Abstract—Technology supported services for achieving a healthy lifestyle have shown their short term effects and are receiving increasing interest from the research community. However, long term adherence to these services is poor. This paper describes research-in-progress regarding the implementation of automated goal-setting and tailored feedback messages into one such technology supported service, which aims to improve the user’s physical activity pattern. Tailored feedback messages for several personas were set up based on theories from behavioral science and categorized by experts during an expert workshop. Results indicate reasonable agreement on the matching of motivational messages to four personas. Additional expert input is discussed descriptively. Future research will focus on examining the effectiveness of the new version of the service under investigation. Index Terms—Accelerometers, Behavioral science, Physical activity, Telemedicine. I. INTRODUCTION EOPLE live an increasingly sedentary lifestyle, resulting in a decrease in health and posing a risk for various diseases (e.g. [1]). On the other hand, a physically active lifestyle has significant positive effects on prevention of chronic diseases, such as cardiovascular disease, diabetes and cancer. Also, a sufficient level of physical activity has positive effects on mental health condition through reduced perceived stress and lower levels of burnout, depression and anxiety [2]. Therefore, a physically active lifestyle may lead to less hospitalization, higher life expectancy and improved well- being in general. Currently, many applications that support people in achieving an active lifestyle are available. One such example is the Ambulant Activity Feedback System (AAFS) [3]. The AAFS measures the activity of users during everyday life using an accelerometer-based sensor. Data is sent from the sensor to a smartphone, which displays the information to its user. The user’s cumulative activity is plotted in a graph that also shows a goal line, based on the average activity of a This publication was supported by the Dutch national program COMMIT (project P7 SWELL). R. Achterkamp is with Roessingh Research and Development, Enschede, OV 7522AH, The Netherlands and the Faculty of Electrical Engineering, Mathematics and Computer Science, University of Twente, Enschede, OV 7522NB, The Netherlands (e-mail: r.achterkamp@rrd.nl). M. Cabrita is with Roessingh Research and Development and with the Physics Department, Faculty of Science and Technology, Universidade Nova de Lisboa, Lisbon, Portugal. H. op den Akker, H. J. Hermens and M. M. R. Vollenbroek-Hutten are with Roessingh Research and Development and the Faculty of Electrical Engineering, Mathematics and Computer Science, University of Twente. group of healthy control subjects. The system is able to coach the user. Based on the percentage of deviation from the goal line, the system provides motivational cues on whether the user needs to become more active (e.g. take a short walk) or take a break, in order to achieve a balanced, daily activity goal. Previous research indicated the potential of this system [3]. However, it also showed that the adherence to the system dropped substantially after a few weeks. It is expected that this lack of adherence to the system and decrease of effectiveness can be overcome by adding 1) context aware goal setting and 2) personalization of information, i.e. tailoring [4]. II. BACKGROUND A. Context aware goal setting When Locke and Latham first proposed the Goal-setting Theory [5], it mainly focused on questions regarding motivation in work settings. However, its promising results made it one of the most commonly used theories to promote a healthy lifestyle. According to the Goal-Setting Theory, people are more likely to change behavior the higher the specificity and (achievable) difficulty of a goal. However, one should always bear in mind personal characteristics of the subject, such as goal importance, self-efficacy and feedback. Research regarding physical activity interventions shows that combining goal-setting and persuasive technologies can significantly improve the results of interventions [6]. When setting goals, baseline level of physical activity of the user should be assessed first, based on which a personal goal can then be set. Setting a goal based on the average activity of a group of healthy controls will in many cases lead to users not reaching their goal. Furthermore, considering the applicability of technology supported services to various domains, patients, for example, would struggle to keep up, possibly never reach their goal and simply give up. Therefore, we recommend to automatically adjust the height of a goal and set it slightly higher than the personal baseline. Another important aspect, next to height of the goal, is the physical activity pattern throughout the day. Research into the physical activity pattern of chronic low back pain patients [7] and patients suffering from chronic fatigue [8] shows that these patients are unable to balance their physical activity pattern throughout the day. Our solution is to incorporate context aware automated goal setting; enabling the technology supported service to automatically detect imbalances in the user’s physical activity pattern, set goals and continuously keep these goals up to date. Promoting a Healthy Lifestyle: Towards an Improved Personalized Feedback Approach R. Achterkamp, M. Cabrita, H. op den Akker, H. J. Hermens, M. M. R. Vollenbroek-Hutten P 2013 IEEE 15th International Conference on e-Health Networking, Applications and Services (Healthcom 2013) 978-1-4673-5801-9/13/$26.00 ©2013 IEEE 719