Preventive Medicine 63 (2014) 96–102

Contents lists available at ScienceDirect

Preventive Medicine journal homepage: www.elsevier.com/locate/ypmed

Gradual versus abrupt quitting among French treatment-seeking smokers Monique Baha a,⁎, Anne-Laurence Le Faou b,c a b c

Univ Paris Descartes, Sorbonne Paris Cité, 15 rue de l'Ecole de Médecine, F-75006 Paris, France Univ Paris Diderot, Sorbonne Paris Cité, EA 4069, Hôtel-Dieu 1, place du Parvis Notre-Dame, F-75004 Paris, France AP-HP, Hôpital Européen Georges Pompidou, 20 rue Leblanc, F-75015 Paris, France

a r t i c l e

i n f o

Available online 20 March 2014 Keywords: Quit attempt Gradual quitting Abrupt quitting Nicotine replacement therapy Smoking cessation Varenicline

a b s t r a c t Objective. This study examined the prevalence and predictors of gradual quitting among treatment-seeking smokers. Method. This study examined quit attempts among 28,156 adult smokers who attended French smoking cessation services nationwide between 2007 and 2010. Predictors of gradual quitting were determined using multivariate regression models. Results. Only 4.4% quit gradually whereas 48.7% quit abruptly and 46.9% continued smoking. 34.1% of abrupt quitters and 31.9% of gradual quitters were abstinent at 1 month post-quit (p = 0.108). Gradual quitting was associated with: older age, heavy smoking at baseline, no previous quit attempts, low self-efficacy, baseline intake of anxiolytics, symptoms of depression and history of depressive episodes. Gradual quitters had a similar anxiodepressive profile than continued smokers but were more educated and more likely to have reported previous quit attempts. Prescription of oral nicotine replacement therapy (NRT) only as opposed to combination NRT doubled the odds of gradual quitting. Likelihood of gradual quitting compared with continued smoking improved with the number of follow-up visits. Conclusion. Our findings suggest that hard-to-treat smokers may be more likely to quit gradually than abruptly. However, intense follow-up with adapted treatment appears to be crucial to achieve cessation gradually in French smoking cessation services. © 2014 Elsevier Inc. All rights reserved.

Introduction Until recently, smoking cessation guidelines in France, the United Kingdom and the United States did not recommend gradual quitting but rather encouraged cessation specialists to assist smokers in quitting abruptly (Fiore et al., 2008; Le Foll et al., 2005; National Institute for Health and Clinical Excellence, 2008). Yet, smokers in the general population appear to be interested in cutting down to quit. 43.5% of quit attempts reported in the 2011 English Smoking Toolkit Survey included cutting down first (West and Brown, 2012). In France, 44.7% of smokers interested in quitting plan to quit gradually (Guignard et al., 2013). Besides, between 2005 and 2010, the prevalence of smokers in the French general population has increased and the percentage of smokers interested in quitting has decreased from 64.8% to 57.6% (Guignard et al., 2013). Given smokers' difficulties to quit, the UK and French guidelines have been complemented in 2013 by tobacco harm reduction guidance encouraging gradual quitting among smokers unable to

⁎ Corresponding author at: Hôpital Européen Georges Pompidou, 20 rue Leblanc, 75015 Paris, France. Fax: +33 1 56 09 36 43. E-mail address: [email protected] (M. Baha).

http://dx.doi.org/10.1016/j.ypmed.2014.03.014 0091-7435/© 2014 Elsevier Inc. All rights reserved.

or not interested in quitting abruptly (Haute Authorité de Santé, 2013; National Institute for Health and Clinical Excellence, 2013). Literature seems to support the promotion of gradual quitting along with abrupt quitting. Indeed, a Cochrane meta-analysis of randomized studies comparing both methods of quitting found no significant difference in subsequent abstinence rates, regardless of use of pharmacotherapy (Lindson-Hawley et al., 2012). Nevertheless, data from the International Tobacco Control (ITC) Policy Evaluation Survey conducted in the United States, Canada, the UK and Australia revealed that abrupt quitters were twice more likely to achieve 1-month abstinence than gradual quitters (Cheong et al., 2007). These apparently mixed findings may be explained by the fact that in real life settings, gradual quitting is chosen by instead of randomly assigned to smokers with characteristics that may thus differ from those of abrupt quitters. The ITC Survey has shown that older smokers (≥40 years), women, smokers with low socioeconomic status as well as smokers who had reported high perceived difficulty to quit were more likely to quit gradually than abruptly (Siahpush et al., 2010). The ITC Survey also uncovered that users of smoking cessation medication and quitline services were more likely to quit gradually. In several European countries, marketing licences for oral forms of nicotine replacement therapy have been updated to include cutting down to quit

M. Baha, A.-L. Le Faou / Preventive Medicine 63 (2014) 96–102

(Beard et al., 2013; Zellweger et al., 2008). Yet, little literature exists on reduction versus abrupt quitting in real-life tobacco treatment settings (Jiménez-Ruiz et al., 2009; Samaan et al., 2012; Wee et al., 2011). In an effort to assess the effectiveness of French smoking cessation services in assisting smokers who experience the most difficulty quitting, the present study examined the proportion of treatment-seeking smokers who quit gradually, along with their profile. We focused on reduction by half because it has been reported to be the most common objective for smokers interested in cutting down to quit (Hughes et al., 2007). We determined predictors of gradual versus abrupt quitting, and then examined the association between method of quitting and abstinence at one month post-quit. Material and methods Population 230 French cessation services nationwide contributed data from 62,508 treatment-seeking smokers to the national smoking cessation on-line database (the “Consultation de tabacologie” — CDT programme). During their first visit in a cessation service between October 2007 and December 2010, smokers filled a standardised paper questionnaire. Anonymised questionnaires are registered in the database by staff in cessation services. This national programme has received the agreement of the French National Auditing Committee on Informatics and Individual Liberty. Over 90% of participating services are outpatient hospital-based cessation services. Staff usually includes at least a nurse and a physician. Smoking cessation training is offered to French health professionals through either a one-

97

year university degree or brief training. The content of training programmes is designed with respect to smoking cessation guidelines and has been presented elsewhere (Le Louarn et al., 2005; McLoughlin, 2006). In routine care, smokers are offered tailored counselling, support and educational information. A previous analysis of this database indicated that cessation specialists discuss and tailor an intervention plan with smokers during their first visit (Baha and Le Faou, 2009). Additionally, cessation specialists can prescribe nicotine replacement therapy (NRT — patch, oral forms or a combination of both) or varenicline. Cessation services are not free of charge. However, 9 out of 10 people in France have at least minimal health insurance, which covers 70% of the cost of cessation visits, along with a fixed partial coverage of NRT (50 euros per year per insured person). Data selection for the present analysis is detailed in Fig. 1. We examined retrospectively records from 28,156 adult smokers (aged ≥ 18 years, not pregnant) who were followed up (attended ≥ 2 visits) in cessation services. Half of smokers systematically registered during their first visit did not return for follow-up. A similar percentage of smokers defaulting from treatment without having set a quit date has also been observed in the UK, although not systematically monitored (Lowey et al., 2002). Table 1 presents a comparison of baseline measures between non-returning smokers and followed-up smokers. Measures In the standard questionnaire routinely used in French cessation services, smokers self-report socio-demographic information, current use of psychotropic medication, history of depression, tobacco-related information, and alcohol and cannabis consumption. To evaluate self-efficacy, smokers are asked to mark from 0 to 10 how confident they are in their ability to quit. Nicotine dependence is assessed using the Fagerström test (Heatherton et al., 1991). The French

62,508 smokers attended at least one visit in cessation services nationwide between October 2007 and December 2010 and were registered in the national database

For 96.8% (n=60,518), expired air carbon monoxide measures were available

84.6% (n = 51,208) smokers were aged ≥ 18, not pregnant still smokers at first visit (we excluded 4,839 who were abstinent for at least 24h at their first visit) and smoked manufactured or hand-rolled cigarettes (as opposed to other tobacco products)

55.0% (n = 28,156) smokers had returned

45.0% (n = 23,052) smokers did

for follow-up

not return for follow-up

(attended ≥ 2 visits)

(attended 1 visit)

Fig. 1. Selection of the study population extracted from the French national smoking cessation database among smokers registered between 2007 and 2010 in smoking cessation services nationwide.

98

M. Baha, A.-L. Le Faou / Preventive Medicine 63 (2014) 96–102

Table 1 Comparison between smokers who did not return for follow-up and smokers who attended ≥2 visits in French smoking cessation services nationwide between 2007 and 2010.

Women Men 18–44 years old ≥45 years old No education Low-level vocational education Secondary school Higher education Employed Retired Unemployed Inactive Trainee or student Disability pension Baseline intake of anxiolytics Baseline intake of antidepressants Symptoms of anxiety Symptoms of depression History of episodes of depression No previous attempt to quit One previous attempt to quit Two or more previous attempts to quit ≤10 cigarettes per day at first visit 11–20 cigarettes per day at first visit ≥21 cigarettes per day at first visit Fagerström score — mean (SD) Expired CO level at first visit — mean (SD) Self-efficacy to quit (1–10) — mean (SD) Use of cannabis during the last 12 months History of alcohol abuse Total

Smokers who returned for follow-up

Smokers who did not return for follow-up

Followed-up smokers vs. Non-returning smokers

n (%)

n (%)

OR (95% CI)

13,236 (47.0) 14,920 (53.0) 12,541 (44.5) 15,615 (55.5) 6192 (22.0) 8019 (28.5) 5834 (20.7) 8111 (28.8) 16,485 (58.5) 3567 (12.7) 2643 (9.4) 2254 (8.0) 550 (2.0) 2657 (9.4) 6210 (22.1) 5872 (20.9) 10,232 (36.3) 6247 (22.2) 8230 (29.2) 8476 (30.1) 8492 (30.2) 11,188 (39.7) 4588 (16.3) 13,967 (49.6) 9337 (33.2) 6.0 (2.4) 17.7 (13.4) 5.5 (2.6) 2984 (10.6) 7471 (26.5) 28,156 (100.0)

9876 (42.8) 13,176 (57.2) 10,609 (46.0) 12,443 (54.0) 6324 (27.5) 6823 (29.6) 4457 (19.3) 5448 (23.6) 13,095 (56.8) 2643 (11.5) 2546 (11.0) 2073 (9.0) 595 (2.6) 2100 (9.1) 4530 (19.7) 4115 (17.9) 8045 (34.9) 4918 (21.3) 5963 (25.9) 8231 (35.7) 7302 (31.7) 7519 (32.6) 4306 (18.7) 11,048 (47.9) 7465 (32.4) 5.8 (2.5) 14.3 (13.0) 5.3 (2.8) 2798 (12.1) 5846 (25.4) 23,052 (100.0)

Reference 0.85 (0.81–0.88) Reference 1.06 (1.02–1.11) Reference 1.16 (1.10–1.22) 1.26 (1.19–1.33) 1.41 (1.33–1.49) Reference 1.21 (1.14–1.29) 0.84 (0.79–0.90) 0.91 (0.85–0.97) 0.80 (0.71–0.91) 1.02 (0.95–1.09) 1.07 (1.02–1.13) 1.07 (1.01–1.13) 0.98 (0.94–1.03) 1.01 (0.96–1.06) 1.06 (1.01–1.11) Reference 1.07 (1.02–1.12) 1.31 (1.25–1.37) Reference 1.04 (0.99–1.10) 0.97 (0.91–1.04) 1.02 (1.01–1.03) 1.02 (1.01–1.02) 1.04 (1.03–1.05) 0.92 (0.86–0.97) 1.06 (1.01–1.11)

All independent bivariate associations were significant.

version of the CAGE test is used, with a score ≥ 2 suggesting alcohol abuse (Malet et al., 2005). The Hospital Anxiety and Depression scale (Zigmond and Snaith, 1983) is used to screen for symptoms of anxiety and depression with a threshold score of 11 for anxiety and 8 for depression. During follow-up visits, cessation specialists measure carbon monoxide (CO) levels in expired air, which verifies smokers' self-reported smoking status. Details of prescribed treatment and records of follow-up visits are routinely registered in the national database by staff in cessation services. By examining follow-up records, we classed as abrupt quitters those who had quit without first reducing their cigarette consumption during follow-up. Smokers who had reduced their baseline consumption by half before quitting were classed as gradual quitters. For both abrupt and gradual quitters, we examined follow-up records over a period of up to 4 weeks after the first recorded quit attempt and we only considered quit attempts validated by CO measures below 10 parts per million (ppm) (Judge et al., 2005). We determined point prevalence abstinence at one month post-quit date for individuals with CO-validated self-reports of no smoking since their previous visit. Although we used 7-day point prevalence abstinence, cessation specialists had also noted the length of abstinence since quitting for 83.4% of abstinent individuals (non-compulsory variable). The median was 45 days of prolonged abstinence. Self-reported gradual quitters for whom CO measures at follow-up were ≥10 ppm, smokers who had reduced their consumption without quitting and smokers who had neither reduced their consumption nor quit during followup were classed as continued smokers. They had been followed up for a median of 6 weeks since their first visit. Statistical analyses Chi-square tests for categorical data and analysis of variance for continuous data were used for descriptive bivariate analysis. Predictors of gradual quitting were assessed using forced-entry multivariate logistic regression models adjusted on baseline information. To examine the effect of cessation treatment, we used stepwise multivariate logistic regression models. These models were adjusted for baseline covariates for which associations

with abrupt quitting, gradual quitting and continued smoking had been significant in bivariate analysis. To examine the effect of gradual versus abrupt quitting on 1-month abstinence, we used a forced-entry multivariate logistic regression model adjusted for baseline covariates for which associations with 1-month abstinence had been significant in bivariate analysis. Two-tailed p-values ≤ 0.05 and confidence intervals (CI) of odds ratios (OR) not inclusive of unity were considered statistically significant. Calculations were performed using SAS software (version 9.2; SAS Institute, Cary, North Carolina, USA).

Results 48.7% (n = 13,709) of smokers followed up in French cessation services quit abruptly, 4.4% (n = 1251) quit gradually and 46.9% (n = 13,196) continued smoking.

Baseline predictors of gradual quitting Table 2 presents a descriptive comparison of baseline measures between gradual quitters, abrupt quitters and continued smokers. Sex did not significantly distinguish between the 3 categories and was therefore excluded from multivariate logistic regression models presented in Table 3. Gradual quitters were more likely to be aged ≥45 and to be retired than abrupt quitters or continued smokers (Table 3). Highly educated smokers (university degree) were more likely to quit gradually than to continue smoking. However, highly educated smokers were not more likely to quit abruptly than gradually. Likewise, comparing unemployment versus employment did not significantly distinguish gradual and abrupt quitters.

M. Baha, A.-L. Le Faou / Preventive Medicine 63 (2014) 96–102

99

Table 2 Baseline characteristics of abrupt quitters, gradual quitters and continued smokers among followed-up smokers registered in French smoking cessation services between 2007 and 2010.

Women Men 18–44 years old ≥45 years old No education Low-level vocational education Secondary school Higher education Employed Retired Unemployed Inactive Trainee or student Disability pension Baseline intake of anxiolytics Baseline intake of antidepressants Symptoms of anxiety Symptoms of depression History of episodes of depression No previous attempt to quit One previous attempt to quit Two or more previous attempts to quit ≤10 cigarettes per day at first visit 11–20 cigarettes per day at first visit ≥21 cigarettes per day at first visit Fagerström score — mean (SD) Expired CO level at first visit — mean (SD) Self-efficacy to quit (1–10) — mean (SD) Use of cannabis during the last 12 months History of alcohol abuse Total

Abrupt quitters

Gradual quitters

Gradual quitters vs. Abrupt quitters

Continued smokers

Gradual quitters vs. Continued smokers

n (%)

n (%)

p-Value

n (%)

p-Value

6517 (47.5) 7192 (52.5) 6258 (45.6) 7451 (54.4) 2650 (19.3) 3886 (28.4) 2880 (21.0) 4293 (31.3) 8726 (63.7) 1854 (13.5) 1064 (7.8) 915 (6.7) 225 (1.6) 925 (6.7) 2243 (16.4) 2168 (15.8) 4426 (32.3) 2369 (17.3) 3322 (24.2) 3319 (24.2) 4123 (30.1) 6267 (45.7) 2483 (18.1) 7149 (52.2) 3898 (28.4) 5.7 (2.4) 15.9 (13.0) 6.0 (2.6) 1151 (8.4) 3061 (22.3) 13,709 (100.0)

603 (48.2) 648 (51.8) 445 (35.6) 806 (64.4) 252 (20.1) 342 (27.3) 276 (22.1) 381 (30.5) 693 (55.4) 208 (16.6) 108 (8.6) 93 (7.4) 23 (1.8) 126 (10.1) 330 (26.4) 313 (25.0) 466 (37.2) 304 (24.3) 415 (33.2) 376 (30.1) 353 (28.2) 522 (41.7) 142 (11.3) 593 (47.4) 514 (41.1) 6.4 (2.3) 20.8 (13.4) 5.3 (2.5) 114 (9.1) 347 (27.7) 1251 (100.0)

Descriptive analysis indicated that all anxio-depressive indicators were more present among gradual quitters than abrupt quitters (Table 2). However, the corresponding multivariate logistic regression model only yielded significant adjusted odds-ratios for baseline intake of anxiolytics, symptoms of depression and history of depressive episodes (Table 3). Interestingly, gradual quitters' anxio-depressive profile did not differ significantly from that of continued smokers. Heavy smokers (≥ 21 cigarettes per day) were more likely to quit gradually than to quit abruptly (Table 3). Models presented in Table 3 also show that smokers who had made several previous quit attempts were more likely to quit abruptly than gradually and also more likely to quit gradually than to continue smoking. At baseline, gradual quitters had reported lower self-efficacy than abrupt quitters. History of alcohol abuse was more frequent among gradual quitters than abrupt quitters but the effect was no longer significant once adjusted in the multivariate model. Cannabis use and history of alcohol abuse were both more frequent among continued smokers (Table 2).

Association between prescription of cessation treatment, intensity of follow-up and method of quitting Abrupt quitters reported quitting on average by their second visit while gradual quitters reported quitting understandably later, on average at their fifth visit. Compared with gradual quitters, abrupt quitters had more often been prescribed nicotine patch only as well as combinations of nicotine patch and oral NRT (Table 4). Prescription of oral NRT only as opposed to combination NRT doubled the odds of gradual quitting versus abrupt quitting (Table 5). Gradual quitters were also more likely not to have been prescribed any pharmacotherapy (Table 5). We verified that there was no significant difference in the number of visits between

0.653 b0.0001 0.642

b0.0001

b0.0001 b0.0001 0.0009 b0.0001 b0.0001 b0.0001

b0.0001

b0.0001 b0.0001 b0.0001 0.383 b0.0001

6116 (46.3) 7080 (53.7) 5838 (44.2) 7358 (55.8) 3290 (24.9) 3791 (28.7) 2678 (20.3) 3437 (26.1) 7066 (53.5) 1505 (11.4) 1471 (11.2) 1246 (9.4) 302 (2.3) 1606 (12.2) 3637 (27.6) 3391 (25.7) 5340 (40.5) 3574 (27.1) 4493 (34.0) 4781 (36.2) 4016 (30.4) 4399 (33.4) 1963 (14.9) 6225 (47.2) 4925 (37.3) 6.2 (2.4) 19.2 (13.5) 5.1 (2.6) 1719 (13.0) 4063 (30.8) 13,196 (100.0)

0.209 b0.0001 b0.0001

b0.0001

0.371 0.600 0.007 0.009 0.532 b0.0001

0.0004

0.040 b0.0001 0.011 b0.0001 0.025

gradual quitters who had been prescribed pharmacotherapy and those who had not. Multivariate models showed no significant difference in NRT prescription as well as in the absence of pharmacotherapy prescription between gradual quitters and continued smokers (Table 5). The key difference was in the number of follow-up visits. Gradual quitters all attended at least 4 visits in a cessation service whereas 69.3% of continued smokers ended their follow-up after 2 or 3 visits (Table 4). A longer follow-up markedly reduced the odds of continued smoking (Table 5). Prescription of varenicline was significantly more frequent among gradual quitters than among continued smokers. Association between method of quitting and subsequent abstinence Overall, 53.3% (n = 14,960) of smokers followed up in French cessation services made a quit attempt. Among them, bivariate analysis showed no significant difference in abstinence rates at one month post-quit date between abrupt and gradual quitters: respectively 34.1% and 31.9% (p = 0.108). This result was confirmed by multivariate analysis. The odds-ratio of abstinence associated with gradual versus abrupt quitting was: 1.01 (95% CI 0.88–1.17). The multivariate regression model was adjusted on educational level, professional status, anxio-depressive indicators, previous quit attempts, Fagerström score, baseline cigarette consumption, cannabis use, baseline CO level, selfefficacy, pharmacotherapy and number of follow-up visits. Discussion Half of smokers followed up in French smoking cessation services made a quit attempt and a small proportion among them quit gradually. Predictors of gradual quitting versus abrupt quitting included: older age, indicators of anxiety and depression, heavy daily consumption

100

M. Baha, A.-L. Le Faou / Preventive Medicine 63 (2014) 96–102

Table 3 Multivariate odds-ratio estimates of baseline predictors of abrupt and gradual quitting among followed-up smokers registered in French smoking cessation services between 2007 and 2010. Gradual quitters vs. Abrupt quitters

18–44 years old ≥45 years old No education Low-level vocational education Secondary school Higher education Employed Retired Unemployed Inactive Trainee or student Disability pension Baseline intake of anxiolytics Baseline intake of antidepressants Symptoms of anxiety Symptoms of depression History of episodes of depression No previous attempt to quit One previous attempt to quit Two or more previous attempts to quit ≤10 cigarettes per day at first visit 11–20 cigarettes per day at first visit ≥21 cigarettes per day at first visit Fagerström score — mean (SD) Expired CO level at first visit — mean (SD) Self-efficacy to quit (1–10) — mean (SD) Use of cannabis during the last 12 months History of alcohol abuse

Gradual quitters vs. Continued smokers

OR (95% CI)

OR (95% CI)

Reference 1.50 (1.30–1.72) Reference 0.99 (0.83–1.18)

Reference 1.38 (1.20–1.59) Reference 1.12 (0.94–1.33)

1.07 (0.89–1.29) 1.13 (0.95–1.35) Reference 1.26 (1.05–1.50) 1.16 (0.93–1.44) 1.14 (0.90–1.45) 1.59 (1.01–2.51) 1.22 (0.99–1.52) 1.32 (1.12–1.57) 1.14 (0.96–1.36) 0.98 (0.86–1.12) 1.17 (1.01–1.37) 1.18 (1.03–1.37) Reference 0.76 (0.65–0.89) 0.74 (0.64–0.86)

1.26 (1.04–1.52) 1.39 (1.16–1.66) Reference 1.29 (1.08–1.55) 0.84 (0.68–1.05) 0.85 (0.67–1.08) 0.98 (0.63–1.54) 0.84 (0.68–1.04) 1.05 (0.89–1.24) 0.97 (0.81–1.15) 0.90 (0.79–1.03) 0.91 (0.78–1.06) 0.98 (0.85–1.13) Reference 1.05 (0.90–1.23) 1.36 (1.17–1.57)

Reference 1.14 (0.93–1.40) 1.41 (1.11–1.80) 1.03 (0.99–1.06) 1.02 (1.02–1.03)

Reference 1.22 (0.99–1.50) 1.33 (1.05–1.70) 1.02 (0.99–1.06) 1.01 (1.00–1.01)

0.93 (0.91–0.95)

1.03 (1.00–1.05)

1.18 (0.95–1.45)

0.82 (0.66–1.01)

1.11 (0.97–1.27)

0.85 (0.74–0.97)

All baseline variables except sex were entered in these multivariate logistic regression models.

prior to follow-up, no previous attempt to quit and low self-efficacy. Some of those predictors had also been identified in previous studies: namely older age and lower self-efficacy (Etter, 2011; Shiffman et al., 2007; Siahpush et al., 2010). Compared with gradual quitters, continued smokers were younger, less educated and less likely to have reported a history of previous quit attempts. According to a literature review on predictors of quit attempts in the general population, smokers who have made several previous attempts to quit are more likely to continue their efforts at smoking cessation (Vangeli et al., 2011). Therefore, despite low self-

efficacy perhaps enhanced by an anxio-depressive profile, gradual quitters may have been committed to quitting smoking albeit not in “one step”. Characteristics shared by both gradual quitters and continued smokers have been associated in literature with failure to quit: namely low self-efficacy (Vangeli et al., 2011), anxio-depressive symptoms (McClave et al., 2009), and cannabis use (Weinberger et al., 2013). Surprisingly, gradual quitters were more often heavy smokers at baseline than both abrupt quitters and continued smokers. Our findings thus illustrate that with a group of potentially hard-to-treat smokers, reduction prior to quitting may be a successful way to achieve abstinence. Gradual quitters were prescribed oral forms of NRT twice more often than abrupt quitters. Several randomized clinical trials have found that compared with placebo, oral forms of NRT (gum or inhaler) are an effective aid for achieving smoking reduction as well as gradual quitting (Batra et al., 2005; Rennard et al., 2006; Shiffman et al., 2009). Previous research has revealed that the odds of cessation improve with the number of follow-up visits (Dorner et al., 2011; Fiore et al., 2008). Unfortunately, continued smokers ended their follow-up significantly sooner than the two other groups of smokers. In our study, smokers who attended more than six follow-up visits were twice more likely to quit gradually than to continue smoking. To understand why gradual quitters attended more follow-up visits than continued smokers, we speculate that smokers may have chosen to quit gradually from the beginning of follow-up. If a treatment plan for gradual quitting was agreed upon between smokers and cessation specialists beforehand, it is possible that the latter may have been more proactive in encouraging attendance to follow-up visits. Moreover, varenicline which is indicated for use before the quit date (Fiore et al., 2008) had been prescribed to gradual quitters more often than to continued smokers. Abrupt quitting was the most prevalent method of quitting among smokers followed up in French smoking cessation services. Only 4.4% quit gradually. This was surprising given that in the general population, almost half of smokers interested in quitting plan to quit gradually (Guignard et al., 2013; Shiffman et al., 2007; West and Brown, 2012). A survey conducted in English smoking cessation services has revealed that 39.3% of tobacco cessation specialists promote abrupt quitting over gradual quitting and 49.2% promote abrupt quitting while allowing gradual quitting (Beard et al., 2012). The low percentage of gradual quitters in our study may therefore be linked to cessation specialists mainly promoting abrupt quitting. Yet, there was no significant difference in 1-month abstinence between abrupt and gradual quitters. Our findings thus agree with randomized clinical studies suggesting that both methods of quitting are equally effective in achieving abstinence (Lindson-Hawley et al., 2012). In agreement with more recent guidelines, cessation specialists should promote gradual quitting among smokers presenting with baseline determinants of failure to quit abruptly. Besides, considering the

Table 4 Prescribed pharmacotherapy and number of visits attended by followed-up smokers registered in French smoking cessation services between 2007 and 2010.

No tobacco treatment pharmacotherapy Prescription of oral NRT only Prescription of nicotine patch only Prescription of nicotine patch and oral NRT Prescription of varenicline 2–3 visits in cessation service 4–6 visits in cessation service ≥7 visits in cessation service Total

Abrupt quitters

Gradual quitters

Gradual quitters vs. Abrupt quitters

Continued smokers

Gradual quitters vs. Continued smokers

n (%)

n (%)

p-Value

n (%)

p-Value

2002 (14.6) 1387 (10.1) 2361 (17.2) 5945 (43.4) 1964 (14.3) 6632 (48.4) 5673 (41.4) 1404 (10.2) 13,709 (100.0)

226 (18.1) 248 (19.8) 148 (11.8) 452 (36.1) 173 (13.8) 0 (0.0) 572 (45.7) 679 (54.3) 1251 (100.0)

0.001 b0.0001 b0.0001 b0.0001 0.630 b0.0001

2471 (18.7) 2970 (22.5) 1722 (13.0) 4900 (37.1) 1100 (8.3) 9142 (69.3) 2807 (21.3) 1247 (9.4) 13,196 (100.0)

0.567 0.029 0.220 0.483 b0.0001 b0.0001

M. Baha, A.-L. Le Faou / Preventive Medicine 63 (2014) 96–102 Table 5 Multivariate odds-ratio estimates associated with prescribed pharmacotherapy and number of visits attended by followed-up smokers registered in French smoking cessation services between 2007 and 2010. Gradual quitters vs. Abrupt quittersa

No tobacco treatment pharmacotherapy Prescription of oral NRT only Prescription of nicotine patch only Prescription of nicotine patch and oral NRT Prescription of varenicline 2–3 visits in cessation service 4–6 visits in cessation service ≥7 visits in cessation service

Gradual quitters vs. Continued smokersb

OR (95% CI)

OR (95% CI)

1.56 (1.30–1.87)

0.86 (0.72–1.04)

2.55 (2.12–3.07) 0.91 (0.74–1.12) Reference

0.86 (0.72–1.04) 0.97 (0.78–1.19) Reference

1.40 (1.15–1.71) N/A Reference 4.29 (3.76–4.90)

1.77 (1.44–2.18) N/A Reference 2.91 (2.54–3.33)

101

Conflict of interest statement Dr Le Faou undertakes consultancy and research for and receives travel funds and hospitality from manufacturers of smoking cessation medications. Dr Le Faou and Dr Baha are currently participating in a research study funded by Pfizer Inc. (2012–2014).

Acknowledgments Funding for this study was provided by Pfizer for an InvestigatorInitiated Research (IIR) (grant number WS 1608433). Pfizer had no role in the study design, collection, analysis or interpretation of the data, writing of the manuscript, or the decision to submit the paper for publication. The authors wish to thank smoking cessation services contributing to the CDTnet programme and Nicolas Rodon for managing the national smoking cessation online database. References

a

The following baseline variables were excluded: sex, education level, and use of cannabis. b The following baseline variables were excluded: sex, baseline intake of anxiolytics, baseline intake of antidepressants, and history of episodes of depression.

issue with retention in smoking cessation services, informing potentially hard-to-treat smokers about the possibility of successfully quitting by cutting down first should be done during the first visit.

Strengths and limitations Our study adds to previous findings on gradual quitters by highlighting additional factors that distinguish them from abrupt quitters. Our study also pointed out what distinguish gradual quitters from continued smokers, thus contributing to the topic of retention in smoking cessation services, which is crucial in order to ensure subsequent quitting. Another strength of our analysis is that it was supported by a large dataset. Records of follow-up visits included CO measures, which allowed for biochemical validation of smoking status. Although training of cessation specialists is organised according to national guidelines, we do not know if treatment delivery and counselling are done homogeneously across all cessation services. Further research may be needed to evaluate variations in treatment delivery across French cessation services and their effect on outcomes. Nevertheless, a previous analysis on treatment effectiveness among 42 English stop smoking services, allowing for variation across services, showed that only 7 services had a result that significantly contradicted the mean effect (Brose et al., 2013). Contrary to what we expected based on literature, the percentage of gradual quitters was very low. This was not representative of the interest for gradual quitting among French smokers in the general population. We tested if a lower reduction threshold of 25% instead of 50% reduction would increase the percentage of gradual quitters: it did not increase the percentage above 5%.

Conclusion French treatment-seeking smokers who are aged ≥45, present with baseline heavy smoking, no previous quit attempts, low self-efficacy and anxio-depressive symptoms are more likely to quit gradually than abruptly. Cessation specialists should not hesitate to promote gradual quitting among smokers with such a hard-to-treat profile, since our findings corroborate that gradual quitting yields similar 1-month abstinence rates than abrupt quitting. Nevertheless, encouraging adherence to follow-up appears to be crucial to achieve successful cessation, especially when quitting gradually.

Baha, M.Y., Le Faou, A.L., 2009. Attitude towards cessation among French pregnant smokers: explaining the poor uptake of specialised support. Eur. J. Obstet. Gynecol. Reprod. Biol. 147 (1), 46–51. Batra, A., Klingler, K., Landfeldt, B., Friederich, H.M., Westin, A., Danielsson, T., 2005. Smoking reduction treatment with 4-mg nicotine gum: a double-blind, randomized, placebo-controlled study. Clin. Pharmacol. Ther. 78 (6), 689–696. Beard, E., McDermott, M., McEwen, A., West, R., 2012. Beliefs of stop smoking practitioners in United Kingdom on the use of nicotine replacement therapy for smoking reduction. Nicotine Tob. Res. 14 (6), 639–647. Beard, E., Bruguera, C., Brown, J., McNeill, A., West, R., 2013. Was the expansion of the marketing license for nicotine replacement therapy in the United Kingdom to include smoking reduction associated with changes in use and incidence of quit attempts? Nicotine Tob. Res. 15 (10), 1777–1781. Brose, L.S., West, R., Stapleton, J.A., 2013. Comparison of the effectiveness of varenicline and combination nicotine replacement therapy for smoking cessation in clinical practice. Mayo Clin. Proc. 88 (3), 226–233. Cheong, Y., Yong, H.H., Borland, R., 2007. Does how you quit affect success? A comparison between abrupt and gradual methods using data from the International Tobacco Control Policy Evaluation Study. Nicotine Tob. Res. 9 (8), 801–810. Dorner, T.E., Trostl, A., Womastek, I., Groman, E., 2011. Predictors of short-term success in smoking cessation in relation to attendance at a smoking cessation program. Nicotine Tob. Res. 13 (11), 1068–1075. Etter, J.F., 2011. Comparing abrupt and gradual smoking cessation: a randomized trial. Drug Alcohol Depend. 118 (2–3), 360–365. Fiore, M., Jaén, C., Baker, T., et al., 2008. Treating tobacco use and dependence: 2008 update. Clinical Practice Guideline (May, Retrieved October 2008, from http://www. surgeongeneral.gov/tobacco/treating_tobacco_use08.pdf). Guignard, R., Beck, F., Richard, J.-B., Peretti-Watel, P., 2013. [Smoking in France: data from the 2010 Health Barometer]. INPES, St-Denis. Haute Authorité de Santé, 2013. Arrêt de la consommation de tabac : du dépistage individuel au maintien de l'abstinence en premier recours. Heatherton, T.F., Kozlowski, L.T., Frecker, R.C., Fagerstrom, K.O., 1991. The Fagerstrom Test for Nicotine Dependence: a revision of the Fagerstrom Tolerance Questionnaire. Br. J. Addict. 86 (9), 1119–1127. Hughes, J.R., Callas, P.W., Peters, E.N., 2007. Interest in gradual cessation. Nicotine Tob. Res. 9 (6), 671–675. Jiménez-Ruiz, C.A., Ulibarri, M.M., Besada, N.A., Guerrero, A.C., Garcia, A.G., Cuadrado, A.R., 2009. Progressive reduction using nicotine gum as a prelude to quitting. Nicotine Tob. Res. 11 (7), 847–850. Judge, K., Bauld, L., Chesterman, J., Ferguson, J., 2005. The English smoking treatment services: short-term outcomes. Addiction 100 (Suppl. 2), 46–58. Le Foll, B., Melihan-Cheinin, P., Rostoker, G., Lagrue, G., 2005. Smoking cessation guidelines: evidence-based recommendations of the French Health Products Safety Agency. Eur. Psychiatry 20 (5–6), 431–441. Le Louarn, A., Jung, F., Kruchen, A., Quoix, E., 2005. An inter-university diploma on tobacco and smoking cessation: pedagogical evaluation and professional impact. Santé Publique 17 (2), 241–253. Lindson-Hawley, N., Aveyard, P., Hughes, J.R., 2012. Reduction versus abrupt cessation in smokers who want to quit. Cochrane Database Syst. Rev. 11, CD008033. Lowey, H., Fullard, B., Tocque, K., Bellis, M., 2002. Are Smoking Cessation Services Reducing Inequalities in Health? An Evaluation Study. North West Public Health Observatory. Liverpool John Moores University, Liverpool (July). Malet, L., Llorca, P.M., Reynaud, M., 2005. Systematic use in hospitals of a alcohol problems detecting questionnaire. The CAGE (cut down, annoyed, guilty, eye opener) questionnaire. Presse Med. 34 (7), 502–505. McClave, A.K., Dube, S.R., Strine, T.W., Kroenke, K., Caraballo, R.S., Mokdad, A.H., 2009. Associations between smoking cessation and anxiety and depression among U.S. adults. Addict. Behav. 34 (6–7), 491–497. McLoughlin, K., 2006. Assessment of smoking cessation training programs in Europe. Retrieved 30 January 2014, from http://www.ensh.eu/docs/71-assessment_of_ smoking_cessation_training_programs_in_europe_report.pdf (October). National Institute for Health and Clinical Excellence, 2008. Smoking cessation services in primary care, pharmacies, local authorities and workplaces, particularly for manual working groups, pregnant women and hard to reach communities. Public Health

102

M. Baha, A.-L. Le Faou / Preventive Medicine 63 (2014) 96–102

Guidance 10 (Retrieved 15 june 2013, from http://www.nice.org.uk/nicemedia/pdf/ PH010guidance.pdf). National Institute for Health and Clinical Excellence, 2013. Tobacco: harm-reduction approaches to smoking. Public Health Guidance 45 (Retrieved 11 october 2013, from http://www.nice.org.uk/nicemedia/live/14178/63996/63996.pdf). Rennard, S.I., Glover, E.D., Leischow, S., et al., 2006. Efficacy of the nicotine inhaler in smoking reduction: a double-blind, randomized trial. Nicotine Tob. Res. 8 (4), 555–564. Samaan, Z., Nowacki, B., Schulze, K., Magloire, P., Anand, S.S., 2012. Smoking cessation intervention in a cardiovascular hospital based clinical setting. Cardiovasc. Psychiatry Neurol. 2012, 970108. Shiffman, S., Hughes, J.R., Ferguson, S.G., Pillitteri, J.L., Gitchell, J.G., Burton, S.L., 2007. Smokers' interest in using nicotine replacement to aid smoking reduction. Nicotine Tob. Res. 9 (11), 1177–1182. Shiffman, S., Ferguson, S.G., Strahs, K.R., 2009. Quitting by gradual smoking reduction using nicotine gum: a randomized controlled trial. Am. J. Prev. Med. 36 (2), 96–104 (e101). Siahpush, M., Yong, H.H., Borland, R., Reid, J.L., 2010. Socioeconomic position and abrupt versus gradual method of quitting smoking: findings from the International Tobacco Control Four-Country Survey. Nicotine Tob. Res. 12 (Suppl.), S58–S63.

Vangeli, E., Stapleton, J., Smit, E.S., Borland, R., West, R., 2011. Predictors of attempts to stop smoking and their success in adult general population samples: a systematic review. Addiction 106 (12), 2110–2121. Wee, L.H., Shahab, L., Bulgiba, A., West, R., 2011. Conflict about quitting predicts the decision to stop smoking gradually or abruptly: evidence from stop smoking clinics in Malaysia. J. Smok. Cessat. 6 (01), 37–44. Weinberger, A.H., Pilver, C.E., Hoff, R.A., Mazure, C.M., McKee, S.A., 2013. Changes in smoking for adults with and without alcohol and drug use disorders: longitudinal evaluation in the US population. Am. J. Drug Alcohol Abuse 39 (3), 186–193. West, R., Brown, J., 2012. Smoking and Smoking Cessation in England 2011: Findings From the Smoking Toolkit Study. Pierre Fabre, London (www.smokinginengland. info). Zellweger, J.P., Coleman, T., Thomas, D., et al., 2008. Nicotine replacement therapies. Consensus of European experts on indications and contraindications. “Les Entretiens du Carla” 18th Conference (Pierre Fabre SA. Retrieved 15 june 2013, from http://www. entretiens-du-carla.com/datas/publications/17/edc18_17_web_17.pdf). Zigmond, A.S., Snaith, R.P., 1983. The Hospital Anxiety and Depression Scale. Acta Psychiatr. Scand. 67, 361–370.