Executive Summary
Clinical research is expanding beyond measurements collected during scheduled site visits. Wearables, smartphones, connected medical devices, and digital health platforms can now generate continuous or frequent information about patients in their everyday environments.
These data streams are contributing to the development of digital biomarkers: measurable, quantifiable characteristics derived from digital technologies that can provide information about physiological, behavioral, or disease-related states. Unlike conventional assessments that may capture a patient’s condition at a limited number of time points, digital biomarkers can provide more frequent observations of how patients function between visits.
For pharmaceutical companies, this creates opportunities to improve disease measurement, identify treatment-response signals, support patient monitoring, and design more patient-centric clinical trials. However, digital biomarkers also introduce challenges around validation, data quality, interoperability, privacy, participant burden, and regulatory acceptance.
The emerging opportunity is therefore not simply to collect more digital data. It is to identify digital measurements that are scientifically meaningful, clinically relevant, and sufficiently validated to support drug development decisions.
Key Themes
- Digital biomarkers can provide more frequent measurements outside clinical sites.
- Wearables and smartphones are expanding the range of measurable patient characteristics.
- AI is helping convert complex digital signals into potentially meaningful biomarkers.
- Digital biomarkers can support both efficacy and safety assessment.
- Validation and clinical relevance remain essential for adoption.
1. Wearable Activity Biomarkers
Wearable devices can continuously measure physical activity, movement, step counts, gait, and other behavioral characteristics. These signals can provide useful information about mobility and functional status across a patient’s daily life.
In clinical research, activity measurements may help assess diseases that affect movement, fatigue, physical function, or recovery. They can also provide longitudinal information that complements periodic clinical assessments.
The key requirement is demonstrating that changes in the digital measurement reliably reflect a clinically meaningful change.
2. Heart Rate and Cardiovascular Biomarkers
Smartwatches, chest sensors, and other connected devices can capture heart rate and related cardiovascular signals over extended periods.
Continuous monitoring can provide a broader view than measurements collected during occasional clinical visits. Researchers can potentially evaluate resting heart rate, activity-related changes, heart-rate variability, and other signals depending on the device and study design.
These measurements may support research across cardiovascular, metabolic, neurological, and other conditions where physiological changes are relevant.
3. Sleep and Circadian Biomarkers
Digital devices can provide estimates of sleep duration, sleep timing, movement during sleep, and other sleep-related characteristics.
Sleep data can be relevant to neurological disorders, psychiatric conditions, metabolic diseases, and other areas where changes in sleep patterns may provide information about disease status or treatment response.
Longitudinal measurements can also help researchers identify changes that may not be apparent from occasional patient-reported assessments.
4. Digital Gait and Movement Biomarkers
Smartphones and wearable sensors can capture information about walking speed, stride characteristics, balance, tremor, and other aspects of movement.
These measurements are particularly relevant to neurological and musculoskeletal research, where subtle changes in mobility can provide important information about disease progression.
Digital movement biomarkers can potentially detect changes earlier or with greater frequency than conventional assessments, although appropriate validation remains necessary.
5. Digital Cognitive Biomarkers
Smartphones, tablets, and other digital interfaces can be used to measure aspects of cognitive performance and behavior.
Researchers are exploring digital measures related to reaction time, memory, attention, speech, typing patterns, and other cognitive functions.
These approaches could support research in neurological and psychiatric disorders by providing more frequent assessments while potentially reducing reliance on lengthy in-person evaluations.
6. Digital Speech and Voice Biomarkers
Voice recordings can contain measurable characteristics related to speech patterns, vocal intensity, timing, articulation, and other features.
AI and signal-processing techniques can analyze these characteristics to identify patterns potentially associated with neurological, respiratory, psychiatric, or other conditions.
Because smartphones can capture voice data remotely, speech biomarkers may offer a relatively accessible method for longitudinal monitoring. However, differences in devices, environments, languages, and patient characteristics can affect measurement consistency.
7. Digital Respiratory Biomarkers
Connected devices and smartphone technologies are increasingly being explored for respiratory measurements. Depending on the technology, digital respiratory signals may include breathing rate, cough characteristics, oxygen saturation, or other physiological indicators.
These measurements could support research into respiratory disease, infection, sleep-related conditions, and other disorders.
The ability to capture respiratory information outside the clinic could provide researchers with additional insight into symptoms and disease fluctuations over time.
8. Digital Symptom and Patient-Reported Biomarkers
Mobile applications can collect frequent patient-reported information about symptoms, pain, fatigue, mood, medication use, and other experiences.
Although these measurements are often patient-reported rather than passively captured, their digital collection can increase measurement frequency and reduce reliance on retrospective recall.
When combined with sensor data, digital symptom measures can provide additional context for interpreting changes in physiological or behavioral signals.
9. Digital Phenotyping
Digital phenotyping involves analyzing patterns of behavior and interaction generated through digital technologies to characterize an individual’s health or disease state.
Potential signals can include mobility patterns, smartphone interaction, sleep, communication behavior, location patterns, and other digital traces, depending on the study and consent framework.
For clinical research, the approach could help identify disease subtypes, monitor progression, or detect changes in patient behavior. Because digital phenotyping can involve highly sensitive information, privacy, consent, data minimization, and governance are particularly important.
10. AI-Derived Composite Biomarkers
The growing volume of digital data is creating opportunities to combine multiple signals into composite biomarkers.
AI and machine learning can analyze combinations of movement, sleep, physiological measurements, patient reports, and other data to identify patterns associated with disease or treatment response.
Rather than relying on one digital signal, composite approaches could provide a more comprehensive representation of a patient’s condition. However, their complexity increases the need for transparent methodology, robust validation, reproducibility, and appropriate clinical interpretation.
How Are Digital Biomarkers Changing Clinical Research?
The most important change is the shift from episodic measurement toward more continuous observation.
Traditional clinical research may capture a patient’s condition during scheduled visits. Digital biomarkers can potentially provide information between those visits, creating a more longitudinal view of disease and treatment response.
This can support:
- More frequent measurements
- Remote patient monitoring
- Reduced dependence on site visits
- Earlier detection of changes
- More patient-centric trial designs
- Additional endpoints and exploratory measures
However, continuous measurement does not automatically mean better measurement. Researchers must establish whether the signal is reliable, clinically meaningful, and relevant to the study’s objectives.
What Challenges Must Pharma Companies Address?
Digital biomarkers require rigorous development and validation before they can become reliable clinical research tools.
Key challenges include:
- Analytical and clinical validation
- Device-to-device variability
- Data quality and missing data
- Patient adherence
- Sensor accuracy
- Interoperability
- Privacy and consent
- Cybersecurity
- Regulatory acceptance
Sponsors also need to consider participant burden. A technically sophisticated biomarker may have limited value if patients find the required technology difficult or inconvenient to use consistently.
What Will Define the Future of Digital Biomarkers?
The next phase will likely involve greater integration of digital biomarkers with clinical, molecular, imaging, and real-world data. AI can help researchers interpret these diverse signals and identify patterns that may be difficult to detect through individual measurements.
Digital biomarkers may also become more closely integrated into decentralized and hybrid clinical trials, allowing sponsors to collect information from participants in their normal environments.
The long-term opportunity is to move from isolated digital measurements toward validated digital endpoints that can contribute meaningfully to clinical development decisions.
Key Takeaways
- Wearables can provide continuous activity and physiological measurements.
- Cardiovascular signals can be monitored beyond scheduled clinical visits.
- Digital sleep measures can provide longitudinal information about patient behavior.
- Gait and movement data can support research into neurological and musculoskeletal conditions.
- Smartphones can enable more frequent cognitive assessments.
- Voice analysis is emerging as a digital biomarker across multiple disease areas.
- Digital respiratory measurements can extend monitoring beyond the clinic.
- Mobile symptom reporting can increase the frequency of patient-reported assessments.
- Digital phenotyping can characterize broader patterns of behavior and health.
- AI-derived composite biomarkers can combine multiple digital signals into richer measures.
Conclusion
Digital biomarkers are changing clinical research by expanding when, where, and how patient health can be measured. Wearables, smartphones, connected devices, and AI are creating new opportunities to capture physiological, behavioral, and patient-reported information between traditional clinical visits.
The strategic value, however, will depend on scientific validity rather than data volume. A digital signal becomes meaningful for clinical research only when researchers can demonstrate that it is reliable, reproducible, interpretable, and relevant to the disease or treatment being studied.
For pharmaceutical and biotechnology leaders, this means digital biomarkers should be approached as scientific measurement capabilities rather than simply technology projects. With stronger validation, interoperability, privacy safeguards, and regulatory experience, they could become an increasingly important component of more continuous, patient-centric, and data-rich clinical development.
What Are Digital Biomarkers in Clinical Research?
Digital biomarkers are measurable indicators of health or disease collected through digital technologies such as wearable sensors, smartphones, connected medical devices, and software applications. In modern Clinical Research, these measures can help researchers track changes in patients’ health, evaluate treatment responses, and collect information outside traditional clinical settings.
Unlike assessments conducted only during scheduled visits, digital biomarkers can provide frequent or continuous measurements of movement, sleep, heart activity, and other health-related signals. A 2026 review of wearable technologies identified growing use across clinical trials, while also highlighting the need for reliable validation and regulatory qualification.
Challenges in Adopting Digital Biomarkers
Despite their potential, digital biomarkers introduce challenges for Clinical Research organizations. Different devices may produce inconsistent measurements, while missing data, poor connectivity, battery limitations, and participant adherence can affect data quality.
Privacy and security are also essential because continuous monitoring may collect sensitive information. Researchers must establish appropriate consent processes, protect participant data, and explain how digital measurements will be used.
Another major consideration is clinical validation. A sensor reading does not automatically qualify as a reliable biomarker or a suitable clinical trial endpoint. Researchers need evidence that the measurement is accurate, reproducible, meaningful to patients, and fit for its intended context. The FDA provides guidance on evaluating digitally derived measures for clinical investigations.
The Future of Digital Biomarkers in Clinical Research
Digital biomarkers are expanding the ways researchers study diseases and evaluate treatments. Wearables, smartphones, connected sensors, and digital assessments can provide complementary information about patients’ health between traditional study visits.
As technologies advance, Clinical Research may increasingly combine digital measurements with laboratory results, imaging, patient-reported outcomes, and conventional clinical assessments. This integrated approach could help researchers build a more complete understanding of disease progression and treatment response.
