The plaintiffs are the parents of a child born with severe brain damage who died one day after delivery by Caesarian section. The mother (plaintiff) sues on her own behalf and as administratrix of her child’s estate. During a previous pregnancy, the mother was found to be Rh negative, a blood condition which, if not promptly treated, can lead to severe fetal *131damage and death to the child. For this reason, those undertaking the plaintiff’s obstetrical care regularly had blood tests performed so as to allow an appropriate response if the blood condition recurred. These tests were performed at the clinical laboratory of the MetroWest Medical Center, Inc. (MetroWest). The suit alleges that, after a particular blood test, two weeks went by before her obstetricians became aware that the test had indicated a dangerous level of antibodies and that, as a result, proper treatment was unduly delayed, resulting in an unnecessary Caesarian section and the death of the child. The plaintiffs sued the obstetricians and the obstetrical group who cared for her, MetroWest, and Dr. Adner, the associate chief of medicine at MetroWest and director of the MetroWest blood bank and hematology laboratory.
At the defendants’ request, a medical malpractice tribunal (tribunal), consisting of a judge in the Superior Court, a physician, and an attorney, was convened, pursuant to G. L. c. 231, § 60B, to which the plaintiff submitted her claims, offering to prove that all of the defendants had acted negligently and that Adner had acted negligently because the laboratory had no policies or procedures for assuring that treating physicians are promptly notified of dangerously abnormal results in tests such as were performed for the plaintiffs. The obstetrical practice and the laboratory were located at separate MetroWest campuses and, according to evidence before the tribunal, it was the laboratory’s practice to have test results delivered to the doctors’ offices at the other campus by an independent courier service. According to a letter from the treating physician to the plaintiff after the death of her child, the physician stated that he had “routinely” been informed by telephone when there was an abnormality, but in this case he “did not receive a phone call or written report.” The tribunal found “sufficient evidence to raise a legitimate question of liability appropriate for judicial inquiry” in respect to the obstetricians, the obstetrical practice, and MetroWest. As to Dr. Adner, however, the tribunal found that there was hot sufficient evidence. The plaintiffs declined to post the bond required by § 60B in case of such an adverse finding, and a judge in the Superior Court granted Adner’s motion to dismiss the case against him. The plaintiffs appealed, and we transferred the case here on our own motion.
I
The plaintiff and the defendant disagree about whether they *132stood as doctor and patient to each other. The plaintiff contends that, if there was no doctor-patient relationship, then the medical tribunal had no jurisdiction over the case and she should have been permitted to bring the case as a simple negligence action in the Superior Court. In that event, replies the defendant, he is not hable as a matter of law, as he owed the plaintiff no duty.4
A
The parties spend considerable effort arguing about whether a doctor-patient relationship existed between the plaintiff and Dr. Adner, the director of the laboratory. They are invited to this exercise by our cases. In Doherty v. Hellman, 406 Mass. 330 (1989), we asked whether the director of a radiation therapy service stood in that relationship to the patient of another doctor in the facility. We held he did not. And in Lambley v. Kameny, 43 Mass. App. Ct. 277 (1997), the Appeals Court asked whether a psychiatrist who examined an applicant for employment at the request of the prospective employer and, by his alleged misdiagnosis, caused the candidate not to be hired, stood in such a relation to the candidate. The court concluded that he did. Section 60B does not require the existence of a doctor-patient relationship as a predicate for its application. It states that “[ejvery action for malpractice, error or mistake against a provider of health care shall be heard by a tribunal consisting of a single justice of the superior court, a physician . . . and an attorney . . . .” The statute goes on to state that “a provider of health care shall mean a person, corporation, facility or institution licensed by the commonwealth to provide health care or professional services as a physician . . . [or] hospital. . . or an officer, employee or agent thereof acting in the course and scope of his employment.” Our early cases involving G. L. c. 231, § 60B, inserted by St. 1975, c. 362, § 5, do not speak of this requirement. See Little v. Rosenthal, 376 Mass. 573 (1978); Campbell v. Thornton, 368 Mass. 528 (1975). In Campbell, supra, the question arose whether the chief of the medical staff of a hospital could be held hable for an accident that occurred on *133one of the hospital’s services, and we said: “The claim against Dr. MacDonald is predicated solely on the fact that . . . the doctor was the chief of the hospital’s medical staff. This alone is not sufficient to impose liability on him for the negligence ... of other doctors or of agents ... of the hospital. . . . His position . . . would not, without more, make Dr. MacDonald liable under the doctrine of respondeat superior . . . .” Id. at 535-536. The term doctor-patient relationship (as a predicate for the applicability of § 60B) crept into the law, it appears, in Kapp v. Ballantine, 380 Mass. 186, 193 (1980), where — in a case which did not turn on the issue — we summarized the law as follows: “A plaintiff’s offer of proof as to negligence will prevail before a malpractice tribunal, under the Little directed-verdict test, (1) if a doctor-patient relationship is shown, (2) if there is evidence that the doctor’s performance did not conform to good medical practice, and (3) if damage resulted therefrom.” We offered as authority Little v. Rosenthal, supra, which nowhere uses the term or adverts to the concept. It has, however, become boilerplate. This is unfortunate. The term has relevance not in establishing the applicability of § 60B but in establishing whether a particular defendant has the requisite duty of care to a particular plaintiff sufficient to lay the predicate for malpractice as negligence, or for breach of a contractual undertaking to the patient, or for a failure to obtain appropriate consent from the patient. We discuss this distinct issue below. In respect to § 60B, however, although there may be difficulties at the margins in determining who is a health care provider within the terms of the statute, see Perez v. Bay State Ambulance & Hosp. Rental Serv., Inc., 413 Mass. 670, 675-676 (1992) (ambulance company not health care provider), the introduction of the term “doctor-patient relationship” may be an example of explaining the obscure by the more obscure. In Lambley, supra, the Appeals Court obviously demoted the requirement of such a relationship, id. at 283 (assuming “without deciding, that such a relationship or its equivalent is a jurisdictional prerequisite”), relative to the threshold set out in the statute itself. Although it was far from clear that the psychiatrist stood in such a relationship to the prospective employee, it was a great deal clearer that, in making his diagnosis, the psychiatrist was a provider of health care. And we also agree with the Appeals Court that this focus accords with at least one important purpose of § 60B, that there be expert screening of cases “which directly implicate *134the professional judgment or competence of a provider.” Id. at 282.
Viewed in this light, there can be no doubt at all that the plaintiffs sued Dr. Adner in his capacity as a provider of health care. The suit against MetroWest, which performed the laboratory services, has been allowed to proceed, and not surprisingly. If Dr. Adner were charged with overbilling or some other such claim having nothing to do with the exercise of medical judgment, we might have a different case, but the complaint against Dr. Adner implicated the same issues of medical judgment as those that must lie at the basis of the plaintiffs’ claim against MetroWest. In modem medical practice, there is a variety of services performed by skilled practitioners who never encounter the patient personally: radiologists read X-rays, ultra-sound recordings, MRI and CAT scans, all of which may have been physically administered by technicians who produce the records that are then interpreted; pathologists examine tissue samples obtained from the patient by others; and laboratory tests of various sorts are performed in complicated organizational settings, where the person actually encountering the patient may be the least skilled member of the system.
We have no difficulty in concluding that, if the suit against Dr. Adner may proceed at all, it must be as he is a “provider of medical services,” and therefore under the procedures set out in § 60B.
B
That this was a suit against Dr. Adner in respect to the provision of medical services does not, however, end the inquiry. The analysis, interpretation, and reporting of the blood tests were medical services within the terms of the statute. The dispute which occupies the parties here, regarding the existence of a doctor-patient relationship, may be better understood as centering on the question whether it was the defendant who provided those services. The claimed malpractice, or wrongful conduct, or breach of duty by Dr. Adner was his failure to institute practices and procedures such that critical test results reached a patient’s physician promptly. What must be shown is that this defendant had such a duty and that it was a duty owed to the *135plaintiff.5 This poses the troubling issue in this case more sharply than the amorphous question whether Dr. Adner stood in a doctor-patient relationship to the plaintiffs.
There is a variety of organizational schemes that obtains in the provision of modem medical services. In the Doherty case, supra, the plaintiff based her claim on negligent treatment and on the failure to inform her of the risks of an innovative course of radiation therapy. The defendant was not her treating physician. Rather, he was the director of the center at which she was treated. He had designed and pioneered the treatment that patients at the center received and set the standards for its administration. He reviewed the results of treatments in the center and reported them in the medical literature. See id. at 334. Because the defendant was not the plaintiff’s treating physician and because her treating physician had undertaken to provide her with appropriate informed consent, we held that the defendant could not be held liable on his own account for any neghgence or failure to inform the plaintiff. Nor could he be held vicariously liable as director of the program for any negligence or failure of the treating physician. This is an important result because medical services are often rendered in settings with some degree of hierarchical organization, at the head of which is a senior, respected figure, even though the patients receive treatment from their own physicians, who are skilled professionals expected to exercise independent judgment. The Campbell case, supra, is an extreme example of this attempt to impose personal liability on one who occupies a place, as chief of a hospital’s medical staff, at the top of that hierarchy. On the other hand, just because so much of modem medical service is rendered in bureaucratized and segmented settings, it is important not to announce a mle in terms of which no human being is ever responsible for failures in the system — although, of course, the organization (here MetroWest) as an entity may be fiable. Returning to the example of the radiologist who interprets an X-ray taken, usually, not by the radiologist but by a technician, it could scarcely be maintained that the *136radiologist does not have a duty to the patient to read the film with care, and, if it seems to have been inadequately produced, to direct that it be taken again. Similarly, the radiologist obviously has a duty to the patient to report significant results to the patient’s treating physician. If he does not do the reporting himself but entrusts it to others, it is a fair question whether the steps he took to have the report reach its destination were reasonable — whether or not they succeeded.
This case does not fall in any obvious way under any of the competing paradigms. Needless to say, Dr. Adner personally did not draw the patient’s blood nor perform the analysis himself. If crucial results were not noticed or not communicated to her treating physician, this may have been the result of negligence on some person’s part. But the work of a unit may be so compartmentalized and routinized that no such person exists, and then the fault — if there is fault — would only be that of the unit and the institution of which it is a part. (As we have noted, a medical tribunal has allowed the suit against Metro-West to proceed.) This would be analogous to tort cases in which only a corporate defendant is liable for negligence and no individual employee can be identified who shares responsibility personally with the corporation. We must avoid holding that in every case where there is institutional liability there must always be a responsible individual as well, because an overly determined search for such an individual may sometimes turn up no one else than some distant senior member of the institutional hierarchy with formal responsibility for the workings of the systems reporting to him in the institutional chain of command. This may be appropriate in the military, but it is not the rule as to civil liability in tort. Such a rule would, in effect, make individual supervisors personally responsible not as employers — which they rarely are — but under a version of respondeat superior, which has regularly been rejected in the medical as well as other contexts. The question in these cases is whether the defendant owed a duty personally to the patient. The radiologist who reads an X-ray and reports his conclusion clearly does; the chief of medicine of a hospital, simply by virtue of his position, does not. The chief’s duty is to his employer, not to patients who are not his patients but those of the treating physicians. See Doherty v. Hellman, 406 Mass. 330 (1989); Campbell v. Thornton, 368 Mass. 527 (1975).
*137II
With these considerations in view, we turn to the plaintiff’s proffer of evidence to determine whether it was sufficient “to raise a legitimate question of liability appropriate for judicial inquiry.” G. L. c. 231, § 60B. We have held that the standard under § 60B is similar to that on directed verdict, see St. Germain v. Pfeifer, 418. Mass. 511, 516 (1994), and it is on that standard that we review the plaintiff’s claim. Once again, we recall that the plaintiff must show that the defendant personally owed her a duty of care which he breached. His position as director of the laboratory will not of itself be sufficient to make out such a duty to the plaintiff, but neither does the fact that he personally did not treat her, perform the analysis of her blood, or undertake personally to report the results automatically absolve him of liability. It depends on the particular facts..
The plaintiff proffered the affidavits of two experts, whose qualifications for these purposes seem entirely adequate. They both agreed that the blood test results were of critical importance. One expert, Dr. Underkofler, a board certified obstetrician, stated that “[a]ny lab running serial Rh titers for an obstetrician should have in place procedural guidelines to insure that such information gets to the physician in an appropriate manner.” He characterized a two-week delay as “indefensible.” The other expert, Dr. Liebman, a board certified specialist in hematology and internal medicine and the medical director of a hemostasis laboratory, went further and stated that “the failure of the MetroWest blood bank and its director, Dr. Marvin Adner, to develop a proper protocol to immediately notify physicians of a significant change in anti-D titers failed to meet the proper standard of good laboratory practice placing [the plaintiff’s] fetus at greater risk of death by hydrops fetalis.” These affidavits certainly support the plaintiff’s claim against MetroWest, but Dr. Liebman’s statement that Dr. Adner was also personally responsible is a legal conclusion which the affidavits do not support.
The plaintiff’s obstetrician reported that he. “could not locate the paper report” of the plaintiff’s blood test and that he “felt he should have had a phone call since this is a significant change.” Although protocols for blood testing in cases like the plaintiff’s were offered in evidence, they made no reference to the reporting of test results. Jeannine Bensette, a member of Dr. Adner’s staff, testified that testing procedures and policies were *138developed in consultation with and approved by Dr. Adner. Although there was no formal protocol, it was the practice of the laboratory to transmit test results to the physicians in the MetroWest campus where the plaintiff’s obstetrician worked by private courier, an independent contractor. She also testified that, after this incident, Dr. Adner instituted a policy requiring telephone notification of significant results in cases such as this one.
We conclude that these proffers are “sufficient to raise a legitimate question of liability appropriate for judicial inquiry.” G. L. c. 231, § 60B. More specifically, there was sufficient evidence to suggest that Dr. Adner’s involvement with the care and routines at the laboratory was sufficiently personal that the case may not as a matter of law be classed with cases such as Doherty, supra; Campbell, supra; or Lyon v. Morphew, 424 Mass. 828, 833 (1997) (“general supervisory role with regard to the engineering department” not sufficient to make hospital’s chief operating officer personally liable for accident when safety railing was improperly removed by engineering department). Counting for holding Dr. Adner personally responsible is the evidence that the personnel of the laboratory looked to him to set the details of the policies and procedures on just this matter that is the basis of the plaintiff’s claim.6 What would count against Dr. Adner’s personal responsibility would be evidence that the organization was a large one and his responsibility for details such as this policy was only formal or hierarchical, or that others in the organization assumed a more direct relation to persons in the plaintiff’s position. But no evidence of this sort is cited; we do not even know the size and organizational structure of the laboratory. In this state of the proof, the plaintiff should have been allowed to go to trial. Certainly the evidence raises a legitimate question that Dr. Adner’s involvement with the care of patients receiving blood tests at the laboratory was at least as personal as that of the official of a college who “designed and supervised the installation of the security system . . . [and] was responsible for the patrol pattern and the network of locks” which proved insufficient to prevent the rape of a student there. Mullins v. Pine Manor College, 389 Mass. 47, 56 (1983).
We conclude only that there was sufficient evidence to preclude dismissal for failure to post the statutory bond required *139by § 60B. The judgment of the Superior Court is vacated and the case is remanded for further proceedings consistent with this opinion.
So ordered.
The defendant also argues that the plaintiff’s jurisdictional argument comes too late, having been raised for the first time in this appeal. Because we conclude that' the case was properly before the tribunal, we need not decide whether this would be the sort of jurisdictional argument that may be raised at any time.
Assuming there was such a duty and it was indeed owed to the plaintiff, there will be further questions. For example, what was the practice for transmitting critical results to treating physicians; had Dr. Adner acted reasonably by allowing the routine (if that was the routine) for delivering test results — the use of an independent courier service — to remain in place; did the results reach the physician’s office in regular course and the failure to take them into account occur thereafter?
We do not intend to foreclose the possibility that in fact Dr. Adner had discharged his responsibilities appropriately and the fault was that of others.