Opinion
CHRISTIAN, J.The State of California has appealed from a judgment for damages in the aggregate sum of $3,052,000 caused by an automobile accident on the Benicia-Martinez Bridge. The judgment is based on jury verdicts finding the state responsible for the wrongful deaths of two victims of the accident and for personal injuries suffered by six other victims. The jury assessed damages against the state on two independent theories: negligence in permitting the driver of the other vehicle to be driving while away from his place of residence at the California Veterans Home at Yountville, and maintenance of the Benicia-Martinez Bridge in a dangerous condition. We affirm the judgment.
At about 4:30 in the afternoon on Sunday, October 26, 1970, respondents were riding in a 1967 Chevrolet pickup with a camper shell. Respondents were then 11 to 13 years of age and were part of a group of Girl Scouts returning to their homes in the San Jose-Santa Clara area from an outing in the Napa Valley. The pickup was being driven by Carl L. Harrington, father of two of the girls.
*480While proceeding south across the Benicia-Martinez Bridge, the Harrington camper was struck head-on by a 1965 Ford station wagon driven by Author Edgmon, who had been returning to his place of residence at the California Veterans Home in Yountville. Edgmon’s automobile had struck the guardrail on the bridge at the right margin of the northbound lanes and was propelled across the center median strip and into the southbound side of the highway; there it collided with the Harrington vehicle, which was entirely on its own side of the highway. Both vehicles caught fire, and were burning wildly when highway patrol officers arrived. Author Edgmon and Mr. and Mrs. Harrington all died in the flaming crash. In addition to sustaining cuts and broken bones, the young girls were severely burned.
Appellant contends that the staff at the Veterans Home did not have a duty to prevent Author Edgmon from driving his automobile. Absent a special relationship, there is ordinarily no duty to control the conduct of a third person so as to prevent him from causing harm to another. (Tarasoff v. Regents of University of California (1976) 17 Cal.3d 425, 435-437 [131 Cal.Rptr. 14, 551 P.2d 334]; Weirum v. RKO General, Inc. (1975) 15 Cal.3d 40, 48-49 [123 Cal.Rptr. 468, 539 P.2d 36]; Wright v. Arcade School Dist. (1964) 230 Cal.App.2d 272, 277 [40 Cal.Rptr. 812]; Rest.2d Torts (1965) § 315.)1 It is respondents’ contention that a special relationship existed in this case between the State of California and Author Edgmon giving rise to a duty of care on the part of employees at the Veterans Home. (See Tarasoff v. Regents of University of California, supra, 17 Cal.3d 425, 435-437; Poncher v. Brackett (1966) 246 Cal.App.2d 769, 772-773 [55 Cal.Rptr. 59]; Rest.2d Torts (1965) § 319.)2 Respondents point out that at all times during which Author Edgmon resided at the Veterans Home (Home) he wassubjectto arule (adopted by the commandant of the Home under authority of Cal. Admin. Code, tit. 12, § 502) that no resident could leave or remain away from the premises without a written pass. Edgmon kept an automobile on the Home premises which he sometimes drove on weekends to Stanislaus County, where his wife *481resided. Edgmon was returning from such a trip, with a written pass, when the accident occurred.
There was evidence that a special relationship did exist between the responsible authorities at the California Veterans Home and Edgmon. The state provides hospitalization or domiciliary care at the Home for necessitous veterans. Edgmon was incapable of sustaining himself outside the Home due to his physical and mental disabilities. Residents of the Home are subject to some control by the staff, as evidenced by published regulations (see Mil. & Vet. Code, §§ 1023, 1044; Cal. Admin. Code, tit. 12, §§ 502-508) and as reflected in the contractual agreement signed by entering veterans as a condition of admission. Any resident of the Home who left without permission was subject to discipline including discharge from the Home. A resident could obtain permission to park his vehicle on the Home premises; Author Edgmon was given such permission.
Members of the staff had on prior occasions prevented residents from driving “[i]f there is a question about an old stroke and a possibility of seizure or a blackout or hypertension over certain ages, . . .” However, at the time of the events in question, there was apparently no procedure in effect for determining whether residents remained capable of driving; there was no communication between the medical and administrative staffs on this matter, and passes were issued as a clerical matter unless the resident’s history card showed some restriction.
Author Edgmon was admitted to the California Veterans Home on January 29, 1968, having come from the federal Veterans Administration Hospital in Palo Alto. He remained at Yountville from 1968 until he died, except for a period of four and a half months from April until mid-August of 1969 when he was returned to the Palo Alto V.A. Hospital. His medical history showed that he had suffered a head injury in 1942, probably causing meningeal adhesions to the brain. He had a long history of headaches of an incapaciting intensity, and suffered frequently from dizziness, nausea, what he described as “blackouts,” blurred vision and diplopia or double vision.
Edgmon was regularly administered four drugs at the Home: (1) Dilatin, an anticonvulsant drug primarily used in the treatment of epilepsy; (2) Darvon, a mild analgesic structurally related to the narcotic analgesic, Methadone; (3) Prolixin, used in the management of schizophrenia and manifestations of psychotic disorders; and (4) Taractan, also *482used for control of schizophrenic manifestations. All four drugs are central nervous system depressants. When taken singly, these drugs have a sedative effect and can impair motor functions. Three of the four drugs regularly administered to Edgmon had specific indications of potential danger in operating motor vehicles. The fourth drug, Dilatin, has even more effect on the central nervous system.
There was thus evidence from which the jury could reasonably conclude that employees of the Veterans Home had notice or knowledge of facts suggesting that Edgmon’s driving presented a risk of harm to himself and others. But the risk was not one that could be guarded against by warning potential victims (cf. Tarasoff v. Regents of University of California, supra, 17 Cal.3d 425). There was uncontradicted evidence supporting an inference that a physician at the Home had warned Edgmon that he should not drive when he was under medication. Edgmon was not under guardianship, and it could not be maintained that by becoming a resident at the Home he had surrendered his civil rights and his responsibility for his own conduct. There is no statute or regulation purporting to authorize the commandant of the Home to order a resident veteran not to drive a car. Apparently the only authoritative procedure available to prevent a veteran from driving his car would be expulsion from the Home for violation of such an order. No lawful basis has been shown for the exercise of such paternalistic powers over a citizen, in the absence of an adjudication of incompetency. Even if such power existed, its exercise would be a discretionary act for which the state has not subjected itself to tort liability. (Gov. Code, § 818.4; Papelian v. State of California ex rel. Dept. of Motor Vehicles (1976) 65 Cal.App.3d 958 [135 Cal.Rptr. 665].) We conclude that the judgment cannot be sustained on respondents’ theory that the state should have prevented Edgmon from driving an automobile.
Appellant contends that there is no substantial evidence that a dangerous condition of the Penicia-Martinez Bridge was a proximate cause of Edgmon’s collision with the Harrington camper.
A reviewing court must view the evidence in the light most favorable to the party prevailing below. (Bakity v. County of Riverside (1970) 12 Cal.App.3d 24, 28 [90 Cal.Rptr. 541].) The weight of the evidence and the credibility of the witnesses are matters for the trier of fact and not for the appellate court. (Crawford v. Southern Pacific Co. (1935) 3 Cal.2d 427, 429 [45 P.2d 183].) “The test is not whether there is substantial conflict in evidence but whether there is substantial evidence in favor of *483the respondent.” (McKinley v. Buchanan (1959) 176 Cal.App.2d 608, 611-612 [1 Cal.Rptr. 573]; Crogan v. Metz (1956) 47 Cal.2d 398, 404 [303 P.2d 1029].)
Under the Tort Claims Act of 1963 (Gov. Code, §§ 830-840.6), a plaintiff must establish by a preponderance of the evidence the elements required by section 835 of the Government Code in order to establish liability of a public entity for a dangerous condition. Section 835 provides:
“Except as provided by statute, a public entity is liable for injury caused by a dangerous condition of its property if the plaintiff establishes that the property was in a dangerous condition at the time of the injury, that the injury was proximately caused by the dangerous condition, that the dangerous condition created a reasonably foreseeable risk of the kind of injury which was incurred, and that either:
“(a) A negligent or wrongful act or omission of an employee of the public entity within the scope of his employment created the dangerous condition; or
“(b) The public entity had actual or constructive notice of the dangerous condition under Section 835.2 a sufficient time prior to the injury to have taken measures to protect against the dangerous condition.” (See Hayes v. State of California (1974) 11 Cal.3d 469, 471 [113 Cal.Rptr. 599, 521 P.2d 855]; Baldwin v. State of California (1972) 6 Cal.3d 424, 427 [99 Cal.Rptr. 145, 491 P.2d 1121]; Sykes v. County of Marin (1974) 43 Cal.App.3d 158, 160-161 [117 Cal.Rptr. 466]; Callahan v. City and County of San Francisco (1971) 15 Cal.App.3d 374, 378 [93 Cal.Rptr. 122]; Bakity v. County of Riverside, supra, 12 Cal.App.3d 24, 29-30; Feingold v. County of Los Angeles (1967) 254 Cal.App.2d 622, 625 [62 Cal.Rptr. 396].)
A “dangerous condition” of public property means a condition of the property which “creates a substantial . . . risk of injury when such property ... is used with due care in a manner in which it is reasonably foreseeable that it will be used.” (Gov. Code, § 830; Bakity v. County of Riverside, supra, 12 Cal.App.3d 24, 30.) Government Code section 830 requires that the physical condition of the public property be in a *484dangerous or defective condition involving reasonably foreseeable risk to the public. (Sykes v. County of Marin, supra, 43 Cal.App.3d 158, 161; see Murrell v. State of California ex rel. Dept. Pub. Wks. (1975) 47 Cal.App.3d 264, 267 [120 Cal.Rptr. 812].) Whether a given set of circumstances creates a dangerous or defective condition is primarily a question of fact. (Fackrell v. City of San Diego (1945) 26 Cal.2d 196, 206 [157 P.2d 625, 158 A.L.R. 773]; Granone v. County of Los Angeles (1965) 231 Cal.App.2d 629, 651 [42 Cal.Rptr. 34]; Briggs v. State of California (1971) 14 Cal.App.3d 489, 496 [92 Cal.Rptr. 433]; Gardner v. City of San Jose (1967) 248 Cal.App.2d 798, 804-805 [57 Cal.Rptr. 176].) As the court stated in De La Rosa v. City of San Bernardino (1971) 16 Cal.App.3d 739, 745 [94 Cal.Rptr. 175], “The existence of a dangerous condition is usually a question of fact and may be resolved as a question of law only if reasonable minds can come to but one conclusion.” (Gov. Code, § 830.2; Callahan v. City and County of San Francisco, supra, 15 Cal.App.3d 374, 379; Bakity v. County of Riverside, supra, 12 Cal.App.3d 24, 30.)
Although the evidence was conflicting, there was evidence from which the jury could reasonably have found the existence of a dangerous condition. The bridge connects three freeways. Respondents’ expert witness testified that in light of the high speed and volume of traffic on the bridge, the bridge should have been but was not built according to freeway standards.3 The south end of the bridge was introduced by a superelevated S-curve. The accident in fact occurred on the curve. There was expert testimony that the use of a curve on the bridge itself was dangerous and contraiy to sound engineering practice.
*485There was also evidence that, because of the height of the bridge, the high volume of traffic and frequent gusts of high wind on the bridge, the shoulder and median areas were too narrow for safe travel. Respondents’ expert testified that the height of the bridge roadway is an intimidating factor to some drivers and that wider shoulders aid persons who fear driving in the air near railings. Moreover, narrow shoulders and median are dangerous because many drivers move out of their normal lane or change speed abruptly where there is constriction, either actual or apparent, in the roadway ahead. There was also expert testimony that the 65-mile-per-hour posted speed limit was too high in view of the other hazards of the bridge. In 1967 or 1968, the Highway Patrol had recommended that the speed limit on the bridge be reduced to 50 or 55 miles per hour. There was evidence that a deflecting guardrail and the lack of a cross-median barrier were additional hazards, in view of the traffic volume on the bridge. The Highway Patrol had through official channels pointed out the hazards and requested the Division of Highways to install a median barrier. Thus, there was evidence that a dangerous condition existed and that the state had notice of that condition. (See Cameron v. State of California, supra, 7 Cal.3d 318, 323-324; Baldwin v. State of California, supra, 6 Cal.3d 424, 428.)
Appellant contends that there is no substantial evidence in the record to support the conclusion that the dangerous condition of the bridge was a proximate cause of the injury. It should be noted that the fact that a third person may negligently have used the highway would not necessarily exonerate the state. (See Bakity v. County of Riverside, supra, 12 Cal.App.3d 24, 32; Murrell v. State of California ex rel. Dept. Pub. Wks., supra, 47 Cal.App.3d 264, 271-272.) The state may be held liable if its negligence in maintaining dangerous property and the negligence of another party concur as proximate causes of the injury. (Murrell, supra, 47 Cal.App.3d at p. 267; see Hayes v. State of California, supra, 11 Cal.3d 469, 472; Baldwin v. State of California, supra, 6 Cal.3d 424, 428, fn. 3.) This was a question of fact for the jury (Bakity v. County of Riverside, supra, 12 Cal.3d at p. 32).
Edgmon’s car was deflected by the guardrail across a roadway which was not protected by a median barrier. The metal guardrail installed on the sides of the bridge was of a design that projected the Edgmon vehicle across the highway into the path of oncoming traffic. A state report on the collision stated that: “The action of October 25th, 1970, might possibly have been avoided, if the metal beam barrier first hit by vehicle No. 1 had been of a different design, so that it did not rebound into *486traffic.” Thus, the jury could reasonably have inferred that respondents’ maintenance of a dangerous condition was a proximate and contributing cause of the accident.
Appellant contends that the trial court erroneously denied its request to give certain instructions designed to advise the jury on the element of due care contained in the definition of “dangerous condition.” (See Gov. Code, § 830, subd. (a).) This contention is without merit.
“Jury instructions are sufficient which in composite supply the jury with a well-balanced statement of the necessary legal principles. (City of Los Angeles v. Frew [1939] 139 Cal.App.2d 859, 872 [294 P.2d 1073].)” (Murrell v. State of California ex rel. Dept. Pub. Wks., supra, 47 Cal.App.3d 264, 270; see Deaile v. General Tel. Co. of California (1974) 40 Cal.App.3d 841, 853 [115 Cal.Rptr. 582]; Fuller v. State of California (1975) 51 Cal.App.3d 926, 944 [125 Cal.Rptr. 586]; 4 Witkin, Cal. Procedure, Trial, § 193.), In this case, the instructions of the trial court adequately and correctly set forth the applicable law with regard to the nature and meaning of a dangerous condition.4 “The trial court is not required to give every instruction offered by a litigant nor is a party entitled to have the substance of instructions given by the court repeated in different language. (Thompson v. Keckler (1964) 228 Cal.App.2d 199, 212 [39 Cal.Rptr. 267].)” (Fuller v. State of California, supra, 51 Cal.App.3d 926, 945; see Deaile v. General Tel. Co. of California, supra, 40 Cal.App.3d 841, 853.) The instructions given here adequately informed the jury of the necessary legal principles.
*487Appellant contends that the trial court erred in denying its motion to have the degree of culpability of General Motors determined.On March 19, 1973, appellant filed cross-complaints for implied equitable indemnity against General Motors Corporation, the manufacturer of the Harrington vehicle, alleging negligent design of the fuel system. Demurrers by General Motors to the cross-complaints were sustained without leave to amend. Judgments were entered on September 24, 1973, dismissing the cross-complaints. Appellant did not appeal from those judgments.
The consolidated cases were set for trial on- June 2, 1975. After the pretrial conference on April 17, 1975, the state filed a motion for orders requiring joinder of General Motors as a party defendant, for leave to file new cross-complaints, and for leave to amend its answer to allege concurrent negligence of General Motors. This relief was sought on the basis of the decision in Li v. Yellow Cab Co. (1975) 13 Cal.3d 804 [119 Cal.Rptr. 858, 532 P.2d 1226, 78 A.L.R.3d 393], which adopted the rule of “pure” comparative negligence and abolished the complete defense of contributory negligence in this state. The motions were denied. Appellant complains of those rulings, arguing that the logical extension of Li *488and of the “pure” comparative negligence system is that the doctrine of “joint and several” liability is no longer the law of this state. But the motions were properly denied on the ground of res judicata because the judgment dismissing the cross-complaint against General Motors was a final determination of the rights asserted by the cross-complainant and the cross-defendant. Since the state did not appeal, the judgment dismissing the cross-complaint became final and the rule of res judicata applies. Moreover, the issues raised by the state’s cross-complaints were determined on September 24, 1973, when judgments of dismissal were entered after the demurrers had been sustained without leave to amend; therefore, Li, which has prospective effect commencing May 1, 1975, is not applicable. (See Rose v. International Brotherhood of Electrical Workers (1976) 58 Cal.App.3d 276, 279 [129 Cal.Rptr. 736].)
Appellant moved for a nonsuit and for the entiy of judgment notwithstanding the verdict at appropriate points in the proceedings below. Both motions were in connection with the causes of action involving the bridge and were based upon the ground that there was no evidence that the bridge was in a dangerous condition, or that any condition of the bridge was a proximate cause of respondents’ injuries. Appellant contends that the trial court erred in denying these motions. As previously discussed, there was substantial evidence to support a jury verdict in favor of respondents; the trial court did not err in denying the motions.
The judgment and the order denying judgment notwithstanding the verdict are affirmed.
Caldecott, P. J., concurred.
Restatement Second of Torts section 315: “There is no duty so to control the conduct of a third person as to prevent him from causing physical harm to another unless
“(a) a special relation exists between the actor and the third person which imposes a duty upon the actor to control the third person’s conduct, or
“(b) a special relation exists between the actor and the other which gives to the other a right to protection."
Restatement Second of Torts section 319: “One who takes charge of a third person whom he knows or should know to be likely to cause bodily harm to others if not controlled is under a duty to exercise reasonable care to control the third person to prevent him from doing such harm.”
Under section 830.6 of the Government Code, the state may be shielded from tort liability for a dangerous condition of public property by the defense of “design immunity” where the state proves: (1) discretionary approval of the plan prior to construction, (2) a- causal relationship between the plan and the accident, and (3) the reasonableness of the design at the time it was approved. (Anderson v. City of Thousand Oaks (1976) 65 Cal.App.3d 82, 88-89 [135 Cal.Rptr. 127]; see Cameron v. State of California (1972) 7 Cal.3d 318, 325 [102 Cal.Rptr. 305, 497 P.2d 777].) If the state proves in a given case that the defense of design immunity should initially attach, a further issue remains. The immunity provided by the statute is not perpetual, and the defense of design immunity can be lost as the result of changed conditions. (Baldwin v. State of California, supra, 6 Cal.3d 424, 431, 434-435, 438.)
In the present case, although the state pleaded the defense of design immunity in its answer, it apparently did not rely on the defense at trial. Evidence was introduced at trial that the bridge was dangerous and defective when it was built. Moreover, there was substantial evidence that traffic and other conditions had materially changed in the eight years between the opening of the bridge and the date the accident occurred, and the state had notice of these changed circumstances. “Once the entity has notice that the plan or design, under changed physical conditions, has produced a dangerous condition of public property, it must act reasonably to correct or alleviate the hazard.” (Baldwin v. State of California, supra, 6 Cal.3d 424, 434.)
In Murrell v. State of California ex rel. Dept. Pub. Wks., supra, 47 Cal.App.3d 264, 269-270, the court stated:
“None of the jury instructions specifically explained the used with due care clause of section 830, subdivision (a), in terms of the difference between two kinds of use—general public use and use at the time and place of the accident. As construed, Government Code section 830, subdivision (a), refers to the former kind of use, not the latter; it does so only in veiled terms; thus an instruction which simply repeats the verbiage of the statute tends to veil the statute’s true meaning from the jury. Where, as here, the concurrent negligence of a third party is a focal issue, the jury should be told expressly what the statute only implies.
“None of the trial court’s instructions—and none of the BAJI instructions—was designed to meet that need directly. A set of instructions would be erroneous which gave the jury the impression that a third party’s negligent use would negate existence of a ‘dangerous condition’ and exonerate the public entity from liability. (Callahan v. City and County of San Francisco, supra.) Had the trial judge supplied the jury with an instruction distinctly explaining the used with due care clause, the instructions would have been improved. Nevertheless, the trial court cannot be charged with error at this point because the instructions, in composite, succeeded in conveying the requisite information to the jury.”
In the present case, the trial court’s instructions with regard to the “used with due *487care” clause of section 830, subdivision (a), directly met the “need” specified in Murrell. The trial court instructed the jury that: “A dangerous condition, as that term is used in these instructions, means a condition of property that creates a substantial risk of injury when such property or adjacent property is used with due care in a manner in which it is reasonably foreseeable that it will be used.
“A public entity is not an insurer of the safety of persons using its property. It is not required to maintain its property free from every defect that might possibly lead to injury.
“The phrase ‘used with due care’ in the definition of a dangerous condition has reference to whether the condition would result in injuries when used with due care by the public generally. That expression does not refer to the care used in connection with this particular accident. The plaintiffs are not required to prove that the bridge was being used with due care by any persons involved in the accident. On the issue of whether a dangerous condition existed, the plaintiffs are required to establish only that the condition created a substantial risk of harm when used with due care by the public generally in a reasonably foreseeable manner. If the condition was dangerous in that sense, then the State is subject to liability for injuries proximately caused by any such condition, even though a third person did not use due care in connection with this particular accident.
“A public entity is not liable for injury caused by a dangerous condition of its property created by an act or omission of a public employee if the public entity establishes that the act or omission that created the dangerous condition was reasonable.
“In determining whether the act or omission which created a dangerous condition was reasonable, you shall weigh the probability and gravity of potential injury to persons the defendant should have foreseen might be injured by the condition against the practicability and cost of having taken alternative action that would not have created the risk of injury or would have protected against it.”