risk of cardiovascular, pulmonary, and neoplastic sequelae that can be associated with specific cancer treatments and aging. A limited number of randomized controlled trials have evaluated smoking cessation interventions in patients with cancer (Gritz et al., 1993; Wewers et al., 1994; Griebel et al., 1998; Browning et al., 2000; Sanderson Cox et al., 2002; Schnoll et al., 2003, 2005; Emmons et al., 2005). These interventions generally employ cognitive-behavioral counseling administered by health educators, nurses (Wewers et al., 1994; Griebel et al., 1998; Browning et al., 2000), dentists (Gritz et al., 1993), physicians (Gritz et al., 1993; Schnoll et al., 2003), or peers (Emmons et al., 2005). The interpretation of study results is limited by a variety of factors, however, including low statistical power (Stanislaw and Wewers, 1994; Wewers et al., 1994; Griebel et al., 1998; Browning et al., 2000), small sample size (Stanislaw and Wewers, 1994; Wewers et al., 1994; Griebel et al., 1998), high attrition rates (Gritz et al., 1993), and lack of long-term follow-up (Stanislaw and Wewers, 1994; Griebel et al., 1998; Schnoll et al., 2005). Consequently, results overall provide little or no evidence to support the effectiveness of behaviorally based smoking cessation interventions. Gritz and colleagues (1993) observed no difference in continuous abstinence rates at 12-month follow-up in patients with head and neck cancers randomized to receive standard advice to quit or surgeon-delivered smoking cessation counseling. Another study likewise found that quit rates did not differ among cancer patients who received standard smoking cessation counseling and those who received a brief smoking cessation intervention from their physician (Schnoll et al., 2003). In a third study, childhood cancer survivors randomized to receive peer-delivered smoking counseling with telephone follow-up were twice as likely to quit smoking as those who received self-help materials. However, the quit rate at 12-month follow-up for both groups was relatively modest (15 versus 9 percent), and the incremental cost of the intervention was substantial ($5,371 per additional quit) (Emmons et al., 2005).
Collectively, the available results of intervention trials in cancer populations, the well-established health risks associated with cancer and its treatment, and the morbidity associated with tobacco use support the need for more research aimed at developing effective, sustainable tobacco control interventions for cancer patients that take behavioral, psychological, and economic factors into account. In the interim, clinicians caring for patients with a past or present diagnosis of cancer should assess their smoking status and counsel those who smoke about the increased health risks they incur in doing so. This recommendation is based on the finding that among the population at large (i.e., without regard to having a particular diagnosis), individual face-to-face counseling by a trained therapist or nurse or brief advice from a physician can be effective in reducing smoking (Lancaster and Stead, 2004, 2005; Rice and Stead, 2004).