The opinion of the court was delivered by
D’ANNUNZIO, JAD.
Plaintiffs1 appeal from a judgment entered on a jury determination that defendant Chrysler Corporation’s vehicle was not defective. A judgment entered against co-defendant Joseph A. Lupo in the amount of $2,750,000 has not been appealed.
[210]*210On March 1,1991, plaintiff was alone, driving home in her 1987 Jeep Grand Wagoneer on Route 208. Plaintiffs vehicle was proceeding in the middle lane of the three northbound lanes when the engine stopped running. Plaintiffs vehicle eventually came to rest in the extreme right-hand lane against the curb. There was no shoulder adjacent to the traveled lanes at this point of the highway and, because of the curb, plaintiff could move no farther to the right. ’
Several vehicles proceeding in the right-hand lane slowed as they approached plaintiffs stopped vehicle and eventually drove around her. A good Samaritan, Carolyn Welch, stopped behind plaintiffs vehicle to see if she could be of assistance. Plaintiff and Welch emerged from their vehicles and walked towards each •other. At that time, co-defendant Joseph Lupo was operating a vehicle in the right-hand lane and was unable to stop. He struck the Welch vehicle in the rear, pushing it into plaintiff and crushing her legs between the front of the Welch vehicle and the rear of plaintiffs vehicle. Plaintiff sustained very serious injuries.
Plaintiffs’ claims against Joseph Lupo and the Chrysler Corporation were tried to a jury. As previously indicated, judgment was entered against Joseph Lupo based on the jury’s determination that he had been negligent in the operation of his vehicle and that his negligence was a proximate cause of plaintiffs injuries. This appeal concerns only plaintiffs’ claim against Chrysler Corporation..
The engine stalled because an overheated check valve burned through a convoluted conduit; a conduit is a covering which protects an engine wire harness consisting of a bundle of wires. After burning through the conduit, the valve also burned through the insulation covering the wires, causing the electrical system to short circuit. No one disputes this sequence of events or that these events were the cause of the engine’s failure.
The electrical harness ran from the front to the rear of the engine along the engine’s right2 bank. As previously indicated, it [211]*211was protected by a nylon conduit which fit over the harness like a sleeve. In the vicinity of the check valve, the conduit was secured through the use of “L” brackets and rosebud clips. The “L” brackets were attached to the engine’s valve cover approximately fourteen to fifteen inches apart. A rosebud clip was attached to each bracket, and the conduit ran through the circular rosebud clips. These clips were approximately fifteen inches apart. As designed by Chrysler, the conduit at the rear of the engine was to be located on the inboard side of the “L” bracket. On the date of plaintiff’s injury, the rear clip and the conduit were located outboard of the “L” bracket. The outboard location brought the conduit closer to the cheek valve than it would have been had the rear clip been mounted inboard.
The engine had two check valves, one on the right side and one on the left side. These valves were part of the engine’s emission control system. They permitted air to flow into the exhaust system through the top of the valve. However, the valve contained a diaphragm which prevented hot exhaust gases from flowing into the top of the valve, i.e., in the opposite direction from the airflow. It was uncontroverted that at the time of plaintiffs injury neither valve was functioning properly because both valves were badly corroded. Accordingly, hot exhaust gases were able to make their way into the top of the valve causing it to overheat. Plaintiff, however, did not contend that the check valves were defectively designed or defectively manufactured. No expert testified to that effect. Although all experts who testified agreed that the valves were not functioning, they did not know why the valves had corroded.
Plaintiffs contended that the harness/conduit had been defectively manufactured and defectively designed. As previously indicated, the rear rosebud clip was found after the accident to have been located outboard of the “L” bracket rather than inboard. Chrysler’s design called for it to be located inboard. Gilbert Wray, plaintiffs’ expert witness, testified that the mislocation of the conduit was a manufacturing defect because it had not been [212]*212assembled according to Chrysler’s design. Wray’s opinion in this regard rested on his inference that a Chrysler employee had erroneously placed the conduit outboard of the engine at the time the vehicle had been assembled by Chrysler.
Wray also testified that Chrysler’s failure to use a third clip to assure that the conduit would not come into contact with the check valve was .a design defect. It appears from the evidence that a properly functioning check valve would not have burned through the conduit even if the valve and the conduit had come into contact with each other; a properly functioning check valve would not have generated the temperature necessary to degrade the conduit material. Wray testified, however, that failure of a check valve was a reasonably foreseeable event that Chrysler should have anticipated and protected against by utilizing a third clip. Wray also opined that the conduit was defective because its melting temperature was not high enough in light of the foreseeability of an overheated check valve.
Michael Cassidy, a Chrysler employee, testified for Chrysler. Cassidy had been involved in the harness design and testified that the harness routing utilized on plaintiffs’ vehicle had been used since 1974. Cassidy had been responsible for durability tests of the harness in 1972 and 1973. He had inspected plaintiffs’ vehicle in preparation for trial. Chrysler established through his testimony that substantial work had been done under the hood of plaintiffs’ vehicle subsequent to its sale by Chrysler.
There had been a substantial modification of the battery mounting frame, and a ground wire had been added to the right fender splash apron. Cassidy noted a number of changes in the heater hose routing and that the radiator top had been re-soldered. Additionally, the drive belt which ran the air pump had been changed. The air pump fed air into the top of the check valves. If the air pump had not been functioning because of a broken belt, then the check valves would not have received the required airflow.
[213]*213Finally, Cassidy testified that a new distributor cap and new ignition wiring had been installed. Installation of new ignition wiring was significant in this case because the ignition wiring for the right bank of the engine had to be installed under the conduit. According to Cassidy, the easiest way to install new ignition wiring on the right side of the engine would be to detach the conduit, install the new wiring, and then reattach the conduit. Cassidy also stated that the routing of the new ignition wires deviated from the design routing. As designed, the wires would run in front of the valve, but on plaintiffs’ vehicle, all the ignition wires ran behind the valve.
Cassidy stated that the engine harness conduit is installed early in the assembly process, before the engine is placed into the vehicle’s engine compartment. According to Cassidy, the harness is easy to see when installed and, therefore, any deviation in its attachment from the place of attachment as designed, would have been easily noted and corrected.
James Wishart testified as an expert for Chrysler. He has a Masters Degree in electrical engineering and had been employed for over thirty years as an engineer by the Cadillac Division of General Motors. Wishart testified that the conduit used by Chrysler was the highest grade available and was rated for 250° of continuous duty with intermittent excursions to 300°. Its melting point was 410°. Contrary to Wray’s testimony, Wishart stated that protection to 600° was not necessary because the wiring is kept “high and cool,” i.e., that the underhood temperature at the top of the engine is kept between 200° and 250°, whereas the temperature at the bottom of the engine, where the exhaust manifold is kept, can rise to 800°. According to Wishart, even if the check valves overheated, the harness would not have melted if it had been maintained in its designed position. The maximum temperature of a properly mounted harness would not exceed 215°, a temperature which would have no effect on the conduit.
Wishart testified that the attachment, routing and location of the conduit was the same across the industry and that the rosebud [214]*214clips were typical in the industry. According to Wishart, two rosebud clips adequately controlled the harness, and a third clip was not necessary in a fifteen inch span.
Wishart opined that the harness location was not defective, relying in part on internal test records, proving grounds and road test records, and warranty records. According to Wishart, these records showed no indication of difficulty with the harness/conduit.
Finally, Wishart testified on direct that the conduit was resting against the overheated valve because it had been misplaced by someone during repairs. However, on cross-examination, Wishart admitted that he did not know when the harness misrouting had occurred and that he did not know the location of the conduit when it left the factory. Wishart agreed that if it had left the factory with the harness against the valve, it would have been improper.
Wishart also testified on cross-examination that a modification of the routing of the heater hose had occurred and that this had been a significant factor in the shutdown because “it forced the engine harness downward ... toward the side of the flange on the valve.” Regarding the replacement of ignition wires, Wishart testified that there was sufficient room under the harness to install those wires without moving the conduit. He stated that was the way that he would do it and that it was not necessary to remove the conduit to install the ignition wires, thereby contradicting Cassidy. Wishart also noted, however, that the rosebud clip appeared to have been disturbed when he examined it. It appeared to have some of the dust removed. Wishart agreed that it was foreseeable that at some point in its life a check valve would cease to function.
Plaintiffs contend that the trial court erred when it permitted Chrysler, over plaintiffs’ objection, to introduce evidence that the National Highway Traffic Safety Administration (NHTSA), see 49 U.S.C.A. § 105, had not opened an investigation regarding any alleged defect in the harness/conduit. Subsequent to the opening statement of Chrysler’s counsel, plaintiffs moved for a mistrial on [215]*215the ground that counsel had told the jury about the lack of a NHTSA investigation. The court denied the motion. Thereafter, during re-cross examination, Chrysler’s lawyer was permitted to ask Wray, over plaintiffs’ objection, if he had inquired of NHTSA “to see if they were conducting any investigation into failures that would result from this ‘defect.’ ” Wray responded that he had called NHTSA regarding this problem and that NHTSA was not involved in any such investigation.
Cassidy testified on direct examination regarding the absence of reports regarding problems with the harness. He stated that 50,000 mile durability tests of Jeep Wagoneers yielded no reports of problems regarding the routing, location or attachment of the harness. These durability tests are performed every year and the routing, location and attachment of the harness have been the same since 1974.
According to Cassidy, warranty repair reports and dealer feedback are additional sources of information regarding problems or potential problems. Neither source reported harness problems.
Cassidy testified, over plaintiffs’ objection, that consumer complaints to NHTSA were a third source of information regarding functional problems. Cassidy stated that he “reviewed the records that he had from NHTSA and there is no information in those records that depict any problems with the engine harness and the senior Jeep vehicle.” Cassidy identified exhibit DC-4 as “a list of all those [NHTSA] investigations that were open based on the complaints that were made to NHTSA.” (Emphasis added.) Cassidy testified that there were “no complaints ... on that vehicle and wiring harness at all.” Although he used the word “complaints,” it appears from the context and his description of DC4i that Cassidy was referring to investigations. Exhibit DC-4 was marked for identification but not offered into evidence.
Regarding NHTSA procedure, Cassidy testified that if NHTSA gets “enough” complaints it opens an investigation and asks the manufacturer for information relevant to the problem. Cassidy, however, did not know how many complaints were “enough.” [216]*216Plaintiffs’ counsel cross-examined Cassidy about NHTSA records and established that as late as 1994, NHTSA had no record of the 1991 harness failure that caused plaintiffs’ engine to quit.
Plaintiffs argue that the testimony regarding NHTSA records was inadmissible because it was hearsay and because its capacity for prejudice outweighed its probative value.
The jury had to apply the risk-utility analysis in evaluating plaintiffs’ claim of a design defect. O'Brien v. Muskin Corp., 94 N.J. 169, 183, 463 A.2d 298 (1983); Cepeda v. Cumberland Engineering Co., Inc., 76 N.J. 152, 173-74, 386 A.2d 816 (1978), overruled on other grounds, Suter v. San Angelo Foundry & Mach. Co., 81 N.J. 150, 406 A.2d 140 (1979). The second risk-utility factor is “the likelihood that [the product] will cause injury, and the probable seriousness of the injury.” Id. at 174, 406 A.2d 140. Our Supreme Court has noted that “[e]vidence of prior similar accidents is relevant and should be admissible as evidence of the risk, or lack thereof, of a product.” Ryan v. KDI Sylvan Pools, Inc., 121 N.J. 276, 290, 579 A.2d 1241 (1990) (hereafter Ryan).
Ryan involved an action for damages for personal injuries resulting from a dive into a residential swimming pool. The plaintiffs product liability claim against KDI, the pool manufacturer, was based on an alleged design defect and inadequate warning. KDI appealed from an adverse judgment, contending that the trial court erred in excluding the testimony of KDI’s expert.
In Ryan, plaintiff contended that the pool, equipped with a diving board, was unsafe for diving in light of the pool’s depth and configuration. He also contended that the pool should have had depth markers or warnings. In support of these contentions, plaintiffs expert testified that depth standards established by the pool industry’s trade association, NSPI, were inadequate.
In response, KDI sought to introduce the testimony of its expert, Joseph Schmerler, regarding the low incidence of residential swimming pool injuries and, in particular, the low incidence of [217]*217cervical spine injuries “over a seventeen-year period in pools with diving boards meeting NSPI 1974 standards.” Id. at 281, 579 A.2d 1241. Schmerler had developed this information from a variety of sources. After a non-jury hearing, the trial court precluded this testimony.
The Supreme Court reversed the judgment and remanded for a new trial based on the exclusion of Schmerler’s testimony. The Court addressed the relevance and probative value of a product’s accident history and ruled that such evidence is admissible. The Court stated:
Evidence of prior similar accidents is relevant and should be admissible as evidence of the risk, or lack thereof, of a product. We have found several factors relevant in a risk-utility analysis, among them “[t]he safety aspects of the producte-the likelihood that it will cause injury, and the probable seriousness of the injury.” O’Brien v. Muskin, 94 N.J. 169, 182, 463 A.2d 298 (1983). Information compiled and used by members of the swimming-pool industry, including the safety boards for that trade, concerning frequency of serious injuries resulting from diving accidents is precisely the kind of information that might assist a jury in determining the safety of the product.
[Id. at 290, 579 A.2d 1241.]
The Court also discussed the foundation required for the admission of survey information regarding accident history and frequency. In this regard, the Court analyzed Evid. Rule 56(2), now N.J.R.E. 703, which permitted an expert witness’ reliance on facts or data not admissible in evidence “[i]f of a type reasonably relied upon by experts in the particular field in forming opinions or inferences upon the subject.” Evid. Rule 56(2); N.J.R.E. 703. The Court stated:
We interpret Evidence Rule 56(2) to require that a court make an inquiry into and a finding on whether experts in the given field rely on certain information. If such reliance be found, then it is presumed to be reasonable.
[Ryan, supra, 121 N.J. at 289, 579 A.2d 1241.]
In the present case, there was no determination whether the lack of a NHTSA investigation regarding a particular defect, or the existence of such an investigation was a fact reasonably relied upon by experts in the particular field. Indeed, there was no evidence regarding that threshold question. In its brief, Chrysler argues that the jury could infer that NHTSA records [218]*218were “important” because plaintiff’s expert went to the trouble of calling NHTSA and inquiring about the harness problem. We are persuaded that this permissible inference is an inadequate substitute for the threshold determination required in Ryan. This is especially so in light of Wray’s testimony that his call to NHTSA had no impact on his opinion that the harness/conduit was defective.
Our primary concern regarding the NHTSA evidence is not based on the rule prohibiting hearsay, but rather, on the evidence’s probative value when compared with its capacity for prejudice. As previously indicated, the Supreme Court in Ryan held that a product’s accident history is relevant to application of the risk-utility analysis. Ryan involved the proffered testimony of an expert who was going to provide specific information regarding the small number of residential swimming pool injuries and cervical spine injuries, to establish the minimal risk involved with residential swimming pools. In the present case, however, the court permitted evidence of the lack of a NHTSA investigation regarding Chrysler’s electrical harness. The purpose of this evidence was to permit a jury to infer that the lack of an investigation meant that there had been no complaints to NHTSA, or a low incidence of complaints, and to further infer from the absence of complaints that there had been no or few problems with the harness/conduit. We are persuaded that the jury may have drawn such inferences. We are also persuaded that the information given to the jury was inadequate to permit it to draw these inferences.
We emphasize that the testimony of Wray and Cassidy was with regard to investigations, not complaints. The lack of an investigation does not necessarily mean that there had been no complaints. The jury was not informed adequately regarding NHTSA’s criteria for mounting an investigation. Cassidy alluded to a criteria based on a number of complaints, but he obviously did not know NHTSA’s specific requirements for instituting an investigation. Similarly, there was no information given to the jury regarding [219]*219the source of “complaints” to NHTSA, or the likelihood that complaints would be made or acted upon by NHTSA, if similar harness failures had occurred but caused no injury or property damage. Moreover, the absence of an investigation could have been the result of NHTSA’s priorities in light of its available resources, but the jury had no information regarding NHTSA’s priorities.
In Cramer v. Kuhns, 213 A.D.2d 131, 630 N.Y.S.2d 128 (1995) the Appellate Division reversed a judgment for plaintiff and ordered a new trial because the trial court had admitted a NHTSA study regarding motorcycle side stands. The theory of plaintiffs case against Harley Davidson Motor Company, Inc. was that the side stand of the motorcycle on which plaintiff had been a passenger failed to retract after dropping down or being left down by the operator. As a result, the side stand struck the pavement causing the operator to lose control.
The Appellate Division considered the admissibility of the study under New York’s exception to the rule against hearsay which admits public documents if trustworthy and reliable. The court concluded that the study did not possess those attributes. The court stated:
Applying this analysis to the document before us, we are of the view that the NHTSA study does not fall within the scope of the public document exception contemplated by CPLR 4520. The study itself was preliminary in nature; no public findings were released and no recalls were issued. Additionally, the study was exceedingly brief in nature, and there was very little detail provided as to the actual tests conducted upon the various motorcycle models. Finally, the “observations” contained in the study were based, in part, upon the owner surveys and accident reports, neither of which were admissible, and such observations were presented in a most eonelusory fashion. Such deficiencies, coupled with the absence of testimony from anyone actually involved in the study, persuade us that the study and the accompanying videotape should not have been admitted into evidence.
[Id. at 131, 630 N.Y.S.2d 128 (footnote omitted).]
The court also noted that even if the study were admissible as a business record or public document, the study should not be admitted because of its limited probative value. “In view of the sweeping and eonelusory nature of the study, and in particular the [220]*220lack of detail provided regarding the actual testing conditions, we are unable to conclude that the study has any relevancy to the matter now before us.” Id. at 132, 630 N.Y.S.2d 128.
We conclude that the absence of a NHTSA investigation, standing alone, was inadequate to support the desired inferences. Without more information regarding NHTSA, its role, procedures, criteria and practices, the jury was unable to evaluate properly the significance of the absence of a NHTSA investigation. Cf. Johnson v. Ford Motor Co., 988 F.2d 573, 580 (5th Cir.1993) (ruling that “NHTSA letters regarding a preliminary inquiry which did not result in any action by the NHTSA” were properly excluded). We are also persuaded that admission of this evidence had the capacity to produce an unjust result. We agree with the district court’s observation in Fowler v. Firestone Tire & Rubber Co., 92 F.R.D. 1, 2 (N.D.Miss.1980), that because a NHTSA investigative report and a report of a committee of the House of Representatives, were
reports promulgated by agencies of the United States government, [their] apparent ‘official’ nature is likely to cause a jury to give the evidence inordinate weight and for this reason, any probative value the evidence might have would be far outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury.
[Ibid.]
Accord Johnson v. Ford Motor Co., supra, 988 F.2d at 580 (quoting Fowler). See City of New York v. Pullman, Inc., 662 F.2d 910, 915 (2d Cir.1981) (holding that the trial court did not err in excluding a report of the Urban Mass Transit Administration because “the report would have been presented to the jury in ‘an aura of special reliability and trustworthiness’ which would not have been commensurate with its actual reliability.”) (quoting United States v. Fosher, 590 F.2d 381, 383 (1st Cir.1979)), cert. denied, 454 U.S. 1164, 102 S.Ct. 1038, 71 L.Ed.2d 320 (1982).
The record demonstrates that Chrysler’s counsel considered the absence of a NHTSA investigation to be powerful evidence. In a summation which is reproduced on thirteen transcript pages, many of which are dedicated to the issue of proximate cause [221]*221rather than to the design defect issue, he refers to the absence of a government record regarding the harness failure on five of those pages, specifically mentioning NHTSA on three of those five pages.
Our concern regarding the potentially inordinate impact of the NHTSA evidence is heightened by the absence of any reference to this evidence in the jury instruction.3 Thus, the court did not inform the jury that the relevance of the NHTSA evidence was limited to only one element of the risk-utility analysis and was not relevant to plaintiffs’ claim of a manufacturing defect.
We conclude that admission of the testimony regarding NHTSA requires a new trial.
Plaintiffs also contend that the court erred in dismissing a claim by Cheryl Adelman’s daughters, Randi and Cori, for loss of their mother’s society and comfort. We affirm the dismissal of those claims because New Jersey does not recognize them as actionable. Russell v. Salem Transp. Co., 61 N.J. 502, 295 A.2d 862 (1972).
In light of our reversal of the judgment based on the NHTSA issue, we deem it unnecessary to address plaintiffs’ contention that the trial court had erred in instructing the jury that there was no direct evidence of a manufacturing defect.
Chrysler filed a cross-appeal, contending that the plaintiffs had failed to establish that any defect in its vehicle was a proximate cause of plaintiffs’ injuries. Chrysler withdrew its cross-appeal prior to oral argument in light of Yun v. Ford Motor Company, [222]*222143 N.J. 162, 669 A.2d 1378 (1996), rev’g on dissent, 276 N.J.Super. 142, 158, 647 A.2d 841 (App.Div.1994).
The judgment in favor of Chrysler is reversed, and the case is remanded for a new trial on the issue of Chrysler’s liability.