Smoking habits and attitudes to smoking 2001 among hospital staff at a Danish hospital—comparison with a similar study 1999

Preventive Medicine 41 (2005) 321 – 327 www.elsevier.com/locate/ypmed

Smoking habits and attitudes to smoking 2001 among hospital staff at a Danish hospital—comparison with a similar study 1999 Pia Nimann Kannegaard, M.D.a,*, Svend Kreiner, M.Sc.b, Per Gregersen, M.D.c, Henri Goldstein, M.D.a a

Department of Preventive Medicine, Roskilde County Hospital Koege, DK-4600 Koege, Denmark b Department of Biostatistics, Copenhagen University, Denmark c Department of Occupational Medicine, Roskilde County Hospital Koege, Denmark Available online 13 February 2005

Abstract Background. Smoking is the main preventable cause of lifestyle-related diseases and smoking cessation interventions have shown to be cost-effective. It is important to know the smoking habits and attitudes towards smoking among the hospital staff because of the impact the staff have on the patients. Two studies were conducted to attain this information. Methods. This study analyzes the results of a survey conducted in the summer of 2001 and compares them to a similar study conducted in 1999. 970 anonymous questionnaires were distributed in 2001. Results. The number of smokers among the staff has significantly decreased from 33% to 26% over the 2-year period. Those who do still smoke are less willing to quit and are reluctant to accept help to stop smoking. Fewer are bothered by passive smoking now compared to 1999 and approximately 2/3 of the responders do not find it acceptable to implement sanctions against the members of staff who break the prohibition. Conclusion. It is worth providing resources to smoking cessation intervention. The intensive preventive work carried out at the hospital may have resulted in significantly fewer smokers among the staff which again facilitate the staff’s effort to help the patients stop smoking. D 2005 Elsevier Inc. All rights reserved. Keywords: Smoking cessation intervention; Hospital staff; Smoking habits; Attitude

Introduction The number of deaths due to the use of tobacco in Denmark is approximately 13,000 on an annual basis [1]. Smoking is the main preventable cause of lifestyle-related diseases such as lung cancer, chronic obstructive pulmonary disease and coronary heart disease. The treatment of tobacco-related diseases makes up an economic burden to the health care system as well as to society due to the fact that almost half of those who dies due to smoking die before the age of 70 [2]. In the process of smoking cessation, the health care staff plays an important role both as advisers and as role models. In several countries such as Denmark, Great Britain and * Corresponding author. E-mail address: [email protected] (P.N. Kannegaard). 0091-7435/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2004.11.017

Mexico, studies have been carried out to reveal the smoking habits of health care staff and their attitude towards smoking and their roles as role models [3–8]. In summer of 1999, a study (postal questionnaire) was carried out among the hospital staff at a Danish hospital revealing the habits of smoking and the attitude of the staff towards smoking. In summer of 2001, a similar study was carried out.

Aim The purposes of this study are the following: (1) to illustrate smoking habits and attitudes to smoking among a hospital staff and (2) to illustrate possible changes in these subjects over a 2-year period before an announced status for the hospital as a non-smoking hospital. From the first of

322

P.N. Kannegaard et al. / Preventive Medicine 41 (2005) 321–327

January 2002, it was no longer permitted for the staff to smoke in the hospital (this date was known to the staff in 1999 as well as in 2001) and (3) make a brief discussion on the causes of changes, if any, can be revealed, and an evaluation of the impact on the hospitals staff’s smoking habits having a department of preventive medicine inside the hospital.

Table 2 The age composition of the responders in 1999 and 2001 Age composition

1999

2001

19–34 years 35–44 years 45–54 years 55–years Total numbers

23.2% 30.9% 32.3% 13.6% N = 721

21.5% 30.2% 31.3% 17.0% N = 722

No significant differences P N 0.05. Mean age 43.14 years (1999) and 43.61 years (2001).

Materials and methods The basis of this paper is the results of the survey conducted in 1999 [9] and the survey conducted in 2001 among the hospital staff. In both of the surveys, an anonymous questionnaire was sent to every member of the staff with an addressed envelope thereby facilitating the return of the questionnaire. Questionnaires were sent by internal post to all who were employed by June 1999 and 2001. This includes full- and part-time staff. The questionnaires in June 1999 and June 2001 were identical except for the addition of one question in 2001. In 2001, the staff’s attitude towards possible sanctions towards members of staff who smoked while at work was examined. In the first study in 1999, 958 questionnaires were distributed and 678 questionnaires were returned. After a reminder, further 51 were returned corresponding to a response rate of 76%. In the study in 2001, 970 questionnaires were distributed and as an immediate response 635 questionnaires were returned. After one reminder, further 94 questionnaires were returned corresponding to a response rate of 75.2%. There is no significant difference as to the distribution of the staff into professional groups as shown in Table 1. The questionnaires comprised questions concerning smoking status, smoking habits and attitudes towards active and passive smoking, as well as questions about age, gender and profession. The staff was also asked if they had experienced any inconveniences caused by smoke in the hospital and whether or not they saw themselves as role models to the patients. Data were analyzed by chain graph models [10] examining the changes in smoking behavior in the light of background variables and attitudes towards smoking. Table 1 Distribution of the responders into professionals groups Job groups

1999

Administration Therapeuts Nursing staff Doctors Service staff Laboratory technician Others Total numbers

14.4% 3.8% 46.7% 10.7% 9.0% 6.4% 8.9% 99.9%

( ) indicates the actual number. No significant differences P N 0.05.

2001 (N (N (N (N (N (N (N (N

= = = = = = = =

105) 28) 341) 78) 66) 47) 65) 730)

15.2% (N 3.8% (N 46.2% (N 11.0% (N 9.7% (N 7.4% (N 6.6% (N 99.9% (N

= 111) = 28) = 337) = 80) = 71) = 54) = 48) = 729)

Statistical significance was evaluated using both v 2-tests and partial gamma coefficients for ordinal data [11].

Results Over the 2-year period, there has been a change in employment period. In 2001, a larger number of the staff has been employed for a short period (0–2 years) compared to the 1999-study. Fewer people have an employment period of 3–10 years, while an increased number of the staff have been employed for more than 11 years ( P b 0.001). The distribution on gender and age has not changed significantly from 1999 to 2001. Approximately 85% of the staff are women and almost 15% were men in both studies. The age distribution is shown in Table 2. In both studies, the mean age was approximately 43 years. Smoking on a daily and non-daily basis has significantly changed from 1999 to 2001 ( P b 0.013). In 1999, 33% of the staff answered that they were smokers, while in 2001 only slightly more than 26% were smoking daily or nondaily. The numbers are listed in Table 3. The numbers of exsmokers and bnever-smokersQ have increased. Tables 4 and 5 show smoking status of the responders compared with gender and job groups. In the 1999-study, 61.5% of the smokers wanted to stop smoking while 38.4% were not interested in smoking cessation. The part of the smokers who wished to stop smoking had decreased from 61.5% in 1999 to 48.4% in 2001. This is a significant decrease ( P b 0.05). In 2001, only 47% of the smokers indicated that they wanted help to quit smoking, while in 1999 this number was 76% (Tables 6a, 6b and 7). In 2001, one new question was included in the questionnaire. It should reveal the staff’s attitude towards Table 3 The smoking status of the responders in 1999 and 2001 Smoking status

1999

Smoker, daily Smoker, non-daily Ex-smoker Never smoked Total

27.4% 5.6% 24.9% 42.0% 99.9%

2001 (N (N (N (N (N

= = = = =

199) 41) 181) 305) 726)

21.7% 4.6% 27.2% 46.5% 100.0%

(N (N (N (N (N

= = = = =

157) 33) 197) 337) 724)

( ) indicates the actual number. The decrease in numbers of smokers (daily and non-daily) are significant P b 0.01.

P.N. Kannegaard et al. / Preventive Medicine 41 (2005) 321–327 Table 4 The smoking status of the responders in 1999 and 2001 distributed on gender 1999

2001

Table 6a Smoking habits among hospital staff compared with gender 1999 Smoking habits Smokes at work

Smoking status male Smoker, daily Smoker, non-daily Ex-smoker Never smoked Total

32.1% 7.5% 30.2% 30.2% 100.0%

(N (N (N (N (N

= = = = =

34) 8) 32) 32) 106)

21.2% 3.8% 35.6% 39.4% 100.0%

(N (N (N (N (N

= = = = =

22) 4) 37) 41) 104)

Smoking status, female Smoker, daily Smoker, non-daily Ex-smoker Never smoked Total

26.7% 5.3% 24.0% 43.9% 100.0%

(N (N (N (N (N

= = = = =

165) 33) 148) 271) 617)

21.6% 4.7% 25.8% 47.8% 100.0%

(N (N (N (N (N

= = = = =

134) 29) 160) 296) 619)

Smokes in own office Smokes in common room Smokes in the canteen Smokes in smoking areas Smokes in other areas Smokes during breaks Smokes after lunch

( ) indicates the actual number. P = 0.051 for males. P = 0.039 for females.

Smokes at meetings

implementing sanctions towards members of staff who broke the prohibition of smoking at the hospital. Not everybody answered this question, only 91.6% answered (663 of the 724 returned questionnaires). Among all responders, 33.5% think that sanctions should be implemented (Table 8). When compared with different variables such as gender, smoking status and job groups, there are differences in the attitude of the staff. Taking gender into consideration, the numbers show a higher level of acceptance of sanctions among men than women, cf. Table 8. A significant ( P b 0.008) higher number of women have a negative attitude towards sanctions. 68.6% of the female staff say bNoQ to sanctions whereas only 54.5% of the male staff say bNo.Q Table 9 shows that there is a clear impact of the responders smoking status on the attitude towards imple-

323

Smokes at any time

yes no yes no yes no yes no yes no yes no yes no yes no yes no yes no

1999 male

1999 female

34.9% (37) 65.1% (69) 17.9% (19) 82.1% (87) 28.3% (30) 71.7% (76) 8.5% (9) 91.5% (97) 4.7% (5) 95.3% (101) 10.4% (11) 89.6% (95) 31.1% (33) 68.9% (73) 22.6% (24) 77.4% (82) 6.6% (7) 93.4% (99) 11.3% (12) 88.7% (94)

27.1% 72.9% 4.2% 95.8% 20.3% 79.7% 4.5% 95.5% 7.9% 92.1% 2.3% 97.7% 24.8% 75.2% 17.9% 82.1% 2.7% 97.3% 5.3% 94.7%

(168) (452) (26) (594) (126) (494) (28) (592) (49) (571) (14) (606) (154) (466) (111) (509) (17) (603) (33) (587)

P 0.099 b0.0005 0.065 0.086 0.248 b0.0005 0.171 0.247 0.040 0.018

( ) indicates the actual number.

mentation of sanctions against members of staff smoking at work. When compared with job groups, there is also a difference in the staff’s attitude towards implementation of sanctions, Table 10. The problem of passive smoking is apparently declining. In 2001, more employees (of all responders) state that passive smoking is not a problem (in 2001 24.5% vs. 16.7% in 1999, ( P b 0.001)). In our study, fewer claims to be bothered by others smoking in 2001 than in 1999 (65% in 2001 against 59.9% in 1999 ( P b 0.05)). Despite the numbers listed above, there are people who are bothered by smoking. The inconveniences from smok-

Table 5 Smoking status of responder compared with job groups Smoker, daily

Smoker, non-daily

Ex-smoker

Never smoked

Total

Smoking status 1999 Administration Therapeuts Nursing staff Doctors Service staff Laboratory technician Others

30.5% 3.6% 26.2% 19.2% 54.5% 22.2% 25.0%

(32) (1) (89) (15) (36) (10) (16)

5.7% 0.0% 5.6% 10.3% 4.5% 6.7% 3.1%

(6) (0) (19) (8) (3) (3) (2)

21.9% 39.3% 25.0% 26.9% 18.2% 20.0% 31.3%

(23) (11) (85) (21) (12) (9) (20)

41.9% 57.1% 43.2% 43.6% 22.7% 51.1% 40.6%

(44) (16) (147) (34) (15) (23) (26)

100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

(105) (28) (340) (78) (66) (45) (64)

Smoking status 2001 Administration Therapeuts Nursing staff Doctors Service staff Laboratory technicians Others

22.7% 0.0% 17.9% 15.0% 56.3% 22.6% 17.0%

(25) (0) (60) (12) (40) (12) (8)

3.6% 0.0% 6.6% 3.8% 2.8% 3.8% 0.0%

(4) (0) (22) (3) (2) (2) (0)

25.5% 28.6% 31.0% 22.5% 19.7% 20.8% 29.8%

(28) (8) (104) (18) (14) (11) (14)

48.2% 71.4% 44.5% 58.8% 21.1% 52.8% 53.2%

(53) (20) (149) (47) (15) (28) (25)

100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

(110) (28) (335) (80) (71) (53) (47)

( ) indicates the actual number. P b 0.0005 in 1999 and 2001. All P values are N5% when 1999 is compared with 2001 for separate job groups.

324

P.N. Kannegaard et al. / Preventive Medicine 41 (2005) 321–327

Table 6b Smoking habits among hospital staff compared with gender 2001 Smoking habits Smokes at work Smokes in own office Smokes in common rooms Smokes in the canteen Smokes in smoking areas Smokes in other areas Smokes during breaks Smokes after lunch Smokes at meetings Smokes at any time

yes no yes no yes no yes no yes no yes no yes no yes no yes no yes no

2001 male

2001 female

21.2% 78.8% 9.6% 90.4% 17.3% 82.7% 5.8% 94.2% 2.9% 97.1% 6.7% 93.3% 20.2% 79.8% 14.4% 85.6% 2.9% 97.1% 5.8% 94.2%

20.9% 79.1% 1.1% 98.9% 13.0% 87.0% 3.5% 96.5% 8.0% 92.0% 4.0% 96.0% 19.3% 80.7% 13.2% 86.8% 1.3% 98.7% 3.7% 96.3%

(22) (82) (10) (94) (18) (86) (6) (98) (3) (101) (7) (97) (21) (83) (15) (89) (3) (101) (6) (98)

(130) (493) (7) (616) (81) (542) (22) (601) (50) (573) (25) (598) (120) (503) (82) (541) (8) (615) (23) (600)

Table 8 Attitude towards sanctions compared with gender among responders 2001 P 0.947 b0.0005 0.236 0.272 0.062 0.211 0.824 0.726 0.216 0.316

( ) indicates the actual number.

ing are, among other things, difficulty in breathing (increasing from 4.0% in 1999 to 6.2% in 2001 ( P b 0.05)) and coughing. There is no change from 1999 to 2001 in the number of staff members who are bothered by smell, headache, irritation of the eyes and concern about passive smoking. In 2001, a significant higher number of responders stated that smoking is a private matter; 81.9% agreed on this statement whereas in 1999 69.7% thought that smoking is a private matter ( P = 0.004). The correlation between the smoking status of the responder and the answer to whether or not smoking is a private matter is insignificant both in 1999 and 2001 ( P N 0.05). Table 11 shows the answers according to this matter. Both in 1999 and 2001, approximately 65% indicate that the patients adhere to the policy of smoking at the hospital. In both studies, about 40% state that the policy is also kept

Gender

Yes to sanctions

No to sanctions

Female Male Female + male

31.4% 45.5% 33.5%

68.6% 54.5% 66.5%

There is a significant difference in the way female and male responders have answered this question, P = 0.006.

by the visitors. To the question of whether or not the policy is being adhered to by the staff, there is no significant difference from the first to the second study. In 1999, 79% answered that the staff did in fact keep the policy while this number had decreased to 76% in 2001. In 1999, 56.9% of the responders recognized that they have a special obligation as a non-smoking example, that is, they must be brole models.Q In 2001, the number was 58.4%. In the first study, 52.8% acknowledge that the smoking habits of the staff affect the patients’ attitudes towards smoking. It was found that among the staff as a whole approximately 20% were satisfied with the implementation of prohibition on smoking in the hospital; the highest level of satisfaction is found among those who have never smoked and the most unsatisfied members of staff are found among daily smokers, see Table 12. The results are summarized in Fig. 1 showing a somewhat simplified version of the graphical model fitting all the variables of the study. We notice that attitudes towards smoking and smoking habits in Fig. 1 include several variables and that variables describing discomfort and other problems caused by active and/or passive smoking have been omitted. The main results of the analysis can be summarized as follows: (1)

Table 7 Wish to stop smoking compared with job groups Wish to stop smoking

1999

Administration

59.5% 40.5% 100.0% 0.0% 62.7% 37.3% 63.6% 36.4% 71.1% 29.9% 45.5% 54.5% 50.0% 50.0%

Therapeuts Nursing staff Doctors Service staff Laboratory technician Others

yes no yes no yes no yes no yes no yes no yes no

( ) indicates the actual number.

2001 (32) (15) (1) (0) (64) (38) (14) (8) (27) (11) (5) (6) (9) (9)

46.4% 53.6% 0.0% 0.0% 53.7% 46.3% 46.7% 53.3% 35.0% 65.0% 69.2% 30.8% 37.5% 62.5%

(2)

P (13) (15) (0) (0) (43) (37) (7) (8) (14) (26) (9) (4) (3) (5)

0.260 – 0.407 0.319

Smoking habits and attitudes are associated in a way that cannot be completely described by variation in age, gender, job group, employment period or calendar year. No assumption has been made concerning a simple causal relationship between attitudes and smoking. Smoking habits are not directly related to job group, employment period, gender or age. Attitudes on the other hand depend on these variables mediating an indirect effect of these variables on smoking habits.

Table 9 Attitude towards sanctions compared with smoking status of responder 2001

0.014 0.403 0.297

Smoker, daily Smoker, non-daily Ex-smoker Never smoked Total

Yes to sanctions

No to sanctions

Total

8.8% 9.4% 36.7% 46.2% 33.5%

91.2% 90.6% 63.3% 53.8% 66.5%

100.0% 100.0% 100.0% 100.0% 100.0%

(13) (3) (65) (140) (221)

( ) indicates the actual number.

(135) (29) (112) (163) (439)

(148) (32) (177) (303) (660)

P.N. Kannegaard et al. / Preventive Medicine 41 (2005) 321–327 Table 10 Attitudes towards sanctions compared with job groups 2001 Job groups

Yes to sanctions

No to sanctions

Administration Therapeuts Nursing staff Doctors Service staff Laboratory technician Others

43.3% 26.9% 27.3% 43.8% 29.0% 40.8% 38.1%

56.7% 73.1% 72.7% 56.2% 71.0% 59.2% 61.9%

(N (N (N (N (N (N (N

= = = = = = =

45) 7) 82) 32) 20) 20) 16)

(N (N (N (N (N (N (N

= = = = = = =

59) 19) 218) 41) 49) 29) 26)

( ) indicates the actual number.

(3)

(4)

Smoking habits are on the other hand directly related to calendar year while attitudes on the other hand are only indirectly related to calendar year due to changes in job profiles and changes in smoking habits. This implies that smoking habits have changed over time in ways that appear not to be the cause of changes in attitudes towards smoking. There has been no change in the gender and age distribution among the hospital staff.

Discussion The two studies from 1999 and 2001 are comparable regarding design and distribution of the responders on professional groups, gender and age. The two main results of the study are the following: (1) the number of smokers among the staff has declined significantly thereby facilitating the staff’s effort to advice and help the patients to quit smoking and (2) a fewer number of staff are bothered by passive smoking. The response rate in both the 1999 and 2001 studies seems rather high. This may be due to several reasons. Firstly, the topic of smoking cessation has been debated continuously from 1998 among the hospital staff, and secondly the unit conducting the survey has a remarkable position among all employees of the hospital. Thirdly, preventive medicine and interventions have been encouraged strongly by the leadership of that specific hospital for several years.

Table 11 Smoking status of responder and attitude towards whether or not smoking is a private matter Smoking is a private matter

Total

Yes

No

1999 Non-smoker Smoker

58.6% (17) 71.6% (149)

41.4% (12) 28.4% (59)

100.0% (29) 100.0% (208)

2001 Non-smoker Smoker

71.4% (15) 83.2% (139)

28.6% (6) 16.8% (28)

100.0% (21) 100.0% (167)

( ) indicates the actual number.

325

Table 12 Satisfaction with prohibition on smoking in the hospital compared with smoking status of responder 1999 Smoker, daily Smoker, non-daily Ex-smoker Never smoked Total 2001 Smoker, daily Smoker, non-daily Ex-smoker Never smoked Total

Satisfied

Not satisfied

Total

48.5% (N = 94) 87.8% (N = 36)

51.5% (N = 100) 100.0% (N = 194) 12.2% (N = 5) 100.0% (N = 41)

88.2% (N = 157) 11.8% (N = 21) 100.0% (N = 178) 95.2% (N = 277) 4.8% (N = 14) 100.0% (N = 291) 80.1% (N = 564) 19.9% (N = 140) 100.0% (N = 704)

21.1% (N = 43) 90.3% (N = 28)

70.9% (N = 105) 100.0% (N = 148) 9.7% (N = 3) 100.0% (N = 31)

87.2% (N = 164) 12.8% (N = 24) 100.0% (N = 188) 96.6% (N = 311) 3.4% (N = 11) 100.0% (N = 322) 79.2% (N = 546) 20.8% (N = 143) 100.0% (N = 689)

( ) indicates the actual number. P b 0.0005 in 1999 and 2001.

It is a known fact that the smoking habits and attitudes to smoking of the staff affect the patients and their attitudes [12]. When the staff’s habits are changed in one direction, that is, more smokers become non-smokers, the staff’s attitude towards smoking cessation will also make more smoking patients interested in smoking cessation [12]. Furthermore, evidence that effective support for smoking cessation can be delivered through the health care system is substantial [13]. Smoking is the largest single preventable cause of death, and smoking cessation intervention makes up relative modest costs compared to other medical interventions. One study showed that the average medical interventions cost approximately 17,000 o per life year gained for smokers compared to the costs of smoking cessation intervention which are about 212–873 o per life year gained [13]. The results of this second questionnaire survey showed two obvious changes: First, there was a significant decrease in the number of smokers from 1999 to 2001. In 2001, 26% were smokers compared to 33% in 1999. In a similar study conducted in 1998 at the County Hospital of Glostrup– another Danish Hospital–it was found that 27% of the female staff and 31% of the male staff were smokers [14]. In a study conducted at Bispebjerg Hospital in Copenhagen in 1992 and 1999, the proportion of smokers decreased from 42% in 1992 to 33% in 1999 [4]. In the Danish population, 34.0% were found to be smokers in 2000 [15]. This number is decreasing, but not as fast as among hospital staff [7]. This fact is assumed to be due to the intensive influence on the hospital staff. Second, there are significant fewer staff members who are bothered by passive smoking. These two tendencies confirm each other. The fact that there is a fewer number of smokers among the staff in 2001 can be due to the fact that the habits of

326

P.N. Kannegaard et al. / Preventive Medicine 41 (2005) 321–327

Fig. 1. Simplified independence summarizing the results of the analysis (employment period = Empl. Per.). Summary independence graph illustrating the main results of the statistical analysis. Each node in the graph represents either single variables or in some cases, e.g. attitude and smoking habits collections of variables. There is no link in the graph if variables are conditionally independent given all concurrent or prior variables. Links between variables therefore refer to direct association between variables. We emphasize that the independence graph shown here is a simplified graph illustrating the main results of the analysis. The real independence graph behind the analysis is too complex with too many variables to be of any help as a visual presentation of the results.

the staff have changed, but it can also be caused by the fact that new staff members primarily are non-smokers and that it is possible that smokers have left the hospital for other jobs. In the Glostrup-study, there was a difference in the number of smokers when divided into job groups. Members of the bService staffQ-group were found to be the most smoking staff members, and the lowest numbers of smokers were found among therapeuts [14]. This difference was confirmed in our study. This study shows that smokers are less willing to accept help to quit smoking in 2001 compared to the smokers in 1999 indicating that probably it is the inveterate smokers who have not stopped yet. When our study was conducted in 2001, only half a year remained before the hospital became a no-smoking hospital. After the first study in 1999, many initiatives were made to focus on the importance of smoking cessation, such as posters, information, competition and free smoking cessation courses for the staff. Not everyone was satisfied with the decision to turn the hospital into a no-smoking workplace. Our study could not show that the staff’s attitude towards smoking has been changed due to the special preventive effort at the hospital over this 2-year period. The aim for the preventive work has been to change the staff’s knowledge on smoking and thereby their smoking habits. Results show that the habits have changed, whereas the data are not able to show any effect on the staff’s attitude.

Although the numbers of smokers have decreased probably in part due to the smoking cessation initiatives at the hospital, it is remarkable that the staff, that are still smokers, are far less willing to accept help to quit smoking. Grizeani et al. found that 37% of hospital staff think that it is their obligation to advice about smoking cessation whereas 81% think that they are brole modelsQ [16]. In the 1999-study, almost 57% agreed that hospital staff has a special obligation as no-smoking brole modelsQ; this number had not changed significantly in the 2001-study where 58.4% agreed on this matter. In regard to whether or not the attitudes towards smoking among the patients are influenced by the smoking habits of the staff, approximately 53% of the responders agreed in both studies. To be aware of ones role as a role model is more common among ex-smokers and non-smokers; these persons are also more aware of their part in counseling [8]. Of the responders in 2001, 24.5% answer that passive smoking is not a problem compared to 16.7% in 1999. This is a significant change ( P b 0.001). Another study conducted in Great Britain showed that the majority (57%) of staff was bothered by passive smoking [17]. As a conclusion, we reached answers to 2 of the 3 aims of the study. However, it cannot from our results be concluded whether the establishment of a hospital department of preventive medicine has an impact on the smoking of the staff. We conclude that there is a significant decrease

P.N. Kannegaard et al. / Preventive Medicine 41 (2005) 321–327

of smokers among the staff, but an explanation of this difference is not answered by our data. In spite of a lack of a clear answer, preventive medicine should be provided with resources, among other reasons, due to the cost-effectiveness of smoking cessation shown in former literature.

References [1] bHealth in the 21st Century. The health policy of WHO in EuropeQ [in Danish: bSundhed i det 21. 3rhundrede. Rammen for sundhed for alle politikken i WHO’s europ&iske region.Q]. Danish Association of Nurses. [in Danish: Dansk Sygeplejer3d]; 1999. [2] Peto R, et al. Mortality from smoking in developed countries. WHO; 1994. [3] Ferguson P, Small WP. Further study of the smoking habits of hospital nurses. Health Bull 1985;43/1:13 – 8. [4] Olsen AD, Fugleholm AM, Rasmussen S, Backer V, Jbrgensen SJ, Tbnnesen H, et al. Smoking behaviour and exposure to passive smoking among hospital employees 1992–1999. Ugeskr L&ger 2000; 162:5623 – 7. [5] Tapia-Conyer R, Cravioto P, De la Rosa B, Galva´n F, Garcı´a-de la Torre G, Kuri P. Cigarette smoking; knowledge and attitudes among Mexican physicians. Salud Pu´blica Mex 1997;39:507 – 12. [6] Kawakami M, Nakamura S, Fumimoto H, Takizawa J, Baba M. Relation between smoking status of physicians and their enthusiasm to offer smoking cessation advice. Intern Med 1997;36:162 – 5.

327

[7] Nielsen PE, Falk J, Danielsen US. Smoking habits and attitudes towards the tobacco issue among health professionals in Denmark in 1996. Ugeskrift 2000;162:4140 – 4. [8] Reeve K, Adams J, Kouzekanani K. The nurse as exemplar. Cancer Pract 1996;4:31 – 3. [9] Grunnet K, Gregersen P, Skov P. Smoking habits and attitudes to smoking. A questionnaire study among the staff at Roskilde County Hospital Koege. Roskilde County Hospital Koege7 Department of Occupational Medicine; 1999 [Free copy available at Department of Occupational Medicine]. [10] Lauritzen SL. Graphical models. Oxford7 Clarendon Press; 1996. [11] Agresti A. Analysis of ordinal categorical data. New York7 John Wiley and Sons; 1984. [12] Willaing I. Smoking and prevention at a hospital [in Danish: Rygning og forebyggelse i sygehusregi]. Copenhagen; 2000. [13] Raw M, McNeill A, West R. Smoking cessation: evidence based recommendations for the healthcare system. BMJ 1999;318: 182 – 5. [14] Heitmann BL, Brink-Kj&r T, Jbrgensen T. Results of a questionnaire survey about smoking among the staff at the County Hospital of Glostrup [in Danish: Resultater fra spbrgeskemaundersbgelsen om rygning blandt personalet p3 Amtssygehuset i Glostrup]. Copenhagen; 1999. [15] Kjbller M, Rasmussen NK, [red]. Health and morbidity in Denmark 2000 [in Danish: Sundhed og sygelighed i Danmark 2000]. National Institute of Public Health; 2002. [16] Grizeau D, Baudier F, Doucet C, Lemarie´ Y. Smoking behaviour of hospital staff. Rev Mal Respir 1998;15:79 – 87. [17] Hurst K, Jones R, Pulleyblank P, Hickling D. Attitudes to smoking among health care staff. Nurs Times 1990;86:52.