Background: Both the 2012 Beers list and the American Geriatric Society 'Choosing Wisely' campaign suggest restraint in the use of sedative-hypnotics for the treatment of insomnia in older people. Sedative hypnotic agents continue to be widely prescribed even though their use in the elderly is associated with an increased risk of falls, fractures, and emergency hospitalizations.

Objective: The aim of this study was to estimate the cost effectiveness of cognitive behavioral therapy (CBT) compared with sedative-hypnotics and no treatment for insomnia in the US Medicare population, adjusting for the risk of falls and related consequences.

Methods: A model-based economic evaluation (decision tree) using the US Medicare perspective and a conservative annual temporal framework was conducted. Simulations were performed in a hypothetical cohort of Medicare beneficiaries suffering from insomnia. The main outcome measure was the incremental cost per quality-adjusted life year (QALY) gained. Sensitivity analyses assessed the robustness of the base-case analysis.

Results: On an annual basis, CBT showed a dominance (cost: US$19,442; QALYs: 0.594) over sedative hypnotics (cost: US$32,452; QALYs: 0.552) and no treatment (cost: US$33,853; QALYs: 0.517). Assuming a willingness to pay of US$50,000, the net monetary benefit was positive for CBT (US$10,287) and negative for sedative hypnotics (-US$4,851) and no treatment (-US$7,993). CBT had a 95% chance of being the dominant strategy, with results most sensitive to an older adult's baseline risk of falling.

Conclusion: Failure to consider drug harms such as drug-induced falls and hospitalization represents a growing public health concern, significantly underestimating the cost of sedative-hypnotic therapy and loss in quality of life for the elderly. Public payers should reconsider reimbursement of sedative-hypnotic drugs as first-line treatment for insomnia in older adults.