Integrating contextual and online self-reported data for personalized healthcare: a tennis elbow case study

Advances in sensors and mobile technology have helped evolve the use of eHealth, especially in the field of chronic pain. Chronic pain is a widespread problem where self-management is important. Current studies tend to collect data at sparse intervals due to the cost involved in collecting data using traditional instruments. We demonstrate how technology enables richer data collection frequencies to analyse the influence of patients’ context on their pain levels. In this paper, we present a case study as an add-on analysis to a clinical trial for lateral epicondylitis (tennis elbow). We explore the usefulness of on-line key data collected at higher frequencies in explaining or discovering changes in pain. This dataset allowed us to learn that there are no associations with temperature and humidity to this type of pain, that patients tend to have different pain experiences, and that pain at night tends to be higher than overall or activity-related pain.