Pirjo K. Garby brought this wrongful death and survival action following the suicide of her husband, Michael Garby. Mr. Garby took his own life approximately six hours after his discharge from the Emergency Room of the George Washington University Hospital ("the Hospital") during the night of November 7-8, 1998. Mrs. Garby contends that her husband's suicide was proximately caused by the professional negligence of the Hospital and of Jeffrey S. Akman, M.D., who was the decedent's attending psychiatrist.

Mrs. Garby, a native of Finland, had been married to her husband for approximately three years.

Following a trial in February and March of 2003, the jurors were unable to agree upon a verdict and the judge declared a mistrial. In a subsequent written order, however, he granted the defendants' motion for judgment as a matter of law, concluding among other things that the evidence was insufficient to reasonably permit a verdict by a jury that any negligence of the defendants had proximately caused Mr. Garby's death.

On appeal, Mrs. Garby argues that the defendants breached the applicable standard of care in several respects, chiefly in that although they were aware that the decedent was depressed and mentally ill and that he had reported recent plans to commit suicide by jumping off a bridge, they released him to go home with his wife without apprising her of his suicidal ideation. She contends that Mr. Garby's leap to his death from his eighth floor balcony hours after leaving the hospital, while she was in the shower, was proximately caused by the defendants' professional negligence. For the reasons that follow, we agree with the trial judge that the evidence was insufficient to support a reasonable inference by a jury that the alleged negligence of the defendants proximately caused Mr. Garby's death.

I.

On Saturday, November 7, 1998, Mr. Garby, an electrical engineer, was in a depressed and seemingly paranoid frame of mind, believing that he might be in legal trouble and that numerous persons, including his wife, were conspiring against him. On the previous day he had gone to see his attorney, who later testified that during their meeting Mr. Garby exhibited such nervousness, and so lacked any sense of proportion, that his manner "could be described as bordering on delusion." In the days before his death, Mr. Garby's suspicions had intensified, as he believed, among other things, that waiters, bartenders, and his wife were working for the police in an effort to set him up for copyright infringement, and that the police were tapping every telephone he used. As a result of his concern about the supposed interception of his conversations, Mr. Garby made calls from various telephones to his sister, Ruth Torres, a police detective in Connecticut. After Mr. Garby had made some twenty calls to Ms. Torres in a single day, she urged him to seek medical assistance. Michael Garby's other sister, a nurse, provided similar advice, as did his wife. Ultimately, Mr. Garby agreed to follow the women's suggestions.

According to Mrs. Garby, her husband was afraid that he might be in serious legal trouble because some blueprints which had been stolen from another company were used by his own company, and because he could be a suspect in the theft or unlawful use of the blueprints.

On the evening of November 7, Mr. Garby presented himself at the Hospital's Emergency Room. His wife accompanied him, but at his request she was not present in the room when he described his problems to the physicians. Mr. Garby reported to Emergency Room personnel that he was, or had been, experiencing anxiety, persecutory delusions, and suicidal thoughts. Craig Norris, M.D., the first doctor to examine Mr. Garby, noted in Mr. Garby's chart that the patient "had been feeling anxious and paranoid [at] work [and at] home for [the] past 2 + weeks (maybe more) and more depressed [and had] some suicidal ideation + plan to jump off bridge." Tenagne Haile Mariam, M.D., the supervising physician in the Emergency Room, also interviewed Mr. Garby and wrote in his chart that the patient was suffering from "mild paranoia depression — `whole world is against me,' has thoughts of suicide — `to jump off a bridge' — but no other more concrete plan."

Dr. Mariam referred Mr. Garby to a psychiatric resident, Alfredo F. Soto, M.D., who spoke with Mr. Garby in some detail. Dr. Soto noted, after examining Mr. Garby, that the patient reported legal problems which he refused to describe in his wife's presence. Dr. Soto wrote that the patient "notes recent [increasing] hopelessness because of [the legal problems]," and that "from this hopelessness, he has had some SI w/plan to jump off bridge." According to Dr. Soto's notes, "[h]is wife, who is unaware of his concerns w/legal prob's, has noted [greater] paranoid ideation X2 days with [greater] awareness of persecutory feelings/concerns on her husband's part." Neither Mr. Garby nor the physicians told Mrs. Garby of his suicidal thoughts.

SI stands for "Suicidal Ideation."

Initially, both Dr. Soto and Mrs. Garby believed that Mr. Garby should remain in the Hospital. Mrs. Garby did not want her husband to sign a document in which he agreed to be released to go home, but despite the advice of the doctors, Mr. Garby stated that he wanted to go home. Dr. Soto telephoned Dr. Akman, the attending psychiatrist who was Dr. Soto's superior, at Dr. Akman's home and the two physicians discussed the case for some time. Dr. Soto ultimately wrote in Mr. Garby's record:

As pt does not meet full criteria for involuntary hospitalization, he has agreed (as has his wife) to be observed by his wife over next 24-48 hours. Both have agreed to call both insurance co. and our outpatient clinic to obtain urgent F/U w/in next wk. Attending (Dr. Akman) agrees.

Mr. Garby was released from the Emergency Room between 2:30 and 3:00 a.m.; his wife was given a prescription for sleeping pills (Ambien) for her husband. According to Mrs. Garby, Dr. Soto told her that "it was my responsibility for the [next] 48 hours" to keep an eye on her husband. The couple then left the Emergency Room and walked home without having the prescription filled.

Mr. and Mrs. Garby went to bed at approximately 3:30 or 4:00 a.m. on Sunday, November 8. Mr. Garby woke up at about 7:00 a.m. After he and his wife had breakfast, Mrs. Garby decided to be the first to take a shower. When she came out of the shower, she discovered, to her horror, that her husband had leaped to his death from the eighth floor balcony of their apartment.

In his written order of July 21, 2003, the trial judge entered judgment as a matter of law in favor of the defendants. He first concluded that, "according to the evidence in the case, defendant doctors adhered to the standard of care required of them, and provided adequate aid and treatment under the circumstances." Although, in the judge's view, this conclusion mooted the issue of proximate causation, he went on to conclude alternatively that Mrs. Garby's position that the defendants had proximately caused Mr. Garby's suicide "amounted to mere speculation" and rested on insufficient evidence to meet her burden of proof by a preponderance of the evidence on that issue.

II.

We agree with the trial court that Mrs. Garby failed as a matter of law to prove that any negligence attributable to the defendants proximately caused the death of her husband. That being so, we need not resolve the parties' dispute over whether Mrs. Garby's medical expert, Dr. Cavanaugh, correctly defined a national standard of care allegedly breached by the defendants. We assume arguendo that in one or more respects Dr. Cavanaugh was correct (or that a jury could properly so find) in opining that the emergency room physicians failed to exercise reasonable care in the manner by which they treated or discharged Mr. Garby. Even so, Mrs. Garby was required to "introduce evidence . . . afford[ing] a reasonable basis for the conclusion that it is more likely than not that the conduct of the defendant[s] was a substantial factor in bringing about the [death of her husband; a] mere possibility of such causation is not enough; and [if] the matter remain[ed] one of pure speculation or conjecture, or the probabilities [were] at best evenly balanced, it [became] the duty of the court to direct a verdict for the defendant[s]." Gordon v. Neviaser, 478 A.2d 292, 296 n. 2 (D.C. 1984). See also Talley v. Varma, 689 A.2d 547, 552 (D.C. 1997) ("To establish causation, the plaintiff must present evidence from which a reasonable juror could find that there was a direct and substantial causal relationship between the defendant's breach of the standard of care and the plaintiff's injuries, and that the injuries were foreseeable."); Twyman v. Johnson, 655 A.2d 850, 854 (D.C. 1995) (directed verdict required where a conclusion that negligence had substantially contributed to the injury "would have rested upon surmise").

Of considerable importance to our conclusion that Mrs. Garby's proof of causation failed is that she has not challenged, either here or in the trial court, the determination by the Hospital and Dr. Akman that they had no basis for detaining Mr. Garby involuntarily for observation under the District of Columbia Hospitalization of the Mentally Ill Act, D.C. Code §§ 21-521 et seq. (2001) (the Ervin Act). Section 21-521 states that "a physician or qualified psychologist of the person in question, who has reason to believe that a person is mentally ill and, because of the illness, is likely to injure himself or others if he is not immediately detained may, without a warrant, take the person into custody, transport him to a public or private hospital, . . . and make application for his admission thereto for purposes of emergency observation and diagnosis" (emphasis added). Section 21-522(a)(2) in turn permits the administrator of a hospital to admit and detain, for purposes of emergency observation and diagnosis, a person certified by a psychiatrist or qualified psychologist to have "symptoms of a mental illness and [who], because of the mental illness, is likely to injure himself or others unless the person is immediately detained." Mrs. Garby presented no expert testimony disputing the reasonableness of the defendants' judgment that Mr. Garby's mental illness, as revealed to them, did not create a likelihood that he would injure himself unless immediately hospitalized under these statutes. In this court, as Mrs. Garby's principal brief was somewhat ambiguous on the point ("Appellant . . . has never contended that involuntary commitment was the only option for properly treating Michael Garby" (emphasis added)), the court pressed her attorney on it at oral argument, and he replied that "[w]e do not contend that [Mr. Garby] met the standard for involuntary commitment." This was in keeping with plaintiff's position at trial. Although her complaint alleged that the defendants had been "negligent in failing to involuntarily hospitalize . . . [Mr.] Garby," by the time of her pretrial statement the claim had changed to one that the defendants had negligently "fail[ed] to adequately urge and insist that [Mr.] Garby accept treatment at the hospital for his illness." At trial the judge inquired twice about the matter. The first time Mrs. Garby's counsel hedged, but when the judge pursued the issue by asking, rhetorically, "I don't expect that you would argue or imply in an argument that the decedent could have been involuntarily committed," counsel responded, "I can't argue facts that aren't in evidence, and I can't ask the jury to speculate." In his testimony regarding involuntary hospitalization, Dr. Cavanaugh, Mrs. Garby's expert, could say only that reasonable care required the emergency room doctors, as one option, to "consider involuntarily committing [Mr. Garby]." Yet it was undisputed that the doctors did consider that option but concluded that they lacked reason to believe he would injure himself unless detained, and Dr. Cavanaugh expressed no opinion that that judgment was negligent or mistaken.

Such detention need only be supported by "probable cause," Williams v. Meredith, 407 A.2d 569, 574 (D.C. 1979), but is limited in duration to forty-eight hours unless extended by court order. See D.C. Code § 21-523.

"[A judge of this court]: . . . [Y]ou are not arguing that if the doctors in the emergency room had been more probing, had taken more time, [and] had been less prepared to take at face value Mr. Garby's assertions that he was okay now[, . . . that] would have provided a basis . . . for seeking involuntary commitment. . . .

[Counsel for appellant]: That is essentially right. . . ."

"[The Court]: . . . [D]o you suggest from your evidence that . . . [Mr. Garby] was subject to involuntary commitment?

[Counsel]: Your Honor, that's going to be a qualified no."

To summarize, then, Mrs. Garby does not claim, and presented no evidence, that the defendants negligently failed to hospitalize Mr. Garby against his will or that any failure on their part to accurately assess and diagnose his condition deprived them of information that would have supported such commitment. Instead, her argument rests on three assertions related to the pivotal issue of causation:

(1) Had the physicians kept Mr. Garby in the emergency room longer for observation, they might have been able to persuade him to agree to voluntary hospitalization overnight or longer.

(2) Had they administered or prescribed anti-depressant or tranquilizing medications rather than send Mr. Garby home with a prescription merely for sleeping pills (Ambien), that might have quelled his suicidal impulses enough to prevent his death six hours later; and

(3) Had Mrs. Garby been informed of her husband's "true condition" rather than have it withheld from her based on the doctors' erroneous reliance on physician-patient confidentiality, she "would have done many things" at home "to further reduce the likelihood of the tragic events whether it was to keep him in her sight at all times, bring others in, such as his sister to help her out, or seek better medical treatment."

We consider these arguments in order.

A. Voluntary Commitment. Mrs. Garby argues that the defendants did not try hard enough to persuade Mr. Garby to check himself into the psychiatric unit on the night in question, relying on Dr. Cavanaugh's testimony that if Mr. Garby "couldn't be involuntarily committed for legal reasons," a reasonable physician "would try quite hard, very hard to get him to come in voluntarily." Only conjecture, however, supports a conclusion that additional efforts to persuade Mr. Garby to agree to voluntary hospitalization would have succeeded. By the time Mr. Garby was sent home, he had been in the emergency unit for nearly four hours, and twice during that time he had been told of and refused the option of voluntary commitment to the Hospital. Holding him there longer against his will, or attempting to persuade him more forcefully to remain there overnight, would have amounted to asserting the very same authority to compel his admission to the Hospital that Mrs. Garby admits the defendants lacked. Nor would enlisting Mrs. Garby's help to convince him to remain there have offered more than speculative assistance. The undisputed testimony was that Mr. Garby viewed her as one of the persons plotting against him. He did not want her present during his communications with the doctors, and when she initially opposed his signing a document agreeing to be released to go home, he insisted that he wanted to be released, and she acceded to his wishes. The inference Mrs. Garby argues that had she been more fully informed of his recent thoughts of suicide she would have persuaded him to stay in the hospital rests on surmise or "at best evenly balanced [probabilities]." Gordon, 478 A.2d at 296 n. 2. Indeed, informing Mrs. Garby of her husband's "true condition" (Br. for App. at 34) would have meant telling her that the doctors did not believe he was presently dangerous enough to himself to need hospitalization, and that information doubtless would have influenced how forcefully, if at all, she sought to have him remain there voluntarily.

Likewise, if the doctors had informed Mrs. Garby that they did not believe he was presently dangerous to himself, it is speculative at best whether such disclosure would have prompted different actions by Mrs. Garby upon taking her husband home.

B. Better Medication. Dr. Cavanaugh faulted the doctors for "send[ing Mr. Garby] home with sleeping pills," which made "no clinical sense." Rather, proper care would have been to "[s]end [Mr. Garby] home with antidepressant [or anti-anxiety] medications," which "might make some sense" because "[y]ou always want to treat the anxiety," something "[y]ou don't treat . . . with sleeping pills." Dr. Cavanaugh's testimony, however, failed to support an inference beyond conjecture that treating Mr. Garby with anti-anxiety or anti-depressant drugs would have prevented his suicide within six hours of his release. Other expert testimony, uncontradicted by Dr. Cavanaugh, was that a substantially longer period of time is required for anti-depressant drugs to take effect. While no similar testimony was given about tranquilizers, Dr. Cavanaugh did not explain how they, any more than anti-depressants, could be expected to work quickly and effectively enough to prevent Mr. Garby's suicide within so short a time. Further, prescribing either tranquilizers or anti-depressants depended on Mr. Garby's cooperation, and the evidence showed that when he and Mrs. Garby passed an all-night CVS pharmacy as they walked home from the Hospital, they did not stop to fill the Ambien prescription he had been given. In these circumstances, Dr. Cavanaugh's opinion that it "might make some sense" to prescribe anti-depressants or tranquilizers did not permit a conclusion by a preponderance of the evidence that doing so would have prevented Mr. Garby's suicide.

Concerning tranquilizers, Dr. Soto of the Hospital testified that there were generally two classes, "antipsychotics and . . . what they call benzodiazepines," that both "have significant side effects in their use," and that Mr. Garby had not been prescribed tranquilizers because on his release he would lack "close monitoring by someone who knows of the side effects."

C. Informing Mrs. Garby. Mrs. Garby's primary argument is that if the physicians had told her of her husband's suicidal ideation, she would have taken measures to insure that he was not in a position to harm himself that night. She relies on Dr. Cavanaugh's testimony that the exercise of proper care by physicians required that Mrs. Garby be made "fully cognizant of the risk that was being assumed, the potential for danger" in taking Mr. Garby home — "that she know exactly what had been going on with this man in the few days or so before he gets into the emergency room" — and that her ignorance in particular of his "suicidal ideation with [a] plan to jump off a bridge" denied her the ability to take precautions against him injuring himself.

This argument highlights the basic tension, not to say contradiction, in Mrs. Garby's position. A common theme of her briefs (opening and reply) is that the defendants erroneously believed that physician-patient confidentiality barred them from informing her of her husband's suicidal impulses. Thus, she states: "There is no doubt in this case that Pirjo Garby was not aware of her husband's suicidal thoughts or his plan to jump off of a bridge." And, citing her own testimony, she adds that "if she had been made aware of her husband's suicidal ideations she would not have taken custody of him" (Br. for App. at 24; Reply Br. for App. at 2; emphasis added). Dr. Cavanaugh too acknowledged that, had Mrs. Garby known the seriousness of her husband's condition, she realistically would not have believed herself able to protect him by any precautions at home:

Q. Now assuming[, Doctor, that] Mr. Garby had gone home and Mrs. Garby had been told that you believed that he was imminently in danger of committing suicide[,] . . . what would you have expected her to do?

A. I honestly would have expected her not to take him home.

Yet, as we have seen, Mrs. Garby presented no evidence and does not argue that the doctors were negligent in concluding that Mr. Garby was not "imminently in danger of committing suicide" such that he could be hospitalized for observation against his will. She similarly does not argue that if they had informed her of the gravity of his condition and learned that she was unwilling to take him home, that fact would have given them adequate reason, otherwise lacking, to involuntarily commit him.

The questions of whether Mr. Garby was mentally ill and dangerous enough to be detained against his will and whether the doctors were ethically at liberty to reveal his confidences to others, including to Mrs. Garby, are conceptually similar. See, e.g., Principles of Medical Ethics of the American Medical Association, No. IV, Anno. ("Psychiatrists at times may find it necessary, in order to protect the patient or community from imminent danger, to reveal confidential information disclosed by the patient.") (emphasis added); MacDonald v. Clinger, 84 A.D.2d 482, 446 N.Y.S.2d 801, 805 (N.Y.App.Div. 1982) ("Disclosure of confidential information by a psychiatrist to a spouse will be justified whenever there is a danger to the patient, the spouse or another person"). Given our resolution of this appeal, we do not consider whether, as the defendants argued at trial, disclosure of Mr. Garby's confidences to his wife when they did not believe he was an immediate danger to himself would have risked breaching their ethical duties.

What Mrs. Garby's position comes down to, rather, is that while her husband was not dangerous enough to be hospitalized against his will, he was "very, very close" to that (to quote her attorney's language at argument in this court), and thus it was foreseeable to the defendants that if she was not informed of his suicidal ideation and able to take measures to guard against it at home, he would attempt to end his life. The problems with this argument begin, however, with the fact that Dr. Cavanaugh himself was skeptical about the efficacy of any measures Mrs. Garby could reasonably have taken to prevent the suicide, as this exchange reveals:

A. . . . I would have expected her to be as vigilant as she possibly could have with some particular attention to [—] . . . which would be very difficult I admit [—] how to block access to the porch in their condominium. . . .

Q. You would not have expected Mrs. Garby to stay up all Saturday evening and Sunday morning to watch her husband, would you?

A. I would have liked that but I understand that would be impossible.

Q. And you would have expected Mrs. Garby, for example, to be able to go into the bathroom on her own and not continuously watch her husband?

A. Well . . . I would tell her that somebody's got to be with him. Now how that could be arranged I honestly do not know and I certainly agree with you. She could take a shower, sure.

Q. . . . Michael Garby was a pretty big guy, so if . . . he wanted to jump off the balcony . . . Mrs. Garby would [not] have been much physical restraint from him, would she?

A. No, it was a problem. . . . [T]he sending of him home was a problem.

Neither Mrs. Garby nor anyone else testified how she would have been able to secure help from relatives (who apparently lived in Connecticut) or friends, on such short notice, to monitor Mr. Garby's actions between 3:00 a.m. and his suicide later that morning.

Moreover, from the standpoint of foreseeability on the defendants' part — an essential component of proximate cause analysis, see, e.g., Psychiatric Inst. of Wash. v. Allen, 509 A.2d 619, 624 (D.C. 1986) — Mrs. Garby's assertion that her husband was not dangerous enough to be committed but "almost so" does not make sense. If the doctors reasonably did not believe that Mr. Garby was an immediate danger to himself, how could they have foreseen that he would take his life within hours of his release unless protected from himself by Mrs. Garby? Either he was dangerous enough to himself to be detained for observation, or he was not; if he was, it would have been irresponsible — negligent or even grossly negligent — of the defendants to release him from the Hospital with or without knowledge of his suicidal intention by Mrs. Garby, untrained in monitoring or preventing actions by someone mentally ill. If he was not, it could not fairly be held foreseeable to them — by a preponderance of the evidence — that releasing him would risk his killing himself that morning save only for knowledge by Mrs. Garby of his "true condition."

Our dissenting colleague points out correctly that Dr. Cavanaugh's credentials as an expert were not challenged and that he was firm in his opinion that failure to inform Mrs. Garby of her husband's suicidal ideation was "proximately, causally related to [Mr. Garby's] . . . subsequent death by suicide." But Dr. Cavanaugh's conclusion undermined itself in the same way that Mrs. Garby's position does. He testified, we have seen, that he would not have expected a fully-informed Mrs. Garby to agree to "take [her husband] home," yet he expressed no opinion that the defendants mistakenly believed they could not hospitalize Mr. Garby against his will, the only practical alternative to Mrs. Garby taking "custody" of him. Dr. Cavanaugh also opined that Mr. Garby's mental illness "was quite treatable . . . [a]nd if treated appropriately over a reasonable period of time almost certainly would have eliminated the future possibility of a completed suicide[,] with the continuation of appropriate treatment." The question for a jury, however, would not have been whether "continued" treatment of Mr. Garby "over a reasonable period of time" would have saved him from himself, but whether better precautions by the defendants on releasing him from the emergency room — specifically telling Mrs. Garby of his suicidal ideations — would reasonably have prevented his suicide within hours of his release. The evidence supporting an affirmative answer to that question did not rise above surmise or conjecture.

The dissent argues that Dr. Cavanaugh's inability to question the defendants' assessment that Mr. Garby was not immediately dangerous to himself is a "red herring," because the issue a jury could fairly decide was whether "a prudent plan for post-discharge treatment" would have sufficed to protect him — instead of a "negligently devised and executed" plan. Post at 534. But the only negligence the dissent points to in this regard is the failure to tell Mrs. Garby "the critical facts" of her husband's condition, and, as we have seen, a key such fact would have been the doctors' judgment — reached after hours-long observation, and unrebutted by Dr. Cavanaugh — that Mr. Garby was not an immediate danger to himself and so did not require hospitalization. The dissent nevertheless assumes throughout the fact of Mr. Garby's present dangerousness and goes on to assert, for example, that had Mrs. Garby known "what the physicians knew" she "doubtless would have . . . attempted to persuade him to remain at the hospital voluntarily and . . . he may have agreed to do just that." Id. at 528, 529-530. What the dissent concedes here is "only a possibility," id. at 530, is pure conjecture given the undisputed evidence, mentioned earlier, that Mr. Garby excluded his wife from his confidences with the doctors because he saw her as part of the conspiracy against him. The dissent's added suggestion that "[i]f [Mr. Garby] had not agreed to remain, he may . . . have reacted in a manner that would have changed the defendants' appraisal of the extent of his dangerousness to himself," id. at 530, merely further compounds the speculation. And the dissent's ultimate conclusion that "[t]he enhanced sense of urgency that would have existed if Mrs. Garby had known of her husband's suicidal planning" — "plans" the doctors, without contradiction by her expert, believed were not active or seriously-entertained enough to warrant emergency hospitalization — "would surely have generated a dramatically different scenario with dramatically different consequences," id., substitutes what the dissent and all of us wish had happened for evidence.

The dissent mistakenly treats Mrs. Garby's concentration on the issue of "post-discharge planning" as a mere tactical choice of her counsel that implies nothing about Mr. Garby's dangerousness. In fact Mrs. Garby's position was necessitated by Dr. Cavanaugh's inability to fault the defendants' judgment that Mr. Garby could safely be released from the Hospital because he was not likely to injure himself.

Merely by posing the question in their briefs, "would Mr. Garby have changed his mind," the defendants cannot remotely be said to have "conceded" the reasonable possibility that he would have, contrary to the dissent's assertion. Post at 528, 529.

Assuming it were to make a difference (which the dissent does not quite say), our colleague opines that "the majority's approach to the issue of causation was introduced into the controversy by [the majority]." Post at 533. That is incorrect. A repeated theme of the defendants' case, at trial and in this court, has been the absence of evidence countering the doctors' conclusion that Mr. Garby was not an immediate danger to himself, and on that basis they argue on brief — as part of the eleven pages they devote to proximate causation — that no causal connection existed between whatever "greater knowledge" Mrs. Garby should have had about her husband's condition and his suicide. See, e.g., Br. for Appellees at 23. Unsurprisingly, causation, and specifically foreseeability, then became a focal point of oral argument. In any case, appellant can scarcely claim "the procedural unfairness" the dissent suggests concerning any aspect of our causation analysis, post at 533, when her opening brief ignored entirely the trial court's determination of no proximate causation.

In the end, the dissent is simply unable to accept as reasonable the defendants' judgment that Mr. Garby was not presently dangerous to himself and so did not require hospitalization. Mrs. Garby herself, it says, "would have been unpersuaded by the notion that her husband was not a danger to himself." Post at 532. But Mrs. Garby has never argued, because the evidence would not support the argument, that the doctors would have had grounds otherwise lacking to commit her husband involuntarily had she only been able to express her disagreement with their diagnosis or unwillingness to accompany him home. The dissent is thus left with the possibilities already discussed — that a better-informed Mrs. Garby might have talked her husband into remaining in the hospital voluntarily or, in any case, would not have "react[ed] in the same way as she did without the information." Post at 42. As we have explained, these suppositions fail as a matter of law to establish a direct and substantial causal relationship between any omissions by the defendants and Mr. Garby's death. See Talley, 689 A.2d at 552.

Affirmed.

SCHWELB, Associate Judge, dissenting:

I regret that I am unable to agree with the majority's affirmance of the judgment of the trial court. In my opinion, the trial judge erred by directing a verdict in the defendants' favor, and Mrs. Garby has been unjustly denied the right to have a potentially meritorious case decided by a jury. Accordingly, I respectfully dissent.

As noted by my colleagues, Mrs. Garby's principal contention on appeal is that the defendants breached the applicable standard of care in that, although they were aware that the decedent was deeply depressed, paranoid, and mentally ill, and that he had reported recent plans to commit suicide by jumping off a bridge, they effectively placed him in the care and custody of his wife without apprising her of his suicidal ideation. Mr. Garby's thoughts of ending it all by leaping to his death were reported to the doctors and fully documented in the hospital records, see maj. op. at p. 513, ante, but no disclosure was made to the woman whom the defendants expected to look after him.

With respect to causation — the issue that divides the court — Mrs. Garby contends that her husband's leap to his death from his eighth floor balcony while she was in the shower was proximately caused by the defendants' professional negligence. In my view, the expert testimony and other evidence presented by Mrs. Garby were sufficient, if credited by the jury, to establish not only the applicable standard of care and its breach, but also to show that the breach proximately and foreseeably resulted in the decedent's suicide. Accordingly, the questions of negligence and causation were for the jury, and I would reverse the award to the defendants of judgment as a matter of law (JMOL).

As I read the majority opinion, if one were to put to one side the "waiver" theory on which that opinion is largely based, my colleagues would not seriously dispute Mrs. Garby's contention that there was sufficient evidence in the record of each element of medical malpractice to warrant the submission of the case to a second jury (the first jury having been unable to agree). Rather, according to the majority, Mrs. Garby effectively surrendered or waived what would otherwise have been a legally sufficient case by not asserting that the defendants were legally obliged to commit Mr. Garby involuntarily. This apparently tactical determination, my colleagues suggest, means that, for purposes of this appeal, the decedent was not dangerous to himself at all, regardless of his documented recent plan to jump off a bridge, and no matter how negligent the defendants' post-release planning may have been.

It is worth noting that the trial judge had initially denied the defendants' motion for a directed verdict because he believed that there were issues of fact which were properly the province of the jury.

I address the majority's insistence that this was not a tactical determination in Part III of this opinion.

This claim by my colleagues that Mrs. Garby's failure to argue for involuntary commitment defeats proximate cause originates entirely with the majority. In some twelve pages devoted to proximate cause in the brief for the defendants, there is not the slightest suggestion of such a contention. The trial judge's brief discussion of the issue of proximate cause likewise fails to connect the issue in any way with the theme that the majority has now adopted. This is not surprising, for if the failure to argue for involuntary commitment is relevant to any issue in the case other than whether Mr. Garby should have been involuntarily committed — and, in my view, it is not — it bears on whether there was a breach of the standard of care rather than on causation.

The defendants assert that because there is no claim that Mr. Garby should have been committed involuntarily, and because there is no evidence that he was incompetent, they could not have been negligent in letting him go home. I do not agree with this contention either, for the plaintiff's expert testified forcefully to the contrary.

In any event, at least as I see it, involuntary commitment is a red herring. It diverts the inquiry from the specific negligence alleged by Mrs. Garby and from the foreseeable consequences of that negligence. An impartial jury, assessing the record in the light most favorable to the plaintiff, could reasonably find, as the plaintiff's expert opined, that the defendants breached the standard of care by placing Mr. Garby in the custody of a woman who knew nothing about his recent plan to jump off a bridge. As I endeavor to show in Part II of this opinion, a jury could likewise reasonably find that this conduct on the part of the defendants proximately and foreseeably resulted in the decedent's death.

I. THE STANDARD OF CARE AND ITS BREACH

Although the disagreement between the majority and myself centers on whether or not the evidence of causation was sufficient to go to the jury, that issue can most readily be understood in the context of the entire case. I therefore address each of the principal legal issues.

A. The standard of review.

The question whether the trial judge properly directed a verdict in favor of a party is one of law. Phillips v. District of Columbia, 714 A.2d 768, 772 (D.C. 1998). Accordingly, this court owes no deference to the trial judge's ruling, and our review of his order is de novo. Washington Metro. Area Transit Auth. v. Jeanty, 718 A.2d 172, 174 (D.C. 1998).

B. The JMOL standard.

Although my colleagues in the majority seem to attach little, if any, significance to the point, a trial court may grant judgment as a matter of law only where "a party has been fully heard on an issue and there is no legally sufficient evidentiary basis for a reasonable jury to find for that party on that issue." Super. Ct. Civ. R. 50(a). "On motion for a directed verdict, the record must be viewed in the light most favorable to the non-moving party, and that party . . . is entitled to the benefit of every reasonable inference from the evidence." Jeanty, 718 A.2d at 174; see also Bauman v. Sragow, 308 A.2d 243, 244 (D.C. 1973) (per curiam). "As long as there is some evidence from which jurors could reasonably find the necessary elements, a trial judge must not grant a directed verdict." Marshall v. District of Columbia, 391 A.2d 1374, 1379 (D.C. 1978); see also Abebe v. Benitez, 667 A.2d 834, 836 (D.C. 1995). Judgment as a matter of law should therefore be granted sparingly, and only in extreme cases. District of Columbia v. Wilson, 721 A.2d 591, 596 (D.C. 1998); King v. Pagliaro Bros. Stone Co., 703 A.2d 1232, 1234 (D.C. 1997). Finally, in ruling on a motion for judgment as a matter of law, the trial court "must take care to avoid weighing the evidence, passing on the credibility of witnesses or substituting its judgment for that of the jury." Pazmino v. Washington Metro. Area Transit Auth., 638 A.2d 677, 678 (D.C. 1994) (quoting Vuitch v. Furr, 482 A.2d 811, 814 (D.C. 1984)).

In the present case, the trial judge found as a fact ("the court is convinced . . .") that the Hospital's doctors "acted reasonably" and "did not . . . deviat[e]" from the standard of care. The judge also found, contrary to the testimony of James Cavanaugh, M.D., the plaintiff's expert, that the standard of care described by Dr. Cavanaugh was "strictly personal and not in keeping with the national standard of care." Missing from the judge's order is any explicit (or apparent) recognition that the record must be viewed in the light most favorable to the plaintiff, and that the jury, not the court, determines witness credibility, "reasonableness," and the existence, vel non, of a deviation from the national standard. Nevertheless, my colleagues appear to agree with the judge's approach, at least as to proximate cause, and perhaps more broadly than that.

This is no minor matter. Adherence to the "light most favorable" standard is not a negligible factor in the calculus, which can be satisfied by rote recitation but ignored in practice. A court's refusal to assess the evidence in the light most favorable to the party against whom JMOL is sought, or to draw every reasonable inference in that party's favor, deprives the non-moving party of the right to trial by jury protected by the Seventh Amendment. In this case, in my opinion, the improvident award of JMOL to the defendants, without any serious attempt to apply the "light most favorable" standard, effectively denied Mrs. Garby that fundamental constitutional right.

Indeed, in their brief description of the legal standard, maj. op. at p. 514, ante, my colleagues do not mention the "light most favorable" standard at all, by rote or otherwise.

C. The standard of care.

Mrs. Garby's evidence regarding the applicable standard of care was contained in the testimony of her expert witness, James L. Cavanaugh, Jr., a physician specializing in general and forensic psychiatry. Dr. Cavanaugh testified, to a reasonable degree of medical certainty, that there was a single national standard that was applicable to Dr. Akman and Dr. Soto on the night of November 7, 1998. According to Dr. Cavanaugh, there are "not different standards for Chicago and Washington and Baltimore and San Francisco." Describing the defendants' responsibilities under the national standard, Dr. Cavanaugh explained that the physicians were obliged to (1) make an accurate assessment and diagnosis of the patient's condition; (2) weigh appropriately the pros and cons of discharging the patient from the hospital; and (3) in the event that the decision was made to discharge the patient, devise a reasonable post-discharge treatment plan.

Dr. Cavanaugh is board-certified in general psychiatry by the American Board of Psychiatry and Neurology, and in forensic psychiatry by the American Board of Forensic Psychiatry. Dr. Cavanaugh's qualifications were undisputed; much of his cross-examination by counsel for the defendants focused on the very substantial amount that he charged for his time. Defense counsel did not question Dr. Cavanaugh regarding whether the standard that he described was, as he asserted, a national standard (as opposed to his personal opinion).

In this case, as the majority points out, the plaintiff did not claim in the trial court, nor does she assert on appeal, that her husband should have been involuntarily committed or that the doctors were negligent in not opting for involuntary commitment. She does argue, however — and this is her most persuasive contention — that the post-discharge treatment plan was deficient, in that she was effectively placed in charge of her husband without being apprised of the nature of his illness, or, in particular, of his suicidal ideation. She contends that this failure to disclose led inexorably to her husband's death.

As previously noted, the trial judge apparently disbelieved Dr. Cavanaugh's testimony that the standard that he had articulated constituted a "national standard of care" and that the defendants' actions were in breach of that standard. Specifically, the judge wrote that "[u]nless the expert's claims about what is or is not `reasonable' within the medical profession are substantiated by accepted authority, those claims will be deemed meritless." The judge so concluded in spite of the following acknowledgment by counsel for the defendants, in response to an inquiry from the judge:

Yes, Your Honor, I believe that Dr. Cavanaugh referenced a national standard of care and at this time it is not our argument that he did not articulate standards that he believed were the national standard of care.

Further, as this court emphasized in Wilson, 721 A.2d at 599-600, "to paraphrase In re Melton, 597 A.2d 892, 903 (D.C. 1991) (en banc), the proper inquiry is not what the court deems the standard of care to be, but what experts in the relevant discipline reasonably deem it to be." (Internal brackets omitted.)

Notwithstanding their concession in the trial court that Dr. Cavanaugh had articulated a national standard of care, the defendants take a contrary position on appeal. They assert that the standard of care described in the testimony of Dr. Cavanaugh conflicts with the District of Columbia Hospitalization of the Mentally Ill Act, D.C. Code §§ 21-521 et seq. (2001), popularly known as the Ervin Act. This is incorrect.

The Ervin Act was enacted for the purpose of enhancing the rights of the mentally ill. See, e.g., Dixon v. Jacobs, 138 U.S.App.D.C. 319, 325, 427 F.2d 589, 595 (1970). It protects a mentally ill patient from involuntary commitment unless he is a danger to himself or others. In re Alexander, 336 F.Supp. 1305 (D.D.C. 1972). It also establishes a broad array of procedural rights in mental health proceedings. Under the Act, a patient has a conditional liberty interest in the least restrictive placement suitable for his affliction. In re Stokes, 546 A.2d 356, 361 (D.C. 1988). The philosophy underlying the Ervin Act is therefore an appropriate factor in the physician's calculus in determining, in cases (unlike this one) in which the issue has been raised, whether a patient should be involuntarily committed. See, e.g., Farwell v. Un, 902 F.2d 282, 289-91 (4th Cir. 1990).

This is a far cry indeed, however, from the defendants' remarkable claim that "if appellee GWU met the requirements of the Ervin Act, then there can be no imposition of liability as a matter of law." In their brief, the defendants point to no expert testimony to the effect that the Ervin Act constitutes or defines the applicable standard of care. On the contrary, a patient in Mr. Garby's circumstances was entitled both to the enjoyment of the rights secured by the Ervin Act and to proper rather than negligent medical care, including post-discharge planning. In this case, as we shall see, Mrs. Garby contends that under any reasonable post-discharge treatment plan, the physicians would have been obliged to inform her of her husband's stated plan to kill himself. In particular, Mrs. Garby claims that before she took her husband home — in this case, to a high-rise apartment — the defendants should have disclosed that he had contemplated jumping off a bridge. I do not discern the slightest tension between Mrs. Garby's position on this issue and the requirements of the Ervin Act.

D. Breach.

Having concluded that Mrs. Garby presented evidence of a national standard of care sufficient to warrant its submission to the jury, I turn now to the evidence of breach of that standard.

Dr. Cavanaugh testified that, in his opinion, the physicians had failed to comply with the standard of care (1) in their assessment of the seriousness of the patient's psychotic condition as documented by them in the clinical record, and (2) in their determination regarding whether and when the patient should be released. His most powerful evidence of negligence, however, and the sole claim by Mrs. Garby that requires substantial discussion related to the defendants' failure to devise or execute a reasonable plan to discharge Mr. Garby. In Dr. Cavanaugh's words:

In Dr. Cavanaugh's view, all of the potential diagnoses documented by Dr. Soto indicated that Mr. Garby was in a psychotic condition. Dr. Cavanaugh also described a troubling information gap in the doctors' assessments:

Dr. Akman believes that when Mr. Garby comes into the emergency room he had suicidal ideas about jumping off a bridge. He says it in his deposition. But two hours later he will say when, I don't know the exact time but a short amount of time later, when he talks with Dr. Soto about the case Dr. Soto is now reporting to him as he has to that there are no longer suicidal ideas.

What happened to the suicidal ideas? If a doctor evaluating a potentially suicidal patient can't answer the question, I have a patient either who was suicidal when they came to my emergency room or was suicidal in some period of time before that of some unknown period but recent is the word that Dr. Soto uses. How can I feel comfortable to accept the patient telling me a couple of hours later I don't have any more suicidal ideas unless I have an explanation [of] what's different because we know that suicidal ideas fluctuate a bit.

Dr. Cavanaugh was of the opinion that, at the very least, it would have been prudent to urge the patient to remain in the Emergency Room while further information was gathered from family members and others.

If I was left then with [a situation in which] I had to let him go, there's no other way out of this, I would want to make absolutely certain that the person that he left with would be a person who was fully informed as to the seriousness of his illness. And in particular, knew about the suicidal thinking that he either actually was having when he came into the emergency room that night about jumping off a bridge or had it in the recent past. There's some controversy about that.

And what I would at a minimum want is that this supervisor, who in this case would have been Mrs. Garby, was fully cognizant of the risk that was being assumed, the potential for danger. And that she knew exactly what had been going on with this man in the few days or so before he gets into the emergency room. All of which in this case she didn't know. . . .

(Emphasis added.) Observing that Mrs. Garby "didn't know about [Mr. Garby's contemplated] suicide," or of her husband's plan to jump off a bridge, Dr. Cavanaugh testified that this was "very important information . . . down the line when we're discharging this . . . man, as to whether [Mrs. Garby] was a competent person to take him home." How, the witness asked rhetorically, "can you supervise somebody if you don't know what's wrong with [him]. That's kind of basic." (Emphasis added.)

I entertain no doubt, on the basis of the quoted testimony, that an impartial juror could reasonably find a violation of the standard of care applicable to post-release planning. Further, such a juror could likewise reasonably find Dr. Akman's response to questions by counsel for Mrs. Garby less than persuasive:

Q. After hearing Mrs. Garby testify about her experience in the emergency room about what she knew and didn't know in the emergency room, do you still agree that she was an appropriate observer of her husband?

A. Yes.

Q. Even though she had no idea that he had any suicidal thoughts, right?

A. Right.

Q. And even though she had no idea even about his past medical history, psychiatric medical or family psychiatric medical history?

Michael Garby's family medical history included bipolar disease, major depression and substance abuse.

A. Right.

Q. And even though she was originally told that Michael was — that the Dr. Soto was recommending admission, correct?

A. Right.

Q. And even though Mrs. Garby said to Dr. Soto I don't know what to do. You heard her say that on the stand under oath?

A. I — I recall something like that, those are the exact words.

Q. So it's your opinion, and you're telling this jury that a spouse who says to a doctor I don't know what to do, [s]he's emotional, [s]he's crying, [s]he's being consoled by random people in the emergency room, everything is going to be okay, that person is an appropriate caregiver, appropriate observer and companion in an emergency room situation like Mr. Garby was in?

A. Yes, and I'd like to explain.

Counsel for Mrs. Garby states in his brief to this court that no explanation was provided. Counsel for the defendants has not contradicted this assertion, and my review of the record has not disclosed any explanation by Dr. Akman. Dr. Akman's trial testimony also differed dramatically from his deposition — on deposition he asserted, but at trial he denied, that Mr. Garby had suicidal ideation at the time he came to the Emergency Room.

The only explanation offered by the defendants in the trial court for not informing Mrs. Garby of her husband's suicidal ideation was that providing this information would have violated Dr. Soto's ethical duty not to disclose matters told to him in confidence by his patient. At trial, Dr. Soto testified that he did not tell Mrs. Garby of her husband's plan to jump off a bridge "because it's doctor-patient confidentiality." Dr. Cavanaugh, expressed the view, on the other hand, that if a physician believes that a patient "could be dangerous to [him]self, suicidal type thing or to somebody else," then "I can break that confidentiality because I also have duty to society for safety and protection." Dr. Cavanaugh's position is consistent with the case law and with other authorities on the subject.

This claim has not been pressed by the defendants in their brief on appeal, and might fairly be treated as waived.

Principle No. IV of the Principles of Medical Ethics of the American Medical Association (AMA), AMERICAN MEDICAL ASSOCIATION, CODE OF MEDICAL ETHICS (2004-2005 ed.), states in pertinent part that a physician shall "safeguard patient confidences and privacy within the constraints of the law." An annotation to Principle No. IV provides that "[p]sychiatrists at times may find it necessary, in order to protect the patient or the community from imminent danger, to reveal confidential information disclosed by the patient." "[A]ny confidentiality between patient and physician is subject to the exceptions . . . where the supervening interests of society or the private interests of the patient intervene." Horne v. Patton, 291 Ala. 701, 287 So.2d 824, 832 (1973) ( citing AMA Principles); accord Hague v. Williams, 37 N.J. 328, 181 A.2d 345, 349 (1962); Saur v. Probes, 190 Mich.App. 636, 476 N.W.2d 496, 499 (1991); see, generally Judy E. Zelin, J.D., Annotation: Physician's Tort Liability for Unauthorized Disclosure of Confidential Information About Patient, 48 A.L.R.4th 668, 708-13 (1986 Supp. 2004).

In MacDonald v. Clinger, 84 A.D.2d 482, 446 N.Y.S.2d 801, 805 (1982), the court stated:

Disclosure of confidential information by a psychiatrist to a spouse will be justified whenever there is a danger to the patient, the spouse or another person; otherwise information should not be disclosed without authorization. Justification or excuse will depend upon a showing of circumstances and competing interests which support the need to disclose.

(Emphasis added.) This standard is, in my view, a fair and reasonable one.

In the present case, the patient had recently contemplated suicide by jumping off a bridge. According to Dr. Cavanaugh, the physicians' entries on the patient's chart revealed signs of a serious psychosis. Nevertheless, Mr. Garby was sent home to his eighth floor apartment in the care of his wife, who knew nothing at all of his suicidal ideation. Assuming that involuntary commitment of Mr. Garby was not necessary so long as the person to whom Mr. Garby was released was reasonably competent to look after him, it was surely dangerous to place the decedent in the care of a woman who had no idea what was wrong with him. At the very least, an impartial jury could reasonably so find; indeed, the question was quintessentially one for the jury. Considerations of confidentiality therefore cannot trump Dr. Cavanaugh's testimony that the defendants' nondisclosure to Mrs. Garby of Mr. Garby's plan to commit suicide by leaping off a bridge contravened the applicable standard of care. Accordingly, as the majority concedes, or at least assumes for purposes of this case, judgment as a matter of law could not properly be granted in favor of the defendants on the basis of insufficient evidence of a breach.

II. PROXIMATE CAUSE

According to the majority, even assuming, arguendo, that Mrs. Garby presented sufficient evidence to go to the jury with respect to the standard of care and its breach, judgment as a matter of law was warranted because the plaintiff's case foundered on the issue of causation. I agree with the majority in part; the only claim with respect to which Ms. Garby presented sufficient evidence of proximate cause was the failure of the doctors to apprise Mrs. Garby that her husband had recently planned to commit suicide by leaping to his death.

Dr. Cavanaugh also faulted the defendants for not attempting to persuade Mr. Garby to remain in the Emergency Room longer and for prescribing Ambien rather than anti-depressant medication. Whatever the merits of these criticisms may be, Mr. and Mrs. Garby remained at the hospital for approximately four hours, and they went home without stopping at an all-night pharmacy near their home in order to obtain the prescribed medicine. There appears to be no substantial evidence that either a slightly longer stay in the Emergency Room or a different prescription would have averted the tragic outcome.

In Psychiatric Inst. of Washington v. Allen, 509 A.2d 619, 624 (D.C. 1986), we set forth the requirements for establishing the requisite causation in a medical malpractice case:

To establish proximate cause, the plaintiff must present evidence from which a reasonable juror could find that there was a direct and substantial causal relationship between the defendant's breach of the standard of care and the plaintiff's injuries and that the injuries were foreseeable.

(Emphasis in original; citations omitted.) The question before us is whether an impartial jury, viewing the record in the light most favorable to Mrs. Garby, could reasonably find that she has satisfied this standard.

The defendants contend, the trial judge held, and my colleagues apparently agree that, as a matter of law, there was insufficient evidence to establish that the defendants' failure to disclose to Mrs. Garby her husband's suicidal ideation proximately caused his death. Although nobody can know for sure what would have happened if the doctors had apprised Mrs. Garby that her husband had been making plans to commit suicide, I am satisfied that an impartial jury, acting reasonably, could properly conclude by a preponderance of the evidence that if disclosure had been made, the suicide probably would not have occurred.

Dr. Cavanaugh was qualified without objection as an expert witness in this case, and his expertise is not in question. In his testimony, Dr. Cavanaugh focused very specifically on the issue of proximate cause:

As a result of those deviations [by Dr. Akman and Dr. Soto from the standard of care], it is . . . my opinion that there is a proxima[te] causal relationship between these deviations and the subsequent, unfortunate death by suicide of Mr. Garby.

Dr. Cavanaugh explained that the person who was asked to look after Mr. Garby following his release from the Emergency Room — here Mrs. Garby — had not been apprised of his suicidal thinking about jumping off a bridge; that the failure to inform her was unreasonable; and that this "substandard care . . . in the discharge decision-making process and in the post-discharge planning process by both Dr. Akman and Dr. Soto" was "proximately, causally related to [Mr. Garby's] subsequent death by suicide." The evaluation of testimony, including expert testimony, and the credibility of witnesses, is a function confided to the jury. To sustain a judgment as a matter of law or a directed verdict for the defense, this court would have to conclude that no reasonable jury could credit Dr. Cavanaugh's testimony or agree with his expert opinion. In the absence of any effective impeachment of Dr. Cavanaugh's credibility, and viewing the record, as we must, in the light most favorable to Mrs. Garby, I do not believe that a directed verdict is supportable.

It is true that Dr. Cavanaugh was unable to describe exactly what Mrs. Garby would have done to prevent her husband from committing suicide if she had known what the physicians knew. See his testimony quoted ante, maj. op. at 518. Plainly, she could not have physically overpowered him. But the witness' inability to proffer specifics — to predict exactly what would have happened if — does not mean that Mrs. Garby, knowing that her husband had recently planned to end his life by jumping off a bridge, would nevertheless have hurried with him to her eighth floor apartment and then left him to his own devices while she took a shower, so that he would have a convenient and unimpeded opportunity to execute his planned leap, albeit from a balcony rather than from a bridge. Such a supposition defies common sense, and a reasonable juror would not be likely to find, and a fortiori would not be required to find, that if Mrs. Garby had been provided with the relevant information, events would nevertheless have turned out just as they did. Judgment as a matter of law is proper only in extreme cases, in which no reasonable person, viewing the evidence in the light most favorable to the party opposing the motion, could reach a verdict for that party. District of Columbia v. Cooper, 445 A.2d 652, 655 (D.C. 1982) (en banc); Wilson, 721 A.2d at 596. This is not such a case.

In their brief on appeal, the defendants ask rhetorically:

If Mrs. Garby had been given additional information about the danger she was assuming, presumably about her husband's thoughts of suicide and of jumping off a bridge, how would that have prevented his suicide on Sunday or at some future time? Would she have made a convincing argument to Mr. Garby that evening to persuade him to sign himself in as an inpatient? If she did that, would Mr. Garby have changed his mind and agreed to admission, or would he have felt betrayed?

The argument implied by this question is an ironic one, for at trial, the defense objected strenuously (and successfully) to any testimony regarding what Mrs. Garby would have done if she had been provided with the information, on the grounds that such evidence would be speculative. Having succeeded in preventing Mrs. Garby from explaining how things would have developed differently if she had been provided with information about her husband's suicidal ideation, the defendants now claim, in effect, that they should prevail because no such testimony was presented.

The questioning by Mrs. Garby's attorney proceeded as follows:

Q. Pirjo, if the doctors had, if anybody had told you that Michael had thoughts of jumping off a bridge would you have allowed Michael to go home?

A. No

[COUNSEL FOR DEFENDANTS]: Objection.

THE COURT: I sustain that objection and strike that answer. Disregard the answer.

Q. Mrs. Garby, if you had been told that Michael had thoughts of jumping off a bridge would you have — what would you have told the doctors?

[COUNSEL FOR DEFENDANTS]: Objection.

THE COURT: Sustained.

The judge explained his ruling at the bench, and he stated, inter alia:

You can certainly argue to the jury that if she had had all the facts, she would have had reasonable alternatives but that doesn't make the case. I mean that, that goes frankly goes to the sympathy, passion and prejudice. Objection is sustained.

In my view, counsel's question was a perfectly legitimate one, relevant to, and probative of, the existence vel non of proximate cause. The plaintiff has not argued, however, that this evidentiary ruling warrants reversal.

Even so, the defendants' rhetorical question — "would Mr. Garby have changed his mind?" — effectively concedes the possibility that the decedent would have done just that. If such a possibility existed — which, obviously, it did — and if, viewing the record in the light most favorable to Mrs. Garby, the jury could reasonably have found that her husband would have changed his mind and agreed to stay at the hospital, then proximate cause would arguably have been established on the basis of the defendants' own implicit concession, without more.

Contrary to the majority's apparent perception, the fact that the defendants posed this possibility in their brief cannot be reconciled with the notion that no such possibility existed. The defendants surely would not spend their time addressing non-possibilities.

But, of course, there was much more. As we shall see, the acknowledged possibility that Mrs. Garby would have persuaded her husband to change his mind was only one of several potential outcomes, other than suicide, each of which the jury could reasonably have found to be quite plausible individually and to be quite probable cumulatively. See note 17, infra.

Moreover, notwithstanding the judge's ruling sustaining the defense objection, evidence was adduced during cross-examination from which at least a part of the answer to the question disallowed by the trial judge may readily be inferred. Counsel for the defendants interrogated Mrs. Garby regarding what she would have done during the days preceding her husband's suicide if she had known that he had been planning to jump off a bridge. Mrs. Garby testified that she would have taken "positive action to convey [these facts] to [her husband's] sisters, the detective and the nurse." She would also have contacted his doctors. This testimony, elicited by the defense, revealed that Mrs. Garby would not have remained passive in the face of the danger to her husband, but would have taken what opposing counsel called "affirmative action" to try to prevent disaster. There is no reason to suppose that she would have done any less if the information had been disclosed to her in the Emergency Room.

As the majority points out, the detective lived in Connecticut. The record shows, however, that Mr. Garby leaned heavily on her for advice, and even telephone contact on an urgent basis during the night of the suicide might well have prevented the tragedy.

Even before taking other "affirmative action" to protect her husband from his suicidal plans, Mrs. Garby could have — and doubtless would have — attempted to persuade him to remain at the hospital voluntarily and, as the defendants effectively concede, he may have agreed to do just that. If he had not agreed to remain, he may (or may not) have reacted in a manner that would have changed the defendants' appraisal of the extent of his dangerousness to himself. To be sure, each of these individual possibilities, viewed in isolation, is only a possibility, in the sense that we cannot know for sure whether it would have occurred. Nevertheless, in my opinion, the notion that, notwithstanding disclosure to Mrs. Garby of her husband's recent plan to jump off a bridge, events would have transpired in the same manner as they did is not only speculative; in my view, the likelihood of such an outcome is surely minimal. Thus, contrary to the majority's assertion, the possibility that one of the scenarios other than suicide would have played out is not speculative at all; indeed, the jury might well find it highly probable. The enhanced sense of urgency that would have existed if Mrs. Garby had known of her husband's suicidal planning would surely have generated a dramatically different scenario with dramatically different consequences. At the very least, a finding to that effect by the jury, viewing the record in the light most favorable to the plaintiff, would not have been unreasonable.

If the defendants had disclosed to Mrs. Garby that her husband had recently planned to jump off a bridge, and if upon learning this information Mrs. Garby had declined to accept responsibility for him, then the doctors' post-discharge plan would obviously have been in tatters, and the situation at the time of Mr. Garby's proposed release from the hospital would thus have been materially different from the one which the defendants had previously expected to arise. There are a number of quite plausible scenarios, other than suicide, that might have ensued, including (but surely not limited to) the following:

1. Mrs. Garby would have withdrawn her agreement to the discharge of her husband, and she would have persuaded Mr. Garby to remain voluntarily in the hospital;

2. Mrs. Garby would have refused to take her husband home, and the doctors, recognizing that their post-discharge plan could no longer be carried out, would have persuaded Mr. Garby to remain voluntarily at the hospital;

3. in light of the change of circumstances, the collapse of the doctors' post-discharge plan, and the lack of any person in whose custody Mr. Garby could prudently be released, the doctors would have reconsidered the involuntary commitment option.

In my opinion, the jury could properly find (and probably would have found) that one of these three outcomes would have become reality, and that if Mrs. Garby had been fully informed, Mr. Garby would not have left the hospital on the evening in question. However,

4. in the unlikely event that none of the foregoing possibilities materialized, Mrs. Garby would have contacted friends on an emergency basis to stay with and watch her husband for the rest of the weekend; and

5. at the very least, knowing of Mr. Garby's plan to jump off a bridge, Mrs. Garby would not have taken him to their eighth floor apartment, which has a balcony, and then taken a shower, leaving him free to leap to his death.

Obviously, we cannot be certain, in the event that the decedent's suicidal ideation had been disclosed to his wife, that any affirmative efforts on Mrs. Garby's part to save her husband's life, either at the hospital or following his release, would have been successful. But certainty is not required. As we stated in Psychiatric Inst. of Washington, 509 A.2d at 624,

the law does not require the expert to testify that he or she is personally certain that the plaintiff would not have sustained the injuries but for the defendant's negligence. "The fact of causation is incapable of mathematical proof, since no one can say with absolute certainty what would have occurred if the defendant had acted otherwise." W. Prosser W. Keeton, The Law of Torts § 41, at 269-270 (5th ed. 1984). The expert need only state an opinion, based on a reasonable degree of medical certainty, that defendant's negligence is more likely than anything else to have been the cause (or a cause) of the plaintiff's injuries. See Fitzgerald v. Manning, 679 F.2d 341, 351 (4th Cir. 1982); W. Prosser W. Keeton, supra at 269.

Dr. Cavanaugh's opinion conformed precisely to the requirements of the foregoing passage.

As I have noted, there are several scenarios other than suicide that could have become reality if the doctors had disclosed to Mrs. Garby what her husband had told them about his suicidal ideation. If the cumulative probability of all of these scenarios exceeded 50% — or more precisely, if the jury, drawing all reasonable inferences in Mrs. Garby's favor, could rationally find that their cumulative probability was over 50% — then there was no basis for granting judgment to the defendants as a matter of law. Surely, such a finding by the jury would not have been unreasonable. In this case, the least likely scenario is that knowing all the facts, Mrs. Garby would have gone home with her husband, without summoning help, to a high-rise apartment with a balcony that might readily serve as the bridge from which Mr. Garby had been planning to jump. Does the majority seriously suggest that Mrs. Garby would then have serenely taken a shower, while the husband whom she knew to be suicidally-oriented leaped to his death? A verdict could not properly be directed against Mrs. Garby on the theory that such an improbable sequence of events would actually have happened, or on the supposition that the plaintiff could not and would not have done anything to save Mr. Garby's life. In sum, this case lacks the stuff of which judgments as a matter of law are made, and, in my opinion, it is a grave injustice to Mrs. Garby to deny her the right to have what I consider to be a very substantial case decided by a jury.

The defendants cite District of Columbia v. Peters, 527 A.2d 1269, 1275 (D.C. 1987) and Washington Metro. Area Transit Auth. v. Johnson, 726 A.2d 172, 177 (D.C. 1999) (en banc), for the proposition that suicide is an intentional and deliberate act, and thus a superseding cause for which the defendants are not legally responsible. These decisions, however, differ in dispositive respects from the present case. In Peters, the decedent was shot and paralyzed by a police officer, and he then killed himself. In Johnson, the decedent deliberately jumped to her death in front of a moving metro train, and the plaintiff then brought suit against WMATA for alleged negligence which enabled the decedent to accomplish her goal — volenti non fit injuria. Neither case involved a situation in which the decedent had come to the defendant for treatment.
In the present case, on the other hand, Mr. Garby went to the Hospital because he had pre-existing suicidal ideation. He asked the physicians, in effect, to treat his illness so that he would not commit suicide. Unlike the decedents in Peters and Johnson, Mr. Garby placed his trust in the defendants, whose very function it was to help him to overcome his problem, and who allegedly failed to do so as a result of their professional negligence. I do not believe that the ratio decidendi of Peters and Johnson has any application to the present set of facts. See also Phillips, 714 A.2d at 774-76 (reversing dismissal of a wrongful death and survival action involving a prisoner who committed suicide).

The majority opinion is rather free with its characterizations of my opposing viewpoint as, e.g.: "pure conjecture," "further compound[ed] speculation," and the like. In the majority's penultimate flourish, my colleagues seem to suggest that I am engaged in wishful thinking. All of these comments are based on the premise — surely a fallacious one — that each possible scenario other than suicide may properly be assessed separately and in a vacuum, and then dismissed as "speculative." To the contrary, at least in my view, the correct inquiry is whether — considering the cumulative probability of all of these possible scenarios — an impartial jury, assessing the record in the light most favorable to Mrs. Garby, could fairly find by a preponderance of the evidence that Mr. Garby would not have jumped to his death from the eighth floor balcony. To say that such a potential finding by the jury would be impermissibly speculative is, in my judgment, unrealistic and contrary to the "light most favorable" standard.

According to the majority, ante, page 516,

informing Mrs. Garby of her husband's "true condition". . . would have meant telling her that the doctors did not believe he was presently dangerous enough to himself to need hospitalization, and that information doubtless would have influenced how forcefully, if at all, she sought to have him remain there voluntarily.

Here, in my judgment, my colleagues are "holding the line against creeping practicality" and being less than realistic regarding what would probably have occurred.

Let us envisage how the scenario envisaged by the majority would be likely to play out. Dr. Soto and his colleagues would presumably tell Mrs. Garby that, in the very recent past, her husband had planned to commit suicide by jumping off a bridge. Then the physicians would add something like: "Don't worry, Mrs. Garby, he is not a danger to himself, you can take him home." A short time earlier, Dr. Soto had recommended, and Mrs. Garby had agreed, that Mr. Garby should be kept at the hospital. Now, suddenly, according to my colleagues, Mrs. Garby would probably have believed that there was nothing to worry about and that her husband's suicidal plans could safely be disregarded.

See also note 7 of this opinion, supra, quoting Dr. Cavanaugh on a related matter.

Surely this perception of what would have happened is less than plausible. It is far more likely that, upon learning of her husband's suicidal plans, Mrs. Garby would have been unpersuaded by the notion that her husband was not a danger to himself. Indeed, to an impartial juror, Mrs. Garby's probable reaction would have been one of incredulity, along the following lines: "My husband has been acting irrationally and in paranoid fashion for several days, he has planned to jump off a bridge, we came here for help, the doctor wanted to keep him in the hospital, I wanted to keep him here, and now you tell me that there is no danger!" At least as I see it, any reassurance by the doctors that Mr. Garby was not a danger to himself would be remarkably unreassuring, if not one of the least reassuring reassurances in reassurance history. To suggest that any jury could possibly believe that Mrs. Garby would have felt sufficiently reassured to react in the same way as she did without the information is remarkable enough. To argue that an impartial jury, viewing the record in the light most favorable to Mrs. Garby, was required to accept such a claim (and the other claims related to it) goes well beyond "remarkable."

III. WAIVER

The majority's fundamental position appears to be that, by failing to assert that involuntary commitment was required, the plaintiff has conceded that Mr. Garby was not foreseeably a danger to himself. My colleagues treat the decision by Mrs. Garby's counsel not to make this argument as equivalent to a stipulation of non-dangerousness — the defendants thought it appropriate to release the decedent, the plaintiff does not contest this decision, and so, according to the majority, everyone agrees that there was no foreseeable danger of suicide. Even though several physicians noted in the hospital record that Mr. Garby had been planning to jump off a bridge, and notwithstanding the fact that his paranoid behavior for several days had led his wife to bring him to the Emergency Room, the majority apparently maintains that all of this evidence is trumped by plaintiff's counsel's choice of legal theories. Under the majority's analysis, all of the evidence of Mr. Garby's suicidal ideation and dangerousness to himself — evidence that led both Dr. Soto and Mrs. Garby to try to persuade the decedent to stay in the hospital — disappears from the calculus because, according to my colleagues, the parties have effectively stipulated that there was no foreseeable danger. There is something wrong — alarmingly wrong — with this picture. A man who very recently has been planning suicide by jumping off a bridge is not non-dangerous, and his attorney's litigation strategy (absent a dispositive stipulation or concession) does not make him so.

According to the majority, the failure to argue that the decedent ought to have been involuntarily committed was not a tactical decision at all, but was forced upon counsel by lack of evidence. Whether or not the decision was tactical or not is not necessarily of decisive importance, but the majority's position cannot be reconciled with the record.
Dr. Cavanaugh did not testify that there was no basis for involuntary commitment; rather, he was never asked. Indeed, the passage from his testimony quoted on page 27 of this opinion strongly suggests that he was assuming, for the sake of argument only, and with considerable reluctance, that Mr. Garby had to be released. This is apparent from his phrasing: "If . . . I . . . had to let him go, [if] there's no other way out of this." Dr. Cavanaugh testified unequivocally that the defendants were negligent in releasing Mr. Garby into Mrs. Garby's care and custody without disclosing to Mrs. Garby her husband's suicidal plans, that this was a dangerous course of action, and that this negligence proximately caused the decedent's death. The strategy not to argue that involuntary commitment was called for was counsel's, and it was not based on any reluctance on the part of the expert to testify that the decedent was a danger to himself.

As I have previously noted, the majority's approach to the issue of causation was introduced into the controversy by my colleagues. The defendants did not even hint at the majority's theory in their twelve-page briefing of the issue of proximate cause. The connection likewise never occurred to the trial judge. This, of course, is not dispositive, for at least in the absence of procedural unfairness, In re Walker, 856 A.2d 579, 586 (D.C. 2004) (per curiam); 1137 19th St. Assocs. v. District of Columbia, 769 A.2d 155, 161 (D.C. 2001), the trial judge's decision must be affirmed if the result is correct, even if the appellate court bases its decision on a different ground from that relied upon by the trial court or urged upon the court by the parties. See, e.g., Marinopoliski v. Irish, 445 A.2d 339, 340 (D.C. 1982). In this case, however, the majority's theory was not argued by the defendants, and the plaintiff never had an opportunity to respond to it. I therefore have reservations about the procedural fairness of relying on that theory now, even though the curious tactics of Mrs. Garby's counsel with respect to this aspect of the case make a claim of procedural irregularity a hard sell.

The defendants in this case were represented by experienced attorneys. The failure of defense counsel to claim a waiver of the kind perceived by the majority itself suggests that the claim of waiver is not an especially persuasive one. Cf. Hunter v. United States, 606 A.2d 139, 145 (D.C.), cert. denied, 506 U.S. 991, 113 S.Ct. 509, 121 L.Ed.2d 444 (1992) (failure of trial counsel to perceive or react to alleged prejudice "is itself suggestive in some measure of lack of prejudice").

In all candor, I must acknowledge doubt that the plaintiff would have responded to this theory even if it had been asserted by the defendants. In fact, the attorneys for the plaintiff failed to address proximate cause at all in their opening brief on appeal, and a respectable argument can be made for the proposition that the plaintiff waived the issue. The majority does not ground its decision on such a waiver, however, and although it is a close call, I would not treat counsel's failure to argue the point as forfeiting it, for there has been no discernible prejudice to the defendants.

On the merits, I do not agree that the plaintiff's failure to argue that Mr. Garby should have been involuntarily committed should be treated as a concession of non-dangerousness. This interpretation of the plaintiff's litigation position is surely the least favorable one possible from the perspective of the plaintiff, and the record must be assessed in the light most favorable to the plaintiff. It is certainly not the only interpretation. Attorneys decline to make apparently available arguments for a host of reasons, mostly tactical. Presenting one's client's position as moderate and restrained, as one that eschews extremes, is a common barristerial strategy. In a case such as the present one, counsel might well wish to take a position which avoids potential conflict with the Ervin Act and with the protection of the civil liberties of the mentally ill, and might prefer to cast the case in terms of negligence in carrying out a plan not involving involuntary commitment. I suspect that this is what occurred here. But few if any attorneys would anticipate, if they framed their position in this way, that they would thereby be waiving the right to argue that Mr. Garby was likely to carry out his intention to leap to his death, and that such a waiver, however unintentional, would cost them the case even if they could prove that the post-discharge plan for his care was negligently devised and executed and that the decedent died as a consequence of this negligence. After all, any competent attorney would know of the obvious danger of suicide reflected in the physicians' own entries in the hospital record. In my opinion, no reasonable practitioner would expect that, as a result of such a perceived waiver, all of the evidence of dangerousness would suddenly become more or less worthless. If the defendants do not claim that this election on the part of the plaintiff's counsel constituted a dispositive concession, surely this court ought not to rule that it did.

Assuming, arguendo (as all parties appear to assume) that involuntary commitment of Mr. Garby was not necessary because measures short of involuntary commitment — e.g., a prudent plan for post-discharge treatment — would be sufficient to protect him, it does not follow that the decedent would not pose a danger to himself if that plan was negligently devised and executed. The importance of sound post-discharge planning, and the defendants' failure in that regard, constituted the gravamen of Dr. Cavanaugh's expert testimony. Based on Dr. Cavanaugh's evidence, the jury might reasonably find, even if involuntary commitment was not mandated, that the doctors were negligent in placing Mr. Garby in the custody of a woman who did not know the critical facts, and that this negligence — not the failure to order involuntary commitment — proximately caused the suicide. I discern nothing unreasonable about such a conclusion.

In sum, the failure to argue that involuntary commitment was appropriate has little to do with the principal theory of negligence and causation on which counsel for Mrs. Garby relied. Dr. Cavanaugh's testimony, and counsel's presentation, focused, as we have seen, on the negligent formulation and implementation of a post-discharge plan. Such a plan could only come into play if Mr. Garby was not involuntarily committed. Fundamentally, arguments about involuntary commitment have nothing to do with the principal issue before us. Dr. Cavanaugh testified, in effect, that the defendants' planning for the period immediately after Mr. Garby left the hospital — i.e., placing the decedent in the custody of his wife, when the wife knew nothing about his suicidal ideation — violated the national standard of care and proximately caused the decedent's suicide. This was powerful testimony, sufficient in my view, for the case to go to the jury. I cannot agree with the majority's apparent position that counsel waived the claim of dangerousness or conceded away Mrs. Garby's case.

IV. CONCLUSION

For the foregoing reasons, I would reverse the judgment and remand the case for a new trial.

Garby v. George Washington University Hosp

Summary

Opinion

I.

II.

I. THE STANDARD OF CARE AND ITS BREACH

II. PROXIMATE CAUSE

III. WAIVER

IV. CONCLUSION

Garby v. George Washington University Hosp

Garby v. George Washington University Hosp

Case Details

Citations

Citing Cases

Garby v. George Washington University Hosp

Summary

Opinion

I.

II.

I. THE STANDARD OF CARE AND ITS BREACH

II. PROXIMATE CAUSE

III. WAIVER

IV. CONCLUSION

Garby v. George Washington University Hosp

Garby v. George Washington University Hosp

Case Details

CitationsCopy Citation

Citing Cases

Citations