concurring in part and dissenting in part.
I join the majority in concluding that a new trial is warranted based on the erroneous and harmful exclusion of testimony regarding the overdiagnosis of amniotic fluid embolus (AFE) that the plaintiff sought to elicit during the cross-examination of the defense’s expert witness. I dissent, however, from the majority’s decision to adopt the same harmless error standard for civil cases as is used in criminal cases, despite the different burdens of proof and constitutional interests that are implicated in the civil and criminal contexts. I also disagree with the majority’s conclusion that the trial court erred in excluding evidence related to the alleged witness tampering of the deputy medical examiner.
My primary disagreement is with the majority’s decision to adopt the criminal “no reasonable possibility” test for harmless error, which is grounded in the “beyond a reasonable doubt” standard of proof that applies in criminal trials. Because a different standard applies in civil cases, and because the attributes of the two types of cases are very different, I would adopt the “more likely than not” standard set forth by the Fourth District Court of Appeal and require the beneficiary of the error in a civil case to show that it is more likely than not that the error did not influence the trier of fact and thereby contribute to the verdict.
I begin by discussing my view concerning the appropriate standard for harmless error in civil cases and then proceed to address each of the specific instances of alleged error in turn.
I. Harmless Error Standard in Civil Cases
While I disagree with the ultimate standard adopted by the majority, I do agree with both the majority and the Fourth District in rejecting the outcome-determinative “but for” test for determining harmless error in civil cases in favor of an “effect on the trier-of-fact” approach, as *1266embodied in the harmless error approach this Court adopted in State v. DiGuilio, 491 So.2d 1129, 1135 (Fla.1986). See majority op. at 1256-57; Special v. Baux, 79 So.3d 755, 770-71 (Fla. 4th DCA 2011). As this Court explained in DiGuilio, “harmless error analysis must not become a device whereby the appellate court substitutes itself for the jury.” DiGuilio, 491 So.2d at 1136. Further, placing the burden of proof on the beneficiary of the error, as the majority properly does, is consistent with our own precedent in civil cases:
Equity and logic demand that the burden of proving such an error harmless must be placed on the party who improperly introduced the evidence. Putting the burden of proof on the party against whom the evidence is used, as the district court did, would simply encourage the introduction of improper evidence. The record in this case shows that the issue of liability was close. We cannot say that the jury’s verdict on liability was not improperly influenced by the evidence of the Gormleys’ insurance claim. The likelihood of improper influence on the liability issue was not rebutted by the one who introduced the evidence.
Gormley v. GTE Prods. Corp., 587 So.2d 455, 459 (Fla.1991); see also Sheffield v. Superior Ins. Co., 800 So.2d 197, 203 (Fla.2001) (“The burden of proving that the admission of the collateral source evidence was harmless rests on [the beneficiary of the collateral source evidence].”).
However, because of the differing burdens of proof and constitutional rights at stake, I disagree with the majority’s adoption of the identical standard for harmless error in civil cases as applies in criminal cases. By adopting the test for harmless error that applies to criminal cases without even referencing the different burdens and interests that apply in the civil context, the majority favors form over substance and offers no compelling explanation as to why the “no reasonable possibility” language from DiGuilio, which is rooted in the “beyond a reasonable doubt” burden of proof, should be used in civil cases. As stated by the Fourth District, the harmless error test for civil cases “should acknowledge the particular attributes of those cases.” Special, 79 So.3d at 770.
Indeed, the most obvious attribute of civil cases that distinguishes them from their criminal counterparts is the “preponderance of the evidence” burden of proof that applies instead of the “beyond a reasonable doubt” standard that is the hallmark of criminal prosecutions. Although the harmless error test articulated by this Court in DiGuilio is based on the “beyond a reasonable doubt” standard that applies in the criminal context, DiGuilio, 491 So.2d at 1138, the majority omits any discussion of the completely different “preponderance of the evidence” standard that applies in the civil context — despite adopting the identical harmless error test.
Specifically, instead of acknowledging this important difference, the majority simply transplants the “no reasonable possibility” language from the criminal harmless error test into the civil test it adopts, while eliminating the “beyond a reasonable doubt” portion of the. test articulated in DiGuilio. Although unstated, the majority would presumably substitute the phrase “more likely than not” in place of “beyond a reasonable doubt,” and thus the test would require the beneficiary of the error to prove “more likely than not” that the error did not contribute to the verdict, or alternatively stated, that there is “no reasonable possibility” that the error contributed to the verdict. Because this test does not account for the different burdens of proof, it simply makes no sense.
*1267While it is of course true that trial level burdens of proof and appellate standards of legal error are different concepts, I respectfully disagree with Justice Lewis’s assertion in his separate opinion that “[t]he level of a factual burden of proof at trial has nothing to do with the standards for appellate legal error.” Specially concurring in part & dissenting in part op. at 1273 (Lewis, J.). As the majority itself makes clear, the harmless error test seeks to determine the effect a particular error had on the trier of fact and is not a device for an appellate court to improperly substitute its judgment for that of the jury. See majority op. at 1256 (“Harmless error is not a device for the appellate court to substitute itself for the trier-of-fact by simply weighing the evidence.” (quoting DiGuilio, 491 So.2d at 1139)). In other words, the harmless error test asks the appellate court to put itself in the shoes of the jury and analyze how the error affected the jury, taking into consideration the permissible evidence upon which the jury could have legitimately relied and the impermissible evidence that may have influenced the jury’s verdict.
In my view, the only way to properly assess how an error would have affected the jury is to apply an appellate standard that accounts for the burden of proof the jury was required to apply at trial. Such an approach does not “commingle[] and confuse[ ]” trial level burdens with appellate standards. Specially concurring in part & dissenting in part op. at 1273-74 (Lewis, J.). Instead, it appropriately acknowledges the proper role of an appellate court performing a harmless error test and recognizes that assessing the impact of an error on the jury — as the majority itself concludes to be the correct approach— requires the test applied by the appellate court to account for the burden applied by the jury.
Section 59.041, Florida Statutes (2003), provides that a court may not set aside a judgment or grant a new trial in any case, whether civil or criminal, unless “the error complained of has resulted in a miscarriage of justice.” As the majority appropriately recognizes, “[ujnder this rule, appellate courts must evaluate harmless error on a case-by-case basis” and determine what constitutes a “miscarriage of justice” based on the record and factors present in each particular case. Majority op. at 1255.
In his separate opinion, Justice Lewis suggests that the Legislature’s use of the same “miscarriage of justice” language for both criminal and civil cases “specifically and unambiguously” indicates that the Legislature intended the same harmless error test to apply in both contexts. Specially concurring in part & dissenting in part op. at 1273-74 (Lewis, J.). However, it is the test this Court adopted in DiGui-lio that gives meaning to the phrase “miscarriage of justice” in criminal cases— while respecting the Legislature’s authority to enact harmless error statutes — and that sets forth a framework for applying the standard on a case-by-case basis. In fact, this point was recognized by the Fourth District, which explained that “the trigger for reversible error is the occurrence of a ‘miscarriage of justice’; how the courts have defined this term has determined the scope of the statute’s application.” Special, 79 So.3d at 761.
Any analysis that is specific to each case must account for the particular attributes and distinctive features of that case. In other words, while “miscarriage of justice” is the standard that the Legislature has chosen to apply to all cases, what constitutes a “miscarriage of justice” in one case is not the same as what constitutes a “miscarriage of justice” in another. It is up to the judiciary to develop a framework for *1268assessing and analyzing this determination in individual cases. If the statute clearly and unambiguously set forth the specific test for courts to apply to determine whether an error is harmless in an individual case, courts would not have struggled for more than a century to interpret the statute, and this Court would hot now be called upon to provide a framework for analyzing what constitutes a “miscarriage of justice” in civil cases.
Instead of adopting the identical test for the sake of consistency between cases that arise in contexts that are completely different, I would adopt the well-reasoned opinion of the Fourth District as to the appropriate standard for harmless error in civil cases. Sitting en banc, the Fourth District cogently articulated why the prop: er test should require the beneficiary of the error to prove “more likely than not that the error did not influence the trier of fact and thereby contribute to the verdict.” Special, 79 So.3d at 771.
In my view, the Fourth District persuasively explained why a different standard for reversal is well-suited to civil cases:
In formulating a harmless error test in civil cases, it is important to recognize that DiGuilio derived its formulation from the elevated burden of proof in criminal cases:
The harmless error test ... places the burden on the state, as the beneficiary of the error, to prove beyond a reasonable doubt that the error complained of did not contribute to the verdict or, alternatively stated, that there is no reasonable possibility that the error contributed to the conviction.
DiGuilio, 491 So.2d at 1135 (emphasis added) (citation omitted). This elevated test acknowledges (1) the higher burden of proof in criminal cases, which amplifies the potential effect of an evidentiary error on the trier of fact, and (2) the special concern for the legitimacy of criminal convictions expressed in the constitutional and statutory protections accorded to criminal defendants. A harmless error test for civil cases should acknowledge the particular attributes of those cases.
Id. at 770. In other words, as DiGuilio makes clear, the “no reasonable possibility” language in the criminal harmless error test cannot be divorced from the “beyond a reasonable doubt” language, since the two phrases are simply alternative formulations of the same test, which is rooted in the “particular attributes” of criminal prosecutions. See Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967) (“There is little, if any, difference between our statement in [a prior case] about “whether there is a reasonable possibility that the evidence complained of might have contributed to the conviction’ and requiring the beneficiary of a constitutional error to prove beyond a reasonable doubt that the error complained of did not contribute to the verdict obtained.”); see also Black’s Law Dictionary 1457 (10th ed. 2009) (defining “reasonable doubt” as “the belief that there is a real possibility that a defendant is not guilty”).
When the Court is required to address harmless error in criminal cases, it is always the State that has obtained a conviction against a defendant based in some measure on an erroneous legal ruling at trial. Therefore, the strictest formulation of the harmless error test, as set forth in DiGuilio, is consistent with the State’s responsibility to ensure that convictions are secured without the assistance of harmful errors, which is an important public policy concern. See DiGuilio, 491 So.2d at 1138-39.
By contrast, an error in a civil case could result in potential harm to either a *1269plaintiff or a defendant. Thus, the use of the same standard for harmless error as applies to the burden of proof in civil cases — that a particular occurrence was “more likely than not” — vindicates the concerns that the majority refers to as “con-serv[ing] judicial resources while protecting the integrity of the process.” Majority op. at 1257. It also “strikes the proper balance between the parties.” Id. In fact, a harmless error test based on the “preponderance of the evidence” standard actually strikes a better and more appropriate balance because it takes into account how each of the parties in a civil case may be affected by an erroneous legal ruling.
As stated by the Ninth Circuit Court of Appeals regarding why that circuit applies a different harmless error standard for civil and criminal cases:
The purpose of a harmless error standard is to enable an appellate court to gauge the probability that the trier of fact was affected by the error. Perhaps the most important factor to consider in fashioning such a standard is the nature of the particular fact-finding process to which the standard is to be applied. Accordingly, a crucial first step in determining how we should gauge the probability that an error was harmless is recognizing the distinction between civil and criminal trials. This distinction has two facets, each of which reflects the differing burdens of proof in civil and criminal cases. First, the lower burden of proof in civil cases implies a larger margin of error. The danger of the harmless error doctrine is that an appellate court may usurp the jury’s function, by merely deleting improper evidence from the record and assessing the sufficiency of the evidence to support the verdict below. This danger has less practical importance where, as in most civil cases, the jury verdict, merely rests on a more probable than not standard of proof.
The second,facet of the distinction between errors in civil and criminal trials involves the differing degrees of certainty owed to civil and criminal litigants. Whereas a criminal defendant must be found guilty beyond a reasonable doubt, a civil litigant merely has a right to a jury verdict that more probably than not corresponds to the truth.
The civil litigant’s lessened entitlement to veracity continues when the litigant becomes an appellant. We conclude that a proper harmless error standard for civil cases should reflect the burden of proof. Just as the verdict in a civil case need only be more probably than not true, so an error in a civil trial need only be mere probably than not harmless. In other words, when an appellate court ponders the probable effect of an error on a civil trial, it need only find that the jury’s verdict is more probably than not untainted by the error.
Haddad v. Lockheed Cal. Corp., 720 F.2d 1454, 1458-59 (9th Cir.1983) (emphasis added) (citations omitted). The Tenth Circuit Court of Appeals has also aptly explained why “absent some countervailing policy ... the harmless error standard should mirror the standard applied at trial,” since logic requires the stringency of the test for judging error on appeal to be the same as the stringency of the standard applied at trial. U.S. Indus., Inc. v. Touche Ross & Co., 854 F.2d 1223, 1252 n. 39 (10th Cir.1988).
Although Justice Lewis in his separate opinion relies on McQueeney v. Wilmington Trust Co., 779 F.2d 916 (3d Cir.1985), for the proposition that the same standard should apply in both contexts, the Third Circuit Court of Appeals has adopted “a moderately stringent, though hot unrea*1270sonably high, standard in civil as well as criminal cases”-unlike the most stringent test for harmless error, that this Court adopted in .DiGuilio for criminal cases. Id. at 927; see also Gov’t of Virgin Islands v. Toto, 529 F.2d 278, 284 (3d Cir.1976) (explaining that the Third Circuit follows the “middle ground” or “highly probable” test for harmless error). The harmless error test adopted by the Third Circuit is simply not the standard in Florida, and the considerations identified by the Third Circuit as a basis for applying the same test are therefore less persuasive. See Roger J. Traynor, The Riddle of Harmless Error 35 (1970) (explaining that an appellate court has three options in choosing a harmless error standard: (1) more probable than not; (2) highly probable; or (3) almost certain); McQueeney, 779 F.2d at 924 (stating that “beyond a reasonable doubt” is “a more stringent standard than ‘high probability,’ ” as followed in the Third Circuit).
The Third Circuit has stated, as a basis for adopting the same test for harmless error in civil and criminal cases, that “broad institutional concerns militate against increasing the number of errors deemed harmless.” McQueeney, 779 F.2d at 927. This Court, however, already follows that policy in applying the most stringent of possible harmless error standards for criminal cases, where the defendant’s constitutional liberty interest is always at stake. Unlike in Florida, the Third Circuit’s test for harmless error in the criminal context is not based on the “beyond a reasonable doubt” standard. To apply a lower standard for civil cases than an already less stringent standard that applies to criminal cases would lead to the kind of potential for an increased number of errors that is simply not an issue in Florida.
For all these reasons, I would adopt the Fourth District’s cogent articulation of the proper test for harmless error in civil cases. Under this approach,
harmless error occurs in a civil case when it is more likely than not that the error did not contribute to the judgment. To avoid a new trial, the beneficiary of the error in the trial court must show on appeal that it is more likely than not that the error did not influence the trier of fact and thereby contribute to the verdict.
Special, 79 So.3d at 771.
II. The Alleged Errors
Having set forth what I believe to be the appropriate test for harmless error in civil cases, I now address the two specific instances of alleged error in this case. In short, I agree with the majority’s conclusion as to the first, but disagree as to the second.
A. Exclusion of Cross-Examination Concerning AFE
Although I disagree with this Court’s adoption of the same test for harmless error in civil cases as the Court applies in criminal cases, I agree with the majority that the error in restricting the cross-examination of the defense expert regarding the overdiagnosis of AFE was not harmless error, regardless of which standard is applied. Because the focus of the entire case revolved around whether Susan Special’s death was caused by medical malpractice or AFE, the limitation on the cross-examination and the acknowledgment by the .defense’s own expert concerning the possibility of overdiagnosis of AFE cannot be said to amount to harmless error. That is, Baux and West Boca have not demonstrated that it is more likely than not that the exclusion of this testimony did not contribute to the jury’s verdict finding the defendants not liable. This testimony directly called into question the credibility of the AFE diagnosis in this *1271case, particularly because AFE is a diagnosis of exclusion, and unduly restricted Special’s ability to undermine the defense expert’s testimony in the minds of the jurors.
Further, I agree that the ability of trial counsel to argue about overdiagnosis in closing is not a substitute for having the defense’s own expert acknowledge the ov-erdiagnosis of AFE, where the competing expert opinions were the focal point of this medical malpractice case. See Linn v. Fossum, 946 So.2d 1032, 1041 (Fla.2006) (“We conclude that the trial court erred in allowing Dr. Weaver-Osterholtz to testify that she consulted with colleagues and that this error was not harmless because the competing expert opinions on the proper standard of care were the focal point of this medical malpractice trial.”); see also Donshik v. Sherman, 861 So.2d 53, 56 (Fla. 3d DCA 2003) (“Where, as here, the competing expert opinions, on both sides, were the focal point of the trial, we cannot deem the error in the introduction of the ACAS report to be harmless.”).
Accordingly, because the trial court’s error in restricting the cross-examination of the defense expert was not harmless, a new trial is warranted.
B. Exclusion of Testimony Concerning Alleged Witness Tampering
Lastly, based on the record before the trial court, I disagree with the majority that there was any error in disallowing testimony of the deputy medical examiner, Dr. Barbara Wolf, on what the plaintiff alleged to be “witness tampering.” I deplore, as does this Court, any threats or attempts at intimidation attributable to the adverse party. The key, however, is that the proper predicate must be laid to show that it was the adverse party or its agent that made the threats or attempted to intimidate a witness. To allow a party to argue witness intimidation where there is insufficient evidence demonstrating that this is what occurred is incredibly prejudicial to the party being accused of the improper conduct. See Penalver v. State, 926 So.2d 1118, 1129-30 (Fla.2006). The majority, in fact, recognizes this critical point in correctly concluding that there was an insufficient factual basis to attribute the Department of Health complaint against Dr. Wolf to the defendants and that this evidence was therefore properly excluded.
However, as to the pre-deposition comments allegedly made to Dr. Wolf by her attorney, neither Dr. Wolfs attorney nor Baux’s defense attorney, the original de-clarant, testified as to what was said. Therefore, as the trial court properly recognized, the double hearsay issue with these alleged statements presents a threshold problem for their admission.
Moreover, simply being provided with photographs of slides taken by another expert, Dr. Factor, does not, in itself, amount to witness intimidation. There would have been nothing wrong with the plaintiff asking Dr. Wolf at trial if her attorney showed her photographs of slides taken by Dr. Factor before her deposition and whether that changed her opinion regarding her conclusion about the cause of death. That is a far cry, though, from allowing the plaintiff to argue to the jury that agents of the defense attempted to intimidate the witness and that those actions are evidence of their culpability.
It may be that in a new trial, a more specific nexus can be established with the defendants’ alleged actions to pressure Dr. Wolf to change her testimony, including testimony from either of the two attorneys who engaged in the alleged conversation. The current record, however, is devoid of anything but hearsay and attenuated connections, with speculation having to fill the *1272missing pieces. While I do not object to the further development of facts on this issue on remand, I would affirm the trial court’s decision to exclude the evidence based on the facts as presented.
III. Conclusion
In sum, I concur in the majority’s conclusion that a new trial is warranted based on the trial court’s error, which was not harmless, in excluding cross-examination testimony from the defense expert concerning the overdiagnosis of AFE. I dissent, however, from the majority’s adoption of the-identical standard for harmless error in civil cases as applies in the criminal context. This decision, in my view, fails to account for the important differences that exist between civil and criminal trials.
LEWIS, J., specially concurs in part and dissents in part with an opinion.