OBJECTIVES: New m-health technologies such as wearable devices allow us to monitor patients like never before and such frequent data collection generates unprecedented amounts of information. This gives the potential to track disease progression and patient experience with great precision, however, the processing, summarizing and understanding of such vast quantities of data is a challenge. This research aims to highlight how effective data visualization allows for quick understanding of large volumes of m-health data.

METHODS: We present a hypothetical case study (i.e., using simulated data) in which patients complete a daily ‘symptom diary’ grading a range of symptoms each day over a length of time and show how an automated RMarkdown report can quickly produce a range of visualizations covering different features of the data. The same report is re-run on a second symptom diary with different items, grading scale and length of administration which is stored in a dataset with the same basic structure.

RESULTS: The report allows us to rapidly gain insights into the patient experience and disease progression at a patient-level, identifying patterns and features which would be nearly impossible to spot on summary tables. The visualizations also allow for inference on the respective performance of competing treatments over the period of administration. Re-running on the second diary highlights the efficiency gained from template code, allowing the same insights to be instantaneously generated for an entirely different diary.

CONCLUSIONS: Efficient template code that produces smart data visualization can substantially increase the speed at which we get to understanding large amounts of m-health data. This work also shows how to synthesize the patient experience to reveal important patterns in the symptom pathway.