reading

60 contexts.org

t trends

Mental health advocacy organizations

and public health offi cials have raised

concerns over calling 911 in a mental

health crisis. Police interactions can turn

confrontational, and individuals may be

transported to emergency inpatient treat-

ment––or even jailed––when other treat-

ment referrals may be more appropriate.

Mental health-related calls account for up

to 20 percent of 911 calls. From 2015 to

2016, 25 percent of fatal police shootings

involved individuals with mental illness.

In 2018, Contexts highlighted that, of

these, approximately 81 percent were

initiated by a 911 call, most commonly

made by a family member or friend.

As communities grapple with how

to safely and effectively respond to men-

tal health crises in the context of calls to

defund police and shift funds toward

social services, alternative models of

response could become more available.

For example, by July 2022, a national

number (988) will be launched, con-

necting callers to Lifeline crisis centers

to deliver support by telephone, assess

for additional needs, and coordinate

additional support services if needed.

However, we know little about potential

differences in demand for mental illness

crisis lines versus calling 911, particularly

with regard to race or trust in police.

In late 2020, we surveyed a diverse

sample of people actively involved in

their communities to understand prefer-

ences for different support types during

a mental health crisis––1,609 individuals

from Christian, Jewish, and Hindu con-

gregations in Washington DC, Maryland,

Virginia, and Texas.

Religious communities provide cru-

cial civic spaces and advice, especially in

times of crisis, while religious leaders play

an important role in providing mental

health support and often act as gate-

keepers to accessing professional mental

health services. Furthermore, particularly

in Black communities, pastors are often

viewed as the most important community

leaders.

Our sample is similar to the gen-

eral U.S. population in terms of race/

ethnicity. However, on average, the

sample is slightly older, is predominantly

women, and has higher education levels

and income than the general U.S. popu-

lation. Just over half the sample identifi es

as Democrat. Therefore, while our sample

is not representative of the general U.S.

population in terms of gender or socio-

economic status, it is similar in racial and

ethnic composition, and provides insight

into how actively involved community

members would perceive the police and

navigate mental health crises.

We investigate the links between

race, trust in police, and preferences for

mental illness crisis resources in this sam-

ple. In our survey, we asked, “If a loved

one was experiencing a mental illness and

race, trust in police, and mental health crisis support by emma frankham, christopher jacobi, and brandon vaidyanathan

B Y-

SA 2

.0 , cc

A memorial vigil for Decynthia Clements—a Black woman killed by an Elgin police offi cer—held in April 2018 in Chicago, Illinois.

10

20

30

40

50

60

70

80

90

100 percent

Across all respondents Of those who would call a mental illness crisis line

Of those who would call police

Of those who would call 911

Call preferences in a mental health crisis

Source: Mental Health in Congregations Study (2020)

I would call 911 I would call a mental illness crisis line I would call the police

61S U M M E R 2 0 2 1 c o n t e x t sContexts, Vol. 20, Issue 3, p. 60-62. ISSN 1536-5042. © American Sociological Association. http://contexts.sagepub.com. 10.1177/15365042211035343.

was threatening to harm themselves or

others, what would you do?” Options

included “I would call 911,” “I would

call the police,” and “I would call a men-

tal illness crisis line.” Respondents were

able to select all that applied––in reality,

people may choose several courses of

action, depending on the urgency of the

situation, so we wanted to capture the

range of options people felt were viable.

The first figure captures the complexity

of respondent’s ability to select all that

applied. We included “mental illness crisis

line” to understand whether respondents

would choose an alternative to the police.

As the National Alliance on Mental Illness

observes, “often crisis situations can be

resolved over the phone, dramatically

reducing the need for law enforcement

intervention.”

We included both calling police and

calling 911 to understand whether there

is a difference in preferences between

calling 911 generally, or specifically call-

ing to request police. Police are typically

first responders to mental health crises

when a 911 call is made, whether or not

the caller requests a medical response.

Examining preferences for both allows

us to identify whether respondents

understand this. We found 65 percent

of respondents would call 911, 19 per-

cent would call the police, and 62 percent

would call a crisis line (see the left side

of the first figure). Of those who would

call police, 88 percent would also call

911. However, the reverse is not true––of

those who would call 911, only 26 per-

cent would also call police––suggesting a

majority of those who would call 911 are

not seeking a police response.

We also asked respondents, “In

general, to what extent do you trust the

following people? (Police officers).” We

created this survey question to directly

assess respondents’ trust in police. Trust

in police was generally high: 11 percent

of respondents trust police “completely,”

64 percent trust police “quite a lot,” 24

percent “a little,” and only 1 percent

“not at all.” In line with national public

opinion polls, we find diverging views

by race (Figure on the top left)––White

respondents trust police “completely”

or “quite a lot” (82 percent combined)

more than other respondents, with Black

or African American respondents trusting

police the least (40 percent combined).

We also examined how trust in

police is associated with respondents’ call

preferences if a loved one was experienc-

ing a mental health crisis. We find lower

levels of trust in police are associated with

fewer respondents willing to call 911

or police––reinforcing that respondents

understand calling 911 leads to a police

response, which some don’t want.

Figure on the bottom left shows

that among respondents who trust police

“completely,” 70 percent would call 911

versus 62 percent of respondents who

trust police “a little.” Of respondents

who trust police “completely,” 21 per-

cent would call the police versus only

13 percent who trust police “a little.”

We don’t examine call preferences of

respondents who trust police “not at all”

because there are too few responses to

reliably provide insight.

Conversely, lower levels of trust in

police are associated with respondents

preferring a mental illness crisis line.

The figure on the top left shows that

among respondents who trust police

“completely,” 50 percent would call a

percent

Trust in the police by race or ethnicity

Source: Mental Health in Congregations Study (2020)

Not at all A little Quite a lot Completely

10 20 30 40 50 60 70 80 90 100

White

Black or African American

Hispanic or Latino

Other

10

20

30

40

50

60

70

80

90

100 percent

I would call 911 I would call a mental illness crisis line

I would call the police

Trust in police and call preferences in a mental health crisis

Source: Mental Health in Congregations Study (2020)

A little Quite a lot Completely

Lower levels of trust in police are associated with fewer respondents willing to call 911 or police

62 contexts.org

mental illness crisis line versus 67 percent

of respondents who trust police “a little.”

Finally, we jointly examined trust in

police and race––is trust in police a proxy

for underlying racial or ethnic differences?

There are too few respondents who trust

police “completely” to include an analysis

of their call preferences by race, so we

examined preferences of those who trust

police “a little” or “quite a lot.”

As shown in the figure above, there

is a positive association between trust in

police and preferences to call police. For

example, while only 15 percent of Afri-

can Americans who trust police “a little”

would call police, 30 percent of African

Americans who trust police “quite a lot”

would do the same. However, there are

important differences by race and eth-

nicity: when taking into account trust in

police, Whites are less likely than African

Americans or Hispanics to want to call

police. For example, only 8 percent of

Whites who trust police “a little” and 20

percent of Whites who trust police “quite

a lot” would call police.

The inverse is true for preferences

to call a mental illness crisis line: higher

levels of police trust are associated with a

lower likelihood that a respondent would

call a mental illness crisis line. However, as

with preferences to call police, there are

differences by race and ethnicity––when

taking into account trust in police, Afri-

can Americans are less likely than Whites

or Hispanics to want to call a mental

illness crisis line.

How policy makers fund, support,

and publicize mental illness crisis lines for

individuals seeking alternatives to a police

response is critical. Communities of color,

in particular, experience a lack of access

to good quality mental health care which

can lead to a cycle of “more severe symp-

toms, greater criminal involvement, and

more frequent arrest.” African Americans

disproportionately face barriers to mental

health care, including a lack of insurance

and not receiving appropriate informa-

tion about services, as well as deterrents

including a lack of culturally competent

providers, and being less likely than

Whites to receive guideline-consistent

care. As a result, African Americans are

less likely than Whites to access mental

health specialists. Only one in three Afri-

can American adults who needs mental

health care receives it, and––as our data

show––African Americans are less likely

than Whites or Hispanics to want to use

a mental illness crisis line.

Therefore, if resources intended to

divert people in crisis away from a police

response will be successful, work needs

to be done to raise awareness of alterna-

tives and encourage conversations about

mental health. As the National Alliance

on Mental Illness argues, “we need bet-

ter access to care and crisis services for

Black people with mental illness. One

way we do that is by giving communities

another option to call when a neighbor or

loved one is experiencing a mental health

crisis… When 988 finally goes live, let’s

all be mobilized and ready to greet it.”

This study was funded by the John

Templeton Foundation (#61107). A full

list of recommended readings can be

found in the online at www.contexts.org.

Emma Frankham is a writer and researcher who

received her PhD from the University of Wisconsin–

Madison. She publishes on mental illness, incarcera-

tion, and policing. Christopher Jacobi is a Research

Scholar in the Sociology Department at The Catholic

University of America and a DPhil student at Nuffield

College, University of Oxford. He is interested in mental

health research, organizational research on the Catho-

lic priesthood, and statistical methodology. Brandon

Vaidyanathan is an Associate Professor and Chair

of Sociology at The Catholic University of America. His

research examines the cultural dimensions of religious,

commercial, medical, and scientific institutions.

trends

B Y-

SA 2

.0 , cc

A protest sign at the April 2018 vigil for Decynthia Clements.

10

20

30

40

50

60

70

80

90

100 percent

54%

A little Quite a lot A little Quite a lot A little Quite a lot A little Quite a lot

White Hispanic or Latino OtherBlack or African American

Racial or ethnic differences in trust in the police and call preferences in a mental health crisis

Source: Mental Health in Congregations Study (2020)

I would call 911 I would call a mental illness crisis line I would call the police

How policy makers fund, support, and publicize mental illness crisis lines for individuals seeking alternatives to a police response is critical