Nurses can foster data-powered health through getting involved in research and ensuring that data collection is in patients’ best interest and accessible to all. Patricia Brennan, PhD, RN, director of the National Library of Medicine, and Suzanne Bakken, PhD, RN, FAAN, a professor at Columbia University, discussed how data can empower health and what nurses can do to advance research during a session at the 43rd Annual Congress in Washington, DC.
Data science has exploded in many fields, including medicine. Brennan discussed the work the National Institutes of Health (NIH) is doing, including the All of Us Research Program, a key element of the Precision Medicine Initiative. This historic effort is gathering data on more than 1 million people in the United States to further research and improve health. She encouraged nurses to join All of Us either as a participant or as a provider with the knowledge and perspective of their profession. Until June 15, 2018, NIH has a request open for input on strategies to more properly match patients to clinical trials.
“This is your opportunity to advocate for your patients,” she said, urging the elevation of the nursing perspective.
She also noted that the program is unique in that the cohort is much more diverse than many clinical trials and the follow-up is long-term. She asked the audience, “What questions would you want to answer if you had all the data in world?”
Research projects used to just answer questions; now they generate ideas, Brennan said, touching on images, laboratory tests, genome sequences, and patient behavior. She also questioned ways to preserve and protect such a large amount of data.
“We are talking about data at a size you can’t even imagine,” said Brennan. She discussed data commons, which are computing environments that support access, utilization, and storage of digital objects that include software services and tools that enable provisioning, indexing, sharing, and connectivity.
“Data science has got to lead us to data-powered health,” she said. Without interpretation, data is useless, she said. “You must interact with data-driven discoveries going on around you,” Brennan said.
“We are challenged to make sure data is flowing freely,” she added. Organizations like the NIH that are collecting data need inspiration from nurses, clinicians, and patients to understand what is important.
Areas where nurses are involved in data collection include signs, symptoms, observations of daily living, images, biomarkers, family dynamics, patient experience, population phenotypes, biological specimens, and everyday living spaces.
“I want you to become more conscious of data elements you experience,” she said. “What would make a difference to your practice if it could be answered? What would make a difference to your patient if it could be answered?”
Bakken then discussed various data streams for nursing science and care, including electronic health records, patient-reported outcomes measurement information systems and other patient-reported data, biomarkers, exposomes, digital biomarkers, wearable tracking devices, and sensors.
Data is moving beyond electronic health records, she said, delving more into the socioeconomic front, and there is a need for innovative ways to handle and record this. She gave an example of a 2013 study by Collins et al., which found that nursing documentation patterns were linked to patient mortality. Among the 15,000 acute care patients and 145 patients who experienced cardiac arrest, those who died had a mean of 0.9–1.5 more optional comments and 6.1–10 more vital signs documented in 48 hours than patients who survived. A higher frequency of comments and vital sign documentation was also associated with a higher likelihood of cardiac arrest.
“Data science is a team sport,” she said, which involves math and statistics, computer science, and subject matter expertise. Competencies of data science include research, practice, and ethics.
“There’s data everywhere, but some data is used in ways it wasn’t intended or consented,” Bakken cautioned. Researchers need to balance of risks and benefits of data. “Is your data being commodified and for purposes you would be against?” she asked, noting a need to consider the people within the dataset being mined.