dissenting.
I agree with the majority’s opinion as it relates to the procedural irregularity in LINA’s handling of Brimers’ claim, and that it was not error for the district court to reject Brimers’ argument that she should be able to introduce additional evidence on the applicability of Exclusion 6. However, I cannot agree with the majority’s conclusion that Brimers’ “forfeiture” of the argument regarding the conflict between Exclusions 6 and 7 prevents this Court from addressing that conflict. I therefore respectfully dissent.
The district court relied solely on Exclusion 6 (“medical treatment”) to uphold LINA’s denial of Brimer’s claim, and Brimers appeal that decision. The de novo standard of review in this ERISA case requires this Court to determine the correctness of that decision — i.e., the applicability of Exclusion 6 — by examining the plan documents as a whole. A party’s reference to, or failure to refer to, a specific provision of the policy before the district court does not alter this Court’s obligation to examine the whole policy. Reading the policy as a whole, and the exclusions in conjunction with one another, I find a conflict between Exclusions 6 and 7 that renders Exclusion 6 ambiguous as to the question before us, i.e., whether James Brimer’s death from an overdose of prescribed drugs, notwithstanding that it was accidental, resulted from “medical treatment.” Resolving this ambiguity in favor of Brimer, as required by the doctrine of contra proferentem, I would hold that LINA may not rely upon Exclusion 6 to deny coverage, and I would reverse and remand with instructions to enter judgment in Brimers’ favor.
The policy does not define “medical treatment.” The interpretation of the undefined terms of an ERISA plan is governed by federal common law. LaAsmar v. Phelps Dodge Corp. Accidental Death & Dismemberment & Dependent Life Ins. Plan, 605 F.3d 789, 817 (10th Cir.2010) (citing Santaella v. Metro. Life Ins. Co., 123 F.3d 456, 461 (7th Cir.1997)); see Miller v. Monumental Life Ins. Co., 502 F.3d 1245, 1249-50 (10th Cir.2007). In interpreting an ERISA plan, this Court must “examine the plan documents as a whole, and, if unambiguous, construe them as a matter of law.” Miller, 502 F.3d at 1250 (internal quotations omitted). Reading the documents as a whole includes reading all exclusions in conjunction. See King v. Hartford Life and Acc. Ins. Co., 414 F.3d 994, 1004 (8th Cir.2005) (“But even if [the insurer’s] reading of [one exclusion] might be a reasonable interpretation of the language standing alone, ... it is not reasonable in the context of this policy, because it renders meaningless other important policy language[, including the next exclusion listed].”) (internal citation omitted); Clark v. Metropolitan Life Ins. Co., 369 F.Supp.2d 770, 778 (E.D.Va.2005) (reading exclusions “in conjunction”). If, employing a de novo standard of review, we find a plan term ambiguous, the doctrine of contra proferentem — construing all ambiguities against the drafter — requires the ambiguity to be resolved in favor of coverage. See LaAsmar, 605 F.3d at 805; Rasenack ex rel. Tribolet v. AIG Life Ins. Co., 585 F.3d 1311, 1318 (10th Cir.2009); Miller, 502 F.3d at 1249.
Whether a plan term or provision is ambiguous depends on the “circumstances presented” and the “question at issue in a given situation.” LaAsmar, 605 F.3d at 803-04 (citing Rasenack, 585 F.3d at 1318); id. at 804 n. 10. A plan provision is ambiguous if it “is reasonably susceptible to more than one meaning, or where there is uncertainty as to the meaning of the term.” Id. at 804 (quoting Rasenack, 585 F.3d at 1318). The inquiry is not into what the insurer, as drafter, “unilaterally *813intended the terms of the Plan to mean, but what a reasonable person in the position of the [insured] would have understood those terms to mean.” Id. at 801 (citing Rasenack, 585 F.3d at 1318).
I disagree with the district court’s conclusion that an overdose of prescription drugs “unambiguously” falls within the scope of the medical treatment exclusion. Rather, I find the plain language of the exclusion is ambiguous in this context. In light of the “question at issue in [this] situation,” id. at 803-04 — i.e., whether “medical treatment” includes an accidentally self-administered overdose of prescribed medication — the term “medical treatment” is “reasonably susceptible to more than one meaning,” see id. at 804. Because at least one of those possible meanings would result in coverage, the doctrine of contra proferentem requires that it be construed against LINA as the drafter. See id. at 805.
Even if Exclusion 6 were not facially ambiguous, it would most certainly become ambiguous when read in conjunction with Exclusion 7, which excludes losses resulting from “[v]oluntary self-administration of any drug or chemical substance not prescribed by, and taken according to the directions of, a licensed physician.” The district court correctly found this provision ambiguous. Consider just the interpretation urged by LINA before the district court, i.e., that the policy should be meant to cover losses related to voluntary, self-administered drug ingestion only if the drug was both prescribed by, and taken according to the instructions of, a licensed physician. This is impossible to square with the district court’s reading of Exclusion 6, which would hold any action by the patient with respect to a prescribed drug as part of “medical treatment.” If the policy is meant to cover losses that result from self-administering prescribed medication in accordance with a doctor’s instructions, then that means that self-administering prescribed medication in accordance with the doctor’s instructions is not “medical treatment” of a “sickness, disease, or bodily infirmity” within the meaning of Exclusion 6. And if self-administration of prescribed medication in accordance with the doctor’s instructions is not “medical treatment” within the meaning of Exclusion 6, it is difficult to see how self-administration of a prescribed medication contrary to a doctor’s instructions could be “medical treatment.” Conversely, if “medical treatment” encompasses the taking of any drug on the advice of a doctor, then there cannot be any exception to the drug exclusion, regardless of whether the prescribed drug is taken “according to the instructions of’ a doctor, contrary to the language of Exclusion 7. All of this places Exclusions 6 and 7 at odds with one another, a conflict that must, in light of our de novo review, be resolved in favor of coverage. See id. at 805; Rasenack, 585 F.3d at 1318; Miller, 502 F.3d at 1249; cf. Scruggs v. ExxonMobil Pension Plan, 585 F.3d 1356, 1366 (10th Cir.2009) (rule of contra proferentem inapplicable under arbitrary and capricious standard).
Although the majority denies the Brim-ers relief in this case on a procedural ground, the next similarly situated beneficiary will avoid this fate by arguing the conflict between Exclusions 6 and 7 from the outset. Insurers would do well to make clear in their policies precisely what they intend to cover and what they intend to exclude. According to the Centers for Disease Control, deaths resulting from accidental prescription drug overdoses are epidemic,1 and even a cursory review of *814the case law shows that such deaths have spurred extensive litigation between beneficiaries and insurers. It would not be unreasonable for an insurer to decide that it does not wish to insure against the risk of such deaths. And nothing prevents an insurer that makes that decision from simply, and explicitly, writing a policy that excludes deaths resulting from accidental drug overdoses. But an insurer should not benefit from poor drafting and ambiguities in its own policy to avoid paying benefits where they are due. Here, LINA drafted a policy that is ambiguous on the question of whether a self-administered accidental overdose of prescription medications is within the exclusion for “medical treatment.” That ambiguity should have been resolved in the Brimers’ favor.
I respectfully dissent.
. Centers for Disease Control, Policy Impact: Prescription Painkiller Overdoses, http://www. cdc.gov/homeandrecreationalsafety/rxbrief/ (last updated Dec. 19, 2011) (noting that drug overdose death rates have more than tripled *814since 1990, and that sales of prescription painkillers have increased more than 300% since 1999).