Development of a discharge ontology to support postanesthesia discharge decision making Lucy L Wang∗, Yong Choi Department of Biomedical Informatics and Medical Education, School of Medicine, University of Washington, 850 Republican St, Seattle, WA 98195, USA ABSTRACT Many clinics have specified their own modified criteria for Postanesthesia discharge decision making is a challenging postanesthesia discharge. We begin by assembling resources process due to the high complexity and variability of care provided to published online by various surgical units. Among these resources, postoperative patients. We built an ontology-based decision support many are based on the Aldrete scoring system, with additional system that generates discharge recommendations for patients who modifications tailored to clinic-specific workflow. Criteria from have undergone surgical procedures. Discharge decisions are made Stanford Hospital and Clinics, Loyola University Medical Center based on patient vitals, symptoms, medical history and details and others are used to construct a global discharge rule set (Stanford of the surgical procedure. The output recommendations of our Hospital and Clinics, 2010; Brown et al., 2008). The Phillips et al., system can aid healthcare providers in discharge decision-making 2011 systematic review of postanesthesia discharge protocols is also and potentially reduce readmissions due to improper discharge. This used to determine levels of evidence for various scoring criteria. project demonstrates the potential uses of ontologies in medical Scoring guidelines present in all or most resources we studied are decision support systems, especially in areas that use specific included as criteria in our ontology-based decision support system. scoring guidelines to aid decision-making. Based on the the criteria in these resources, we build a set of SWRL rules, which in turn guides our development of an OWL 1 INTRODUCTION ontology. We first create a full set of postanesthesia discharge criteria using information from our source documents. We then Evidence-based discharge decision making and planning is a translate these criteria into SWRL rule syntax to facilitate reasoning. critical care process that can improve patient outcomes and reduce Afterwards, we use these rules to guide the creation of OWL classes, readmission rates. Inappropriate discharge can cause additional as well as the definition of object and data properties. pain and suffering for patients and their families and consume The modified Aldrete score, for example, consists of five primary unnecessary hospital resources (Anderson et al., 2011). For surgical criteria: consciousness, respiration, circulation, movement and pain. procedures, the risks associated with early discharge may be even Some of these, such as respiration and circulation, can be broken higher. When planning for discharge, healthcare professionals have down further. For example, respiration consists of breathing quality, to account for multiple variables such as age, vitals, comorbidities, breath rate and oxygen saturation. medications and social issues. A tool like the Aldrete scoring Aldrete subscores, along with the total score, are used as criteria system is commonly used to help healthcare professionals determine for discharge. For each Aldrete subscore, a patient receives a score when patients can be safely discharged (Aldrete, 1995). However, on a 2 point scale, where 0 means a low functional level and 2 there are no standardized guidelines routinely used by healthcare means a normal functional level. We create a data property that professionals to assist them in making postoperative discharge corresponds to each primary criteria. The value of this data property decisions. A knowledge-based decision support tool based on is determined using SWRL rules and assigned to a patient based standardized procedures can enhance discharge decision making on his/her current status in the system. An example data property and reduce errors. In Bouamrane et al., 2010, the authors built an hasAldreteScoreConsciousness may take on the values of 0, 1 or 2 ontology to model preoperative domain knowledge. In this paper, if the patient is unresponsive, responsive but drowsy, and responsive we use a similar approach to create a postoperative ontology-based and fully alert respectively. decision support system to assist discharge decision making. Another example is the circulation subscore, where the patient’s blood pressure must fall within a pre-specified range from the 2 METHODS baseline blood pressure. Within ± 20 mmHg yields a score of 2, Our goal is to create an ontology to aid in post-surgery discharge within ± 20-50 mmHg yields a score of 1, and anything outside decision-making. Following surgery, patients generally go from of that range yields a score of 0. These specific differences can phase I postanesthesia care to phase II before being discharged be automatically calculated by our reasoner, which then assigns a to home. Phase I care immediately follows surgery and involves score to the patient. This reduces the need for healthcare workers to intensive monitoring of patient status. Phase II care is less intensive perform time-consuming numerical calculations. and sees the patient recovering well from anesthesia. The goals An example SWRL rule for oxygen saturation is: of our system are (i) to detect patients who may be suitable for P atient(?pt), hasSpO2(?pt, ?SpO2), greaterT han(?SpO2, 95) discharge, (ii) to determine the appropriate discharge workflow, and (iii) to generate a list of additional recommendations for physicians. → hasAldreteScoreOxygen(?pt, 2) which assigns 2 points to the Aldrete oxygen subscore if a ∗ To whom correspondence should be addressed: lucylw@uw.edu patient’s oxygen saturation is greater than 95%. The sum of points Copyright c 2015 for this paper by its authors. Copying permitted for private and academic purposes 1 Choi and Wang assigned to all Aldrete criteria is then calculated and used to work are the semantic plurality of clinical data representation and determine whether a patient fits the basic criteria for discharge. If non-standard data exchange protocols between platforms. a patient satisfies this condition, she/he is assigned into the class DISCHARGE FROM PHASE I POSTANESTHETIC CARE. Our ontology classifies patients for discharge from phase I to phase II care, as well as from phase II care to home. Additionally, our discharge ontology makes recommendations for healthcare provider actions. For example, a patient receiving a sciatic block may require clutches at discharge (Figure 1), or a patient with high pain levels may require additional pain management. These recommendations can be used by healthcare providers to prioritize patient care and to generate discharge notes. Fig. 2. Integrating information from electronic health records and patient monitoring devices into the decision support ontology. Another future direction is to expand our system to create automated discharge summary notes to assist in the transition of care. The discharge summary notes can be generated for two groups of users: (1) healthcare professionals, and (2) patients and caregivers. Discharge notes generated for healthcare professionals can be used to facilitate continuity of care. Notes generated for patients and caregivers can contain care instructions specifically tailored to the patient to help guide them through the complex post-discharge care process. 4 CONCLUSION The discharge decision-making process relies on a set of predefined Fig. 1. Example patient who is 30 minutes post surgery with stable vitals. Output classifications based on SWRL rules are given. clinical criteria that must be interpreted correctly to reach the appropriate discharge decision. Our ontology integrates patient 3 RESULTS & DISCUSSION vitals, symptoms, and surgical and medical information, and outputs recommendations for discharge and healthcare provider actions. In this project, we demonstrated our work in building a knowledge- This decision support tool could simplify postanesthesia discharge based decision support system that generates decision support procedures and may help reduce adverse events based on improper recommendations to determine patient discharge eligibility after or early discharge. surgical procedures. We were able to model appropriate discharge decision making in several example patients (Figure 1). In addition to making the correct discharge decision, our system also generates ACKNOWLEDGEMENTS a list of recommendations for clinicians which should be followed This study was supported in part by National Library of Medicine before the actual discharge. (NLM) Training Grant T15LM007442. Our decision support system operated with a number of limitations. First of all, the recommendations and guidelines issued REFERENCES by our system are constrained by the accuracy of the guidelines that Aldrete, J. A. (1995). The post-anesthesia recovery score revisited. J. Clin. Anesth., 7, we modeled. Therefore, any errors or flaws present in the model 89–91. guidelines will be systematically replicated by our system. Also, due Anderson, D., Price, C., Golden, B., Jank, W., and Wasil, E. (2011). Examining the discharge practices of surgeons at a large medical center. Health Care Manag. Sci., to the lack of a standard discharge protocol, we could only capture 14, 338–347. a representative set of criteria. Our ontology, therefore, may need to Bouamrane, M. M., Rector, A., and Hurrell., M. (2010). Experience of using be modified for use in any specific clinical environment. owl ontologies for automated inference of routine pre-operative screening tests. Additionally, we should align our system with pre-existing International Semantic Web Conference 2010, pages 50–65. medical ontologies for morbidity classification such as ICD-10 or Brown, I., Jellish, W. S., Kleinman, B., Fluder, E., Sawicki, K., Katsaros, J., and Rahman, R. (2008). Use of postanesthesia discharge criteria to reduce discharge SNOMED-CT. We believe that such integration is critical for the delays for inpatients in the postanesthesia care unit. J. Clin. Anesth., 20, 175–179. future interoperability of our system. Future work will also involve Phillips, N. M., Haesler, E., Street, M., and Kent, B. (2011). Post-anaesthetic discharge extracting information from electronic health records or in-room scoring criteria: a systematic review. JBI Library of Systematic Reviews, 9, 1679– patient sensors (Figure 2) to increase the accuracy, timeliness and 1713. Stanford Hospital and Clinics (2010). Discharge criteria for phase I & II post anesthesia reliability of patient medical data. Some obvious challenges to this care. 2 Copyright c 2015 for this paper by its authors. Copying permitted for private and academic purposes