Stopping smoking cessation treatments too soon may reduce odds of success for 45 percent of smokers

Their findings published online in the journal Addiction suggest that current treatment theories that maintain any smoking after the planned target quit day predicts treatment failure need to be expanded to take into account a more dynamic quitting process. The team's analysis points to two types of successful quitters: those who quit immediately and remain abstinent through the end of treatment and those who are "delayed" in attaining abstinence but achieve success by the end of treatment.

"In 'real-world' clinic settings, health care providers must decide whether or not to continue a specific treatment based on their clinical judgment and the published reports in the scientific literature. They can lose confidence that a specific cessation treatment is effective when the patient is unable to quit on the recommended target quit day or if the patient is unable to maintain total abstinence within the early weeks of treatment," said David Gonzales, Ph.D., the study's lead author and a senior clinical investigator in Pulmonary and Critical Care Medicine at the OHSU Smoking Cessation Center, OHSU School of Medicine.

"Patients also can become discouraged that a treatment is not working and worry about continuing to pay for treatments they believe do not work. As a result, cessation treatment for some patients may be discontinued before the prescribed treatment period is completed and the patient and/or the treatment considered a failure."

In this study, however, the data show a substantial proportion of smokers who became 'successful quitters' by the end of 12 weeks of treatment smoked in one or more weeks during the first eight weeks and were delayed in achieving a period of continuous abstinence. This was true of successful quitters treated with varenicline, bupropion and with counseling alone [placebo], Gonzales explained, and appears to be a previously unreported and natural pattern of quitting for motivated smokers who seek treatment to quit.

"Had treatment been interrupted or discontinued for these 'delayed quitters,' opportunities for achieving continuous abstinence could have been lost for up to 45 percent of quitters who were ultimately successful," Gonzales said.

Gonzales and colleagues analyzed data from two identically designed, published studies (Gonzales et al. JAMA 2006 and Jorenby et al. JAMA 2006) conducted between June 2003 and April 2005. Participants included 2,052 generally healthy adult smokers who randomly received either a smoking cessation drug -- varenicline or bupropion -- or a placebo for 12 weeks of treatment plus 40 weeks of follow-up. All participants received brief smoking cessation counseling at clinic visits and investigators were blinded to the treatment assignments.

Successful quitters were defined as smokers who achieved continuous abstinence, not even one puff, for the last four weeks of treatment (weeks nine through 12). Among successful quitters, two groups were identified: "immediate quitters," smokers who quit and remained abstinent from their target quit date through the end of week 12; and "delayed quitters," smokers who had periods of smoking prior to attaining continuous abstinence for at least the last four weeks of treatment.

The overall end-of-treatment quit rates for the two studies were previously shown to be higher for varenicline, but in this analysis, the researchers found cumulative continuous abstinence increased similarly for all treatments during weeks three through eight. They also found quitting patterns among delayed quitters were similar regardless of whether they took varenicline, bupropion or received counseling only (placebo).

While delayed quitters did not fare quite as well as immediate quitters following the end of active treatment, they still accounted for approximately one-third of those who remained continuously abstinent at 12 months regardless of treatment group.

"Based on these findings, we believe that treatment failure, or success for that matter, should not be assessed until the recommended period of treatment is completed. An analogy with antibiotic treatment, while not totally appropriate, is, nevertheless, a useful framework for illustrating some of the dynamics of the quitting process," explained Gonzales. "We know that some patients quit taking antibiotics when there is relief of symptoms [success] and others quit taking medication if symptoms don't seem to be resolving [failure]. In both cases discontinuing treatment prematurely risks treatment failure. Stopping smoking cessation treatment seems to have similar risks."

The take-home message for clinicians and patients, according to Gonzales, is that 'real-world' quit rates may be significantly increased by just continuing cessation treatments without interruption for patients who remain motivated to quit despite lack of success during the first eight weeks of treatment.