Journal Information
Vol. 55. Issue 12.
Pages 654-656 (December 2019)
Vol. 55. Issue 12.
Pages 654-656 (December 2019)
Scientific Letter
Full text access
When should we measure self-efficacy as an aid to smoking cessation?
¿En qué momento deberíamos medir la autoeficacia para ayudar a dejar de fumar?
José Ignacio de Granda-Orivea,
Corresponding author

Corresponding author.
, José Francisco Pascual-Lledób, Segismundo Solano-Reinac, Santos Asensio-Sánchezb, Marcos García-Ruedad, Manuel Ángel Martínez-Muñize, Lourdes Lázaro-Aseguradof, Susana Luhningg, Daniel Bujulbasichh, Rogelio Pendinoh, Isabel Cienfuegos-Agustína, Carlos A. Jiménez-Ruizi
a Servicio de Neumología, Hospital Universitario 12 de Octubre, Madrid, Spain
b Servicio de Neumología, Hospital General Universitario de Alicante, Alicante, Spain
c Servicio de Neumología, Hospital General Universitario Gregorio Marañón, Madrid, Spain
d Servicio de Neumología, Hospital Carlos Haya, Málaga, Spain
e Servicio de Neumología, Hospital San Agustín, Avilés (Asturias), Spain
f Servicio de Neumología, Hospital Universitario de Burgos, Burgos, Spain
g Instituto Médico Humana, Centro Asistencial de Consulta Externa, Córdoba, Argentina
h Servicio de Neumología, Sanatorio Nuestra Señora del Rosario, Centro Médico IPAM, Rosario, Argentina
i Unidad Especializada de Tabaquismo de la Comunidad de Madrid, Madrid, Spain
Ver más
This item has received
Article information
Full Text
Download PDF
Tables (1)
Table 1. Frequency distribution of the categories of question 4 from the Richmond test and question 3 from the Khimji-Watts test, by outcome variable and comparison between success and failure.
Full Text
To the Editor,

In order to improve interventions to assist in smoking cessation (SC), the multiple facets that make up this addiction and the factors behind relapses must be better understood.1,2 Self-efficacy (SE), understood as a person’s belief in their ability to succeed in a particular situation, has been reliably associated with smoking abstinence and relapse, and is therefore an important target in anti-smoking interventions.3,4 The objective of our study was to assess the association between SE, measured at the baseline visit, and abstinence at 12 months.

To this end, we performed a multicenter observational study of consecutive patients who attended SC clinics between October 2014 and October 2015. Patient demographic variables and smoking status were collected. SE was assessed from 2 questions included in the Richmond and Khimji-Watts questionnaires for measuring SC motivation5 (Table 1). The statistical analysis was performed using the IBM SPSS 20.0 package for Windows. Logistic regression models were constructed to assess the association between SE and the outcome variable (SC success or failure), using the outcome variable as the dependent variable, the SE variable as the independent variable, and the remaining variables as control variables and adjustment. The level of statistical significance was set at p<0.05.

Table 1.

Frequency distribution of the categories of question 4 from the Richmond test and question 3 from the Khimji-Watts test, by outcome variable and comparison between success and failure.

Variable  Overall series  Outcomep 
  N (%)  Failure  Success   
Question 4 of the Richmond test: Do you think that you will still be a smoker in 6 months? (N=274)
Definitely not/perhapsa  52 (19.0)  19 (15.0)  33 (22.4)   
Yes  134 (48.9)  67 (52.8)  67 (45.6)   
Definitely yes  88 (32.1)  41 (32.3)  47 (32.0)  0.255 
Definitely not/perhaps  52 (19.0)  19 (15.0)  33 (22.4)   
Yes/definitely yes  222 (81.0)  108 (85.0)  114 (77.6)  0.115 
Question 3 of the Khimji-Watts test: What chance of success would you have if you tried to quit? (N=275)
Doubtful  27 (9.8)  17 (13.3)  10 (6.8)   
Medium  109 (39.6)  50 (39.1)  59 (40.0)   
Great  139 (50.5)  61 (47.7)  78 (53.1)  0.188 
Doubtful/medium  136 (49.5)  67 (52.3)  69 (46.9)   
Great  139 (50.5)  61 (47.7)  78 (53.1)  0.371 

N: sample size; p: degree of significance.


The categories of “definitely not” and “perhaps” were grouped, given the limited number of data points in the first category.

The study included 275 subjects, 130 men (47.3%), with an overall average age of 51.2 years (SD 10.7). In total, 53.5% (147 subjects) were successful in SC. The mean values of the different motivation tests were: visual analogue scale, 8 (1.9); Richmond questionnaire, 7.9 (1.5); the Henry Mondor Hospital test, 13 (2.7); and the Khimji-Watts questionnaire, 11.4 (2.5). The average overall degree of dependency, as measured by the Fagerström test (FTCD) was 5.9 (2.2), with no differences in the degree of dependency between those who quit smoking (mean 5.9, SD 2.2), and those who failed to quit (mean 5.9, SD 2.4). We found no significant differences in the analyzed variables between those who were successful in SC compared to those who were not, whether quantitative (age, age of onset, daily consumption of cigarettes, number of years smoked, accumulated consumption in pack-years, number of previous SC attempts, motivation, and FTCD questionnaires] or qualitative (sex, marital status, level of education, and employment status), except in educational level, where subjects with a secondary education had a 17.1% (95% CI: 4.8%–29.4%; p=0.004) greater rate of success, while those with university studies had a 12.4% (95% CI: 0.7%–24.0%; p=0.026) greater rate of failure. We did not find a statistically significant association between SE and the outcome variable (Table 1) in any of the possible logistic regression models constructed when the SE variable was examined both qualitatively and quantitatively, and or when controlling for sex, age, daily consumption of cigarettes, number of years smoked, number of previous attempts to quit, FTCD, and educational level. Nor did we find any differences in the multivariate analysis between men and women in the association between SE and the outcome variable. Cigarette dependence as measured by the FTCD score was not associated with the probability of SC. We conclude that neither SE nor the degree of dependence were predictive of the likelihood of SC success or failure, and only the educational level of participants showed a statistically significant association with the SC outcome.

The most important finding of our study is that we found no association between SE and SC. Gwaltney et al.3 found that SE is a poor predictor of abstinence when it is measured before making an attempt to quit (as observed in our study) and that the association is rather more robust if it is measured after “D-Day” (after cessation). SE varies in its ability to predict abstinence over time3,6: its association with abstinence is greater when measured 1 week after “D-Day”,3 but this relationship weakens when measured later; that is to say, SE provides a better prediction of proximal behavior.3 Many of the smokers who quit smoking have relapses and setbacks in their attempt to quit; thus, SE diminishes during these episodes, and a bidirectional association was identified between the degree of SE at the beginning and the time to the first relapse: a significant decrease in SE is followed by a return to consumption and vice versa.4,7–9 If the association between SE and abstinence is only modest, it is unlikely that fostering interventions in SE will have a substantial impact on final abstinence, because this factor clearly varies in response to the consumption of tobacco and is a better predictor of proximal conduct.3,10 Indeed, SE correlates positively with success in maintaining abstinence, but negatively in cessation attempts.11 Believing that “I can give up whenever I want” leads to a deferral of the decision to quit.11,12

In our study, the only variable that showed an association with abstinence at 1year was educational level: individuals with secondary education were more likely to quit than subjects with basic or university studies. In general, there is no unanimity in considering SE as a good predictor of abstinence. Nerin et al.13 found that smokers with basic studies among a Spanish population had higher cessation rates than those with average or higher educational levels, although statistical significance was not reached. Hymowitz et al.14 found no differences in the cumulative time spent in formal education; in contrast, Diemert et al.15 observed that individuals with a low educational level made more attempts to quit, and those with a high educational level achieved better rates of abstinence, as already reported by Holm et al.16 and Walker et al.17

Our study has several limitations: (1) we did not use specific scales for measuring SE; (2) the study might not have sufficient statistical power to identify differences; (3) our findings were collected from smokers who voluntarily attended our SC clinics, so would probably be more motivated; (4) the tests were performed in different settings and locations, which might not reflect what would happen if the test had been administered to the general population of smokers; and (5) the use of questionnaires in patients can elicit results that are not always accurate. All this variability could lead to different outcomes.

We have not been able to show that SE prior to cessation has a predictive association with SC success or failure. We believe, therefore, that SE should be measured after the SC intervention, when it is based on the successful execution of the SC attempt.


This study was supported with a SEPAR 2013 grant.

Conflict of interests

JIG-O has received honoraria for presentations, participation in clinical studies, and publications from (alphabetical order): Astra Zeneca, Esteve, Gebro, Menarini, Pfizer, and Rovi. CRC has received honoraria for presentations, participation in clinical studies, and publications from (alphabetical order): Esteve, Gebro, Menarini, and Pfizer. LL-A has received honoraria for presentations, participation in clinical studies, and publications from (alphabetical order): Astra-Zeneca, Boehringer, Chiesi, Esteve, Ferrer, Grifols, GSK, Menarini, Novartis, and Pfizer. SS-R has received honoraria for presentations, participation in clinical studies, and publications from (alphabetical order): Boehringer, Esteve, Pfizer, and Sandoz. The other authors state that they have no conflict of interests.

R. West.
ABC of smoking cessation. Assessment of dependence and motivation to stop smoking.
R. West.
The multiple facets of cigarette addiction and what they mean for encouraging and helping smokers to stop.
C.J. Gwaltney, J. Metrik, C.W. Kahler, S. Shiffman.
Self-efficacy and smoking cessation: a meta-analysis.
Psychol Addict Behav, 23 (2009), pp. 56-66
M. Clyde, A. Pipe, R. Reid, C. Els, H. Tulloch.
A bidirectional path analysis model of smoking cessation self-efficacy and concurrent smoking status: impact on abstinence outcomes.
M.Á Martínez-Muñiz, et al.
Is there an association between the degree of nicotine dependence and the motivation to stop smoking?.
C. Taniguchi, H. Tanaka, H. Saka, I. Oze, K. Tachibana, Y. Nozaki, et al.
Changes in self-efficacy associated with success in quitting smoking in participants in Japanese smoking cessation therapy.
Int J Nurs Pract, 24 (2018),
C.J. Gwaltney, S. Shiffman, M.H. Balabanis, J.A. Paty.
Dynamic self-efficacy and outcome expectancies: prediction of smoking lapse and relapse.
J Abnorm Psychol, 114 (2005), pp. 661-675
S. Shiffman.
Dynamic influences on smoking relapse process.
K.A. Perkins, C. Parzynski, M. Mercincavage, C.A. Conklin, C.A. Fonte.
Is self-efficacy for smoking abstinence a cause of, or a reflection on, smoking behavior change?.
Exp Clin Psychopharmacol, 20 (2012), pp. 56-62
A. Spaulding, D.R. Haley, M. Zhao.
Highly self-efficacious or simply delusional?.
Int J Health Sci, 2 (2014), pp. 01-17
J.J. Moreno Arnedillo, F. Herrero García de Osma.
Factores asociados al éxito en programas para dejar de fumar.
Adicciones, 12 (2000), pp. 329-341
J.R. Hughes, S. Naud.
Perceived role of motivation and self-efficacy in smoking cessation: a secondary data analysis.
Addict Behav, 61 (2016), pp. 58-61
I. Nerín, P. Novella, A. Cruzelaegui, A. Beamonte, N. Sobradiel, P. Gargallo.
Factores predictores de éxito a lo 6 meses en fumadores tratados en una unidad de tabaquismo.
Arch Bronconeumol, 40 (2004), pp. 558-562
N. Hymowitz, K.M. Cummings, A. Hyland, W.R. Lynn, T.F. Pechacek, T.D. Hartwell.
Predictors of smoking cessation in a cohort of adult smokers followed for five years.
Tob Control, (1997), pp. S57-62
L.M. Diemert, S.J. Bondy, K.S. Brown, S. Manske.
Young adult smoking cessation: predictors of quit attempts and abstinence.
Am J Public Health, 103 (2013), pp. 449-453
M. Holm, L. Schiöler, E. Andersson, B. Forsberg, T. Gislason, C. Janson, et al.
Predictors of smoking cessation: a longitudinal study in a large cohort of smokers.
Respir Med, 132 (2017), pp. 164-169
J.F. Walker, P.D. Loprinzi.
Longitudinal examination of predictors of smoking cessation in a national sample of U.S. adolescent and young adult smokers.
Nicotine Res Tob, 16 (2014), pp. 820-827

Please cite this article as: de Granda-Orive JI, et al. ¿En qué momento deberíamos medir la autoeficacia para ayudar a dejar de fumar? Arch Bronconeumol. 2019;55:654–656.

Copyright © 2019. SEPAR
Archivos de Bronconeumología
Article options

Are you a health professional able to prescribe or dispense drugs?