VOL: 102, ISSUE: 19, PAGE NO: 32
Paul Slater, BSc, MSc, is research associate, joint post, Royal Group of Hospitals, Belfast, Institute of Nursing Research, University of Ulster, Northern Ireland
Gerard McElwee, BSc, PGCE, DASE, PDHP, MSc, is head of cancer prevention, Ulster Cancer Foundation, Belfast; Paul Fleming, BEd, MSc, PDHE, FRIPH, MIHPE, ILTM, is assistant dean, School of Life and Health Sciences; Hugh McKenna, RGN, RMN, DipN, BSc, AdvDipEd, RNT, PhD, FFN RCSI, FEANS, FRCN, is dean; both are at the School of Life and Health Sciences, University of Ulster, Northern Ireland
Smoking tobacco is the main cause of preventable illness and premature death in the UK (Department of Health, 1998). In 1990 smoking was the underlying cause of 20% of deaths and one third of all cancer cases (Peto et al, 1994), while in 1997 it caused an estimated 117,400 deaths in the UK (Royal College of Physicians, 2000). Nurses’ smoking behaviour is important in relation to their role in smoking cessation. Several studies have compared the prevalence of smoking among nurses with that in the general population (Spencer, 1984; Swenson and Dalton, 1983). They show a decrease in prevalence over time for both groups. More recently, Elkind (1988) showed that nurses’ smoking behaviour was representative of women of the same age and from the same socioeconomic group in the general population.
In the 1980s, several studies suggested that smoking rates were influenced by the nurse role. Becker et al (1986) and Feldman and Richard (1986) showed that senior nurses smoke less than staff nurses, students and nursing auxiliaries. In the US Tagliocozzo and Vaughan (1982) found the lowest prevalence among paediatric nurses (12.2%) and the highest among psychiatric nurses (28.6%). These studies also report a higher prevalence in the hospital setting compared with the community. Smoking cessation treatments are among the most cost-effective of all healthcare interventions (Parrott et al, 1998). Improved cessation rates depend not only on the individual’s motivation but also on the availability of health professionals to help the smoker choose the appropriate level of support (Fagerstrom, 2002).
Smoking Kills - A White Paper on Tobacco (DH, 1998) identified a number of health professionals, including practice nurses, midwives and health visitors who have ‘an important role to play in giving the kind of smoking cessation advice which a modern health service ought to provide’. Rigotti et al (2002) analysed the literature on smoking cessation interventions for hospitalised patients. They found that intensive intervention with follow-up support after discharge was associated with a significantly higher rate of cessation than that achieved in a control group.
Rice and Stead (2002) conducted a meta-analysis of nurses’ effectiveness in smoking cessation. They found a significant increase in the liklihood of people quitting with nursing intervention. Hollis (1993) found a quit rate of 2% in those advised by a physician, which increased to 4% when a nurse delivered additional interventions. Given the high prevalence of smoking, the potential scope of nurse-led programmes and the benefits of cessation, the impact on the population’s health status could be substantial. Rice and Stead (2002) concluded: ‘The results indicate the potential benefits of smoking cessation advice and counselling given by nurses. The challenge will be to incorporate smoking cessation advice as part of standard practice.’
Nurses accept that they have a role in smoking cessation (Faulkner and Ward, 1983), yet it is argued that they have not fulfilled their potential in preventive healthcare (Robson et al, 1989).
Research aims and objectives
A research project was undertaken to explore the links between nurses’ smoking behaviour, their knowledge of, and attitudes to smoking, and their perceived willingness to play a health promotion role in facilitating smoking cessation with patients. In addition a number of research objectives were identified. These were to:
- Establish smoking prevalence among nurses;
- Identify patterns in the smoking careers of nurses and their desire to quit;
- Assess nurses’ knowledge of and attitudes to the health risks of smoking;
- Measure nurses’ knowledge of the benefits of cessation;
- Establish nurses’ attitudes to their health-promotion role in relation to smoking cessation;
- Identify cessation training received and expressed needs for further training.
According to the World Health Organization a smoker is a person who has smoked at least 100 cigarettes in their lifetime (WHO, 1998). In this study an operational definition of a smoker was a person who had reached this level of consumption and continued to smoke. An ex-smoker was a person who had stopped smoking for at least three months. A non-smoker was someone who had consumed fewer than 100 cigarettes in their lifetime.
A questionnaire using multiple-choice and Likert-type questions was developed. Four focus groups were used to validate the questionnaires. In addition, face and content validity were assured by seeking expert validation in the field of smoking research. A pilot study (n=30) was conducted to identify potential problems. From a target population of all qualified nurses in Northern Ireland (13,040), a stratified sample of 2,000 nurses (15.3% of the nurse population) were chosen from acute hospital trusts (73.7%), community trusts (19.7%), private nursing homes and voluntary organisations (6.6%). Survey packs were then distributed. Anonymity and confidentiality were assured.
From the 2,000 registered nurses chosen to take part in the study, 1,199 (60%) returned their questionnaires. Following quality monitoring of the data 1,074 questionnaires were considered usable, representing a response rate of 54%. Of these 72.2% (n=777) were from the acute sector, 23% (n=247) the community sector and 3% (n=30) from the private and voluntary sectors (missing values 1.9%; n=20).
The findings were analysed using SPSS 11.0 for Windows extracting frequency and cross-tabulation of smoking classification and demographic details. Mean scores for Likert scales were generated for the three classifications of smoker, ex-smoker and non-smoker to explore differences. Analysis of Variance (ANOVA) and the Scheffe post hoc test were used to measure the statistical significance of the differences in mean score.
The results showed that 55.2% (n=594) of respondents were non-smokers, 19% (n=204) were ex-smokers and 25.8% (n=278) were smokers. The smokers consumed an average of 14 cigarettes per day and 82.5% of them (n=226) had attempted to give up the habit. For the total sample, the average number of ‘attempts to quit’ was five (mean=4.63). When respondents were asked about their desire to stop smoking, 33.6% indicated that they intended to do so within the next month. When the remainder were asked about their desire to quit, 59.4% stated that they would like to do so within six months. Overall, only 27.7% of the smoking sample did not want to stop within the following six months.
While 92% of the sample was female, smoking prevalence was significantly higher among male nurses (38.1%) than females (24.6%). The highest overall prevalence rates were evident in the age groups 20-25 years (33.9%) and 26-30 years (39.0%). Prevalence rates in the remaining age groups ranged from 20% to 27%. The prevalence of smoking among acute psychiatric nurses, addiction nurses and mental health nurses was 46%, the highest among the various specialties. This was followed by community psychiatric nurses (37.0%) and nurses caring for older people (35.0%). At 16.4% and 16.7% respectively, rates among paediatric nurses and district nurses were below average.
Knowledge of dangers of smoking
Respondents were asked to rate smoking as a risk factor for the following: lung cancer, heart disease, emphysema, bad breath, chronic bronchitis, stomach ulcer, oral cancer, breast cancer, diabetes, early menopause, osteoporosis and low birth weight. All three groups correctly gave lung cancer, heart disease, emphysema and chronic bronchitis top scores and rated early menopause as least likely to be caused by smoking.
When compared with non-smokers and ex-smokers, smokers underestimated the danger of smoking for all of the conditions. In all but two of the conditions (breast cancer and early menopause), a statistically significant difference in mean scores was observed. Two conditions, lung cancer and emphysema, were significant to a p(lt)0.05 level of probability. The remaining conditions had a probability level of p(lt)0.01, indicating a very strong group difference in how smokers responded to the risks.
Post hoc analysis of the results showed that in the 10 conditions that showed statistical significance in the one-way analysis of variance, these differences were between the non-smokers and smokers. In each case, smokers rated the risk of smoking causing the conditions lower than non-smokers. Smokers rated the risks of smoking significantly lower than ex-smokers for diabetes, chronic bronchitis, oral cancer, osteoporosis and low birthweight.
Benefits from smoking cessation
Respondents were asked to what extent they felt stopping smoking would improve the prognosis for a patient with lung cancer, heart disease, emphysema, bad breath, chronic bronchitis, gastric ulcer, oral cancer and osteoporosis. Significant differences in response were seen in bad breath, emphysema, gastric ulcers, lung cancer and osteoporosis. Heart disease and chronic bronchitis were borderline and showed a strong difference between responses across the groups but fell short of statistical significance. Oral cancer was found to have a response rate that did not differ significantly between the three groups.
Scheffe post hoc tests were performed on the conditions that demonstrated statistical difference. In lung cancer, non-smoking nurses rated the benefits of cessation higher than both ex-smokers and smokers (p(lt)0.05). There was no difference in responses between smokers and ex-smokers. In condition three, emphysema, there was a strong difference in response between smokers and non-smokers (p(lt)0.05). No other statistical differences were recorded for this condition. The trend was the same for bad breath and stomach ulcer.
In osteoporosis, two statistical differences were observed: smokers and non-smokers differed in their rating, as did smokers and ex-smokers (p(lt)0.05). In all of the conditions where significant differences were observed, smokers rated the benefits of cessation lower than non-smokers or ex-smokers.
Attitude to risk
Attitudes to the risks associated with smoking were gauged with the aid of seven statements. Responses to three of the seven statements were considered statistically significant. The statements were:
- ‘People have enough problems without adding to them by giving up smoking’;
- ‘Too much fuss is made of the dangers of smoking’;
- ‘The evidence linking smoking to ill health does not really amount to that much.’
The Scheffe test showed that in all three statements the same relationships existed between the three groups. There were significant differences between the two non-smoking groups and those who smoked (p(lt)0.001). For all statements the non-smokers and ex-smokers had more negative attitudes to smoking than the smokers. There were no differences between the non-smokers and ex-smokers.
Attitude to roles
Three statements (c, g, i) (Table 1) dealt with the nurse role as an exemplar. While respondents agreed that they had such a role, in each case smokers rated the nurses’ exemplar role lower than the other two groups, while non-smokers rated this role highest. Two statements (a, e) were related to the nurse role as educator. There was general agreement that nurses should educate patients to the dangers of smoking and the benefits of stopping. Non-smokers rated this role highest and smokers rated it lowest.
Six statements (b, d, f, h, j, k) addressed the nurse role in smoking cessation and health promotion. The findings present a picture of respondents being willing to engage in smoking cessation and health promotion (b, j, k) but feeling that they could be better prepared for doing so (f, h) and that the impact of their interventions had limited success (d). In five statements (b, d, f, h, j) respondents who smoked rated the need for and potential of the nurse’s role in smoking cessation lower than non-smokers and ex-smokers. However, smoking and ex-smoking nurses rated their ‘responsibility’ (k) to help patients who wished to quit more highly than non-smokers. Responses to nine of the ten statements were statistically significant (Table 1).
A comparison of the post hoc analysis showed that in all but one case (k) in which there were significant differences, the smokers had less positive attitudes to cessation than non-smokers and ex-smokers. In the two cases where significant differences existed between ex-smokers and non-smokers, the latter had more positive attitudes.
Training for cessation support
Nurses were asked whether they had received training in helping patients to stop smoking and whether they would like to receive more training in this area. Data analysis shows that 8.5% of the sample had received training in smoking cessation and 91.5% had not. While most of the sample (61.2%) said that they would like more training in smoking cessation, 38.8% did not want training. To analyse the training received and attitudes to further training of the three groups (smokers, ex-smokers and non-smokers), cross tabulations were carried out.
A greater proportion of ex-smokers (11.2%) had received cessation training followed by non-smokers (9.8%), with fewer smokers (3.9%) having received training. A higher percentage of non-smokers (65.7%) said they would like further training, followed by ex-smokers (59.4%). Fewer smokers (54.1%) wanted more training despite the fact that lower numbers of this group had already had some training.
To summarise, fewer smokers than ex-smokers or non-smokers had been trained or wanted further training in smoking cessation. This reflects the differing knowledge and attitudes of the three groups to the benefits of cessation and their roles in it. However, it is encouraging that most respondents (62%) wanted further training.
A prerequisite for smoking cessation support is a sound knowledge of the dangers of smoking. Overall, the results show that nurses have a good knowledge of the dangers of smoking with the major conditions rated highly in terms of risk. However, nurses who smoked rated the dangers below that of the non-smokers and the ex-smokers. This supports the literature regarding nurses’ attitudes to smoking, which shows that smoking status strongly affects attitudes towards smoking (Padula, 1992; Becker et al, 1986).
This study found significant differences between attitudes of smoking and non-smoking respondents with the latter having a more positive stance. All three groups had a sound knowledge of the benefits of quitting smoking, regardless of their own smoking status. However, non-smokers rated cessation benefits significantly higher than smokers for five conditions and ex-smokers rated cessation benefits higher than smokers for two conditions.
Overall, it was clear that nurses accepted their responsibilities as educators in smoking cessation programmes. However, in both the relevant statements, smokers rated their role as educator lower than the others. The literature supports this, showing that smoking status has a mixed effect on nurses’ efficacy as health educators (Nagle et al, 1999; Rowe, 1998). It may be that smoking, if undertaken discreetly, would have no effect on the smoking behaviour of others or on the nurses’ ability to do a good job in smoking cessation. What remains uncertain is why, given their willingness to act as educators, nurses do not do so to the extent expected of them. Further research into the effect of smoking status and the role of the patient in getting help to stop smoking is needed.
Three statements were aimed at measuring nurses’ opinions of their function as a role model. All three groups in this study felt that whether they smoked or not was a personal matter (mean=7.11). However, they saw their smoking behaviour as influencing opinions on smoking and agreed that nurses have a responsibility to set a good example. There was a divergence between smokers and the other two groups. This supports the general theory that nurses resent the image of perfection with which they are associated (Adriaanse et al, 1991), with those who smoke tending to see nursing as just another job and non-smoking nurses tending to adhere to a more idealistic view of nursing. These results support those of Nelson et al (1994), who stated that smoking among nurses was seen to be a significant health problem, not only because of the personal health risk but also because of the negative influence their smoking may have on the smoking behaviour of patients.
Olive and Ballard (1992) also argued that the smoking behaviour of nurses might affect others through modelling. When attitudes to the nurse health promotion role were examined, those who smoked held a more negative view with regard to helping a patient to stop smoking. All three groups rated the success of nurse-led cessation as low (Table 1, statement d) and were uncertain about whether nurses had the necessary skill and expertise (statement f). It is encouraging that most of the nurses across all three groups felt they had an active role to play (statement j) and were interested in hearing about new cessation methods (statement h).
Non-smokers and ex-smokers felt they should proactively encourage cessation (statement b). While smokers scored much lower (mean=5.55), it is interesting to contrast this with their response to statement k: ‘It is the nurse’s responsibility to help patients who wish to stop smoking’. In this case smokers responded more positively than non-smokers and ex-smokers. This result does not contradict previous findings, which showed that smokers had a more negative view of their educator role. Instead it highlights the fact that all nurses recognise the responsibilities they have in helping patients who want to stop smoking. All respondents scored it highly with a total average score of 7.8 out of 10. The key may be that in this case the patient wishes to stop.
However, Goldstein et al (1987) argued that to offer cessation only to motivated smokers is to miss opportunities for effective interventions for all smokers. Thus nurses felt they had a responsibility for cessation but seemed to find it difficult to personally facilitate this. This fits the description of the paradox that Rowe (1998) described. Rowe posited that there are nurses who are smokers, ex-smokers and non-smokers who do not believe that they are well placed to provide help and information on smoking cessation, while there are other nurses in the same three groups who feel their status is an asset when providing support.
These results also support the findings of Nagle et al (1999) who found that nurses believed helping patients quit smoking should be part of their job, particularly if the patient wants to stop. They found that over half of the sample thought their smoking status to be beneficial in providing smoking cessation advice. The patient’s request for help in stopping smoking was considered by respondents to be a key factor in prompting help being offered by nurses.
The findings further reinforce the message that a willingness exists in the nursing community to help patients stop smoking. It is encouraging that while only a small proportion of nurses had received cessation training, the majority in all three groups of respondents expressed a wish to receive such training. Those professionals who receive training are more likely to provide effective cessation support (Godfrey et al, 1993). This study’s findings are in marked contrast to those of Hodson et al (2002) who found that only a minority of midwives had been trained and three-quarters felt they did not need training.
This was a systematic study designed to address research aims and objectives. However, it was subject to the following limitations:
- The data was obtained by self-reporting. There is a tendency to give researchers answers they may wish to hear. However, anonymity and confidentiality were employed in an attempt to reduce this bias;
- The questionnaire was conducted over the busy Christmas period and this may have contributed to the low response rate (54%).
Conclusions and recommendations
Based on the findings, the following conclusions and recommendations seem plausible. There should be:
- Effective smoking prevention and cessation programmes in schools. Most nurses start smoking in their mid-teens before joining the profession;
- Provision of evidence-based smoking cessation support for nurses in colleges and while working. This should be tailored to the student or nurse’s level of dependence and motivation to quit;
- Dedicated time and resources for nurses to integrate smoking cessation support for patients into their practice;
- Further research into what factors prevent most nurses who accept they have a role in providing cessation support from realising this role in practice.
This study has revealed clear links between nurses’ smoking status and their knowledge and attitudes to issues relating to smoking cessation. These differences are underlined by attitudes to the value of smoking cessation, their role in providing it and their willingness to undergo training. Nurses who smoke rate the health risks of smoking lower than non-smokers and ex-smokers. They are more likely to have negative attitudes to smoking-related issues and rate the benefits of cessation much lower. However, most nurses who smoke want to quit and have tried to do so several times.
There is a need to increase knowledge of the health risks of smoking among nurses and address attitudes to these risks in nurses who smoke. It is important to improve smokers’ awareness of the benefits of cessation and appropriate support should be provided to smoking nurses to motivate and enhance their own cessation. This would produce an improvement in their own health and improve the status of the nurse as exemplar and role model. In addition, it would produce a larger cohort of ex-smokers. It could be expected they would have more positive views about undertaking cessation training and providing cessation support to all smokers on a proactive basis. It appears that many nurses are willing to support smokers but only if the smoker expresses a need.
A culture shift is needed whereby all nurses address smoking proactively. There is a concurrent need to raise smokers’ awareness of potential support mechanisms and to encourage them to seek these. Most respondents, including smokers, feel that they should be an educator and to a lesser extent, a role model. Those who smoke rate their role in cessation support lower than those in the other two groups. However, most of those in the three groups felt strongly that there was a role for them in health promotion and that they had a responsibility to encourage cessation.
All three groups seemed unclear as to whether nurses have the adequate skills and expertise to provide cessation support. This, alongside the fact that only a small minority of nurses have received training - and this number is very low for smokers (3.9%), means that there is an urgent need to provide cessation training and refresher courses. In this context it is especially encouraging that most respondents in all three groups want future cessation training. This should address nurses’ knowledge, attitudes and behaviour along with awareness of the effectiveness of cessation support interventions and their role in these.
It is also important that nurses are provided with opportunities to put training into practice. All of this would necessitate changes to the curriculum and a reassessment of nurses’ contracts to provide dedicated time and additional resources for smoking cessation activities.
This article has been double-blind peer-reviewed. For related articles on this subject and links to relevant websites see www.nursingtimes.net