The study used a controlled before-and-after study design, based on the TREND framework
]. The intervention consisted of a 6-month, social marketing campaign costing £8,000, delivered between late September 2010 and March 2011. A group of nine local residents and local community and health workers acted as an advisory group to the study. They were consulted on a quarterly basis and at key points of the campaign. No intervention occurred in the control area other than normal delivery of existing sessions. A range of physical activity sessions was already available at the leisure centre in the intervention area. These activities were unrelated to the intervention and were running prior to it beginning; they are referred to here as pre-existing sessions. Data were collected at baseline, mid-point and at the end of the intervention at the five weekly intervention sessions, at baseline in the five weekly physical activity sessions running in the control area and at the nine weekly pre-existing sessions in the intervention area.
The intervention area was Southmead, and the control area, Filwood, both suburbs of Bristol, UK. The two areas have similar population levels (11,000-12,000) with 5.2% and 3.8% Black Minority Ethnic (BME) residents respectively. Both have low life expectancy (75.3 years and 76 years respectively) compared to the UK average of 80 years
]. The areas have similar Index of Multiple Deprivation scores (69.60, 69.23), based on income, crime, employment, health and education statistics
]. They are the second and third most deprived wards in Bristol
] and despite priority investment, have below average percentages of residents who exercise at least once a week
Components of the Campaign
The intervention was designed to comply with the National Social Marketing Centre’s Benchmark Criteria
]. Formative research was conducted, the results of which have already been published
]. Briefly, this work suggested that there are some key issues relating to engagement in physical activity sessions that have a greater impact on low-income groups than the general population. Issues of particular importance are cost and childcare; communication/session awareness; the social support particularly required by women to attend organised exercise sessions; the importance of socialising, fun and enjoyment; and concerns regarding perceived competence.
A summary of the input from the formative research into campaign planning is shown in Table
Summary input from formative research into campaign planning
Social marketing prescribes that a target market should be defined in terms that relate to behaviour. The target market selected comprised of individuals with an awareness of the benefits of exercise and positive attitudes towards it; anxiety-related barriers to joining exercise sessions; a fear of arriving alone (lack of relatedness); and a perceived lack of competence and autonomy. The formative research indicated that this target group was likely to be substantial in size and largely female.
The intervention marketing mix and branding strategy are detailed below.
The product development was based on findings from the formative research, and the literature relating to recruitment and retention. Three kinds of dance session (Line Dancing, Zumba and Salsa), an instructor-led gym session, and Body Tone, a balance and stretching session, were selected to form the product offering
]. These five sessions each took place once per week at 9.15 am, so a different session ran every weekday morning. Sessions were all held at the local leisure centre, which is described under ‘Place’ below. It was hoped that parents would attend sessions immediately after ‘the school run’, prior to returning home and engaging with their established morning routine. Socialising is associated with adherence to exercise, so free tea and coffee was available before and after sessions and socialising was actively encouraged
Branding is central to social marketing. It builds strong bonds between the product and its target market by creating a tone, feeling or emotional response to a product or service
]. A number of brand names were tested with the target group. Fit and Fab was selected because fitness was a commonly used term and was considered less ‘worthy’ than health or exercise. Using ‘Fab’ with ‘Fit’ diluted its seriousness, suggested an inclusiveness of women, was not associated with slimness and was easy to remember and pass on.
The branding strategy was designed to generate the feeling of a local movement, to create momentum, convey a sense of fun and enjoyment and to imply that this was a programme for ‘people like me’. It emphasised session accessibility for beginners, the unfit, the overweight and people of any age. Branding approaches were tested with the advisory group and members of the target group. The final Fit and Fab logo is shown in Figure
Fit and Fab Outdoor banner (8ft x 3ft).
All the five sessions were branded as Fit and Fab. The intention was to create one offering within which participants were offered choices. The different sessions were the product mechanisms which delivered the brand values of localism, accessibility, fun, enjoyment and social norm and which enabled the intervention’s overall objective (increased participation in physical activity) to be achieved.
People, Place and Price
Session leaders were encouraged to maximise fun and enjoyment. They were issued with a session delivery guide and provided with regular feedback.
Fit and Fab was delivered in an old school building which was converted into a leisure centre 25 years ago. It is run by a charitable trust for the benefit of local residents and is geographically close to a large proportion of the intervention area’s population.
The price of a product is ‘the cost that the target market associates with adopting the new behaviour’
]. This incorporates both monetary and non-monetary costs, such as time, energy and any psychological costs
]. An effective pricing strategy tips the balance between the costs and benefits of the product such that the target market chooses to exchange their current behaviour for the target behaviour. Monetary cost is frequently presented as a barrier to engaging in organised physical activity and has a proportionately greater impact on low-income groups
]. While some argue that free activities are less valued
], the impact of a very low cost approach in a low-income area has not been fully tested. For the first six weeks of this intervention all sessions were free, from then on the cost was £1.
Issues of time, and fitting exercise into the daily routine, were often cited as barriers to exercising in the formative research. As a result a 9.15 am start time and short 45 minute sessions were used. As tackling issues of childcare was impractical for this campaign, those with children of at least school age were targeted.
Issues of low self-esteem, confidence to exercise and confidence to attend an activity session were addressed though the promotional campaign.
Promotion is the development and deployment of persuasive materials and activities to convincingly communicate the product benefits and its value, particularly in relation to competing products or activities
]. The key campaign promotional messages emerged from the analysis of barriers, enablers and motivations to change behaviour.
The selection of media channels was based on the target group’s media habits, the communication objectives, the available local media and their geographic coverage, and budget constraints. Promotional activities were designed not to spill over into neighbouring, affluent areas to avoid the usual bias towards engaging the middle classes
]. As Fit and Fab was previously unknown, and the intention was to move the consumer rapidly towards action, the promotional campaign aimed to reach the target group a minimum of three times, preferably more. The key promotional elements included 17 outdoor banners (8 ft × 3 ft); 3 door drops (4,500 residences each drop); street leafleting; leaflet distribution via schools, community groups and GPs; a poster campaign; face to face recruitment; local press; a campaign blog; a text campaign; two taster sessions; and a loyalty scheme.
The main promotional objectives were to build high levels of awareness, tackle barriers and support engagement, promote the Fit and Fab brand and its values, generate word of mouth promotion, create a ‘buzz’ around the area and offer session trialability.
Fit and Fab materials are shown in Figures
and the campaign budget is detailed in Table
Social marketing campaign budget
During the period of the intervention the pre-existing and control area sessions were promoted through posters at the session venues and inclusion in the programme of events published by each venue. Details of the pre-existing sessions in the intervention venue were included in a flyer that was delivered to c4,500 homes in the intervention area.
The data collection, analysis and results were conducted in four main sections, participants, recruitment, attendance and adherence.
Participant data were collected to assess equivalence across the intervention and control areas. Attendance data enabled assessment of the effect of the campaign on participant recruitment, attendance and adherence in the intervention and pre-existing sessions, and as compared to the control area. Attendance data were collected weekly and presented and analysed monthly. This study was designed to mirror usual community practice; as such participants were allowed to join the sessions throughout the intervention. Therefore monthly data analysis was considered to be the most informative method of presenting the results. This method provided an overview of participation patterns across the intervention period rather than just at a beginning and end point, and enabled the calculation of adherence rates. Data from the participant surveys were collected to enable a comparison of the characteristics of pre-existing attendees and those recruited through the campaign, and to compare the characteristics and motivations of adherers and drop outs.
The intervention site was a local leisure centre. It offered a range of physical activity sessions and operated a gym. The leaders of these pre-existing sessions, and the gym manager at the leisure centre, agreed to provide their attendance data. All session leaders and the gym manager at the leisure centre and health park in the control area also agreed to provide their session attendance data.
To provide baseline data, attendance data at these pre-existing sessions in the study area (nine sessions per week), and sessions in the control area (five sessions per week across two centres) were collected for one month prior to the intervention and for the whole of the intervention period. Attendance data at all the intervention (Fit and Fab) sessions were collected using a paper-based register, for the whole of the intervention period.
Three different questionnaires were used to collect demographic data and reasons for participation from five different participant groups.
Group 1 were participants at physical activity sessions already running at the leisure centre where the intervention was conducted (pre-existing sessions). At baseline, Group 1 completed Questionnaire A which recorded name, age, gender, postcode, height, weight, ethnicity, attendance duration, attendance frequency, attendance with a friend, confidence to start attending alone, reasons for attending and awareness of the intervention.
Group 2 were participants in physical activity sessions in the control area. They also completed Questionnaire A at baseline, as described above. Group 3 were participants in the intervention (Fit and Fab) sessions during month 1 and they completed Questionnaire A on recruitment.
Group 4 were those who had participated in the intervention (Fit and Fab) from month 1 and were still attending after 3–4 months. During months 3 and 4 they completed Questionnaire B. This recorded name, age, gender, postcode, height, weight, ethnicity, attendance duration, attendance frequency, attendance with a friend and investigated reasons for adherence and how this was affected by the campaign.
Group 5 were participants who were recruited into the intervention during month 6. Group 5 completed Questionnaire C on recruitment which recorded how the campaign influenced attendance, and reasons for attendance.
For recruitment into groups 1 and 2, information sheets were distributed at sessions and 47 consent forms and questionnaires were completed in the intervention area and 52 in the control area. Group 3, 4 and 5 participants were recruited at Fit and Fab sessions during month 1, months 3 and 4 and month 6 of the intervention respectively. Consent forms and the questionnaire were completed by 46 participants in month 1, 24 participants in months 3 and 4, and 41 participants in month 6.
The study was approved by the University of Bristol School of Applied Community Health Studies Research Ethics Committee (Ref 015/10).
Participants in pre-existing sessions in the intervention and control areas (Groups 1 & 2) were compared. Participation data from intervention, pre-existing, and control area sessions were compared to assess the effect of the campaign on participant recruitment, attendance and adherence. Recruitment, attendance and adherence rates at the different types of Fit and Fab sessions were also compared. All analyses were conducted in SPSS (version 16.0) and alpha was set at p
Chi-squared tests of independence were used to examine if participants attending pre-existing physical activity programmes in the intervention and control areas differed by age category, BMI range, ethnicity, attendance frequency, attendance with a friend or confidence to attend alone.
There are no widely accepted definitions of the terms recruitment, retention or adherence (retention and adherence appear to be largely interchangeable), particularly relating to public health interventions. Editors appear to be reluctant to publish articles where recruitment and adherence are the prime focus
], and details of recruitment and adherence procedures are often not reported in published studies
]. In this study recruitment reflects whether the target group overcame any pertinent barriers and were sufficiently motivated to engage with the intervention at least once. Attendance is the number of participants at any one session, and indicates the scale of the impact of the intervention.
In most published studies, adherence has been defined as attendance at a certain percentage of the available sessions. Due to the extended length of this intervention, the large number of sessions available and the on-going recruitment throughout the intervention, this approach could have led to misleading results. In this study, two levels of adherence were set. Both required current, on-going participation, defined as at least one attendance during the final month of the intervention. In addition, ‘Low adherence’ required 6–12 session attendances over the 6 months of the intervention, and ‘High adherence’ >12 attendances. The total number of adherers is defined as the sum of the low and high adherers.
Recruitment and attendance
To compare recruitment and attendance rates between the Fit and Fab, pre-existing and control area sessions, analysis of variance tests were used with mean monthly recruitment rates or mean monthly attendance rates as the outcome and session type (intervention/pre-existing/control) as a factor. Significant main effects were further explored using Bonferroni post-hoc multiple comparisons. In addition, χ2 tests of independence were used to compare the relative importance of different recruitment mechanisms in the early and late stages of the intervention.
Percentages were used to compare adherence rates at Fit and Fab, pre-existing and control area sessions, and between the different types of Fit and Fab session. Additionally, χ2 tests of independence were used to examine if intervention drop-outs and adherers, differed by age category, gender, postcode, BMI range, ethnicity, attendance duration, attendance frequency, attendance with a friend or motivations to participate.