This action arose out of the suicide of plaintiff's husband while a patient at the Veterans Administration Hospital in West Haven, Connecticut.1 The only question on appeal is the validity of the district court's finding that the hospital officials and staff were negligent in permitting Mr. Dinnerstein's death. Both parties agree that under the Tort Claims Act, Connecticut law applies, and that under that law the proper standard of care is that standard which other hospitals in the area have and afford similar patients. The parties disagree, however, as to whether the evidence presented at trial-most notably the expert psychiatric testimony-sufficiently supports the court's finding of negligence. Since we find the evidence to be sufficient, we affirm.

For as long as six years before his death on January 4, 1969, Howard Dinnerstein had suffered recurrent periods of depression serious enough to require voluntary psychiatric care. On January 31, 1968, he ran his car into a highway bridge abutment. Though little damage was done to the car and none to himself, Dinnerstein told his wife that the "accident" had actually been a suicide attempt. The same day he was admitted -again voluntarily-to the V.A. Hospital in West Haven; he remained there under observation until February 12. The hospital records indicate that at the time its staff doctors considered Dinnerstein a "suicide risk." Though the records also indicate that Dinnerstein was gradually accorded greater freedom of movement-from complete restriction on January 31 to full privileges on February 2-his release on February 12 was nevertheless "against medical advice."

In the fall of 1968, business pressures apparently drove Dinnerstein back into his depression. Dr. Gottlieb, Dinnerstein's new private psychiatrist, convinced him to admit himself to St. Raphael's Hospital in New Haven on November 5, 1968. Dr. Gottlieb testified that his reasons for hospitalizing Dinnerstein in November included protecting him from possible suicide. On November 9, Dinnerstein's condition took a turn for the worse: He would not leave his bed nor do anything for himself. However, by November 14 he had improved enough to be granted walking privileges; and on November 28, he was again discharged.

But in December the cycle again turned downward, and Dinnerstein was treated by Dr. Gottlieb no less than seven times. As he was going steadily downhill, Dr. Gottlieb again recommended hospitalization. At the V.A. Hospital, the physician on duty, Dr. Talan, informed Dinnerstein that the individual therapy he was seeking was not available. Depressed further by this, Dinnerstein refused to admit himself. However, on the following day, January 3, 1969, he decided to accept the V.A. offer of group therapy. During the admission session, Dr. Clark of the V.A. staff diagnosed Dinnerstein's condition as a "Depressive Reaction." However, he later testified that he did not consider Dinnerstein to be suicidal.

Though Dr. Clark was aware of Dinnerstein's prior history from both the V.A. records of his February hospitalization and from a telephone conversation between Dr. Talan and Dr. Gottlieb on January 2, he did not order any restrictions or special supervision for Dinnerstein in assigning him to a seventhfloor ward. Dr. Clark also discontinued a regime of anti-depressant drugs Dinnerstein had followed up to that time in order to obtain a more accurate reading of his psychological state.

The V.A. records indicate that on the evening of January 3, Dinnerstein complained to Dr. Clark that he had become even more depressed since arriving at the hospital because of the inadequacy of the group therapy program. However, despite these complaints, Dr. Clark issued no additional orders either to supervise Dinnerstein more closely or to give him anti-depressant medication. At approximately 3:00 p. m. on the following day, January 4, Dinnerstein walked to the lavatory on the seventh floor and leaped to his death from an unsecured window.

The government's first contention is that as a matter of Connecticut law, the district court's finding of negligence cannot stand because no psychiatric expert testified that the appropriate standard of medical care was not satisfied. This argument misstates the law. Rather it is clear that expert medical testimony is necessary only to provide the trier of fact with the proper standard of medical care against which to measure the defendant's actions. It is the trier of fact, and not the expert witness, who must then evaluate the evidence to determine whether or not that standard was actually met. Slimak v. Foster, 106 Conn. 366, 370-71, 138 A. 153, 155 (1927). Indeed, the trier of fact may even base a finding of negligence on the expert testimony of the defendant-doctor. Snyder v. Pantaleo, 143 Conn. 290, 294, 122 A.2d 21, 24 (1956); Allen v. Giuliano, 144 Conn. 573, 575, 135 A.2d 904, 906 (1957); Console v. Nickou, 156 Conn. 268, 274, 240 A.2d 895, 898 (1968); Levett v. Etkind, 158 Conn. 567, 575, 265 A.2d 70, 73 (1969).

Here there was clear testimony from three psychiatric experts-Dr. Gottlieb, Dr. Talan, and Dr. Rubenstein, a government witness-that the hospital should have restricted Dinnerstein if there was reason to believe he was suicidal. Indeed, this standard of care is so obvious that expert testimony seems hardly necessary.

Nevertheless, when one remembers the fact that unlike blood pressures or pulse rates, emotional states cannot be calibrated with precision, medical expertise again becomes relevant. That is, the question for expert guidance-but still for ultimate determination by the trier of fact-was whether it was reasonably foreseeable on the basis of Dinnerstein's prior history that he would commit suicide. The government correctly notes that all of the expert witnesses, including Dr. Gottlieb, testified that they did not consider Dinnerstein to be suicidal on his admission to the hospital. However, despite this testimony on the narrow question of whether Dinnerstein exhibited suicidal tendencies on the day he was admitted, the trial court-relying primarily on the hospital's own records and the conversations between Dr. Gottlieb and Dr. Talan-nevertheless found:

[T]hat the Hospital knew or should have known of the real possibility of a suicide attempt, that it failed to take appropriate measures to guard Dinnerstein against that danger, that it was its duty to do so, and that, having failed to act with due care in the discharge of that duty, the United States is liable to the plaintiff for the losses caused thereby. . . .

At the least, for the first few days of Dinnerstein's admission his movements should have been restricted so that he could be closely watched. As he was assigned to a ward on the seventh floor, measures should have been taken to see that he could not jump from a window. His own denial upon admission of suicidal ideation and even Dr. Gottlieb's belief that he was not imminently suicidal, cannot excuse the complete absence of precautions to insure the safety of a patient with a suicidal gesture in his past, a long history of psychiatric treatment for recurrent and severe depression, previous hospitalization to protect against possible suicide. . . .

In our review of this finding, the government urges that we reconsider the evidence unrestrained by the "clearly erroneous" test of Rule 52(a).2 However, while we agree that a finding of negligence is reviewable as a matter of law, we cannot accept the government's invitation to conduct a broad, almost de novo, inquiry into the basic facts of this case. As we have previously stated:

The standard of review to be applied to a district court's finding of negligence is not the "clearly erroneous" standard . . . for a finding of negligence is reviewable as a matter of law. Nevertheless, the lower court finding "[will] ordinarily stand unless the [lower] court manifests an incorrect conception of the applicable law." Cleary v. United States Lines Co., 411 F.2d 1009 (2 Cir. June 2, 1969), citing Radovich v. Cunard Steamship Co., 364 F.2d 149, 152 (2 Cir. 1966) and Esso Standard Oil S.A. v. S.S. Gasbras Sul., 387 F.2d 573 (2 Cir.), cert. denied, 391 U.S. 914, 88 S. Ct. 1808, 20 L. Ed. 2d 653 (1968).

Hendry v. United States, 418 F.2d 774, 784 (2d Cir. 1969).

Here there is no doubt that the trial court applied the proper legal standard: If Dinnerstein's suicide was reasonably foreseeable, the hospital staff was clearly negligent in assigning him to an unsupervised and unrestricted room on the seventh floor.

We further hold that even without the benefit of the clearly erroneous rule, the finding that Dinnerstein's suicide was sufficiently foreseeable to demand greater supervision is amply supported by the record. The hospital officials should certainly have been aware of Dinnerstein's prior hospitalization history contained in their own records. Similarly, Dr. Gottlieb, who testified that he considered Dinnerstein's condition in late December to be the same or worse than it was in November when he was hospitalized partially to prevent a suicide attempt, informed Dr. Talan that Dinnerstein had exhibited "negativism and unpredictability" as recently as November 9th. This diagnosis certainly should have been made available to Dr. Clark, the admitting physician.

Moreover, Dr. Clark himself admitted that soon after his admission, Dinnerstein expressed dissatisfaction and further depression with the treatment he was receiving. Yet despite these complaints, Dr. Clark continued to deprive Dinnerstein of anti-depressant drugs he had been receiving, so as to evaluate his psychological condition.

From these facts, when combined with a six-year history of deepening depression, it was not at all unreasonable for the trial court to conclude that Dinnerstein should have been supervised more closely. This is not to suggest that every potential suicide must be locked in a padded cell. The law and modern psychiatry have now both come to the belated conclusion that an overly restrictive environment can be as destructive as an overly permissive one.3 But while we must accept some calculated risks in order to insure the patient's legal rights and provide him with the most effective therapy,4 we must also admit that errors in judgment do occur and that when they do, medical authorities must assume their rightful share of the responsibility.5 Here, in short, we cannot say the trial court erred in concluding that the hospital staff was negligent in allowing Dinnerstein-an emotional unknown quantity-unrestricted movement on the seventh floor.

The judgment is affirmed.