Event Abstract Back to Event Quantified-self for obesity: Physical activity behaviour sensing to improve health outcomes Jennifer Murphy1*, David Taylor1, Mian Ahmad1, Suzanne I. Alsters2, Sanjay Purkayastha1, Samantha Scholtz3, Ramin Ramezani4, Ivo Vlaev5, Ahmed R. Ahmed6, Harvinder Chahal6, Alexandra I. Blakemore*2 and Ara Darzi*1 1 Imperial College London, Department of Surgery, Cancer and Investigative Medicine, United Kingdom 2 Imperial College London, Section of Investigative Medicine, Division of Diabetes, Endocrinology, and Metabolism, Department of Medicine, United Kingdom 3 Imperial College London, Department of Investigative Medicine, United Kingdom 4 University of California, Los Angeles, Wireless Health Institute, United States 5 University of Warwick, Behavioural Science Group, Warwick Business School, United Kingdom 6 Imperial College Healthcare NHS Trust, St. Mary’s Hospital, Imperial Weight Centre, United Kingdom Background Physical activity is effective for long-term weight loss maintenance following calorie restriction diets; however, there is a lack of research on its effects on weight loss following bariatric surgery1. Previous studies objectively measuring physical activity have shown that while there is a wide range of activity levels in bariatric patients, the majority achieve significantly less than the recommended 30min/day moderate-to-vigorous physical activity (MVPA) guidelines for health-related benefits2-5. It is important to gain a better understanding of physical activity profiles in this heterogeneous group over time and to ascertain whether increased physical activity is associated with better weight loss outcomes, improved quality of life and resolution of comorbid conditions. Aims The primary aim of this study is to assess the feasibility of continuously monitoring physical activity via a smartphone app with minimal user interaction required, and to characterise the physical activity profiles of bariatric surgery patients. Method 255 patients aged 18-65years with a BMI >35kg/m2 and who were either awaiting or had previously underwent bariatric surgery were recruited from the Imperial Weight Centre. Physical activity [i.e. step count, walk time (mins) and distance walked (km)] was recorded using the free, commercial ‘Moves’ app (Protogeo), with an adjunct app recording weight, mood and wakefulness data. Physical activity data tracked by the Moves app for at least 8 hours per day between the hours of 6am-10pm was categorised as follows: 1) Average walking time (mins) of light, moderate and vigorous intensity per day, 2) ‘Moderate to Vigorous physical activity (MVPA)’; average daily walking time with >80steps/min; 3) average light and MVPA in bouts of 10 minutes per day, and 4) average MVPA in 10 minute bouts of activity per day. Results Data from 147 participants (mean age 42.95, SD±11.50; 70.7% female) met inclusion criteria for data analysis. The average BMI of patients awaiting surgery (n=52) was 47.1kg/m2 (SD±9.53) and average weight loss percentage for post-surgery patients (n=95) at the time of recruitment was 14.62% (SD±10.71) encompassing a post-surgery period from 10 days to >2 years. Participants were tracked for a median of 26 days each (range 1-139 days, partly depending on when they were recruited). 61% of participants (n=89) achieved an average of at least 30 minutes of light and MVPA per day. 57% (n=84) achieved one or more bouts of light and MVPA for a minimum of 10 minutes per day but only averaging 15.31 minutes (SD±2.89). 66% of participants (n=97) walked at a moderate-to-vigorous pace at >80 steps/min, but only for an average of 6 minutes (SD±2.83) per day. 54% (n=80) performed MVPA in bouts of 10 minutes, but only averaging at 15 minutes (SD±3.26). Conclusions We have shown that it is possible to passively monitor physical activity in a large patient population in a cost-effective way. The results demonstrate that while two thirds of bariatric patients achieved an average of 30 minutes walking per day, this was not of sufficient intensity to gain health-related benefits. Further analysis will examine whether increased activity is associated with successful weight loss outcomes, improved mood and psychological functioning, and increased quality of life. We will also employ machine-learning techniques to identify the factors that are critical for a successful outcome following bariatric surgery. Recruitment will continue to the end of the project (April 2016) and tracking will continue into 2017.