Using Ontologies to Enhance Data on Intimate Partner Violence S. Clint Dowland 1and William R. Hogan 1 1 Department of Health Outcomes & Biomedical Informatics, University of Florida, Gainesville, FL, USA Abstract Intimate partner violence is among the social determinants of health, and for this reason is addressed in instruments intended to gather data on social determinants of health. Anticipating the need to integrate data on intimate partner violence in the OneFlorida+ Clinical Research Consortium, we propose an ontology-based approach to representing the referents of data on intimate partner violence gathered via two widely used instruments. Keywords 1 ontology, intimate partner violence, domestic abuse, social determinants of health 1. Introduction instruments. Newly developed terms are to be housed within the Ontology of Medically Related Social Entities (OMRSE), which focuses on the The negative health outcomes for victims of intersection of the medical and the social [9,10]. intimate partner violence (IPV) are both acute and OMRSE uses the Basic Formal Ontology (BFO) chronic, they include physical and mental health as its upper-level ontology, and here we re-use outcomes, and they include increased risks for classes only from other BFO-compliant unhealthy behaviors [1,2]. For these reasons, IPV ontologies [11]. Unless otherwise stated, all is considered to be among the social determinants relations (object properties) used here are from of health (SDoH). SDoH are “non-medical factors BFO or the Relation Ontology (RO) [12]. that influence health outcomes,” and they include “the conditions in which people are born, grow, work, live, and age,” as well as the “forces and systems shaping the conditions of daily life” [3]. 2. IPV Data in PRAPARE and Epic Recognition of the health impacts arising from socioeconomic factors has led to an increased We focus here on the sort of data gathered as emphasis in recent decades on researching and responses to IPV-related prompts in two SDoH- addressing SDoH [4,5], and on developing screening instruments. instruments to gather SDoH data [6,7]. These data First, the Protocol for Responding to and gathered by different instruments are organized in Assessing Patients’ Assets, Risks, and different ways and may not be easily integrated Experiences (PRAPARE) is an assessment tool without additional work. Axiomatically rich designed to assist health care organizations and ontologies would provide a means to enhance providers with collecting and using SDoH data meaningful integration of data organized by about their patients [13]. Next, Epic Systems has heterogenous data models by providing the data an EHR software system with different modules with a computable semantics in order to achieve for different domains of data, with Epic Healthy semantic interoperability [8]. Planet being the module intended for gathering We present here an ontology-based approach SDoH data [14]. to modeling and enhancing IPV-related data derived from questions on two SDoH-focused ICBO 2022, September 25-28, 2022, Ann Arbor, MI, USA EMAIL: clintdowland@gmail.com (S.C. Dowland); hoganwr@ufl.edu (W.R. Hogan) ORCID: 0000-0003-1909-9269 (S.C. Dowland); 0000-0002- 9881-1017 (W.R. Hogan) ©️ 2022 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR Workshop Proceedings (CEUR-WS.org) CEUR Workshop ceur-ws.org ISSN 1613-0073 1 Proceedings 2.1. PRAPARE 3. Intimate Partnerships PRAPARE categorizes two questions under What makes it true that the person is either a “Safety and Domestic Violence” [15]: current or former partner of the respondent is some current or past relationship between the Q1: Do you feel physically and emotionally respondent and that person. We represent the type safe where you currently live? of relationship in question—intimate partnership—as a subtype of BFO: relational Q2: In the past year, have you been afraid of quality, a type of quality that has a plurality of your partner or ex-partner? bearers. Similarly, Arp, Smith, and Spear [11] give marriage bond as an example in their In both cases, the response options include Yes; overview of BFO: relational quality. To specify No; Unsure; and I choose not to answer this the nature of intimate partnerships, we draw upon question. For Q2, there is an additional option: as the way the meaning of “intimate partner” is I have not had a partner in the past year. explained by the Centers for Disease Control and Prevention (CDC) [2]: 2.2. Epic Healthy Planet intimate partnership=def. Relational quality inhering in persons by virtue of being each other’s spouse, boyfriend/girlfriend, dating Epic includes within Healthy Planet four partner, or ongoing sexual partner. questions that it places under the heading ‘Intimate Partner Violence.’ One question is the Axiomatically, each instance of intimate same in meaning as one of PRAPARE’s, and is partnership inheres in at least two persons. phrased as follows with these response options: Q3: Within the last year, have you been afraid of your partner or ex-partner? 4. Referents of ‘the past year’ Possible values are Yes; No; and Patient refused. Taken literally, the referent of “the past year” The other IPV-related questions from Epic or “the last year” is an interval with a length of have the same value options as the preceding one, exactly 365 days (exactly 8760 hours). But there and each asks whether the respondent has been are reasons to represent the referent of those abused in some way by a partner or ex-partner: phrases, as found in the questions considered here, as an interval lasting longer than a year. Q4: Within the last year, have you been Suppose that at noon on July 27 of the current humiliated or emotionally abused in other year, I ask you whether some type of event has ways by your partner or ex-partner? occurred within the past year. If we take “the past year” to denote an interval precisely 365 days long Q5: Within the last year, have you been kicked, and ending at the moment the question is asked, hit, slapped, or otherwise physically hurt by then the question is about an interval extending your partner or ex-partner? back to noon on July 27 of last year (setting aside Leap Years). But we are not sure that anyone Q6: Within the last year, have you been raped really interprets such phrases that way. For or forced to have any kind of sexual activity by example, if you recall that sort of event happening your partner or ex-partner? most recently on July 27 of last year, we should not be surprised if you answer, “Yes,” even if it From here on, we shorten “partner or ex- occurred earlier than noon that day. To account partner” to “ex/partner.” Having introduced the for this while positing an interval of precise length questions, we may now proceed with as the referent of “the past year,” we might then ontologically representing the entities and add twelve hours to the year in order to account relations entailed by a ‘Yes’ or ‘No’ response to for the question being asked at noon. But if we the first PRAPARE question and to a ‘Yes’ wish to account for all cases no matter how late in response to the others. the day the question is asked, we might posit the interval is one year, twenty-three hours, fifty-nine 2 minutes, and fifty-nine seconds long. Or we could a postal delivery route denoted by a ZIP code is simply round up to one year and twenty-four realized. That ICE and the appraisal of the hours. respondent each stand in the is about relation to Applying the above reasoning to “the past that location. Instead of IAO: is about, the domain year” or “the last year” in the questions considered of which is restricted to ICEs, we instead use the here, we stipulate the referent is an interval that is about of Smith and Ceusters [17], the domain ends precisely at the end of the day on which the of which includes not only ICEs but also question is asked, and extends back precisely 366 representations. The region is an instance of GEO: days (that is, precisely 8,784 hours). A further geographical region.3 advantage is that this accounts for cases in which the available data include only the date for a given set of a responses, and not a time. 7. Referents of ‘Yes’ or ‘No’ to Q1 To better connect this interval to the respondent, we demarcate a proper temporal part of the respondent’s BFO: history that exactly We have outlined the basic elements for occupies that interval. Since that part of the representing the referents of a response to Q1. Figure 1 below represents the referents of an person’s history is temporally extended, it is not a answer to question Q1 about whether the process boundary; and thus, like any part of a respondent feels safe where they live. process that is not a process boundary, it too is a process. Its relation to the relevant interval is occupies temporal region. Other entities may be related to that interval via the part of the respondent’s history that occupies it. For an entity inhering in the respondent, the relevant history part has participant at some time that entity. For a process such an instance of abuse, the relevant history part has proper occurrent part that process. 5. Appraisals of Safety For representing judgements of danger or safety, we make use of a set of terms in the Emotion Ontology [16]: appraisal and its subtype Figure 1: Referents of ‘Yes’ or ‘No’ to Q1 appraisal of dangerousness, which “represents an evaluation of how threatening an object or If the response is a ‘Yes,’ then the appraisal in situation is.” The latter has appraisal as Fig. 1 is an instance of appraisal as not dangerous and appraisal as not dangerous, dangerous. If ‘No,’ it is an instance of appraisal differing with respect to whether they represent as dangerous. The rest is the same for each. the thing in question as a threat to the bearer. 8. Referents of ‘Yes’ to Q2 or Q3 6. Referent of ‘where you live’ We turn next to representing the referents of a We use OMRSE: disclosure of residence to ‘Yes’ about whether the respondent has been represent the process that outputs (OBI: has afraid of an ex/partner. Figure 2 shows what they specified output) a patient or other respondent’s are and how they relate. This is the same for both residential data.2 Here we focus on the ZIP code the PRAPARE (Q2) and Epic (Q3) versions. alone as the output of that process. It is represented with OMRSE: residence ZIP code ICE, which is about a geographic region in which 2 3 http://purl.obolibrary.org/obo/OBI_0000299. http://purl.obolibrary.org/obo/GEO_000000372. 3 The domain of each is abusive behavior. The range of the former can be restricted to human beings, while that of the latter could be more general since there are nonhuman abuse victims. We make use of these relations in axioms of abusive behavior: each instance stands in has aggressor to at least one person, and stands in the is abuse of relation to at least one person or animal. We draw upon CDC definitions [2] in defining subclasses of abusive behavior corresponding to the types of abuse described in Q4-Q6. psychologically abusive behavior =def. Abusive behavior in which the aggressor does or attempts to do the following: mentally or emotionally harm or exert control over another. physically abusive behavior =def. Abusive behavior in which the aggressor does Figure 2: Referents of ‘Yes’ to Q2 or Q3 or attempts to do the following: harm, restrain, or coerce another through physical force. 9. Types of Abuse sexually abusive behavior =def. Abusive behavior in which the aggressor does The three remaining questions Q4-Q6 are or attempts to do the following: force or coerce similar in structure to one another. Each asks another to participate in a sexual act to which whether the respondent has been abused by an the latter has not freely given consent. ex/partner within the last year. They differ with respect to which type of abuse is the focus. As a It is worth noting that a single episode of abuse starting point for representing these types of abuse can be or include an instance of multiple subtypes. and grouping them together under a more general We include “does or attempts to do” in the classification, we draw upon the ‘abuse’ entry in three preceding definitions because cases of the the APA Dictionary of Psychology [18] in same type of abuse may vary greatly concerning defining a general class of abuse: both outcome and intent. We mention attempts because failed attempts at abuse—a dodged punch abusive behavior=def. Behavior that is cruel, for example—can be abusive. violent, demeaning, or invasive. While “attempts to” would suffice to cover both successful and failed attempts, we must also In order to connect an act of abusive behavior account for unintentionally abusive behavior. If to the involved persons in OWL, we make use of for example one partner is repeatedly careless in a object properties that represent how that act way that puts the other at risk of harm, then even relates to those persons. We introduce two object without intent to harm, the pattern of neglecting to properties to distinguish the abuser’s sort of account for the other’s safety may be abusive. We participation from that of the abused: must then account for attempts at abuse whether or not they are successful, as well as for behavior has aggressor =def. Relation between an that has abusive effects whether or not there is abusive behavior and one who inflicts it upon abusive intent. someone or something else. Of course, to avoid disjunctive definitions, we could have instead developed a set of terms for is abuse of =def. Relation between an abusive each subtype of abuse, for example one for behavior and one upon whom it is inflicted. attempting to harm via physical force and another simply for causing harm via physical force. But 4 such terms would be less useful here, since a ‘Yes’ health-relevant factors in the respondent’s life. to one of the questions in consideration does not That representation may be even larger if it provide us with details such as whether the abuse incorporates the referents of EHR data about the was intentional or whether the respondent was same person. But this, too, requires additional indeed harmed. ontological engineering that results in compatible, ontology-based models. 10. Referents of ‘Yes’ to Q4, Q5, or Q6 12.Acknowledgements We use the above terms to specify the type of abuse depending on which questions receive a This work was supported by the OneFlorida+ ‘Yes.’ Apart from the type of abuse, the rest of the Clinical Research Network, funded by Patient- picture is the same for a ‘Yes’ to any of Q4-Q6: Centered Outcomes Research Institute (PCORI) awards CDRN-1501-26692, RI-CRN-2020-005 and RI-FLORIDA-01-PS1; by a Florida Department of Health’s (DOH’s) James and Esther King Biomedical Research Program award 4KB16; and by the University of Florida Clinical and Translational Science Institute (CTSI), which is supported in part by the NIH National Center for Advancing Translational Sciences (NCATS) under award number UL1TR001427 and UL1TR000064. The content is the responsibility of the authors and does not represent the views of PCORI, OneFlorida+, the CTSI, the Florida DOH, or the NIH. 13.References [1] World Health Organization. (2012). Understanding and addressing violence against women: Intimate partner violence (No. WHO/RHR/12.36). [2] Breiding, M., Basile, K. C., Smith, S. G., Black, M. C., & Mahendra, R. R. (2015). Figure 3: Referents of ‘Yes’ to Q4, Q5, or Q6 Intimate partner violence surveillance: Uniform definitions and recommended The entity labeled ‘temporal interval 1’ in Fig. 3 data elements. Version 2.0. is the interval occupied by the instance of abuse. [3] World Health Organization. Social determinants of heath. https://www.who.int/health-topics/social- 11. Discussion and Conclusion determinants-of-health. [4] McGinnis, J. M., Williams-Russo, P., & Knickman, J. R. (2002). The case for While each diagram above shows only the more active policy attention to health referents of a single response, the end result of promotion. Health affairs, 21(2), 78-93. multiple responses would be several such sets of [5] Braveman, P., & Gottlieb, L. (2014). The referents all relating to the same respondent. As social determinants of health: it's time to ontology-based models of responses to additional consider the causes of the causes. Public prompts are developed, they too may be health reports, 129(1_suppl2), 19-31. transformed in the same manner and connected to [6] Gold, R., Cottrell, E., Bunce, A., the other data elements about the same Middendorf, M., Hollombe, C., Cowburn, respondent, yielding a larger representation of S., ... & Melgar, G. (2017). Developing 5 electronic health record (EHR) strategies Conference on Biomedical Ontology related to health center patients' social (ICBO). CEUR vol. 1515. pp. 1-5. determinants of health. The Journal of the [18] VandenBos, G. R. (2007). APA American Board of Family Dictionary of Psychology. American Medicine, 30(4), 428-447. Psychological Association. [7] LaForge, K., Gold, R., Cottrell, E., Bunce, A. E., Proser, M., Hollombe, C., ... & Clark, K. D. (2018). How 6 organizations developed tools and processes for social determinants of health screening in primary care: an overview. The Journal of ambulatory care management, 41(1), 2. [8] Brochhausen, M., Bona, J., & Blobel, B. (2018). The Role of Axiomatically Rich Ontologies in Transforming Medical Data to Knowledge. Studies in Health Technology and Informatics, 249, 38-49. [9] Hogan, W. R., Garimalla, S., Tariq, S. A. (2011). Representing the Reality Underlying Demographic Data. International Conference on Biomedical Ontology: ICBO 2011. Buffalo, NY. [10] Hicks, A., Hanna, J., Welch, D., Brochhausen, M., & Hogan, W. R. (2016). The Ontology of Medically Related Social Entities: recent developments. Journal of Biomedical Semantics, 7(1), 1-4. [11] Arp, R., Smith, B., & Spear, A. D. (2015). Building ontologies with Basic Formal Ontology. MIT Press. [12] Huntley, R. P., Harris, M. A., Alam- Faruque, Y., Blake, J. A., Carbon, S., Dietze, H., ... & Mungall, C. J. (2014). A method for increasing expressivity of Gene Ontology annotations using a compositional approach. BMC bioinformatics, 15(1), 1-11. [13] PRAPARE. Who We Are. https://prapare.org/who-we-are/. [14] OCHIN. Building the Foundation for Population Health at OCHIN. https://ochin.org/blog/population-health- at-ochin. [15] National Association of Community Health Centers. (2016). PRAPARE implementation and action toolkit. [16] Hastings, J., Ceusters, W., Mulligan, K., & Smith, B. (2011). Dispositions and processes in the Emotion Ontology. [17] Ceusters, W. & Smith, B. (2015). Aboutness: Towards Foundations for the Information Artifact Ontology. In Proceedings of the Sixth International 6