SELYA, Circuit Judge.

Plaintiff-appellant Rafaela Cortes-Irizarry (Cortes), suing on behalf of her minor child, Rafael Jose Musiz Cortes (Jose), challenges an order granting summary judgment to Corporacion Insular de Seguros (CIS) and its insured, Juan Ramon Gonzalez Aristud (Dr. Gonzalez), in a medical malpractice action. See Irizarry v. CIS, 928 F. Supp. 141, 147-48 (D.P.R. 1996). We vacate the order and remand for trial.

I. BACKGROUND

Although the accepted summary judgment protocol calls for us to cast the facts in the light most complimentary to the plaintiff's position, consistent with record support, see, e.g., Garside v. Osco Drug, Inc., 895 F.2d 46, 48 (1st Cir. 1990), we temper that protocol here to the extent that we set off, as point and counterpoint, conflicting evidence where the clash helps to illuminate pertinent legal issues. For simplicity's sake we omit any further reference to CIS and treat its insured as if he were the sole defendant.

Dr. Gonzalez, a specialist in obstetrics, provided prenatal care to Cortes after she became pregnant with Jose. On December 15, 1979, Cortes related to Dr. Gonzalez that her last menstrual cycle prior to conception began on November 2 and lasted only two days. The length of her immediately preceding menses was three days, and her periods typically had lasted two or three days during the year prior to her current pregnancy. Based on this data, Dr. Gonzalez calculated Cortes' estimated delivery date (EDD) to be August 9, 1980. He delivered Jose by cesarean section on July 30, 1980. The newborn weighed eight pounds, eight and three-quarter ounces (two pounds more than Cortes' first child) and exhibited no fetal distress.

According to the defendant's computations, Cortes was in her thirty-ninth week of pregnancy when the baby arrived. This calculation forms the nub of the case. The plaintiff's theory is that Dr. Gonzalez misfigured the baby's fetal age and, consequently, allowed the pregnancy to continue beyond forty-two weeks, thus bringing into play a risk factor known as "post-datism" or "post-maturity." A post-dated fetus is at risk of oxygen deprivation during its extended stay in the mother's womb, and brain damage is a predictable result. While Jose, at birth, displayed no detectable symptoms suggesting a post-dated delivery, the circumstances of the delivery revealed some indications of potential perinatal difficulties; for instance, the cesarean section took twenty-one minutes (roughly twice as long as the norm), and, on one view of the proof, a tracheal catheter was used to incubate the newborn.

Time resolved these mixed signals. Jose showed signs of neurologic abnormality at three months and was diagnosed with impaired motor development and hearing loss at fourteen months. His condition worsened as the years passed. As an adolescent, he was diagnosed as severely brain damaged, epileptic, and profoundly deaf. At that juncture, Cortes, then a citizen of Florida, sued Dr. Gonzalez in Puerto Rico's federal district court, see 28 U.S.C. §(s) 1332(a) (diversity jurisdiction), alleging that the physician's negligence caused her son's infirmities.

Cortes' case rests primarily on the opinions of two experts. An obstetrician, Dr. Bernard Nathanson, opined that a competent obstetrician, rather than relying upon a reported two-day menstrual period to calculate a gravid woman's EDD, would have launched a more detailed gynecologic investigation. Had Dr. Gonzalez done so, the witness stated, he would have discovered that Cortes' actual EDD was July 9, 1980, and he would have recognized that a substantial risk of post-datism arose when her pregnancy extended past the EDD (a risk which he presumably could have negated by performing the cesarean section earlier). In reaching these conclusions, Dr. Nathanson stressed the unusual brevity of the reported period (especially as contrasted with Cortes' previous menses) and Dr. Gonzalez' failure to confirm the EDD by performing various tests which the witness stated were available in 1979-1980 (e.g., a B-scan ultrasound examination). In Dr. Nathanson's opinion, the pregnancy was post-dated, and the defendant's failure to realize it and take corrective action violated the prevailing standard of care.

Dr. Nathanson also disputed Dr. Gonzalez' assertion that he in fact performed a manual pelvic examination at Cortes' initial appointment and subsequently measured her uterus throughout her pregnancy to corroborate the EDD. Dr. Nathanson saw no evidence that these steps had been taken. Moreover, Dr. Gonzalez' office record did not mention either the periodic uterine measurements or their results. Although some of Cortes' prenatal charts apparently had been lost, Dr. Nathanson stated that these data "are so vital that they should be in [Dr. Gonzalez'] record in any case had he done them."

The plaintiff's second expert, Dr. Allan Hausknecht, a neurologist, diagnosed Jose as suffering from Lennox Gasteault Syndrome (LGS). This neurological condition is caused roughly fifty percent of the time by perinatal brain damage (resulting from a lack of sufficient oxygen to the fetal brain). Doctor Hausknecht stated that, in his experience, this percentage increases sharply when, as in this instance, no evidence of any other known cause exists. Noting that the gradual development of Jose's condition was characteristic of a post-mature fetus, Dr. Hausknecht rendered an opinion that Jose's brain damage resulted from the post-datism which Dr. Nathanson had identified. This opinion was bolstered in some degree by Dr. Nathanson's statement that, while some post-dated infants will show immediate signs of placental senescence, such as meconium-stained amniotic fluid or peeling of the skin (Jose had neither), many others will appear asymptomatic at birth yet manifest the effects of post-datism at a later time.

To be sure, the plaintiff's evidence was hotly contested. The defendant claimed that he had figured the EDD accurately and that many of the tests suggested by Dr. Nathanson were unnecessary, or impracticable, or both. He also presented experts who offered an alternative theory of causation: intrauterine cytomegalovirus (CMV) infection, a rare condition which occurs in 0.2 to 2.2 percent of all live births. The results of blood tests performed on Jose at age fifteen revealed previous or latent CMV infection, but did not indicate whether the infection had been contracted in utero. This is a significant omission because, while infants who suffer from CMV may be asymptomatic at birth and thereafter develop mental retardation or deafness, CMV can be transmitted in various ways and affects most individuals during their lifetimes.

II. THE SUMMARY JUDGMENT STANDARD

A court may grant summary judgment "if the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law." Fed.R.Civ.P. 56(c). We have expounded this standard and its particulars in a symphony of cases, see, e.g., McCarthy v. Northwest Airlines, Inc., 56 F.3d 313, 315 (1st Cir. 1995) (collecting cases), and we refrain from rehearsing this jurisprudential chorus here. For our purposes, it suffices briefly to describe the rule's operation.

The objective of summary judgment "is to pierce the boilerplate of the pleadings and assay the parties' proof in order to determine whether trial is actually required." Wynne v. Tufts Univ. Sch. of Med., 976 F.2d 791, 794 (1st Cir. 1992). To defeat a motion for summary judgment, the nonmoving party must demonstrate the existence of a trial-worthy issue as to some material fact. See Coyne v. Taber Partners I, 53 F.3d 454, 457 (1st Cir. 1995). A fact is "material" if it potentially could affect the suit's outcome. See Garside, 895 F.2d at 48. An issue concerning such a fact is "genuine" if a reasonable factfinder, examining the evidence and drawing all reasonable inferences helpful to the party resisting summary judgment, could resolve the dispute in that party's favor. See National Amusements, Inc. v. Town of Dedham, 43 F.3d 731, 735 (1st Cir.), cert. denied, 115 S.Ct. 2247 (1995).

Exercising de novo review, see Coyne, 53 F.3d at 457, we hold that the record in this case presents triable issues as to whether Dr. Gonzalez violated his duty of care, and, if so, whether his actions caused Jose's injuries. Consequently, the district court erred in granting the motion for brevis disposition.

III. ANALYSIS

We first survey the junction where summary judgment principles and the standards governing the admissibility of expert scientific evidence intersect. We then evaluate the lower court's ruling.

A.

The defendant asserts on appeal that the entry of judgment should be affirmed because the district court had the power to exclude the plaintiff's expert evidence pursuant to Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), and that, without such evidence, the plaintiff has no case. Cortes parries this thrust by contending that Daubert does not apply at the summary judgment stage. The truth lies somewhere in between.

The Daubert Court formulated a regime for use in ascertaining the admissibility of expert scientific evidence under Fed.R.Evid. 702. This regime contemplates that trial judges will perform a gatekeeping function, determining "whether the reasoning or methodology underlying [proffered expert] testimony is scientifically valid and . . . whether that reasoning or methodology properly can be applied to the facts in issue." Daubert, 509 U.S. at 592-93; see United States v. Sepulveda, 15 F.3d 1161, 1183 (1st Cir. 1993) (discussing this function).

The plaintiff posits that Daubert is strictly a time-of-trial phenomenon. She is wrong. The Daubert regime can play a role during the summary judgment phase of civil litigation. If proffered expert testimony fails to cross Daubert's threshold for admissibility, a district court may exclude that evidence from consideration when passing upon a motion for summary judgment. See Cavallo v. Star Enter., 100 F.3d 1150, 1159 (4th Cir. 1996), petition for cert. filed, 65 U.S.L.W. 2399 (U.S. Mar. 19, 1997) (No. 96-1493); Peitzmeier v. Hennessy Indus., Inc., 97 F.3d 293, 297-99 (8th Cir. 1996), petition for cert. filed, 65 U.S.L.W. 3539 (U.S. Jan. 29, 1997) (No. 96-1212); Claar v. Burlington N.R.R., 29 F.3d 499, 502-05 (9th Cir. 1994); Porter v. Whitehall Lab., Inc., 9 F.3d 607, 612, 616-17 (7th Cir. 1993).

The fact that Daubert can be used in connection with summary judgment motions does not mean that it should be used profligately. A trial setting normally will provide the best operating environment for the triage which Daubert demands. Voir dire is an extremely helpful device in evaluating proffered expert testimony, see Sepulveda, 15 F.3d at 1184 n. 15, and this device is not readily available in the course of summary judgment proceedings. Moreover, given the complex factual inquiry required by Daubert, courts will be hard-pressed in all but the most clear-cut cases to gauge the reliability of expert proof on a truncated record. Because the summary judgment process does not conform well to the discipline that Daubert imposes, the Daubert regime should be employed only with great care and circumspection at the summary judgment stage.

We conclude, therefore, that at the junction where Daubert intersects with summary judgment practice, Daubert is accessible, but courts must be cautious — except when defects are obvious on the face of a proffer — not to exclude debatable scientific evidence without affording the proponent of the evidence adequate opportunity to defend its admissibility. See Margaret A. Berger, Procedural Paradigms for Applying the Daubert Test, 78 Minn. L. Rev. 1345, 1379-80, 1381 (1994).

Having rejected the plaintiff's broadcast contention that Daubert can never be used at the summary judgment stage, we turn to the defendant's case-specific argument that Daubert necessitates the exclusion of the opinions advanced by the plaintiff's experts. This asseveration suffers from a very basic shortcoming: the defendant never asked the district court to exclude this evidence from consideration, and the district court made no effort to do so on its own initiative. If trial courts should be slow to employ Daubert at the summary judgment stage, appellate courts should be even more hesitant to head in that direction where there has been no development of the issue below. After all, the bifurcated inquiry into reliability and relevance which Daubert requires is best performed by trial judges who, unlike appellate judges, have a broad array of tools which can be brought to bear on the evaluation of expert testimony. Hence, we can envision few, if any, cases in which an appellate court would venture to superimpose a Daubert ruling on a cold, poorly developed record when neither the parties nor the nisi prius court has had a meaningful opportunity to mull the question.

This case falls squarely into the maw of these general principles. The defendant, notwithstanding the animadversions that he spouts on appeal, never asked in the district court to strike or otherwise defenestrate the statements of Drs. Nathanson and/or Hausknecht. The district court's rescript neither cites Daubert nor purposes to exclude the expert evidence submitted on the plaintiff's behalf. And, moreover, the record as it stands is wholly inadequate to permit a reasoned Daubert determination. For these reasons, we decline the defendant's odd invitation that we start from scratch and undertake a Daubert analysis in the context of this appeal. This means, of course, that we must consider the entire record, including the opinions of Drs. Nathanson and Hausknecht, as we ponder the merits of the district court's dispositive ruling.

B.

In this diversity suit, the substantive law of Puerto Rico controls. See Erie R.R. v. Tompkins, 304 U.S. 64, 78 (1938); Rolon-Alvarado v. Municipality of San Juan, 1 F.3d 74, 77 (1st Cir. 1993). The Puerto Rico Civil Code states that "[a] person who by an act or omission causes damage to another through fault or negligence shall be obliged to repair the damage so done." P.R. Laws Ann. tit. 31, Section(s) 5141 (1991). Under this proviso, three elements comprise a prima facie case of medical malpractice; a plaintiff must establish (1) the duty owed (i.e., the minimum standard of professional knowledge and skill required in the relevant circumstances), (2) an act or omission transgressing that duty, and (3) a sufficient causal nexus between the breach and the claimed harm. See Lama v. Borras, 16 F.3d 473, 478 (1st Cir. 1994); Rolon-Alvarado, 1 F.3d at 77. On whole-record review, we conclude that the plaintiff produced sufficient evidence to establish a genuine factual controversy as to each element.

1. Duty and Breach. In this case, the elements of duty and breach are inextricably intertwined. Thus, we address them in the ensemble.

Puerto Rico holds health care professionals to a national standard of care. See Oliveros v. Abreu, 101 P.R. Dec. 209, 226-27, translated in 1 P.R. Sup. Ct. Off'l Trans. 293, 313 (1973). Accordingly, a health care provider has "a duty to use the same degree of expertise as could reasonably be expected of a typically competent practitioner in the identical specialty under the same or similar circumstances, regardless of regional variations in professional acumen or level of care." Rolon-Alvarado, 1 F.3d at 77-78. Nevertheless, because Puerto Rico law presumes that physicians exercise reasonable care, a plaintiff bent on establishing a breach of a physician's duty of care ordinarily must adduce expert testimony to limn the minimum acceptable standard and confirm the defendant doctor's failure to meet it. See id. at 78.

Cortes' proffer is sufficient to this end. Dr. Nathanson, a specialist in the same field as Dr. Gonzalez, clearly delineated the standard of care and identified what he believed to be Dr. Gonzalez' departures from it. He stated categorically that an "average gynecologist" would not rely on a reported two-day menstrual period — unusually short even if relatively common to that particular individual — and that the failure to perform corroborating tests then available violated "the prevailing medical standard." For purposes of summary judgment, affiants and witnesses need not be precise to the point of pedantry. Thus, we treat Dr. Nathanson's references to the "average gynecologist" and to "the prevailing medical standard" as meaning the national standard of care. Cf. Lama, 16 F.3d at 479 n. 7.

The district court advanced three principal grounds in support of its conclusion that these issues — duty and breach — could be resolved against the plaintiff at the summary judgment stage, notwithstanding Dr. Nathanson's opinion evidence. All of these grounds lack persuasive force.

First, the court observed that Cortes' menstrual periods had lasted "an average of two to three days," and, accordingly, "a two-day period was not abnormal or unusually short for her." Irizarry, 928 F. Supp. at 146. But Dr. Nathanson's testimony supported the opposite conclusion; thus, whether the menses was abnormal and whether it triggered a duty to inquire further became questions of fact not properly resolved on summary judgment. See, e.g., Greenburg v. Puerto Rico Maritime Shipping Auth., 835 F.2d 932, 936 (1st Cir. 1987).

Second, the court determined that, because Dr. Gonzalez measured the uterus periodically throughout the pregnancy, yielding results consistent with the EDD on which he relied, he had no reason to suspect an earlier date of conception or to order any additional tests. See Irizarry, 928 F. Supp. at 146. While Dr. Gonzalez so testified, the court erred in treating that testimony as conclusive. When Dr. Gonzalez' and Dr. Nathanson's assertions are juxtaposed, the net result is a factual issue as to whether the defendant made the measurements, and, if so, whether this procedure satisfied the applicable standard of care.

Third, the court damned Dr. Nathanson's opinion with the faintest of praise, characterizing it as nothing more than one doctor's assertion that he would have acted differently in identical circumstances than did another, and, consequently, denying it effect in the summary judgment calculus. See id. at 147. We accept the court's premise that a mere disagreement in medical judgment, without more, does not prove duty or breach in a medical malpractice case brought under Puerto Rico law. See Rolon-Alvarado, 1 F.3d at 78. But we reject the court's conclusion; Dr. Nathanson's declarations, read in context, amount to a satisfactory statement of the standard of care and the defendant's deviation from it which, if credited by a jury, could support a finding for the plaintiff on these elements of her cause of action. And in the absence of a Daubert determination excluding the Nathanson evidence as scientifically untenable, the trial court was not at liberty on summary judgment to ignore that evidence merely because it deemed other evidence more credible.

2. Causation. Notwithstanding proof of both duty and breach, a plaintiff also must offer competent evidence of causation in a medical malpractice case. See Rolon-Alvarado, 1 F.3d at 77. The lower court found the plaintiff's submissions on this element wanting. See Irizarry, 928 F. Supp. at 147. We demur.

A medical malpractice plaintiff can — and often does — establish causation through expert testimony. See Lama, 16 F.3d at 478. Cortes took that route. Dr. Nathanson offered an opinion to a reasonable degree of medical certainty that Cortes' EDD was actually July 9, not August 9, and that this one-month discrepancy had dire consequences. If accepted, this testimony meant that Dr. Gonzalez did not perform the cesarean section until the forty-third week of a post-dated pregnancy.

Relatedly, Dr. Hausknecht diagnosed Jose as suffering from LGS, which, in the absence of any genetic or other known explanation, is generally thought to be caused by perinatal brain damage. It is undisputed that the adverse effects of post-datism include oxygen deprivation, and thus can lead to brain damage. Finding no evidence of any hereditary etiology and observing a pathology consistent with post-datism, Dr. Hausknecht opined that Jose's cerebral damage probably was caused by the post-datism which Dr. Nathanson identified. Both physicians also noted likely indications of complications at birth, and these findings buttress the plaintiff's theory of causation. Drawing reasonable inferences from this evidence, a rational jury could find that Dr. Gonzalez' negligent reliance upon a reported two-day menstrual period and his eschewal of further (available) tests caused a post-dated pregnancy, the effects of which included perinatal brain damage which manifested itself in the form of LGS.

Of course, the defendant's experts debunked the plaintiff's proof and offered an alternative causal theory — the presence of a CMV infection — which the district court found "more compelling." Irizarry, 928 F. Supp. at 147. But such comparisons are invidious at the summary judgment stage. Even at trial, a plaintiff in a medical malpractice suit need not prove a causal connection with mathematical accuracy nor eliminate all other possible causes of damage. See Cruz Rodriguez v. Corporacion de Servicios del Centro Medico, 113 P.R. Dec. 719, 744, translated in 13 P.R. Sup. Ct. Off'l Trans. 931, 960-61 (1983). Legal rules of this sort acquire added significance on summary judgment because Rule 56 "contemplates an abecedarian, almost one dimensional, exercise geared to determining whether the nonmovant's most favorable evidence and the most flattering inferences which can reasonably be drawn therefrom are sufficient to create any authentic question of material fact." Greenburg, 835 F.2d at 936.

In this case, the defendant's evidence on the issue of causation, as compelling as it might have seemed, did not warrant the entry of summary judgment. The plaintiff articulated an alternative theory of causation and backed it up with expert testimony as to the causal nexuses between LGS and perinatal damage, and between perinatal damage and post-datism. At the same time, she cast doubt on the defendant's theory of causation, establishing the low incidence of intrauterine CMV infection and suggesting an alternate origin of any CMV detected in Jose's system (related to a history of sexual molestation at school, thereby opening up the possibility that any CMV infection was sexually transmitted). This evidence sufficed to create a trial-worthy issue vis-a-vis the element of causation. See Coyne, 53 F.3d at 460 (explaining that "when the facts support plausible but conflicting inferences on a pivotal issue in the case, the judge may not choose between those inferences at the summary judgment stage"); see also United States v. Kayne, 90 F.3d 7, 12 (1st Cir. 1996) (stating that disagreements among experts are "properly the subject of searching cross-examination" at trial), cert. denied, 117 S.Ct. 681 (1997).

IV. CONCLUSION

We need go no further. Scrutinizing the entire record in the light most congenial to the plaintiff, rational jurors could find all the elements of medical malpractice. Though the plaintiff's evidence may appear thin to some, it establishes factual disagreements as to which reasonable minds may differ. No more is exigible. See Greenburg, 835 F.2d at 936 (explaining that the ground rules associated with summary judgment practice "admit of no room for credibility determinations, no room for the measured weighing of conflicting evidence such as the trial process entails, no room for the judge to superimpose his own ideas of probability and likelihood (no matter how reasonable those ideas may be) upon the carapace of the cold record"). Right or wrong, the plaintiff is entitled to present her case to a jury.

The order granting summary judgment is vacated and the case is remanded for trial. Costs to appellant.