- Email: [email protected]
Hospitalized smokers: Compliance with a nonsmoking policy and its predictors
Preventive Medicine 43 (2006) 113 – 116 www.elsevier.com/locate/ypmed
Hospitalized smokers: Compliance with a nonsmoking policy and its predictors Meritxell Sabidó a,⁎, Jordi Sunyer b , Cristina Masuet a , Joan Masip a a
Department of Preventive Medicine and Public Health, Hospital Universitari de Bellvitge, Crta. Feixa Llarga s/n, L'Hospitalet de Llobregat 08907, Spain b Environmental Respiratory Research Unit, Institut Municipal d'Investigació Mèdica (IMIM), Barcelona, Spain Available online 21 April 2006
Abstract Background. Factors associated with compliance with smoke-free policies among hospitalized smokers are poorly described. A better understanding of these factors may improve smoking cessation during admission and in the long-term. Methods. Two cross-sectional studies were conducted in an urban teaching hospital in Spain during 2002 and 2004. We interviewed 229 admitted smokers gathering data on smoking history, admission diagnosis, belief that hospitalization is related to smoking, policy's awareness, and smoking during admission and place of smoking. Results. Among hospitalized patients, approximately a third were current smokers. The compliance with the nonsmoking policy in 2002 and 2004 was respectively 71.9% (IC95%: 63.9–79.9) and 60.1% (IC95%: 50.9–69.3). In the multivariate regression model, factors significantly associated with compliance were: contemplation stage, confidence in quitting after discharge, belief that current symptoms or illness were related to smoking, and mild withdrawal symptoms. Conclusions. Admission in a smoke-free hospital does not guarantee that patients will refrain from smoking. Factors associated with compliance identified may be modified by tailored smoking cessation interventions. Our results might help physicians to understand inpatients' difficulties to abstain from cigarettes and enhance their efforts to take advantage of the hospitalization as a window opportunity to quit. © 2006 Elsevier Inc. All rights reserved. Keywords: Smoking cessation; Hospitals; Patient compliance
Introduction Admission to a smoke-free hospital provides a favorable environment to stop smoking influencing: the severity of nicotine withdrawal syndrome (NWS); the stages of change (Prochaska and DiClemente, 1983); perceived health threat and self-efficacy (Barth and Bengel, 2000); and social norms. Little information is available about smoking behavior among inpatients. More than 75.0% of them comply with nonsmoking policies, which is associated with long-term cessation (Rigotti et al., 2000). A better understanding of factors associated with compliance could improve the effectiveness of cessation interventions maximizing the opportunity to quit. This was the first such study within the Spanish health system. Primary objectives were to assess the prevalence of
compliance with a nonsmoking policy and to identify its associated factors among inpatient in a smoke-free hospital. In order to guide policy, we also aimed to analyze potential time differences in the policy's compliance. Therefore, we recruited smokers during two separated time-periods: October 2002 and January to April 2004. Methods Design Cross-sectional study conducted in Hospital Universitari de Bellvitge, Barcelona, a teaching hospital (627 beds). The smoke-free policy was implemented in April 2001: smoking is prohibited in all indoor areas but it is permitted outside the building.
Subjects ⁎ Corresponding author. Fax: +34 932 607 878. E-mail address: [email protected] (M. Sabidó). 0091-7435/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2006.02.012
All admitted smokers were considered for enrollment. Current smokers were those who smoked at least one cigarette within the 6 months prior to admission
114
M. Sabidó et al. / Preventive Medicine 43 (2006) 113–116
(daily or occasionally). We classified them as: compliant, those who did not smoked indoors or who only smoked outdoors; and noncompliant, those who smoked indoors.
Data collection Daily admissions were identified in the hospital records on a 5 evenings/ week bases. Patients were approached in their rooms to assess smoking status that was self-reported. Whenever the patient was cognitively impaired, refused to answer or was missing from the room after 3 attempts, smoking status was obtained from medical records, or from a patient's next of kin. After that, current smokers were interviewed with a structured questionnaire previously tested. Those current smokers cognitively impaired, not willing to participate, or absent from their rooms after 3 attempts were excluded from the study. Items assessed were sociodemographic characteristics (Regidor, 2001); belief that hospitalization is related to smoking; awareness of the policy; smoking during admission (amount and place); confidence to quit after discharge; and smoking history: nicotine dependence (Fagerström and Schneider, 1989), the stages of change scale (DiClemente et al., 1991), and NWS (Hughes et al., 1984). The Admission Unit provided total number of patients, date of birth, sex, and admission's diagnosis, date, and type. The primary outcome was the prevalence of compliance with the nonsmoking policy (number of hospitalized current smokers who did not smoke indoors among hospitalized smokers). The secondary outcome was the percentage change in compliance between 2004 and 2002.
Data analysis Data were analyzed using the SPSS.12. After describing the sample, bivariable analysis compared compliant and noncompliant smokers on baseline variables using: the chi-squared test (categorical variables) and t tests and Mann–Whitney U test (continuous variables). Crude odds ratios with 95% confidence interval were calculated. Among those who smoked while hospitalized, a two-related-sample Wilcoxon test determined the amount of cigarettes reduced compared with their preadmission consumption. All significance tests were two sided. A P value < 0.05 was considered statistically significant. Multivariable analysis identified factors independently related to compliance. The logistic regression model was backward fitted and included all the explanatory variables statistically associated with the outcome.
Results Study population We identified a total of 874 patients (434 in 2002, and 440 in 2204), from which 618 were ex-/never smokers (313 in 2002, and 305 in 2004). Current smokers included (121 (27.8%) in 2002, and 135 (30.7%) in 2004), from which 27 were excluded: two (0.7%) unwilling to participate, 15 (5.9%) not present in their room, and 10 (3.9%) cognitively impaired. Finally, 229 (89.5%) current smokers were interviewed. There were no significant time differences in relevant baseline characteristics, smoking history and compliance between 2002 and 2004. Baseline characteristics and smoking history The sample was 77% male, mean age 50 years (SD:16.9), 78.6% married, 79.9% manual workers, 45.0% had not completed primary school, 27.1% electively admitted, and a quarter underwent surgery. Median length of stay was 6.0 days
(range, 2–84). Despite the fact that most patients (62%) had a smoking-related disease (reclassifying admission diagnosis following medical criteria), only a third reported that current symptoms or disease was related to smoking (Table 1). Compliance with the nonsmoking policy Compliance was 71.9% (95%CI: 63.9–79.9) in 2002 and 60.1% (95%CI: 50.9–69.3) in 2004. This 11.8% reduction was Table 1 Smoking characteristics of admitted smokers in a smoke-free hospital a 2002, N (%)
2004, N (%)
Total, N (%)
Age of smoking 15.0 (8–27) 15.0 (9–35) 15 (9–35) initiation (median, range) Prior brief counseling intervention Yes 68 (56.2) 56 (51.9) 124 (54.1) No 53 (43.8) 52 (48.1) 105 (45.9) Living with other smokers No 55 (45.5) 36 (33.3) 91 (39.7) Yes 66 (54.5) 72 (66.7) 138 (60.3) Quit attempts Yes 60 (49.6) 49 (45.4) 109 (47.5) No 61 (50.4) 59 (54.5) 120 (52.5) Prior use of nicotine replacement therapy Yes 6 (10.0) 2 (4.1) 8 (3.5) No 54 (90.0) 47 (95.9) 101 (96.5) Stage of change c Precontemplation 60 (49.6) 62 (57.4) 122 (53.3) Contemplation/ 61 (50.4) 46 (42.6) 107 (46.7) preparation Cigarettes per day 20.0 (2–90) 20.0 (2–80) 20.0 (2–90) (median, range) Nicotine dependence Mild (0–3) 42 (34.7) 21 (19.5) 63 (27.5) Moderate (4–6) 30 (24.8) 35 (32.4) 65 (28.3) Severe (7–10) 49 (40.5) 52 (48.1) 101 (44.2) Nicotine withdrawal symptoms Mild (0–6) 54 (44.7) 62 (57.4) 116 (50.6) Moderate (7–12) 43 (35.5) 32 (29.6) 75 (32.8) Severe (13–18) 24 (19.8) 14 (13.0) 38 (16.6) Confidence in the ability to quit after discharge Extremely confident 58 (47.9) 55 (50.9) 113 (49.4) Not confident 42 (34.7) 38 (35.2) 80 (34.9) Have doubts 21 (17.4) 15 (13.9) 36 (15.7) Aware of the smoke-free hospital policy Yes 118 (97.5) 107 (99.1) 225 (98.3) No 3 (2.5) 1 (0.9) 4 (1.7) Smoked during hospitalization Yes 36 (29.8) 43 (39.8) 79 (34.5) No 85 (70.2) 65 (60.2) 150 (65.5) Cigarettes per day during hospitalization ≤3 cigarettes 22 (61.1) 18 (41.9) 40 (50.6) >3 cigarettes 14 (38.9) 25 (58.1) 39 (49.4) Place of smoking during hospitalization Patient's room, bathroom 5 (13.9) 11 (25.6) 16 (20.3) Stairs 22 (61.1) 32 (74.4) 54 (68.4) Other indoor areas 7 (19.4) 0 (0.0) 7 (8.8) Hospital outdoors 2 (5.6) 0 (0.0) 2 (2.5) a
P value b 0.64
0.59
0.07
0.59
0.29
0.28
0.28
0.01
0.13
0.76
0.15
0.11
0.11
0.06
Based on smokers interviewed in both 2002 and 2004. Not statistically significant (P ≥ 0.05). c Stages of change further categorized as smokers not planning to quit (precontemplators), and smokers planning to quit (contemplators/preparators). b
M. Sabidó et al. / Preventive Medicine 43 (2006) 113–116 Table 2 Multivariable analysis: factors associated with compliance with the nonsmoking policy among hospitalized smokers Compliant, Noncompliant, Adjusted N (%) N (%) OR (95%CI) a Sex Female 28 (54.9) 23 (45.1) Male 122 (68.5) 56 (31.5) Belief that current disease is related to smoking Yes 58 (78.4) 16 (21.6) No 92 (59.4) 63 (40.6) Nicotine dependence Mild (0–3) 49 (77.0) 14 (23.0) Moderate (4–6) 47 (70.0) 18 (30.0) Severe (7–10) 54 (53.5) 47 (42.9) Nicotine withdrawal symptoms Mild (0–6) 101 (87.1) 15 (12.9) Moderate (7–12) 29 (38.7) 46 (61.3) Severe (13–18) 20 (52.6) 18 (47.4) Stage of change Contemplation 90 (84.1) 17 (15.9) Precontemplation 60 (49.2) 62 (50.8) Confidence in quitting after discharge Extremely 97 (85.8) 16 (14.2) confident Not confident 31 (38.8) 49 (61.3) Have doubts 22 (61.1) 14 (38.9)
P value
1 2.49 (0.99–6.31)
0.05
1 0.35 (0.15–0.82)
0.15
1 0.60 (0.20–1.73) 0.41 (0.15–1.10)
0.07 0.01
1 0.11 (0.44–0.25) <0.001 0.26 (0.90–0.74) 0.01 1 0.39 (0.18–0.86)
0.01
1 0.21 (0.48–0.90) <0.001 0.34 (0.12–0.96) 0.041
a
Adjusted odds ratio with corresponding 95% confidence interval. Adjusted by multiple logistic regression for all listed covariates, prior brief counseling intervention, living with other smokers, daily cigarette consumption, age, and smoking-related disease as admission diagnosis.
not statistically significant (OR 0.64; 95%CI: 0.37–1.10, P = 0.12). In both years, inpatients who smoked consumed a median of 3 cigarettes/day (range, 1–20), which was substantially lower than their preadmission consumption (P < 0.001). Among the 79 patients who smoked, only 2 (2.5%) went outside the building. Noncompliance was significantly more likely: women, ≤50 years; have a smoking-related disease; live with smokers; did not received brief counseling; do not believe that admission was related to smoking; precontemplators; greater mean of cigarette/day (P = 0.003); severe nicotine dependence; moderate or severe NWS; and not confident in the ability to quit or to doubt it. Factors associated with compliance In the multivariable regression model (Table 2), compliance was associated with the contemplative stage, believing that current symptoms or disease was related to smoking, confidence in the ability to quit, and mild NWS. Discussion This is the first study to examine compliance among patients in a Spanish smoke-free hospital. Although not statistically significant, compliance in 2004 was lower than in 2002 suggesting some lost in its impact. This may reflect a proximal effect of the policy's introduction in 2002 (Jeffery et al., 1994), low involvement of health personnel, and high
115
proportion of the sample of low socioeconomic status who are less likely to absorb health messages (Townsend et al., 1994). The factors associated with compliance identified are consistent with other studies (Emmons et al., 1998; Goodman et al., 1998; Rigotti et al., 2000; Sciamanna et al., 2000) and may be potentially modifiable by cessation interventions (Rigotti et al., 2003). Approximately half of our sample were precontemplators. Despite this group being less receptive to health information, some computer interventions providing self-help material have shown promising results (Dijkstra et al., 1999). Study's strengths are large sample, inclusion of smokers with low motivation to quit, wide range of admission diagnosis, high response rate, and data consistency. Limitation is that no biochemical validation of smoking, thereby, underreporting cannot be excluded (Ustim et al., 1997). However, responses were confidential and self-reported smoking was likely to be true (Patrick et al., 1994). We cannot be certain that the nicotine withdrawal symptoms we measured represent NWS although the use of a validated scale supports our results. Data on smoking after the interview were not available, with the risk of overestimating compliance. However, median length of stay was long enough to lead smokers to violate the policy. Smoke-free hospitals provide a supportive environment for cessation and reduce exposure to passive smoking. With little resources available to implement these interventions, and with evidence suggesting that compliance greatly increases long-term cessation, stronger efforts should be made to improve compliance and to take advantage of the opportunity provided by hospitalization to foster cessation. However, these efforts should go beyond hospitalization in order to maintain cessation after discharge.
Acknowledgments We gratefully acknowledge the help provided by Gemma Binefa Rodriguez (M.D.) and by Inés Suarez (M.D., Ph.D.) with their meticulous and critical revision of the paper. References Barth, J., Bengel, J., 2000. Prevention to fear? The State of Fear Appeal Research. Series in Research and Practice of Health Promotion. Federal Centre for Health Education, Cologne. vol. 8. DiClemente, C.C., Prochaska, J.O., Fairhurst, S.K., Velicer, W.F., Velasquez, M.M., Rossi, J.S., 1991. The process of smoking cessation: an analysis of precontemplation, contemplation, and preparation stages of change. J. Consult. Clin. Psicol. 59, 295–304. Dijkstra, A., De Vries, H., Roijackers, J., 1999. Targeting smokers with low readiness to change with tailored and nontailored self-help materials'. Prev. Med. 28, 203–211. Emmons, K.M., Cargill, B.R., Hecht, R.N., Goldstein, M., Milman, R., Abrams, D., 1998. Characteristics of patients adhering to a hospital's no-smoking policy. Prev. Med. 27, 846–853. Fagerström, K.O., Schneider, N., 1989. Measuring nicotine dependence: a review of the FTND. J. Behav. Med. 12, 159–182.
116
M. Sabidó et al. / Preventive Medicine 43 (2006) 113–116
Goodman, J.M., Nadkarni, M., Schorling, J.B., 1998. The natural history of smoking cessation among medical patients in a smoke-free hospital. Subst. Abuse 19, 71–79. Hughes, J.R., Hatsukami, D.K., Pickens, R.W., Svikis, D.S., 1984. Consistency of the tobacco withdrawal syndrome. Addict. Behav. 9, 409–412. Jeffery, R.W., Kelder, S.H., Forster, J.L., French, S.A., Lando, H.A., Baxter, J.E., 1994. Restrictive smoking policies in the workplace: effects on a smoking prevalence and cigarette consumption. Prev. Med. 23, 78–82. Patrick, D.L., Cheadle, A., Thompson, D.C., Diehr, P., Koepsell, T., Kinne, S., 1994. The validity of self-reported smoking: a review and metanalysis. Am. J. Public Health 84, 1086–1093. Prochaska, J.O., DiClemente, C.C., 1983. Stages and process of self-change of smoking: towards an integrative model of change. J. Consult. Clin. Psychol. 51, 390–395. Regidor, E., 2001. La clasificación de clase social de Goldthorpe: Marco de referencia para la propuesta de medición de la clase social del grupo de
trabajo de la Sociedad Española de Epidemiología. Rev. Esp. Salud Publica 75, 13–22. Rigotti, N., Arnsten, J., McKool, K.M., Wood-Reid, K.M., Pasternak, R.C., Singer, D. E., 2000. Smoking by patients in a smoke-free hospital: prevalence, predictors, and implications. Prev. Med. 31, 159–166. Rigotti, N.A., Munafò, M., Murphy, M.F.G., Stead, L.F., 2003. Interventions for smoking cessation in hospitalised patients. Cochrane Database Syst. Rev. 1 (CD001837). Sciamanna, C.N., Stillman, F.A., Hoch, J.S., Butler, J.H., Gass, K.G., Ford, D.E., 2000. Opportunities for improving inpatients smoking cessation programs: a community hospital experience. Prev. Med. 30, 496–503. Townsend, J., Roderick, P., Cooper, J., 1994. Cigarette smoking by socioeconomic group, sex, and age: effects of price, income, and health publicity. BMJ 309, 923–927. Ustim, B., Compton, W., Mager, D., et al., 1997. WHO study on the reliability and validity of the alcohol and drug use disorder instruments: overview of methods and results. Drug Alcohol Depend. 147, 161–169.
Hospitalized smokers: Compliance with a nonsmoking policy and its predictors Meritxell Sabidó a,⁎, Jordi Sunyer b , Cristina Masuet a , Joan Masip a a
Department of Preventive Medicine and Public Health, Hospital Universitari de Bellvitge, Crta. Feixa Llarga s/n, L'Hospitalet de Llobregat 08907, Spain b Environmental Respiratory Research Unit, Institut Municipal d'Investigació Mèdica (IMIM), Barcelona, Spain Available online 21 April 2006
Abstract Background. Factors associated with compliance with smoke-free policies among hospitalized smokers are poorly described. A better understanding of these factors may improve smoking cessation during admission and in the long-term. Methods. Two cross-sectional studies were conducted in an urban teaching hospital in Spain during 2002 and 2004. We interviewed 229 admitted smokers gathering data on smoking history, admission diagnosis, belief that hospitalization is related to smoking, policy's awareness, and smoking during admission and place of smoking. Results. Among hospitalized patients, approximately a third were current smokers. The compliance with the nonsmoking policy in 2002 and 2004 was respectively 71.9% (IC95%: 63.9–79.9) and 60.1% (IC95%: 50.9–69.3). In the multivariate regression model, factors significantly associated with compliance were: contemplation stage, confidence in quitting after discharge, belief that current symptoms or illness were related to smoking, and mild withdrawal symptoms. Conclusions. Admission in a smoke-free hospital does not guarantee that patients will refrain from smoking. Factors associated with compliance identified may be modified by tailored smoking cessation interventions. Our results might help physicians to understand inpatients' difficulties to abstain from cigarettes and enhance their efforts to take advantage of the hospitalization as a window opportunity to quit. © 2006 Elsevier Inc. All rights reserved. Keywords: Smoking cessation; Hospitals; Patient compliance
Introduction Admission to a smoke-free hospital provides a favorable environment to stop smoking influencing: the severity of nicotine withdrawal syndrome (NWS); the stages of change (Prochaska and DiClemente, 1983); perceived health threat and self-efficacy (Barth and Bengel, 2000); and social norms. Little information is available about smoking behavior among inpatients. More than 75.0% of them comply with nonsmoking policies, which is associated with long-term cessation (Rigotti et al., 2000). A better understanding of factors associated with compliance could improve the effectiveness of cessation interventions maximizing the opportunity to quit. This was the first such study within the Spanish health system. Primary objectives were to assess the prevalence of
compliance with a nonsmoking policy and to identify its associated factors among inpatient in a smoke-free hospital. In order to guide policy, we also aimed to analyze potential time differences in the policy's compliance. Therefore, we recruited smokers during two separated time-periods: October 2002 and January to April 2004. Methods Design Cross-sectional study conducted in Hospital Universitari de Bellvitge, Barcelona, a teaching hospital (627 beds). The smoke-free policy was implemented in April 2001: smoking is prohibited in all indoor areas but it is permitted outside the building.
Subjects ⁎ Corresponding author. Fax: +34 932 607 878. E-mail address: [email protected] (M. Sabidó). 0091-7435/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2006.02.012
All admitted smokers were considered for enrollment. Current smokers were those who smoked at least one cigarette within the 6 months prior to admission
114
M. Sabidó et al. / Preventive Medicine 43 (2006) 113–116
(daily or occasionally). We classified them as: compliant, those who did not smoked indoors or who only smoked outdoors; and noncompliant, those who smoked indoors.
Data collection Daily admissions were identified in the hospital records on a 5 evenings/ week bases. Patients were approached in their rooms to assess smoking status that was self-reported. Whenever the patient was cognitively impaired, refused to answer or was missing from the room after 3 attempts, smoking status was obtained from medical records, or from a patient's next of kin. After that, current smokers were interviewed with a structured questionnaire previously tested. Those current smokers cognitively impaired, not willing to participate, or absent from their rooms after 3 attempts were excluded from the study. Items assessed were sociodemographic characteristics (Regidor, 2001); belief that hospitalization is related to smoking; awareness of the policy; smoking during admission (amount and place); confidence to quit after discharge; and smoking history: nicotine dependence (Fagerström and Schneider, 1989), the stages of change scale (DiClemente et al., 1991), and NWS (Hughes et al., 1984). The Admission Unit provided total number of patients, date of birth, sex, and admission's diagnosis, date, and type. The primary outcome was the prevalence of compliance with the nonsmoking policy (number of hospitalized current smokers who did not smoke indoors among hospitalized smokers). The secondary outcome was the percentage change in compliance between 2004 and 2002.
Data analysis Data were analyzed using the SPSS.12. After describing the sample, bivariable analysis compared compliant and noncompliant smokers on baseline variables using: the chi-squared test (categorical variables) and t tests and Mann–Whitney U test (continuous variables). Crude odds ratios with 95% confidence interval were calculated. Among those who smoked while hospitalized, a two-related-sample Wilcoxon test determined the amount of cigarettes reduced compared with their preadmission consumption. All significance tests were two sided. A P value < 0.05 was considered statistically significant. Multivariable analysis identified factors independently related to compliance. The logistic regression model was backward fitted and included all the explanatory variables statistically associated with the outcome.
Results Study population We identified a total of 874 patients (434 in 2002, and 440 in 2204), from which 618 were ex-/never smokers (313 in 2002, and 305 in 2004). Current smokers included (121 (27.8%) in 2002, and 135 (30.7%) in 2004), from which 27 were excluded: two (0.7%) unwilling to participate, 15 (5.9%) not present in their room, and 10 (3.9%) cognitively impaired. Finally, 229 (89.5%) current smokers were interviewed. There were no significant time differences in relevant baseline characteristics, smoking history and compliance between 2002 and 2004. Baseline characteristics and smoking history The sample was 77% male, mean age 50 years (SD:16.9), 78.6% married, 79.9% manual workers, 45.0% had not completed primary school, 27.1% electively admitted, and a quarter underwent surgery. Median length of stay was 6.0 days
(range, 2–84). Despite the fact that most patients (62%) had a smoking-related disease (reclassifying admission diagnosis following medical criteria), only a third reported that current symptoms or disease was related to smoking (Table 1). Compliance with the nonsmoking policy Compliance was 71.9% (95%CI: 63.9–79.9) in 2002 and 60.1% (95%CI: 50.9–69.3) in 2004. This 11.8% reduction was Table 1 Smoking characteristics of admitted smokers in a smoke-free hospital a 2002, N (%)
2004, N (%)
Total, N (%)
Age of smoking 15.0 (8–27) 15.0 (9–35) 15 (9–35) initiation (median, range) Prior brief counseling intervention Yes 68 (56.2) 56 (51.9) 124 (54.1) No 53 (43.8) 52 (48.1) 105 (45.9) Living with other smokers No 55 (45.5) 36 (33.3) 91 (39.7) Yes 66 (54.5) 72 (66.7) 138 (60.3) Quit attempts Yes 60 (49.6) 49 (45.4) 109 (47.5) No 61 (50.4) 59 (54.5) 120 (52.5) Prior use of nicotine replacement therapy Yes 6 (10.0) 2 (4.1) 8 (3.5) No 54 (90.0) 47 (95.9) 101 (96.5) Stage of change c Precontemplation 60 (49.6) 62 (57.4) 122 (53.3) Contemplation/ 61 (50.4) 46 (42.6) 107 (46.7) preparation Cigarettes per day 20.0 (2–90) 20.0 (2–80) 20.0 (2–90) (median, range) Nicotine dependence Mild (0–3) 42 (34.7) 21 (19.5) 63 (27.5) Moderate (4–6) 30 (24.8) 35 (32.4) 65 (28.3) Severe (7–10) 49 (40.5) 52 (48.1) 101 (44.2) Nicotine withdrawal symptoms Mild (0–6) 54 (44.7) 62 (57.4) 116 (50.6) Moderate (7–12) 43 (35.5) 32 (29.6) 75 (32.8) Severe (13–18) 24 (19.8) 14 (13.0) 38 (16.6) Confidence in the ability to quit after discharge Extremely confident 58 (47.9) 55 (50.9) 113 (49.4) Not confident 42 (34.7) 38 (35.2) 80 (34.9) Have doubts 21 (17.4) 15 (13.9) 36 (15.7) Aware of the smoke-free hospital policy Yes 118 (97.5) 107 (99.1) 225 (98.3) No 3 (2.5) 1 (0.9) 4 (1.7) Smoked during hospitalization Yes 36 (29.8) 43 (39.8) 79 (34.5) No 85 (70.2) 65 (60.2) 150 (65.5) Cigarettes per day during hospitalization ≤3 cigarettes 22 (61.1) 18 (41.9) 40 (50.6) >3 cigarettes 14 (38.9) 25 (58.1) 39 (49.4) Place of smoking during hospitalization Patient's room, bathroom 5 (13.9) 11 (25.6) 16 (20.3) Stairs 22 (61.1) 32 (74.4) 54 (68.4) Other indoor areas 7 (19.4) 0 (0.0) 7 (8.8) Hospital outdoors 2 (5.6) 0 (0.0) 2 (2.5) a
P value b 0.64
0.59
0.07
0.59
0.29
0.28
0.28
0.01
0.13
0.76
0.15
0.11
0.11
0.06
Based on smokers interviewed in both 2002 and 2004. Not statistically significant (P ≥ 0.05). c Stages of change further categorized as smokers not planning to quit (precontemplators), and smokers planning to quit (contemplators/preparators). b
M. Sabidó et al. / Preventive Medicine 43 (2006) 113–116 Table 2 Multivariable analysis: factors associated with compliance with the nonsmoking policy among hospitalized smokers Compliant, Noncompliant, Adjusted N (%) N (%) OR (95%CI) a Sex Female 28 (54.9) 23 (45.1) Male 122 (68.5) 56 (31.5) Belief that current disease is related to smoking Yes 58 (78.4) 16 (21.6) No 92 (59.4) 63 (40.6) Nicotine dependence Mild (0–3) 49 (77.0) 14 (23.0) Moderate (4–6) 47 (70.0) 18 (30.0) Severe (7–10) 54 (53.5) 47 (42.9) Nicotine withdrawal symptoms Mild (0–6) 101 (87.1) 15 (12.9) Moderate (7–12) 29 (38.7) 46 (61.3) Severe (13–18) 20 (52.6) 18 (47.4) Stage of change Contemplation 90 (84.1) 17 (15.9) Precontemplation 60 (49.2) 62 (50.8) Confidence in quitting after discharge Extremely 97 (85.8) 16 (14.2) confident Not confident 31 (38.8) 49 (61.3) Have doubts 22 (61.1) 14 (38.9)
P value
1 2.49 (0.99–6.31)
0.05
1 0.35 (0.15–0.82)
0.15
1 0.60 (0.20–1.73) 0.41 (0.15–1.10)
0.07 0.01
1 0.11 (0.44–0.25) <0.001 0.26 (0.90–0.74) 0.01 1 0.39 (0.18–0.86)
0.01
1 0.21 (0.48–0.90) <0.001 0.34 (0.12–0.96) 0.041
a
Adjusted odds ratio with corresponding 95% confidence interval. Adjusted by multiple logistic regression for all listed covariates, prior brief counseling intervention, living with other smokers, daily cigarette consumption, age, and smoking-related disease as admission diagnosis.
not statistically significant (OR 0.64; 95%CI: 0.37–1.10, P = 0.12). In both years, inpatients who smoked consumed a median of 3 cigarettes/day (range, 1–20), which was substantially lower than their preadmission consumption (P < 0.001). Among the 79 patients who smoked, only 2 (2.5%) went outside the building. Noncompliance was significantly more likely: women, ≤50 years; have a smoking-related disease; live with smokers; did not received brief counseling; do not believe that admission was related to smoking; precontemplators; greater mean of cigarette/day (P = 0.003); severe nicotine dependence; moderate or severe NWS; and not confident in the ability to quit or to doubt it. Factors associated with compliance In the multivariable regression model (Table 2), compliance was associated with the contemplative stage, believing that current symptoms or disease was related to smoking, confidence in the ability to quit, and mild NWS. Discussion This is the first study to examine compliance among patients in a Spanish smoke-free hospital. Although not statistically significant, compliance in 2004 was lower than in 2002 suggesting some lost in its impact. This may reflect a proximal effect of the policy's introduction in 2002 (Jeffery et al., 1994), low involvement of health personnel, and high
115
proportion of the sample of low socioeconomic status who are less likely to absorb health messages (Townsend et al., 1994). The factors associated with compliance identified are consistent with other studies (Emmons et al., 1998; Goodman et al., 1998; Rigotti et al., 2000; Sciamanna et al., 2000) and may be potentially modifiable by cessation interventions (Rigotti et al., 2003). Approximately half of our sample were precontemplators. Despite this group being less receptive to health information, some computer interventions providing self-help material have shown promising results (Dijkstra et al., 1999). Study's strengths are large sample, inclusion of smokers with low motivation to quit, wide range of admission diagnosis, high response rate, and data consistency. Limitation is that no biochemical validation of smoking, thereby, underreporting cannot be excluded (Ustim et al., 1997). However, responses were confidential and self-reported smoking was likely to be true (Patrick et al., 1994). We cannot be certain that the nicotine withdrawal symptoms we measured represent NWS although the use of a validated scale supports our results. Data on smoking after the interview were not available, with the risk of overestimating compliance. However, median length of stay was long enough to lead smokers to violate the policy. Smoke-free hospitals provide a supportive environment for cessation and reduce exposure to passive smoking. With little resources available to implement these interventions, and with evidence suggesting that compliance greatly increases long-term cessation, stronger efforts should be made to improve compliance and to take advantage of the opportunity provided by hospitalization to foster cessation. However, these efforts should go beyond hospitalization in order to maintain cessation after discharge.
Acknowledgments We gratefully acknowledge the help provided by Gemma Binefa Rodriguez (M.D.) and by Inés Suarez (M.D., Ph.D.) with their meticulous and critical revision of the paper. References Barth, J., Bengel, J., 2000. Prevention to fear? The State of Fear Appeal Research. Series in Research and Practice of Health Promotion. Federal Centre for Health Education, Cologne. vol. 8. DiClemente, C.C., Prochaska, J.O., Fairhurst, S.K., Velicer, W.F., Velasquez, M.M., Rossi, J.S., 1991. The process of smoking cessation: an analysis of precontemplation, contemplation, and preparation stages of change. J. Consult. Clin. Psicol. 59, 295–304. Dijkstra, A., De Vries, H., Roijackers, J., 1999. Targeting smokers with low readiness to change with tailored and nontailored self-help materials'. Prev. Med. 28, 203–211. Emmons, K.M., Cargill, B.R., Hecht, R.N., Goldstein, M., Milman, R., Abrams, D., 1998. Characteristics of patients adhering to a hospital's no-smoking policy. Prev. Med. 27, 846–853. Fagerström, K.O., Schneider, N., 1989. Measuring nicotine dependence: a review of the FTND. J. Behav. Med. 12, 159–182.
116
M. Sabidó et al. / Preventive Medicine 43 (2006) 113–116
Goodman, J.M., Nadkarni, M., Schorling, J.B., 1998. The natural history of smoking cessation among medical patients in a smoke-free hospital. Subst. Abuse 19, 71–79. Hughes, J.R., Hatsukami, D.K., Pickens, R.W., Svikis, D.S., 1984. Consistency of the tobacco withdrawal syndrome. Addict. Behav. 9, 409–412. Jeffery, R.W., Kelder, S.H., Forster, J.L., French, S.A., Lando, H.A., Baxter, J.E., 1994. Restrictive smoking policies in the workplace: effects on a smoking prevalence and cigarette consumption. Prev. Med. 23, 78–82. Patrick, D.L., Cheadle, A., Thompson, D.C., Diehr, P., Koepsell, T., Kinne, S., 1994. The validity of self-reported smoking: a review and metanalysis. Am. J. Public Health 84, 1086–1093. Prochaska, J.O., DiClemente, C.C., 1983. Stages and process of self-change of smoking: towards an integrative model of change. J. Consult. Clin. Psychol. 51, 390–395. Regidor, E., 2001. La clasificación de clase social de Goldthorpe: Marco de referencia para la propuesta de medición de la clase social del grupo de
trabajo de la Sociedad Española de Epidemiología. Rev. Esp. Salud Publica 75, 13–22. Rigotti, N., Arnsten, J., McKool, K.M., Wood-Reid, K.M., Pasternak, R.C., Singer, D. E., 2000. Smoking by patients in a smoke-free hospital: prevalence, predictors, and implications. Prev. Med. 31, 159–166. Rigotti, N.A., Munafò, M., Murphy, M.F.G., Stead, L.F., 2003. Interventions for smoking cessation in hospitalised patients. Cochrane Database Syst. Rev. 1 (CD001837). Sciamanna, C.N., Stillman, F.A., Hoch, J.S., Butler, J.H., Gass, K.G., Ford, D.E., 2000. Opportunities for improving inpatients smoking cessation programs: a community hospital experience. Prev. Med. 30, 496–503. Townsend, J., Roderick, P., Cooper, J., 1994. Cigarette smoking by socioeconomic group, sex, and age: effects of price, income, and health publicity. BMJ 309, 923–927. Ustim, B., Compton, W., Mager, D., et al., 1997. WHO study on the reliability and validity of the alcohol and drug use disorder instruments: overview of methods and results. Drug Alcohol Depend. 147, 161–169.