Gloria Miller and her parents, Gary and Julie Miller (Millers), appeal from a judgment entered on a jury verdict dismissing their medical malpractice claim against Young II Kim, M.D., the Wisconsin Health Care Liability Insurance Plan and the Wisconsin Patients Compensation Fund.1 They also appeal from an order denying their motion for a new trial. We conclude that the trial court erred when it instructed the jury that it should not find Dr. Kim negligent if he merely made a choice between alternative methods of diagnosis. The issue in this case is not whether Dr. Kim chose between two recognized methods of diagnosis but whether he negligently failed to recognize that Gloria's symptoms required that he perform a spinal tap to rule out spinal meningitis as the cause of those symptoms. We therefore reverse and remand for a new trial.
On May 10, 1986, Gloria, who was then three months old, was treated in the emergency room of Iowa County Memorial Hospital. Dr. Kim examined Gloria, concluded that she did not have an acute illness, and discharged her. The next day her parents returned Gloria to the hospital and another doctor performed a spinal tap which revealed that Gloria was ill with spinal meningitis. Gloria suffered permanent brain damage producing profound mental retardation and requiring total care. She died on June 6, 1993. The death certificate lists the cause of death as pulmonary aspiration due to a recent respiratory infection related to meningitis during infancy.
*191A jury found that Dr. Kim was not negligent in his care and treatment of Gloria. The trial court denied the Millers1 motion for a new trial. On appeal, the Millers argue that the trial court committed prejudicial error when it instructed the jury that it should not find Dr. Kim negligent "merely because he made a choice of a recognized alternative method of diagnosis . . . ." They contend that the medical experts were unanimous that if Gloria exhibited symptoms of spinal meningitis, a lumbar puncture (spinal tap) was the only diagnostic method which could rule out the condition. The Millers argue that the "alternative method of diagnosis" instruction probably misled the jury. We agree.
The instruction given by the trial court was an amended version of WlS J I — CIVIL 1023, Medical Malpractice:
If you find that more than one method of diagnosis for Gloria Miller's symptoms is recognized, then Dr. Kim was at liberty to select any of the recognized methods. Dr. Kim was not negligent merely because he made a choice of a recognized alternative method of diagnosis if he used the required care, skill, and judgment in administering the method. This is true even though other medical witnesses may not agree with him on the choice that was made.
(Emphasis added.)
The trial court substituted "diagnosis" for "treatment" in the pattern medical malpractice instruction. We conclude that the trial court erred in giving the amended instruction because there was no medical evidence that there was more than one method of diagnosing Gloria's symptoms. The medical experts were unanimous that if Gloria exhibited symptoms *192consistent with spinal meningitis, a spinal tap was the only proper diagnostic test.
BACKGROUND
Gloria was first seen in the emergency room by nurse Kathy Arneson. She recorded that Gloria had a fever of 103.2 degrees, a pulse of 180 and respiration of 50. She took the following history from Gloria's parents:
Felt warm last PM... fever broke with Tylenol. Felt warm once again during the noc. Awoke off and on during the noc. Has [decreased] appetite. Vomited x 1 this AM. Had mild cough last PM. Stools looser. Appears pale. Voided while temp, being taken. Urine appeared cone [ent rated]. Lung sounds clear.
Dr. Kim made a physical examination of Gloria and reviewed the laboratory report on Gloria's blood sample. His "impression" was that Gloria had a "viral illness." He concluded: "The patient does not have any acute illness." Dr. Kim sent Gloria home with instructions for her parents to administer Tylenol and Pedialyte and to return if her condition worsened.
The next morning the Millers found Gloria lethargic and returned her to the emergency room. Dr. Strickler examined Gloria and recorded her recent symptomatic history, which her mother recounted, including fever, diarrhea, vomiting, irritability and lethargy. He performed a spinal tap. The test revealed spinal meningitis. Gloria was immediately started on antibiotics and transferred to a Madison hospital.
At trial, the Millers presented the testimony of two medical experts, Dr. Frank Baker, a specialist in internal and emergency medicine, and Dr. Sidney Sussman, a specialist in pediatrics and pediatric infections. Dr.
*193Baker testified that Dr. Kim was negligent in his diagnosis of Gloria. He testified that Dr. Kim should have performed a septic work-up including a spinal tap to rule out the possibility of meningitis and further stated that any child under one year of age with an unexplained fever should have a spinal tap. Dr. Sussman testified that symptoms like Gloria's in one so young should have indicated to Dr. Kim the possibility of meningitis.
Dr. Kim presented four medical experts, each of whom testified that the decision to perform a spinal tap was a judgment call requiring individualized assessment of the patient. They also explained that because Gloria did not appear septic or irritable, a spinal tap was not indicated and therefore Dr. Kim could not be considered negligent for failing to perform such a test. However, all of Dr. Kim's experts agreed that if symptoms of meningitis were present, the only way to rule out the illness was to perform a spinal tap.
At the instructions conference, the Millers objected to the amended medical malpractice instruction. They contended that the medical experts unanimously agreed that the only way to diagnose spinal meningitis is by a spinal tap. Thus, they argued, any question as to whether the standard of care required Dr. Kim to perform the test was a question of fact for the jury and use of an "alternative method" instruction would only confuse the jury. The trial court disagreed and gave the instruction.
The jury found Dr. Kim not negligent in his diagnosis and treatment of Gloria. The Millers moved the court for a new trial.2 The court denied the Millers' *194motion and entered judgment dismissing the Millers' complaint with prejudice.
STANDARD OF REVIEW
An appellate court will reverse and order a new trial if it finds that a challenged jury instruction, taken as a whole, was prejudicial in that it probably and not merely possibly misled the jury, or if it finds that the meaning communicated by the instruction as a whole was an incorrect statement of the law. Fischer v. Ganju, 168 Wis. 2d 834, 849-50, 485 N.W.2d 10, 16 (1992).
"ALTERNATIVE METHOD OF DIAGNOSIS" INSTRUCTION
Dr. Kim makes three arguments explaining why the trial court did not err when it chose to give an "alternative method of diagnosis" instruction to the jury. We address each and explain why we reject his arguments.
First, Dr. Kim argues that the evidence warranted giving the "alternative method" instruction. He claims there are two methods of diagnosis of spinal meningitis: the first requires a doctor to do a spinal tap for any child under age one with an unexplained fever; the second makes the decision to perform a spinal tap an *195individualized one, required only when a child appears irritable or toxic or septic. However, these are not two different methods of diagnosis but two different opinions as to what symptoms exhibited by a young child require a spinal tap. The experts were unanimous that when the symptoms of spinal meningitis are present in a young child, the only diagnostic method to rule out the illness is a spinal tap.
Dr. Benjamin Wedro, defendants' expert, was asked the following question and gave the following answer:
Q Now Doctor, if we looked at what this text has said are indications of potential meningitis, and before we go any farther, you would agree that the only way to definitively diagnose meningitis is with a lumbar puncture or spinal tap?
A That's the only way you know 100percent for sure.
(Emphasis added.) Dr. Sydney Gellis, defendants' expert, testified:
Q That is not to say, Doctor, that if there is an indication for the procedure, that you should not do the tap because of these potential risks?
A If there is an indication, you have to go ahead.
Q Then it's mandatory?
A Mandatory.
(Emphasis added.) Dr. Starkey Davis, defendants' expert, testified:
*196Q And the only way to rule out meningitis once it's included in the differential is to do a lumbar puncture, is that not correct?
A Yes, sir.
(Emphasis added.) Dr. Mark Bishop, defendants' expert, testified:
Q "The critical diagnostic procedure is the spinal tap." Do you agree with that?
A Yes.
Q And you absolutely agree that the only way to tell whether a child does or does not have meningitis is to do the spinal tap if your index of suspicion for meningitis is high enough to justify it?
A That's the goal standard.
Q And you told . . . Mr. Liddle despite these possible risks that you described, if the index of suspicion is there, then the spinal tap is done?
A Yes.
(Emphasis added.) Dr. Frank Baker, plaintiffs' expert^ testified:
Q If the physical examination won't allow the physician to rule out meningitis, will anything obtained by laboratory means allow him to do that?
Yes. The only test that will allow you to rule out meningitis definitively one way or the other is a spinal tap also known as a lumbar puncture. ►
(Emphasis added.) Dr. Sydney Sussman, plaintiffs expert, testified:
*197Q And what kinds of changes occur?
A There is an inflammation here. And with that inflammation and the germs, pus develops and it gets under this membrane into this space and it can circulate down here. This is where the spinal fluid circulates. It can circulate down here all along the top here, down here, and this is why we have to do a spinal tap. It really is for all [intents and] purposes the only way you can tell for sure whether the patient does or does not have meningitis ....
(Emphasis added.) Thus, the medical experts agreed that a spinal tap should be performed if the patient's symptoms indicated spinal meningitis. We therefore reject Dr. Kim's argument that "individualized" observation is an alternative diagnostic technique when the patient's symptoms present an "index of suspicion" high enough to suggest meningitis.
Next, Dr. Kim argues that the "alternative method" instruction has been approved by the Wisconsin Supreme Court. See Treptau v. Behrens Spa, Inc., 247 Wis. 438, 443-44, 20 N.W.2d 108, 111 (1945).3 We agree that the instruction given in Treptau may be given in any case in which the medical evidence estab*198lishes that there is more than one method of diagnosis of a medical condition. However, in this case, it was error to give an "alternative method" instruction because the medical evidence was unanimous that where the "index of suspicion" is sufficiently high, a spinal tap is the only diagnostic tool which can rule out spinal meningitis.
The medical malpractice pattern instruction, WlS J I — CIVIL 1023, does not specifically instruct the jury as to negligent diagnosis. Diagnosis is considered "care and treatment."4 The "alternative method" instruction is optional and is only to be given by the trial court when the evidence allows the jury to find that more than one method of treatment of the patient is recognized by the average practitioner. The trial court's amendment of the pattern instruction would have been appropriate had there been medical expert testimony that there were available to the average practitioner alternative methods of diagnosing Gloria's spinal meningitis.5
*199One of the cases cited by the Civil Jury Instructions Committee in its comment to the pattern instruction is Francois v. Mokrohisky, 67 Wis. 2d 196, 226 N.W.2d 470 (1975), which was a negligent diagnosis case. In that case, the court said that "[a] physician is not an insurer of the results of his diagnosis or procedures. He is obliged to conform to the accepted standard of reasonable care, but he is not liable for failing to exercise an extraordinary degree of care." Id. at 201, 226 N.W.2d at 472. The court said that medicine is not an exact science,
and even the very best of [physicians] can be wrong in diagnosis or procedure. The question, however, is not whether a physician has made a mistake; rather, the question is whether he was negligent. Unless the untoward result was caused by the failure to conform to the accepted standard of care, he is not liable in negligence for damages.
Id.
The Francois court said "[t]he accuracy of diagnosis of gallbladder pathology is not a matter of common knowledge, and no expert testimony was adduced that could lead to a reasonable inference that a misdiagnosis was probably the result of negligence." Id. at 204, 226 N.W.2d at 473. Unlike Francois, here there was ample expert testimony that Dr. Kim was negligent in his diagnosis of Gloria's condition. Whether Dr. Kim *200was negligent in failing to recognize the symptoms of spinal meningitis or whether Gloria did not exhibit those symptoms sufficiently to create the necessary "index of suspicion" of meningitis is a question for the jury. "[W]here medical experts in a case agree as to the recognized and established proper treatment for a particular type injury but there is a dispute as to whether the plaintiff had that type of injury, the latter question is one of fact for the jury." Morganstein v. House, 547 A.2d 1180, 1183 (Pa. Super. Ct. 1988).
Finally, Dr. Kim argues that the "alternative method" instruction did not prejudice the Millers. The Wisconsin Supreme Court stated in Lutz v. Shelby Mutual Insurance Co., 70 Wis. 2d 743, 750, 235 N.W.2d 426, 431 (1975), that" [i] t is error for a court... to give an instruction on an issue which finds no support in the evidence." It is prejudicial error if an erroneous instruction probably and not merely possibly misleads the jury. Fischer, 168 Wis. 2d at 849-50, 485 N.W.2d at 16; Lutz, 70 Wis. 2d at 751, 235 N.W.2d at 431. We conclude that the "alternative method of diagnosis" instruction probably misled the jury.
By the Court. — Judgment reversed and cause remanded with directions.
1n this opinion, "Dr. Kim" includes all defendants. Gloria Miller died after this action was begun. We continue to include her in our references to the "Millers."
The Millers requested a new trial bn the following grounds:
*1941. That the jury's verdict was contrary to the great weight and clear preponderance of the evidence.
2. That the trial court erroneously instructed the jury concerning "alternative method[s] of diagnosis."
3. That the trial court erroneously refused to submit the issue of informed consent in light of its decision to give the "alternative method of diagnosis" instruction.
4. In the interest of justice.
The instruction given to the jury in Treptau read as follows:
You are further instructed that if you find from all the credible evidence in the case that the school of medicine recognizes more than one method of diagnosis or treatment of the condition which affected the plaintiffs right foot, it was not required, at its peril, to select one or the other of such methods and was at liberty to select either of said methods and may not be considered wanting in the required degree of care and skill merely because expert witnesses give their opinion that some other method would have been preferable.
See Taber's Cyclopedic Medical Dictionary 535 (17th ed. 1993) (defining diagnosis as "[t]he use of scientific and skillful methods to establish the cause and nature of a person's illness. . . . The value of establishing a diagnosis is to provide a logical basis for treatment and prognosis."); see Rivard v. McGinnis, 454 N.W.2d 453, 455 (Minn. Ct. App. 1990) ("One of the obvious purposes of a diagnosis, or determining the nature of a patient's condition, is to assist in determining the appropriate course of therapy.").
Were we to approve the "alternative method of diagnosis" instruction, there would be few "failure to diagnose" cases where the instruction would not be appropriate. Such cases invariably involve an assertion that a physician failed to recognize that an observed symptom or symptoms indicated the presence of a particular disease or injury. The question in the usual "failure to diagnose" case is whether the physician was *199negligent in failing to recognize the significance of the symptom or symptoms. That is because the alleged negligence lies in failing to do something, not in negligently choosing between courses of action. Doing something and doing nothing are not two methods of diagnosing a disease or injury. In any event, those were not the choices the expert witnesses testified were available in this case.