Health Education Research, Vol. 14, No. 4, 519-531,
August 1999
© 1999 Oxford University Press
Applying a contingency model of strategic decision making to the implementation of smoking bans: a case study
Department of Health Education and Promotion, and
1 Department of Health Organization Policy and Economics, Maastricht University, Maastricht, The Netherlands
Correspondence to: M. C. Willemsen, Dutch Foundation on Smoking and Health (Stivoro), PO Box 84370, 2508 AJ The Hague, The Netherlands
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
A model of strategic decision making was applied to study the implementation of worksite smoking policy. This model assumes there is no best way of implementing smoking policies, but that `the best way' depends on how decision making fits specific content and context factors. A case study at Wehkamp, a mail-order company, is presented to illustrate the usefulness of this model to understand how organizations implement smoking policies. Interview data were collected from representatives of Wehkamp, and pre- and post-ban survey data were collected from employees. After having failed to solve the smoking problem in a more democratic way, Wehkamp's top management choose a highly confrontational and decentralized decision-making approach to implement a complete smoking ban. This resulted in an effective smoking ban, but was to some extent at the cost of employees' satisfaction with the policy and with how the policy was implemented. The choice of implementation approach was contingent upon specific content and context factors, such as managers' perception of the problem, leadership style and legislation. More case studies from different types of companies are needed to better understand how organizational factors affect decision making about smoking bans and other health promotion innovations.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Recently, more attention has been paid to the settings in which health education is conducted (Mullen et al., 1995
; Poland et al., 1998). Settings are `social structures that provide channels and mechanisms of influence for reaching defined populations' (Mullen et al., 1995). Important settings are communities, schools, worksites and health care institutes. An important question is which implementation approaches are successful in particular settings. The present paper examines this question with regard to implementing smoking control activities in worksites.
Smoking restrictions in worksites are important because they limit or eliminate non-smokers' exposure to environmental tobacco smoke by separating smokers and non-smokers, and by reducing tobacco consumption during working hours (Petersen et al., 1988; Biener et al., 1989; Borland et al., 1990; Gottlieb et al., 1990; Daughton et al., 1992; Brigham et al., 1994; Jeffery et al., 1994). As a result of a smoking ban, on average, employees smoke four cigarettes per day less. Furthermore, some studies show an effect of smoking bans on worksite smoking prevalence [e.g. (Millar, 1988; Borland, 1991; Sorensen et al., 1991; Stave and Jackson, 1991; Woodruff et al., 1993)], but not all (Petersen et al., 1988; Biener et al., 1989; Borland et al., 1990; Gottlieb et al., 1990; Daughton et al., 1992; Jeffery et al., 1994). As part of a smoking policy, employers may offer their smoking employees methods to quit smoking.
Ready-to-use implementation protocols for smoking policies have existed for about 10 years (American Lung Association, 1985; National Heart Foundation of Australia, 1987; Action on Smoking and Health, 1988; USDHHS, 1996). Such implementation protocols roughly contain the following steps: orientation (assessing the current situation), deciding on the policy and developing a plan of implementation, informing employees about the policy changes, and announcing consolidation of the policy. They recommend obtaining support from top management, adjusting the anti-smoking policy to employee preferences by means of a survey, and setting up a working team to plan and implement the policy. Smoking policy options are not put to a vote, but rather a lot of effort is taken to ensure that employees can express their opinion about the policy and that there will be broad employee support. In terms of decision-making approaches, one could say that a rather decentralized and information-driven approach is recommended. If we look at the results from studies in the field of smoking control, we see that the effectiveness of this approach is equivocal. Several recent large-scale studies reported considerable variability in effect sizes resulting from smoking control and/or smoking cessation programmes across worksites, despite standardization of implementation strategies (Jeffery et al., 1993; Glasgow et al., 1995; Sorensen et al., 1996). This variability has also been noted in a meta-analysis of 20 worksite smoking cessation programmes (Fisher et al., 1990). Glasgow et al. (Glasgow et al., 1995) recommend that future research examines how worksite characteristics determine programme success. It has been suggested that such variability may partly originate from a good or bad `fit' between an organization and the programme (Fisher et al., 1990). This is also our own experience with Dutch worksites: factors such as company culture and personal characteristics of managers seem to determine to a large extent whether a programme will be successful or not (Willemsen, 1997).
Literature is scarce on how worksite characteristics affect implementation of worksite smoking control programmes (Fielding, 1991), especially literature that makes use of a theoretical model. Gottlieb et al. (Gottlieb et al., 1992) proposed a model based on organizational theory. Their `Innovation Implementation Model' assumes that the implementation process and the resulting outcomes are determined by the interaction of what they call `concept' (i.e. the innovation being implemented) and the (organizational) context within which the innovation is being introduced. They found support for their model in a case study of the implementation of a smoking policy in a large decentralized company, and suggested that more research should be carried out to examine the relationships between concept, context, implementation process and outcomes. Crump et al. (Crump et al., 1996) examined employee participation in worksite health promotion programmes. They found evidence that participation is the result of both the characteristics of the programme (including the way it is being implemented) and the organizational context within which the programme is introduced.
We propose a more elaborate model that adds one important intermediary factor not taken into account before, i.e. dimensions of decision making. We have also made a clearer conceptualization of what Gottlieb et al. (Gottlieb et al., 1992) call `concept factors' and Crump et al. (Crump et al., 1996) call `implementation process'. We propose a distinction between the content of the problem (e.g. how important, complex, new, politically sensitive, etc., is smoking at work for company management?) and the characteristics of the solution that is proposed (e.g. how restrictive must the smoking policy be?). In this paper, a model of strategic decision making is presented and applied to the implementation of smoking policies, but we believe that the model is applicable to other worksite health promotion programmes as well. To illustrate the usefulness of the model, a case study is presented. Data were collected both qualitatively (i.e. through interviews with those involved with the programme) and quantitatively (i.e. through a survey among the employees).
A model of strategic decision making
A specific implementation strategy that works in one organizational context might not work in another. In our opinion, a central role is played by the top managers. Confronted with a specific problem, managers must decide how they will deal with it: they have to decide on how decisions are going to be made. For each possible solution that arises, e.g. a smoking ban, management has to assess the level of acceptance in the organization, and it has to estimate how effective, efficient and satisfactory this specific solution will be. Based on these considerations, a specific decision-making approach is chosen. This approach will determine the smoothness and length of the implementation process and the resulting outcomes. Hence, the central issue is the decision-making approach of managers, and how this is determined and affected by how managers perceive the organizational context.
In terms of decision making, the `standard' approach, which is very much in line with the community organization theory, is a highly information-driven decision-making style: the results from an employee survey directly dictate which smoking policy option is feasible. Moreover, a fairly democratic decision-making style is normally recommended. Employee representatives are advised to take part in the decision-making process from the start. We are interested in discovering under which conditions this approach works well and in which organizational conditions other decision-making styles may be more appropriate.
The decision-making model that is used as a theoretical framework in the present study, was developed by Pool and Koopman [(Pool and Koopman, 1992); see also (Koopman and Pool, 1994)]. It is an integration of elements of previous models on strategic decision making [e.g. (Hickson et al., 1986; Heller, 1988)]. The model of Pool and Koopman is a research model, not a normative or prescriptive one, as most models in the management literature are, and is tested in various settings (health care, industry) in the Netherlands.
The model is founded in the so-called contingency approach in organization theory, which states that organizational performance is related to or `contingent upon' a proper fit between the decision-making style and the organization and its environment. The basic assumption is that there is not one most effective way of organizing. What works in one organization and what works with respect to one specific problem, might not work in another organization or with respect to another problem: it strongly depends on the circumstances. It is therefore crucial to study how these circumstantial factors affect the decision-making process and its outcomes.
Decisions are regarded as strategic if they are of substantial significance to the future of an organization, such as reorganizations, relocation and product-innovation decisions (Pool and Koopman, 1992). We assume that factors that affect strategic decision making also affect decision making with regard to topics that are of less importance to an organizations' survival, but that do require top management attention and decision making (such as a smoking ban).
Dimensions of decision making
In the literature on strategic decision making, various types of decision-making styles have been distinguished [e.g. (Grandori, 1984; Mintzberg et al., 1976)]. Pool and Koopman have reduced these to four central dimensions on which decision-making processes can be controlled (Pool and Koopman, 1992, 1994). The first is centrality: the extent to which top management involves lower levels in the decision-making process. A second dimension is formalization: the extent to which the decision-making process is formalized (following standard procedures) or more informal and ad hoc. The third dimension is information: the extent to which decisions are based on the collecting of information and a consideration of pros and cons based on this information. The fourth dimension is confrontation: the extent to which decisions are the result of a political process in which a manager has to confront other parties that have opposing interests.
The freedom of managers to decide how they are going to handle a specific problem (i.e. which dimension of decision making) is restricted by both content and context factors. For example, if a problem is a key issue for the company (a content factor), chances are that managers will opt for a highly centralized approach. In a highly bureaucratic organization (a context factor), a formal approach is most appropriate. Once a manager has decided on a specific approach, the end result will again depend on content and context factors: it may be necessary to adjust the decision-making style during the process or it may fail, because unforseen factors do not fit with the chosen style.
Figure 1 shows the complete model. Content and context factors are independent variables, dimensions of decision making and the decision-making/implementation process are intermediary variables, and the outcomes are the dependent variables. The various factors will be described below.
|
Content factors
An important content factor is the content of the subject matter. Pool and Koopman (Pool and Koopman, 1992
) distinguish the extent to which managers perceive a problem to be simple or complex, old (familiar) or new (unfamiliar) and important or unimportant. For example, complex and unfamiliar problems often require more information-based decision processes, whereas problems that are very important to a company's survival may be more centrally decided upon. Another factor is whether a subject matter is controversial (involving specific conflicts of interest within the company). The second content factor is the proposed solution to the problem. (This factor was not mentioned in Pool and Koopman's original model, but is added by the present authors.) Gottlieb et al. (Gottlieb et al., 1992) and Crump et al. (Crump et al., 1996) call this characteristics of the (proposed) intervention. In the case of smoking control, an important characteristic would be the restrictiveness of the policy, and whether the smoking policy will be accompanied by a smoking cessation programme and by health education activities.
Context factors
Managers' decision making may be affected at three levels: the level of the individual decision maker or decision-making body, the organizational context and the environment of the organization (Pool and Koopman, 1992). At the individual level, the important factors are the personality traits of managers, specific beliefs with respect to the topic under discussion, social factors such as how other managers think about the issue at hand and the proposed solutions (social norms), and managers' abilities to solve the problem, e.g. their negotiation skills. At the organizational level, the important factors are the `rules of the game', i.e. organizational structure and organizational culture prescribe to a large extent the way problems are dealt with (Trice and Beyer, 1993). Important factors in an organization's environment are competing companies, governmental regulations, economic factors, customs and societal norms.
The decision-making process
The decision-making process itself follows a specific course, which may be short or lengthy, orderly or chaotic. The time period between problem identification and effectuation of a solution (e.g. a smoking ban) may take several months to several years and involves a small or large number of distinct phases, that may include delays and feedback loops (backslides).
Outcomes
The outcomes of the decision-making process can be measured in terms of its effectiveness (e.g. reduction of environmental tobacco smoke), efficiency (i.e. cost-effectiveness and cost-benefits) and satisfaction: how satisfied are employees with the new ban and with the way the ban came into effect (e.g. with respect to employee participation).
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Setting
The study was conducted at Wehkamp, which is an especially interesting case because Wehkamp was the first large Dutch worksite to implement a complete smoking ban. Wehkamp is the largest of three companies that make up GUS Holland Group. Wehkamp is a well-known mail-order company in the Netherlands, employing 1053 employees and consisting of a head office (575 employees) and two warehouses where goods are stored and processed. This paper describes the implementation of a smoking policy in the head office.
Data collection
Data were obtained through individual interviews with company representatives who were involved with implementing the smoking ban, and cross-sectional pre- and post-ban surveys among employees at Wehkamp's head office. The post-ban survey was carried out about 1 year after the smoking ban was implemented.
Interviews
The chronological sequence of the decision-making process and all distinct factors in the model were systematically assessed by means of semi-structured interviews with all people who had been involved in the decision-making process. Respondents were the director of Wehkamp, a representative from the workers' council (workers' councils in the Netherlands are the main bodies for employee representation), the manager/head of the PR department and the media production manager. The average interviewing time was 60 min. In addition, relevant minutes of meetings, memos, reports and newspaper clippings were collected. All raw data were brought together and categorized by two independent researchers, according to the factors in the model. Parts of the transcribed interviews were categorized differently by two researchers. These were discussed until consensus was reached. Two separate case descriptions were written based on the transcriptions. These were compared and differences were again discussed. To optimize the reliability of the final descriptions, they were then checked by one of the interviewees.
Surveys
Before the decision-making process started and after the ban was completed, representative surveys were conducted among the employees at Wehkamp. The pre-test survey was distributed to all 575 employees. The response was 78% (N = 450). Relevant questions referred to hindrance from environmental tobacco smoke (ETS) as measured with the question `Do you ever experience hindrance from ETS at your worksite?' (yes/no), age, sex and whether the employee smoked during working-hours.
The post-ban questionnaire was distributed to a representative sample of 44% of all employees (N = 251). The response was 53% (N = 133). The objective of this survey was to assess the programme's effectiveness (i.e. exposure to ETS and reduction of smoking during working hours) and employee satisfaction with the new policy. Exposure to ETS was measured with the question `How much tobacco smoke, on average, is there during a day in your office?' (0 = no smoke, 6 = a lot of smoke). This self-reported measure is a valid measure of actual exposure to ETS in work offices (Willemsen et al., 1997). Satisfaction was measured with the question `How satisfied are you with the new smoking policy?' (–2 = very dissatisfied to +2 = very satisfied) and `How satisfied are you with how the new policy came about?' (–2 = very dissatisfied to +2 = very satisfied).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The majority of employees were female (58%) and a large proportion (56%) were in the 35–50 age category. Proportions of male workers and age distributions did not differ between pre- and post-surveys, and were comparable to data from the personnel department of Wehkamp. At both surveys, about 29% of employees smoked.
The qualitative results from the interviews are presented below, according to the variables in the model. Figure 2 summarizes the main results.
|
Content factors
Problems with ETS
Before the ban, hindrance from ETS was most serious in a number of very large offices where tobacco smoke was re-introduced via the air-conditioning system. All kinds of solutions had been tried, but none proved satisfactory for all workers concerned. Smoking at work became a concern for top management, when they realized that this problem could not be solved easily. They called for a drastic `once and for all' solution.
Conflict of interests
Management knew that the problem was a very sensitive one, with smokers and non-smokers at loggerheads. Because of this conflict of interests and the sensitivity of the issue, top management said that this was a problem that could not be solved democratically. The director decided that a central decision-making process was required. Another content factor was that Wehkamp had never implemented a policy that was so controversial before, meaning that there was no standard procedure they could follow. This ruled out the possibility of a highly formalized decision-making strategy.
Underestimated conflict potential
Part of the programme was an intensive campaign to make the employee attitude towards the ban more positive. Popular activities were organized, such as a photo-contest. Moreover, hard-hitting anti-smoking posters were distributed throughout the building. Management expected much conflict and opposition from smoking employees, but they did not expect that they would try to organize themselves, which is what happened at one point. Some smokers started to organize a petition to make a stand against the upcoming ban. Management reacted by removing posters and changing the campaign into a low-profile one.
Proposed intervention: complete ban
Top management specified the objective, i.e. to achieve a complete smoking ban within 2 years (for legal and technical reasons; see context variables). They realized that this would mean a confrontation with smoking employees and with employee representatives (workers' council). Hence, they realized that decision making would be highly confrontational:
The decision was in no way a compromise, because a compromise is what people want the most. We cannot have that. You cannot make compromises with a topic such as this. [Wehkamp's CEO]
Context variables
Various contextual factors made it possible for top management to choose to implement a complete smoking ban.
Legal issues
Dutch Labor Conditions Act obliges employers to provide their employees with a healthy working environment and can be interpreted to include protection from ETS as well. An important issue for the director was that he was afraid non-smoking employees might hold him legally responsible for any health damages due to passive smoking. He knew that already in the US and UK, employers had been held responsible for smoking-related diseases of their workers. This was the main argument the CEO used when he convinced the board of directors that Wehkamp should become completely smoke free:
...the decision to ban smoking was accelerated by changes in legislation. In 10 years from now, an employee can say to his employer that he hesitated too long with making his workplace smoke-free. As a result, he may sue his boss. I see this happening already. [Wehkamp's CEO].
Technical restrictions
A second reason why top management chose a complete smoking ban without exceptions was a technical reason: there was not enough space in the buildings to create well-ventilated separate smoking areas. Management exploited this point to their advantage.
Lack of space played a confrontational role, because the organization took a firm stand by saying they did not have enough room that could be designated to smokers. As a result, the standpoints were not flexible. [Representative form workers' council]
Strong leadership
The director of this company was a very strong leader, with a clear vision. He first convinced the other members of the board of directors to back him up. These were all non-smokers, which made it easier to come to an agreement at top-level to go ahead with the ban. Then he confronted the workers' council and succeeded in getting them to accept a smoking ban, after having made one concession (i.e. a sheltered smoking area outside the main office building).
Looking back, I can see that we quite deliberately made our own choices. We did not want to talk about special facilities for smokers in the company. We even went against the advice from the V.v.I.K. (i.e. the external advisors: authors). [Wehkamp's CEO]
Company culture
As part of the campaign to prepare employees for the upcoming ban, all kinds of playful activities were organized. These were not very well received at the head office, but the same activities were well received at other sites. Interviewees explained this difference was due to variations in organizational culture and type of employees. The head office employed mostly white-collar workers who did not appreciate playful activities.
Negative media attention
One smoking employee contacted a national newspaper which wrote an article about the upcoming ban, saying that there was some opposition from the workers. As a result, Wehkamp was confronted with sudden media attention.
All of a sudden, a big article appeared in De Telegraaf. It was a terribly agitating piece. The funny thing was that from that moment on, the smoking policy became a real issue in the eyes of the workers. [Wehkamp's head of PR]
Wehkamp reacted by ordering a media-stop for their employees. This worked very well, because after several weeks the media attention stopped.
External pressure groups
The Foundation for the Right of Smokers, a group supported by the tobacco industry, started a lobbying campaign against Wehkamp. They advised smoking employees to try to take legal steps to prevent the smoking ban. The tobacco industry was probably very concerned that if this company became smoke-free, other worksites would follow. In a response to this, Wehkamp's director made his advisers find out what his position would be if a smoking employee took legal steps. It become clear that as an employer he had the Labor Conditions Act on his side. Furthermore, the workers' counsel had approved the decision to go smoke-free and there would still be the possibility for hard-core smokers to smoke outside the buildings. He also made it clear that if smokers did not accept the new ban, he was prepared to take the ultimate step, which would be to fire them.
Dimensions of decision making
Content and context factors determined the latitude of decision-making styles from which the management could choose. The smoking problem could not be dealt with in a highly formalized way. A democratic solution was also not chosen in light of management perception that smoking and non-smoking employees had not been able to reach a satisfactory solution. An approach firmly based on information contradicted the fact that the final solution had already been decided upon (i.e. a complete ban). After a period in which a more democratic and information-based approach had been tried (see below), a highly confrontational and centralized decision-making process was started, in which the director himself invested much time. He became head of the committee that planned and implemented the programme. He persuaded the board of directors that a complete ban was the best solution.
The decision-making process
The decision-making process took a rather lengthy course. It took almost 3 years (34 months) from the moment top management started to realize they had to do something about the smoking problem to the moment Wehkamp's head office became smoke-free. The decision-making process went through the following three distinct phases.
(1) The orientation phase (December 1993–May 1995: 17 months) started when top management realized that smoking at work was a problem that needed to be dealt with. Management and workers' representatives discussed the issue regularly. Management sought advice from external organizations (e.g. the Dutch Foundation for Smoking and Health and the Dutch Collaborating Cancer Centers) about which smoking policy would be best and how this could be implemented. So, in this stage, top management chose a fairly democratic and information-based decision style. This led to an employee survey to assess hindrance from ETS and employee support for smoking bans at Wehkamp's head office. This showed that two-thirds (68%) of non-smokers reported experiencing hindrance from ETS at their workplace. However, only 17% of all employees supported a complete smoking ban. Furthermore, 77% of non-smokers, but only 8% of smokers, reported they would have `no problem' when a complete ban on smoking in the building would be proclaimed. The external organizations advised against implementing a complete smoking ban without designated smoking areas, because they expected that this would result in too much resistance from smoking employees. Top management ignored this advice and chose to implement a complete smoking ban instead. An important reason for this was that they felt that smokers and non-smokers had conflicting interests, and that democratic procedures would not result in a policy in which unwanted ETS exposure would be eliminated.
In this period, one of the external organizations prepared a protocol for the implementation of a smoking cessation programme and accompanying mass-media type activities. Also, a working group was chosen that could direct the implementation.
(2) A negotiation phase (May 1995–May 1996: 12 months). During this second phase, management discussed the plan to go smoke-free with the workers' counsel. Formally, regulations oblige the employer to get approval from the workers' council for drastic policy changes. The workers' council decided against the complete ban, because they wanted designated smoking areas. After a new period of negotiation, a compromise was reached in which there would be a shelter outside the building where hard-core smokers could smoke during lunch time. In total there was 4 months delay, because top management wanted to have the workers' approval by January 1996, but this was delayed until 1 May of that year.
(3) An implementation phase (May 1996–October 1996: 5 months) in which the smoking cessation programme (i.e. primarily smoking cessation classes) and an awareness campaign were executed, resulting in a complete smoking ban per 1 October 1996. During this phase, there was much commotion regarding the upcoming ban. The activities that were organized to try to make employees' attitudes towards the upcoming ban more positive had a negative effect on the smoking employees. Many protested when they realized that there would be a complete smoking ban and external pressure groups made a public media event out of it. However, management succeeded in going through with their intentions without deviating from the set time schedule. They did this by keeping to their confrontational decision-making style. Because they were prepared for some tough opposition, they were able to respond quickly and effectively to this.
To summarize: although the implementation of the intervention had no delays, the whole decision-making process, especially the orientation phase, took a long time. Two reasons for this were detected. First, there was a rather lengthy orientation period, in which recommendations from external organizations were not taken up by top management. In this period, a democratic and information-based decision-making approach changed into a more centralized and confrontational style. Second, there was one feedback loop during a period in which top management had to deal with opposition from the workers' council, causing unforeseen delay.
Outcomes
Effectiveness
Smoking prevalence during work hours was reduced from 28% (125 of 450) before the smoking ban to 23% (31 of 133) after the ban (
2 = 0.83; d.f. = 1; NS). One year after the ban, 121 (91%) of non-smokers reported that there was no tobacco smoke in their office. One employee reported very little tobacco smoke, seven reported little tobacco smoke and two reported `not much, not little' ETS. There were two missing cases with respect to this question. Before the ban, only 32% of non-smokers were not exposed to tobacco at their workplace.
Efficiency
It took Wehkamp almost 3 years to get a smoking ban, which is not very fast. Moreover, some of the working-group members spent a lot of time on it. Thus, the efficiency was probably moderate, as indicated by `+/–' in Figure 2
.
Satisfaction
Post-ban survey data revealed that about 1 year after the smoking ban was implemented, smokers were less satisfied with the policy than the non-smokers (t = 7.62; d.f. = 129; P < 0.0001). One-third of smokers was dissatisfied (Figure 3). More interestingly, smokers were also less satisfied with the process how the policy had come about (t = 6.79; d.f. = 134; P < 0.0001). See Figure 4.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Complete smoking bans for worksites are still rare in the Netherlands. One reason is that employers expect that a smoking ban has insufficient support from employees. At Wehkamp this was also the case. Only 17% of employees supported a complete smoking ban. A recent study among eight large Dutch companies showed that only 8% of non-smoking employees and 2% of smoking employees supported a complete smoking ban (Willemsen et al., 1996
). Specific conditions that were present in Wehkamp made it possible for top management to get the ban through without having to make substantial compromises. The most important conditions were a strong leadership, and that the management perceived the problem as being important and as impossible to solve in a more democratic way. However, in a rather lengthy orientation period, a more democratic and information-based approach was first tried out. Due to the unsatisfactory experiences with this, the decision-making style then changed. Specific context and content factors, such as lack of room for designated areas and fear for litigation, made it more `logical' for top management to choose a more centrally controlled and confrontational style. This style worked out quite well, as far as effectiveness was concerned, since exposure to ETS was almost eliminated. However, it seems that this result was at the cost of the satisfaction of a substantial number of smokers and at the cost of smooth implementation. There was some delay and falling back to previous phases, and the company came under severe attack from smoking employees and outside pressure groups. They survived this period, basically by keeping to their centralized decision-making style and by making the confrontational approach less visible. Top management reacted to the unexpected negative media attention by ordering a media-stop, thus strengthening their top-down and centralistic approach. Thus, a more democratic approach had been tried out, failed and was replaced by a more authoritarian approach. In the US and Australia, complete smoking bans have replaced separate smoking areas as the standard recommendation. This is not yet the case in most European countries. Wehkamp illustrates this transition from an area in which the restrictiveness of a smoking policy is still open to discussion to a time when there is enough impetus for managers to enforce a complete ban.
The central intermediary factor in our model was `dimensions of decision making', which was not included in previous models (Gottlieb et al., 1992; Crump et al., 1996). The present study confirmed the importance of this factor, by showing that the observed stages in the lengthy implementation process could be understood by changes in decision-making dimensions (from decentralized and information-based to centralized and confrontational). Dimensions of decision making describe how top managers go about making a specific intervention work in their organization. What is important is top management's perception of the intervention and how they are going to dealt with it, as well as their own personal characteristics. This study illustrates that there is no a priori `best way' of organizing a smoking policy, but that several approaches may be effective, depending on such organizational circumstances. The type of decision-making approach that works in a specific organization is contingent upon context and content variables. If these variables change over time, the decisional style may also change. The challenge for health educators is to determine what type of intervention is most feasible, given specific conceptual and contextual variables. To this end, it is recommended that more studies such as this should be carried out, preferably in different types of companies, of different sizes and so on. This will give us a rich set of examples of real-life `fit' between smoking policy and organizational factors. The next step would be to validate these findings by conducting larger scale quantitative studies. This would enable us to write possible scenarios for different types of companies.
At Wehkamp, a rather authoritative implementation style was chosen, which seemed rather effective. This observation may seem to conflict with the stress on participation and empowerment in current health promotion. However, participation is most important when a programme's success depends on winning people's hearts. For example, when employee participation in a life style modification programme (nutrition, fitness, alcohol, smoking cessation, etc.) is all important to a programme's success. This is less crucial in the case of legislative programmes, such as a smoking ban. It would therefore be very interesting to do this type of study with a programme in which behavioral modification is more crucial, e.g. a smoking cessation programme or a life-style health promotion programme.
Practitioners who want to implement health education programs in worksites could benefit from the model presented here. Most importantly, the contingency model can make practitioners more sensitive to the variation in decision-making styles they may encounter in organizations. Implementation protocols should reflect this variation. Clearly, more research is needed to better understand how exactly content and context variables influence decision making with respect to implementation of health promotion programs.
This study has generated several hypotheses that could be further examined by means of quantitative research. One hypothesis is that to realize a complete smoking ban, a confrontational and centralized decision-making style may be required, all other factors being equal. Another hypothesis would be that for a confrontational/centralized approach to result in a smoking ban, several conditions must be present: a strong leadership, management perception of the problem as impossible to solve in a more democratic way, a shortage of space to make designated smoking areas possible and managers being afraid that non-smokers will start legal action if their workplace is not adequately protected against ETS.
|
Acknowledgments |
|---|
This research was made possible by grants from the Dutch Foundation on Smoking and Health. We thank Nell Gottlieb for her comments on the paper.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
American Lung Association (1985) Creating your Company Policy. ALA, New York.
Action on Smoking and Health (1988) How to Achieve a Smoking Policy at Work. Action on Smoking and Health, London.
Biener, L., Abrams, D. B., Follick, M. J. and Dean, L. (1989) A comparative evaluation of a restrictive smoking policy in a general hospital. American Journal of Public Health, 79, 192–195.
Borland, R., Chapman, S., Owen, N. and Hill, D. (1990) Effects of workplace smoking bans on cigarette consumption. American Journal of Public Health, 80, 178–180.
Borland, R., Owen, N., Hill, D. and Schofield, P. (1991) Predicting attempts and sustained cessation of smoking after the introduction of workplace smoking bans. Health Psychology, 10, 336–342.[ISI][Medline]
Borland, R., Owen, N. and Hocking, B. (1991) Changes in smoking behavior after a total workplace smoking ban. Australian Journal of Public Health, 15, 130–134.[ISI][Medline]
Brigham, J., Gross, J., Stitzer, M. L. and Felch, L. J. (1994) Effects of a restricted work-site smoking policy on employees who smoke. American Journal of Public Health, 84, 773–738.
Borland, R., Owen, N. and Hocking, B. (1991) Changes in smoking behavior after a total workplace smoking ban. Australian Journal of Public Health, 15, 130–134.
Crump, C. E., Earp, J. A. L., Kozma, C. M. and Hertz-Picciotto, I. (1996) Effect of organizational-level variables on differential employee participation in 10 federal worksite health promotion programmes. Health Education Quarterly, 23, 204–223.[ISI][Medline]
Daughton, D. M., Andrews, C. E., Orona, C. P., Kashinath, D. and Rennard, S. I. (1992) Total indoor smoking ban and smoking behavior. Preventive Medicine, 21, 670–676.[ISI][Medline]
Fisher, K. J., Glasgow, R. E. and Terborg, J. R. (1990) Work-site smoking cessation: a meta-analysis of long-term quit rates from controlled studies. Journal of Occupational Medicine, 32, 429–439.[ISI][Medline]
Fielding, J. E. (1991) Smoking control at the workplace. Annual Review Public Health, 12, 209–234.
Glasgow, R. E., Terborg, J. R., Hollis, J. F., Severson, H. H. and Boles, S. M. (1995) Take Heart: results from the initial phase of a worksite wellness programme. American Journal of Public Health, 85, 209–216.
Gottlieb, N. H., Eriksen, M. P., Lovato, C. Y., Weinstein, R. P. and Green, L. W. (1990) Impact of a restrictive worksite smoking policy on smoking behavior, attitudes and norms. Journal of Occupational Medicine, 32, 16–23.[ISI][Medline]
Gottlieb, N. H., Lovato, C. Y., Weinstein, R., Green, L. W. and Eriksen, P. (1992) The implementation of a restrictive worksite smoking policy in a large decentralized organization. Health Education Quarterly, 19, 77–100.[ISI][Medline]
Grandori, A. (1984) A prescriptive contingency view of organizational decision making. Administrative Science Quarterly, 29, 192–209, 1984.
Green, L. W. and Kreuter, M. W. (1991) Health Promotion Planning: An Educational and Environmental Approach. Mayfield, Mountain View, CA.
Heller, F. A., Drenth, P. J. D., Koopman, P. L. and Rus, V. (1988) Decisions in Organizations: A Three Country Comparative Study. London, Sage.
Hickson, D. R., Butler, R. J., Cray, D., Mallory, G. R. and Wilson, C. D. (1986) Top Decisions: Strategic Decision-making in Organizations. Basil Blackwell, Oxford.
Jeffery, R. W., Forster, J. L., French, S. A., Kelder, S. H., Lando, H. A., McGovern, P. G., Jacobs, D. R. and Baxter, J. E. (1993) The Healthy Worker Project: a worksite intervention for weight control and smoking cessation. American Journal of Public Health, 83, 395–401.
Jeffery, R. W., Kelder, S. H., Forster, J. L., French, A. S., Lando, H. A. and Baxter, J. E. (1994) Restrictive smoking policies in the workplace: effects on smoking prevalence and cigarette consumption. Preventive Medicine, 23, 78–82.[ISI][Medline]
Koopman, P. L. and Pool, J. (1994) Decision making in organizations. In Cooper, C. L. and Robertson, I. T. (eds), Key Reviews in Managerial Psychology: Concepts and Research for Practice. Wiley, Chichester.
Millar, W. J. (1988) Evaluation of the impact of smoking restrictions in a government work setting. Canadian Journal of Public Health, 79, 379–82.[ISI][Medline]
Mintzberg, H. D., Raisinghani, D. and Theoret, A. (1976) The structure of `unstructured' decision processes. Administrative Science Quarterly, 21, 246–275.[ISI]
Mullen, P. D., Evans, D., Gottlieb, N. H., Kreuter, M., Moon, R., O'Rourke, T. and Strecher V. J. (1995) Settings as an important dimension in health education/promotion policy, programmes, and research. Health Education Quarterly, 22, 329–345.[ISI][Medline]
National Heart Foundation of Australia (1987) Going Smoke-free: A Guide for Workplaces. National Heart Foundation of Australia, Newcastle.
Petersen, L. R., Helgerson, S. D., Gibbons, C. M., Calhoud, C. R., Ciacco, K. H. and Pitchford, K. C. (1988) Employee smoking behavior changes and attitudes following a restrictive policy on worksite smoking in a large company. Public Health Reports, 103, 115–122.[ISI][Medline]
Poland, B., Green, L. W. and Rootman, I. (eds) (1998) Settings for Health Promotion: Linking Theory and Practice. Sage, London.
Pool, J. and Koopman, P. L. (1992) Strategic decision making on organizations: a research model and some initial findings. In Hosking, D. M. and Anderson, N. (eds), Organizational Change and Innovation: Psychological Perspectives and Practices in Europe. Routledge, London.
Sorensen, G., Rigotti, N., Rosen, A. and Prible, R. (1991) Effects of a worksite nonsmoking policy: evidence for increased cessation. American Journal of Public Health, 81, 202–204.
Sorensen, G., Thompson, B., Glanz, K., Feng, Z., Kinne, S., DiClemente, C., Emmons, K., Heimendinger, J., Probart, C. and Lichtenstein, E. (1996) Work site-based cancer prevention: primary results from the Working Well Trial. American Journal of Public Health, 86, 939–947.
Stave, G. M. and Jackson, G. W. (1991) Effect of a total work-site smoking ban on employee smoking and attitudes. Journal of Occupational Medicine, 33, 884–890.[ISI][Medline]
Trice, H. M. and Beyer, J. M. (1993) The Cultures of Work Organizations. Prentice-Hall, Englewood Cliffs, NJ.
USDHHS (1996) Making your Workplace Smokefree: A Decision Maker's Guide. US Department of Health and Human Services, Office on Smoking and Health.
Willemsen, M. C. (1997) Kicking the habit: the effectiveness of smoking cessation programmes in Dutch worksites. PhD thesis, Maastricht, Maastricht University.
Willemsen, M. C., Brug, J., Uges, R. A. and Vos de Wael, M. L. (1997) Validity and reliability of self-reported exposure to environmental tobacco smoke in work offices. Journal of Occupational and Environmental Medicine, 39, 1111–1114.
Willemsen, M. C., De Vries, H. and Genders, R. (1996) Annoyance from environmental tobacco smoke and support for no-smoking policies at eight large Dutch workplaces. Tobacco Control, 5, 132–138.[Abstract]
Woodruff, T. J., Rosbrook, B., Pierce, J. and Glantz, S. A. (1993) Lower levels of cigarette consumption found in smoke-free workplaces in California. Archives of Internal Medicine, 28, 1485–1493.
Received on February 23, 1998; accepted on July 31, 1998
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. M. Rogers and J. C. Peterson Diffusion of Clean Indoor Air Ordinances in the Southwestern United States Health Educ Behav, October 1, 2008; 35(5): 683 - 697. [Abstract] [PDF]
|
|
|
|
 |
|
 D. Segaar, C Bolman, M. Willemsen, and H De Vries Identifying determinants of protocol adoption by midwives: a comprehensive approach Health Educ. Res., February 1, 2007; 22(1): 14 - 26. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||