Is self-efficacy for smoking abstinence a cause of, or a reflection on, smoking behavior change?

Social learning theory considers self-efficacy as a causal factor in behavior change. However, in line with behavioral theory, recent clinical research suggests self-efficacy ratings may reflect, rather than cause, behavior change. To test these two disparate views, self-efficacy was related to actual smoking abstinence on the next day (i.e., self-efficacy causes change), and abstinence status over 1 day was tested as a predictor of rated self-efficacy for quitting the next day (i.e., reflects change). All data were from two similar crossover studies evaluating the short-term effects of both placebo versus medication, nicotine patch (n = 209) or varenicline (n = 123), on smoking abstinence during week-long practice quit attempts. Placebo and active medication periods were separated by an ad lib smoking washout, and analyses were controlled for prior-day's abstinence or self-efficacy values. Results were very consistent between studies in showing essentially bidirectional associations: daily self-efficacy predicted next-day's abstinence, and current-day's abstinence status predicted self-efficacy for abstinence the next day. However, secondary factors differentially predicted abstinence and, to a lesser extent, self-efficacy, between these two medication studies. These data provide some support for both social learning and behavioral theories of smoking behavior change, although self-efficacy may only briefly predict subsequent short periods of abstinence as assessed in these studies. Nonetheless, because self-efficacy has long been assumed to cause behavior change, including smoking cessation, the notion of self-efficacy as a reflection of recent smoking behavior change in these studies warrants greater attention in clinical research on smoking cessation treatment.