Hughes v. Blue Cross of Northern California

215 Cal. App. 3d 832, 263 Cal. Rptr. 850, 1989 Cal. App. LEXIS 1174

• Court: California Court of Appeal
• Decided: November 14, 1989
• Docket: A032025
• Status: Published

Opinion

Opinion

NEWSOM, J.

This appeal arises from an action brought by Sally Hughes (hereafter Mrs. Hughes or respondent) and her estranged husband, Jurdy Hughes (hereafter Mr. Hughes), against Blue Cross of Northern California (hereafter Blue Cross) alleging several causes of action arising from the denial of insurance benefits for their son’s hospitalization. Upon petition of Blue Cross, the plaintiffs’ right to the disputed benefits was referred to arbitration and the arbitrator ordered payment. Mrs. Hughes then proceeded to trial to recover damages for Blue Cross’s alleged breach of its implied covenant of good faith and fair dealing. The jury rendered a special verdict awarding her $150,000 in compensatory damages and $700,000 in punitive damages. Although not covered by the Blue Cross insurance contract, Mr. Hughes also proceeded to trial on theories of breach of the implied covenant of good faith and fair dealing, unfair insurance claims practices, and negligent infliction of emotional distress. Finding that he had no standing to sue, the trial court directed a verdict against him and later entered a judgment of dismissal.

On September 6, 1981, Patrick Hughes, a young man 21 years old, took an overdose of about 30 aspirins and stabbed himself repeatedly in the abdomen with a screwdriver. After discovering the boy’s condition, his father and mother took him to the emergency ward of St. Luke’s Hospital. It was Patrick’s first episode of serious mental disorder. In the first grade and again in the eighth grade, he had briefly undergone psychological counseling after showing signs of nervousness. Later, he appeared to fare well in his high school years. Electing to live with his father so that he could attend St. Ignatius Preparatory School, he received good grades and seemed “to fit right in.” But he had difficulty coping after graduation and dropped out of a junior college. In recent months, he had been fired from his job and quarreled with his father and his girlfriend.

A psychiatrist, Dr. Hector De Lorente, arranged for Patrick’s transfer to the psychiatric ward of Mary’s Help Hospital where he remained for about six weeks. Dr. De Lorente found that Patrick was “extremely anxious” and “overwhelmed by images which had the contents of knives, axes as well as fire.” He did not respond well to treatment. After several weeks he confided to Dr. De Lorente a grandiose delusion characteristic of schizophrenia. In an excited condition, he explained that God’s eyes were within him. If he could witness the sunrise with his girlfriend, there would be less crime and less pressure for people in the world. On Dr. De Lorente’s recommendation, Patrick was taken on October 25th to Belmont Hills Psychiatric Center, an institution providing acute psychiatric care.

At Belmont Hills Patrick was put under the care of Dr. William E. Lofthouse. He continued to be “withdrawn, behaving inappropriately, was depressed, was talking about subjects like death and dying and violence.” In time Dr. Lofthouse also learned of his obsessive delusion about saving the world by witnessing the rising sun. Dr. Lofthouse put him on a “rigorous trial on an antipsychotic medication.”

The “Nursing Referral and Care Plan” at Belmont Hills documents an early and continued concern with planning for Patrick’s discharge. But he did not respond well to the antipsychotic medication. “Throughout his stay in the hospital he was somewhat isolated, related in a superficial manner, [and] had major difficulties using any individual or group psychotherapy.” Nevertheless, Dr. Lofthouse was concerned “to avoid institutionalizing him.” By December his symptoms “were beginning to be less active,” although “his instability was always apparent.” While recognizing that “he was not well,” Lofthouse arranged for Patrick to be released to his family on December 18th for the Christmas holiday and to enter a half-way house in January.

In about two weeks Patrick suffered a crisis similar to that which had first precipitated his hospitalization. A day after his admission to a half-way house, he ran away, returned to his mother’s home, and took an overdose of four aspirin and seventeen tetracycline. He was taken again to Mary’s Help Hospital on January 6, 1982. Dr. De Lorente recalls, “He was extremely agitated. He was extremely frightened. I think he was in panic. He had not slept for one or two days.” After two days he was readmitted to Belmont Hills where Dr. Lofthouse noted that “he had a strong belief of a delusional nature that his soul was going to hell.”

Patrick’s second hospitalization at Belmont Hills lasted until February 12, 1982, when he was discharged to his parents with the recommendation that he receive treatment as an outpatient. During the next four months, he continued to receive psychotherapy once a week but was mute and withdrawn and often refused to take medication. In an effort to lift his mood of depression, the Belmont Hills clinic administered electroconvulsive therapy on four occasions. Late in May his condition deteriorated. He was reluctant to rise from bed in the morning and neglected his personal appearance, refusing to dress, bath, shave, or comb his hair. Late at night he could be seen staring at religious programs on television, with the volume on high and his face “right in front of [the television].” His parents appealed to Dr. De Lorente who visited Patrick at home to persuade him to return to the hospital. After getting no response, Dr. De Lorente suggested the option of calling the police to place him involuntarily in a mental ward.

On the morning of June 11th, Patrick was seen wandering the streets near his home, clad in a bathrobe, with a catatonic demeanor. That afternoon his father took him to Belmont Hills for a previously scheduled outpatient visit. Dr. Lofthouse directed that he be placed in a locked ward for 14 days—the period of involuntary confinement permitted under Welfare and Institutions Code section 5254—and observed, “[t]he patient was not talking and seemed extremely frightened and in a trance.” Patrick vigorously resisted this hospitalization. Showing uncharacteristic resourcefulness, he succeeded in retaining an attorney and applied for a writ of habeas corpus. Pleased by this unexpected display of assertiveness, Dr. Lofthouse decided not to contest the writ. Patrick was discharged on June 23th.

Within three days, Patrick began to talk “about fears of hurting himself’ and was again admitted to Mary’s Help Hospital for emergency hospitalization. Dr. De Lorente noted, “[h]e was agitated. He appeared to be frightened. He was hallucinating. He admitted to visual images of knives and axes.” He had apparently remained insomniac for about two days before hospitalization and continued to be unable to sleep despite extra dosage of hypnotic medicine. He remained under treatment at Mary’s Help until a bed became available at Belmont Hills on July 6th.

After being placed under the 72-hour detention permitted by Welfare and Institutions Code section 5150, Patrick angrily denied the facts of his past psychiatric history and soon managed to escape from the hospital. Returning home, he had difficulty sleeping and seemed nervous and quite frightened. On July 13th he was taken for a fifth time to Belmont Hills. Dr. Lofthouse testified that he clearly needed hospitalization “[b]ecause his behavoir [s7c] beyond any doubt indicated that he was incapable of functioning as an outpatient at home. He just could not cope. His illness was too active.” Lofthouse signed the certification required by Welfare and Institutions Code section 5252 to hold Patrick involuntarily for 14 days. The certification, which was cosigned by Dr. Ronald Hayman, the director of the Belmont Hills center, attested that the patient was “gravely disabled” and in need of “intensive care.” When Patrick again filed a writ of habeas corpus, Dr. Lofthouse appeared in court in opposition. He explained, “[h]e wasn’t getting help, he was acting bizarrely. We could rarely get him into what I would consider a stable outpatient condition and I felt that he probably needed much longer period of hospitalization.” The court denied the writ, and Patrick was transferred on August 15th for long-term care at the Institute for Living in Hartford, Connecticut.

As an employee of Ralph K. Davies Medical Center, Mrs. Hughes and her dependents were covered by group medical insurance offered by Blue Cross. This action concerns the periods of hospitalization at Belmont Hills from October 26 to December 18, 1981, and July 13 to August 15, 1982. For these periods, Mrs. Hughes submitted claims for hospital expenses totalling $23,698.69. Blue Cross consented to pay only $6,598.69, disallowing the balance on the ground that the hospitalizations were not medically necessary. The insurance company denied coverage for the portion of the 1981 hospitalization between November 21st and December 18th, and for all of the period of hospitalization in July-August 1982.

Blue Cross asserts that there was no substantial evidence to support the jury’s verdict. The contention calls for examination of the record dealing with the processing of the respondent’s claims. In viewing this record, the court must bear in mind the familiar rule of Crawford v. Southern Pacific Co. (1935) 3 Cal.2d 427, 429 [45 P.2d 183]: “[W]hen a verdict is attacked as being unsupported, the power of the appellate court begins and ends with a determination as to whether there is any substantial evidence, contradicted or uncontradicted, which will support the conclusion reached by the jury. When two or more inferences can be reasonably deduced from the facts, the reviewing court is without power to substitute its deductions for those of the trial court.”

Mrs. Hughes submitted her first claim for the October-December hospitalization in November 1981. After a preliminary clerical screening, the claim was referred to the medical review department of Blue Cross. This department allows certain claims that fall within very restrictive guidelines. The remaining claims are referred to a psychiatric consultant. The function of the medical review department is limited to procuring the necessary records, handling claim correspondence, and implementing the medical consultant’s recommendation.

Shortly after receiving her claim, the medical review department sent a form to the Belmont Hills clinic that indicated, by checks in a series of boxes, a request for the following records: “History and Physical Examination (Complete),” “Physician’s Orders,” “Progress Notes,” “Discharge Summary,” “Consultation Reports,” and “Nurses Notes.” At trial the claim file contained only three categories of records from Belmont Hills: physician’s orders, progress notes, and nurses’ notes. It did not include the discharge summary, a detailed statement by the attending physician of the patient’s history, treatment, and diagnosis. The medical consultant was confident he remembered reading the summary, but the clerk in the medical review department thought that it might not have been obtained. In any event, it is undisputed that the claim file did not contain the record of the patient’s earlier treatment at Mary’s Help Hospital or the “Nursing Referral and Care Plan” that documented discharge planning. Moreover, the file contained no record of a concurrent utilization review program at the Belmont Hills clinic. Under this program, an independent psychiatric consultant, Dr. Burton White, visited the clinic each week and reviewed the charts of all patients to determine the necessity of continued hospitalization. Dr. White had consistently approved Patrick’s stay at Belmont Hills.

Upon receiving this limited medical record, the Blue Cross consultant, Dr. Ronald Hayman, requested a letter from the utilization review committee at the Belmont Hills clinic explaining the need for “acute level of care” after November 15, 1981. The medical review department encountered delays in obtaining the requested letter. Shortly before making the request, Dr. Hayman had himself assumed the position of director of the Belmont Hills clinic. On his suggestion, the attending physician, Dr. Lofthouse, finally wrote a one-page letter dated May 13, 1982, advising among other things that within two weeks of his discharge Patrick “became progressively more despondent and suicidal” and was again hospitalized. After the letter was sent, Dr. Hayman notified the medical review department that the case should be sent to another psychiatric consultant.

The case was next reviewed by a registered nurse on the Blue Cross staff who transmitted the file to Dr. Ronald Mintz, a psychiatric consultant in Southern California, with the recommendation that coverage be disapproved after December 1, 1981. On July 9, 1982, Dr. Mintz jotted the following recommendation on the medical consultant review report: “By 11/21, following 11 weeks intensive inpt. treatment a lower level of care is medically appropriate. Discharge planning was seriously delayed, and is the reason for the custodial period 11/21 - discharge. Accept 10/26 - 11/20 and non-cover 11/21—discharge.”

Dr. Mintz was a person of undoubted professional credentials who had 14 years of experience working for Blue Cross. But in the course of adverse examination under Evidence Code section 776 respondent raised questions about the care and standards that he employed in reviewing claims. Certain testimony, seized upon in respondent’s oral argument, suggested that he devoted an average of only 12 minutes per claim. He didn’t know of the program of concurrent utilization review at the Belmont Hills clinic. He considered that Dr. Lofthouse’s letter contained no helpful information even though it alerted him to the patient’s suicide attempt two weeks after the disputed discharge. Most damaging, Dr. Mintz disclaimed any responsibility to secure all relevant records before making a recommendation. He testified, “I felt that it was the responsibility of the treating doctor to put any information that he wanted considered in the review of these records into the records.” Conceding that “the patient is the one that pays” if the records are incomplete, he still affirmed that it was not his practice to contact the utilization review committee of the treating hospital before denying a claim. If the hospital had additional information pertinent to the claim, it was the hospital’s option to submit it to Blue Cross with a request for reconsideration. He would make no effort to obtain the information.

The jury could reasonably infer that Dr. Mintz employed a standard of medical necessity markedly at variance from that of the psychiatric community in California. He testified that over the years he recommended disapproval of about 30 percent of the claims he reviewed. He was unswayed by the fact that his recommendation conflicted with that of other psychiatrists, such as Dr. Lofthouse, Dr. De Lorente, and Dr. Hayman, who were familiar with the case, and evinced little interest in the opinion of Dr. White who reviewed the patient’s charts to make the same determination of medical necessity each week of his hospitalization. “It is not an accounting contest,” he insisted. In repeated questioning, respondent sought to elicit an admission that Dr. Mintz’s definition of medical necessity differed from profession standards. While conceding that the consensus of the psychiatric community was “one of the things to be considered,” the latter confirmed that he employed independent judgment and once seemed to allow that his standard of medical necessity might be more restrictive than the generally accepted professional standard. The wording of his recommendation, stating that a “lower level of care is medically appropriate,” appeared to reflect independent opinion at odds with the judgment of the treating physicians.

According to its normal procedures, Blue Cross sent the treating physician, Dr. Lofthouse, a standard form letter notifying him of its intention to deny the claim for hospitalization after November 20th. The letter stated: “Based on our review of the medical records we have received to date, it is our determination that it was not medically necessary for this patient to have been hospitalized at the acute level after November 20, 1981. [fl] To allow all possible contract coverage, we are asking if you have any further medical information that can be provided regarding this patient’s admission.” The letter did not inform the physician of the medical grounds for the denial or identify the records Blue Cross had reviewed. Dr. Lofthouse testified that he was “very unclear” how to respond and chose not to answer. He explained, “I didn’t know what to believe, because I had written an initial letter and still I was getting more letters.” Receiving no response from Dr. Lofthouse, Blue Cross sent a letter dated October 5, 1982, to the patient, Patrick Hughes, denying the claim for hospitalization between November 20th and December 18th.

The processing of respondent’s claim for the period of hospitalization of the Belmont Hills clinic from July 13 to August 15, 1982, followed a very similar pattern. The Blue Cross medical review department evidently conveyed to Dr. Mintz only the physician’s orders, progress notes, and nursing notes for the patient’s hospitalization during this period. At trial, the file evidently contained an admission summary dated July 13th but no discharge summary. Although the progress notes alluded to the patient’s involuntary confinement, the file did not include the certification of need for intensive care signed by Dr. Lofthouse and Blue Cross’s own consultant, Dr. Hayman. The testimony of Dr. Mintz contains no indication that he considered—or received—any records pertaining to the patient’s earlier hospitalization at Belmont Hills on June 11th and July 6th, the clinic’s concurrent utilization review, or the two periods of hospitalization at Mary’s Help Hospital.

Dr. Mintz arrived at a recommendation based on the records he received without requesting additional documentation or making further inquiries. In a report dated September 14, 1982, he concluded, “[tjhis appears to be a custodial stay for the purpose of disposition planning. . . . Noncover.”

Before denying the claim, Blue Cross sent the attending physician, Dr. Lofthouse, the same form letter informing him of its determination that the hospitalization was not medically necessary. The letter again did not disclose the medical grounds for the determination or the records on which it had been based. Dr. Lofthouse complained, “[t]here was the sense of frustration of how to respond to the letters. I didn’t know quite what did they want.” Again, he failed to reply. On November 10, 1982, Blue Cross sent the patient a letter denying the claim for his hospitalization.

The covenant of good faith and fair dealing implied in all contracts requires each contracting party to refrain from doing anything to impair “ ‘the right of the other to receive the benefits of the agreement.’ ” (Murphy v. Allstate Ins. Co. (1976) 17 Cal.3d 937, 940 [132 Cal.Rptr. 424, 553 P.2d 584].) As applied to insurance contracts, it does not merely “connote the absence ... of positive misconduct of a malicious or immoral nature . . . .” (Neal v. Farmers Ins. Exchange (1978) 21 Cal. 3d 910, 922, fn. 5 [148 Cal.Rptr. 389, 582 P.2d 980].); it demands that the insurer act reasonably. (Id. at p. 925.) As stated in Gruenberg v. Aetna Ins. Co. (1973) 9 Cal.3d 566, 573 [108 Cal.Rptr. 480, 510 P.2d 1032], the covenant entails “a duty not to withhold unreasonably payments due under a policy.” Although the record here contains much conflicting testimony as to the merits of the insured’s claim, viewing the entire record in a light most favorable to the judgment, we find ample evidence to support a finding that the insurer acted unreasonably in denying benefits. (See Neal v. Farmers Ins. Exchange, supra, 21 Cal. 3d 910, 921.)

The presence of good faith implies “ ‘consistency with the justified expectations of the other party.’ ” (Neal v. Farmers Ins. Exchange, supra, 21 Cal.3d at p. 922, fn. 5.) As observed in Egan v. Mutual of Omaha Ins. Co. (1979) 24 Cal.3d 809, 819 [169 Cal.Rptr. 691, 620 P.2d 141], “[t]he insured in a contract like the one before us does not seek to obtain a commercial advantage by purchasing the policy—rather, he seeks protection against calamity. . . . The purchase of such insurance provides peace of mind and security . ...” In a medical insurance policy, the insured’s expectation of security is relevant to the interpretation of medical necessity. The issue is framed by Sarchett v. Blue Shield of California (1987) 43 Cal.3d 1 [233 Cal.Rptr. 76, 729 P.2d 267], where the court considered whether the duty of good faith requires an insurer to defer to the judgment of the treating physician on the issue of medical necessity. It refused to adopt this extreme position, holding that “ ‘medical necessity’ or similar policy language is an objective standard to be applied by the trier of fact, . . .” (Id. at p. 9.) But the court emphasized that the “ ‘reasonable expectation of the insured’ ” requires that the policy language be construed liberally “so that uncertainties about the reasonableness of treatment will be resolved in favor of coverage.” (Id. at p. 10.) The court emphasized “that, with doubts respecting coverage resolved in favor of the subscriber, there will be few cases in which the physician’s judgment is so plainly unreasonable, or contrary to good medical practice, that coverage will be refused.” (Id. at p. 13.)

If the insurer employs a standard of medical necessity significantly at variance with the medical standards of the community, the insured will accept the advice of his treating physician at a risk of incurring liability not likely foreseen at the time of entering the insurance contract. Such a restricted definition of medical necessity, frustrating the justified expectations of the insured, is inconsistent with the liberal construction of policy language required by the duty of good faith. It is true that the practice of retroactive review, which the Sarchett decision sanctions, will inevitably introduce a degree of uncertainty as to insurance coverage. But good faith demands a construction of medical necessity consistent with community medical standards that will minimize the patient’s uncertainty of coverage in accepting his physician’s recommended treatment. Here, the jury could reasonably infer from the testimony of Dr. Mintz that Blue Cross employed a standard of medical necessity sufficiently at variance with community standards to constitute bad faith.

Equally relevant to the present case, the Supreme Court has held “that an insurer may breach the covenant of good faith and fair dealing when it fails to properly investigate its insured’s claim.” (Egan v. Mutual of Omaha Ins. Co., supra, 24 Cal.3d 809, 817.) In reviewing the medical necessity of hospitalization, this duty of investigation surely entails an obligation to make reasonable efforts to obtain all medical records relevant to the hospitalization. The record discloses a failure of the appellant’s staff to forward all significant documents to the medical consultant, Dr. Mintz, who in turn exhibited an apparent lack of concern for securing complete documentation. It might be argued that these failures can be ascribed to the fallibility inherent in all organizations, falling short of actual bad faith. But the two Blue Cross letters to the treating physician plainly bring its conduct within the sphere of bad faith. By not identifying the records on which the consultant’s recommendation was based, the letters tended to assure that the staff’s earlier failure to secure all relevant records would go undetected. And by omitting any explanation of the medical grounds for the intended denial of coverage, the letters placed an undue burden of inquiry on the insured’s physician. The Blue Cross witnesses, in fact, defended the letters on the ground that the physician was free to write or call the medical review department to gain more information. The covenant of good faith and fair dealing, however, places the burden on the insurer to seek information relevant to the claim. This requires that the necessary letters to a treating physician be drafted in a manner calculated to elicit an informed response.

Former Civil Code section 3294, subdivision (a) authorizes punitive damages where “the defendant has been guilty of oppression, fraud, or malice, . . .” Each of these terms is defined in subdivision (c). As fraud is not at issue here, only the definitions of malice and oppression are pertinent: “(1) ‘Malice’ means conduct which is intended by the defendant to cause injury to the plaintiff or conduct which is carried on by the defendant with a conscious disregard of the rights or safety of others. (2) ‘Oppression’ means subjecting a person to cruel and unjust hardship in conscious disregard of that person’s rights.” Under these statutory definitions, punitive damages thus may be predicated either on an intent to harm or a conscious disregard of another’s rights. (Colonial Life & Accident Ins. Co. v. Superior Court (1982) 31 Cal.3d 785, 792 [183 Cal.Rptr. 810, 647 P.2d 86].) Since there is no evidence here of an intent to harm, the issue narrows the question whether there is substantial evidence that Blue Cross acted “in conscious disregard” of the insured’s rights.

Punitive damages are appropriate only in cases of egregiously tortious conduct. “The mere carelessness or ignorance of the defendant does not justify the imposition of punitive damages. Unhappily, as a society, we must tolerate without added retribution these all too common lapses in ourselves. Punitive damages are proper only when the tortious conduct rises to levels of extreme indifference to the plaintiff’s rights, a level which decent citizens should not have to tolerate.” (Flyer’s Body Shop Profit Sharing Plan v. Ticor Title Ins. Co. (1986) 185 Cal.App.3d 1149, 1154 [230 Cal.Rptr. 276].) But the award of punitive damages finds a justification where it serves to deter socially unacceptable corporate policies. In Egan v. Mutual of Omaha Ins. Co., supra, 24 Cal. 3d 809, 820, the court observed, “[i]n the present context, the principal purpose of section 3294 is to deter acts deemed socially unacceptable and, consequently, to discourage the perpetuation of objectionable corporate policies. [Citation.] Traditional arguments challenging the validity of exemplary damages lose force when a punitive award is based on this justification.” Thus, in Neal v. Farmers Ins. Exchange, supra, 21 Cal. 3d 910, 923, a judgment of punitive damages was upheld where evidence indicated that an insurer’s actions “were all part of a conscious course of conduct, firmly grounded in established company policy, . . .” (Fn. deleted.)

In the case at bar, there was evidence that the denial of respondent’s claim was not simply the unfortunate result of poor judgment but the product of the fragmentary medical records, a cursory review of the records, the consultant’s disclaimer of any obligation to investigate, the use of a standard of medical necessity at variance with community standards, and the uninformative follow-up letters sent to the treating physicians. The jury could reasonably infer that these practices, particularly the reliance on a restrictive standard of medical necessity and the unhelpful letter to the treating physician, were all rooted in established company practice. The evidence hence was sufficient to support a finding that the review process operated in conscious disregard of the insured’s rights.

Blue Cross objects to the definition of “medical necessity” incorporated in the order compelling arbitration, and contends that the court erred in failing to refer to arbitration the question whether the insured was suffering from an “acute phase” of mental illness within the meaning of the insurance contract. However, we need not consider the merits of these issues since the order was not appealable. (9 Witkin, Cal. Procedure (3d ed. 1985) Appeal, §71, p. 95.) An appeal may be taken only from a judgment confirming an arbitration award or from an order refusing to enter such a judgment. (Code Civ. Proc., § 1294.) In the present case, no judicial proceedings followed the arbitration award. Blue Cross willingly paid the disputed benefits, and the respondents did not seek to enter a judgment. The question subject to arbitration—the amount of benefits due under the policy—did not enter into the trial resulting in the judgment.

Finally, we reject Blue Cross’s claim that the trial court erred in instructing the jury that the insurer’s duty to process claims fairly and in good faith was a nondelegable duty. The instruction properly states the holding of Garner v. American Mut. Liability Ins. Co. (1973) 31 Cal.App.3d 843 [107 Cal.Rptr. 604], with which we agree.

After the opening briefs had been filed in this appeal, the United States Supreme Court rendered its decision in Pilot Life Ins. Co. v. Dedeaux (1987) 481 U.S. 41 [95 L.Ed.2d 39, 107 S.Ct. 1549], dealing with the scope of federal preemption of state law under the Employee Retirement Income Security Act of 1974 (ERISA). Relying on Pilot Life, Blue Cross raised the issue of federal preemption for the first time in its reply brief. We held that the issue had not been timely raised, and Blue Cross petitioned the California Supreme Court for review of this portion of the decision. Our high court has remanded the case to this court “with directions to vacate [our] opinion and to reconsider in light of Commercial Life Ins. v. Superior Court (1988) 47 Cal. 3d 473 [253 Cal.Rptr. 682, 764 P.2d 1059]; Rizzi v. Blue Cross of So. Calif. (1988) 206 Cal.App.3d 380, 382-391 [253 Cal.Rptr. 541]; Barnick v. Longs Drug Stores, Inc. (1988) 203 Cal.App.3d 377, 379-380 [250 Cal.Rptr. 10]; Molina v. Retail Clerks Unions etc. Benefit Fund (1980) 111 Cal.App.3d 872, 878-879 [168 Cal.Rptr. 906].” We now present a more detailed analysis of the issue, giving particular attention to the precedents the Supreme Court has recommended for our consideration.

Under Pilot Life, ERISA clearly preempts causes of action for breach of the implied covenant of good faith in an employee welfare benefit plan as defined in 29 United States Code section 1002 (1). Subsequent California decisions have in fact extended the reach of ERISA preemption to statutory causes of action under Insurance Code section 790.03. (Commercial Life Ins. Co. v. Superior Court (1988) 47 Cal.3d 473 [253 Cal.Rptr. 682, 764 P.2d 1059]; Rizzi v. Blue Cross of So. California (1988) 206 Cal.App.3d 380, 382-391 [253 Cal.Rptr. 541].) But while the scope of federal preemption is now clear, the right of a litigant to raise the issue for the first time on appeal turns on other complex questions. We must first inquire whether federal preemption under the facts of the present cases alfects subject matter jurisdiction. Arguing that this is the correct analysis, Blue Cross relies on the rule that “[l]ack of subject matter jurisdiction . . . is such a basic defect that it can be raised at any time by any available procedure.” (Barnick v. Longs Drug Stores, Inc. (1988) 203 Cal.App.3d 377, 379 [250 Cal.Rptr. 10]; DeTomaso v. Pan American World Airways, Inc. (1987) 43 Cal.3d 517, 520, fn. 1 [235 Cal.Rptr. 292, 733 P.2d 614], cert. den. 484 U.S. 829 [98 L.Ed.2d 60, 108 S.Ct. 100]; Hartenstine v. Superior Court (1987) 196 Cal.App.3d 206, 214 [241 Cal.Rptr. 756].)

Subject matter jurisdiction has been defined as the authority of a court to try actions of the type or class to which a suit belongs. (2 Witkin, Cal. Procedure (3d ed. 1985) § 39, p. 405; Barnick v. Longs Drug Stores, Inc., supra, 203 Cal.App.3d at p. 379.) A question of subject matter jurisdiction necessarily involves choice of forum; since all legally cognizable grievances may be considered in some forum, a claim that a particular agency has or does not have authority to adjudicate a case inevitably implies that another court or agency does not have or has such power. (See Longshoremen v. Davis (1986) 476 U.S. 380, 391 [90 L.Ed.2d 389, 400-401, 106 S.Ct. 1904].) In the field of federal preemption of state law, questions of subject matter jurisdiction arise in those categories of cases over which the federal courts have exclusive jurisdiction, such as bankruptcy, patent and copyright law, penalties and forfeitures under the law of the United States, and some types of cases within the area of admiralty and maritime law. (Wright, The Law of Federal Courts (4th ed. 1983) § 10, p. 36.)

As in the field of admiralty and maritime law, federal preemption under ERISA sometimes affects subject matter jurisdiction and sometimes does not. Title 29 United States Code section 1132 (e)(1) provides: “Except for actions under subsection (a)(1)(B) of this section, the district courts of the United States shall have exclusive jurisdiction of civil actions under this subchapter brought by the Secretary or by a participant, beneficiary, or fiduciary. State courts of competent jurisdiction and district courts of the United States shall have concurrent jurisdiction of actions under subsection (a)(1)(B) of this section.” The exception of subsection (a)(1)(B) is a broad one applying to civil actions brought by a participant or beneficiary “to recover benefits due to him under the terms of his plan, to enforce his rights under the terms of the plan, or to clarify his rights to future benefits under the terms of the plan.”

Within the sphere of concurrent federal and state jurisdiction under subsection (a)(1)(B), the issue of federal preemption presents a question of choice of law rather than choice of forum. As explained in Scott v. Gulf Oil Corp. (9th Cir. 1985) 754 F.2d 1499, 1502, “Congress intended for the courts, borrowing from state law where appropriate, and guided by the policies expressed in ERISA and other federal labor laws, to fashion a body of federal common law to govern ERISA suits.” But within the sphere of exclusive federal jurisdiction under subsection (e), the issue of federal preemption of state law concerns the competency of the state court to hear and determine the case, that is, subject matter jurisdiction. Such cases include actions for breach of fiduciary obligations under pension and welfare benefit plans covered by ERISA and actions asserting statutory rights conferred by ERISA. (See Commercial Life Ins. Co. v. Superior Court, supra, 47 Cal.3d 473, 485; Lembo v. Texaco, Inc. (1987) 194 Cal.App.3d 531, 537-538 [239 Cal.Rptr. 596].)

We consider that an action, such as the present one, concerning the denial of a claim for benefits under a group insurance policy falls within the concurrent jurisdiction of federal and state courts under subsection (a)(1)(B) as an action to “recover benefits” and to enforce “rights . . . under the terms of the plan.” It is an obvious tautology to say that an action for breach of the provisions of a plan instrument, including breach of the implied covenant of good faith and fair dealing, is predicated on “rights [arising] under the terms of the plan . . . .” A companion case to Pilot Life supports this conclusion. In Metropolitan Life Ins. Co. v. Taylor (1987) 481 U.S. 58 [95 L.Ed.2d 55, 107 S.Ct. 1542], the plaintiff had sued in state court for the denial of disability benefits under an employee group insurance policy, alleging causes of action for breach of contract, retaliatory discharge, and wrongful termination of benefits. Relying chiefly on the analogy between the Labor Management Relations Act (29 U.S.C. § 141 et seq.) and ERISA (29 U.S.C. § 1001 et seq.), the Supreme Court held that, being subject to ERISA, the plaintiff’s action was removable to federal courts. The holding implies that the state court had subject matter jurisdiction over the action. Federal removal jurisdiction has long been regarded as derivative; only cases properly within the jurisdiction of state court are subject to removal. (Lambert Co. v. Balt. & Ohio R.R. Co. (1922) 258 U.S. 377, 382 [66 L.Ed. 671, 675, 42 S.Ct. 349]; Avco Corp. v. Aero Lodge 735 (1968) 390 U.S. 557 [20 L.Ed.2d 126, 88 S.Ct. 1235]; Wamp v. Chattanooga Housing Authority, etc. (6th Cir. 1975) 527 F.2d 595, 597.)

Dueringer v. General American Life Ins. Co. (5th Cir. 1988) 842 F.2d 127 is directly in point. There, the plaintiff brought an action alleging breach of contract and bad-faith denial of his insurance claim against an insurance company under a group policy of his former employer. After noting the relevance of Pilot Life Ins. Co. v. Dedeaux, supra, 481 U.S. 41, the court observed, “preemption in this case clearly involves a choice-of-law question . . . .” (Dueringer v. General American Life Ins. Co., supra, 842 F.2d at p. 130.)

Blue Cross relies, however, on Barnick v. Longs Drug Stores, Inc., supra, 203 Cal.App.3d 377. The decision concerned an action brought by an employee against his employer alleging a series of causes of action. Certain of these causes of action related to a benefit plan subject to ERISA. The plaintiff apparently assumed that, if these causes of action were preempted by ERISA, they would fall within the exclusive jurisdiction of the federal courts. On appeal, he relied on the defendant’s failure to raise federal preemption as an affirmative defense and argued that, in any event, “his entire lawsuit [did] not fall within the exclusive jurisdiction of the federal courts.” (Id. at p. 378.) For purpose of analysis, the court assumed without discussion that the issue of federal preemption concerned subject matter jurisdiction. “Considering the fundamental nature of subject matter jurisdiction,” the court held that the defense of federal preemption was not waived by failure to raise it as an affirmative defense. (Id. at p. 380.)

Respondent suggests that the assumption in Barnick of exclusive federal jurisdiction, affecting subject matter jurisdiction, results from a simple failure to recognize the applicable federal statute. Certainly, the fact that the court did not cite 29 United States Code section 1132 lends credit to this interpretation. But since the parties apparently agreed that the issue of preemption pertained to subject matter jurisdiction, we believe it can be inferred that the case involved claims, not specifically described in the opinion, such as breach of fidicuary obligations, that indeed fell within the exclusive jurisdiction of the federal courts under 29 United States Code section 1132(e)(1) and therefore involved subject matter jurisdiction.

Under federal law, where the defense of federal preemption involves choice of law rather than subject matter jurisdiction, it may be waived by failure to raise it properly in the trial court. The distinction is drawn in the leading case of Longshoremen v. Davis, supra, 476 U.S. 380. As the Supreme Court phrased the issue, the case concerned the question whether preemption of state proceedings by the National Labor Relations Act “is in the nature of an affirmative defense that must be asserted in the trial court or be considered forever waived or whether it is in the nature of a challenge to a court’s power to adjudicate that may be raised at any time.” (Id. at pp. 381-382 [90 L.Ed.2d at p. 395].) The court reasoned that “Congress intended for the [National Labor Relations] Board generally to exercise exclusive jurisdiction in this area.” Hence, federal preemption under the National Labor Relations Act entailed “a choice-of-forum rather than a choice-of-law question,” that is, subject matter jurisdiction. (Id. at p. 391 [90 L.Ed.2d at p. 401].) As an issue bearing on subject matter jurisdiction, the claim of federal preemption under the act could not be waived by failure to raise it as an affirmative defense. The court made clear, however, that its “decision today does not apply to pre-emption claims generally but only to those pre-emption claims that go to the State’s actual adjudicatory or regulatory power as opposed to the State’s substantive laws.” (Id. at p. 391, fn. 9 [90 L.Ed.2d at p. 401].)

In the converse factual situation where federal preemption involves a choice-of-law question, Ninth Circuit decisions following Davis have regarded federal preemption as a defense that is waived if not timely raised. (Gilchrist v. Jim Slemons Imports, Inc. (9th Cir. 1986) 803 F.2d 1488, 1497; Johnson v. Armored Transport of California, Inc. (9th Cir. 1987) 813 F.2d 1041, 1043.) As discussed earlier, Dueringer v. General American Life Ins. Co., supra, 842 F.2d 127, 129, characterized an issue of ERISA preemption very similar to that on appeal, as involving a choice-of-law question. Proceeding on this premise, the court held that the defense was waived where the defendant failed to raise it as an affirmative defense. Metropolitan Life Ins. Co. v. Taylor, supra, 481 U.S. 58, 63 [95 L.Ed.2d 55, 63], similarly observed, “Federal pre-emption is ordinarily a federal defense to the plaintiff’s suit.” Elaborating on this point, Kanne v. Connecticut General Life Ins. Co. (9th Cir. 1988) 859 F.2d 96, 99, footnote 4, stated, “[b]ecause Connecticut General’s claim of ERISA preemption is a federal defense in this lawsuit, Metropolitan Life Ins. Co. v. Taylor, 107 S.Ct. at 1546, the burden is on the defendant to prove the facts necessary to establish it.”

Blue Cross contends, however, that a state court decision, Molina v. Retail Clerks Unions etc. Benefit Fund (1980) 111 Cal.App.3d 872 [168 Cal.Rptr. 906], requires us to extend the sphere of federal preemption more broadly than required by federal precedents. In Molina, the plaintiffs, who brought suit for personal injuries against several defendants, received benefit payments from a third party, the Retail Clerks Union Benefit Fund. In the course of the litigation, they filed a complaint for declaratory relief against the benefit fund to determine its right to reimbursement for its payments in the event they should recover from the other defendants. The court ruled that the benefit fund was entitled to such reimbursement. On appeal, the benefit fund raised for the first time the question of ERISA preemption, arguing that federal law entitled it to reimbursement. Analyzing this issue, the Court of Appeal assumed that state courts had concurrent jurisdiction over the case “pursuant to 29 United States Code section 1132(e)(1), [szc] . . . .” (Id. at p. 879.) The court invoked the state procedural rule that a new legal theory may be raised on appeal when it rests on factual issues litigated in the trial court. Accordingly, the court held that the Benefit Fund had not waived the issue of federal preemption by failure to raise it in the trial court: “We have concluded that there is no dispute with respect to the underlying facts of this case. The respondent merely raises a new question of law.” (Id. at p. 878.)

The Molina decision raises a number of questions. First, what was the basis for the court’s assumption that it had subject matter jurisdiction? Concurrent state and federal jurisdiction under ERISA can normally be found under 29 United States Code section 1132(a)(1)(B). This provision, however, applies only to actions by a participant or beneficiary of a pension or welfare benefit plan; Molina concerned a claim by a trustee of a welfare benefit plan. Second, the Molina decision appears to imply that, even when federal preemption pertains only to choice of law, it is not waived by failure to raise it as an affirmative defense where the record on appeal provides an adequate basis to consider the issue. If this is so, the decision enlarges the sphere of federal preemption beyond that required by federal precedents. What interest has the state in extending the scope of federal preemption? While we think the decision in Molina is both fair and logical on its facts, which involved the selection of a state or federal rule for determination of a narrow technical point, we find no ready basis for extending its rationale beyond those facts.

In deference to our high court’s remand order, however, we will consider pursuant to Molina whether the record in this case establishes the existence of federal preemption. Though Molina cites no authority for its standard of appellate review, we construe it as adopting the rule that “[wjhere . . . the facts with reference to the contention newly made on appeal appear to be undisputed and [it appears] that probably no different showing could be made at a new hearing it is deemed appropriate to entertain the contention as a question of law on the undisputed facts and pass on it accordingly.” (Williams v. Mariposa County Unified Sch. Dist. (1978) 82 Cal.App.3d 843, 850 [147 Cal.Rptr. 452]; Pena v. Municipal Court (1979) 96 Cal.App.3d 77, 81; Leslie Salt Co. v. San Francisco Bay Conservation etc. Com. (1984) 153 Cal.App.3d 605, 611 [200 Cal.Rptr. 575].)

Federal preemption under ERISA applies to those employee benefit programs falling within the definition of an employee welfare benefit plan as defined in 29 United State Code section 1002(1). The application of the statutory definition to group insurance programs has engendered much litigation in recent years. We do not find it necessary, however, to review the relevant law here; that task was ably performed in Rizzi v. Blue Cross of So. California, supra, 206 Cal.App.3d 380, 382-390. We wish to observe only that the excellent survey of the decisional and regulatory law in Rizzi ends with a conclusion of disquieting ambiguity that reveals the uncertainty and controversy that still attends efforts to apply the statutory definition to particular facts: “Although factual differences and varying analytical approaches we have recited do not always coincide, the following general conclusions can be drawn from federal case precedent and the Department of Labor regulation. An employer may have (but has not necessarily) established an ERISA plan if his main involvement is to contribute to the payment of the premiums. If the employer does not contribute to the premiums, but merely transmits the employee’s premiums without endorsing the policy or profiting from the program, no ERISA plan has been established. On the other hand, whether the employer contributes premiums or not, if the employer has endorsed the program in the sense that he is significantly involved in it, an ERISA plan may have been established. The ultimate determination depends on an evaluation of all the circumstances, keeping in mind ERISA’s purpose of regulating an employer’s administration of benefit plans.” (Id. at pp. 389-390.)

Apart from the statute itself, the principal points of reference in the case law have been the decision of Donovan v. Dillingham (11th Cir. 1982) 688 F.2d 1367 and the Department of Labor regulations, 29 Code of Federal Regulations section 2510.3-1Q. Donovan adopted an analysis of the statutory language that enjoys widespread acceptance. (E.g., Dodd v. John Hancock Mut. Life Ins. Co. (E.D.Cal. 1988) 688 F.Supp. 564; Tucker v. Employers Life Ins. Co. (N.D.Ala. 1988) 689 F.Supp. 1073.) Approaching the matter from a different perspective, the Department of Labor sought to identify certain practices that do not constitute employee welfare benefit plans. The department was careful to avoid the implication that its regulations serve to define welfare benefit plans: “[T]his section should not be read as implicitly indicating the Department’s views on the possible scope of section 3(1).” (29 C.F.R. 2510.3-l(a)(4).) But the regulations have been used “as an analytic starting point in determining the existence of an ERI-SA plan.” (Lambert v. Pacific Mutual Life Ins. Co. (1989) 211 Cal.App.3d 456, 464 [259 Cal.Rptr. 398].)

Much of the controversy concerning the application of ERISA to group insurance contracts concerns the issue of the possible requirement of employer involvement in the administration of the programs. The issue has its origin in Fort Halifax Packing Co. v. Coyne (1987) 482 U.S. 1 [96 L.Ed.2d 1, 107 S.Ct. 2211]. Holding that ERISA did not apply to a one-time severance payment to employees in the event of plant closing, the Supreme Court observed that the purpose of ERISA preemption and the regulatory concerns of ERISA itself applied only where a benefit program entails “an ongoing administrative program” or at least “some type of administrative activity . . . .” (Id. at pp. 11, 15 [96 L.Ed.2d at pp. 11, 14].) Applying Fort Halifax to group insurance programs, Rizzi v. Blue Cross of So. California, supra, 206 Cal.App.3d 380, 389, concluded: “the line of demarcation between ERISA plans and non-ERISA plans depends on the level of involvement by the employer in the program so as to warrant federal regulation of the administrative integrity of the program.” This language must be read, however, in light of the actual holding of the case. The court affirmed a summary judgment dismissing the action on the ground of ERISA preemption on a showing only that the employer procured the group insurance policy for the benefit of its employees, maintained control over amendment or termination of the policy, and performed certain incidental services, such as transmission of subscription charges and notices. The analysis in Rizzi of the employer involvement issue was closely followed on very similar facts in Lambert v. Pacific Mutual Life Ins. Co., supra, 211 Cal.App.3d 456, 462.) 1

Arguing that the undisputed facts bring this case within these vague and controversial standards, Blue Cross adopts as a first line of defense a drastic oversimplification of existing case law: “[U]nder Donovan and its progeny,” it maintains, “ERISA is broadly construed to regulate group health insurance arrangements provided as a benefit of employment where the employer pays all or a portion of the policy premiums.” Alternatively, Blue Cross argues that the present record reveals sufficient employer involvement in the administration of the group health insurance program to bring the case within the standards of Rizzi.

In support of these contentions, Blue Cross relies chiefly on exhibit 14, a copy of a group insurance contract between Blue Cross and Mrs. Hughes’s employer, Ralph K. Davies Medical Center, which Hughes—not Blue Cross—introduced in the cross-examination of a Blue Cross supervisory employee. Exhibit 14 unquestionably refers to a group medical insurance policy for the benefit of the employees of Ralph K. Davies Medical Center. It reveals that the employer exercised much the same administrative control over the insurance program as the employer in Rizzi: it had the power to receive notices and terminate the policy, and it assumed responsibility for forwarding employee applications and “up-to-date elibigility data” to Blue Cross. But the agreement does not clearly reveal who paid premiums under the policy. While the section on “Termination of Coverage” does refer to the “failure of the Employer to pay subscription charges on the Member’s behalf,” this language is ambiguous with respect to the source of the payments; it could for example apply to cases where the employer merely transmits payroll deductions of its employees.

Blue Cross argues that any ambiguity regarding payment of premiums was resolved by Mrs. Hughes’s own testimony. At the beginning of her direct testimony, Hughes stated that one of the reasons she worked at the Ralph K. Davies Medical Center was because of the medical benefits. Blue Cross asserts that this statement establishes that the benefits were extended “at no cost to her.” The assertion appears to rest on the rather tenuous logic that Hughes could hardly have found the benefits attractive if she were required to pay for them herself. More convincingly, Blue Cross points to Hughes’s answer to a compound question on cross-examination: “Was it your understanding that it was only at Ralph K. Davies that you could have the benefit of an insurance policy paid for by an employer that would cover a dependent? A. I really didn’t think about another place. I only knew that was a plus where I was working and that meant a lot to me.”

Mrs. Hughes vigorously contests these indications that the employer paid premiums under the insurance agreement. In addition, she argues that the record is silent with respect to several points recognized in the decisional law and department of labor regulations as being relevant to the statutory definition of an employee benefit plan. Some of the points she raises are largely ancillary to the question of payment of premiums. For example, she argues that there is no evidence regarding whether the employer “endorsed” the policy or whether participation in the group insurance was voluntary. If the employer actually paid the premiums, these factors could almost inevitably be inferred. She also points to a rather obscure criterion in the Department of Labor regulations: did the employer profit by endorsing the policy or receive consideration in connection with the Blue Cross policy? (See 29 C.F.R. § 2510.3-l(j)(4).) Whatever may be the significance of this factor, the record is indeed silent on the point. Finally, she contends that the record fails to establish that Ralph K. Davies falls within the broad category of employers to which ERISA applies: there is no evidence that the employer engaged in interstate commerce or is not an exempted church or government hospital. (See 29 U.S.C. § 1003(a) and (b).) While the record is silent on these points, “at least a minimal effect on interstate commerce” can nearly always be inferred from such a group insurance contract. (Comprehensive Care Corp. v. Doughtry (S.D.Fla. 1988) 682 F.Supp. 516, 517.) Moreover, nothing suggests affirmatively that the employer belonged to the exempted categories of employers. 2

If we were faced, as was the court in Rizzi, with a question merely of substantial evidence, we would not hesitate to hold that this evidence, properly introduced and with an adequate foundation for relevance, would suffice to support a finding that the subject group insurance policy is an employee welfare benefit plan within the meaning of ERISA. But we can reverse the judgment here only if the fact of an employee welfare benefit plan is established beyond dispute. The inferences to be drawn from the record suggesting the existence of an ERISA plan, though highly plausible, in our view as a matter of fairness fall short of removing the question from the realm of reasonable objection and argument.

The record on this issue, moreover, includes other more serious defects. First, we can find no adequate foundation for the relevance of exhibit 14, the document on which Blue Cross bases most of its case. The document appears on its face to be a photocopy of an agreement between Blue Cross and Ralph K. Davies Medical Center dated August 1, 1981. The cover page contains the signature of two officers of Blue Cross but none of the officers of Ralph K. Davies Medical Center; it recites that the agreement “shall remain in effect for the term of one month from said date [Aug. 1, 1981], and, with the consent of Blue Cross, from month to month thereafter.” Mrs. Hughes introduced the exhibit in the cross-examination of two claims reviewers of Blue Cross, Nancy Lynn Larsen and Bonnie Lee Browning, in an effort to show that they denied the claim without any precise knowledge of the terms of the insurance contract. Under the pressure of cross-examination, both women replied evasively, displaying only tentative recognition of the document, and generally disclaimed any reliance on it.

The record, in short, contains no evidence serving to authenticate exhibit 14 or to establish its relevance; that is, to show that it was actually in effect at Ralph K. Davies Medical Center during the relevant time period. Blue Cross contends, however, that Mrs. Hughes vouched for the relevance of the document by introducing it herself. The question reaches this court in a novel procedural posture. Normally, we would review a question of proper evidentiary foundation following a proper objection in the trial court. Here, Blue Cross seeks on appeal to use evidence, introduced by Mrs. Hughes, for a purpose entirely different from that for which it was offered; and the proponent of the evidence, Hughes, objects to this new use of the evidence. We think that Hughes vouched only for the relevance of the document in cross-examination to test the knowledge of claims reviewers. She remains free to object to the new use of the document by Blue Cross on the ground that there is no foundation for this use.

Secondly, reliance on the fragmentary record in this case to find the existence of an employee welfare plan would violate the basic procedural rule that “evidence cannot determine an issue that the parties have left out of their pleadings.” (1 Witkin, Cal. Evidence (3d ed. 1986) § 286, p. 256.) As stated in Trafton v. Youngblood (1968) 69 Cal.2d 17, 32 [69 Cal.Rptr. 568, 442 P.2d 648], “ ‘[i]t is frequently the case that evidence which is admissible to establish one issue may tend to establish another issue than that for which it is offered, and it is a rule that evidence so introduced is available to establish any of the issues in the case. This rule is, however, limited to the issues which are to be tried. If the other issue that the evidence may tend to establish is not before the court the evidence must be limited to the actual issue. The fact of its introduction cannot be used to establish an issue that the parties have not made in their pleadings. The court would not be authorized to consider it as establishing an issue that was not before it for trial.’ ” (Quoting Riverside Water Co. v. Gage (1895) 108 Cal. 240, 245 [41 P. 299], italics deleted; accord: Crescent Lbr. Co. v. Larson (1913) 166 Cal. 168, 171 [135 P. 502]; Marvin v. Marvin (1981) 122 Cal.App.3d 871, 875 [176 Cal.Rptr. 555]; Stockton v. Ortiz (1975) 47 Cal.App.3d 183, 193 [120 Cal.Rptr. 456].)

The rule arises from the fundamental consideration, implied by due process, that a party must have notice of a claim and an opportunity to defend against it. (George Arakelian Farms, Inc. v. Agricultural Labor Relations Bd. (1986) 186 Cal.App.3d 94, 104 [230 Cal.Rptr. 428].) If evidence is used to resolve a claim not before the court for trial, both notice of the claim and an opportunity to defend against it will be deficient. In J.R. Norton Co. v. Agricultural Labor Relations Bd. (1987) 192 Cal.App.3d 874 [238 Cal.Rptr. 87], the Agricultural Labor Relations Board based an order on a finding of illegal failure to rehire. Finding that this charge was not included in the original charge against the defendant, the court stated, “Because Norton was not advised that failure to rehire was the activity it needed to defend against, it is not surprising the Board found Norton failed to present evi dence justifying a failure to rehire. Consequently Norton had no opportunity to gather evidence or prepare legal arguments refuting the occurrence of such violations. Fundamental fairness includes both the right to adequate notice and the right to defend against charged violations.” (Id. at p. 888.)

Similarly, in the instant case, Mrs. Hughes had no opportunity in the trial court to gather evidence and marshall arguments against the claim that the group insurance agreement is covered by ERISA. We may surmise from the sparse record that the agreement very probably is a welfare benefit plan, but the issue involves complex questions of law and fact. It would be fundamentally unfair to find the existence of a welfare benefit plan without having given Hughes notice of this defense and an opportunity to refute it.

We find nothing in Molina or in federal precedents that should cause us to abandon principles so well established in California law, and essential for the proper resolution of factual disputes, as the rules requiring a foundation to establish the relevance of documentary evidence and limiting findings to the issues before the court for trial. Blue Cross urges that this court remand the case for a further hearing to remedy any deficiency in the record. Such a remand does lie within the discretion of an appellate court where an issue of subject matter jurisdiction is first raised on appeal. In Consolidated Theatres, Inc. v. Theatrical Stage Employees Union (1968) 69 Cal.2d 713 [73 Cal.Rptr. 213, 447 P.2d 325], the record generally indicated that the case fell within the exclusive jurisdiction of the National Labor Relations Board but was unclear on a particular issue relating to this jurisdiction—whether the National Labor Relations Board had declined or would decline to exercise its exclusive jurisdiction. The court remanded the case for a hearing on this limited issue. But as we have seen, the issue of federal preemption here concerns only the question of choice of law. We know of no precedent authorizing this court to remand a case for a further hearing to determine the possible application of a new legal theory not raised in the trial court. We decline to do so.

Cross-appeal of Jurdy Hughes *

The judgment is affirmed.

Racanelli, P. J., and Holmdahl, J., concurred.

A petition for a rehearing was denied December 14, 1989, and the petition of appellant Blue Cross of Northern California for review by the Supreme Court was denied March 1, 1990. Lucas, C. J., and Panelli, J., were of the opinion that the petition should be granted.

1 A United States Supreme Court decision rendered after Rizzi appears to be broadly consistent with its treatment of the issue of employer involvement in administration. In Massachusetts v. Morash (1989) 490 U.S. 107, 115 [104 L.Ed.2d 98, 109, 109 S.Ct. 1668], the Supreme Court considered the application of ERISA to a policy of paying discharged employees for their unused vacation time. Holding that the payment of these benefits did not constitute an “employee welfare benefit plan,” the court noted that “Congress’ primary concern [in enacting ERISA] was with the mismanagement of funds accumulated to finance employee benefits and the failure to pay employees benefits from accumulated funds.” “The distinguishing feature of most of [the benefits enumerated in section 3(1) of ERISA] is that they accumulate over a period of time and are payable only upon the occurrence of a contingency outside of the control of the employee.” (Ibid.) While the decision does not provide any precise guidance for the case at bar, the court’s analysis can probably be read as confirming the relevance of employer involvement in plan administration as a factor in determining whether a group insurance contract is an employee welfare benefit plan.

2 Upon remand from the Supreme Court, Blue Cross has attempted to answer these points by filing two requests for judicial notice under Evidence Code section 452, subdivisions (c) and (h). The documents attached to these requests—all of limited relevance—consist of papers filed with state and federal agencies. We find that the documents do not come within the terms of subdivision (c). “We have found no authority and none has been cited for the proposition that materials prepared by private parties and merely on file with state agencies may be judicially noticed pursuant to subdivision (c).” (People v. Thacker (1985) 175 Cal.App.3d 594, 598 [221 Cal.Rptr. 37]; South Shore Land Co. v. Petersen (1964) 226 Cal.App.2d 725, 746 [38 Cal.Rptr. 392].) They also fall well outside the intended scope of subdivision (h). The Law Revision Commission commentary to Evidence Code section 452 states that subdivision (h) is intended to allow judicial notice of “[fjacts which are accepted as established by experts and specialists in the natural, physical, and social sciences, if those facts are of such wide acceptance that to submit them to the jury would be to risk irrational findings.”

The parties also engage in a largely irrelevant argument over whether the group insurance agreement and the insurer’s claim handling procedures complied with ERISA. With reference to another insurer, Comprehensive Care Corp. v. Daughtry, supra, 682 F.Supp. 516, 517, observed, “[t]he court must decide not whether Aetna complied with ERISA, but whether the insurance policy falls under the purview of ERISA.” At most, compliance with ERISA has some relevance to the issue of employer involvement in administration.

* See footnote, ante, page 832.