This article provided a systematic literature review of social marketing interventions to promote PA among over 60-year-olds. We found very few published studies on this topic; only nine interventions emerged from our search of five scientific databases. Of the nine interventions, eight concluded that social marketing was effective for increasing PA among the elderly.
The eight studies that reported a positive assessment of their programs believed their success to be mainly due to three factors. First, the use of social marketing techniques through the implementation of a number of criteria. A better understanding of core marketing concepts supports better planning, implementation, and effectiveness. Second, the funding of the activities, which made it possible to tackle financial barriers. Third, the building of social ties between participants in the programs, which provided them with support and interaction. For instance, Varma et al. (2016) suggest that walking levels increased among women because they were taking part in a new voluntary activity.
Beyond effective social marketing programs, one failed. Kamada et al. (2013, 2015) state that their program did not succeed in demonstrating an increase in PA levels at 1 and 3 years because it was not comprehensive enough. For that reason, they suggested that a strategy of modifying the environment by improving the public transport system and building facilities specifically for PA could facilitate the meeting of recommended PA levels. Five years after the launch of the study, Kamada et al. (2018) showed that their intervention had had a positive effect.
Eight interventions identified in our systematic review used the marketing mix and its "4Ps": Price, Place, Promotion, and Product. The 4Ps represent a group of operational areas for developing strategies and tactics. This theory has been consensually accepted. However, Gallopel-Morvan et al. (20) advance five criteria they deem more suited to social marketing and behavioural change: Influencers, Behaviour, Cost, Ease of Access, and Communication. Notably, social marketers have a range of marketing tools at their disposal and use them in interventions.
The segmentation variables reported in the PA studies focused on basic descriptors of demographics (who?) and geography (where?). Although tried and tested, a narrow focus on markers as sex and gender can limit the kinds of insight required to develop an effective marketing strategy (44). For instance, psychographic segmentation addresses the motivational concerns of target consumers and is widely used to develop user personas and profiles. Research on consumer personality traits, values, attitudes, interests, and lifestyles can identify what motivates targeted behaviours. Such insights have informed interventions that target segments of young adults who use tobacco (45,46), for example Lisha et al.’s (46) study that identified distinct tobacco use risk patterns across different groups of young adults (Hipster, Country, Hip Hop, Partier, Homebody, and Young Professional). Research on the mindsets of older people and their views on physical activity are both lacking and needed.
Furthermore, implementing segmentation schemes that place a greater emphasis on actual consumer behaviour holds great potential (21). Marketing involves exchanges, and consumers make decisions not only about what they "get" (e.g., products or services) but also about what they "give up" (e.g., allocation of time, effort, money, comfort, autonomy, privacy, social reputation). Social marketers ought to pay close attention to the nuances of behaviour and heed segment-level differences in benefits sought and willingness to pay (47), purchase and use occasions (48) or even patterns of social media sharing (49). Used together, psychographic and behavioural segmentation can help to better understand and target messages for high-risk subgroups (46) and inform the development of targeted messages that echo a target audience's existing views and practices, and therefore, produce more powerful persuasive effects (50).
Future research might consider looking at the dynamics of influence within the social networks of older people, given that exercise is often initiated and enacted with other people. A networked view of consumption can yield innovative intervention strategies. For example, consider the people implicated in mothers' breastfeeding decisions (51). Breastfeeding initiation and cessation decisions are shaped by varied aspects of the mother-infant relationship, guided by different healthcare professionals, muddled by mother-other child duties, enabled by spousal or caregiver support, and impacted by the (re)actions of other people in the workplace and public sphere. In a similar vein, research has leveraged insights about peer crowd affiliation to promote smoking cessation (52). A consumption ensemble approach (53) could inform and inspire the development of sociologically-based intervention strategies.
The measurement of “daily step counts” by Varma and al. (2016) is an excellent example of how technology (wearable devices) can generate performance feedback to both motivate participants and to assess the success of interventions. Applications of wearable technologies in healthcare had recently gained considerable interest from researchers and industry alike. Similarly, the retailing and services marketing literature increasingly look to self-service technologies as ways to improve service experiences while achieving operational efficiencies.
It is difficult to say whether social marketing is useful in promoting PA among seniors since none of the nine interventions selected used the entire approach (i.e., all seven benchmark criteria). Haruka Fujihira et al. (25) showed that the more a program uses the social marketing benchmark criteria defined by A.R. Andreasen, the more it becomes effective at changing the behavioural habits of target audiences. Literature reviews such as those by Julia Carins and Sharyn Rundle-Thiele (54), Bo Pang et al. (55), and Martine Stead et al. (56) have classified social marketing interventions according to the same benchmark criteria as Andreasen. They have reported similar findings to those of Fujihira (2015).
Our literature review is based on the classic and widely-cited social marketing benchmark criteria defined by A.R. Andreasen (21). Moreover, we added a seventh criterion, evaluation. Actually, other researchers, however, have expanded the theory. Observers have noted that many social marketing programs are poorly assessed or are not evaluated at all (49). Accordingly, Gallopel-Morvan et al. (20) add a seventh criterion, evaluation, as a fundamental step in the process of a campaign using social marketing theory (20). Among other things, program evaluation can provide valuable information about program design and implementation, and whether or not the program succeeded in changing the targeted behaviour (58).
Still, to improve the effectiveness of prevention campaigns using social marketing, other researchers adopt the eight criteria recommended by the UK National Social Marketing Centre (44, 45). Among the additional criteria, one relates to the use of behavioural theory in social marketing intervention. Thus, recent work shows that when the theory of behaviour change is integrated into the construction of interventions, additional positive effects of the changes are noted (55,60). It is in this sense that the work of Sharyn Rundle-Thiele et al. (61) also goes, explaining that current research focuses on the explanation of individual behaviours and not on behaviour change. Moreover, the authors proposed ten social marketing theory development goals to “assist social marketers to develop new ways of thinking that will deliver the theory and evidence base needed to outline what practitioners and policymakers should do to effect change” (61). The social marketing method is a young discipline. Today, several theories are in development. The most recent theories state that social marketing criteria should no longer be used as a simple tick-box checklist, but as a set of interrelated concepts (59).
Additionally, the papers that we selected in this review comprise certain limitations in terms of the assessment method, group heterogeneity, and contamination risk. The assessment models chosen preclude any definite direct attribution of the positive results to any one part of the intervention or any combination of activities. In some studies, discrepancies between the intervention and control groups interfered with the assessment of change between them. Withall et al. (2012) suggested that the groups had differences (particularly in terms of age and ethnic origin), which prevented them from making direct comparisons during the assessment. Kamada et al. (2015) thought that the results of their study were contaminated because the control group may have been exposed to the social marketing campaign developed for the intervention group. This unintended exposure may have been due to the geographical proximity of the groups and word-of-mouth. Moreover, the criteria selected for assessing the prevention programs were not always the same, making it difficult to compare the interventions, or were insufficient to measure effectiveness. For instance, in the study by Withall et al. (2012), the criteria only gauged PA program participation without measuring PA levels, PA intensity, or biological markers.
Some limitations of this study are to be highlighted. First, there is a publication bias because negative studies were probably less likely to be published. Second, our systematic literature review excludes public health and community-based interventions, which perhaps use social marketing techniques.
Different age groups (children, adults, seniors) have specific characteristics and guidelines in terms of public health and necessitate that separate social marketing interventions be designed for each one. Given both the lack of studies on using social marketing principles to promote PA in seniors, there is a need to develop more social marketing intervention promoting physical activity and targeting older people (25).