healthscotland.com

Workplace Smoking Policies in Scotland

Contents:

Executive Summary Executive Summary 1. Background to study and method 1. Background to study and method 2. Views about smoking and the risks of passive smoking 2. Views about smoking and the risks of passive smoking 3. Attitudes to smoking and smoking policies at work 3. Attitudes to smoking and smoking policies at work 4. Smoking policies in the workplace 4. Smoking policies in the workplace 5. Implementation, communication and enforcement of workplace smoking policies 5. Implementation, communication and enforcement of workplace smoking policies 6. Views of current smoking policies 6. Views of current smoking policies 7. Cessation support 7. Cessation support 8. Views of legislation and policy initiatives 8. Views of legislation and policy initiatives Appendix A Appendix A Appendix B Appendix B Appendix C Appendix C Appendix D Appendix D

1. Background to study and method

In 2004, NHS Health Scotland commissioned BMRB Social Research to conduct a survey of workplace smoking policies in Scotland on behalf of the Scottish Executive. This research was commissioned in the context of an increased priority for action on smoking in public places as set out in the UK White Paper Smoking Kills (1998) and the Scottish Executive Tobacco Control Action Plan (2004). Emphasis had also been placed on health in the workplace by the Scottish Executive in Our National Health: a plan for action, a plan for change (2000). The Tobacco Control Action Plan which was published in January 2004 announced a public consultation on smoking in enclosed public places. This consultation programme was launched in June 2004, and this research project is part of a series of projects designed to support the programme.

The research also addresses issues raised by the Regulation of Smoking (Scotland) Bill proposed by Stewart Maxwell MSP, which aims to restrict smoking by employees and the general public in enclosed areas where food is served.

The findings from this research add to the findings of research commissioned in 2003 by HEBS and ASH, to explore the impact of the Voluntary Charter of May 2000. The Charter, launched by the Scottish Licensed Trade Association and the Scottish Tourist Forum, set targets for the licensed trade, tourism and hospitality sectors to extend the provision of smoke-free areas for the general public.

The objectives of this study were:

• to determine the patterns of provision of smoking policies in commercial workplaces in Scotland and to measure the extent of policy enforcement
• to determine what smoking cessation support is currently in place for employees and how this could be extended
• to identify the barriers and facilitators to establishing employee smoking policies
• to assess the potential impact of the introduction of smoking policies and the perceived need for legislation.

1.1 Objectives

In 2004, NHS Health Scotland commissioned BMRB Social Research to conduct a survey of workplace smoking policies in Scotland on behalf of the Scottish Executive. This research was commissioned in the context of an increased priority for action on smoking in public places as set out in the UK White Paper Smoking Kills (1998) and the Scottish Executive Tobacco Control Action Plan (2004). Emphasis had also been placed on health in the workplace by the Scottish Executive in Our National Health: a plan for action, a plan for change (2000). The Tobacco Control Action Plan which was published in January 2004 announced a public consultation on smoking in enclosed public places. This consultation programme was launched in June 2004, and this research project is part of a series of projects designed to support the programme.

The research also addresses issues raised by the Regulation of Smoking (Scotland) Bill proposed by Stewart Maxwell MSP, which aims to restrict smoking by employees and the general public in enclosed areas where food is served.

The findings from this research add to the findings of research commissioned in 2003 by HEBS and ASH, to explore the impact of the Voluntary Charter of May 2000. The Charter, launched by the Scottish Licensed Trade Association and the Scottish Tourist Forum, set targets for the licensed trade, tourism and hospitality sectors to extend the provision of smoke-free areas for the general public.

The objectives of this study were:

• to determine the patterns of provision of smoking policies in commercial workplaces in Scotland and to measure the extent of policy enforcement
• to determine what smoking cessation support is currently in place for employees and how this could be extended
• to identify the barriers and facilitators to establishing employee smoking policies
• to assess the potential impact of the introduction of smoking policies and the perceived need for legislation.

1.2 Overview of Method

The study was commissioned in three stages. First, there was a survey of workplaces in Small and Medium Enterprises or SMEs (organisations with under 250 employees). This was followed by a survey of workplaces in large organisations (250+ employees). The fieldwork for these two surveys was carried out at different times, but the methodology and the questionnaire used were the same. Following the two surveys, qualitative case studies were conducted in seventeen of the workplaces that had participated in the surveys.

1.2.1 The surveys

Sample selection

The surveys included commercial organisations with at least one employee. Sole traders were excluded as many of the issues were not directly relevant to this group. The sample was split into two groups, workplaces that were classified as belonging to SMEs (organisations with at least one, but less than 250 employees) and workplaces that were part of large organisations (with 250 or more employees).

The sampling unit for the surveys was workplace rather than organisation, and it is important to note, therefore, that a large number of workplaces were branches of larger organisations. This means that the size of the workplace is not the same as the size of the organisation; indeed, it is possible to find very small workplaces that belong to very large organisations.

Databases of businesses tend to vary in their coverage; therefore, to ensure the best coverage of all sizes of organisation, sample was purchased from two commercial databases: Dun and Bradstreet and Experian. Organisations from the public sector, private households and sole traders were removed from the sample before selection.

The sample for SMEs was selected at branch level, meaning that organisations with several branches could be selected more than once. The sample was selected to over-represent the larger SME organisations (with 50-250 employees) in order to ensure that enough businesses of this size were available for analysis. The final data were then weighted back to the population profile.

The sample for larger organisations was selected to include branches and head offices as well as large stand-alone organisations with over 250 employees. The sample for larger organisations was stratified alphabetically before selection to prevent too many branches of one large organisation from being selected, which would have over-represented one organisation in the data. Equal numbers of branches with head offices in Scotland and head offices elsewhere were selected. The final data were weighted back to the natural proportions of branches whose head offices were in Scotland, branches whose head offices were elsewhere in the UK, and head offices and single site, stand-alone organisations.

Data collection method

Each organisation in the final selected sample was sent an advance letter to explain the purpose of the study and to give them the opportunity to opt-out. Fifteen-minute telephone interviews were conducted using CATI (Computer Assisted Telephone Interviewing) technology. The interviewer asked to speak to the person responsible for smoking policies at the workplace, and this included those responsible either for deciding the policy or for implementing and enforcing the policy at the workplace. The focus of the study was the workplace contacted rather than the organisation, although, for SMEs, the workplace contacted was often the only one in the organisation.

Interviews with SME organisations took place between Monday 19 th January and Friday 12 th March 2004, and interviews with the larger organisations took place from Thursday 1 ^st April until Friday 30^th April 2004.

Response rates
1066 interviews were achieved with SME workplaces, and 538 interviews were achieved with workplaces in larger organisations. In total, 1604 interviews were achieved with Scottish organisations, with an estimated response rate of 57%. In calculating the response rate, we have taken into account the high rate of ineligible sample amongst SME workplaces (32%), due to the inclusion in the initial sample of both sole traders and organisations with too many employees. On this basis, the response rate among SMEs was estimated to be 59%. Since eligibility was much higher among the larger organisations, no adjustment was required, and a response rate of 54% was achieved for this group.

For further details of response rates and weighting details see Appendix B.

Sample profile
After weighting to return the sample to the natural proportions of workplaces, workplaces in SMEs made up 87% of the final sample, with 13% of workplaces in large organisations. Within the SME sample, 61% were stand-alone organisations. In contrast, most workplaces from large organisations (94%) were branches of a parent company.

As already noted, the size of workplace does not necessarily relate to its classification as belonging to an SME or large organisation. The table clearly illustrates that whilst the average workplace in a larger organisation has more employees, there are workplaces of similar sizes within both SMEs and large businesses.

Table 1.1: Number of employees at workplace (%)

Number of employees at workplace	All (1604)	Large (538)	SME (1066)
2 to 4 employees	42%	8%	47%
5 to 9 employees	25%	27%	25%
10 to 49 employees	25%	42%	22%
50 to 250 employees	7%	16%	6%
251+ employees	1%	7%	-
Base: All respondents

When interpreting the results of this survey, it is also important to bear in mind that a third of workplaces had no employees who smoked, and a further third had only a few employees who smoked. The majority of employees smoked in only one in five workplaces. This is likely to have some impact on smoking policies. For further details of sample profile see Appendix A.

SIC classification
In this report, some of the results have been broken down to show the different attitudes and behaviour of key sub-groups. One of the key analysis variables is Standard Industry Classification (SIC) or sector. The table below shows which types of business are included in each grouping.

Table 1.2: Percentage of businesses in each industry sector (%)

Category	% of weighted sample	Definition
Finance, real estate and other business	18%	Also includes renting, legal activities, advertising and research, cleaning services, security etc
Retail & wholesale	28%	Wholesale, retail and certain repairs
Social and personal services	6%	Services such as hairdressing, dry cleaning, funeral trade; waste and sewage; membership organisations etc
Leisure and hospitality	14%	Hotels, restaurants, tourism trade etc
Manufacturing	11%	All manufacturing trades
Agriculture, forestry etc	6%	Agriculture, hunting, forestry & fishing, mining & quarrying etc
Construction	9%	All aspects of construction
Transport & storage	5%	All transport and storage.
Base: All respondents (1604)

In general terms, those in the retail and wholesale, finance and real estate, and social and personal services sectors tended to be smaller workplaces, generally based indoors with a mix of manual and non-manual work. These made up around half of workplaces interviewed. The manufacturing and leisure and hospitality businesses tended to be medium to large workplaces, again based largely indoors, but with mostly manual work. These made up a quarter of the workplaces. The remaining sectors (agriculture and forestry, construction and transport) tended to be small to medium businesses, often outdoors and off-site, and with mostly manual work. These made up a fifth of the workplaces.

These correlations between sector and other business demographics mean that there will be a strong relationship between analysis by sector, and by other analysis variables such as whether work is indoors/outdoors, manual/non-manual, on-site/off-site, and size of business.

Reporting differences
Throughout the report, when differences in the data are discussed, these have been tested for significance using a two-tailed T-test on column proportions. Where observed differences between different sub-groups are significant, these are specifically noted in the text and reported at the 95% confidence level. This means that, if 20 samples were selected, the difference would appear in 19 out of 20 samples. If the differences observed are not significant, there will be no comment in the text. Finally, it should be noted that significance testing assumes a 100% response rate and normally distributed data and should, therefore, only be treated as indicative for this data set.

Questionnaire wording
Where total prohibition of smoking or a smoke-free workplace was explored in the questionnaire, the decision was taken to refer to this as a 'ban on smoking' since this was considered to be the term that respondents would be most familiar with and should, therefore, not result in any ambiguity. To ensure that the results of the survey are reported accurately, this report often refers to a 'ban' rather than using terms that may be in use elsewhere. This terminology was carried through to the qualitative research.

Table conventions
In tables, the following conventions have been used:

• percentages for single-response questions do not always add up to 100% exactly due to the effect of rounding
• a "*" symbol denotes a percentage of less than 0.5%
• a "-" symbol denotes zero.

1.2.2 The qualitative case studies

Seventeen case studies involving about three interviews were carried out with businesses that had taken part in the survey and agreed to be re-contacted. The case studies were chosen to reflect a range of characteristics including:

• number of employees
• number of sites
• how many employees smoked
• what kind of smoking policy was reported in the survey
• the type of business
• when the smoking policy was introduced.

In each case study, one interview was carried out with the person who had filled in the survey - this was usually the owner in small businesses and a manager or Human Resources (HR) contact in larger ones. In addition to this, two interviews were carried out with other staff in the business, wherever possible, with one smoker and one non-smoker. The qualitative study explored the situation in practice in companies and the rationale behind owners' or managers' decisions. In addition, the study was able to examine the views of staff about these decisions, and the impact they were felt to have. The full structure of the qualitative case studies, and the interviews carried out within them, is provided in Appendix A.

The interviews were carried out at the respondents' workplaces, but care was taken to ensure that the interviews were held individually with staff in private. Interviews were conducted by experienced qualitative researchers using a topic guide to ensure that key issues were discussed. A total of 55 interviews were carried out. The topic guide is included in Appendix D. The fieldwork was conducted in May and June 2004.