Clinical context to take your research a step further
You probably know the frustration of realizing crucial insights you want to include in your research — data on disease progression, symptom severity, caregiver impressions of the patient — are missing from or underreported in traditional structured clinical and claims data. But this valuable information isn’t lost; these details are often just trapped in unstructured clinical notes.
By directly accessing de-identified clinical notes, you can learn more about the patient’s journey across specialists, settings and sites of care. This gives you a more holistic view of:
- Patient behaviors
- Provider decisions
- Health outcomes
With this granular information, you can answer more complex research questions and propel your work forward to help patients, no matter the therapeutic area you’re working in.
Enable meaningful exploration of patient cohorts
Access a secure clinical notes platform
With Optum Clinical Notes Lab, you can review de-identified clinical notes data autonomously.
Data de-identified in accordance with HIPAA’s Expert Determination method. This helps safeguard patient privacy while maintaining as much context as possible about the patients in your population of interest.
By accessing the cloud-based software platform yourself — rather than having a vendor apply their own artificial intelligence (AI) to notes data and only provide you the output — you can independently review notes to gain a thorough understanding of patient-clinician interactions.
Use AI tools to support your learning and analysis
A suite of natural language processing (NLP), large language models (LLMs) and AI tools — along with pre-loaded libraries — are available to support the exploration and annotation of the clinical notes.
These resources enable you to easily extract clinically relevant concepts and annotate both the concepts and their relationships within the text. For more granular control, additional tools for manual annotation are available as well.
You can also upload and train your own models on the notes using Amazon SageMaker. Because this capability allows you to develop and refine custom models, the extracted data outputs are uniquely tailored to your company's specific needs.
Plus, you can adjust and iterate upon your own models as needed, leading to a more agile and targeted workflow.
Work with a curated notes cohort
With a notes cohort specifically defined by your team’s research criteria — medications, age, ICD-10 diagnosis codes, etc. — you know you’re working with patient data that meet your needs.
All patients in the data set are sourced solely from electronic health record (EHR) notes. This means you can review real-world patient and provider behaviors and observations pulled directly from the clinical record.
This lets you study meaningful concepts you might not otherwise see in structured data fields as well as supplement the variables reliably represented in structured data.
Key benefits
Optum Clinical Notes Lab can help you gain new, propriety insights and get an even greater return on your real-world data investments.
Analyze high-value information
Our unstructured notes are derived from the EHRs of millions of real-world patients spanning geographies, demographics and sites of care.
Conduct research with greater confidence
With more data points, you can develop insights based on a more holistic view of the condition, from pre-diagnosis to the present day.
Enjoy the ease of a self-service solution
Train models and explore data on your own time. For example, compare data outputs to source notes — without waiting for the vendor to do so.
Integrate with other structured clinical data
Optum can compliantly link Clinical Notes Lab outputs back to your existing clinical data licensed from us at a 100% patient match rate.
Trace the journey from request to research-ready data set
Clinical Notes Lab FAQ
You can:
- Determine rationale for medication switching and factors in patient adherence and nonadherence
- Better understand and evaluate care and disease progression
- Evaluate lifestyle factors – such as physical activity and diet – and their impact on clinical outcomes
- Monitor physician prescribing patterns around clinical events
- Explore patient cohorts for conditions not well-identified or underrepresented with diagnosis codes
With our clinical notes platform, you can:
- Understand the patient story underpinning treatment changes
- Profile patients based on symptomology to compare treatment pathways and outcomes based on different symptoms and clinical presentations
- Develop detailed phenotypes to support the development and commercialization of products
- Inform medical communication strategies for healthcare providers
- Generate tailored treatment recommendations based on patient-specific information
- Study rare and underdiagnosed diseases
You can gain a deeper understanding of the Alzheimer’s patient journey, for example. Through clinical notes, you get:
- Context behind cognitive test administration
- Perspective on the patient’s condition and lifestyle — from neurologist observations to caregiver dynamics
- Better ability to characterize patients by disease severity with various test scores
- More nuanced understanding of how disease severity affects outcomes
Related healthcare insights
Article
What more is there to the patient story? Get deeper data insight into patient-clinician interactions with de-identified clinical notes.
Article
Using real-world data (RWD) that are not fit for purpose can waste time and money. Review 6 simple steps to guide your research towards the right data.
White paper
Natural language processing (NLP) systems allow us to draw insights from billions of unstructured medical notes. Learn more about our methodology.
Product information and resources
-
Download fact sheet
PDF
Fact sheet
Reap the research benefits of Clinical Notes Lab
Get an overview of features and potential use cases to fuel your research.
-
Download fact sheet
PDF
Fact sheet
Mining clinical notes to understand Alzheimer’s disease
Review the art of the possible with this Alzheimer’s disease example and walk step-by-step through the process of deriving value from clinical notes data.