摘要：Objectives. We examined smoking cessation rate by education and determined how much of the difference can be attributed to the rate of quit attempts and how much to the success of these attempts. Methods. We analyzed data from the National Health Interview Survey (NHIS, 1991–2010) and the Tobacco Use Supplement to the Current Population Survey (TUS-CPS, 1992–2011). Smokers (≥ 25 years) were divided into lower- and higher-education groups (≤ 12 years and > 12 years). Results. A significant difference in cessation rate between the lower- and the higher-education groups persisted over the last 2 decades. On average, the annual cessation rate for the former was about two thirds that of the latter (3.5% vs 5.2%; P < .001, for both NHIS and TUS-CPS). About half the difference in cessation rate can be attributed to the difference in quit attempt rate and half to the difference in success rate. Conclusions. Smokers in the lower-education group have consistently lagged behind their higher-education counterparts in quitting. In addition to the usual concern about improving their success in quitting, tobacco control programs need to find ways to increase quit attempts in this group. It is well established that smoking prevalence is much higher among those with lower education than among those with higher education. 1–6 However, the literature on the difference in cessation rate by education level is inconsistent. 7,8 Given that the smoking prevalence of any group is determined by the rate at which nonsmokers take up cigarettes and current smokers quit smoking, it is important to understand if the disparity in smoking prevalence comes from uptake or cessation or both. 9 This study examined cessation. Some studies have reported that smokers with less education find it more difficult to quit smoking. 4,10–13 It has also been suggested that the disparity in cessation rate by education has increased over time. 4 Other studies, however, have suggested that the smoking cessation rates are not significantly different between education groups. 8,14–18 These studies suggest that once people have become established smokers, they find it equally difficult to quit regardless of education level. One study even reported the reverse association between education and cessation; smokers with less education were more successful at quitting than were those with more education. 19 The inconsistency in these reports may stem partly from the use of different samples. Some studies were based on clinical samples 13,20,21 and others on population surveys. 11,14,15 Some had larger samples, 10,16 and others had relatively small samples. 4,17 In addition, some studies adjusted for covariates such as family or personal income 11,19 and motivation 14,15 in their analysis, whereas others did not. 13,17 These adjustments may help researchers understand what factors are correlated with education level, but they divert attention from the simpler question of whether a difference in the cessation rate is seen between education groups. In short, heterogeneity in study samples and analytical approaches contributed to inconsistencies in reports of whether cessation rates differed between groups with different levels of education. This study attempted to resolve this issue by analyzing data from 2 US nationally representative surveys with very large samples collected over 2 decades. The strength of large, nationally representative samples is the ability to provide statistically reliable estimates. Also, the long period of study allowed us to check for trends in the difference between education groups over time. We used 2 national surveys to determine whether the difference found in 1 survey can be replicated in the other. In addition, we separately examined the quit attempt rates and success rates of those quit attempts. We further quantified the difference in cessation rates, if any, by partitioning the difference into the difference in the rate of making quit attempts and the difference in the success of these quit attempts.
