Irene Ducharme, Etc. v. United States

BREYER, Circuit Judge. Leo Ducharme’s wife and his estate sued the United States under the Federal Tort Claims Act, 28 U.S.C. §§ 1346(b), 2671-80 (1982), claiming that negligence at the Veterans Administration Medical Center (“V.A.”) in Manchester, New Hampshire caused Leo Ducharme’s death in late 1979. After a trial (without jury), the district court found for the plaintiffs. The government appeals, arguing that the record evidence does not support those of the court’s findings upon which its liability determination rests. Our detailed review of the record leads us to conclude that the government is correct; and, we must reverse the district court’s judgment. I Background A. The basic facts are as follows. Sometime on December 22, 1979, Leo Duc-harme’s son brought him to the V.A. hospital. Ducharme, 66, complained of chest pains, fatigue, shortness of breath, chills and sweats. The doctors whom he saw that evening thought a heart attack was possible, but, because of a fever, a heart murmur, and certain other symptoms, they concluded it was more likely that Duc-harme had a bacterial heart disease called bacterial endocarditis. Ducharme’s breathing problem and low blood oxygen levels led them to consider as well another possibility, that he had blood clots in his lungs (pulmonary emboli). They put Ducharme in intensive care and began to treat him for bacterial endocarditis. On December 23, Ducharme had trouble breathing; his blood oxygen level was low; his son felt it was difficult to get the staff to pay enough attention to him; the doctors administered oxygen through nasal prongs and they provided other treatment suitable for the bacterial disease. On December 24, Dr. Gillie, the physician in charge, ordered a “ventilation-perfusion” lung scan to see whether there were blood clots in Ducharme’s lungs, preventing oxygen from getting to his blood. Since the V.A. hospital had no lung scan machine, Ducharme had to be taken to the Catholic Medical Center (“C.M. C.”), a step that Dr. Gillie delayed until December 26 in part (he said) because he feared Ducharme was too sick to move. On December 26, the lung scan at C.M.C. revealed probable blood clots. Doctors at the V.A. began treatment with a drug called Heparin, which dissolves blood clots. Because of the severity of his condition, the doctors also decided to transfer Ducharme to a V.A. hospital in West Roxbury, Massachusetts, which was capable of performing a difficult but more immediately effective procedure called urokinase therapy. The therapy involves serious risks, so before giving this therapy, doctors check the heart to be absolutely certain that blood clots are the problem. They perform the check by inserting a small tube (catheter) into the heart through a blood vessel. When they did so, the catheter pierced Ducharme’s heart wall (a well-known risk of the procedure). The doctors immediately withdrew the catheter, hoping (as sometimes happens) that the wall would then seal itself. It did not do so, and Ducharme died. B. The plaintiffs did not criticize the care provided at C.M.C. or at the V.A. hospital in West Roxbury. Rather, they attacked (1) the Manchester V.A.’s delay in diagnosing the blood clots, and (2) the way in which the Manchester V.A.’s doctors administered oxygen. Under their delay theory, they said that, had the Manchester doctors on December 22, 23, or 24 noticed that Ducharme had phlebitis (a vein disease in the legs leading to blood clots that may break off and travel to the lungs), or had *29 they read chest x-rays or electrocardiograms more carefully or administered other tests, they would have diagnosed the lung clots sooner, thereby either avoiding the need for the urokinase treatment, or at least leading to earlier use of urokinase treatment. Earlier use of the treatment may have meant a heart wall that had not been weakened by several days without enough oxygen, a heart wall more likely to have sealed itself once the doctors withdrew the piercing catheter. In any event, they said, the Manchester V.A. doctors should have sent Ducharme for a lung scan sooner, and that could have increased his chances of survival for the reasons just mentioned. Alternatively, under the deprivation-of-oxygen theory, plaintiffs said that, had the Manchester V.A. administered more oxygen through better methods from December 22 to December 26, Ducharme’s heart wall would have been stronger, increasing the likelihood of its resealing. The district court, after hearing from the experts of both sides, made the following findings in respect to the doctors involved in Ducharme’s treatment: The individual care by the VA doctors involved in this case, Doctors Gillie and Folland[,] cannot be impugned as far as reasonable medical care is concerned or relevant. The VA doctors in diagnosing plaintiff’s illness had various options open to them and their actions cannot be considered to be acts, or omissions to act, constituting malpractice. Rec.App. at 561. The court made the following additional findings in respect to negligence: When [Ducharme] ... entered the VA [H]ospital on December 22, 197[9] it was during the holiday season and the hospital was shorthanded as far as personnel was concerned.... The VA was not equipped to conduct a lung scan.... It is the finding of the court that the VA Hospital was negligent in not having sufficient personnel available.... There is also evidence that there was a delay in the performance of the lung profusion [sic] scan. Further, the VA Hospital impliedly, if not expressly, represented that it had the facilities of the average hospital.... Two important medical aids were missing or absent from the Manchester VA Hospital: means of conducting a lung scan and angiogram with urokinase therapy. The lack of the latter in 1979 is understandable, the lack of the former with a two day delay or hiatus is not to be condoned.... The court finds for the plaintiff and further finds that none of the doctors from the VA are chargeable with malpractice. ... [T]he Manchester VA Hospital was negligent and its negligence caused or contributed to cause the death of the plaintiff. Rec.App. at 560-564. II The government notes that New Hampshire law, which governs this case, 28 U.S.C. § 1346(b), requires a plaintiff to show a hospital’s negligence through expert evidence, Martin v. Wentworth-Douglass Hospital, 130 N.H. 134, 136, 536 A.2d 174, 175-76 (1987); Folger v. Corbett, 118 N.H. 737, 738, 394 A.2d 63, 63-64 (1978), unless the claimed negligence is so obvious or nontechnical that it is within the knowledge and experience of an ordinary person. Martin, 130 N.H. at 136, 536 A.2d at 175; April v. Peront, 88 N.H. 309, 311-12, 188 A. 457, 459 (1936). The government says that the expert evidence in this record does not permit the court to conclude that the hospital was negligent in respect to the actions (and omissions) upon which the court based its judgment. We agree. We have found no evidence at all, either expert or nonexpert, that suggests the Manchester V.A. was negligent in failing to have a lung scan on the premises. The record notes that the hospital did not have the machine, but plaintiffs have not directed us to, nor have we found, any testimony or other evidence suggesting that it was *30 unreasonable for the hospital not to have had one. Indeed, plaintiffs did not argue that the hospital was negligent in this respect in the court below. The record does contain evidence suggesting that the hospital was short staffed during the Christmas week. Ducharme’s son testified that, when he requested a diagnosis on December 24, the doctor said, “Look, right now I can’t get ahold of his records and it’s Christmastime. We’re not —we’re shorthanded and I don’t have an answer for you.” Rec.App. at 96. But, the record does not explain how a staff shortage could have been linked to any mistake in the administration of oxygen. Nor does it explain how it could have led to any delay in Ducharme’s treatment. To the contrary, Ducharme’s doctor, Dr. Gillie, testified that he ordered a lung scan on December 24 and consciously decided to delay it until December 26, largely in light of Ducharme’s condition. The doctor testified, “he was acutely ill. It was my opinion that he was still acutely ill and to move him that far ... is ... trouble with somebody who is acutely ill. So between ... how sick he was and the possibility of causing further complications [I decided not to move him].” Rec.App. at 296. With one exception, the availability or nonavailability of staff had nothing to do with his decision. The exception consists of the following testimony: Dr. Gillie: [We did not send Mr. Duc-harme to Catholic Medical Center because] the man was very acutely ill. We were on an emergency status as far as holiday. The people in ... nuclear medicine at CMC were also on a holiday. This acutely ill man could get into problems if you shipped him in an ambulance, and there were some problems over there with just a doctor and whoever went with him and a single technician. Rec.App. at 294. This testimony, however, refers to no special shortage at Manchester V.A.; in fact, it refers to Christmas crews at Catholic Medical Center; and, we do not see how it alone provides sufficient basis for concluding that an (assumedly) negligent failure to maintain adequate staff at the Manchester V.A. brought about any relevant delay. Of course, it is logically possible that a staff shortage could have led Dr. Gillie to order the lung scan on December 24 instead of, say, on December 23 or the night of December 22. But there is not a word of testimony suggesting that is so. Indeed, the facts that Dr. Gillie says he thought of ordering the scan on the 23rd, that Ducharme was very sick, and that the doctors were treating the symptoms based on alternative theories, all suggest that “staff shortage” had nothing to do with Dr. Gillie’s delaying the order from December 22 to December 24 (assuming for the sake of argument that one could characterize that delay as unreasonable.) Conversely, we recognize that there is some (disputed) evidence that the lung scan delay, from December 23 or December 24 to December 26, could have caused a weakening of Ducharme’s heart wall (through inadequate oxygen) and thereby contributed to his death. But, we cannot find any evidence to link this delay to the lack of the lung scan or to the short staffing described above. The record makes clear that Dr. Gillie intended the delay; he could have ordered the scan or made certain the scan took place sooner, but he did not. The court accepted Dr. Gillie’s medical explanation of why he waited two days; it expressly found he was not negligent. And, that being so, it is not possible to find the hospital negligent, for the hospital, in this respect, acted though Dr. Gillie. Thus, we find inadequate support, indeed next to no support, in the record for the findings of the district court in respect to the hospital’s negligent causation of Duc-harme’s death. These findings are “clearly erroneous” and we must set them aside. Ill We see no reason to remand this case for new findings. For one thing, unlike most juries, when a district court finds facts, it spells out the reasons for its findings. Here, its opinion, read in light of the record, makes clear that the doctors were not negligent in any relevant respect. We *31 do not see how our discussion, about staff shortages and the presence of a lung scan machine, could, as a matter of logic, affect any of the district court’s other, unap-pealed findings in respect to medical negligence. See Anderson v. City of Bessemer City, 470 U.S. 564, 575, 105 S.Ct. 1504, 1512, 84 L.Ed.2d 518 (1985) (findings based on testimony and not contradicted by extrinsic evidence, “if not internally inconsistent, can virtually never be clear error”); compare Kitchen v. Chippewa Valley Schools, 825 F.2d 1004, 1013-14 (6th Cir.1987) (only so much of district court findings of liability as are inconsistent with jury verdict reversed, others unaffected), with Emerson Electric Co. v. Farmer, 427 F.2d 1082 (5th Cir.1970) (contradictory Rule 52(a) findings made after Fed.R.Civ.P. 41(b) motion; case reversed and remanded for new, comprehensive findings). For another thing, the record contains strong support for the court’s “no negligence” conclusion. The plaintiffs’ main expert, Dr. Bricker, argued that the V.A. doctors should have recognized phlebitis from Ducharme’s swollen leg, should have investigated pulmonary blood clots sooner in light of the oxygen deficiency, should not have delayed two days to get the lung scan, and should have used more effective oxygenating techniques such as a face mask or respirator. Defendant’s experts, Doctors Akey and Bousvaros, responded, however, that the leg swelling was insignificant (a circumference difference of 1.5 centimeters), that fever and a heart murmur suggested the bacterial infection, that to start treatment for the other, blood clot diagnosis (dissolving the clots) could be dangerous if there was an infection (because it might prevent the body’s normal control of bleeding), that the delay was a ‘judgment call’ by the treating doctor in light of the risks of movement, and that different oxygenating techniques would not likely have made much difference for the problem was not insufficient oxygen in the lungs but clots stopping the blood from getting the oxygen. Defendant’s experts also strongly argued that the delay and arguably inadequate oxygenation almost certainly had nothing to do with Duc-harme’s death, in light of other factors (e.g., chronic hypertension, severe coronary artery disease) that likely weakened his resistance and the fact that insufficient oxygen, making the heart work harder for several days, was not likely to have “thinned” the heart wall. In addition, the district court said that, in judging this disagreement among the experts, it would weigh their qualifications. The plaintiffs’ main expert, comparatively speaking, had little experience in the area of medicine at issue. He was not qualified to practice at any New Hampshire hospital. His applications for hospital staff privileges had apparently been rejected. He had not qualified for his medical boards in his fields of study (though he eventually qualified in “legal medicine”). He had become a doctor by completing medical school; but he had not completed a residency or a “preceptorship” (an apprenticeship to a particular doctor). He made up to half his living testifying as an expert in malpractice cases, which he had done 400 to 500 times for plaintiffs and something like 3 times for defendants. Defendant’s experts had more traditional and directly relevant qualifications. They were specialists in the fields at issue; they were on the staffs at Concord Hospital in New Hampshire and the Metropolitan Hospital Center in New York, and one was a professor at New York Medical College. The district court could properly credit the opinions of these experts while giving significantly less weight to those of plaintiffs’ experts. Doing so would lead to a conclusion of “no liability” on the part of the doctors. Appellate courts can rarely second-guess such district court determinations. Anderson, 470 U.S. at 575, 105 S.Ct. at 1512. Given the striking difference in qualifications, in light of the specific testimony, we believe it most unlikely that the district court would, or could, change its opinion in respect to negligence. Finally, assuming the doctors were not negligent, we can find no other basis in the record for a finding that the hospital itself was negligent in a manner that caused Ducharme’s death. All this is to say that, *32 having reviewed the record, we have a “definite and firm conviction,” United States v. United States Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 542, 92 L.Ed. 746 (1948), that the findings that the hospital negligently caused Ducharme’s death are “clearly erroneous.” The findings as to the doctors’ negligence have more than adequate record support and remain unaffected by those in respect to the hospital. For these reasons, the judgment of the district court is Reversed.