dissenting.
I respectfully dissent. As Judge Andrews explains in his dissent, a plaintiff in a case such as this must prove (1) a legal duty, (2) a breach of that duty, i.e., negligence, (3) a “legally attributable causal connection” between the breach and the resulting injury, and (4) loss or damage as a result of the breach of the duty. Analysis of all lawsuits alleging negligence, and especially analysis of psychiatric malpractice cases, would be easier and more just if, when discussing item (3) above, we carefully distinguish between cause-in-fact and proximate cause.10
As explained by our Supreme Court in a scholarly opinion,
a holding that a defendant’s conduct is not the proximate cause of the plaintiff’s injury does not constitute a determination that the defendant’s conduct is not a cause in fact of the plaintiff’s injury, but rather is in the nature of a policy decision by the court that, for a variety of reasons . . . , the defendant’s conduct and the plaintiff’s injury are too remote for the law to countenance a recovery.11
The scattershot affidavit of Dr. Reid, supplemented and modified by his deposition and that of Dr. Alpern, asserts that Dr. Peterson breached his professional duty of care in several ways. For example, Dr. Alpern asserts negligence by Dr. Peterson in not ordering the involuntary commitment of Mrs. Reeves, or not assessing her properly, or not being present when she was discharged from Horizons.
Of course, a physician owes a duty to his or her patient, indeed, a high duty. But he is not an insurer of the patient’s health in all circumstances.
As argued in Judge Andrews’s dissent, the circumstances in which a duty exists is a matter of law, not an issue of fact for a jury. *382Where the General Assembly has spoken about duty, our task is merely ministerial. But when statutes are silent on the issue, we must nonetheless, like all common law judges, decide the dispute in front of us. Doing so may oblige us to say whether a duty exists or not in the circumstances presented. And our decision is precedent until and unless changed by the representatives of the people.12
I would be reluctant to create, or, to use a less argumentative term, to ratify, several of the duties argued for in the majority and specially concurring opinions. Those opinions and the plaintiffs experts cite as negligent Peterson’s failure on August 26, 2005, to perform a “suicide or self-injury assessment” and an “adequate psychiatric assessment.” We are not informed as to the time required for these assessments, and hence we do not know how much time would be added to the usual time spent by a psychiatrist with a hospitalized patient with whom he is already familiar. Dr. Peterson did not know on August 26, 2005, that he would not see Mrs. Reeves again at Horizons. Imposing a duty on a psychiatrist to prevent self-harm by a patient five days after the physician last sees her by means of administering these additional assessments would mean that every time a psychiatrist sees any patient with a history of suicidal attempts or thoughts, he must again administer those assessments. Every psychiatrist-patient encounter might be the last encounter before the five-day event.
Similarly, I would not create nor find that Georgia law already implies a duty incumbent on a psychiatrist to prevent a patient from harming herself two days after discharge by being available for consultation at discharge or having another psychiatrist be available. Having a psychiatrist be available for consultation would obviously increase the costs of community mental health facilities such as Horizons, or the cost to the patient’s insurance company, if any. More importantly, it would make the psychiatrist unavailable to other patients while he was doing discharge consultations. Such involuntary rearrangements of the practice of the profession of psychiatry would be better left to a more-fully informed legislature.
One allegation of a plaintiff’s expert might invoke a duty which does already exist. Dr. Alpern asserted that Dr. Peterson breached a duty of care by “failing to stabilize Reeves in a proper medication regimen.” The implication is that this failure breached a duty to prevent a patient from harming herself five days later. Dr. Peterson’s *383notes of the August 26 visit say “meds will take a while.” Whether we find Dr. Alpern’s testimony plausible is, of course, not relevant to our decision. Unlike the alleged duty to do things he did not do, e.g., be available for a discharge consultation, we know that Dr. Peterson did attempt to adjust Reeves’s medications. Whether any medication would prevent attempted suicide five days later is a dispute among experts.
Taking Alpern’s testimony as correct, as we must, Peterson’s alleged failure would be a cause-in-fact of Mrs. Reeves’s injuries. Georgia uses the conservative “but for” test to determine cause-in-fact. Under that analysis, the horrific injuries of the plaintiff would not have occurred five days later but for the negligence of the defendant and, of course, the negligence of others.13
But was the alleged negligence the proximate cause? Proximate cause is a policy decision, to be made by the jury except in plain cases, as to whether it is reasonable to impose liability for resulting injury, injury clearly “caused,” by the defendant, when the connection between the negligence and the injury is attenuated.14 Here, the injury may have been foreseeable, but it occurred five days later, and Peterson did not know that Reeves would leave Horizons in the interval. Under those circumstances, I would find that Peterson’s negligence was not the proximate cause of the injuries.
I would reverse.
See generally C. Mikell, Jury Instructions and Proximate Cause: An Uncertain Trumpet in Georgia, 27 Ga. St. Bar J. 60 (1990); W. Prosser, The Law of Torts, 244 et seq. (4th Ed. 1971); L. Green, Causal Relationship in Negligence Law, 60 Mich. L. Rev. 543 (1962).
McAuley v. Wills, 251 Ga. 3, 7 (5) (303 SE2d 258) (1983), cited with approval in Atlanta Obstetrics &c. v. Coleman, 260 Ga. 569 (398 SE2d 16) (1990). But see Justice Weltner’s special concurrence in Coleman, supra at 574 (4) (b) (“McAuley v. Wills . . . should be disapproved”).
Or overruled by the Supreme Court of Georgia. The legislature can make an informed decision after committees hear from the concerned profession, from consumer advocates, etc. Our decisions could perhaps he called uninformed. We have only the information furnished us by the parties.
King v. Zakaria, 280 Ga. App. 570, 576 (4) (b) (634 SE2d 444) (2006) (in medical malpractice action, plaintiff must show that purported violation was both cause-in-fact and proximate cause of injury, which would not have occurred but for violation); see also Black v. Ga. Southern & Fla. R. Co., 202 Ga. App. 805, 807 (1) (415 SE2d 705) (1992) (malfunctioning signal may have been cause-in-fact of accident under theory that but for conditions created by malfunction, accident would not have occurred, but this does not mean railway’s negligence was proximate cause of accident as there may have been intervening acts of third parties). See generally Gen. Motors Corp. v. Davis, 141 Ga. App. 495, 496 (1) (233 SE2d 825) (1977).
McAuley, supra. The controversial decision in Palsgraf v. Long Island R. Co., 248 N.Y. 339 (162 NE 99) (1928), can be explained only as a finding that there was no proximate cause (or that the guards were not negligent in the first place). Clearly, Mrs. Palsgraf was within the “ambit” of those to whom a duty was owed. She had already bought a ticket, so she was a passenger. The railroad was a common carrier and owed her a duty of extraordinary care.