Skip to main content
CCTSI

Informatics

Go Search
CCTSI
About Us
CTRC
Research Resources
Informatics
Education & Training
Community Engagement
Funding Opportunities
Child & Maternal Health
Contact Us
  

CCTSI > Informatics > What is Translational Informatics?
 

What is Translational Informatics?

Translational InformaticsTranslational informatics is simply research informatics applied to translational research: research designed to accelerate translation of scientific discoveries from the bench to medical care at the bedside, improving the health of the community.

 
   

 What is Biomedical Informatics?

Informatics in general

A reasonable general definition of the academic field of informatics is provided by the Wikipedia entry for the term (accessed 2009-09-29): "Informatics is the science of information, the practice of information processing, and the engineering of information systems. Informatics studies the structure, algorithms, behavior, and interactions of natural and artificial systems that store, process, access and communicate information." It incorporates academic disciplines including artificial intelligence, cognitive science, computer science, information science, and social science. The general goal of informatics is to develop systems that assist users by making it easy to store, retrieve, and use data, information, and knowledge.

Biomedical informatics

Biomedical informatics (or more broadly, biomedical and health informatics) is the application of informatics in the biomedical realm, concerned with "the optimal use of information, often aided by the use of technology, to improve individual health, health care, public health, and biomedical research" (Hersh 2009). Friedman's "fundamental theorem" of biomedical informatics is that a patient, clinician, or researcher working in partnership with an information resource is more effective than that same person unassisted:

As an example, consider the biomedical examples of data, information, and knowledge in the table below. Note how representations of data developed by informaticists (gene number, NDC value) allow simple data to be linked with useful information and a rich set of knowledge resources developed by the National Library of Medicine for researchers (OMIM) and clinicians and patients (DailyMed).

Data (single points of observation)

Gene number 606822

NDC (National Drug Code) 0006-0661-68

Information (providing meaning for data)

This gene encodes for a copper-transporting ATPase

This is trientine, a drug used to treat Wilson disease

Knowledge (interpretation of information)

A defect in this gene causes Wilson disease, which causes liver and neuropsychiatric disease: OMIM: Online Mendelian Inheritance in Man

Pharmacology, administration, dosage, and warnings associated with trientine: DailyMed

While this is a simple example, informatics makes it possible to develop much more sophisticated tools for researchers, clinicians, and patients to make sense of data (such as laboratory values or gene expression data) and to draw new inferences that will advance science and medical care.

An academic discipline and a service

It may not be surprising that as a field encompassing such heterogeneous domains (such as computer science, social science, and biostatistics), has indistinct boundaries with other academic disciplines. The other "fuzzy boundary" for biomedical informatics is the fact that, like other disciplines such as biostatistics, it is both an academic discipline of its own and a service that is provided for researchers, clinicians, and patients. Informatics professionals can provide assistance in diverse domains such as setting up databases; analyzing high-throughput genomic, proteomic, and metabolomic studies; extracting data from clinical information systems; and designing applications for use by the general public. However, typically a distinction is drawn between the work of informaticists (how data and relationships among data should be represented, stored, and transmitted) and the work of information systems professionals (designing, purchasing, installing, maintaining, and supporting computers, networks, and software).

 Biomedical and Health Informatics

William Hersh has taken an especially active role in defining "biomedical and health informatics" (Hersh 2009). This broad term is rarely in use, with most informaticists using more specific adjectives to describe their domain of interest. Hersh provides a schematic of the major subcategories of informatics:

Some major categories of biomedical and health informatics are described below, with local contacts among the CCTSI partners:

Bioinformatics addresses issues related to the analysis of data collected on biological processes, particularly high throughput data collected through processes such as DNA microarrays.

Clinical informatics (such as medical informatics and nursing informatics) addresses issues related to the provision of medical care, such as the electronic medical record, computerized provider order entry for testing (laboratory, radiology) and therapy (pharmacy, nursing, etc.), clinical decision support systems, and telemedicine. The related field of consumer health informatics addresses issues related to patients and other consumers, such as personal health records and other applications to support better health.

Public health informatics addresses issues related to public health, such as outbreak preparation, detection (biosurveillance), and response. Given the need to coordinate information systems from public health agencies on the local, state, and national level (e.g. the formation of the National Electronic Disease Surveillance System and the Public Health Information Network) public health informatics has played a leading role in health information exchange efforts.

  • Dennis Lezotte is the primary contact for public health informatics in the Colorado School of Public Health.
  • Arthur Davidson is the primary contact for public health informatics at Denver Public Health.

Consumer health informatics is focused on empowering patients, by linking patients with health professionals and combining data from self-monitoring, personal health information and consumer health information, using the Internet as well as devices such as touchscreen kiosks and mobile phone applications.

  • Steve Ross in the School of Medicine and Diane Skiba in the College of Nursing are local contacts for consumer health informatics.

Research informatics is "the use of informatics to facilitate biomedical and health research" (Hersh, 2009). As shown in the figure above, research informatics crosses domains of bioinformatics, medical (clinical) informatics, and public health informatics. A particularly active active area of research informatics is the development of information systems to support clinical research. Since much clinical research takes place in the practice setting, Integrating clinical information systems with research information systems remains an important but difficult challenge. (Kahn 2007).

 Translational Informatics

Translational informatics is simply research informatics (itself an overarching term) applied to translational research: research designed to accelerate translation of scientific discoveries from the bench to medical care at the bedside, improving the health of the community.

While translational research properly applies to both type 1 and type 2 translation, translational bioinformatics in type 1 research has been particularly active. The American Medical Informatics Association has defined translational bioinformatics as

“. . . the development of storage, analytic, and interpretive methods to optimize the transformation of increasingly voluminous biomedical data into proactive, predictive, preventative, and participatory health. Translational bioinformatics includes research on the development of novel techniques for the integration of biological and clinical data and the evolution of clinical informatics methodology to encompass biological observations. The end product of translational bioinformatics is newly found knowledge from these integrative efforts that can be disseminated to a variety of stakeholders, including biomedical scientists, clinicians, and patients."

 Key Issues in Translational Informatics

Standards for Representation and Transmission of Information

The importance of standards for representation of data can be demonstrated in Figure 1. Different researchers could refer to the same gene as "copper gene," or "Wilson disease gene," and different clinical systems could refer to trientine by the string name "Syprine" or a myriad of other proprietary codings (such as those from First Databank, Medispan, etc.). All of these might work perfectly well within a single research laboratory or a single clinical system, but without common representations (such as the gene number in NCBI) or a method of "normalizing" codes to a common meaning (such as the National Library of Medicine's RxNorm), the information will remain siloed and its full utility will not be realized. In contrast, when data is encoded for interoperability and reusability:

  • The research team generating the data can link it to other information resources to gain a more sophisticated understanding of its meaning and implications.
  • The data can also be shared more easily with collaborators and the research community.
  • The data can be machine-readable, and subjected to logic (such as decision support tools that can alert, for instance, when two incompatible drugs are being ordered for the same patient, or when contraindications such as kidney failure exist).

The more data that are stored using common representations, the more critical mass there will be for the development of sophisticated research and clinical tools. Some commonly used medical informatics standards are shown in the table below:

Standard

Domain

LOINC (Logical Observation Identifiers and Codes)

Laboratory data

ICD-9 (international Classification of Diseases)

Diagnoses

DICOM (Digital Imaging and Communications in Medicine)

Radiographic and other imaging data

SNOMED CT (Systematized Nomenclature of Medicine - Clinical Terms)

Medical terms and concepts

Standards are also useful for transmission of data, particularly in the clinical realm. Clinical systems in research hospitals are rarely (if ever) monolithic, single-vendor systems. Standards such as Health Level 7 (HL7) ensure that information from radiology, the clinical laboratory, the pharmacy, and the hospital floor can be synthesized into a cogent whole for the researcher or the clinician.

Identity Matching and De-Identification

Matching identities across systems can be helpful in drawing new inferences about diseases. Consider the potential knowledge that could be gained from knowing that the same person seen today at University of Colorado Hospital for one condition was seen at Denver Health years ago for a potentially predisposing condition, and has proteomic data and frozen specimens on hand from a previous research study. Since no single national patient identifier exists (and Social Security Numbers are not adequate proxies), probabilistic algorithms must be used to automatically match patient identities across systems.

At the same time, HIPAA (the Health Information Portability and Accountability Act) guarantees patients the rights to control the flow of their personal health information. Protecting privacy through deidentification of datasets, while still enabling cross-institutional linkages, remains a major challenge for translational informaticists.

Data Organization and Storage

The construction of research and clinical databases can enable or constrain research. For instance, clinical databases may be very efficient at pulling all the information for a single patient on demand, but very inefficient at cross-patient queries (for instance, identifying all patients with juvenile rheumatoid arthritis). Meeting both needs may require supplementing clinical databases with data warehouses that facilitate reports and data mining techniques.

Visualization of Data

To make data meaningful, we should take full advantage of the 30 percent of neurons in the cerebral cortex devoted to visual processing. Beyond simple line and bar graphs, it is possible to identify and communicate patterns through techniques such as three dimensional rendering of proteins and geocoding public health data. An elegant and inspiring demonstration of effective visualization is provided by Hans Rosling.

 References and Resources

Shortliffe EH, Cimino JJ (eds). Biomedical Informatics: Computer Applications in Health Care and Biomedicine (3rd Ed). New York: Springer, 2006. ISBN-13: 978-0-387-28986-1

Butte AJ. Translational bioinformatics: coming of age. J Am Med Inform Assoc 2008; 15:709-14. PMID: 18755990

Friedman CP. A "fundamental theorem" of biomedical informatics. J Am Med Inform Assoc 2009; 16:169-70. PMID: 19074294

Hersh W. A stimulus to define informatics and health information technology. BMC Medical Informatics and Decision Making 2009, 9:24. PMID: 19445665

Hersh W. Medical informatics: improving health care through information. JAMA 2002; 288:1955-1958. PMID: 12387634.

Kahn MG, Kaplan D, Sokol RJ, DiLaura RP. Configuration challenges: implementing translational research policies in electronic medical records. Acad Med 2007: 82:661-9. PMID: 17595562.