=Paper=
{{Paper
|id=Vol-1747/IP05_ICBO2016
|storemode=property
|title=An Ontological Representation for the Transtheoretical Theory
|pdfUrl=https://ceur-ws.org/Vol-1747/IP05_ICBO2016.pdf
|volume=Vol-1747
|authors=Hua Min,Robert H. Friedman,Julie Wright
|dblpUrl=https://dblp.org/rec/conf/icbo/MinFW16
}}
==An Ontological Representation for the Transtheoretical Theory ==
https://ceur-ws.org/Vol-1747/IP05_ICBO2016.pdf
An Ontological Representation for the Transtheoretical Theory
Hua Min1, Julie Wright2 ,Robert H. Friedman3
1
Department of Health Administration and Policy, George Mason University, Fairfax, VA 22030
2
Department of Exercise and Health Science, University of Massachusetts, Boston, MA 02125
3
Medical Information System Unit, Boston University, Boston, MA 02118
Abstract—Ontologies are widely used in computer science and
medicine. Ontologies may be useful in health promotion and II. METHODS AND RESULTS
disease prevention for intervention development. Interventionists An ontology of the TTM was created. The description of
usually use theory to guide intervention design and evaluation,
the TTM was gathered from books [5, 11], literature [7-10],
but there is no standard vocabulary for health behavior theory.
A formal mechanism for converting theory to a computer-based
and domain experts. Domain experts were asked to respond to
representation may provide a tool that can assist in the a list of core competency questions (e.g., “What variables in
development of computer-based interventions. This paper the theory may be involved in each stage transition?”). A list
demonstrates how ontology can be used to represent a health of important terms was created (see Table 1). Those terms are
behavior theory using the Transtheoretical Model (TTM) of organized based on parent-child relationship. Protégé was
behavior change as an example. used to construct the TTM ontology [12]. The ontology’s class
and logical consistency was evaluated with the FaCT++ [13].
Keywords—Ontology; Health Behavior Theory; Transtheoretical The resulting ontology defined 82 classes including stages of
Model
change, processes of changes, and self-efficacy and 12 types
of relationships including parent-child, hasComponent, and
I. INTRODUCTION hasApplication.
Table 1: Important terms in the TTM ontology
Ontology is crucial in data exchange, integration, and reuse
in biomedical research [1-3]. Another potential ontology Stages of Change Processes of Change
Precontemplation (PC) Consciousness Raising (CR)
application is in the field of health promotion and disease
Contemplation (C) Dramatic Relief (DR)
prevention, specifically in designing effective theory-driven
Preparation (PR) Environmental Re-evaluation (ER)
computer-based interventions. Health behavior theories Action (A) Social Liberation (SO)
provide an organized and efficient tool to design and evaluate Maintenance (M) Self-reevaluation (SR)
health behavior interventions [4-6]. Most health behavior Termination (T) Reinforcement Management (RM)
theories, however, have not been formally defined using Decisional Balance Counter Conditioning (CC)
ontologies. The interpretation of health behavior theories and Pros of changing Helping Relationships (HR)
their constructs often vary across research labs, making Cons of changing Stimulus Control (SC)
comparisons difficult. This paper discusses how ontology can Temptation Self-liberation (SL)
assist interventionists with designing effective theory-guided Habit Addictive Self-efficacy
Negative Affective Confidence
interventions, and provides an example of a health behavior
Positive Social Temptation
ontology using Transtheoretical Model (TTM) [7-10].
The TTM is comprised of 15 constructs: (a) Stages of III. DISSCUSION
Change, (b) Pros and Cons, (c) Situational Self-efficacy and
A. TTM Ontology
Temptation, and (d) the 10 Processes of Change. The central
organizing construct of the model is the Stages of Change. This work demonstrated how to use an ontology application
Stage is defined by an individual’s intention to meet a specified to represent a health behavior theory. An ontology provides a
health criterion (e.g., Perform 150 minutes/week of moderate way to communicate between domain experts and ontology
intensity exercise or not smoke). The pros and cons of users. Domain experts provide knowledge of the TTM during
changing are the two decision making constructs. Situational ontology construction. Users (e.g., interventionists and IT
professionals) can implement the TTM knowledge with
Self-efficacy is an individual’s confidence in successfully
assistance of the ontology. For example, ontology can be used
performing a healthful behavior whereas Situational to guide the development of computer-based behavior
Temptation is one’s ability to resist engaging in the unhealthful interventions such as an automated telephony system [14].
behavior in challenging situations. The processes of change are
the cognitive and behavioral activities or strategies that B. Standards Efforts in Behavior Medicine
promote behavior change. These constructs are expressed to a
greater or lesser extent depending on the individual’s stage, Ontology has been widely adopted in science especially in
and can be targeted in an intervention to facilitate progression biomedical filed. Scientists rely on definitions and taxonomy
through the stages of change. in order to communicate with each other, disseminate their
work and advance scientific knowledge. Behavioral scientists
have realized the importance of standards and ontologies.
Several ongoing projects aim to establish repositories for [8] J.O. Prochaska, and W.F. Velicer, The transtheoretical model of health
behavior change. Am J Health Promot, 1997. 12(1): p. 38-48.
standard behavioral measures including grid-enabled measures
[9] J.M. Lipschitz, et al., Transtheoretical Principles and Processes for
(GEM)[18], consensus measures for phenotypes and Adopting Physical Activity: A Longitudinal 24-Month Comparison of
exposures (PhenX) [19], patient-reported outcomes Maintainers, Relapsers, and Nonchangers. J Sport Exerc Psychol, 2015.
measurement information system (PROMIS) [20], NIH 37(6): p. 592-606.
toolbox for the assessment of neurologic and behavioral [10] W.F. Velicer, et al., Using the Transtheoretical Model for Population-
functioning [21], and the national collaborative on childhood based Approaches to Health Promotion and Disease Prevention.
Homeostasis in Health and Disease, 2000. 40: p. 174-195.
obesity research [22]. The National Cancer Institute has
[11] K. Glanz, F.M. Lewis, and B.K. Rimer, Health Behavior and Health
developed a grid infrastructure to share behavioral data Education: Theory, Research, and Practice. 1990, San Francisco: Jossey-
through the GEM and registered them in the cancer data Bass, Inc.
standards registry and repository (caDSR) [23]. [12] Protégé. Available from: http://protege.stanford.edu/
A theory-linked taxonomy of behavior change techniques [13] FACT++. Available from: http://owl.man.ac.uk/factplusplus/
(BCTs) used in interventions has been developed [24-27]. [14] R. Farzanfar, T. Hereen, J. Fava, J. Davis, L. Vachon, and R Friedman,
This taxonomy provides standard definitions for 93 BCTs. It Psychometric properties of an automated telephone-based PHQ-9.
provides a foundation to identify content of complex BCTs Telemed J E Health, 2014 Feb;20(2):115-21
and facilitates the development of more effective interventions [15] N.F. Noy, and D.L. McGuinness, Ontology Development 101: A Guide
to Creating Your First Ontology, in Technical Report KSL-01-05. 2001,
to improve health. The taxonomy, however, does not represent Stanford Knowledge Systems Laboratory.
theories, only techniques of BCT. In this paper, we developed [16] M. Appelbaum, H. Cooper, S. Maxwell, A. Stone, and K. J. Sher.
an ontology for one behavioral model using description logic. Reporting standards for research in psychology: why do we need them?
The difference between taxonomy and ontology is that What might they be? Am Psychol, 2008. 63(9): p. 839-51.
taxonomy contains hierarchical relationship of concepts (i.e. [17] CONSORT. Consolidated Standards of Reporting Trials. Available
from: www.consort-statement.org.
parent/child, or subClass/superClass, or broader/narrower)
[18] R.P. Moser, et al., Grid-enabled measures using science 2.0 to
while an ontology has arbitrary complex standardize measures and share data. Am Journal Prev
relations between concepts. Med 2011;40: S134–43.
[19] T. Hendershot, et al. Using the PhenX Toolkit to add standard measures
C. Limitations to a study. Curr Protoc Hum Genet 2011:1.21. 1–1. 18.
This study explores the ontology representation (OWL) for [20] D. Cella, et al., The Patient-Reported Outcomes Measurement
one health behavior theory. Thus, it lacks testing for general Information System (PROMIS) developed and tested its first wave of
ontology representation of other health behavior theories. adult self-reported health outcome item banks: 2005–2008. J Clin
Epidemiol2010;63:1179–94.
IV. CONCLUSION [21] R.C. Gershon , D. Cella , N.A. Fox , R.J. Havlik, H.C. Hendrie and M.V.
Wagster, Assessment of neurological and behavioural function: the NIH
This work demonstrated that an ontology can be utilized to Toolbox Lancet Neurol 2010;9:138–9.
represent the TTM knowledge. The TTM Ontology provides a [22] R.A. McKinnon , J. Reedy, D. Berrigan, S.M. Krebs-Smith, and
starting point for ontology research in behavior theories. NCCOR Catlogue and Registry Working Groups, The National
Collaborative on Childhood Obesity Research catalogue of surveillance
References systems and measures registry: new tools to spur innovation and
increase productivity in childhood obesity research. Am J Prev
[1] O. Bodenreider, Biomedical ontologies in action: role in knowledge Med 2012;42:433–5.
management, data integration and decision support. Yearb Med Inform, [23] H. Min, et al., Sharing behavioral data through a grid infrastructure
2008: p. 67-79. using data standards. J Am Med Infrom Assoc. 2014 Jul-Aug;21(4):642-
[2] J.A. Blake and C.J. Bult, Beyond the data deluge: data integration and 9.
bio-ontologies. J Biomed Inform, 2006. 39(3): p. 314-20. [24] S. Michie, C.E. Wood, M. Johnston, C. Abraham, J.J. Francis, and W.
[3] J.G. Klann, A. Abend, V. A. Raghavan, K. D. Mandl, and S. N. Murphy, Hardeman. Behaviour change techniques: the development and
Data interchange using i2b2. J Am Med Inform Assoc, 2016. evaluation of a taxonomic method for reporting and
describing behaviour change interventions (a suite of five studies
[4] S.M. Noar and R.S. Zimmerman, Health Behavior Theory and involving consensus methods, randomised controlled trials and analysis
cumulative knowledge regarding health behaviors: are we moving in the of qualitative data). Health Technol Assess. 2015 Nov;19(99):1-188
right direction? Health Educ Res, 2005. 20(3): p. 275-90.
[25] S. Michie, et al., Towards An Ontology of Behacior Change: an
[5] K. Glanz, B.K. Rimer, and S.M. Su, Theory at a Glance: A Guide for inonovative approach to intervention development, Annals of behavioral
Health Promotion Practice. 2005, United States National Cancer medicine. (2015) , 49 p. S138 - S138
Institute.
[26] L.An, An application of a behavior change ontology to identify
[6] M. Eccles, J. Grimshaw, A. Walker, M. Johnston and N. Pitts, compoinetns wihtin intervreiosn repositorires, Annals of behavioral
Changing the behavior of healthcare professionals: the use of theory in medicine. (2015) , 49 p. S139 - S139
promoting the uptake of research findings. Journal of Clinical
Epidemiology, 2005. 58(2): p. 107-112. S. Michie, Johnston M, Rothman A, Kelly M, and de Bruin M.
Developing methodology for designing and evaluating theory-based
[7] J.O. Prochaska, and C.C. DiClemente, Stages and processes of self- complex interventions: an ontology for linking behaviour change
change of smoking: toward an integrative model of change. J Consult techniques to theory. Medical Research Council.
Clin Psychol, 1983. 51(3): p. 390-5.