Impact of Medicaid Institution for Mental Diseases Exclusion on Serious Mental Illness Outcomes

ABSTRACT

Objectives: Medicaid’s Institution for Mental Diseases (IMD) exclusion bars federal funding for treatment in facilities with more than 16 psychiatric beds, but some states have obtained waivers under Section 1115 of the Social Security Act to increase options for treating serious mental illness (SMI). This retrospective claims study assessed health care resource utilization, costs, homelessness, and incarceration among Medicaid beneficiaries with SMI in states with and without waivers.

Study Design: Patients were 18 years and older and had at least 1 diagnosis of SMI and 12 months of continuous enrollment pre– and post index date.

Methods: Fixed-effect models, adjusted for patient and state characteristics, estimated the waivers’ impact on outcomes.

Results: The odds of having psychiatric-specific inpatient admissions and emergency department (ED) visits were lower by 14% and 26%, respectively, in the waiver cohort (n = 130,224) vs the nonwaiver cohort (n = 3,102,971). Odds of all-cause inpatient admissions and ED visits were also lower (9% for both) in the waiver cohort, but the odds of having all-cause outpatient visits were 19% greater in the waiver cohort. States with waivers had 11% fewer incarcerations, or about 250 fewer cases per year, based on an average of 23,592 incarcerations.

Conclusions: Our findings underscore the beneficial impact of IMD exclusion waivers on psychiatric-specific and all-cause health care resource utilization and costs as well as on incarceration rates for individuals with SMI. CMS may want to consider the results of this study in addition to other available data when granting waivers to states and potentially removing this exception from the Medicaid law.

Am J Manag Care. 2025;31(12):In Press

Takeaway Points

• This research underscores the beneficial impact of the Medicaid Institution for Mental Diseases (IMD) exclusion waivers on psychiatric-specific and all-cause health care resource utilization and costs, as well as incarceration rates, for individuals with serious mental illness.
• Although the IMD exclusion was put into place to protect patients from prolonged institutionalization, our findings indicate that state waivers allowing treatment in IMDs do not lead to increased psychiatric hospitalization.
• Additional state-level policies such as Medicaid Health Homes and Medicaid expansion should be considered by state policy makers and Medicaid administrators when assessing the risk-benefit trade-offs of investing in mental health services.

Serious mental illness (SMI) refers to a category of mental health conditions that encompasses major depressive disorder, schizophrenia, bipolar disorder, mania and its related disorders, and specific personality disorders.1-3 Approximately 5.5% of all US adults experience SMI, and 26% of them rely on Medicaid for access to essential behavioral health services.4,5

The Institution for Mental Diseases (IMD) exclusion is a Medicaid law that prohibits the federal government from providing federal funds to states for services rendered to Medicaid enrollees aged 21 to 65 years who receive treatment in an IMD. IMDs are facilities with more than 16 beds that primarily provide psychiatric or psychological care and are under the jurisdiction of state mental health authorities.6-8 It is the only part of the Medicaid law that prohibits funding for particular facilities.

The intent of the IMD exclusion was to promote a transition from large institutional settings to community-based care.9 However, because states could no longer receive federal funding for care provided in IMDs, many facilities were downsized or closed altogether. This policy ultimately reduced the overall supply of psychiatric hospital beds available for patients with SMI. By 2016, the US had fewer than 12 psychiatric hospital beds per 100,000 patients.10 A lack of psychiatric beds has led to situations in which individuals faced prolonged stays in emergency departments (EDs), awaited admission to hospitals, or were confined in correctional facilities.11 Furthermore, many individuals with SMI experience homelessness or inadequate treatment, which contributes to poor mental and physical health.8

Recognizing the complex needs of individuals with SMI, many states have pursued waivers for the IMD exclusion. These waivers, granted under Section 1115 of the Social Security Act, allow states to seek federal reimbursement for certain psychiatric services provided in IMDs.6 As of July 2025, 11 states had approved waivers and 7 had pending waivers.6,12

Despite the potential benefits of IMD exclusion waivers in expanding access to acute psychiatric care, questions remain about their broader impact on health care resource utilization (HCRU) and costs. This retrospective claims-based study addressed these questions by examining the association of IMD exclusion waivers with HCRU and cost within the Medicaid population. Exploratory analyses also assessed the impact of the waivers on homelessness and incarceration. By understanding the effects of IMD waivers, progress can be made toward a more equitable and effective behavioral health care system.

METHODS

A retrospective cohort study was conducted using Kythera Labs Medicaid data from January 2016 to July 2023. Specifically, this study utilized the Medicaid closed portion, which included 44,470,509 patients and 901,136,733 claims for the study period. The data set contains deidentified patient-level data, including demographic data (age, sex, insurance type, zip code) and treatment information (International Statistical Classification of Diseases, Tenth Revision diagnoses; Current Procedural Terminology codes; and National Drug Codes for medications).13 Unique patient identifiers linking all encounters allowed for longitudinal analyses. The validity and consistency of health care outcomes derived from these data were compared with other data sets.14-16

Patients were included in the study if they had at least 1 medical claim for SMI (index date) during the identification period between January 1, 2017, and July 31, 2022; were 18 years or older at the index date; and had continuous enrollment for 12 months pre– and post index date. Patients were classified into 2 cohorts: those in the 11 states with an IMD exclusion waiver (waiver cohort) and those in the other 39 states without an IMD exclusion waiver (nonwaiver cohort).

Primary outcomes included the impact of IMD waivers on HCRU and associated costs for inpatient, ED, outpatient, and pharmacy visits. Exploratory analyses were also conducted to assess the effects of IMD waivers on homelessness and incarceration. Homelessnesswas defined as the estimated number of people experiencing homelessness. A homeless person was defined as one who lacks a fixed, regular, and adequate nighttime residence. Homelessness numbers were obtained from the US Department of Housing and Urban Development, which has the 2016-2023 state estimates of homelessness.17 Jail incarcerationwas defined as the number of people held in the custody of the state jail system (usually sentenced to < 1 year).18 This variable was obtained from the Annual Survey of Jails Data Series—funded by interagency agreements through the Bureau of Justice Statistics, the National Institute of Justice, and the Office of Justice Programs’ Office of Juvenile Justice and Delinquency Prevention—which contains estimates of jail incarceration from 2016 to 2023.18

Descriptive analyses compared patients in each cohort on demographic characteristics (age, sex), clinical characteristics (concomitant mental and physical health conditions), and HCRU and cost prior to adjusting. To proxy differences in patient severity among the cohorts, Elixhauser Comorbidity Index19 scores and flags for any mental health comorbidities and systematic comorbidities before the initial SMI diagnosis were created. Numeric and percentage distributions were presented for dichotomous and polychotomous variables, and means and SDs were reported for continuous variables. P values were computed using the χ2 test for dichotomous variables and t tests for continuous variables. Standardized differences were also calculated alongside P values for each variable.

Analyses were then performed to adjust for differences in demographic and clinical characteristics and state-level policy considerations. For cost estimation, the following fixed-effects model was used:

log(Total Cost or PSY Cost)i,s = β0 + β1waves + Pi,s + Xs + ui,s

where log(Total Cost or PSY Cost)i,s is the all-cause health care cost or psychiatric-specific cost for individual i in state s. Wave is 1 if the state has a waiver and 0 otherwise at the index year; Pi,s is a vector of individual characteristics such as age, sex, and comorbidities; and Xs is a vector of state-level characteristics, depending on data availability, including a binary indicator for a prescription drug monitoring program, a binary indicator for a medical marijuana law,20 a binary indicator for a Medicaid Health Home for mental illness and/or substance use disorder (SUD),20 a binary indicator for Affordable Care Act (ACA)–related Medicaid expansion,12 unemployment rate,21 number of behavioral health care providers per 100,000 state residents,22 and Substance Abuse and Mental Health Services Administration (SAMHSA) block grants for mental illness and SUD prevention and treatment per capita.23 The term u represents the error term, which captures all unobserved factors affecting the outcome variable for individual i in state s that are not explained by the included model covariates. Coefficients of β1 showed the percentage change of all-cause or psychiatric-specific costs due to the waiver. SEs were clustered by state.

The same set of explanatory variables was used for the other outcome measures. Logistic models were employed to assess the probability of all-cause or psychiatric-specific inpatient, outpatient, ED, and pharmacy visits. A standard α level of .05 was used to determine statistical significance.24 All statistical analyses were conducted using the R 4.4.1 statistical package(R Foundation for Statistical Computing).25

For incarceration and homelessness analysis, the following autoregressive model of order 1 was considered:

yt = β0 + β1yt – 1 + β2wavet + Xt + ut

where yt is the incarceration (or homelessness) number at year t, yt – 1 is the incarceration (or homelessness) number at year t – 1, wavet is a binary variable (1 if the state has a waiver at year t), and Xt is the vector of state-level characteristics. The sign and significance of the estimated β2 showed whether the waiver had a positive or negative effect on the incarceration (or homelessness) number, controlling for the incarceration (or homelessness) number for the previous year. All significant autocorrelation terms in the regression model were controlled for, and the Durbin-Watson statistic in the final adjusted models was examined to determine the nonsignificance of first-order autocorrelation of the regression residuals.

RESULTS

After applying inclusion criteria, 130,224 patients were identified in states with an IMD waiver and 3,102,971 patients were identified in states without an IMD waiver (Table 1).

Unadjusted Descriptive Analysis

Patients in the nonwaiver cohort were older than those in the waiver cohort (mean age, 43.18 vs 41.27 years; P < .0001) and more likely to be female (66.50% vs 65.12%; P < .0001). Elixhauser Comorbidity Index, Chronic Disease Score, and Charlson Comorbidity Index scores were significantly lower in the waiver states vs nonwaiver states (1.71 vs 2.04, 1.71 vs 1.77, and 0.54 vs 0.67, respectively; all P < .0001). Additionally, systemic and mental health comorbidities were lower in the waiver states (Table 2).

Unadjusted utilization of all-cause inpatient and emergency care was higher in the nonwaiver cohort. However, states with waivers had higher utilization of outpatient care and pharmacy. The opposite trend was noted for psychiatric-specific HCRU, where inpatient hospitalization was higher in the waiver cohort and outpatient and pharmacy visits were higher in the nonwaiver cohort.

The unadjusted cost analysis (eAppendix Table 1 [eAppendix available at ajmc.com]) showed that all-cause total cost was higher in the nonwaiver cohort than in the waiver cohort ($12,774.68 vs $12,327.16; P < .0001). This was driven by higher inpatient ($2065.35 vs $1712.52; P < .0001), outpatient ($5378.40 vs $5280.08; P = .0068), and ED ($1062.76 vs $939.31; P < .0001) costs. However, pharmacy costs were higher in states with waivers ($4395.25 vs $4268.17; P = .0006). Total psychiatric-specific health cost ($4033.89 vs $2447.17; P < .0001) was higher in the nonwaiver cohort, driven by higher inpatient ($2011.58 vs $706.27; P < .0001), outpatient ($578.52 vs $459.32; P < .0001), ED ($37.93 vs $15.72; P < .0001), and pharmacy ($1405.86 vs $1265.86; P = .0001) costs.

Adjusted Multivariable Analysis

Table 3 shows the adjusted ORs for the likelihood of health care utilization after controlling for patient and state differences. In terms of psychiatric-specific utilization, the waiver cohort was 14% less likely to have inpatient admissions and 26% less likely to have ED visits than the nonwaiver cohort (both P < .0001). The differences in odds of having psychiatric-specific outpatient and pharmacy visits were not significant between the 2 cohorts. Similarly, the waiver cohort was 9% less likely to have all-cause inpatient admissions and 9% less likely to have all-cause ED visits (both P < .0001). There was no significant difference in the likelihood of having an outpatient or pharmacy visit between the 2 cohorts.

Table 4 presents the coefficient estimates for health care costs, adjusting for variations in patient characteristics and state factors. Psychiatric-specific total costs were 41% lower in the waiver cohort than in the nonwaiver cohort (P = .0094). This difference was driven by 38% lower inpatient (P < .0001) and 28% lower ED (P = .0014) costs in the waiver cohort. There were no significant differences in psychiatric-specific outpatient and pharmacy costs or any all-cause health care expenditures between states with IMD waivers and those without waivers.

In addition to age, sex, and comorbid conditions, the largest contributing state policy factors to differences in psychiatric-specific inpatient cost and utilization were having Medicaid Health Homes and SAMHSA block grants for mental health/SUD, which were both significantly associated with lower inpatient utilization and costs. Only age, sex, and comorbid conditions were factors in the other cost and utilization types. In terms of all-cause utilization and cost, a higher number of behavioral health providers was associated with lower inpatient, ED, and outpatient costs, and having mental health homes was also associated with lower all-cause inpatient costs (eAppendix Table 2).

Effects on Incarceration and Homelessness

States with waivers showed an 11% reduction in incarcerations vs those without waivers, translating to approximately 250 fewer incarcerations annually based on an average of 23,592 incarcerations. States with Medicaid Health Homes for mental illness and/or SUD and ACA Medicaid expansion showed negative effects on incarceration numbers (ie, fewer incarcerations). However, states that had patients with elevated comorbidity scores had elevated incarceration rates. Additionally, having more behavioral health care providers was associated with a state having fewer incarcerations, although these effects were only marginally significant (eAppendix Table 3).

The results indicated that waivers had no significant effect on homelessness. However, ACA Medicaid expansion was associated with a decrease in homelessness, but Medicaid Health Homes for mental health/SUD did not have an effect. In addition, states with a higher percentage of men exhibited increased homelessness (eAppendix Table 4).

DISCUSSION

This study examined the effect of Medicaid IMD exclusion waivers on HCRU and costs in patients with SMI. Although the IMD exclusions were enacted to prevent inappropriate institutionalization of patients with mental illness,7 they have made access to care more difficult for the neediest of patients with SMI. The Medicaid IMD exclusion policy may affect access to inpatient behavioral health services, leading to delayed care, lack of access to inpatient beds, or premature discharge for patients.11,26 Previous research indicated that IMD waivers for patients with SUD improved access to care, treatment use, and outcomes.27,28 To our knowledge, this is the first study to analyze the impact of Medicaid IMD exclusion waivers on outcomes for patients with SMI.

Unadjusted analysis revealed higher utilization of all-cause outpatient care and pharmacy services along with decreased use of inpatient and emergency care in states with IMD waivers. This suggests a shift toward preventive and community-based health care delivery models. However, psychiatric-specific HCRU demonstrated a contrasting trend, with higher inpatient hospitalizations observed in waiver states. Patients in states with IMD waivers were older and exhibited significantly lower comorbidity scores than their counterparts in nonwaiver states. This suggests potential differences in patient populations and health care needs between waiver and nonwaiver states, which could influence HCRU and costs.

After adjusting for these patient characteristics and state-level policies related to mental health treatment, both all-cause and psychiatric-specific inpatient and ED utilization were higher in nonwaiver states. These findings were unexpected. Because the IMD waiver allows federal funding to be used for the treatment of patients in IMDs, an increase in psychiatric-specific inpatient care would be expected. However, this analysis found that states with waivers had reduced acute care utilization. Interestingly, incarceration rates were also lower in states with IMD waivers, highlighting the connection between mental health treatment and incarceration. Although CMS put the IMD exclusions into Medicaid law to protect patients with SMI from extended institutionalization in state hospitals, these results indicate that hospitalizations actually decrease in states with waivers where treatment in IMDs is allowed. The lower utilization of all-cause and psychiatric-specific inpatient and ED services in states with waivers also decreased costs, leading to cost savings for those states.

These differences could be attributed to waiver states’ commitment to expanding nonresidential and outpatient community-based mental health services when applying for the waiver. Previous research has shown that IMD waivers for SUD have a similar effect in expanding the range of SUD services available to Medicaid enrollees.29 Results of this study also showed that differences in psychiatric-specific inpatient utilization were correlated with the presence of Medicaid Health Homes and SAMHSA block grants for mental health/SUD and that lower incarceration was associated with the presence of Health Homes and Medicaid expansion. An increase in the number of behavioral health providers was also associated with lower all-cause resource utilization and costs. These findings may be indicative of improvements in outcomes due to further investment in programs to support patients with mental health disorders and SUDs. This investment in the care delivery system may have a positive impact on the care of patients with SMI and reduce the need for acute care.

Strengths and Limitations

Strengths of this study include its large sample size and comprehensive coverage of Medicaid enrollees without time lag. The inclusion of state-level policy levers on outcomes also makes this study unique. However, being an observational study, it is susceptible to bias and confounding factors.

Retrospective outcomes research studies provide valuable insights from existing data but come with limitations that can affect the validity and reliability of findings. Because patient selection was not random, potential exists for selection bias. To mitigate this bias, multivariable regression analysis was used to control for as many factors as possible based on the available data. Additionally, because the HCRU data were collected for administrative rather than research purposes, some records may be incomplete, inaccurate, or inconsistently recorded. A diagnosis code on a medical claim may not always confirm the presence of a disease, as it could be incorrectly coded or included as rule-out criteria rather than an actual diagnosis. Certain clinical and disease-specific parameters, which could have an impact on study outcomes, may not be readily available in claims data. However, range checks, format checks, and cross-validation procedures were conducted before data analysis to ensure the claims data were as accurate as possible.

Likewise, the incarceration and homelessness data were not collected for research purposes and may also contain inaccuracies. It was not possible to link the homelessness and incarceration data to specific patients contained in the claims data set; therefore, assumptions were made to estimate the percentage of people with SMI experiencing homelessness and incarceration. The effect of IMD waivers on these outcomes may be overestimated or underestimated as a result, but the findings point to the need for further research in this area.

CONCLUSIONS

This study underscores the beneficial impact of IMD exclusion waivers on psychiatric-specific and all-cause HCRU and costs as well as incarceration rates for individuals with SMI. The findings also highlight the importance of other policy investments in mental health (eg, Medicaid expansion, block grants, Health Homes) on patient outcomes. IMD waivers and other policy levers may not only enhance patient outcomes but also offer potential cost savings to states, demonstrating their crucial role in improving the overall health care systems’ effectiveness and sustainability. CMS may want to consider the results of this study in addition to other available data when granting waivers to states and potentially removing this exception from the Medicaid law. Although the IMD exclusion was put into place to protect patients from prolonged institutionalization, the findings indicate that waivers to allow treatment in IMDs do not lead to increased psychiatric hospitalization. Additional state-level policies such as Medicaid Health Homes and Medicaid expansion should be taken into consideration by state policy makers and Medicaid administrators when assessing the risk-benefit trade-offs of investing in mental health services.

Author Affiliations: Boğaziçi University (OB), İstanbul, Türkiye; City University of New York (OB), New York, NY; Otsuka Pharmaceutical Development & Commercialization, Inc (HCW, XH), Princeton, NJ; Columbia Data Analytics (NY, KR, LI, DF), New York, NY; University of Cambridge (RP), Cambridge, England.

Source of Funding: Otsuka Pharmaceutical Development and Commercialization, Inc.

Author Disclosures: Drs Waters and Han are employed by Otsuka, the study sponsor. Mr Yapar, Ms Rodchenko, and Ms Freedman are employed by Columbia Data Analytics, which is a paid consultant to Otsuka. Ms Isenman was employed by Columbia Data Analytics at the time of the study. Dr Patel has received consulting or advisory fees from Columbia Data Analytics and Otsuka. Dr Baser reports no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (OB, HCW, NY, KR, LI, XH, DF, RP); acquisition of data (OB); analysis and interpretation of data (OB, HCW, NY, KR, LI, XH, DF); drafting of the manuscript (OB, HCW, NY, KR, LI, DF, RP); critical revision of the manuscript for important intellectual content (OB, HCW, NY, KR, LI, DF, RP); statistical analysis (OB, NY); provision of patients or study materials (OB); obtaining funding (HCW); administrative, technical, or logistic support (HCW); and supervision (OB, HCW, XH, RP).

Address Correspondence to: Onur Baser, PhD, MS, City University of New York, 55 W 125th St, New York, NY 10027. Email: onur.baser@sph.cuny.edu.

REFERENCES

1. Mental illness. National Institute of Mental Health. Updated September 2024. Accessed November 9, 2023. https://www.nimh.nih.gov/health/statistics/mental-illness

2. Evans TS, Berkman N, Brown C, et al. Disparities Within Serious Mental Illness. Technical brief 25. Agency for Healthcare Research and Quality; May 2016. Accessed September 18, 2024. https://www.ncbi.nlm.nih.gov/books/NBK368427/

3. Rasic D, Hajek T, Alda M, Uher R. Risk of mental illness in offspring of parents with schizophrenia, bipolar disorder, and major depressive disorder: a meta-analysis of family high-risk studies. Schizophr Bull. 2014;40(1):28-38. doi:10.1093/schbul/sbt114

4. National Alliance on Mental Illness. Mental health by the numbers. Updated April 2023. Accessed July 1, 2024. https://www.nami.org/about-mental-illness/mental-health-by-the-numbers/

5. Zur J, Musumeci M, Garfield R. Medicaid’s role in financing behavioral health services for low-income individuals. KFF. June 29, 2017. Accessed July 14, 2024. https://www.kff.org/mental-health/medicaids-role-in-financing-behavioral-health-services-for-low-income-individuals/

6. Musumeci M. Key questions about Medicaid payment for services in “Institutions for Mental Disease.” KFF. June 2018. Accessed July 7, 2024. https://files.kff.org/attachment/Issue-Brief-Key-Questions-about-Medicaid-Payment-for-Services-in-Institutions-for-Mental-Disease

7. Musumeci M, Chidambaram P, Orgera K. State options for Medicaid coverage of inpatient behavioral health services. KFF. November 6, 2019. Accessed July 17, 2024. https://www.kff.org/mental-health/state-options-for-medicaid-coverage-of-inpatient-behavioral-health-services/

8. Geller JL. Excluding Institutions For Mental Diseases from federal reimbursement for services: strategy or tragedy? Psychiatr Serv. 2000;51(11):1397-1403. doi:10.1176/appi.ps.51.11.1397

9. National Association of Medicaid Directors. The IMD exclusion. April 2022. Accessed July 3, 2024. https://medicaiddirectors.org/wp-content/uploads/2022/04/IMD-NAMD-Federal-Policy-Briefs.pdf

10. Psychiatric bed supply need per capita. Treatment Advocacy Center. September 2016. Accessed November 12, 2025. https://www.tac.org/wp-content/uploads/2023/11/bed-supply-need-per-capita.pdf

11. Fuller DA, Sinclair E, Geller J, Quanbeck C, Snook J. Going, Going, Gone: Trends and Consequences of Eliminating State Psychiatric Beds. Treatment Advocacy Center; June 2016. Accessed July 9, 2024. https://www.tac.org/reports_publications/going-going-gone-trends-and-consequences-of-eliminating-state-psychiatric-beds/

12. Status of state action on the Medicaid expansion decision. KFF. Accessed November 9, 2024. https://www.kff.org/affordable-care-act/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act

13. Kythera Labs. Accessed September 9, 2023. https://www.kytheralabs.com/

14. Baser O, Samayoa G, Yapar N, Baser E, Mete F. Use of open claims vs closed claims in health outcomes research. J Health Econ Outcomes Res. 2023;10(2):44-52. doi:10.36469/001c.87538

15. Baser O, Samayoa G, Rodchenko K, Isenman L, Baser E, Yapar N. The association between weight loss medications and cardiovascular complications. Obesity (Silver Spring). 2024;32(7):1401-1409. doi:10.1002/oby.24037

16. Baser O, Isenman L, Baser S, Samayoa G. Impact of semaglutide on osteoarthritis risk in patients with obesity: a retrospective cohort study. Obes Sci Pract. 2024;10(3):e762. doi:10.1002/osp4.762

17. Office of Policy Development and Research. Annual Homelessness Assessment Report: 2023 AHAR: part 1 - PIT estimates of homelessness in the US. December 2023. Accessed October 2, 2024. https://www.huduser.gov/portal/datasets/ahar/2023-ahar-part-1-pit-estimates-of-homelessness-in-the-us.html

18. Carson EA. Prisoners in 2021 — statistical tables. Bureau of Justice Statistics. December 2022. Accessed November 9, 2023. https://bjs.ojp.gov/library/publications/prisoners-2021-statistical-tables

19. Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care. 1998;36(1):8-27. doi:10.1097/00005650-199801000-00004

20. Health home information resource center. Medicaid.gov. Accessed November 9, 2023. https://www.medicaid.gov/resources-for-states/medicaid-state-technical-assistance/health-home-information-resource-center

21. State unemployment rates, seasonally adjusted. US Bureau of Labor Statistics. Accessed December 20, 2024. https://www.bls.gov/charts/state-employment-and-unemployment/state-unemployment-rates-map.htm#

22. Mental health providers in United States. America’s Health Rankings. September 2024. Accessed November 14, 2024. https://www.americashealthrankings.org/explore/measures/MHP

23. SAMHSA grants dashboard. Substance Abuse and Mental Health Services Administration. Accessed July 15, 2024. https://www.samhsa.gov/grants/grants-dashboard

24. Fox J, Weisberg S. Time-series regression and generalized least squares in R. Updated September 26, 2018. Accessed July 9, 2024. https://www.john-fox.ca/Companion/appendices/Appendix-Timeseries-Regression.pdf

25. An introduction to R. The Comprehensive R Archive Network. Updated October 31, 2024. Accessed November 12, 2024. https://cran.r-project.org/doc/manuals/r-release/R-intro.html

26. American Psychiatric Association Presidential Report on the Assessment of Psychiatric Bed Needs in the United States. The psychiatric bed crisis in the United States: understanding the problem and moving toward solutions. Am J Psychiatr. 2022;179(8):586-588. doi:10.1176/appi.ajp.22179004

27. Medicaid Section 1115 substance use disorder (SUD) demonstrations: an in-depth look into pre-demonstration measures of SUD need, treatment use, availability, and outcomes across states. CMS. November 2022. Accessed July 15, 2024. https://www.medicaid.gov/medicaid/section-1115-demonstrations/downloads/sud-115-rcr-baseline.pdf

28. Tormohlen KN, Krawczyk N, Feder KA, Riehm KE, Crum RM, Mojtabai R. Evaluating the role of Section 1115 waivers on Medicaid coverage and utilization of opioid agonist therapy among substance use treatment admissions. Health Serv Res. 2020;55(2):232-238. doi:10.1111/1475-6773.13250

29. Maclean JC, Wen H, Simon KI, Saloner B. Institutions for Mental Diseases Medicaid waivers: impact on payments for substance use treatment facilities. Health Aff (Millwood). 2021;40(2):326-333. doi:10.1377/hlthaff.2020.00404