1
How Acceptable Do Older Adults Find the Concept of Being Physically Active? A Systematic Review and Meta-Synthesis
Abstract
Despite the significant health benefits of regular physical activity for older adults, only a minority achieve recommended levels. To develop effective interventions, the reasons for the low levels of physical activity in this population must be understood. The present review identifies and synthesises qualitative studies concerning the acceptability of physical activity to community dwelling older adults. A systematic search of four electronic databases identified 10 studies meeting inclusion criteria. These were appraised and findings combined and compared using Thematic Synthesis. Older adults construed physical activity as a by-product of other activities, rather than as a purposeful activity within itself. This seemed to be linked to their self-perception as an ageing member of society, with physical activity considered irrelevant and competing roles and responsibilities, e.g. family, taking precedence. Additionally, older adults appeared to experience conflict between maintaining their autonomy and accepting the physical and social vulnerabilities associated with ageing. As older adults do not see physical activity as purposeful within itself, interventions promoting moderate or vigorous physical activity are likely to have limited success. As even small increases in physical activity benefit older adults, future interventions may wish to target the reduction of sedentary behaviour in this population.
Key words: Physical activity, older adults, acceptability, meta-synthesis, interventions, systematic review
Introduction
The health benefits of engagement in regular physical activity have been well documented and a lack of physical activity is one of the leading global risk factors for mortality (Lee et al., 2012).Physical activityis defined as “any bodily movement produced by skeletal muscles that results in energy expenditure” (Caspersen, Powell, & Christenson, 1985, p126). Though the terms are sometimes used interchangeably, ‘physical activity’ is distinct from ‘exercise’, which instead refers to “a subset of physical activity that is planned, structured, and repetitive and has as a final or an intermediate objective the improvement or maintenance of physical fitness” (Caspersen et al., 1985, p126). Physical activity entails significant health advantages for older adults, includingreduced rates of coronary heart disease, type 2 diabetes, and colon and breast cancer(Bauman, Lewicka, & Schöppe, 2005; Paterson, Jones, & Rice, 2007; US Department of Health, 2008). Physical activity additionallyhelps to prevent frailty, loss of independence(Paterson & Warburton, 2010; Peterson et al., 2009),and cognitive decline (e.g. Guiney & Machado, 2013), andimproves mood, sleep, and quality of life (Reid et al., 2010) in older adults.
International recommendations currently state that adults aged 65 years and above should engage in a minimum of 150 minutes of moderate-intensity physical activity each week(World Health Organization, 2010a). In addition, older adults should perform physical activities to improve both muscle strength on at least two daysand balance on at least three days per week respectively, and should minimise extended periods of sedentariness, which is defined as “any waking behaviour characterised by an energy expenditure ≤1.5 METs while in a sitting or reclining posture”(Sedentary Behaviour Research Network, 2012, p540). Many older adults fail to achieve the amount of physical activity advised by these guidelines andphysical inactivity increases with age.In the UK just 58% of men and 52% of women aged 65-74 years reported achieving the recommended amount of physical activity in 2012 (NHS Information Centre, 2014). This proportion declined further for older age groups, with just 43% of men and 21% of women aged 75-84, and 11% of men and 8% of women aged 85 and above meeting the guidelines. This decline in physical activity with age appears to be consistent throughout the developed world (e.g. Craig, Russel, Cameron, & Bauman, 2004; Casperson, Pereira, & Curran, 2000).As the world population is ageing rapidly (World Health Organization, 2012),increasing physical activityand thereby reducing the financial burden of morbidity on public funding represents animportant target for health-related behaviour change.
Physical activity interventions targeting non-clinical older adult populations are often developed based on studies involving general adult populations. However, it is likely that different types of interventions are needed to successfully increase physical activity in older adults specifically, as their motivations and capabilities may differ. One group of intervention techniques that may differ in efficacy in populations of different ages are self-regulation techniques, such as goal setting,self-monitoring, and feedback).Self-regulation involves a continual cycle of goal setting, assessing the current state of the goal, identifying the discrepancy between the current state and goal state, and modifying behaviour in an attempt to reduce this discrepancy (Carver & Scheier, 2001).Arecent systematic review foundthat interventions including self-regulation techniqueswere associated with smaller improvements in physical activityof non-clinical adult populations aged 60 years or over relative to those interventions that did not include self-regulation techniques (French, Olander, Chisholm, & McSharry, 2014).In direct contrast,aprevious review with people under 60 yearsfound interventions including self-regulatory techniques to be associated with larger increases in physical activity than those not including self-regulatory techniques(Williams & French, 2011). Importantly, these reviews used identical methods, including the same taxonomy of behaviour change techniques (BCTs; Coventry, Aberdeen & London – Refined [CALO-RE]; Michie et al., 2011), enabling direct comparison of results. This suggests that interventions designed to increase physical activity in older adults should differ from those based on younger adult populations.
It has been suggested that physical activity interventions based on self-regulatory approaches may be less effective for older adults because such approaches are less acceptable to this population (French et al., 2014). Many BCTs involving self-regulatory processes are used to help people achieve a particular level of physical activity.Older adults, however, may be less concerned about reaching goals to meetphysical activity guidelines, and more concerned with finding enjoyable activities and achieving social goals (Kassavou, Turner, Hamborg, & French, 2014).
Acceptability to the target population is a key element in the development of health-related interventions, as proposed by the UK Medical Research Council (MRC) Framework for Developing and Evaluation Complex Interventions (Campbell et al., 2000; Craig et al., 2008).As older adults’ needs and expectations concerning physical activity are likely to differ to general adult populations (e.g. they may have different motivations and capabilities than younger adults) the most acceptable approach to increase physical activity in older adults must be determined. Thereis therefore a clear rationale for conducting qualitative research to ascertain what is viewed as acceptable or unacceptable to older adults when developing interventions to increase physical activity in this population. This may include older adults’ perceptions, beliefs and attitudes regarding physical activity, in addition to their perceived barriers and facilitators to being physically active. To date, numerous individual qualitative studies have investigated such factors. However, in order to gain a more comprehensive understanding of this topic, there is a need to take into account all relevant qualitative studies, and consider the similarities and differences across the findings of each.
The meta-synthesis of qualitative data is a relatively new approach developed within healthcare research.Using this approach, researchers endeavour to yield new insights and understanding from synthesising collections of qualitative studies (Walsh & Downe, 2005), generating new theoretical perspectives of a phenomenon that are ‘greater than the sum of parts’ (Campbell et al., 2003, p672). Meta-synthesis is therefore a particularly useful method of optimising learning from existing qualitative research.
A recent meta-synthesis has identifiedolder adults’ barriers and facilitators tophysical activity (Franco et al., 2015). Six themes were identified from 132 primary studies: social influences; physical limitations; competing priorities; difficulties in access; personal benefits of physical activity; and motivation and beliefs. Thisreview provides valuable insight into specificissues relating to older adults’ engagement in physical activity. However, the analysis involved quantitative aggregation of study findings, therefore producing a largely superficial analysis. Failure to address the origin of the identified barriers, or suggest feasible methods of addressing them, means there remains a large gap in knowledge and understanding of how to better engage older adults in physical activity. Furthermore, both clinical and non-clinical populations were included in this review, thus contextual differences,such asdifferent barriers and facilitators andmotivations to be active, were overlooked.
Another meta-synthesis has recently beenpublishedinvestigating how the role of acceptability influences older adults’ engagement in physical activity interventions (Devereux-Fitzgerald, Powell, Dewhurst & French, 2016). This synthesis addressed some of the limitations of previous work through investigating the acceptability of physical activity interventions specifically targeting older adults from non-clinical populations living independently in the community. Notable findings werethat the enjoyment of social interaction was a key motivating factor for older adults to engage in physical activity and that experiencing relevant short-term benefits increased its perceived value.
The studies included in the meta-synthesis by Devereux-Fitzgerald et al. (2016) used populations of older adults who hadrecently undertaken interventions targeting increases in physical activity. Therefore, these participants were,for the most part, willing to contemplate increasing their levels of physical activity. In contrast, the current reviewseeks toexamine theperspectives of older adults who are not involved in physical activity interventions and therefore not necessarily active or willing to be active.By looking at the views of a broader population of older adults, a range of perspectives may be obtained that better represents the inactive older adult population, who may be less motivated to increase their physical activity behaviour.Such knowledge can facilitate the development of more effective interventions to increase physical activity levels of inactive older adults.
The present researchtherefore aims to systematically review the existing qualitative literature concerning how acceptable older adults find the concept of being physically active, and perform a meta-synthesis of the results across studies.
Method
Inclusion Criteria
Studies were included in which the primary focus was to explore older adults’ perceptions of performing physical activity as defined by Caspersen et al. (1985), including the reduction of sedentary behaviour. Eligible studies employed older adult participants who were all aged 65 years and above living independently in the community and were not recruited on the basis of having a particular clinical condition (e.g.diabetes). The age criterion of 65 years and above was selected as this age signifies the beginning of older adulthood according to definitions of numerous national physical activity guidelines (e.g. UK Department of Health, 2011; US Department of Health and Human Services, 2008; World Health Organization, 2010a). Studies had to be written in English, to have used qualitative methods of both data collection and data analysis, and to have been published from the year 1970 onwards. This date was chosen as a cut-off as little qualitative research concerning physical activity existed prior to 1970. Additionally, qualitative data from mixed-method studies, in which the qualitative data were analysed and reported separately to quantitative data, were included.
Exclusion Criteria
Studies were ineligible where participants were participating in a physical activity-related intervention as part of a research study, which aimed to evaluate an intervention participants had previously participated in, or in which participants were affiliated with community services or programmes related to physical activity. Studies presenting findings as percentages, frequencies of concepts, or in other quantitative formats were excluded, as were those not published in a peer-review journal.
Search Method
Search terms were mapped using the SPIDER Tool (Cooke, Smith, & Booth, 2012), shown inAppendix 1.This tool facilitates retrieval of qualitative work in systematic database searches by breaking down the research questions into five components: Sample, Phenomenon of Interest, Design, Evaluation, Research type.Searches were conducted in April2015 using the following electronic databases: PsychInfo (1806 – April Week 2 2015), Medline (1946 – April Week 2 2015), CINAHL (Cumulative Index to Nursing & Allied Health Literature; 1937 – April 20th 2015), and AMED (Allied & Complementary Medicine Database; 1985 – April 2015). The full search strategy for PsychInfo, Medline, and AMED is included as Appendix 2.
Further to the electronic database searches, forward and backward citation searches were conducted on all included papers and the reference list of an existing related review (Franco et al., 2015) was searched.
Screening
Title and abstract screening was conducted by the first author on all records to exclude any obviously irrelevant papers, with a random sample of 20% double screened by a second researcher. The chance-corrected kappa value calculated to assess inter-rater reliability was κ= 0.86, which is considered to be ‘almost perfect’ agreement according to conventional criteria (Landis & Koch, 1977).Full texts of papers included at the first stage were retrieved and screened in detail by the first author and a co-author, withκ = 0.72 indicating ‘substantial’ agreement (Landis & Koch, 1977).
Quality Appraisal
Appraising qualitative studies in a systematic review determines the trustworthiness of included papers (Dixon-Woods et al., 2004). This review utilised the Critical Appraisal Skills Programme (CASP) tool for qualitative research (Public Health Resource Unit, 2006), which addresses rigour, credibility, and relevance of qualitative work. Appraisal of studies was undertaken independently by the first author. Studies were categorised as high, medium, or low quality on the basis of their CASP rating. Four of the ten included studies were also independently rated by a co-author with prior experience of using the CASP tool, with good agreement obtained. Quality assessment ratings did not determine inclusion of studies, rather they served as a useful tool to assist the analysis (e.g. where contradictory findings could be a result of poor quality studies).
Data Extraction
Study details concerning research aims, sample size, sample ethnicity, country of study, data collection method, and type of qualitative analysis for each included article were extracted and recordedon a standard data extraction form. The findings of each included study were extracted from the text and transferred verbatim into a data extraction table. All text contained under the heading ‘findings’ or ‘results’ was considered as data, including both participant quotes and author interpretations, in addition to information regarding study findings reported in the abstract or discussion.For mixed-method studies, only qualitative findings were extracted: quantitative findings were not.
Data Synthesis
This review employed Thomas and Harden’s (2008) method of Thematic Synthesis. This method was chosen as it originated from a demand for research reviews to address the appropriateness, acceptability, and effectiveness of interventions relating to health behaviour change (Barnett-Page & Thomas, 2009).
The analysis process consisted of three main stages. First, inductive line-by-line coding was undertaken on the findings of the primary studies, whereby each line of text was assigned one or more codes that encapsulated its meaning. Codes were split or merged where concepts were identified as being semantically different or similar, as each subsequent study was coded(Appendix 3). In the second stage of analysis, codes were further refined and grouped according to conceptual similarities, producing ‘descriptive’ themes (Appendix 4). These first two stageswere inductive and provided an organised summary of the content of the primary studies.
In contrast, the third stage involved generation of new interpretative conclusions, thus taking the findings beyond the content of the primary data. This was achieved by further analysis of the descriptive themes with direct consideration to the research question,and identification of underlying processes within and between themes according to the reviewer’s own insights and judgements. Preliminary analytical themes were discussed within the research team and refined according to further insight from all reviewers to produce the final analytical themes. This enabled the data to be transformed to construct a larger narrative ofolder adults’ views of physical activity.
Results
Database searches yielded 1198 records. Figure 1 outlines the number of studies identified, excluded, and included at each stage of the screening process. Ten studies were included in the synthesis(see Table 1).Five studies were adjudged to be of high quality (Annear et al., 2009; Dye & Wilcox, 2006; Graham & Connelly, 2013; Janssen & Stube, 2014; Leavy & Åberg, 2010); three were judged as medium quality (Ceria-Ulep et al., 2011; Chastin et al., 2014; Kalavar et al., 2004); and two were deemed low quality (Aronson & Oman, 2004; Jancey et al., 2009).Many of thelower quality primary studies were vague in the methodological information they reported and did not consider the relationship between the researcher and participants. Some lacked rigour in the reporting of analysis or provided insufficient raw data to support the findings,and one study did not audiotape and transcribe interviews (Chastin et al., 2014).
Study Perspectives
Studies were conducted across seven different countries: USA(5); New Zealand; Australia; Canada; Scotland; Republic of Ireland; and Sweden. Two studies included participants from specific ethnic groups, onewith people of Asian Indian origin living in the USA (Kalavar et al., 2004) and one with Filipinos living in the USA (Ceria-Ulep et al., 2011). Just one study explored differences between high and low socio-economic neighbourhoods (Annear et al., 2009). The primary focus of all studies was older adults’ perceptions of physical activity, with one exception that investigated determinants of and motivations to reduce sedentary behaviour in older adults (Chastin et al., 2014).
Synthesis Findings
Seven descriptive themes were identified: personal motivations for physical activity; intrapersonal constraints for physical activity; perceptions of ageing; provision; external sources of encouragement; knowledge and beliefs about physical activity;and influence of environmental factors. From the seven descriptive themes, three analytical themes were derived: Older Adults’ Construal of Physical Activity; Self-Identity and Roles within the Wider Society; and Perceived Vulnerability vs. Maintaining Control. These analytical themesconstitute the findings presented in this paper. Appendix 4 shows the relationships between each analytic theme and the descriptive themes and sub-themes from which they are derived. In the following discussion of results, author quotes from the primary studies are denoted by use of italics, and participant quotes are indicated by quotation marks. Use of both italics and quotation marks represents a participant quote within an author quote.