Opinion
SHABO, J.
This is an appeal from a judgment of conviction (order granting probation) following a jury trial, in which appellant was convicted of a single count of violating Unemployment Insurance Code section 2101.1 Following the denial of his oral motion for new trial, the court placed appellant upon summary probation upon certain terms and conditions, among which were that he serve 120 days in the county jail and pay a fine of $180.
Contentions on Appeal
Appellant contends on appeal that the trial court committed reversible errors (1) in failing properly to instruct the jury as to the substantive offense [Supp. 32]*Supp. 32charged in the complaint, (2) in admitting evidence concerning other disability claims submitted by appellant and by others upon which inadmissible statistical probability evidence was based, (3) in permitting copies of the medical chart of the principal prosecution witness to be used by the jury during the prosecution’s cross-examination of appellant, and (4) in allowing a disability claim form for this witness (People’s exh. 1) to be sent into the jury room because the claim form contained a statement as to the penalty for violating section 2101. Because we agree with appellant’s first two contentions, we reverse.
Trial Testimony
The prosecution’s evidence at trial involved principally the testimony of Fred Argosino, a special investigator on his first undercover assignment for the California Employment Development Department. Mr. Argosino testified to going to appellant’s Gardena clinic in order to bring a disability benefit claim form to appellant, a general practitioner, for his certification. Mr. Argosino made various visits to the clinic, the first being on September 30, 1981, when he appeared without a prior appointment and spoke to the receptionist. He told the receptionist at first that there was nothing wrong with him but then changed his story; he told her that he was experiencing “frequent dizziness.” Because he had no appointment, the receptionist told him to come back in the afternoon. That afternoon, when he returned, Mr. Argosino completed a personal history form on which he falsely listed the name of his employer, as Robert Shaw Controls, listed his job as an assembler, and gave a fabricated residential address with no telephone number. When he first saw appellant, he told appellant that there was really nothing wrong with him, that the only reason for his visit was that he wanted appellant to give him a medical leave, and that the witness had heard that he was going to be laid off from his job. Mr. Argosino told appellant that a coworker had advised him that he could get a medical leave from appellant. Appellant responded, according to the witness, that Mr. Argosino “should” tell him that there was something wrong with him before appellant could or would fill out a disability form, that Mr. Argosino’s problem was “financial,” not medical, and, at one point, that Mr. Argosino “should tell” appellant that he was “dizzy” or “nervous.” Explaining that the only reason that he had come to see appellant was to “get my medical slip,” and that really there was “nothing wrong,” Mr. Argosino then asked appellant, “What if I tell you, doc, that I was dizzy or that I am dizzy, but, you know, I am not?” Appellant, according to the witness, did not respond. Mentioning the strictness of the State of California, appellant told the witness to think the matter over and call him back if Mr. Argosino could think of “something or anything” that he thought was wrong with him. Mr. Argosino immediately left the clinic. He was not charged for his visit.
[Supp. 33]*Supp. 33In his defense, appellant testified with regard to Mr. Argosino’s afternoon visit on September 30 that Mr. Argosino had come in requesting that appellant put him on medical leave of absence. When appellant asked him what his problems were, Mr. Argosino was “very evasive” and was acting “very nervous,” was standing up even when told to sit down, was “constantly rubbing his hands” and had overall facial expressions and mannerisms indicative to appellant of Argosino’s being “extremely nervous.” Upon appellant’s inquiry as to what Mr. Argosino’s problems were, the latter replied that he actually had no problem. Appellant asked, “Don’t you have any problems?” to which Mr. Argosino then replied that he was “dizzy.” When appellant asked him concerning his dizziness, Argosino denied that he was dizzy. At the time of this visit appellant explained that his back office medical assistant had suddenly been taken ill, appellant was all alone in his back office, and that he also had an emergency to see. He explained that Mr. Argosino was taking up “a lot of my time. First he said he was all right. Then he said he wasn’t. He was very evasive ... I had a feeling this was going to go on forever. So, I said, ‘Don’t you have any problems at all?’ ” Because of this evasiveness, appellant told Mr. Argosino to leave. Appellant denied telling him “to think about it or call me back if you have anything wrong with you.”2
Appellant testified with regard to the September 30 visit that when Mr. Argosino asked, “Suppose I say I’m dizzy” and then said he had no problem, appellant told him “if he didn’t have any valid medical reasons, I couldn’t treat him” and did not give him a certification on September 30. Appellant observed no faintness nor did Mr. Argosino complain of nightmares, diarrhea, blurred vision, insomnia, tingling of his hands or feet, or shortness of breath, which is found in many people with heart conditions. [Supp. 34]*Supp. 34Appellant, unaware of any history of illness, anything physical, psychological, emotional, mental or of any injuries that Mr. Argosino might have suffered, described as “inaccurate” Mr. Argosino’s testimony that appellant “should” tell him what was wrong with him. Appellant explained, “Now, I know better. I’m supposed to be treating the patient. It is inaccurate—not a patient treating me. How am I to know what is wrong with him? I have asked him that over and over. ...”
On the afternoon of the following day, according to Mr. Argosino’s testimony, he returned to the clinic but found it closed. Nonetheless, he repeatedly banged on appellant’s door, which appellant finally opened halfway. Appellant stated to Mr. Argosino, “We are closed” and pointed to a sign reflecting office hours. The sign was posted on the door; however, Mr. Argosino testified that he had not noticed it. He told appellant, “I am your patient, the day before, and you told me to come back if I could think of something that was wrong with me. I think I have an ulcer or some stomach problems. That’s why I came back.” Mr. Argosino was given an appointment for the following day.
With regard to this incident appellant testified that he thought Mr. Argosino’s banging on the closed office door was “a little strange” because there was sign very prominently posted right in front of the door with the office hours written on it. Appellant was reluctant to let Mr. Argosino in because appellant, having suffered two prior robberies, was “very afraid that he might rob me or do bodily harm.” Appellant explained that Mr. Argosino was not then his patient.
According to Mr. Argosino, he returned on October 2, 1981. At this timé appellant asked, “What’s really wrong with you?” and said, “You have got to have something wrong with you, now.” Mr. Argosino replied, “I think I got some ulcer or stomach pains, stomach problem.” At this appellant “grabbed” Mr. Argosino’s right arm and took his blood pressure; Mr. Argosino, though denying that he offered resistance, admitted that he did pull his arm back in response. As appellant was taking his blood pressure, Mr. Argosino again mentioned that there was nothing wrong with him and that he had only come to see appellant in order to get his medical slip for disability. Appellant nonetheless continued to examine Mr. Argosino, rolled up his T-shirt and pressed a stethoscope to his stomach, appellant moving the stethoscope around the stomach. Mr. Argosino testified that it was “as if he didn’t hear me” with regard to his denial that there was anything wrong with him and repeated to appellant that he had only come to get his medical slip.
According to Mr. Argosino’s testimony, appellant asked if he was nervous, to which Mr. Argosino responded, “No,, doctor, not really.” At the [Supp. 35]*Supp. 35end of the five-or six-minute visit, Mr. Argosino asked appellant how his blood pressure was; appellant replied, “Your blood pressure is good. You’re okay.” Appellant told Mr. Argosino to give his disability form to his receptionist. Leaving the forms with her, the witness obtained an appointment for October 23, 1981.3
Mr. Argosino testified that he was in good physical and mental health when he visited appellant, denied being nervous at these times but did admit that he made an habitual movement with his hand while waiting for appellant and that he talked continuously while appellant was checking his blood pressure and stomach on October 2. He further testified that appellant never examined his neck area, throat, ears, or reflexes nor did appellant discuss with him his dizziness or other physical or mental symptoms, such as fainting, unusual fatigue, depression, or phobias. Appellant never discussed with Mr. Argosino chest pains, sexual problems, nor the latter’s work at Robert Shaw. Appellant took no X-rays, laboratory tests nor did he prescribe medication or a change in the witness’s way of life. Finally, Mr. Argosino denied standing up while in appellant’s office, talking fast or moving around continuously.
On the other hand, appellant testified that he had been in general medical practice for 24 years, saw a lot of Hispanic patients and actually saw about 50 patients a day, including 1 to 3 disability patients. His patients represented a “cross-section of Los Angeles,” and included people from Palos Verdes, Chinese and Black patients and many factory workers from Robert Shaw, the canneries and Max Factor. He testified that he had experience with Robert Shaw assemblers as patients. These patients had disability problems and described their work as “very monotonous” and their working conditions as being “very noisy;” he described the workers at Robert Shaw as having acute anxiety states, by which he meant to describe a condition occurring “when a person is subjected to situational difficulties under constant stress and strain.” He described his Robert Shaw factory patients as being under “constant stress and strain” and described the acute anxiety state as having an acute onset, like the death of a family member or social problems. He further testified that many minority patients live in substandard housing conditions or suffer illness in the family or other problems with the family, such as unemployment, drinking, child abuse, financial problems or accidents.4
[Supp. 36]*Supp. 36Appellant testified that he frankly was scared of Mr. Argosino and did not want to see him, but had thought that if a person had gone through the trouble to see him, that person might have a “problem.” Thus, appellant had consented to see Mr. Argosino, who complained of stomach pain on October 2. According to appellant, Mr. Argosino was nervous and said that he had been worrying because he thought he might be laid off. Mr. Argosino told appellant that he had some financial problems. However, Mr. Argosino continued to be very evasive. He was very suspicious and did not answer the doctor’s questions regarding an oral history. For this reason, appellant thought Mr. Argosino might be a drug abuser and, on the pretense of taking his blood pressure, appellant had looked for needle marks on Mr. Argosino’s arm. Mr. Argosino “kind of resisted” the taking of his blood pressure and mumbled very repetitiously, “Doc, I’m OK. I am OK, Doc.” Suspecting that Mr. Argosino was trying to hide something, appellant tried not to listen to him; appellant wanted to concentrate on what he was doing because he thought he might be dealing with a drug addict. Mr. Argosino was not acting “like the normal person, he kept resisting me, his muscles were tight to me, indicative of a nervous person.” After carefully examining for any needle marks or any signs of a scar on Mr. Argosino’s arms, appellant then examined the latter’s stomach because he had complained that he thought that he might have an ulcer. Appellant checked Mr. Argosino’s liver, using his stethoscope and his hand in palpating. Appellant found no pain in the stomach and checked Mr. Argosino’s pancreas. At this point appellant was not “very convinced” that Mr. Argosino had an ulcer because he did not say anything about pain. Because of his evasiveness, appellant thought that Mr. Argosino might have an alcohol or even have a mental problem. He explained that alcoholics will deny that there is anything wrong with them. Appellant did not know whether Mr. Argosino actually had an ulcer or something else; appellant thought that Mr. Argosino might possibly have had a sexual problem.
Appellant did not find anything that was physically wrong with Mr. Argosino, but remained convinced that Mr. Argosino was “covering up something” because of his evasiveness. Appellant testified: “It was simply impossible to get any history from him. He was so inconsistent. First, he would say he’s nervous, he’s dizzy. Then, he would say he’s all right. When I am examining him I believe he said, T don’t want to be examined. I am all right.’” Because of Mr. Argosino’s uncooperativeness, inconsistent statements and refusal to reply when questioned about an oral history, appellant had no definite plan of treatment but did put Mr. Argosino on disability for the reason that “I felt that he may have had some problem, maybe not physical but a mental problem or a drinking problem.”
Appellant testified that he did check Mr. Argosino’s heart with a stethoscope and checked for palpitations and for arythmic beats. Although he did [Supp. 37]*Supp. 37not take Mr. Argosino’s temperature and did not check his eyes “close up” nor his ears, appellant did notice sweating but not excessive breathing. Appellant testified that Mr. Argosino was flushed, very fidgety, always moving, was very “hyper” like a “hyperactive child,” was talking fast but did not complain of inability to concentrate or tightness in his throat. Appellant testified that Mr. Argosino did complain about dizziness on September 30, 1981, and explained, “How can I treat anyone if they don’t tell me what’s wrong with him?. ...” Appellant denied suggesting, on September 30, dizziness to Mr. Argosino and explained that he did not order tests nor prescribe medication or a special diet for Mr. Argosino or refer him to a psychiatrist or psychologist or for counselling because Mr. Argosino was “so evasive, it was rather hard to actually arrive at the correct diagnosis.” For this reason appellant ordered no blood tests to check for anemia or physical injury or infection. “Oftentimes, people will tell you that they don’t want it. I think there is too much emphasis placed on diagnosis, laboratory tests and procedures to arrive at a correct diagnosis.” The reason that appellant did not take a blood test from Mr. Argosino was because he believed he might have a “mental problem.”
According to Mr. Argosino’s testimony, despite an appointment for October 23 with appellant, he returned to the clinic on October 21. The receptionist refused his request to see appellant for this reason. He then rescheduled the appointment for October 28, when again he reported to appellant’s office. At this time, according to Mr. Argosino, appellant asked him how he was feeling. Mr. Argosino raised two hands and replied, “I’m great, fine.” He told appellant that he had just returned from Las Vegas, where he had been looking for work, and asked appellant for an extension of his medical leave. Appellant asked, “How about if I give you two weeks?” After Mr. Argosino thanked appellant, he exclaimed, “Doctor, you are beautiful.” Appellant then offered to give Mr. Argosino more than two weeks. Mr. Argosino accepted the offer. Appellant directed him to submit his “pink slip” and appellant would take care of it.5 The October 28 visit lasted less than three minutes, according to Mr. Argosino.
Nancy Sargent, an employee of the clinic, testified for the defense with regard to the October 28 visit. On Mr. Argosino’s chart for that date, she entered, “In for checkup. Feels fine” and wrote his blood pressure reading which she had taken. She testified, however, that Mr. Argosino “seemed very nervous, very fidgety in the room, when I went to take his blood pressure; at first he was hesitant to take off his jacket and that I noticed right off the bat.” She stated that she had suspected that Mr. Argosino was [Supp. 38]*Supp. 38a drug user, and looked at his forearm to see if there were any needle marks or any tracks.
Appellant, on October 28, looked at the blood pressure reading taken by Ms. Sargent and asked Mr. Argosino how he was doing. Mr. Argosino stated that he was feeling better, pointed to his stomach and said, “Better.” Appellant testified: “He looked better .... I honestly felt that the rest I gave him was really helping him . . . .” According to appellant’s testimony, he told Mr. Argosino that he should be able to work if he was better. Mr. Argosino replied at this point, “Not better.” Appellant asked, “What?” and Mr. Argosino repeated, “Not better.” According to appellant, Mr. Argosino was still nervous. Appellant extended his leave of absence for this reason.
According to Mr. Argosino, he returned on November 11 to appellant’s clinic. When appellant asked how Mr. Argosino was doing, the latter replied that he was feeling better but wanted more time off. Appellant replied, “Now that you are better, you should be able to return to work.” According to appellant, Mr. Argosino then changed his mind, said he was not better, and displayed inconsistencies again. Appellant felt that Mr. Argosino was still “acting a little strange” and therefore granted him an extension of disability. Appellant testified that Mr. Argosino’s facial expressions appeared very nervous, that he kept constantly rubbing his nose, and there was “something about him [with] which I was uncomfortable.”
Ms. Sargent testified in corroboration of appellant: On November 11, she entered on Mr. Argosino’s file, “In for checkup, needs form filled out, and feels pretty good.” She noted that he was “very, very fidgety with his hands, always moving them, constantly rubbing them together and . . . you don’t see that too often with patients.” She wrote on the chart exactly what Mr. Argosino had said but did not write down “because it’s up to the doctor” that he seemed nervous and very fidgety.
Mr. Argosino asked for more disability; appellant thought that it was in the latter’s best interests to have this disability. Appellant explained that if he returned Mr. Argosino to work in that condition and if something happened to him, and being mindful of the media coverage on the John Hinckley case, which had occurred during that period, and having seen pictures of Jim Brady, the Press Secretary who was shot, appellant feared Mr. Argosino’s “strange behavior” and therefore believed he should be kept under “close observation.”
Another appointment was arranged for Mr. Argosino, which he failed to keep. Because appellant was concerned, he called Mr. Argosino’s employer [Supp. 39]*Supp. 39and found that he did not work at Robert Shaw. Appellant then went over to the place that Mr. Argosino had listed as his residence, which was near appellant’s office and the nursing home where appellant had patients. Appellant could not find Mr. Argosino’s name on the directory at the listed place of residence.
Appellant admitted that this visit came after he had received a state disability inquiry form on November 17 and that he had gone to the place of residence on December 2, but still thought that Mr. Argosino was a bona fide patient, not an undercover agent for the State Disability Fraud Unit. Appellant testified that his plan was to return Mr. Argosino directly to work if he did not follow appellant’s advice to see another doctor, primarily a psychiatrist. Appellant explained that he would have felt responsible if he had returned the man to work without any psychiatric counseling should something happen to him or should he become a menace to society. Appellant thought it “strange and inappropriate” that Mr. Argosino had been very repetitious, calling him “Doc, doc, and doc,” being very evasive and, when appellant had extended his leave, saying, “Doc, you are beautiful” and waving his arm “like a hyperactive child.”
Appellant admitted signing three disability forms for Mr. Argosino. Concerning medical entries on Mr. Argosino’s chart, appellant denied altering the chart (People’s exh. 22) following his arrest, though he did admit to writing additional notations in the chart because of his concern that he might be sued for malpractice should Mr. Argosino do something dangerous to himself or to others. Appellant acknowledged that he had made these notations after learning that Mr. Argosino did not work at Robert Shaw or live where he had indicated. He explained that he was “very confused” and “annoyed” by Mr. Argosino’s behavior and evasiveness because appellant had “really sick” patients to treat and that it was simply impossible to treat Mr. Argosino.
Over defense objection, the prosecution was permitted to present evidence concerning a statistical comparison of diagnoses contained on 196 disability claims certified by appellant as compared to the same number of diagnoses on disability claims certified by other health practitioners. Seventy-five percent of these claims came from the Long Beach Disability Insurance Office. These claims were divided into five categories. Category 1 indicated “anxiety states” and/or “headaches” and also included “acute anxiety states;” category 2 included “anxiety states” and/or “headaches;” category 3 included “depression” or any other mental problem; this grouping was broader than categories 1 or 2.6 Category 4 was divided into subparts: category [Supp. 40]*Supp. 404-A corresponded to category 1 but required a statement of physical injury or illness; 4-B corresponded to category 2 with the addition of a statement indicating a physical injury and/or illness; 4-C corresponded to category 3 plus any statement about physical injury and/or illness. Category 5 included all other diagnoses not covered in categories 1 through 4 and necessarily meant a purely physical injury or illness diagnosis.
In category 1 there were 46 claims from Dr. Louie and none from the sampling of other doctors’ diagnoses: Category 2 contained six claims from appellant and none from the other sampling. Category 3 had one claim from appellant; and three from the other sampling. Category 4-A had 41 from appellant and three from the sampling of other doctors’ diagnoses; 4-B had seven claims from appellant, two from the other group; 4-C had three from appellant and nine from the other group sampled; category 5 had 92 from appellant and 179 from the other sampling. The total number of all claims of appellant involving categories 1, 2 and 3 were 53 as compared to three from the other sampling. The total for the first four categories was 104 for appellant, 17 from the other sampling, which converted to 53 percent for appellant and 8.7 percent for the other sampling. As to categories 1 through 3, appellant’s percentage was 23.5 while that for the sample was 1.5.
All of the samples of the other doctors’ diagnoses came from the Long Beach office; appellant’s claims came from various offices in the Southern California area, including 25 from Los Angeles offices, one from San Bernardino, 15 from Whittier, two from Van Nuys, three from Santa Ana, and two from Santa Monica. The sample group did not consider Spanish-surname patients, the percentage of black patients, or women, the patients’ histories or their occupations, nor did the sample group include the patients’ areas of residence.
In addition, no statistical analysis was conducted concerning the percentage of patients in the two groups compared who worked for Robert Shaw Controls, the percentage of patients who were assemblers in factories, or the percentage of patients who were on disability for pregnancy. In fact, nothing was known about the patients included in the sample.
Moreover, no determination was made in the random sample concerning the type of practice or type of the doctors (whether chiropractic or medical) who were certifying people for disability. At least two were chiropractors. There was no percentage breakdown of the doctors who were general practitioners as opposed to doctors in other areas of medical expertise.7
[Supp. 41]*Supp. 41According to a prosecution witness the reason why these factors were not included was “basically that they were irrelevant. We were testing just what the diagnosis was as stated by the certifying doctor.” This witness testified that there was no way of knowing a patient’s race based on the information on the claims form. With regard to the Spanish-surname patient “it wasn’t felt that whether a person had a Spanish surname or not would make any difference to what the doctor’s diagnosis would be.” He did admit that the kind of work that patients did might affect the diagnosis but explained: “Well, it would depend on their history. The history that they would give the doctor may or may not have a bearing, actually, on the kind of work they do. If the history they gave the doctor was related to their work, they would not or should not be certified for state disability, but, rather, for Worker’s Compensation because if their disability is in any way arising out of [or] in the course of their employment, it would come under the purview of the Worker’s Compensation.”
Every 50th disability certification was pulled in numerical sequence based upon the last four digits of the Social Security number shown on the certification. The bulk of the claims were for the period between July, 1981 and June, 1982. Six pregnancies, included in category 5, were shown in appellant’s certifications and 26 on the random sample. Twenty fractures were shown on the random sample and none on any of appellant’s 196 certifications.
Based upon the foregoing comparison, the prosecution called a mathematics professor, who testified that the likelihood that “random chance” alone rather than some other factor accounted for the difference of 25 percentage points between the category 1-3 diagnoses of Dr. Louie as compared to the same categories of diagnoses in the other-group sampled was three in 10 million, “which is, for a statistician, a zero likelihood.” The witness further testified that the likelihood of “random chance” alone accounting for a difference between 53 percent and 8.7 percent was “off the tables. I really don’t know what the number is, but it exceeds—probably in many billions, one over many billions, the probability of that happening. One chance in many billions, I would guess. The number is so small I don’t think you are going to find it in any table. ...”
As to category 1, the comparison of 23.5 percent for Dr. Louie and zero for the other-group sample, the witness testified, “the probability is less than four times in a million” that that discrepancy would happen “by chance.” Denying that for statistical purposes there could be any variation depending upon the subjective perceptions of the doctors involved, the witness testified “the samples don’t care who wrote the numbers down. The numbers are there and it doesn’t matter if you had 10,000 different doctors. [Supp. 42]*Supp. 42When you pull those numbers out, there is some true of all the files. If 1.5 percent was the correct number, if you took every file and checked you would find 150 of the files would have that category. It’s irrelevant how those files were written and who wrote them.” He further testified that absolutely “some variations” will result if there is any bias between the sample taken and the one against which one is actually checking.8
The prosecution’s witness stated that statistics are not concerned with the subject matter but simply the likelihood of producing certain proportions when one samples. He explained that his testimony concerned only the likelihood that such a difference would be due to “random chance” and to no other cause. As to appellant, the witness had not the faintest idea what the explanation for the difference was. In his professional opinion, however, the explanation would have to be “other than chance. ” He did acknowledge that it was certainly “possible” that appellant might have a patient population that had, for example, 23.5 percent anxiety states or headaches and that “there could be any other explanation other than chance.” He further acknowledged that the “numbers seem to say to me that there is an explanation other than chance.”
The defense presented a statistical expert who testified that the methodology used in compiling the randoms suffered from the following errors: (1) there was a mixture of physicians other than physicians in general practice such as appellant; (2) there were no “controls” on the patient population of the sample (such as age or occupation) to assure comparability with appellant’s patient population; and (3). there was no statement explaining how cases were selected when there was more than one in a folder. The witness conceded that it would not be necessary to match appellant’s patient population with a sample in every variable or factor, and that as the size of the sample is increased, if done correctly, the laws of chance alone would achieve the representation of that factor correspondingly closer to the actual percentage of the population.
Discussion
1. Instructional Errors Require Reversal of the Judgment.
Appellant contends that the trial court was duty bound to instruct the jury that a certification prepared by defendant in good faith and in the [Supp. 43]*Supp. 43belief that it was true was not a violation of section 2101 of the Unemployment Insurance Code even though the certification was in fact mistaken.
Section 2101 of the Unemployment Insurance Code declares that wilfully making a false statement or representation, or knowingly failing to disclose a material fact, to obtain any benefit or payment constitutes a misdemeanor. (See fn. 1.) Under section 2101 the prosecution must prove two elements beyond a reasonable doubt: (1) An intent to defraud and (2) the wilful making of a false statement or representation. (People v. Lustman (1970) 13 Cal.App.3d 278, 286 [91 Cal.Rptr. 548] [overruled on another ground by People v. Ruster (1976) 16 Cal.3d 690, 697 [129 Cal.Rptr. 153, 548 P.2d 353, 80 A.L.R.3d 1269]].) It is, of course, a defense to an allegation of intent to defraud that the accused acted without such intent but in the good faith belief that his conduct was justified.
It is well settled that in a criminal prosecution, even in the absence of a request, the trial court must instruct on the general principles of law relevant to the issues raised by the evidence. The general principles governing the case are those closely and openly connected with the facts before the court, and which are necessary for the jury’s understanding of the case. Included within the trial court’s duty is the obligation to instruct on defenses and on the relationship of those defenses to the elements of the charged offense, where it appears that the defendant is relying on such a defense or if there is substantial evidence which supports the particular defense. (People v. Stewart (1976) 16 Cal.3d 133, 140 [127 Cal.Rptr. 117, 544 P.2d 1317]; see also People v. Flannel (1979) 25 Cal.3d 668, 680-681 [160 Cal.Rptr. 84, 603 P.2d 1].)
We agree with appellant’s contention that the trial court failed to discharge its duty to instruct on appellant’s theory of defense. Although the trial court properly instructed the jury in the statutory language of section 2101 and expressly defined the specific intent required in order to convict under the section, the trial court failed to instruct the jury that if appellant possessed a good faith belief in the disability of Mr. Argosino in certifying the latter’s disability claims, appellant was not acting with the required intent to defraud.
In People v. Stewart, supra, 16 Cal.3d 133, our Supreme Court reversed a conviction of nine counts of grand theft based upon prosecution evidence tending to show that the defendant had embezzled funds from his employer. Despite the defendant’s testimony that he thought that he had been authorized to use the funds for his own investments, the trial court failed to give sua sponte an instruction informing the jury that a good faith belief in the authority to appropriate and use property which a defendant is accused of [Supp. 44]*Supp. 44embezzling eliminates fraudulent intent, a necessary element of the crime of embezzlement. (Id., 16 Cal.3d at pp. 139, 140, 142.) In the case at bench the trial court similarly failed, despite substantial evidence supportive of the theory of this defense, to advise the jury that appellant’s good faith belief in Mr. Argosino’s disability constituted a complete defense to the charge of violating section 2101 of the Unemployment Insurance Code. A finding of such good faith belief would, of course, negate the possibility of finding that appellant acted with the required intent to defraud.
Compounding the trial court’s error, the court below also gave, over defendant’s objection, CALJIC No. 7.22, which derives from Penal Code section 125 and provides that an unqualified statement of that which one does not know to be true is equivalent to a statement of that which one knows to be false. Such an instruction has been held to be improper in a prosecution for perjury because it absolutely eliminates the whole question of criminal intent, which is “the vital element in every crime of perjury.” (People v. Von Tiedeman (1898) 120 Cal. 128, 134 [52 P. 155].) As the Von Tiedeman court aptly observed with regard to the perjury prosecution involved in that case, “Under the charge as given, a person with the most honest intentions could be convicted. A man honestly mistaken could be guilty of the crime [of perjury].” (Id., at p. 134.)
The foregoing analysis applies with equal force to the case at bench. Although, as noted, the trial court properly instructed on the requirement of specific intent, the court’s further instruction based upon CALJIC No. 7.22 effectively permitted the jury to find such specific intent to defraud based only upon appellant’s disability certifications of Mr. Argosino, when appellant himself admitted in his testimony that he did not definitely know what was wrong with the patient. Thus, if appellant’s diagnosis were mistaken, as it was shown to be through the prosecution evidence, the jury could find, based upon CALJIC No. 7.22, the specific intent to defraud solely on account of appellant’s mistake, no matter how honestly made. The instruction under CALJIC No. 7.22, without a good faith instruction accompanying it, in this case deprived appellant of his right to a jury determination of the issue of appellant’s specific intent to defraud. (See CALJIC No. 7.24.)9
The People urge that appellant was not prejudiced by the instructional errors committed by the trial court “since all the evidence herein manifestly [Supp. 45]*Supp. 45indicates that appellant obviously knew that quick and facile certifications of the undercover agent’s ‘disability’ was utterly without any medical basis other than totally unfounded and unprofessional speculation put to rest (and belied) by the candor of the undercover agent himself.” We disagree. Unless instructional error is otherwise cured by other properly given instructions which enable the trier of fact properly to resolve the factual issue in dispute, it cannot be concluded that such instructional error which goes to the heart of the defense case is harmless. (Cf. People v. Stewart, supra, 16 Cal. 3d 133, 141-142.)
The People’s reliance on People v. Hill (1967) 67 Cal.2d 105, 119 [60 Cal.Rptr. 234, 429 P.2d 586] is wholly misplaced. In Hill the trial court erroneously gave both a specific and a general intent instruction. Under the evidence presented in that case, however, the jury could only conclude that the defendant had the required specific intent. By contrast, the evidence at bench does not point irresistibly to the conclusion that appellant acted with the required fraudulent intent. Under proper instructions the jury could well have reasonably acquitted him on the evidence presented.
2. The Introduction of Statistical Probability Evidence Based Upon Comparative Diagnoses of Other Health Practitioners Constituted Prejudicial Error.
Appellant contends that the trial court improperly admitted evidence concerning the comparative diagnoses of appellant vis-a-vis 196 other health practitioners, and that the admission of this evidence, accompanied by the statistical probability testimony of the prosecution’s expert, constituted prejudicial error. Appellant is correct.
Over defense objection, the evidence complained of was admitted following an oral offer of proof by the prosecution. The court held no evidentiary hearing out of the jury’s presence, as the court was authorized to do under Evidence Code section 403, subdivision (a)(1).10, 11 Instead, the court be[Supp. 46]*Supp. 46low admitted the statistical comparison evidence solely upon the prosecutor’s representation that the evidence came within the exception of Evidence Code section 1101, subdivision (b) with regard to other-acts evidence.12 The prosecution contended that it was proper to admit the statistical comparison evidence arid to adduce statistical expert testimony to eliminate the possibility that chance could have caused the discrepancies in diagnoses of anxiety states by Dr. Louie and those diagnoses of other health practitioners. We disagree.
The admissibility of evidence is, of course, governed by the provisions of the Evidence Code. Evidence Code section 350 provides that no evidence is admissible unless it is relevant.13 Relevant evidence “means evidence having any tendency in reason to prove or disprove any disputed fact that is of consequence to the determination of the action.” (Evid. Code, § 210.)14 In the case at bench the crucial issue for the jury’s determination was the question whether Dr. Louie had certified Mr. Argosino for disability with the specific intent to defraud. Evidence which tended to prove or disprove this issue was clearly relevant under section 210 of the Evidence Code.
Having reviewed the complained of evidence in light of the prosecution’s offer of proof, we fail to perceive how the evidence of statistical comparison and the prosecution’s expert opinion based thereon had any tendency in reason to prove the issue of specific intent.
As the proponent of the evidence the prosecution had the burden of establishing preliminarily that appellant fraudulently diagnosed his other pa[Supp. 47]*Supp. 47tients for whom disability claims were submitted (Evid. Code § 403, subd.(a)(1), see fn. 11, ante). The prosecution made no such showing and therefore the evidence that appellant’s diagnosis in the present case was fraudulent was not capable of proof by the means which the prosecution employed. Simply stated, the statistical probability evidence that “random chance” could not explain the discrepancy between appellant’s diagnoses and those of others lacked relevance because the absence of proof that appellant’s diagnoses had been fraudulently made rendered the inference sought to be established without evidentiary support. (Cf. People v. Riser (1956) 47 Cal.2d 566, 577 [305 P.2d 1]; People v. Henderson (1976) 58 Cal.App.3d 349, 360 [129 Cal.Rptr. 844]; see People v. De La Plane (1979) 88 Cal.App.3d 223, 240 [151 Cal.Rptr. 843].) As Justice Jefferson has observed in 1 California Evidence Benchbook (2d ed. 1982) section 21.3, page 502: “Evidence is irrelevant if it has a tendency to prove or disprove a disputed fact of consequence only by reason of drawing speculative or conjectural inferences from such evidence. The concept of a ‘tendency in reason’ to prove a disputed fact, required by Evid C § 210, necessarily means that the deduction or inference to be drawn from the proffered evidence to the existence or nonexistence of a disputed fact is a reasonable inference. If the inference of the existence or nonexistence of a disputed fact which is to be drawn from proffered evidence is based on speculation, conjecture, or surmise, the proffered evidence cannot be considered relevant evidence. If proffered evidence can cause the trier of fact only to speculate from such evidence as to the existence or nonexistence of a disputed fact, such evidence is irrelevant and inadmissible, since it does not come within the definition of ‘relevant evidence’ set forth in Evid C § 210.” (Italics added.) Even the prosecution’s own expert established that the statistical discrepancy in diagnosis of “anxiety states” and/ or headaches could be attributable to the particular characteristics of appellant’s patient population or to other causes wholly unrelated to fraudulent intent. This statistical comparison evidence therefore only had the effect of permitting the trier of fact to find fraudulent intent by the impermissible means of drawing speculative or conjectural inferences. (Cf. People v. Albertson (1944) 23 Cal.2d 550, 576-579 [145 P.2d 7].)
In addition to the prosecution’s failure to establish as a preliminary fact that appellant’s 196 other disability diagnoses of anxiety state were fraudulently made, the statistical probability evidence, and the prosecution’s expert opinion based thereon, lacked probative value and should not have been admitted for yet another reason: The statistical comparison of diagnoses was compiled from health practitioners whose practices were not shown to be similar to that of appellant; the comparison was based upon populations of patients not shown to be comparable by way of ethnic background, physical and mental condition, socioeconomic status, work, sex, age or geographical [Supp. 48]*Supp. 48area. Such variables must be quantified “exactly” if mathematical probabilities are to have any utility in the courtroom. (.People v. Celia (1983) 139 Cal.App.3d 391, 405 [188 Cal.Rptr. 675].) The absence of quantification of crucial variables in the case at bench rendered the evidence useless in terms of proving the issue of appellant’s intent. (Cf. People v. Collins (1968) 68 Cal.2d 319, 329-331 [66 Cal.Rptr. 497, 438 P.2d 33, 36 A.L.R.3d 1176].) The lack of exact quantification of variables in the case at bench was as scientifically inaccurate as the “estimates” improperly employed in Collins.
As our Supreme Court has cautioned, application of statistical probability theory, particularly in a criminal case, must be “critically examined” in view of the substantial unfairness to a defendant which may result from ill-conceived techniques with which a trier of fact is not technically equipped to cope. (People v. Collins, supra, at p. 332.) The critical examination which Collins enjoins leads us to conclude that the failure to quantify all of the crucial variables rendered the statistical probability evidence and the prosecution testimony based on it utterly useless as proof of criminal intent.15
Even assuming that the comparative statistical analysis evidence and probability theory testimony offered by the prosecution were relevant under section 210, we would be constrained to hold that the trial court nonetheless abused its discretion under Evidence Code section 352 in determining that the probative value of this evidence outweighed its prejudicial impact. 16
The law is clear that the admissibility of other-acts evidence must be carefully examined and that merely finding an admissible purpose for the evidence does not automatically authorize its reception. (People v. Guerrero (1976) 16 Cal.3d 719, 724 [129 Cal.Rptr. 166, 548 P.2d 366]; see also People v. Green (1980) 27 Cal.3d 1, 24-26 [164 Cal.Rptr. 1, 609 [Supp. 49]*Supp. 49P.2d 468]; People v. Gibson (1976) 56 Cal.App.3d 119, 126-131 [128 Cal.Rptr. 302].) Such evidence must be received with “extreme caution” and if its connection with the charged offense is not clearly perceived, the doubt should be resolved in favor of the accused. {People v. Guerrero, supra, 16 Cal.3d 719.)
The admission of the evidence in the case at bench inevitably resulted in prejudice to appellant. Depending essentially on a single witness, whose credibility was impeached in large measure by defense evidence, the case against appellant on the critical issue of intent was wholly circumstantial and depended entirely on the jury’s determination of appellant’s and his witness’ credibility against that of the undercover agent, Mr. Argosino. On this basis, it is likely that the seductively superficial statistical evidence tipped the balance on the credibility issue in the prosecution’s favor despite the absence of any indication of fraudulent intent in certifying the 196 disability claims certified by appellant. (Cf. People v. Collins, supra, 68 Cal.2d at p. 332.)
On the basis of the inadmissible evidence the jury must have concluded that appellant was engaged in a large scale scheme of defrauding the state by means of filing false disability claims—a belief fostered by the prosecution’s misuse in closing argument of the statistical data to impugn appellant’s character: Thus, although acknowledging, on the one hand, that he was not arguing that the statistical evidence proved that appellant wrote “phoney anxiety state certifications,” the deputy district attorney emphatically argued, on the other, that appellant was “simply overreaching, that he was simply running an acute anxiety mill, issuing very easy leaves from work, increasing that patient load a little bit more,” and that the evidence “does show that the doctor simply was playing fast and loose with disability certifications for acute anxiety states.” The prosecution referred to the Argosino diagnosis of acute anxiety state as “the doctor’s old favorite diagnosis” and emphasized that there was a “fairly obvious explanation” for the statistical discrepancy, that explanation being that appellant “must be a little bit known as a doctor who will grant a vacation if you need one” and that “charitably” appellant had “been a soft touch for acute anxiety certifications.”17
[Supp. 50]*Supp. 50Although the trial court did give a modified version of CALJIC No. 2.50,18 we conclude that the prosecution’s argument effectively undermined whatever protection may have been available to appellant under that instruction. The deputy district attorney’s assertions deprived appellant of the right to a jury determination of the question of guilt or innocence upon a proper consideration of the evidence before it and invited the jury to convict appellant upon its conception of him as a person of criminal disposition and of bad character; thus, the prosecution’s misuse of the evidence conflicted with Evidence Code section 1101, subdivision (a), which bars evidence of bad character and criminal disposition to prove conduct on a specified occasion.
Having reviewed the entire record, we are of the opinion that the inadmissible evidence, the prejudicial impact of which was compounded by the prosecution’s closing argument, deprived appellant of his right to a fair trial and requires a reversal of the judgment of conviction (order granting probation).
3. Other Contentions.
We have examined the record with regard to appellant’s other contentions and conclude that no error occurred in permitting the jury to have copies of Mr. Argosino’s medical chart during cross-examination of appellant. This procedure permitted the jury to follow the testimony more clearly and permitted the jury to conclude for itself whether appellant had altered the document and, if so, whether this action constituted evidence of a consciousness of guilt. (People v. Isby (1947) 30 Cal.2d 879, 892 [186 P.2d [Supp. 51]*Supp. 51405]; see also People v. Boggs (1967) 255 Cal.App.2d 693, 707 [63 Cal.Rptr. 430].) Moreover, the trial court extensively admonished the jury not to give undue emphasis to the medical chart.
Concerning People’s exhibit 1, we find no indication that appellant could have been prejudiced by the inclusion on the exhibit of the penalty for a violation of Unemployment Insurance Code section 2101. The trial court’s instruction that the jury was not to consider penalty or punishment was presumably followed by the jurors (Evid. Code, § 664); appellant has presented no evidence to rebut this presumption.
The judgment (order granting probation) is reversed.
Reese, P. J., and Cooperman, J., concurred.