Transforming Nursing

Nursing_Needs_Big_Data_and_Big.pdf

Home >Nursing homework help >Transforming Nursing

PROFESSION AND SOCIETY

Nursing Needs Big Data and Big Data Needs Nursing Patricia Flatley Brennan, RN, PhD, FAAN, FACMI1 & Suzanne Bakken, RN, PhD, FAAN, FACMI2

1Beta Eta at large, Lillian S. Moehlman-Bascom Professor of Nursing and Industrial Engineering, University of Wisconsin-Madison, Madison, WI, USA 2Alpha Eta, The Alumni Professor of Nursing an Professor of Biomedical Informatics, Columbia University, New York, NY, USA

Key words Big data, clinical information, data science

Correspondence Dr. Patricia Flatley Brennan, University of

Wisconsin-Madison – Living Environments

Laboratory, 330 N. Orchard St., Madison, WI

53715. E-mail: [email protected]

Accepted: June 19, 2015

doi: 10.1111/jnu.12159

Abstract

Purpose: Contemporary big data initiatives in health care will benefit from greater integration with nursing science and nursing practice; in turn, nursing science and nursing practice has much to gain from the data science initiatives. Big data arises secondary to scholarly inquiry (e.g., -omics) and everyday ob- servations like cardiac flow sensors or Twitter feeds. Data science methods that are emerging ensure that these data be leveraged to improve patient care. Organizing Construct: Big data encompasses data that exceed human com- prehension, that exist at a volume unmanageable by standard computer sys- tems, that arrive at a velocity not under the control of the investigator and possess a level of imprecision not found in traditional inquiry. Data science methods are emerging to manage and gain insights from big data. Methods: The primary methods included investigation of emerging federal big data initiatives, and exploration of exemplars from nursing informatics re- search to benchmark where nursing is already poised to participate in the big data revolution. We provide observations and reflections on experiences in the emerging big data initiatives. Conclusions: Existing approaches to large data set analysis provide a neces- sary but not sufficient foundation for nursing to participate in the big data rev- olution. Nursing’s Social Policy Statement guides a principled, ethical perspec- tive on big data and data science. There are implications for basic and advanced practice clinical nurses in practice, for the nurse scientist who collaborates with data scientists, and for the nurse data scientist. Clinical Relevance: Big data and data science has the potential to provide greater richness in understanding patient phenomena and in tailoring inter- ventional strategies that are personalized to the patient.

Nursing and the people nursing serves stand to benefit from the application of data science methods to the mas- sive amount of data now emerging from environmental sensors, clinical assessments, and imaging and laboratory studies. This massive amount of data, referred to as “big data,” may hold insights about the patient experience heretofore not available through traditional research ap- proaches. Nursing’s participation in the big data and data science initiatives now underway is essential to ensure that the discoveries not only be shaped by our profes- sion’s unique understanding of the patient experience but also that the discoveries lead to knowledge that is

useful to nursing. As nurse thought leaders in biomedical informatics, we present observations and reflections based on our experiences in the emerging big data initiatives.

What can big data provide nurses? Big data can illu- minate the health consequences of climate change that lead to molecular changes in fish that, once consumed, cause novel infections in humans. It can reveal patterns shared by thousands of people. It can suggest hypotheses that in turn can be answered through the evaluation of multiple data streams. Our key premise in this article is that nursing needs big data and big data needs nursing.

Nursing Needs Big Data Brennan & Bakken

As context to providing evidence to support our premise, we briefly define big data and data science and examine national initiatives. We then draw parallels between methods familiar to nursing and those employed by data science. We close with recommendations for capacity building in nursing.

Big Data and Data Science

The phrase “big data” emerged over the past decade to encompass the phenomenon of large amounts of data emerging from sensors, novel research techniques, and ubiquitous information technologies. Big data is first and foremost a metric of size, reflecting initially the idea that data sets far in excess of those commonly found in research or business practices are now available and require new ways to manage and new methods for analysis. What has emerged is the realization that data are “big” not only because of size (the volume of the data), but also due to other characteristics. These char- acteristics include variety, velocity, veracity, and value (Laney, 2001). Variety reflects the diversity of data types, including but not limited to alphanumeric data, image data, and continuous flow data, such as streaming video or blood flow monitoring. Velocity is that characteristic that depicts the unprecedented speed at which data are generated and received. Veracity refers to the level of uncertainty associated with data elements and their source, and alert the user to the inherent uncertainties in data that may have been collected under unanticipated or untraceable conditions. A fundamental shift brought about by big data thinking is that relevance of data to a question occurs at the point of use rather than at the point of collection. What is important is not just the data, but the insights and scientific discoveries enabled by it.

Mention “big data” in health and one’s mind turns to the “omic” sciences (e.g., proteomics, genomics, and metabolomics). Masys (National Institutes of Health Big Data to Knowledge [NIH BD2K], 2014) characterizes big data as that which exceeds the capacity of unaided human cognition and strains the computer processing units (CPUs), bandwidth, and storage capabilities of modern computers. Big data may emerge from a primary data collection, such as the genomewide association studies, as by-products from some otherwise purposeful effort (think of Twitter feeds), or as a mix of the two. The boundary between “big data” and any other data may be arbitrary and leveraged for purpose outside of inquiry. Fortunately, a principled, scientific process for handling big data is emerging.

“Data science” complements the slang term “big data” and reflects this realization that principled exploitation

avoids the problems associated with uncharted examina- tion of data. Data science is both a philosophy and a set of techniques to address data that are openly accessible and distributed across multiple locations, explored and analyzed with sharable routines, and fraught with uncer- tainty, such that data provenance (the trace of the data source and all subsequent modifications) is as important in big data explorations as precise variable definitions are in traditional inquiry. The National Consortium for Data Science has defined data science as the systematic study of the organization and use of digital data in order to accelerate discovery, improve critical decision-making processes, and enable a data-driven economy (Ahalt et al., 2014). Data science encompasses the principled acquisition, curation, exploration, manipulation, and interpretation of big data sets.

Data science draws upon theories and techniques from multiple fields, as well as multiple methodological tradi- tions. These include mathematics, probability, statistics, predictive analytics, uncertainty modeling, and infor- mation science, incorporating computer programming and signal processing. Computer engineering strategies also employed are visualization, data warehousing, and high-performance computing (Wang & Krishnan, 2014). Despite its opportunistic origins, data science now commands strong attention from the scientific sector. For the remainder of this article we will employ the term “big data” to refer to the sociological revolution underway, and “data science” when we specifically refer to the scholarly dimensions.

The process of using big data begins with posing questions or recognizing opportunities. Nurse scientists are well aware of this familiar starting point for investiga- tions. Data science methods are useful for both primary inquiry and secondary analysis. In addition, data science includes both special purpose methods as well as meth- ods that re-use existing data similar to those employed in Knowledge Discovery in Databases (KDD). Hilary Mason from bitly.com proposes a general structure for data sci- ence inquiry: obtain, scrub, explore, model, and interpret (Mason & Wiggins, 2010). After a question or focus has been determined, one must obtain the data. Often the data emerge as secondary to some other process, such as sending Twitter feeds about the flu or evaluating the pres- ence of a specific allele in a genotype exploration. Rarely are all of the data employed in a data science investigation located in a single site, and, because of volume, these data are rarely moved and stored in a single newly created data set. Thus, distributed data management techniques are needed. Due to the messiness of real-world data, substan- tial effort is required in the scrub phase (i.e., review and clean) to ensure that the data employed in subsequent ef- forts is of sufficient integrity for subsequent use. In most

Brennan & Bakken Nursing Needs Big Data

data science investigations, the data schema is defined at the point of use, rather than a priori, known as the distinction of “schema on read” vs. “schema on write.” This “schema on read” approach leads to an ever-evolving data model that grows and is modified as new data are encountered and obtained, rather than rejecting data because they do not meet an existing schema. Data explo- ration includes visualizing, clustering, and dimensionality reduction techniques, and employs human-mediated as well as machine-driven approaches. Modeling strate- gies, again driven by the primary questions, are iterative and formative rather than summative, and designed from an exploratory, convergence perspective rather than an inferential, hypothesis-driven perspective. Finally, and perhaps most important, the purpose of in- terpretation of these huge data sets is insight rather than prediction.

There are four features common to data science inves- tigations: (a) highly distributed data sources that remain under the control of the original owner of the data, (b) employment of provenance and security measures that address both data at rest and data in motion, (c) engagement of a large community of investigators and collaborators who share methods and leverage each other’s results, and (d) a focus on accelerating insights that can be gleaned by re-use of data generated through many strategies—with an emphasis on the integrity of the integration of the data more than on the original collection of the data.

National Initiatives in Big Data and Data Science

The National Science Foundation, the National Aero- nautics and Space Administration, and the Department of Defense have all launched data science initiatives in the past few years. The NIH established the BD2K ini- tiative (http://bd2k.nih.gov/) to advance methodologies, discoveries, and workforce development in the quest for using data science to accelerate biomedical discovery. The BD2K initiative capitalized on scientifically generated data from three types of research programs: (a) projects funded to produce important resources for the research community that yield large amounts of data, (b) large data sets useful for individual projects that might be broadly useful to the research community, and (c) small data sets whose value can be amplified by aggregating or integrating them with other data.

Developments are occurring rapidly, and already the major focus of federal investment moves away from basic methods development into clinical and research application of data science. In January 2015, President Obama announced the Precision Medicine Initiative. This

multi-million dollar program will create a million-person genomic resource possible only through data science methods, which again can benefit, and benefit from, nursing (Collins & Varmus, 2015).

Also at the national level, the Patient Centered Out- comes Research Institute (PCORI) has developed PCOR- net, a large, highly representative, national network for conducting clinical outcomes research (Fleurence et al., 2014). PCORI contributes to all phases of the data science process: generating new data sources, stimulating the development of novel methods, and creating utilities for data storage, integration, visualization, and interpre- tation. This has increased the availability of data from a variety of sources through its clinical research data networks and patient-powered research networks.

Drivers of Big Data and Data Science in Health and Nursing

There are multiple drivers for the increased attention to big data and data science for health. Data are increasingly available. “-omics” analysis strategies generate volumes of data. Data science methods may open the way to supporting inquiry at the level of complexity needed by nurses whose practice relies on understanding health in everyday living and delivering contextual interventions. The emerging Internet of Things has the potential to accelerate the data explosion through sensors around the world embedded in things we wear and the appliances we use. Similarly, electronic health data have increased in volume and variety. Consumer-generated data from mo- bile apps and medical devices along with multiple sources of environmental (e.g., biosensors, smart homes) com- plement traditional electronic sources of research data such as surveys, data warehouses, and transaction data.

Nursing’s traditions of inquiry relying on electronic health records (EHRs), claims data exploration, or public health data sets truly provides a good foundation for thinking about the rubric of data science. It is critical to remember that nursing must use all methods, familiar and emerging, for generating nursing knowledge. It is useful to ground what we must do in the future with what we have accomplished in the present.

Nursing Needs Big Data and Data Science

The relevance of big data and data science to nursing science is multifaceted. First, as these unique and large data sources are more accessible, big data has the poten- tial to illuminate nursing phenomena in a more rich and vibrant manner. Traditional data sources about symptom

Nursing Needs Big Data Brennan & Bakken

Figure 1. The role of theory in traditional and data science–driven nursing research.

status, management strategies, and outcomes such as patient reports, standardized survey measures, and EHR data can now be complemented by other data sources to enhance understanding of the symptom experience and tailor intervention strategies. These include biosensors, mobile apps, and “-omics” such as data related to the genome (e.g., SNP mutations), metabolome, and micro- biome. Second, distributed networks enable processing and analysis of big data across hundreds of nodes using query and analytic platforms supported by thousands of developers. Thus, there is a large modeling and analytic resource. Third, big data provide new pathways to knowledge informed but not constrained by theory (Figure 1). Instead of theory bounding the research process in a way that constrains the data obtained and the ways it is analyzed, in data science, theory provides a lens that is applied in a variety of ways in the pro- cess of obtaining, scrubbing, exploring, modeling, and interpreting the data.

Nursing inquiry has and will continue to follow a sys- tematic process in which data are selected and acquired, stored and structured in accordance with the theory driving the question resolution process. Exploratory data analyses increasingly relying on visualization ap- proaches precede the manipulation using inferential or nonparametric models. Knowledge results from the process. In Figure 1, the left image depicts a traditional approach to research and reflects the familiar-to-nursing theory-driven approach, in which the complex interplay between questions and theory resolves with the selection of a theory to guide the study. Theory then drives what data are sought and what are rejected (Brennan, 2008), and how data are stored, structured, explored through visualization and other approaches, and manipulated in

an analysis process from which knowledge ultimately results and is interpreted.

In contrast, the image on the right in Figure 1 depicts a data science approach in which the data already exist and the storage of the data is out of the control of the investigating team. Theory enters the process at the point of structuring the data for inquiry, and at subsequent points. Theory ensures an efficient approach to exploration through visualization and manipulation, preventing the proverbial random walk through data that seeks and finds serendipitous discovery only after expending large resources.

Nurse scientists are taking up the opportunities af- forded by big data and data science in some interesting ways. For several decades, nurse scientists, working in interdisciplinary teams, have applied some of the mining and modeling techniques now associated with big data and data science to phenomena central to nursing and health. These experiences lay the foundation for application and refinement of such techniques using big data rather than traditional data sources, including claims databases, single-site EHR data, and surveys. For example, Abbott et al. (1998) conducted early research using KDD approaches to predict patient transition from long-term care to acute care using the minimum data set. Clancy et al. (2007) have led a series of studies that apply the systems science perspective as well as computational models and simulation for purposes such as predicting the impact of the EHR on practice patterns. In terms of survey data, Merrill, Keeling, and Carley (2010) collected data from 11 public health departments and used organizational network analysis, a computational method derived from social network analysis, to examine the linkage between network structure and performance

Brennan & Bakken Nursing Needs Big Data

on essential public health services. Using social media as a big data source, Yoon and colleagues applied data mining and other computational techniques to a large corpus of Tweets related to physical activity to assess content, sentiments, and network structures (Yoon & Bakken, 2012; Yoon, Elhadad, & Bakken, 2013).

The key starting point to take advantage of the oppor- tunities afforded by big data and data science methods as well as to advance the methods is to begin with asking good, well-focused questions. For example, to better understand how context shapes health information management, Brennan, Ponto, Casper, Tredinnick, and Broecker (2015) used point cloud data to build high-fidelity three-dimensional (3D) replications of 20 homes. Each house generates a data set of over 950,000 data points, which exceeds the capacity of cur- rent CPU processing and graphics card capabilities. This vizHOME team will later visualize these 3D replications in an immersive virtual reality environment (Casper, Brennan, Perreault, & Marvin, 2015) and conduct a series of experiments involving professionals and lay people. Exploration of data related to intimate physical spaces facilitates the generation of research questions and new insights related to the study of personal health information management systems and extends the foundational work of scholars examining context and space. Moreover, the work has advanced data science methods through the development of a distributed model for processing point cloud data (Brennan et al., 2015).

Documentation of patient status, nursing interven- tions, and patient outcomes in EHRs extends the patient phenotypes generated from such systems beyond de- mographics, medical diagnoses, medical procedures, and laboratory values. For example, Dykes et al. (2010) have characterized patients in terms of their risk for falls and used the falls risk score to apply interventions tailored to risk level. Nurses have led efforts to establish the necessary standards to support inclusion of nursing data in a variety of EHRs and establish comparability across sites and settings at the national (Harris et al., 2015; Westra et al., 2015) and international level (Hardiker, Kim, Bartz, Coenen, & Jansen, 2013). As part of in- terdisciplinary teams, nurses have also contributed to integration of genomic data and knowledge into EHRs (Jing, Kay, Marley, Hardiker, & Cimino, 2012).

Big data and data science could be of value to nursing across the profession from clinicians to investigators. For example, effective use of big data can make real the promises of the Learning Health System through applications such as enhancing comparative effective- ness surveillance, enabling opportunity monitoring, supporting adverse event identification, and enhancing

public health surveillance (D. Meeker, personal com- munication, 2014).

Big Data and Data Science Need Nursing

The tsunami of data brings with it so many temptations of hidden insights and new knowledge. Such insights and knowledge may emerge through opportunistic exploration driven by random scurrying through large data sets. This approach is not only time consum- ing but may be distracting, teasing investigators with spurious insights, or harmful, consuming scarce re- sources. Nursing’s long tradition of theory-driven science provides the frameworks that can guide explorations towards promising phenomena and leverage insights into new knowledge, thereby avoiding the distractions of opportunistic exploration.

The premise that big data and data science need nurses follows from the American Nurses Association’s (ANA’s) Social Policy Statement, which delineates nursing expertise as the diagnosis and treatment of human re- sponse, and advocacy in the care of individuals, families, communities, and populations (ANA, 2003). The 2010 update specifically delineates the role of nurses in the generation and application of knowledge and technology to healthcare outcomes and planning for health policy and regulation that is responsive to consumer needs and provides for best resource use in the provision of health care for all (ANA, 2010). Consequently, it is vital that nurses apply their expertise in multiple areas of relevance to big data and data science, including defining important questions, extending data sources, applying data mining and modeling methods, and addressing ethical, legal, and social implications (ELSI).

Nurses bring extensive expertise related to ELSI in working with populations at high risk for health disparities and can provide insights related to the rep- resentativeness of big data sources and the potential scientific biases that can occur. Consider the earlier example of the Learning Health System within the context of creating a fabric of trust for the digital infras- tructure for such a system. The Institute of Medicine (IOM) suggests that for the vision of the Learning Health System to be achieved, patients should (a) participate in the development of a robust data utility; (b) use new clinical communication tools, such as personal portals, for self-management and care activities; and (c) be involved in building new knowledge through patient- reported outcomes and other knowledge processes (IOM Committee on the Learning Health Care System in America, 2012). However, racial and ethnic minori- ties are under-represented in clinical research (Gwadz et al., 2010), biorepositories (Bussey-Jones et al., 2010;

Nursing Needs Big Data Brennan & Bakken

Table 1. Roles of Data Intensive Nurses in Practice and Nurse Researchers

Data-intensive nurses in practice Data-intensive nurse researchers

Bachelor’s in nursing Advanced practice nurse Data-intensive nursing PhD Nurse data scientist

Training BS/BA/BSN with courses in

data science in the context

of evidence-based nursing

practice

DNP with courses in data

science methods

PhD in nursing; minor,

concentration in or

post-doc in data science

PhD in methodological

specialty; minor,

concentration in or

post-doc in nursing

Roles Nursing expert in

interprofessional teams

Critical assessment and

application of findings

generated through data

science methods

Clinical expert in

interdisciplinary teams

Evaluator and user of findings

generated through data

science methods

Lead a program of research

supported by data science

methods

Lead program of research in

data science informed by

discipline

Activities Implement data policies

Contribute to knowledge

development from the

bedside

Contribute to devising

pathways of informed

practice

Oversee and implement data

policies

Initiate knowledge

development from the

bedside

Devise pathways of informed

practice

Conduct inquiry into basic

nurse phenomena

supported by data science

methods

Generate new methods

informed by the discipline’s

phenomena of concern and

knowledge building

traditions

Green et al., 2006), use of the Internet for health-related purposes (Lee, Boden-Albala, Larson, Wilcox, & Bakken, 2014), and some types of mobile apps (Fox & Dug- gan, 2013), thus influencing the big data available for analysis to inform the Learning Health System or to conduct research in a particular topic area. To enhance understanding of this issue in a Latino community survey cohort, Bakken, Yoon, and Suero-Tejeda (2015) have examined the predictors of consent to provision and use of biospecimens in four different contexts, linkage of community survey data with EHR data, and agreement to future contact for research by other investigators.

Nursing’s understanding of the whole person com- plements the reductionist approach taken by many data scientists. It is useful to remind our data science colleagues that those Tweets, alleles, and video streams are attached to a person, and their interpretation in light of that individual’s life goals and priorities is something that nursing can guide. The value to people and society of a discovery that a single-gene mutation predicts the occurrence of Alzheimer’s disease 40 years in the future (Darst et al., 2015) will be of greater value if complemented by nursing interventions that help indi- viduals with knowledge of the mutation to enact a mean- ingful life in light of information that may be troubling or disturbing.

In addition to enriching data science processes with theoretical frameworks that avoid spurious insights and make efficient use of research resources, data science can

benefit from nursing by (a) introduction to the unique data types and sets that nurses are more familiar with, such as symptom inventories and parenting dynamics, (b) familiarity with novel data types such as videos of family interaction and self-tracking and self-monitoring information, and (c) bringing patient–family and com- munity engagement perspectives. Nursing’s contribution to big data and data science target three areas: (a) offering definition and context to data elements, (b) expertise in the use of theories to organize variables and interpret analysis results, and (c) creating interventions that assist patients in interpreting and acting on the information afforded through data science investigations. Nursing’s long history of work in standardized terminologies and formal concept representation identified patient- centered concepts and indicators as well as nursing and self-management interventions that can be populated by sensor data or by newly defined data collection strategies. While creating novel computer architectures lies largely outside of the purview of nursing, establishing the data models that organize data once it is acquired clearly lies within nursing’s purview. The meaningful, comprehen- sive interpretation of the results of data science inquiries can be clearly informed by nursing.

How Does Nursing Participate in Big Data and Data Science?

Given our premise that nursing needs big data and big data needs nursing, where do we want to end up

Brennan & Bakken Nursing Needs Big Data

and how do we get there? We strive for data-informed nursing practice, in which clinicians caring for patients have greater understanding of the patient experience through data science and can use the more comprehen- sive view of the person to devise creative approaches to interventions and monitor the effectiveness of the interventions. These nurses also have a unique point of contact for assuring patient understanding of data rights and assisting in data collection. This is best accomplished by the integration of data science principles across the nursing practice curricula.

We envision the growth of nursing knowledge enabled by a cadre of nurse researchers with skills in data science. Data-intensive nurse researchers have two pathways for action in the big data era: as nurse researchers with a particular focus on data science and participation in big data initiatives and as nurse data scientists. Table 1 delineates practice and research roles for nursing and training pathways.

The challenge to nurse educators is to design curricula at the baccalaureate and graduate levels that integrates data science throughout. Judicious referral to courses from supporting disciplines (e.g., computer science, statistics) should be guided by the learning needs of the student.

Conclusions

Nursing needs big data, and big data needs nursing. As a profession we have much to gain and much to con- tribute to a healthcare system informed by the discover- ies enabled by data science. As not every nurse will be, or should be, prepared as a data scientist, but all nursing practice and research should be informed by data science, the pathway to the future is through partnerships and team science. We offer these observations and reflections to advance the discussion among nurses in practice, edu- cation, and research.

Acknowledgments

Manuscript preparation was partially supported by vizHome: a context sensitive health information needs assessment strategy (R01HS02254; Brennan); Smart Asthma Management (NSF 1343969; Brennan); WICER 4 U (R01HS022961; Bakken); and the New York City Hispanic Dementia Caregiver Research Program (R01NR14430; Bakken). The insights and generosity of Daniella Meeker, PhD, UCSD, and the editorial as- sistance of Andrew Moreland are acknowledged with gratitude.

Clinical Resources � Johns Hopkins Data Scientist Toolbox – This free

on-line course will help you identify and clas- sify data science problems. It is the first course in the Data Science Specialization: https://www. coursera.org/course/datascitoolbox

� ChartBuilder is an online chart making tool. You can enter your data manually and the program will organize it into rows and columns and create charts on the fly: http://quartz.github.io/Chartbuilder/

� Nursing Knowledge 2015 – This site is the hub of the University of Minnesota School of Nurs- ing’s annual big data conference. It includes case studies and reports from the meeting: http:// www.nursing.umn.edu/about/calendar-of-events/ 2015-events/nursing-kno–wledge-2015-big-data- science-conference/index.htm

References

Abbott, P. A., Quirolgico, S., Candidate, D., Manchand, R.,

Canfield, K., & Adya, M. (1998). Can the US minimum

data set be used for predicting admissions to acute care

facilities? Studies in Health Technology and Informatics, 52(Part

2), 1318–1321.

Ahalt, S., Bizon, C., Evans, J., Erlich, Y., Ginsberg, G.,

Krishnamurthy, A., . . . Wilhelmsen, K. (2014). Data to

discovery: Genomes to health. A white paper from the National

Consortium for Data Science. Chapel Hill, NC: Renaissance

Computing Institute. Retrieved from

http://data2discovery.org/dev/wp-content/uploads/2014/

02/NCDS-Summit-2013.pdf

American Nurses Association. (2003). Nursing’s social policy

statement: The essence of the profession (2nd ed.). Silver Spring,

MD: Author.

American Nurses Association. (2010). Nursing’s social policy

statement: The essence of the profession (3rd ed.). Silver Spring,

MD: Author.

Bakken, S., Yoon, S., & Suero-Tejeda, N. (2015). Factors

influencing consent for electronic data linkage in urban Latinos.

Presented at MedInfo 2015, August 19–23, 2015, San

Paolo, Brazil.

Brennan, P. F. (2008). Standing in the shadows of theory.

Journal of the American Medical Informatics Association, 15(2),

263–264. doi:10.1197/jamia.M2691

Brennan, P. F., Ponto, K., Casper, G. R., Tredinnick, R., &

Broecker, M. (2015). Virtualizing living and working

spaces: a proof of concept. Journal of Biomedical Informatics,

57, October 2015, pp 53–61. doi: 10.1016/j.jbi.2015.07.007

Bussey-Jones, J., Garrett, J., Henderson, G., Moloney, M.,

Blumenthal, C., & Corbie-Smith, G. (2010). The role of

race and trust in tissue/blood donation for genetic research.

Nursing Needs Big Data Brennan & Bakken

Genetics in Medicine, 12(2), 116–121.

doi:10.1097/GIM.0b013e3181cd6689

Casper, G. R., Brennan, P. F., Perreault, J. O., & Marvin, A. G.

(2015). vizHOME—A context-based home assessment:

Preliminary implications for informatics. Presented at MedInfo

2015, August 19–23, 2015, San Paolo, Brazil.

Clancy, T. R., Delaney, C. W., Segre, A., Carley, K., Kuziak,

A., & Yu, H. (2007). Predicting the impact of an electronic

health record on practice patterns using computational

modeling and simulation. AMIA Annual Symposium

Proceedings, 2007, 145–149. Retrieved from http://www.

ncbi.nlm.nih.gov/pmc/articles/PMC2813672/

Collins, F. S., & Varmus, H. (2015). A new initiative on

precision medicine. New England Journal of Medicine, 372(9),

793–795. doi:10.1056/NEJMp1500523

Darst, B. F., Koscik, R. L., Hermann, B. P., La Rue, A., Sager,

M. A., Johnson, S. C., & Engelman, C. D. (2015).

Heritability of cognitive traits among siblings with a

parental history of Alzheimer’s disease. Journal of

Alzheimer’s Disease, 45(4). doi:10.3233/jad-142658

Dykes, P. C., Carroll, D. L., Hurley, A., Lipsitz, S., Benoit, A.,

Chang, F., . . . Middleton, B. (2010). Fall prevention in

acute care hospitals: A randomized trial. Journal of the

American Medical Association, 304(17), 1912–1918.

doi:10.1001/jama.2010.1567

Fleurence R. L., Curtis L. H., Califf R. M., Platt, R., Selby, J. V.,

& Brown, J. S. (2014). Launching PCORnet, a national

patient-centered clinical research network. Journal of the

American Medical Informatics Association, 21(4), 578–582.

doi:10.1136/amiajnl-2014-002747

Fox, S., & Duggan, M. (2013). Tracking for health. Retrieved

from http://www.pewinternet.org/2013/01/28/tracking-

for-health/

Green, D., Cushman, M., Dermond, N., Johnson, E.A., Castro,

C., Arnett, D., . . . Manolio, T.A. (2006). Obtaining

informed consent for genetic studies. The multiethnic study

of atherosclerosis. American Journal of Epidemiology, 164(9),

845–851. doi:10.1093/aje/kwj286

Gwadz, M. V., Colon, P., Ritchie, A. S., Leonard, N. R.,

Cleland, C. M., Riedel, M., . . . Mildvan, D. (2010).

Increasing and supporting the participation of persons of

color living with HIV/AIDS in AIDS clinical trials. Current

HIV/AIDS Report, 7(4), 194–200.

doi:10.1007/s11904-010-0055-3

Hardiker, N., Kim, T. Y., Bartz, C. C., Coenen, A., & Jansen,

K. (2013). Collaborative development and maintenance of

health terminologies. AMIA Annual Symposium Proceedings,

2013, 572–577. Retrieved from http://www.ncbi.nlm.nih.

gov/pmc/articles/PMC3900200/

Harris, M.R., Langford, L. H., Miller, H., Hook, M., Dykes, P.

C., & Matney, S. A. (2015). Harmonizing and extending

standards from a domain-specific and bottom-up approach:

An example from development through use in clinical

applications. Journal of the American Medical Informatics

Association, 22(3), 545–552. doi:10.1093/jamia/ocu020

Institute of Medicine Committee on the Learning Health Care

System in America. (2012). Best care at lower cost: The path to

continuously learning health care in America. Washington, DC:

National Academies Press.

Jing, X., Kay, S., Marley, T., Hardiker, N. R., & Cimino J. J.

(2012). Incorporating personalized gene sequence variants,

molecular genetics knowledge, and health knowledge into

an EHR prototype based on the Continuity of Care Record

standard. Journal of Biomedical Informatics, 45(1), 82–92.

doi:10.1016/j.jbi.2011.09.001

Laney, D. (2001). 3D data management: Controlling data volume,

velocity, and variety. Retrieved from http://blogs.gartner.

com/doug-laney/files/2012/01/ad949-3D-Data-

Management-Controlling-Data-Volume-Velocity-and-

Variety.pdf

Lee, Y. J., Boden-Albala, B., Larson, E., Wilcox, A., & Bakken,

S. (2014). Online health information seeking behaviors of

Hispanics in New York City: A community-based

cross-sectional study. Journal of Medical Internet Research,

16(7), e176. doi:10.2196/jmir.3499

Mason, H., & Wiggins, C. (2010). A taxonomy of data science.

Retrieved from http://www.dataists.com/2010/09/a-

taxonomy-of-data-science/

Merrill, J., Keeling, J. W., & Carley, K. M. (2010). A

comparative study of 11 local health department

organizational networks. Journal of Public Health

Management Practice, 16(6), 564–576.

doi:10.1097/PHH.0b013e3181e31cee

National Institutes of Health Big Data to Knowledge. (2014).

Workshop on enhancing training for biomedical big data.

Retrieved from http://bd2k.nih.gov/pdf/bd2k_training_

workshop_report.pdf

Wang, W., & Krishnan, E. (2014). Big data and clinicians: A

review on the state of the science. JMIR Medical Informatics,

2(1), e1. doi:10.2196/medinform.2913

Westra, B. L., Latimer, G. E., Matney, S. A., Park, J. I.,

Sensmeier, J., Simpson, R. L., . . . Delaney, C.W. (2015). A

national action plan for sharable and comparable nursing

data to support practice and translational research for

transforming health care. Journal of the American Medical

Informatics Association, 22(3), 600–607.

doi:10.1093/jamia/ocu011

Yoon, S., & Bakken, S. (2012). Methods of knowledge discovery in

tweets. Nursing Informatics: Proceedings of the

International Congress on Nursing Informatics, 2012, p.

463. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/

articles/PMC3799133/

Yoon, S., Elhadad, N., & Bakken, S. (2013). A practical

approach for content mining of Tweets. American Journal of

Preventive Medicine, 45(1), 122–129.

doi:10.1016/j.amepre.2013.02.025

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.