Magazine article Developments in Mental Health Law

Understanding and Applying Virginia's New Statutory Civil Commitment Criteria

Magazine article Developments in Mental Health Law

Understanding and Applying Virginia's New Statutory Civil Commitment Criteria

Article excerpt

In 2008, Virginia's General Assembly enacted significant amendments to the Commonwealth's civil commitment statute, based on the recommendations of the Commission on Mental Health Law Reform (the "Commission"). This Article is designed to review the statutory language that modified the civil commitment criteria, provide examples of how the new language in the statute might be applied, and promote a common understanding of the commitment criteria across the Commonwealth.

I. Background

Previous commitment criteria (from VA. CODE [section] 37.2-817B) (key phrases in bold): (1)

   After observing the person and obtaining
   the necessary positive certification and
   considering any other relevant evidence
   that may have been offered,

   if the judge or special justice finds by clear
   and convincing evidence that

      (i) the person presents an imminent
      danger to himself or others as a
      result of mental illness OR has been
      proven to be so seriously mentally
      ill as to be substantially unable to
      care for himself and ...

      (ii) ... there is no less restrictive
      alternative to involuntary inpatient

   the judge or special justice shall order that
   the person be admitted involuntarily to a
   facility for a period of treatment not to
   exceed 180 days ...

New commitment criteria (from VA. CODE [section] 37.2-817C):

The revised statute amended the language of both prongs of the previous civil commitment criteria. The new statute provides the following:

   After observing the person and
   considering (i) the recommendations of
   any treating physician or psychologist
   licensed in Virginia, if available, (ii) any
   past actions of the person, (iii) any past
   mental health treatment of the person, (iv)
   any examiner's certification, (v) any health
   records available, (vi) the preadmission
   screening report, and (vii) any other
   relevant evidence that may have been

   if the judge or special justice finds by clear
   and convincing evidence that

      (a) the person has a mental illness
      and there is a substantial likelihood
      that, as a result of mental illness,
      the person will, in the near future,

        (1) cause serious physical harm
        to himself or others as
        evidenced by recent behavior
        causing, attempting, or
        threatening harm and other
        relevant information, if any, OR

        (2) suffer serious harm due to
        his lack of capacity to protect
        himself from harm or to provide
        for his basic human needs, and.

      (b) all available less restrictive
       treatment alternatives to involuntary
      inpatient treatment have been...
      determined to be inappropriate,

   the judge or special justice shall order that
   the person be admitted involuntarily to a
   facility for a period of treatment not to
   exceed 30 days ..."

Why were the civil commitment criteria revised?

The 2008 General Assembly made several changes to the civil commitment legislation designed to address two key problems.

* First, research conducted by the Commission documented striking variations on civil commitment procedures and outcomes throughout the Commonwealth. (2) This variability raises serious questions of fairness as well as how well the state was addressing the needs of persons with serious mental illness. It also suggested the need for greater statutory specificity to guide the various professionals involved with civil commitment proceedings.

* Second, the phrase "imminent danger" to oneself or others (used in the previous statute) was widely regarded as unduly restrictive.

To promote more uniform application of the civil commitment criteria as well as broadening the circumstances that could lead to civil commitment, the General Assembly modified the criteria for civil commitment based on proposals recommended by the Commission. (3)

What is the expected impact of these changes?

Some have expressed concerns that the changes in the criteria will significantly increase the number of requests for emergency custody orders (ECOs) and temporary detention orders (TDOs) and the number of petitions, hearings, and commitment orders. Obviously, we will have to wait and see what happens, but a substantial increase in such proceedings or in commitment orders appears unlikely to occur in our opinion. For one thing, empirical research in other states has repeatedly shown that changes in the wording of commitment criteria, standing alone, are not associated with major changes in the number or rate of commitment orders. This finding is generally thought to indicate that the major determinants of involuntary hospitalization rates are system and resource factors, such as number of available beds and the availability of suitable alternatives to hospitalization, not the legal criteria for commitment. If outcomes change as a result of modifying statutory criteria, these changes are likely to occur at the margins.

Second, the changes enacted by the General Assembly in 2008 may have the effect in many localities of tightening the current criteria in some respects while loosening them in others, adding further support to the idea that the overall impact of these changes will be felt at the margins in close cases rather than in a wholesale lowering of the threshold for involuntary treatment. Third, incremental increases in funding for crisis stabilization programs, and outpatient services should help, over time, to reduce pressure on the commitment process as these services come on line. It would be gratifying if those outcomes begin to emerge in the coming year.

Finally, one key source of uncertainty about the effect of the 2008 reforms concerns the new provisions relating to mandatory outpatient treatment (MOT) orders. Even though MOT is still available only as a less restrictive alternative for people who meet the inpatient commitment criteria, the detailed new procedures under the statute are likely to lead to more such orders than were issued under the prior statute. (In May, 2007, such orders were entered in about 6% of hearings, mostly in a few jurisdictions). However, the effect of any increase in the number of MOT orders on the number of in-patient commitments remains to be seen.

II. Some Considerations Regarding the Meaning of the New Language

One of the major goals of the civil commitment reforms adopted in 2008 is to promote more consistent interpretation of the law throughout the Commonwealth. In order to help achieve that goal, the Supreme Court and the Department of Mental Health, Mental Retardation and Substance Abuse Services intend to conduct ongoing training activities for all participants in the process. The purpose of this Article is to highlight key questions that can be expected to arise concerning the meaning of the civil commitment criteria and, on occasion, to offer some opinions regarding the interpretation intended by the General Assembly.

1. A necessary condition for involuntary commitment under both the previous and revised statute is the finding that the person has a "mental illness" and that he or she presents a risk of harm "as a result of mental illness." Although this statutory language remains unchanged, promoting a common understanding of the meaning of this language will support more uniform application of the statute. As a result, it is important to review some of the conditions that might affect a determination of whether an individual has a mental illness and is covered by the civil commitment statute in the first place.

Like most state commitment statutes, Virginia's commitment statute defines "mental illness" relatively broadly to mean "a disorder of thought, mood, emotion, perception, or orientation that significantly impairs judgment, behavior, capacity to recognize reality, or ability to address basic life necessities and requires care and treatment for the health, safety, or recovery of the individual or for the safety of others" (VA. CODE [section] 37.2-100). In general terms, any psychiatric diagnosis of a major mental disorder that is listed in Axis I of the American Psychiatric Association's diagnostic manual (DSM-IV-TR) would meet this definition. (Axis I basically includes all mental disorders except personality disorders and mental retardation, and encompasses schizophrenia, bipolar disorder, depression, anxiety disorders, and eating disorders.) It must be remembered, though, that even if a person has a mental illness, the symptoms must be severe enough to meet the above definition. For example, the symptoms of depression (such as sadness, nihilistic thinking, suicidal thoughts, and cognitive impairment) in major depressive disorder can range in severity, from being so mild that the individual is able to continue to meet all social and occupational demands to being so severe that the individual is acutely psychotic or catatonic. In addition, some mental illnesses (such as panic disorder) can present with symptoms that are more circumscribed, such that they are severe but nonetheless do not impair judgment, behavior, the capacity to recognize reality, etc. Therefore, an individual would not be subject to civil commitment unless (1) he or she has a mental illness, (2) the symptoms of the illness are significant enough to impair the individual's functioning as described above, and (3) he or she presents a risk of harm, specifically "as a result of mental illness" (as opposed to posing a chronic threat of harm for unrelated reasons).

Issues that sometimes arise in assessing whether some action is "a result of mental illness" are whether a person whose primary diagnosis is personality disorder, substance abuse or dependence, or certain neurological conditions has a "mental illness" and meets the threshold required for the civil commitment statute. Consider the following examples:

* Personality disorders. The issue of personality disorder is an important one. A severe personality disorder, such as borderline personality disorder, is associated with marked instability in interpersonal relationship, self-image, moods, and impulse-control. While most individuals with the diagnosis of borderline personality disorder are treated as outpatients, during periods of interpersonal crisis and/or in the context of other superimposed psychiatric problems such as mood disorder or substance abuse, they pose an increased risk of engaging in potentially harmful behavior toward themselves or others. Twenty percent of psychiatric inpatients meet the diagnostic criteria for borderline personality disorder, and 10% of individuals with borderline personality disorder ultimately die by suicide. An individual with a more severe form of personality disorder who is experiencing impairment in "judgment, behavior, capacity to recognize reality, or ability to address basic life necessities," therefore, would be potentially appropriate for civil commitment. However, if the personality disorder contributes to a chronically increased risk of engaging in violent behavior (but the increased risk is not attributable to the types of impairment just mentioned, as is the case for many individuals with antisocial personality disorder), the person would not be appropriate for civil commitment.

* Substance-related disorders. The fact that an individual has a history of or current substance-related disorder (alcohol or drug abuse or dependence) would not in itself constitute a basis for civil commitment. However, chronic substance use, acute substance intoxication, and/or substance withdrawal all constitute important risk factors in assessing an individual's risk either of causing serious physical harm to himself or others or suffering serious harm due to a lack of capacity to protect himself from harm or provide for his basic human needs. As with mood disorders, anxiety disorders, or psychotic disorders, the symptoms of substance-related disorders occur along a continuum of severity, from non-problematic social drinking to "problem drinking" and ultimately all the way to severe substance addiction. Substance abuse in its more severe forms can cause mood swings similar to those seen in major depressive disorder (including hopelessness and suicidal ideation), can cause psychotic symptoms (including voices telling one to kill himself), and can cause cognitive impairment as severe as that seen in other forms of dementia. In addition, other psychiatric illnesses, such as mood disorders, psychotic disorders, or personality disorders, can be dramatically exacerbated by substance abuse. In summary: a person's "status" as a substance abuser per se is not a sufficient predicate for commitment but (a) acute and chronic medical complications of drinking could lead to an increased risk of harm to oneself or others, and (b) substance abuse can complicate other psychiatric illnesses, thereby contributing to an increased risk of violence.

* Medical conditions with psychiatric features. Another important point to consider relates to the relation between mental and physical disorders, such as Alzheimer's disease. Medical conditions and psychiatric diagnoses are not mutually exclusive under the modern understanding that mental illnesses (the more severe ones at least) have a biological basis. Alzheimer's disease or brain injury would qualify as a mental illness under the commitment statute if the patient has impaired "judgment, behavior, capacity to recognize reality, or ability to address basic life necessities." The issue sometimes presented in these cases is whether a mental health facility is the proper placement for a person with a neurological or other medical condition with psychiatric features. In practice, such patients are admitted to acute care psychiatric hospitals when they are (1) medically stable enough to be managed on a psychiatric unit rather than a medical unit and (2) when the primary problem leading to admission is emotional or behavioral problems that need to be addressed, similar to any other mental illness being admitted. Sometimes, the primary treatment provided to such an individual while on the psychiatric unit is medical. For example, it is quite common for Alzheimer's patients in a nursing home to acutely become more agitated and to be admitted to a psychiatric unit. Rather than simply starting medications to treat the agitation, the first step of treatment is to elucidate the cause. Often a medical problem such as a bladder infection is enough to trigger worsening of cognitive impairment and increased aggression and the primary treatment is to prescribe an antibiotic rather than a psychotropic medication. Even though this is "medical care," it's a part of overall psychiatric treatment. So, psychiatric treatment is defined by the nature of the presenting complaints and not whether the underlying cause is medical or psychiatric. The decision about whether care should be delivered on a medical unit or a psychiatric unit is a medical triage decision based on where treatment can most safely be provided, rather than simply a categorical distinction with a bright line. For example: does the patient require intensive medical monitoring due to medical instability? Is the patient an elopement risk who would best be treated on a locked psychiatric unit rather than an open medical unit?

Once an individual is found to meet the threshold of having a "mental illness" under the civil commitment statute, the criteria for commitment must then be applied to the facts of the case to determine the appropriate course of action.

2. The revised statute replaces the term "imminent danger to himself or others" with the phrase "substantial likelihood that ... he or she will cause serious physical harm to himself or others...."

   The basis for this change was that the
   term "danger" was considered to be
   excessively vague on two crucial grounds.
   First, it provided no indication of how likely
   the anticipated harm must be. Second, it
   provided no indication of how serious that
   harm must be in order for commitment to
   be justified. In contrast, the revised
   statutory language specifies that the harm
   must have a "substantial likelihood" of
   occurring (not just any likelihood, no
   matter how small). In other words, the
   potential harm to oneself or others must
   be regarded as probable, not simply
   possible. In this section of the statute, the
   new language also specifies that the harm
   must be of a "serious physical" nature--trivial
   injury or emotional harm will not
   qualify. But neither is it necessary that the
   predicted harm be lethal, as in suicide or

The revised statute does not spell out a specific percentage risk as to what would constitute a "substantial likelihood," and this remains a legal term of art. Nor does the statute specify or define the level of injury to oneself or others that would amount to "serious" physical harm. In actual practice, clinicians and legal decision-makers tend to employ a sliding scale model, in which the more serious the harm, the lower its likelihood needs to be in order to trigger civil commitment.

   Example: suppose the clinician is
   evaluating a patient with a likely
   diagnosis of borderline personality
   disorder who recently has cut himself
   superficially on the forearm. The
   clinician is aware that this patient has
   cut himself repeatedly over the years
   in order to relieve tension--without
   suicidal intent or major personal injury.
   Under these circumstances, the
   clinician might reach the conclusion
   that there is a very high likelihood that
   the patient will cause physical harm to
   himself in the near future, but that
   there does not appear to be a
   substantial likelihood of serious
   physical harm.

However, suppose this same individual, following a recent relationship breakup, cuts himself more deeply, to the point that sutures were required, and also overdoses on medications. Suppose further that, in the emergency room, he describes having few social supports and describes his outlook as still being pretty hopeless. He then declines voluntary hospitalization and says that his actions were impulsive and that he "thinks he'll be safe at home." Now the examiner is seriously concerned about the risk of serious physical harm. He feels that it is not a remote risk, but that it is difficult to quantify. When pushed to make a "ballpark" estimate, he says that there probably is about a one-in-four risk of another such incident occurring over the next few days. Is this a "substantial likelihood" of serious harm? What if he said one-in-five? One-in-ten? What level of risk of this kind of harm warrants involuntary hospitalization?

   Another example: suppose a man
   gets into an argument with his wife and
   then shoots himself in the chest? In
   the Intensive Care Unit (once off the
   ventilator and able to speak), he
   reports that he was very angry at his
   wife and just "wanted to get her
   attention" by pointing the loaded gun at
   his chest, but that the gun then went
   off accidentally. Let us assume that
   the evaluator concludes that (1) the
   patient is telling the truth, (2) the
   patient was and remains seriously
   depressed, and (3) his risk that he will
   engage in another act of serious self
   harm over the next few days is about
   one-in-ten. Given the severity of his
   recent self-harm, however, should this
   risk be considered a substantial
   enough likelihood of harm to justify
   involuntary hospitalization, at least for
   a day or two of further assessment?

Our opinion: admittedly, the examples are a bit artificial because clinicians do not have the ability to make quantitative probability estimates as precise as "one-in-four" or "one-in-ten" in these situations; at best, they are able to sort cases into risk levels based on qualitative clinical judgments. However, it is useful heuristically to attach probability estimates to such qualitative judgments. In our view, a "one-in-four" estimated risk of serious harm in the near future is sufficient, particularly when the harm being threatened is potentially fatal, as opposed to cutting, burning, or punching oneself. A "substantial risk" is not meant to mean "more likely than not" (51%). On the other hand, a very remote chance of serious harm is not sufficient. The estimates presented in the examples (one-in-ten or one-in-five) are meant to illustrate that the seriousness of the harm and the acuity of the danger inevitably affect judgments regarding whether there is a "substantial likelihood" of serious harm. Certainly a genuine short-term risk of serious harm, as in the second example, justifies detention for further evaluation.

3. The revised statute replaces the term "imminent" with the phrase "in the near future."

   The basis for this change was the
   evidence that some clinical evaluators and
   legal decision-makers were interpreting
   the term "imminent" to mean that the
   feared harm was expected to occur
   "immediately" or "within 24 hours." (4) In
   addition, the narrow interpretation of
   "imminent" has been a major target of
   criticism by clinicians and families of
   people with mental illness over the years.
   In fact, the Commission's research found
   that Virginia's statute was among the most
   restrictive in the country. Very few states
   require a showing of "imminent" danger,
   and several states that previously used the
   "imminent danger" standard have
   loosened their criteria in recent years. The
   Commission therefore concluded that
   "immediate" was an unduly narrow
   criterion, and that the term "imminent"
   should be omitted from the statute in order
   to assure that this restrictive approach
   would be discarded.

However, the Commission did not wish to leave the time frame for anticipated harm open-ended. Thus, the language proposed by the Commission, and adopted by the General Assembly, specifies that the harm must be anticipated to occur "in the near future," indicating that harm believed likely to occur in the more distant future (weeks to months) would not provide a sufficient predicate for commitment. Exact specificity (e.g. "in the next 48 hours") was deemed to be unworkable. So, what does "in the near future" mean? A significant consideration in interpreting this phrase is that mental health experts generally concede their inability to predict an individual's dangerous behavior related to acute mental illness beyond a period of about a week. Accordingly, a reasonable interpretation of "near future" would involve a time frame, generally speaking, of up to about one week. At the same time, assessment of a risk of violence inevitably involves fact-bound clinical judgments regarding the individual's clinical course within the context of his or her environment, especially interactions with other people. It would therefore be a mistake to embrace an absolute rule; periods slightly longer than a week are not precluded by the statutory language.

In general, the intended meaning of these statutory phrases is best understood by grounding their interpretation in a clinical context. Requests for involuntary treatment typically arise when people with serious mental illness are experiencing a significant decline or deterioration of functioning associated with impaired judgment, emotional distress, diminished grasp of reality, loss of self-control, and other symptoms. The question posed by the commitment criteria is whether this downward spiral, as evidenced by recent behavior as well as by mental and emotional symptoms, raises serious concern about harm "in the near future" if the deterioration were to continue without therapeutic intervention and without major amelioration of stresses in the environment. The assessment of risk must always be grounded in an understanding of the person's recent clinical course and in an assessment of the most likely clinical course in the near term--a horizon of about one week.

4. The revised statute specifies that the finding that there is a substantial likelihood of serious physical harm ... must be "evidenced by recent behavior causing, attempting or threatening harm and other relevant information, if any."

   Under the previous statute, there was no
   language indicating what constitutes an
   acceptable evidentiary basis for
   concluding that a person is "dangerous."
   The revised statute specifies that a clinical
   judgment that someone presents a
   "substantial likelihood" of causing harm in
   the near future must be "evidenced by
   recent behavior causing, attempting, or
   threatening harm." This requirement is
   designed to anchor the clinical risk
   assessment in the person's "recent
   behavior" and thereby avoid unfettered

A. The phrase "recent behavior" implies that harmful acts occurring long ago, although providing an important context, do not in themselves provide a sufficient evidentiary basis for civil commitment at the present time. A recent overt act or statement must be documented. However, actual harm need not already have occurred in order for commitment to be justified--recent acts or statements attempting or threatening harm will also suffice.

B. Recent behavior "causing" or "attempting" harm is likely to be easy to identify and document. However, the phase "threatening harm" is broader and more subtle, and several issues regarding the meaning of this phrase are likely to arise:

   Example: consider a person with a
   documented history of paranoid
   schizophrenia who voices the belief
   that her neighbors actually are foreign
   agents who are spying on her, has
   called 911 repeatedly to complain
   about them, and now has purchased a
   hunting knife and a rifle in order to
   "defend myself against them if it comes
   to that." Is she subject to commitment
   at the present time?

Our opinion: admittedly, this woman has not caused or attempted to harm her neighbors. However, has she engaged in conduct "threatening harm"? The first point to be noted is that the revised statute does not require evidence that the individual has made a specific threat against a particular identifiable individual; a generalized expression of intention or inclination to cause serious harm to anyone as a result of mental illness would be sufficient. (5) In this case, the woman's conduct would provide a sufficient behavioral basis for commitment as long as the totality of the evidence supports the necessary finding.

The woman in the hypothetical case is subject to involuntary commitment if, as a result of her illness, there is a "substantial likelihood" that, if not treated, she will cause serious harm to the neighbors or someone else in the near future.

Purchasing the weapon and making these statements under the circumstances would suffice to establish a recent behavioral basis for the prediction ("recent behavior. .. threatening harm") even though she has not yet caused or attempted harm and has not yet identified a specific victim of an increasingly likely dangerous act. Whether this woman can be shown, by clear and convincing evidence, to present a "substantial likelihood" of causing serious harm in the near future would depend on the full clinical picture, including her history of violence. The point being made here is that the statements and assembling of weapons would provide a sufficient behavioral basis for such an otherwise supported clinical judgment.

   Example: suppose an individual who
   has a well-documented history of
   mania has just this afternoon fired a
   gun into the air in his yard "as a
   warning" to the world at large that he is
   'in charge." However, he is not at this
   point verbally threatening to shoot any
   specific individual. Assume that the
   clinician has a high level of concern
   that the person will, if not treated, fire
   his weapon impulsively and recklessly
   when other people would be at risk.
   Has the person engaged in "recent
   behavior causing, attempting or
   threatening harm?" His recent
   behavior did not cause harm. Did he
   attempt to cause harm? Is a concrete
   "verbal threat" necessary under these

Our opinion: it is possible, of course, that the person described in this vignette could be subject to criminal charges for endangerment or discharging a firearm, but let us assume that civil commitment is sought instead. The statutory phrase "recent behavior ... threatening harm" does not require evidence of a specific verbal threat or physical menacing (such as swinging a tire iron and pointing it at someone). The language indicates that the behavior itself can constitute a threat. Suppose a person started carrying around a baseball bat without verbally threatening or without suggestively swinging it in the direction of his father. If he had previously attacked his father with a baseball bat, this behavior would properly be considered "threatening."

To summarize, the most reasonable interpretation of the overall phrase "as evidenced by recent behavior causing attempting, or threatening harm" is that it refers to any recent behavior that evidences a threat of harm; it is designed to anchor the clinical judgment that "there is a substantial likelihood that the person will cause serious physical harm" in recent threatening conduct. Thus the phrase should not be read as if it were referring to the elements of a criminal offense that require a "specific intent" to cause injury, such as attempt, or a purpose to put someone in fear of such harm, such as extortion. Any behavior that is "threatening" when seen in the context of the person's symptoms provides an ample basis for the risk assessment even if it does not amount to a specific verbal threat.

C. What is meant by the phrase "and other relevant information, if any"? This phrase is designed to make it clear that "any relevant evidence" may be introduced and used by the decision-maker to support the finding that "there is a substantial likelihood that the person will cause serious physical harm ..." as long as the finding is supported at least by "recent behavior causing, attempting of threatening harm." (6)

   Example: Mr. E, an individual with a
   longstanding history of schizophrenia,
   who lives with his father, has a history
   of violent assault against his father
   when he is ill, most recently about one
   year ago, when he attacked his father
   with a knife in their home. He was
   hospitalized voluntarily about three
   weeks ago and was released from the
   hospital after one week. This most
   recent hospital stay was much briefer
   than most of his past hospital stays
   (one week, compared to previous
   stays that lasted about three months
   on average). Since his hospital
   discharge, it appears to Mr. E's father
   that he has not been taking his
   medication. (His father cannot say for
   sure, since Mr. E resists any
   supervision or outside monitoring of his
   medication administration.) Mr. E
   refuses to attend the outpatient
   appointment at the local Community
   Services Board (CSB) that was
   arranged upon his discharge from the
   hospital. Over the past week, he has
   become almost completely mute, looks
   around him constantly, as if perceiving
   things that are not there, and has been
   glaring intensely at his father with
   clenched fists. There is no evidence
   that he is failing to eat or is losing
   weight. He refuses all offers of
   outreach services and refuses to go
   with his father to the emergency room.
   His father ultimately calls the police,
   who transport him on an ECO to the
   emergency room where he is seen by
   the CSB evaluator. Is there a sufficient
   evidentiary basis for commitment of
   Mr. E under the revised statute?

Our opinion: the revised statute requires a threshold finding that the person has engaged in recent behavior causing, attempting, or threatening harm. Conceivably, Mr. E's recent behavior (glaring at his father with clenched fists) does in fact "threaten harm." However, this finding is not, by itself, a sufficient basis for commitment. The ultimate question under the statute is whether there is a "substantial likelihood" that this man will cause serious physical harm to his father or someone else "in the near future." The statute doesn't require one to draw this conclusion solely based upon consideration of the recent behavior in isolation. The modern practice of risk assessment for violence involves looking at a variety of relevant data, including the individual's past history of violence while ill, his current clinical symptoms, and even certain demographic factors, such as age.

In this particular case, Mr. E has begun to demonstrate all of the symptoms that he has demonstrated in the past when ill. In the past while ill, he has become violent toward his father. His current behavior includes actions that indicate persecutory ideas about his father, and in fact he resides with his father. (It might be another matter had he been discharged from the hospital to a group home or shelter.) His most recent hospital stay was much briefer than previous hospital stays (allowing less time for full recovery) and he is refusing to attend outpatient follow-up and likely isn't taking his medication. All of these factors would constitute "other relevant information," and might serve to heighten one's ultimate level of certainty that there is "substantial likelihood of serious physical harm." The evidence, taken as a whole, strikes us as legally sufficient for commitment.

5. In the second set of criteria that can be used as the basis for civil commitment, the revised statutory language replaces the phrase "substantially unable to care for self' with the phrase "suffer serious harm due to his lack of capacity to protect himself from harm or to provide for his basic human needs."

   The previous statute did not specify what it
   meant for a person to be "unable to care
   for himself." The goal of the new language
   was to provide reater s ecificit
   regarding the circumstances under which
   a protective intervention would be justified.
   The new language focuses on the
   outcome that this prong of the commitment
   standard seeks to avoid, i.e., "harm" to the
   individual. It specifies that the predicted
   harm must be a "serious harm," whether
   it is attributable to a failure to protect
   oneself from harm or to a failure to provide
   for one's basic needs. We think that these
   two phrases should be read together since
   the various types of incapacity due to
   mental illness tend to overlap.

A. What does "serious harm" mean? Note that unlike the "danger to self or others" criterion discussed above, which requires a substantial risk of serious physical harm, this provision requires evidence of "serious harm." A risk of serious physical injury or death obviously qualifies. However, the omission of the requirement that the harm be physical was intentional. The "suffer serious harm" criterion was originally proposed by the Commission after deliberations in which supporters of the proposed language explicitly indicated that it was intended to cover harms other than physical harm, such as financial harm. Moreover, the special Mental Health Subcommittee of the Courts of Justice Committee of the House of Delegates rejected a proposal that would have limited the criterion to physical harm. Thus, the key interpretive issues arising under this prong of the commitment criteria relate to the meaning of "serious harm."

If attributable to mental illness, and likely to occur in the near future, the following predicted harms might amount to serious harms under this portion of the statute:

* Serious financial harm that could result from a person spending his or her life savings while in a manic state.

* Serious medical harm due to failure to seek medical care or take prescribed medications. Failure to take insulin in an individual with longstanding history of diabetes with a past history of life-threatening diabetic ketoacidosis following a previous discontinuation of insulin, likely would qualify, as would the failure to take antibiotics in the context of a current severe pneumonia. By contrast, failure to take antihypertensive medications, which might result in a heart attack or stroke at some point in the next decade, likely would not qualify.

* Eviction from lodging due to the person's grossly inappropriate behavior.

* Loss of custody of one's children because of grossly inappropriate or dangerous parenting.

* Loss of employment due to grossly inappropriate workplace behavior.

* Engaging in illness-related criminal behavior that would be highly likely to lead to arrest and incarceration if the police were to decide when confronted with such behavior to initiate the criminal process.

B. The revised statute states that the individual must "lack capacity" to protect himself from harm or to provide for his basic human needs. When does substantial impairment of judgment, cognition, or emotional control symptomatic of mental illness amount to a "lack of capacity" to protect oneself? A person who is unconscious or catatonic obviously lacks capacity to protect him or herself. But the cases that typically arise involve people who are both conscious and mobile. In applying this criterion, the focus should be on deficits in capacities relating to those activities of daily life that, if not carried out, can lead to "serious harm." In the context of emergency civil commitment, the emphasis is likely to be on a recent change in the person's functioning and an associated decline in relevant capacities for self-protection (whether due to symptoms of an acute illness, such as mania, or to the marked decline of capacities in a person with a chronic condition, such as dementia). Its assessment is therefore likely to be focused on whether the person has recently exposed him or herself to serious harm and on whether interventions designed to prevent harmful behavior have been attempted and failed. If so, this would amount to evidence of "lack of capacity" for self-protection.

C. The problem of homelessness: every state has to grapple with problems relating to people with mental illness who are homeless. Being chronically homeless and on the street, for example, likely would not be regarded by most evaluators or decision-makers as demonstrating a lack of capacity to protect oneself from harm or provide for one's basic needs, even though such a person is chronically at risk of harm of one kind or another. However, the scenario might be different in the middle of the winter if the individual is not agreeing to accept shelter and the lack of self-protection is attributable to mental illness and would otherwise provide a basis for intervention by Adult Protective Services. Moreover, if the person had previously been a high-functioning individual who has recently experienced a severe functional decline over the past few weeks (e.g., left his job, left his home) and is now disoriented and wandering the streets, most evaluators likely would consider this to constitute a significant likelihood of "suffering serious harm" due to severe incapacity attributable to mental illness.

The differences between these two cases lie in the time frame (the latter case is more acute in terms of the decline in functioning), in recent behavioral evidence of this decline in functioning, and in a high likelihood of a downward trajectory. In the latter case, a reasonable observer might conclude that the behavioral change crosses a "threshold of serious concern" and that precautionary action is indicated.

   Discussion problem: what about a
   homeless woman who, generally, is
   getting by on the street but is now
   pregnant? There is, clearly, a risk to
   the pregnancy--both the woman's and
   her fetus's health is in jeopardy. Is she

   Discussion problem: (7) a sixty-year-old
   woman diagnosed with schizoaffective
   disorder has been in and out of
   psychiatric hospitals for the past fifteen
   years. For the past year, the patient
   has lived in a residential facility and by
   all accounts has been functioning quite
   well. However, several weeks ago,
   she began to intermittently decline
   medication and has become
   increasingly agitated and bizarre in her
   behavior. She ultimately was
   "discharged" from the residential
   facility because she was
   "unmanageable." At that time, she
   refused to take her medications
   entirely and also refused voluntary
   psychiatric hospitalization.

   Upon leaving the facility, she
   immediately spent seventy dollars and
   then turned up at her daughter's
   apartment broke and without a place to
   stay. Her daughter convinced her to
   go to the hospital, where she was
   voluntarily admitted. However, she
   signed herself out several hours later.
   At midnight the police called her
   daughter to inform her that her mother
   had ordered a lobster dinner and then
   had left the restaurant without paying.
   The police had transported her
   (voluntarily, not on an ECO) to the
   local CSB, where the CSB evaluator
   was able to persuade her to sign
   herself into the hospital voluntarily.
   However, the next day, she again
   signed herself out. She remained
   broke and homeless. Two days later,
   the police again called her daughter,
   reporting that she had again ordered a
   dinner for which she did not pay. They
   again took her to the Emergency
   Room but at that point she refused
   voluntary admission, and a TDO is
   sought. Is the patient committable?

   Aftermath: assume that (1) the
   prescreener concluded that the patient
   was not committable, (2) the
   magistrate refused to issue a TDO,
   and (3) the patient was released. That
   afternoon she phoned her daughter
   from the cemetery, insisting that her
   deceased husband was out of his
   grave and causing her a lot of trouble.
   She was arrested for loitering. She
   appeared unkempt and dirty and was
   carrying a bag full of garbage. A nurse
   at the jail called the daughter
   requesting background information.
   Because the patient was continuing to
   decline all psychiatric treatment, she
   remained off her psychotropic
   medications while at the jail and was
   housed in segregation throughout her
   time in jail. Her daughter feared
   bonding her out, as her mother at least
   was in a sheltered setting at that point.
   A CSB evaluation requested by the jail
   psychiatrist determined that she did
   not meet commitment criteria while in
   her current sheltered setting, as she
   was eating adequately and had not
   engaged in assaultive behavior.
   Ultimately she went to court and the
   charges were dismissed. She again
   was homeless after this, sleeping
   primarily in bus depots. Was this the
   correct response?

Comment: this case study demonstrates the problems associated with a more restrictive interpretation of "serious harm." Obviously, decision makers may differ in how they would approach this case. Of note, nobody involved in this case disagreed that (1) this woman had a mental illness, (2) her illness could potentially benefit from psychiatric treatment, and (3) she was incapable of providing (or refusing to provide) valid informed consent for psychiatric treatment. What should be done in cases of this kind?

III. Conclusion

This review of Virginia's revised civil commitment criteria is designed to begin the iterative process of developing a common understanding of the new criteria, and thereby minimize the variability in its application across the state. As with any legal innovation, however, unanticipated questions about the meaning and application of the new provisions will continue to arise, and every effort will be made to establish a mechanism for sharing ideas and information as experience accumulates over the coming months and years. This paper is meant to initiate that ongoing process.

(1) Under the old and new versions of the statute, the same criteria apply to involuntary admission to a facility for inpatient treatment and to mandatory outpatient treatment as a less restrictive alternative to inpatient treatment. The quoted portions of the statute pertain to involuntary admission to a facility.

(2) The Commission conducted an interview study of 210 stakeholders and participants in the commitment process in Virginia. The report of that study, entitled Civil Commitment Practices in Virginia: Perceptions, Attitudes and Recommendations, was issued in April 2007. The study is available at us/cmh/civil_commitment_practices_focus_groups.pdf. A second major research project was a study of commitment hearings and dispositions (the "Commission's Hearings Study"). In response to a request by the Chief Justice, the special justice or district judge presiding in each case filled out a 2page instrument on every commitment hearing held in May 2007. (There were 1,526 such hearings). The Commission's Hearings Study can be found at 2007_05_civil_commitment_hearings.pdf.

(3) Based on its research and the reports of its Task Forces and Working Groups, the Commission issued its Preliminary Report and Recommendations of the Commonwealth of Virginia Commission on Mental Health Law Reform ("Preliminary Report") in December, 2007. The Preliminary Report, which is available on-line at minary_report.pdf, outlined a comprehensive blueprint for reform ("Blueprint") and identified specific priorities for consideration by Virginia's General Assembly in 2008.

(4) This is certainly a plausible interpretation of the phrase. Indeed, the definition for "imminent" provided by the Merriam-Webster's Collegiate Dictionary is "ready to take place; especially: hanging threateningly over one's head ."

(5) A threat to a specific person would be required to trigger a duty to take precautionary action under VA. CODE [section] 54.1-2400.1, but such a specific threat has never been required as a predicate for civil commitment in Virginia or elsewhere.

(6) The revised statute requires the magistrate, Community Services Board (CSB) evaluator, and Independent Examiner to consider a wide array of other information in making his or her determination. In addition, the statute makes clear that the judge or special justice is also expected to consider all records, reports, and relevant information admitted at the hearing.

(7) Adapted from Darold A. Treffert, The Obviously III Patient in Need of Treatment: A Fourth Standard for Civil Commitment, 36 HOSP. & COMMUNITY PSYCHIATRY 259, 261 (1985).

By Bruce J. Cohen, Richard J. Bonnie, and John Monahan *

* The authors thank Katherine Acuff and Jane Hickey for their comments and suggestions. However, the opinions expressed represent the authors' own views, and should not be attributed to any one else.

Author Advanced search


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.