- Email: [email protected]
One-Year Longitudinal Study of a No-Smoking Policy in a Medical Institution
special report One- Year LonQitudinal Study of a No-Smoking Policy in a Medical Institution* Leonard G. Rudzinski, Ph.D.; and Edward D. Frohlich, M.D., F.C.C.P.
Smoking and health are increasingly understood to be incompatible. To evaluate, prospectively and retrospectively, the attitudes of employees, staff physicians, and patients of a medical institution, a questionnaire was administered before and after implementation of a nosmoking policy. or many questions, select ones reviewed here focused on the following concerns: (I) how tobacco smoke affects employees and patients, (2) employee acceptance of a no-smoking policy before and after its implementation, and (3) the consequences of the policy on employee smokers. Open-ended questions about smoking were constructed by a committee comprised of clinicians, investigators, and administrators. The questionnaire was given to 2,000 randomly selected patients and the institution's entire staff of 4,200 employees and 225 staff physicians. Data were obtained on three occasions: six months before, six months after, and one year after the implementation of the no-smoking policy. The majority of patients, employees, and physicians indicated that the smoke of others bothered them
the first national smoking poll was conW hen ducted in 1944, 41 percent of adults aged 17 years and older indicated they were smokers. In 1987, the US Office on Smoking and Health national survey indicated that the overall smoking rate in the United States decreased from 36.7 to 30.4 percent (52.4 to For editorial comment see page 1027 51.1 million adults) from 1976 to 1985. Although these data suggest a sharp decline in smoking, they do not indicate that the largest growing segment of the smoking population is people under 17 years of age. Over the past 20 years, the health consequences of smoking have crystallized so that, at present, smoking is considered the major preventable cause of death and disability in this country, the major preventable risk factor for coronary heart disease, and the major cause of emphysema, lung cancer, and other forms of cancer. As a result of these dramatic statistics, health care *Department of Psychiatry (Dr. Rudzinski) and Vice President for Academic Affairs (Dr. Frohlich), Ochsner Clinic and Alton Ochsner Medical Foundation, New Orleans.
1198
and ranked the following as most offensive: smell, eye irritation, provocation of sinus problems, coughing, and headache. Approximately 80 percent of employees and patients favored the policy before its inception; and employees increasingly favored it through the year after its implementation with a favorable attitude increasing by nearly 10 percentage points. One year after implementation, 80 percent of patients were in favor of the policy. In the final survey, 74 percent of respondents indicated the policy had helped them. One year after policy implementation, employee smoking was reduced significantly from 22 to 14 percent; and of those employees who continued to smoke, 81 percent smoked less than eight cigarettes per day. This study underscores the benefit of a work-site nosmoking policy in a health care provision setting. This overwhelming approval of smoke restriction supports the conclusion that employees, patients, and staff welcome a smoke-free environment. (Chest 1990; 97:1198-1!02)
providers, volunteer health agencies, private employers, and federal, state, and local governments have established smoking regulations aimed to protect nonsmokers from exposure to tobacco. These regulations and changes then are particularly based on evidence supporting the passive smoking argument: nonsmokers exposed to tobacco smoke absorb a variety of toxic inhalants that affect health as is articulated in the Surgeon General's Report on "The Health Consequences of Involuntary Smoking, 1986." The Ochsner Medical Institutions (New Orleans, La) have been a prime health care provider advocating this health risk ever since one of our founders, Alton Ochsner, first reported the association of smoking and lung cancer in 1939. 1 More recently, we have established one of the first health care institution policies that enforced a comprehensive campus-wide smokefree environment. (Smoking is permitted on the acute psychiatry inpatient unit by physician approval.) A questionnaire was used to measure attitudes of patients, employees, and staff before and one year after implementation of the no-smoking policy. As part of the questionnaire we specifically asked the following: No-Smoking Polley in a Medic:allnslllutlon (Hudzlnskl, Frohlich)
(1) How does the smoke of others bother you in the institution? (2) How favorably would you rate a nosmoking policy for this institution? (3) How has the policy changed your smoking frequency both in and outside the institution? (4) How much is the policy a help to you at work? METHODS
lblicy Development Early in 1986, after the creation of a Smoke-Free 'Illsk Force that was directed to learn about the attitudes of all concerned about smoking and the possible promulgation of a comprehensive policy, 4,200 employees, 225 staff physicians, and 2,000 patients were invited to respond to a questionnaire designed by the 'Illsk Force to provide opinions about an institution-wide "no-smoking" policy. Six months thereafter, following analyses of these data, extensive discussions, and planning, the policy was implemented. The same individuals were invited again to respond to a nearly identical questionnaire used to evaluate the policy six and 12 months thereafter. For purposes of analyses, all physician data were collapsed into the employee response category. The 'Illsk Force included clinicians, psychologists, and administrative personnel from the public affairs and employee relations departments. Open-ended questions about smoking were used to interview employees and patients. Pretest responses were then integrated into the subsequent standardized Likert, multiple alternative attitude scale questionnaire that was administered on a voluntary basis to randomly selected patients and our entire medical and scientific staff. Likert scale items permit the expression of several (usually five) degrees of agreement-disagreement. The questionnaire was reviewed and approved by our institutional review board for human investigation. The questionnaire satisfied several goals: it assisted the 'Illsk Force in formulating the smoking policy, it served as an evaluation tool across time measuring changes that resulted from the policy, and it was selectively used for the investigative interests of the group. In this respect, three select areas of research interest and inquiry, which we report on in this article, involved how tobacco smoke affects employees or patients while at the institution, the acceptance of a no-smoking policy before and after its implementation, and the consequences of the policy on the smoker. The questionnaire was introduced with a statement of the purpose and with instructions for completion. The questionnaire was mailed to all employees and to more than 2,000 randomly selected patients. Respondents were assured that their answers would be kept confidential and were instructed to complete the questionnaire
within one week, returning it in a confidential envelope. Responses to the questionnaire, involving nominal and ordinal levels of data, were coded and the data were analyzed using survey statistical methods.• The results of this study will be particularly confined to responses of employees who experience the major impact of the institution policy.
REsuLTS In all, three surveys were obtained: the initial survey, six months before policy implementation; the second, six months after policy implementation; and the third, after the policy had been in effect for one year. In the prepolicy baseline study, of 4,200 distributed employee questionnaires, 1,946 (46 percent) were returned. In the second survey, 1,608 (38 percent) questionnaires were returned, but for the third study, one year after policy implementation, only 684 (16 percent) questionnaires were completed, perhaps possibly indicating general acceptance of the program. This latter lower response rate of the ,most recent survey was expected as a function of the policy being less controversial with employees feeling that they had voiced their opinion on two previous occasions.
Demography There was a significant reduction in the smoking habits of our employees one year after implementation of the no-smoking policy. Thus, six months before and after the policy was instituted, 22 and 20 percent, respectively, of our employees indicated that they smoked, but after one year this was reduced by 30 percent to 14 percent for all employees (x2 = 11.53; p<0.003). These data, howe-ver, failed to show an impact by gender of the employee (Thble 1). Most nonsmokers were under the age of 35 years (52 percent), and almost two thirds of smokers (63 percent) were 35 years and older; but of these, more older individuals stopped smoking. These data further indicated that women were more likely than men to be smokers (79 percent of smokers were women).
Table I- Demographic Profile of Employees' Smoking Status One Year After Policy
,.,
Exsmoker,
,.,
Total,
N*
,.,
Nonsmoker,
300 202 182
37 30 33
52 26 22
25 39 36
44 29 27
199 485
21 79
31 69
30 70
29
71
562 122
85
85
15
15
73 27
82 18
Smoker, Age, yr <35 35to44 c:!!45 Gender Male Female Position Employee Physician
,.,
Statistics
x•
df
p
34.46
4
0.0000
3.490
2
0.1746
11.91
2
0.0026
*N=684. CHEST I 97 I 5 I MAY, 1990
1199
Table 2-Effect ofNo-Smolcing Policy on Choice oflleolth Care Facility• Smokers(%) Nonsmokers(%) Exsmokers (%) More likely to seek institution's healthcare No effect on decision Less inclined to use Ochsner
2 (1.7)
51 (44.3)
19 (16.5)
8 (7.0)
9 (7.8)
6 (5.2)
20 (66.7)
*N = 115; x•=78.8; df=4; p=O.Ol.
Acceptance Nearly 80 percent of employees, a clear majority, favored the no-smoking policy; this acceptance persisted one year after its implementation. Within the last six months of the analysis, this favorable attitude increased nearly 10 percentage points between this study's last two surveys (p<0.001). It was of interest to learn, albeit with a relatively small subsample of 115 patients of whom 20 were smokers, that when asked how this policy would affect their future choice of a medical facility, 67 percent indicated that they would be less likely to use a "nosmoking" facility in the future (Table 2). In other words, the majority of patients who smoked indicated less inclination to use a health care facility that prohibited smoking. However, of those patients who
.--.. ~
0 ........
•atcv ....c ...•u • a.
(684) (1946)
77
(1608)
75
........... ················ ,•,•,•,•,•,•,•,•,•,•, .......... ........... ,•,•,•,•,•,•,•,•,•,•, ·:·:·:·:·:·:·:·:·:·:· ........... .......... ::::::::::::::::::::: ........... ·:·:·:·:·:·:·:·:·:·:· ........... ,•,•,•,•,•,•,•,•.•,•,
.......... ........... .......... .......... ............ ......... ........... ·········•··········· ........... ........... .......... .·•···•········•····· ........... .·.·.················ .......... .•.·.·.·.·.·····••··· ........... ·····•···•··········• ·:·:·:·:·:·:·:·:·:·:· ........... .......... ........... .......... ........... ::::::::::::::::::::: ,•,•,•,•,•,•,•,•,•,•,
,•,•,•,•,•,•,•,•,•,•,
'•'•'•'i'•'•'•'•'•'•'
.·.·•·········•······
Employees FIGURE
1200
Consequences In the survey conducted prior to implementation of the policy, both employees and patients were asked to what extent the smoke of others bothered them within the confines of the medical institution. The majority of employees indicated that the smoke of others bothered them. Of these individuals, 35 percent indicated that the smoke of coworkers affected them a great deal; 22 percent responded that the smoke of others did not bother them. Of those who indicated that the smoke of others had an adverse effect, 66 percent were smokers. Patient response rates were nearly identical to the employee response rates {Fig 1). In this regard, the most offensive effects were smell, eye irritation, and provocation of sinus problems, cough, and headache. One major consequence of interest was the possible change in the pattern of smoking while employees were at work. In both surveys, six and 12 months after implementation of the policy, approximately one fourth of smokers indicated that they no longer smoked cigarettes at work. However, approximately 40 percent of smokers indicated at both the six- and 12-month surveys that their cigarette consumption remained unchanged after leaving the medical facility at the end
I!] 6 Months Post-Policy
[]Pre-Policy
100· 90· 80· 70· 60· 50· 40· 30· 20· 10· 0
were not in favor of the policy, 20 percent were themselves smokers.
84
(607)
82
m1
Year Post-Policy
(397)
93
.......... ........... .·.··················
•,•,•,•,•,•,•,•,•,•,•,
·:·:·:·:·:·:·:·:·:·:· ........... ·:·:·:·:·:·:·:·:·:·:·
(600)
80
::::::::::::::::::::: .·.·········•········ ·::::::::::::::::::::: :·:·:·:·:·:·:·:·:·:·: ::::::::::::::::::::: ::::::::::::::::::::: :·:·:·:·:·:·:·:·:·:·: .·.···········•··••·· ::::::::::::::::::::: ·:·:·:·:·:·:·:·:·:·:· :·:·:·:·:·:·:·:·:·:·: •,•,•,•,•,•,•,•,•,•,• :·:·:·:·:·:·:·:·:·:·: ·:·:·:·:·:·:·:·:·:·:· ·:·:·:·:·:·:·:·:·:·:·: ·.·.·.·.·.·•·.·•·.·.·
...........
..........
......... .
.
,•,•,•,•.•,•,•,•,•,•, ,•,•,•,•,•,•,•,•,•,•,
::::::::::::::::::::: .·.········•···•····· ·:·:·:·:·:·:·:·:·:·:· •·•··••··•·••·•••·•·• •'•'•'•'•'•'•'•'•'•'•
........... ..........
Patients
1. Employee and patient attitude toward policy (percent agreeing with policy). No-Smoking Polley in a Medical Institution (Hudzinsld, Frohlich)
of a workday. In this regard, 23 percent of employees who smoked reported smoking less, but 35 percent smoked more after work hours since the policy went into effect. Those who indicated that they smoked more were in the 35- to 44-year-old age group, had smoked for ten or more years, and were women. Notwithstanding the foregoing frustrating responses, after one year there was a significant decrease in the number of employees who smoked (20 vs 14 percent; p<0.01). Moreover, the tobacco users indicated that they smoked fewer cigarettes after one year in comparison with the previous years data. Thus, 81 percent of smokers reported using less than eight cigarettes per day after one year. These data are comparable with the national finding that significant numbers of smokers had either quit using tobacco or reduced the number of cigarettes. 3 These national trends run parallel to the first survey taken one year earlier, when more than one fourth of smokers (28 percent) indicated that they intended to stop smoking ifour institution implemented a policy; and the most recent survey indicated that most who expressed that interest had attempted to do so. Indeed, 25 percent of employees reported that they physically tried to stop smoking at the time of the six-month postpolicy study, and 21 percent reported that they had physically tried to stop smoking one year after the policy was begun. Furthermore, at six and 12 months, approximately 60 percent of smokers indicated that they wished to stop smoking in the future. The majority of our employees reported the nosmoking policy had helped them by decreasing irritants while at work. Most employees indicated eliminating smoking stopped their experiencing burning eyes, sinus problems, cough, headaches, and the offensive smoking smell. In the final survey, 74 percent indicated that the policy had helped them; and of these, more than 50 percent indicated that the policy had helped a great deal. Thus, 7 percent of employee exsmokers indicated they had stopped smoking directly because of our institution's no-smoking policy; and 30 percent of employees believed that their work performance had improved as a result of this policy, whereas 60 percent believed it had no effect at all. DISCUSSION
The results of this report provide the most comprehensive analysis to date of a no-smoking work-site policy. They indicate that both employee and patients overwhelmingly favored the restriction of smoking at our health care institution. It could be argued that the lower response rate to the third administration of the questionnaire, 684 returned, does not support a conclusion that the policy was widely accepted. Wide acceptance is concluded based on overwhelmingly favorable ratings at both the first survey six months
before and the second survey six months after policy implementation. This conclusion is also based on no verbal or demonstrated opposition to this wide-reaching policy. As noted earlier, the lower response rate is particularly based on employees verbalizing they had completed the questionnaire twice before and had voiced their opinion. This study is concordant with smoking cessation trends in national surveys that indicate a significant number of smokers were in favor of, and wanted to stop smoking. Moreover, our results indicate that of those who had continued to smoke, the majority reduced their daily cigarette consumption. Our findings suggest that a no-smoking policy may stimulate if not actually provoke those who intended to stop smoking to cease their smoking. Although national data indicate that maintaining cessation from smoking may involve multiple attempts, a no-smoking policy may play a pivotal role in reducing the frequency of smoking, maintaining cessation, and improving both the health of smokers and nonsmokers alike. These results are also similar to the findings of the national smoking survey conducted by Gallup, 4 that found that 85 percent of nonsmokers, 78 percent of former smokers, and 62 percent of smokers agreed that smokers should not smoke near nonsmokers. This trend underscores the contention that, as a nation, we welcome a smoke-free environment. Industrial studies have assessed employee attitude in planning and implementing their own smoking policies. Pacific Telephone and Texas Instruments corporation are two of many that have undertaken employee surveys. They indicate on average that eight of ten nonsmoking employees were bothered by the smoke of others at work; and most smokers favor a smoke-free policy. 5 Although these studies surveyed employees only once (ie, preceding policy implementation), they provide support for efforts in establishing a smoke-free work environment. Assessment of employees has also been undertaken in health care settings. 6 These studies support establishing smoke restricting policies; and the only employee group that has not accepted the policy is the employee who smokes.7 In one study, Rigotti and associates& noted that although only 35 percent of smokers supported the ban at baseline, one year after implementation of a no-smoking policy this support had increased to 67 percent. Rosenstock" indicated similar trends when hospital no-smoking policies were instituted and that the smokers' initial reluctance diminished with time. Two recent studies, one conducted at Mayo Clinic and the other at the Oregon Health Sciences University, have reported on their methods for developiDg their smoke-free policies. 10•11 Six months after implementing its policy, each institution reported wide CHEST I 97 I 5 I MAY, 1990
1201
acceptance of the smoking ban by both staff and patients. Each concluded that a no-smoking policy CaQ be implemented if it is well planned and supported by staff. Our findings are in accord with this conclusion. As with all longitudinal surveys, a studys findings must be tempered by the possibility that uncontrolled factors may have inHuenced the results. In our study, repetitive questionnaires may have sensitized employees and patients in their responses to questions. This may have inH_'!lenced their tobacco use patterns or the honesty of answers irrespective of the policy's impact. Our institutions concurrent ongoing 2o-yeaN>Id, individual, group, nicotine replacement smoking cessation programs may have been another factor or interacting force that inftuenced employees' attitudes rather than the policy itself: Still another factor may be the national trend, rigorously supported by the former Surgeon General,·in stopping this, the major national caustl of preventable death and disability. No-smoking policies and their. growing prevalence in the workplace, and in health care facilities in particular, oiay have dramatic consequences. They increase general knowledge of health risks about smoking and may promote nonsmoking as a more normative behavior. Prohibition of smoking may help reduce the peer pressure to smoke and, conversely, increase social support for nonsmoking. No-smoking policies could, in fact, increase the success of a smoker's intent to quit and to decrease daily cigarette consumption. No-smoking policies are rapidly emerging. Since we addressed and implemented ours, all health care facilities in our community have implemented similar policies. Smokers consequently are perceiving little choice in finding health care providers who permit smoking within our community. This is becoming a growing and national trend. Smo~g policies may have other significant benefits beyond the issue itself: A smoke-free policy can directly impact on employers by decreasing the costs resulting from employees who smoke with respect to increased indirect costs from absenteeism, increased health care utilization, and higher health insurance rates. 12 Clearly, nonsmokers and exsmokers may be expected to favor smoking restrictions; it requires no change in behavior and offers the benefits of reduced health risks. Opinions of smokers are expected to be less favorable toward a restricted policy that, ifanything, is inconvenient. Smokers may argue that they have the least to gain from a no-smoking policy and they are valued employees who should not be confronted with a punitive policy. Smoking is an undesir-
1202
able habit and a major preventable cause of death and disability; and this institution continues to emphasize that message as often as possible. Likewise, our institution remains committed to its employees, to the smoker and nonsmoker, to job satisfaction, and continued quality performance although we unequivocally support non-smoking efforts. To this end, we provide an alternative for the smoker-an off-the-main-campus setting where those who smoke can meet the need of their dependency-while at the same time we emphas~e smoking cessation and smoking treatment programs, including group counseling and nicotine replacement therapy. In conclusion, we strongly advocate that medical institutions should take the lead as responsible proponents of sociomedical policy by setting an example for the public as it relates to health care issues. We have tested a no-smoking policy and found it to be desirable for our medical institution and for the nonsmoker and smoker alike. ACKNOWLEDGMENTS: The assistance of Mr. Charles Madanick, Director, DiviSion of Marketing/Public Affairs, and his staff, including Ms. Connie Baer and Ms. CoUene Terro for statistical help and advice, is gratefuUy acknowledged. REFERENCES
1 Ochsner AD, DeBakey M. Primary pulmonary malignancy: treatment by total pneumonectomy: analysis of 79 collected cases and presentation of7 personal cases. Surg Gynecol Obstet 1939; 68:435-51 2 Rosenberg M. The strategy of survey analysis: the logic of surver analysis. New York: Basic Books Inc; 1968; 75 3 Department of Health and Human Services. The health consequences of smoking, nicotine addiction. Report of the Surgeon General; Washington, DC. US Government Printing Office, 1988; 16 4 Gallup Organization. Survey of attitudes towards smoking: American Lung Association survey. Princeton, NJ: Gallup Or~n. July 1985 5 Pacific Telephone. Employee smoking study. January 1983 6 Rosenstock IM, Stergachis A, Heaney C. Evaluation of smoking prohibition policies in a health maintenance organization. Public Health Briefs 1986; 76:10-14 7 Koop CE: Smoking in the work place. Corporate Fitness and Recreation; June-July 1986; 35 8 Rigotti NA, Hill PB, Cleary P, Singer DE, Mulley AG. The impact of banning smoking on a hospital ward: acceptance, compliance, air quality and smoking behavior. Clin Res 1986;
34:833A 9 Rosenstock IM. Evaluation of smoking prohibition policy in a health maintenance organization. Public Health Briefs 1986; 76:1014-15 10 Hurt RD, Berge KG, Offord KP, Leonard DA, Gerlach DK, Renquist CL. The making of a smoke-free medical center. JAMA 1989; 261:91).7 11 Barker AF, Moseley JR, GlideweU BL. Components of a smokefree hospital program. Arch Intern Med 1989; 149:1357-59 12 Hallitt R. Smoking intervention in a work place: review and recommendations. Prev Med 1986; 15:213-17
No-Smoking Poley in a Macllc:allnstllullon (Hudzlnald, Frohlich)
Smoking and health are increasingly understood to be incompatible. To evaluate, prospectively and retrospectively, the attitudes of employees, staff physicians, and patients of a medical institution, a questionnaire was administered before and after implementation of a nosmoking policy. or many questions, select ones reviewed here focused on the following concerns: (I) how tobacco smoke affects employees and patients, (2) employee acceptance of a no-smoking policy before and after its implementation, and (3) the consequences of the policy on employee smokers. Open-ended questions about smoking were constructed by a committee comprised of clinicians, investigators, and administrators. The questionnaire was given to 2,000 randomly selected patients and the institution's entire staff of 4,200 employees and 225 staff physicians. Data were obtained on three occasions: six months before, six months after, and one year after the implementation of the no-smoking policy. The majority of patients, employees, and physicians indicated that the smoke of others bothered them
the first national smoking poll was conW hen ducted in 1944, 41 percent of adults aged 17 years and older indicated they were smokers. In 1987, the US Office on Smoking and Health national survey indicated that the overall smoking rate in the United States decreased from 36.7 to 30.4 percent (52.4 to For editorial comment see page 1027 51.1 million adults) from 1976 to 1985. Although these data suggest a sharp decline in smoking, they do not indicate that the largest growing segment of the smoking population is people under 17 years of age. Over the past 20 years, the health consequences of smoking have crystallized so that, at present, smoking is considered the major preventable cause of death and disability in this country, the major preventable risk factor for coronary heart disease, and the major cause of emphysema, lung cancer, and other forms of cancer. As a result of these dramatic statistics, health care *Department of Psychiatry (Dr. Rudzinski) and Vice President for Academic Affairs (Dr. Frohlich), Ochsner Clinic and Alton Ochsner Medical Foundation, New Orleans.
1198
and ranked the following as most offensive: smell, eye irritation, provocation of sinus problems, coughing, and headache. Approximately 80 percent of employees and patients favored the policy before its inception; and employees increasingly favored it through the year after its implementation with a favorable attitude increasing by nearly 10 percentage points. One year after implementation, 80 percent of patients were in favor of the policy. In the final survey, 74 percent of respondents indicated the policy had helped them. One year after policy implementation, employee smoking was reduced significantly from 22 to 14 percent; and of those employees who continued to smoke, 81 percent smoked less than eight cigarettes per day. This study underscores the benefit of a work-site nosmoking policy in a health care provision setting. This overwhelming approval of smoke restriction supports the conclusion that employees, patients, and staff welcome a smoke-free environment. (Chest 1990; 97:1198-1!02)
providers, volunteer health agencies, private employers, and federal, state, and local governments have established smoking regulations aimed to protect nonsmokers from exposure to tobacco. These regulations and changes then are particularly based on evidence supporting the passive smoking argument: nonsmokers exposed to tobacco smoke absorb a variety of toxic inhalants that affect health as is articulated in the Surgeon General's Report on "The Health Consequences of Involuntary Smoking, 1986." The Ochsner Medical Institutions (New Orleans, La) have been a prime health care provider advocating this health risk ever since one of our founders, Alton Ochsner, first reported the association of smoking and lung cancer in 1939. 1 More recently, we have established one of the first health care institution policies that enforced a comprehensive campus-wide smokefree environment. (Smoking is permitted on the acute psychiatry inpatient unit by physician approval.) A questionnaire was used to measure attitudes of patients, employees, and staff before and one year after implementation of the no-smoking policy. As part of the questionnaire we specifically asked the following: No-Smoking Polley in a Medic:allnslllutlon (Hudzlnskl, Frohlich)
(1) How does the smoke of others bother you in the institution? (2) How favorably would you rate a nosmoking policy for this institution? (3) How has the policy changed your smoking frequency both in and outside the institution? (4) How much is the policy a help to you at work? METHODS
lblicy Development Early in 1986, after the creation of a Smoke-Free 'Illsk Force that was directed to learn about the attitudes of all concerned about smoking and the possible promulgation of a comprehensive policy, 4,200 employees, 225 staff physicians, and 2,000 patients were invited to respond to a questionnaire designed by the 'Illsk Force to provide opinions about an institution-wide "no-smoking" policy. Six months thereafter, following analyses of these data, extensive discussions, and planning, the policy was implemented. The same individuals were invited again to respond to a nearly identical questionnaire used to evaluate the policy six and 12 months thereafter. For purposes of analyses, all physician data were collapsed into the employee response category. The 'Illsk Force included clinicians, psychologists, and administrative personnel from the public affairs and employee relations departments. Open-ended questions about smoking were used to interview employees and patients. Pretest responses were then integrated into the subsequent standardized Likert, multiple alternative attitude scale questionnaire that was administered on a voluntary basis to randomly selected patients and our entire medical and scientific staff. Likert scale items permit the expression of several (usually five) degrees of agreement-disagreement. The questionnaire was reviewed and approved by our institutional review board for human investigation. The questionnaire satisfied several goals: it assisted the 'Illsk Force in formulating the smoking policy, it served as an evaluation tool across time measuring changes that resulted from the policy, and it was selectively used for the investigative interests of the group. In this respect, three select areas of research interest and inquiry, which we report on in this article, involved how tobacco smoke affects employees or patients while at the institution, the acceptance of a no-smoking policy before and after its implementation, and the consequences of the policy on the smoker. The questionnaire was introduced with a statement of the purpose and with instructions for completion. The questionnaire was mailed to all employees and to more than 2,000 randomly selected patients. Respondents were assured that their answers would be kept confidential and were instructed to complete the questionnaire
within one week, returning it in a confidential envelope. Responses to the questionnaire, involving nominal and ordinal levels of data, were coded and the data were analyzed using survey statistical methods.• The results of this study will be particularly confined to responses of employees who experience the major impact of the institution policy.
REsuLTS In all, three surveys were obtained: the initial survey, six months before policy implementation; the second, six months after policy implementation; and the third, after the policy had been in effect for one year. In the prepolicy baseline study, of 4,200 distributed employee questionnaires, 1,946 (46 percent) were returned. In the second survey, 1,608 (38 percent) questionnaires were returned, but for the third study, one year after policy implementation, only 684 (16 percent) questionnaires were completed, perhaps possibly indicating general acceptance of the program. This latter lower response rate of the ,most recent survey was expected as a function of the policy being less controversial with employees feeling that they had voiced their opinion on two previous occasions.
Demography There was a significant reduction in the smoking habits of our employees one year after implementation of the no-smoking policy. Thus, six months before and after the policy was instituted, 22 and 20 percent, respectively, of our employees indicated that they smoked, but after one year this was reduced by 30 percent to 14 percent for all employees (x2 = 11.53; p<0.003). These data, howe-ver, failed to show an impact by gender of the employee (Thble 1). Most nonsmokers were under the age of 35 years (52 percent), and almost two thirds of smokers (63 percent) were 35 years and older; but of these, more older individuals stopped smoking. These data further indicated that women were more likely than men to be smokers (79 percent of smokers were women).
Table I- Demographic Profile of Employees' Smoking Status One Year After Policy
,.,
Exsmoker,
,.,
Total,
N*
,.,
Nonsmoker,
300 202 182
37 30 33
52 26 22
25 39 36
44 29 27
199 485
21 79
31 69
30 70
29
71
562 122
85
85
15
15
73 27
82 18
Smoker, Age, yr <35 35to44 c:!!45 Gender Male Female Position Employee Physician
,.,
Statistics
x•
df
p
34.46
4
0.0000
3.490
2
0.1746
11.91
2
0.0026
*N=684. CHEST I 97 I 5 I MAY, 1990
1199
Table 2-Effect ofNo-Smolcing Policy on Choice oflleolth Care Facility• Smokers(%) Nonsmokers(%) Exsmokers (%) More likely to seek institution's healthcare No effect on decision Less inclined to use Ochsner
2 (1.7)
51 (44.3)
19 (16.5)
8 (7.0)
9 (7.8)
6 (5.2)
20 (66.7)
*N = 115; x•=78.8; df=4; p=O.Ol.
Acceptance Nearly 80 percent of employees, a clear majority, favored the no-smoking policy; this acceptance persisted one year after its implementation. Within the last six months of the analysis, this favorable attitude increased nearly 10 percentage points between this study's last two surveys (p<0.001). It was of interest to learn, albeit with a relatively small subsample of 115 patients of whom 20 were smokers, that when asked how this policy would affect their future choice of a medical facility, 67 percent indicated that they would be less likely to use a "nosmoking" facility in the future (Table 2). In other words, the majority of patients who smoked indicated less inclination to use a health care facility that prohibited smoking. However, of those patients who
.--.. ~
0 ........
•atcv ....c ...•u • a.
(684) (1946)
77
(1608)
75
........... ················ ,•,•,•,•,•,•,•,•,•,•, .......... ........... ,•,•,•,•,•,•,•,•,•,•, ·:·:·:·:·:·:·:·:·:·:· ........... .......... ::::::::::::::::::::: ........... ·:·:·:·:·:·:·:·:·:·:· ........... ,•,•,•,•,•,•,•,•.•,•,
.......... ........... .......... .......... ............ ......... ........... ·········•··········· ........... ........... .......... .·•···•········•····· ........... .·.·.················ .......... .•.·.·.·.·.·····••··· ........... ·····•···•··········• ·:·:·:·:·:·:·:·:·:·:· ........... .......... ........... .......... ........... ::::::::::::::::::::: ,•,•,•,•,•,•,•,•,•,•,
,•,•,•,•,•,•,•,•,•,•,
'•'•'•'i'•'•'•'•'•'•'
.·.·•·········•······
Employees FIGURE
1200
Consequences In the survey conducted prior to implementation of the policy, both employees and patients were asked to what extent the smoke of others bothered them within the confines of the medical institution. The majority of employees indicated that the smoke of others bothered them. Of these individuals, 35 percent indicated that the smoke of coworkers affected them a great deal; 22 percent responded that the smoke of others did not bother them. Of those who indicated that the smoke of others had an adverse effect, 66 percent were smokers. Patient response rates were nearly identical to the employee response rates {Fig 1). In this regard, the most offensive effects were smell, eye irritation, and provocation of sinus problems, cough, and headache. One major consequence of interest was the possible change in the pattern of smoking while employees were at work. In both surveys, six and 12 months after implementation of the policy, approximately one fourth of smokers indicated that they no longer smoked cigarettes at work. However, approximately 40 percent of smokers indicated at both the six- and 12-month surveys that their cigarette consumption remained unchanged after leaving the medical facility at the end
I!] 6 Months Post-Policy
[]Pre-Policy
100· 90· 80· 70· 60· 50· 40· 30· 20· 10· 0
were not in favor of the policy, 20 percent were themselves smokers.
84
(607)
82
m1
Year Post-Policy
(397)
93
.......... ........... .·.··················
•,•,•,•,•,•,•,•,•,•,•,
·:·:·:·:·:·:·:·:·:·:· ........... ·:·:·:·:·:·:·:·:·:·:·
(600)
80
::::::::::::::::::::: .·.·········•········ ·::::::::::::::::::::: :·:·:·:·:·:·:·:·:·:·: ::::::::::::::::::::: ::::::::::::::::::::: :·:·:·:·:·:·:·:·:·:·: .·.···········•··••·· ::::::::::::::::::::: ·:·:·:·:·:·:·:·:·:·:· :·:·:·:·:·:·:·:·:·:·: •,•,•,•,•,•,•,•,•,•,• :·:·:·:·:·:·:·:·:·:·: ·:·:·:·:·:·:·:·:·:·:· ·:·:·:·:·:·:·:·:·:·:·: ·.·.·.·.·.·•·.·•·.·.·
...........
..........
......... .
.
,•,•,•,•.•,•,•,•,•,•, ,•,•,•,•,•,•,•,•,•,•,
::::::::::::::::::::: .·.········•···•····· ·:·:·:·:·:·:·:·:·:·:· •·•··••··•·••·•••·•·• •'•'•'•'•'•'•'•'•'•'•
........... ..........
Patients
1. Employee and patient attitude toward policy (percent agreeing with policy). No-Smoking Polley in a Medical Institution (Hudzinsld, Frohlich)
of a workday. In this regard, 23 percent of employees who smoked reported smoking less, but 35 percent smoked more after work hours since the policy went into effect. Those who indicated that they smoked more were in the 35- to 44-year-old age group, had smoked for ten or more years, and were women. Notwithstanding the foregoing frustrating responses, after one year there was a significant decrease in the number of employees who smoked (20 vs 14 percent; p<0.01). Moreover, the tobacco users indicated that they smoked fewer cigarettes after one year in comparison with the previous years data. Thus, 81 percent of smokers reported using less than eight cigarettes per day after one year. These data are comparable with the national finding that significant numbers of smokers had either quit using tobacco or reduced the number of cigarettes. 3 These national trends run parallel to the first survey taken one year earlier, when more than one fourth of smokers (28 percent) indicated that they intended to stop smoking ifour institution implemented a policy; and the most recent survey indicated that most who expressed that interest had attempted to do so. Indeed, 25 percent of employees reported that they physically tried to stop smoking at the time of the six-month postpolicy study, and 21 percent reported that they had physically tried to stop smoking one year after the policy was begun. Furthermore, at six and 12 months, approximately 60 percent of smokers indicated that they wished to stop smoking in the future. The majority of our employees reported the nosmoking policy had helped them by decreasing irritants while at work. Most employees indicated eliminating smoking stopped their experiencing burning eyes, sinus problems, cough, headaches, and the offensive smoking smell. In the final survey, 74 percent indicated that the policy had helped them; and of these, more than 50 percent indicated that the policy had helped a great deal. Thus, 7 percent of employee exsmokers indicated they had stopped smoking directly because of our institution's no-smoking policy; and 30 percent of employees believed that their work performance had improved as a result of this policy, whereas 60 percent believed it had no effect at all. DISCUSSION
The results of this report provide the most comprehensive analysis to date of a no-smoking work-site policy. They indicate that both employee and patients overwhelmingly favored the restriction of smoking at our health care institution. It could be argued that the lower response rate to the third administration of the questionnaire, 684 returned, does not support a conclusion that the policy was widely accepted. Wide acceptance is concluded based on overwhelmingly favorable ratings at both the first survey six months
before and the second survey six months after policy implementation. This conclusion is also based on no verbal or demonstrated opposition to this wide-reaching policy. As noted earlier, the lower response rate is particularly based on employees verbalizing they had completed the questionnaire twice before and had voiced their opinion. This study is concordant with smoking cessation trends in national surveys that indicate a significant number of smokers were in favor of, and wanted to stop smoking. Moreover, our results indicate that of those who had continued to smoke, the majority reduced their daily cigarette consumption. Our findings suggest that a no-smoking policy may stimulate if not actually provoke those who intended to stop smoking to cease their smoking. Although national data indicate that maintaining cessation from smoking may involve multiple attempts, a no-smoking policy may play a pivotal role in reducing the frequency of smoking, maintaining cessation, and improving both the health of smokers and nonsmokers alike. These results are also similar to the findings of the national smoking survey conducted by Gallup, 4 that found that 85 percent of nonsmokers, 78 percent of former smokers, and 62 percent of smokers agreed that smokers should not smoke near nonsmokers. This trend underscores the contention that, as a nation, we welcome a smoke-free environment. Industrial studies have assessed employee attitude in planning and implementing their own smoking policies. Pacific Telephone and Texas Instruments corporation are two of many that have undertaken employee surveys. They indicate on average that eight of ten nonsmoking employees were bothered by the smoke of others at work; and most smokers favor a smoke-free policy. 5 Although these studies surveyed employees only once (ie, preceding policy implementation), they provide support for efforts in establishing a smoke-free work environment. Assessment of employees has also been undertaken in health care settings. 6 These studies support establishing smoke restricting policies; and the only employee group that has not accepted the policy is the employee who smokes.7 In one study, Rigotti and associates& noted that although only 35 percent of smokers supported the ban at baseline, one year after implementation of a no-smoking policy this support had increased to 67 percent. Rosenstock" indicated similar trends when hospital no-smoking policies were instituted and that the smokers' initial reluctance diminished with time. Two recent studies, one conducted at Mayo Clinic and the other at the Oregon Health Sciences University, have reported on their methods for developiDg their smoke-free policies. 10•11 Six months after implementing its policy, each institution reported wide CHEST I 97 I 5 I MAY, 1990
1201
acceptance of the smoking ban by both staff and patients. Each concluded that a no-smoking policy CaQ be implemented if it is well planned and supported by staff. Our findings are in accord with this conclusion. As with all longitudinal surveys, a studys findings must be tempered by the possibility that uncontrolled factors may have inHuenced the results. In our study, repetitive questionnaires may have sensitized employees and patients in their responses to questions. This may have inH_'!lenced their tobacco use patterns or the honesty of answers irrespective of the policy's impact. Our institutions concurrent ongoing 2o-yeaN>Id, individual, group, nicotine replacement smoking cessation programs may have been another factor or interacting force that inftuenced employees' attitudes rather than the policy itself: Still another factor may be the national trend, rigorously supported by the former Surgeon General,·in stopping this, the major national caustl of preventable death and disability. No-smoking policies and their. growing prevalence in the workplace, and in health care facilities in particular, oiay have dramatic consequences. They increase general knowledge of health risks about smoking and may promote nonsmoking as a more normative behavior. Prohibition of smoking may help reduce the peer pressure to smoke and, conversely, increase social support for nonsmoking. No-smoking policies could, in fact, increase the success of a smoker's intent to quit and to decrease daily cigarette consumption. No-smoking policies are rapidly emerging. Since we addressed and implemented ours, all health care facilities in our community have implemented similar policies. Smokers consequently are perceiving little choice in finding health care providers who permit smoking within our community. This is becoming a growing and national trend. Smo~g policies may have other significant benefits beyond the issue itself: A smoke-free policy can directly impact on employers by decreasing the costs resulting from employees who smoke with respect to increased indirect costs from absenteeism, increased health care utilization, and higher health insurance rates. 12 Clearly, nonsmokers and exsmokers may be expected to favor smoking restrictions; it requires no change in behavior and offers the benefits of reduced health risks. Opinions of smokers are expected to be less favorable toward a restricted policy that, ifanything, is inconvenient. Smokers may argue that they have the least to gain from a no-smoking policy and they are valued employees who should not be confronted with a punitive policy. Smoking is an undesir-
1202
able habit and a major preventable cause of death and disability; and this institution continues to emphasize that message as often as possible. Likewise, our institution remains committed to its employees, to the smoker and nonsmoker, to job satisfaction, and continued quality performance although we unequivocally support non-smoking efforts. To this end, we provide an alternative for the smoker-an off-the-main-campus setting where those who smoke can meet the need of their dependency-while at the same time we emphas~e smoking cessation and smoking treatment programs, including group counseling and nicotine replacement therapy. In conclusion, we strongly advocate that medical institutions should take the lead as responsible proponents of sociomedical policy by setting an example for the public as it relates to health care issues. We have tested a no-smoking policy and found it to be desirable for our medical institution and for the nonsmoker and smoker alike. ACKNOWLEDGMENTS: The assistance of Mr. Charles Madanick, Director, DiviSion of Marketing/Public Affairs, and his staff, including Ms. Connie Baer and Ms. CoUene Terro for statistical help and advice, is gratefuUy acknowledged. REFERENCES
1 Ochsner AD, DeBakey M. Primary pulmonary malignancy: treatment by total pneumonectomy: analysis of 79 collected cases and presentation of7 personal cases. Surg Gynecol Obstet 1939; 68:435-51 2 Rosenberg M. The strategy of survey analysis: the logic of surver analysis. New York: Basic Books Inc; 1968; 75 3 Department of Health and Human Services. The health consequences of smoking, nicotine addiction. Report of the Surgeon General; Washington, DC. US Government Printing Office, 1988; 16 4 Gallup Organization. Survey of attitudes towards smoking: American Lung Association survey. Princeton, NJ: Gallup Or~n. July 1985 5 Pacific Telephone. Employee smoking study. January 1983 6 Rosenstock IM, Stergachis A, Heaney C. Evaluation of smoking prohibition policies in a health maintenance organization. Public Health Briefs 1986; 76:10-14 7 Koop CE: Smoking in the work place. Corporate Fitness and Recreation; June-July 1986; 35 8 Rigotti NA, Hill PB, Cleary P, Singer DE, Mulley AG. The impact of banning smoking on a hospital ward: acceptance, compliance, air quality and smoking behavior. Clin Res 1986;
34:833A 9 Rosenstock IM. Evaluation of smoking prohibition policy in a health maintenance organization. Public Health Briefs 1986; 76:1014-15 10 Hurt RD, Berge KG, Offord KP, Leonard DA, Gerlach DK, Renquist CL. The making of a smoke-free medical center. JAMA 1989; 261:91).7 11 Barker AF, Moseley JR, GlideweU BL. Components of a smokefree hospital program. Arch Intern Med 1989; 149:1357-59 12 Hallitt R. Smoking intervention in a work place: review and recommendations. Prev Med 1986; 15:213-17
No-Smoking Poley in a Macllc:allnstllullon (Hudzlnald, Frohlich)