05. Statistical analyses were conducted using the Statistical Package for Social Sciences (SPSS) Version 15.0 (SPSS, 2007) and Stata version 9.2 (StataCorp, 2005). Results Participants Demographic nevertheless characteristics of participants are shown in detail in Table 1. Of the 548 participants, 51.8% (n = 282) were male, 92.8% (n = 504) were Jordanian, 92.0% (n = 494) were single, and 85.7% (n = 460) lived with family. Because nearly 100% were Arab (99.1%) and Muslim (98.7%), these variables were not included in subsequent analyses. Mean age was 21.7 years (SD = 2.9), and mean monthly income was 652 JD (ca. 925 US dollars). Students roughly equally represented each of the four study sites (Table 1). Most participants indicated that they were majoring in arts (36.3%), general sciences (30.4%), or medicine (24.

8%). Table 1. Demographic characteristics by water pipe tobacco smoking Prevalence of water pipe tobacco smoking Of the 548 participants, 61.1% (n = 335) had ever smoked tobacco from a water pipe. This was slightly higher than the 56.6% of participants (n = 309) who had ever smoked a cigarette. Use of water pipe at least monthly was reported by 42.7% (n = 227). Of those who had smoked tobacco from a water pipe, pattern of use was approximately equal across categories: yearly (25.6%), monthly (24.3%), weekly (30.5%), and daily (19.7%). Of ever water pipe tobacco smokers, mean age of initiation was 18.1 years (SD = 3.6). First use was most commonly with friends (60.7%) or family (30.9%) but less commonly alone (8.4%). Place of first use was most commonly at home (37.

3%) or a caf�� (32.6%). Associations between water pipe tobacco smoking and sociodemographic variables As indicated in Table 1, ever use was significantly associated only with gender and income. With regard to gender, men were substantially more likely to have ever used a water pipe to smoke tobacco (86.5% vs. 37.4%, p < .001). Use increased for the first three levels of income, peaking at 73.1% of those earning 500�C999 per year, but use dropped to 63.2% among the most wealthy. Use at least monthly was also significantly associated with gender and income (Table 1). Additionally, non-Jordanians were somewhat more likely to be smokers at least monthly (60.5% vs. 41.3%, p = .02). Table 2 displays unadjusted and adjusted multivariable logistic regression models that controlled for all covariates with a univariable relationship with the outcome of at least p < .

20. These analyses demonstrated significant associations between ever use and gender (for Brefeldin_A women compared with men, odds ratio [OR] = 0.11, 95% CI = 0.07�C0.17) and income (for those earning 500�C999 vs. those earning <250, OR = 2.37, 95% CI = 1.31�C4.31). A similar pattern was noted for the outcome of at least monthly use (see Table 2). When models were conducted using stepwise backward regression, results were similar in terms of levels of significance. Table 2.