Syllabus by the Court

         Proximate cause to support a negligence claim is a cause that results in the claimed injuries as part of a natural, continuous, and probable sequence of events. When a one-car accident occurs that does not block the roadway, a second accident caused by the distraction of police or ambulance personnel in the area 35 minutes later is not sufficiently connected to the first accident for the first accident to be a proximate cause of the second.

         Appeal from Shawnee District Court; FRANKLIN R. THEIS, judge.          Paul D. Post, of Topeka, for appellant.

         Gregory S. Young, of Hinkle Elkouri Law Firm L.L.C., of Wichita, for appellees.

         Before MALONE, P.J., BUSER and Leben, JJ.

         LEBEN, J.

         Accepting plaintiff's allegations as true, Jason Packard's negligence caused his car to run off a highway in Topeka and into a tree some distance off the paved surface of the roadway and its shoulder area. Emergency workers arrived shortly after the accident, and traffic on the roadway began to back up. About 35 minutes after Packard's accident, another driver, Judy Brown, failed to respond quickly enough to the traffic congestion; Brown's car struck the plaintiff's car from behind. Plaintiff, Mary Hale, now seeks to recover from Packard and his employer for the injuries she sustained when Brown's car hit Hale's. But liability for negligence is not unlimited--longstanding caselaw holds that liability for negligence arises only when the consequences of an act are probable under normal human experience, not a mere possibility. Applying this rule, we agree with the district court that Packard's negligence is not sufficiently connected to Brown's negligent driving[38 Kan.App.2d 496] to allow Hale to recover from Packard or his employer for her injuries.

         We assume the facts stated by the plaintiff are true because we must. Like the district court, on a motion to dismiss, we must accept the facts that the plaintiff has alleged; we must then determine whether those facts and any inferences reasonably drawn from them state a claim for relief under any possible legal theory. Jones v. State, 279 Kan. 364, 366, 109 P.3d 1166 (2005). The only legal theory discussed by the parties is negligence, and we find no other possible legal theory on these facts. Thus we examine the viability of plaintiff's claim for negligence.

         Packard's negligence, if indeed he was negligent, was most unfortunate. He felt lightheaded and considered pulling over but instead continued to drive along the highway toward his home. He soon passed out, drove off the road, and ran into a tree. The highway was I-470 in Topeka; the accident occurred at 4:57 p.m. on a weekday. As might be expected at that time on a weekday, traffic backed up in the area once the police and an ambulance were on the scene. Judy Brown's collision with plaintiff Hale's car occurred at about 5:35 p.m., and Hale was injured.

         A valid negligence claim requires that a plaintiff meet four elements: the existence of a duty to the plaintiff, a breach of that duty, an injury, and proximate cause. D.W. v. Bliss, 279 Kan. 726, 734, 112 P.3d 232 (2005). The disputed element here is proximate cause. Proximate cause requires more than mere cause in fact. A proximate cause is one that caused the injury " 'in natural and continuous sequence, unbroken by an efficient intervening cause, . . . the injury being the natural and probable consequence of the wrongful act.' " Yount v. Deibert, 282 Kan. 619, 624-25, 147 P.3d 1065 (2006) (quoting St. Clair v. Denny, 245 Kan. 414, 420, 781 P.2d 1043 [1989]).

         Just how likely a consequence must be for it to be considered a "natural and probable consequence" of an act has been stated in a number of ways. The most frequently cited standard in Kansas holds that "[a] defendant is not responsible for all possible consequences of his or her negligence, only those consequences which are probable according to ordinary and usual experience." (Emphasis [38 Kan.App.2d 497] added.) Aguirre v. Adams , 15 Kan.App.2d 470, 472, 809 P.2d 8 (1991). Accord Sly v. Board of Education of Kansas City, 213 Kan. 415, 424, 516 P.2d 895 (1973); Hickert v. Wright, 182 Kan. 100, 108, 319 P.2d 152 (1957); Shideler v. Habiger, 172 Kan. 718, 722, 243 P.2d 211 (1952); Beldon v. Hooper, 115 Kan. 678, 683, 224 P. 34 (1924); Clark v. Powder Co., 94 Kan. 268, 273-74, 146 P. 320 (1915). Although this foreseeability test is stated in terms of events that are "probable," proximate cause may sometimes be found even for events likely to occur less than half the time, especially when the defendant has created a particularly dangerous condition. See, e.g., Wahwasuck v. Kansas Power & Light Co., 250 Kan. 606, 610-12, 828 P.2d 923 (1992); Cooper v. Eberly, 211 Kan. 657, 665, 508 P.2d 943 (1973). Even so, as one court has phrased it, the consequence must occur "with reasonable probability from the negligent act of the defendant." Matter of Rhone-Poulenc Rorer Inc., 51 F.3d 1293, 1303 (7th Cir. 1995). Like many tests found in law, the rule provides a general standard, but its contours must be filled in as specific cases are considered.

         A viable argument can be made in this case for the plaintiff. The starting point for that case is the general rule that proximate cause is normally a question for the jury. Miller v. Zep Mfg. Co., 249 Kan. 34, 50, 815 P.2d 506 (1991). So in most cases the application of this liability-limiting rule will be left to the jury. It is only when the evidence is so clear that it is "susceptible of only one inference" that proximate cause is treated as a question of law for the court and is not submitted to the jury. Cullip v. Domann, 266 Kan. 550, 556, 972 P.2d 776 (1999).

         Further, a Kansas Supreme Court case suggests that the class of cases that should be decided as a matter of law has been narrowed in recent decades by the adoption in Kansas of comparative fault principles in 1974. Reynolds v. Kansas Dept. of Transportation, 273 Kan. 261, 268-69, 43 P.3d 799 (2002). Under prior law, the existence of some fault on the part of the plaintiff, called contributory negligence, barred recovery altogether. In response, courts sometimes found that a plaintiff's fault was not the proximate cause of his injury, lessening the harshness of what was then an all-or-nothing recovery. Similarly, when multiple defendants were involved, [38 Kan.App.2d 498] one's acts might be found the sole proximate cause--or one's negligence might be labeled "active" and the other's "passive"--so as to avoid rules that otherwise would have required equal contribution between them. When comparative fault was adopted, the fact-finder could compare and apportion the fault of all parties and, in Kansas, the plaintiff could recover as long as his or her fault was less than 50 percent. Based on this, the court in Reynolds said that "[w]ith the adoption of comparative fault, Kansas has moved beyond the concept of proximate cause in negligence." 273 Kan. at 269, 43 P.3d 799.

         As we understand it, that statement did not mean that proximate cause was no longer a requirement of a negligence claim in Kansas. The court separately recognized that "[p]roximate cause is not an obsolete concept in Kansas law," 273 Kan. at 268, 43 P.3d 799, and that intervening causes still would cut off liability in "extraordinary cases." 273 Kan. at 269, 43 P.3d 799. And both the Supreme Court and our court have continued to cite and apply proximate cause as an element of negligence claims after Reynolds. See, e.g., Yount 282 Kan. at 624-25, 147 P.3d 1065; D.W. 279 Kan. at 734, 112 P.3d 232; Williamson v. City of Hays, 275 Kan. 300, Syl. ¶ 6, 64 P.3d 364 (2003); Crowe v. True's IGA, 32 Kan.App.2d 602, 612-13, 85 P.3d 1261, rev. denied 278 Kan. 844 (2004); Miller v. Westport Ins. Corp., No. 95,768, 2007 WL 519043, unpublished opinion, filed February 16, 2007, rev. granted June 21, 2007 (review pending). But the Reynolds court was acknowledging what others have called "the logic that changing to a regime of comparative responsibility affects the scope of proximate cause." Restatement (Third) of Torts: Liability for Physical Harm, § 34, comments, p. 682 (Proposed Final Draft No. 1, 2005).

         Consistent with this potentially diminished role for proximate cause, an example provided in the latest Restatement of Torts supports plaintiff's claim. The American Law Institute approved the Restatement (Third) of Torts: Liability for Physical Harm in 2005. Under section 29 of the Third Restatement, the black-letter rule for proximate cause (a term the Restatement avoids) is that "[a]n actor's liability is limited to those physical harms that result from the risks that made the actor's conduct tortious." Restatement (Third) of Torts § 29, p. 575 (Proposed Final Draft No. 1, 2005). [38 Kan.App.2d 499] The Restatement sets forth the view that the distraction arising around an accident scene is within the risks that should be anticipated by a negligent driver:

"Tortious conduct may be wrongful because of a variety of risks to a number of different classes of persons. Thus, driving a vehicle negligently poses risks to persons and property who might foreseeably be harmed in a number of ways--by a collision with another vehicle or pedestrian, by the vehicle leaving the road, by the consequences of a narrowly averted collision, by the confusion and distraction of an accident scene, or by other consequences. Some of those risks may be more prominent than others, but all are relevant in determining whether the harm is within the scope of liability of the actor's tortious conduct." (Emphasis added.) Restatement (Third) of Torts§ 29, comment d, p. 582. The Restatement then provides an example quite similar to the case now before us:

"[Illustration] 6. Parker's automobile was run off a narrow, hilly road by Wilson, who was driving a semitrailer negligently. Because the accident scene involved an unusual configuration of the semitrailer and Parker's vehicle, Deborah, who was driving by, stopped her car at the side of the road to observe the scene. While parked at the side of the road, Deborah was hit by another vehicle driven carelessly into Deborah's car. Whether Deborah's harm is within the scope of liability created by Wilson's negligence in causing the accident with Parker is a question for the factfinder." Restatement (Third) of Torts § 29, comment d, illus. 6.

         Under the view of Third Restatement, then, a plaintiff injured in a follow-on accident apparently caused by the distraction of an earlier accident would be able to sue the negligent driver of the first accident and have that claim submitted to a jury.

         Where to draw the line under legal rules that lack bright-line clarity is a matter of judgment. We do not hide the existence of that judgment call here by presenting only one side of the argument. But though we have sketched out some viable arguments for the plaintiff, we do not find them persuasive.

         The Third Restatement view may one day gain acceptance, but even its authors concede that they have staked out a basis of analysis that is different than the one actually employed by the courts today (or for the past 100 years for that matter). The Restatement reporters recognized that they have proposed a scope-of-risk analysis--rather than the standard foresee ability test applied in Kansas and most other states--for proximate cause. Restatement (Third) of [38 Kan.App.2d 500] Torts§ 29 , comments d and j, pp. 579-81, 594-96 & Reporters' Notes to comments d and j, pp. 610-14, 627-29 (Proposed Final Draft No. 1, 2005). And while they suggest that these different standards are "quite compatible," Restatement (Third) of Torts § 29, Reporters' Note to comment d, p. 610, they do not suggest they are the same.

         Significantly, no case citation is provided as a basis for the Restatement's Illustration 6. This is not because cases involving follow-on automobile accidents are scarce--there were enough cases for a lengthy compilation limited to the proximate-cause issue in follow-on auto accidents back in 1958, and a supplement to that collection lists many cases since then. Annot., Negligence Causing Automobile Accident, or Negligence of Driver Subsequently Approaching Scene of Accident, as Proximate Cause of Injury by or to the Approaching Car or its Occupants, 58 A.L.R.2d 270 (1958 & Supp. 2003).

         The annotator correctly notes that it is difficult to wrest general rules from these fact-specific cases that will universally predict the results. 58 A.L.R.2d 270 § 2. Yet the general trend in the cases cited there does not support the conclusion of Illustration 6, and the cases certainly do not support a finding of proximate cause in Hale's case here against Packard.

         Two major factors seem to provide fault lines for the cases in the annotation. First, in the cases finding proximate cause, there generally was some debris from the first accident still blocking the roadway, while there was not in the cases finding no proximate cause. Compare Morrison v. Frito-Lay, Inc., 546 F.2d 154, 163 (5th Cir. 1977) (finding that the act of parking a truck partially on the roadway could serve as proximate cause both for one car hitting the truck and a second accident in which two other cars collided while trying to avoid the first accident); Anderson v. C.E. Hall & Sons, Inc., 131 Conn. 232, 238, 38 A.2d 787 (1944) (finding that the first accident could be proximate cause for the second when the first accident left the vehicle crosswise on the highway); Johnson v. Sunshine Creamery Co., 200 Minn. 428, 433-34, 274 N.W. 404 (1937) (finding that the first accident, which left two vehicles blocking the highway, could be proximate cause for the second [38 Kan.App.2d 501] accident) with Atkinson v. Allstate Ins. Co., 361 So.2d 32, 34 (La.App. 1978) (finding that the first accident was not cause for the second when the vehicles in the first accident were off the roadway before the second accident occurred); Natell v. Taylor-Fichter Steel Construction Co., 257 A.D. 764, 765-66, 15 N.Y.S.2d 327 (1939) (finding that since cars in the first accident had come to rest off the traveled portion of the highway it was not proximate cause of the second accident), aff'd 283 N.Y. 737, 28 N.E.2d 966 (1940); and Kukacka v. Rock, 154 Or. 542, 545-46, 61 P.2d 297 (1936) (finding negligence of the driver whose car ended up in a ditch not proximate cause of the accident in which the passenger who went onto the roadway to wave for help was struck by another vehicle). Second, in the cases finding proximate cause, there generally was only a short time gap between the first accident and the follow-on accident, while in the cases finding no proximate cause there was a substantial time lapse between them. Compare Sawdey v. Producers' Milk Co., 107 Cal.App. 467, 480, 290 P. 684 (1930) (finding that a 3-minute gap between accidents did not eliminate proximate-cause question); and Anderson, 131 Conn. at 238-39, 38 A.2d 787 (finding that a gap of 2 to 4 minutes between accidents did not eliminate proximate-cause question); and Krumvieda v. Hammond, 71 S.D. 544, 548-49, 27 N.W.2d 583 (1947) (finding that 2 minutes lapsed time between the first and second accidents did not take proximate-cause issue away from jury consideration) with Millirons v. Blue, 48 Ga.App. 483, 484-85, 173 S.E. 443 (1934) (finding that the first accident was not proximate cause of the second accident occurring 20 minutes later when the damaged vehicle was on the roadway with lights on); and Anderson v. Jones, 66 Ill.App.2d 407, 411-12, 213 N.E.2d 627 (1966) (finding that a gap of 3 to 10 minutes between accidents was an important factor in finding no proximate-cause relationship between them). But see Johnson, 200 Minn. 428, 274 N.W. 404 (finding that 1-hour gap did not eliminate proximate-cause relationship when disabled vehicles continued to block the roadway at the time of the second collision). Thus the cases that are found in the annotation suggest that liability would not arise for a follow-on accident to one that did not block the roadway and that occurred quite some time earlier. And that's our case.

         [38 Kan.App.2d 502] The two cases most prominently relied upon by Packard and his employer are fully in line with this analysis. One of the cases involved an initial one-car accident in which the car ran off the roadway, ending up in the untravelled median. O'Connor v. Nigg, 254 Mont. 416, 417-18, 838 P.2d 422 (1992). Passers-by stopped to render assistance, and a highway patrol trooper also arrived; all of them left their vehicles off the traveled portion of the roadway. About 10 minutes after the first accident, a car traveling on the roadway rear-ended another car, even though the lane was unobstructed and the patrol car's lights were flashing. The Montana Supreme Court held that the initial accident could not be considered a proximate cause of the second one because it was not sufficiently probable to be foreseeable. 254 Mont. at 421, 838 P.2d 422. The other case involved a second accident that occurred 20 to 45 minutes after the first one when an inattentive driver ran into an officer directing traffic around the first collision. Williams v. Smith, 68N.C.App.71, 73, 314 S.E.2d 279 (1984). The North Carolina Court of Appeals held that there was not an unbroken connection between the initial accident and the striking of the officer; it also held that the second driver's negligence was not foreseeable in this situation. 68N.C.App.at 73, 314 S.E.2d 279. Collectively, these cases support the conclusion of the district court in our case that follow-on accidents caused by the distraction of an initial wreck or inattention of a later driver are not sufficiently probable to support probable cause: "Rear end collisions, although a foreseeable possibility from such a slow-down of traffic, are not a likely or probable consequence at each one."          The district court also fairly concluded that the primary cases relied upon by plaintiff Hale were readily distinguishable. These cases generally involved disabled vehicles that remained in the road and were then directly involved in a second accident. See Flaharty v. Reed, 167 Kan. 319, 205 P.2d 905 (1949) (finding that defendant's parked car, which obstructed a highway lane, was the proximate cause of a pedestrian's injuries when another vehicle struck the parked car); Walton v. Tull, 234 Ark. 882, 356 S.W.2d 20 (1962) (finding the driver's negligence in an initial accident that left one of two vehicles in the roadway and another immediately behind it [38 Kan.App.2d 503] on the shoulder was the proximate cause of a second accident involving a drunk driver); and Hill v. Wilson, 216 Ark. 179, 224 S.W.2d 797 (1949) (finding a truck driver's unexpected stop in the roadway was the proximate cause of an accident between two vehicles following the truck). These cases are consistent with those we have already discussed and thus do not suggest a different result.

         One final consideration leads us to follow this existing case law rather than the new academic theory found in the Third Restatement. The adoption of a new theory that would subject large numbers of additional drivers to jury trials for follow-on accidents would have substantial real-world costs. Lawsuits are intrusive and expensive. Here, Packard would be subjected to interrogatories, requests for documents, and deposition. His medical records would be combed for signs that he should have anticipated blacking out while driving. And there would be no possibilities for resolution of the suit other than jury trial or settlement. Considering both these realities and existing case law, we believe that a plurality of the Texas Supreme Court got it right when they "decline[d] the invitation to abandon decades of case law" by adopting section 29 of the Third Restatement. Dew v. Crown Derrick Erectors, Inc., 208 S.W.3d 448, 452 n.4 (Tex. 2006).

         We do not believe that the average Kansan whose negligence caused her own car to run completely off the roadway would expect to be liable to someone injured 35 minutes later when a third driver was distracted by the commotion. Nor should she. Such an event is not sufficiently foreseeable for liability under traditional foreseeability analysis, which is fully consistent with commonsense expectations. The judgment of the district court granting defendants' motion to dismiss is therefore affirmed.