Skip navigation

Health Care Needs of Adults Involved With the Criminal Justice System 2021

Download original document:
Brief thumbnail
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
(U MACPAC

lssueBrief
August 2021

Advising Congress on Medicaid and CHIP Policy

Access in Brief: Health Care Needs of Adults
Involved with the Criminal Justice System
In 2018, an estimated 6.4 million individuals were under the supervision of the adult correctional system,
including 4.4 million under community supervision (e.g. on parole or probation) and 2.1 million under the
custody of state or federal prisons or local jails (BJS 2020a). 1 The majority of adults under community
supervision have committed non-violent offenses (e.g., crimes related to property, drugs, or traffic offenses
such as driving under the influence) (BJS 2020b).
While Medicaid’s role is limited with respect to those who are incarcerated, it plays an important role in the
treatment of mental illness and substance use disorders (SUD) for adults under community supervision.
Historically, most justice-involved adults were uninsured. But, with the expansion of Medicaid to the new
adult group under the Patient Protection and Affordable Care Act (ACA, P.L. 111-148, as amended), more
individuals involved with the criminal justice system became eligible for Medicaid (MACPAC 2018). From
2015 – 2019, 28 percent of adults under community supervision were enrolled in Medicaid. Even so, about
one quarter (26 percent) of adults under community supervision remain uninsured (MACPAC 2021).
This issue brief uses five years (2015 – 2019) of data from the National Survey on Drug Use and Health
(NSDUH) to analyze the treatment needs and access to behavioral health services for adults age 18–65 who
reported that they were under community supervision in the past 12 months. 2 Specifically, our analysis
examines selected demographic and health characteristics, and prevalence and treatment rates for
behavioral health conditions among these adults, comparing the experience of adults with Medicaid to
adults with other forms of coverage. Where sample size permits, we also report estimates by race, ethnicity,
sex, and sexual orientation.
The vast majority of adults under community supervision (54 percent) were either enrolled in Medicaid or
lacked health insurance. Further, we found:
•

Relative to their privately insured peers, Medicaid beneficiaries under community supervision were more
likely to be Black or Hispanic. They were also more likely to be female.

•

When compared to their peers with other forms of coverage, Medicaid beneficiaries under community
supervision were more likely to have Hepatitis B or C, chronic bronchitis, or asthma.

•

Medicaid beneficiaries under community supervision reported changes in housing more often than their
peers with private coverage and at similar rates to those who are uninsured. Across all racial and ethnic
groups, Medicaid beneficiaries reported moving at similar rates. However, female beneficiaries moved
more frequently than their male peers.

•

With few exceptions, Medicaid beneficiaries under community supervision reported higher rates of
behavioral health conditions than their privately insured or uninsured peers. They also reported receiving
mental health or SUD treatment at higher rates. However, Black beneficiaries with behavioral health
conditions reported receipt of treatment at lower rates than their white peers.

•••
Medicaid and CHIP Payment
and Access Commission

1800 M Street NW
Suite 650 South
Washington, DC 20036

www.macpac.gov
202-350-2000 l
202-273-2452 I&

2

Characteristics of Adults on Parole or Probation
Below we discuss the demographic, physical health, and housing-related characteristics of adults under
community supervision.

Demographic characteristics
From 2015 – 2019, over half of adults under community supervision were either uninsured or covered by
Medicaid (Table 1). Relative to their privately insured peers, Medicaid beneficiaries are more likely to be
Black or Hispanic. However, a higher proportion of adults on parole or probation who lack coverage identify
as Hispanic (28.8 percent) compared to their peers with Medicaid (21.2 percent). Medicaid beneficiaries
were also more likely to be female, compared to their peers with private insurance or those without
coverage. Finally, Medicaid beneficiaries are more than twice as likely to identify as bisexual than their
privately insured peers. Among beneficiaries, a higher percentage (19.6 percent) of female beneficiaries on
parole or probation identify as bisexual than males (2.7 percent) (MACPAC 2021). (Additional demographic
data for adults under community supervision is discussed in Appendix A.)
TABLE 1. Characteristics of Non-Institutionalized Adults Age 18 – 64 Under Community Supervision in the
Past Year, by Insurance Status, 2015 – 2019
Percentage of adults age 18–64 in each
coverage category
Percentage of
adults age 18–
64

Medicaid

Private
coverage

Uninsured

100%

28.1%

34.4%*

25.5%*

18 – 25

24.8

23.8

30.2*

22.7

26 – 34

27.9

33.3

24.9*

33.2

35 – 49

30.4

27.6

30.0

31.9

50 – 64

16.9

15.3

14.9

13.1

Male

70.1

59.4

74.6*

75.7*

Female

29.9

40.6

25.4*

24.3*

Total
Age

Sex

Sexual orientation

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

3
Heterosexual

91.4

88.6

92.7*

91.7*

Lesbian or gay

2.7

1.9

3.1

2.9

Bisexual

5.9

9.5

4.2*

5.4*

White, non-Hispanic

54.3

50.2

61.9*

48.6

Black, non-Hispanic

18.6

21.7

16.1*

17.8*

Hispanic

20.4

21.2

15.5*

28.8*

Asian American, non-Hispanic

1.5

1.1

2.6

0.5

American Indian, Alaskan Native, Native
Hawaiian, or Pacific Islander, non-Hispanic

2.1

2.5

1.6

1.0*

Two or more races, non-Hispanic

3.2

3.3

2.2

3.3

Race and ethnicity

Notes: We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private,
Medicaid, other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
* Difference from Medicaid is statistically significant at the 0.05 level.
– Dash indicates that estimate is based on too small of a sample or is too unstable to present.
Source: MACPAC analysis of the 2015 – 2019 National Survey on Drug Use and Health (NSDUH), 2021.

Chronic physical health conditions
Adults involved in the criminal justice system have higher rates of chronic diseases than the general
population (MACPAC 2018). 3 For many chronic health conditions, Medicaid beneficiaries under community
supervision reported higher rates of such conditions when compared to their peers with private coverage
(Table 2). For example, Medicaid beneficiaries under community supervision were more than eight times as
likely to report that they had Hepatitis B or C in their lifetime. They were also more likely to have chronic
bronchitis or asthma (MACPAC 2021).
TABLE 2. Lifetime Rates of Chronic Conditions among Non-institutionalized Adults Under Community
Supervision, Age 18–64, by Insurance Status, 2015 - 2019
Percentage of adults age 18–64 in each coverage
category
Condition
Ever had any chronic condition

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

Percentage of
adults age 18–64

Medicaid

Private coverage

Uninsured

34.2%

39.7%

30.2%*

29.4%*

4
Ever had a heart condition

5.9

6.0

5.5

Ever had diabetes

7.3

7.0

5.8

Ever had chronic bronchitis

4.8

6.5

3.5*

3.2*

Ever had hepatitis B or C

4.3

8.5

1.0*

4.4*

Ever had asthma

10.1

13.3

8.4*

8.8*

Ever had high blood pressure

9.3

9.9

7.9

6.2*

I

I

4.8

I

6.8

j

Notes: Any chronic condition includes HIV or AIDS, heart conditions, diabetes, chronic bronchitis, cirrhosis of the liver, Hepatitis B or C,
kidney disease, asthma, cancer, high blood pressure, and sexually transmitted diseases. Respondents were asked whether they had
any of the chronic conditions listed in this table over their lifetime (SAMHSA 2019a).
We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private, Medicaid,
other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
* Difference from Medicaid is statistically significant at the 0.05 level.
– Dash indicates that estimate is based on too small of a sample or is too unstable to present.
Source: MACPAC analysis of the 2015 – 2019 National Survey on Drug Use and Health (NSDUH), 2021.

Beneficiaries under community supervision who were white or identified as two or more races experienced
chronic physical health conditions at higher rates than others (Figure 1). However, those identifying as Black,
reported having a sexually transmitted infection within the past year at more than twice the rate (9.9
percent) of their white peers (4.6 percent). 4 Similarly, female beneficiaries under community supervision had
higher rates of sexually transmitted infections in the past year (8.9 percent) than their male counterparts
(4.7 percent).

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

5
FIGURE 1. Reported Lifetime Rates of Chronic Physical Health Conditions among Adults Covered by
Medicaid Under Community Supervision, Age 18 –64, by Race and Ethnicity, 2015 – 2019

Percentage of beneficaries on parole or
probation

70%
59%

60%
50%

46%
37%

40%
30%

28%

29%

Hispanic*

AI/AN/NH/PI*

20%
10%
0%

White

Black*

Two or more races

Race and ethnicity
Notes: Hispanic is anyone of Hispanic, Latino, or Spanish origin. AI/AN/NH/PI combines data for respondents who identified as
American Indian, Alaska Native, Native Hawaiian, or other Pacific Islander and are not of Hispanic origin. White, Black, and two or
more races do not include respondents of Hispanic origin. Due to issues with sample size, we were unable to produce estimates for
lifetime rates of chronic physical health conditions among Asian Americans under community supervision.
We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private, Medicaid,
other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
*Difference from white beneficiaries is statistically significant at the 0.05 level.
Source: MACPAC analysis of the 2015 – 2019 National Survey on Drug Use and Health (NSDUH), 2021.

Changes in housing status
Individuals leaving prison or jail move more frequently than the general public and are almost 10 times more
likely to be homeless (CJCC 2020). According to U.S. Department of Health and Human Services, more than
10 percent of people released from prisons and jail face homelessness upon reentry. In large urban areas,
the share may be as high as 50 percent (HHS 2020).
Access to affordable housing substantially increases the likelihood that a person returning home from
prison or jail will be able to find and retain employment, abstain from drug use, and refrain from committing
additional crimes. Among the general population, poor housing conditions can worsen health outcomes and
make it difficult for individuals to obtain health care and manage complex medical conditions. Moreover,
frequent moves (moving three or more times within a one-year period), put adults at greater risk for suicidal
outcomes compared to those with stable housing (Forman-Hoffman, Glasheen, and Ridenour 2017).

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

6
Medicaid beneficiaries under community supervision reported moving more often than their peers with
private coverage and at similar rates to their uninsured peers (Table 3). From 2015 – 2019, beneficiaries
under community supervision reported that they moved at least once in the past year at higher rates (52.2
percent) compared to their peers with private coverage (38.1 percent). Moreover, they were nearly twice as
likely to report that they moved three or more times in the past year when compared to their privately insured
peers. While there were no differences in rates of moving across racial and ethnic groups, a higher
percentage of female beneficiaries under community supervision moved at least one time (58.4 percent)
compared to their male peers (47.9 percent) (MACPAC 2021).
TABLE 3. Changes in Housing Status among Non-Institutionalized Adults Age 18–64 Under Community
Supervision in the Past Year, by Insurance Status, 2015 – 2019
Percentage of adults age 18–64 in each coverage
category

Percentage
of adults 18 –
64

Medicaid

Private coverage

Uninsured

No moves

54.5%

47.8%

61.9%*

50.1%

One move

25.6

26.8

24.7

27.6

Two moves

11.0

13.3

7.8*

13.2

Three or more moves

8.9

12.1

5.6*

9.1

Moved at least one time

45.5

52.2

38.1*

49.9

Change in housing status

i

Notes: We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private,
Medicaid, other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private, Medicaid,
other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
* Difference from Medicaid is statistically significant at the 0.05 level.
Source: MACPAC analysis of the 2015 – 2019 National Survey on Drug Use and Health (NSDUH), 2021.

Behavioral Health: Prevalence, Treatment Rates, and
Disparities Among Adults on Probation or Parole
For adult respondents, the NSDUH captures prevalence of mental health conditions among adults age 18 to
64 that vary in terms of severity. Prevalence estimates for mental health conditions are reported in four
categories:
•

Any mental illness – This category includes those who currently have or at any time in the past year
reporting having had a diagnosable mental, behavioral, or emotional disorder. Mental illness in this
category can vary in severity. 5

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

7
•

Mild to moderate mental illness – This category includes those with any mental illness except serious
mental illness who currently have or at any time in the past year reporting having had a diagnosable
mental, behavioral, or emotional disorder resulting in less than substantial impairment in carrying out
major life activities. 6

•

Serious mental illness – This category includes those who currently have or at any time in the past year
reported having had a diagnosable mental, behavioral, or emotional disorder resulting in substantial
impairment in carrying out major life activities. 7 Major life activities include activities of daily living, such
as eating or dressing; instrumental activities of daily living, including managing money and taking
prescribed medication; and functioning in social, family, and vocational or education contexts (SAMSHA
2019).

•

Major depressive episode – This category includes adults who reported experiencing certain symptoms
for two weeks or longer in the past 12 months. 8

Below we discuss the prevalence of mental health and SUDs among adults under community supervision.
We also examine the rates at which they receive treatment.

Prevalence of behavioral health conditions
With few exceptions, Medicaid beneficiaries under community supervision report higher rates of behavioral
health conditions than their privately insured peers, or those without insurance (Table 4). In part, this may be
because many individuals qualify for Medicaid based on a disability, including those with serious mental
illness, such as schizophrenia. 9 From 2015 – 2019, nearly 40 percent of Medicaid beneficiaries under
community supervision reported experiencing any mental illness. They are also more likely to experience
mild to moderate mental illness, major depressive episodes, or co-occurring conditions compared to those
with private coverage. Generally, beneficiaries report higher rates of drug dependence or abuse than their
privately insured peers, as well as their peers without insurance. They are also more likely to report
experiencing co-occurring mental health and SUD. Rates of serious mental illness are similar across types of
coverage. (See Appendix B for additional information on prevalence of selected SUDs.)
TABLE 4. Prevalence of Behavioral Health Conditions Among Non-Institutionalized Adults Under Community
Supervision, by Insurance Status, 2015 – 2019
Percentage of adults age 18–64 in each
coverage category
Condition

Percentage of
adults age 18–
64

Medicaid

Private
coverage

Uninsured

34.6%

39.2%

30.8%*

31.5%*

27.2

21.1*

22.1*

12.0

9.7

9.4

Mental health
Any mental illness, past year
Mild to moderate mental illness, past
year
Serious mental illness, past year

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

,-

,-

23.3

..

11.3

..

,-

8
Major depressive episode, past year

13.2

14.9

11.5*

11.2*

Nicotine dependence, past year

26.9

32.3

20.8*

27.9

Illicit drug or alcohol dependence, past
year

23.7

28.3

20.7*

23.1*

Illicit drug dependence, past year

15.3

21.4

11.9*

14.8*

Illicit drug dependence or abuse, past
year

19.5

26.3

15.6*

18.2*

Illicit drug or alcohol dependence or
abuse, past year

33.8

36.3

32.7

32.8

Substance use disorders

Co-occurring mental health and substance use disorders
Any mental illness and illicit drug or
alcohol dependence or abuse, past
year

17.2

20.2

16.0*

15.6*

Serious mental illness and illicit drug
or alcohol dependence or abuse, past
year

6.3

7.4

5.7

6.1

Notes: Estimates of any mental illness, mild to moderate mental illness, and serious mental illness are based on a statistical model of
a clinical diagnosis and responses to questions in the main National Survey on Drug Use and Health (NSDUH) interview on distress,
using the Kessler-6 scale; impairment, which is assessed through an abbreviated version of the World Health Organization Disability
Assessment Schedule; past year major depressive episode; past year suicidal thoughts; and age. Mental illnesses in this category can
vary in severity, ranging from no impairment, to mild or moderate, to severe impairment.
Estimates for dependence or abuse questions for alcohol and illicit drugs were based on criteria in the Diagnostic and Statistical
Manual of Mental Disorders, 4TH edition. Illicit drugs include marijuana, cocaine, heroin, hallucinogens, inhalants, methamphetamine,
and the misuse of prescription psychotherapeutic drugs (i.e., pain relievers, tranquilizers, stimulants, and sedatives). The NSDUH
instrument included items asking about symptoms of dependence or abuse related to the use of a specific substance in the past 12
months.
With the 2015 – 2019 NSDUH survey, a diagnosable mental, behavioral, or emotional disorder, and dependence and abuse are defined
based on the Diagnostic and Statistical Manual of Mental Disorders, 4th edition and excludes developmental and substance use
disorders (2019).
We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private, Medicaid,
other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
* Difference from Medicaid is statistically significant at the 0.05 level.
Source: MACPAC analysis of the 2015 – 2019 National Survey on Drug Use and Health (NSDUH), 2021.

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

9

Prevalence across racial and ethnic groups. The prevalence of behavioral health conditions in this
population varied across racial and ethnic groups (Table 6). Among Medicaid beneficiaries, reported rates of
any mental illness are highest for those who identify as either white or two or more races. White
beneficiaries under community supervision also had significantly higher rates of nicotine dependence than
their peers. Those identifying as American Indian, Alaska Native, Native Hawaiian or Pacific Islander
reported significantly higher rates of illicit drug or alcohol dependence than their white peers. While Black
beneficiaries report significantly lower rates of illicit drug or alcohol dependence than their white
counterparts.
TABLE 6. Prevalence of Behavioral Health Conditions Among Non-Institutionalized Beneficiaries Under
Community Supervision, by Race and Ethnicity, 2015 – 2019
Percentage of Medicaid beneficiaries age 18–64 in each racial and
ethnic group
White

Black

Hispanic

AI/AN/NA/PI

Two or more
races

47.8%

25.4%*

35.7%*

23.5%*

37.0%

Mild to moderate mental illness, past
year

31.6

21.2*

26.0

22.9%

16.2*

Serious mental illness, past year

16.2

4.2*

9.8*

–

–

19.3

6.9*

12.2*

–

–

Nicotine dependence, past year

45.5

18.8*

16.3*

29.0*

29.8

Illicit drug or alcohol dependence, past
year

30.9

23.2*

28.0

50.5*

–

Illicit drug dependence, past year

23.8

14.9*

23.1

32.6

–

Illicit drug dependence or abuse, past
year

28.8

18.6*

29.2

34.5

–

Illicit drug or alcohol dependence or
abuse, past year

37.5

31.6

38.3

53.4

–

Mental health
Any mental illness, past year

Major depressive episode, past year

I

I

Substance dependence or abuse

Co-occurring mental health and substance use disorders

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

10
Any mental illness and illicit drug or
alcohol dependence or abuse, past
year

26.0

11.1*

17.9*

–

–

Notes: Hispanic is anyone of Hispanic, Latino, or Spanish origin. AI/AN/NH/PI combines data for respondents who identified as
American Indian, Alaska Native, Native Hawaiian, or other Pacific Islander and are not of Hispanic origin. White, Black, and two or
more races do not include respondents of Hispanic origin. Due to issues with sample size, we were unable to produce estimates for
behavioral health conditions among Asian Americans under community supervision.
Estimates of for any mental illness, mild to moderate mental illness, and serious mental illness are based on a statistical model of a
clinical diagnosis and responses to questions in the main National Survey on Drug Use and Health (NSDUH) interview on distress,
using the Kessler-6 scale; impairment, which is assessed through an abbreviated version of the World Health Organization Disability
Assessment Schedule; past year major depressive episode; past year suicidal thoughts; and age. Mental illnesses in this category can
vary in severity, ranging from no impairment, to mild or moderate, to severe impairment.
Estimates for dependence or abuse questions for alcohol and illicit drugs were based on criteria in the Diagnostic and Statistical
Manual of Mental Disorders, 4TH edition. Illicit drugs include marijuana, cocaine, heroin, hallucinogens, inhalants, methamphetamine,
and the misuse of prescription psychotherapeutic drugs (i.e., pain relievers, tranquilizers, stimulants, and sedatives). The NSDUH
instrument included items asking about symptoms of dependence or abuse related to the use of a specific substance in the past 12
months.
With the 2015 – 2019 NSDUH survey, a diagnosable mental, behavioral, or emotional disorder, and dependence and abuse are defined
based on the Diagnostic and Statistical Manual of Mental Disorders, 4th edition and excludes developmental and substance use
disorders (2019).
We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private, Medicaid,
other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
*Difference from Medicaid is statistically significant at the 0.05 level.

– Dash indicates that estimate is based on too small of a sample or is too unstable to present.
Source: MACPAC analysis of the 2015 – 2019 National Survey on Drug Use and Health (NSDUH), 2021.

In interpreting these data, it is important to consider differences in how people of color report and
experience mental health conditions. For several decades, research has demonstrated that Black Americans
often have higher rates of psychological distress than white Americans. Moreover, when Black and Hispanic
people experience mental illness, their episodes tend to be more severe, result in higher levels of
impairment, persist for longer periods of time, and be more debilitating than for any other racial or ethnic
group (Williams 2019).

Thoughts and plans of suicide. The rate of suicide among adults involved in the criminal justice system is
significantly higher than the general population. From 2015 – 2019, roughly 10 percent of adults on parole or
probation reported that they seriously thought about committing suicide in the past year. Among those that
seriously thought about suicide, 43.0 percent made plans to commit suicide, and 24.4 percent attempted
suicide. Reported rates of past year suicidal ideation and suicide attempts were generally similar across
coverage groups (MACPAC 2021).
These rates varied across racial and ethnic groups. White beneficiaries reported serious thoughts of suicide
at more than twice the rate (11.7 percent) of their Black peers (5.2 percent). However, Black beneficiaries
who had thoughts of suicide were more likely to make plans of suicide (49.7 percent), or report attempting
suicide (48.2 percent) compared to white peers. 10 Similarly, Hispanic beneficiaries reported rates of suicidal

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

11
ideation at similar rates to their white peers (10.1 percent); however, they were significantly more likely to
make plans of suicide (66.9 percent), or report attempting suicide (51.7 percent) (MACPAC 2021).

Access to behavioral health treatment
Judges and parole boards often require that individuals participate in mental health or SUD treatment as a
condition of probation or release (Skeem and Louden 2006). Treatment could occur as a condition of
incarceration, probation, or pretrial release, or be required in-lieu of incarceration (NIDA 2018).
Generally, Medicaid beneficiaries under community supervision with behavioral health conditions received
treatment at higher rates than their privately insured and uninsured peers. However, treatment rates among
beneficiaries under community supervision varied significantly based on race and ethnicity (MACPAC 2021).

Mental health treatment. Medicaid beneficiaries under community supervision reported both experiencing
behavioral health conditions and receiving treatment at higher rates than their peers with private insurance
or those without insurance. From 2015 – 2019, 57.8 percent of beneficiaries under community supervision
with any mental illness reported receiving mental health treatment, compared to 42.4 percent of their peers
with private coverage, and 35.2 percent of those without insurance (Table 7). They most frequently reported
taking medication for their mental health condition (49.1 percent) and receiving some form of outpatient
treatment (37.3 percent). However, nearly one third (31.0 percent) of those with mental illness reported that
they needed mental health treatment or counseling but did not receive it. These rates did not differ by
coverage type. Female beneficiaries with a mental health condition who were under community supervision
reported higher rates (41.0 percent) of unmet need than their male counterparts (21.3 percent) (MACPAC
2021).
TABLE 7. Mental Health Treatment among Non-institutionalized Adults Age 18–64 Under Community
Supervision with Any Mental Illness, by Insurance Status, 2018
Percentage of adults age 18–64 in each
coverage category
Percentage
of adults age
18 – 64

Medicaid

Private
coverage

Uninsured

48.9%

57.8%

42.4%*

35.2%*

Needed mental health treatment or counseling but
did not received it

27.0

31.0

24.8

24.1

Received treatment in an inpatient hospital

9.1

11.0

5.5*

9.3

Received any outpatient treatment

31.1

37.3

25.6*

21.7*

22.4

9.4*

11.6*

9.3

12.8

6.1

Treatment setting
Received any mental health treatment

Received treatment in an outpatient mental health
center or day treatment program
Received treatment in a private therapist’s office

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

I

15.4
10.7

I

12
Took any prescription medication for a mental
health condition

I

39.2

I I
49.1

33.4*

22.4*

Notes: Inpatient treatment settings for mental health include a public or private psychiatric hospital, a psychiatric unit or medical unit
of an acute care hospital, a residential treatment facility, or some other inpatient setting. Estimates of for any mental illness are based
on a statistical model of a clinical diagnosis and responses to questions in the main National Survey on Drug Use and Health
(NSDUH) interview on distress, using the Kessler-6 scale; impairment, which is assessed through an abbreviated version of the World
Health Organization Disability Assessment Schedule; past year major depressive episode; past year suicidal thoughts; and age.
Mental illnesses in this category can vary in severity, ranging from no impairment, to mild or moderate, to severe impairment.
With the 2015 – 2019 NSDUH survey, a diagnosable mental, behavioral, or emotional disorder, and dependence and abuse are defined
based on the Diagnostic and Statistical Manual of Mental Disorders, 4th edition and excludes developmental and substance use
disorders (2019).
We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private, Medicaid,
other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
*Difference from Medicaid is statistically significant at the 0.05 level.
Source: MACPAC analysis of the 2015 – 2019 National Survey on Drug Use and Health (NSDUH), 2021.

As with prevalence, treatment rates varied across racial and ethnic groups (Table 8). Beneficiaries with
mental illness who identify as Hispanic or as two or more races, reported receipt of mental health treatment
at similar rates as their white peers. However, Black beneficiaries with mental illness under community
supervision received mental health treatment at significantly lower rates than their white counterparts
(MACPAC 2021).
TABLE 8. Reported Use of Mental Health Treatment among Non-Institutionalized Adult Medicaid
Beneficiaries Age 18 – 64 Under Community Supervision with Past Year Mental Illness, by Racial and Ethnic
Group, 2015 - 2019
Percentage of Medicaid beneficiaries age 18–64 in each
racial and ethnic group

Treatment characteristics
Needed but did not receive mental health
treatment, past year

White

Black

Hispanic

Two or more
races

33.0%

24.0%

27.6%

45.1%

41.8*

57.0

44.5

32.0*

51.5

33.9

Received any mental health treatment, past year

62.4

Took any prescription medication for a mental
health condition, past year

53.1

j

I

Notes: Estimates of for any mental illness are based on a statistical model of a clinical diagnosis and responses to questions in the
main National Survey on Drug Use and Health (NSDUH) interview on distress, using the Kessler-6 scale; impairment, which is
assessed through an abbreviated version of the World Health Organization Disability Assessment Schedule; past year major
depressive episode; past year suicidal thoughts; and age. Mental illnesses in this category can vary in severity, ranging from no
impairment, to mild or moderate, to severe impairment.

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

13
With the 2015 – 2019 NSDUH survey, a diagnosable mental, behavioral, or emotional disorder, and dependence and abuse are defined
based on the Diagnostic and Statistical Manual of Mental Disorders, 4th edition and excludes developmental and substance use
disorders (2019).
We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private, Medicaid,
other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
*Difference from Medicaid is statistically significant at the 0.05 level.
Source: MACPAC analysis of the 2015 – 2019 National Survey on Drug Use and Health, 2021.

Substance use treatment. Beneficiaries under community supervision with SUD were more likely to engage
in treatment compared to those with private insurance, or adults who are uninsured (Table 9). Specifically,
they were nearly twice as likely as privately insured or uninsured peers to report receiving such treatment in
the past year. They were also more likely to participate in support groups, such as Alcoholics Anonymous,
and receive treatment in jail or prison. However, male beneficiaries with SUD who were under community
supervision reported receiving SUD treatment in their lifetime at significantly higher rates (73.0 percent) than
their female counterparts (62.0 percent) (MACPAC 2021).
Table 9. Substance Use Treatment among Non-institutionalized Adults Age 18–64 Under Community
Supervision with Past Year Substance Use Disorder, by Insurance Status, 2015 – 2019
Percentage of adults age 18–64 under
community supervision
Percentage of
adults age 18–64

Medicaid

Private
coverage

Uninsured

21.7%

31.1%

17.8%*

17.3%*

Received any substance use treatment,
past year

38.7

49.6

33.9*

34.7*

Ever received alcohol or drug treatment

58.0

68.7

53.6*

50.5*

Perceived the need for treatment or
counseling for alcohol or drug use

71.4

61.2

75.0*

76.9*

Received treatment in a hospital overnight
as an inpatient

8.0

10.7

8.0

-

Received treatment in a residential drug or
alcohol rehabilitation facility

11.5

17.7

9.7*

7.6*

Received treatment in a drug or alcohol
rehabilitation facility as an outpatient

21.4

29.9

17.2*

18.4*

Condition
Currently receiving treatment or counseling

During previous 12 months

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

14
Received treatment in a mental health
center or facility as an outpatient

13.4

19.1

12.5*

9.1*

Received treatment in prison or jail

8.3

12.0

5.9*

8.7

Participated in a mutual aid group such as
Alcoholics Anonymous or Narcotics
Anonymous

23.1

31.0

21.1*

18.6*

Received treatment in some other place

13.5

17.3

9.8*

12.6

Notes: Estimates for dependence or abuse questions for alcohol and illicit drugs were based on criteria in the Diagnostic and
Statistical Manual of Mental Disorders, 4TH edition. Illicit drugs include marijuana, cocaine, heroin, hallucinogens, inhalants,
methamphetamine, and the misuse of prescription psychotherapeutic drugs (i.e., pain relievers, tranquilizers, stimulants, and
sedatives). The National Survey on Drug Use and Health (NSDUH) instrument included items asking about symptoms of dependence
or abuse related to the use of a specific substance in the past 12 months.
With the 2015 – 2019 NSDUH survey, a diagnosable mental, behavioral, or emotional disorder, and dependence and abuse are defined
based on the Diagnostic and Statistical Manual of Mental Disorders, 4th edition and excludes developmental and substance use
disorders (2019).
We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private, Medicaid,
other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
*Difference from Medicaid is statistically significant at the 0.05 level.

– Dash indicates that estimate is based on too small of a sample or is too unstable to present.
Source: MACPAC analysis of the 2015 – 2019 National Survey on Drug Use and Health (NSDUH), 2021.

Except for Black beneficiaries, Medicaid beneficiaries under community supervision with past year SUD
received treatment at similar rates as their white peers (Table 10). White beneficiaries under community
supervision with past year SUD were more than twice as likely to report that they were currently receiving
SUD when compared to their Black peers. Moreover, Black beneficiaries under community supervision with
SUD were less likely to receive treatment for their SUD in the past year, or at any point in their lifetime
compared to their white counterparts (MACPAC 2021).
TABLE 10. Reported Use of Substance Use Treatment among Non-Institutionalized Adult Medicaid
Beneficiaries Age 18 – 64 Under Community Supervision with Past Year Substance Use Disorder, by Racial
and Ethnic Group, 2015 - 2019
Percentage of Medicaid beneficiaries age 18– 64 in each
racial and ethnic group

Treatment characteristics
Perceived the need for treatment or counseling
for alcohol or drug use, past year

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

l

White

Black

Hispanic

53.6%

71.6%*

66.9%

AI/AN/NH
/PI

l

67.8

Two or
more
races
61.9

15
Received any substance use disorder treatment,
past year

56.0

36.3*

47.7

54.7

44.8

Ever received substance use disorder treatment

78.2

44.0*

68.3

72.2

53.2

Currently receiving substance use disorder
treatment

37.4

16.5*

–

49.0

–

Notes: Hispanic is anyone of Hispanic, Latino, or Spanish origin. AI/AN/NH/PI combines data for respondents who identified as
American Indian, Alaska Native, Native Hawaiian, or other Pacific Islander and are not of Hispanic origin. White, Black, and two or
more races do not include respondents of Hispanic origin. Due to issues with sample size, we were unable to produce estimates for
behavioral health treatment among Asian Americans under community supervision.
Estimates for dependence or abuse questions for alcohol and illicit drugs were based on criteria in the Diagnostic and Statistical
Manual of Mental Disorders, 4TH edition. Illicit drugs include marijuana, cocaine, heroin, hallucinogens, inhalants, methamphetamine,
and the misuse of prescription psychotherapeutic drugs (i.e., pain relievers, tranquilizers, stimulants, and sedatives). The National
Survey on Drug Use and Health (NSDUH) instrument included items asking about symptoms of dependence or abuse related to the
use of a specific substance in the past 12 months.
With the 2015 – 2019 NSDUH survey, a diagnosable mental, behavioral, or emotional disorder, and dependence and abuse are defined
based on the Diagnostic and Statistical Manual of Mental Disorders, 4th edition and excludes developmental and substance use
disorders (2019).
We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private, Medicaid,
other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
*Difference from Medicaid is statistically significant at the 0.05 level.

–Dash indicates that estimate is based on too small of a sample or is too unstable to present.
Source: MACPAC analysis of the 2015 – 2019 National Survey on Drug Use and Health (NSDUH), 2021.

Data and Methods
Data sources
Data for this report comes from the 2015–2019 NSDUH, an annual survey sponsored by the Substance
Abuse and Mental Health Services Administration, that conducts interviews with approximately 70,000
randomly selected, civilian, non-institutionalized individuals age 12 and older in the United States. NSDUH
respondents are residents of households and individuals in non-institutional group quarters, such as
shelters, rooming houses, college dorms, and halfway houses. Individuals with no fixed household address,
such as individuals who are homeless and not in shelters; active-duty military personnel; and residents of
institutional group quarters, including congregate settings for youth in foster care, correctional facilities,
nursing homes, and mental institutions, are excluded. The NSUDH is a primary source of national and statelevel estimates on use of tobacco products, alcohol, illicit drugs, SUDs, mental health status, and related
treatment (SAMHSA 2020b).

Insurance coverage
The following hierarchy was used to assign individuals with multiple coverage sources to a primary source:
Medicare; private; Medicaid/CHIP; other type of insurance (e.g., TRICARE, military health care); or uninsured.

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

16
Coverage source is defined as primary coverage at the time of the interview. Private health insurance
coverage excludes plans that paid for only one type of service, such as accidents or dental care.
Point estimates were calculated using sample weights, and corresponding variances accounted for the
complex sample design of NSUDH. All estimates in this brief have a relative standard error of less than or
equal to 30 percent. All differences discussed were computed using t-tests and are significant at the 0.05
level.
Endnotes

Individuals on parole include people released through discretionary or mandatory supervised release from prison. In
comparison, probation is a court-ordered period of correctional supervision in the community, typically viewed as an
alternative to incarceration.

1

NSDUH respondents are residents of households and non-institutionalized group quarters (e.g., shelters, rooming houses,
dormitories) and civilians living on military bases age 12 and older. The survey excludes individuals experiencing
homelessness who are not residing in shelters; military personnel on active duty; and residents of institutional group quarters,
including jails, nursing homes, mental institutions, and long-term care hospitals.

2

Adults involved with the criminal justice system have a higher prevalence of HIV/AIDS, tuberculosis, and sexually transmitted
diseases than the general population (NCCHC 2002). They also have higher rates of chronic conditions such as asthma,
diabetes, and hypertension, as well as behavioral health conditions.

3

In part, this may reflect differences in age between Black and white beneficiaries on parole or probation. Roughly 30 percent
of Black beneficiaries on parole or probation are 18 – 25 years in age, compared to 18.8 percent of white beneficiaries.

4

Estimates for any mental illness are based on a statistical model of a clinical diagnosis and responses to questions in the
main NSDUH interview on: distress, using the Kessler-6 scale; impairment, which is assessed through an abbreviated version
of the World Health Disability Assessment Schedule; past year major depressive episode; past year suicidal thoughts; and age
(SAMHSA 2019).

5

Estimates for mild to moderate mental illness are based on a statistical model of a clinical diagnosis and responses to
questions in the main NSDUH interview on: distress, using the Kessler-6 scale; impairment, which is assessed through an
abbreviated version of the World Health Disability Assessment Schedule; past year major depressive episode; past year
suicidal thoughts; and age. Less than substantial impairment is defined based on clinical interview Global Assessment of
Functioning scores of 50 or less (SAMHSA 2019).

6

Estimates for serious mental illness are based on a statistical model of clinical diagnosis and responses to questions in the
main NSDUH interview on: distress, using the Kessler-6 scale; impairment, which is assessed through an abbreviated version
of the World Health Organization Disability Assessment Schedule; past year major depressive episode; past year suicidal
thoughts; and age. Within the 2019 NSDUH survey, a diagnosable mental, behavioral, or emotional disorder is defined based
on the Diagnostic and Statistical Manual of Mental Disorders, 4th edition and excludes developmental and substance use
disorders. Substantial impairment is defined based on clinical interview Global Assessment Functioning scores of 50 or less
(SAMHSA 2019).
7

The 2019 NSDUH defined individuals as having an MDE if they reported at least five or more of the following symptoms in the
same two-week period in the past year (with at least one of the symptoms being a depressed mood or loss of interest or
pleasure in daily activities): (1) depressed mood most of the day, nearly every day; (2) markedly diminished interest or pleasure
in all or almost activities most of the day; (3) significant weight loss when not dieting or weight gain or decrease or increase in
appetite nearly every day; (4) insomnia or hypersomnia nearly every day; (5) psychomotor agitation or retardation at a level
that is observable by others nearly every day; (6) fatigue or loss of energy nearly every day; (7) feelings of worthlessness or
excessive or inappropriate guilt nearly every day; (8) diminished ability to think or concentrate or indecisiveness nearly every
day; and (9) recurrent thoughts of death or recurrent suicidal ideation (SAMHSA 2019).

8

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

17

In 2019, among those qualifying for Supplemental Security Income, 6 out of 10 were diagnosed with a mental disorder (SSA
2020).

9

Among white beneficiaries on parole or probation who had seriously thought about suicide, 38 percent made plans to
commit suicide, and 19 percent attempted suicide.

10

References
Criminal Justice Coordinating Council for the District of Columbia (CJCC). 2020. Housing for criminal justice involved
individuals in the District of Columbia: Research brief. Washington, DC: CJCC.
https://cjcc.dc.gov/sites/default/files/dc/sites/cjcc/Housing%20for%20criminal%20justice%202020.pdf.
Forman-Hoffman, V., C. Glasheen, and T.A. Ridenour. 2018. Residential transience and substance use disorder are
independently associated with suicidal thoughts, plans, and attempts in a nationally representative sample of U.S. adults.
Suicide and life-threatening behavior, 48(4): 401-412. American Association of Suicidology.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5701885/.
Medicaid and CHIP Payment and Access Commission (MACPAC). 2021. State Health Access Data Assistance Center
(SHADAC), University of Minnesota, analysis for MACPAC of the 2015-2019 National Survey on Drug Use and Health (NSDUH).
Washington, DC: MACPAC.
Medicaid and CHIP Payment and Access Commission (MACPAC). 2018. Issue brief: Medicaid and the criminal justice system.
Washington, DC: MACPAC. https://www.macpac.gov/wp-content/uploads/2018/07/Medicaid-and-the-Criminal-JusticeSystem.pdf.
National Institute on Drug Abuse (NIDA). 2018. What role can the criminal justice system play in addressing drug addiction?
https://www.drugabuse.gov/publications/principles-drug-addiction-treatment-research-based-guide-third-edition/frequentlyasked-questions/what-role-can-criminal-justice-system-play-in-addressing.
Social Security Administration (SSA). 2020. SSI annual statistical report, 2019. SSA publication no. 13-11827. Washington, DC:
SSA. https://www.ssa.gov/policy/docs/statcomps/ssi_asr/2019/ssi_asr19.pdf.
Skeem, J.L., and J.E. Louden. Toward an evidenced-based practice for probationers and parolees mandated to mental health
treatment. Psychiatric Services, 57(3): 334-342. https://ps.psychiatryonline.org/doi/full/10.1176/appi.ps.57.3.333.
Substance Abuse and Mental Health Services Administration (SAMHSA), U.S. Department of Health and Human Services.
2020. 2019 National Survey on Drug Use and Health (NSDUH): Methodological summary and definitions. Rockville, MD:
SAMHSA. https://www.samhsa.gov/data/report/2019-methodological-summary-and-definitions.
Williams, D.R. 2018. Stress and the mental health of populations of color: advancing our understanding of race-related
stressors. Journal of Health and Social Behavior, 59(4): 466-485.
https://journals.sagepub.com/doi/10.1177/0022146518814251?url_ver=Z39.882003&rfr_id=ori%3Arid%3Acrossref.org&rfr_dat=cr_pub%3Dpubmed&.
U.S. Department of Health and Human Services (HHS). 2020. Report to Congress: Innovative state initiatives and strategies for
providing housing-related services and supports under a state Medicaid program to individuals with substance use disorders
who are experiencing or at risk of experiencing homelessness. Washington, DC: HHS.
https://www.medicaid.gov/medicaid/benefits/downloads/rtc111320-1017.pdf.

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

18

Appendix A: Education and Employment by
Insurance Status
TABLE A-1. Education and Employment among Non-institutionalized Adults Age 18–64 Under Community
Supervision, by Insurance Status, 2015 – 2019
Percentage of adults age 18–64 in each
coverage category
Percentage of
adults age 18–
64

Medicaid

Private
coverage

Uninsured

100%

28.1%

34.4%*

25.5%*

Less than high school

23.1

28.1

13.7*

31.7

High school graduate

37.0

38.8

34.3

38.6

Some college or associate degree

32.5

29.8

38.4*

25.5

College graduate

7.4

3.3

13.5*

4.2

Working full time

49.2

30.7

70.0*

51.0*

Working part time

11.2

11.9

9.9

11.0

Unemployed

13.1

19.1

7.5*

15.6

12.6*

22.4*

Total
Education

Employment

Other

I

26.5

I

38.3

I

Notes: We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private,
Medicaid, other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
* Difference from Medicaid is statistically significant at the 0.05 level.
Source: MACPAC analysis of the 2015 – 2019 National Survey on Drug Use and Health (NSDUH), 2021.

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

19

Appendix B: Substance Misuse, Abuse, and
Dependence by Insurance Status
TABLE B-1. Substance Misuse, Abuse, and Dependence among Non-institutionalized Adults Age 18–64
Under Community Supervision, by Insurance Status, 2015 – 2019
Percentage of adults age 18–64 in each
coverage category
Percentage of
adults 18–64

Medicaid

Private
coverage

Uninsured

21.1%

17.8%

23.4%*

21.3%

7.5

8.8

8.2

5.1*

Used oxycontin not directed by a doctor,
past year

3.6

4.7

3.2

3.6

Heroin use, past year

3.8

6.0

2.6*

3.7*

Opioid dependence or abuse, past year

7.1

12.0

3.0*

6.6*

Pain reliever dependence or abuse, past year

5.0

8.3

2.4*

4.7*

Heroin dependence or abuse, past year

3.6

7.3

1.3*

3.2*

Cocaine use, past year

9.1

10.0

8.9

9.2

Cocaine dependence or abuse, past year

3.4

5.3

1.9*

3.4

8.4

12.5

5.1*

9.1*

Substance use
Alcohol
Alcohol dependence or abuse, past year
Marijuana
Marijuana dependence or abuse, past year
Opioids

I

Cocaine

Methamphetamines
Methamphetamine use, past year

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov

20
Methamphetamine dependence or abuse,
past year

6.3

9.5

3.9*

6.7*

Psychotherapeutic misuse, past year

22.0

25.0

20.4*

21.8

Psychotherapeutic dependence or abuse,
past year

6.6

9.7

4.7*

5.9*

Psychotherapeutics

Notes: Estimates for dependence or abuse questions for alcohol and illicit drugs were based on criteria in the Diagnostic and
Statistical Manual of Mental Disorders, 4TH edition. Illicit drugs include marijuana, cocaine, heroin, hallucinogens, inhalants,
methamphetamine, and the misuse of prescription psychotherapeutic drugs (i.e., pain relievers, tranquilizers, stimulants, and
sedatives). The National Survey on Drug Use and Health (NSDUH) instrument included items asking about symptoms of dependence
or abuse related to the use of a specific substance in the past 12 months.
With the 2015 – 2019 NSDUH survey, a diagnosable mental, behavioral, or emotional disorder, and dependence and abuse are defined
based on the Diagnostic and Statistical Manual of Mental Disorders, 4th edition and excludes developmental and substance use
disorders (2019).
We used the following hierarchy to assign individuals with multiple coverage sources to a primary source: Medicare, private, Medicaid,
other, or uninsured. Coverage source is defined as of the time of the most recent survey interview.
* Difference from Medicaid is statistically significant at the 0.05 level.
Source: MACPAC analysis of the 2015 – 2019 National Survey on Drug Use and Health (NSDUH), 2021.

•••
Medicaid and CHIP Payment
and Access Commission
www.macpac.gov