Summary
declining to consider evidence not provided to the plan administrator regarding the reliability of post-mortem blood-alcohol testing
Summary of this case from Baptiste v. Securian Fin. Grp.Opinion
CIVIL ACTION NO. 3:06CV-344-DW.
September 30, 2009
MEMORANDUM OPINION
This ERISA case is before the Court on cross-motions for judgment filed by each of the parties. The central question raised by the parties' motions is whether the Defendant insurance companies correctly denied the claim of the Plaintiffs for accidental death benefits following the death of their mother, Donna Cornish.
Defendants, United States Life Insurance Company of the City of New York (US Life) and American General Life Companies (AIG), denied the claim of the deceased's surviving children, Brinon and Heather, for accidental death and dismemberment (AD D) benefits based on the language of the intoxication exclusion provision of the policy. In doing so, US Life and AIG relied exclusively on a death certificate, medical examiner's autopsy report and police report, which indicated that the late Ms. Cornish drowned in her home bathtub "during ethanol intoxication." US Life and AIG now argue that their reliance on the exclusion to deny AD D benefits was justified given the findings contained in the various reports. The insurers, therefore, ask that judgment be entered in their favor.
The Court earlier determined in an extended order entered on June 4, 2009 (DN 42) that the appropriate standard of review in this ERISA action is the de novo standard.
Plaintiffs challenge the denial of AD D benefits. They argue that the Defendant insurance companies, in their rush to deny the claim, erroneously chose to rely on an wildly inaccurate, post-mortem blood alcohol test result, when minimal investigation would have revealed that medical and forensic publications condemn putrid blood alcohol test results as being inherently unreliable due to the endogenous production of alcohol in a decaying body. Because a heart blood sample was not taken from the deceased and analyzed until almost seven days after she was discovered dead in her bath tub, Plaintiffs argue that the results could not have accurately revealed the amount of alcohol in the blood of Ms. Cornish at the time of her death. Therefore, the blood alcohol level obtained from the testing was not reliable.
Plaintiffs further maintain that none of the reports relied on by US Life or AIG state anywhere that intoxication "caused" the drowning death of Donna Cornish. The reports instead reflect at most that she drowned "during" or "while" intoxicated. Accordingly, Plaintiffs insist that US Life and AIG have failed to carry the burden of causation placed on them by the very language of the exclusion that they now rely on to defeat the Plaintiffs' claim for benefits.
Finally, Plaintiffs assert that the term "intoxication" as used in the policy at issue and is itself so inherently ambiguous that it must be interpreted in accordance with Kentucky state law in a fashion most favorable to the Plaintiffs. When Kentucky case law is consulted, plaintiffs insist that decisions such as Old Equity Life Ins. Co. v. Combs, 437 S.W.3d 173 (Ky. 1969) and Healthwise of Kentuck y, Inc. v. Anglin, 956 S.W.2d 213 (Ky. 1997) require the defendants to prove that Ms. Cornish was physically and/or mentally incapacitated by alcohol at the time of her death in order for the intoxication exclusion to apply.
Because the intoxication exclusion required the Defendants to adequately show a causal relationship between Donna Cornish's death and her alleged intoxication, and because the Defendant insurance companies failed to meet this burden for the above reasons, Brinon and Heather Cornish ask the Court to enter judgment in their favor and award them $225,000 in AD D benefits, along with their attorney's fees and costs.
The Material Facts.
The basic circumstances of the death of Donna Cornish are not genuinely disputed. On Jan. 14, 2004, the late Ms. Cornish, a 55-year-old employee of the National Processing Company (NPC), was discovered at approximately 7:45 p.m. by her son, who found her face down in the bathtub of her home at 3405 Nandina Drive in Louisville, Kentucky. Local police who responded to the scene estimated that Ms. Cornish had died in her bathtub at approximately 4 p.m. on the preceding day, Jan. 13, 2004. At the time of her death, Donna Cornish had in force the aforementioned AD D policy with the Defendant insurance companies, which she had purchased through her employer, NPC.
On Jan. 15, 2004, two days after the death, Drs. William Ralston and Angela Wetherton performed an autopsy on the body of Donna Cornish at the Urban County Government Center located in Jefferson County, Kentucky. The autopsy revealed a well nourished, well developed 55-year-old, white female, 5 6 tall and weighing 208 lbs. Medical examination revealed no evidence of peripheral edema or fracture, nor any apparent evidence of medical treatment for prior injuries. The examination report, however, did reflect evidence of decomposition of the body that included "multiple areas of skin slippage and dark pink, green and brown discoloration" along with "`washerwoman's changes . . . [in] the hands and feet" with "early marbling . . . of the upper midline chest." (DN 46, Administrative Record (A.R.) USL 9). Likewise, all of the internal organs and viscera demonstrated mild to moderate decomposition evidenced by dark discoloration and diffuse softening. ( Id. at USL 10).
Examination of the heart revealed no abnormality in either the valves or the muscles; however, the doctors noted grade IV coronary atherosclerosis in the left anterior descending coronary artery with 80% to 85% occlusion and grade III atherosclerosis with 50% to 75% occlusion in the left circumflex coronary artery. Otherwise, the systemic aorta were found to be normal and elastic. Inspection of the large veins revealed no evidence of a blood clot prior to the deceased's death. ( Id.).
Drs. Ralston and Wetherton's examination of the deceased's remaining internal organs produced no remarkable results. The lungs revealed no evidence of any pre-death thrombo-emboli, calcification or friability formation. No local abnormal markings were demonstrated in the liver or gallbladder. The pancreas was "normally lobulated" with no calcification. Both the adrenals and the kidneys were found to be of normal size and shape with no abnormality. The neck and musculoskeletal systems revealed no fractures to the skull, the spine, the ribs, the shoulders or the pelvis, and no soft tissue hemorrhage in the neck. The doctors likewise failed to find any soft tissue hemorrhage within the scalp or any abnormalities in the deceased's brain.
As part of the autopsy examination, the doctors extracted a sample of blood from Ms. Cornish's heart, which along with her other internal organs had already begun to decompose. In fact, the autopsy report noted the heart displayed "post-mortem autolysis and bacterial colonization." ( Id. at USL 14). The sample of heart blood extracted on Jan. 15, along with a separate urine sample taken from the bladder, were transported to the state toxicology lab located in Frankfort, Kentucky, where they were later tested on Jan. 20, 2004. Only the heart blood sample was tested for the presence of alcohol, however. The urine sample tested negative for the presence of any opiates, cocaine, metabolites, PCP, cannavinoids, amphetamines or penzodiazephine. The toxicology report for the heart blood sample noted "apparent putrificiation" of the sample. ( Id. at USL 16). The same report indicated a blood alcohol level of 0.21% at the time of testing, seven days after the estimated time of death on Jan. 13.
On March 8, 2004, Chief Medical Examiner Ralston rendered his final diagnosis on the death of Donna Cornish. ( Id. at USL 8). In his final report, Dr. Ralston concluded that the death of Donna Cornish was "due to drowning during ethanol intoxication." ( Id.). Also among the findings of the medical examiner's final report was a "history of recent and chronic ethanol abuse per coroner." ( Id.). Otherwise, the report reflected only the presence of ischemic heart disease, "moderate decomposition" of the body, and the presence of a blood citalopram level of .2 mg/l (therapeutic range 0.081 — 0.16 mg/l).
Citalopram is an antidepressant found in selective serotonin re-uptake inhibitors (SSRIs) such as the antidepressant Celexa, among others. Citalopram is noted to have a moderate drug interaction with alcohol, and when used in combination therewith "may result in additive central nervous system depression and impairment of judgment, thinking, and psychomotor skills." See, http:\\www.drugs.com\drug-interactions\ citalopram-with-ethol-alcohol-679-0-1034-6570.html (visited Sept. 24, 2009).
Because the Defendant insurers did not cite any other form of intoxication other than ethanol intoxication as a basis for the denial of the Plaintiffs' claim, this information concerning the citalopram level determined by the postmortem toxicology testing is included in this opinion solely for the sake of completeness, rather than as a basis on which to support a judgment for or against either party.
On March 16, 2004, the Kentucky Cabinet for Health Services, Register of Vital Statistics, entered the official death certificate for Donna Cornish. The death certificate lists the "IMMEDIATE CAUSE (Final disease or condition resulting in death)" as being "drowning during ethanol intoxication." (DN 46, A.R. USL 4).
Based on the final diagnosis of the chief medical examiner, the immediate cause of death listed on the certificate of death, and the police report on the circumstances of the decedent's death, the Defendant insurers denied the Plaintiffs' claim in reliance on the exclusion for any loss that results from or is caused directly, indirectly, wholly or partly by being intoxicated or under the influence of any drug, unless taken as prescribed by a physician. (DN 46, A.R. USL 90). Specifically, investigative specialist Bruce Carter of AIG informed Brinon Cornish on June 25, 2004, by a denial of benefits letter that
Information developed in our investigation revealed that Donna M. Cornish died on Jan. 14, 2004, due to drowning during ethanol intoxication. Toxicology results reveal the blood alcohol level of 0.21%. Based upon our review of the Louisville Metro Police Department Uniform Incident Report and the Commonwealth of Kentucky Office of the Chief Medical Examiner Autopsy and Toxicology Report, we feel that this exclusion is applicable to this claim. As a result, we must decline any benefits under the accidental death benefit provision of the policy.
(DN 8, Plaintiff's Memorandum on Subject Matter Jurisdiction, Exh. 5, AIG letter of June 25, 2004).
Subsequently, Brinon Cornish filed suit against the Defendant insurance companies in the Circuit Court of Jefferson County, Kentucky, first on behalf of his mother's estate, and subsequently on his own behalf and that of his sister, alleging that the Defendants breached their contract of insurance by refusing to pay AD D benefits to the deceased's estate, and engaged in unfair claims settlement practices by failing to make prompt payment of the Plaintiffs' claim for AD D benefits. This lawsuit was removed to federal court where the undersigned determined by a prior order that the correctness of the decision to deny such benefits must be reviewed under the de novo standard of review (DN 42).
The Standard of Review.
The Court addressed this question in detail in its order of June 4, 2009 (DN 42). In that order, the Court determined that the "due proof of loss" language of the insurance policy at issue did not vest sufficient discretion with the plan administrator so as to justify the use of the arbitrary and capricious standard of review (DN 42, p. 3). In so doing, the Court rejected that line of cases arising from Perez v. Aetna Life Ins. Co., 150 F.3d 550, 557 (6th Cir. 1998) in favor of Hoover v. Provident Life and Acc. Ins. Co., 290 F.3d 801, 808 (6th Cir. 2001) and Perrin v. Hartford Life Ins. Co., 616 F.Supp.2d 652 (E.D. Ky. 2007). Accordingly, the Court will review the denial of benefits de novo.As Hoover explains, the role of a federal court when applying the de novo standard is simply to determine whether the plan administrator made the correct decision. Hoover, 290 F.3d at 808-09. The decision of the administrator in such circumstances is not accorded any presumption of correctness or other form of deference. Id; Perry v. Simplicity Engineering, 900 F.2d 963, 966-67 (6th Cir. 1990) ("When a court reviews a [benefits] decision de novo, it simply decides whether or not it agrees with the decision under review.").
Under the de novo standard, the Court not only reviews the legal conclusions of the plan administrator anew, but also the factual determinations, as well. See, Miller v. Hartford Life Ins. Co., 348 F. Supp.2d 815, 187 (E.D. Mich. 2004) (citing Wilkins v. Baptist Healthcare Systems, Inc., 150 F.3d 609, 613 (6th Cir. 1998)). As Hoover puts the matter, the role of the Court is "to determine whether the administrator properly interpreted the plan and whether the insured was entitled to benefits under the plan." Hoover, 290 F.3d at 809.
When the Court performs such review, it "applies `general rules' of contract law as part of the federal common law. . . ." Miller, 348 F. Supp. at 317 (citing Cassidy v. Akzo Nobel Salt, Inc., 308 F.3d 613, 165 (6th Cir. 2002). Further, "the federal common law may draw on state law principles, but state law is not controlling authority." Id. The Court must examine the terms of the insurance policy at issue according to "their plain meaning, in an ordinary and popular sense." Marquette General Hospital v. Goodman Forest Industries, 315 F.3d 629, 633 (6th Cir. 2003). The Cassidy decision explains that "in applying this `plain meaning analysis' the court `must give effect to the unambiguous terms of an ERISA plan.'" Cassidy, 308 F.3d at 618 (quoting Lake v. Metro Life Ins. Co., 73 F.3d 1372, 1379 (6th Cir. 1996)).
Finally, when it conducts a de novo review of the decision of the plan administrator, "the court generally considers only that evidence presented to the plan administrator at the time he or she determines the employee's eligibility in accordance with the plan's terms." Clark v. Aetna Life Ins. Co., 395 F. Supp.2d 589, 608-09 (W.D. Mich. 2005) (citing Smith v. Ameritech, 129 F.3d 857, 863 (6th Cir. 1997)). In other words, in the Sixth Circuit, the Court's de novo review is limited to the administrative record,. See, Wilkins, 150 F.3d at 618. The administrative record in an ERISA case is considered to include all documentation submitted during the administrative appeals process, as well as that information considered by the plan administrator at the time of the challenged decision to deny benefits. Id. (citing Kalish v. Liberty Mutual/Liberty Life Assur. Co. of Boston, 419 F.3d 501, 511 (6th Cir. 2005)). See also, Perrin, 616 F. Supp.2d at 660 ("The review is limited to the record before the administrator and the court must determine whether the administrator properly interpreted the plan and whether the insured was entitled to benefits under the plan.").
The Sixth Circuit has explained the rationale for its limitation of consideration to the administrative record in Perry. Perry, 900 F.2d at 966. To quote the Court:
In the ERISA context, the role of the reviewing federal court is to determine whether the administrator or fiduciary made a correct decision, applying a de novo standard. Nothing in the legislative history suggests that Congress intended that federal district courts would function as substitute plan administrators, a role they would inevitably assume if they received and considered evidence not presented to administrators concerning an employee's entitlement to benefits. Such a procedure would frustrate the goal of prompt resolution of claims by the fiduciary under the ERISA scheme.Id. Accord Hoover, 290 F.3d at 808-09 ("The review is limited to the record before the administrator. . . ."). Thus, while other federal courts outside the Sixth Circuit may take a broader view so as to permit consideration of certain matters outside the administrative record, for example, when necessary to define a complex medical term, the Sixth Circuit appears to have taken a more restrictive view of the matter.
For other decisions that permit consideration of matters outside the administrative record, see Kansky v. Coca-Cola Bottling Co. of New York, 492 F.3d 54, 60 (1st Cir. 2007) (District court provided a reasonable explanation for its use of a single medical article that was not part of the administrative record, where the article was considered as background information); Elswick v. Life Ins. Co., 2007 WL 2745358 at *8-9, 13 (S.D. W.Vir. Sept. 20, 2007) (District court in exceptional circumstances, such as when claims require consideration of complex medical questions or issues related to the credibility of medical experts, may exercise its discretion to allow additional evidence beyond that found in the administrative record); Franz v. New England Life Ins. co., 2007 WL 2238388 (N.D. Okla. Aug. 1, 2007) (Supplementation of the administrative record is permitted under Quesinberry v. Life Ins. Co. of North Amer., 987 F.2d 1017, 1021-27 (4th Cir. 1993) in six exceptional circumstances, but the district court must only permit the supplementation if the party seeking to introduce such evidence can demonstrate that it could not previously have been submitted to the plan administrator at the time that the challenged decision was made); Ducra v. SBC-Southwestern Bell, 2007 WL 128900 at *2 (W.D. Tex. Jan. 12, 2007) ("Court may not normally look beyond the administrative record when evaluating ERISA claims, [however] the court may refer to extrinsic information to explain medical terminology within the record.").
Position of the Plaintiffs.
Plaintiffs argue in their memorandum in support of summary judgment that the decision of AIG to apply the intoxication exclusion under the policy was flawed in multiple respects. First and foremost, the post-mortem blood sample relied on to determine the blood alcohol content of Donna Cornish was putrid at the time of the toxicology testing on Jan. 20, 2004. Post-mortem autolysis and bacterial colonization of Ms. Cornish's heart resulted in the endogenous production of ethanol due to the putrification and fermentation of her blood sample.This endogenous production of alcohol potentially resulted in a post-mortem blood alcohol level of between .01 to .22 percent simply by the natural process of putrification, even if no alcohol has been imbibed by the deceased before her death. See, William Anderson, Medical-Legal Aspects of Alcohol, 11.3 at pp 240-41 (Lawyers Pub. Co. 2003) ("The formation of ethanol in human cadavers has been suspected and documented for some time"); Yale Kaplan, Ph.D., "The Determination Of Alcohol In Blood And Breath" Forensic Science Handbook, Chap. 12, p. 608 (Prentice-Hall 1982) ("Post-mortem blood is, on the other hand, subject to putrification and fermentation whereby the destruction and neoformation of alcohol is possible."); M.F. Gilliland, M.D. and Robert Borst, Ph.D., "Alcohol In Decomposed Bodies: Post-mortem Synthesis And Distribution," Journal of Forensic Sciences, Vol. 38, No. 6 at pp. 1266-74 (Nov. 1993) ("Blood alcohol concentrations as high as .15 percent have been contributed to the endogenous production of alcohol alone, as opposed to ingestion.").
Based on the above-cited medical literature, which is not included in the administrative record, Plaintiffs maintain that the putrid sample of heart blood analyzed seven days after the death of Donna Cornish was in no fashion a reliable factual basis on which to determine that Ms. Cornish was intoxicated at the time of her death. Yet, U.S. Life and AIG relied only on the autopsy report of the chief medical examiner and the certificate of death, without consulting any medical or toxicology expert to assist in understanding the impact of the circumstances at the time of the testing on the accuracy of the blood alcohol test results. In the Plaintiffs' words, "It was patently unreasonable for AIG to deny coverage based solely upon the blood alcohol level obtained from the obviously putrified [heart blood] sample." (DN 48, p. 9).
Aside from the improper reliance of the Defendant insurers on the suspect toxicology report, Plaintiffs also argue that the language of the exclusion places the burden squarely on the Defendants to show that alcohol intoxication caused the death of Donna Cornish, not merely that she may have had some level of alcohol in her system on Jan. 13. Given the inherently ambiguous and undefined nature of the term "intoxication," Plaintiffs further maintain this Court must construe the term "intoxication" in a fashion most favorable to the Plaintiffs using Kentucky case law to limit intoxication to those instances where the deceased was demonstrably under the influence of alcohol so as to substantially impair her mental and physical abilities. See, Old Equity Life Ins. Co. v. Combs, 437 S.W.2d. 173 (Ky. 1969); Healthwise of Kentucky, Ltd. v. Anglin, 956 S.W.2d 213, 217-18 (Ky. 1997).
On the first point, regarding causation, Plaintiffs maintain that neither the death certificate nor the autopsy report expressly state that the death of Donna Cornish was due to intoxication. Rather, the documents indicate merely that Cornish drowned "while" or "during" intoxicated. Plaintiffs cite to a number of state court decisions from jurisdictions outside Kentucky that recognize the burden of the insurer to establish causation in order to properly apply the intoxication exclusion. See, Harris v. Carolina Life Ins. Co., 233 So.2d 833, 834 (Fla. 1970) (Use of the terminology "directly or indirectly, wholly or partially" indicated some causal connection was intended by the parties to the contract so that the insurance company had the burden to show a causal relationship between the insured's death and his intoxication in order for the exclusionary provision to be effective"); Hastie v. J. C. Penney Life Ins. Co., 115 F.3d 895-97 (11th Cir. 1997) ("The insurance company had the burden of showing some causal connection between the intoxication and the insured's death in order for the exclusion to be effective"); Olson v. American Bankers Ins. Co., 35 Cal. Rptr.2d 897 (Cal. App. 1 Dist. 1994) (same).
Because the mere presence of alcohol in the blood to some degree is not itself proof of intoxication, as defined in Kentucky, much less proof that alcohol intoxication caused the death of the insured, Plaintiffs conclude that U.S. Life and AIG failed to carry their burden under the terms of the policy. Consequently, this Court must award Plaintiffs the AD D benefits due them, as well as their attorney's fees and costs, given the Defendants' abuse of discretion in denying their claim for benefits.
LEGAL ANALYSIS
The Court begins its legal analysis with the language of the insurance policy at issue. Firestone Tire Rubber Co. v. Bruch, 489 U.S. 101, 112-13 (1989) (The court "review[s] the employee's plan as it would any other contract claim, by looking to the terms of the plan and other manifestations of the parties' intent.) The Court when performing this review of the plan is charged with the duty to construe the employee benefit plan with a view toward effecting its general purpose. Kolkowaski v. Goodrich Corp., 448 F.3d 843, 850 (6th Cir. 2006). Here, the language of the exclusion provision at issue provides that
No benefits will be paid for any loss that results from or is caused directly, indirectly, wholly or partly by:
. . .
6. Being intoxicated or under the influence of any drug, unless taken as prescribed by a physician.
(DN 33, Exh. C, Policy Exclusions). It is this language that US Life and AGI rely on as the basis of the decision to deny the Plaintiffs' claim.
First, the Court has no difficulty in light of the quoted policy language in concluding that the Defendant insurance companies bear the burden to establish a causal relationship between the intoxication of the late Donna Cornish and her resulting death by drowning in her home bath tub on or about Jan. 13, 2004. The case law cited by the Plaintiffs involving similar exclusion language repeatedly confirms the existence of such a duty. For example, in Harris v. Carolina Life Ins. Co., 233 So.2d 833, 834-35 (Fla. 1970), the Florida Supreme Court construed a similar exclusion in an accidental death insurance policy, which contained an exception for the death of an insured that resulted directly or indirectly, wholly or partially from . . . bodily injury while under the influence of alcohol or drug." Id. at 834.
In Harris, the deceased was a passenger in an automobile collision that resulted in his death. No dispute existed between the parties that the insured was under the influence of alcohol at the time of the accident, "but all parties concede[d] that there was no causal connection between such condition and the insured's death." Id. After the trial court granted summary judgment for the defendant insurer, based on its finding that no causal connection need be established, the Florida Supreme Court took the case on petition for certiorari.
In its opinion, the Florida Supreme Court in Harris observed that "the average person would assume that the language under consideration excluded liability for death related directly or indirectly to the use of drugs or alcohol." Id. at 834. The court immediately continued to reject the interpretation offered by the insurance company, an interpretation of the exclusion provision which would have removed the duty to establish causation, instead holding that
[B]eing mindful of the rule requiring that ambiguous provisions of an insurance policy be liberally construed in favor of the insured, we hold that the insurance company has the burden to show some causal relationship between the death and the intoxication in order for the exclusionary provision to be effective.Harris, 233 So.2d at 834-35.
Twenty-seven years later, the U.S. Court of Appeals for the Eleventh Circuit cited Harris in Hastie v. J. C. Penney Life Ins. Co., 115 F.3d at 896-97. The 11th Circuit in Hastie rejected the argument that the mere intoxicated status of the insured at the time of death triggered similarly-worded exclusions in two accidental death policies. Hastie found no significant difference between the facts of the case before it, which involved a fatal motorcycle accident that occurred while the insured was intoxicated, and those of Harris. In Hastie, the death certificate of the insured listed multiple blunt traumatic injuries as being the underlying cause of death, along with acute alcohol intoxication as a "significant condition contributing to death, but not resulting in the underlying cause." Id. at 896. Under these circumstances, the 11th Circuit held that a genuine issue of material fact existed as to whether a causal connection between the insured's intoxication and the fatal accident had been proven.
Plaintiffs also cite several California cases that involve the denial of accidental death benefits based on the interpretation of a policy exclusion for a loss caused wholly or in part, directly or indirectly by intoxicants. The primary case cited is Olson v. American Bankers Ins. Co. of Florida, 35 Cal. Rptr. 2d 897 (Ca. App., 1st Dist. 1995). In Olson, the deceased drowned in her home bath tub where friends discovered her floating face-down on the Fourth of July. After attempts to revive the deceased proved unsuccessful, an autopsy was performed. The autopsy determined death to be the result of asphyxia due to drowning. Id. A toxicology report revealed a .14 blood alcohol level with therapeutic amounts of Diazepam, commonly sold under the trade name Valium.
Based on this information, the defendant insurer, American Bankers Insurance Company, denied the claim of the insured's estate for benefits. Following a jury trial, a verdict was returned for the insured's spouse and beneficiary, Ivan Olson. On appeal, American Bankers challenged an instruction by the trial court regarding the intoxication exclusion based on California statutory law, which the court instructed the jury required the similarly-worded intoxication exclusion to require the defendant insurance company to show that the death of the insured, Phyllis Olson, was "in consequence of" the influence of alcohol. Olson, 35 Ca. Rptr. at 902. In reliance on the California statutorily required language, "in consequence of being intoxicated," the court in Olson citing Rivers v. Conger Life Ins. Co., 229 So.2d 625, 628 (Fla. App. 1969) held that the policy language required the insurer to establish a causative connection between the intoxication and the death or injury of the insured in order to deny benefits. The Rivers decision, itself citing 10 Couch on Insurance, § 41:466, p. 519 (2nd Ed. 1982), likewise holds that in those insurance policies that contain exclusions for injuries that are the "result" or "in consequence" or "caused by" intoxication, some causal connection must be shown. Rivers, 229 So.2d at 628 ("We therefore hold that in order to avoid liability in this case, [the insurer] was required to demonstrate a causative connection between Jefferson's intoxication and death."). See gen., 10 Couch on Insurance, § 143:88 (3rd ed. 2006).
Under section 10369.12 of the California Insurance Code, a policy may contain an exclusion for intoxicants or controlled substances, but such exclusion is required by state law to provide that "the insurer shall not be liable for any loss sustained or contracted in consequence of the insured's being intoxicated or under the influence of any controlled substance unless administered on the advice of a physician." Olson, 35 Ca. Rptr.2d at 902. Thus, as the California Appellate Court explained in Olson, "the legislature has made a statutory determination on the limits of an alcohol or controlled substance exclusion in an insurance policy issued in California." Id.
While one might reasonably debate the factual circumstances of any of the cited cases, or the influence of the particular state law involved therein, the general conclusion holds true — that language of the present intoxication exclusion requires that a defendant insurance company establish a causal link between the insured's intoxication and the loss at issue in order to properly assert the exclusion as a basis on which to deny benefits under the policy. Indeed, little possibility of another reasonable interpretation of the exclusion exists, given the "results from or is caused directly, indirectly, wholly or partly" by being intoxicated language of the policy. A different conclusion would require the Court to read out the term "caused" contained in the exclusion. Accordingly, US Life and AIG, as Plaintiffs argue, must establish causation in order to avoid contractual liability for AD D benefits.
This conclusion, while important, is not conclusive. It does not put to rest the nature of the causation that must be shown. The policy exclusion at issue does not require, nor does it contain any language, that puts the burden on AIG or US Life to show that intoxication was the "sole," "exclusive," or even "primary," cause of the drowning death of Donna Cornish. To the contrary, the policy language expressly provides that intoxication may be at most an indirect or partial cause of the subject loss without rendering the exclusion inapplicable. The result is that the Defendants must indeed establish a causation, but so long as the intoxication of the deceased is shown to have played any role in bringing about her death, the causal requirement will be satisfied.
The above discussion leaves open the significant question of just what "intoxication" is intended to mean under the policy. Plaintiffs correctly point out that nowhere in the policy at issue is the term "intoxication" defined. In the absence of a contractual definition, they maintain that the term is inherently ambiguous; therefore, the Court should rely on Kentucky case law such as Old Equity Life Ins. Co. v. Combs, 437 S.W.2d 173, 174-75 (Ky. 1969) and Healthwise of Kentucky, Ltd. v. Anglin, 956 S.W.2d 213, 217-18 (Ky. 1997) to define the term to the benefit of the insured.
In Old Equity, the Supreme Court of Kentucky, then known as the "Kentucky Court of Appeals," addressed an exclusion provision of a life insurance policy that provided that the insurer "shall not be liable for any loss sustained or contracted in consequence of the insured's being intoxicated or under the influence of any narcotic unless administered on the advice of a physician. Id. at 174. After finding that such exclusion provisions are not void as being contrary to public policy, the court in Old Equity explained that "in order to avoid liability the insurance company had the burden of showing that the insured was intoxicated to a degree that his judgment and coordination were impaired." Id. Under the circumstances of the case then at issue, the court in Old Equity held that the testimony of several witnesses concerning their observations of the deceased drinking moonshine whiskey and staggering about cursing while trespassing into the home of a neighbor were sufficient to entitle the insurance company to a directed verdict on the question of its liability. Id. at 175.
Likewise, the Kentucky Supreme Court in the Healthwise decision interpreted an intoxication exclusion provision in a group health insurance policy which excluded "treatment for injuries sustained as a result of being under the influence of alcohol (legal intoxication as defined by Kentucky law) or the illegal use of drugs." In reviewing this exclusion in Healthwise, the Supreme Court of Kentucky found there to be ambiguity in the language from the fact that several definitions of legal intoxication under state law could apply. Therefore, due to the ambiguity in the contract provision, it was liberally construed to afford coverage to the insured by the court adopting the statutory definition of public intoxication under KRS 222.202(1), which provides that
A person is guilty of alcohol intoxication when he appears in a public place manifestly under the influence of alcohol to the degree that he may endanger himself or other persons or property or unreasonably annoy persons in his vicinity.Healthwise, 956 S.W.2d at 217. In the view of the court in Healthwise, the ambiguity in the intoxication exclusion "required a legal adjudication of intoxication before the exclusion could be applied." Id.
The two Kentucky cases discussed above are certainly not irrelevant to the present inquiry. ERISA does permit a federal court to develop a body of federal common law when necessary to resolve an unsettled issue contained in an employee benefit plan. The Sixth Circuit recognized this possibility in Regents of the Univ. of Mich. v. Employees of Agency Rent-A-Car Hosp. Assoc., 122 F.3d 336, 339 (6th Cir. 1977) ("In those limited situations where ERISA fails to address a particular issue, federal courts are expected to develop a body of federal common law to fill in the interstitial gap in the statutory mandate."). Regents continues to add that the federal courts also may take instruction from the law of the state in which they sit in developing this federal common law, or may generally review the current state of the law on the issue and adopt a federal rule. Id. Therefore, Old Equity and Healthwise are not irrelevant, and may be instructive if the Court concludes that the term "intoxication" as used in the present policy is inherently ambiguous.
The problem with such a conclusion is that it runs contrary to the most recent federal ERISA case law in Kentucky, Brauer v. Prudential Ins. Co. of Amer., Case No. 04-531-JBC, 2006 WL 39058 (E.D. Ky. 2006). Brauer involved an insured who died from accidental alcohol poisoning while covered by a group AD D employee benefit plan that excluded payment if death occurred from intoxication. In Brauer, the insurance company, Prudential, denied the benefits claim of the deceased employee's spouse in reliance on the policy's intoxication exclusion. Brauer, 2006 WL 39058 at *1. The surviving spouse filed suit in an effort to obtain payment. On motion for summary judgment, the District Court for the Eastern District of Kentucky directly addressed the question of whether the undefined term "intoxication" as used in the policy must be interpreted in a manner most favorable to the insured. Prudential in denying the claim in Brauer relied on Kentucky's statutory presumption of intoxication arising from the legal blood alcohol limit of 0.08 percent set forth in KRS 189A.010 to define intoxication.
Judge Coffman in Brauer first noted in her legal analysis that the construction of the terms of an insurance plan under ERISA are governed by federal common law rules of contract interpretation that dictate that the provisions of such a plan "must be given their plain meaning in an ordinary and popular sense." Id. at *2 (citing Perez v. Aetna Life Ins. Co., 150 F.3d 550, 556 (6th Cir. 1998) and Turner v. SAFECO Life Ins. Co., 17 F.3d 141, 145 (6th Cir. 1994)). From this starting point, Judge Coffman continued to observe that "federal law gives . . . no right to torture language in an attempt to force particular results or to convey delitescent [hidden] nuances the contracting parties never intended or imagined." Id. (citing Turner, 17 F.3d at 145). Judge Coffman then concluded that the term intoxication "is not ambiguous" and that it includes alcohol poisoning. Because the death certificate for the insured listed his cause of death as alcohol poisoning, Judge Coffman concluded that the decision of the defendant insurer to deny benefits was neither arbitrary nor capricious.
This Court must concur with Judge Coffman in her conclusion that the term "intoxication" is not inherently ambiguous so as to trigger the principle of contract construction that requires the Court to adopt a definition of the term most favorable to the insured. Intoxication is presumed in Kentucky when one's blood alcohol level meets or exceeds 0.08 percent under KRS 189A.010(1) (Michie Cum. Supp. 2008). This well-established statutory presumption of intoxication is routinely relied on in virtually every court in the state when faced with motor vehicle offenses arising under KRS Chapter 189A. The Court accordingly must agree with Judge Coffman in her analysis of the matter in Brauer.
Further, we note that this Court is not the only one to rely upon a state statute that establishes a legal presumption of intoxication based on a blood alcohol level to determine when an individual is intoxicated. See, Ober v. CUNA Mut., 645 So.2d 231, 235-36 (La. App., 2nd Cir. 1994) ("In Louisiana, a person with a blood alcohol level of .10 is legally intoxicated. LRS 14:98A(2). Even if the term intoxication is ambiguous, it is still sufficiently specific to include one who is "decidedly drunk," or legally intoxicated.").
It is true that more demanding definitions of intoxication in various contexts can be found in other statutes in Kentucky. For example, public intoxication under KRS 525.110(1) (Michie 2008) defines an individual to be guilty of public intoxication
when he appears in a public place manifestly under the influence of a controlled substance, or other intoxicating substance, excluding alcohol (unless the alcohol is present in combination with any of the above), not therapeutically administered, to a degree that he may endanger himself or other persons or property or unreasonably annoy persons in his vicinity.
KRS 525.100(1) (emphasis added). Similarly, under KRS 222.202(1), the offense of alcohol intoxication in a public place occurs when a person "appears in a public place manifestly under the influence of alcohol to the degree that he may endanger himself or other persons or property, or unreasonably annoy persons in his vicinity." KRS 222.202(1) (Miochie 2007). Plaintiffs would have the Court apply either of these more stringent definitions of intoxication so as to require proof that at the time of her death Donna Cornish was manifestly under the influence of alcohol to the degree that she unreasonably endangered herself so as to bring about her own death.
The problem with this approach, beyond the holding of the Brauer decision, is that it writes into the exclusion at issue, additional requirements not presently found in the contract. In other words, Plaintiffs would have the Court through the vehicle of interpretation create, in effect, the additional requirement that the insured be essentially incapacitated in order for the exclusion to apply. This the Court cannot do. The Court may not substantially revise the parties' agreement under the guise of interpretation, as Judge Coffman recognized in Brauer.
To do so under Kentucky's statutes regarding public intoxication would be particularly inappropriate in light of the requirement in both statutes that the intoxicated individual manifest their diminished state so that an observer could determine that the affected individual was a threat to himself, others, or to their property. If such a statutory definition were deemed to be the proper standard by which to define intoxication, then insurers in the position of US Life and AIG would never be able to rely upon the intoxication exclusion under the present circumstances, regardless of the blood alcohol level of the deceased, in those cases such as the present one where the insured dies alone with no one present to observe their physical condition. Under such a scenario, even if a blood sample had been immediately taken, and had registered well in excess of the legally presumed limit for intoxication, the defendant insurers nonetheless would have been stripped of the protection of the intoxication exclusion as no witness was present to observe Donna Cornish to determine whether she was "manifestly under the influence of alcohol" so as to present a danger to herself or others.
Because the term "intoxication" is not inherently ambiguous in light of Kentucky's well-established statutory presumption for intoxication arising under KRS 189A.010, and because to adopt the statutory standard urged by the Plaintiffs would be to rewrite the contract at issue so as to render the exclusion clause ineffective regardless of the deceased's blood alcohol level, the Court must respectfully decline to apply either KRS 222.202(1) or KRS 525.100(1) to establish the definition of intoxication. We conclude that Brauer and reasoning set forth above forecloses the adoption of either of these alternative statutory measures of intoxication.
This brings the Court to the final and most important question of the issues raised by the Plaintiffs: May the Court properly rely upon the blood alcohol test results obtained on Jan. 20, 2004, from the post-mortem blood sample of Donna Cornish, a full week after her death on Jan. 13, 2004? Plaintiffs have offered up a number of scholarly articles and treatises that call into question the accuracy of such blood alcohol tests after the blood sample has begun to deteriorate, resulting in the endogenous production of alcohol. The Court certainly does not presume to take issue with the conclusions of the medical doctors in these articles, which the Court has previously cited in this opinion. The natural decay and deterioration of the blood itself can produce alcohol. Indeed, such a process may well have been at least in part responsible for the 0.21 percent blood alcohol level obtained from the test results.
The problem in this instance, however, is that none of the scholarly articles nor the treatises, nor the opinions of the authors offered therein, were presented to the Defendant insurance companies prior to their determination to deny the claim for benefits. Sixth Circuit case law previously discussed directly limits this Court to the contents of the administrative record. Indeed, the Sixth Circuit takes one of the more conservative views on this matter, unlike other federal circuit courts that permit the administrative record to be supplemented for various reasons not presently at issue in this case. Because the Plaintiffs' concerns regarding the putrification of the blood sample were never raised, as far as the Court can determine, prior to the final denial of the Plaintiffs' claim for benefits, this Court cannot now reopen the administrative record for a de novo hearing. While Plaintiffs are entitled to de novo review of the denial of benefits, they are not entitled, nor will the Court provide, a de novo hearing. See, Perry v. Simplicity Engineering, 900 F.2d at 996 ("Nothing in the legislative history [of ERISA] suggests that Congress intended that federal district courts should function as substitute plan administrators, a role they would inevitably assume if they received and considered evidence not presented to administrators concerning and employee's entitlement to benefits.").
In light of Perry and the other Sixth Circuit decisions cited, we may not reopen the administrative record to consider the aforementioned medical treatises and texts. A contrary holding would mean that the Court not only would be required to consider the Plaintiffs' new medical information, but would also be required to entertain the prospect that the Defendant insurers would seek to supplement the record with additional medical information concerning the potential synergistic effects of alcohol and Citalopram, an anti-depressant found in the postmortem blood sample of Donna Cornish in excess of the maximum therapeutic range. The Court previously has made little mention of Citalopram because the Defendant insurance companies did not rely upon its presence or possible interaction with alcohol, as an additional basis on which to apply the intoxication exclusion. Were the Court to now reopen the record for a de novo hearing and begin again from square one, fundamental fairness would require that the Defendant insurance companies at least have the opportunity to purse this aspect of the case as it relates to the relevant exclusion should they choose to do so. The Court, however, will not begin "from square one." Sixth Circuit case law expressly excludes the consideration of matters outside the administrative record. Even in those ERISA cases which a denial of benefits receives de novo review, a de novo hearing is not permitted.
When one considers the administrative record, the documents relied upon by the Defendant insurers appear to the Court to set forth an ample basis on which to correctly apply on the intoxication exclusion. The final diagnosis of Chief Medical Examiner Ralston reveals that Donna Cornish was a woman with a "history of recent and chronic ethanol abuse. . . ." Her autopsy results reveal no other apparent reason for her drowning death other than her intoxication. She showed no evidence of any prior invasive surgical procedures indicative of a pre-existing health condition which might have brought about her unfortunate and untimely death. She had no fractured bones in her skull, neck or other main weight-bearing joints which might have indicated a fall. Autopsy of her brain tissue was essentially normal with no indication of any stroke or other process which would have brought about her death.
The only indication of any potentially serious medical condition was the existence of coronary artery disease. One of Ms. Cornish's coronary arteries, her left anterior descending coronary artery was noted to be 85% occluded, or blocked. Neither the autopsy report, nor any other source, however, gives any indication that this coronary artery disease was even symptomatic, much less a potential cause of death. Nowhere in the administrative record is there revealed any treatment history for coronary artery disease. Neither party has cited the Court to any portion of the record which would indicate that Ms. Cornish prior to her death was under active treatment for heart disease. Simply put, the findings of the autopsy report leave little other cause for the drowning death of Ms. Cornish than alcohol intoxication.
Both the chief medical examiner and the certificate of death indicate that Donna Cornish drowned during ethanol intoxication. In fact, the official certificate of death specifically states that the immediate cause resulting in death was drowning during ethanol intoxication. US Life and AIG cannot be faulted for their natural reliance upon these conclusions. As they note, in Kentucky a death certificate is presumed to be prima facie evidence of the cause of death. Ford v. Commonwealth Life Ins. Co., 267 S.W.2d 950, 951 (Ky. 1934). Although the Plaintiffs now criticize the Defendant insurers for not conducting additional inquiry, the information before them in the administrative record including the death certificate, autopsy report, toxicology report and final diagnosis of the chief medical examiner, were themselves sufficient to support the denial of benefits under the intoxication exclusion in the Court's view. It was reasonable for the Defendants to rely on the findings of the medical examiner who had the benefit of the blood analysis and, presumably, considered the effect that putrificiation had on the validity of the blood/alcohol reading. With that knowledge and the benefit of a full autopsy, the medical examiner concluded that Ms. Cornish was intoxicated at the time of her death by drowning. Moreover, the autopsy results offered no plausible cause for drowning other than alcohol intoxication. Upon review of all of the well-drafted motion papers of the parties and the administrative record, the Court is left with the firm conviction that the correct result was reached in denying benefits under the AD D policy at issue. Accordingly, the Court shall enter a separate judgment on behalf of the Defendants and dismiss the claims of the Plaintiffs with prejudice.