Harnessing consumer smartphone and wearable sensors for clinical cancer research

Growing evidence suggests that consumer wearable and smartphone sensor data are related to symptoms, quality of life, and risk for adverse outcomes in cancer. Larger studies that use sensor data to inform and personalize clinical care are needed.

While advances in early detection and treatment have yielded significant improvements in survival rates, cancer remains the second leading cause of death worldwide and a major public health problem.  Cancer and its treatment can cause debilitating symptoms such as nausea, fatigue, and pain, weaken immunity, impair functioning and ability to engage in valued activities, and lead to repeated unplanned hospitalizations and emergency room visits.  

As smartphones and consumer wearable devices have become more common and powerful, they offer an opportunity to harness sensor data to improve clinical cancer research and care.  Mobile sensors can quantify important factors such as physical activity, smartphone use habits, sleep, heart rate, and location, and they can capture these data continuously, in real-world settings, between clinical encounters, and with minimal patient burden.

A paper published today in npj Digital Medicine reviews the growing evidence linking smartphone and wearable sensor data to clinical outcomes in adult cancer patients. We identified 14 small prospective studies suggesting that mobile sensing data may reflect meaningful variation in patient-reported symptom burden, quality of life, functional status, and risk for adverse outcomes including unplanned hospital readmissions and mortality. These studies support the feasibility and potential clinical value of mobile sensing in oncology, but larger studies that use sensor data to inform clinical care are needed.  

Expanding this research and making it more clinically impactful requires overcoming significant barriers, including translating sensor data into actionable clinical insights, supporting patient and clinician engagement and co-designing systems that meet their evolving needs and are integrated into their routines, and ensuring that patient privacy is protected. 

In the future, these passive sensor measures could be integrated into clinical trials, to predict the likelihood of a patient tolerating a new therapy or to better understand how treatments affect patient lives.  For example, in our research at the University of Pittsburgh and UPMC Hillman Cancer Center, we have used Fitbit devices to better understand recovery trajectories following open versus robotic pancreatic cancer surgery. 

Mobile sensor data could be used for remote real-time patient monitoring and combined with algorithms to trigger clinician or caregiver notifications or patient education instructions. In another study funded by the National Cancer Institute, we are collecting sensor data from patients starting a new chemotherapy regimen and aim to develop algorithms that detect worsening side effects and trigger clinical action. 

Wearable device and smartphone data can also be used to personalize supportive or lifestyle interventions, tailoring content, dose, timing, or goals and helping patient identify patterns in their own data to better understand how behaviors and other variables affect their health.  In a pilot study of persistently fatigued cancer survivors, we are using data from consumer wearable devices to help patients identify and change activity behavior patterns that seem to worsen their fatigue. 

The COVID-19 pandemic has highlighted the value of remote patient monitoring and care, especially for vulnerable groups such as cancer patients. Clinical oncology experts, computer and data scientists, and behavioral scientists will need to work together to realize the potential of mobile sensing to transform cancer care and research.

Please sign in or register for FREE

If you are a registered user on Healthcare and Nursing Community, please sign in