Dr. David Herzog was the obstetrician who delivered Amanda Costantino. During the delivery, Amanda's shoulder got trapped behind her mother's pubic bone, a condition known as "shoulder dystocia." While attempting to remedy the condition, Dr. Herzog performed: (1) the McRoberts maneuver: pulling Mrs. Costantino's legs toward her head and applying pressure to the area above her pubic bone; (2) the Woods corkscrew: reaching into the womb and rotating baby Amanda to release her trapped shoulder; and (3) the Posterior Arm Sweep: delivering Amanda's free posterior arm to create more space. Ultimately, Dr. Herzog delivered Amanda, but she was born with "Erb's Palsy," an impairment to the nerves running to the arm.

The Costantinos filed a diversity action against Dr. Herzog in the United States District Court for the Eastern District of New York (Gleeson, J.) alleging that by pulling and rotating Amanda's head during the delivery, he had caused her Erb's Palsy. They claimed that Dr. Herzog had deviated from accepted standards of obstetrical practice, and had therefore committed malpractice under governing New York law. The defense denied any malpractice, asserting that Amanda's Erb's Palsy was caused by the normal forces of labor.

The case was tried to a jury. Plaintiffs' first witness was the defendant, Dr. Herzog. Counsel questioned him on an excerpt from a medical treatise edited by Steven G. Gabbe, entitled Obstetrics, that stated: "Once a vaginal delivery has begun, the obstetrician must resist the temptation to rotate the head to a transverse axis." Dr. Herzog acknowledged attempting to rotate Amanda's head, but disagreed with the statement read from the Gabbe treatise.

Plaintiffs' medical expert was Dr. Bernard Nathanson. Among his qualifications, Dr. Nathanson testified that he was a fellow of the American College of Obstetricians and Gynecologists ("ACOG"). ACOG, according to Dr. Nathanson, "is an organization of thirty thousand obstetricians and gynecologists," that "sets up courses for doctors who are in practice so that they will continue to be current with ongoing research." Dr. Nathanson added that ACOG "publishes a great deal of material which serve[s] to contribute to setting a standard of care for obstetricians and gynecologists."

Relying in part on various journals published by ACOG, as well as on the Gabbe treatise, Dr. Nathanson proceeded to testify that it was a departure from the standard of medical care to engage in "any manipulation of the head" during a shoulder dystocia delivery because it does nothing to relieve the trapped shoulder and greatly increases the risk of causing Erb's Palsy.

The defense sought to rebut this theory in several ways. Primarily, the defense relied on another learned treatise - Williams Obstetrics - which stated that "downward traction . . . to the fetal head" was among "the most popular techniques" used to remedy shoulder dystocia. Dr. Nathanson conceded that the Williams treatise was authoritative and that applying traction to the fetal head was indeed the "most popular" treatment technique for shoulder dystocia. He continued, nevertheless, to insist that use of that technique constituted malpractice.

The defense also sought to justify Dr. Herzog's management of Amanda's delivery by introducing a 15-minute videotape from ACOG's audiovisual library, entitled "Shoulder Dystocia." The tape was, according to the defense, "put out by [ACOG] to educate physicians" and portrayed the various techniques recommended to remedy shoulder dystocia.

Both the parties and Judge Gleeson recognized that the ACOG video was hearsay under Federal Rule of Evidence 801. The defense nevertheless sought to introduce it pursuant to the "learned treatise" exception to the hearsay rule set forth in Rule 803(18). Plaintiffs objected, arguing that Rule 803(18) enumerates only "published treatises, periodicals, or pamphlets" as learned treatises, and therefore could not encompass videotapes. Plaintiffs also argued that no foundation had been laid for the video.

After an in camera review of the videotape, Judge Gleeson ruled it admissible. With respect to whether a video could qualify as a learned treatise under Rule 803(18), Judge Gleeson reasoned: "I think . . . focusing on the distinction between . . . something in the form of a periodical or a book, as opposed [to] a videotape is just overly artificial."

As to the foundation, Judge Gleeson found, based on his in camera review, that the ACOG video "was a dissemination to the doctors in the relevant medical community of how they should go about dealing with this problem [of shoulder dystocia]." He also found that it had been "well established" through trial testimony that ACOG was "the source of authoritative information regarding the practices to be used by obstetricians in these circumstances." Included in the trial testimony regarding ACOG and the videotape, were Dr. Nathanson's concessions that he had: (1) viewed the videotape at a staff conference some years ago; and (2) testified in a prior action that he generally accepted "the standards promulgated by ACOG" within the field of obstetrics as "authoritative."

The ACOG video was played twice during trial. It was played in its entirety during the cross-examination of plaintiffs' expert, Dr. Nathanson, and portions were replayed during the direct examination of defendants' own expert Dr. James Howard.

In graphic detail the video portrays actual deliveries, complicated by shoulder dystocia, and demonstrates the recommended obstetrical responses to it. These portrayals are accompanied by a narrative given by a Dr. Young, who the video reveals is from Dartmouth College's Hitchcock Medical Center. The various procedures recommended, Dr. Young explains, were chosen "following a careful review of the available literature." Several times during its 15 minute duration, the video cautions that "unfortunately, babies cannot always be delivered without injury even when the management is optimal," and that "sometimes . . . injuries cannot be avoided."

As the first step in treatment, the video recommends coordination of the mother's efforts to expel the baby with what Dr. Young characterizes as application of "a limited" or "appropriate" amount of traction to the baby's head by the obstetrician. Dr. Young goes on to warn, however, that "some of the problems that occur with shoulder dystocia may result from more forceful and prolonged efforts at pushing and pulling." The video instructs that where an initial application of traction to the baby's head has been unsuccessful, "it is important to stop and take a different approach to relieve the problem." The video then portrays application of the McRoberts, Woods and Posterior Arm Sweep maneuvers, which Dr. Herzog had testified to performing during Amanda's troubled delivery.

The video concludes with a portrayal of what Dr. Young acknowledges are two rare maneuvers, apparently to be tried only when all else fails. First, the video depicts placing the mother on all fours to facilitate delivery. Second, the video portrays pushing the baby back into the uterus to permit delivery by a Caesarian section.

In its closing credits, the video scrolls across the screen a printed disclaimer, stating:

This video does not define a standard of care, nor is it intended to dictate an exclusive course of management. It presents recognized techniques of clinical practice for consideration by health care providers for incorporation into their practices.

The closing credits also reveal that the video was awarded the Scientific Program Award at the 1995 Annual Clinical Meeting of ACOG.

In addition to the ACOG video, the defense introduced as learned treatises, over objection, two articles published in the American Journal of Obstetrics and Gynecology. The first was entitled, Shoulder Dystocia, an Analysis of Risks and Obstetrical Maneuvers, and was written by Dr. James Nocon of the University of Indiana Medical School. Before this article was read to Dr. Herzog on cross-examination, he testified on voir dire that he: (1) did not know of the article's author, Dr. Nocon; and (2) had not read the specific article until "two minutes" before beginning his testimony about it.

Later, during cross-examination of plaintiffs' expert, Dr. Nathanson, defense counsel sought to introduce a different article written by Robert Gherman, an obstetrics professor at the University of Southern California and also published in the American Journal of Obstetrics and Gynecology. Though Dr. Nathanson refused to concede the authoritativeness of that journal (based on his apparent belief that "no journal is authoritative"), Judge Gleeson allowed the defense to read the following statement from Dr. Gherman's article:

[Erb's] palsy is commonly attributed to excessive lateral traction applied to the fetal neck by the treating physician during attempts to free the shoulder . . . . Some authors have argued that with worsening severity of the shoulder dystocia, and increased efforts by the operators, the impact of the traction force becomes more pronounced . . . . We have presented six cases in which it is virtually certain that this mechanism played no role . . . . Our data also imply that some cases of [Erb's palsy] may be related to events antedating the actual delivery, especially in light of the fact that one patient underwent a Caesarian section while in early labor.

Defense counsel also read a different portion of the same article to defendant's expert, Dr. Howard, during direct examination. Dr. Howard did not testify that he relied on the article in any way.

After trial, the jury found for Dr. Herzog on the issue of liability. The Costantinos now appeal, arguing that Judge Gleeson erred in admitting: (1) the ACOG video; and (2) the two articles from the American Journal of Obstetrics and Gynecology.

The Costantinos argue that because videotapes are not mentioned in Rule 803(18), they can never be learned treatises. Alternatively, they maintain that even if videos can be learned treatises, reversal is still required because Judge Gleeson: (1) erred in admitting the ACOG video without proper foundation; and (2) should have excluded the video as unduly confusing and prejudicial under Rule 403. None of these contentions warrant reversal.1

The primary question presented is whether videotapes can be admitted as learned treatises pursuant to Rule 803(18). We are the first federal Court of Appeals to address this question, though various state courts have considered it under their cognate learned treatise exceptions, and have forged no consensus. Compare Loven v. State, 831 S.W.2d 387, 396-97 (Tex. Ct. App. 1992) (admissible), and Schneider v. Cessna Aircraft Co., 722 P.2d 321, 327 (Ariz. Ct. App. 1986) (same), with Simmons v. Yurchak, 551 N.E.2d 539, 542-43 (Mass. App. Ct.) (inadmissible), review denied, 554 N.E.2d 851 (Mass. 1990), and Morrison v. Stallworth, 326 S.E.2d 387, 392-93 (N.C. Ct. App. 1985) (same).

We review Judge Gleeson's legal conclusion that videos can constitute learned treatises de novo. See Borawick v. Shay, 68 F.3d 597, 601 (2d Cir. 1995).

The following are not excluded by the hearsay rule, even though the declarant is available as a witness:

(18) Learned treatises. To the extent called to the attention of an expert witness upon crossexamination or relied upon by the expert witness in direct examination, statements contained in published treatises, periodicals, or pamphlets on a subject of history, medicine, or other science or art, established as a reliable authority by the testimony or admission of the witness or by other expert testimony or by judicial notice. If admitted, the statements may be read into evidence but may not be received as exhibits.

Fed. R. Evid. 803(18).

The rationale for this exception is self-evident: so long as the authority of a treatise has been sufficiently established, the factfinder should have the benefit of expert learning on a subject, even though it is hearsay. See Christopher B. Mueller & Laird C. Kirkpatrick, Evidence 8.52, at 997 (1995).

Emphasizing plain language, the Costantinos argue that videos cannot fall within the scope of Rule 803(18) because unlike "published treatises, periodicals, or pamphlets," they are not specifically listed in the Rule. They rely on Simmons v. Yurchak, 551 N.E.2d 539 (Mass. App. Ct.), review denied, 55 N.E.2d 851 (Mass. 1990), which accepted this contention, and affirmed a trial court's refusal to recognize videotapes as learned treatises under the Massachusetts version of Rule 803(18). According to the Simmons court: "adding videotapes to the list of materials in [the Massachusetts learned treatise exception] would constitute judicial legislation." 551 N.E.2d at 543.

Uttering the dark incantation of "judicial legislation" is to substitute a slogan for an analysis. Indeed, we are exhorted in Rule 102 to interpret the Rules of Evidence to promote the "growth and development of the law . . . to the end that the truth may be ascertained." Fed. R. Evid. 102. In this endeavor a certain measure of legislative judgment is required. As Justice Holmes, in his Boston Brahmin prose conceded: "I recognize without hesitation that judges do and must legislate, but they can do so only interstitially; they are confined from molar to molecular motions." Southern Pac. Co. v. Jensen, 244 U.S. 205, 221 (1917) (Holmes, J., dissenting). Or as a graduate of N.Y. City College, Justice Frankfurter, put it more prosaically: "legislatures make law wholesale, judges retail." Joseph P. Lash, From the Diaries of Felix Frankfurter 67 (1975) (quoting unidentified letter from Felix Frankfurter to Hugo Black). Because judges "cannot escape the responsibility of filling in gaps which the finitude of even the most imaginative legislation renders inevitable," the problem is "not whether the judges make the law, but when and how much." Id.

In this case, we are compelled to "make law." For we agree with Judge Gleeson that it is just "overly artificial" to say that information that is sufficiently trustworthy to overcome the hearsay bar when presented in a printed learned treatise loses the badge of trustworthiness when presented in a videotape. We see no reason to deprive a jury of authoritative learning simply because it is presented in a visual, rather than printed, format. In this age of visual communication a videotape may often be the most helpful way to illuminate the truth in the spirit of Rule 102. See, e.g., Schneider v. Cessna Aircraft Co., 722 P.2d 321 (Ariz. Ct. App. 1986) (training film admissible under Arizona Rule 803(18) in action against an aircraft manufacturer arising from crash, where the film depicted a crash similar to the manufacturer's version of what happened).

In sum, we agree with the Texas Court of Appeals that "[v]ideotapes are nothing more than a contemporary variant of a published treatise, periodical or pamphlet." Loven v. State, 831 S.W.2d 387, 397 (Tex. Ct. App. 1992). Accordingly, we hold that videotapes may be considered learned treatises under Rule 803(18).

The Costantinos argue that even assuming that the ACOG video may be a learned treatise, Judge Gleeson erred in admitting it without a proper foundation. We disagree.

Rule 803(18) explicitly requires trial judges to act as gatekeepers, ensuring that any treatise admitted is "authoritative." Schneider v. Revici, 817 F.2d 987, 991 (2d Cir. 1987). Thus, trial judges must first determine that the proffered treatise is "trustworth[y] . . . as viewed by professionals in [the relevant] field." Id.; see Fed. R. Evid. 803(18) Advisory Committee Note. In making this evaluation, trial judges need not be draconian. "Since the object of [Rule 803(18)] is to make valuable information available to the trier of fact, trial judges should not insist on a quantum of proof . . . that the proponent cannot meet." See Weinstein's Federal Evidence 803.23[4] (2d ed. 1997) (citation omitted).

Judge Gleeson's finding that the authoritativeness of the ACOG video had been sufficiently established is reviewed for abuse of discretion and will not be disturbed absent "manifest error." See Starter Corp. v. Converse, Inc., 170 F.3d 286, 292 (2d Cir. 1999) (quoting Proteus Books Ltd. v. Cherry Lane Music Co., 873 F.2d 502, 514 (2d Cir. 1989)).

The Costantinos invite us to apply the reasoning of Meschino v. North American Drager, Inc., 841 F.2d 429 (1st Cir. 1988). There, the First Circuit ruled that it was not enough that the trade magazine in which an article appeared was reputable; the author of the particular article must also to be shown to be an authority before the article could be used as a learned treatise pursuant to Rule 803(18). See id. at 433-34; see also Twin City Fire Ins. Co. v. Country Mut. Ins. Co., 23 F.3d 1175, 1184 (7th Cir. 1994) (same). The Court reasoned that "[i]n these days of quantified research, and pressure to publish, an article does not reach the dignity of a 'reliable authority' merely because some editor, even a most reputable one, sees fit to circulate it." Meschino, 841 F.2d at 434.

The Costantinos argue that Meschino requires reversal because Judge Gleeson's sole basis for accepting the video as authoritative was its sponsorship by ACOG. They claim that no testimony was offered that the video itself had received acceptance in the medical profession, or that its narrator, Dr. Young, was considered an authority in the field. And these omissions were particularly devastating, the Costantinos maintain, because as the Simmons court noted, it is unclear that the "careful, professional criticism" which ensures the reliability of printed treatises, also attends the production and dissemination of videotapes. 551 N.E.2d at 543.

These arguments are unavailing. Even assuming that the potential authoritativeness of videotapes is somehow more suspect than that of materials like "pamphlet[s]" explicitly listed by the Rule, we cannot fault the district court's performance as a gatekeeper in this case.

We, of course, agree with the Meschino court that the contents of a periodical cannot be automatically qualified "wholesale" under Rule 803(18) merely by showing that the periodical itself is highly regarded. 841 F.2d at 434. We do not, however, read Meschino to say that the reputation of the periodical containing the proffered article is irrelevant to the authoritativeness inquiry. Publication practices vary widely, and an article's publication by an esteemed periodical which subjects its contents to close scrutiny and peer review, obviously reflects well on the authority of the article itself. Indeed, because the authoritativeness inquiry is governed by a "liberal" standard, good sense would seem to compel recognizing some periodicals - provided there is a basis for doing so - as sufficiently esteemed to justify a presumption in favor of admitting the articles accepted for publication therein. See generally Weinstein's Federal Evidence 803.23[4] (2d ed. 1997); Allen v. Safeco Insurance Co., 782 F.2d 1517, 151920 (11th Cir.), vacated on other grounds, 793 F.2d 1195 (1986); McCarty v. Sisters of Mercy Health Corp., 440 N.W.2d 417, 419-21 (Mich. Ct. App. 1989).

Turning to the instant case, there is no small irony in plaintiffs' complaint that ACOG sponsorship was insufficient to establish the video as an authority. After all, while plaintiffs' own expert Dr. Nathanson refused to concede the authoritativeness of the video itself, he nevertheless: (1) touted himself as a member of ACOG; (2) praised ACOG as an organization that "publishes a great deal of material which serve[s] to contribute to setting a standard of care for obstetricians and gynecologists;" and (3) testified in a prior action that he generally accepted "the standards promulgated by ACOG" within the field of obstetrics as "authoritative." Moreover, Dr. Nathanson relied in part on ACOG publications in reaching his opinion that Dr. Herzog had committed malpractice.

In any event, other factors - quite apart from ACOG's status as a reputable organization - established the authoritativeness of the video. In particular, Dr. Nathanson recalled seeing a version of the ACOG video at a staff conference, "inferentially conced[ing]" that it was exactly what the defense said it was: a training resource for the continuing medical education of obstetricians and gynecologists. Dawson v. Chrysler Corp., 630 F.2d 950, 960-61 (3d Cir. 1980). And the video's use as a training resource - "written primarily and impartially for professionals, subject to scrutiny and exposure for inaccuracy, with the reputation [of its producers and sponsors] at stake" - is clearly an important index of its authoritativeness under Rule 803(18). Fed. R. Evid. 803(18) Advisory Committee Note.

Moreover, Judge Gleeson himself took the additional precaution of reviewing the ACOG video in camera. Through that review, Judge Gleeson knew that the tape's narrator, Dr. Young, was a physician at Dartmouth College's Hitchcock Medical Center, and that the video itself had won an ACOG award, credentials which compared favorably with those of any expert who testified at trial. And after the same review, Judge Gleeson found that the video was what the defense represented it to be: a training resource - with recommendations culled from the "available literature" - used to show doctors "how they should go about dealing with this problem [of shoulder dystocia]." Having viewed the videotape ourselves, and having observed its clinical format, as well as its calm and instructional tone, we cannot say this finding amounts to "manifest error." Starter Corp., 170 F.3d at 297 (internal quotation marks omitted); see Loven, 831 S.W.2d at 397 (affirming trial court's finding of authoritativeness after viewing challenged video).

In sum, we conclude that Judge Gleeson's determination that the ACOG video was sufficiently authoritative to deserve admission rested on an appropriate foundation. This was not a case in which there was "no basis" for finding the proffered treatise trustworthy. Schneider, 817 F.2d at 991. And while some of the indicia of the video's reliability came to light through Judge Gleeson's independent in camera review, rather than through testimony, the authoritativeness inquiry is a freewheeling one and may be conducted by "any means." Fed. R. Evid. 803(18) Advisory Committee Note; see also Weinstein's Federal Evidence 803.23[4] (2d ed. 1997) ("trial judge[s] should be liberal in allowing other proof of . . . authoritativeness, so long as it indicates that the [treatise] is recognized by the medical profession") (citing Ward v. United States, 838 F.2d 182, 187 (6th Cir. 1988)).

Judge Gleeson did not abuse his discretion in finding that the ACOG video was sufficiently authoritative to be presented to the jury.

Even though the ACOG video qualified as a learned treatise under Rule 803(18), Judge Gleeson retained the discretion to exclude it if its "probative value [was] substantially outweighed by the danger of unfair prejudice [or] confusion" under Rule 403. See Li v. Canarozzi, 142 F.3d 83, 88 (2d Cir. 1998); Schneider, 817 F.2d at 991. The Costantinos argue that the video should have been excluded under Rule 403. We find no basis for reversal.

Because the trial judge is in the best position to evaluate the evidence and its effect on the jury, his Rule 403 rulings are entitled to considerable deference and will not be overturned absent a clear abuse of discretion. See Li, 142 F.3d at 88; United States v. Smith, 727 F.2d 214, 220 (2d Cir. 1984). Moreover, the improper admission of evidence is grounds for reversal only where it affects "a substantial right" of one of the parties. Fed. R. Evid. 103(a); see 28 U.S.C. 2111; Fed. R. Civ. P. 61. "We will not conclude that a substantial right was affected unless it is likely that in some material respect the factfinder's judgment was 'swayed by the error.'" Perry v. Ethan Allen, Inc., 115 F.3d 143, 150 (2d Cir. 1997) (quoting Phoenix Assocs. III v. Stone, 60 F.3d 95, 10405 (2d Cir. 1995)).

There can be no doubt that much of the ACOG videotape represented highly probative evidence. To prove that Dr. Herzog committed malpractice under governing New York law, plaintiffs had to show that he deviated from the accepted obstetrical practices of the medical profession. See Pellegrino v. Cunanan, 643 N.Y.S.2d 844, 844 (4th Dep't. 1996); Walsh v. Staten Island Obstetrics & Gynecology Assocs. P.C., 598 N.Y.S.2d 17, 18 (2d Dep't. 1993). As Judge Gleeson explained in ruling that the video was admissible, one "dimension" of plaintiffs' malpractice claim was that "the standard of care . . . prohibits a doctor from engaging in any traction" to remedy shoulder dystocia. Measured against this claim, those portions of the video portraying application of "a limited" or "appropriate" amount of traction to the baby's head were directly probative of whether such a technique is an accepted obstetrical practice. And those portions of the video portraying application of the Woods, McRoberts and Posterior Arm Sweep maneuvers were probative of whether Dr. Herzog properly used those maneuvers in delivering Amanda.

We reject the plaintiffs' claim that the ACOG video should have been excluded because of the danger of undue confusion. The Costantinos maintain that the ACOG video must have caused the jurors to confuse what they saw on the tape with what was being orally reconstructed for them throughout the trial - i.e., the birth of Amanda. They argue that because the video shows appropriate amounts of traction being applied to babies' heads, the jurors must have unfairly concluded that Dr. Herzog's management of Amanda's delivery was appropriate.

This argument is pure speculation. It was obvious that Amanda's birth was not what was depicted in the video. Even if it were not obvious to the jury, any ephemeral danger of confusion was more than outweighed by the highly probative value of the video.

Plaintiffs' second Rule 403 contention is similarly unavailing. The Costantinos point out that in addition to portraying the Woods, McRoberts and Posterior Arm Sweep maneuvers (as well as the application of traction to babies' heads), the ACOG video portrays the two rarely performed procedures of: (1) placing the mother on all fours to facilitate delivery; and (2) pushing the baby back into the uterus to facilitate a Caesarian section. The Costantinos assert that these portrayals were not probative of anything at trial, and were highly prejudicial because "the only inference the jury could have drawn from those parts of the tape . . . is that shoulder dystocia can be a complicated problem, requiring seemingly drastic measures to deliver children."

It is true that Dr. Herzog never testified that he even contemplated performing these maneuvers during the birth of Amanda. However, to the extent that these portrayals prejudiced the Costantinos by emphasizing the inherent dangers of shoulder dystocia, such prejudice did not warrant exclusion.

Because virtually all evidence is prejudicial to one party or another, to justify exclusion under Rule 403 the prejudice must be unfair. See Weinstein's Federal Evidence 403.04[1][a] (2d ed. 1997) (citing cases). The unfairness contemplated involves some adverse effect beyond tending to prove a fact or issue that justifies admission. See Perry, 115 F.3d at 151.

We fail to see how plaintiffs were unfairly prejudiced by the portrayal of the challenged procedures here. The testimony at trial established that shoulder dystocia is indeed a "complicated problem, requiring seemingly drastic" responses by the treating obstetrician. To cite but one example, plaintiffs' own expert Dr. Nathanson testified that shoulder dystocia can be a life-threatening "emergency" to both mother and child, and that it is necessary to perform an episiotomy (a vaginal incision), "as deepely and widely as possible," frequently "extend[ing the incision] right into the rectum, so that the entire vagina is laid open."

In light of this and similar testimony, any prejudice arising from the video's portrayal of the challenged procedures was not unfair within the meaning of Rule 403. To the contrary, that portrayal was probative of whether Amanda Costantino's shoulder dystocia presented an "emergency" and whether Dr. Herzog responded appropriately in attempting to remedy it.

II. The American Journal of Obstetrics and Gynecology

The Costantinos final argument is that the admission of the two articles from the American Journal of Obstetrics and Gynecology also requires reversal. Once again, plaintiffs argue lack of proper foundation. First, they complain that because Dr. Herzog testified that he had not seen Dr. James Nocon's article until shortly before testifying about it, he could not have "relied" upon that article as required by Rule 803(18). Second, they protest that there was no testimony establishing the authoritativeness of either Dr. Nocon's article, or the second article written by obstetrics professor Robert Gherman of the University of Southern California.

We will not tarry long on these arguments. It is not essential that a testifying expert rely on a learned treatise on direct examination. If he did not, a sufficient foundation is laid if the treatise is "called to the attention of [the] expert upon-cross examination." Fed. R. Evid. 803(18). The essential notion is that the jury enjoys "explanation, context and perspective" on the treatise's contents. Mueller & Kirkpatrick, Evidence 8.52, 998. That is exactly what happened on defense counsel's cross-examination of Dr. Herzog.

As to authoritativeness, Dr. Herzog testified that the American Journal of Obstetrics and Gynecology is "probably the most reputable journal in our field and if it's accepted for publication, then we know it's quality information being presented." Plaintiffs' own expert, Dr. Nathanson, backed Dr. Herzog on this point, testifying that articles are published in the American Journal of Obstetrics and Gynecology only after "other obstetricians and gynecologists . . . review them for accuracy and believability." The district court was properly satisfied with this foundation.

Even if this testimony were insufficient to establish a foundation for the two challenged articles, any error was harmless. Plaintiffs do not cite any specific prejudice arising from presentation of Dr. Nocon's article. And the essential point of Dr. Gherman's article was that Erb's Palsy can be caused even without excessive traction being applied to the fetal head. There was ample other evidence in support of this very proposition, including the ACOG video. See Paradise v. CCI Warden, 136 F.3d 331, 338 (2d Cir. 1998) (admission of potentially objectionable testimony harmless where other evidence proved issue). The admission of the two articles from the American Journal of Obstetrics and Gynecology does not warrant reversal.

We are troubled by one aspect of this case. As noted above, the ACOG videotape's closing credits include the statement: "This video does not define a standard of care." Other learned treatises read at trial contain similar legalistic language, or actually allude to litigation, such as Dr. Gherman's article in the American Journal of Obstetrics and Gynecology, which refers to unfounded assertions made by "plaintiffs' expert witnesses."

Such references tend to suggest that shoulder dystocia births are the source of much unjustified litigation, and that the instant case was merely an example of such litigiousness rather than a legitimate attempt to achieve redress for real injury. Although plaintiffs have not directly challenged these references, we feel compelled to point out that they should have been redacted under Rule 403. See generally Outley v. City of New York, 837 F.2d 587, 592 (2d Cir. 1988) (because "[t]he charge of litigiousness is a serious one, likely to result in undue prejudice" it may be cause for exclusion under Rule 403); see, e.g., Staley v. Bridgestone/ Firestone, Inc., 106 F.3d 1504, 1513 (10th Cir. 1997) (redacting prejudicial material under Rules 403 and 404(b)).

The district court's failure to perform such a redaction, however, does not warrant reversal. Throughout trial, Judge Gleeson carefully policed both counsel and the witnesses to ensure that the evidence was - in the main - properly focused on the discrete issue of whether Dr. Herzog committed malpractice in delivering Amanda Costantino. On this record, it is clear that the isolated admission of a few objectionable legal references could not have swayed the jury's verdict. See Perry, 115 F.3d at 150. We raise this issue only to alert district courts to be wary of admitting such material in the future.

For the reasons set forth above, the judgment of the district court is AFFIRMED.