Date of this Version
ERISA requires that plan administrators provide consumers with understandable health plan documents. The present study assessed the readability and comprehensibility of medical necessity and claims procedure clauses. For Study 1, I collected 40 summary plan descriptions from a diverse sample of employers and ran readability tests on the medical necessity and claims procedure clauses. Scores on the Flesch Reading Ease, Flesch Grade Level, and Fog Index indicated that the clauses were, in violation of ERISA’s disclosure requirement, written at reading levels beyond those one might expect the average plan participant to possess.
In Studies 2 and 3, employees read either original or redrafted versions of the clauses that received low readability scores in Study 1. Participants completed a comprehension test regarding the clauses. In both studies, participants’ overall comprehension accuracy scores (.15 in Study 2 and .19 in Study 3) indicated that participants did not understand the clauses. Contrary to hypotheses, participants who received the redrafted versions of the clauses did not perform better on the comprehension test than participants who received the original versions of the clauses did.
In Study 3, employees read the clauses either as a reading-to-learn or a reading-to-do task. Contrary to hypotheses, participants in the reading-to-do condition did not perform better on the comprehension test than participants in the reading-to-learn condition. The strength of the medical necessity claim also was manipulated in Study 3, and participants were informed that coverage for a treatment they sought had been denied. Consistent with hypotheses, participants were less likely to appeal a claim the more they felt the health plan was procedurally fair and the more they were satisfied with the health plan (ß=-.22, p<.001). In addition, participants were less likely to appeal a claim the more they comprehended the health plan (ß = -.24, p < .01), especially when they had a weak claim (ß=.21, p<.05), R2=.34, F(4, 204)=26.04, p<.001. Therefore, better comprehension led to more appropriate appeal decisions. Findings from this study have implications for enforcing ERISA’s disclosure requirement and for reducing healthcare expenditures by reducing the number of lawsuits over plan coverage.
Supervisor: Dr. Brian Bornstein