An ideal physical activity app for adults – what should it be like? Requirements of adult users Kerli Mooses Institute of Computer Sciences University of Tartu Tartu, Estonia Kerli.Mooses@ut.ee Abstract—Despite the numerous health benefits provided by process. Early identification of the needs, preferences and physical activity, people struggle to meet healthy activity levels. problems of end users helps to address them during early There are numerous apps aiming to support the physical activity development phase and thereby ensure the acceptance and of adults. However, it is unclear whether and at what extent is usability of the end product. The findings about the features the end user been involved in the development process. In that adults would like to have in a physical activity app are addition, there is a lack of research which examines which functional, quality and emotional goals are preferred and sparse. According to a qualitative study [12] young adults expected by the user of physical activity app. To address this prefer physical activity apps that coach and motivate them, issue, do-be-feel workshops were carried out with 38 adults and provide tailored feedback, enable to set personal goals and the results were combined into goal model. These results offer a allow competition with friends. User-friendliness and good overview of the expectations of end user which can be used flexibility has also been reported [8]. Most published data in further development. focus on describing existing apps, trying to identify most effective and/or most frequently used features [9], [11], [13]– Keywords—requirements elicitation, physical activity, goal [15]. For example, it has been found that giving feedback, modelling, adults setting goals, receiving rewards, social sharing and I. INTRODUCTION competition can positively influence the physical activity of adults [13]. These studies mostly focus on functional The positive impact of physical activity on mental, social requirements, but in addition to functional requirements it is and physical health are widely recognized [1], [2]. Countries important to consider the quality and emotional requirements make efforts to support and promote healthy physical activity which also influence the acceptance and usability of end levels among their citizens. It has been shown that if all product [16]. Therefore, knowledge of quality and emotional inactive people became active, more than 5 million deaths per requirements needed by the end user can increase the year could be avoided [2]. Recent research indicates that effectiveness and attractiveness of the app. every move counts as health benefits occur even when The aim of current study is to find out the needs and physical activity is below the international recommended expectations of adults for an app that would support their levels [1], [3]. At the same time, higher physical activity physical activity. In addition, we combine the elicited levels and reduced sedentary time substantially reduces the requirements into hierarchical goal model to give a good risk for premature mortality [4]. Despite the facts and visual overview of gathered requirements. research reassuring the positive influence of physical activity on health, a lot of people struggle to achieve healthy II. METHODS physically active levels as third of adult population do not To identify the functional, quality and emotional goals of meet the physical activity recommendations by World Health an app that would support the physical activity of adults, nine Organisation [5]. According to the literature reviews, the online workshops with a total of 38 adults (22 female, 16 most common barriers for physical activity reported by adults male, aged 35-45) were conducted. The participants were are lack of time, motivation and self-discipline [6], [7]. This recruited through the requirements elicitation course of the indicates that a tool supporting the motivation and self- university. The workshops were carried out online due to the discipline could have beneficial effects on physical activity COVID-19 situation and to reduce the personal contact. The levels of adults. do-be-feel method was used in the workshop. At first the The rapid advantages in information and communication participants were given a following description: “It is advised technologies (ICTs) and their extensive use worldwide make by World Health Organisation that adults should accumulate ICTs valuable tools in health promotion, including 10 000 steps per day or 150 minutes of moderate activity per supporting physical activity. There are numerous apps week in order to stay healthy. If there would be an app that available in app stores for physical activity. However, it has helps you to fill these activity goals, what would it be like?” been shown, that their effect on physical activity behaviour is The participants were asked to think on four aspects: 1) what modest [8]–[10]. It could be partly explained by the fact that should such app do? (functional goals), 2) what should such the involvement of end user during the development process app be like? (quality goals), 3) how should such app make seems to be modest or even non-existing as no physical you feel? (emotional goals). In addition, the participants were activity app has reported users involvement during the asked to think which roles are associated with the app. The development process [11]. In addition, to the best of our answers were all gathered in one table where participant knowledge there is no study describing the involvement of could add additional ideas. The results from nine workshops the end user during the requirements elicitation phase of were divided into subthemes and combined into one summary physical activity app. At the same time end user is a crucial table presented in Results section. The frequency of goals stakeholder who should be involved in the development Copyright © 2022 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0) mentioned by different workshop groups was also added to rate, time, velocity, distance and activity minutes. In addition the table to identify the most popular needs. to the possibility that the app itself tracked the activity, it was In order to give a good visual and systematic overview of important that activity data could be imported from watches gathered requirements agent-oriented goal model [17]–[21] and activity trackers. As for the reports, different input was created. In this hierarchical goal tree, the functional goals variables, such as activity minutes, steps, activity types, and are presented in tilted rectangles. Roles, quality and emotional time periods, such as day, week, month, was mentioned. The goals are attached to the functional goal that they are goal setting was considered an important feature. All groups associated with. The roles are presented as stick man, quality wanted to set personal goals and some groups preferred the goals with clouds and emotional goals with heart symbols. app to make suggestions for next goals based on previous Goal modelling was used as it presents early functional, goals and activity. Together with goal setting, the participants quality and emotional requirements in a holistic and coherent wanted to get notification and praise for meeting the goal. way, which can be further developed into user stories [22]. Reminders, motivating tips and messages were considered as User stories in turn are often used in agile software engineering [23], [24]. In addition, it has been shown that goal encouraging and motivating which help them to meet goals models are also helpful when there is a need to involve or and overcome barriers. The participants also considered communicate with nontechnical stakeholders or experts [25]. notifications that would remind activity in case of prolonged passive periods useful. Another popular features were III. RESULTS possibility to participate in challenges and share and compare Measuring activity and making reports were two your progress with others within app or in social media. functional goals that were mentioned by all groups (Table 1). Participants were interested the app to suggest nearby tracks A variety of indicators for measuring physical activity by the for different activities, such as hiking, orienteering, running app was mentioned – for example, the number of steps, heart etc. TABLE I. FUNCTIONAL, QUALITY AND EMOTIONAL REQUIREMENTS IDENTIFIED WITH DO-BE-FEEL WORKSHOPS To-do To-be To-feel Roles Measure activity Safe (7) Healthy (9) User (9) Track activity (9) Intuitive (6) Satisfied (7) Friend (9) Use GPS (6) Modern (6) Motivated (6) Sponsor (5) Import data from other apps/ watches/ activity trackers (5) Simple (5) Happy (6) Doctor (3) Insert activity manually User friendly (5) Inspired (4) Scientist, Health expert (2) Make adjustments to activity data Precise, accurate (5) Self-confident (4) Athlete (2) Goals Reliable (5) Aware (4) Sport psychologist Set personal goals (5) Compatible (4) Proud (3) Sport club Suggests new goals based on previous activity (3) Motivating (3) Victorious (3) Coach Feedback Free (3) Proud (3) Make reports (9) Reasonably priced (3) Informed (2) Get points for activity / virtual medals / awards (5) Energy-saving (3) Addicted (2) Visualise the distance from goal (4) Inspiring (2) Supported (2) Present distance virtually (e.g. home to Rome/ move around a globe) Playful (2) Recognised (2) Get personal advice from coach Adaptive (2) Cared for (2) Reminders and notifications Comfortable (2) Positive (2) Get notification when filling a goal (7) Fast (2) In control Get notifications when passive (6) Spectacular Hopeful Get reminders, motivating messages and tips (4) Innovative Success Turn off notifications and reminders Unnoticed Amused Challenges Practical Special Participate in challenges (5) Interesting Belong to Win awards from sponsors (2) Realistic Active Create challenges (2) Adds-free Excited Invite friends to join a challenge Energetic Create leader boards based on activity type Guided Socialize Share and compare results with others (5) Invite friend(s) to training session (3) Invite friend to join the app (2) Find activity Save and rate track (with GPS) (4) Show/suggest tracks nearby (4) Add training sessions to calendar (2) Other Insert personal data (5) Call help Keep a creature alive with your activity Read motivating info about the benefits of physical activity () – number of groups who mentioned this requirement. Presented if N > 1. Fig. 1. Goal model for app supporting the physical activity levels of adults based on most frequently mentioned requirements Safety, simplicity, user-friendliness, accurate rewards and enabling social comparison [9], [13], [14]. measurements and reliability were most frequently Although rewards have a potential to support engagement in mentioned quality goals. As for emotional goal, the physical activity, they have to be aimed at intrinsic participants wanted to feel healthy, satisfied, motivated and motivation, which supports long term changes [28]. Caution happy when using the app. should be taken when incorporating social comparison The gathered requirements were transformed into goal feature to the app as the scientific findings are contradictory models which give a good visual of the most popular needs – some people find motivating comparing with more active of adults who want to be physically active with the help of an people, whereas the others demotivating [12]. app (Fig. 1). Our participants also pointed out the need to invite a friend to training session and share your results with friends IV. DISCUSSION or in social media. Social support and enjoyment have been There is a limited evidence in scientific literature about reported as one of the main motives for physical activity [6] the needs and expectations of adults who use physical activity and being able to train with a friend will certainly support app. In current study we addressed this knowledge gap and these motives. Being active with friends helps to fulfil identified functional, quality and emotional requirements for abundance of emotional goals which emerged from our such app with the help of do-be-feel workshops. workshops. For example, supported, cared for and recognised Most common behaviour change techniques such as are only few of them. measuring activity, setting goals and getting feedback [11], Possibility to create challenges and participate in them [14], [15] were mentioned by most of our workshop groups. was favoured by half of our workshop groups. Similarly, Being aware of one’s activity levels can be considered a participating in challenges has been previously reported by prerequisite for evaluating the sufficiency of physical young adults [12]. However, not all people are competitive activity. In addition, it helps to increase the awareness of and interested in challenges [13], thus this feature should be actual activity levels and review the changes over time [26]. kept optional. According to studies, these three features have a potential to Interestingly a possibility to find activity has not been increase the effectiveness of the app [27]. For our participants mentioned in previous studies with adults or in existing apps, it was equally important to use the app for performing activity whereas this feature was reported by almost half of the measurements as well as importing the data from watch or workshop groups. Previously a shortage of facilities has been activity tracker. As different sporting disciplines can make reported as one of the barriers for physical activity [6]. In restrictions on the measurement device, it would be useful to some cases, there could be a lack of information about ensure compatibility and connectivity with a variety of different opportunities for in- and outdoor physical activity. devices. As for feedback, only one group mentioned the In such cases, including a possibility to search or get possibility to consult a coach although personalised feedback suggestions about nearby tracks and facilities can help to has been identified one factor that encourages PA change [9]. overcome this barrier. This feature would also be beneficial Our participants preferred to get feedback and view progress in a case of a holiday or a trip to a new and foreign over time with graphical and numeric reports. surrounding. Another quite popular behaviour change techniques that Safety was the most highly valued quality goal by our are used in existing physical activity apps and which were participants. Participants wanted to share their personal data mentioned several times by our participants, were providing as it enhances user experience and improves personalisation. As the privacy and security of existing physical activity apps [2] I. M. Lee, E. J. Shiroma, F. Lobelo, P. Puska, S. N. Blair, and P. T. Katzmarzyk, “Effect of physical inactivity on major non- has been rated as low [11], [29], developers should pay extra communicable diseases worldwide: an analysis of burden of disease attention to meet this requirement. and life expectancy,” Lancet, vol. 380, 2012, doi: 10.1016/S0140- Motives for being physically active are health benefits, 6736(12)61031-9. social support [6], [7] and enjoyment [6]. These are in line [3] J. M. Jakicic et al., “Association between Bout Duration of Physical Activity and Health: Systematic Review,” Med. Sci. Sports with emotional goals reported by our participants. Exerc., vol. 51, no. 6, pp. 1213–1219, Jun. 2019, doi: A. Limitations 10.1249/MSS.0000000000001933. [4] U. Ekelund et al., “Dose-response associations between One limitation of our requirements elicitation method is accelerometry measured physical activity and sedentary time and that the workshops were online which could have hindered all cause mortality: systematic review and harmonised meta- some co-creation that would have emerged during face-to- analysis,” BMJ, vol. 366, pp. l4570–l4570, Aug. 2019, doi: 10.1136/bmj.l4570. face meetings. However, we believe this effect was minimal [5] R. Guthold, G. A. Stevens, L. M. Riley, and F. C. Bull, “Worldwide as all participants participated actively during the workshops trends in insufficient physical activity from 2001 to 2016: a pooled and due to the prolongation of COVID-19 restrictions people analysis of 358 population-based surveys with 1·9 million have become more familiar, experienced and comfortable participants,” Lancet Glob. Health, vol. 6, no. 10, pp. e1077–e1086, Oct. 2018, doi: 10.1016/S2214-109X(18)30357-7. with online meetings. [6] M. R. L. Pedersen, A. F. Hansen, and K. Elmose-Østerlund, Another limitation of our model is, that it presents the “Motives and Barriers Related to Physical Activity and Sport across viewpoint of one stakeholder group – the end user. Inclusion Social Backgrounds: Implications for Health Promotion,” Int. J. of health and behaviour experts could also be beneficial to Environ. Res. Public. Health, vol. 18, no. 11, p. 5810, May 2021, doi: 10.3390/ijerph18115810. ensure the correctness and safety of the content, and long- [7] A. Baillot et al., “Physical activity motives, barriers, and term usage of the app which in turn supports healthy and preferences in people with obesity: A systematic review,” PloS lifelong physical activity. Despite the benefits of experts One, vol. 16, no. 6, pp. e0253114–e0253114, Jun. 2021, doi: involvement, experts are rarely involved during the 10.1371/journal.pone.0253114. [8] S. S. Coughlin, M. Whitehead, J. Q. Sheats, J. Mastromonico, and development process [11], [14], [29], [30]. We have S. Smith, “A review of smartphone applications for promoting previously created goal models for supporting the physical physical activity,” Jacobs J. Community Med., vol. 2, p. 021, 2016. activity of adolescents [25]. These goal models were based [9] J. Bort-Roig, N. D. Gilson, A. Puig-Ribera, R. S. Contreras, and S. on the literature review and inclusion of health experts and G. Trost, “Measuring and Influencing Physical Activity with Smartphone Technology: A Systematic Review,” Sports Med., vol. overall resulted in similar themes as in current study with 44, no. 5, pp. 671–686, May 2014, doi: 10.1007/s40279-014-0142- adults. Therefore, we could argue that in an early 5. requirements elicitation phase the involvement of experts [10] R. Gal, A. M. May, E. J. van Overmeeren, M. Simons, and E. M. might not be that urgent. At the same time, we agree that Monninkhof, “The effect of physical activity interventions comprising wearables and smartphone applications on physical behavioural and physical activity experts should be involved activity: a systematic review and meta-analysis,” Sports Med. in later development phases and during validation of the early Open, vol. 4, p. 42, 2018, doi: 10.1186/s40798-018-0157-9. requirements in order to create an evidence-informed mobile [11] S. Paganini et al., “Quality of Physical Activity Apps: Systematic app [30]. Such expert help to ensure that suggestions and Search in App Stores and Content Analysis,” JMIR MHealth UHealth, vol. 9, no. 6, pp. e22587–e22587, Jun. 2021, doi: information provided to the users is healthy and evidence- 10.2196/22587. based and presented in a way that they support intrinsic [12] A. Middelweerd et al., “What features do Dutch university students motivation. This in turn supports the long-term behavior prefer in a smartphone application for promotion of physical change [28]. activity? A qualitative approach,” Int. J. Behav. Nutr. Phys. Act., vol. 12, p. 31, 2015, doi: 10.1186/s12966-015-0189-1. When interpreting the results, it should be kept in mind [13] M. Hosseinpour and R. Terlutter, “Your personal motivator is with that participants aged 35-45 were involved in the workshop you: A systematic review of mobile phone applications aiming at and their present physical activity levels are unknown. increasing physical activity,” Sports Med., vol. 49, no. 9, pp. 1425– Whether and to what extent created goal model apply to 1447, 2019, doi: 10.1007/s40279-019-01128-3. [14] P. Simões, A. G. Silva, J. Amaral, A. Queirós, N. P. Rocha, and M. younger or older end users warrants further investigation. Rodrigues, “Features, Behavioral Change Techniques, and Quality of the Most Popular Mobile Apps to Measure Physical Activity: V. CONCLUSION Systematic Search in App Stores,” JMIR MHealth UHealth, vol. 6, no. 10, pp. e11281–e11281, Oct. 2018, doi: 10.2196/11281. Do-be-feel workshops conducted with adults helped to [15] A. Middelweerd, J. S. Mollee, C. N. van der Wal, J. Brug, and S. J. discover the needs and preferences that adults have for Te Velde, “Apps to promote physical activity among adults: a physical activity app. These findings are valuable input for review and content analysis,” Int. J. Behav. Nutr. Phys. Act., vol. developers who want to improve their existing app or develop 11, pp. 97–97, Jul. 2014, doi: 10.1186/s12966-014-0097-9. [16] A. Mendoza, T. Miller, S. Pedell, and L. Sterling, “The role of a new one. users’ emotions and associated quality goals on appropriation of systems: two case studies,” ACIS 2013 Proc., p. 57, 2013. [17] K. Taveter, L. Sterling, S. Pedell, R. Burrows, and E. M. Taveter, ACKNOWLEDGMENT “A method for eliciting and representing emotional requirements: two case studies in e-Healthcare,” in IEEE 27th International We thank all the participants of workshops for their Requirements Engineering Conference Workshops (REW), Sep. valuable input and time. 2019, pp. 100–105. doi: 10.1109/REW.2019.00021. [18] L. Sterling and K. Taveter, The art of agent-oriented modelling. REFERENCES MIT Press, 2009. [1] F. C. Bull et al., “World Health Organization 2020 guidelines on [19] A. L. Lorca, R. Burrows, and L. Sterling, “Teaching motivational physical activity and sedentary behaviour,” Br. J. Sports Med., vol. models in agile requirements engineering,” in IEEE 8th 54, no. 24, pp. 1451–1462, Dec. 2020, doi: 10.1136/bjsports-2020- International Workshop on Requirements Engineering Education 102955. and Training (REET), Aug. 2018, pp. 30–39. doi: 10.1109/REET.2018.00010. [20] T. Miller, S. Pedell, L. Sterling, F. Vetere, and S. Howard, Med. Internet Res., vol. 23, no. 5, pp. e24810–e24810, May 2021, “Understanding socially oriented roles and goals through doi: 10.2196/24810. motivational modelling,” J. Syst. Softw., vol. 85, no. 9, pp. 2160– [26] L. Dennison, L. Morrison, G. Conway, and L. Yardley, 2170, 2012, doi: 10.1016/j.jss.2012.04.049. “Opportunities and challenges for smartphone applications in [21] T. Miller, S. Pedell, A. A. Lopez-Lorca, A. Mendoza, L. Sterling, supporting health behavior change: qualitative study,” J. Med. and A. Keirnan, “Emotion-led modelling for people-oriented Internet Res., vol. 15, no. 4, p. e86, 2013, doi: 10.2196/jmir.2583. requirements engineering: The case study of emergency systems,” [27] J. M. Murray, S. F. Brennan, D. P. French, C. C. Patterson, F. Kee, J. Syst. Softw., vol. 105, pp. 54–71, 2015, doi: and R. F. Hunter, “Effectiveness of physical activity interventions 10.1016/j.jss.2015.03.044. in achieving behaviour change maintenance in young and middle [22] T. Tenso, A. H. Norta, H. Rootsi, K. Taveter, and I. Vorontsova, aged adults: A systematic review and meta-analysis,” Soc. Sci. “Enhancing requirements engineering in agile methodologies by Med., vol. 192, pp. 125–133, 2017, doi: agent-oriented goal models: two empirical case studies,” in 2017 10.1016/j.socscimed.2017.09.021. IEEE 25th International Requirements Engineering Conference [28] S. J. G. Ahn, K. Johnsen, and C. Ball, “Points-based reward Workshops (REW), Sep. 2017, pp. 268–275. doi: systems in gamification impact children’s physical activity 10.1109/REW.2017.24. strategies and psychological needs,” Health Educ Behav, vol. 46, [23] M. Cohn, User stories applied: For agile software development. no. 3, pp. 417–425, 2019, doi: 10.1177/1090198118818241. Addison-Wesley Professional, 2004. [29] P. Bondaronek, G. Alkhaldi, A. Slee, F. L. Hamilton, and E. [24] F. Paetsch, A. Eberlein, and F. Maurer, “Requirements engineering Murray, “Quality of Publicly Available Physical Activity Apps: and agile software development,” in WET ICE 2003. Proceedings. Review and Content Analysis,” JMIR MHealth UHealth, vol. 6, no. Twelfth IEEE International Workshops on Enabling Technologies: 3, pp. e53–e53, Mar. 2018, doi: 10.2196/mhealth.9069. Infrastructure for Collaborative Enterprises, 2003., 2003, pp. 308– [30] E. Knight, M. I. Stuckey, H. Prapavessis, and R. J. Petrella, “Public 313. doi: 10.1109/ENABL.2003.1231428. health guidelines for physical activity: is there an app for that? A [25] K. Mooses and K. Taveter, “Agent-Oriented Goal Models in review of android and apple app stores,” JMIR MHealth UHealth, Developing Information Systems Supporting Physical Activity vol. 3, no. 2, pp. e43–e43, May 2015, doi: 10.2196/mhealth.4003. Among Adolescents: Literature Review and Expert Interviews,” J.