The United States Court of Appeals for the First Circuit (Coffin, Senior Circuit Judge) has certified the following question of law, see SUP. CT. R. 34:
Under New Hampshire law, in a crashworthiness or enhanced injury case, does the plaintiff bear the burden of demonstrating the specific nature and extent of the injuries attributable to the manufacturer, or does the burden of apportionment fall on the defendant once the plaintiff has proved causation?
Trull v. Volkswagen of America, Inc., 187 F.3d 88, 103 (1st Cir. 1999). We conclude that the defendants bear the burden of apportionment once the plaintiffs prove causation.
We adopt the court of appeals’ recitation of the facts. In February 1991, the plaintiffs, David and Elizabeth Trull, and their two sons, Nathaniel and Benjamin, were traveling in New Hampshire when their Volkswagen Vanagon slid on black ice and collided with an oncoming car. Id. at 91. Both parties agree that Nathaniel and Benjamin were seated in the rear middle bench seat of the Vanagon, which was equipped with lap-only seatbelts, and were wearing the available lap belts. Benjamin died in the accident, and both Elizabeth and Nathaniel suffered severe brain injuries. Id.
In this diversity products liability action, the plaintiffs sought damages from the defendants “on the ground that defects in the design of the Vanagon made their injuries more severe than they otherwise would have been.” Id. “Plaintiffs had two primary theories of recovery: (1) the Vanagon was defective because it was a forward control vehicle constructed in such a way that it lacked sufficient protection against a frontal impact, and (2) the Vanagon was defective because the rear bench seats, on which Nathaniel and Benjamin were seated, did not have shoulder safety belts as well as lap belts.” Id. at 92. The plaintiffs contend that the defendants are liable in, inter alia, negligence and strict liability because the automobile was not crashworthy. See id.
The United States District Court for the District of New Hampshire granted summary judgment for the defendants on a breach of warranty claim, and both Elizabeth and David Trull’s claims were dismissed with prejudice. Id. The trial proceeded with Nathaniel’s *261and Benjamin’s claims, and the jury found for the defendants. Id. at 92-93.
The plaintiffs appealed to the United States Court of Appeals for the First Circuit, arguing, among other things, that the district court “improperly imposed on plaintiffs the burden of proving the nature and extent of the enhanced injuries attributable to the Vanagon’s design.” Id. at 92. Recognizing that the question “of who, under New Hampshire law, should bear the burden in a so-called ‘crashworthiness’ case, poses sophisticated questions of burden allocation involving not only a choice of appropriate precedent but also an important policy choice,” the court of appeals granted the plaintiffs’ motion to certify the question to this court. Id. at 92, 103.
The plaintiffs’ theory of liability for defective design is commonly referred to as the “crashworthiness,” “second collision,” or “enhanced injury” doctrine. See Caiazzo v. Volkswagenwerk A. G., 647 F.2d 241, 243 n.2 (2d Cir. 1981) (defining “crashworthiness” as “the protection that a passenger motor vehicle affords its passengers against personal injury or death as a result of a motor vehicle accident”); Larsen v. General Motors Corporation, 391 F.2d 495, 502 (8th Cir. 1968) (defining “second collision” as that occurring between the passenger and the interior of the vehicle); Smith v. Ariens Co., 377 N.E.2d 954, 956-57 (Mass. 1978) (permitting recovery where design defect “enhanced injuries” plaintiff received in collision).
The crashworthiness doctrine “extends the scope of liability of a manufacturer to the situations in which the construction or design of its product has caused separate or enhanced injuries in the course of an initial accident brought about by an independent cause.” Bass v. General Motors Corp., 150 F.3d 842, 847 (8th Cir. 1998) (quotation omitted). The doctrine is implicated, not because the design caused the accident, 63A Am. Jur. 2D Products Liability § 1020, at 175 (1997), but because, as a result of the second collision, the plaintiffs suffered either a more severe injury or an injury they otherwise would not have received due to the defective design, see Lee v. Volkswagen of America, Inc., 688 R2d 1283, 1286 (Okla. 1984). Consequently, the plaintiffs seek damages from the defendants for at least a portion of their injuries.
In order to answer the certified question, we first must decide whether a manufacturer may be held liable for enhanced injuries arising from a defective design. See Smith, 377 N.E.2d at 956-57. If a manufacturer can be held liable, then we must determine under what circumstances, if any, the burden shifts to the defendant manufacturer.
*262Two divergent approaches have been developed to analyze whether a manufacturer may be held liable for enhanced injuries arising from a defective design. The first concludes that a product’s intended purpose does not include its involvement in collisions with other objects, and thus refuses to hold a manufacturer liable for enhanced injuries due to defective design resulting from such collision. See, e.g., Evans v. General Motors Corporation, 359 F.2d 822, 825 (7th Cir.), cert. denied, 385 U.S. 836 (1966), overruled by Huff v. White Motor Corp., 565 F.2d 104 (7th Cir. 1977). While the continued vitality of Evans is questioned, subsequent decisions continue to recognize it as a possible approach to this issue. See Smith, 377 N.E.2d at 957 n.3. The second approach concludes that enhanced injuries arising from collisions are foreseeable in the normal use of automobiles aqd imposes liability on manufacturers for such injuries. Larsen, 391 F.2d at 502. Although Larsen was a negligence case, courts have applied its interpretation of “intended use” to the strict liability area. Turcotte v. Ford Motor Company, 494 F.2d 173, 181 (1st Cir. 1974).
Under New Hampshire ■ law, the duty of a manufacturer “is limited to foreseeing the probable results of the normal use of the product or a use that can reasonably be anticipated.” McLaughlin v. Sears, Roebuck, 111 N.H. 265, 268, 281 A.2d 587, 588 (1971). We do not; however, restrict ■ this rule to the “intended” purpose of the product. See Price v. BIC Corp., 142 N.H. 386, 390, 702 A.2d 330, 333 (1997). “Manufacturer liability may . . . attach even if the user employs the product in an unintended but foreseeable manner.” Id. Furthermore, we have previously recognized that an enhanced injury theory “may be applicable in a products liability action against a manufacturer.” Cusson v. Beauregard, 143 N.H. 410, 412, 727 A.2d 979, 981 (1999); cf. Chellman v. Saab-Scania AB, 138 N.H. 73, 76, 637 A.2d 148, 152 (1993) (recognizing defective design claim against automobile manufacturer due to, inter alia, driver-side door opening during rollover).
We conclude, therefore, that our case law supports the Larsen approach. While we do hot hold that manufacturers are “insurers” for defectively designed vehicles, see Price, 142 N.H. at 390, 702 A.2d at 333, we do hold that in a crashworthiness case, a “manufacturer should be liable for that portion of the damage or injury caused by the defective design over and above the damage or injury that probably would have occurred as a result of the impact or collision absent the defective design.” Larsen, 391 F.2d at 503.
*263Thus, in crashworthiness cases,
where the injuries sustained are separate and divisible . . ., the burden of proof remains solely upon the plaintiff[s], including the burden of proving “enhancement,” i.e., the plaintiff[s] must prove which of the several injuries are attributable to the manufacturer’s defective product and the degree of “enhancement” occasioned by the product as distinguished from the injuries flowing from the third party’s acts of negligence.
Lee, 688 P.2d at 1288. The question of actual apportionment of damages among several causes is one of fact for the jury. Mitchell v. Volkswagenwerk, AG, 669 F.2d 1199, 1209 (8th Cir. 1982). In essence, the normal principles of torts apply.
When, however, the plaintiffs receive injuries that are indivisible, courts are split as to whether the plaintiffs or the defendants bear the burden of segregating the injuries caused by the automobile’s defect. See, e.g., Polston v. Boomershine Pontiac-GMC Truck, Inc., 952 F.2d 1304, 1310 (11th Cir. 1992), certified question answered by 423 S.E.2d 659 (Ga. 1992); RESTATEMENT (THIRD) OF TORTS: PRODUCTS LIABILITY § 16 comment d at 243-53 (1998).
The defendants urge us to adopt the minority approach referred to as the “Huddell-Caiazzo” approach, which places the burden on the plaintiffs to prove the nature and extent of their enhanced injuries. See Huddell v. Levin, 537 F.2d 726, 737-38 (3d Cir. 1976) (applying New Jersey law); Caiazzo, 647 F.2d at 250 (applying New York law).
Under the Huddell-Caiazzo approach,
[f]irst, in establishing that the design in question was defective, the plaintiff[s] must offer proof of an alternative safer design, practicable under the circumstances. Second, the plaintiff[s] must offer proof of what injuries, if any, would have resulted had the alternative, safer design been used. Third, the plaintiff[s] must offer some method of establishing the extent of enhanced injuries attributable to the defective design.
Caiazzo, 647 F.2d at 250 (ellipses and quotation omitted).
The plaintiffs, conversely, urge us to adopt the majority approach referred to as the “Fox-Mitchell” approach, derived from Fox v. Ford Motor Co., 575 F.2d 774, 786-88 (10th Cir. 1978) (applying Wyoming law), and Mitchell, 669 F.2d at 1206-08 (applying Minne*264sota law). See RESTATEMENT (THIRD) OF TORTS: PRODUCTS LIABILITY § 16 reporter’s note at 244 (noting that Fox-Mitchell approach is adopted in majority of jurisdictions).
Under the Fox-Mitchell approach, the plaintiffs must “prove only that the design defect was a substantial factor in producing damages over and above those which were probably caused as a result of the original impact or collision.” Trull, 187 F.3d at 101 (quotation omitted). This approach provides that once the plaintiffs carry the burden of proving that the defective design of the car was a substantial factor in causing the enhanced injury, the burden of proof shifts to the tortfeasors to apportion the damages between them. See Polston v. Boomershine Pontiac-GMC Truck, 423 S.E.2d 659, 660 (Ga. 1992).
The principles that guide our answer to the question of which approach New Hampshire should adopt are derived from products liability law grounded in both negligence and strict liability. We therefore review fundamental principles of these areas.
A design defect exists “when the product is manufactured in conformity with the intended design but the design itself poses unreasonable dangers to consumers.” Thibault v. Sears, Roebuck & Co., 118 N.H. 802, 807, 395 A.2d 843, 846 (1978). A strict liability action based upon a theory of defective design may be joined with an action groundéd in negligence. Id. at 811, 395 A.2d at 849. “To maintain a products liability claim based on defective design, a plaintiff must prove: (1) that the design of the product created a defective condition unreasonably dangerous to the user; (2) that the condition existed when the product was sold by a seller in the business of selling such products; (3) that the use of the product was reasonably foreseeable by the manufacturer; and (4) that the condition caused injury to the user or the user’s property.” Chellman, 138 N.H. at 77, 637 A.2d at 150.
“[Cjausation is a necessary element in both negligence and strict liability actions.” LeFavor v. Ford, 135 N.H. 311, 313, 604 A.2d 570, 572 (1992). The negligent design of a product must be a proximate cause of the harm to the plaintiff. Cf. Brookline School Dist. v. Bird, Inc., 142 N.H. 352, 356, 703 A.2d 258, 261 (1997) (failure to warn must be proximate cause of plaintiff’s injuries). “An actor’s negligent conduct is a proximate cause if it is a substantial factor in bringing about the harm, and if the harm would not have occurred without that conduct.” Id. at 354, 703 A.2d at 260 (quotation and citation omitted).
While “the plaintiff has the burden of proving that the defendant’s negligent'act. . . contributed to cause the plaintiff’s damages,” id. *265at 356, 703 A.2d at 260, the plaintiff need not show that the defendant’s design was the sole or dominant cause of the injuries, see LeBlanc v. American Honda Motor Co., 141 N.H. 579, 586, 688 A.2d 556, 562 (1997); cf. Peterson v. Gray, 137 N.H. 374, 378, 628 A.2d 244, 246 (1993).
In crashworthiness cases involving indivisible injuries, we conclude that the plaintiffs must prove that a “design defect was a substantial factor in producing damages over and above those which were probably caused as a result of the original impact or collision. Once the plaintiff[s] make[] that showing, the burden shifts to the defendants] to show which injuries were attributable to the initial collision and which to the defect.” Trull, 187 F.3d at 101-02 (quotation and citation omitted).
This answer is supported by our treatment of products liability actions, where we have, based upon a “compelling reason of policy,” abandoned the higher burden of proof of negligence actions in lieu of adopting the less stringent burden of proof of strict liability. Bagley v. Controlled Environment Corp., 127 N.H. 556, 560, 503 A.2d 828, 826 (1986). Our rationale has been that the plaintiff’s burden “had proven to be, and would continue to be, a practically impossible burden.” Id. Similar policy reasons compel us to allocate the burden of apportionment to the defendants once the plaintiffs prove causation.
In contrast, “[a]doption of the Huddell[-Caiazzo] position takes away the incentive of automobile manufacturers to design their products in a responsible fashion.” Polston, 423 S.E.2d at 661. Furthermore, “application of the Huddell[-Caiazzo] standard might impair the promotion of ‘safer products’ design by weakening the deterrent value of products action.” Id,, at 661-62 (quotation and ellipsis omitted).
We agree with the Mitchell court’s rejection of Huddell on the basis that a plaintiff would be “relegated to an almost hopeless state of never being able to succeed against a defective designer,” Mitchell, 669 F.2d at 1204, and that the Huddell-Caiazzo approach “requires obvious speculation and proof of the impossible,” id. at 1205. To hold otherwise would result in the implicit adoption of the view that
it is better that a plaintiff, injured through no fault of his own, take nothing, than that a wrongdoer pay more than his theoretical share of the damages arising out of a situation which his wrong has helped to create. In other words, the rule is a result of a choice made as to where a loss due to *266failure of proof shall fall — on an innocent plaintiff or on defendants who are clearly proved to have been at fault.
Id. at 1208; see RESTATEMENT (THIRD) OF TORTS: PRODUCTS LIABILITY § 16 comment d at 240, We have adopted similar reasoning in holding.that two dr more tortfeasors may be jointly and severally liable where their negligence, through their independent acts, produces a single, indivisible injury. Carpenter v. Company, 78 N.H. 118, 121-22, 97 A. 560, 561-62 (1916). Consequently, the defendants should bear the burden of apportioning their respective liability once the plaintiffs have made a prima facie showing that the defendants’ conduct contributed as a proximate cause to the harm suffered. See 3 S. SPEISER ET AL., THE AMERICAN LAW OF TORTS § 11:32, at 499 (1986).. '
We do not accept the defendants’ contention that the following language of Pillsbury-Flood v. Portsmouth Hospital, 128 N.H. 299, 512 A.2d 1126 (1986), is dispositive:
[T]he plaintiff -retains the ultimate burden of persuasion ■. ■. ... This rule ensures that defendants will not have to defend against improbable claims, and that defendants will not have to disprove the element of causation. Proving a negative poses practical and theoretical problems that should not be placed on defendants. Moreover, this rule ensures that juries will not have to engage in speculation and conjecture in determining whether liability exists.
Id. at 304, 512 A.2d at 1129-30 (citations omitted).
As the court of appeals recognized, this rule addresses whether “the burden of negating causation should have been shifted to the defendants.” Trull, 187 F.3d at 102. “The Fox-Mitchell approach does not relieve a plaintiff of the threshold obligation of proving causation, and thus liability; it is only after a plaintiff has demonstrated that the design defect was a ‘substantial factor’ in producing damages over and above those that otherwise would have occurred that the burden shifts to the defendant to apportion damages.” Id. Therefore, Pillsbury-Flood is not dispositive of the question before us.
The defendants finally contend that because a manufacturer will normally defend a crashworthiness claim by arguing that no defect in the'vehicle existed, or that the plaintiffs’ injuries could not have been prevented due to the severity of the accident, requiring them to apportion the plaintiffs’ injuries would illogically “require the manufacturer to prove what it vehemently denies, that the plain*267tiff[s’] injuries were enhanced by a defect which the manufacturer claims does not exist.”
This argument misses the mark. Defendants regularly contest the extent of plaintiffs’ injuries while at the same time denying liability for the underlying causes. Requiring the defendants to apportion the plaintiffs’ injuries does not require them to prove that which they deny. The burden of proving causation is on the plaintiffs. The only burden on the defendants is to apportion the plaintiffs’ injuries between those caused by the first collision, by the alleged design defect, and by another source, if any.
In summary, the Fox-Mitchell approach and subsequent burden shifting is necessary only where the plaintiffs’ injuries are indivisible. See, e.g., Mitchell, 669 F.2d at 1206 (death and paraplegic injury are indivisible). Whether the plaintiffs’ injuries are indivisible or divisible is a question of law for the trial judge. Mitchell, 669 F.2d at 1209. If the plaintiffs’ injuries are established or are stipulated to as indivisible, the plaintiffs must prove only “that a design defect was a substantial factor in producing damages over and above those which were probably caused as a result of the original impact or collision.” Polston, 423 S.E.2d at 662. If the plaintiffs meet this burden, the burden of proof shifts to the defendants to apportion the damages between them. See id.
Remanded.
DALIANIS, J., concurred; GROFF, MANGONES, and MOHL, JJ., superior court justices, specially assigned under RSA 490:3, concurred.