Journal of Mental Health

ISSN: 0963-8237 (Print) 1360-0567 (Online) Journal homepage: http://www.tandfonline.com/loi/ijmh20

The relationships among depression, physical health conditions and healthcare expenditures for younger and older Americans Sunha Choi, Sungkyu Lee, Jason Matejkowski & Young Min Baek To cite this article: Sunha Choi, Sungkyu Lee, Jason Matejkowski & Young Min Baek (2014) The relationships among depression, physical health conditions and healthcare expenditures for younger and older Americans, Journal of Mental Health, 23:3, 140-145 To link to this article: http://dx.doi.org/10.3109/09638237.2014.910643

Published online: 06 May 2014.

Submit your article to this journal

Article views: 85

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ijmh20 Download by: [University of Otago]

Date: 09 November 2015, At: 00:28

http://informahealthcare.com/jmh ISSN: 0963-8237 (print), 1360-0567 (electronic) J Ment Health, 2014; 23(3): 140–145 ! 2014 Shadowfax Publishing and Informa UK Limited. DOI: 10.3109/09638237.2014.910643

ORIGINAL ARTICLE

The relationships among depression, physical health conditions and healthcare expenditures for younger and older Americans Sunha Choi1, Sungkyu Lee1, Jason Matejkowski2, and Young Min Baek3 College of Social Work, The University of Tennessee, Knoxville, TN, USA, 2The Center for Mental Health Research and Innovation and School of Social Welfare, University of Kansas, Lawrence, KS, USA, and 3Department of Communication, Yonsei University, Seoul, Korea

Downloaded by [University of Otago] at 00:28 09 November 2015

1

Abstract

Keywords

Background and aims: Little is known about the extent depression adds to the costs of treatment for physical health conditions. This study examined the paths and the extent to which depression in conjunction with a physical health problem is associated with an increase in healthcare expenditures and how that is different between younger and older adults. Methods: Data from the 2007 Medical Expenditure Panel Survey (MEPS) were analyzed. Depression status and physical health conditions were identified through ICD-9 codes. The multiple group structural equation modeling (SEM) was employed to examine the moderated mediation effects. Results: Approximately 11% of adults had clinical depression. The multiple group SEM for both younger and older adult groups supports not only a direct effect of depression on expenditures but also an indirect effect via comorbid health conditions. Furthermore, the indirect effect was significantly more prominent among older respondents than among younger ones, indicating significant moderated mediation by age. Conclusions: Depression has greater effects on comorbid health conditions and an increase in total healthcare expenditures through comorbid conditions among older adults. Findings of this study suggest that proper detection and treatment of depression is beneficial in reducing overall healthcare expenditures, especially among older adults.

chronic health conditions, comorbidity, depression, healthcare expenditures

Introduction Individuals with depression and individuals with chronic health conditions independently incur higher healthcare costs than their counterparts without such conditions (Gameroff & Olfson, 2006; Kim & Lee, 2006; Luber et al., 2001). Accordingly, people with both depression and comorbid health conditions can be assumed to have higher healthcare expenditures when compared to those who have either depression or a health condition alone or who have neither a health condition nor depression (Unu¨tzer et al., 2009). The purpose of this study was to examine the extent to which depression in conjunction with a physical health problem is associated with an increase in healthcare expenditures. However, depression is also a risk factor for developing physical health problems such as diabetes and heart disease (Druss & Walker, 2011; Egede et al., 2002; Katon, 2003; Patten et al., 2008), and the presence of depression can therefore have a substantial influence on healthcare costs by

History Received 19 September 2013 Revised 7 January 2014 Accepted 20 March 2014 Published online 6 May 2014

increasing the chances of experiencing a co-occurring chronic health condition. Thus, rather than hypothesizing independent effects of depression and comorbid health conditions on healthcare expenditures, we hypothesized a mediating relationship and tested the direct and indirect effects of depression through physical comorbidity on healthcare expenditures (Figure 1). Furthermore, to inform targeted healthcare policies and practices, this mediating relationship was tested in different age groups (i.e. younger versus older adults; moderated mediation). Even though depression and other psychiatric conditions tend to be less prevalent as age increases (Kessler et al., 1994), among depressed individuals, the prevalence of comorbid health condition increases with age (Finkelstein et al., 2003). Thus, we hypothesized that the magnitude of the mediating relationship of depression on healthcare expenditures through comorbid health conditions would be greater among older adults.

Methods Data source and sample Correspondence: Sungkyu Lee, PhD, College of Social Work, The University of Tennessee, 301 Henson Hall, 1618 Cumberland Ave., Knoxville, TN 37996, USA. Tel: 865-974-3164. Fax: 865-9743701. E-mail: [email protected]

Data were obtained from the 2007 Medical Expenditure Panel Survey (MEPS). The MEPS is a nationally representative survey of the U.S. civilian non-institutionalized population on

Healthcare expenditures

DOI: 10.3109/09638237.2014.910643

Moderator: Age (Younger vs. Older adult groups)

a

Comorbid Health Conditions

Depression Status

c

b

Healthcare Expenditures

Downloaded by [University of Otago] at 00:28 09 November 2015

Figure 1. The hypothesized moderated mediating model.

health care use, expenditures, and insurance coverage (AHRQ, 2009b). The sampling frame for the MEPS was drawn from the participants of the previous year’s National Health Interview Survey using a stratified multistage area probability sampling design with ethnic minorities (such as Hispanics and Asians) and low-income families oversampled (AHRQ, 2008). In this study, we analyzed data collected from this nationally representative sample of 16 384 individuals including 13 207 younger adults (18–64) and 3177 older adults (65+). Measures Healthcare expenditures Total all-cause healthcare expenditures served as the main dependent variable. The MEPS covers a broad range of healthcare expenditures, including office-based visits, outpatient visits, emergency room visits, dental visits, inpatient stays, home healthcare, prescription drugs, and other expenditures including vision aids and medical supplies. For each healthcare category, expenditures were defined as the sum of payments for care provided regardless of payment source (Mohanty et al., 2005). Finally, the total expenditures were calculated as the sum of expenditures from all health service categories during a given year. The MEPS excludes payments for non-prescription drugs and for alternative medicine. To account for positive skewness of healthcare expenditure, a log transformation was conducted for statistical tests. Depression and comorbid physical conditions The MEPS asked participants to verbally describe their health status if the conditions were associated with service utilization or disability days, or if a respondent was bothered by the conditions. Then expert coders assigned ICD-9-CM codes to each reported condition per participant. Since its initiation in 1996, the MEPS has used ICD-9-CM codes (AHRQ, 2009a). The ICD-9-CM was the official medical diagnosis codes used in the U.S. when the 2007 MEPS data were collected (AHRQ, 2009a; Centers for Disease Control and Prevention, 2013). For this study, following guidelines described in previous studies using the MEPS (Chung, 2005; Egede et al., 2002; Harman et al., 2004b), the ICD-9 code of 311 was used to identify people with clinical depression. Since the publicly available MEPS dataset does not provide detailed diagnosis, the ‘‘296’’ codes were excluded because it could also include people with bipolar disorder. However, in most cases, people

141

in this code also tended to have a code of ‘‘311’’ (Harman et al., 2004b). Self-reported physical health conditions were also matched with ICD-9 codes and the Charlson Comorbidity Index (CCI) was employed to account for the severity of physical health conditions per participant (Charlson et al., 1987; McGregor et al., 2005; Sundararajan et al., 2004). The CCI was used to rank physical conditions by assigning weights of 1, 2, 3, or 6 based on the risk of mortality associated with each condition. The scores were then summed per individual and the total score was used to represent the severity of physical health conditions for an individual. For example, an individual would have a total score of 7 if she or he suffered from rheumatologic disease (¼1 point) and metastatic solid tumor (¼6 points). The CCI has been tested valid and reliable with various populations (Charlson et al., 1987; McGregor et al., 2005; Sundararajan et al., 2004). Socioeconomic and demographic characteristics Socioeconomic and demographic covariates comprising the final models included age, gender, race (Caucasian, African American, and other), education, individual income, and insurance status. Age (in years), income (including both taxable and non-taxable income), and education (years completed) were employed as continuous variables. We operationalized insurance status using four dichotomous variables (yes/no): having Medicare, Medicaid, and private insurance, and being uninsured for all of the past 12 months. Statistical analysis Descriptive statistics were computed to describe and compare the sample’s socioeconomic, demographic, and clinical characteristics by age (younger versus older adults) and depression status (depressed versus non-depressed). In addition, healthcare expenditures were compared by age, depression status, and CCI score (0 versus 1+). To test the hypothesized moderated mediation effect, we employed multiple group structural equation modeling (SEM) and controlled for the effects of socioeconomic and demographic characteristics. To account for the complex survey design of the MEPS, survey procedures in SAS 9.3 were utilized for calculating descriptive and bivariate statistics. In SEM, the maximum likelihood estimation with robust standard errors (MLR) was used in conjunction with sample weights to adjust for missing data (Muthe´n & Muthe´n, 2009). To examine the goodness-of-fit in models, we used chi-square test, Comparative Fit Index (CFI; good model  0.95), and Root Mean Square Error of Approximation (RMSEA; good model  0.06). SEM analyses were conducted using M-Plus 6.11 version.

Results Findings indicate that 11.3% of the non-institutionalized U.S. adult population is estimated to have clinical depression based on the ICD-9 criteria. Table 1 shows the socioeconomic, demographic, and comorbid health characteristics of the sample (N ¼ 16 384). The mean age of the sample was 47.9 years old. While 54.4% of the sample was female, a larger

142

S. Choi et al.

J Ment Health, 2014; 23(3): 140–145

Table 1. Sample characteristics by age and depression status: weighted mean or % (SE). Younger adults (n ¼ 13 207)

Downloaded by [University of Otago] at 00:28 09 November 2015

Variables

All (N ¼ 16 384)

Socioeconomic and demographic characteristics Age 47.9 (0.2) Gender Female 54.4 (0.3) Male 45.6 (0.3) Race Caucasians 83.0 (0.6) African Americans 10.6 (0.5) Other 6.4 (0.4) Education (years) 14.8 (0.04) Income ($) 35 628 (418.9) Insurance status (¼yes) Medicare 20.8 (0.5) Medicaid 5.6 (0.2) Private 61.3 (0.6) Uninsured 11.7 (0.4) Comorbid conditions Charlson Comorbidity Index (CCI) 0 82.0 (0.4) 1 12.5 (0.3) 2 4.1 (0.2) 3 or more 1.5 (0.1) CCI score (Mean, (SE)) 0.26 (0.01)

Older adults (n ¼ 3177)

Depressed (n ¼ 1582)

Non-depressed (n ¼ 11 625)

Depressed (n ¼ 355)

Non-depressed (n ¼ 2822)

44.1 (0.4)

41.3 (0.2)

74.1 (0.3)

74.8 (0.2)

63.6 (1.2) 36.4 (1.2)

52.2 (0.5) 47.8 (0.5)

69.1 (1.7) 30.9 (1.7)

56.7 (0.8) 43.3 (0.8)

86.2 8.7 5.1 14.8 30 055

(1.0) (0.7) (0.6) (0.1) (964.4)

81.4 11.6 7.0 15.1 38 345

(0.7) (0.5) (0.4) (0.05) (504.2)

90.2 4.5 5.4 13.8 24 379

(0.9) (0.7) (0.6) (0.1) (919.3)

86.9 8.5 4.6 14.0 28 495

(0.8) (0.6) (0.6) (0.1) (721.3)

10.5 11.5 53.0 15.7

(0.8) (0.9) (1.6) (1.1)

2.8 4.6 66.8 14.3

(0.2) (0.2) (0.6) (0.5)

92.2 9.1 44.5 0.4

(0.7) (1.1) (1.7) (0.01)

93.8 6.3 44.2 0.4

(0.5) (0.5) (1.3) (0.1)

76.3 18.4 3.7 1.7 0.32

(1.3) (1.1) (0.5) (0.3) (0.02)

87.0 9.9 2.5 0.7 0.17

(0.4) (0.3) (0.2) (0.1) (0.01)

50.5 25.0 13.7 10.8 0.87

(2.1) (1.5) (1.8) (1.5) (0.04)

68.2 18.8 9.7 3.3 0.49

(1.0) (0.8) (0.7) (0.4) (0.02)

Table 2. Healthcare expenditures by service type, age group, and depression status: weighted mean (SE). Younger adults (n ¼ 13 207) Service type Total healthcare expenditures ($) Office-based visits Outpatient visits Emergency room visits Inpatient stays Dental visits Home healthcare Prescription drugs Other expenditures

All (N ¼ 16 384) 5388.2 1266.8 473.5 170.1 1727.8 307.3 160.1 1181.1 101.6

(121.6) (45.5) (22.5) (7.6) (82.7) (8.3) (12.2) (27.1) (5.8)

Depressed (n ¼ 1582) 7829.4 1631.8 757.8 281.9 2271.0 294.0 193.4 2310.9 88.6

Non-depressed (n ¼ 11 625)

(556.2) (90.8) (128.9) (34.9) (382.0) (20.9) (51.7) (142.5) (6.5)

majority of depressed younger adults (63.6%) as well as of depressed older adults (69.1%) were female. In addition, among both younger and older age groups, disproportionately higher percentages of depressed individuals were Caucasians and had Medicaid, compared with their counterparts without depression. It is also worthwhile to note that among younger adults, 10.5% of those who met criteria for depression were covered by Medicare, while Medicare coverage was only 2.8% among younger, non-depressed adults. The majority of the sample (82.0%) reported that they did not have any physical health condition in a given year. However, when disaggregated by age and depression status, about half of older adults with depression reported physical health conditions, while only about a quarter of younger adults with depression reported physical health conditions. Overall, older adults spent $9786 for healthcare costs, while younger adults spent $4354 on healthcare (p50.001; results not shown in table). Further breakdown of the total healthcare expenditures by age and depression status

3917.0 1085.6 395.6 159.3 1085.4 302.2 51.4 761.9 75.5

(123.7) (59.5) (21.9) (8.5) (75.9) (9.3) (7.8) (23.1) (5.5)

Older adults (n ¼ 3177) Depressed (n ¼ 355) 17 147.0 2617.9 894.4 273.7 7754.4 424.0 883.1 3838.4 461.3

(781.6) (258.8) (58.8) (69.7) (541.6) (22.0) (133.3) (322.2) (70.0)

Non-depressed (n ¼ 2822) 8794.1 1664.6 597.4 141.8 3377.1 320.7 510.2 2010.6 171.7

(269.8) (58.2) (43.5) (10.5) (203.4) (22.1) (52.9) (65.7) (21.2)

(Table 2) indicates that for both groups, depressed individuals had greater health care expenditures when compared to those without depression ($7829 versus $3917 for younger group, p50.001; $17 147 versus $8794 for older groups, p50.001). In terms of the healthcare expenditures on different types of services, prescription drugs and inpatient care were the two greatest contributors to the total healthcare expenditures among depressed individuals in both younger and older groups. In addition, the amounts spent on these two service types were much greater than the amounts spent by their nondepressed counterparts. For example, while the mean amount spent on prescription drugs among non-depressed younger adults was $761.9, it was $2310.9 among their depressed counterparts (p50.001). The same relationship was observed when comparing older adults without depression to older adults with depression ($2010.6 versus $3838.4, p50.001). However, there were no differences between individuals with and without depression in the amounts spent on dental care both among older and younger individuals, while no

Healthcare expenditures

DOI: 10.3109/09638237.2014.910643

143

Downloaded by [University of Otago] at 00:28 09 November 2015

Figure 2. Bivariate comparison of the total healthcare expenditures ($) by Charlson Comorbidity Index (CCI) score, age group, and depression status: Mean.

Table 3. Results of multiple group structural equation modeling: standardized path coefficients.

Younger adults Causal paths Depression to Healthcare expenditure (logged) 0.093*** Comorbidity 0.184*** Comorbidity to Healthcare expenditure (logged) 0.061*** Decomposing depression’s effect on healthcare Indirect effect via Comorbidity 0.011*** Direct effect 0.093*** Total effect 0.104***

Older adults

Equality test between two groups

0.138*** 0.271***

0.021 13.192***

0.138*** expenditure 0.037*** 0.138*** 0.176***

40.938*** 6.108* 0.021 0.692

*p50.05, ***p50.001.

differences were found by depression status in the expenditures spent for outpatient and emergency room visits only among older adults (p40.05). Figure 2 displays the comparison of healthcare expenditures further by comorbidity, depression status, and age group. For example, among depressed individuals, the presence of comorbid health condition increased health care expenditure for both the younger group ($5853 versus $9827, p50.001) and the older group ($11 548 versus $19 378, p50.001). Table 3 shows the results of SEM after controlling for covariates. The multiple group SEM fits the data well (CFI ¼ 1.000 and RMSEA50.001). Although the chi-square test of model fit was significant, 2ð54Þ ¼ 2899.735, p50.001, this significance may be due to the large number of parameters estimated and the large sample size (Bollen, 1989). Individuals with depression had higher healthcare

expenditures compared to people without depression among both age groups (i.e. a direct effect; younger ¼ 0.093 and older ¼ 0.138, p50.001, respectively). Likewise, depression was associated with higher levels of healthcare expenditures through the presence of comorbid health conditions (i.e. an indirect effect; younger ¼ 0.011 and older ¼ 0.037, p50.001, respectively). In other words, people with depression were more likely to have comorbid health condition (i.e. younger ¼ 0.184 and older ¼ 0.271, p50.001, respectively) and having comorbid health conditions tended to increase healthcare expenditures (i.e. younger ¼ 0.061 and older ¼ 0.138, p50.001, respectively). When the magnitudes of the mediating relationship was compared for statistical differences between younger and older adult groups, the indirect effect of depression on expenditures through comorbid medical conditions was greater among older adults than among younger adults, 2ð1Þ ¼ 6.108, p50.05, although no difference was found in the magnitude of the direct effect, 2ð1Þ ¼ 0.021, and the total effect, 2ð1Þ ¼ 0.692, between the two groups. Both paths that consist of indirect effects (i.e. paths between depression to comorbidity and between comorbidity to health care expenditures) represented greater coefficients among older adults than among younger adults, 2ð1Þ ¼ 13.192, p50.001; 2ð1Þ ¼ 40.938, p50.001, respectively.

Discussion In a nationally representative survey of the U.S. civilian non-institutionalized population, depression was found to be associated with higher total healthcare costs both among younger and older adults. Moreover, this study contributes to prior research on this topic by revealing that depression increases healthcare costs through its relationship with co-occurring health conditions. These relationships were significantly stronger among older, rather than younger,

Downloaded by [University of Otago] at 00:28 09 November 2015

144

S. Choi et al.

adults (i.e. indirect.younger ¼ 0.011 and indirect.older ¼ 0.037). On average, the presence of depression increased annual healthcare costs 85% (from $6227.5 to $11 548.00) in older adults without co-occurring health conditions. These increases were reflective of higher expenditures for depressed individuals across most types of healthcare services, particularly the expenditures for inpatient stays and prescription drugs. The addition of one or more co-occurring health conditions more than tripled healthcare costs (to $19 378.00). Findings suggest significant practice and policy implications involving treatment for depression. The impact of depression on healthcare can negatively affect the financial well-being of older adults through wealth depletion in later life (Kim & Lee, 2006) and add substantially to Medicare claims. As the U.S. population continues to age, approaches for reducing depression-related morbidity will become increasingly important methods for reducing Medicare expenditures. Although late-life depression is a treatable condition (Ciraulo et al., 2011; Thompson et al., 2001), disparities in depression treatment for older adults have been reported (Harman et al., 2004a). Early detection and treatment of depression could reduce these financial burdens if this identification can result in measures that prevent early symptoms from progressing to full-blown depressive episodes and from contributing to the development of other chronic health conditions (Delaney et al., 2011). The review of the 14 randomized controlled trials which tested the effectiveness of screening on depression identification and treatment revealed positive effects of early screening and detection especially when they were combined with appropriate follow-up treatment (Pignone et al., 2002). As such, physical healthcare providers should take advantage of the more frequent physician visits recommended for older adults to not only assess for physical ailments but to also screen for depression and refer those at risk for depression to appropriate treatment (Fiske et al., 2009; O’Connor et al., 2009). For those who may already be experiencing a depressive episode, the use of collaborative care or care mangers approaches have been found to be cost-effective, especially in addressing the complex needs of patients with chronic, co-occurring conditions (Druss & Walker, 2011). The effectiveness of those treatment approaches has been supported in numerous randomized controlled trials (Gilbody et al., 2006; Katon et al., 2005; Schoenbaum et al., 2001; Unu¨tzer et al., 2002). Among younger adults, the presence of depression and at least one co-occurring health condition more than tripled average annual healthcare expenditures, in the same manner it did for older adults (from $3023.30 to $9826.60). Younger adults with depression were much more likely to have their healthcare covered by Medicaid and Medicare than younger adults without depression, suggesting that methods for improving the identification and treatment of depression among younger adults could incur substantial cost savings for public programs in much the same way they could for older adults. Limitations of the study include the utilization of the CCI to examine physical health conditions and its severity among the study sample. Originally, Charlson et al. (1987) developed

J Ment Health, 2014; 23(3): 140–145

this measure to predict one-year mortality of comorbid medical conditions based on administrative data. Although this measure has been broadly used in health care research and clinical practice, the CCI only accounts for 19 medical conditions and excludes some medical conditions, such as non-malignant hematologic disease (Hall et al., 2004). Second, future studies with longitudinal data would strengthen the implications of study findings regarding the relationships between depressions, comorbid health conditions, and accompanying healthcare expenditures. In addition, depression measure in this study was based on self-report and did not capture symptom severity.

Declaration of interest The authors declare no conflicts of interests. The authors alone are responsible for the content and writing of this article.

References AHRQ. (2008). Methodology report #22: Sample design of the Medical Expenditure Panel Survey household component, 1998–2007. Rockville (MD): AHRQ. AHRQ. (2009a). MEPS HC-112: 2007 Medical Conditions. Rockville (MD): AHRQ. AHRQ. (2009b). MEPS HC-113: 2007 full-year consolidated data file. Rockville (MD): AHRQ. Bollen KA. (1989). Structural equations with latent variables. New York (NY): John Wiley & Sons. Centers for Disease Control and Prevention. (2013). International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). Available from: http://www.cdc.gov/nchs/icd/ icd10cm.htm#icd2014 [last accessed 3 Jan 2014]. Charlson ME, Pompei P, Ales KL, MacKenzie CR. (1987). A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chron Dis, 40, 373–83. Chung S. (2005). Does the use of SSRIs reduce medical care utilization and expenditures? J Ment Health Pol Econ, 8, 119–29. Ciraulo DA, Evans JA, Qiu WQ, et al. (2011). Antidepressant treatment of geriatric depression Pharmacotherapy of depression, 2nd ed. New York (NY): Springer Science + Business Media, 125–83. Delaney C, Fortinsky R, Doonan L, et al. (2011). Depression screening and interventions for older home health care patients: Program design and training outcomes for a train-the-trainer model. Home Health Care Manage Pract, 23, 435–45. Druss BG, Walker ER. (2011). Mental disorders and medical comorbidity. Princeton (NJ): Robert Wood Johnson Foundation. Egede LE, Zheng D, Simpson K. (2002). Comorbid depression is associated wtih increased health care use and expenditures in individuals with diabetes. Diabetes Care, 25, 464–70. Finkelstein EA, Bray JW, Chen H, et al. (2003). Prevalence and costs of major depression among elderly claimants with diabetes. Diabetes Care, 26, 415–20. Fiske A, Wetherell JL, Gatz M. (2009). Depression in older adults. Ann Rev Clin Psychol, 5, 363–89. Gameroff MJ, Olfson M. (2006). Major depressive disorder, somatic pain, and health care costs in an urban primary care practice. J Clin Psychiatr, 67, 1232–9. Gilbody S, Bower P, Fletcher J, et al. (2006). Collaborative care for depression: A cumulative meta-analysis and review of longer-term outcomes. Arch Intern Med, 166, 2314–21. Hall W, Ramachandran R, Narayan S, et al. (2004). An electronic application for rapidly calculating Charlson comorbidity score. BMC Cancer, 4, 94. Harman JS, Edlund MJ, Fortney JC. (2004a). Disparities in the adequacy of depression treatment in the United States. Psychiatr Serv, 55, 1379–85. Harman JS, Kelleher KJ, Reynolds CF, Pincus HA. (2004b). Outof-pocket healthcare expenditures of older Americans with depression. J Am Geriatr Soc, 52, 809–13.

Downloaded by [University of Otago] at 00:28 09 November 2015

DOI: 10.3109/09638237.2014.910643

Katon WJ. (2003). Clinical and health services relationships between major depression, depressive symptoms, and general medical illness. Biol Psychiatr, 54, 216–26. Katon WJ, Schoenbaum M, Fan MY, et al. (2005). Cost-effectiveness of improving primary care treatment of late-life depression. Arch Gen Psychiatr, 62, 1313–20. Kessler RC, McGonagle KA, Zhao S, et al. (1994). Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: Results from the National Comorbidity Survey. Arch Gen Psychiatr, 51, 8–19. Kim H, Lee J. (2006). The impact of comorbidity on wealth changes in later life. J Gerontol Series B: Psychol Sci Soc Sci, 61, S307–14. Luber MP, Meyers BS, Williams-Russo PG, et al. (2001). Depression and service utilization in geriatric primary care patients. Am J Geriatr Psychiatr, 9, 169–76. McGregor JC, Kim PW, Perencevich EN, et al. (2005). Utility of the chronic disease score and Charlson comorbidity index as comorbidity measures for use in epidemiologic studies of antibiotic-resistant organisms. Am J Epidemiol, 161, 483–93. Mohanty SA, Woolhandler S, Himmelstein D, et al. (2005). Health care expenditures of immigrants in the United States: A nationally representative analysis. Am J Publ Health, 95, 1431–8. Muthe´n LK, Muthe´n BO. (2009). Mplus user’s guide, 5th ed. Los Angeles (CA): Muthe´n & Muthe´n.

Healthcare expenditures

145

O’Connor EA, Whitlock EP, Beil TL, Gaynes BN. (2009). Screening for depression in adult patients in primary care settings: A systematic evidence review. Ann Intern Med, 151, 793–803. Patten SB, Williams J, Lavorato DH, et al. (2008). Major depression as a risk factor for chronic disease incidence: Longitudinal analyses in a general population cohort. Gen Hospt Psychiatr, 30, 407–13. Pignone MP, Gaynes BN, Rushton JL, et al. (2002). Screening for depression in adults: A summary of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med, 136, 765–76. Schoenbaum M, Unu¨tzer J, Sherbourne C, et al. (2001). Cost-effectiveness of practice-initiated quality improvement for depression: Results of a randomized controlled trial. JAMA, 286, 1325–30. Sundararajan V, Hendersona T, Perrya C, et al. (2004). New ICD-10 version of the Charlson Comorbidity Index predicted in-hospital mortality. J Clin Epidemiol, 57, 1288–94. Thompson LW, Coon DW, Gallagher-Thompson D, et al. (2001). Comparison of desipramine and cognitive/behavioral therapy in the treatment of elderly outpatients with mild-to-moderate depression. Am J Geriatr Psychiatr, 9, 225–40. Unu¨tzer J, Katon W, Callahan CM, et al. (2002). Collaborative care management of late-life depression in the primary care setting: A randomized controlled trial. JAMA: J Am Med Assoc, 288, 2836–45. Unu¨tzer J, Schoenbaum M, Katon WJ, et al. (2009). Healthcare costs associated with depression in medically ill fee-for-service Medicare participants. J Am Geriatr Soc, 57, 506–10.

The relationships among depression, physical health conditions and healthcare expenditures for younger and older Americans.

Little is known about the extent depression adds to the costs of treatment for physical health conditions. This study examined the paths and the exten...
482KB Sizes 0 Downloads 3 Views