The growth rate of health-related data is set to outstrip that in financial services and manufacturing industries, honing in on human body analysis at a higher resolution and more in real time than ever before.
Technology allowing us to measure our health has traditionally ridden on single measurement points like vital signs, such as heart or respiratory rate, and biochemical assays, such as glucose or CRP levels in blood. This provides only a snapshot of the patient in several limited dimensions.
Now health data technology has evolved to more complex, multi-dimensional data types, such as high-resolution CT imaging for emphysema, genome sequencing for rare disease and tailoring treatment for cancer patients.
As a result, health-related data is projected to have an annual growth rate of 36% in the period towards 2025, outstripping rates in other industries such as manufacturing and financial services1. These advancements in medical technology allow the human body to be analyzed and described digitally at resolutions never previously possible.
The medical data of tomorrow will display several new traits that explain its high projected growth rate. Spearheading this acceleration is the continuous, real-time data being produced from health sensors, either from external wearables such as non-invasive continuous glucose monitoring devices, or implanted devices currently under development to measure cardiovascular parameters.
Wellness data from lifestyle sensors are also increasing in volume and availability, such as the such as the Apple watch ECG (electrocardiogram) which measures electrical signals that make the heart beat. However, while an intention of these data is to provide healthcare services with better insight into a patient’s health, it remains to be seen how and to what extent this data is relevant and can realistically be integrated into clinical flows.
Sensors in development which are already relevant to existing clinical flow are those that measure disease-specific traits, such as tremors for Parkinson’s disease2. The opportunities that these sensors provide also bring the necessity of thinking critically about what data is collected.
Connectivity is another common trait underpinning the next generation of health sensors and will be at a higher degree than ever before, defined in part by a range of protocols dependent on the specific needs of the device, but also by the opportunities utilized and risks managed of remote monitoring of a patient’s condition. One real-time remote patient monitoring system being piloted in the UK, Current, helps staff to prioritize home visits and reduce readmission to hospitals, such as by detecting a decline in oxygen saturation in a patient and intervening sooner than standard care.
Connectivity will also enable another trait: actionability. Through the integration of transmitted data streams with relevant knowledge bases and decision support, clinical decisions can be made and even implemented remotely through feedback to the medical device. To effectively take advantage of connectivity, we must define the scope of the data we take action on and the knowledge underlying it. In context, we can think of how data on an individual’s cholesterol level is integrated with knowledge of their genetic profile and their lifestyle. The scope on which data is collected will impact the health recommendation for that individual.
Widespread impact on the healthcare ecosystemPatients, clinicians and healthcare systems in general will be significantly impacted by widespread use of medical sensors and devices displaying one or more of these traits.
For patients, real-time remote monitoring enables actionability on their health condition on a shorter timescale and in some cases may allow preventive measures to be taken, such as through closed-loop epilepsy detecting and controlling devices. Self-monitoring empowers a patient, creates a framework for prevention and provides healthcare systems with better insight at a potentially lower cost. However, these advantages ride on a patient’s willingness and ability to self-monitor, investment in staff training and new technologies, and consideration of how greater monitoring and less or more healthcare thereafter may impact a patient’s mental health.
For clinicians, increased availability of real-time and real-world data streams will shape space for tailored clinical decisions and the implementation of precision medicine for individual patients, but also aggregation of data across larger cohorts of patients. Combined with secure data sharing and high-performance computing, this will enable clinicians to gain deeper insights into disease mechanisms and allow for more accurate decision making, such as in the example of a lung cancer clinicogenomic database populated with data from routine patient care3.
Implementation of next-gen sensor technology will impact healthcare systems as a whole by narrowing the distance between a patient, healthcare staff and treatment. Remote monitoring in combination with improved virtual patient-doctor communication interfaces will facilitate low-threshold consultations, potentially at earlier stages of illness, thereby reducing the pressure on hospitals. Within hospitals tagged patient bracelets can also help track patient flow, thereby identifying bottlenecks and improving efficiency. Finally, integrating data sources will allow assessment of patient trajectories across cohorts, enabling analysis of both cost and quality of care, and subsequent evidence-based changes.
ContributorsMain author: Sharmini Alagaratnam
Editor: Tiffany Hildre
- The Digitization of the World: From Edge to Core, IDC White paper, 2018
- Wearable sensors for Parkinson’s disease: which data are worth collecting for training symptom detection models. Lonini et al., Nature Digital Medicine, 2018
- Association of patient characteristics and tumor genomics with clinical outcomes among patients with non–small cell lung cancer using a clinicogenomic database, Singal et al., JAMA, 2019