- Email: [email protected]
Motivational interviewing versus prescriptive advice for smokers who are not ready to quit
Patient Education and Counseling 83 (2011) 129–133
Contents lists available at ScienceDirect
Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou
Short communication
Motivational interviewing versus prescriptive advice for smokers who are not ready to quit Melinda F. Davis a,*, Dan Shapiro a, Richard Windsor b, Patrick Whalen a, Robert Rhode a, Hugh S. Miller a, Lee Sechrest a a b
University of Arizona, USA George Washington University Medical Center, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 September 2009 Received in revised form 14 April 2010 Accepted 28 April 2010
Objective: Smokers who are not ready to quit are a very difficult group to treat. Physicians, nurses, and nurse practitioners are in a unique position to encourage patients to quit smoking. However, the best approach to do so is not clear. Methods: A two-group randomized controlled trial with 218 pack-a-day precontemplative and contemplative smokers recruited from the community. The laboratory-based study was designed to simulate outpatient visits to general practitioners. Participants were randomized to a 15-min intervention to compare the effectiveness of brief motivational or prescriptive counseling by a health professional. Thirteen outcome variables included intentions to quit and verbal reports at 1 and 6 months with biological verification. A composite outcome measure was constructed to provide greater power to detect study differences. Results: Approximately 33% of the sample reported at least one 24-h quit period during the 6 months they were followed after the trial. Results suggest that while neither treatment was superior, there were subgroup differences. Participants in the motivational condition were also more likely to respond to follow-up calls. Conclusions and practice implications: Motivational interviewing and prescriptive advice were equally effective for precontemplative and contemplative smokers. Practitioners should use the method that appeals to them. ß 2010 Elsevier Ireland Ltd. All rights reserved.
Keywords: Smoking Stages of change Motivational interviewing Brief advice Treatment effectiveness evaluation Behavior change
1. Introduction Physicians and nurses are uniquely situated to have an important impact on the behavior of smokers. Even minor changes in quit attempts could have a significant effect on patient health. Those who are the farthest away from quitting, however, are the least likely to be urged to quit by health professionals [1,2]. How should health professionals counsel precontemplative or contemplative smokers? The literature on substance abuse and addiction presents two possible approaches; prescriptive or motivational. Prescriptive advice has been the dominant approach for smoking cessation, while motivational interviewing
* Corresponding author at: Department of Psychology, University of Arizona, 1503 E University Blvd., PO Box 210068, Tucson, AZ 85721-0068, USA. Tel.: +1 520 626 7820; fax: +1 520 621 9306. E-mail addresses: [email protected], [email protected] (M.F. Davis), [email protected] (D. Shapiro), [email protected] (R. Windsor), [email protected] (P. Whalen), [email protected] (R. Rhode), [email protected] (H.S. Miller), [email protected] (L. Sechrest). 0738-3991/$ – see front matter ß 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pec.2010.04.024
encourages patient participation in health care. In the prescriptive approach, health professionals maintain a firm, authoritative approach with smokers. Some patients want health professionals to take charge and firmly tell them what to do. Research suggests that this is particularly true of older, poorer, and less educated patients [3,4]. Motivational interviewing (MI) seeks to establish a supportive and empathic alliance [5–7]. Ambivalence is acknowledged as part of the normal change process. The patient takes responsibility for major decisions about care [8]. Practice guidelines for health professionals tend to adopt major facets of either the prescriptive or motivational enhancement approaches to smoking cessation [9,10]. 1.1. Review of motivational interviewing Meta-analyses on motivational interviewing have reached different conclusions regarding its effectiveness for smoking reduction [11,12]. We identified eight randomized controlled trials comparing MI against prescriptive advice [13]. While most of
130
M.F. Davis et al. / Patient Education and Counseling 83 (2011) 129–133
the trials found some support for the MI condition, the type of treatment (MI versus prescriptive advice) was confounded with amount of treatment, making it nearly impossible to disentangle the results [14–19]. Only two studies had equal intervention times [20,21]. At the end of a community-based study, any observed effects may be due to the treatment or study characteristics. In this lab-based study, we expected MI to be superior to the prescriptive approach for precontemplative and contemplative smokers. However, there were two other possible outcomes of the study. The methods may be equivalent; or, one method may work best for identified sub-groups. 2. Methods 2.1. Study design Smokers were recruited through advertisements and direct recruitment for a study designed ‘‘to investigate health professional communication with smokers’’. A detailed screening protocol asked smokers to indicate their current quit plans before reimbursement was discussed. Smokers ready to quit were not enrolled in the study. For participation, subjects were offered $25. In the lab, participants completed informed consent, baseline assessments, and were randomized to receive either a 15-min motivational or prescriptive interview. The intervention was designed to match the time available in the average health professional–patient interaction. Smokers who made a plan to quit or reduce were phoned the day prior to their quit/reduction day. Follow-ups at 1 and 6 months were completed by phone. Subjects reporting significant reductions were invited into the lab to provide urine samples. 2.1.1. Fidelity of interventions This efficacy study was designed to be a strong test of MI and prescriptive advice. Neither treatment had a ‘‘home court’’ advantage in that expert consultants provided training for the interventions (RR and RW). All tapes were reviewed for adherence to the protocol and weekly meetings were held with the study nurses. Sessions not reaching criterion were removed from the analyses. 2.2. Assessments The Fagerstrom Test of Nicotine Dependence [22], the Smoking Hazards Scale [23], self-reported smoking behavior, and urine cotinine were collected at baseline, 1 and 6 months. Smoking history and spirometry were collected at baseline.
Smoking cessation was defined as cotinine less than 100 mm; significant reduction was defined as a decrease of 50% which was biochemically verified. Participants lost to follow-up were assumed not to have changed. Rasch analyses were used to create a continuous measure from the 13 outcomes and to reflect the composite smoking reduction score [24]. Rasch modeling is a probabilistic model that makes use of all available data to jointly estimate an individual’s progress towards smoking cessation and calibrate items [25]. Individuals are at different levels in the quitting process, and smoking items can be regarded as hurdles of varying heights. For example, quits are harder to achieve than reductions, and both are much harder to achieve than planning to quit. To receive the maximum score of twelve, a smoker needed to be abstinent at both the 1-month and the 6-month follow-up. 2.3. Analyses Demographic characteristics and outcomes were examined using t-tests and x2 statistics. A generalized linear model was the primary outcome analysis. The dependent variable was the composite outcome measure for smoking reduction. The independent variables and their order of entry were (1) gender, (2) age, (3) ethnicity, (4) cigarettes per day (to assure the groups were similar at baseline), (5) Treatment assignment (to evaluate the differential treatment effect), and interaction terms (to examine subgroup differences). 3. Results Two hundred and thirty precontemplative and contemplative smokers were randomized to the brief treatments and 218 cases were included in the analyses. Of the 218 smokers, 71% were available at 1 month, and 56% at 6 months (Table 1). 3.1. Equivalency of groups at baseline The two groups were comparable at baseline on age, gender, total years smoked, age at first cigarette, lifetime packs and spirometry (Table 2). The average Fagerstrom score was 6.2, indicating the study participants had a high dependence on nicotine. There were no observed relationships between risk perception and smoking reduction. Caucasians were overrepresented in the motivational group. Smokers in the prescriptive group had higher Fagerstrom scores (Table 2). 3.2. Follow-up status by treatment group
2.2.1. Outcome measures The study included 13 indicators of intentions to quit, selfreport, and biological verification (see Table 3). All interventions were videotaped and coded for intentions to reduce or quit.
There was no difference in follow-up status at 1 month. However, 63% of the motivational group responded to the 6-month follow-up, compared to 49% of the prescriptive group.
Table 1 Recruitment and retention of participants by treatment group. Procedures
Total n
Prescriptive approach
Motivational interviewing
Randomized to treatment Discontinued during treatment Excluded from the analysesa One-month follow-upb Six-month follow-upc Both one- and six-month follow-up Included in the analysis
230 0 12 154 122 111 218
114 0 5 77 53 51 109
116 0 7 77 69 61 109
a One case did not meet inclusion criteria, two cases did not meet the treatment standard, one individual was deceased, and the remaining nine cases had missing baseline data. b One-month follow-up (x2(1,n = 218) = 0.0, p = 1.00). c Six-month follow-up (x2(1,n = 218) = 4.77, p = .03).
M.F. Davis et al. / Patient Education and Counseling 83 (2011) 129–133
131
Table 2 Baseline characteristics by treatment assignment (n = 218). Prescriptive approach
Motivational interviewing
Total
Gender (n, %) Males Females
61 (56%) 48 (44%)
59 (54%) 50 (46%)
120 (55%) 98 (45%)
x2 = 0.74
.785
Ethnicity (n, %) Caucasian Hispanic African American Other
74 16 10 9
92 7 1 8
166 23 11 17
x2 = 8.18
.003
Fagerstrom score (M, SD) Age started smoking (M, SD) Years smoked (M, SD) Cigarettes smoked a day (M, SD) Average age (M, SD) Normalized spirometry percent (M, SD)
6.6 13.7 22.4 25.4 38.2 78.5
t = 2.64 t = 0.37 t = 1.37 t = 1.33 t = 0.74 t = 1.11
.009 .710 .174 .186 .460 .267
(68%) (15%) (9%) (8%) (2.2) (4.4) (11.6) (13.1) (12.4) (21.0)
(85%) (7%) (1%) (7%)
5.8 13.9 20.1 25.4 37.0 81.5
(2.1) (4.2) (12.3) (8.4) (11.7) (18.5)
p
(76%) (11%) (5%) (8%)
6.2 13.8 21.1 25.4 37.6 80.0
(2.2) (4.3) (12.0) (11.0) (12.0) (19.8)
x2 tests were used for dichotomous variables; t-tests were used for continuous variables.
Table 3 Intention, verbal report, and biologically verified smoking outcomes by treatment assignment (n = 218).
Indicates intention to quit within 6 months Indicates intention to reduce within 1 month Indicates intention to quit within 1 month Indicates intention to reduce within 1 week of intervention Indicates intention to quit within 1 week of intervention Verbal report—one 24-h 50% reduction at 1 or 6 months Verbal report—one 72-h 50% reduction at 1 or 6 months Verbal report—one 24-h quit at 1 or 6 months Verbal report—one 72-h quit 1 or 6 months Biologically verified 50% reduction at 1 or 6 months Biologically verified 50% reduction at 1 and 6 months Biologically verified quit at 1 or 6 months Biologically verified quit at 1 and 6 months Composite smoking reduction score (Mean, SD) a
Prescriptive n (%)
Motivational n (%)
Total n (%)
x2
p
58 (53) 54 (50) 18 (17) 35 (32) 1 (1) 44 (40) 36 (33) 29 (27) 6 (6) 17 (16) 5 (5) 5 (5) 0 (0) 6.16 (1.5)
60 (55) 47 (43) 25 (23) 26 (24) 2 (2) 44 (40) 32 (29) 28 (26) 4 (4) 19 (17) 2 (2) 7 (6) 1 (1) 6.09 (1.5)
118 (54) 101 (46) 43 (20) 61 (28) 3 (1) 88 (40) 68 (31) 57 (26) 10 (5) 36 (17) 7 (3) 12 (6) 1 (1) 6.13 (1.5)
0.07 0.90 1.42 1.84 –a 0.00 0.34 0.24 0.42 0.13 –a 0.35 –a 0.33
.79 .34 .23 .18 1.00 1.00 .56 .88 .52 .72 .45 .55 1.00 .75
Fishers exact test was used when estimated cell counts were below 5.
3.3. Behavioral and biological outcomes On follow-up, 26% reported at least one 24-h quit and 17% had a biologically verified reduction of 50% or more. Six percent of the sample had a biologically verified quit during follow-up. There were no differences by treatment group assignment on any outcome measure (Table 3). Hierarchical regression was used to evaluate the effect of the intervention with the composite smoking measure. Gender, age, ethnicity (Caucasian), and number of cigarettes smoked at baseline
Table 4 Hierarchical regression model to predict smoking reduction (n = 218). Predictor
F
p
Gender Age Caucasian Number of cigarettes a day at baseline Treatment group Treatment gender Treatment age Treatment Caucasian Treatment cigarettes a day at baseline Treatment gender age Treatment gender Caucasian Treatment gender cigarettes a day at baseline Treatment age Caucasian Treatment age cigarettes a day at baseline Treatment Caucasian cigarettes a day at baseline
10.88 0.22 0.00 1.70 0.28 0.09 0.24 0.45 0.27 3.02 3.19 1.39 0.35 4.86 1.01
.001 .64 .99 .19 .60 .77 .62 .51 .60 .05 .04 .25 .70 .009 .37
were entered first, to assure that the treatment and control groups were similar at baseline. There was no effect of age, Caucasian ethnicity, or number of cigarettes smoked at baseline on the composite smoking measure. However, there was an effect of gender; women showed greater improvement regardless of treatment condition. There was no differential effect of treatment; participants in the prescriptive and motivational groups did equally well (Table 4). Several interaction terms were also significant. Caucasian women showed greater progress towards smoking reduction than Caucasian males, while non-Caucasian men and women had better responses to MI (Fig. 1). Older, heavier smokers showed better progress towards smoking reduction in MI (Fig. 2). Older females showed greater progress towards cessation in MI (Fig. 3). The study
Fig. 1. Smoking progress: gender, ethnicity and treatment assignment.
132
M.F. Davis et al. / Patient Education and Counseling 83 (2011) 129–133
away, while an empathic approach may keep the door open for future interventions. 4.2. Conclusion
Fig. 2. Smoking progress: age, baseline smoking and treatment assignment.
Medical visits provide a window of opportunity to offer brief smoking interventions. The largest contribution of this study may be the methodological innovation of evaluating more than cessation rates when focusing on heavily addicted smokers; the composite measure enabled us to examine subgroup differences that were not detected with dichotomous outcomes. 4.3. Practice implications Few trials have attempted to encourage cessation in precontemplative and contemplative, heavily addicted smokers. Neither approach was superior overall. These results indicate that, in general, health professionals should use the intervention that appeals to them. We speculate that MI may provide an added benefit in conjunction with other smoking cessation treatment components. Acknowledgement Supported by a grant from The Arizona Disease Control Research Commission.
Fig. 3. Smoking progress: gender, age (<40 and >40) and treatment assignment.
References was powered to detect a 15% difference in proportions in selfreported quit rates. Because women showed greater gains in this study, we examined gender differences across the thirteen outcome variables. Women were more likely to state they would quit within 6 months (x2(1,n = 218) = 5.99, p = .01), report at least one 24-h 50% reduction (x2(1,n = 218) = 5.49, p < .02), at least one 3-day 50% reduction (x2(1,n = 218) = 7.68, p = .006), and at least one 24-h quit at 1 or 6 months (x2(1,n = 218) = 6.74, p = .009). 4. Discussion and conclusion 4.1. Discussion This study focused on smokers not ready to quit, a group that comprises nearly 90% of smokers. Our primary study hypothesis, that motivational interviewing would increase biologically verified quits by 15%, was not supported by the data. MI and prescriptive advice were equivalent in this 15-min intervention designed to simulate the real pressures faced by health care professionals. However, 33% of the subjects reported making a 24-h quit attempt during the 6 months of the trial, and women showed a greater response than males to both interventions. These results demonstrate the usefulness of brief smoking cessation interventions. Smokers make multiple quit attempts before succeeding [26], and any attempt to quit represents progress. We found modest subgroup differences. Non-Caucasian men and women showed greater progress in the MI condition; Caucasians responded equally well to both treatments, and women showed greater progress than men. Results from a meta-analysis [21] reported larger effect sizes for MI in ethnic minority populations. Older, heavier smokers and older women showed greater reductions in the MI condition; these interactions may be due to differences in education or status levels. More smokers were lost to follow-up in the prescriptive condition. A more authoritative approach may drive participants
[1] Sesney JW, Kreher NE, Hickner JM, Webb S. Smoking cessation interventions in rural family practices: an UPRNet study. J Fam Practice 1997;44:578–85. [2] Pollak KI, McBride CM, Scholes D, Grothaus LC, Civic D, Curry SJ. Women’s reports of smoking cessation advice during reproductive health visits and subsequent smoking cessation. Am J Manag Care 2002;8:837–44. [3] Pendelton L, House WC. Preferences for treatment approaches in medical care: college students versus diabetic outpatients. Med Care 1984;22:644–6. [4] Strull WM, Lo B, Charles G. Do patients want to participate in medical decision making? J Amer Med Assoc 1984;252:2990–4. [5] Rollnick S, Healther N, Bell A. Negotiating behaviour change in medical settings: the development of brief motivational interviewing. J Ment Health 1992;1:25–37. [6] Rollnick S, Miller WR. What is motivational interviewing? Behav Cogn Psychother 1995;23:325–34. [7] Schneider RJ, Casey J, Kohn R. Motivational versus confrontational interviewing: a comparison of substance abuse assessment practices at employee assistance programs. J Behav Health Ser R 2000;27:60–74. [8] Kaplan SH, Greenfield S, Martin E. Expanding patient involvement in care: the epic program. Washington, DC: United States Public Health Service; 1983. [9] The Agency for Health Care Policy and Research. Smoking cessation clinical practice guideline. J Amer Med Assoc 1996;275:1270–80. [10] American Psychiatry Association. Practice guidelines for the treatment of patients with nicotine dependence. Am J Psychiat 1996;153:S1–31. [11] Hettema J, Steele J, Miller WR. Motivational interviewing. Annu Rev Clin Psychol 2005;1:91–111. [12] Rubak S, Sandbæk A, Lauritzen T, Christensen B. Motivational interviewing: a systematic review and meta-analysis. Brit J Gen Pract 2005;55:305–12. [13] Wagner C, Conners W. MI Bibliography 1983–2009. Retrieved September 1, 2009 from http://motivationalinterview.org/library/biblio.html. [14] Butler CC, Rollnick S, Cohen D, Russell I, Bachmann M, Stott N. Motivational consulting versus brief advice for smokers in general practice: a randomised trial. Brit J Gen Pract 1999;49:611–6. [15] Borrelli B, Novak S, Hecht J, Emmons K, Papandonatos G, Abrams D. Home health care nurses as a new channel for smoking cessation treatment: outcomes from project CARES (Community-Nurse Assisted Research and Education on Smoking). Prev Med 2005;41:815–21. [16] Colby SM, Monti PM, Barnett NP, Rohenow DJ, Weissman K, Spirito A, et al. Brief motivational interviewing in a hospital setting for adolescent smoking: a preliminary study. J Consult Clin Psych 1998;66:574–8. [17] Colby SM, Monti PM, O’Leary TT, Barnett NP, Spirito A, Rohsenow DJ, et al. Brief motivational intervention for adolescent smokers in medical settings. Addict Behav 2005;30:865–74. [18] Haug NA, Svikis DS, DiClemente C. Motivational enhancement therapy for nicotine dependence in methadone-maintained pregnant women. Addict Behav 2004;18:289–92. [19] Soria R, Legido A, Escolano C, Lopez-Yeste A, Montoya J. A randomised controlled trial of motivational interviewing for smoking cessation. Brit J Gen Pract 2006;56:768–74.
M.F. Davis et al. / Patient Education and Counseling 83 (2011) 129–133 [20] Steinberg ML, Ziedonis DM, Krejci JA, Brandon TH. Motivational interviewing with personalized feedback: a brief intervention for motivating smokers with schizophrenia to seek treatment for tobacco dependence. J Consult Clin Psych 2004;72:723–8. [21] Ahluwalia JS, Okuyemi K, Nollen N, Choi WS, Kaur H, Pulvers K, et al. The effects of nicotine gum and counseling among African American light smokers: a 2 2 factorial design. Addiction 2006;101:883–91. [22] Fagerstrom KO, Schneider NG. Measuring nicotine dependence: a review of the Fagerstrom Tolerance Questionnaire. J Behav Med 1989;12:159–82.
133
[23] Davis MF, Shapiro D, Sechrest L. The smoking hazards scale. J Ariz Nevada Acad Sci 2006;38:102–6. [24] Davis MF, Sechrest L, Shapiro D. Measuring progress toward smoking cessation. J Appl Meas 2005;6:164–72. [25] Rasch G. Probabilistic models for some intelligence and attainment tests. Copenhagen: Danish Institute for Educational Research. (1960, Expanded edition, 1980). Chicago: University of Chicago Press; 1980. [26] Hughes JR. Motivating and helping smokers to stop smoking. J Gen Intern Med 2003;18:1053–7.
Contents lists available at ScienceDirect
Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou
Short communication
Motivational interviewing versus prescriptive advice for smokers who are not ready to quit Melinda F. Davis a,*, Dan Shapiro a, Richard Windsor b, Patrick Whalen a, Robert Rhode a, Hugh S. Miller a, Lee Sechrest a a b
University of Arizona, USA George Washington University Medical Center, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 4 September 2009 Received in revised form 14 April 2010 Accepted 28 April 2010
Objective: Smokers who are not ready to quit are a very difficult group to treat. Physicians, nurses, and nurse practitioners are in a unique position to encourage patients to quit smoking. However, the best approach to do so is not clear. Methods: A two-group randomized controlled trial with 218 pack-a-day precontemplative and contemplative smokers recruited from the community. The laboratory-based study was designed to simulate outpatient visits to general practitioners. Participants were randomized to a 15-min intervention to compare the effectiveness of brief motivational or prescriptive counseling by a health professional. Thirteen outcome variables included intentions to quit and verbal reports at 1 and 6 months with biological verification. A composite outcome measure was constructed to provide greater power to detect study differences. Results: Approximately 33% of the sample reported at least one 24-h quit period during the 6 months they were followed after the trial. Results suggest that while neither treatment was superior, there were subgroup differences. Participants in the motivational condition were also more likely to respond to follow-up calls. Conclusions and practice implications: Motivational interviewing and prescriptive advice were equally effective for precontemplative and contemplative smokers. Practitioners should use the method that appeals to them. ß 2010 Elsevier Ireland Ltd. All rights reserved.
Keywords: Smoking Stages of change Motivational interviewing Brief advice Treatment effectiveness evaluation Behavior change
1. Introduction Physicians and nurses are uniquely situated to have an important impact on the behavior of smokers. Even minor changes in quit attempts could have a significant effect on patient health. Those who are the farthest away from quitting, however, are the least likely to be urged to quit by health professionals [1,2]. How should health professionals counsel precontemplative or contemplative smokers? The literature on substance abuse and addiction presents two possible approaches; prescriptive or motivational. Prescriptive advice has been the dominant approach for smoking cessation, while motivational interviewing
* Corresponding author at: Department of Psychology, University of Arizona, 1503 E University Blvd., PO Box 210068, Tucson, AZ 85721-0068, USA. Tel.: +1 520 626 7820; fax: +1 520 621 9306. E-mail addresses: [email protected], [email protected] (M.F. Davis), [email protected] (D. Shapiro), [email protected] (R. Windsor), [email protected] (P. Whalen), [email protected] (R. Rhode), [email protected] (H.S. Miller), [email protected] (L. Sechrest). 0738-3991/$ – see front matter ß 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pec.2010.04.024
encourages patient participation in health care. In the prescriptive approach, health professionals maintain a firm, authoritative approach with smokers. Some patients want health professionals to take charge and firmly tell them what to do. Research suggests that this is particularly true of older, poorer, and less educated patients [3,4]. Motivational interviewing (MI) seeks to establish a supportive and empathic alliance [5–7]. Ambivalence is acknowledged as part of the normal change process. The patient takes responsibility for major decisions about care [8]. Practice guidelines for health professionals tend to adopt major facets of either the prescriptive or motivational enhancement approaches to smoking cessation [9,10]. 1.1. Review of motivational interviewing Meta-analyses on motivational interviewing have reached different conclusions regarding its effectiveness for smoking reduction [11,12]. We identified eight randomized controlled trials comparing MI against prescriptive advice [13]. While most of
130
M.F. Davis et al. / Patient Education and Counseling 83 (2011) 129–133
the trials found some support for the MI condition, the type of treatment (MI versus prescriptive advice) was confounded with amount of treatment, making it nearly impossible to disentangle the results [14–19]. Only two studies had equal intervention times [20,21]. At the end of a community-based study, any observed effects may be due to the treatment or study characteristics. In this lab-based study, we expected MI to be superior to the prescriptive approach for precontemplative and contemplative smokers. However, there were two other possible outcomes of the study. The methods may be equivalent; or, one method may work best for identified sub-groups. 2. Methods 2.1. Study design Smokers were recruited through advertisements and direct recruitment for a study designed ‘‘to investigate health professional communication with smokers’’. A detailed screening protocol asked smokers to indicate their current quit plans before reimbursement was discussed. Smokers ready to quit were not enrolled in the study. For participation, subjects were offered $25. In the lab, participants completed informed consent, baseline assessments, and were randomized to receive either a 15-min motivational or prescriptive interview. The intervention was designed to match the time available in the average health professional–patient interaction. Smokers who made a plan to quit or reduce were phoned the day prior to their quit/reduction day. Follow-ups at 1 and 6 months were completed by phone. Subjects reporting significant reductions were invited into the lab to provide urine samples. 2.1.1. Fidelity of interventions This efficacy study was designed to be a strong test of MI and prescriptive advice. Neither treatment had a ‘‘home court’’ advantage in that expert consultants provided training for the interventions (RR and RW). All tapes were reviewed for adherence to the protocol and weekly meetings were held with the study nurses. Sessions not reaching criterion were removed from the analyses. 2.2. Assessments The Fagerstrom Test of Nicotine Dependence [22], the Smoking Hazards Scale [23], self-reported smoking behavior, and urine cotinine were collected at baseline, 1 and 6 months. Smoking history and spirometry were collected at baseline.
Smoking cessation was defined as cotinine less than 100 mm; significant reduction was defined as a decrease of 50% which was biochemically verified. Participants lost to follow-up were assumed not to have changed. Rasch analyses were used to create a continuous measure from the 13 outcomes and to reflect the composite smoking reduction score [24]. Rasch modeling is a probabilistic model that makes use of all available data to jointly estimate an individual’s progress towards smoking cessation and calibrate items [25]. Individuals are at different levels in the quitting process, and smoking items can be regarded as hurdles of varying heights. For example, quits are harder to achieve than reductions, and both are much harder to achieve than planning to quit. To receive the maximum score of twelve, a smoker needed to be abstinent at both the 1-month and the 6-month follow-up. 2.3. Analyses Demographic characteristics and outcomes were examined using t-tests and x2 statistics. A generalized linear model was the primary outcome analysis. The dependent variable was the composite outcome measure for smoking reduction. The independent variables and their order of entry were (1) gender, (2) age, (3) ethnicity, (4) cigarettes per day (to assure the groups were similar at baseline), (5) Treatment assignment (to evaluate the differential treatment effect), and interaction terms (to examine subgroup differences). 3. Results Two hundred and thirty precontemplative and contemplative smokers were randomized to the brief treatments and 218 cases were included in the analyses. Of the 218 smokers, 71% were available at 1 month, and 56% at 6 months (Table 1). 3.1. Equivalency of groups at baseline The two groups were comparable at baseline on age, gender, total years smoked, age at first cigarette, lifetime packs and spirometry (Table 2). The average Fagerstrom score was 6.2, indicating the study participants had a high dependence on nicotine. There were no observed relationships between risk perception and smoking reduction. Caucasians were overrepresented in the motivational group. Smokers in the prescriptive group had higher Fagerstrom scores (Table 2). 3.2. Follow-up status by treatment group
2.2.1. Outcome measures The study included 13 indicators of intentions to quit, selfreport, and biological verification (see Table 3). All interventions were videotaped and coded for intentions to reduce or quit.
There was no difference in follow-up status at 1 month. However, 63% of the motivational group responded to the 6-month follow-up, compared to 49% of the prescriptive group.
Table 1 Recruitment and retention of participants by treatment group. Procedures
Total n
Prescriptive approach
Motivational interviewing
Randomized to treatment Discontinued during treatment Excluded from the analysesa One-month follow-upb Six-month follow-upc Both one- and six-month follow-up Included in the analysis
230 0 12 154 122 111 218
114 0 5 77 53 51 109
116 0 7 77 69 61 109
a One case did not meet inclusion criteria, two cases did not meet the treatment standard, one individual was deceased, and the remaining nine cases had missing baseline data. b One-month follow-up (x2(1,n = 218) = 0.0, p = 1.00). c Six-month follow-up (x2(1,n = 218) = 4.77, p = .03).
M.F. Davis et al. / Patient Education and Counseling 83 (2011) 129–133
131
Table 2 Baseline characteristics by treatment assignment (n = 218). Prescriptive approach
Motivational interviewing
Total
Gender (n, %) Males Females
61 (56%) 48 (44%)
59 (54%) 50 (46%)
120 (55%) 98 (45%)
x2 = 0.74
.785
Ethnicity (n, %) Caucasian Hispanic African American Other
74 16 10 9
92 7 1 8
166 23 11 17
x2 = 8.18
.003
Fagerstrom score (M, SD) Age started smoking (M, SD) Years smoked (M, SD) Cigarettes smoked a day (M, SD) Average age (M, SD) Normalized spirometry percent (M, SD)
6.6 13.7 22.4 25.4 38.2 78.5
t = 2.64 t = 0.37 t = 1.37 t = 1.33 t = 0.74 t = 1.11
.009 .710 .174 .186 .460 .267
(68%) (15%) (9%) (8%) (2.2) (4.4) (11.6) (13.1) (12.4) (21.0)
(85%) (7%) (1%) (7%)
5.8 13.9 20.1 25.4 37.0 81.5
(2.1) (4.2) (12.3) (8.4) (11.7) (18.5)
p
(76%) (11%) (5%) (8%)
6.2 13.8 21.1 25.4 37.6 80.0
(2.2) (4.3) (12.0) (11.0) (12.0) (19.8)
x2 tests were used for dichotomous variables; t-tests were used for continuous variables.
Table 3 Intention, verbal report, and biologically verified smoking outcomes by treatment assignment (n = 218).
Indicates intention to quit within 6 months Indicates intention to reduce within 1 month Indicates intention to quit within 1 month Indicates intention to reduce within 1 week of intervention Indicates intention to quit within 1 week of intervention Verbal report—one 24-h 50% reduction at 1 or 6 months Verbal report—one 72-h 50% reduction at 1 or 6 months Verbal report—one 24-h quit at 1 or 6 months Verbal report—one 72-h quit 1 or 6 months Biologically verified 50% reduction at 1 or 6 months Biologically verified 50% reduction at 1 and 6 months Biologically verified quit at 1 or 6 months Biologically verified quit at 1 and 6 months Composite smoking reduction score (Mean, SD) a
Prescriptive n (%)
Motivational n (%)
Total n (%)
x2
p
58 (53) 54 (50) 18 (17) 35 (32) 1 (1) 44 (40) 36 (33) 29 (27) 6 (6) 17 (16) 5 (5) 5 (5) 0 (0) 6.16 (1.5)
60 (55) 47 (43) 25 (23) 26 (24) 2 (2) 44 (40) 32 (29) 28 (26) 4 (4) 19 (17) 2 (2) 7 (6) 1 (1) 6.09 (1.5)
118 (54) 101 (46) 43 (20) 61 (28) 3 (1) 88 (40) 68 (31) 57 (26) 10 (5) 36 (17) 7 (3) 12 (6) 1 (1) 6.13 (1.5)
0.07 0.90 1.42 1.84 –a 0.00 0.34 0.24 0.42 0.13 –a 0.35 –a 0.33
.79 .34 .23 .18 1.00 1.00 .56 .88 .52 .72 .45 .55 1.00 .75
Fishers exact test was used when estimated cell counts were below 5.
3.3. Behavioral and biological outcomes On follow-up, 26% reported at least one 24-h quit and 17% had a biologically verified reduction of 50% or more. Six percent of the sample had a biologically verified quit during follow-up. There were no differences by treatment group assignment on any outcome measure (Table 3). Hierarchical regression was used to evaluate the effect of the intervention with the composite smoking measure. Gender, age, ethnicity (Caucasian), and number of cigarettes smoked at baseline
Table 4 Hierarchical regression model to predict smoking reduction (n = 218). Predictor
F
p
Gender Age Caucasian Number of cigarettes a day at baseline Treatment group Treatment gender Treatment age Treatment Caucasian Treatment cigarettes a day at baseline Treatment gender age Treatment gender Caucasian Treatment gender cigarettes a day at baseline Treatment age Caucasian Treatment age cigarettes a day at baseline Treatment Caucasian cigarettes a day at baseline
10.88 0.22 0.00 1.70 0.28 0.09 0.24 0.45 0.27 3.02 3.19 1.39 0.35 4.86 1.01
.001 .64 .99 .19 .60 .77 .62 .51 .60 .05 .04 .25 .70 .009 .37
were entered first, to assure that the treatment and control groups were similar at baseline. There was no effect of age, Caucasian ethnicity, or number of cigarettes smoked at baseline on the composite smoking measure. However, there was an effect of gender; women showed greater improvement regardless of treatment condition. There was no differential effect of treatment; participants in the prescriptive and motivational groups did equally well (Table 4). Several interaction terms were also significant. Caucasian women showed greater progress towards smoking reduction than Caucasian males, while non-Caucasian men and women had better responses to MI (Fig. 1). Older, heavier smokers showed better progress towards smoking reduction in MI (Fig. 2). Older females showed greater progress towards cessation in MI (Fig. 3). The study
Fig. 1. Smoking progress: gender, ethnicity and treatment assignment.
132
M.F. Davis et al. / Patient Education and Counseling 83 (2011) 129–133
away, while an empathic approach may keep the door open for future interventions. 4.2. Conclusion
Fig. 2. Smoking progress: age, baseline smoking and treatment assignment.
Medical visits provide a window of opportunity to offer brief smoking interventions. The largest contribution of this study may be the methodological innovation of evaluating more than cessation rates when focusing on heavily addicted smokers; the composite measure enabled us to examine subgroup differences that were not detected with dichotomous outcomes. 4.3. Practice implications Few trials have attempted to encourage cessation in precontemplative and contemplative, heavily addicted smokers. Neither approach was superior overall. These results indicate that, in general, health professionals should use the intervention that appeals to them. We speculate that MI may provide an added benefit in conjunction with other smoking cessation treatment components. Acknowledgement Supported by a grant from The Arizona Disease Control Research Commission.
Fig. 3. Smoking progress: gender, age (<40 and >40) and treatment assignment.
References was powered to detect a 15% difference in proportions in selfreported quit rates. Because women showed greater gains in this study, we examined gender differences across the thirteen outcome variables. Women were more likely to state they would quit within 6 months (x2(1,n = 218) = 5.99, p = .01), report at least one 24-h 50% reduction (x2(1,n = 218) = 5.49, p < .02), at least one 3-day 50% reduction (x2(1,n = 218) = 7.68, p = .006), and at least one 24-h quit at 1 or 6 months (x2(1,n = 218) = 6.74, p = .009). 4. Discussion and conclusion 4.1. Discussion This study focused on smokers not ready to quit, a group that comprises nearly 90% of smokers. Our primary study hypothesis, that motivational interviewing would increase biologically verified quits by 15%, was not supported by the data. MI and prescriptive advice were equivalent in this 15-min intervention designed to simulate the real pressures faced by health care professionals. However, 33% of the subjects reported making a 24-h quit attempt during the 6 months of the trial, and women showed a greater response than males to both interventions. These results demonstrate the usefulness of brief smoking cessation interventions. Smokers make multiple quit attempts before succeeding [26], and any attempt to quit represents progress. We found modest subgroup differences. Non-Caucasian men and women showed greater progress in the MI condition; Caucasians responded equally well to both treatments, and women showed greater progress than men. Results from a meta-analysis [21] reported larger effect sizes for MI in ethnic minority populations. Older, heavier smokers and older women showed greater reductions in the MI condition; these interactions may be due to differences in education or status levels. More smokers were lost to follow-up in the prescriptive condition. A more authoritative approach may drive participants
[1] Sesney JW, Kreher NE, Hickner JM, Webb S. Smoking cessation interventions in rural family practices: an UPRNet study. J Fam Practice 1997;44:578–85. [2] Pollak KI, McBride CM, Scholes D, Grothaus LC, Civic D, Curry SJ. Women’s reports of smoking cessation advice during reproductive health visits and subsequent smoking cessation. Am J Manag Care 2002;8:837–44. [3] Pendelton L, House WC. Preferences for treatment approaches in medical care: college students versus diabetic outpatients. Med Care 1984;22:644–6. [4] Strull WM, Lo B, Charles G. Do patients want to participate in medical decision making? J Amer Med Assoc 1984;252:2990–4. [5] Rollnick S, Healther N, Bell A. Negotiating behaviour change in medical settings: the development of brief motivational interviewing. J Ment Health 1992;1:25–37. [6] Rollnick S, Miller WR. What is motivational interviewing? Behav Cogn Psychother 1995;23:325–34. [7] Schneider RJ, Casey J, Kohn R. Motivational versus confrontational interviewing: a comparison of substance abuse assessment practices at employee assistance programs. J Behav Health Ser R 2000;27:60–74. [8] Kaplan SH, Greenfield S, Martin E. Expanding patient involvement in care: the epic program. Washington, DC: United States Public Health Service; 1983. [9] The Agency for Health Care Policy and Research. Smoking cessation clinical practice guideline. J Amer Med Assoc 1996;275:1270–80. [10] American Psychiatry Association. Practice guidelines for the treatment of patients with nicotine dependence. Am J Psychiat 1996;153:S1–31. [11] Hettema J, Steele J, Miller WR. Motivational interviewing. Annu Rev Clin Psychol 2005;1:91–111. [12] Rubak S, Sandbæk A, Lauritzen T, Christensen B. Motivational interviewing: a systematic review and meta-analysis. Brit J Gen Pract 2005;55:305–12. [13] Wagner C, Conners W. MI Bibliography 1983–2009. Retrieved September 1, 2009 from http://motivationalinterview.org/library/biblio.html. [14] Butler CC, Rollnick S, Cohen D, Russell I, Bachmann M, Stott N. Motivational consulting versus brief advice for smokers in general practice: a randomised trial. Brit J Gen Pract 1999;49:611–6. [15] Borrelli B, Novak S, Hecht J, Emmons K, Papandonatos G, Abrams D. Home health care nurses as a new channel for smoking cessation treatment: outcomes from project CARES (Community-Nurse Assisted Research and Education on Smoking). Prev Med 2005;41:815–21. [16] Colby SM, Monti PM, Barnett NP, Rohenow DJ, Weissman K, Spirito A, et al. Brief motivational interviewing in a hospital setting for adolescent smoking: a preliminary study. J Consult Clin Psych 1998;66:574–8. [17] Colby SM, Monti PM, O’Leary TT, Barnett NP, Spirito A, Rohsenow DJ, et al. Brief motivational intervention for adolescent smokers in medical settings. Addict Behav 2005;30:865–74. [18] Haug NA, Svikis DS, DiClemente C. Motivational enhancement therapy for nicotine dependence in methadone-maintained pregnant women. Addict Behav 2004;18:289–92. [19] Soria R, Legido A, Escolano C, Lopez-Yeste A, Montoya J. A randomised controlled trial of motivational interviewing for smoking cessation. Brit J Gen Pract 2006;56:768–74.
M.F. Davis et al. / Patient Education and Counseling 83 (2011) 129–133 [20] Steinberg ML, Ziedonis DM, Krejci JA, Brandon TH. Motivational interviewing with personalized feedback: a brief intervention for motivating smokers with schizophrenia to seek treatment for tobacco dependence. J Consult Clin Psych 2004;72:723–8. [21] Ahluwalia JS, Okuyemi K, Nollen N, Choi WS, Kaur H, Pulvers K, et al. The effects of nicotine gum and counseling among African American light smokers: a 2 2 factorial design. Addiction 2006;101:883–91. [22] Fagerstrom KO, Schneider NG. Measuring nicotine dependence: a review of the Fagerstrom Tolerance Questionnaire. J Behav Med 1989;12:159–82.
133
[23] Davis MF, Shapiro D, Sechrest L. The smoking hazards scale. J Ariz Nevada Acad Sci 2006;38:102–6. [24] Davis MF, Sechrest L, Shapiro D. Measuring progress toward smoking cessation. J Appl Meas 2005;6:164–72. [25] Rasch G. Probabilistic models for some intelligence and attainment tests. Copenhagen: Danish Institute for Educational Research. (1960, Expanded edition, 1980). Chicago: University of Chicago Press; 1980. [26] Hughes JR. Motivating and helping smokers to stop smoking. J Gen Intern Med 2003;18:1053–7.