Peters, C. J.
The principal issue in this medical malpractice case is whether the “continuous treatment” or “continuing course of conduct” doctrine tolls the repose section of the statute of limitations, General Statutes § 52-584,1 which disallows any action brought more than three years from the date of a physician's álleged misconduct. The named plaintiff, Evelyn E. Blanchette,2 filed a complaint charging the defendant, [259]*259Frederick C. Barrett, a physician, with having negligently failed to exercise reasonable medical care in his diagnosis and treatment of her breast condition, which was subsequently found to have been cancerous.3 The [260]*260defendant denied the allegations in the complaint and raised two special defenses: the plaintiff’s contributory negligence and the expiration of the statute of limitations. The case was tried to a jury, which found for the plaintiff on her allegation of negligence and for Ronald Blanchette, the plaintiff’s husband, on his derivative claim for loss of consortium.4 The trial court denied the defendant’s motions to set aside the verdict and for a new trial, and for judgment notwithstanding the verdict, and rendered judgment for the plaintiff. The defendant filed a timely appeal to the Appellate Court, which we transferred to this court pursuant to General Statutes § 51-199 (c) and Practice Book § 4023. We reverse the judgment of the trial court and order a new trial.
The undisputed facts are as follows. The defendant became the plaintiff’s family physician in 1973, and saw her professionally at various intervals between 1973 and 1985. These professional consultations routinely included a breast examination. The plaintiff’s last professional consultation with the defendant took place on January 10, 1985. In accordance with the defendant’s direction at that consultation, the plaintiff had a mammogram. On January 15, 1985, the defendant received the mammogram report, which was negative, and so informed the plaintiff.
In May, 1987, concerned about the condition of her breast, the plaintiff called the defendant’s office to [261]*261make an appointment to see him. Because the defendant was unavailable, someone in the defendant’s office referred the plaintiff to Paul Deutsch, another family practitioner. Deutsch examined the plaintiff on May 19, 1987, discovered a mass in the plaintiff’s left breast, and referred the plaintiff to a surgeon, first for a biopsy and shortly thereafter for surgery, which confirmed the presence of breast cancer. The plaintiff commenced her action against the defendant by complaint dated May 17,1989. Further facts will be detailed as relevant.
The defendant challenges the validity of the judgment rendered against him on numerous grounds. First, he maintains that the trial court improperly denied his motions to set aside the verdict and for judgment notwithstanding the verdict, “because the jury could not have reasonably and legally concluded that [the] plaintiff’s action was brought within three years of the allegedly tortious conduct.” Second, he claims that the trial court improperly instructed the jury on the continuing course of conduct doctrine when it stated that the defendant had a duty to notify or warn the plaintiff of a negligent misdiagnosis. Third, he contends that the trial court’s instructions, on numerous occasions, misstated the date of the plaintiff’s last personal consultation with the defendant, and that such misstatements “misled the jury to believe that the defendant’s alleged negligence [had] occurred in 1987 rather than 1985.” Fourth, he argues that, in highlighting the statistical life expectancy table introduced by the plaintiff, the trial court misled the jury to believe that the plaintiff, in 1985, had the life expectancy of an average, normal, healthy white female. Fifth, he maintains that the trial court abused its discretion when it submitted an interrogatory to the jury that called for a conclusion of law.
We disagree with the defendant’s contention that the trial court could not have found sufficient evidence to [262]*262uphold the jury’s verdict, but we agree that the trial court misstated the continuing course of conduct doctrine and that this error in the jury instructions gave the jury inadequate guidance, as a matter of law, to decide whether the plaintiff’s action had been brought within the three year period allowed by the repose section of the statute of limitations, § 52-584. We therefore reverse the judgment in the plaintiff’s favor and order a new trial.5 This disposition makes it unnecessary for us to consider the defendant’s other grounds for reversal of the judgment against him.6
[263]*263I
The defendant claims that the trial court improperly denied his motions to set aside the verdict and for a new trial, and for judgment notwithstanding the verdict. He contends that the plaintiffs lawsuit, filed on May 17,1989, was barred by the repose section of the statute of limitations, § 52-584. That statute provides that an action for the negligence of a physician must be brought “within two years from the date when the injury is first sustained or discovered or in the exercise of reasonable care should have been discovered . . . The statute’s repose section, however, imposes a further time limitation: “except that no such action may be brought more than three years from the date of the act or omission complained of . . . .’’General Statutes § 52-584. The defendant maintains that the three year period of the repose section began to run on January 10, 1985,7 the last date on which the plain[264]*264tiff consulted him personally. From this predicate, the defendant posits that he cannot be held liable for any alleged professional omissions thereafter. The defendant recognizes that, under some circumstances, the continuous treatment doctrine or the continuing course of conduct doctrine serves to toll the statute of limitations, but denies the applicability of either doctrine to his professional relationship with the plaintiff.
The crux of the defendant’s contention is that he engaged in no subsequent wrongful conduct with respect to the plaintiff after January 10,1985. Because wrongful conduct may include acts of omission as well as affirmative acts of misconduct, we disagree that the evidence supports his contention.
We undertake only limited appellate review of a trial court’s denial of a motion for judgment notwithstanding the verdict and of a motion to set aside the verdict. In each case, we accord great deference to the trial court’s superior opportunity to view the trial in its entirety. “In reviewing the decision of the trial court, we consider the evidence in the light most favorable to the sustaining of the verdict. . . .” (Citations omitted; internal quotation marks omitted.) Ford v. Blue Cross & Blue Shield of Connecticut, Inc., 216 Conn. 40, 57, 578 A.2d 1054 (1990); Novella v. Hartford Accident & Indemnity Co., 163 Conn. 552, 555, 316 A.2d 394 (1972). Our function is “to determine whether the trial court abused its discretion in denying [either] motion . . . .” Mather v. Griffin Hospital, 207 Conn. 125, 139, 540 A.2d 666 (1988); Skrzypiec v. Noonan, 228 Conn. 1, 11, 633 A.2d 716 (1993); Ginsberg v. Fusaro, 225 Conn. 420, 425, 623 A.2d 1014 (1993); State v. Hammond, 221 Conn. 264, 270, 604 A.2d 793 (1992). “ ‘The trial court’s [denial of each motion] is entitled to great weight and every reasonable presumption should be indulged in favor of its correctness. . . .’ ” Skrzypiec v. Noonan, supra, 10-11; Ginsberg v. [265]*265Fusaro, supra, 430-31; Bartholomew v. Schweizer, 217 Conn. 671, 687, 587 A.2d 1014 (1991); Mather v. Griffin Hospital, supra, 139.
With these principles in mind, we address the defendant’s claim that the three year repose section of § 52-584 had expired prior to the plaintiff’s commencement of this medical malpractice lawsuit. Concededly, the relevant “date of the act or omission complained of,” as that phrase is used in § 52-584, is “the date when the negligent conduct of the defendant occurs and . . . not the date when the plaintiff first sustains damage.” Vilcinskas v. Sears, Roebuck & Co., 144 Conn. 170, 173, 127 A.2d 814 (1956) (interpreting General Statutes [1949 Rev.] § 8324, predecessor to § 52-584); Prokolkin v. General Motors Corp., 170 Conn. 289, 294-97, 365 A.2d 1180 (1976) (utilizing interpretation of § 52-584 to construe same language in General Statutes § 52-577). In the medical malpractice context, we have specifically determined that a lawsuit commenced more than three years from the date of the negligent act or omission complained of is barred by the statute of limitations, § 52-584, regardless of whether the plaintiff had not or, in the exercise of care, could not reasonably have discovered the nature of the injuries within that time period. Stein v. Katz, 213 Conn. 282, 285, 567 A.2d 1183 (1989); Catz v. Rubenstein, 201 Conn. 39, 49-50, 513 A.2d 98 (1986); McDonald v. Haynes Medical Laboratory, Inc., 192 Conn. 327, 334, 471 A.2d 646 (1984). We have also recognized, however, that the statute of limitations, in the proper circumstances, may be tolled under the continuous treatment or the continuing course of conduct doctrine, thereby allowing a plaintiff to commence his or her lawsuit at a later date. See, e.g., Connell v. Colwell, 214 Conn. 242, 252-55, 571 A.2d 116 (1990); Fichera v. Mine Hill Corp., 207 Conn. 204, 209-10, 541 A.2d 472 (1988); Cross v. Huttenlocher, 185 Conn. 390, 400, 440 [266]*266A.2d 952 (1981); Handler v. Remington Arms Co., 144 Conn. 316, 321, 130 A.2d 793 (1957); Giambozi v. Peters, 127 Conn. 380, 385, 16 A.2d 833 (1940), overruled on other grounds, Foran v. Carangelo, 153 Conn. 356, 360, 216 A.2d 638 (1966). Because the plaintiff last consulted the defendant on January 10, 1985, the defendant maintains that there is no basis for applying either the continuous treatment doctrine or the continuing course of conduct doctrine to hold him liable in this case.
A
The applicability of the tolling doctrines to the facts of this case depends upon the factual circumstances that were presented to the jury.8 The jury reasonably could have found the following additional facts.
The professional relationship between the plaintiff and the defendant, which began in 1973, was one between a patient and a regular family physician. The plaintiff consulted the defendant at various intervals between 1973 and 1985. Both the plaintiff and the defendant recognized the existence of a physician-patient relationship. The defendant was aware of the plaintiffs loyalty to him, her trust in him and her dependence on his professional opinions and advice. Between 1973 and 1979, the plaintiff consulted the defendant at approximately six month intervals for check-ups, prescriptions, monitoring of medications and other illnesses and ailments. These consultations became less frequent after 1979.
At every consultation, the defendant performed a breast examination on the plaintiff. The defendant first observed a notable condition in the plaintiff s left breast [267]*267during an office consultation in 1979. He diagnosed this condition as fibrocystic disease,9 and accordingly entered a finding of “fibrocystic disease left breast” into the plaintiffs medical records. The defendant neither mentioned nor discussed this finding with the plaintiff.
During the course of a regular check-up in July, 1983, the defendant felt a small lump in the plaintiffs left breast.10 This time, he informed the plaintiff of his finding and attempted unsuccessfully to have her feel this lump, by placing her hand over the spot where he had felt it. He told the plaintiff that he had found a cyst and that she should come back if she noticed any change in the cyst. The defendant also made a conscious decision at that time not to advise the plaintiff to have a mammogram although the plaintiff was then forty-four years old.
The next consultation, a regular check-up, occurred on January 10,1985. Shortly before that check-up, the plaintiff had been able to feel a small, hard lump in her left breast, approximately the size of a pea. During the 1985 visit, the defendant also felt a lump in the plain[268]*268tiff’s left breast and told the plaintiff to have a mammogram because the lump had become larger. The plaintiff immediately obtained a mammogram. After receiving a negative mammogram report on January 15, 1985, the defendant informed the plaintiff of this report and told her that she was fine.
Between the 1985 consultation and 1987, the plaintiff considered the defendant to be her family physician and relied on the defendant’s opinion that she merely had a cyst. Because she felt well during that period, she scheduled no further consultations with the defendant. Apart from informing her of the negative results of the 1985 mammogram, the defendant never discussed further care with her and made no arrangements either for another mammogram or for a second opinion with another doctor. The defendant never told the plaintiff anything about monitoring or following up on her breast condition, including the need for a repeat mammogram on an annual basis or the need to return to see the defendant if she noticed any change, and never instructed her on how to perform a breast self-examination. After the 1985 mammogram report, the defendant initiated no contact with the plaintiff in any fashion with regard to any medical matter.
In the early spring of 1987, the plaintiff became concerned about her health because the lump in her left breast had grown to the size of a golf ball and because she was experiencing symptoms that resembled breast cancer symptoms about which she had recently read. Accordingly, in May, 1987, the plaintiff called the defendant’s office to schedule an appointment with the defendant. Because the defendant was unavailable, the receptionist at the defendant’s office informed the plaintiff that she could be seen the next day by Deutsch.11
[269]*269On May 19, 1987, which was the following day, the plaintiff consulted Deutsch with regard to her breast mass. Deutsch found a dominant mass in the plaintiff’s left breast. According to the plaintiff, the lump that Deutsch then found was the same lump that the defendant had felt in 1983 and that she and the defendant had felt in 1985. After unsuccessfully attempting to aspirate the lump, Deutsch referred the plaintiff to Jeffrey Goldblatt, a surgeon, for a biopsy and a mammogram.
Goldblatt examined the plaintiff on May 29,1987. He informed her that the mass in her left breast was probably cancerous and recommended a biopsy to confirm his diagnosis. Goldblatt also sent the defendant a letter that described his findings and diagnosis that the plaintiff had cancer. On June 8, 1987, after having learned of the probable diagnosis but before agreeing to the breast biopsy, Ronald Blanchette telephoned the defendant to ask his opinion of Goldblatt’s advice concerning his wife’s breast condition. The defendant told Ronald Blanchette that he concurred in Goldblatt’s opinion that the plaintiff should have a breast biopsy.
The biopsy that the plaintiff underwent on June 10, 1987, confirmed that she had breast cancer. The next day, a modified radical mastectomy was performed. A subsequent analysis revealed the presence of cancer in twenty-seven of the thirty-one lymph nodes in the plaintiff’s breast. After the surgery, Goldblatt referred the plaintiff to an oncologist for treatment.
To assist the jury’s evaluation of her claim of medical malpractice, the plaintiff presented the testimony of Robert Schneider, a medical expert witness, who addressed the issue of the standard of care of a family physician. He concluded that the defendant had failed to meet the standard of care with regard to breast examinations, record keeping, management and diagnosis of a finding in the breast and follow-up treatment [270]*270of a breast condition. In his opinion, “the [defendant’s] entire pattern of care, especially as to [the above mentioned] points, represented a significant deviation from the standard of care.” He explained that reliance by a physician on a negative mammogram to diagnose a patient as not having breast cancer did not meet the standard of care of the medical profession.
Schneider testified that follow-up on a patient with a breast condition is a very important part of the standard of care. When dealing with a patient with a lump that is suspected to be fibrocystic disease rather than cancer, Schneider stated that the patient should be examined just before and just after her menstrual cycle to determine whether the size of the cyst has changed, which would indicate that it had been filled with fluid and was, therefore, benign. Even with such a finding and a negative mammogram, the patient may still need to be brought back in a short period of time, such as a month, for further examination and testing, including ultrasound, needle aspiration and a consultation for a biopsy, so that the physician can conclusively rule out the possibility of cancer. He explained that although fibrocystic disease feels different than a lump, the physician should not rely simply on his or her sense of touch. Schneider opined that the defendant had deviated from the standard of care by failing to bring the plaintiff back and perform further diagnostic procedures.
Schneider stated that the defendant also had failed to monitor the plaintiff’s condition at sufficiently frequent intervals. In his view, even after definitively diagnosing a patient as having either fibrocystic disease or a totally benign mass in one or both of her breasts, the physician should see the patient at either three or six month intervals.
[271]*271Once a physician determines that follow-up is required, Schneider testified that it is “incumbent upon the [physician] to ensure that the patient returns” for the follow-up examination. He continued: “If the patient does not return and there is a potentially important disease process, the standard of care requires that the doctor calls, writes or in some other way communicates with the patient to indicate to her that she should return and the reason why.” A physician should keep a calendar of future appointments, use a tickler system or otherwise ensure that the patient schedules her next appointment before leaving the physician’s office. If the patient fails to schedule or to keep the indicated appointment despite the physician’s instructions for follow-up, then someone from the physician’s office needs to communicate with the patient by phone, letter or certified mail about the importance of returning and should make a reasonable effort to have the patient return for the follow-up examination. In Schneider’s expert opinion, the failure to monitor and follow up on the condition of a patient with either fibrocystic disease or a benign mass is itself a breach of the required standard of care. He noted that he had found no indication that the defendant had complied with this standard of care.12
The defendant challenged the plaintiff’s case in many respects. He not only disagreed with the plaintiff’s recollection of the events that had transpired but presented expert evidence in his own behalf.
Factually, the defendant disputed the plaintiff’s allegations that he had discovered a lump or a mass during any of their office visits. He testified that he had never felt a dominant mass or lump between 1979 and 1985, had never noticed a change in her breasts and [272]*272had found nothing that he could have aspirated. He stated that he had diagnosed the plaintiff as having fibrocystic disease in 1979 and that the plaintiff’s breast had remained that way. Concerning the plaintiff’s 1983 office visit, the defendant testified that he had attempted to show the plaintiff how fibrocystic disease felt and to show her how to perform a breast self-examination so that she would better be able to detect any change.
He testified that the plaintiff’s 1985 office visit had been a check-up. He stated that the plaintiff had never told him during that visit that she had been feeling a lump in her left breast. His rationale for sending the plaintiff to obtain a mammogram was that the plaintiff was in her forties and needed a baseline mammogram. The defendant testified that he specifically had told the plaintiff, during the 1985 telephone call regarding the mammogram results, that she needed to come back to see him at her usually scheduled time of every six months or to contact him if she felt anything. He continued by stating that, because she had not returned for two and one-half years, he had no way of knowing what was happening in her breast. If she had followed her usual pattern of returning to his office in six months, then he would have been able to reexamine her and order another mammogram if necessary.
He presented his own and other expert testimony to sustain the reasonableness of his standard of care. He explained that fibrocystic disease is a misnomer, a non-disease, a benign diffuse process within the breast with no masses, dominant lesions or major cysts and that his notations concerning fibrocystic disease meant exactly that. The defendant admitted, however, that many physicians refer to fibrocystic disease as one that includes cysts and that he had similarly done so on occasion. He also stated that, while one can be reasonably sure of differentiating clinically between fibrocystic dis[273]*273ease and cancer, a physician can never be 100 percent certain of a breast condition without performing a biopsy. He stated that he would have reexamined the plaintiff in six months to a year and would have ordered another mammogram if indicated. He further testified that it had been his standard practice to do a followup on a patient even when faced with a negative mammogram. The defendant’s medical expert witness, Malcolm Beinfield, testified that the defendant’s conduct had not deviated from the appropriate standard of care for management of a condition like that of the plaintiff’s. He stated that fibrocystic disease is not a disease, but a process of maturation that occurs in a woman’s breast and that it is typically found in the breasts of women over forty years of age.
Further, Beinfield testified about the absence of a causal relationship between the plaintiff’s cancer and any alleged diagnostic failure on the defendant’s part. Relying on the plaintiff’s testimony concerning the growth of her tumor from the size of a pea in 1985 to that of a golf ball by May, 1987, and the analysis of the mastectomy that revealed twenty-seven of thirty-one lymph nodes in the plaintiff’s breast being cancerous, Beinfield opined that the cancer must have metastasized by 1985, and probably by 1983. Thus, the defendant’s failure to diagnose the plaintiff as having cancer in 1985 or thereafter could not have caused any injury to the plaintiff since she already had cancer in her body by 1983.13
B
We must decide whether, in light of this evidentiary record, the continuous treatment doctrine or the con[274]*274tinuing course of conduct doctrine tolled the statute of limitations in this case. Both of these doctrines are well established in the jurisprudence of this state.
The continuous treatment doctrine was first recognized in a medical malpractice context in Giambozi v. Peters, supra, 127 Conn. 380.14 We stated in Giambozi that “[t]he term malpractice itself may be applied to a single act of a physician or surgeon or, again, to a course of treatment. The Statute of Limitations begins to run when the breach of duty occurs. When the injury is complete at the time of the act, the statutory period commences to run at that time. When, however, the injurious consequences arise from a course of treatment, the statute does not begin to run until the treatment is terminated.” Id., 385. We quoted approvingly from a Minnesota case that defined the cessation of treatment as follows: “So long as the relation of physician and patient continues as to the particular injury or malady which [the physician] is employed to cure, and the physician continues to attend and examine the patient in relation thereto, and there is something more to be done by the physician in order to effect a cure, it cannot be said that the treatment has ceased. That does not mean that there must be a formal discharge of the physician or any formal termination of his [or [275]*275her] employment. If there is nothing more to be done by the physician as to the particular injury or malady which he [or she] was employed to treat or if he [or she] ceases to attend the patient therefor, the treatment ordinarily ceases without any formality.” (Internal quotation marks omitted.) Id., 384, quoting Schmit v. Esser, 183 Minn. 354, 358, 236 N.W. 622 (1931). We recently reaffirmed the existence of the continuous treatment doctrine in Connell v. Colwell, supra, 214 Conn. 252-54, even though we again held the doctrine to be factually inapplicable.
As an alternative, the continuing course of conduct doctrine has also been applied to toll the statute of limitations. In its modern formulation, we have held that in order “[t]o support a finding of a ‘continuing course of conduct’ that may toll the statute of limitations there must be evidence of the breach of a duty that remained in existence after commission of the original wrong related thereto. That duty must not have terminated prior to commencement of the period allowed for bringing an action for such a wrong. . . . Where we have upheld a finding that a duty continued to exist after the cessation of the ‘act or omission’ relied upon, there has been evidence of either a special relationship between the parties giving rise to such a continuing duty or some later wrongful conduct of a defendant related to the prior act.” Fichera v. Mine Hill Corp., supra, 207 Conn. 209-10 (no evidence to support continuing duty on part of defendant after property sold); see, e.g., Connell v. Colwell, supra, 214 Conn. 242 (improper reliance on theory); Beckenstein v. Potter & Carrier, Inc., 191 Conn. 150, 464 A.2d 18 (1983) (no continuing duty on defendant’s part after completion of roof installation); Prokolkin v. General Motors Corp., supra, 170 Conn. 289 (continuing course of conduct theory inappropriate in strict product liability action); Handler v. Remington Arms Co., supra, 144 Conn. 316 [276]*276(applying continuing course of conduct doctrine to toll statute of limitations on the basis of continuing duty to warn of defective cartridge by manufacturer); Vilcinskas v. Sears, Roebuck & Co., supra, 144 Conn. 174 (continuing course of conduct inapplicable where act completed by sale of air rifle).
Although the continuing course of treatment and the continuing course of conduct doctrines are analytically separate and distinct, their relevance to any particular set of circumstances, such as those involved in this appeal, may overlap. In underlying policy, the doctrines have considerable similarity. In application, both doctrines are conspicuously fact-bound.
These doctrines share similar supporting rationales. The continuing course of conduct doctrine reflects the policy that, during an ongoing relationship, lawsuits are premature because specific tortious acts or omissions may be difficult to identify and may yet be remedied. Similarly, “ ‘[t]he policy underlying the continuous treatment doctrine seeks to maintain the physician/patient relationship in the belief that the most efficacious medical care will be obtained when the attending physician remains on a case from onset to cure.’ ” Connell v. Colwell, supra, 214 Conn. 253.
The similarity and overlap between the continuous treatment doctrine and the continuing course of conduct doctrine can be seen in Cross v. Huttenlocher, supra, 185 Conn. 390, the only Connecticut medical malpractice case in which we determined that the statute of limitations would be tolled because of the continuing duty of a physician toward his patient. The defendant, a pediatric neurologist, negligently failed to warn his patient that the taking of medication that he had prescribed could produce the harmful side-effect of blindness. The physician-patient relationship began in 1965. Although the defendant last saw the patient [277]*277in 1968, he included a prescription for the medication for the plaintiff in his May, 1969 response to correspondence from the plaintiffs mother in which she requested a copy of the plaintiff’s latest prescriptions. Id., 391-92. The patient thus continued to use the medication until 1970. We held that the statute of limitations would not bar that part of the plaintiff’s case alleging negligent failure by the defendant to warn the plaintiff and her parents of the harmful side-effect of blindness, although not all of the claim would have been brought within the applicable statute of limitations period. Our determination that the statute of limitations should be tolled was based upon the continuous treatment of the plaintiff patient by the defendant physician through his prescribing of medication to her in order to treat her condition. In reaching our decision, we also stated that “[bjecause the negligent failure to warn is a continuing course of conduct, the statute of limitations does not begin to run until the course of conduct is completed.” Id., 400.
The facts of this case demonstrate the general applicability of these doctrines to an ongoing relationship between a physician and a patient. It may be impossible to pinpoint the exact date of a particular negligent act or omission that caused injury during a course of treatment. Alternatively, the negligence may have consisted of a series of acts or omissions. Thus, it is appropriate to allow the course of treatment to terminate before allowing the repose section of the statute of limitations to run, rather than having the parties speculate and quarrel over the date on which the act or omission occurred that caused the injury during a course of treatment. See Comstock v. Collier, 737 P.2d 845, 848-49 (Colo. 1987) (applying continuous treatment doctrine to toll statute of limitations until course of treatment for particular condition that included one [278]*278or more negligent acts or omissions causing injury terminated); 1D. Louisell & H. Williams, Medical Malpractice (1993) 113.02.
Specifically, either the continuous treatment doctrine or the continuing course of conduct doctrine, under the circumstances of this case, would require the jury to find: (1) an ongoing physician-patient relationship that had not terminated in January, 1985, the last time the plaintiff had consulted with the defendant; (2) negligence by the defendant in 1985; and (3) some form of treatment or required conduct that continued beyond 1985.
The determination of whether the physician-patient relationship has terminated depends upon several factors. These factors include the subjective views of the parties as to whether their relationship had terminated; the length of their relationship; the frequency of their interactions; the nature of the physician’s practice; whether the physician had prescribed a course of treatment for or was monitoring the condition of the patient; whether the patient was relying upon the opinion and advice of the physician with regard to a particular injury, illness or medical condition; and whether the patient had begun to consult with another physician concerning the same injury, illness or medical condition. See Adams v. Luros, 406 N.E.2d 1199, 1203 (Ind. App. 1980) (using similar multifactor test to determine termination of physician-patient relationship); 1 D. Louisell & H. Williams, supra, 113.02 (requiring either recovery of patient from illness, injury or condition being treated; authorized transfer or referral of patient to another physician; or express or implied discharge of physician by patient to terminate relationship).
If the jury credited the evidence presented by the plaintiff in this case, it reasonably could have found that the physician-patient relationship continued until at [279]*279least the middle of 1987, when the plaintiff was diagnosed by Goldblatt as having cancer. The long-term family physician-patient relationship between the plaintiff and the defendant began in 1973 and continued uninterruptedly with regular visits until the plaintiff was definitively diagnosed. The defendant was aware of the plaintiff’s trust and confidence in his abilities. The plaintiff relied on the defendant’s opinion and advice after the 1985 visit regarding her breast condition. The defendant did not tell her that a further follow-up was required. Especially significant is the fact that, when the plaintiff telephoned the defendant’s office in May, 1987, attempting to schedule an appointment with him, it was only the defendant’s temporary unavailability that led to the referral to Deutsch. In June, 1987, after the consultation with Deutsch, the defendant still possessed all of the plaintiff’s medical records, including a letter from Goldblatt that contained his findings and diagnosis. Further, the defendant responded to a telephone consultation initiated by the plaintiff’s husband about the medical inferences to be drawn from Goldblatt’s examination.
The testimony of the plaintiff’s expert witness Schneider, which the jury might have found credible, was sufficient for the jury to find not only the existence of continuous treatment and a continuing duty on the part of the defendant with regard to the patient’s breast condition, but also continuing negligence on the part of the defendant based upon a breach of his professional duty of care to the plaintiff. Schneider testified that the defendant had been negligent, not only by failing to perform or refer the plaintiff for immediate further diagnostic procedures after having received the negative results of the mammogram, but also by failing to pursue follow-up contacts with the plaintiff regarding her breast condition. According to Schneider, even if the defendant’s original diagnosis of fibrocystic disease [280]*280was not negligent, the absence of further monitoring of this breast condition was actionable since this later wrongful conduct continued until the plaintiff discovered her injury. It is this continuous failure to monitor, similar to the continuous failure to warn that we found actionable in Cross v. Huttenlocher, supra, 185 Conn. 390, that requires the application of the continuous treatment doctrine and the continuing course of conduct doctrine, thereby tolling the running of the statute of limitations. Since the plaintiff commenced her action within three years of May, 1987, the jury reasonably could have found that the plaintiffs action had not been barred by the statute of limitations.
II
The defendant next claims that, even if there was sufficient evidence before the jury to allow it to conclude that the plaintiff’s cause of action was not time-barred, the trial court’s instructions did not provide proper guidance to the jury on this issue. The defendant maintains that the charge included a misstatement of the law concerning the continuing course of conduct doctrine; namely, that the defendant had a duty to notify or warn the plaintiff of a negligent misdiagnosis. We agree that the charge was improper and order a new trial for this reason.
In reviewing the defendant’s claim, we adhere to the well settled rule that “a charge to the jury is to be considered in its entirety, read as a whole, and judged by its total effect rather than by its individual component parts.” Van Steensburg v. Lawrence & Memorial Hospitals, 194 Conn. 500, 507, 481 A.2d 750 (1984). “[T]he test of a court’s charge ‘is not whether it is as accurate upon legal principles as the opinions of a court of last resort but whether it fairly presents the case to the jury in such a way that injustice is not done to either party under the established rules of law.’ ...” (Citations [281]*281omitted.) Holbrook v. Casazza, 204 Conn. 336, 351-52, 528 A.2d 774 (1987), cert. denied, 484 U.S. 1006, 108 S. Ct. 699, 98 L. Ed. 2d 651 (1988); Konover Development Corp. v. Zeller, 228 Conn. 206, 219, 635 A.2d 798 (1994); Siladi v. McNamara, 164 Conn. 510, 515, 325 A.2d 277 (1973). We do not critically dissect the charge in order to discover possible inaccurate statements. Amato v. Desenti, 117 Conn. 612, 617, 169 A. 611 (1933). Rather, we “see if [the jury instructions] gave the jury a reasonably clear comprehension of the issues presented for their determination under the pleadings and upon the evidence and were suited to guide the jury in the determination of those issues. . . . [I]n our task of reviewing jury instructions, we view the instructions as part of the whole trial.” (Citations omitted; internal quotation marks omitted.) Champagne v. Raybestos-Manhattan, Inc., 212 Conn. 509, 566, 562 A.2d 1100 (1989). As “long as [the instructions] are correct in law, adapted to the issues and sufficient for the guidance of the jury”; Kelley v. Bonney, 221 Conn. 549, 584, 606 A.2d 693 (1992); we will not view the instructions as improper. Even if instructions are found to be improper, we must further determine whether they have been prejudicial to the claiming party by adversely affecting the trial’s outcome. Id., 586; Blancato v. Randino, 30 Conn. App. 810, 816, 622 A.2d 1032 (1993).
The defendant argues that the jury instructions with regard to the continuing course of conduct doctrine misstated the law and misled the jury by imposing on the defendant a duty to warn, a duty that he did not and could not have had. While conceding that, under the continuing course of conduct doctrine, a fiduciary would be required to disclose what he knows, the defendant contends that the plaintiff neither alleged nor presented any evidence that would suggest that he concealed the proper diagnosis of breast cancer from the plaintiff. Because he did not diagnose the plaintiff as having [282]*282breast cancer at the 1985 visit, he had no knowledge whatsoever that she had breast cancer and thus could not have breached an ongoing duty to warn the plaintiff about this matter. Thus, the defendant argues, the jury, in deciding that the statute of limitations had been tolled, reasonably could have based their decision on this erroneous statement of the duty of the defendant to warn the plaintiff of his misdiagnosis.
The trial court charged the jury with respect to this issue as follows. “[TJhere is what is called an exception to that three year rule, and that is where the statute [of limitations] is what we call tolled, that means that the statute is suspended, the running of the time is suspended. And I will explain that to you. That means that it may be tolled or suspended, that is the running of the statute, if you find that the plaintiff has proven . . . a continuing course of conduct on the part of the defendant.
“So to determine whether there was a continuing treatment, you must find that the plaintiff was either treated after that time or that there was a continuing course of conduct. And to prove that, there must be evidence of a breach of a duty that remained in existence after the commission of the original wrong.
“So if you assume the original wrong occurred in ’85 in that he failed to discover the cancer, then to toll the statute, the plaintiff must prove a special relationship such as a physician-patient relationship between the parties giving rise to a continuing duty between them so that the statute would toll and would not be cut off as of the last time he saw her in 1989, I am sorry, in 1989, which I think everybody agrees is the last time they saw was 1985. So whether—where there is a negligent failure to make a correct diagnosis and there is an ongoing physician-patient relationship, the three year period would not begin to run until the patient [283]*283discovered the correct diagnosis. And in this case, I think everybody agrees that in May of ’87 she found out that she did have breast cancer. That is because the physician has a continuing duty to notify his patient of the correct diagnosis. . . .
“Now, you must make this determination, that is the law, but you have to decide the factual question. If you find that there was an ongoing physician-patient relationship between Evelyn Blanchette and Doctor Barrett, that would give rise to a continuing duty on the part of Doctor Barrett to warn the Blanchettes of his misdiagnosis, then the Statute of Limitations would not start to run until the condition was discovered or should have been discovered, and we know it was discovered in May or June of 1987. . . .
“If you find either of these situations to be the case, then you may find that the statute, that this action, rather, was commenced within the period of the Statute of Limitations. If, on the other hand, you find that neither one of these situations, that is that there was not an ongoing patient-physician relationship, or that there was not a continuing course of treatment of Evelyn Blanchette, then you must find that the action is barred by the Statute of Limitations, and your verdict must be for the defendant.” (Emphasis added.)
We agree with the defendant that this jury charge misstated the law with respect to the continuing course of conduct doctrine. The complaint merely alleged a failure on the defendant’s part to warn the plaintiff with regard to follow-up care and observation of her left breast; see footnote 3; not a failure to warn her of his misdiagnosis. See Matthews v. F.M.C. Corp., 190 Conn. 700, 705, 462 A.2d 376 (1983) (plaintiff can only recover upon allegations contained in complaint). Furthermore, we can find no evidence that would support the giving of a charge concerning the defendant’s continuing duty [284]*284as a physician to notify his plaintiff patient of the “correct diagnosis.” See, e.g., Goodmaster v. Houser, 225 Conn. 637, 648, 625 A.2d 1366 (1993) (court has duty not to submit issue to jury if issue is not reasonably supported by the evidence); Bell v. Bihary, 168 Conn. 269, 273, 362 A.2d 963 (1975) (harmful error if jury charge permitted consideration of issue unsupported by fact).
More fundamentally, we disagree with the premise that a physician who has performed a misdiagnosis has a continuing duty to correct that diagnosis in the absence of proof that he subsequently learned that his diagnosis was incorrect. While there may be instances in product liability situations where a continuing duty to warn may emanate from a defect, without proof that the manufacturer actually knew of the defect; see, e.g., General Statutes § 52-572q(b); Giglio v. Connecticut Light & Power Co., 180 Conn. 230, 236, 429 A.2d 486 (1980); Tomer v. American Home Products Corp., 170 Conn. 681, 689, 368 A.2d 35 (1976); the same principle does not apply to a physician’s misdiagnosis. To apply such a doctrine to a medical misdiagnosis would, in effect, render the repose part of the statute of limitations a nullity in any case of misdiagnosis. We do not think that the language or policy of the statute permits such a reading.
In addition, the jury instructions concerning the continuing course of conduct doctrine reasonably can be understood as merely requiring the jury to find the existence of an ongoing physician-patient relationship in order to toll the running of the three year repose section of the statute of limitations. The court appears to have assumed that the existence of the physician-patient relationship automatically places a continuing duty on the physician to notify his or her patient of the correct diagnosis. Such an instruction is inaccurate and does not properly inform the jury on the correct prin[285]*285ciple of law. See Mazzucco v. Krall Coal & Oil Co., 172 Conn. 355, 357, 374 A.2d 1047 (1977); Velardi v. Selwitz, 165 Conn. 635, 639, 345 A.2d 527 (1974). Finally, the court emphasized the defendant’s alleged negligent misdiagnosis and failed to direct the jury to consider whether the defendant’s continuing duty consisted of monitoring the plaintiff’s breast condition.
In light of the inaccurate charge on the continuing course of conduct doctrine, we conclude that the jury was not properly guided in its deliberations on a critical matter. Alaimo v. Royer, 188 Conn. 36, 40, 448 A.2d 207 (1982). We cannot assume that this inaccurate instruction would not have adversely affected the jury’s determination on the defendant’s affirmative defense that the statute of limitations barred the plaintiff’s action. Because this issue was crucial to the jury’s verdict and the court’s subsequent judgment, the misinstruction cannot be viewed as harmless.
The judgment is reversed and the case is remanded for a new trial.
In this opinion the other justices concurred.