, Data Bulletin of the Center for Studying Health System Change (HSC), No 25, June2003.
- http://www.cms.hhs.gov/NationalHealthExpendData/downloads/highlights.pdf. Accessed November 25, 2009.
- Medical Cost Increase to Accelerate Worldwide Over Next Five Years, Watson Wyatt Poll Finds. http://www.watsonwyatt.com/news/press.asp?ID=18622. Accessed November 25, 2009.
- Keeping Healthcare Affordable, Healthcare Trends in America: A Reference Guide from Blue Cross Blue Shield Association. 2009.
- Thomas MR, Waxmonsky JA, Gabow PA, Flanders-McGinnis G, Socherman RS, Rost K. Prevalence of psychiatric disorders and cost of care among adult enrollees in a Medicaid HMO. Psychiatr Serv. 2005;56(11):1394–1401. doi:10.1176/appi.ps.56.11.1394 [CrossRef]
- Carney RM, Rich MW, Friedland et al. Major depressive disorder predicts cardiac events in patients with coronary artery disease. Psychosom Med. 1988;50(6):627–630.
- Kahl KH, Rudolf S, Stoeckelhuber BM, et al. Bone mineral density, markers of bone turnover, and cytokinase in young women with borderline personality disorder with and without comorbid major depressive disorder. Am J Psychiatry. 2005;162(1):168–174. doi:10.1176/appi.ajp.162.1.168 [CrossRef]
- Mandell DS, Guevara JP, Rostain AL, Hadley TR. Economic Grand Rounds: Medical Expenditures Among Children With Psychiatric Disorders in a Medicaid Population. Psychiatr Serv. 2003;54(4):465–467. doi:10.1176/appi.ps.54.4.465 [CrossRef]
- Johnson J, Weissman M, Klerman G. Service utilization and social morbidity associated with depressive symptoms in the community. JAMA. 1992;267(11):1478–1483. doi:10.1001/jama.267.11.1478 [CrossRef]
Worldwide Employer Medical Cost Increases
|Middle East, Africa, Russia|
|Saudi Arabia||South Africa||Russia|
|North and Central America|
|General Medical Diagnostic Groups||Medical Specialized Diagnostic Groups||Psychiatric Diagnostic Groups|
|Endocrine||Coronary Artery Disease (CAD)||Major Depressive Disorder|
|Ear, Nose, and Throat (ENT)||Congestive Heart Failure (CHF)||Personality Disorder|
|Fractures/Injuries||Chronic Obstructive Pulmonary Disorder (COPD)||Schizophrenia|
|Gastrointestinal (GI)||Diabetes||Substance Abuse|
|Genitourinary/Gynecologic (GU/GYN)||Hypertension (HTN)|