Health promotion

State Laws on Emergency Holds for Mental Health Stabilization Leslie C. Hedman, J.D., John Petrila, J.D., L.L.M., William H. Fisher, Ph.D., Jeffrey W. Swanson, Ph.D., Deirdre A. Dingman, Dr.P.H., Scott Burris, J.D.

Objective: Psychiatric emergency hold laws permit in- voluntary admission to a health care facility of a person with an acute mental illness under certain circumstances. This study documented critical variation in state laws, identified important questions for evaluation research, and created a data set of laws to facilitate the public health law research of emergency hold laws’ impact on mental health outcomes.

Methods: The research team built a 50-state, open-source data set of laws currently governing emergency holds. A protocol and codebook were developed so that the study may be replicated and extended longitudinally, allowing fu- ture research to accurately capture changes to current laws.

Results: Although every state and the District of Columbia have emergency hold laws, state law varies on the duration of emergency holds, who can initiate an emergency hold,

the extent of judicial oversight, and the rights of patients during the hold. The core criterion justifying an involuntary hold is mental illness that results in danger to self or others, but many states have added further specifications. Only 22 states require some form of judicial review of the emergency hold process, and only nine require a judge to certify the commitment before a person is hospitalized. Five states do not guarantee assessment by a qualified mental health professional during the emergency hold.

Conclusions: The article highlights variability in state law for emergency holds of persons with acute mental illness. How this variability affects the individual, the treatment system, and law enforcement behavior is unknown. Research is needed to guide policy making and implementation on these issues.

Psychiatric Services 2016; 67:529–535; doi: 10.1176/appi.ps.201500205

The reforms in civil commitment statutes that occurred in the late 1960s and early 1970s led to profound changes in both substantive and procedural aspects of involuntary hospitalization (1). One such change, the addition of the requirement that persons affected bymental illness be either a danger to themselves or others or gravely disabled, re- quired that this determination be made before initiation of long-term commitment proceedings and that evidence of the determination be available in a commitment hearing (2). (The term “gravely disabled” refers to a person who, because of a mental illness, is unable to meet his or her basic needs, including the ability to meet the need for food, shelter, and basic self-care.) Toward that end, most states included an emergency hold period as part of the commitment process, during which a person could be placed in custody while the required determinations were made.

Emergency holds potentially play an important role as a bridge between people in crisis and emergency mental health services that individuals may not have otherwise been willing or able to access. Over the past three decades, this pathway has coexisted with a range of new approaches to the management of people with mental illnesses, including

the proliferation of police-based crisis intervention models and other forms of jail diversion (3). This article examines the current state of emergency hold law and identifies im- portant questions about the emergency hold mechanism in contemporary U.S. mental health systems that today bear little resemblance to the mental health systems in existence when many of these laws were enacted.

The pathway between people in crisis and the portals of local mental health services requires critical examination because of the serious health and social problems worldwide caused by undertreatment of mental illness (4). In the United States, 40% of people with a severe mental illness are untreated (5). People with severe mental illness who do not receive treatment are the most likely to end up in an acute mental health crisis in need of emergency hospitalization. Individuals facing a mental health crisis who do not receive treatment may go without care or, in the case of violent or disruptive behavior, be arrested (6).

An emergency hold (also called a 72-hour hold, a pick-up, an involuntary hold, an emergency commitment, a psychi- atric hold, a temporary detention order, or an emergency pe- tition) is a brief involuntary detention of a person presumed to

Psychiatric Services 67:5, May 2016 ps.psychiatryonline.org 529

ARTICLES

have a mental illness in order to determine whether the individual meets criteria for involuntary civil commitment; an emergency hold does not necessarily entail involuntary treatment (7). Under an emergency hold, a person may be confined in a health care facility at the behest of one or more categories of requestor. Generally, a requestor must fill out an affidavit or go before a judge to testify that a person has a mental illness and meets the state’s specified criteria for a hold as a result of that mental illness. Neither presentation of the matter to a judge nor prompt judicial review is uniformly required. The duration of an emergency hold is typically a few days, but there is significant variation among states.

Emergency holds are distinct from civil inpatient or outpatient commitment, which entails the involuntary treatment of mental illness over a period of days or weeks. An emergency hold is the shortest form of civil restriction on liberty and is often triggered in anticipation of a commitment proceeding (8). Emergency holds, and all forms of in- voluntary commitment, implicate constitutional rights of autonomy, liberty, and due process. Under the “danger standard” articulated in the Supreme Court’s 1975 decision in O’Connor v. Donaldson, a state cannot involuntarily com- mit people for treatment simply because they have a mental illness; instead, the state can exercise its police powers to coerce treatment only when individuals present a danger to themselves or others (9). O’Connor v. Donaldson established that the state may not confine a nondangerous individual who is “capable of surviving safely in freedom by himself or with the help of willing and responsible family members or friends.” Because of the short-term, emergency nature of emergency hold law, however, the statutory requirements typically differ from those set forth for long-term in- voluntary commitment (10).

In theory, emergency holds reduce harm and increase treatment access for people with mental illnesses, but the actual impact of these policies, applied to tens of thousands of individuals each year, has not been evaluated. Indeed, it is not even known reliably how many people are exposed to this intervention every year. Variation in provisions across

the states constitutes different “conditions” and a natural experiment for evaluation purposes. This article describes an available online open-source data set designed for mul- tistate evaluations of the current state of law governing emergency holds and raises important research questions.

METHODS

Using the methods set out in Anderson and colleagues (11), we conducted a comprehensive survey of current emergency hold laws effective on November 1, 2014. “Emergency hold laws”weredefined as statutes concerning the length, duration, criteria, and regulation of involuntary short-term psychiatric hospitalizations. The researchers worked iteratively and redundantly to develop a research protocol that reliably identified the target statutes. The final search terms included mentally ill, civil commitment, emergency commitment, emergency hold, mental illness procedures, firearm rights, and institutionalization procedures. Using Westlaw Next, the team searched for laws in all 50 states and the District of Columbia. The team used state legislature Web sites to ob- tain text of the current law. A coding scheme was developed to capture key operational features of the law and accom- modate cross-jurisdictional variation. The team used an it- erative process of duplicate coding and resolved discrepancies through discussion. Subject matter experts (JP and JWS) helped define the variables and the coding scheme and reviewed changes in the coding scheme. A detailed protocol is available at www.lawatlas.org. The final coding scheme consists of 11 variables, including circumstances triggering emergency hold, duration of emergency hold, who initiates an emergency hold, whether judicial review of an emergency hold is required, and the effect of an emergency hold on firearm rights.

RESULTS

All states and Washington, D.C., allow a person to be placed and held in a health care facility for treatment, observation, or stabilization without consent. Current laws vary on how and for what reason a person can be held, whether or not judicial review of the emergency hold is required, how long a hold can last, and the rights to which a person is entitled during and after the emergency hold. The most prevalent reason for an emergency hold is being a danger to oneself or others, and the most common maximum length of time permitted for the emergency hold is 72 hours (Table 1).

There is considerable variation in the categories of indi- viduals who may initiate a hold. Police in all jurisdictions have the authority to detain a person who appears to pose an imminent danger, and 38 states explicitly authorize police and peace or parole officers to initiate the emergency hold process. Police officers are the only legal initiators of emergency holds in two states (Wisconsin and Kansas). In 31 states mental health practitioners (such as psychologists, psychiatrists, and mental health workers) can initiate

TABLE 1. Duration of emergency holds and states’ ability to extend holds without a court order

Duration No court order

required Court order required

23 hours ND 24 hours AZ, DE, IL, ME, MI, MT, NC,

SC, UT 30 hours MD 48 hours GA, HI, IA DC, TX 72 hours LA, NY, TN, VT, WA AK, AR, CA, CO, CT, FL,

IN, KY, MA, MN, MS, NJ, NV, OR, VA, WI, WY

96 hours MO, OH 5 days ID, OK, PA, SD 7 days AL, NM 10 days NH, RI Unspecified KS, NE, WV

530 ps.psychiatryonline.org Psychiatric Services 67:5, May 2016

STATE LAWS ON EMERGENCY HOLDS FOR MENTAL HEALTH STABILIZATION

an emergency hold, in 22 states medical personnel (including physicians and nurses) can initiate an emer- gency hold, and in 22 states any interested person may ini- tiate the process. Most states allow more than one type of initiator (Table 2).

Eight triggering criteria appear with varying frequency across jurisdictions. Forty-five states and the District of Co- lumbia allowemergency holds when a person is a danger to him- or herself or to others due tomental illness. The five remaining states allow an emergency hold when a per- son is a danger to self or others without specifying that the danger is due to mental illness. Nineteen states al- low an emergency hold if the person is gravely disabled or unable to meet his or her ba- sic needs. Five states specify that a person who has re- cently attempted suicide may be held, even in the absence of ongoing suicidal ideation. Georgia is an outlier: the only criterion for an emergency hold is having amental illness and being in need of treat- ment (Table 3).

Twenty-two states require judicial approval for an emergency hold (Figure 1). In nine of these states, judicial approval is required before the admission, and whoever initiates the commitment must show probable cause before a judge or magistrate that the emergency commitment criteria have been met. The hearing may be conducted ex parte (that is, without the person subject to the hold being represented or present). If the judge or magistrate believes that there is probable cause to believe that the person meets one or more criteria for a hold, the judge in- structs the police to bring the person into custody and to a hospital where he or she will be examined. If the health care professional believes that the person meets the criteria, the person is placed under an emergency hold. In the other 13 states, judicial review and approval are required after ad- mission. These hearings are conducted ex parte and are based primarily on observations of the patient since his or her hospitalization. If the health care practitioner does not report satisfaction of the hold criteria, then the judge orders the immediate release of the patient.

The length of emergency holds varies by state (Table 2). The maximum time a person can be held ranges from 23 hours (N=1) to ten days (N=2). Twenty-two states have a 72-hour hold. In eight states, practitioners can extend an emergency hold without a court order.

Kansas, Nebraska, and West Virginia do not specify a maximum length for an emergency hold (Table 1). Kansas requires a health care professional to evaluate the patient within 17 hours and either release the individual or initiate involuntary commitment proceedings. InNebraska, a person who is taken into custody must be seen by a health pro- fessional within 48 hours. If the health care professional finds commitment to be medically justified, he or she will notify the county attorney, and long-term commitment procedures may begin; otherwise, the person must be re- leased. West Virginia requires a commitment hearing within 24 hours of the person’s being placed on an emergency hold.

Every state but Utah has emergency hold statutes that guarantee specific rights for a person being detained (Table 4). Twenty-one states require the hospital to allow

TABLE 2. Who can initiate emergency commitment and judicial review requirements, by state

Initiator No requirement Predetention ex parte hearing

Postdetention ex parte hearing

Any interested person AZ, DE, LA, MA, MN, MO, NC, SD, UT, WV

AR, CO, MD, MS, VA, VT

IA, IN, ME, NH, TX

Relative AZ, OK MS, NY NV Friend AZ Police officer AL, CT, DE, FL, HI, LA, MA,

MO, MT, OH, RI, WI NY KS, NV, TN, WY

Peace officer AK, AZ, CA, CO, DE, IL, KY, LA, MD, MI, MT, NE, NM, OK, OR, PA, SD, TX, UT

NY ME, MI, NH

Parole officer OH Physician AK, AZ, CT, DE, FL, GA, HI, KY,

LA, MA, MD, MO, MN, NC, NJ, OH, OR, PA, RI, UT

NV DC, ND, NH, NV, TN, WY

Nurse AZ, MA, MO, NJ, RI CO, FL, NY ND Advanced practice

registered nurse CT, GA, HI, LA, MD, MN NH, WY

Physician assistant HI, MN WY Psychologist AK, CT, DE, GA, HI, LA, MA,

MD, MN, MO, NC, NJ, OH, RI

FL, NY DC, ND, NV, TN, WY

Psychiatrist AK, AZ, DE, HI, MO, NJ, OH, RI, UT

VA ND, NV, WY

Mental health professional AL, CA, CO, DE, GA, HI, MA, MD, MN, MO, NE, RI, UT, WA

FL, KY DC, ME, ND, NV, WY

Medical directors CA, OR Hospital staff ID Attorney HI MS Judge HI, IL, NJ FL, VA Social worker CT, GA, IL, HI, MA, MN, NJ, RI CO, FL, NY ND, NV, WY Clergy HI Government employee DE, HI County-appointed

professional HI, MD, MS, PA TN

Mental health program MO, NJ Guardian ID, OK MS, NY NV, TX

Psychiatric Services 67:5, May 2016 ps.psychiatryonline.org 531

HEDMAN ET AL.

the patient to make phone calls, 26 states offer the held person the ability to see an attorney, 12 states require that a hospital allow the refusal of treatment, and eight states guarantee the right to appeal the emergency hold. Twenty- nine states require the hospital to provide written notifica- tion of the reason for the hold. Ten state laws require discharge transportation for the patient after the hold.

The full LawAtlas data set, the text of the law, and the report, codebook, and research protocol are available for pub- lic use at http://lawatlas.org/ query?dataset=short-term-civil- commitment. The LawAtlas Web site includes a “contact us” feature, and people are encouraged to make contact in regard to errors found or updates needed.

DISCUSSION

Every state and the District of Columbia provide for tem- porary, involuntary hospitali- zation of people suffering from acute mental illness. The legitimacy and value of these interventions depend on several factors: the statu- tory criteria and their appli- cation, the accuracy of the process for triggering an emergency hold, the degree to which the intervention facilitates (or interferes with) access to care, and the re- lationship of holds and hold procedures to health and treatment outcomes. There is little research aimed at measuring these factors. This survey of mental health laws creates the foundation for studies to evaluate how emergency hold laws are being used and to assess the impact of the laws on care, community safety, and the treatment system.

The results of the study demonstrate the diversity of criteria that justify holds under state law. Current law generally reflects the stan-

dard established in O’Connor v. Donaldson that people cannot be forced into treatment unless they are at risk of serious harm or of seriously harming another (12). Many states initially defined dangerousness quite narrowly in state commitment laws. However, over time, concern about lack of treatment has led some policy makers to call for a less stringent standard for involuntary commitment generally,

TABLE 3. Reasons for emergency commitment, by state

State Danger to self

Danger to others

Mentally ill

Danger to self due to mental illness

Danger to others due to mental illness

Recently attempted suicide

Gravely disabled

Unable to meet basic needs

AK ✓ ✓ ✓

AL ✓ ✓

AR ✓ ✓

AZ ✓ ✓

CA ✓ ✓ ✓

CO ✓ ✓ ✓

CT ✓ ✓ ✓

DC ✓ ✓

DE ✓ ✓

FL ✓ ✓ ✓

GA ✓

HI ✓ ✓

IA ✓ ✓

ID ✓ ✓ ✓

IL ✓ ✓

IN ✓ ✓

KS ✓ ✓ ✓

KY ✓ ✓

LA ✓ ✓ ✓

MA ✓ ✓

MD ✓ ✓

ME ✓ ✓

MI ✓ ✓ ✓

MN ✓ ✓ ✓

MO ✓ ✓ ✓ ✓

MS ✓ ✓

MT ✓ ✓ ✓

NC ✓ ✓ ✓ ✓

ND ✓ ✓ ✓

NE ✓ ✓

NH ✓ ✓ ✓ ✓

NJ ✓ ✓

NM ✓ ✓ ✓

NV ✓ ✓

NY ✓ ✓

OH ✓ ✓

OK ✓ ✓

OR ✓ ✓

PA ✓ ✓ ✓ ✓

RI ✓ ✓

SC ✓ ✓

SD ✓ ✓

TN ✓ ✓

TX ✓ ✓

UT ✓ ✓

VA ✓ ✓ ✓

VT ✓ ✓

WA ✓ ✓ ✓

WI ✓ ✓ ✓

WV ✓ ✓

WY ✓ ✓

532 ps.psychiatryonline.org Psychiatric Services 67:5, May 2016

STATE LAWS ON EMERGENCY HOLDS FOR MENTAL HEALTH STABILIZATION

such as “lacking the capacity to consent to treatment,” based on the assumption that the longer people go without treatment after a first episode of psychosis, the worse their illness will become (12). The data presented here document the expansion of emergency hold criteria outside of the danger standard. Nineteen states allow emergency holds when a person is “gravely disabled” or unable to meet basic needs, which allows the emer- gency hospitalization of people who do not present an immediate danger to themselves. Most of these additional criteria are consis- tent with the standard in O’Connor, but both legal and clinical questions remain unan- swered. One study of commitment decisions made in California general hospital emer- gency rooms suggests that clinicians generally adhere to statutory criteria in their decisions (13), but there is no research on how criteria are being applied and whether the available criteria are influencing the incidence or ap- propriateness of emergency holds (and longer- term commitment).

Whether the criteria have an impact on future access to treatment is also unknown. Emergency holds are applied in an environment of chronic deficiencies in treatment capac- ity. Because there are far more people whomeet criteria for a psychiatric hold than there is space to accommodate them in short-term psychiatric facilities (14–16), it is certain that the hold process cannot be counted on (or justified) as a reliable gateway to treatment. A lack of short-term inpatient capacity invites the question of whether statutory provisions in many states’ involuntary hold laws can be implemented under these conditions.

Emergency hold laws also may influence community safety. States differ on who can initiate a hold, with police officers, trained mental health professionals, judges, and lay people among the choices.Whomaymake the decision to hold someone may in turn result in different interpretations of whether a person constitutes a danger to self or others. The effect of these differences on incidence of holds, the hold process, or hold outcomes has not been objectively evaluated. Similar knowledge gaps concern the required procedures and the rights of individuals subject to a hold. Available research suggests that if individuals believe they have been treated fairly and given a voice, their satisfaction and willingness to adhere to treatment may be enhanced (17–19), but the comparative effect on public safety is largely unexplored.

The effectiveness of the emergency hold as a tool to sta- bilize acute symptoms of mental illness also is largely un- known. Ideally, a patient placed on an emergency hold is discharged with a long-term care strategy. Unfortunately, many patients on emergency holds are discharged without a mental health care strategy or lack the resources to follow

through on the plan advised and find themselves in a cycle of crisis care (20). Emergency hold laws do not require the implementation of a long-term treatment strategy, and, remarkably, Alabama, Arkansas, Colorado, and Utah do not mandate that a person on an emergency hold be seen by a health care professional at all. Notably, Pennsylvania does not require an evaluation by a health care professional but mandates that a physician certify the treatment, which must be based on an assessment either before or during the hospitalization.

The difficulty of measuring these statutes in a scientifically valid manner has long presented a barrier to rigorous evalu- ation of emergency hold policy and, more broadly, of involuntary civil commitment. This research provides an open-source, regularly updated database for capturing the variation in these laws across states. The protocol and coding scheme may be used to create longitudinal data to facilitate quasi-experimental designs with a capacity to support credible causal inferences (21). The database facilitates future scientific research exploring these myriad, highly important, although presently uninvestigated, documented variations.

Optimal use of involuntary hold laws involves the balance of competing concerns: the welfare of adults with incapacitating mental health conditions, the civil rights of such adults, the public’s concern with safety, the high direct cost of acute inpatient psychiatric services, and the (perhaps even higher) indirect, deferred cost of not providing such services in a timely way to the people who need them. Finding such a balance may implicate public policies and en- cumber resources at the interface of behavioral health, law enforcement, civil courts, and criminal justice systems. That the features of the relevant state laws vary state by state,

FIGURE 1. State variation in requiring judicial approval before emergency holds

Judicial approval required (N=22) AR, CO, DC, FL, IA, ID, IN, KS, KY, MD, ME, MS, ND, NH, NV, NY,

SC, TN, TX, VA, VT, WY

Postdetention adversarial hearing (N=13) DC, IA, ID, IN, KS, ME, ND, NH, NV,

SC, TN, TX, WY

Predetention ex parte hearing (N=9) AR, CO, FL, KY, MD, MS, NY, VA, VT

Psychiatric Services 67:5, May 2016 ps.psychiatryonline.org 533

HEDMAN ET AL.

in ways that would seem to require correspondingly varying policy solutions and legislative reforms, amounts to a strong argument for more solid state-comparative research on the ef- fectiveness and fairness of emergency commitment practices and their legal underpinnings. The legal database described herein makes such research possible, feasible, and imperative.

This study focused on state statutes dictating the explicit standards regulating emer- gency hold law. A notable limitation of this study is that it considered only emergency hold laws and did not address the relationship between emergency hold law criteria and statutory criteria for longer-term involuntary com- mitment. We are developing a database to codify involuntary commitment statutes beyond the emergency hold criteria, which will make this type of analysis possible in the future. Hospital protocols and local law enforcement practices also have a significant effect on whether an emergency hold takes place (7). Court cases related to emergency hold standards may inform, or even control, how the le- gal standards are applied. Understanding the impact of state law on local practices is a key area of future re- search. In order to map out trends, the research team created categories that nec- essarily lost some finer dis- tinctions in the law; to address this limitation, the database includes a detailed protocol capturing the assumptions the coders implanted dur- ing the research phase.

CONCLUSIONS

Recent events, such as mass shootings, highlight the need for both acute and long-term mental health care, espe- cially as a means to prevent escalation of harm during mental health crises. Emer- gency holds may be one of

the portals through which people with untreated mental illnesses receive stabilizing and consistent mental health services. This study highlights the variability in state law and the acute lack of research on variations in law and practice that influence the incidence and outcomes of emergency holds. In addition, this article illustrates the methods by

TABLE 4. Rights of persons under emergency commitment, by state

State

Right to know reason for

commitment

Right to refuse

medication

Right to refuse

treatment

Right to make a phone call

Right to see an attorney

Right to see a health care professional

for an assessment

Right to appeal the emergency commitment

AK ✓ ✓ ✓ ✓ ✓ ✓ ✓

AL ✓

AR ✓ ✓

AZ ✓ ✓ ✓

CA ✓ ✓ ✓ ✓ ✓

CO ✓ ✓

CT ✓ ✓ ✓

DC ✓

DE ✓ ✓ ✓

FL ✓

GA ✓ ✓ ✓ ✓

HI ✓ ✓

IA ✓ ✓ ✓

ID ✓

IL ✓ ✓ ✓ ✓

IN ✓

KS ✓ ✓

KY ✓ ✓

LA ✓ ✓ ✓ ✓

MA ✓ ✓ ✓ ✓

MD ✓

ME ✓ ✓ ✓

MI ✓ ✓ ✓

MN ✓ ✓ ✓ ✓ ✓ ✓

MO ✓ ✓ ✓ ✓

MS ✓ ✓ ✓ ✓

MT ✓ ✓ ✓ ✓

NC ✓ ✓ ✓ ✓

ND ✓ ✓ ✓

NE ✓ ✓ ✓

NH ✓ ✓ ✓ ✓

NJ ✓ ✓ ✓ ✓

NM ✓ ✓ ✓ ✓

NV ✓

NY ✓ ✓ ✓

OH ✓ ✓ ✓ ✓ ✓

OK ✓ ✓ ✓

OR ✓ ✓ ✓

PA ✓ ✓

RI ✓ ✓ ✓ ✓

SC ✓

SD ✓ ✓ ✓

TN ✓

TX ✓ ✓ ✓ ✓

UT VA ✓ ✓

VT ✓ ✓

WA ✓

WI ✓ ✓ ✓ ✓ ✓

WV ✓ ✓

WY ✓ ✓ ✓ ✓

534 ps.psychiatryonline.org Psychiatric Services 67:5, May 2016

STATE LAWS ON EMERGENCY HOLDS FOR MENTAL HEALTH STABILIZATION

which comprehensive, systematic legal data sets can be created.

AUTHOR AND ARTICLE INFORMATION

Mrs. Hedman is with Legal Science, LLC (e-mail: [email protected]), and Dr. Dingman and Mr. Burris are with Public Health Law Research and Policy Surveillance Programs, Temple University Beasley School of Law, Philadelphia. Mr. Petrila is with the Department of Health Policy and Management, College of Public Health, University of South Florida, Tampa. Dr. Fisher is with the School of Criminology and Justice Studies, University of Massachusetts, Lowell. Dr. Swanson is with the Department of Psychiatry and Behavioral Sciences, Duke University School of Medicine, Durham, North Carolina.

The study was funded by the Public Health Law Research and Policy Surveillance Programs, both of which are funded by the Robert Wood Johnson Foundation.

The authors report no financial relationships with commercial interests.

Received May 26, 2015; revision received August 25, 2015; accepted October 6, 2015; published online March 1, 2016.

REFERENCES 1. Fisher WH, Grisso T: Commentary: civil commitment statutes:

40 years of circumvention. Journal of the American Academy of Psychiatry and the Law Online 38:365–368, 2010

2. Brooks RA: Psychiatrists’ opinions about involuntary civil com- mitment: results of a national survey. Journal of the American Academy of Psychiatry and the Law Online 35:219–228, 2007

3. Bonnie RJ, Reinhard JS, Hamilton P, et al: Mental health system transformation after the Virginia Tech tragedy. Health Affairs 28: 793–804, 2009

4. Kessler RC, Berglund PA, Bruce ML, et al: The prevalence and correlates of untreated serious mental illness. Health Services Research 36:987–1007, 2001

5. Bijl RV, de Graaf R, Hiripi E, et al: The prevalence of treated and untreated mental disorders in five countries. Health Affairs 22: 122–133, 2003

6. Lamb HR, Weinberger LE: Persons with severe mental illness in jails and prisons: a review. Psychiatric Services 49:483–492, 1998

7. Faulkner LR, McFarland BH, Bloom JD: An empirical study of emergency commitment. American Journal of Psychiatry 146: 182–186, 1989

8. Testa M, West SG: Civil commitment in the United States. Psychiatry 7:30–40, 2010

9. O’Connor v Donaldson. United States Reports: US; 1975. p 563 10. Lamb HR, Mills MJ: Needed changes in law and procedure for the

chronically mentally ill. Hospital and Community Psychiatry 37: 475–480, 1986

11. Anderson E, Tremper C, Thomas S, et al: Measuring statutory law and regulations for empirical research; in Public Health Law Re- search: Theory and Methods. Edited by Wagenaar A, Burris S. San Francisco, Jossey-Bass, 2013

12. Large MM, Ryan CJ, Nielssen OB, et al: The danger of danger- ousness: why we must remove the dangerousness criterion from our mental health acts. Journal of Medical Ethics 34:877–881, 2008

13. Segal SP, Laurie TA, Segal MJ: Factors in the use of coercive retention in civil commitment evaluations in psychiatric emer- gency services. Psychiatric Services 52:514–520, 2001

14. Bloom JD, Krishnan B, Lockey C: The majority of inpatient psy- chiatric beds should not be appropriated by the forensic system. Journal of the American Academy of Psychiatry and the Law 36: 438–442, 2008

15. Lamb HR, Weinberger LE: The shift of psychiatric inpatient care from hospitals to jails and prisons. Journal of the American Academy of Psychiatry and the Law 33:529–534, 2005

16. Liptzin B, Gottlieb GL, Summergrad P: The future of psychiatric services in general hospitals. American Journal of Psychiatry 164: 1468–1472, 2007

17. Cascardi M, Poythress NG, Hall A: Procedural justice in the con- text of civil commitment: an analogue study. Behavioral Sciences and the Law 18:731–740, 2000

18. Fondacaro M, Frogner B, Moos R: Justice in health care decision- making: patients’ appraisals of health care providers and health plan representatives. Social Justice Research 18:63–81, 2005

19. Murphy-Berman V, Cross T, Fondacaro M: Fairness and health care decision making: testing the group value model of procedural justice. Social Justice Research 12:117–129, 1999

20. Christy A, Petrila J, McCranie M, et al: Involuntary outpatient commitment in Florida: case information and provider experience and opinions. International Journal of Forensic Mental Health 8: 122–130, 2009

21. Wagenaar AC, Komro KA: Natural experiments: research design elements for optimal causal inference without randomization; in Public Health Law Research: Theory and Methods. Edited by Wagenaar A, Burris S. San Francisco, Jossey-Bass, 2013

Psychiatric Services 67:5, May 2016 ps.psychiatryonline.org 535

HEDMAN ET AL.