delivered the opinion of the Court,
in which PHILLIPS, Chief Justice, ENOCH, SPECTOR, BAKER and ABBOTT, Justices, joined.This original proceeding arises out of the breast implant litigation. The issue before us is whether the trial court abused its discretion in consolidating for trial the claims of nine plaintiffs against three defendants. Based on the record before us, we cannot say that the trial court abused its discretion.
I
The underlying suits are four of thousands of breast implant eases pending in Dallas County. At a pretrial hearing that encompassed all breast implant eases pending in the 14 th District Court, the trial court sua sponte set twenty-three of those cases for a single trial. Due to subsequent events not relevant here, only four cases that collectively encompassed the claims of nine separate plaintiffs remain to be tried pursuant to that order. Although there are technically twelve defendants, the parties agree that many of these entities are affiliated and that there are essentially three groups of manufacturers as defendants.
The trial court denied various motions objecting to the consolidation, and the relators unsuccessfully sought a writ of mandamus from the court of appeals. Relators have now petitioned this Court for relief. We granted leave to file in this case and in a mandamus proceeding challenging the consolidation of asbestos-related claims for trial, which we also decide today. See In re Ethyl Corp., 975 S.W.2d 606 (Tex.1998).
II
Before proceeding to the merits, there is one procedural issue that we briefly address. At oral argument in this Court, the plaintiffs contended for the first time that relators, who for ease of identification will be called the defendants, have waived any right to complain about the consolidation. Six of the nine plaintiffs at issue are parties to a single suit. The plaintiffs contend that trying the claims of nine women would be essentially the same as trying the claims of six. The plaintiffs assert that since the defendants never requested a severance of the six claims into single suits, we should not reach the merits of the petition for mandamus. This contention is not well-founded.
The four suits at issue (encompassing claims of nine plaintiffs) were originally assigned to the 14 th District Court. At one time, these cases and most if not all other breast implant eases in Dallas County were consolidated into a special docket for pretrial purposes before a single judge. The Honorable Frank Andrews originally presided over that docket and was succeeded by the Honorable John Marshall. Subsequently, Judge Marshall either remanded or permitted an opt-out of all breast implant cases that had not originally been assigned to the 14111 District Court, over which he presides. As a consequence, all breast implant cases were returned to the courts to which they were originally assigned and the cases at issue in this Court remained in the 14 th District Court. Another special docket was then created for the breast implant litigation in Dallas County, but the 14 th District Court and two other district courts declined to participate in that process and were not required to do so by the local administrative judge in Dallas County or the regional presiding judge.
The order at issue consolidated a number of separate suits pending in the 14th District *603Court for a single trial. No one disputes that fact, and the caption of the order reads “Silicone Gel Breast Implant Litigation.” The order was issued the same day that a hearing was held at which the trial court announced what it called a “little surprise” for the parties. At the hearing, the trial court revealed that “about 20 cases” would be consolidated for trial, but the court would not identify whieh cases. When asked if “cases” meant 20 plaintiffs or 20 different suits in which there were multiple plaintiffs, counsel were informed that it was the latter. At least one of the defendants objected at the hearing to the setting of cases “with any other defendant in any type of consolidated trial” and to “the setting of more than one plaintiff at a time.”
Following this hearing, the trial court issued a written order identifying the cases set for trial, and there were further proceedings and hearings at whieh further objections to the trial of the claims of more than one plaintiff at a time were raised both orally and in written motions. Defendants specifically argued in their motions that each plaintiff’s case should be tried separately. For example, one motion stated:
In short, the consolidation of even a few Plaintiffs, with similar implants but varying claims and histories, would make a trial sufficiently difficult to conduct, without consolidating cases of different manufacturers of different products at different periods of time.
That same motion requested that the order “consolidating these cases for trial be vacated and that these individual causes proceed in separate trials.” The defendants’ objections to trials other than separate trials of no more than one plaintiff at a time were clearly and repeatedly urged in the trial court and were overruled.
The plaintiffs’ argument that no challenge was made in the trial court regarding the suit with six plaintiffs is without merit.
Ill
In Ethyl, we explained that mass torts typically have a life cycle and over time will mature. See 975 S.W.2d at 610. Like the asbestos litigation in Ethyl, the breast implant litigation involves a mass tort. Professor McGovern has defined a mature mass tort as one in which
“there has been full and complete discovery, multiple jury verdicts, and a persistent vitality in the plaintiffs’ [contentions]. Typically at the mature stage, little or no new evidence will be developed, significant appellate review of any novel legal issues has been concluded, and at least one full cycle of trial strategies has been exhausted.”
McGovern, An Analysis of Mass Torts for Judges, 73 Tex.L.Rev. 1821, 1843 (1995) (quoting McGovern, Resolving Mature Mass Tort Litigation, 69 B.U.L.Rev. 659, 659 (1989)).
A court confronted with the task of setting cases for trial should assess the developmental stage of the mass tort. Unlike the asbestos litigation in Ethyl, the breast implant litigation is not a mature tort. Until enough trials have occurred so that the contours of various types of claims within the breast implant litigation are known, courts should proceed with extreme caution in consolidating claims. See generally Ethyl, 975 S.W.2d at 614. We do not have extensive experience with the very diverse classes of claims that have been asserted.
Nevertheless, we cannot say, based on the record before us, that the trial court abused its discretion in consolidating these claims. In Ethyl, we utilized the so-called Maryland factors in analyzing the trial court’s aggregation of claims for trial. See 975 S.W.2d at 615. Those factors are (1) common worksite; (2) similar occupation; (3) similar time of exposure; (4) type of disease; (5) whether plaintiffs were living or deceased; (6) status of discovery in each case; (7) whether all plaintiffs were represented by the same counsel; and (8) type of cancer alleged. Id. at 611. The Maryland factors do not fit breast implant litigation in all particulars, but they do provide some guidance as to the kinds of criteria that should be considered. When we apply criteria comparable to the Maryland factors to the record before us, we conclude that the implant manufacturers have not established how the dif*604ferences among the consolidated claims would materially affect the fairness of a trial.
The first inquiry under the Maryland factors is whether there was a common work-site. Worksites are not at issue in this case, but there is another, analogous consideration not present in asbestos litigation. That is the existence of “learned intermediaries.” Physicians were responsible for implanting the devices at issue, and the warnings and information given to them by the manufacturers and in turn, the warnings and information given to each plaintiff by her physician presumably will vary from patient to patient and from product to product. While some of the package inserts for the implants are part of the record, they are not self-explanatory to one who is not a physician, and the record is otherwise silent as to what warnings were given and how they differ.
Another consideration analogous to the Maryland factors is whether a given plaintiff was implanted with a particular manufacturer’s product. In this ease, seven of the plaintiffs received implants from only one manufacturer, although the two other plaintiffs allege that they received implants from more than one manufacturer. Implants differ from manufacturer to manufacturer, and even implants manufactured by the same company differ from product to product. For example, some implants may be filled with various types of silicone gel, saline, or both gel and saline. Another difference is that some of the plaintiffs’ implants were coated with polyurethane, which allegedly degrades into carcinogens. Here, the plaintiffs received implants with widely varying characteristics at different points in time ranging from 1974 to 1991. There is also the question of rupture. The implants of some plaintiffs allegedly ruptured while the implants of others did not. Of those that did, some ruptures may have been caused by acute trauma while others were not. While the foregoing differences do indicate that the evidence regarding each claim may differ substantially, we cannot conclude with any certainty from this record that those differences are material and would result in prejudice or confuse the jury. There is no evidence explaining what a jury is likely to hear about some but not other implants that would substantially prejudice the trial of these claims.
An additional consideration is whether the types of injuries differ. The assortment of ailments alleged in these consolidated eases include atypical neurological disease, systemic connective tissue disease, flu-like symptoms, and scores of other maladies, as well as “local injuries” such as scarring, capsular contracture, and chest pains. The breast implant litigation does not appear to involve a cluster of conditions or diseases that have related or similar etiologies.
The wide variety of injuries allegedly caused by the implants at issue are similar to the variety of injuries alleged in In re Repetitive Stress Injury Litigation, 11 F.3d 368 (2d Cir.1993). Unlike that ease however, this record does not indicate that the cause of many of the injuries could be, much less is likely to be, something other than the implants. Nor is there is any evidence in this record that the testimony regarding causation with regard to each product and each ailment will substantially and materially differ.
In sum, if the defendants had demonstrated that the disparities among the claims would actually affect the trial of the case and the evidence that the jury would consider, we would be inclined to agree with the manufacturers that a jury would be confused and that the parties would be prejudiced by a single trial of these nine claims. But evidence has not been provided to support the claims of prejudice and confusion. For example, it is alleged that “state-of-the-art” evidence will differ depending on when each implant was received. But there is no evidence in the record to support these assertions, much less evidence of when changes in the state-of-the-art occurred. The manufacturers gave the trial court no guidance as to which particular claims would require different state-of-the-art evidence. Likewise, as we have already alluded, the variety of injuries suffered might require separate trials to prevent injustice and confusion, but there is no evidence from which we can make that determination.
We recognize that the trial court made no attempt to assess whether confusion *605or prejudice would result if these claims were tried together. The only criterion utilized by the trial court was that these cases were all of the breast implant cases pending at the time in its court. We further recognize that even counsel for the plaintiffs stated repeatedly at oral argument that the selection process employed by the trial court was “arbitrary.” But in determining whether mandamus should issue, we cannot plumb the subjective reasoning of the trial court. We must focus on the record that was before the court and whether the decision was not only arbitrary but also amounted “ ‘to a clear and prejudicial error of law.’ ” Walker v. Packer, 827 S.W.2d 833, 839 (Tex.1992) (quoting Johnson v. Fourth Court of Appeals, 700 S.W.2d 916, 917 (Tex.1985)). This record does not demonstrate the prejudice that will result from aggregating these claims for trial.
* ‡ ‡ ‡
Accordingly, we cannot conclude that the trial court abused its discretion, and we deny the request for mandamus relief.
HECHT, Justice, filed a dissenting opinion, in which GONZALEZ, Justice, joined. HARKINSON, Justice, did not participate in the decision.