On the morning of March 22,2008, the plaintiff, Crystal Washington, went to the emergency department at Brigham and Women’s Hospital (BWH). There, she complained of various symptoms, and the defendant, Hilarie Cranmer, M.D., examined, treated, and discharged her. After she returned home, Washington suffered a stroke overnight, which caused permanent neurological damage. She filed this action in Superior Court alleging that Dr. Cranmer caused her injuries by not complying with the applicable *675standard of care.2 In accordance with G. L. c. 231, § 60B, the matter was referred to a medical malpractice tribunal.3 After a hearing, the tribunal concluded that Washington’s offer of proof, even if properly substantiated, was insufficient to raise a “legitimate question of liability appropriate for judicial inquiry.” G. L. c. 231, § 60B, inserted by St. 1975, c. 362, § 5. Washington did not post the $6,000 bond required by the statute, and her action therefore was dismissed.4 See ibid. (“[i]f [the] bond is not posted within thirty days of the tribunal’s finding the action shall be dismissed”). Because we agree with Washington that her offer of proof was adequate, we reverse.
Standard of review. “Before a medical malpractice tribunal, a plaintiff’s offer of proof must (1) show that the defendant is a provider of health care as defined in G. L. c. 231, § 60B; (2) demonstrate that the health care provider [in question] did not conform to good medical practice; and (3) establish resulting damage.” Saunders v. Ready, 68 Mass. App. Ct. 403, 403-404 (2007), citing Santos v. Kim, 429 Mass. 130, 132-134 (1999). The relevant standard of care is the one that applies to “the average qualified physician in his or her area of specialty” (in this case, an emergency medicine physician). Medina v. Hochberg, 465 Mass. 102, 106 (2013). Whether the physician met the applicable standard of care generally can be answered only with the aid of expert opinion. See Kapp v. Ballantine, 380 Mass. 186,190 & n.4 (1980). The expert opinion must be rooted in the record evidence and not be based on speculation, conjecture, or assumptions not supported by the evidence. Blood v. Lea, 403 Mass. 430, 434 (1988).
In determining whether a plaintiff’s offer of proof is sufficient, “[t]he question to be decided ... by the tribunal is a factual one.” Kopycinski v. Aserkoff, 410 Mass. 410, 413 (1991). However, the tribunal’s task is “akin to a trial judge’s evaluation of a motion for a directed verdict.” Cooper v. Cooper-Ciccarelli, 77 Mass. App. *676Ct. 86, 91 (2010), citing Little v. Rosenthal, 316 Mass. 573, 578 (1978). Thus, “the tribunal may not examine the weight or credibility of the evidence.” Ibid., citing Perez v. Bay State Ambulance & Hosp. Rental Serv., Inc., 413 Mass. 670, 676 (1992). Instead, it must consider the proof in the light most favorable to a plaintiff.5 Blake v. Avedikian, 412 Mass. 481, 484 (1992). “An offer of proof is sufficient if ‘anywhere in the evidence, from whatever source derived, any combination of circumstances could be found from which a reasonable inference could be drawn in favor of the plaintiff.’ ” Thou v. Russo, 86 Mass. App. Ct. 514, 516 (2014), quoting from St. Germain v. Pfeifer, 418 Mass. 511, 516 (1994).
Background, a. Facts. At 9:51 a.m., on March 22, 2008, Washington, then thirty-seven years old, arrived by ambulance at the BWH emergency department in Boston. Washington reported that for the past twenty-four hours she had been experiencing body weakness, left arm weakness, difficulty keeping her balance, increased blood sugars, dizziness, and blurry vision. A triage nurse conducted the initial patient intake, and recorded Washington’s blood pressure at 234/153. Intake records also show that Washington had a “gait disturbance” that made her a “fall risk,” that she was “slow to respond” to questions, and that she had some difficulty with finding words (described in a later record as “mild dysarthria”). In addition, despite her relative youth, Washington already had a rich medical history that included diabetes, chronic hypertension, high cholesterol, and — in 2004 — a stroke.
Dr. Cranmer apparently first saw Washington at 10:10 a.m. After conducting physical and neurological examinations, Dr. Cranmer concluded that Washington was alert, ambulatory, and oriented to person, place, and time; that her neurological exam was “normal” apart from her being hypertensive; and that she had a “regular” heart rate and rhythm and “[n]o obvious. . . motor or sensory deficits.” Washington also exhibited no chest pain, shortness of breath, headache, or fever. Washington confirmed that she had not taken her blood pressure medication that morning (March 22).
In response, Dr. Cranmer directed the assisting nurses to give Washington two medications (labetalol and hydrochlorothiazide) *677to lower her blood pressure. At 10:45 a.m., Washington received the prescribed medication orally. Dr. Cranmer also ordered a computer tomography (CT) scan of Washington’s head and various “lab work,” and placed Washington on a cardiac monitor and electrocardiogram (EKG) monitor. Two hours later, at 12:45 p.m., Washington’s blood pressure had fallen to 126/73.
The radiology report from the CT scan did not reveal any intracranial hemorrhage or other acute abnormalities.6 Except for Washington’s elevated blood sugar (258), her laboratory results were found to be within normal limits. At 1:07 p.m., Dr. Cranmer ordered “MRI/A” (i.e., magnetic resonance imaging/angiography) scans in order to evaluate Washington for a “Head Stroke.” In addition, Dr. Cranmer admitted Washington to the emergency department observation section (OBS) and ordered that she be monitored for a transient ischemic attack (TIA)7 and for continued blood pressure control. Dr. Cranmer instructed that Washington’s OBS admission was subject to the “transient neurologie” protocol, and she identified aspirin, MRI/A scans, and “consultation” on an as-needed basis as the appropriate interventions.
By the early afternoon, Dr. Cranmer had made plain that her plan of continued treatment and care for Washington was contingent on the results of the MRI/A scans that she had ordered. Specifically, Dr. Cranmer intended to discharge Washington after her dinner meal, unless there were “new CVA [cerebral vascular accident] findings on the MRI/A” scans. Dr. Cranmer noted that there was “no need for urgent neuro unless MRI/A is as above [i.e., new CVA findings].” The nurses continued to monitor Washington while she rested in the OBS, and they administered aspirin to her at 1:45 p.m. Meanwhile, Washington’s blood pressure had begun to rise again. By 2:00 p.m., it had risen to 153/101, and at 2:45 p.m., it was at 165/107.
At approximately 3:30 p.m., in advance of the scheduled MRI/A scans, the nursing staff administered medication to Washington in an effort to treat her anxiety regarding the proposed procedure. Ultimately, Washington’s claustrophobia prevented her from going through with the MRI/A scans. After the failed MRI/A attempt, Washington was returned to the OBS. The same “E[mer-*678gency] Department] progress note” that stated that Washington was “unable to do the MRI” also noted that Washington reported that “her speech still isn’t baseline.”
By 4:00 p.m., Washington’s blood pressure had risen to 174/106, and it remained at that elevated level (174/105) at 6:00 p.m. when she was discharged.8 Before she was discharged, Washington received another dose of blood pressure medication orally, and Dr. Cranmer issued instructions to “TAKE YOUR MEDS” and to “RETURN FOR ANY WORSENING ANYTHING.” Dr. Cranmer also directed Washington to follow up with a neurologist and her primary care physician within two weeks.
At home that night and the following morning (March 23), Washington’s condition declined. She returned to the BWH emergency department by ambulance at 9:45 a.m. on March 23, presenting with left-sided facial droop as well as left-sided weakness. Her “mild dysarthria” from the day before had “worsened,” and her speech was now “slurred.” In addition, her blood pressure had risen to 218/139 and she was hyperglycemic. Washington was given labetalol to lower her blood pressure and insulin to lower her blood sugar. A new CT scan did not evidence signs of a stroke. However, follow-up tests, including MRI/A scans, revealed that Washington had suffered one.9
On April 2, 2008, Washington was discharged from BWH and transferred to Spaulding Rehabilitation Hospital (SRH). Eventually, she was discharged from SRH. Washington now is confined to a wheelchair and, although she can feed herself, “[s]he needs assistance for all other activities of daily living.”
b. Expert opinion. Before the tribunal, Washington submitted her medical records and an expert opinion letter that had been prepared by Kenneth C. Fischer, M.D., a board-certified neurologist. Among his other conclusions, Dr. Fischer opined that Washington had sustained a “stroke and resultant severe and permanent neurological injury as the direct result of the substandard care and treatment” rendered by Dr. Cranmer on March 22, 2008. According to Dr. Fischer, the average qualified emergency medicine physician would (a) “recognize and appreciate” the complaints and medical *679history presented here as signs and symptoms of a “hypertensive emergency” and TIA; (b) “administer intravenous antihyperten-sive agents” to the patient in order to lower her blood pressure; and (c) order an “immediate neurology consultation” and admit the patient to the neurology intensive care unit (ICU) for monitoring and treatment. Dr. Fischer further opined that if Dr. Cranmer had acted in conformance with this course of treatment, then Washington would have received blood thinner medication “at the new onset of symptoms and, more likely than not, she would not suffer from the severe and permanent neurological injury with which she lives today.”
Discussion. As the record amply reveals, when Washington arrived at the BWH emergency department on March 22, she presented with multiple causes for serious concern. In sum, a patient who already had experienced a stroke at age thirty-three — and who had chronic hypertension and poorly controlled diabetes — was experiencing severely high blood pressure, difficulty walking and finding words, and paresthesia. However, the record also reveals that Dr. Cranmer’s initial course of action in responding to Washington’s symptoms and medical history generally conformed to the standard of care set forth in Dr. Fischer’s opinion letter. Far from ignoring the serious concerns that Washington’s case presented, Dr. Cranmer understood that Washington was in a hypertensive crisis that presented attendant risks for a TIA or stroke. This is well demonstrated by the initial actions she took: conducting a neurological examination, prescribing blood pressure medication, ordering laboratory work and a CT scan of Washington’s head, and scheduling MRI/A scans to evaluate her for a stroke.10 Indeed, Dr. Cranmer’s own appellate brief highlights that the actions she pursued were done in an effort to address the TLA and stroke risks that Washington presented.
*680By midday, the medication that Dr. Cranmer ordered had reduced Washington’s blood pressure to normal limits, and the testing that was done did not reveal any significant abnormalities. However, Dr. Cranmer herself recognized the inconclusiveness of these promising indicators and was awaiting the results of the MRI/A scans.11 Her contemporaneous notes indicate that she considered the MRI/A scans critical, stating that the patient could be discharged based on the MRI/A findings and that no urgent neurological care was needed unless the MRI/A results were abnormal.
Once Washington’s claustrophobia prevented her from going forward with the MRI/A scans, this in turn prevented Dr. Cranmer from having the benefit of the MRI/A results that she considered critical. Nevertheless, Dr. Cranmer went ahead and discharged Washington even though her blood pressure had risen significantly since its midday low, and at midaftemoon, Washington reported that she continued to have difficulty speaking. In our view, there is a sufficient dispute over whether Dr. Cranmer did enough prior to Washington’s discharge to preclude the case *681from being dismissed at the initial tribunal stage.12
To be sure, Dr. Fischer’s opinion letter does not grapple directly with Washington’s inability to go forward with the MRI, and that omission diminishes the letter’s force. However, it does state that “imaging studies” should be conducted in order to rule out stroke. Although one imaging study was performed (the initial CT scan), Dr. Cranmer discharged Washington without ever having received the additional benefit of another imaging study (the MRI/A) that she herself deemed important in order to rule out a stroke. Thus, Dr. Cranmer’s own course of action provides support for Dr. Fischer’s assertions that she diverged from the applicable standard of care.
Moreover, Washington’s offer of proof is based on the overarching theory that BWH discharged a high-risk patient in the throes of a “hypertensive emergency” before that emergency was adequately resolved, and there was record support that the crisis had not been resolved.13 As noted, Dr. Fischer specifically opined, inter alla, that Dr. Cranmer should have admitted Washington to the neurology intensive care unit for “close monitoring and treatment,” and that had this been done, the administration of blood thinner medication likely would have prevented Washington’s injuries. In contrast, Dr. Cranmer discharged Washington after giving her more blood pressure medications, and telling her to come back if she presented new symptoms.
We emphasize that the details of what transpired between the failed MRI/A attempt and Washington’s discharge are not well developed in the current record. In the end, the evidence at trial may show that Dr. Cranmer acted entirely reasonably under the difficult circumstances presented and that she fully complied with *682the applicable standard of care.14 However, that is not something that can be resolved on the current record without straying into the inappropriate role of weighing the evidence. See Cooper v. Cooper-Ciccarelli, 77 Mass. App. Ct. at 91. We conclude that Washington has satisfied her initial burden of “rais[ing] a legitimate question of liability appropriate for judicial inquiry.” G. L. c. 231, § 60B. Accordingly, we reverse the judgment of dismissal.
So ordered.
The claims were based on negligence and related theories.
General Laws c. 231, § 60B, the medical malpractice tribunal statute, applies broadly to “all treatment-related claims” involving a “provider of health care.” Vasa v. Compass Med., P.C., 456 Mass. 175, 177 (2010), quoting from Little v. Rosenthal, 376 Mass. 573, 576 (1978). See generally Jacobs & Laurence, Professional Malpractice §§ 5.1-5.7 (2007).
Washington filed a motion to reduce the bond to $100 on the ground that she was indigent. That motion was denied by a judge in the Superior Court, and Washington did not appeal that ruling. See Faircloth v. DiLillo, 466 Mass. 120, 124 (2013).
Section 60B “requires a physician member on the tribunal so that he or she may lend expertise in medical matters and assist in screening out nonmeritorious claims.” Blood v. Lea, 403 Mass, at 435.
The scan did show indicia — “lacunes” in the “left thalamus and caudate head” — of the “old” 2004 stroke.
A transient ischemic attack is an “acute neurologie deficit resulting from circulatory impairment that resolves within 24 hours." Steadman’s Medical Dictionary 1849 (28th ed. 2006).
Although Washington was not formally discharged until after she had completed her dinner at approximately 6:00 p.m., BWH records indicate that Dr. Cranmer signed the discharge order at 4:47 p.m.
Specifically, an initial MRI/A scan revealed “brainstem infarction in the right pons and bilateral medulla,” and a second MRI/A scan showed additional damage.
There are some remaining potential discrepancies between the initial actions that Dr. Cranmer took and the standard of care that Dr. Fischer described. For example, Dr. Cranmer administered blood pressure medication orally instead of intravenously, and there is no evidence that she consulted with a neurologist. Given the grounds on which we rest, we need not decide whether Dr. Fischer’s opinion letter adequately explained how any such differences mattered. Additionally, we note that neither party has addressed the potential inconsistency between Dr. Cranmer’s conclusion that Washington’s neurological examination was normal, and near contemporaneous observations made by the triage nurse that Washington had both a “gait disturbance” and difficulty finding words. We do not rest on that ground.
In his opinion letter, Dr. Fischer stated that “[a] hypertensive emergency is a severe and persistent elevation in the blood pressure with acute impairment of an organ system (end-organ dysfunction) including, but not limited to, arm and leg weakness, paresthesia, gait disturbance, blurry vision, and difficulty with word finding.” Based on that statement, the dissent posits that Washington could not have presented with “hypertensive emergency” because “[t]he laboratory work, imaging studies, and [EKG] results here furnished no indication of acute end-organ damage.” Post at 683. Thus, according to the dissent, “Dr. Fischer’s opinion (and by extension the plaintiff’s malpractice claim) rests on an ill-based factual assumption, namely, the presence of a hypertensive emergency, which, in the end, is not supported by anything in the record.” Post at 684. This conclusion rests on a misreading of what Dr. Fischer actually said. He characterized a hypertensive emergency principally by a set of symptoms, all of which were present on Washington’s arrival at the emergency department. He did not state, or even imply, that a patient who exhibits those very symptoms cannot present with a hypertensive emergency absent independent proof of “end-organ dysfunction” through laboratory tests and the like. The dissent reads Dr. Fischer’s opinion with a gloss that, at a minimum, is unduly demanding at the tribunal stage. In addition, it passes over the fact that Dr. Cranmer herself recognized that the testing results were inconclusive and that MRI/A scans should be done. Further, it bears noting that the additional CT scan done after the stroke was itself negative (unlike the MRIs), thus corroborating Dr. Cranmer’s own views regarding the limited diagnostic role that CT imaging plays in this context. Especially given our duty to read the record in the light most favorable to Washington, we disagree with the dissent’s conclusion that, as a matter of law, no reasonable jury could have concluded that she presented a “hypertensive emergency.”
Dr. Cranmer highlights that one hospital record characterized Washington’s discharge from BWH on March 22, 2008, as her having “left against medical advice.” However, that statement was not contemporaneous, but instead appears in a much later BWH report that followed Washington’s stroke. If a patient who faced an obvious stroke risk was in fact discharged against the medical advice of the attending physician, a reasonable juror might expect to see that point noted in a hospital record that accompanied her discharge.
In one BWH record, Dr. Cranmer suggested that she was comfortable discharging her patient because Washington had become “asymptomatic.” However, that statement manifestly was not true at the time that Washington was discharged given the elevation in her blood pressure over the course of the afternoon. A different BWH record more accurately states that Washington “was discharged after getting some symptomatic relief’ (emphasis added).
We do not mean to suggest that this case necessarily should proceed to trial. Depending on how the facts are developed in discovery, the case might be appropriate for summary judgment.