Andy R. Allen (“appellee”), as personal representative of the estate of his wife, Patricia Allen, and joined below at trial by his daughter in her individual capacity, sued R.J. Reynolds Tobacco Company and Philip Morris USA Inc. (“appellants,” or “RJR” and “PM USA,” respectively), alleging appellants were liable for the wrongful death of Mrs. Allen at the age of sixty, from chronic obstructive pulmonary disease (“COPD”). The evidence estab*469lished that Mrs. Allen had been a lifelong smoker of both RJR and PM USA’s cigarettes. Following a two-phase Engle1 trial, the jury found both appellants liable under theories of negligence; strict liability for placing defective and unreasonably dangerous cigarettes on the market; fraudulent concealment or omission of the health effects or addictive nature of smoking cigarettes; and conspiracy to conceal or omit material information concerning the health effects or addictive nature of smoking cigarettes. As a result, the jury awarded appellant $6 million in compensatory damages and $34 million in punitive damages.2 The trial court’s final judgment confirmed the compensatory damages awarded, but reduced the punitive damages to $8.1 million against RJR, and $2.7 million against PM USA. The parties now bring this appeal and cross-appeal from the final judgment, raising several issues and sub-issues, each serving as a potential basis for the reversal of all or a portion of the final judgment. We affirm without comment appellants’ Points I, III, and IV. We also affirm appellants’ Point V on the authority of the Florida Supreme Court’s recent decision in Philip Morris USA, Inc. v. Douglas, 110 So.3d 419 (Fla.2013).3 We reverse, however, on appellants’ Point II, which obviates the need to reach appellee’s issues raised on cross-appeal because, for the reasons that follow, our decision requires remand for a new trial.
Under Point II, appellants assert the trial court erred in directing a verdict in favor of appellee on the issue of appel-lee’s membership in the Engle class, by removing the “addiction causation” requirement from the Engle class definition. We review a ruling on a motion for directed verdict de novo, reading the evidence and inferences of fact in a light most favorable to the non-moving party. See Morales v. Weil, 44 So.3d 173, 178 (Fla. 4th DCA 2010). In other words, a trial court should grant a motion for directed verdict only “when the evidence, viewed in the light most favorable to the non-moving party, shows that a jury could not reasonably differ about the existence of a material fact and the movant is entitled to judgment as a matter of law.” Meruelo v. Mark Andrew Of the Palm Beaches, Ltd., 12 So.3d 247, 250 (Fla. 4th DCA 2009); see also Kowkabany v. Home Depot, Inc., 606 So.2d 716, 719 (Fla. 1st DCA 1992) (“[I]n reviewing the propriety of a directed verdict, an appellate court must weigh the facts and inferences to be drawn therefrom in the light most favorable to the person against whom judgment has been granted. A directed verdict can be upheld only if there is no evidence or inference from the evidence which will support the non-moving party’s position.”).
Early in Phase I of the trial, the trial court entered a partial summary judgment in which it found as a matter of law that Mrs. Allen had been addicted to cigarettes containing nicotine. Accordingly, near the close of the evidence, appellee moved for a *470directed verdict on the issue of Engle class membership, asserting that the combined effect of the partial summary judgment and appellants’ concession that smoking had caused Mrs. Allen’s COPD left nothing for the jury to decide on that issue. Appellants countered that Engle class membership requires more than the fact of addiction and evidence that smoking caused the disease; instead, Engle required appellee to produce evidence that Mrs. Allen’s addiction was the legal came of her COPD. Appellants pointed to evidence that Mrs. Allen’s COPD was first diagnosed in 1996, and the evidence was disputed on the issue of whether her addiction had manifested itself prior to that date and was the cause of her developing COPD, as opposed to other factors that explained her decision to smoke, such as her desire to lose weight and for reasons of stress relief. The trial court adopted appellee’s argument, but directed a verdict against RJR only, reasoning that the class membership issue remained a jury question insofar as PM USA was concerned because of evidence that Mrs. Allen had only used PM USA’s brand when she first began smoking.
As a result, the trial court instructed the jury, in pertinent part, as follows:
This action arises out of a case known as the Engle class action. I have determined that Patricia Allen is a member of the Engle class as a matter of law. Certain findings from that action are binding upon you, the court and the parties....
.... [Reading the eight Engle common liability findings.]
However, notwithstanding the foregoing, these findings will not be applicable to the plaintiffs claims against Philip Morris USA, Inc. if you find by the greater weight of the evidence that Patricia Allen’s smoking of Philip Morris USA Inc. cigarettes was not a legal came of her death.
The court has determined that Mrs. Allen was addicted to R.J. Reynolds Tobacco Company cigarettes containing nicotine and that her addiction was a legal cause of her death. The Engle findings will apply to plaintiffs claims against R.J. Reynolds Tobacco Company-
(Emphasis added.)
Thus, as to PM USA, the jury was instructed as follows:
The court has determined as a matter of law that at all times material to this case, Patricia Allen was addicted to cigarettes containing nicotine.
The first issue for your determination on the claim of Andy Allen, Sr., as the personal representative of the estate of Patricia Allen is whether Patricia Allen’s smoking of Philip Morris cigarettes containing nicotine was a legal cause of her death.
Smoking Philip Morris USA, Inc. cigarettes containing a [sic] nicotine is a legal cause of Mrs. Allen’s death if it directly and in a natural and continuous sequence produced or contributed substantially to producing her death so that it can be reasonably be said that, but for that smoking, her death would not have occurred.
We agree with appellants’ position that the trial court’s rationale for the directed verdict, and its above-quoted instructions, which emphasized smoking instead of addiction, reflect a critical misunderstanding of the supreme court’s definition of Engle class membership, which, in turn, resulted in fundamentally flawed instructions to the jury.
In order to gain the advantage of the res judicata effect of the Phase I Engle findings, a plaintiff first must produce suffi-*471dent evidence that he or she is an Engle class member. See Douglas, 110 So.3d at 427; R.J. Reynolds Tobacco Co. v. Martin, 53 So.3d 1060, 1064-69 (Fla. 1st DCA 2010). In Engle, the supreme court clearly and concisely defined the plaintiff class as all Florida “ ‘citizens and residents, and their survivors, who have suffered, presently suffer or have died from diseases and medical conditions caused by their addiction to cigarettes that contain nicotine.’ ” Engle, 945 So.2d at 1256 (emphasis added). In Martin, when this Court was confronted with RJR’s assertion that “the trial court did not require Mrs. Martin to prove legal causation on her negligence and strict liability claims,” we expressly disagreed, observing, “[o]n the contrary,” the trial court instructed the jury that
[t]he first issue for your determination ... is whether Benny Martin was a member of the Engle class. In order to be a member of the Engle class, the plaintiff must prove that Benny Martin was addicted to R.J. Reynolds cigarettes containing nicotine, and, if so, that his addiction was the legal cause of his death .... Addiction is a legal cause of death if it directly and in a natural and continuous sequence produces or contributes substantially to producing such death ... so that it can reasonably be said that, but for the addiction to cigarettes containing nicotine, the death would not have occurred.
53 So.3d at 1064-65 (emphasis added). After reviewing the evidence adduced by Mrs. Martin at trial, we concluded that “Mrs. Martin was required to prove legal causation, and she produced sufficient evidence for a jury to find that Mr. Martin’s addiction to RJR’s cigarettes was the legal cause of his death.” Id. (emphasis added). Thus, contrary to appellee’s claim, class membership is far more than the fact of addiction and the fact that smoking caused the disease; it is, instead the fact that the addiction caused the disease.4
We find the supreme court’s analysis in Douglas highlights and confirms this conclusion. In Douglas, the court was not called upon solely to determine the certified question on due process. It also examined the defendants’ claim that the trial court “misapplied Engle by using the Phase I findings to establish the defect and conduct elements of the plaintiffs claims,” arguing that “because the Engle jury did not adopt a common theory of liability for why their cigarettes were defective or for why their conduct was tor-tious, the Phase I findings are too general to be binding in individual actions.” 110 So.3d at 427. In rejecting this claim, the supreme court explained:
[T]he Phase I common liability jury determined general causation (the connection between the Engle defendants’ addictive cigarettes and the diseases in question), which leaves specific or individual causation (the connection between the Engle defendants’ addictive cigarettes and the injury that an individual *472plaintiff actually sustained) to be determined on an individual basis. The En-gle defendants may defend against the establishment of individual causation, for example, by proving that the disease at issue was the result of a genetic predisposition, exposure to an occupational hazard, or something unrelated to the plaintiffs addiction to smoking the En-gle defendants’ cigarettes.
Id. (emphasis added). Of relevance to the present case, the supreme court referenced the trial court’s instruction to the jury, that it was to determine first whether Mrs. Douglas was a member of the Engle class, explaining, as did the trial court in Martin, that “addiction ... is a legal cause of a disease or medical condition if it directly and in natural and continuous sequence produces or contributes substantially to producing such disease or medical condition so that it can reasonably be said that but for an addiction to cigarettes containing nicotine, such disease or medical condition would not have been suffered.” Id. at 426, n. 4. With this in mind, the supreme court held the Second District “properly applied Engle when holding that legal causation for the strict liability claim was established by proving that addiction to the Engle defendants’ cigarettes containing nicotine was a legal cause of the injuries alleged.” Id. at 429. It went on to state, “When an Engle class member makes this showing, injury as a result of the Engle defendants’ conduct is assumed based on the Phase I common liability findings.” Id. (citing Martin, 53 So.Sd at 1069). More important, in finding that the record in Douglas “conclusively countered] the argument that the Engle defendants are being arbitrarily deprived of their property,” id. at 431, the supreme court emphasized that “individual plaintiffs do not simply walk into court, state that they are entitled to the benefit of the Phase I findings, prove their damages and walk away with a judgment against the Engle defendants.” Id. “Instead, to gain the benefit of the Phase I findings in the first instance, individual plaintiffs must prove membership in the Engle class,” which “often hinges on the contested issue of whether the plaintiff smoked cigarettes because of addiction or for some other reason (like stress relief, enjoyment of cigarettes, and weight control ...).” Id. (emphasis added). The supreme court continued, “Once class membership is established, individual plaintiffs use the Phase I findings to prove the conduct elements of the six causes of action this Court upheld in Engle[.]” Id. at 432. Thus, as illustrated by the jury instructions approved in Douglas and in Martin, addiction causation, not smoking causation, is key to Engle class membership.
In the present case, the directed verdict against RJR, and the instruction given to the jury regarding appellee’s class membership run counter to the analyses of class membership employed in Douglas and Martin. The evidence at the time of the trial court’s decision was mixed on the issue of when Mrs. Allen became addicted to cigarettes in relation to her diagnosis of COPD in 1996. The parties’ experts diametrically disagreed on that subject. Thus, it cannot be said “there is no evidence or inference from the evidence which will support the non-moving party’s position.” Kowkabany, 606 So.2d at 719; see also Phillips v. Van’s Elec. of Lake Worth, Inc., 620 So.2d 253, 253 (Fla. 4th DCA 1993) (observing that “[d]ireeted verdicts in negligence cases must be granted in an especially cautious manner”) Here, RJR introduced evidence sufficient to create a jury issue on whether Mrs. Allen’s addiction caused her COPD. Moreover, because of the improper instructions to the jury, we reject appellee’s claim that the erroneous entry of the directed verdict against RJR was harmless error as re*473gards PM USA. In both instances, the trial court removed from the jury’s consideration the threshold issue on appellee’s class membership for reasons inconsistent with Engle as reaffirmed in Douglas.
In so holding, we also reject appel-lee’s suggestion that the error was harmless given the jury’s finding on the issue of comparative negligence. Douglas is clear that one of the essential components to protecting Engle defendants’ due process rights is the initial inquiry into a plaintiffs membership in the Engle class. Without a proper instruction and finding that Mrs. Allen was a member of the Engle class, appellee, again, was not entitled to rely on the Engle Phase I findings to establish the conduct elements of his claims. After that, the separate questions on comparative fault merely asked the jury to determine who was responsible, and to what degree, for Mrs. Allen’s death; but the Engle class definition makes addiction causation a necessary precursor to the question concerning the degree to which a defendant’s alleged misconduct caused a smoker’s disease. Comparative fault in this ease focused on an entirely different question. Therefore, the jury’s findings on that subject cannot cure the omission of any jury finding on the crucial addiction causation element.
Consequently, for the reasons stated, we reverse the final judgment in favor of ap-pellee, and remand the cause for a new trial to be conducted consistent with this opinion.
AFFIRMED, in part, REVERSED, in part, and REMANDED for further proceedings.
VAN NORTWICK and LEWIS, JJ., concur.. See Engle v. Liggett Grp., Inc., 945 So.2d 1246 (Fla.2006).
. The jury awarded $3 million in compensatory damages each to appellee and to his daughter. The award of punitive damages was likewise evenly split between the two plaintiffs.
. In Douglas, the supreme court answered in the negative the following question certified by the Second District Court of Appeal: "Does accepting as res judicata the eight Phase I findings approved in Engle[ ] violate the [Engle defendants'] due process rights guaranteed by the Fourteenth Amendment of the United States Constitution?” See 110 So.3d at 422 (citing Philip Morris USA, Inc. v. Douglas, 83 So.3d 1002, 1011 (Fla. 2d DCA 2012)). The parties in the present case were given the opportunity to submit supplemental briefs addressing Douglas.
. While our decision in Lorillard Tobacco Co. v. Mrozek, 106 So.3d 479, 480 (Fla. 1st DCA 2012), might be read to suggest otherwise, where we held that "class definition requires only that the smoker is/was addicted to cigarettes containing nicotine, and contracted or died from a disease caused by cigarette smoking," id., we emphasized in the opinion that the "addiction causation” issue was undisputed, as evidenced by the trial court’s entry of a partial summary judgment on that point. Regarding the addicted decedent's continued smoking after learning of the "deleterious health effects” of cigarettes in Mrozek, we considered that fact to raise "a question of comparative fault, and thus, of liability to be determined at trial,” as indicated by the jury’s finding the decedent to be thirty-five percent responsible for her death. Id.