Self-efficacy as a pathway to long-term smoking cessation among low-income parents in the multilevel Kids Safe and Smokefree intervention

Abstract

Background: This study investigated the effects of a multi-level smoking intervention on mediators of long-term abstinence in parental smokers, including smoking cessation self-efficacy, smoking urge coping, and perceived support to quit smoking.

Methods: This is a secondary analysis of data from a randomized trial that recruited parental smokers from pediatric clinics in low-income communities (N = 327, 83% women, 83% African American, 79% below poverty level). Following clinical practice guidelines for tobacco intervention ("Ask, Advise, Refer" [AAR]), pediatricians asked all parents about child tobacco smoke exposure (TSE), advised about TSE harms and benefits of reducing TSE, and referred smokers to cessation resources. Eligible parents were then randomized to additional telephone-based smoking behavior counseling (AAR + counseling) or nutrition education (AAR + control). Bioverified 7-day point prevalence smoking abstinence and perceived counselor support were assessed at 12-month follow-up; cessation self-efficacy and urge coping were assessed at 3-month follow-up.

Results: Relative to AAR + control, AAR + counseling was associated with higher self-efficacy, urge coping, and perceived support to quit (all p's<.001). Self-efficacy, but no other mediators, had a significant positive effect on 12-month bioverified smoking abstinence (p < .001). The indirect effect of intervention on 12-month abstinence via self-efficacy suggested mediation via this pathway (p = .002).

Conclusion: Results suggest that all putative treatment pathways were improved more by the multi-level AAR + counseling than the clinic-level AAR + control intervention. Further, self-efficacy at end-of-treatment prospectively predicted long-term cessation, suggesting that building of self-efficacy through treatment may be key to sustained cessation.

Trial registration: ClinicalTrials.gov NCT01745393.

Keywords: Behavioral intervention; Mediation; Minority; Self-efficacy; Smoking cessation; Treatments.

Conflict of interest statement

Conflict of interest

No conflict declared.

Copyright © 2019 Elsevier B.V. All rights reserved.

Figures

Figure 1.

Hypothesized mediation model testing the direct and indirect effects of KiSS intervention participation on bioverified abstinence at 12-month follow-up (N = 327). Notes: T2 = 3-month end-of-treatment follow-up; T3 = 12-month follow-up. KiSS intervention: 0 = AAR + control vs. 1 = AAR + counseling. T3 Bioverified abstinence: 0 = smoker; 1 = abstinent. All estimates are unstandardized. Solid lines indicated significant pathways; dashed lines indicate non-significant pathways. *p < .05 **p < .01 ***p < .001