This study seeks to understand whether people substitute between recreational cannabis and conventional over-the-counter (OTC) sleep medications. UPC-level grocery store scanner data in a multivariable panel regression design were used to compare the change in the monthly market share of sleep aids with varying dispensary-based recreational cannabis access (existence, sales, and count) in Colorado counties between 12/2013 and 12/2014. We measured annually-differenced market shares for sleep aids as a portion of the overall OTC medication market, thus accounting for store-level demand shifts in OTC medication markets and seasonality, and used the monthly changes in stores’ sleep aid market share to control for short-term trends. Relative to the overall OTC medication market, sleep aid market shares were growing prior to recreational cannabis availability. The trend reverses (a 236% decrease) with dispensary entry (-0.33 percentage points, 95% CI -0.43 to -0.24, p < 0.01) from a mean market share growth of 0.14 ± 0.97. The magnitude of the market share decline increases as more dispensaries enter a county and with higher county-level cannabis sales. The negative associations are driven by diphenhydramine- and doxylamine-based sleep aids rather than herbal sleep aids and melatonin. These findings support survey evidence that many individuals use cannabis to treat insomnia, although sleep disturbances are not a specific qualifying condition under any U.S. state-level medical cannabis law. Investigations designed to measure the relative effectiveness and side effect profiles of conventional OTC sleep aids and cannabis-based products are urgently needed to improve treatment of sleep disturbances while minimizing potentially serious negative side effects.
Copyright © 2019 Elsevier Ltd. All rights reserved.