Low-Level Evidence Used to Substantiate Insurance Coverage Policies for Knee and Hip Arthroplasty.

Background: Access to total knee arthroplasty (TKA) or total hip arthroplasty (THA) has become increasingly regulated by health care insurance providers that require specific criteria be met prior to authorizing surgery as medically necessary. The purpose of this study was to examine references from coverage policies and assess whether they justify the presurgery criteria mandated by insurance providers for approval of total joint arthroplasty in patients who have symptomatic knee and hip degenerative disease.

Methods: The largest private commercial insurance providers in the United States were identified, of which nine had publicly accessible coverage policies for TKA and THA. Coverage criteria for procedural approval and respective references were retrieved. There are three coverage criteria identified: (1) diagnosis of osteoarthritis, (2) nonsurgical treatment (e.g., preoperative physical therapy, nonsteroidal anti-inflammatories, etc.), and (3) exclusion criteria (e.g., body mass index thresholds < 40). Reviewers graded references cited in coverage policies by level of evidence, type of reference, and relationship to the three criteria groupings.

Results: In total, out of 824 references, only 450 (54.6%) references were relevant to primary TKA and THA. Of the 824, 259 (31.4%) contained information pertinent to the diagnosis of osteoarthritis, 84 (10.2%) to nonsurgical treatment, and 107 (13.0%) to exclusion criteria. Of the 84 references relevant to nonsurgical treatment, only 16 (19.1%) had a level of evidence I to III. Among all references related to nonsurgical treatment, only four specifically tested the efficacy of nonoperative modalities, representing 0.5% of all references.

Conclusions: Current prior authorization criteria found in insurance coverage policies for TKA and THA are unsubstantiated. Patients presenting for total joint arthroplasty would benefit from prior authorization criteria with recommendations grounded in evidence-based medicine. This is currently not the case.