The promise of precision medicine is growing as fast as the volumes of data required for its application. Precision medicine structures a person’s disease prevention and treatment in relation to their genomics, environment, and lifestyle. The data-driven insights needed to deliver individualized care are obtained from a host of data sets, such as clinical repositories, genomic libraries, published peer-reviewed research, and data from both the structured and unstructured portions of the electronic health record.
Precision medicine customizes medical treatment to the individual characteristics of a patient. The field of precision medicine is at the forefront of battling complex diseases, e.g., cancer, by supporting clinicians’ ability to determine treatments for patients and life sciences companies’ development of more targeted medications for specific cancer populations.
Many types of data are needed to fully realize precision cancer treatment strategies. Genomic and biomarker data libraries provide insights into which cancers will respond to certain treatments, or which drugs in development may be most effective. Even the non-genic portion of a person’s DNA, the “dark genomic matter,” shows promise in identifying cancer subtypes.1
Other sources of comparative data include scored representational unbiased image libraries, cancer registries, and a rapidly growing body of published scientific articles that provide the support needed for understanding cancer development, diagnosis, and treatment. De-identified longitudinal databases of EHR or claims data allow for the study of populations of people with cancer, which can provide valuable insights into disease development and treatment over time.
These rich sources of data serve as a backdrop to understanding what is gleaned from existing patient records. Health history and medical encounters are required to set the baseline for prior treatment efficacy and disease progression. Patient data can originate from multiple sources, making it important to leverage technologies that bring together these types of data over time.
These sources of data are most effective when they are:
The presence of data is not enough, however. A health information technology (IT) infrastructure that curates, standardizes, and shares high quality data is necessary to advance precision medicine.
The challenge is daunting. In 2018, HIMSS reported that the average hospital had 16 disparate EMR vendors in use at affiliated practices.2 Combine the interoperability and coding challenges with the capture and curation of the variety of data inputs, throw in unstructured patient data and non-standardized data sources, and the result are significant bumps in the road to applying precision medicine to cancer care. The outputs will only be as good as the inputs.
The Office of the National Coordinator for Health Information Technology (ONC) article, Health IT Advances Precision Medicine, states that “Standardizing data generally results in improved data quality and consistency, but the use of data standards does not always result in data standardization due to inconsistent implementation of current standards and the use of proprietary standards, creating a constant need for harmonization.” 3
Harmonization efforts can be assisted by laws like the 21st Century Cures Act, which paves the way for more standardized infrastructure and information sharing. However, as noted above, “the use of data standards does not always result in data standardization.”
This is why IT solutions, such as Verinovum’s Data Curation as a ServiceSM platform or DCaaS, that ingest, standardize, and curate data so insights into a person with cancer episodes of care are fully understood are essential. Verinovum’s DCaaS platform can help pave the way to seamless data harmonization.
These types of automated transformations can make time for what the researchers and clinicians do best — personalizing the treatments and care for the individual as opposed to attempting to gain better insights from imperfect data.
1 Parida L et al.Dark-matter matters: Discriminating subtle blood cancers using the darkest DNA. PLoS Comput Biol. 2019 Aug 30;15(8):e1007332.
2 Sullivan T. Why EHR data interoperability is such a mess in 3 charts. Healthcare IT News. 2018 May 16.
3 Chaney K, Caban TZ, Rogers CC, Denny JC and White J. Health IT Advances Precision Medicine. ONC: Health IT Buzz. 2021 April 20.