dissenting.
Because the majority misuses this Court’s precedents, the summary judgment record, and even Doe’s own pleadings against her on the question of when a “duty” is raised, I dissent.
I.
This is an appeal from a summary judgment for the defendant. SmithKline as mov-ant had the burden to negate as a matter of law the facts supporting Doe’s cause of action. Wilcox v. St. Mary’s University, 531 S.W.2d 589, 592-93 (Tex.1975). It is not enough that Doe failed to prove or present evidence of elements of her case. Smith-Kline had to disprove at least one element as a matter of law. Gibbs v. General Motors Corp., 450 S.W.2d 827, 828 (Tex.1970). In stating the summary judgment “facts,” the court must take all facts, evidence and inferences in the non-movant’s favor; if there is any uncertainty or ambiguity, it must be resolved in favor of Doe. Nixon v. Mr. Property Management Co., 690 S.W.2d 546, 548-49 (Tex.1985); Hudnall v. Tyler Bank & Trust Co., 458 S.W.2d 183, 185 (Tex.1970). The majority refuses to follow these standards.
The majority states there was “no evidence” that Quaker ever saw SmithKline’s advertising brochure statement that “a positive result from SKBL can be accepted with virtual certainty as evidence of drug use.” (Emphasis in original.) The summary judgment standard requires SmithKline to negate as a matter of law that Quaker saw it. SmithKline made no such showing. The absence of proof one way or the other must be taken in favor of Doe.
Beyond that, however, the majority purposefully ignores the reasonable inferences in favor of Doe that SmithKline did make the misrepresentation to Quaker, and that Quaker relied on it. The summary judgment record reflects that Julie Hanson, the manager of health services for Quaker, and John Mutchler, director of health, safety and environmental programs for Quaker, were primarily responsible for hiring SmithKline to process all of Quaker’s drug screening samples. Hanson and Mutchler worked as part of a task force convened in order to consider drug screening and to hire a suitable testing company. Hanson testified that part of the task force’s investigation was to review brochures from three different testing companies. SmithKline showed videotapes and made personal presentations to the task force. When asked whether SmithKline gave any promotional material, Hanson replied, “Oh, yes.” She added that “They said that their tests were very accurate.” She further asserted that Quaker selected Smith-Kline because it wanted to have “absolute certainty that the tests were accurate.” Mutchler similarly testified that he recalled receiving SmithKline promotional literature, although he states he did not “spend much time on [it].” Doe’s response to SmithKline’s motion for summary judgment included copies of each of these depositions as permitted by Tex.R.Civ.P. 166a(d). While Doe may not have presented any direct evidence that Quaker saw the “virtual certainty” language in any of SmithKline’s promotional materials, the evidence she did present raises the rea*357sonable inference that Quaker saw and relied on this claim.
The majority mentions that SmithKline’s later advertising materials contain the disclaimer that “there is no way to completely eliminate this potential problem” that “certain types of poppy seeds, if consumed in sufficient quantity, can produce a positive result for opiates.” There is no evidence that SmithKline gave this subsequent brochure to Quaker, nor that it told Quaker about the uncertainty before Doe’s claim arose, and there is a reasonable inference this did not occur. According to the depositions of several Quaker employees, if Smith-Kline had told Quaker before Doe’s test, the Doe incident would not have occurred. The reasonable inference from this summary judgment record, taken favorably to Doe, is that SmithKline itself discovered its error from Doe’s misfortune and tried to correct it as to future customers. SmithKline did not use this brochure to disprove anything as a matter of law. Other than its potential for being taken as an admission against Smith-Kline, this brochure has no legal relevance to the summary judgment proceeding.
II.
Our pleading rules also apparently do not apply in addressing Doe’s petition. First, Doe makes a general negligence assertion, to which SmithKline failed to pursue its special exceptions. The specific negligence allegations are preceded by “in, at least, the following ways.” (Emphasis added.) This allegation preserves every general negligence claim reasonably related to the identified incident that Doe may wish to make. Boyles v. Kerr, 855 S.W.2d 593, 601 (Tex.1993); Roark v. Allen, 633 S.W.2d 804, 809-810 (Tex.1982) (general negligence in delivery sufficient to raise negligent use of forceps by doctor); Gulf, Colorado & Santa Fe Ry. Co. v. Bliss, 368 S.W.2d 594, 599 (Tex.1963); Higdon v. Shelton Motor Co., 138 Tex. 121, 157 S.W.2d 627, 627-28 (1942); Scott v. Gardner, 137 Tex. 628, 156 S.W.2d 513, 515 (Tex.1942).
What Doe does plead, liberally construed as we are supposed to do in a summary judgment case, adequately raises Smith-Kline’s misrepresentation as a negligence claim. Subparagraphs 2 and 5, taken together, reasonably allege that SmithKline made the misrepresentation to Quaker that its drug test result meant the subject tested was using illegal drugs. What could be more specific than (5), which states that Smith-Kline was negligent in “[flailing to report the level of opiates found in plaintiffs urine test and failing to inform third persons, including plaintiffs prospective employer, that this level was consistent with poppy seed ingestion and not illicit drug use”? If the level of opiates found was consistent with poppy seed ingestion, it necessarily raises the misrepresentation issue as to the “virtual certainty” statement. In fact, the raw test data from SmithKline’s laboratory, favorably interpreted for Doe, suggest that the lab technicians first found a borderline negative result for Doe, but that it was so close to their threshold value that they tested another portion of the sample, which went over the “line” set for positive results. SmithKline set the standards for a “positive” test, and should not have misrepresented what they meant. This pleading gave fair notice that the “virtual certainty” misrepresentation was raised. See State Fidelity Mortgage Co. v. Varner, 740 S.W.2d 477, 479-80 (Tex.App.—Houston 1st Dist.] 1987, writ denied); Tex.R.Civ.P. 45, 47. The majority abandons these established rules for liberally construing the pleading of the non-movant in a summary judgment proceeding.
The majority goes on to acknowledge there is a reasonable inference rule, but then observes that in Boyles the court held the pleading was not sufficient to raise the factual theory, whereas in Roark and Gulf it was. Without further explanation, the majority declares, “The case before us is more like Boyles than Roark or Gulf.” 903 S.W.2d at 355. The misrepresentation ground here is either expressly raised or, at the least, reasonably related to the incident identified in the pleadings to raise it under the liberal pleading construction rule. The majority simply declines to apply the rule.
Further, as an alternative holding, the majority discards the “reasonably related” rule by citation to Cook v. Brundidge, Fountain, Elliott & Churchill, 533 S.W.2d 751, 759 *358(Tex.1976), which supposedly states the defendant moving for summary judgment need only “meet the plaintiffs case as pleaded,” which the majority suggests means narrowly pleaded. The majority quotes the Brun-didge case out of context, and it simply does not stand for the proposition claimed. Brun-didge does not state that grounds reasonably related are not deemed to be in issue. In the first place, in Brundidge, this Court reversed a summary judgment granted the defendants in the trial court and affirmed by the court of civil appeals. We held that under the liberal pleading and favorable inferences requirements, plaintiffs did raise a fact issue. In the second place, the whole quotation from Brundidge makes clear that “reasonably related” grounds must be negated by defendants:
The summary judgment burden was that of establishing the negative of the statutory issues, namely, that Lyon was not apparently carrying on in the usual way the business of the law firm, i.e., was not acting in the ordinary course of its business and hence was not acting within the scope of apparent authority by force of statute. See Torres v. Western Casualty & Surety Co., 457 S.W.2d 50 (Tex.1970); Gibbs v. General Motors Corp., 450 S.W.2d 827 (Tex.1970). The defendant is required to meet the plaintiffs case as pleaded and to demonstrate that the plaintiff cannot prevail. Glenn v. Prestegord, 456 S.W.2d 901 (Tex.1970); Guidry v. Neches Butane Products Co., 476 S.W.2d 666 (Tex.1972). There can be no further burden upon the plaintiff if the requisite facts for summary judgment are not established by the summary judgment record. See Torres, supra, and Box v. Bates, 162 Tex. 184, 346 S.W.2d 317 (1961).
Cook v. Brundidge, 533 S.W.2d at 759. Clearly the context indicates that “meet the plaintiffs case as pleaded” means construed by the “reasonably related” rule for liberally construing the non-movant’s pleadings.
The majority goes further and declares, without citation to authority, that “[w]hen a plaintiffs petition alleges specific claims but does not limit itself to those claims, a defendant is entitled to summary judgment if he disproves at least one element of each claim specifically pleaded unless plaintiff in response raises a genuine issue of material fact as to some other claim that might be brought within the general language of the petition.” 903 S.W.2d at 355. In one fell swoop the majority reverses the summary judgment presumptions this Court has applied for more than a century. The majority shifts the burden of production of evidence to the plaintiff in situations heretofore requiring the defendant to produce evidence to negate claims. No longer, as the Brundidge case cited by the majority states, is it true that “There can be no further burden on the plaintiff’ if defendant fails to negate an element of every cause of action raised by the pleading under the reasonably related test. Suddenly, and arbitrarily, defendants win, without necessity for special exceptions or other express rulings currently and historically envisioned by our rules, unless plaintiffs come back and expressly present proof of factual theories not explicitly in their pleadings. This is a major and significant change in the law, and it is calculated to disadvantage those seeking redress in our courts.
III.
Contrary to the majority’s analysis, this Court should recognize a duty from the drug testing company to Doe to not affirmatively misrepresent the significance of the test results to Quaker, her employer. The majority miscasts the controversy. It analyzes this case as a traditional negligent failure to disclose case, and discusses Restatement (Second) of TORTS § 551 (1977). Section 551, by its context and express terms, is meant to apply only to commercial and business transactions that do not fit the employment drug testing context. Although it distinguishes the negligent misrepresentation ground, which the majority here expressly fails to reach, one case explains, perhaps better than any other, why neither negligent misrepresentation, nor its related theory of negligent failure to disclose, provides a proper vehicle for determining whether to recognize a duty in the employment testing situation. Hall v. United Parcel Service, cited by the majority for the Hall court’s other conclusion that a *359tort duty should not be recognized for polygraph testing, states:
This line of cases [negligent misrepresentation], however, specifically concerns the extent to which a party who has negligently misrepresented facts may be held liable to those who have relied to their detriment on the misrepresentations. The cited cases are of limited analytical utility in resolving a case such as this, where the injury arose not as a result of the injured’s reliance on negligently made statements but rather as a result of the direct impact that those statements had on the injured party’s business and personal life. In that regard, this case is more analogous to one in which the test subject suffered physical injury by virtue of a negligently administered test than it is to the fact pattern presented in Ossining and its predecessors (cf., People v. Hamilton, 125 A.D.2d 1000, 511 N.Y.S.2d 190 [civil damage suit on behalf of women victimized by “unnecessary touching” and unnecessary, sexually suggestive questioning by polygraph examiner] ).
Hall v. United Parcel Service, 76 N.Y.2d 27, 556 N.Y.S.2d 21, 555 N.E.2d 273, 276 (1990).
The majority next erroneously concludes there is no “duty” from SmithKline to Doe. Conceding the foreseeability of great harm to Doe, the majority bases its conclusion on “other factors.” The best that can be said for this analysis is that it gives precedence to imaginary business interests of Quaker, all absent from this summary judgment record, over the obvious personal interests of Doe. The summary judgment evidence raises a strong inference that Quaker did care about the meaning of the test results and wanted “absolute accuracy” from SmithKline. The majority lets purely theoretical concerns, such as “second hand” marijuana smoke or an employer who wants to be arbitrary and not inquire about the meaning and accuracy of the test, control disposition of this case, where such concerns are totally absent. The court should confine its consideration to whether the summary judgment facts in this case raise a duty on the testing laboratory not to misrepresent the meaning of its results.
For its conclusion that this court should not recognize a duty to Doe under these facts, the majority admits there is an analogy to polygraph testing cases but relies on what it says is the one “highest state court” decision on the issue, which rejects a common law polygraph testing duty — Hall v. United Parcel Service, 76 N.Y.2d 27, 556 N.Y.S.2d 21, 555 N.E.2d 273, 276 (1990). The analysis is more than faulty. First, the New York Court of Appeals in Hall is not the only highest state court to rule on the issue. In a summary judgment context, the Arkansas Supreme Court has ruled polygraph testing can raise a duty for a negligence cause of action. Mechanics Lumber Co. v. Smith, 296 Ark. 285, 752 S.W.2d 763, 765 (1988). The majority dismisses this case with a parenthetical notation that “summary judgment on negligence claim reversed, although the grounds are not clear.” What is clear is that the Arkansas Supreme Court could not have reversed the negligence summary judgment if there were “no duty” from the polygraph testing company to the employee-plaintiff. The very annotation cited by the majority lists this Arkansas case as a “contrary” case to Hall, stating it “impliedly acknowledged a cause of action for such negligence” by reversing the summary judgment. Claudia G. Catalano, Annotation, Employee’s Action in Tort Against Party Administering Polygraph, Drug, or Similar Test at Request of Actual or Prospective Employer, 89 AL.R.4th 527, § 3 at 540 (1991). In addition, as cited below, the highest courts of Colorado and Louisiana denied certiorari review to decisions by their courts of appeals holding such causes of action exist. In the admittedly limited number of jurisdictions where the polygraph issue has arisen, more states have recognized a cause of action than have rejected it. Id. § 3 at 540-45. In cases involving drug testing, the jurisdictions appear to be about equally divided. Id. § 15, at 560-62 & 1994 Supp. 9. No case in any other jurisdiction relied upon Hall for its “no duty” holding.
Moreover, Hall’s rationale would require that this court recognize the existence of a duty. The New York Court of Appeals reasoned that the state legislature had provided *360statutory remedies for abuse by polygraph examiners, Hall, 555 N.E.2d at 276-77, and that the Federal Employee Polygraph Protection Act of 1988, 29 U.S.C. § 2001 et seq., prohibited employer use of polygraphs except under limited circumstances and gave adequate statutory protection. Id. 556 N.Y.S.2d at 25, 555 N.E.2d at 277. Essentially, the court concluded that the legislatures, state and federal, had occupied the field and that it should decline to recognize common law actions which might conflict with statutory schemes. Polygraph testing statutes are not drug testing statutes. No one suggests there are such state statutes as to drug testing in Texas, nor have we found any such comprehensive federal statutes. The reasoning of Hall simply does not apply. To the contrary, the implication from Hall is that New York would recognize a duty for drug testing companies under the summary judgment facts presented here. Furthermore, the rule in Texas, contrary to the New York analysis in Hall, is that state legislative action or inaction does not preempt the common law landscape in the absence of an expressed intention to do so. See El Chico Corp. v. Poole, 732 S.W.2d 306, 310-11 (Tex.1987).
Under this record, the court should address the question as straightforward negligence. “The common law doctrine of negligence consists of three elements: 1) a legal duty owed by one person to another; 2) a breach of that duty; and 3) damages proximately resulting from that breach.” Greater Houston Transp. Co. v. Phillips, 801 S.W.2d 523, 525 (Tex.1990); see El Chico Corp. v. Poole, 732 S.W.2d 306, 311 (Tex.1987); Rosas v. Buddies Food Store, 518 S.W.2d 534, 536 (Tex.1975).
Whether SmithKline owed a duty to Doe not to misrepresent the test’s meaning to Quaker is a question of law. Bird v. W.C.W., 868 S.W.2d 767, 769 (Tex.1994); Greater Houston, 801 S.W.2d at 525; Otis Engineering Corp. v. Clark, 668 S.W.2d 307, 309 (Tex.1983) (quoting Abalos v. Oil Dev. Co., 544 S.W.2d 627, 631 (Tex.1976)). When deciding whether a duty should be recognized this court considers “several interrelated factors, including the risk, foreseeability, and likelihood of injury weighed against the social utility of the actor’s conduct, the magnitude of the burden of guarding against the injury, and the consequences of placing the burden on the defendant.” Greater Houston, 801 S.W.2d at 525; see also Otis, 668 S.W.2d at 309; Graff v. Beard, 858 S.W.2d 918, 920 (Tex.1993); Corbin v. Safeway Stores, Inc., 648 S.W.2d 292, 296 (Tex.1983).
Foreseeability of the risk is “the foremost and dominant consideration.” Greater Houston, 801 S.W.2d at 525 (quoting El Chico, 732 S.W.2d at 311). The majority is correct that foreseeability alone is not sufficient to create a new duty. Bird, 868 S.W.2d at 769, citing Boyles v. Kerr, 855 S.W.2d 593, 599 (Tex.1993); Graff, 858 S.W.2d at 920. Even where harm is foreseeable, as a general rule, “a mere bystander who did not create the dangerous situation is not required to become the good Samaritan and prevent injury to others.” Buchanan v. Rose, 138 Tex. 390, 159 S.W.2d 109, 110 (1942). Only where the party created the dangerous situation or where the party enjoys a special relationship with the other party giving rise to a duty will this general rule not apply. See Buchanan, 159 S.W.2d at 110; Otis, 668 S.W.2d at 309; El Chico, 732 S.W.2d at 312. In my view, SmithKline both “created” the harmful situation by its misrepresentation and “enjoyed” a “special relationship” to those employees and prospective employees whose urine it tested at their risk of loss of employment. Contrary to the majority’s faulty analogy to Buchanan v. Rose, SmithKline was not “a mere bystander” to the situation. SmithKline itself created the risk to Doe by its affirmative misrepresentation to Quaker that its test was a “virtually certain” indication of illicit drug use and its setting of the “positive” threshold for the drug testing at a level that could not exclude poppy seed foodstuff ingestion.
Several courts, recognizing the harm that may foreseeably result from an incorrectly conducted test, have held that a drug testing laboratory owes a duty to perform its services in a reasonable manner. See, e.g., Stinson v. Physicians Immediate Care, Ltd., 269 Ill.App.3d 659, 207 Ill.Dec. 96, 646 N.E.2d 930 (Ill.App.1995) (holding drug testing company owes prospective employee duty not to *361contaminate sample and report a false result); Nehrenz v. Dunn, 593 So.2d 915, 917-18 (La.App.1992); Elliott v. Laboratory Specialists, Inc., 588 So.2d 175, 176 (La.App.1991, writ denied); Lewis v. Aluminum Co. of America, 588 So.2d 167, 170 (La.App.1991, writ denied) (holding that a drug testing company owes a duty to the employee to protect against false positive results in drug screening conducted for an employer or potential employer). A similar result was reached in several jurisdictions regarding polygraph tests where the results would be a factor in whether the employer would hire or continue to employ the test subject. Ellis v. Buckley, 790 P.2d 875, 877 (Colo.App.1989), cert. denied, 498 U.S. 920, 111 S.Ct. 296, 112 L.Ed.2d 249 (1990); Lawson v. Howmet Aluminum Corp., 449 N.E.2d 1172, 1177 (Ind.App.1983); Lewis v. Rodriguez, 107 N.M. 430, 432, 759 P.2d 1012, 1014 (N.M.App.1988); Zampatori v. United Parcel Service, Inc., 479 N.Y.S.2d 470, 473-73 (N.Y.Sup.Ct.1984).
The situation in this case is somewhat different. Here, SmithKline performed the test properly, and reported the scientific results accurately, but allegedly harmed the test subject by virtue of its representations regarding the significance of those results. SmithKline engaged in at least two acts that interpret the results of drug tests. First, SmithKline did not merely report a number that indicated the amount of a specific chemical in Doe’s urine. Instead, it created a benchmark above which a sample was labeled as “positive” and below which a sample was labeled as “negative.” SmithKline used its professional judgment in order to decide where that benchmark should be. Second, its promotional literature advertised that a positive finding indicated, with “virtual certainty,” evidence of drug use. The “virtual certainty” statement is a direct interpretation and an affirmative representation about the meaning of a positive result. The majority cites and quotes from Caputo v. Compuchem Labs., Inc., 1994 WL 100084 (E.D.Pa., Feb. 23, 1994), aff.d, 37 F.3d 1485 (3d Cir.1994), cert. denied, — U.S. -, 115 S.Ct. 733, 130 L.Ed.2d 636 (1995), as a case “with facts very similar.” The facts are not similar, however, because Caputo does not involve an affirmative misrepresentation of the meaning of a “positive” result. Furthermore, since Caputo is an unpublished Pennsylvania Federal District Court decision with a “noted only” affirmance by the Third Circuit Court of Appeals and “noted only” cer-tiorari denial by the Supreme Court, it has limited or no precedential value for Texas jurisprudence.
I would hold that a drug testing laboratory with superior knowledge about its procedures has a duty to not make affirmative representations about the accuracy of those results to an employer in a manner that substantially distorts the significance that should reasonably be attached to them. Under this record, SmithKline failed to meet its burden to negate as a matter of law its breach of this duty. In this summary judgment case the majority errs in holding otherwise.
For the foregoing reasons, I dissent.