Underreporting of non-study cigarette use by study participants confounds the interpretation of results from ambulatory clinical trial of reduced nicotine cigarettes

BackgroundAs part of its comprehensive plan to significantly reduce the harm from tobacco products, the US Food and Drug Administration is establishing a product standard to lower nicotine in conventional cigarettes to make them "minimally addictive or non-addictive". Many clinical studies have investigated the potential impact of such a standard on smoking behavior and exposure to cigarette constituents. These ambulatory studies required participants who smoke to switch to reduced nicotine study cigarettes. In contrast to clinical trials on pharmaceuticals or medical devices, participants had ready access to non-study conventional nicotine cigarettes and high rates of non-study cigarette use were consistently reported. The magnitude of non-compliance, which could impact the interpretation of the study results, was not adequately assessed in these trials.MethodsWe conducted a secondary analysis of data from a large, randomized trial of reduced nicotine cigarettes with 840 participants to estimate the magnitude of non-compliance, i.e., the average number of non-study cigarettes smoked per day by study participants assigned to reduced nicotine cigarettes. Individual participants' non-study cigarette use was estimated based on his/her urinary total nicotine equivalent level, the nicotine content of the study cigarette assigned and the self-reported number of cigarettes smoked, using a previously published method.ResultsOur analysis showed that (1) there is a large variation in the number of non-study cigarettes smoked by participants within each group (coefficient of variation 90-232%); (2) participants in reduced nicotine cigarette groups underreported their mean number of non-study cigarettes smoked per day by 85-91%; and (3) the biochemical-based estimates indicate no reduction in the mean number of total cigarettes smoked per day for any group assigned to reduced nicotine cigarettes after accounting for non-study cigarettes.ConclusionsHigh levels of non-compliance, in both the rate and magnitude of non-study cigarette use, are common in ambulatory reduced nicotine cigarette trials where participants have access to conventional nicotine non-study cigarettes. The potential impact of high non-compliance on study outcomes should be considered when interpreting the results from such ambulatory studies.