LEAVY, Circuit Judge:
OVERVIEW
The appellee, Marie Yako (Mrs. Yako), brought this action for medical malpractice under the Federal Tort Claims Act (the FTCA), 28 U.S.C. § 1346(b) and §§ 2671-
2680 (1982 & Supp. V 1987). She claimed Dr. Donn G. Kruse (Dr. Kruse), a physician at a United States Public Health Service hospital in Bethel, Alaska, failed to diagnose meningitis in her two and one-half year old son, Daniel Yako (Daniel), which resulted in severe brain damage, partial blindness and deafness, and permanent mental retardation.
The court found Dr. Kruse breached the applicable standard of care in failing to diagnose meningitis, thereby causing a delay in treatment that exacerbated Daniel’s injuries. The court awarded total damages of $5,188,379 to Daniel and his mother. The United States appeals.
We affirm the district court’s decisions on both liability and damages.
FACTS
On Saturday, March 3, 1984, Daniel suddenly became ill. He vomited a number of times and had a temperature of 101 degrees. When he did not improve by the following day, Mrs. Yako took him to the outpatient clinic at Bethel Hospital shortly after noon on Sunday, March 4,1984. Daniel had become too weak to sit or stand without assistance and was unresponsive to his mother’s voice. Mrs. Yako first spoke to a nurse. Then, Dr. Kruse spoke with Mrs. Yako about Daniel’s symptoms and physically examined Daniel. He noted Daniel was lethargic. Dr. Kruse concluded Daniel had the flu, prescribed Tylenol, and told Mrs. Yako to bring Daniel back to the hospital on Monday morning if he did not improve.
By Monday, March 5, Daniel’s condition had worsened. At 8:30 a.m. on that day, his mother returned with him to Bethel Hospital. By that time, Daniel was semi-comatose. Suspecting meningitis, the doctors immediately administered antibiotics and performed certain tests. Daniel was transported to Anchorage and admitted to the Alaska Native Medical Center. The test results from Bethel Hospital revealed that Daniel had contracted pneumococcal meningitis.
Southwest Alaska, where Bethel Hospital is located, has the highest rate of bacterial meningitis in the United States. Meningitis occurs in Southwest Alaska at ten times the national average. Bacterial meningitis of the type contracted by Daniel is diagnosed positively only by analysis of the cerebral spinal fluid, obtained by performing a spinal tap, or lumbar puncture. Symptoms of meningitis can vary and depend partly on the patient’s age. However, fever, vomiting, and lethargy are common symptoms.
The court concluded that within a reasonable degree of medical probability, Dr. Kruse’s failure to diagnose meningitis immediately, which aggravated Daniel’s mental and physical condition, was a substantial factor in causing Daniel’s injuries. The court found that an adverse outcome from this disease is significantly more likely if a child is semi-comatose when examined, rather than lethargic. The court found Dr. Kruse and Bethel Hospital negligent in that they created an “unnecessary delay in the prompt diagnosis and treatment of [Daniel’s] condition of bacterial meningitis.”
The court found the United States liable for damages under Alaska Statute (AS) 09.55.540
and section 1346 of the FTCA. Daniel was awarded $1,090,179 for lost earning capacity, $1,100,000 in future care, $1,345,000 in residential care, $53,200 in respite care, and $1,300,000 in general damages. Daniel’s award totalled $4,888,379. In addition, Mrs. Yako received $300,000 in general damages to compensate her for the loss of Daniel’s love and companionship.
The United States timely appealed. On appeal, the United States argues the district court erred by concluding there was malpractice, since Daniel’s symptoms were fully consistent with the flu. The United States also contends the damage award was excessive because: (1) the award for pain and suffering was inappropriate in
light of applicable precedent; (2) double compensation was awarded for living accommodations; and (3) Mrs. Yako should not have been given an award for loss of love and companionship of a child, because Alaska has no law allowing such a recovery.
DISCUSSION
Whether The District Court Clearly Erred In Finding Malpractice
According to expert testimony at the trial, meningitis is an inflammation of the membranes that cover the spinal cord and the brain. Without treatment, it is nearly always fatal or disabling. Around the time of World War II, the treatment of meningitis was revolutionized by the discovery of antibiotics. However, it is vital that meningitis be diagnosed at the earliest possible moment, because diagnosis and treatment make a crucial difference to the patient’s recovery. The bacteria and the toxins produced by meningitis act rapidly. If left untreated, the bacteria can inflame not only the covering of the brain and spinal cord, but their substance as well, destroying nerve cells and nerve attachments, thus interfering with vital functions such as respiration and intellectual activity.
Because treatment is available, failure to diagnose meningitis immediately is regarded as catastrophic. Due to the danger level of meningitis, the medical profession has established that a physician dealing with sick children should have a high index of suspicion for meningitis. The standard of care in the profession is that if there is a
possibility
of meningitis, a spinal tap is performed. The only means to prove a diagnosis for meningitis is by a spinal tap.
The government disputes the trial court’s finding that the standard of care is that “when a doctor suspects,bacterial meningitis in a young child, a lumbar puncture should be performed immediately.” Upon review of the record, we find that there was expert testimony by Dr. Mendelsohn that a spinal tap should be performed upon the mere possibility of meningitis where the symptoms in a young child include high fever, lethargy, and vomiting. While there was contrary testimony by Dr. Wehrle as to the appropriate standard of care, we cannot find that the conclusion of the trial court was clearly erroneous.
The symptoms of meningitis vary with age. For a child of two and one-half years, they include fever, vomiting, and lethargy. While a stiff neck is a symptom of meningitis in adults, it is not always present in children between infancy and adulthood. If the disease has produced enough pressure on the brain, the optic nerve swells, producing a condition called papilledema.
The government contends the evidence does not support a finding of malpractice. It argues Daniel’s symptoms were fully consistent with influenza of the H-l-N-1 variety that was prevalent in the Bethel area at the time. The government also contends that since Daniel lacked certain symptoms and was beyond the age when meningitis is most likely to strike, a physician’s suspicions of the possibility of meningitis would not be raised. The government further argues that no evidence could demonstrate that a delay in diagnosis and treatment would have increased Daniel’s injuries, because by the time meningitis is diagnosed, most of the damage has been done.
The government maintains Daniel was stricken with meningitis after Dr. Kruse correctly diagnosed his symptoms as the flu. According to the government, this is the only reasonable explanation for the flu symptoms Daniel exhibited on Sunday, March 4, 1984, and the meningitis that was diagnosed one day later. Since Daniel’s symptoms were so severe by March 5, 1984, the government contends the meningitis was of the rapidly fulminating variety.
We find that the record does not support the government’s arguments. Instead, the record reveals substantial support for the district court’s finding that Dr. Kruse was negligent when he diagnosed the flu and sent Daniel home with Tylenol. That record consists of Marie Yako’s testimony about her son’s condition on March 4, 1984; Dr. Kruse’s scant notes of the examination, his testimony and admissions; and the ex
pert testimony of Dr. Robert Mendelsohn and Dr. Paul Wehrle.
Marie Yako’s Testimony
Marie Yako, a Yupik native Alaskan, lived alone with Daniel in a one-room home, where they shared the same bed. She testified that on Friday, March 2, 1984, Daniel’s health was normal. However, when she arose about 10:00 a.m. on Saturday, March 3, 1984, Daniel started to vomit, and remained in bed. His temperature was 101 degrees. He had a diminished appetite, and although she tried to feed him, he could not keep food down. He had to be assisted to go to the bathroom, and “didn’t do much” all day. By Saturday evening, his condition had worsened. He vomited all night, and Mrs. Yako stayed up to watch him. On Sunday, according to Mrs. Yako, he was even worse. She took him to the hospital in her arms, stating at trial that “[h]e wasn’t responding. It was — he didn’t have any — it was like he didn’t have any feelings.” At the hospital, his temperature was 101 degrees.
Mrs. Yako testified she told Dr. Kruse Daniel’s symptoms: “that he was throwing up, that he had a temp, and that he just laid in bed. And Daniel at that time, he couldn’t really sit down or hold his body up by himself.” Daniel remained in her arms during Dr. Kruse’s examination. Mrs. Yako testified that Daniel physically resisted Dr. Kruse’s physical examination, but “would lay right back down” after the doctor touched him.
Dr. Kruse’s Testimony
Dr. Kruse has no recollection of his examination of Daniel. The screening nurse’s notes reflect only complaints of vomiting and dizziness.
Dr. Kruse’s notes do not show that he asked Mrs. Yako to elaborate on the vomiting or dizziness. His notes on the history of Daniel’s illness state that: (1) no one else in the family had the symptoms; (2) Daniel ate chicken with his mom the previous evening; and (3) as yet, there were no upper respiratory symptoms. Dr. Kruse admitted these notes do not give a full history because they do not reflect how long Daniel had been lethargic, vomiting, and not eating properly.
Dr. Kruse’s notes on his physical examination of Daniel show that he examined Daniel’s nose, ears, throat, chest, abdomen and neck. He found Daniel’s neck was not stiff “when calm.” The lack of a stiff neck led Dr. Kruse away from a diagnosis of meningitis, because he thought a child of thirty months would show signs of one if afflicted with meningitis. However, Dr. Kruse knew symptoms of meningitis vary with age. On cross examination, he admitted that not all children over twenty-four months of age with meningitis have a stiff neck.
He agreed that a stiff neck is seldom seen in the early stages of meningitis.
Dr. Kruse’s notes state that Daniel was “lethargic.” He testified he used “lethargic” to indicate that Daniel was “a low energy, a sleepy child.” However, this testimony is suspect because Dr. Kruse had no recollection of how Daniel really appeared during the examination on March 4, 1984.
Dr. Kruse also admitted he would have performed a spinal tap if given the history Mrs. Yako testified she gave him. He stated: “if the history of the patient is that they do not respond to their mother’s voice and they can’t sit up on their own and they vomited 15 or 16 times, that person is going to get a lumbar puncture unless there is evidence ... that it’s dangerous to do a lumbar puncture.”
On direct examina
tion, he agreed with the defense attorney that if a child is unresponsive or keeping his eyes closed and apparently sleeping, and if such behavior appears with a fever and vomiting, a spinal tap should be administered. He also testified that native Alaskan parents often are vague in describing symptoms and that Yupiks have difficulty thinking in terms of numbers. Nonetheless, Dr. Kruse admitted it is a physician’s responsibility to obtain a thorough and accurate history. He admitted he did not write down a full history in 1984 and that he has no specific recollection of what Mrs. Yako told him on March 4 of that year.
Dr. Kruse stated he did not obtain spinal fluid because:
The child was sleepy but not inappropriately sleepy given his fever and the fact that he’d been vomiting. His neck was supple. He negatively responded to negative things in the examination. He positively responded to positive things in the examination and the record does not include any of the severe complaints in history brought forward by Mrs. Yako in her testimony. That’s why he didn’t get tapped.
Dr. Kruse admitted that a physician must maintain a high level of suspicion for meningitis. He admitted that he was aware of the high rates of meningitis in the Bethel area. He admitted that a spinal tap is a safe method to diagnose meningitis, “easier than drawing blood,” and that he could perform one. He admitted that a spinal tap is the only method to substantiate a diagnosis of meningitis.
According to Dr. Kruse’s testimony, the symptoms of the H-l-N-1 flu epidemic are fever and vomiting for anywhere from twenty-four to forty-eight hours, followed by upper respiratory tract infections, including coughing, sneezing, and/or a stuffy nose that lasts from four to ten days. The fever and the upper respiratory tract infections do not occur at the same time. Dr. Kruse testified that Daniel’s symptoms, as described in his notes, supported his diagnosis of the flu. His notes state “No URI [upper respiratory infection] symptoms
yet.
” (Emphasis added.) According to the testimony of expert witness Dr. Robert Mendelsohn, this observation should have raised Dr. Kruse’s suspicions that he was confronted not with the flu, but a with possible case of meningitis. Therefore, according to Dr. Mendelsohn, Dr. Kruse should have performed a spinal tap.
Dr. Mendelsohn’s Testimony
Dr. Mendelsohn, a pediatrician,
reviewed Daniel’s medical records from the time he was newborn through the time he was admitted to the hospital in Anchorage. Dr. Mendelsohn testified that Daniel’s symptoms on March 4,1984 presented two possible diagnoses: flu or meningitis. He testified that with these symptoms, the physician must look for signs that the child has the more serious illness. He stated “the job of the doctor is to ... winnow out from all the patients with flu the ones who do not have flu and who have meningitis.” Although Dr. Mendelsohn admitted he was not familiar with the H-l-N-1 flu virus, he defended his position as follows:
Q. All right. So you don’t know as you sit here today that the diagnosis that
Donn Kruse made of these symptoms in this record are not consistent with the H-l-N-1 influenza epidemic that was there at the time.
A. Oh, I think I know very well that they are not consistent. Because what Dr. Kruse was doing was trying to predict what would happen in the next couple of days. He was crystal ball-gazing. He had no support for [the flu diagnosis] other than epidemiologic statistical evidence. And epidemiologic statistical evidence does not count when one has a patient in front of them.
Dr. Mendelsohn observed there was no supporting evidence in Daniel’s case for a diagnosis of the flu:
Q. None, Doctor?
A. Well, lets take a look at the record.
The oropharynx was clear. The tym-panic membranes were normal. The chest was clear. And there were no upper respiratory symptoms. Now those are supporting symptoms and findings of flu.
The testimony continued:
Q. If it were represented to you that the presenting clinical signs [of H-l-N-1] were vomiting, fever, no upper respiratory infection symptoms appeared until resolution of the vomiting and the fever, would these — the information you have in here be consistent with that?
A. No, and I’ll tell you why. If it were presented to me as you just said, if the presenting symptoms of flu are vomiting, fever and no respiratory symptoms, in this child, you still have to contend with the dizziness and the lethargy.
Dr. Mendelsohn’s critical testimony occurred when the defense attempted to impeach him by asking him how he knew Dr. Kruse had not seen hundreds of cases presenting exactly the same symptoms as Daniel’s. Dr. Mendelsohn responded:
I assume that Dr. Kruse had seen lots of patients....
Now I would predict that if one went through all the cases, and I’ve had an opportunity to look at some of them, but if one went through all the cases, I would predict that you will not find this combination of vomiting, dizziness, lethargy and absence of a supportable diagnosis.
Dr. Mendelsohn then examined several case histories, some selected by defense counsel at the court’s invitation, purportedly to show case histories identical to Daniel’s. Dr. Mendelsohn succinctly distinguished the symptoms and diagnosis of each one from Daniel’s case history.
At this point, the government was unable to present any other extrinsic evidence to contradict Dr. Mendelsohn’s testimony.
When a witness like Dr. Mendelsohn has given coherent and facially plausible testimony that is not contradicted by extrinsic evidence, a finding based on that testimony can virtually never be clear error.
Anderson v. City of Bessemer,
470 U.S. 564, 575, 105 S.Ct. 1504, 1512, 84 L.Ed.2d 518 (1985);
Pallia v. Hawaii Dep’t of Land & Natural Resources,
852 F.2d 1106, 1110 (9th Cir.1988). The district
court’s conclusions that Daniel’s symptoms were not consistent with the flu, that Dr. Kruse should have been suspicious of meningitis and performed a spinal tap, and that Daniel’s injuries were aggravated by the delay in treatment, have support in the record. Consequently, the government’s contention on appeal that Daniel’s flu developed into a rapidly fulminating meningitis after Dr. Kruse’s March 4, 1984 diagnosis is a red herring.
Further, the government’s argument that no evidence could demonstrate that a delay in diagnosis and treatment would have increased Daniel’s injuries, because by the time meningitis is diagnosed most of the damage has been done, is not supported by the record. Dr. Kruse himself admitted that early diagnosis of meningitis has a profound impact on the patient’s prognosis. Dr. Mendelsohn agreed that early treatment is vital: “the outcome in meningitis is measured in terms of hours of delay in treatment.”
The government’s attempts to impeach Mrs. Yako’s testimony regarding her son’s symptoms on March 4, 1984, with evidence that she incorrectly recalled the location and the length of the examination are diminished by a medical history taken from her the following day. Dr. Freudenthal, who examined Daniel on March 5, 1984, recorded Marie Yako’s description of Daniel’s symptoms from Saturday, March 3, 1984. His notes mirror what she testified she told Dr. Kruse on Sunday, March 4, 1984. Dr. Freudenthal’s notes state that Daniel was normal on Friday and that on Saturday he just lay in bed and vomited 15 or 16 times, did not seem to respond to verbal stimuli, could not sit or stand without help, and had a temperature of 101 degrees. The notes reflect the vomiting began on Saturday. This is strong evidence either that Mrs. Yako told Dr. Kruse what she testified she told him, or that he failed to elicit from her the history of Daniel’s case as she described it to Dr. Freu-denthal. Dr. Kruse admitted it is a physician’s responsibility to obtain a thorough and accurate history.
After examining the entire available record of this case on appeal, we are not left with the definite and firm conviction that a mistake has been committed. Accordingly, the trial court did not clearly err in finding malpractice.
Whether The District Court Erred in Determining The Amount of Damages
Standard of Review
In FTCA cases the clearly erroneous standard governs our review of factual determinations, including damages.
Shaw v. United States,
741 F.2d 1202, 1205 (9th Cir.1984).
A. Pain and Suffering
Under the FTCA, suits against the United States are governed by the substantive law of the place where the act or omission complained of occurred.
See, e.g., Richards v. United States,
369 U.S. 1, 6-10, 82 S.Ct. 585, 589-91, 7 L.Ed.2d 492 (1962);
Trevino v. United States,
804 F.2d 1512, 1516 (9th Cir.1986),
cert. denied,
484 U.S. $16, 108 S.Ct. 70, 98 L.Ed.2d 34 (1987);
Shaw,
741 F.2d at 1205.
Tbe district court awarded Daniel $1.3 million for physical and emotional pain and suffering and loss of enjoyment of life. Alaska law entitles a plaintiff to damages for these injuries.
American Nat’l Watermattress Corp. v. Manville,
642 P.2d 1330, 1341 (Alaska 1982);
Beaulieu v. Elliott,
434 P.2d 665, 674-76 (Alaska 1967). The Alaska Supreme Court has held: “there is no fixed measure of compensation in awarding damages for pain and suffering, and such an award necessarily rests in the good sense and deliberate judgment” of the trial court.
Beaulieu,
434 P.2d at 676.
The government contends the award of $1.3 million is excessive. To determine whether an award is excessive, this court looks to the relevant state’s case law on excessive awards.
Trevino,
804 F.2d at 1515. The Alaska Supreme Court will set
aside an award for pain and suffering as excessive only if “it is so large as to strike us that it is manifestly unjust, such as being the result of passion or prejudice or a disregard of the evidence or rules of law.”
Beaulieu,
434 P.2d at 676 (footnote omitted). To make that determination, this court must compare the challenged award to awards in similar cases in the same jurisdiction.
Trevino,
804 F.2d at 1515 (citing
Shaw,
741 F.2d at 1209).
We have very recently examined a medical malpractice case from Alaska under the FTCA that involved a child injured at birth. In that case, we affirmed an award of $2 million in noneconomic damages.
Scott v. United States,
884 F.2d 1280, 1283-84 (9th Cir.1989). Based on a comparison with this precedent from Alaska, the $1.3 million award to Daniel is not excessive under Alaska law, because the child in
Scott
and Daniel suffer similar injuries.
Cf id.
at 1284 (finding, by comparing injuries, that a $5 million nonpecuniary award in an Alaska case is evidence that the $2 million award in
Scott
was not excessive).
The child in
Scott
and Daniel have similar life-long prognoses. As a result of medical malpractice, the child in
Scott
is severely deformed and suffers from spastic quadriplegia. He will not be able to work in his adult life; he has abnormal speech and eye movement; he cannot walk and is wheelchair bound; and he will never be able to feed or independently care for himself.
Id.
at 1281-82, 1283. He is aware of his environment, is capable of feeling pain and can perceive the fact that he is disabled.
Id.
at 1282.
Daniel’s I.Q. is only 53. His fine and gross motor functions and hearing and vision are seriously impaired. He is now about seven years old and cannot list the alphabet, cannot identify certain colors, cannot copy a sloping line, cannot throw a ball, or count past the number five. He will need long-term special education and will be dependent on adult supervision the rest of his life.
Accordingly, we affirm the $1.3 million award for Daniel’s pain, suffering, and loss of the enjoyment of life.
B. The Alleged Double Compensation for Living Accommodations
The district court awarded Daniel $1,090,179 for the loss of earning capacity and $1,345,000 to cover residential care in a facility for handicapped adults
after his twenty-first birthday, for his life expectancy of 50.46 years. Daniel’s residential care must include specialized treatment and therapy. Alaska law allows recovery for lost future earnings and for medical and custodial care costs.
See, e.g., Alaska Airlines, Inc. v. Sweat,
568 P.2d 916, 934 (Alaska 1977) (loss of future earnings);
Morrison v. State,
516 P.2d 402, 406 (Alaska 1973) (anticipated medical and custodial care).
The government contends this award is duplicative, in that “much of the earnings of a healthy, working person will be used to provide his own residential care, including such expenses as food and lodging.” The government argues the amount of earnings that would normally be spent on residential care should be subtracted from the total lost earnings figure. The district court rejected this argument at trial.
For support, the government cites
Flannery v. United States,
718 F.2d 108, 112-13 (4th Cir.1983),
cert. denied,
467 U.S. 1226, 104 S.Ct. 2679, 81 L.Ed.2d 874 (1984), a FTCA case in which the Fourth Circuit considered the prohibition in 28 U.S.C. § 2674 against “punitive damages.” Considering the issue
sua sponte,
a divided panel of that court held that since the plaintiff in
Flannery
was comatose and unable to spend any of the money awarded to him, he was unlike the usual cases where
a living plaintiff must pay his own living expenses, though an award for lost earn
ings may be the source of his funds. In this case, however, the plaintiff will be required to pay nothing, for the award of future medical expenses includes all of the personal expense that the plaintiff will incur.
718 F.2d at 112. The Fourth Circuit found that awards of future care expenses and future lost earnings required the government to pay twice for a plaintiff’s personal living expenses, are thereby punitive, and must be offset.
Id.
at 112-13.
Aside from the fact that Daniel Yako is not comatose, every court outside the Fourth Circuit, when faced with this issue, has rejected Flannery's approach, including this circuit.
Shaw,
741 F.2d at 1208.
See, e.g., Reilly v. United States,
863 F.2d 149, 163-66 (1st Cir.1988);
Rufino v. United States,
829 F.2d 354 (2d Cir.1987) (citing this circuit’s decision in
Shaw).
The courts reject
Flannery
for two main reasons: (1) they do not believe separate awards for future medical and maintenance costs and for lost earnings are in fact punitive,
see, e.g., Kalavity v. United States,
584 F.2d 809, 811 (6th Cir.1978) (the FTCA’s prohibition against punitive damages is designed to prevent “use of retributive theory of punishment against the government”); and (2) the refusal to award separate costs for these items may interfere with the FTCA’s provision for recovery in accordance with local law.
This circuit is widely quoted for providing the latter reason why
Flannery
should not be followed. In
Shaw,
this circuit explicitly rejected
Flannery,
stating: “we recognized that such an expansive view of federal law would ‘impinge seriously upon the architecture of the Act which provides for recovery according to the
lex loci delictus.’
” 741 F.2d at 1208 (quoting
Felder v. United States,
543 F.2d 657, 675 (9th Cir.1976)). This circuit also approved of the Sixth Circuit’s refusal in
Kalavity
to “adopt the view that any award of damages in excess of a plaintiffs direct or out-of-pocket loss is punitive.”
Shaw,
741 F.2d at 1208.
Because Alaska permits recovery for future medical and maintenance costs and lost future earnings, the precedent in
Shaw,
and the general rejection of
Flan-nery
by other courts, the district court’s award of separate amounts for future residential care and lost earnings capacity is not clearly erroneous and is affirmed.
C. Loss of Love And Companionship of A Child
Mrs. Yako was awarded $300,000 for the loss of Daniel’s love and companionship. The district court stated: “[t]he applicable law allows Marie Yako to receive reasonable compensation for the loss of Daniel Yako’s love, society and companionship.”
The government claims there is no precedent in Alaska that permits damages for the parent’s loss of consortium of a child. However, we recently decided that under Alaska law, a parent may receive damages for the loss of love and companionship of a child.
Scott,
884 F.2d at 1282. We consequently affirm the district court’s award for Marie Yako’s loss of consortium of her child.
CONCLUSION
We are not left with the “definite and firm conviction that a mistake has been committed” as to either the appellant’s liability or the amount of the damages.
United States v. United States Gypsum Co.,
333 U.S. 364, 395, 68 S.Ct. 525, 542, 92 L.Ed. 746 (1948). The district court did not clearly err, and its decision is AFFIRMED.