Dependence measures based on hookah smokers’ experiences and context are needed

We read with interest Sidani et al.’s 1 paper ‘Behavioral associations with waterpipe tobacco smoking dependence among US young adults’, which aimed to develop a dependence measure for adult waterpipe (WP) smokers in the United States 1. Although the authors recognize the importance of developing WP-specific dependence measures that conform to its use behaviors and patterns, in our opinion their approach fails to do so. First, the authors’ selection of WP dependence measures was influenced heavily by the cigarette literature rather than by WP's unique use patterns and behaviors. For example, the six items assessed were based on two scales—the Fagerström Test for Cigarette Dependence and the Lebanon Waterpipe Dependence scale (LWDS-11). The fact that some of the items were derived from the LWDS does not make this approach WP-sensitive, because the LWDS was also developed largely from cigarette instruments and did not follow a systematic approach that involves WP smokers for item generation 2. For example, the item ‘…how soon after waking up did you smoke your first hookah’ does not have a sound rationale for a predominantly intermittent tobacco use method such as the WP that requires time to set up and smoke. This item's value is based mainly on the bioavailability of nicotine (half-life of 2 hours), which makes sleep-time a period of ‘forced abstinence’ for most daily smokers driving nicotine levels to their lowest levels in the morning and creating a strong urge to smoke that correlates with dependence 3-6; but for a predominantly intermittent tobacco use method, night-time does not represent a period of ‘forced abstinence’ for most WP smokers, rendering time of smoking after waking up meaningless. In fact, only 5.1% of the sample studied by Sidani and colleagues smoked at least once in the past month and perhaps very few, if any, were daily smokers 1. Moreover, for a behavior that takes time to set up and smoke (average 1 hour), the authors’ extension of response time to 10 hours or more led perhaps to measuring opportunity to smoke rather than dependence. Other than the length of smoking, WP use is a seated behavior, making it much less accessible than cigarettes, which can lead the more dependent WP smokers to resort to the more readily available cigarettes to deal with their smoking urges 7 rather than smoke WP in the morning. The final concern about Sidani and colleagues’ study is the lack of any information on concomitant cigarette/other tobacco use among the study's sample. Given that the majority of young WP smokers in the United States also use other forms of tobacco 8, it is uncertain whether WP or cigarette smoking is driving the responses to the dependence items assessed. We agree with the authors that what is needed to move forward the assessment of WP dependence is careful consideration of WP-specific use patterns and context 9. Specifically, psychometrically sound instruments are needed that are grounded in dependence theory and informed by WP smokers’ experiences and setting.