Journal of Mental Health
ISSN: 0963-8237 (Print) 1360-0567 (Online) Journal homepage: http://www.tandfonline.com/loi/ijmh20
Medicaid provisions and the US mental health industry composition Lawrence C. Pellegrini & Rosa Rodriguez-Monguio To cite this article: Lawrence C. Pellegrini & Rosa Rodriguez-Monguio (2014) Medicaid provisions and the US mental health industry composition, Journal of Mental Health, 23:6, 312-316 To link to this article: http://dx.doi.org/10.3109/09638237.2014.951486
Published online: 02 Sep 2014.
Submit your article to this journal
Article views: 28
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 Birmingham]
Date: 06 November 2015, At: 00:55
http://informahealthcare.com/jmh ISSN: 0963-8237 (print), 1360-0567 (electronic) J Ment Health, 2014; 23(6): 312–316 ! 2014 Shadowfax Publishing and Informa UK Limited. DOI: 10.3109/09638237.2014.951486
ORIGINAL ARTICLE
Medicaid provisions and the US mental health industry composition Lawrence C. Pellegrini1 and Rosa Rodriguez-Monguio1,2 School of Public Health and Health Sciences, University of Massachusetts, Amherst, MA, USA and 2The Institute for Global Health, University of Massachusetts, Amherst, MA, USA
Downloaded by [University of Birmingham] at 00:55 06 November 2015
1
Abstract
Keywords
Background: Medicaid is the largest payer for mental health (MH) services. Aims: This study examines associations between Medicaid provisions and the MH industry composition. Methods: Medicaid data derived from the Centers for Medicare and Medicaid Services. MH facility gross payroll and occupational employment data derived from the Bureau of Labor Statistics. State fixed-effects regression models are performed to examine associations. Results: In the 1999–2009 period, per-capita gross payroll gains are largest for residential MH and substance abuse (SA) facilities and MH practitioner offices, followed by MH clinics and physician offices. Likewise, occupational employment gains per 100 000 people are largest for MH and SA social workers and MH counselors, followed by psychiatrists and psychologists. The Medicaid beneficiary rate is related with gross payroll gains at residential MH and SA facilities (p50.001) and MH clinics (p50.001), and with employment gains for MH and SA social workers (p50.001) and MH counselors (p50.001). Smaller effect sizes exist with MH physician offices (p50.05) and psychiatric hospitals’ (p50.01) gross payroll. No statistically significant relationship exists between the Medicaid beneficiary rate and psychiatrist and psychologist employment. Conclusion: Medicaid provisions are related with the MH industry composition. An imbalanced MH industry may lead to inadequate management of MH disorders.
Health outcomes, medicaid, medical care, mental health services, mental health industry
Introduction Prevalence of mental health (MH) disorders in the United States (US) increased over the past decade (Mojtabai, 2011; Whitaker 2010). In 1999, the average number of mentally unhealthy days per month was three days, and the percentage of population reporting 14 or more mentally unhealthy days per month was 8.9% (Centers for Disease Control and Prevention, 2013). By 2009, the average number of mentally unhealthy days increased to 3.5 days, and the percentage of the population reporting mental distress lasting 14 or more days went up to 10.6%. In the US, individuals may seek treatment for MH disorders through inpatient and outpatient settings. Inpatient psychiatric and substance abuse (SA) hospitals provide extended-stay diagnostic, treatment and MH service monitoring. Residential MH and SA facilities provide residents with room and board, in addition to supervision, counseling and related MH support services. There are three types of outpatient establishments: (1) MH physician offices, (2) MH
Correspondence: Rosa Rodriguez-Monguio, PhD, Associate Professor, University of Massachusetts, Amherst-School of Public Health and Health Sciences, 715 N. Pleasant Street, 322 Arnold House, Amherst, MA 01003, USA. Tel: 1 413 545 7427. Fax: 1 413 545 1645. E-mail: [email protected]
History Received 24 July 2013 Revised 28 May 2014 Accepted 30 May 2014 Published online 2 September 2014
practitioner offices and (3) MH clinics. MH physician offices are private facilities that provide psychiatric services and/or psychoanalysis as performed by a physician (i.e. psychiatrist). MH practitioner offices are also privately operated; these facilities employ non-physician practitioners who specialize in the diagnosis and treatment of MH disorders. Finally, MH clinics employ physician and non-physician MH professionals who collectively provide diagnostic and treatment-related services (Bureau of Labor Statistics, 2013a). MH treatment modalities may include counseling and/or pharmaceutical management. Psychiatrists provide therapy and prescribe pharmaceuticals. Psychologists, MH and SA social workers and MH counselors provide therapeutic services. The 2011 average hourly wage for psychiatrists was $83.73, followed by psychologists at $35.14, MH and SA social workers at $20.50 and MH counselors at $20.48 (Bureau of Labor Statistics, 2013b). Individuals who meet means-based testing criteria may access care through state Medicaid programs. Medicaid is the largest payer of MH treatment services (Centers for Medicare and Medicaid Services, 2013). In 2005, $113 billion was spent on MH services in the US; the share of MH spending funded by Medicaid was 28% (Substance Abuse and Mental Health Services Administration, 2010). Driving forces reshaping Medicaid MH provisions include reimbursement rates and cost-shifting opportunities through federal matching funds, expansion of community-based providers, managed
Medicaid and mental health industry
DOI: 10.3109/09638237.2014.951486
313
and Prevention, 2013). Proxy variables for MH industry are selected as they represent the majority of facilities and professionals qualified to treat MH disorders in the US. Secular trends in Medicaid provisions (i.e. Medicaid beneficiary rates and overall and MH expenditures per beneficiary) are compared with proxy variables for the MH industry (i.e. MH facility gross payroll per-capita and MH professionals per 100 000 people) to identify possible associations. State fixed-effects regression models are performed to assess the relationship between Medicaid provisions and the MH industry. The regression model is: Yit ¼ 0 þ 1 Xit þ i þ "it
Downloaded by [University of Birmingham] at 00:55 06 November 2015
Figure 1. Medicaid provisions and the US mental health industry composition.
care arrangements and new pharmacological treatment (Frank et al., 2003; Gronfein, 1985; Rowland et al., 2003; Sullivan, 2006; Taube et al., 1990). Previous studies assess the relationship between economic conditions and Medicaid MH spending (Ellis et al., 2012; Frank, 1985; Mark et al., 2011; Minoiu & Andres, 2008; Pellegrini & Rodriguez-Monguio, 2013) and the effect of Medicaid provisions on the US healthcare workforce composition (Pellegrini et al., 2014). However, the extent to which Medicaid provisions are related with privately delivered MH services has not been studied on a national level. Thus, this study assesses the relationship between Medicaid provisions and the MH industry composition. We hypothesize that as a primary payer for the MH services, Medicaid provisions shape the supply of MH facilities and professionals (Figure 1). Furthermore, we hypothesize that increasing Medicaid beneficiaries and MH service demand, along with Medicaid funding constraints, results in MH industry structural changes and provider imbalances to effectively treat individuals with MH disorders.
Data and methods Data are collected from readily available datasets for 50 states and the District of Columbia for the 1999–2009 time period. Medicaid beneficiary and overall and MH spending data are collected from the Centers for Medicare and Medicaid Services (Centers for Medicare and Medicaid Services, 2013). MH facilities data are gathered from the Bureau of Labor Statistics’ Quarterly Census of Employment and Wages program, including: (1) psychiatric and SA hospitals, (2) residential MH and SA facilities, (3) MH physician offices, (4) MH practitioner offices and (5) MH clinics (Bureau of Labor Statistics, 2013a). Occupational employment data are collected from the Bureau of Labor Statistics’ Occupational Employment Statistics program. Four MH-related occupations are included in the study: (1) psychiatrists, (2) psychologists, (3) MH and SA social workers and (4) MH counselors (Bureau of Labor Statistics, 2013b). Statewide annual average number of mentally unhealthy days per month data are obtained from the Centers for Disease Control and Prevention’s Behavioral Risk Factor Surveillance System (Centers for Disease Control
where Yit represents MH facility or professional type, Xit represents Medicaid provisions and "it represents the error term. The state fixed-effects model is performed as Medicaid is a state-managed program. Thus, the state fixed-effects dummy variable, ai, represents omitted variables that vary across states but not over time. Expenditure data are adjusted to 2010 dollars using the consumer price index for all urban consumers. Count data are converted to rates per-capita (i.e. facility gross payroll), per beneficiary (i.e. Medicaid overall and MH spending) or per 100 000 people (i.e. Medicaid beneficiaries and occupational employment). All statistical tests are based on a two-sided a significance level of p50.05. Statistical analyses are performed with STATA version 12.0 statistical software (College Station, TX).
Results In the 1999–2009 study period, the average Medicaid beneficiary rate per 100 000 people for 50 states and DC is 13 625; increasing from 11 164 in 1999 to 15 453 in 2009 (38.42% increase). The study period average Medicaid MH spending per beneficiary is $445; decreasing from $534 in 1999 to $384 in 2009 (28.09% decrease). The average number of mentally unhealthy days also increased from 3 days in 1999 to 3.37 in 2009 (Table 1). There are changes in MH facility gross payroll per-capita. MH clinics represent the highest spending per-capita over the study period; increasing from $23.79 in 1999 to $26.34 in 2009 (10.72% increase). Likewise, in 1999 and 2009, the smallest number of per-capita dollars is spent at MH practitioner offices, $6.93 and $7.87, respectively, followed by MH physician offices, $7.70 and $7.97, respectively. Growth rates differ for each of the five examined facility types. Residential MH and SA facilities experienced the largest increase in per-capita gross payroll; from $13.66 percapita in 1999 to $19.05 in 2009 (39.46% increase). To the contrary, psychiatric and SA hospitals show a study period decrease (11.62%; Table 1). There are also changes to MH occupational employment in the 1999–2009 study period. In 1999, psychologist employment represents the largest occupational group (32.79 per 100 000). However, of the four examined occupations, psychologist employment shows the smallest growth rate in occupational employment per 100 000 people (4.27%).
314
L. C. Pellegrini et al.
J Ment Health, 2014; 23(6): 312–316
Table 1. Descriptive statistics and trends. 1999–2009 time period
Downloaded by [University of Birmingham] at 00:55 06 November 2015
Variable name Medicaid provisions Medicaid beneficiaries Medicaid spending Medicaid MH spending Mentally unhealthy days Mentally unhealthy days Mental health facilities Psychiatric hospitals Residential MH and SA facilities MH physician offices MH practitioner offices MH clinics Mental health professionals Psychiatrists Psychologists MH and SA social workers MH counselors
Trends
No. of obs.
Mean
SD
Minimum
Maximum
Mean 1999
Mean 2009
Average % change 1999–2009
561 561 533
13 625 7473 445
4474 1939 418
5097 3852 3
25 112 13 481 2573
11 164 7148 534
15 453 7622 384
38.42% 6.63% 28.09%
531
3.27
0.50
0.70
5.70
3.00
3.37
12.36%
348 521 542 544 492
12.50 15.93 7.96 7.49 25.09
7.77 8.96 6.00 5.09 21.77
1.23 2.26 0.74 1.13 0.87
42.07 52.49 51.49 35.81 135.13
14.29 13.66 7.70 6.93 23.79
12.63 19.05 7.97 7.87 26.34
11.62% 39.46% 3.51% 13.56% 10.72%
466 557 541 531
8.12 35.12 41.46 31.68
7.71 16.41 25.25 19.47
0.87 5.39 5.95 4.70
64.04 125.42 211.84 131.18
5.96 32.79 29.61 23.82
8.08 34.19 50.35 35.96
35.57% 4.27% 70.04% 50.97%
Medicaid beneficiaries and mental health professionals’ data are in rates per 100 000 people. Medicaid overall and MH spending data are per beneficiary. Mental health facilities’ data are gross-payroll per-capita. Source: Medicaid data derived from the Centers for Medicare and Medicaid Services. Mentally unhealthy days data derived from the Centers for Disease Control and Prevention’s Behavioral Risk Factor Surveillance System. Mental health facilities and professionals’ data derived from the Bureau of Labor Statistics’ Quarterly Census of Employment and Wages (QCEW) and Occupational Employment Statistics (OES) survey, respectively.
The largest study period employment growth rates are for MH and SA social workers (70.04%) and MH counselors (50.97%). In 2009, MH and SA social workers and MH counselors represent the largest occupational group at 50.35 and 35.96 per 100 000 people, respectively, followed by psychologists at 34.19 per 100 000 people. Psychiatrist employment represents the smallest occupational employment group both in 1999 and 2009. In 1999, there are 5.96 psychiatrists per 100 000 people, increasing to 8.08 in 2009 (35.57% increase; Table 1). Results show that the Medicaid beneficiary rate is associated with per-capita gross payroll gains at all examined facilities excluding MH practitioner offices. However, the effect size is largest for MH clinics followed by residential MH and SA facilities (p50.001). Per-capita gross payroll growth at MH physician offices (p50.05) and psychiatric hospitals (p50.01) is least related with the Medicaid beneficiary rate. Similar effect sizes are found for Medicaid overall and MH spending per-beneficiary. Nevertheless, the relationships between Medicaid MH spending per beneficiary and MH facilities’ gross payroll per-capita are not statistically significant. Study results also evidence that MH and SA social worker employment growth per 100 000 people is most related with Medicaid provisions. A one percentage point increase in the Medicaid beneficiary rate and Medicaid overall, and MH spending per beneficiary is associated with a 0.003 (p50.001), and 0.002% (p50.05) increase, and 0.011% (p50.05) decrease in the number of MH and SA social workers per 100 000 people, respectively. To the contrary, the association between Medicaid beneficiary rate, MH spending, and psychologist employment per 100 000 people is not statistically significant. Likewise, there is no statistically significant relationship between psychiatrist employment and either the
Medicaid beneficiary rate or overall and MH spending per beneficiary. However, an increase in Medicaid overall spending spending per beneficiary is associated with a 0.003% increase in the number of psychologists per 100 000 people (p50.05) (Table 2).
Discussion Access to MH services is often challenging for underserved individuals with MH disorders (Carbaugh et al., 2006; Ellis et al., 2012; Mojtabai, 2011). State Medicaid programs play an important role in the configuration of MH facilities and professionals serving beneficiaries with mental disorders. In the study period, there are significant changes in the composition of the MH industry. Results show that MH clinics continue to maintain a dominant role. However, growth in residential MH and SA facilities further evidences trends in managing the chronically mentally ill in the community as opposed to psychiatric institutions; as of 2009, residential facilities’ gross-payroll spending per-capita is higher than psychiatric hospitals’ spending. Medicaid provisions have a strong relationship with residential MH and SA gross-payroll spending. Nevertheless, increases in residential spending have not been met with an equally significant, and positive, association between Medicaid provisions and MH physician or practitioner offices’ grosspayroll spending; both of these facilities employ significant numbers of outpatient psychiatrists and psychologists (Bureau of Labor Statistics, 2013b). The analysis of MH professionals reveals that there is significant growth in employment of MH and SA social workers and MH counselors relative to psychologists; both of these professionals earn approximately 58% the market wage of psychologists. The Medicaid beneficiary rate is significantly related with employment growth for MH and SA
Medicaid and mental health industry
DOI: 10.3109/09638237.2014.951486
315
Table 2. Medicaid provisions and the US mental health industry composition. Mental health facilities Psychiatric hospitals
Downloaded by [University of Birmingham] at 00:55 06 November 2015
Medicaid beneficiaries 0.000222b Standard error (0.0000764) Sample size 348 0.919 R2 Medicaid spending 0.000160 Standard error (0.000234) Sample size 348 0.917 R2 Medicaid MH spending 0.00135 Standard error (0.000997) Sample size 334 0.918 R2
Mental health professionals
Residential MH MH physician MH practitioner and SA facilities offices offices 0.000637a (0.0000776) 521 0.900 0.000704b (0.000229) 521 0.887 0.0000614 (0.00101) 494 0.884
0.000144c (0.0000562) 542 0.875 0.000277 (0.000168) 542 0.874 0.000826 (0.000676) 514 0.875
0.0000311 (0.0000550) 544 0.830 0.000339c (0.000154) 544 0.832 0.000373 (0.000556) 516 0.851
MH clinics
MH and SA social MH Psychiatrists Psychologists workers counselors
0.000814a 0.0000679 0.000154 0.00330a 0.000865b (0.000106) (0.0000768) (0.000217) (0.000355) (0.000278) 492 466 557 541 531 0.971 0.888 0.746 0.717 0.729 0.00156a 0.000357 0.00260a 0.00218c 0.000764 (0.000313) (0.000216) (0.000598) (0.00108) (0.000783) 492 466 557 541 531 0.968 0.888 0.754 0.670 0.724 c 0.000139 0.000310 0.00133 0.000283 0.0108 (0.00129) (0.000954) (0.00246) (0.00450) (0.00319) 466 441 531 513 504 0.967 0.888 0.747 0.663 0.720
Standard errors in parentheses a p50.001, bp50.01, cp50.05. Medicaid beneficiaries and mental health professionals’ data are in rates per 100 000 people. Medicaid overall and MH spending data are per beneficiary. Mental health facilities’ data are gross-payroll per-capita. Source: Medicaid data derived from the Centers for Medicare and Medicaid Services. Mental health facilities and professionals’ data derived from the Bureau of Labor Statistics’ Quarterly Census of Employment and Wages (QCEW) and Occupational Employment Statistics (OES) survey, respectively.
social workers and MH counselors; whereas, the association is not statistically significant for psychologists. In addition, there is an increase in psychiatrist employment during the study period. Nevertheless, the relationship between the Medicaid beneficiary rate and overall and MH spending per beneficiary and psychiatrist employment is not statistically significant. The associations between Medicaid provisions and the MH industry composition may be related with underlying Medicaid reimbursement mechanisms. State Medicaid programs have often set reimbursement levels below that of the private sector, financially discouraging MH providers (Frank et al., 2003). As a result, publicly funded state Medicaid programs may be constrained in their ability to provide more costly private sector psychiatric and/or psychological services to their beneficiaries. Nevertheless, from a public health standpoint, each type of healthcare facility and professional provides a unique array of health services amongst the continuum of care of MH disorders. The MH industry composition becomes increasingly important as current clinical practice further evidences that adjunctive therapy and pharmacological maintenance is often the preferred treatment modality for MH disorders. However, of all examined MH professionals, only psychiatrists are licensed to prescribe pharmaceuticals in all fifty states and DC. Furthermore, other associated MH professionals possess differing levels of pharmaceutical training. While Medicaid provisions include coverage for psychotropic drugs, it is also associated with growth of specific types of facilities and professionals, as influenced by state-specific reimbursement rates. Thus, Medicaid reimbursement rates along with trends toward pharmacological treatment of mental disorders in the US present additional challenges to the delivery of optimum MH services to patient populations. In addition, chronically mentally ill individuals are likely to be Medicaid beneficiaries, thus; reduced access to the breadth of MH services becomes increasingly worrisome in a postdeinstitutionalization era.
Limitations This study has some potential limitations worth noting. First, our proxy variable for mental distress, mentally unhealthy days, is a self-reported measure that may be influenced by sampling methodologies and response rates. Furthermore, an increase in the average number of mentally unhealthy days does not necessarily indicate a need for MH services. Second, this study examines facilities and professionals that are consistent across states as classified by the Bureau of Labor Statistics. Thus, our MH industry proxy variables may exclude additional state-specific facilities and professionals that play an instrumental role in state MH industries. In addition, examined MH facilities and professionals are not exclusively Medicaid providers. Third, Medicaid is a stateadministered program with provisions varying from state to state. As such, state-specific reimbursement policy likely influences the configuration of each state’s MH industry. Medicaid overall and MH spending per beneficiary is used to proxy these reimbursement policies. Fourth, our model examining the relationship between Medicaid provisions and the MH industry may be subject to omitted variable bias. For example, state Medicaid provisions may be correlated with state economic conditions, and such conditions could influence mental distress and gross-payroll spent at examined MH facilities. Furthermore, state-specific MH occupational employment may be related with other factors not examined in this study, including availability of graduate training programs and state specific licensing requirements. Finally, our model may also be subject to reverse causality; the MH industry may affect Medicaid provisions much like Medicaid provisions may affect the MH industry. However, this study does not aim to evaluate causality but to assess associations between study measures.
Conclusion In the period 1999–2009, there are increases in the statewide average number of mentally unhealthy days per month. There
Downloaded by [University of Birmingham] at 00:55 06 November 2015
316
L. C. Pellegrini et al.
are also significant changes in the MH industry composition as related to changes in Medicaid provisions. The Medicaid beneficiary rate is related with per-capita gross payroll spending growth at MH clinics and residential MH and SA facilities and with occupational employment gains per 100 000 people for MH and SA social workers and MH counselors. Furthermore, the Medicaid beneficiary rate is also related with per-capita gross payroll spending growth at MH physician offices and psychiatric hospitals. The relationship between psychologist employment per 100 000 people and the Medicaid beneficiary rate and MH spending per beneficiary is not statistically significant. Likewise, there is no statistically significant association between psychiatrist employment and either the Medicaid beneficiary rate or Medicaid overall and MH spending per beneficiary. Each type of facility and health professional provides a unique array of services in the MH continuum of care. As a result, imbalances in the US MH industry composition may lead to barriers in access to MH services and inadequate management of MH disorders.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.
References Bureau of Labor Statistics. (2013a). Labor force statistics from the Quarterly Census of Employment and Wages Survey. Available from: http://www.bls.gov/qcew/ [last accessed 15 Feb 2013]. Bureau of Labor Statistics. (2013b). Labor force statistics from the Occupational Employment Statistics Survey. Available from: http:// www.bls.gov/oes/ [last accessed 15 Feb 2013]. Carbaugh AL, Elias R, Rowland D. (2006). Aid to people with disabilities: Medicaid’s growing role. Annu Rev Publ Health, 27, 417–42.
J Ment Health, 2014; 23(6): 312–316
Centers for Disease Control and Prevention. (2013). Behavioral Risk Factor Surveillance System (BRFSS). Available from: http:// www.cdc.gov/brfss [last accessed 15 Feb 2013]. Centers for Medicare and Medicaid Services. (2013). Medicaid and Statistical Information System (MSIS). Available from: http:// www.medicaid.gov [last accessed 15 Feb 2013]. Ellis WR, Huebner C, Vander Stoep A, Williams M. (2012). Washington State exhibits wide regional variation in proportion of Medicaideligible children who get needed mental health care. Health Affairs, 31, 990–9. Frank RG. (1985). A model of state expenditures on mental health services. Public Financ Quart, 13, 319–38. Frank RG, Goldman HH, Hogan M. (2003). Medicaid and mental health: Be careful what you ask for. Health Aff, 22, 101–13. Gronfein W. (1985). Incentives and intentions in mental health policy: A comparison of the Medicaid and community mental health programs. J Health Soc Behav, 26, 192–206. Mark TL, Levitt KR, Vandivort-Warren R, et al. (2011). Changes in US spending on mental health and substance abuse treatment, 1986–2005, and implications for policy. Health Aff, 30, 284–92. Minoiu C, Andres AR. (2008). The effect of public spending on suicide: Evidence from US state data. J Soc Econ, 37, 237–61. Mojtabai R. (2011). National trends in mental health disability, 1997-2009. Am J Public Health, 101, 2156–63. Pellegrini LC, Rodriguez-Monguio R. (2013). Unemployment, Medicaid provisions, the mental health industry, and suicide. Soc Sci J, 50, 482–90. Pellegrini LC, Rodriguez-Monguio R, Qian J. (2014). The US healthcare workforce and the labor market effect on healthcare spending and health outcomes. Int J Health Care Finance Econ, 14, 127–41. Rowland D, Garfield R, Elias R. (2003). Accomplishments and challenges in Medicaid mental health. Health Aff, 22, 73–83. Substance Abuse and Mental Health Services Administration, Center for Mental Health Services and Center for Substance Abuse Treatment. (2010). Spending Estimates Project 2010. Available from: http:// www.samhsa.gov/ [last accessed 15 Feb 2013]. Sullivan WP. (2006). Mental health services in the 21st century: The economics and practice challenges on the road to recovery. Adv Soc Work, 6, 193. Taube CA, Goldman HH, Salkever DS. (1990). Medicaid coverage for mental illness: Balancing access and costs. Health Aff, 9, 5–18. Whitaker R. (2010). Anatomy of an epidemic: Magic bullets, psychiatric drugs, and the astonishing rise of mental illness in America. 1st ed. New York, NY: Crown Publishers.
ISSN: 0963-8237 (Print) 1360-0567 (Online) Journal homepage: http://www.tandfonline.com/loi/ijmh20
Medicaid provisions and the US mental health industry composition Lawrence C. Pellegrini & Rosa Rodriguez-Monguio To cite this article: Lawrence C. Pellegrini & Rosa Rodriguez-Monguio (2014) Medicaid provisions and the US mental health industry composition, Journal of Mental Health, 23:6, 312-316 To link to this article: http://dx.doi.org/10.3109/09638237.2014.951486
Published online: 02 Sep 2014.
Submit your article to this journal
Article views: 28
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 Birmingham]
Date: 06 November 2015, At: 00:55
http://informahealthcare.com/jmh ISSN: 0963-8237 (print), 1360-0567 (electronic) J Ment Health, 2014; 23(6): 312–316 ! 2014 Shadowfax Publishing and Informa UK Limited. DOI: 10.3109/09638237.2014.951486
ORIGINAL ARTICLE
Medicaid provisions and the US mental health industry composition Lawrence C. Pellegrini1 and Rosa Rodriguez-Monguio1,2 School of Public Health and Health Sciences, University of Massachusetts, Amherst, MA, USA and 2The Institute for Global Health, University of Massachusetts, Amherst, MA, USA
Downloaded by [University of Birmingham] at 00:55 06 November 2015
1
Abstract
Keywords
Background: Medicaid is the largest payer for mental health (MH) services. Aims: This study examines associations between Medicaid provisions and the MH industry composition. Methods: Medicaid data derived from the Centers for Medicare and Medicaid Services. MH facility gross payroll and occupational employment data derived from the Bureau of Labor Statistics. State fixed-effects regression models are performed to examine associations. Results: In the 1999–2009 period, per-capita gross payroll gains are largest for residential MH and substance abuse (SA) facilities and MH practitioner offices, followed by MH clinics and physician offices. Likewise, occupational employment gains per 100 000 people are largest for MH and SA social workers and MH counselors, followed by psychiatrists and psychologists. The Medicaid beneficiary rate is related with gross payroll gains at residential MH and SA facilities (p50.001) and MH clinics (p50.001), and with employment gains for MH and SA social workers (p50.001) and MH counselors (p50.001). Smaller effect sizes exist with MH physician offices (p50.05) and psychiatric hospitals’ (p50.01) gross payroll. No statistically significant relationship exists between the Medicaid beneficiary rate and psychiatrist and psychologist employment. Conclusion: Medicaid provisions are related with the MH industry composition. An imbalanced MH industry may lead to inadequate management of MH disorders.
Health outcomes, medicaid, medical care, mental health services, mental health industry
Introduction Prevalence of mental health (MH) disorders in the United States (US) increased over the past decade (Mojtabai, 2011; Whitaker 2010). In 1999, the average number of mentally unhealthy days per month was three days, and the percentage of population reporting 14 or more mentally unhealthy days per month was 8.9% (Centers for Disease Control and Prevention, 2013). By 2009, the average number of mentally unhealthy days increased to 3.5 days, and the percentage of the population reporting mental distress lasting 14 or more days went up to 10.6%. In the US, individuals may seek treatment for MH disorders through inpatient and outpatient settings. Inpatient psychiatric and substance abuse (SA) hospitals provide extended-stay diagnostic, treatment and MH service monitoring. Residential MH and SA facilities provide residents with room and board, in addition to supervision, counseling and related MH support services. There are three types of outpatient establishments: (1) MH physician offices, (2) MH
Correspondence: Rosa Rodriguez-Monguio, PhD, Associate Professor, University of Massachusetts, Amherst-School of Public Health and Health Sciences, 715 N. Pleasant Street, 322 Arnold House, Amherst, MA 01003, USA. Tel: 1 413 545 7427. Fax: 1 413 545 1645. E-mail: [email protected]
History Received 24 July 2013 Revised 28 May 2014 Accepted 30 May 2014 Published online 2 September 2014
practitioner offices and (3) MH clinics. MH physician offices are private facilities that provide psychiatric services and/or psychoanalysis as performed by a physician (i.e. psychiatrist). MH practitioner offices are also privately operated; these facilities employ non-physician practitioners who specialize in the diagnosis and treatment of MH disorders. Finally, MH clinics employ physician and non-physician MH professionals who collectively provide diagnostic and treatment-related services (Bureau of Labor Statistics, 2013a). MH treatment modalities may include counseling and/or pharmaceutical management. Psychiatrists provide therapy and prescribe pharmaceuticals. Psychologists, MH and SA social workers and MH counselors provide therapeutic services. The 2011 average hourly wage for psychiatrists was $83.73, followed by psychologists at $35.14, MH and SA social workers at $20.50 and MH counselors at $20.48 (Bureau of Labor Statistics, 2013b). Individuals who meet means-based testing criteria may access care through state Medicaid programs. Medicaid is the largest payer of MH treatment services (Centers for Medicare and Medicaid Services, 2013). In 2005, $113 billion was spent on MH services in the US; the share of MH spending funded by Medicaid was 28% (Substance Abuse and Mental Health Services Administration, 2010). Driving forces reshaping Medicaid MH provisions include reimbursement rates and cost-shifting opportunities through federal matching funds, expansion of community-based providers, managed
Medicaid and mental health industry
DOI: 10.3109/09638237.2014.951486
313
and Prevention, 2013). Proxy variables for MH industry are selected as they represent the majority of facilities and professionals qualified to treat MH disorders in the US. Secular trends in Medicaid provisions (i.e. Medicaid beneficiary rates and overall and MH expenditures per beneficiary) are compared with proxy variables for the MH industry (i.e. MH facility gross payroll per-capita and MH professionals per 100 000 people) to identify possible associations. State fixed-effects regression models are performed to assess the relationship between Medicaid provisions and the MH industry. The regression model is: Yit ¼ 0 þ 1 Xit þ i þ "it
Downloaded by [University of Birmingham] at 00:55 06 November 2015
Figure 1. Medicaid provisions and the US mental health industry composition.
care arrangements and new pharmacological treatment (Frank et al., 2003; Gronfein, 1985; Rowland et al., 2003; Sullivan, 2006; Taube et al., 1990). Previous studies assess the relationship between economic conditions and Medicaid MH spending (Ellis et al., 2012; Frank, 1985; Mark et al., 2011; Minoiu & Andres, 2008; Pellegrini & Rodriguez-Monguio, 2013) and the effect of Medicaid provisions on the US healthcare workforce composition (Pellegrini et al., 2014). However, the extent to which Medicaid provisions are related with privately delivered MH services has not been studied on a national level. Thus, this study assesses the relationship between Medicaid provisions and the MH industry composition. We hypothesize that as a primary payer for the MH services, Medicaid provisions shape the supply of MH facilities and professionals (Figure 1). Furthermore, we hypothesize that increasing Medicaid beneficiaries and MH service demand, along with Medicaid funding constraints, results in MH industry structural changes and provider imbalances to effectively treat individuals with MH disorders.
Data and methods Data are collected from readily available datasets for 50 states and the District of Columbia for the 1999–2009 time period. Medicaid beneficiary and overall and MH spending data are collected from the Centers for Medicare and Medicaid Services (Centers for Medicare and Medicaid Services, 2013). MH facilities data are gathered from the Bureau of Labor Statistics’ Quarterly Census of Employment and Wages program, including: (1) psychiatric and SA hospitals, (2) residential MH and SA facilities, (3) MH physician offices, (4) MH practitioner offices and (5) MH clinics (Bureau of Labor Statistics, 2013a). Occupational employment data are collected from the Bureau of Labor Statistics’ Occupational Employment Statistics program. Four MH-related occupations are included in the study: (1) psychiatrists, (2) psychologists, (3) MH and SA social workers and (4) MH counselors (Bureau of Labor Statistics, 2013b). Statewide annual average number of mentally unhealthy days per month data are obtained from the Centers for Disease Control and Prevention’s Behavioral Risk Factor Surveillance System (Centers for Disease Control
where Yit represents MH facility or professional type, Xit represents Medicaid provisions and "it represents the error term. The state fixed-effects model is performed as Medicaid is a state-managed program. Thus, the state fixed-effects dummy variable, ai, represents omitted variables that vary across states but not over time. Expenditure data are adjusted to 2010 dollars using the consumer price index for all urban consumers. Count data are converted to rates per-capita (i.e. facility gross payroll), per beneficiary (i.e. Medicaid overall and MH spending) or per 100 000 people (i.e. Medicaid beneficiaries and occupational employment). All statistical tests are based on a two-sided a significance level of p50.05. Statistical analyses are performed with STATA version 12.0 statistical software (College Station, TX).
Results In the 1999–2009 study period, the average Medicaid beneficiary rate per 100 000 people for 50 states and DC is 13 625; increasing from 11 164 in 1999 to 15 453 in 2009 (38.42% increase). The study period average Medicaid MH spending per beneficiary is $445; decreasing from $534 in 1999 to $384 in 2009 (28.09% decrease). The average number of mentally unhealthy days also increased from 3 days in 1999 to 3.37 in 2009 (Table 1). There are changes in MH facility gross payroll per-capita. MH clinics represent the highest spending per-capita over the study period; increasing from $23.79 in 1999 to $26.34 in 2009 (10.72% increase). Likewise, in 1999 and 2009, the smallest number of per-capita dollars is spent at MH practitioner offices, $6.93 and $7.87, respectively, followed by MH physician offices, $7.70 and $7.97, respectively. Growth rates differ for each of the five examined facility types. Residential MH and SA facilities experienced the largest increase in per-capita gross payroll; from $13.66 percapita in 1999 to $19.05 in 2009 (39.46% increase). To the contrary, psychiatric and SA hospitals show a study period decrease (11.62%; Table 1). There are also changes to MH occupational employment in the 1999–2009 study period. In 1999, psychologist employment represents the largest occupational group (32.79 per 100 000). However, of the four examined occupations, psychologist employment shows the smallest growth rate in occupational employment per 100 000 people (4.27%).
314
L. C. Pellegrini et al.
J Ment Health, 2014; 23(6): 312–316
Table 1. Descriptive statistics and trends. 1999–2009 time period
Downloaded by [University of Birmingham] at 00:55 06 November 2015
Variable name Medicaid provisions Medicaid beneficiaries Medicaid spending Medicaid MH spending Mentally unhealthy days Mentally unhealthy days Mental health facilities Psychiatric hospitals Residential MH and SA facilities MH physician offices MH practitioner offices MH clinics Mental health professionals Psychiatrists Psychologists MH and SA social workers MH counselors
Trends
No. of obs.
Mean
SD
Minimum
Maximum
Mean 1999
Mean 2009
Average % change 1999–2009
561 561 533
13 625 7473 445
4474 1939 418
5097 3852 3
25 112 13 481 2573
11 164 7148 534
15 453 7622 384
38.42% 6.63% 28.09%
531
3.27
0.50
0.70
5.70
3.00
3.37
12.36%
348 521 542 544 492
12.50 15.93 7.96 7.49 25.09
7.77 8.96 6.00 5.09 21.77
1.23 2.26 0.74 1.13 0.87
42.07 52.49 51.49 35.81 135.13
14.29 13.66 7.70 6.93 23.79
12.63 19.05 7.97 7.87 26.34
11.62% 39.46% 3.51% 13.56% 10.72%
466 557 541 531
8.12 35.12 41.46 31.68
7.71 16.41 25.25 19.47
0.87 5.39 5.95 4.70
64.04 125.42 211.84 131.18
5.96 32.79 29.61 23.82
8.08 34.19 50.35 35.96
35.57% 4.27% 70.04% 50.97%
Medicaid beneficiaries and mental health professionals’ data are in rates per 100 000 people. Medicaid overall and MH spending data are per beneficiary. Mental health facilities’ data are gross-payroll per-capita. Source: Medicaid data derived from the Centers for Medicare and Medicaid Services. Mentally unhealthy days data derived from the Centers for Disease Control and Prevention’s Behavioral Risk Factor Surveillance System. Mental health facilities and professionals’ data derived from the Bureau of Labor Statistics’ Quarterly Census of Employment and Wages (QCEW) and Occupational Employment Statistics (OES) survey, respectively.
The largest study period employment growth rates are for MH and SA social workers (70.04%) and MH counselors (50.97%). In 2009, MH and SA social workers and MH counselors represent the largest occupational group at 50.35 and 35.96 per 100 000 people, respectively, followed by psychologists at 34.19 per 100 000 people. Psychiatrist employment represents the smallest occupational employment group both in 1999 and 2009. In 1999, there are 5.96 psychiatrists per 100 000 people, increasing to 8.08 in 2009 (35.57% increase; Table 1). Results show that the Medicaid beneficiary rate is associated with per-capita gross payroll gains at all examined facilities excluding MH practitioner offices. However, the effect size is largest for MH clinics followed by residential MH and SA facilities (p50.001). Per-capita gross payroll growth at MH physician offices (p50.05) and psychiatric hospitals (p50.01) is least related with the Medicaid beneficiary rate. Similar effect sizes are found for Medicaid overall and MH spending per-beneficiary. Nevertheless, the relationships between Medicaid MH spending per beneficiary and MH facilities’ gross payroll per-capita are not statistically significant. Study results also evidence that MH and SA social worker employment growth per 100 000 people is most related with Medicaid provisions. A one percentage point increase in the Medicaid beneficiary rate and Medicaid overall, and MH spending per beneficiary is associated with a 0.003 (p50.001), and 0.002% (p50.05) increase, and 0.011% (p50.05) decrease in the number of MH and SA social workers per 100 000 people, respectively. To the contrary, the association between Medicaid beneficiary rate, MH spending, and psychologist employment per 100 000 people is not statistically significant. Likewise, there is no statistically significant relationship between psychiatrist employment and either the
Medicaid beneficiary rate or overall and MH spending per beneficiary. However, an increase in Medicaid overall spending spending per beneficiary is associated with a 0.003% increase in the number of psychologists per 100 000 people (p50.05) (Table 2).
Discussion Access to MH services is often challenging for underserved individuals with MH disorders (Carbaugh et al., 2006; Ellis et al., 2012; Mojtabai, 2011). State Medicaid programs play an important role in the configuration of MH facilities and professionals serving beneficiaries with mental disorders. In the study period, there are significant changes in the composition of the MH industry. Results show that MH clinics continue to maintain a dominant role. However, growth in residential MH and SA facilities further evidences trends in managing the chronically mentally ill in the community as opposed to psychiatric institutions; as of 2009, residential facilities’ gross-payroll spending per-capita is higher than psychiatric hospitals’ spending. Medicaid provisions have a strong relationship with residential MH and SA gross-payroll spending. Nevertheless, increases in residential spending have not been met with an equally significant, and positive, association between Medicaid provisions and MH physician or practitioner offices’ grosspayroll spending; both of these facilities employ significant numbers of outpatient psychiatrists and psychologists (Bureau of Labor Statistics, 2013b). The analysis of MH professionals reveals that there is significant growth in employment of MH and SA social workers and MH counselors relative to psychologists; both of these professionals earn approximately 58% the market wage of psychologists. The Medicaid beneficiary rate is significantly related with employment growth for MH and SA
Medicaid and mental health industry
DOI: 10.3109/09638237.2014.951486
315
Table 2. Medicaid provisions and the US mental health industry composition. Mental health facilities Psychiatric hospitals
Downloaded by [University of Birmingham] at 00:55 06 November 2015
Medicaid beneficiaries 0.000222b Standard error (0.0000764) Sample size 348 0.919 R2 Medicaid spending 0.000160 Standard error (0.000234) Sample size 348 0.917 R2 Medicaid MH spending 0.00135 Standard error (0.000997) Sample size 334 0.918 R2
Mental health professionals
Residential MH MH physician MH practitioner and SA facilities offices offices 0.000637a (0.0000776) 521 0.900 0.000704b (0.000229) 521 0.887 0.0000614 (0.00101) 494 0.884
0.000144c (0.0000562) 542 0.875 0.000277 (0.000168) 542 0.874 0.000826 (0.000676) 514 0.875
0.0000311 (0.0000550) 544 0.830 0.000339c (0.000154) 544 0.832 0.000373 (0.000556) 516 0.851
MH clinics
MH and SA social MH Psychiatrists Psychologists workers counselors
0.000814a 0.0000679 0.000154 0.00330a 0.000865b (0.000106) (0.0000768) (0.000217) (0.000355) (0.000278) 492 466 557 541 531 0.971 0.888 0.746 0.717 0.729 0.00156a 0.000357 0.00260a 0.00218c 0.000764 (0.000313) (0.000216) (0.000598) (0.00108) (0.000783) 492 466 557 541 531 0.968 0.888 0.754 0.670 0.724 c 0.000139 0.000310 0.00133 0.000283 0.0108 (0.00129) (0.000954) (0.00246) (0.00450) (0.00319) 466 441 531 513 504 0.967 0.888 0.747 0.663 0.720
Standard errors in parentheses a p50.001, bp50.01, cp50.05. Medicaid beneficiaries and mental health professionals’ data are in rates per 100 000 people. Medicaid overall and MH spending data are per beneficiary. Mental health facilities’ data are gross-payroll per-capita. Source: Medicaid data derived from the Centers for Medicare and Medicaid Services. Mental health facilities and professionals’ data derived from the Bureau of Labor Statistics’ Quarterly Census of Employment and Wages (QCEW) and Occupational Employment Statistics (OES) survey, respectively.
social workers and MH counselors; whereas, the association is not statistically significant for psychologists. In addition, there is an increase in psychiatrist employment during the study period. Nevertheless, the relationship between the Medicaid beneficiary rate and overall and MH spending per beneficiary and psychiatrist employment is not statistically significant. The associations between Medicaid provisions and the MH industry composition may be related with underlying Medicaid reimbursement mechanisms. State Medicaid programs have often set reimbursement levels below that of the private sector, financially discouraging MH providers (Frank et al., 2003). As a result, publicly funded state Medicaid programs may be constrained in their ability to provide more costly private sector psychiatric and/or psychological services to their beneficiaries. Nevertheless, from a public health standpoint, each type of healthcare facility and professional provides a unique array of health services amongst the continuum of care of MH disorders. The MH industry composition becomes increasingly important as current clinical practice further evidences that adjunctive therapy and pharmacological maintenance is often the preferred treatment modality for MH disorders. However, of all examined MH professionals, only psychiatrists are licensed to prescribe pharmaceuticals in all fifty states and DC. Furthermore, other associated MH professionals possess differing levels of pharmaceutical training. While Medicaid provisions include coverage for psychotropic drugs, it is also associated with growth of specific types of facilities and professionals, as influenced by state-specific reimbursement rates. Thus, Medicaid reimbursement rates along with trends toward pharmacological treatment of mental disorders in the US present additional challenges to the delivery of optimum MH services to patient populations. In addition, chronically mentally ill individuals are likely to be Medicaid beneficiaries, thus; reduced access to the breadth of MH services becomes increasingly worrisome in a postdeinstitutionalization era.
Limitations This study has some potential limitations worth noting. First, our proxy variable for mental distress, mentally unhealthy days, is a self-reported measure that may be influenced by sampling methodologies and response rates. Furthermore, an increase in the average number of mentally unhealthy days does not necessarily indicate a need for MH services. Second, this study examines facilities and professionals that are consistent across states as classified by the Bureau of Labor Statistics. Thus, our MH industry proxy variables may exclude additional state-specific facilities and professionals that play an instrumental role in state MH industries. In addition, examined MH facilities and professionals are not exclusively Medicaid providers. Third, Medicaid is a stateadministered program with provisions varying from state to state. As such, state-specific reimbursement policy likely influences the configuration of each state’s MH industry. Medicaid overall and MH spending per beneficiary is used to proxy these reimbursement policies. Fourth, our model examining the relationship between Medicaid provisions and the MH industry may be subject to omitted variable bias. For example, state Medicaid provisions may be correlated with state economic conditions, and such conditions could influence mental distress and gross-payroll spent at examined MH facilities. Furthermore, state-specific MH occupational employment may be related with other factors not examined in this study, including availability of graduate training programs and state specific licensing requirements. Finally, our model may also be subject to reverse causality; the MH industry may affect Medicaid provisions much like Medicaid provisions may affect the MH industry. However, this study does not aim to evaluate causality but to assess associations between study measures.
Conclusion In the period 1999–2009, there are increases in the statewide average number of mentally unhealthy days per month. There
Downloaded by [University of Birmingham] at 00:55 06 November 2015
316
L. C. Pellegrini et al.
are also significant changes in the MH industry composition as related to changes in Medicaid provisions. The Medicaid beneficiary rate is related with per-capita gross payroll spending growth at MH clinics and residential MH and SA facilities and with occupational employment gains per 100 000 people for MH and SA social workers and MH counselors. Furthermore, the Medicaid beneficiary rate is also related with per-capita gross payroll spending growth at MH physician offices and psychiatric hospitals. The relationship between psychologist employment per 100 000 people and the Medicaid beneficiary rate and MH spending per beneficiary is not statistically significant. Likewise, there is no statistically significant association between psychiatrist employment and either the Medicaid beneficiary rate or Medicaid overall and MH spending per beneficiary. Each type of facility and health professional provides a unique array of services in the MH continuum of care. As a result, imbalances in the US MH industry composition may lead to barriers in access to MH services and inadequate management of MH disorders.
Declaration of interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.
References Bureau of Labor Statistics. (2013a). Labor force statistics from the Quarterly Census of Employment and Wages Survey. Available from: http://www.bls.gov/qcew/ [last accessed 15 Feb 2013]. Bureau of Labor Statistics. (2013b). Labor force statistics from the Occupational Employment Statistics Survey. Available from: http:// www.bls.gov/oes/ [last accessed 15 Feb 2013]. Carbaugh AL, Elias R, Rowland D. (2006). Aid to people with disabilities: Medicaid’s growing role. Annu Rev Publ Health, 27, 417–42.
J Ment Health, 2014; 23(6): 312–316
Centers for Disease Control and Prevention. (2013). Behavioral Risk Factor Surveillance System (BRFSS). Available from: http:// www.cdc.gov/brfss [last accessed 15 Feb 2013]. Centers for Medicare and Medicaid Services. (2013). Medicaid and Statistical Information System (MSIS). Available from: http:// www.medicaid.gov [last accessed 15 Feb 2013]. Ellis WR, Huebner C, Vander Stoep A, Williams M. (2012). Washington State exhibits wide regional variation in proportion of Medicaideligible children who get needed mental health care. Health Affairs, 31, 990–9. Frank RG. (1985). A model of state expenditures on mental health services. Public Financ Quart, 13, 319–38. Frank RG, Goldman HH, Hogan M. (2003). Medicaid and mental health: Be careful what you ask for. Health Aff, 22, 101–13. Gronfein W. (1985). Incentives and intentions in mental health policy: A comparison of the Medicaid and community mental health programs. J Health Soc Behav, 26, 192–206. Mark TL, Levitt KR, Vandivort-Warren R, et al. (2011). Changes in US spending on mental health and substance abuse treatment, 1986–2005, and implications for policy. Health Aff, 30, 284–92. Minoiu C, Andres AR. (2008). The effect of public spending on suicide: Evidence from US state data. J Soc Econ, 37, 237–61. Mojtabai R. (2011). National trends in mental health disability, 1997-2009. Am J Public Health, 101, 2156–63. Pellegrini LC, Rodriguez-Monguio R. (2013). Unemployment, Medicaid provisions, the mental health industry, and suicide. Soc Sci J, 50, 482–90. Pellegrini LC, Rodriguez-Monguio R, Qian J. (2014). The US healthcare workforce and the labor market effect on healthcare spending and health outcomes. Int J Health Care Finance Econ, 14, 127–41. Rowland D, Garfield R, Elias R. (2003). Accomplishments and challenges in Medicaid mental health. Health Aff, 22, 73–83. Substance Abuse and Mental Health Services Administration, Center for Mental Health Services and Center for Substance Abuse Treatment. (2010). Spending Estimates Project 2010. Available from: http:// www.samhsa.gov/ [last accessed 15 Feb 2013]. Sullivan WP. (2006). Mental health services in the 21st century: The economics and practice challenges on the road to recovery. Adv Soc Work, 6, 193. Taube CA, Goldman HH, Salkever DS. (1990). Medicaid coverage for mental illness: Balancing access and costs. Health Aff, 9, 5–18. Whitaker R. (2010). Anatomy of an epidemic: Magic bullets, psychiatric drugs, and the astonishing rise of mental illness in America. 1st ed. New York, NY: Crown Publishers.
