The Two Ewing Cases and Tarasoff (The California Therapists, March/April 2005) By David G. Jensen Staff Attorney for the California Association of Marriage and Family Therapists

 

 

GO BACK

 

On July 16, 2004 and July 27, 2004 moderate earthquakes rocked the otherwise staid world of psychotherapy in California. On July 16, 2004 the Court of Appeal, Second District, issued an opinion in the case of Ewing v. Goldstein, Ph.D. (2004) 120 Cal. App. 4th 807 ("Ewing I"), and on July 27, 2004 the same appellate court issued an opinion in the case of Ewing v. Northridge Hospital Medical Center (2004) 120 Cal. App. 4th 1289 ("Ewing II"). By abandoning the literal reading of California Civil Code section 43.92, Ewing I and Ewing II have sent shockwaves throughout California's legal and therapeutic communities. Understanding these cases begins with a working knowledge of the factual and procedural backgrounds of them.

The Factual Underpinnings of Ewing I and Ewing II
The facts stated herein are taken from the facts as reported in Ewing I and Ewing II; however, these cases may not recount the facts as they will be testified to at trial. This does not mean that someone is not telling the truth, or has not told the truth. For tactical reasons occurring before trial, attorneys will argue certain facts or avoid bringing up certain facts to try to effectuate some desired outcome, i.e., an early dismissal of the lawsuit. Consequently, at trial, individuals may testify to facts not included here.

Ewing I and Ewing II grew out of a tragic murder-suicide that occurred in the Los Angeles area on June 23, 2001. On that day, Geno Colello ("Colello"), a former member of the Los Angeles Police Department, shot and killed Keith Ewing ("Ewing"), the new boyfriend of Colello's former girlfriend, Diana Williams ("Williams").¹ The murder occurred as Ewing was washing his car in the driveway of his home.² Colello then turned the gun on himself and committed suicide.³ It seems, however, that Colello's life had been unraveling for years.

In 1997, Colello began receiving counseling from David Goldstein, Ph.D. ("Goldstein") for work-related emotional problems and problems concerning Williams.⁴ In early 2001, Colello became increasingly depressed and despondent over the termination of his seventeen-year relationship with Williams.⁵ His feelings escalated in June of 2001 when he learned that she was romantically involved with another man.⁶

Goldstein last saw Colello for treatment on June 19, 2001, but he spoke with Colello via the telephone on June 20, 2001 and again on June 21, 2001.⁷ During the June 21 conversation, Goldstein asked Colello if he was suicidal and Colello admitted to thinking about suicide.⁸ Goldstein discussed voluntary hospitalization with Colello, (presumably to avert a suicide) and Goldstein obtained permission from Colello to speak with Colello's father, Victor Colello.⁹

At dinner that evening, Colello asked Victor Colello for a gun so that Colello could shoot himself, but Victor Colello refused to give him one.¹⁰ Colello then said that his alternative was to get a gun and go kill Williams' new boyfriend and then himself.¹¹ Some type of altercation ensued between Colello and his father, and Colello ended up punching his father in the face.¹² Victor Colello then called Goldstein and reported what Colello had said about harming Williams' new boyfriend,¹³although, at trial, it is expected that Goldstein will deny that Victor Colello told him that Colello had threatened Ewing. Goldstein urged Victor Colello to take Colello to Northridge Hospital Medical Center ("Northridge"), where Goldstein arranged for him to obtain psychiatric care,¹⁴ again, presumably to prevent a suicide.

At Northridge, Art Capilla ("Capilla"), a licensed clinical social worker, assessed Colello.¹⁵ During the assessment, Colello's father told Capilla about the threat Colello made concerning Williams' new boyfriend,¹⁶ although Capilla will likely deny that Victor Colello told him that Colello had threatened Ewing. Capilla was initially going to have Colello involuntarily committed, presumably for being a threat to self, but fearful of the effect such an action would have on his career as a policeman, Colello agreed to voluntarily enter Northridge on June 21, 2001.¹⁷ He then came under the care of Dr. Gary Levinson ("Levinson"), a staff psychiatrist. ¹⁸ Levinson did not believe that Colello was suicidal, and over Goldstein's remonstrations, on June 22, 2001 he discharged Colello from Northridge.¹⁹

No one ever warned Ewing that Colello was dangerous to him, and, tragically, on June 23, 2001, one day after being discharged from Northridge, Colello murdered Ewing and then took his own life.²⁰

In February 2002, the Ewing family filed a wrongful death action for professional negligence against Goldstein, which resulted in Ewing I, and a wrongful death action for professional negligence against Northridge, which resulted in Ewing II. The Ewing family also filed suit against the Colellos and Dr. Levinson. This tragic tale of despondency, lost love, and rage is certainly heart rending for all those impacted by it. But, the unique factors of this tragedy have spawned legal ramifications that affect California psychotherapists and their Tarasoff obligations.

Ewing I and Ewing II and California Civil Code Section 43.92.
In Ewing I, ²¹ at the trial court level and in harmony with the literal reading of California Civil Code section 43.92, Goldstein contended that he could not be held liable for failing to warn Ewing and the police about the danger that Colello posed to Ewing because Colello (patient) had not directly disclosed to Goldstein (psychotherapist) his intention to harm Colello. The Ewing family countered Goldstein's argument by claiming that by virtue of Colello's statements to Goldstein, Colello's interactions with Goldstein, and the information Victor Colello allegedly communicated to Goldstein about the threat of harm that Colello posed to Ewing, Goldstein was aware of the threat of harm that Colello posed to Ewing. Consequently, Goldstein should have warned Ewing about such threat.

The Ewing family's contention runs countercurrent to the express, literal language of California Civil Code Section 43.92, which generally immunizes psychotherapists for failing to warn of, protect against, or predict a patient's violent behavior except in cases where the "patient has communicated to the psychotherapist a serious threat of physical violence against a reasonably identifiable victim or victims" and the therapist fails to make reasonable attempts to notify the intended victim and law enforcement. The trial court sided with Goldstein because Colello did not tell Goldstein personally that he intended to harm Ewing and it dismissed the case, a decision that was subsequently appealed by the Ewing family to the Court of Appeal, Second District.

In Ewing I, the Court of Appeal, Second District examined the question of whether a communication from a patient's family member, made for the purpose of advancing the patient's therapy, is a "patient communication" within the meaning of Civil Code section 43.92. The Court of Appeal, Second District, in Ewing I, and then again in Ewing II, said yes, communications from family members are patient communications within the meaning of Civil Code section 43.92. With Ewing I, the literal reading of section 43.92 was turned on its head.

Why would the Court of Appeal, Second District rule this way? After all, the decision creates enormous confidentiality problems for psychotherapists, who do not even acknowledge the identities of their clients to third parties, including family members, without written authorization. The Court of Appeal, Second District in Ewing I arrived at its conclusion for four reasons:

1. California law requires confidentiality to give way to disclosures of otherwise confidential information to protect third parties from physical harm.²²
2. California law protects as privileged communications any communications between patients and third parties that are made in furtherance of the patient's therapy.²³
3. There is no good reason why a threat that a parent shares with his or her child's therapist about the risk of grave bodily injury the patient poses to another should not be considered a patient communication in determining whether the therapist's duty to warn is triggered under 43.92.²⁴
4. A narrow, literal reading of 43.92 leads to the possibility of murder or grave bodily injury occurring when such violence could have been prevented had the therapist warned the intended victim and law enforcement.²⁵

Goldstein and Northridge appealed to the California Supreme Court to overrule Ewing I and Ewing II, but on November 10, 2004 the California Supreme Court declined to review the cases. Consequently, Ewing I and Ewing II have become part of California's legal landscape. These cases raise five questions for consideration.

DOES PATIENT JUST MEAN PATIENT?
The term "patient" in Civil Code section 43.92, post Ewing I and Ewing II, does not literally mean just a therapist's actual patients, but the term also includes "family members of patients."²⁶ Therefore, a communication from a patient's "family member" to the patient's therapist, made for the purpose of advancing the patient's therapy, may create a duty upon the therapist to warn an intended victim of the patient's threatened violent behavior.

This expansive interpretation of the term "patient" in section 43.92 has two wrinkles. The first being that Ewing I and Ewing II expressly limit this rule to "family members" or to the patient's "immediate family." But, what exactly is a family? Does the court mean a family in the "Leave It To Beaver" sense, or in the "Mrs. Doubtfire" sense? Must there be an intact marriage with a mother and a father living under the same roof? What if there has been a divorce? What if "mom" and "dad" were never married and have never lived together? Can a second cousin be a family member? What about domestic partners? Ewing I and Ewing II do not answer these questions, which is unfortunate because the lack of a definition for the key concept of "family" only creates more questions with which to wrestle.

We believe that the emphasis on family members is misplaced anyway because the information conveyed, i.e., the threat, is more crucial than the family relationship. A boyfriend or girlfriend, a college roommate, or a best friend may have more accurate information about a patient's mental state than a member of the patient's family.

Moreover, what about communicating with third parties who are not family members? This is a thorny issue for therapists because of their obligation to maintain confidentiality. In a footnote to Ewing I, the Court of Appeal, Second District declined to consider the question of what occurs when a third party who is not a member of the patient's "immediate family" is involved in the patient's therapy and that third party discloses a threat made by the patient to the patient's therapist.²⁷ Consequently, we are left to postulate about what should be done in such a situation.

Since the threat is more important than the family relationship, the analysis set forth herein should be followed, although this is an issue on which there is no definitive answer. One of the problems stemming from Ewing I and Ewing II is that it opens the door to possible liability if a therapist fails to act on a threat from a credible third party and the patient then harms the intended victim. The family of the victim will certainly want to hold the therapist liable for failing to warn and Ewing I and Ewing II give such family more ammunition for doing so. Hence, we believe that threats from credible third parties, in addition to threats from family members, may also give rise to a duty to warn. By credible, we mean someone who knows that the therapist is treating the patient and is providing information to the therapist that is consistent with the therapist's knowledge of the patient. This rationale comports more closely with the state's goal of preventing murder or grave bodily injury than a rule that does not allow for such contact.

But, what happened to patient confidentiality? The Court of Appeal, Second District was not unmindful of a patient's right to confidentiality and the need to maintain confidentiality in therapy. In Ewing II, the court opined that assurances of confidentiality are important for three reasons: (1) to avoid the stigma that results from seeking mental health care; (2) to effectuate counseling; and, (3) to facilitate trust between the patient and the therapist.²⁸ However, it remains the public policy of the state to limit confidentiality in order to protect individuals from physical harm.²⁹ Although our recommendation takes another chunk out of the wall of confidentiality, any lawyer that defends therapists will tell you that it is easier to defend a breach of confidentially action than a wrongful death action. And, in reality, choosing between the two does not necessarily lead to liability. There is room for a therapist to successfully navigate both options by acting competently under the circumstances.

Although, therapists should not be acknowledging the identities of their clients to third parties, in a situation where a patient is dangerous to himself or herself, or dangerous to the person or property of others, the law permits the therapist, under section 1024 of the Evidence Code, to contact whomever is necessary to prevent the threatened danger from occurring. Patient suicides or patient homicides rarely "come out of the blue," like lightning on a sunny day. Therapists generally have a context, in light of the treatment relationship, for gauging whether a particular patient is potentially suicidal or even potentially homicidal. This context is best developed through conducting thorough assessments, reviewing previous treatment records, taking thorough patient histories, and rendering good clinical work. This context will be the impetus for the need to communicate with family members or credible third parties.

The second wrinkle is that hearing a threat, whether from a patient, a family member, or even a credible third party, does not necessarily mean that a Tarasoff obligation has been created. The operative word is the word "may." Just because a patient, a family member of a patient, or a credible third party calls and tells a therapist that his or her patient has threatened to kill someone or harm someone does not mean that a Tarasoff obligation has been created. Further analysis is required.

THE KINDLING AND THE SPARK: WHEN DO I HAVE A DUTY TO WARN?
Ewing I and Ewing II require threats of patients, again, whether from the patient, a family member of a patient, or, we believe, even a credible third party, to be analyzed in light of what the therapist already knows about his or her patient. A therapist will have a duty to warn if the information communicated to the therapist by the patient, a family member of the patient, or a credible third party leads the therapist to believe that his or her patient poses a serious risk of grave bodily injury to another.³⁰ A good metaphor for this concept is the kindling and the spark, with the kindling being what you already know about your patient and the spark being the information that is communicated to you by the patient, a family member of your patient, or a credible third party.

For instance, you may already know that your patient has a previous history of violence, or that your patient has never harmed anyone; you may know that your patient has an inability to control his or her anger, or that your patient never loses his cool; you may know that your patient has command hallucinations and he does or does not do what the voices inside his head are telling him to do; you may know that your patient is abusing alcohol or drugs and that while under the influence of such substances he or she has become violent before, or you may know that your patient has never abused alcohol or drugs; or, you may know that your patient is impulsive and violent, or impulsive but not violent. What you know about your patient, or what you should know about your patient, based on your review of the patient's records, your assessments, and your clinical work, constitutes the kindling, but kindling without a spark is just small pieces of wood.

Kindling ignited by a spark, however, can turn into a destructive fire. What you are told by your patient, a family member of your patient, or a credible third party about your patient is the spark, which may or may not ignite the kindling. The law requires you to analyze the information communicated to you by your patient, the family member of your patient, or a credible third party in light of your knowledge of your patient. Does the information communicated lead you to believe that your patient is dangerous to another? Does the spark ignite the kindling?

For instance, suppose the mother of your male client calls and tells you that your client has just said that he is going to kill his girlfriend because she wrecked his new car. Although there is a threat, or a spark to complete the metaphor, the threat alone does not mean that you have a Tarasoff obligation. What you have to do is ask yourself the next question: does the threat cause you, as the therapist, to believe that your client is going to harm the girlfriend? If, after hearing the threat, you determine or believe that your client is dangerous to the girlfriend, you would then make reasonable efforts to notify the intended victim and law enforcement; conversely, if you did not determine or believe that your client is going to harm his girlfriend, then you would not have to make such attempts at notification. The key, however, is having reasons that you can articulate for your belief. Why do you believe he is dangerous? Is it because your patient has hurt women who have betrayed him? Is it because you know he values his car more than life itself and has threatened to harm anyone who even scratches it?

On the other hand, why do you believe he is not dangerous? Is it because he's made similar threats in the past and never carried out any of them? Is it because there is no history of violence? Whatever the rationale, it should be documented in the patient file. But what if I judge wrongly? What if I determine that a patient is not dangerous to a third party and then that patient harms the third party? Although this looks like a situation where the therapist would be liable, in reality he or she may not be. In Tarasoff, the California Supreme Court recognized that it is difficult to forecast violence. Hence, the court stated that:

"[O]bviously, we do not require that the therapist, in making that determination, render a perfect performance; the therapist need only exercise that reasonable degree of skill, knowledge, and care ordinarily possessed and exercised by members of [that professional specialty] under similar circumstances. Within the broad range of reasonable practice and treatment in which professional opinion and judgment may differ, the therapist is free to exercise his or her own best judgment without liability; proof, aided by hindsight, that he or she judged wrongly is insufficient to establish negligence."³¹

The key to avoiding liability is having good reasons for the decisions you make. We cannot stress enough that such reasons should come from your review of prior treatment records, i.e., you have a duty to obtain the treatment records of a patient with a history of violence;³² your assessments and findings; and, your interactions with your patients.

What about so-called conditional threats? A conditional threat is one in which the patient says "I am going to kill X if...," or "I might kill X if…" Ewing I opines that even conditional threats can trigger a duty to warn,³³ assuming the therapist believes the patient is dangerous to the intended victim. The condition appears to be irrelevant. A therapist may believe that a patient is dangerous to another even though the threat was made conditionally. Since the therapist will likely have no knowledge about whether the condition occurs, it seems prudent to warn the intended victim if the therapist believes his or her patient is dangerous to another. What if, in the previous example, the mother had told the therapist that she believes her son is dangerous to the girlfriend, but the son had never actually threatened the girlfriend? The Court of Appeal in Ewing I pointed out that a belief, unaccompanied by a statement of a threat, would not give rise to liability under section 43.92. There must be an actual threat for section 43.92 to be triggered. Hence, the mother's mere belief would not give rise to a duty to warn under section 43.92.³⁴

WHAT IS GRAVE BODILY INJURY?
A therapist has a duty to warn an intended victim when information is communicated to the therapist that leads the therapist to believe that his or her patient poses a serious risk of physical violence. The concepts of serious physical violence and grave bodily injury are synonymous. But, what exactly is grave bodily injury? Fortunately, Ewing I clarifies this concept. A threat to take another's life, if believed by the therapist, is sufficient to trigger the therapist's duty to warn, but the duty to warn can also be triggered if the patient "intends to commit an act or acts of grave bodily injury short of murder, but akin to "mayhem" or "serious bodily injury."³⁵

Mayhem includes such things as depriving a human being of a member of his or her body; disabling, disfiguring, or rendering a part of his or her body useless; cutting or disabling the tongue; putting out an eye; or, slitting the nose, ear or lip.³⁶ So, if your patient threatens to cut off his girlfriend's ear, and you believe him, you would have a duty to warn the girlfriend even though the patient's intent was not to kill the girlfriend.

Serious bodily injury means a serious impairment of physical condition, including, but not limited to, loss of consciousness; concussion; bone fracture; protracted loss or impairment of function of any bodily member or organ; a wound requiring extensive suturing; and serious disfigurement.³⁷ So, if your patient threatens to beat up his girlfriend, and you believe him, since the attack could cause loss of consciousness or concussion, you would have a duty to warn the girlfriend even though the patient's intent was not to kill the girlfriend.

Gently slapping or pinching a victim is not mayhem or grave or serious bodily injury.³⁸ Neither is grabbing and kissing.³⁹

IS EXPERT TESTIMONY REQUIRED?
Expert testimony is not required to establish liability for a psychotherapist's failure to warn under section 43.92; rather, the mind-set of a therapist about whether a particular patient is dangerous or not dangerous can be evaluated by resorting to common knowledge without the aid of an expert witness.⁴⁰

The key question is whether the therapist actually believes that his or her patient poses a serious risk of inflicting grave bodily injury.⁴¹ In Ewing II, the Court of Appeal, Second District concluded, "it is not beyond the layperson's ken to understand that a patient's threat to take another's life, if believed, is serious." ⁴² Basically, the court is saying that it does not take any specialized knowledge or insight into the human psyche to make these kinds of determinations. Laypeople can understand whether and why a therapist believed his or her patient was dangerous without the need of expert testimony.

CAN A PATIENT BE DANGEROUS WITHOUT EVER THREATENING SOMEONE?
Suppose you have a patient that has a history of beating his children, although the last of his children has long ago left the home, whenever he has a problem with his boss at work. In session this patient has just disclosed that his boss has written him up again and is threatening to fire him for poor performance. You also know that he has recently remarried, and that he now has a thirteen year-old stepdaughter living with him. As you are listening to your patient, the thought that the stepdaughter is going to be beaten enters your mind, like an unwelcome visitor, and you believe that your patient is going to harm the girl. You may even inquire about the girl's safety and your patient may deny that he will harm her, but you still have this gnawing belief that the girl is not safe. In essence, you have made a determination that your patient is dangerous to another, although the patient never said a word about harming her, and, in fact, denied that he would hurt her. Under these facts, do you have a Tarasoff obligation? Yes!

But, let's change the facts of the hypothetical a bit to further illustrate the concept. Same facts as above except that you believe that your patient is not now a threat to his stepdaughter because he has made tremendous progress in therapy. He is not the threat to his children that he was ten years ago. Do you still have an obligation under Tarasoff, the case, to do reasonable things to protect the girl? No! In light of your clinical work, you have determined that he does not pose a danger to her. The difference in the two outcomes is your knowledge of your patient.

Under Tarasoff, when a therapist determines that a patient poses a serious danger of violence to others, the therapist bears a duty to exercise reasonable care to protect the foreseeable victim from such danger.⁴³ The key word is "determines." A therapist can determine that a patient is dangerous to others in one of two ways: (1) the patient may tell the therapist directly that he or she intends to harm someone and the therapist then believes, that such patient will harm such individual or (2) the therapist may determine that a patient is dangerous to another from the therapist's knowledge of the patient's history or propensities and the patient's present situation.⁴⁴ These are very different determinations, and they are treated differently under Tarasoff. It is a mistake to think that section 43.92 has replaced Tarasoff in all instances; rather, it is more accurate to think of Tarasoff as having two faces. One face is concerned with acutal threats of patients, which are then governed byTarasoff and section 43.92; the other face is concerned with determinations that therapists make about their patients in the absence of actual threats, with such determinations then being governed just by Tarasoff.

Ewing I distinguishes between Tarasoff, the case, and section 43.92, the statute, by saying that the "resulting statutory provision, section 43.92, was not intended to overrule Tarasoff or Hedlund, but rather to limit the psychotherapist's liability for failure to warn to those circumstances where the patient has communicated an actual threat of violence against an identified victim…"⁴⁵

Moreover, Ewing I also states that section 43.92 "refers only to a patient's communication to his or her psychotherapist."⁴⁶ Section 43.92 was created by the state legislature to limit liability when patients threaten others. The legislative history of section 43.92 is clear, however, that it was not enacted to replace Tarasoff.47 Consequently, section 43.92 functions as an immunity statute in situations where patients threaten to harm others. But, we have just seen that a patient can be dangerous to others without ever making a threat. This second way of being dangerous is not covered by section 43.92 because section 43.92 only deals with threats communicated to therapists by patients, or post-Ewing I and II to threats communicated to therapists by family members of patients or even credible third parties. Tarasoff has two faces, and because it has two faces there are two rules that must be followed. Those two rules are:

• If a patient threatens to kill or harm a third party, the patient has made an actual threat, which then allows the therapist to tap into the immunity afforded by section 43.92 of the Civil Code, which requires therapists to make reasonable attempts to contact law enforcement and the intended victim. If the therapist does these things, and his or her patient harms the third party, the therapist is immunized from liability.
• If, in the absence of a threat, the therapist determines that a patient is dangerous to another, under Tarasoff, but not under section 43.92, the therapist must do reasonable things to protect such person from harm, including notifying the police, warning the victim, warning others likely to apprise the victim, arranging for the patient to be hospitalized, or really do anything that is reasonable under the circumstances to protect the intended victim.
  
  For a schematic outline of this information,please see "The Two Faces of Tarasoff " diagram to the right.

WHAT HAPPENED TO GOLDSTEIN?
The Court of Appeal's decision in Ewing I does not mean that Goldstein did anything wrong or that he has lost the case. The decision in Ewing I simply means that the case can proceed to trial, where a jury will decide whether Goldstein had a duty to warn Ewing because Goldstein believed Colello intended to kill or cause serious physical harm to Ewing. In making such a determination, the jury will take into account all of the relevant evidence, including the threat to kill that Victor Colello allegedly communicated to Goldstein. Alternatively, the case may settle out of court and we will never know what a jury would have decided in this case. As of the writing of this article, the parties were proceeding to trial.

Since Tarasoff situations can be frightening and perplexing, we recommend calling CAMFT, or your own attorney, for a consultation whenever the situation arises.

 

This article appeared in the March/April 2005 issue of The Therapist, the publication of the California Association of Marriage and Family Therapists, headquartered in San Diego, California. The information contained in this article is intended to provide guidelines for addressing legal dilemmas. It is not intended to address every situation that could potentially arise, nor is it intended to be a substitute for independent legal advice or consultation. When using such information as a guide, be aware that laws, regulations and technical standards change over time, and thus one should verify and update any references or information contained herein.

 

THESE EXCERTPS ARE REPRODUCED BY LAWRENCE E. HEDGES FOR THE PURPOSE OF CONTUNING EDUCATION WITH PERMISSION OF THE AUTHOR AND CAMFT.