Translating Information to Transform Health Care
Arkalgud Ramaprasad, Annette L. Valenta
University of Illinois at Chicago, Chicago, IL 60607, U.S.A.
Ian Brooks
National Center for Supercomputing Applications,Urbana, IL 61801, U.S.A.
Keywords: Clinical and translational science, Informatics, Ontology, Ontological analysis.
Abstract: Clinical and Translational Science (CTS) is a new and emerging academic discipline which seeks to reduce
(a) the time-to-application of research to, and (b) the time-to-research of, health care problems. Translating
information within and between the research and practice silos is central to CTS. The role of CTS
Informatics (CTSI) can be stated as ‘translating information to transform health care’. We present an
ontological analysis of the transformation of health care by CTSI. The five dimensions of the ontology are
derived by parsing the above definition of CTSI. They are: (a) information, (b) semiotic translation, (c)
spatial translation, (d) temporal translation, and (e) health care. Each dimension is defined by a taxonomy.
Each sentence, formed by concatenating categories across the five dimensions using appropriate prefixes
and conjunctive words and phrases, is a natural language descriptor of CTSI. The set of all such sentences is
a closed description of CTSI.
Clinical and Translational Science, at its core, is
information intensive. The guiding mission of CTS
Informatics (CTSI) is to develop, support, and
continuously improve the research workflow to
reduce the time-to-application of basic research and
the time-to-research of clinical and community
health problems. CTS’ success will depend upon the
“bi-directional information flow between basic and
translational scientists….” (Zerhouni 2005, p. 1621)
It is expected that the “CTSA [CTS Awards]
institutions will work as part of a national effort to
expand and improve clinical research informatics to
share data across disciplines and across
institutions….New and expanded IT will help to
solve issues related to workflow, usability, and
interoperability with collaborating organizations,
along with the need to ensure privacy and
confidentiality of human subjects.” (Zerhouni 2007,
p. 127) The informatics infrastructure will be
essential “[t]o develop the preemptive, predictive
medicine of the 21
century.” (Zerhouni 2005, p.
1355) Overall “the NIH aims to develop a national
system of interconnected clinical research networks
capable of more quickly and efficiently mounting
large-scale clinical studies. As currently conceived,
this system of networks will integrate and expand
extant research networks using common or
interoperable infrastructure, including harmonized
informatics, governance, terminology, and training.”
(Zerhouni 2005, p. 1356)
Thus, we can succinctly encapsulate the above
visions and objectives of CTSI in the statement
‘translating information to transform health care.’
Using this statement we will deconstruct the
complexity of CTSI.
The following ontological analysis (Ladkin
2005) of the transformation is a step to capture
CTSI’s complexity with natural language
descriptions using a structured terminology. We
address the complexity using an ontology derived by
parsing the above stated definition of CTSI. We
synthesize the extant literature on CTSI using the
ontology. The analysis is used to develop a strategy
for CTSI. At a time when CTS is still developing a
CTSI strategy will be critical to its wellbeing.
This method of logical analysis, synthesis of the
literature, interpretation, and application is
systematic, repeatable, and extensible. It is also
novel. In the following we will first present the
ontology derived by parsing our definition of CTSI.
Then, we will discuss the topics corresponding to the
Ramaprasad A., L. Valenta A. and Brooks I. (2009).
CLINICAL AND TRANSLATIONAL SCIENCE INFORMATICS - Translating Information to Transform Health Care.
In Proceedings of the International Conference on Health Informatics, pages 135-141
DOI: 10.5220/0001431701350141
ontology in the following sequence: (a) ontological
analysis, (b) CTSI ontology, (c) translating
information between research and practice; (d)
translating information semiotically, spatially, and
temporally; and (e) translating information to
transform health care.
Ontologies are used in informatics systems design to
standardize terminologies, map requirements,
organize them systematically, facilitate integration
of systems, promote information exchange between
systems, etc. (Gruber 1995; Gruber 2008)
Ontologies are related to but different from
taxonomies, typologies, concept hierarchies,
thesauri, and dictionaries. (Gilchrist 2003) They are
tools for systematizing the description of complex
systems (Cimino 2006); a way of deconstructing the
architecture of complexity (Simon 1962). Such
systematization, in turn, facilitates analysis and
design of these systems.
There is no standard definition of ontology of an
informatics system. We will define it as a logically
constructed n-dimensional natural language
description of the system. The dimensions are
derived from the problem statement, system’s
definition, or objectives. Each dimension is
independent of the other and is a taxonomy of
discrete categories. Each taxonomy may be flat or
hierarchical. Further, the order of categories in a
particular dimension at a particular level of the
taxonomy may be nominal (no particular order) or
ordinal (based on some parameter). The stages of
progression along the dimension, the sequence of
evolution, the progressive part-whole relationships,
the scale, etc. are some bases for ordering the
categories. Last, a dimension may have sub-
dimensions with their own taxonomies. That is, a
dimension itself may be hierarchical.
A combination of categories across all the
dimensions, with appropriate conjunctive words, is a
natural language descriptor of the system in the form
of a sentence – albeit a little awkward grammatically
sometimes. The set of all combinations across all
categories – that is all possible sentences – is a
closed description of the system. The full set can
have a very large number of descriptors (individual
combinations). However, many of the combinations
may be uninterpretable or empirically unviable –
they may be discarded from further analysis. At the
same time some combinations may be novel and
creative, providing valuable insights into the
properties and possibilities of the system.
The ontology we use for the analysis of CTSI has
five dimensions, namely: (a) information, (b)
semiotic translation, (c) spatial translation, (d)
temporal transportation, and (e) health care. The
logic of the derivation of the dimensions from the
definition of ‘translating information to transform
health care’ should be intuitively clear. We have
deconstructed translation into three dimensions: (a)
semiotic translation, (b) spatial translation, and (c)
temporal transportation. Information and health care
have been retained as such.
The five dimensions and their corresponding
taxonomies are shown in Figure 1 and discussed
below. The conjunctive prefixes, words, and phrases
to concatenate the columns are shown before and
between the columns. They make the concatenations
across dimensions natural and understandable. Five
illustrative combinations are shown at the bottom of
the figure. The ontology as presented can be
expanded into 7*6*4*6*7 = 7,056 combinations.
The above representation is a concise way of
representing them and analyzing them
systematically. A listing of all the combinations
would likely take about 100 pages.
This section focuses on the first and the last
dimensions of the ontology. They mirror each other
with the order reversed – the first dimension is
focused on information and the last on health care.
In a sense they can be seen as the input and the
output of CTSI, cross-linking research to practice
and practice to research, with the translation process
in between.
The dichotomy of research and practice in health
care (and other disciplines) is historical and
persistent. While there is some overlap between the
two, they are often seen as polar opposites. CTS
seeks to transform the relationship between the two
into a seamless continuum. Ideally perhaps they
should be like the Yin and the Yang, separate but
simultaneously containing the other – research
HEALTHINF 2009 - International Conference on Health Informatics
Information Semiotic Spatial Temporal Health Care
Physical Minutes Practice
Basic Data Internal Hours Clinical
Animal Analysis External Days Public health
Clinical Interpretation Virtual Weeks Community health
Public health Conclusion Internal Months Research
Practice Recommendation External Years Basic
Clinical Animal
Public health Clinical
Community health Public health
Illustrative combinations
Translating basic research conclusion in external virtual space in months to transform clinical practice.
Translating clinical practice observation in external physical space in months to transform basic research.
Translating public health research data in external virtual space in days to transform public health
Translating clinical practice recommendation in external virtual space in months to transform community
health research.
[to transform]
[space in]
Figure 1: CTSI Ontology.
embedding practice and practice embedding
We will use the research-practice dichotomy for
the taxonomy of information and of health care in
the ontology. There are many other taxonomies; any
one of them could be used in the ontology. A
different taxonomy would naturally yield a different
perspective on the issue. Our choice would be in
keeping with the actuality (Cicmil, Williams et al.
2006) of CTSI. Research information and health
care research are further categorized as basic,
animal, clinical, or public health; practice
information and health care practice as clinical,
public health, or community health. The four
categories of research have been construed to
connote two stages – from basic research to clinical
trials, and from clinical trials to clinical practice
(Sung, Crowley et al. 2003). The two-level
taxonomy shown in Figure 1 is adequate for the
present discussion. It can be extended or refined
subsequently if necessary.
At the core of CTSI is the translation of research
information into health care practice, and health care
practice information into research (Sung, Crowley et
al. 2003). The absence of these two types of
translation results in the “knowledge to action gap”
(Graham, Logan et al. 2006) or the “Evidence-to-
Practice Gap” (Lang, Wyer et al. 2007, p. 355) and
the “action to knowledge gap” – the last has not
received as much attention as the first two in health
care literature (Westfall, Mold et al. 2007). While
there is a lot of concern about the accumulation of
basic research that does not get translated into
clinical trials and then to clinical practice (Sung,
Crowley et al. 2003), there is some but not the same
amount of concern about the accumulation of
practice knowledge that does not get fed back to
inform the basic research and be validated by it. For
example, the “Institute of Medicine Strategies for
Knowledge Translation Related to Health Care
Quality” cited by Hedges (Hedges 2007) has no
feedback loop from practice/outcomes to
research/literature. Translation has to be “a 2-way
connection between the interstates of academic
scientific discoveries and the patients receiving care
in the ambulatory practice.” (Westfall, Mold et al.
2007, p. 404) Interestingly, the CTSA programs are
supposed to create “two-way synergies with local
and regional communities by reaching out to
underserved populations, community-based groups,
and healthcare providers.” (Zerhouni 2007, p. 127)
However their purpose is to help “deliver improved
medical care to the entire population, helping to
disseminate new technologies and new advances
into clinical practice.” (Zerhouni 2007, p. 127) This
could also take place by translating practice
knowledge into research and back to practice – thus
completing the feedback loop (Ramaprasad 1983).
There can be many barriers to and facilitators of
translation of research information to practice and
vice-versa (Sung, Crowley et al. 2003; Ghosh and
Ghosh 2005; Gaughan 2006; Anderson, Lee et al.
2007; Westfall, Mold et al. 2007). We categorize
them along the three dimensions of the ontology as
semiotic, spatial, or temporal. In the next section, we
discuss these three dimensions of translation and
how they can inhibit or engender translation, starting
with a lexical and linguistic discussion of translation
CLINICAL AND TRANSLATIONAL SCIENCE INFORMATICS - Translating Information to Transform Health Care
Translating information – we use information to
generically connote data, information, and
knowledge – is the key to CTS and hence to
bringing about the transformation it seeks. We
address the complexity of translation by breaking it
down into three dimensions: (a) semiotic, (b) spatial,
and (c) temporal translation. We discuss each of
these three and their taxonomies below.
5.1 Semiotic Translation
The process of translating research information to
practice and practice information to research to
improve health care is semiotic. It is an ongoing
series of iterative cycles of generation and
dissipation cutting across the semiotic layers of
morphology, syntax, semantics, and pragmatics
(Ramaprasad and Rai 1996; Ramaprasad and
Ambrose 1999; Ambrose, Ramaprasad et al. 2003;
Payne, Mendonca et al. 2007). In lay terms, it is the
process of moving from observation to data to
analysis to interpretation to conclusion to
recommendation, then feeding back into
observation. These steps form the taxonomy of the
semiotic dimension in the ontology (Figure 1).
The CTSI should support the semiotic
translation of information (a) by researchers and
practitioners, and (b) between and among
researchers and practitioners. The traditional
research process is one of semiotic translation by
researchers; the semiotic translation by practitioners
when it occurs is akin to the grounded research
(Glaser and Strauss 1964) process. The exchange of
information among and between researchers and
practitioners may occur at a semiotic level or cut
across many of them. Thus, for example, two
researchers may simply exchange data or
conclusions. Or, one researcher may send his data to
another researcher who may analyze it and send her
results back to the first researcher. The network of
researchers and practitioners involved in CTS is
likely to be far more complex than the above dyadic
examples; correspondingly the semiotic translation
entailed and hence CTSI has to support will be
complex too.
The informatics tools and techniques required to
move across the semiotic levels vary. For example,
in some types of laboratory research all the steps up
to and including analysis can be automated; on the
other hand in some qualitative field research none of
the steps can be automated. Similarly, a simple
interpretation may be communicated succinctly
without loss of fidelity, while communicating a
complex interpretation may require a
correspondingly complex explanation.
The study of the semiotics of translation is not
new. “Medical semiotics in the 18th century was
more than a premodern form of diagnosis. Its
structure allowed for the combination of empirically
proven rules of instruction with the theoretical
knowledge of the new sciences, employing the
relation between the sign and the signified.” (Hess
1998, p. 203) The semiotic dimension of translation,
however, has received little explicit attention
recently (seeGraham and Tetroe 2007, for example).
As Scott et al. discuss “the challenge of translating
evidence from SRs [systematic reviews], while
maintaining a sufficient level of validity and
relevance to satisfy both clinicians and researchers,
is rarely discussed.” (Scott, Moga et al. 2007, p.
681) The complexity of the semiotics is indicated by
their conclusion: “The key elements for creating
clinically relevant knowledge from SRs are: a
flexible, consistent and transparent methodology;
credible research; involvement of renowned content
experts to translate the evidence into clinically
meaningful guidance; and an open, trusting
relationship among all contributors.” (Scott, Moga et
al. 2007, p. 681) The lack of attention to semiotics is
particularly glaring at a time when the so called
semantic web (perhaps better called the semiotic
web (Ramaprasad and Kashyap 2008)) is under
development and is likely to have a major impact on
5.2 Spatial Translation
Researchers and practitioners who have to be
networked for CTS are likely to be distributed
internally within an organization, locally, regionally,
nationally, and even globally. Among them, the silos
of research and practice, and of the different
categories of research and practice, may not just be
artifacts of the mind fostered by specialized
disciplines but also manifest in their physical
location, their offices, and labs. The challenge for
CTSI is to spatially translate the information (a)
from research to practitioners, (b) from practice to
researchers, (c) from research to researchers, and (d)
from practice to practitioners, across the virtual and
physical silos.
The broad presumption of spatial translation is to
eliminate the physical location dependence of CTS
(Ambrose, Ramaprasad et al. 2003). The NECTAR
HEALTHINF 2009 - International Conference on Health Informatics
Network (Zerhouni and Alving 2006), The Family
Practice Inquiries Network and other practice-based
research-networks (Westfall, Mold et al. 2007), the
International Clinical Epidemiology Network,
(Tugwell, Robinson et al. 2006), and the Oklahoma
Physicians Resource/Research Network (OKPRN)
(Nagykaldi and Mold 2007) are examples of systems
set up for this purpose. In the same vein, the CTSA
mandates that all the centers should be networked.
Not only should any researcher or practitioner be
able to access the information ‘anywhere’, but
should be able to process it ‘anywhere’. Today, with
the internet, there is a rising expectation that
information be available ‘anytime, anywhere’.
Today’s information and communication
technologies – the internet is one of them – have not
only altered the constraints of physical space but
have also created an entirely new virtual space. The
dynamics of the physical and virtual spaces affect
and are affected by each other. The capabilities of
the virtual space can be used to overcome the
constraints of the physical space and vice-versa.
Thus, CTSI can be used to create a virtual space that
complements the capabilities and constraints of the
physical space in which its users operate. It must be
noted that the exchange of information in the virtual
space has not obviated the need for exchange in the
physical space – despite e-mails and webinars face-
to-face conversations and meetings continue to be
important. The barriers to and facilitators of spatial
translation for CTS have to be understood in the
context of the convergence of the physical and
virtual space.
There is almost always a distinction between the
‘internal’ and the ‘external’ in discussing physical
space and virtual space. The boundary between the
two can be an important barrier to or filter of the
information translated. The rules governing the
translation internally – like security, privacy, and
confidentiality rules – are different from those for
translation externally. The boundary separating the
two may itself be arbitrary or adaptive to the
context. Thus for some information everybody in the
organization may be internal, but for others only the
members of the research group may be internal.
Despite the fuzziness and variability of the boundary
the internal-external distinction is an important
consideration in the spatial translation in CTSI.
Thus, there are four categories of spatial
translation in the ontology: (a) internal-physical, (b)
external-physical, (c) internal-virtual, and (d)
external-virtual. Each of these can play a different
role in the translation of research to practice and
practice to research. In a given context a mix of
them may be used. The CTSI should facilitate and
remove barrier to the use of all four.
5.3 Temporal Translation
The temporal dimension is intrinsic in the objectives
of CTS to minimize the time-to-practice and the
time-to-research. It is also implicit in the concept of
preemptive and predictive medicine (Zerhouni
2005). The scale of these times varies by context.
Bringing the current best research evidence to bear
upon the diagnosis of a patient in the emergency
room may have to be done in minutes (Holroyd,
Bullard et al. 2007); research on and response to an
epidemic such as SARS may spread over weeks and
months; response to Avian Flu (Eysenbach 2003)
can be planned months or years ahead and activated
in hours or days; and taking a drug from discovery
to clinical deployment may take over ten years. For
example, “[i]n the first documented instance of bird-
to human infection with the H5N1 flu virus in 1997,
Hong Kong reacted by destroying its entire poultry
population of 1.5 million birds within three days.”
(Webster and Hulse 2005, 415) For another
example, in the case of SARS the first “[u]nusual
atypical pneumonia was documented in Foshan,
Guangdong Province, China” in November 2002;
the virus was identified in March 21-27, 2003; and
the full genome was mapped by April 12, 2003.
(Peiris, Yuen et al. 2003, p. 2432)The total time was
less than six months. The role of a CTIS in a SARS-
like epidemic is highlighted by the recommendation
for an “efficient information technology systems that
provide a means to link clinical, epidemiological,
and laboratory data on SARS cases and to
disseminate this information locally, nationally, and
globally, and systems that allow rapid identification,
tracking, evaluation, and monitoring of contacts of
SARS cases.” (Parashar and Anderson 2004, p. 632)
Thus, in the temporal dimension of the ontology
we categorize time by the common units of minutes,
hours, days, weeks, months, and years. The
categories are ordinal and the progression is not
linear. To continuously improve the time-to-practice
and time-to-research it will be necessary to map and
measure the corresponding workflows. The
workflows are likely to be complex, fragmented, and
widely distributed in physical and virtual space. To
date the whole process of translation has not been
conceptualized as a system; it has been an
agglomeration of a number of ad hoc systems. The
CTSAs are compelling the (re)conceptualization of
the entire system. In that context, the CTSI should
reflect the requirements of these workflows and
CLINICAL AND TRANSLATIONAL SCIENCE INFORMATICS - Translating Information to Transform Health Care
reengineer them to make them more efficient and
Consider the four illustrative combinations at the
bottom of Figure 1. These four sentences are natural
language descriptions of the functions of CTSI.
They encapsulate the translation requirements in the
context of the emergency medicine, SARS, and
Avian Flu discussed earlier. Although a little
awkward grammatically, they make sense and can
be related to specific requirements of CTSI. There
are 7,052 similar sentences that can be constructed
from the ontology. Each of these sentences can
connote a number of requirements. No one system is
likely to fulfil all the requirements connoted by all
the sentences. On the other hand, a selection of
sentences can be a high level description of the
requirements of a CTSI.
Ambrose, P., A. Ramaprasad, et al. (2003). "Managing
thin and thinly distributed knowledge in medical
genetics using the Internet." Logistics Information
Management 16(3/4): 207-214.
Anderson, N. R., E. S. Lee, et al. (2007). "Issues in
Biomedical Research Data Management and Analysis:
Needs and Barriers." Journal of the American Medical
Informatics Association 14(4): 478-488.
Cicmil, S., T. Williams, et al. (2006). "Rethinking Project
Management: Researching the actuality of projects."
International Journal of Project Management 24(8):
Cimino, J. J. (2006). "In defense of the Desiderata."
Journal of Biomedical Informatics 39(3): 299-306.
Eysenbach, G. (2003). "SARS and Population Health
Technology." Journal of Medical Internet Research
Gaughan, A. (2006). "Bridging the divide: the need for
translational informatics." Pharmacogenomics 7(1):
Ghosh, A. K. and K. Ghosh (2005). "Translating evidence-
based information into effective risk communication:
Current challenges and opportunities." Journal of
Laboratory and Clinical Medicine 145(4): 171-180.
Gilchrist, A. (2003). "Thesauri, taxonomies and ontologies
-- an etymological note." Journal of Documentation
59(1): 7-18.
Glaser, B. G. and A. Strauss (1964). The discovery of
grounded theory: Strategies for grounded research,
New York: Aldine.
Graham, I. D., J. Logan, et al. (2006). "Lost in knowledge
translation: time for a map." J Contin Educ Health
Prof 26(1): 13-24.
Graham, I. D. and J. Tetroe (2007). "Some Theoretical
Underpinnings of Knowledge Translation." Academic
Emergency Medicine 14(11): 936-941.
Gruber, T. R. (1995). "Toward Principles for the Design of
Ontologies Used for Knowledge Sharing."
International Journal Human-Computer Studies 43(5-
6): 907-928.
Gruber, T. R. (2008). Ontology. Encyclopedia of Database
Systems. L. Liu and M. T. Ozsu, Springer-Verlag.
Hedges, J. R. (2007). "The Knowledge Translation
Paradigm: Historical, Philosophical, and Practice
Perspectives." Academic Emergency Medicine 14(11):
Hess, V. (1998). "Medical semiotics in the 18th century: A
theory of practice?" Theoretical Medicine and
Bioethics 19(3): 203-213.
Holroyd, B. R., M. J. Bullard, et al. (2007). "Decision
Support Technology in Knowledge Translation."
Academic Emergency Medicine 14(11): 942-948.
Ladkin, P. B. (2005). "Ontological Analysis." Safety
Systems 14(3): 1-7.
Lang, E. S., P. C. Wyer, et al. (2007). "Knowledge
Translation: Closing the Evidence-to-Practice Gap."
Annals of Emergency Medicine 49(3): 355-363.
Nagykaldi, Z. and J. W. Mold (2007). "The Role of Health
Information Technology in the Translation of
Research into Practice: An Oklahoma Physicians
Resource/Research Network (OKPRN) Study." J Am
Board Fam Med 20(2): 188-195.
Parashar, U. D. and L. J. Anderson (2004). "Severe acute
respiratory syndrome: review and lessons of the 2003
outbreak." Int. J. Epidemiol. 33(4): 628-634.
Payne, P. R. O., E. A. Mendonca, et al. (2007).
"Conceptual knowledge acquisition in biomedicine: A
methodological review." Journal of Biomedical
Informatics 40: 582-602.
Peiris, J. S. M., K. Y. Yuen, et al. (2003). "The Severe
Acute Respiratory Syndrome." N Engl J Med 349(25):
Ramaprasad, A. (1983). "On the Definition of Feedback."
Behavioral Science 28(1): 4-13.
Ramaprasad, A. and P. J. Ambrose (1999). The Semiotics
of Knowledge Management. Ninth Workshop on
Information Technology and Systems, Charlotte,
North Carolina.
Ramaprasad, A. and V. Kashyap (2008). Semiotic Web for
Translational Medicine (Poster). AMIA 2008 Annual
Symposium. Washington D.C.
Ramaprasad, A. and A. Rai (1996). "Envisioning
management of information." Omega-International
Journal of Management Science 24(2): 179-193.
Scott, N. A., C. Moga, et al. (2007). "Creating clinically
relevant knowledge from systematic reviews: the
challenges of knowledge translation." Journal of
Evaluation in Clinical Practice 13(4): 681-688.
Simon, H. A. (1962). "The architecture of complexity."
Proceedings of the American Philosophical Society
106(6): 467-482.
HEALTHINF 2009 - International Conference on Health Informatics
Sung, N. S., W. F. Crowley, Jr., et al. (2003). "Central
Challenges Facing the National Clinical Research
Enterprise." JAMA 289(10): 1278-1287.
Tugwell, P., V. Robinson, et al. (2006). "Systematic
reviews and knowledge translation." Bulletin of the
World Health Organization 84(8): 643-651.
Webster, R. and D. Hulse (2005). "Controlling avian flu at
the source." Nature 435(7041): 415-416.
Westfall, J. M., J. Mold, et al. (2007). "Practice-Based
Research--" Blue Highways" on the NIH Roadmap."
JAMA: The Journal of the American Medical
Association 297(4): 403.
Zerhouni, E. A. (2005). "Translational and clinical
science—Time for a new vision." N Engl J Med
353(15): 1621-1623.
Zerhouni, E. A. (2005). US Biomedical Research Basic,
Translational, and Clinical Sciences, Am Med Assoc.
294: 1352-1358.
Zerhouni, E. A. (2007). "Translational Research: Moving
Discovery to Practice." CLINICAL
Zerhouni, E. A. and B. Alving (2006). "Clinical and
Translational Science Awards: a framework for a
national research agenda." Translational Research
148(1): 4-5.
CLINICAL AND TRANSLATIONAL SCIENCE INFORMATICS - Translating Information to Transform Health Care