ROBERT M. PARKER, Circuit Judge:
Daniel Battle, Jr. (“Daniel”), a minor, his mother Zeta Battle (“Mrs. Battle”), and his father Daniel Battle, Sr. (“Mr. Battle”) brought suit alleging that negligent medical treatment by David L. Reeves, M.D., Dennis W. Aust, M.D. and Emergency Care Specialists of Mississippi, Ltd. resulted in injuries to Daniel Battle, Jr. Plaintiffs further alleged that Memorial Hospital at Gulfport (“Memorial Hospital”) was liable to Daniel under Mississippi tort law and that it violated the Emergency Medical Treatment and Active Labor Act, 42 U.S.C. § 1395dd (1994) (“EMTALA”). Defendants prevailed on all claims and Plaintiffs appeal. We affirm in part, vacate in part and remand for further proceedings.
I. FACTS
In December 1994, fifteen-month-old Daniel suffered from viral encephalitis, an inflammation of the brain, which resulted in extensive neurological injury. Daniel, now six years old, was characterized at trial as “about as damaged as a human being can be and still be alive.”
Daniel, born on September 8, 1993, was healthy and normal until December 22, 1994, when he developed a fever and sores on his tongue. Mrs. Battle took Daniel to his pediatrician, Dr. Reeves, who diagnosed an ear infection and tonsillitis and prescribed a course of antibiotics. Daniel’s condition did not improve. Shortly before midnight on December 24, 1994, Mrs. Battle called and left a message with Dr. Reeves’s answering service because Daniel’s jaws were snapping shut. Mrs. Battle then called 911 because Daniel’s face began to twitch and his eyes rolled back. When Dr. Reeves called back, the paramedics had arrived and they informed him that Daniel had seizures, fever and that one hand and his face were twitching.
Daniel was taken to Memorial Hospital and seen in the emergency room by Dr. Graves and Dr. Sheffield. Dr. Sheffield performed a lumbar puncture, which Dr. Graves interpreted as normal. After x-rays and some blood work, Daniel was diagnosed with febrile seizures, pneumonia and an ear infection. He was discharged and went home with a new set of antibiotics.
In the afternoon of December 25, Mrs. Battle called Dr. Reeves again and told him that Daniel was continuing to have seizures. Dr. Reeves instructed her to take Daniel back to the Memorial Hospital emergency room where he was seen by Dr. Aust. On this second trip, Mrs. Battle put “self-pay” on the emergency room paper work. Dr. Aust diagnosed Daniel with “seizure disorder” and pneumonia and administered Dilantin for the seizures. As Mrs. Battle took Daniel home with a prescription for additional Dilantin, Dr. Aust instructed her to “not bring that child right back in here because Dilantin takes time to work.”
When the Dilantin wore off, Daniel’s seizures returned and continued on and off throughout the day on December 26. That afternoon, Mrs. Battle called Dr. Reeves again. Dr. Reeves instructed her to take Daniel to Memorial Hospital and have him admitted, which she did. Drs. Aust and Reeves ordered a CT scan, without contrast, which was read as negative. They also ordered an EEG, which was not read until seven days later. When read, it was grossly abnormal.
At 9:00 p.m. on December 26, Dr. Reeves saw Daniel for the first time since December 22. Daniel’s condition continued to deteriorate. At 5:00 p.m. on December 27, Dr. Reeves’s partner, Dr. Akin, saw Daniel. She diagnosed viral encephalitis, possibly the rare and dangerous herpes simplex encephalitis (“HSE”), and initiated treatment with Acyclovir, a drug that can halt the progression of HSE in some patients. She then arranged for a helicopter to transport Daniel to Tulane Medical Center where he could receive care from an infectious disease specialist. When Daniel arrived at Tulane around midnight of December 27, health care personnel immediately did a lumbar puncture which was grossly abnormal. They also performed a CT scan, with and without contrast, and an MRI. All the tests revealed abnormal results consistent with HSE.
A positive diagnosis of HSE requires a brain biopsy or a DNA test called PCR (“polymerase chain reaction”). Daniel’s spinal fluid, obtained from the lumbar puncture on December 27, 1994, was tested at Tulane as well as being sent to the Whitley laboratory at the University of Alabama, which specializes in HSE research. Tulane’s test was negative for HSE. On January 19, 1995, Dr. Fred Lakeman in the Whitley lab obtained a positive result on the PCR test, indicating that Daniel had HSE.
Despite the fact that the suspicion of HSE was unconfirmed until January 19, Daniel remained on Acyclovir throughout his treatment at Tulane. Daniel was discharged from Tulane on February 1, 1995, in a near vegetative state. He will require 24-hour-a-day care for the rest of his life.
II. PROCEDURAL HISTORY
Plaintiffs filed medical malpractice claims against Dr. Reeves, Dr. Aust, Dr. „Aust’s practice group, Emergency Care Specialists of Mississippi, Ltd. and Memorial Hospital on October 1, 1996, in Mississippi Circuit Court. After Plaintiffs amended their complaint to allege an EM-TALA claim against Memorial Hospital, Defendants removed the case to federal court on May 1, 1997. After extensive discovery, the case was set for trial on September 14,1998.
Prior to trial, the district court granted summary judgment for Memorial Hospital on Plaintiffs’ state law claims, finding that the claims had not been filed within the controlling Mississippi one-year statute of limitations.
Approximately three weeks prior to trial, Plaintiffs informed Defendants that expert witness Lowell Young, M.D., would not be available for trial and noticed the videotape deposition of Dr. Young for September 3, 1998, in San Francisco, California. On September 2, 1998, Plaintiffs moved for a continuance based on the unavailability for trial of another expert, Dr. Richard Whitley. On September 3, 1998, Plaintiffs noticed the deposition of Whitley for September 9, 1998. The district court granted Plaintiffs’ motion for continuance and reset the trial for January 25, 1999.
Due to a death in District Judge Bram-lette’s family several days before trial, the parties consented to trial before Chief Magistrate Judge John Roper. Before trial, Magistrate Judge Roper denied Plaintiffs’ motion
in limine
to prohibit evidence of Mr. Battle’s incarceration during Dan'iel’s illness and prohibited Plaintiffs from introducing into evidence the deposition of Plaintiffs’ expert Dr. Fred Lakeman. Dr. Young’s video deposition was admitted, but Plaintiffs were not allowed to call him live. Plaintiffs challenge each of these rulings on appeal.
Trial commenced on January 25, 1999. At the close of Plaintiffs’ case, the magistrate judge granted judgment for Memorial Hospital on the EMTALA claims and dismissed it from the case, finding that there was no evidence of disparate treatment or failure to stabilize
Daniel’s condition.
A unanimous jury verdict in favor of Defendants Reeves, Aust and Emergency Care Specialists of Mississippi was entered on February 8, 1999.
III. ANALYSIS
A. EVIDENTIARY RULINGS
Plaintiffs assign as error three eviden-tiary rulings and argue that the cumulative impact of these errors resulted in prejudice and requires reversal of the judgment for Defendants.
1. Mr. Battle’s Incarceration
[2] Mr. Battle was incarcerated from June 1993 to June 1996. Plaintiffs filed a motion
in limine
seeking to exclude evidence of the past criminal acts of Mr. Battle under Federal Rules of Evidence 401, 402, 403 and 609 as irrelevant, unduly prejudicial and outweighing any probative value. Defendants responded that Mr. Battle, who was seeking his own individual damages in the case, might be called as a witness on the issue of damages and should be subject to impeachment under Federal Rule of Evidence 609, which allows evidence that a witness has been convicted of a crime punishable by imprisonment for more than one year for the purpose of attacking that witness’s credibility. The magistrate judge ruled that the danger of undue prejudice did not outweigh the probative value of evidence of the fact and duration of Mr. Battle’s incarceration. However, Defendants were ordered not to name or refer to the specific felonies for which Mr. Battle was incarcerated. After the jury was selected, Plaintiffs indicated to the court that they were considering dropping Mr. Battle’s individual claims and again requested the court to exclude all evidence of his felony conviction. The magistrate judge informed Plaintiffs that if Mr. Battle’s individual claims were dismissed, all evidence of his connection to this case would be excluded, including any reference to the felony conviction. In the end, Plaintiffs decided not to drop Mr. Battle’s individual claims.
Ultimately, Mr. Battle chose not to testify and nothing was mentioned concerning his criminal conviction or prison term. The only evidence admitted on this issue is a hand written note on a social services report on page 98 of approximately 1000 pages of Tulane medical records stating that Daniel’s father was not a part of the family unit because he was in jail and had been denied leave to visit. No one referred to this note in front of the jury.
On appeal, Plaintiffs contend that the magistrate judge erred in refusing to exclude this note because the evidence was wholly irrelevant to any issue in the case and that any probative value was outweighed by its prejudicial impact. They note that Mr. Battle did not take the stand during the trial, so that his criminal conviction was not admissible for purposes of impeachment of his credibility.
See
Fed. R.Evid. 609(a). We review trial court evidentiary rulings for abuse of discretion.
See Jon-T Chemicals, Inc. v. Freeport Chem. Co.,
704 F.2d 1412, 1417 (5th Cir.1983). Under this standard, we cannot say that the magistrate judge erred. The fact that Mr. Battle was not part of the family unit during Daniel’s illness and its aftermath was relevant to Mr. Battle’s individual claim for damages and its admission was not an abuse of discretion.
%. Exclusion of Lakeman’s deposition
Fred Lakeman, a Ph.D. microbiomedical researcher and virologist who runs the Whitley lab at the University of Alabama, has done extensive research on HSE. Lakeman was responsible for HSE testing at the Whitley lab in 1994-95, although the record is not clear whether Lakeman personally ran Daniel’s test or had it run by an assistant under his supervision. One critical issue at trial was Defendants’ contention that Daniel did not have HSE. This point is material because HSE is the only form of encephalitis treatable by Acyclovir. Plaintiffs’ claims hinge on their contention that delay in administering Acyclovir was the cause of Daniel’s injuries. The evidence showed that Lakeman’s test of Daniel’s cerebral spinal fluid (“CSF”) extracted on December 27, 1994, was positive for HSE, while Tulane’s test on the same sample of CSF was negative. To prevail, Plaintiffs needed to convince the jury that Lakeman’s positive result was accurate and Tulane’s negative result was erroneous.
Counsel for Defendant Aust noticed Lakeman’s deposition “pursuant to Rule 30, Federal Rules of Civil Procedure and other applicable provisions of said Rules.” During the deposition Defendants posited an objection, taking the position that the deposition could be used for “discovery purposes only.” Plaintiffs countered that they intended to use it “for all purposes allowed by the Federal Rules of Civil Procedures.” Subsequently, Plaintiffs listed Lakeman in the pretrial order as a “may call” witness, as well as a “may call by deposition” witness. Defendants objected in the pretrial order “to the use of depositions of Defendants and other witnesses available live. Most of these depositions are hearsay and do not meet criteria necessary to substitute for live testimony.” At a hearing just before trial started, Defendants submitted that the purpose of Lakeman’s deposition was to develop Plaintiffs’ expert’s opinion and discover the basis of that opinion. Defendants contended that they asked open-ended questions to produce answers to submit to their own experts for review and that they were not prepared and did not cross-examine the witness to challenge or discredit any of his opinions. Defendants further asserted that Lakeman was under Plaintiffs’ control and that Plaintiffs did not demonstrate that he was unavailable. The magistrate judge held that because Plaintiffs had not demonstrated that Lakeman was unavailable and had not noticed a “trial” deposition of Lakeman, the “discovery” deposition of Lakeman was not admissible during Plaintiffs’ case-in-chief.
On appeal, Plaintiffs challenge the distinction between trial and discovery depositions. Rule 32 of the Federal Rules of Civil Procedure provides:
(a) Use of Depositions.
... (3) The deposition of a witness, whether or not a party, may be used by any party for any purpose if the court finds:
... (B) that the witness is at a greater distance than 100 miles from the place of trial or hearing, or is out of the United States, unless it appears that the absence of the witness was procured by the party offering the deposition ...
Fed.R.Civ.P. 32(a). This court has held that nothing prohibits the use of a discovery deposition at trial, particularly against the party who conducted it.
See Savoie v. Lafourche Boat Rentals, Inc.,
627 F.2d 722, 724 (5th Cir.1980).
Dr. Aust defends the trial/discovery dichotomy used by the trial court by reference to Rule 26(b)(4), which states, “[a] party may depose any person who has been identified as an expert whose opinion may be presented at trial.” Fed.R.Civ.P. 26(b)(4)(A). Aust cites the comment to-this subdivision wherein the drafters recognized that effective cross-examination of an expert witness requires advanced preparation, especially in cases which present intricate and difficult issues as to which expert testimony is likely to be determina
tive.
See
Fed.R.Civ.P. 26 cmt. Subdivision (b)(4)-Trial Preparation: Experts. However, that same comment notes that an expert who was an “actor or viewer with respect to transactions or occurrences that are part of the subject matter of the lawsuit” is to be treated as an ordinary witness.
See id.
We venture no opinion concerning whether Rule 26 supports the distinction between trial and discovery depositions of experts made by the trial court because that distinction simply does not apply in this case to Lakeman, who, because he was a fact witness as well as an expert, must be treated as an ordinary witness for purposes of Rule 26 analysis.
Dr. Reeves takes a different tack on appeal, arguing that the deposition was properly excluded because it was inadmissible hearsay. Rule 804 of the Federal Rules of Evidence provides:
(b) Hearsay exceptions.
The following are not excluded by the hearsay rule if the declarant is unavailable as a witness:
(1) Former testimony.
Testimony given as a witness at another hearing of the same or a different proceeding, or in a deposition taken in compliance with law in the course of the same or another proceeding, if the party against whom the testimony is now offered, or, in a civil action or proceeding, a predecessor in interest, had an opportunity and similar motive to develop the testimony by direct, cross, or redirect examination.
Dr. Reeves contends that Defendants did not have the requisite similar motive to develop Lakeman’s testimony as would be the case at trial. Defendants argued to the trial court that the deposition was taken for the limited purpose of developing the expert’s opinion and its basis. ■Defendants asked open ended questions to produce answers to submit to their own experts for review and were neither prepared for, nor did they attempt, cross-examination. Finally, Defendants asserted that Plaintiffs had not demonstrated that Lakeman was unavailable. Magistrate Judge Roper concurred with that position and excluded the deposition, but made no specific finding concerning similar motive.
There is no dispute that Lake-man was more than 100 miles from the place of trial. This issue thus turns on Rule 804’s similar motive requirement. In
United States v. Salerno,
505 U.S. 317, 112 S.Ct. 2503, 120 L.Ed.2d 255 (1992), the Supreme Court held that a party has no right to introduce former testimony under Rule 804 without showing similar motive.
See id.
at 322, 112 S.Ct. 2503. Because similar motive does not mean identical motive, the similar-motive inquiry is inherently a factual inquiry, depending in part on the similarity of the underlying issues and on the context of the questioning.
See id.
at 326, 112 S.Ct. 2503 (Blackmun, J., concurring). Moreover, like other inquiries involving the admission of evidence, the similar-motive inquiry appropriately reflects narrow concerns of ensuring the reliability of evidence admitted at trial.
See id.
The Fifth Circuit has not addressed how a court is to determine similarity of motive for purposes of Rule 804(b)(1). The Second Circuit has held that the test must turn not only on whether the questioner is on the same side of the same issue at both proceedings, but also on whether the questioner had a substantially similar interest in asserting and prevailing on the issue.
See United States v. DiNapoli,
8 F.3d 909, 912 (2nd Cir.1993). The availability of cross-examination opportunities that were forgone is one factor to be considered, but is not conclusive because examiners will virtually always be able to suggest lines of questioning that were not pursued at a prior proceeding.
See id.
at 914. We find this fact-specific test for determining similar motive valuable.
Defendants in this case were clearly on the same side of the same issues at the deposition and at the trial and had the
same interest in asserting and prevailing on those issues. The core of their argument is that they did not aggressively test Lakeman’s answers with cross-examination type questions. They claim their deposition questions were motivated only by the desire to understand Plaintiffs’ case, not to test it with cross examination. Defendants posit no argument that Lake-man’s deposition testimony lacked reliability. They do not suggest a single question or line of questioning that would have added reliability to the deposition. In fact, they characterize Lakeman’s deposition testimony as cumulative of Whitley’s testimony which was admitted at trial. Based on the foregoing, we conclude that Defendants’ motive in questioning Lakeman at his deposition was similar to their motive at trial and consequently, Lakeman’s deposition was admissible pursuant to Rule 804.
Defendants next argue that if the exclusion of Lakeman’s testimony was error, it was harmless error. Lakeman tested Daniel’s CSF under the auspices of Whitley’s research facility. Defendants contend that Whitley’s testimony regarding the lab, the testing procedures and the results covered similar ground, and because Lakeman’s deposition added nothing essential, its exclusion does not rise to the level of affecting Plaintiffs’ substantial rights.
See Polythane Systems, Inc. v. Marina Ventures Int'l, Ltd.,
993 F.2d 1201, 1209 (5th Cir.1993). We disagree.
Dr. Whitley testified summarily that Tulane’s PCR test was done using a “different set of primers and a different essay [sic], by a laboratory that doesn’t have experience doing it.” Lakeman, on the other hand, went into great detail about the differences between the practices of the two laboratories, identifying three variables that could account for the different results,- all of which indicated that the 'Whitley lab result was correct. First, he explained that Tulane used an extraction technique that could fail to pick up all the nucleic acid in a particular sample, while the Whitley lab used the straight spinal fluid. Second, he discussed dangers that arose from the handling of the sample. If the specimen was improperly stored, the target breaks down and yields a false negative. On the other hand, if the specimen was contaminated by the introduction of herpes simplex it would yield a false positive. However, although herpes is a rather common virus, the number of people who are capable of transmitting herpes simplex "for such a contamination at a given time is very small. Third, he explained at length the controls his lab used to guard against false positives and negatives.
Much was made at trial of the fact that the PCR test was not licensed for diagnosing HSE. Lakeman testified that marketing drives the licensing process much more than science. Although the HSE test is as reliable as HIV testing, for example, there is little market for a HSE test, because the disease is so rare. Whitley testified summarily on this point as well, stating only that the PCR test is the diagnostic method of choice, but that licensure has not been pursued “because of the difficulty required.”
Given these differences, we' conclude that Lakeman’s testimony was not merely cumulative of Whitley’s deposition. In fact, it added information that, if the jurors found it credible, might have been determinative of the question of whether Daniel had HSE. Therefore, the exclusion of Lakeman’s deposition testimony was not harmless error.
3. Live Testimony of Dr. Young
On November 21,1997, the district court set this case for trial on its September 1998 calendar. On August 26, 1998, Plaintiff noticed the video deposition of an expert witness, Dr. Lowell Young, in San Francisco, California, due to his unavailability for trial. Defendants objected and the court ruled that Dr. Young’s trial deposition should be taken in the interest of justice.
On September 2, 1998, one day before Dr. Young’s scheduled deposition, Plaintiffs filed a motion for continuance of the trial, citing the unavailability of Dr. Whitley, another expert, for trial or deposition. Defense counsel objected to revealing their cross-examination strategy if Dr. Young was deposed and later allowed to testify live. The court held a hearing on the motion for continuance by telephone conference call on September 3, 1998, just prior to the start of Dr. Young’s deposition, but no record was made of the hearing. On September 9, 1998, the court entered an order finding Plaintiffs’ actions dilatory, granting the motion to continue, resetting the trial for January 25, 1999, and ruling that Dr. Young’s testimony could be presented only by his video deposition taken on September 3,1998.
On appeal, Plaintiffs contend that the district court abused its discretion in requiring Dr. Young’s testimony to be presented by video deposition rather than live. Plaintiffs rely on
Jauch v. Corley,
830 F.2d 47 (5th Cir.1987), which held that a trial court erred in allowing the introduction of a witness’s deposition when the record showed that the witness worked less than a mile from the courthouse, because “a deposition is an acceptable substitute for oral testimony when in-court observation of the witness is extremely difficult or virtually impossible.”
See id.
at 50. Federal Rule of Civil Procedure 32(a)(3), limiting the use of a deposition unless a witness is unavailable or exceptional circumstances justify its admission, formed the basis of
Jauch’s
preference for live testimony at trial.
Defendants respond that Dr. Young lives and works more than 100 miles from the location of the trial, thus satisfying Rule 32(a)(3)(B)’s unavailability requirement and rebutting the preference for live testimony over deposition. There is nothing in the record to otherwise establish his availability. Defendants also point out that the preference for live testimony over depositions is strongest when the deposition is presented to the jury in the form of a cold transcript. A videotaped deposition, on the other hand, allows jurors to gauge the witness’s attitude reflected by his motions, facial expressions, demeanor and voice inflections.
See United States v. Tunnell,
667 F.2d 1182, 1188 (5th Cir.1982).
Even assuming Dr. Young was available for trial, we conclude that the district court did not abuse its discretion in requiring Plaintiffs to use the video deposition rather than live testimony in this case. The district court attempted to balance the competing interests of Defendants in protecting their cross examination strategy against Plaintiffs’ need for Dr. Young’s testimony in light of Plaintiffs dilatory tactics. The district court’s discretion is broad enough to allow the remedy fashioned here — the use of a video deposition of one expert witness in place of live testimony. We find no abuse of discretion in this evidentiary ruling.
B. JURY ARGUMENT AND INSTRUCTIONS
1. Note, not in evidence, read to jury during closing argument
A trial court’s decisions regarding closing argument are reviewed for abuse of discretion.
See Nissho-Iwai Co. v. Occidental Crude Sales, Inc.,
848 F.2d 613, 619 (5th Cir.1988). In reviewing a closing argument, we consider the argument in conjunction with the jury charge and any corrective measures taken by the court.
See Grizzle v. Travelers Health Network, Inc.,
14 F.3d 261, 269-70 (5th Cir.1994).
During closing argument, Dr. Aust’s attorney stated: “Dennis [Aust] wrote me a note and I’m going to read it to you. I am not going to edit it.” Plaintiffs objected that the note was not in evidence and would constitute additional testimony. The magistrate judge overruled the objection and the note was read to the jury:
If all these experts have such difficulty in agreeing with each other, how could a general pediatrician who has never seen herpes simplex encephalitis before supposed to know how to proceed?
The magistrate judge gave no cautionary instruction to the jury that the note was not evidence. However, over the course of the trial the magistrate judge instructed the jury several times that argument and statements of counsel were not to be considered evidence. This instruction was set out in the written instructions to the jury as well.
Plaintiffs argue on appeal that the note constituted testimony by Aust when he was not in court or under oath that was not tested by cross examination or other method of impeachment. Stated differently, the note was hearsay which was inadmissible under Federal Rule of Evidence 801. Plaintiffs argue that the note was particularly harmful because it addressed the burden of proof, essentially stating that a dispute between experts meant Plaintiffs had not proven their case.
Defendants point out that counsel is generally allowed “reasonable latitude” in making argument.
See Whitehead v. Food Max of Miss., Inc.,
163 F.3d 265, 275 (5th Cir.1998). Further, they argue that any error was cured by the jury instructions and was therefore harmless. The question, if stated by the attorney rather than attributed to Dr. Aust, would have been appropriate argument. Defendants contend that framing an appropriate argument as a note from a defendant did not “impair the calm and dispassionate consideration of the case by the jury” and therefore it does not justify reversal.
Dixon v. Int’l Harvester Co.,
754 F.2d 573, 586 (5th Cir.1985). We disagree. A comment by a party made out of court and not under oath is inadmissible hearsay. We conclude that the magistrate judge abused his discretion in allowing defense counsel to circumvent the rules of evidence by reading the note to the jury verbatim in closing argument.
2. Comparative negligence instruction
The magistrate judge granted Defendants’ request for a comparative negligence instruction advising the jury that it could reduce the amount of Mrs. Battle’s damages if it found that she was negligent and that her negligence was a proximate cause of Daniel’s condition. The instruction was clear that, as a matter of law, an infant cannot be comparatively negligent and that Daniel’s damages could not be reduced due to any comparative negligence of his mother.
We review challenges to jury instructions for abuse of discretion.
See United States v. Monroe,
178 F.3d 304, 307 (5th Cir.1999). A judgment will be reversed only if the charge as a whole creates substantial and ineradicable doubt whether the jury has been properly guided in its deliberations.
See Batts v. Tow-Motor Forklift Co.,
978 F.2d 1386, 1389 (5th Cir.1992).
As a prerequisite for the comparative negligence instruction, there must be evidence in the record that Mrs. Battle was comparatively negligent.
See Jackson v. Southern Ry. Co.,
317 F.2d 532 (5th Cir.1963). On appeal, Plaintiffs contend that the instruction was error because no such evidence was admitted at trial. A review of the record belies Plaintiffs’ position. Plaintiffs’ ease hinges on the theory that Daniel’s neurological sequella would have been decreased or prevented if HSE had been diagnosed and Acyclovir antiviral therapy initiated sooner. There is evidence, albeit disputed, that Mrs. Battle refused Dr. Aust’s recommendation that Daniel be admitted to the hospital during his second emergency room visit. Further, there is evidence that Daniel experienced seizure activity on and off all day before she brought him back to the emergency room in the late afternoon of December 26, after Dr. Aust had given her
instructions to call or return to the hospital if there were any changes in the child’s condition. (Again, it is disputed whether this was attributable to Mrs. Battle’s negligence. There was also evidence that Dr. Aust told her not to bring the child right back but to allow the anticonvulsant medication time to work, which could have been interpreted by the jury as accounting for Mrs. Battle’s delay in returning Daniel to the emergency room.)
We conclude, based on the evidence of Mrs. Battle’s decisions to refuse to allow Daniel to be admitted to the hospital earlier and to delay returning to the hospital, that the magistrate judge did not err in giving the jury a comparative negligence instruction.
C. CUMULATIVE EFFECT OF TRIAL ERRORS
Finding merit in two of Plaintiffs’ grounds of error — the exclusion of Lake-man’s deposition and error in jury argument — we conclude that the substantial rights of Plaintiffs were affected. Therefore, judgment for Defendants must be vacated and this case remanded for further proceedings.
D. MEMORIAL HOSPITAL’S LIABILITY
We review
de ,novo
the district court’s grant of summary judgment in favor of Memorial Hospital on Plaintiffs’ state law claims, as well as dismissal of the EMTALA claims,
see Fields v. Hallsville Indep. Sch. Dist.,
906 F.2d 1017, 1019 (5th Cir.1990), and the magistrate judge’s denial of Plaintiffs’ motion to reconsider for abuse of discretion.
See Calpetco 1981 v. Marshall Exploration, Inc.,
989 F.2d 1408, 1414-15 (5th Cir.1993).
1. Statute of Limitations for Mississippi Tort Claims
The district .court granted summary judgment for Memorial Hospital on Plaintiffs’ Mississippi tort claims, finding that Plaintiffs had not brought suit within the applicable one-year statute of limitations. In determining the start date of the one-year time limit, the district court began with the date Daniel was diagnosed with HSE, December 27,1994.
The district court rendered its opinion on December 1, 1998. On January 21, 1999, the Mississippi Supreme Court ruled that the “discovery rule” controls the calculation of the statute of limitations for the Mississippi Tort Claims Act, Miss.Code ANN. § 11-46-11, et seq.
See Barnes v. Singing River Hosp. Sys.,
733 So.2d 199 (Miss.1999). The “discovery rule” provides that the one year statute of limitations begins when the injured party is aware of (1) the injury and (2) that an act or omission of the negligent party caused the injury.
See id.
at 204. The
Barnes
decision was handed down on January 21, 1999.
See id.
On January 25, 1999, trial commenced in this case. On February 1, 1999, after completing their case-in-chief on liability with only a damage witness remaining, Plaintiffs filed a motion for reconsideration of the summary judgment dismissing their state law claims against Memorial Hospital, citing
Barnes.
The magistrate judge initially denied the motion from the bench as untimely and followed up with a written order denying the motion on the merits. The order considers the
Barnes
decision, then notes that the Mississippi Supreme Court issued another opinion concerning the application of the discovery rule to § 11-46-11(3) on February 4, 1999. In the latter decision,
Robinson v. Singing River Hosp. Sys.,
732 So.2d 204 (Miss.1999), the Mississippi Supreme Court unequivocally restricted the discovery rule to latent injuries.
See id.
at 208. On appeal, Plaintiffs argue that Daniel’s injury was latent until February 1996, when they received a letter from Dr. Young stating that, in his opinion, Defendants were responsible for Daniel’s injury because they failed to timely diagnose and treat his infection. Plaintiffs’ argument is without merit. Under
Robinson,
a Mississippi tort claim accrues when a claimant
was aware or should have been aware of his condition.
See id. ■
It cannot be reasonably argued that Daniel’s injury was latent from the time encephalitis was diagnosed in December 1994 until February 1996.
2. Dismissal of EMTALA claims against Memorial Hospital
Judgment as a matter of law is proper if, under the governing law, there is only one reasonable conclusion as to the verdict.
See Deus v. Allstate Ins. Co.,
15 F.3d 506, 513 (5th Cir.1994). In considering whether there is sufficient evidence to submit the case to the jury, the court must examine all evidence in the light and with all reasonable inferences most favorable to the nonmovant.
See Turner v. Purina Mills, Inc.,
989 F.2d 1419, 1421 (5th Cir.993).
Plaintiffs’ complaint alleged liability against Memorial Hospital under EMTALA. Congress enacted EMTALA “to prevent ‘patient dumping,’ which is the practice of refusing to treat patients who are unable to pay.”
Marshall v. East Carroll Parish Hosp.,
134 F.3d 319, 322 (5th Cir.1998). The act requires that participating hospitals give the following care to an individual who is presented for emergency medical care: (1) an appropriate medical screening, (2) stabilization of a known emergency medical condition, and (3) restrictions on transfer of an unstabilized individual to another medical facility.
See
42 U.S.C. § 1395dd(a)-(c). Plaintiffs alleged that Memorial Hospital violated the screening and stabilization prongs of EMTALA. “Because hospitals can act and know things only vicariously through individuals, any EMTALA violation by ... a physician [who treat patients in fulfillment of their contractual duties with the hospital] is also a violation by the hospital.”
See Burditt v. U.S. Dep’t of Health & Human Services,
934 F.2d 1362, 1374 (5th Cir.1991).
a. Screening
An appropriate medical screening examination is determined “by whether it was performed equitably in comparison to other patients with similar symptoms,” not “by its proficiency in accurately diagnosing the patient’s illness.”
Marshall v. East Carroll Parish Hosp. Serv.
Dist., 134 F.3d 319, 322 (5th Cir.998). A hospital’s liability under EMTALA is not based on whether the physician misdiagnosed the medical condition or failed to adhere to the appropriate standard of care.
See id.
Instead, the plaintiff must show that the hospital treated him differently from other patients with similar symptoms.
See id.
at 324. In
Marshall,
the Fifth Circuit found the evidence of an EMTALA violation insufficient where the record contained no description or identification of other patients who allegedly came to the hospital’s emergency room with symptoms similar to those of the plaintiff.
See id.
Plaintiffs identify three parts of the record to support their position that the EMTALA screening claim should have been submitted to the jury. First, they compare the screening performed on Daniel during his first emergency room visit with the screening provided during his second and third visits to establish disparate treatment. Plaintiffs allege that Daniel was subjected to disparate treatment because he was given a lumbar puncture on his first emergency room visit but not on his second visit because Mrs. Battle revealed on the second visit that Daniel was uninsured. The decision that a patient who had a normal lumbar puncture approximately sixteen hours earlier in the same hospital does not require a repeat of that procedure, while arguably an error in medical judgment, does not constitute disparate treatment under EMTALA. Similarly, Plaintiffs’ complaint that Defendants failed to order an EEG, MRI or a CT scan with contrast, does not inform the query relevant to EMTALA liability, that is, how
Memorial Hospital treated other patients with similar symptoms.
Second, the evidence showed that the EEG done on December 27 was not read until January 4, when it was found to be grossly abnormal. While the evidence may support Plaintiffs’ view that the delay was wholly unacceptable, there is no evidence that the hospital afforded Daniel disparate treatment in this respect. Daniel had been admitted by this time and there is no evidence in this record concerning how long it takes to read EEGs for other inpatients at Memorial Hospital.
Third, Memorial Hospital’s Emergency Department Nursing Care Standards provide that “[i]nfants and elderly are usually hospitalized if no definitive source for fever/infection” is determined. Plaintiffs argue that Defendants had not determined a definitive source for Daniel’s fever and infection but discharged him anyway. Evidence that a hospital did not follow its own screening procedures can support a finding of EMTALA liability for disparate treatment.
See Repp v. Anadarko Mun. Hosp.,
43 F.3d 519 (10th Cir.1994). Defendants respond that they had diagnosed Daniel with pneumonia and an ear infection, thus identifying a definitive source of his fever and infection. Further, they argue that the Nursing Care Standards do not embody the hospital’s screening procedures because they are written for use by nurses who have no decision-making authority in hospital admissions. Also, the standards do not dictate admissions but, by use of the word “usually,” simply describe the usual course of events for the information of the nursing staff. Finally, Defendants point to evidence that Dr. Aust would have admitted Daniel but for Mrs. Battle’s refusal.
Defendants’ explanations for Memorial Hospital’s failure to follow its own published standards in Daniel’s case, while perhaps persuasive to a jury, require credibility determinations that preclude judgment as a matter of law. A rational jury may have concluded, based on the notations concerning Daniel’s seizure disorder, that the source of his fever and infection was not determined at the time he was released. Further, a jury could have concluded that Daniel was sent home sooner than other similarly situated patients. The evidence does not support Defendants’ false dichotomy that Defendants had to release Daniel immediately to go home or to admit him as an inpatient. Memorial Hospital’s policy may have been satisfied by further screening — that is, continued observation in the emergency room until the source of Daniel’s fever and infection was confirmed. Finally, we note that the jury heard evidence concerning an alleged motivation for Memorial Hospital’s disparate treatment of Daniel. He was Black, poor, uninsured and presented at the emergency room during the Christmas holidays. Based on the conflicting evidence in the record, we hold that the judgment as a matter of law on the screening prong of their EMTALA claim was error.
b. Stabilization
EMTALA requires stabilization of a known emergency medical condition.
See
42 U.S.C. § 1395dd(b)(l). The duty to stabilize does not arise unless the hospital has actual knowledge that the patient has an unstabilized medical emergency.
See Marshall,
134 F.3d at 325. The statute defines emergency medical condition as
a medical condition manifesting itself by acute symptoms of sufficient severity (including severe pain) such that the absence of immediate medical attention could reasonably be expected to result in—
(i) placing the health of the individual ... in serious jeopardy,
(ii) serious impairment to bodily functions, or
(iii) serious dysfunction of any bodily organ or part[.]
42 U.S.C. § 1395dd(e)(l)(A).
If the hospital has actual knowledge of the emergency medical condition, it must
then provide either “within the staff and facilities available at the hospital, for such further medical examination and such treatment as may be required to stabilize the medical condition, or for transfer of the individual to another medical facility § 1395dd(b)(l)(A) & (B). Under EMTALA, “to stabilize” means “to provide such medical treatment of the condition as may be necessary to assure, within reasonable medical probability, that no material deterioration of the condition is likely to result from or occur during the transfer of the individual from a facility....” 42 U.S.C. § 1395dd(e)(3)(A). The Fifth Circuit has defined “to stabilize” as “[tjreatment that medical experts agree would prevent the threatening and severe consequence of’ the patient’s emergency medical condition while in transit.
See Burditt v. United States Dep’t of Health & Human Servs.,
934 F.2d 1362, 1369 (5th Cir.1991).
In order to prevail on appeal, Plaintiffs must identify evidence from which a jury could conclude that Memorial Hospital had actual knowledge that Daniel had an emergency medical condition and, if so, that he was not stabilized prior to discharge. Plaintiffs point out Dr. Aust’s written diagnosis of “seizure disorder” on the emergency room chart. Plaintiffs’ experts testified that a “seizure disorder” is an emergency medical condition because deterioration is likely to occur, and in fact, in this case did occur. There is evidence in this record from which a jury could conclude that, particularly by the second emergency room visit, Memorial Hospital released Daniel even though the doctors knew he was suffering from seizures that had not been stabilized and were of an unknown etiology. We therefore conclude that the magistrate judge erred in granting judgment as a matter of law on the stabilization prong of Plaintiffs’ EMTALA claim.
IV. CONCLUSION
Based on the foregoing, we affirm summary judgment for Memorial Hospital on Plaintiffs’ state law tort claims, vacate the judgment for Defendants on the negligence claims, vacate the judgment as a matter of law for Defendants on the EM-TALA claims and remand for further proceedings consistent with this opinion.
AFFIRMED in part, VACATED in part and REMANDED.