A recent ruling from the 9th U.S. Circuit Court of Appeals illustrates a sharp difference in the way different federal circuits handle ERISA claims under what is known as the de novo standard of review. In Muniz v. Amec Construction Management Inc., 2010 U.S.App.LEXIS 22112 (9th Cir. Oct. 27, 2010), the plaintiff, Dierro Muniz, was diagnosed with HIV in 1989 and stopped working in 1991. Shortly thereafter, Muniz began receiving disability benefits under a policy issued by Connecticut General Life Insurance Co. (a CIGNA company). Those benefits continued without interruption until August 2006, when benefits were terminated despite the absence of any material medical improvement and in the face of the treating doctor's ongoing certification of disability. The treating doctor also notified CIGNA that in addition to the disease itself, Muniz's medications caused marked fatigue and left him unable to concentrate. Muniz was also repeatedly afflicted with methicillin-resistant staph infections. Dissatisfied with the treating doctors' records, the insurer demanded that Muniz undergo a functional capacity evaluation (FCE). However, the treating doctor refused to authorize the FCE, explaining that Muniz "suffer[ed] from wasting, fatigue [and being] unable to participate in any functional evaluation." CIGNA thereupon terminated benefit payments based on a file review performed by an in-house associate medical director.
Muniz appealed, but after another file review, a second medical director determined there was insufficient evidence demonstrating "functional deficits" that would prevent Muniz's performance of sedentary work. Hence, the appeal was denied, as was a second appeal. Litigation ensued.
The court applied the de novo standard; and after an initial review of the claim, the court deemed the record insufficient and notified the parties that in accordance with Federal Rule of Evidence 706, the court was prepared to appoint an independent expert to evaluate Muniz and present an opinion as to his functional capacity. Muniz agreed, believing the court was requesting the appointment of a physician with expertise in AIDS/HIV. However, the result was that Muniz was directed to undergo an FCE test. The test results were interpreted to support a conclusion that Muniz could perform sustained light to light-medium duty work, although the examiner characterized the plaintiff's activity tolerance and endurance as "fair to poor," body mechanics, pain behavior correlation and upper/lower strength were rated as "fair to good," and coordination and pace object control were deemed "good." Upon receipt of the FCE findings, the district court concluded that Muniz was not totally disabled and affirmed the benefit termination. Muniz appealed.
The appeals court reviewed the district court's findings under the clearly erroneous standard of appellate review and upheld the lower court's determination. The court began its discussion of the merits by examining which party had the burden of proof. The court ruled the burden remained on the plaintiff under the de novo standard — even though the plaintiff argued that once he met his burden by submitting the treating doctor's finding, the burden should have shifted to the defendant to justify its decision to terminate benefits. The court explained, "That benefits had previously been awarded and paid may be evidence relevant to the issue of whether the claimant was disabled and entitled to benefits at a later date, but that fact should not itself shift the burden of proof." *13. Although the court mentioned Saffon v. Wells Fargo & Co. Long Term Disability Plan, 522 F.3d 863, 871 (9th Cir. 2008), which found that as to a chronic condition one would expect improvement before benefits could be terminated ("In any event, assuming that the MRIs document no 'progression in degeneration,' MetLife does not explain why further degeneration is necessary to sustain a finding that Saffon is disabled. After all, MetLife had been paying Saffon long-term disability benefits for a year, which suggests that she was already disabled. In order to find her no longer disabled, one would expect the MRIs to show an improvement, not a lack of degeneration."), the court characterized that ruling as merely requiring a "meaningful dialogue" with the beneficiary regarding his or her claim before a final denial of the claim.
Ultimately, the court ruled the plaintiff failed to provide "sufficient evidence" to overcome the district court's ruling. First, the appeals court found the district court was not required to give deference to the treating doctor's opinion in accordance with Black & Decker Disability Plan v. Nord, 538 U.S. 822, 834 (2003). The court also upheld the district court's finding that the treating physician's records "were inconsistent, incomplete and did not ultimately support Muniz's claim that he met the definition of total disability under the CGLIC plan." Second, the court rejected the plaintiff's argument that an FCE performed in 2009 was irrelevant to Muniz's condition when benefits were terminated in 2006. The court pointed to other 9th Circuit cases permitting the admission of additional evidence in order to conduct an adequate de novo review of the benefit denial decision (Mongeluzo v. Baxter Travenol Long Term Disability Benefit Plan, 46 F.3d 938, 944 (9th Cir. 1995)), including the opinion of an independent expert. Walker v. American Home Shield Long Term Disability Plan, 180 F.3d 1065 (9th Cir. 1999). Nor did the court agree with the plaintiff that the plan was limited to the reasons given in the formal denial. The court also held the plaintiff's objection to the validity of FCE testing was waived and that the plaintiff had agreed to both the facility and examiner who performed the examination. Finally, the court rejected the plaintiff's argument as to the relevancy of the examination to Muniz's condition three years earlier. The court explained: "[W]hile not conclusive, the 2009 FCE potentially provided insight as to Muniz's previous condition, for Muniz had many of the same symptoms and same activity levels as he did in 2006, and Muniz does not contend that his underlying condition changed substantially."
The outcome in this case starkly demonstrates a marked circuit split in the scope of de novo "review" of ERISA claims. While the 9th Circuit conducted a review of the evidence, the 7th U.S. Circuit Court of Appeals, in Krolnik v. Prudential Ins.Co. of America, 570 F.3d 841 (7th Cir. 2009), made it clear the term "review" is a misnomer."
[T]he word "review" simply has to go. For what Firestone [v. Bruch, 489 U.S. 101 (1989)] requires is not "review" of any kind; it is an independent decision rather than "review" that Firestone contemplates. The court repeatedly wrote that litigation under ERISA by plan participants seeking benefits should be conducted just like contract litigation, for the plan and any insurance policy are contracts. 489 U.S. at 112-13. In a contract suit the judge does not "review" either party's decision. Instead the court takes evidence (if there is a dispute about a material fact) and makes an independent decision about how the language of the contract applies to those facts.
This ruling highlights why Krolnik's approach is so important. The plaintiff, Muniz, was deemed disabled by CIGNA for 15 years due to a chronic condition that is incurable and progressive, even though medical advances have been able to slow the progression of the disease. See, mayoclinic.com/health/hiv-aids/DS00005. Both the disease itself, and the medication necessary to treat the condition, imposed significant functional restrictions on Muniz; and recurrent drug-resistant staph infections only further complicated the plaintiff's status. The fact that Muniz was still alive is itself miraculous, but not dying is hardly the equivalent of capable of working.
Moreover, the known deficiencies of functional capacity evaluations even further weakened the determination in this case. According to Stup v. Unum Life Insur.Co. of America, 390 F.3d 301 (4th Cir. 2004), FCE testing, which is usually only a few hours in duration, cannot accurately predict functionality over an eight hour day or work week. But even if the test in this case was accurate, it is hard to square the test's conclusion with its finding that Muniz's activity tolerance and endurance was "fair to poor." Had a trial been conducted, all of these issues could have carefully been examined. Surely, the same outcome could have occurred; and if the treating doctor's opinion were indeed deficient, that would have been revealed in cross-examination. However, it would have been fascinating to learn at trial why CIGNA deemed Muniz disabled for 15 years, yet suddenly found him able to work — without any consideration as to what work he was capable of doing after being out of the workforce for 15 years.
Unfortunately, what appears to have occurred here is that while the district court and appeals court both used the term "de novo review," without a trial, the review and burden of proof considerations effectively conflated that standard with a deferential standard and elevated the burden of proof to require the plaintiff to prove the insurer's determination was not just wrong but unreasonable. Since there is but one form of civil action according to Rule 2 of the Federal Rules of Civil Procedure, the absence of a plenary proceeding in the 9th Circuit, while one is readily available in the 7th Circuit, demonstrates that as a matter of fairness, the tens of millions of workers covered by ERISA-governed disability and health benefits plans should all receive equal treatment.
