The impact of a social norms approach on reducing levels of misperceptions around smokefree hospital entrances amongst patients, staff, and visitors of a NHS hospital: a repeated cross-sectional survey study

Smoking is a major cause of inequalities in health [1, 2] and an avoidable cause of morbidity and mortality in the United Kingdom [3, 4]. The prevalence rate of smoking in England is 15.5% in the adult population [5]. Nationally the cost of smoking to the NHS has been calculated to be between £2.5 billion [6, 7] and £5.2 billion [8].

Evidence suggests that smoke-free legislation has been effective in protecting people from the harmful effects of second-hand smoke exposure while in public places and buildings [9–11]. There is however evidence for mixed support for smoke-free policies in some NHS settings (e.g. psychiatric services) [12]. Internationally, countries report mixed results after implementing smokefree policies within hospital settings. For example, Greece struggled to implement their 2002 ban of smoking in enclosed places [13].

The introduction and implementation of the smokefree legislation in England in 2007 has had public health benefits by helping smokers to quit, reducing cigarette consumption and smoking prevalence [6]. While the national smoking ban inside buildings is understood and adhered to, there is a perception this has led to staff, patients and visitors smoking at hospital entrances and within hospital grounds [14, 15]. Unintended consequences of smoking bans that inadvertently see a concentration of smokers at entrances or perimeters remain a public health concern [16, 17].

National Institute for Health and Care Excellence (NICE) guidance PH48 on smoking and acute, maternity and mental health services includes recommendations on developing, implementing and communicating, smokefree policies; this includes smokefree grounds [18]. However, only one in sixteen hospitals have completely enforced the ban on smoking on hospital premises with calls for trust boards to be held to account to enforce smoke free hospital policies and support quit attempts [14, 19].

In March 2017 the national planning document ‘Next steps on the NHS Five-Year Forward View’ set a deadline for all of the NHS estate to be smokefree by 2019/20 [20]. In July 2017 the Department of Health published ‘Towards a Smokefree Generation – a Tobacco Control Plan for England’ [6] which included a commitment to make all mental health inpatient service sites smokefree by 2018. It recommended that Public Health England and NHS England work together to implement NICE guidance on helping pregnant smokers to quit; to provide access to training for all health professionals on how to help patients to quit; and that NHS Trusts encourage smokers using, visiting and working in the NHS to quit, with the goal of creating a smokefree NHS by 2020.

Widening tobacco control policies could help shift social norms about the acceptability of exposing others to second-hand smoke and changing cultural attitudes and norms to smoking.

Reducing smoking requires action at an individual and environmental level. Intervention is most effective when it is part of a portfolio of work aimed at denormalisation of smoking (e.g. [21]). For the NHS, this portfolio includes formal policies on smoking (e.g. smokefree hospital grounds) and awareness raising campaigns designed to change the social acceptability of smoking within a NHS premises. Research into normative misperceptions has given rise to a new form of intervention, known as the social norms approach [22, 23]. The premise of this approach is simple; if individuals’ misperceptions of behaviours and/or attitudes can be corrected then the perceived social support for them to engage in positive health behaviours (e.g. keeping hospital entrances smokefree) will be increased and their own engagement in the positive behaviour will increase.

The social norms approach has been used to increase health behaviours in schools, colleges, and universities but there remains a paucity of research investigating the feasibility and effectiveness of implementing social norms approach campaigns in healthcare settings [23]. The current study investigated the outcomes associated with a social norms approach campaign designed to reduce misperceptions of the number of people who smoke in the hospital entrance and who support smoking in the hospital entrance.

Mean age of participants 49.12 years (SD = 16.07). The majority of participants were white/white British (n = 846, 90%), with 6% Asian/Asian British (n = 52) and < 1% Black/African/Caribbean/Black British (n = 7). In total 68% (n = 636) were female. The majority of participants reported being non-smokers (n = 794, 85%).The overwhelming majority of participants reported not smoking in hospital entrances (n = 916, 97%). Not smoking on hospital grounds was also reported as a majority behaviour (n = 867, 92%).

ANOVA revealed an overall significant effect of time across the three attitudes towards smoking in the entrance variables (F(3, 9.14) = 10.72, p < 0.001). With significant differences between the two data collection points (before and after) in the perceived percentage of patients (F (1, 916) = 29.14, p = < 0.001), staff (F(1,916) = 15.10, p < 0.001), and visitors (F(1,916) = 12.25, p < 0.001) respondents perceived agreed that hospital entrances should be smokefree. In all instances the mean of the percentage respondents thought agreed that hospital entrances should be smokefree was higher for those responding after, compared with before, campaign implementation (see Table 1).

Results illustrate high levels of support for smokefree hospital entrances. The majority of participants agreed that patients (before n = 435, 91% agreed; after n = 414, 90% agreed), staff (before n = 437, 92% agreed; after n = 426, 93% agreed) and visitors (before n = 431, 92% agreed; after n = 419, 92% agreed) should not smoke in the hospital entrance. Although the majority supported smokefree hospital grounds, support for not smoking on hospital grounds was lower. More specifically, between 60 and 72% agreed that patients (before n = 288, 60% agree; after n = 315, 69%), staff (before n = 300, 64% agree; after n = 330, 72% agree) and visitors (before n = 298, 63%; after n = 329, 72%) should not smoke on hospital grounds.

The majority (98%) of patients, visitors, and staff reported not smoking in the hospital entrance. The perception amongst respondents completing the survey before the campaign was that 46, 68, and 48% of patients, staff and visitors respectively did not smoke in the hospital entrance. This was despite the fact that the majority of respondents reported not smoking and believed that the entrance to the hospital should be smokefree.

The social norms marketing campaign was designed to promote the message that the majority of people reported keeping the entrance smokefree. This campaign, was associated with a reduction in the percentage of people participants’ believed smoked at the hospital entrances. Specifically, we observed a small yet significant increase of people believing that patients, staff and visitors kept the entrances smokefree. The social norms theory predicts that highlighting that the majority of patients, staff, and visitors report not smoking in the hospital entrance would encourage those that do smoke in the entrance to conform to the ‘norm’ and make the hospital entrance smokefree.

Like many NHS hospitals in England, Pinderfields Hospital has a no smoking policy. The current study highlighted misperceptions of the social norm as one of the challenges faced when implementing and enforcing no smoking polices. Specifically, the perception that the majority of people will smoke in entrances being at odds with the majority of people reporting supporting smokefree hospitals and reporting that they themselves keep the entrance smokefree. The results from this study provided staff at the hospital with evidence that the majority of people report not smoking in the hospital entrance and showed support for keeping the entrance smokefree. This may encourage other people to conform to the ‘norm’ and either not smoke or smoke elsewhere and give staff the confidence to challenge smokers at the entrance knowing that the majority of people agree with keeping the entrance smokefree.

The success of the implementation of the intervention was aided by the team’s ability to build relationships with hospital management and staff. These relationships also enabled researchers to gain access throughout the hospital, this included access to inpatients on the ward. Given access to a wide variety of areas in the hospital building helped to ensure there was equal representation from the three target groups of staff, visitors, and patients.

Due to restrictions on respondents being given a financial incentive to complete the survey a charity donation of £1 was offered for each survey completed. Respondents picked from a choice of three charities, this proved a good recruitment incentive and encouraged engagement as evidenced by the relatively high response rate.

Limitations of the project include the sample being one of convenience rather than randomly selected. While researchers collected data from a variety of locations around the hospital there were areas of restricted access which researchers could not enter and the timing of data collection from each location was one of convenience. It is therefore likely that the sample is not representative of all those present on the hospital grounds during the data collection period. The sample included more women than men (68% vs 32%), and a self-reported smoking prevalence of 15%. In 2012 25% of adults 16 years or over in the Wakefield area reported smoking [25]. It is not possible to ascertain if the sample was representative of those present on the hospital grounds during the data collection period or if the demographic profile is a result of sampling biases introduced as a result of the decision to use a convenience sample. In addition, the current project used data from two cross-sectional data collection points. This before-and –after design means changes across time cannot be attributed with confidence to the campaign. Data was collected using cross sectional surveys at two timepoints, the design would be strengthened had we been able to follow participants across time. Respondents smoking status was self-reported and we were unable to verify smoking status or smoking behaviour using objective measures. Despite responses being anonymous it is possible that social desirability bias influenced participant’s willingness to identify as someone who smoked in hospital entrances and on hospital grounds. Social desirability is likely to have been felt most acutely by hospital staff. The design would have been strengthened from objective observations of smoking behaviour at and around the hospital entrance. The campaign message ‘99% of patients, staff, and visitors keep our hospital entrance free from cigarette smoke’ may have led to some participants believing that the data had been collected using objective measures of smoking at the hospital entrance. A more accurate campaign message may have been ‘99% of patients, staff, and visitors report keeping our hospital entrance free from cigarette smoke’. The current data can say nothing about how the wording was received nor of the extent to which the distinction between ‘self-report’ and ‘actual’ behaviour is of importance to the intended audience.

Hospitals have a duty to protect the health and wellbeing of staff, visitors, and patients. Public Health England chief executive Duncan Selbie has called on hospital managers for greater leadership on tobacco control [14, 26]. Smoking kills 79,000 people each year in the UK. It was responsible for 474,000 hospital admissions in 2015/2016. In spite of smokefree legislation, over 30% of men and 26% of women still report some exposure to second-hand smoke [27]. People smoking at the entrance to NHS trusts gives a poor impression, leads to increased litter of cigarettes butts and also means that those entering and leaving buildings have to pass through tobacco smoke. For people using secondary care services, smoking is associated with increased risk of complications and delays to recovery [18]. Having a smokefree policy in a NHS hospital which is not adhered to or enforced undermines public health messages around the dangers of smoking.

The social norms intervention at Pinderfields Hospital was valuable in raising the discussion and awareness of keeping the hospital entrance smokefree amongst staff, patients, and visitors. To capitalise on the interest and exposure this campaign has generated it is advisable that other measures are now rolled out to achieve a smokefree hospital included improved signage, staff training, policy revision, enforcement, and a revised communication strategy to ensure the smokefree message is disseminated through all external and internal communication.

This study showed high levels of support amongst the target audience for keeping hospital entrances smokefree. It also highlighted misperceptions that participants had in relation to other people smoking in the hospital entrance. Despite only 1% of people smoking in the hospital entrance participants believed this figure to be much higher. This apparent misperception could be due to the high visibility of smokers and the lingering tobacco smoke at the hospital entrance. This study should encourage the hospital to maintain and enforce the smokefree policy on the hospital site and encourage the development of other areas to ensure the hospital does become completely smokefree. Social norms messages alone are unlikely to change all smokers’ behaviour. The approach should be combined with smokefree policies, training, and enforcement to help encourage smokers to conform to the behaviour of the majority, which in this case is not smoking in the hospital entrance. This study offers a platform for further work to reinforce the smokefree policy throughout the hospital with all audiences, building on the awareness raised by the social norms campaign.