Mental Health Diagnoses and Services Utilization Vary by Wage Level

Sherman,Bruce;Lawrence,Debra;Kuharic,Maja;Chrones,Lambros;Patel,Sheetal;Touya,Maëlys;

Graphical Abstract: “Mental Health Diagnoses and Services Utilization Vary by Wage Level”

Infographic: “Effects of Employee Wage on Mental Health Care Use”

Objectives: The relationship between employee wage status and mental health care utilization has not been characterized in large-scale analyses. This study assessed health care utilization and cost patterns for mental health diagnoses according to wage category among employees with health insurance.

Study Design: This was an observational, retrospective cohort study for the year 2017 among 2,386,844 adult full-time employees (254,851 with mental health disorders; subgroup of 125,247 with depression) enrolled in self-insured plans in the IBM Watson Health MarketScan research database.

Methods: Participants were stratified into annual wage categories: $34,000 or less; more than $34,000 to $45,000; more than $45,000 to $69,000; more than $69,000 to $103,000; and more than $103,000. Health care utilization and costs were analyzed via regression analyses.

Results: Prevalence of diagnosed mental health disorders was 10.7% (9.3% in the lowest-wage category); prevalence of depression was 5.2% (4.2% in the lowest-wage category). Severity of mental health, and specifically depression episodes, was greater in lower-wage categories. All-cause utilization of health care services was higher in patients with mental health diagnoses vs the total population. Among patients with mental health diagnoses, specifically depression, utilization was highest in the lowest- vs highest-wage category for hospital admissions, emergency department visits, and prescription drug supply (all P < .0001). All-cause health care costs were higher in the lowest- vs highest-wage category among patients with mental health diagnoses ($11,183 vs $10,519; P < .0001), specifically depression ($12,206 vs $11,272; P < .0001).

Conclusions: Lower mental health condition prevalence and greater use of high-intensity health care resources highlight the need to more effectively identify and manage mental health conditions among lower-wage workers.

Mental health disorders place a substantial burden on the workplace and health care system. In the United States, 19% of adults had a mental illness in 2018.¹ Depression is common among employees and poses challenges in the workplace.^2-4 Employees with mood disorders often lose workdays^3,5 and productivity,^4,6 and mental illness is associated with physical health challenges and disability.⁴ Multiple studies have linked low income levels with greater mental health challenges.^7-11

Although treatment can improve outcomes for employees with mental health disorders,^6,12 many do not receive treatment.^4,6,13 In a national survey, American adults seeking mental health services reported that cost, stigma, and difficulty getting an appointment were barriers to seeking care.¹⁴ Difficulties with access are more acute at lower income levels and are linked to disparities in obtaining health care services. In one study, the proportion of employees receiving any medical service increased with wage level, although emergency department (ED) visits and hospitalizations were more frequent among lower-wage earners.¹⁵

Despite evidence that income levels affect both mental health status and health care utilization, the relationship between wage status and mental health service utilization has not been characterized in large-scale data analyses of commercially insured populations. Comprehensive evidence from such studies may provide a basis for policy and practice changes to reduce disparities in access to mental health services.

The objective of this study was to assess health care utilization and cost patterns for mental health diagnoses according to wage category among employees with employer-provided health insurance.

Additional details of the study methodology are provided in the eAppendix Methods (eAppendix available at ajmc.com). Takeda Pharmaceuticals U.S.A., Inc, and Lundbeck LLC funded the study and participated in study design and interpretation.

This was an observational retrospective cohort study using data from January 1, 2017, to December 31, 2017 (study period). Informed consent was not required because deidentified information from a claims database was used. The analysis included active employees in the United States who were continuously enrolled in self-insured, employer-sponsored health plans in the IBM Watson Health MarketScan research database.

Included individuals were employed benefits enrollees 18 years or older with employer-provided wage information during the study period.

Three participant groups were defined: (1) the total population: all participants; (2) patients with mental health diagnoses: participants who had an International Classification of Diseases, Tenth Revision (ICD-10) code for a psychiatric or substance use disorder and met IBM Watson Health’s Medical Episode Grouper¹⁶ criteria (Mental Health Conditions in eAppendix Methods); and (3) a subgroup of patients with depression: those patients from the prior group with a principal diagnosis of depression. The Medical Episode Grouper was used to identify patients meeting eligibility criteria for the respective mental health conditions (eAppendix Methods).¹⁶ Examples of Medical Episode Grouper use in other research include estimation of annual attributable medical cost analysis of substance use disorder in US hospitals with diagnoses by ICD-10, Clinical Modification (ICD-10-CM) code¹⁷ and assessment of changes in health care services during the initial phase of the COVID-19 pandemic using the IBM Watson Health procedure categories.¹⁸ In prior research, the Medical Episode Grouper was also used to assign disease severity to each episode based solely on ICD-9-CM diagnosis codes for condition-related complications (eAppendix Methods).¹⁹ The following wage categories were defined based on each participant’s annual wage in US$ at the midpoint of 2017: $34,000 or less; more than $34,000 to $45,000; more than $45,000 to $69,000; more than $69,000 to $103,000; and more than $103,000. The wage categories were intended to create quartiles, with the lowest-wage quartile split into 2 wage categories (≤ $34,000 and > $34,000-$45,000) for more detailed analysis.

End points are reported for the total population, patients with mental health diagnoses, and the subgroup of patients with depression. Demographic information was captured as described in the eAppendix Methods.

Prevalence of diagnosed mental health disorders was calculated as the proportion of participants meeting IBM Watson Health’s Medical Episode Grouper criteria for a mental health condition as a percentage of the total study population. Severity of a mental health episode was analyzed based on the highest disease severity (from 1.0 to 4.0) recorded for that episode (eAppendix Methods).

Mental health–specific utilization and costs were calculated for services with a principal diagnosis of mental health according to ICD-10 codes. Health care utilization was analyzed for the following outcomes: hospital admissions, length of hospital stays, outpatient services, ED visits, and supply of a prescription drug (eAppendix Methods). Costs were reported based on the amount allowed (paid from all sources: total costs and costs for inpatient, outpatient, ED, and prescription drug services) and total net payment per patient.

The number of patients prescribed mental health–specific medications was recorded for the following drug types: antidepressants, antipsychotics, anxiolytics, substance use disorder medications, stimulants, and other mental health–specific medications (eAppendix Methods), and medication use is reported. Adherence and discontinuation are reported for antidepressants only. Adherence was analyzed as the percentage of patients with the proportion of days covered (PDC) at least 80% (eAppendix Methods). Discontinuation was analyzed based on the proportion of patients with a gap of at least 90 days in medication supply.²⁰ Provider status was coded as either in network or out of network.

All analyses were performed by IBM Watson using WPS version 4.1.2 (World Programming). Regression models were used to examine the adjusted relationship between key outcomes and wage categories, controlling for demographic and clinical variables (Statistical Analysis section within eAppendix Methods).

A total of 2,386,844 adults were included in the analysis and formed the total population (eAppendix Table 1). Of these, 254,851 (10.7%) were patients with mental health diagnoses. Among patients with mental health diagnoses, 125,247 (5.2% of the total population) experienced depression (subgroup of patients with depression). The mean (SD) age of participants was 44.5 (11.3) years and was higher in higher-wage categories. Overall, 58% of participants were men; most patients with mental health diagnoses, and specifically depression, were women (eAppendix Table 1). Overall, the mean annualized wage was $86,079; this ranged from $24,606 in the lowest-wage category (< $34,000) to $166,868 in the highest-wage category (> $103,000) (eAppendix Table 1).

About half of participants (49%) were enrolled in a preferred provider organization. Fewer than half (39%) of participants had employee-only coverage; most had coverage that included family members.

Analysis revealed that 10.7% of the total population had mental health disorders and 5.2% of the total population had depression. Prevalence of diagnosed mental health disorders, and specifically depression, was highest in the mid-wage categories (Figure 1 [A]).

The adjusted mean severity of depression episodes ranged from 1.46 to 1.63 (on a scale from 1.0 to 4.0) across wage categories (Figure 1 [B]). Depression severity was inversely related to wage, with greater severity in the lower-wage categories. A similar but less pronounced pattern was observed for all mental health episodes (Figure 1 [B]).

All-cause utilization of health care services was greater among patients with mental health diagnoses than in the total population (eAppendix Table 2). Among patients with mental health diagnoses, and specifically depression, all-cause hospital admissions, ED visits, and drug supply were higher in lower-wage categories than in the highest-wage category. This pattern was similar for mental health–specific health care services (eAppendix Table 2). In addition, the proportion of all hospital admissions that were mental health related was numerically greater in the lower-wage than in the higher-wage categories.

Among patients with mental health diagnoses, the adjusted mean rate of mental health–specific hospital admissions per 1000 in the lowest-wage category (32.5) was twice the rate in the highest-wage category (16.1) (eAppendix Table 2). This discrepancy was even greater in the subgroup of patients with depression, for whom those in the lowest-wage category were almost 2.5-fold more likely to have mental health–specific hospitalizations than those in the highest-wage category. Among patients with depression, the rate of all-cause ED visits in the lowest-wage category was also twice as high as in the highest-wage category.

Rates of all-cause outpatient service utilization among patients with mental health diagnoses and patients with depression showed no consistent pattern in relation to wage, with rates appearing greatest in the lowest- and highest-wage categories (eAppendix Table 2).

The total cost of care was higher for individuals with mental health conditions than for the total population (Figure 2 [A] and eAppendix Table 3); costs were even higher for patients with depression. In the total population, all-cause costs declined in association with increasing wages (Figure 2 [A]). A similar pattern occurred for inpatient costs and net payments (eAppendix Table 3). Among patients with mental health diagnoses, and specifically depression, mental health–specific costs were highest in the lowest-wage category (Figure 2 [B] and eAppendix Table 3).

In each analysis group, antidepressants were the most prescribed class of mental health–specific medications, followed by anxiolytics and stimulants (eAppendix Table 4). Among patients with depression, 61.5% received antidepressants during the study period.

Among those with mental health diagnoses, the proportion of patients adherent (based on PDC ≥ 80%) to antidepressant therapy was lower (Figure 3 [A]) and the 90-day discontinuation rate was higher (Figure 3 [B]) in the 2 lower-wage quartiles (eAppendix Table 5).

Use of out-of-network mental health services was highest in the highest-wage category and lowest in the middle-wage categories ($34,000-$69,000) (eAppendix Table 6).

In this study of more than 2 million employees with employer-sponsored health insurance and meeting Medical Episode Grouper criteria for diagnosis, 10.7% of participants were found to have mental health conditions in 2017, including 5.2% of the study population with depression. Following adjustment for demographic and other variables, the proportion of patients with mental health disorders as a percentage of the total population ranged from 9% to 11% across wage categories. For depression, these proportions were 4.2% to 5.1%. Mean severity of depression episodes ranged from 1.46 to 1.63 across wage categories, with higher severity observed in lower-wage groups.

In comparison with available prevalence data in the general population,¹ prevalence of diagnosed mental health disorders based on insurance claims in this study was substantially lower than expected. Individuals may use employee assistance programs (EAPs), pay cash for services, or use other sources of support not captured in claims data. Individuals may also experience cost or access barriers,^4,6,13 preventing use of services that would be documented in claims data. Stigma^14,21 (public or personal) concerns also prompt individuals to avoid seeking care. In addition, this analysis was based on diagnosis of mental health disorders in the context of episodes of care, which may underestimate prevalence. Use of the Medical Episode Grouper may effectively reduce the overall prevalence rates because of its more restrictive methodology.

In this analysis, the highest prevalence of diagnosed mental health disorders was observed in the middle-wage categories, with the lowest prevalence rates in the lowest- and highest-wage categories. In contrast, in a previous analysis, prevalence of depression was greatest at the lowest income level and declined at each higher income level, although this association was not significant among older adults (≥ 55 years).⁸ We also found that patients with mental health diagnoses in the lower-wage categories had more severe mental health episodes and more hospital admissions, and this was particularly true for patients with depression.

The lower apparent prevalence observed in our data for the lowest-wage category may be due to individuals with milder cases not seeking care. This is also apparent by an increased severity of the disease among lower-wage individuals, raising the possibility that care is being sought by patients with the most severe manifestations of depression. In our study population, insurance deductible as a percentage of wage was much higher in the lowest-wage category (12%) than in higher-wage categories (1%-3%), highlighting the cost pressure against seeking medical care among lower-wage earners. In our sample, patients with mental health diagnoses in the highest-wage category were much more likely to use out-of-network providers, presumably reflecting their willingness and ability to seek and pay for these services. Interestingly, there was also an apparent higher rate of out-of-network use in the lowest-wage category than in the middle-wage categories, which may reflect the greater disease severity and need for prompt treatment observed in this group or exposure to out-of-network providers during in-network hospital admissions.^22,23

In this study, costs of care were higher for patients with mental health diagnoses than for the total population, consistent with earlier studies.⁴ All-cause and mental health–specific health care costs were higher for participants in the lowest-wage quartile compared with the highest-wage quartile. Among patients with mental health diagnoses, rates of hospital admissions and ED visits were also higher in the lower-wage categories. After adjustment for demographic variables, the rate of mental health–specific hospital admissions in the lowest-wage category was twice as high as in the highest-wage category, and this discrepancy was even greater among patients with depression. Further, the proportion of mental health–related hospital admissions was greater among lower-income than higher-income workers. These findings suggest that management of mental health conditions in this lower-income population may be suboptimal. It is possible that lower-wage workers delay seeking care until their symptoms are severe enough to warrant inpatient care; however, the relative similarity of outpatient services utilization observed across wage categories suggests that access differences may not have been significant.

These utilization findings are consistent with an earlier study by Sherman and colleagues, which found higher rates of hospital admissions and ED visits among lower-wage earners.¹⁵ These researchers speculated that low-wage-earning adults may exhibit a more reactive approach to health care, either by necessity or choice.¹⁵ Furthermore, low-wage-earning adults may also face difficulties in taking time off to schedule appointments because of workplace conditions less conducive to employees having daytime personal appointments, uncertain work schedules, and unavailability of appointments outside regular business hours.²⁴ The tendency of the lower-wage earners to obtain care in high-intensity settings appears to drive the overall higher cost of care.

There is a need to systematically evaluate employee benefit design considerations and communication approaches to improve mental health service use and effectiveness. Sensitivity to patient privacy has limited the analysis of mental health claims and EAP data to more fully characterize this issue. Even as mental health benefits are offered, utilization of these benefits is not systematically tracked for impact. There is a need to capture clinical outcomes to assess treatment effectiveness; however, data from insurance claims alone are insufficient. Also, self-reported disease may be much more prevalent than claims data indicate,²⁵ as our unexpectedly low prevalence numbers appear to demonstrate.^10,11

Employers’ awareness of the significance of mental health concerns appears to underestimate the importance of the problem.^4,6,13 Despite implementation of the 2008 Mental Health Parity and Addiction Equity Act, our data indicate that disparities in access to and use of mental health services remain, particularly for lower-income individuals.

Further, our analysis again demonstrates that health care costs for patients with mental health disorders are higher than those in the overall population. Therefore, employers have strong incentives to ensure that all employees, including those with mental health issues, have equitable access to health care. Greater remuneration for mental health care providers can increase the number of in-network providers. Integration of mental health services with primary care is another option to increase access to and utilization of mental health services.²⁶ Technology-based care delivery methods²⁷ and innovative content should also be explored.

This study has several limitations. The analysis was limited to employees with wage information who were continuously enrolled in self-insured plans in the IBM Watson Health MarketScan Research Database during 2017, which does not constitute a population-based sample and does not include lower-wage-earning workers who rely on public health insurance. Because the analysis was based on insurance claims, employees with mental health symptoms who did not seek care or receive a formal diagnosis were not identified. Our data also do not include employees paying out of pocket for mental health services outside the delivery system. Patients with mental health conditions were identified using the Medical Episode Grouper,¹⁶ which may undercount the number of patients relative to use of the corresponding ICD-10 codes as inclusion criteria. EAP data were not included in this analysis, making it difficult to know the extent to which individuals with mental health concerns received care via this benefit offering.

Use of claims data–based analysis also precludes our understanding of specific barriers to using mental health services. Consequently, we are unable to determine the relative importance of access, cost, stigma, cultural beliefs, or other factors in the observed mental health services utilization patterns. Our analysis did not incorporate other social determinants of health data beyond wage status that could further inform data interpretation and recommendations. Our analysis was also unable to identify employees who had poor access to mental health services. Additionally, we cannot determine whether low wages caused the observed patterns of mental health disorders and health care utilization, or whether lack of mental health care and resulting exacerbation of mental health disorders may have resulted in lower earning potential.

We did not include health savings account or health care reimbursement arrangement balances in our analysis, which may have influenced individuals’ willingness to use mental health services. Because of our large sample size, differences between groups sometimes achieved statistical significance even when they were numerically small.

Strengths of the study design include the large size and national distribution of the population included in the analysis, increasing the likelihood that these results are generalizable to other commercially insured populations.

Based on our findings, lower-wage-earning workers have a greater severity of mental health disorders and more frequent ED visits and rates of hospitalization. Medication adherence and discontinuation data are also consistent with suboptimal management. Increased and more proactive use of ambulatory mental health services may help to lessen the need for high-intensity services.

Many factors likely contribute to the observed findings, including out-of-pocket cost considerations and barriers to care; additional research can help to better characterize these concerns. Our findings have particular relevance in relation to the COVID-19 pandemic, which has led to an increase in mental health disorders.^10,11 Employers should critically review the design of their company’s mental health benefits to ensure that affordable and accessible treatment is available for all enrollees.

Author Affiliations: Case Western Reserve University School of Medicine (BWS), Cleveland, OH; National Alliance of Healthcare Purchaser Coalitions (BWS), Washington, DC; Takeda Pharmaceuticals U.S.A., Inc (DFL, MK, LC), Lexington, MA;College of Pharmacy, University of Illinois at Chicago (MK), Chicago, IL;Lundbeck LLC (SP, MT), Deerfield, IL.

Source of Funding: Takeda Pharmaceuticals U.S.A., Inc, and Lundbeck LLC funded the study and participated in the study design and interpretation.

Author Disclosures: Dr Sherman is a consultant to the National Alliance of Healthcare Purchaser Coalitions, received grants from Takeda and Lundbeck funding this research, and received honoraria from IBM Watson for conference presentation of preliminary findings. Drs Lawrence and Chrones are employees of Takeda. Ms Kuharic is an employee of the University of Illinois at Chicago and was supported by a Takeda fellowship during the writing of this manuscript. Drs Patel and Touya were employees of Lundbeck at the time of the study.

Authorship Information: Concept and design (BWS, DFL, MK, SP, MT); acquisition of data (BWS); analysis and interpretation of data (BWS, DFL, MK, LC, SP, MT); drafting of the manuscript (BWS, LC, MT); critical revision of the manuscript for important intellectual content (BWS, DFL, MK, LC, SP, MT); statistical analysis (BWS, SP, MT); obtaining funding (BWS, MT); administrative, technical, or logistic support (BWS, MK); and supervision (BWS, DFL).

Address Correspondence to: Bruce W. Sherman, MD, Case Western Reserve University School of Medicine, 117 Kemp Rd E, Greensboro, NC 27410. Email: bws@case.edu.

1. Key substance use and mental health indicators in the United States: results from the 2018 National Survey on Drug Use and Health. Substance Abuse and Mental Health Services Administration. August 2019. Accessed February 10, 2020. https://www.samhsa.gov/data/sites/default/files/cbhsq-reports/NSDUHNationalFindingsReport2018/NSDUHNationalFindingsReport2018.pdf

2. The state of mental health and stress in the workplace 2019. HR.com. October 2019. Accessed February 10, 2020. https://www.hr.com/en/resources/free_research_white_papers/hrcom-mental-health-and-stress-in-the-workplace-20_k11dy184.html

3. Lim D, Sanderson K, Andrews G. Lost productivity among full-time workers with mental disorders. J Ment Health Policy Econ. 2000;3(3):139-146. doi:10.1002/mhp.93

4. Goetzel RZ, Roemer EC, Holingue C, et al. Mental health in the workplace: a call to action proceedings from the Mental Health in the Workplace–Public Health Summit. J Occup Environ Med. 2018;60(4):322-330. doi:10.1097/JOM.0000000000001271

5. Ammerman RT, Chen J, Mallow PJ, Rizzo JA, Folger AT, Van Ginkel JB. Annual direct health care expenditures and employee absenteeism costs in high-risk, low-income mothers with major depression. J Affect Disord. 2016;190:386-394. doi:10.1016/j.jad.2015.10.025

6. Dewa CS, Thompson AH, Jacobs P. The association of treatment of depressive episodes and work productivity. Can J Psychiatry. 2011;56(12):743-750. doi:10.1177/070674371105601206

7. Goldman N, Glei DA, Weinstein M. Declining mental health among disadvantaged Americans. Proc Natl Acad Sci U S A. 2018;115(28):7290-7295. doi:10.1073/pnas.1722023115

8. Sareen J, Afifi TO, McMillan KA, Asmundson GJG. Relationship between household income and mental disorders: findings from a population-based longitudinal study. Arch Gen Psychiatry. 2011;68(4):419-427. doi:10.1001/archgenpsychiatry.2011.15

9. Platt J, Prins S, Bates L, Keyes K. Unequal depression for equal work? how the wage gap explains gendered disparities in mood disorders. Soc Sci Med. 2016;149:1-8. doi:10.1016/j.socscimed.2015.11.056

10. Wu T, Jia X, Shi H, et al. Prevalence of mental health problems during the COVID-19 pandemic: a systematic review and meta-analysis. J Affect Disord. 2021;281:91-98. doi:10.1016/j.jad.2020.11.117

11. Xiong J, Lipsitz O, Nasri F, et al. Impact of COVID-19 pandemic on mental health in the general population: a systematic review. J Affect Disord. 2020;277:55-64. doi:10.1016/j.jad.2020.08.001

12. Hilton MF, Scuffham PA, Sheridan J, Cleary CM, Vecchio N, Whiteford HA. The association between mental disorders and productivity in treated and untreated employees. J Occup Environ Med. 2009;51(9):996-1003. doi:10.1097/JOM.0b013e3181b2ea30

13. Thornicroft G, Chatterji S, Evans-Lacko S, et al. Undertreatment of people with major depressive disorder in 21 countries. Br J Psychiatry. 2017;210(2):119-124. doi:10.1192/bjp.bp.116.188078

14. America’s mental health 2018. Cohen Veterans Network. October 10, 2018. Accessed February 27, 2023. https://gwdocs.com/sites/g/files/zaskib551/files/2021-06/Research-Summary-10-10-2018.pdf

15. Sherman BW, Gibson TB, Lynch WD, Addy C. Health care use and spending patterns vary by wage level in employer-sponsored plans. Health Aff (Millwood). 2017;36(2):250-257. doi:10.1377/hlthaff.2016.1147

16. The IBM Medical Episode Grouper: applications and methodology. IBM Watson Health. 2020. Accessed September 15, 2020. https://www.ibm.com/downloads/cas/EZALXAMB

17. Peterson C, Li M, Xu L, Mikosz CA, Luo F. Assessment of annual cost of substance use disorder in US hospitals. JAMA Netw Open. 2021;4(3):e210242. doi:10.1001/jamanetworkopen.2021.0242

18. Whaley CM, Pera MF, Cantor J, et al. Changes in health services use among commercially insured US populations during the COVID-19 pandemic. JAMA Netw Open. 2020;3(11):e2024984. doi:10.1001/jamanetworkopen.2020.24984

19. Landsman PB, Smith DG, Fendrick AM. Healthcare utilization in community-acquired pneumonia episodes of care. Med Care. 2009;47(10):1084-1090. doi:10.1097/MLR.0b013e3181a8116d

20. Steiner JF, Prochazka AV. The assessment of refill compliance using pharmacy records: methods, validity, and applications. J Clin Epidemiol. 1997;50(1):105-116. doi:10.1016/s0895-4356(96)00268-5

21. Tsai J, Mota NP, Pietrzak RH. U.S. female veterans who do and do not rely on VA health care: needs and barriers to mental health treatment. Psychiatr Serv. 2015;66(11):1200-1206. doi:10.1176/appi.ps.201400550

22. Song Z, Johnson W, Kennedy K, Biniek JF, Wallace J. Out-of-network spending mostly declined in privately insured populations with a few notable exceptions from 2008 to 2016. Health Aff (Millwood). 2020;39(6):1032-1041. doi:10.1377/hlthaff.2019.01776

23. Sun EC, Mello MM, Moshfegh J, Baker LC. Assessment of out-of-network billing for privately insured patients receiving care in in-network hospitals. JAMA Intern Med. 2019;179(11):1543-1550. doi:10.1001/jamainternmed.2019.3451

24. Lewis C, Abrams MK, Seerval S. Listening to low-income patients: obstacles to the care we need, when we need it. The Commonwealth Fund. December 1, 2017. Accessed April 5, 2022. https://www.commonwealthfund.org/blog/2017/listening-low-income-patients-obstacles-care-we-need-when-we-need-it

25. Collins JJ, Baase CM, Sharda CE, et al. The assessment of chronic health conditions on work performance, absence, and total economic impact for employers. J Occup Environ Med. 2005;47(6):547-557. doi:10.1097/01.jom.0000166864.58664.29

26. Klein S, Hostetter M. In focus: integrating behavioral health and primary care. The Commonwealth Fund. August 28, 2014. Accessed February 7, 2020. https://www.commonwealthfund.org/publications/newsletter-article/2014/aug/focus-integrating-behavioral-health-and-primary-care

27. Langarizadeh M, Tabatabaei MS, Tavakol K, Naghipour M, Rostami A, Moghbeli F. Telemental health care, an effective alternative to conventional mental care: a systematic review. Acta Inform Med. 2017;25(4):240-246. doi:10.5455/aim.2017.25.240-246

Mental Health Diagnoses and Services Utilization Vary by Wage Level

Related Content