The plaintiffs commenced this medical malpractice action against the defendant in the Superior Court for his alleged negligence in connection with a “radio frequency ablation” (REA) procedure he performed on the leg of their minor son, which caused severe burning and eventually resulted in the amputation of the child’s leg. The jury did not reach the issue of negligence because they found that, more than three years before the plaintiffs filed the action, they knew or reasonably should have known that the child had been harmed by the defendant’s conduct, so the action was barred by the statute of limitations for medical malpractice claims.

The plaintiffs contend that the jury should have been instructed on the so-called “continuing treatment doctrine” applicable to medical malpractice claims, a doctrine that heretofore has not been recognized under Massachusetts law. Generally speaking, the doctrine states that a cause of action does not accrue, and therefore the statute of limitations clock does not begin to run, for medical malpractice claims during the period that an allegedly negligent physician continues to treat the patient for the same or a related condition. See, e.g., Borgia v. New York, 12 N.Y.2d 151, 156-157 (1962). We now recognize the doctrine under Massachusetts law and hold that the statute of limitations for a medical malpractice claim generally does not begin to run while the plaintiff and the defendant physician continue to have a doctor-patient relationship and the plaintiff continues to receive treatment from the physician for the same or a related condition. We also hold that the continuing treatment exception to the discovery rule terminates once a patient (or the parent or guardian of a minor patient) learns that the physician’s negligence was the cause of his or her injury. We further hold that, once the allegedly negligent physician no longer has any role in treating the plaintiff, the continuing treatment doctrine does not apply even if the physician had at one time been part of the same “treatment team” as the physicians who continue to provide care. Here, where the *370jury found that the plaintiffs knew or reasonably should have known more than three years before commencing suit that they had been harmed by the conduct of the defendant, and where the defendant’s participation in treating the plaintiff ended more than three years before the suit was filed, the cause of action accrued more than three years before the action was commenced and therefore was not timely under the statute of limitations. We thus affirm the judgment in favor of the defendant.4

Background. We recite the facts in the light most favorable to the plaintiffs. See Lipchitz v. Raytheon Co., 434 Mass. 493, 499 (2001). William Parr was born on September 3, 1994.5 At birth, he had a large lump at the back of his right calf. Within a few weeks, he was taken by his parents, Michele Parr and Michael Parr, to Massachusetts General Hospital (MGH), where he was referred to the “sarcoma group” for imaging studies and a biopsy. The sarcoma group is a team of orthopedic surgeons, general surgeons, radiation oncologists, medical oncologists, and others who treat tumors of the connective tissues, including bones, muscles, fat, nerves, and other tissues. The sarcoma group works on an interdisciplinary model. The team members meet twice weekly and have “very close interdisciplinary relationships.”

Initially, William’s lump was diagnosed by the sarcoma group as a “hamartoma.”6 By January, 2003, when William was eight years old, the size of the lump had increased, causing William occasionally to limp. At that time, Dr. Mark Gebhardt, a member of the sarcoma group at MGH, performed a biopsy in which he removed pieces of the lump for the pathologist to examine. Gebhardt determined that the lump was engulfing much of William’s calf muscle and was having an impact on his nerves and blood vessels. At this time, it was determined that the lump was a “desmoid tumor.” Desmoid tumors are relatively rare, benign tumors but can grow in such a way as to infiltrate normal tissue and impair bodily functions.

Soon after the biopsy, Gebhardt left MGH. William’s care was *371assumed by Dr. David Ebb, a pediatric oncologist, and Dr. Kevin Raskin, an orthopedic surgeon, both of whom were on the staff at MGH and were members of the sarcoma group. At some point prior to November, 2005, Raskin and Ebb proposed and scheduled a surgery on William’s tumor, which at this point had caused abnormality in his gait. Meanwhile, Michele continued to research other options, and she discussed the possibility of RFA treatment with Ebb and Raskin.7 After one of the meetings of the sarcoma group, Raskin and Ebb approached the defendant, Dr. Daniel Rosenthal, about the possibility of performing REA on William’s desmoid tumor. Rosenthal was a board certified radiologist on the staff at MGH and had been a member of the sarcoma group since 1978. He “invented” REA, meaning that he was the first physician to use REA to treat a tumor, and was a recognized leader in the field.8 Through the sarcoma group, he was generally familiar with William’s case. Raskin and Ebb then put Michele in touch with Rosenthal. They told Michele that Rosenthal “was the best doctor in the business basically. He was . . . one of the founders of radiofrequency ablation and had worked at [MGH] for a long time.” Rosenthal eventually agreed *372to perform the procedure and it was presented and approved at subsequent sarcoma group meetings. Rosenthal had never performed RFA on a desmoid tumor before performing the procedure on William, and as of the date of trial had not performed another RFA on a desmoid tumor.

Prior to the procedure, Rosenthal told Michele that the procedure was reasonable and could help William. Michele testified that Rosenthal told her that RFA could “kill” the tumor, but he did not explain any risks of the procedure. Rosenthal said the procedure would be a day surgery, that William would come out with “band-aids” at the sites where the probe had gone in, and that he would be home by the afternoon.

Michael brought William to MGH on the morning of November 4, 2005, for the RFA procedure, and Michele arrived soon thereafter. Rosenthal briefly showed Michael and William a drawing describing the procedure, demonstrating the location of the tumor and other areas he was going to treat. Michael signed a consent form, which listed the risks of the procedure, including bleeding, infection, nerve damage, and failure to cure. The form did not disclose any risk of burns to the skin, blood vessels, or other vital structures. Moreover, the risks associated with the use of a tourniquet were not mentioned.

Rosenthal completed the first three of his planned four ablations when he noticed what he described as “superficial skin blisters” in the area behind William’s knee. At that point, despite not having completed all of the planned ablations, Rosenthal realized that he had already burned more than the entire planned treatment area. On seeing the burned area behind William’s knee, Rosenthal then stopped the procedure and called two other sarcoma group members, Ebb and Raskin, to the operating room. A decision was made to discontinue the procedure.

Ebb explained to Michael and Michele that there had been a complication during the procedure, and that William had suffered a burn above the tumor site. Michele testified that she was not told the cause of the burn or how serious it was, but was told that William “would recover and be fine.” Michael testified that he and Michele did not know how serious the burn was at first and that he “never knew” how bad the burn was. Rosenthal originally described it to them as a “superficial burn.” Raskin referred to the burn as a “superficial blister” in his notes on the day of the RFA procedure.

William was admitted to MGH for one week after the RFA procedure and was then transferred to Spaulding Rehabilitation *373Hospital (Spaulding) for an additional five weeks. Rosenthal visited William every day during his week-long stay at MGH and several more times at Spaulding. Rosenthal’s last note in William’s medical file, made during a visit on November 7, 2005, states that there was “clear improvement in his nerve function.” In fact, by that time, the nerves had been irreparably damaged from the burn. The burned area ultimately grew to full thickness, creating a very foul smelling, necrohc blackened hole in the back of William’s knee that spanned the entire area of the knee from medial to lateral. The nerves were destroyed.

When William returned home from Spaulding, he received in-home physical therapy, and a visihng nurse provided medical care. He also continued to receive care from the sarcoma group. The burn did not heal during this period despite efforts throughout the winter that were directed by Raskin. The burn eventually became infected, and William was readmitted to MGH in February, 2006. Raskin performed debridements of the burn. On March 19, 2006, after the seventh debridement, it became clear that William’s leg could not be saved, and his parents were told that amputation below the knee was necessary. On March 20, 2006, William’s leg was amputated below the knee.

About two years later, a second amputation, this one above the knee, became necessary because of continued infections, and because there was insufficient muscle preserved to enable use of a prosthesis or to make the knee funchonal. Thus, on March 12, 2008, Raskin performed an amputation above the knee on William’s right leg. According to Rosenthal’s testimony at trial, the need for the amputations was a direct result of the complication that occurred during the RFA procedure. Neither Ebb, Raskin, nor Rosenthal at any time described to the Parrs what had caused the burn and the resulhng injuries. Michael was told that it was simply an “anomaly.” The Parrs did not know what had happened, despite asking repeatedly. As Michael testified, “We trusted them, we worked with them and we did not know.”

Michele and Michael, as parents and next friends of William, filed a civil complaint in the Superior Court on March 9, 2009, alleging malpractice by Rosenthal with respect to the RFA procedure performed by him on November 4, 2005.9 The case proceeded to a jury trial. Because, under G. L. c. 231, § 60D, a *374medical malpractice action brought on behalf of a minor who is at least six years old must “be commenced within three years from the date the cause of action accrues,” the defendant argued that he was entitled to judgment because the action was not timely filed. At the charge conference and in his proposed jury instructions, the plaintiffs’ counsel argued that the statute of limitations was tolled while William’s treatment was continuing. Relying on the continuing treatment doctrine, he argued that a cause of action for medical malpractice does not accrue until treatment of a plaintiff by a defendant doctor, or doctors with whom he works, has terminated.10

The judge declined to give such an instruction. He correctly stated that Massachusetts had not yet recognized the continuing treatment doctrine. He further stated that he “would suggest” that Massachusetts would not “adopt that theory,” and that, in any event, the doctrine would not apply to the facts of this case. The judge said the defendant “rendered a very specific treatment” and “[tjhat was it”; “[h]e was not involved in the treatment of William after that.” In response to counsel’s argument that the cause of action did not accrue while the plaintiff was being treated by the “treatment team” of which the defendant was a member, the judge noted that no Massachusetts case had taken that position in a medical malpractice case.

As to the question whether the claim was timely brought within the statute of limitations, the judge instructed the jury as follows:

“Ordinarily a personal injury claim must be brought within three years of the date the cause of action accrues or arises. Here, this case was commenced on March 6, 2009.[11] The question is whether the claim was brought within three years *375after the date on which the cause of action arose. The general rule is that a cause of action accrues on the date of the plaintiffs injuryf] in this case, William’s injury. However, that rule does not apply where the plaintiff did not know or could not reasonably have known of the cause of action. . . . [T]he question comes down to whether the plaintiffs knew or should have known that William Parr had been harmed to an appreciable or not insignificant extent by Dr. Rosenthal’s conduct.”

The judge also explained the meaning of ‘“should have known” in this context:

‘“An action for medical malpractice accrues when a reasonably prudent person in the plaintiff’s position reacting to any suspicious circumstances for which they might have been aware should have discovered that his medical care given by the physician may have caused . . . William appreciable or not insignificant injury or harm. Certainty of causation is not required. Rather, notice of likely cause is sufficient to start the statute running [,] imposing on the potential litigant the duty to discover from legal, scientific and medical communities whether a theory of causation supports a legal claim.”

Plaintiffs’ counsel timely objected to the judge’s decision not to give a continuing treatment instruction regarding the statute of limitations.

The jury answered ‘“yes” to the first special verdict question: ‘“Did the plaintiffs know or should they reasonably have known prior [to] March 6th, 2006, that they had been . . . harmed by the conduct of the defendant?” Because they answered ‘“yes” to this question, they did not reach the other questions, including whether the defendant was negligent and, if so, whether his negligence was a substantial contributing factor in causing William’s injury. Judgment entered for the defendant. The plaintiffs moved for a new trial, claiming that the judge erred by failing to furnish the jury with the continuing treatment instruction. The judge denied the motion. The plaintiffs appealed from the judgment, and from the denial of their motion for a new trial.

The Appeals Court reversed the judgment and remanded the case for a new trial. Parr v. Rosenthal, 87 Mass. App. Ct. 787 *376(2015). Relying by analogy on this court’s adoption of the “continuing representation” doctrine in legal malpractice cases, see, e.g., Murphy v. Smith, 411 Mass. 133, 137 (1991), the Appeals Court recognized the “continuing treatment” doctrine in medical malpractice cases. The court held that the limitations period does not begin to run during the continuing treatment of a patient for the same injury on which the action for medical malpractice is based. See Parr, supra at 788, 792-793. The court also declared that the continuing treatment doctrine “will toll the statute of limitations so long as the patient remains in continuous treatment for the injury by the same physician or group, or under the general control of that physician or group, subject to the statute of repose.” Id. at 797.

In cases alleging legal malpractice, the statute of limitations, although tolled under the continuing representation doctrine, nevertheless begins to run once a client acquires actual knowledge that he or she has suffered appreciable harm as a result of the attorney’s conduct. See, e.g., Lyons v. Nutt, 436 Mass. 244, 249-250 (2002). The Appeals Court, however, held that in medical malpractice cases “actual knowledge should not bar application of the continuing treatment doctrine so long as the patient is continuing treatment in good faith and not solely to allow more time to develop their malpractice case.” Parr, supra at 798. We granted the defendant’s application for further appellate review.

Discussion. To state a claim for medical malpractice, a plaintiff must demonstrate that (1) the plaintiff suffered harm; (2) the harm was caused by the defendant physician’s conduct; and (3) the defendant physician was negligent, which in medical malpractice cases means that the physician committed a breach of the “standard of care and skill of the average member of the profession” practicing in his or her specialty. See Bradford v. Baystate Med. Ctr., 415 Mass. 202, 206-208 (1993), quoting Brune v. Belinkoff, 354 Mass. 102, 109 (1968). Where a defendant raises the statute of limitations as an affirmative defense, the plaintiff also bears the burden of proving that the action was timely commenced. See Franklin v. Albert, 381 Mass. 611, 619 (1980). The limitation and repose periods for medical malpractice claims brought on behalf of minors over the age of six are established by G. L. c. 231, § 60D, which provides in relevant part:

“[A]ny claim by a minor against a health care provider stemming from professional services or health care rendered, *?whether in contract or tort, based on an alleged act, omission or neglect shall be commenced within three years from the date the cause of action accrues . . . , but in no event shall such action be commenced more than seven years after occurrence of the act or omission which is the alleged cause of the injury upon which such action is based except where the action is based upon the leaving of a foreign object in the body.”12

The statute of repose is not at issue in this case. The plaintiffs’ claim clearly was brought within seven years of William’s RFA treatment. The defendant does not claim otherwise. The key question is whether the claim was timely brought within the statute of limitations, i.e., within three years of when the cause of action accrued.

A statute of limitations typically prescribes the time period when an action must be commenced after the cause of action ‘“accrues.” The statute sets the limitations period, but in the absence of explicit legislative direction, it is our common law that determines when a cause of action accrues, and hence when the limitations period actually begins to run. See Franklin, 381 Mass. at 617 (‘“Absent explicit legislative direction, the determination of when a cause of action accrues, causing the statute of limitations to run, has long been the product of judicial interpretation in this Commonwealth”). In Franklin, supra at 619, we held that the discovery rule applies to the statute of limitations for medical malpractice claims.13 Under the discovery rule, medical malpractice claims ‘“accrue” “when the plaintiff learns, or reasonably should have learned, that he has been harmed by the defendant’s *378conduct.” Id. ‘“In determining whether a party has sufficient notice of causation, our inquiry is whether, based on the information available to the plaintiff, a reasonably prudent person in the plaintiffs position should have discovered the cause of his or her injuries.” Lindsay v. Romano, 427 Mass. 771, 774 (1998), quoting McGuinness v. Cotter, 412 Mass. 617, 628 (1992). “We do not require that a plaintiff have notice of a breach of a duty before a cause of action may accrue . . . .” Bowen v. Eli Lilly & Co., 408 Mass. 204, 208 (1990). See Lindsay, supra (“It is not necessary that the plaintiff have notice that the defendant was actually responsible for the injury, only that she have knowledge or sufficient notice that the medical care given by the defendant may have caused the injury”). In essence, under our common law, once a patient knows or reasonably should know that he or she has suffered harm and that the harm was caused by the physician’s conduct, the statute of limitations clock starts to run, and the patient then has three years to discover whether the physician committed a breach of the standard of care and whether the theory of causation is supported by the evidence, and, if so, to commence a civil suit. See Bowen, supra at 208, quoting Fidler v. Eastman Kodak Co., 714 F.2d 192, 199 (1st Cir. 1983) (“Thus on notice, the potential litigant has the duty to discover from the legal, scientific, and medical communities whether the theory of causation is supportable and whether it supports a legal claim”).

1. Continuing treatment doctrine. The plaintiffs argue that Massachusetts should recognize the continuing treatment doctrine, which provides that a cause of action does not accrue while the patient is continuing to receive treatment for the same or related injury or illness from the same physician who allegedly caused the patient harm. See Otto v. National Inst. of Health, 815 F.2d 985, 988 (4th Cir. 1987), and cases cited (in claims brought under Federal Tort Claims Act, “where there has been a course of continuous medical treatment, a claim may not accrue until the end of that course of treatment, if the treatment has been for the same illness or injury out of which the claim for medical malpractice arose”). See also Borgia, 12 N.Y.2d at 155-156. The rationale for the doctrine appears to be two-fold. First, a patient who continues a physician-patient relationship impliedly continues to have trust and confidence in the physician, and this trust and confidence put “the patient at a disadvantage to question the doctor’s techniques,” Barrella v. Richmond Mem. Hosp., 88 A.D.2d 379, 384 (N.Y. 1982), and impair “the patient’s ability to *379make an informed judgment as to negligent treatment.” Harrison v. Valentini, 184 S.W.3d 521, 525 (Ky. 2005). See Otto, supra (“The continuous treatment doctrine is based on a patient’s right to place trust and confidence in his physician. . . . [T]he patient is excused from challenging the quality of care being rendered until the confidential relationship terminates”). Second, where there is a poor medical result from a physician’s treatment or procedure, a patient is entitled to allow the physician an adequate opportunity to remedy or mitigate the poor result without needing to risk interruption of that course of treatment by exploring whether the poor result arose from that physician’s negligence. See id. (“the doctrine permits a wronged patient to benefit from his physician’s corrective efforts without the disruption of a malpractice action”); Barrella, supra (patient is entitled “to rely upon the doctor’s professional skill without the necessity of interrupting a continuing course of treatment by instituting suit”).

In Murphy v. Smith, 411 Mass. 133, 137 (1991), we adopted an analogous “continuing representation” rule for legal malpractice claims. The plaintiffs in that case had received a letter from their neighbors’ attorney in 1983, informing them that they did not have good title to property they had purchased. Id. at 135. The plaintiffs then contacted the defendant, the bank’s attorney who had certified good record title when they purchased the property. Id. The defendant assured them that the letter “did not present a cause for concern and that he would take care of it.” Id. at 137. The court determined that the defendant’s legal representation of the plaintiffs began with this assurance, and ended in 1985, when they retained a new attorney. Id. at 135, 137. They commenced a legal malpractice action against the defendant in 1987. Id. at 135. If the discovery rule applied without exception, the three-year statute of limitations would have begun to run at the time the plaintiffs received the letter from their neighbors’ attorney, i.e., when they were put on notice of the alleged defect in their title, and their malpractice claim against the defendant would have been time barred.

The court in Murphy adopted the continuing representation doctrine as an exception to the discovery rule, holding that the doctrine “tolls the statute of limitations in legal malpractice actions where the attorney in question continues to represent the plaintiff’s interests in the matter in question.” Id. at 137. The statute of limitations in that case thus did not begin to run until 1985, when the defendant’s representation of the plaintiffs ended. The *380continuing representation doctrine “recognizes that a person seeking professional assistance has a right to repose confidence in the professional’s ability and good faith, and realistically cannot be expected to question and assess the techniques employed or the manner in which the services are rendered.” Id., quoting Cantu v. Saint Paul Cos., 401 Mass. 53, 58 (1987).

The reasoning we embraced in Murphy, supra, in adopting the continuing representation exception to the discovery rule in legal malpractice claims also justifies the adoption of a continuing treatment exception to the discovery rule in medical malpractice claims. Under the discovery rule, we ordinarily start the clock when the patient knows or has reason to know that he or she has been harmed by the physician’s conduct; we consider such knowledge or reason to know sufficient to trigger the patient’s “duty to discover” within the three-year limitations period whether the physician committed a breach of the standard of care and was the legal cause of the patient’s injury. See Bowen, 408 Mass. at 208-210. However, while that physician continues to treat the patient for the same or related injury or illness, the physician’s patient, like an attorney’s client, “realistically cannot be expected to question and assess the techniques employed or the manner in which the services are rendered.” Murphy, supra at 137. Just as we recognize that a represented party is entitled to retain confidence in his or her legal counsel’s “ability and good faith” while the representation continues, so, too, do we recognize that a patient is entitled to retain confidence in his or her physician’s ability and good faith while continuing treatment with that physician. The legal client is disadvantaged in learning whether his or her attorney has committed a breach of the standard of care while that attorney continues to represent the client, and so, too, is a patient disadvantaged in learning whether a physician has committed a breach of the standard of care while the physician continues to treat the patient. And just as a wronged client is permitted to benefit from his or her attorney’s efforts to correct a problem without the disruption of exploring the viability of a legal malpractice action, so, too, is a patient permitted that same benefit without the disruption of exploring the viability of a medical malpractice action.

Moreover, there is no “explicit legislative direction” that precludes us from recognizing a continuing treatment exception in determining when a medical malpractice cause of action accrues. See Franklin, 381 Mass. at 617. As the Appeals Court correctly *381noted, the Legislature used almost identical language to describe the limitations period for medical malpractice claims and for legal malpractice claims. See Parr, 87 Mass. App. Ct. at 793. Compare G. L. c. 231, § 60D (medical malpractice claim by minor who is at least six years old “shall be commenced within three years from the date the cause of action accrues”), and G. L. c. 260, § 4, second par. (medical malpractice claim by adult “shall be commenced only within three years after the cause of action accrues”), with G. L. c. 260, § 4, first par. (legal malpractice claim “shall be commenced only within three years next after the cause of action accrues”). See generally Harlfinger v. Martin, 435 Mass. 38, 49 (2001) (noting that discovery rule principles applicable to other types of tort claims also applied to medical malpractice claims). The only differences in the language of the statutes are slight and insignificant. Given the nearly identical wording of the provisions, there is no reason to conclude that our interpretation of when a cause of action “accrues” in legal malpractice cases should not have its analog in medical malpractice claims.

The defendant argues that the adoption of the continuing treatment doctrine would constitute “improper judicial legislation,” urging us to infer from the absence of legislation on the doctrine that the Legislature has rejected it. The defendant points to nothing in the record of the Legislature, however, that suggests that its silence on the subject reflects a conscious choice to reject the continuing treatment doctrine. We decline to interpret the absence of legislative action as an affirmative rejection of the doctrine that bars us from adopting the continuing treatment doctrine as a common-law interpretation of when a cause of action “accrues” in a medical malpractice case.14

Our adoption of the continuing treatment doctrine does not affect the statute of repose that applies to medical malpractice *382claims, which provides that “in no event shall such action be commenced more than seven years after occurrence of the act or omission which is the alleged cause of the injury upon which such action is based except where the action is based upon the leaving of a foreign object in the body.” G. L. c. 231, § 60D. The effect of a statute of repose “is to place an absolute time limit on the liability of those within [its] protection and to abolish a plaintiffs cause of action thereafter, even if the plaintiffs injury does not occur, or is not discovered, until after the statute’s time limit has expired” (citation omitted). Rudenauer v. Zafiropoulos, 445 Mass. 353, 357 (2005). Therefore, in this case, the plaintiffs’ cause of action would have been barred by the statute of repose had it been brought more than seven years after November 4, 2005, the date of the RFA procedure, even if the cause of action had yet to accrue under the continuing treatment doctrine.15,16

*3832. Effect of actual knowledge on continuing treatment doctrine. In Lyons v. Nutt, 436 Mass. 244, 250 (2002), we held that the continuing representation exception to the discovery rule in a legal malpractice case terminates once “the client actually knows that he suffered appreciable harm as a result of his attorney’s conduct.” We reasoned that once “the client has such knowledge, then there is no ‘innocent reliance which the continued representation doctrine seeks to protect.’ ” Id., quoting Cantu, 401 Mass. at 58. The defendant urges us to apply an analogous rule for the continuing treatment exception in medical malpractice cases.

In deciding whether to adopt this “actual knowledge” rule, we first consider the logic and purpose behind it. We declared in Lyons, 436 Mass. at 247, quoting Williams v. Ely, 423 Mass. 467, 473 (1996), that “[t]he statute of limitations applicable to a legal malpractice claim begins to run when a client ‘knows or reasonably should know that he or she has sustained appreciable harm as a result of the lawyer’s conduct.’ . . . This is the so-called discovery rule.” The consequence of the termination rule in Lyons is that, even where the client continues to be represented by the attorney, the statute of limitations clock for legal malpractice claims begins when a client actually knows that he or she has sustained appreciable harm as a result of the lawyer’s conduct. If a client reasonably should know that the attorney has caused the client appreciable harm, but does not actually know it, the continuing representation rule continues to apply.

However, in the practice of law, actual knowledge that an at*384torney caused a client appreciable harm generally means actual knowledge that the attorney committed legal malpractice. Indeed, in Lyons, 436 Mass. at 247-248, 251, we affirmed the judge’s ruling that the continuing representation exception did not apply after the client realized that his law firm “didn’t know what they were doing” once the offeror “walked away from the deal.” See Hendrickson v. Sears, 365 Mass. 83, 91 (1974) (“A client’s cause of action against an attorney for negligent certification of title to real estate does not ‘accrue’ . . . until the misrepresentation is discovered or should reasonably have been discovered, whichever first occurs ...”).

But with medical malpractice, a patient’s actual knowledge that the physician has caused the patient appreciable harm does not necessarily mean that the patient knows that the physician was negligent, because every medical procedure carries with it a risk of complications that may occur naturally without any breach of the standard of care by the physician. The instant case is a classic example: there was no question that the defendant’s RFA procedure caused appreciable harm to William, but actual knowledge of that fact shed little light on whether the harm arose from a mere complication or from the defendant’s breach of the standard of care. Therefore, in contrast with an attorney’s client, it is simply incorrect to say that, once a physician’s patient knows that the physician has caused the patient appreciable harm, there can be no “innocent reliance” that the continuing treatment doctrine seeks to protect. A patient who continues under the care of the same physician will still have the same challenges in learning whether the harm the patient suffered from the physician’s treatment arose from the physician’s negligence. Thus, we conclude that the continuing treatment exception to the discovery rule terminates only when the plaintiff has actual knowledge that his or her treating physician’s negligence has caused the patient’s appreciable harm, because it is only then that there can no longer be the kind of “innocent reliance” that the continuing treatment doctrine seeks to protect. Once a patient learns that the physician’s negligence was the cause of his or her injury, the patient has acquired sufficient information to initiate litigation, and there is no longer adequate reason to continue to toll the statute of limitations.17

*385To be clear, by declaring that the tolling of the statute of limitations ends under the continuing treatment doctrine only when a plaintiff obtains actual knowledge of a physician’s negligence, we are not revising the discovery rule in medical malpractice. Where the continuing treatment doctrine does not apply, the statute of limitations clock begins to run on a medical malpractice claim when the plaintiff learns, or reasonably should have learned, that he or she has been harmed by the defendant’s conduct. Franklin, 381 Mass. at 619. Where the continuing treatment doctrine does apply and, but for the application of the doctrine the statute of limitations clock would have started under the discovery rule, the tolling arising from the doctrine ends once the plaintiff has actual knowledge that the physician’s negligence was the cause of his or her injury.

3. Applicability of continuing treatment doctrine during treatment by physicians other than defendant. In this case, there is no evidence that Rosenthal continued to treat William at any point after William returned home from Spaulding in December, 2005. If the continuing treatment doctrine applies to Rosenthal’s treatment only, then the doctrine would not toll the statute of limitations period long enough to render the plaintiffs’ action timely. The question becomes whether the doctrine continued to apply, and continued to toll the statute of limitations for a claim against Rosenthal, for the additional period that Raskin and Ebb treated William thereafter, as they continued to try to remedy the damage done during the RFA procedure.

*386The plaintiffs in their proposed jury instruction claimed that the continuing treatment doctrine applies during the ongoing treatment by a defendant physician “or doctors with whom he works.” See note 8, supra. On appeal, the plaintiffs contend that it applies to medical personnel who were (1) in some “relevant association with the [initial treating] physician,” Kelly v. State, 110 A.D.2d 1062, 1063 (N.Y. 1985) (Hancock, J.P., dissenting); or (2) “part of the same team,” Tausch v. Riverview Health Inst. L.L.C., 187 Ohio App. 3d 173, 182 (2010). The Appeals Court recognized that “[t]he case law in other jurisdictions does not clearly establish a single rule for when treatment by an associated doctor can be imputed to the alleged negligent doctor.” Parr, 87 Mass. App. Ct. at 794, citing Tolliver v. United States, 831 F. Supp. 558, 560 (S.D. W. Va. 1993). See Parr, supra at 795 n.20, and cases cited. The Appeals Court adopted a variation of the plaintiffs’ argument and held that, “[o]n retrial, if the jury conclude that William was a group patient of all three doctors and not an individual patient of Drs. Raskin and Ebb, or that the defendant was still providing input to Drs. Raskin and Ebb on William’s care as part of the group prior to the amputation, then their continuing treatment for the burn can be imputed to the defendant.” Id. at 795-796.

We agree that the continuing treatment doctrine would apply where an allegedly negligent physician continues to supervise, advise, or consult with other physicians who are treating the patient for the same or a related injury. See Otto, 815 F.2d at 989 (in medical malpractice case where National Institute of Health [NIH] was sole defendant, continuing treatment doctrine applied where “additional treatment was rendered at the advice and under the direction of the NIH physicians”); Stephenson v. United States, 147 F. Supp. 2d 1106, 1112 (D.N.M. 2001) (continuing treatment doctrine applied where negligent primary care physician had “continued direct involvement in evaluating [the patient]’s progress” and exercised “control over the treatment of [the patient] by the other health-care providers”); Echols v. Keeler, 735 P.2d 730, 732 (Wyo. 1987) (continuing care doctrine did not apply where allegedly negligent doctor did “not continue as [the patient’s] doctor nor was he associated with or engaged in assisting the doctors thereafter treating [the patient]”). There was no evidence here, however, that after December, 2005, Rosenthal supervised the treatment of William, or advised or consulted with Raskin and Ebb regarding their treatment of him. Consequently, for the plaintiffs’ medical malpractice cause of action to be *387timely, the continuing treatment doctrine would need to apply to the period following the surgery in which William was being treated by Raskin and Ebb alone.

We need not determine here whether to follow the case law in other jurisdictions that have applied the continuing treatment doctrine to the continuing care of other physicians in the same medical group partnership or medical clinic where a patient is considered by the physicians and the patient to be a patient of the group or clinic rather than of an individual physician. See Offerdahl v. University of Minn. Hosps. & Clinics, 426 N.W.2d 425, 428 (Minn. 1988); Watkins v. Fromm, 108 A.D.2d 233, 239 (N.Y. 1985). In Offerdahl, supra, the plaintiff was a student at the University of Minnesota who “did not seek treatment from any particular University physician but employed the University clinic generally as her physician.” The Supreme Court of Minnesota held that although the claim was based upon negligence by a particular physician of the clinic, “under these unique facts where the patient sought treatment from a clinic as a whole rather than an individual physician, the treatment of the clinic as a whole, rather than that of the individual physician alleged to have committed the act of malpractice, is relevant for purposes of determining when treatment terminated and the statute of limitations began to run.” Id. Similarly, in Watkins, supra at 234-235, the Appellate Division of the Supreme Court of New York held that the continuing treatment doctrine tolled the statute of limitations where ongoing treatment was provided by members of the negligent physician’s medical group because, according to deposition testimony, the plaintiff “was considered to be a patient of the entire medical group, rather than of any one of the individual doctors, and that it was the practice of the defendant doctors to discuss, as a group, the diagnosis and treatment of all of the patients under their care.” Those factual circumstances are not presented here. There is no evidence in the record that William was treated as a patient of the sarcoma group rather than of the particular physicians providing treatment, that the physicians in the group discussed the diagnosis or treatment of all their patients with the group, or that William’s parents believed him to be a patient of the group rather than of individual physicians.

We have considered whether the reasons that underlie the continuing treatment doctrine justify the application of the doctrine where the allegedly negligent physician and the physician who continues to treat the patient once were together part of a *388patient’s “treatment team.” We recognize that, in these circumstances, there is a risk that a patient’s continued trust and confidence in the physician providing continuing care might put the patient at a disadvantage in making an informed judgment as to whether a former team member provided negligent treatment, especially where the physician providing the continuing care effectively brought the allegedly negligent physician onto the team. We also recognize that there is a risk that, if a plaintiff were to contemplate a medical malpractice action against a former team member, the plaintiff might reasonably fear that exploring an action against that physician would interrupt or otherwise interfere with the patient’s continuing treatment with other members of the team because of their respect for and close connection with that physician.

We are reluctant, however, to extend the continuing treatment doctrine to a “treatment team” for two reasons. First, tolling the statute of limitations while the plaintiff continues to be treated by a “treatment team” that once included the allegedly negligent physician poses the risk that what was intended to be a narrow exception may be interpreted so broadly as to devour the discovery rule in medical malpractice cases. Second, given the multitude of different ways in which patients receive medical treatment in this Commonwealth, it is difficult to define with precision a patient’s “treatment team.” The absence of a precise definition means not only that it would be difficult at trial to instruct a jury regarding the statute of limitations but, more importantly, it would be difficult to determine whether a case should be dismissed before trial on statute of limitations grounds. The clarity and precision of a limitations period is important to the interests of justice, because it enables untimely filed cases to be dismissed before trial, thus sparing all parties the needless time, expense, and burden of a trial where the jury will never reach an adjudication on the merits.

Because, having balanced the competing considerations, we are unwilling to apply the continuing treatment doctrine to the plaintiff’s continued treatment by a “treatment team” that once included the defendant, the doctrine does not apply in this case after December, 2005. And without the tolling of the statute of limitations beyond that date under the continuing treatment doctrine, the plaintiffs’ medical malpractice claim was not timely filed.

Conclusion. We affirm the judgment in favor of the defendant *389and the order denying the plaintiffs’ motion for a new trial.

So ordered.

(dissenting in part). The court’s decision today fails to consider several factors that strongly militate against adopting a continuing treatment exception to our settled discovery rule for medical malpractice claims. Instead, the court imprudently intrudes into a critically important sphere of health care policy-making and makes its own preferred policy judgment without any inkling of the effect it might have on the cost of health care in Massachusetts, a matter of acute concern to the executive and legislative branches of government. These branches are far better equipped to balance the benefits of a prolonged statute of limitations with the cost and access issues it implicates. Just because the court can act to change the law does not mean that it should. Therefore, I respectfully dissent from the court’s adoption of the continuing treatment doctrine for medical malpractice cases.

For nearly forty years, our law has been clear: a cause of action for medical malpractice “accrue [s] when the plaintiff learns, or reasonably should have learned, that he has been harmed by the defendant’s conduct.” Franklin v. Albert, 381 Mass. 611, 619 (1980). See G. L. c. 231, § 60D. Once the harm and its causal relationship to acts of the physician is known or reasonably should have been learned, the statute of limitations clock starts to run, and the patient has three years to determine whether to file suit. See Bowen v. Eli Lilly & Co., 408 Mass. 204, 208 (1990).

Although I agree with the court’s articulation of our rule that, in the absence of explicit legislative direction, it may determine, as a matter of common law, when a cause of action accrues, and hence when the limitation period begins to run, see Franklin, 381 Mass. at 617, the absence of explicit statutory language does not mean that the court should act to change settled law in a manner inconsistent with legislative objectives. See Rosenbloom v. Kokofsky, 373 Mass. 778, 780 (1977). In this case, the adoption of the continuing treatment doctrine runs contrary to the legislative aims undergirding the Commonwealth’s medical malpractice statutory framework and ignores decades of work and study by the executive and legislative branches regarding reducing the cost of health care in the Commonwealth, ensuring both affordability *390and access.1 Indeed, the court’s ignorance of the impact on the cost of health care of its sudden change of mind on the accrual of malpractice claims is staggering.2 If ever there was a case that *391cried out for judicial restraint and deferral to the branches of government best equipped to strike the proper balance between ensuring affordable and available health care with the protection of injured patients, this is it.

1. Legislative intent. After our adoption of the discovery rule in Franklin, the Legislature amended G. L. c. 231, § 60D, regarding the limitations period during which a minor might bring a claim for medical malpractice. See St. 1986 c. 351, § 23. The legislative history is clear that the Legislature knew that we had adopted the discovery rule, and this knowledge informed the course of the statute’s amendment. See Annual Report of the Special Commission Relative to Medical Professional Liability Insurance and the Nature and Consequences of Medical Malpractice, 1987 House Doc. No. 5262.3 In addition, our prior cases have carefully considered the intent of the Legislature in enacting major pieces of medical malpractice legislation, St. 1975, c. 362, and St. 1986 c. 351, and we have repeatedly acknowledged its concern regarding the costs associated with medical malpractice litigation and its efforts to ameliorate the costs of medical malpractice insurance. See, e.g., Darviris v. Petros, 442 Mass. 274, 283-284 (2004) (describing medical malpractice act of 1986 as “an ex*392haustive statutory scheme governing medical malpractice claims” and concluding that ‘“[e]xpanding the scope of damages available to plaintiffs who are victims of medical malpractice, and the period within which to make such claims, is contrary to the express intent of the Legislature in enacting St. 1986, c. 351”); McGuiggan v. New England Tel. & Tel. Co., 398 Mass. 152, 163 (1986) (Lynch, J., concurring) (”[T]he General Court has recently limited the amount that may be recovered by victims of medical malpractice in an attempt to solve what it perceives as a crisis in the medical profession brought about by the burgeoning cost of malpractice insurance. St. 1986, c. 351. The Legislature has acted to restrict recovery by injured litigants in order to limit the expense of practicing medicine”). See also Paro v. Longwood Hosp., 373 Mass. 645, 647 (1977) (tribunal requirement of medical malpractice act of 1975 was enacted ‘“as part of a comprehensive package designed to ensure the continued availability of medical malpractice insurance at a reasonable cost)”.

In sum, the result reached by the court today is anomalous in light of the legislative history and intervening decisions of this court, which recognize that the medical malpractice statutory framework is intended to moderate the cost and expense of medical malpractice litigation and that such a purpose is accomplished, in part, by the statute of limitations period. The court notes that the absence of legislative action cannot be interpreted as an affirmative rejection of the continuing treatment doctrine. In reaching this conclusion, however, the court ignores the fact that the statutory scheme was developed in tandem with the common law, and that expanding the period in which a medical malpractice claim may be brought markedly departs from the clear policy aims the Legislature sought to accomplish by repeatedly enacting legislation addressing malpractice claims, insurance, and the objective of reducing the time of exposure to such malpractice claims. See note 2, supra.

Finally, it is notable that the Legislature did include express ‘“exceptions” to the limitations period in G. L. c. 231, § 60D. First, there is an exception so any child under the age of six “shall have until his ninth birthday” to bring a claim. Second, the seven-year statute of repose has an exception for “the leaving of a foreign object in the body.” Id. Given the Legislature’s consideration and inclusion of these exceptions, I cannot conclude that a “continuing treatment” exception should be inferred where it was not included by the Legislature. “The fact that the Legislature *393specified one exception ... strengthens the inference that no other exception was intended.” Joslyn v. Chang, 445 Mass. 344, 350 (2005), quoting LaBranche v. A.J. Lane & Co., 404 Mass. 725, 729 (1989).

Thus, contrary to the court’s conclusion, it is apparent that, in the medical malpractice context, the Legislature has concurred with, and maintained, our uniformly applied ‘“accrual” standard, as articulated in Franklin.4 The statutory history and framework reflect a legislative choice to balance the goals of protecting defendant health care providers from extended tort exposure from stale claims, and of eliminating the “manifest injustice” which would result without the discovery rule for plaintiffs who are “blameless[ly] ignoran[t]” of information which might have put them on inquiry notice for purposes of investigating and possibly pursuing a claim. See Franklin, 381 Mass. at 618.

The decision today elevates this latter policy concern over the former, based on the court’s belief that the continuing treatment exception to the discovery rule would benefit patients by addressing a shortcoming it perceives in our current law, namely that patients are unable to make informed judgments as to negligent treatment while such treatment is ongoing. See, e.g., Harrison v. Valentini, 184 S.W.3d 521, 524 (Ky. 2005). In adopting the continuing treatment exception, however, the court fails to consider future impacts to the health care industry — impacts to which the Legislature has dedicated decades of study and, in response, carefully crafted legislation that reflects an effort to best balance competing policy concerns. Where the Legislature’s policy determinations are fairly clear, the court should defer to those judgments.

2. Adopting the exception by analogy. I also disagree with the court’s reasoning that our adoption of the continuing representation doctrine to the discovery rule in legal malpractice claims, see Murphy v. Smith, 411 Mass. 133, 137-138 (1991), justifies the adoption of a continuing treatment exception to the discovery rule in medical malpractice claims. I disagree with the proposition that, “just as a wronged client is permitted to benefit from his or her attorney’s efforts to correct a problem without the disruption of exploring the viability of a legal malpractice action, so, too, is a patient permitted that same benefit without the disruption of *394exploring the viability of a medical malpractice action.” Ante at 380. This statement both mischaracterizes the rationale underlying the continuing representation doctrine, and downplays the significant differences and interests at stake in those two arenas.

First, with respect to legal malpractice, as we have held, the continuing misrepresentation doctrine ‘“recognizes that a person seeking professional assistance has a right to repose confidence in the professional’s ability and good faith, and realistically cannot be expected to question and assess the techniques employed or the manner in which the services are rendered” (citation omitted). Murphy, 411 Mass. at 137. Implicit in the doctrine is an understanding that a person seeking legal services may not recognize that certain acts or omissions by an attorney constitute malpractice. Our justification of the doctrine says nothing, however, about permitting the representation to continue so that an attorney may correct an error arising from the attorney’s conduct; to the contrary, the facts in Murphy suggest that the attorney accused of malpractice did nothing to correct his alleged error. Id.

Moreover, the rationale for adopting the continuing representation doctrine is largely distinguishable from any analogous rule in the medical malpractice context. The ‘“continuing representation” principle that we recognized in the context of legal malpractice arose from assurances given by an attorney that he had attended to a legal issue that had no perceptible manifestation to the client. See id. at 136. In the field of legal malpractice, there are situations, such as the one presented by Murphy, where the attorney may assure the client that a certain task has been carried out correctly and where the client should be able to accept such representations in the absence of information to the contrary. Such a rule makes sense in the legal malpractice context because the alleged act or omission which gives rise to a claim and causes an injury to the plaintiff is caused somewhere other than in the plaintiff’s own body, often under circumstances remote from a plaintiff’s ability to detect circumstances which might put him or her on notice of a claim. See, e.g., Murphy, 411 Mass. at 137 (plaintiffs unaware of attorney’s purportedly improper certification of good record title until receipt of letter from neighbor’s attorney). The same cannot be said about the injury in a medical malpractice case, especially where the defendant’s actions had a direct and perceptible effect on the patient’s body.

The court also justifies its adoption of the continuing treatment exception by analogy to the continuing representation doctrine in *395legal malpractice on the ground that the Legislature otherwise used almost identical language to describe the limitations period for medical malpractice claims and for legal malpractice claims. It is apparent, however, that the Commonwealth’s legislation governing medical malpractice was enacted in light of a number of competing policy concerns that are unique to the health care industry — concerns simply not present in the practice of law — a practice that we as a court regulate.5

3. Conclusion. The court’s adoption of the continuing treatment exception to the discovery rule is inconsistent with the apparent legislative objectives underlying the Commonwealth’s medical malpractice statutory regime, particularly G. L. c. 231, § 60D. In my view, the court should apply the settled discovery rule to the facts of this case. As the court acknowledges, the defendant’s treatment ceased in December, 2005. Therefore, the plaintiffs’ action, brought in 2009, was not timely.