This is a medical malpractice action brought against two doctors as a result of the death of their eleven-year-old patient, Ron Adams. Plaintiffs are the parents of the child. From a nine to three jury verdict for defendants, plaintiffs appeal. We reverse.
Ron Adams died on Sunday, July 22, 1973, of peritonitis resulting from an undiagnosed case of appendicitis. Three days prior to his death, on Thursday, July 19, Ron had begun to complain of severe stomach pains. His father apparently thought the boy was constipated and gave him a laxative tablet. The child continued to experience stomach pain, so his mother took him to the medical emergency room of the Eugene Hospital and Clinic where she saw Dr. Dunn. Mrs. Adams told the doctor that her child was complaining of severe "crampy” abdominal pains, that he had vomited that morning and that since having been given the laxative he had had diarrhea. Dr. Dunn examined the child, performing a standard ear, eyes, nose and throat exam and an examination of the chest and abdomen. Dr. Dunn’s findings, after palpating the abdominal area, included mild upper left quadrant tenderness, but no lower right quadrant tenderness— the most common indicator of an inflamed appendix. The child’s temperature and blood pressure were normal. The doctor did not take a blood sample nor give a rectal examination. Because of the symptoms of abdominal pain, vomiting, diarrhea and no fever, Dr. Dunn made a diagnosis of stomach flu and told Mrs. Adams to bring the child back in the morning if he was still experiencing the pain, vomiting or diarrhea.
Mrs. Adams took the child back the next morning. Dr. Roe, a partner of Dr. Dunn, examined the child on the second occasion and took a somewhat different history of the patient. The child was still experiencing "crampy” abdominal pain, vomiting, diarrhea and loss *36of appetite (anexoria). Dr. Roe ordered a blood test1 and a urine sample, but did not take the child’s temperature. Dr. Roe reached the same conclusion as Dr. Dunn—that the child had gastroenteritis—and told Mrs. Adams that the child would probably continue to experience more vomiting and diarrhea and not to expect a rapid recovery. He did advise her, however, to call him if Ron got any worse over the weekend.
The child died Sunday morning. The report of the medical examiner indicated that the cause of death was acute peritonitis following a ruptured appendix.
The basis of plaintiffs’ complaint was defendants’ failure to diagnose the child’s appendicitis. At trial, Dr. Dunn testified that the "classic symptom” for diagnosing appendicitis was lower right quadrant pain and tenderness which may be accompanied by anexoria, vomiting and diarrhea. He insisted that, in order for a doctor to diagnose appendicitis, there must first be a finding of lower right quadrant pain. Dr. Roe testified in response to questioning that the classic symptoms of an acute appendicitis are abdominal pain, vomiting and fever and a high white blood cell count. He then stated, however, that the symptoms Ron exhibited—crampy abdominal pain with vomiting and diarrhea—were not consistent with a diagnosis of appendicitis.
On the other hand, plaintiff’s expert witness, Dr. Fox, testified that these same symptoms are the usual indicators of appendicitis. He also testified that lower right quadrant pain is not always present, and that in some cases, especially where the appendix is located in the pelvic area (as the autopsy showed Ron’s to have been), the child will complain of generalized abdominal pain. Dr. Fox indicated that, in his opinion, in cases of undiagnosed abdominal pain, to meet the *37appropriate standard of care the giving of a blood test and a rectal examination would be mandatory. He also testified that, considering the nature of the child’s complaints and the results of the incomplete blood test that was given (which indicated an elevated white blood cell count), the child should have been hospitalized and kept under observation.
Both defendants testified that an elevated white blood cell count is not a very reliable indicator of appendicitis, nor a very "helpful” diagnostic tool. Dr. Dunn testified that it was not customary to perform a rectal examination on children because they object to it, and that it is usually only used as a confirmatory aid.
Dr. Roe stated that he had treated children with appendicitis both before and after this incident, and that he had never had a situation of this sort occur before.
Plaintiffs make five assignments of error. We consider four. Plaintiffs first assign the trial court’s instruction to the jury to apply a standard of care based on that degree of care, skill and prudence
"* * *which the ordinary, careful, prudent and skillful physician engaged in the practice of medicine in the community of Eugene, Oregon, or similar communities would have used under the same or similar circumstances.” (Emphasis added.)
Plaintiff had requested that the standard of care be based upon that of a specialist in pediatrics.
Plaintiff is correct that Oregon law requires a specialist to exercise that degree of skill common to his specialty. Rayburn v. Day; 126 Or 135, 142-43, 268 P 1002 (1928). Further, expert testimony is generally required to establish what that appropriate standard of care is, as this would not normally be within the knowledge of the jury. Getchell v. Mansfield, 260 Or 174, 179, 489 P2d 953 (1971). Here, however, plaintiffs neglected to single out and discuss, through the *38questioning and testimony of its expert witnesses, the higher standard of care which might be required of pediatricians in this case. The trial court’s instruction was thus not error, because plaintiffs failed to establish any distinction between the standard of care for pediatricians and the standard of care for physicians in general.
Plaintiffs’ second assignment of error is that the trial court erred in giving its instructions on oral admissions. The trial judge instructed the jury:
"Statements made by the plaintiffs or any witness * * * are called oral admissions * * *. The testimony by a witness about such statements is to be viewed with caution * * *. (Emphasis added.)
Plaintiffs argue they were prejudiced by the court’s instruction because it unfairly draws attention to the "admissions” of plaintiffs, and because it fails to note the admissions of the defendants.
We agree with plaintiffs that the instruction as given was not correct. The instruction should have referred to admissions by all the parties. We do not agree, however, that plaintiffs were thereby prejudiced. At worst, the instruction was abstract, because it appears that no admissions were made by any of the parties. An admission is a statement made or an act done by a party to a lawsuit (or his predecessor or representative) which is or which amounts to a prior-to-trial acknowledgement that some fact is not as he (the party) now at trial claims it to be. See Oxley v. Linnton Plywood Association, 205 Or 78, 98, 284 P2d 766 (1955). We have not been directed to, nor have we found, evidence in the record of any statements made by defendants which amount to an acknowledgement that defendants in fact did not believe that they had correctly diagnosed and treated Ron Adams. The instruction, although error, did not require reversal. ORS 19.125(2).
Plaintiffs’ third assignment of error is that the trial court erred by sustaining defendants’ objection to the *39introduction of plaintiffs’ exhibit 5, a medical history of Ron Adams which contained only those portions of his medical records which had been prepared prior to the child’s death. This assignment is not well taken: The trial court admitted exhibit 5 when plaintiffs offered it a second time.
Plaintiffs fourth assignment of error is that the trial court erred by failing to grant plaintiffs’ motion for a new trial, which was based in part upon newly discovered evidence.
Plaintiffs’ motion for a new trial was accompanied by an affidavit of a Mrs. Mentzer, who had read about the Adams trial in the newspaper and had then called plaintiffs’ attorney. Her affidavit indicates that in January, 1970, her six-year-old daughter, Sarah, became quite ill with intermittent abdominal pain, diarrhea and vomiting. Mrs. Mentzer took Sarah to the Eugene Hospital and Clinic where Dr. Roe examined her. Upon examination, Sarah experienced "questionable tenderness” in both her right lower and left upper quadrant. Roe did not perform a blood test on Sarah. Sarah became increasingly ill with more vomiting and diarrhea and increasingly severe abdominal cramping pain. Mrs. Mentzer took her back to the clinic a second time. Dr. Roe examined her and indicated that Sarah had a bad case of flu. When Sarah did not improve, Mrs. Mentzer called Dr. Roe a third time and he told her to bring Sarah into the clinic immediately as he suspected appendicitis. Dr. Roe told Mrs. Mentzer that he should have given Sarah a blood test during the first examinations.
The child was seen at the Eugene Hospital and Clinic by Dr. Galloway (who was called in by Dr. Roe). Dr. Galloway examined and hospitalized the child and concluded that she had a ruptured appendix. Dr. Galloway’s examination included the giving of a blood test and a rectal examination.
Sarah was hospitalized for approximately four to six weeks. Both Dr. Roe and Dr. Dunn were called in *40for consultation after she was hospitalized and were presumably aware of everything that had occurred with the child since she was initially seen by Dr. Roe.
Plaintiffs point to the marked similarities between the two cases: Sarah, like Ron, had complained of intermittent crampy pain and diarrhea, vomiting and upper left quadrant abdominal pain. As with Ron Adams, Dr. Roe had diagnosed Sarah’s illness as gastroenteritis.
Plaintiffs maintain that the new evidence provided by Mrs. Mentzer indicates specifically that: (1) The defendants had been involved with a case similar to that of Ron Adams. (Dr. Roe testified at trial he had never had a situation "of this sort” occur before.) (2) The defendants were aware that a white blood count is a significant diagnostic tool in diagnosing appendicitis. (Both defendants had testified that a white blood cell count is "not helpful” and is "an unreliable indicator of an appendicitis.”) (3) In order to meet the standard of care, defendants should have given Ron a rectal examination. (Dr. Dunn had testified that a rectal exam was not appropriate in Ron’s case because children do not like it.) (4) Intermittent, crampy abdominal pain accompanied by vomiting and diarrhea is not inconsistent with the existence of appendicitis. (Defendants had testified that those symptoms generally indicated gastroenteritis, not appendicitis.) In sum, plaintiffs assert that the information concerning Sarah Mentzer would indicate that "defendants were aware, or should have been aware, of a more demanding standard of care for the diagnosis of appendicitis than was used in the treatment of Ron Adams.”
Applications for a new trial upon the ground of newly discovered evidence are always "entertained with reluctance and granted with caution.” Lane County Escrow Service, Inc., v. Smith, 211 Or 273, 560 P2d 608 (1977). In order to justify the granting of a new trial on the basis of newly discovered evidence, certain requirements must be met. Initially, the evi*41dence must be such that it could not have been discovered before trial by the exercise of due diligence. Grillo v. Burke’s Pain Co., 275 Or 421, 551 P2d 449 (1976). The evidence must be material to the issue and must not be merely cumulative, contradictory or impeaching of former evidence. Finally, the evidence must be of such a nature as would be likely to produce a different result if a new trial is granted with caution. See Newbern v. Exley Produce Express, Inc., 208 Or 622, 631, 303 P2d 231 (1956), and cases cited therein. See also Lane County Escrow Service, Inc. v. Smith, 277 Or 273, 560 P2d 608 (1977).
We think in this instance the requirements for granting a new trial have been met. Plaintiffs’ affidavit details there post-trial discovery of the evidence and makes it clear that the information could not have been discovered before trial by the exercise of due diligence.
Defendants argue that each of the specific points raised by plaintiffs would at best constitute impeachment. As for establishing a higher standard of care, defendants assert that the evidence would be merely cumulative to the expert testimony. We do not think the evidence of Dr. Roe’s treatment of Sarah Mentzer’s illness under similar circumstances is "merely” impeaching or "merely” cumulative. Rather, we view the evidence as material to the issue of negligent misdiagnosis.
As a general rule, evidence of prior acts by a party is not admissible to prove a specific act of negligence. Such evidence is, however, admissible to prove the existence of a continuing course of negligent conduct, that the course of conduct is in fact dangerous, or that the defendant had notice of its dangerous character, provided that the prior acts occurred under similar circumstances. Rader v. Gibbons and Reed Company, 261 Or 354, 359, 494 P2d 412 (1971). See also Kelty v. Fisher et al, 101 Or 110, 117-18, 199 P 188 (1921). Thus in Radar where the issue was the general contractor’s negligence in permitting rocks to fall from *42a road under construction to the highway below, resulting in the death of a motorist, we held that testimony by witnesses that they had frequently encountered falling rocks and boulders on the same road and had complained to the contractor was admissible because the circumstances were sufficiently similar to those of the fatal accident.
The problem of admissibility of prior acts of negligence arose in an early malpractice case. Rayburn v. Day, 126 Or 135, 268 P 1002 (1928), involved a malpractice action against a surgeon for leaving a sponge in plaintiffs body after an operation, which plaintiff had alleged was the cause of her ill health. Plaintiff sought to introduce a statement by the defendant that he had left a sponge in a patient once before. This court held that the proferred evidence was inadmissible to prove defendant’s negligence. We noted, however, that the evidence would have been admissible if plaintiff had sought to prove that defendant had knowledge that leaving a sponge in plaintiff’s body violated the established practice. Id., at 144-45. Because defendant never contended that leaving a sponge in the body was proper practice, this was not an issue for which plaintiff’s proferred evidence would have been material, and it was properly refused.
In the case at hand, defendants contended that their diagnosis and treatment of Ron Adams was proper practice. The evidence offered by Mrs. Mentzer suggests otherwise. It also suggests that defendants engaged in a continual course of negligent conduct with regard to misdiagnosing appendicitis in children and, most importantly, that defendants had notice of the dangerous character of the very symptoms which they misdiagnosed. In short, the evidence suggests that defendants were aware, or should have been aware, of a more demanding standard of care for the diagnosis of appendicitis than was used in the treatment of Ron Adams. We think this evidence would be likely to produce a different result in a new trial. The trial judge’s refusal to grant a new trial was error. *43Plaintiffs’ fifth assignment of error is that the trial court erred in refusing to instruct the jury that possible negligence by Mr. and Mrs. Adams was not an issue in the case. Plaintiffs argue that the instruction was necessary to counter remarks by defendants’ counsel during closing arguments where he suggested that the negligence of Ron’s parents was a greater source of fault than the doctors’ misdiagnosis.
The question of plaintiffs’ negligence was not an issue framed by the pleadings, and was raised only by defendants’ remarks in closing argument concerning causation. The trial court’s refusal to instruct was not error. The trial judge is not required to give, upon request, instructions as to all possible issues raised by the evidence, whether such issues were pleaded or not. Moe v. Jolly Joan Howser, 239 Or 531, 534, 399 P2d 22 (1965).
Reversed and remanded.
The blood test was ordered to insure that the child was not suffering from sickle cell anemia, not to check the white cell count which might indicate the presence of an infection.