Background
An active lifestyle provides powerful benefits in the general health and wellbeing of the older adults aged 65 years and above. An adequate level of physical activity, combining aerobic activity, muscle-strength and balance training, improves the overall physical function, delaying functional decline and supporting independent living. To be physically active includes, but it is not limited to, participate in structured physical exercise. It also means to be active throughout the day, for example, by avoiding long periods of inactivity. Low-intensity walking activity is also suggested to be associated with better health [
1]. The ‘Global Recommendations on Physical Activity for Health’ from the World Health Organization, highlight the importance of daily activities, such as household chores, games and transportation [
2]. At the same time, older adults should maintain an active lifestyle also in terms of being engaged with their community and environment [
3]. Some older adults can achieve an active lifestyle by themselves; however, others might benefit from an external nudge to become active [
4]. Technology can play an important role here, by incorporating several strategies to support people in being physically active and actively engaged in their social environments.
The research presented in this manuscript is the first step towards the development of a tailored approach to promote physical activity and mental wellbeing in the daily lives of the older adults, through the recommendation of pleasurable, or enjoyable, activities. However, before designing an intervention, it is important to deepen the current knowledge on the interactions between physical activity, positive emotions and daily activities, being this the focus of the presented study. There are several reasons for this. First, participation in pleasurable activities in older age is associated with better physiological function and better sleep [
5], improving general health and wellbeing. Second, there is growing evidence supporting the link between positive emotions and general health [
6‐
10]. Third, according to the Broaden-and-Build Theory, those who experience higher levels of positive emotions are more likely to build a variety of resources, such as environmental mastery and social support [
11]. These resources improve resilience to change, a very important characteristic for this population, as biological and social changes are likely to occur. Furthermore, the ‘upward spiral of lifestyle changes’ [
12], a derivation from the previous theory, defends that positive emotions act as openers for acceptance and adoption of new behaviours, key characteristics when promoting behaviour change. Finally, according to the Self-Determination Theory, the enjoyment or pleasure experienced is an intrinsic motivator to repeat a certain activity [
13], also already evaluated in the physical activity context [
14]. One can thus hypothesize that people are more inclined to do what they like to do. This means that, when suggested an activity that they are familiar with, and have previous positive experiences with, they are more likely to follow the recommendation.
When aiming at increasing physical activity through promotion of pleasurable activity, it is important to take a deep look at the individual daily life contexts, and their impact on physical activity and emotional experience while performing the regular daily activities. Gaining this insight is only possible by looking at the routine over several weeks. Mobile technology provides the means to gather real-time information in daily life for long periods of time [
15]. Accelerometers and experience sampling have been successfully used in the past to evaluate the contexts of sedentary behaviour among older adults [
16], in which most of the sedentary time was performed within the home environment and alone. Also, the influence of contexts on positive affect during physical activity has been evaluated using a similar method with the adult population in which it was reported that social activities might enhance positive emotions while performing physical activity [
17]. However, none of the studies above mentioned looked at several weeks period within individual. Our study intends to extend the previous studies by investigating: (1)
how do daily activities relate to physical activity, (2)
how do daily activities relate to the experience of pleasure, and (3)
how do daily activities influence the relation between physical activity and pleasure in the daily lives of the older adults. Combining the information gathered on the influence of daily activities on physical activity and experience of pleasure in daily life, we can further work on developing technology-ased interventions that will support older adults becoming more active through the promotion of pleasurable activities. Moreover, this research contributes to the understanding of emotions in the daily lives of the older population.
Research Hypothesis
We establish that daily activities are defined by five categories: location (where the person is), activity (what the person is doing), social companion (with whom is the person doing the activity), emotional experience (how is the person feeling) and physical activity (amount of movement performed). Similar categorization is adopted in other studies relating daily context information to physical activity in daily life (e.g. [
16,
17]). We have four hypotheses that are investigated in the current study.
-
H1. Social activities are more pleasurable than activities performed alone. The social environment plays a clear role on the wellbeing of the population. Participation in social activities has constantly been associated with higher experience of general wellbeing [
18] and positive emotions [
19‐
22].
-
H2. Outdoors activities are more pleasurable than activities performed at home. Previous research suggests that outdoor activities are associated with higher levels of positive emotions in the older population [
19,
23,
24]. However, Gagliardi et al., when comparing German and Italian population, found a significant relation in the German population but not in the Italian, suggesting that there are cultural differences [
19].
-
H3. Leisure activities are more pleasurable than basic activities of daily living. Time spent in recreational/hobby activities is a predictor of daily mood independently on level of cognitive impairment [
23]. Participation in leisure activities increases overall levels of positive emotions [
20] and delays functional decline [
25].
-
H4. There is no relation between physical activity and pleasure. There is solid evidence for the benefits of exercise programs on mental health [
26,
27]. However, the relation between physical activity and positive emotions is much less explored. There is some small evidence for relation between positive emotions and physical activity [
28] but further work needs to be done to understand the mechanisms influencing this relation. We hypothesize that there is no relation because we choose an emotion, pleasure, that is not,
per se, associated with high or low arousal emotions. Pleasure can be experienced with very relaxing activities but also with very exciting ones.
Discussion
The objectives of this study were to perform an exploratory investigation on (1) how daily activities relate to physical activity, (2) how daily activities relate to the experience of pleasure, and (3) to which extent these daily activities explain the interaction between pleasure and physical activity, in the daily lives of community-dwelling older adults. Data was collected with an accelerometer and experience sampling on a smartphone. No feedback was provided to the participants during the study. The results of our study confirm the first three hypothesis which stated that social activities, outdoor activities and leisure activities are more pleasurable than activities performed alone, indoor activities and basic activities of daily living, respectively. However, the last hypothesis, stating that there is no relation between physical activity and pleasure, is partially rejected. This result suggest that the type of daily activity as a moderator effect in the relation between physical activity and pleasure.
This research was designed to gather information for the design of strategies to promote physical activity through recommendation of pleasurable activities. Preliminary recommendations for technology development can be drawn from the presented study.
First, our results support the idea of tailoring interventions when promoting pleasurable activities. Although general effects can be taken from the full sample, looking at the results of the within-individual analysis, we see that, as expected, the predictors of pleasure are highly personal. By gathering data over approximately 30 consecutive days (ranging from 24 to 38, according to the availability of each individual), we can conclude that there are individual differences. Taking into account the market growth of mobile devices to monitor physical activity, also outside the scientific domain, in the future, individual differences might be detected automatically using data mining techniques. Interventions can thus be tailored to the preferences and needs of the individuals, even in the cases that the preferences change over time.
Second, older adults spend most of their time at home and alone. This fact is certainly not surprising, but the proportions are, by the fact that our sample was relatively healthy, and active in the community, representing, what we named as, the role models. The study took place during Winter time, and therefore, people are more likely to spend time at home. However, it is still remarkable that, for example, in one of the subjects, 88% of the activities reported during 1 month took place at home and 96% alone. The World Health Organization emphasizes the importance of being engaged in the community and environments for a healthy lifestyle [
3]. Interventions that stimulate social inclusion of older adults are highly recommended as well as interventions that coach the individual to go outdoors, as the home setting is where the older adults spend most of their sedentary time [
16].
Third, motivation of physical activity
by proxy is recommended based on our results, expanding the results from [
28]. By motivation by proxy, a coaching strategy that motivates people to engage in outdoor- or social activities, increasing physical activity indirectly. For example, instead of recommend an individual to go for a walk, one can inform about a new exhibition in the local museum. By going to this exhibition, the individual needs to move. Combining the results from objective 1 and 2, we see that promotion of outdoors activities are the most valuable considering that these activities result both in higher experience of pleasure and more physical activity. Promotion of leisure activities is also highly recommended, as the experience of pleasure increases with physical activity when individuals are engaged in leisure activities, but not in bADL.
Post-hoc analysis suggests that this effect in bADL is due to household activities. It is known that household activities are a source of physical activity in the daily living [
38], however, our study suggests that it is not the most pleasurable one, and therefore, not likely to be a good motivator to perform physical activity.
Fourth, we can see that 30 days of measurement, with approximately six events per day, generates enough data to analyse the influence of daily environments on the experience of pleasure. However, this is a very demanding procedure which should be reduced in the future. Future research could investigate whether it is possible to obtain the same degree of information with a shorter study. Participants of the study reported that answering questions every hour for 1 month is an annoying task. However, identifying what is pleasurable for each individual without becoming cumbersome remains challenging. New technological developments such as emotion recognition tools (either using facial expression recognition or bio sensing) might provide the means for less obtrusive research in this area. Automatic assessments of emotions assume even higher importance when seeing that the interpretation of the feeling pleasure and quantification in a scale is highly personal. While some subjects made use of the full scale (0 to 10) others limited themselves to a short range. This might have to do with personality or with other factors. This means that looking at the exact place of the VAS chosen is not a good measure. Instead, in our data analysis we normalized the values to correspond to a deviance from the median. Rocke et al. reported that older adults report low variability rates of positive affect when compared to younger adults [
39]. The use of the hip-worn accelerometer can also become obtrusive. This is likely to be overcome rapidly with the consumer oriented lifestyle devices to promote physical activity becoming smaller, and being used by more people every year. Objective monitoring of physical activity of the older population might be a promising addition to conventional questionnaires, as there is evidence that the objective measurement provides more reliable information [
40].
To the best of our knowledge, our exploratory study is innovative for the variety of data gathered, and the combination of methods used to gather information during the course of 1 month among the older population. From each participant, we obtained health related information, lifestyle behaviours, emotional- and context-information for a period of approximately 30 days. Furthermore, we made use of three distinct data acquisition methods: conventional standardized questionnaires, on-body sensing and experience sampling. These factors combined provide very valuable knowledge, because, contrarily to most of the studies developed in the past, our data was acquired in real-time, in the daily life of the participants, instead of using a questionnaire that asks previous experiences. We believe that only in this way one can get a reliable view of the daily behaviours. Further research should be performed by analysing more distinct categories of the properties of daily environments, instead of dichotomous variables. For example, in terms of social companion, one could look at how the experience of pleasure and physical activity are influenced by the fact that an activity is performed with the partner, relatives or even specific friends. Future research could also look separately at routine and non-routine activities. Bouisson & Swendsen suggest that breaks in the routine improve the wellbeing of the older adults, even the ones who claim that they prefer routine [
41]. Finally, we encourage researchers to verify if tailored interventions for promotion of physical activity based on the preferences and enjoyment of the individuals do increase adherence, as suggested by the Self-Determination Theory [
13].
Our study has limitations. First, although the subjects were told to adapt the measurement period to their own routine, the battery of the phone and the accelerometer limited the measurement period to a maximum of 12 h. Considering that mostly the subjects would start measuring in the morning, the evening period was not considered in our studies. Assuming that in the evening period, people are more likely to do relaxation activities within the home environment, this means that people spend more time at home than reported. Secondly, the visual analogue scale was experienced as being difficult to use, as the interpretation of the scale differed per participant, as well as where to locate the finger in the scale. For future studies we suggest the use of Likert scales instead of visual analogue scales. Still regarding the outcome variable, in this study we were only interested in investigating the experience of positive emotions. To avoid over complicate the study, we have only looked at the valence dimension (positive axis “Pleasant”) and did look at different experiences of the arousal, or activation, dimension. We recommend researchers to look at the several discrete emotions corresponding to different arousals, from deactivation to activation, following the circumplex models of emotion [
42,
43]. Regarding the data analysis there are two points for discussion. First, we reduced the granularity of the data grouping variables in dichotomous variables. Although this grouping provides already very interesting results, further analysis should be performed with the original categories. Second, in the calculation of the physical activity related to a daily activity, we considered a time window of 10 min centred in the moment of answering the question. This is likely to be arguable, however, our sensitivity analysis with several lengths and time shifts did not show any significant difference. Finally, there is clear the issue of the small sample size. However, as stated in our objectives, we were aiming at getting deep insight in within person analysis, and that objective was met. Further studies should be performed with a larger sample and look at possible predictors of pleasure and physical activity, such as depression or years after retirement.