Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 Expert Systems with Applications xxx (2014) xxx–xxx 1

Contents lists available at ScienceDirect

Expert Systems with Applications journal homepage: www.elsevier.com/locate/eswa 5 6

Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge

3 4 7 8 9 10 11 1 2 3 0 14 15 16 17 18 19

Q1

N. Lasierra a,⇑, F. Roldán b, A. Alesanco a, J. García a a b

Communications Technologies Group (GTC), Aragón Institute of Engineering Research (I3A), University of Zaragoza, Spain Hospital Management Research Unit, University of Zaragoza, Spain

a r t i c l e

i n f o

Keywords: Ontologies Health catalog Items management

a b s t r a c t The goal of this work is to contribute to an improvement in the management and usage of medical items in hospitals by developing an ontology-driven solution that organizes and describes clearly related knowledge. Experts in the purchasing and management of hospital supplies (administrative and clinical) were invited to participate at different stages of the ontology-based system development. As a result of the first stage, the HealthCatalog ontology was developed. This ontology models generic items included in a health catalog and their management. Secondly, a further refinement of the ontology was conducted by specifically studying the case of gloves. Twenty-seven references of different glove types were modeled and included in the ontology during the stage two. A prototype was then developed as a proof of concept and for the evaluation of the ontology. Finally, a usability evaluation was planned to improve the ontology and obtain feedback from experts after testing the system. Experts involved in the evaluation stressed its potential use in a real clinical environment and the benefits it would bring in terms of cost and sharing knowledge among clinical personnel. Our proposed ontology-based system provides an understandable and organized solution to capture knowledge regarding item management and usage. It addresses the integration challenge of health catalogs while providing a framework for collaborative sharing and knowledge acquisition among clinicians. Ó 2014 Published by Elsevier Ltd.

21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

39 40

1. Introduction

41

The rising cost of healthcare is a major concern for governments today. Motivated by demographic changes, e.g. the increase of the aging population with chronic conditions and disabilities in western countries (ECDA, 2010), different methods need to be studied for reducing the cost of providing health care services while ensuring that they continue to be accessible to all citizens (Heinrich et al., 2008; Wickstrøm, Serup-Hansen, & Kristiansen, 2002). The development of telemedicine systems (Monteagudo & Moreno, 2007) and reducing unnecessary diagnostic tests (Vegting et al., 2012) have both been studied with the aim of improving the effectiveness of health care delivery. Effective use of medical products, intelligent purchasing and new management strategies can also positively contribute to reducing cost while maintaining the welfare state (Songthung et al., 2012). Carrying out these tasks

42 43 44 45 46 47 48 49 50 51 52 53 54

⇑ Corresponding author. Address: Communications Technologies Group, Aragón Institute of Engineering Research (I3A), University of Zaragoza, C/María de Luna Ed. Ada Byron, 50018 Zaragoza, Spain. Tel./fax: +34 976762698. E-mail addresses: [email protected] (N. Lasierra), fernando.roldan@gmail. com (F. Roldán), [email protected] (A. Alesanco), [email protected] (J. García).

efficiently requires effective management of a key element within a health system organization: the health catalog. Professionals from different disciplines (clinical, informatics, administrative) using a wide range of products, equipment and services, generate the core of daily activity in hospitals and medical centres. All these centres are managed by means of an ERP (Enterprise Resource Planning). This software application integrates a set of management actions and policies that operates over a structured database containing a list of items. The ERP together with this database are used to manage the available resources of the hospital. This list of available items in the previous database is known as the health catalog. The item forms the basic unit of a health catalog, an item being any type of product (e.g. needles, syringes, gloves or bandages), equipment (e.g. ultrasound scanner) or service (e.g. cataract surgery) used in a hospital. This is a big and complex structure that may include around 20,000 references (there is one reference for each type of item and for each available size of that item). Purchasing procedures, stock management, services provision or the authorization for using the items are examples of procedures done according the health catalog. Nowadays, as stated before, these health catalogs are implemented in big databases. These include information about each item regarding purchasing

http://dx.doi.org/10.1016/j.eswa.2014.04.023 0957-4174/Ó 2014 Published by Elsevier Ltd.

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023

55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 2 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141

N. Lasierra et al. / Expert Systems with Applications xxx (2014) xxx–xxx

information (e.g. cost), classification within each family product, item movements (e.g. stock in warehouses or consumption) and basic descriptive text information. Currently, health catalogs are used in hospitals by technicians (management personnel) in charge of purchasing and control of materials. Two years ago, the situation in Spain was that multiple health catalogs were used in different public health organizations even in each area of one single region of the country. This model enhanced the autonomy of each individual centre that comprised a health area. However, other limitations emerged such as poor product management in terms of costs and usage. While knowledge in terms of cost-efficiency remains in a single area of the health structure, all the costs were assumed by the public health system. As a consequence, motivated also by the global economic crisis, current efforts in hospital management now focus on the centralization of the health catalog and consequently its management. Nowadays, there is no single standard or universal health catalog. Current trends aim to align items from different hospitals in an area (corresponding to a local authority) using as a reference the catalog of the main hospitals. Furthermore, there is an increasing interest in centralizing pharmacy purchases. The health catalog includes a wide diversity of each type of item. An example is medical gloves. The use of gloves has dramatically increased in medical practice to prevent the spread of bloodborne pathogens (WHO 1, 2006). Choosing the appropriate medical glove is critical for effective protection against infections and hazardous materials (Stoessel, 2008). As pointed out by the WHO (World Health Organization), both patient and clinician safety depends on the appropriate usage of gloves (WHO 2, 2009). The classic examination glove is the latex glove. It is elastic, comfortable and has a high degree of sensitivity. Latex gloves are appropriate for techniques that require some precision such as cures and they are appreciated due to the perceived superior durability of the material (Kerr et al., 2004). However, other alternatives are available for allergic people. One example is the nitrile glove which is less elastic than the latex glove, or the vinyl glove which is more elastic but easy to break. The last alternative is the surgical glove which is mainly made of neoprene, is resistant and provides good sensitivity. However, the cost of surgical gloves is considerable higher than the other gloves. The eligibility criterion is then conditioned by tactile sensitivity, elasticity, allergies, price, suitability for the task in question and especially barrier integrity protection (e.g. strength and permeability) (Moore, Dunnill, Peter, & Wilson, 2013). Furthermore, the characteristics of medical gloves usually vary from one manufacturer to another. Thus, around 40 references to gloves can be included in a health catalog. As a consequence of the wide diversity of items that can be included in a health catalog and also their availability in health centres, it is the experience of the personnel in charge of supply management (administrative and clinical staff) that professionals from different organizations may use different items for the same medical practice. It is clear, therefore, that only one of them is using the best option for a specific technique in terms of cost-efficiency and safety. In fact, this is basically a safety problem. Being aware and properly informed is crucial for reducing risks e.g. of infections. The adequacy of each item for a specific technique is learned and reported through experience over time. Hence, an interesting idea would be to collect this knowledge and share it in a clear manner to take advantage colleagues’ experiences. In this paper we present an ontology-based solution in order to contribute to the sharing of expert knowledge and also to unify health catalogs. Our purpose was to develop an ontology-based system to gather and formalize knowledge regarding available items in health catalogs, their usage and their management. This tool could be used as a complement to current ERP systems. Not only are the

items included in the health catalog formalized, but also their relation with management tasks that belong to the ERP. Ontologies are a recognized vehicle for knowledge representation and proven technology for solving problems of understanding. On the one hand, we propose the use of ontologies to represent each item included in a health catalog. Providing a clear description of items will help to align different health catalogs while making clear their content and management. On the other hand, we propose that the health catalog-tool should also be used by clinical staff as a means of sharing knowledge and learning. Our idea is to contribute to a dynamic environment managed by clinical staff where information regarding items included in the ontology-based system may be updated in the light of clinicians’ experience and knowledge. This represents an innovation in hospital supply management given that such information is not available in current management systems. This paper reports the main components of the ontology and the methodology followed to study its usage and evaluation. It was not our purpose to address the size and full complexity of the health catalog, but simply to establish the basis for an ontology-based solution and its feasibility. This is an innovative solution towards improving medical supply management and usage in medical centres. The remainder of the paper is structured as follows. Section 2 provides the background and information on related work. Section 3 describes the materials and methods involved in each stage of the methodology. The results are presented in Section 4. The main advantages and drawbacks of the proposed solution are discussed in Section 5. Finally conclusions are summarized in Section 6.

142

2. Background and related work

170

Ontologies have been successfully used in recent decades in the world of computer science and particularly in the semantic web domain for the purpose of achieving a comprehensive and common machine-readable understanding (Berners-Lee, Hendler, & Lassila, 2001). An ontology is an abstract model that provides a controlled vocabulary for the description of concepts, each with an explicitly defined semantics in machine-readable language. Ontologies aim to capture consensual knowledge by a group of people and may be reused across different applications (Corcho, Fernández-López, & Gómez-Pérez, 2003; Studer, Benjamins, & Fensel, 1998). By means of rules it is possible to merge knowledge from them. Also, their usage is popular for the implementation of decision support tools applied in many areas (Bouamrane, Rector, & Hurrell, 2009; Saa, Garcia, Gomez, Carretero, & Garcia-Carballeira, 2012). Ontologies have proven to be useful tools in many areas including the healthcare and management domain (Becker, Heine, Herrler, & Krempels, 2003; Campana, Moreno, Riaño, & Varga, 2008; Jara et al., 2010; Lasierra, Alesanco, Guillén, & García, 2013; Valls, Gibert, Sánchez, & Batet, 2010). See, for example, Valls et al. (2010) where an ontology is used to describe organizational knowledge in a health care institution as the base to support its management in terms of actors, services and actions. This ontology was part of the European project K4Care (Campana et al., 2008). This project, which focused on personalized home care assistance, included a set of ontologies to capture and integrate knowledge and experience from different centres and professionals. As another example, see the OntHoS domain ontology which gathers a set of terms and definitions for modeling scenarios of hospital logistics (Becker et al., 2003). An interesting application of a knowledge-based system dealing with pharmaceutical products is presented in Jara et al. (2010). This work proposes the combination of the Internet of Things (IoT) and an ontology-based system to detect and assess patients regarding adverse drugs reactions.

171

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023

143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169

172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 N. Lasierra et al. / Expert Systems with Applications xxx (2014) xxx–xxx 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265

Ontologies are widely used for representing shared knowledge from heterogeneous sources, thus addressing the issue of interoperability. This is a critical challenge for information exchanged in different domains in order to guarantee collaboration. Regarding product management, see for example the works presented in Giménez, Vegetti, Leone, and Henning (2008), Lee, Chae, Kim, and Kim (2009), Panetto, Dassisti, and Tursi (2012). The ProductONTOlogy model Giménez et al. (2008), provides consensual knowledge for modeling products. Modelling product structures, processes and features is proposed for addressing problems related to the management of products, from design to sales, from diverse manufacturers and between different enterprises. A similar approach is presented in Panetto et al. (2012). They propose an ontological model of products to facilitate interoperability among applications that share information during the different phases of the physical product lifecycle (e.g. design, sales) and make use of standards for its design. Some of the features for describing these generic products have also been considered in our approach, such as family classification, authorization management and the description of technical features. The role of ontologies has also been highlighted for the development of decision support systems. See, for example, the work presented by Haghighi, Burstein, Zaslavsky, and Arbon (2012) where an ontology is designed for describing the domain knowledge for planning and managing medical services in mass gatherings. Ontologies are used as the knowledge and main core upon which the system makes the decisions (Burstein & Carlsson, 2008). While the use of ontologies in hospitals and healthcare institutions has already been reported (Becker et al., 2003; Jara et al., 2010; Valls et al., 2010) and the use of ontologies for modeling products has been studied (Giménez et al., 2008; Lee et al., 2009; Panetto et al., 2012), few works have examined the use of these semantic technologies for modeling items (e.g. medical products or services), their management in hospital environments (Songthung et al., 2012) and the value they offer for making decisions regarding the correct usage of materials in hospitals. As an example, ontologies available at the BioPortal from NCBO (National Center for Biomedical Ontology) collect a classification and hierarchy of terminology regarding equipment and supplies for the healthcare domain including instruments that are used for different medical procedures and purposes e.g. diagnostic, therapeutic or surgical.1 These items are also classified according to its utilization in different medical services. Regarding product management (purchases and acquisitions) in hospitals, the work presented in Songthung et al. (2012) presents a knowledge decision support system for sharing knowledge among hospitals and specifies the best options for improving purchasing and cost savings for drugs. Although the use of ontologies and the idea of sharing knowledge to improve the management of medical items are also addressed in this latest work Songthung et al. (2012), our approach presents additional contributions in terms of product description and the applications proposed. Noteworthy that ontologies have been proposed in other works to address activities that involve the entire supplies chain management also in hospitals including organizational issues, actors and equipment descriptions (Chandra and Kachhal, 2004; Chandra and Tumanyan, 2007). Besides, ontologies have already been used to organize knowledge regarding service-oriented aspects of logistics that could be specifically applied in the context of hospital management (Scheuermann & Hoxha, 2012). In contrast to previous solutions, our approach deals with ontologies to model items (not only medical products and pharmaceutical items) in hospitals as the basis of a system for health catalog

1

Equipment and supplies ontology in BioPortal: ontology/MSH/D004864. Last visited: April 2014.

http://purl.bioontology.org/

3

integration and the training of clinicians regarding the appropriate use of materials. Hence, while a good classification of items is reported in mentioned ontologies, our work presents additional contributions including within the ontology description functional and non-functional characteristics of the modeled items. These additional features enable us to enhance the functionality of the modeled health catalog and then use it for improving supplies chain management and also educational purposes by collaborative sharing knowledge. Our solution not only reports the utilization of the items within a medical context but provides with a set of practical hints useful for domain experts in daily practice in hospitals that goes beyond technical issues commonly known. This is actually an innovative contribution of our work. Note also, that was not the purpose of this work to provide with a formalization of organizational issues applying the entire supplies chain management but to include in the model information collected from domain experts that may help to its improvement.

266

3. Materials and methods

283

This work was carried out within the four stages depicted in Fig. 1. All the stages provided feedback to contribute to the dynamic and continuous refinement of the ontology. There are multiple evaluation criteria and formal methodologies available in the literature that can be used to evaluate an ontology. An overview of these techniques can be seen in (Haghighi et al., 2012). To the best of our knowledge, there is no standard ontology of the health catalog, hence, its evaluation was focussed on three different studies: (1) ontology features evaluation, (2) application-based study evaluation and (3) human evaluation of quality criteria. Indeed, these tasks were conducted during the first, second and third stages of the methodology development. Based on the stages proposed in Lasierra et al. (20130, the ontology engineering and the application-based study dealt with the design, evaluation and implementation of the ontology. The third stage consisted of a proof of concept of the prototype implementation required to use the ontology-based tool in practice. Finally, the last stage of our methodology consisted of an evaluation both of the ontology and the software application by external users, experts in the field. As has been proposed Bright, Yoko Furuya, Kuperman, Cimino, and Bakken (2012), usability and usefulness were formally measured.

284

3.1. Ontology engineering

305

The ontology engineering stage addressed the specification, conceptualization and formalization of the ontology and then its implementation (Kuziemsky & Lau, 2010; Noy & McGuinness, 2001; Pinto & Martins, 2004). The main objective was to model, in general, a health catalog, the main core of which is the item. Our ontology aims to answer the following competency questions: (1) how can an item be classified within the whole structure of items available in a hospital? (2) where is it authorized to be used and physically available within the areas and services of the health organization? (3) what are the characteristics of this item that dictate its appropriate usage? (4) how should it be managed in terms of purchasing and authorization? In the application-based study stage, this model was refined according to the specific items considered for which the model was validated. Two engineers with extensive knowledge of hospital management (materials, purchasing and organization) and ontology engineering took part during the first and second stage. Note that an initial phase of analysis of requirements was conducted by interviewing personnel from hospitals regarding the idea of unifying health catalogs and the information that should be included in it. Both administrative and health staff were asked about this issue and their responses

306

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023

267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282

285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304

307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 4

N. Lasierra et al. / Expert Systems with Applications xxx (2014) xxx–xxx

Fig. 1. Workflow of the methodology proposal.

327 328 329 330 331 332 333 334 335 336 337 338 339 340 341

were considered in the initial design of the ontology. Catalogs of products from different manufacturers were also reviewed for this task. The OWL 2 language (Ontology Web Language) proposed by the W3C was chosen to describe the ontology model (OWL 2, 2009). OWL 2 adds new functionalities to those contained in OWL 1 (OWL 1, 2004). Some of its features used in this work are: keys, qualified cardinality restrictions and multiple disjoint of classes. Keys are very important for applications in order to uniquely identify an individual by values that are key properties. Qualified cardinality restrictions permit specification, apart from the number of instances, of the class or data range of the instances to be counted. The ontology was implemented by using the Protégé-OWL v.4.2 ontology editor and its consistency was checked using the FACT ++ reasoner (Protégé-OWL Editor, 2004).

342

3.2. Application based-study: the glove case study

343

The application-based study technique provides good results for assessing the capability of an ontology to respond to the competency questions used for its design and then meet the objectives for which the ontology is developed (Brank et al.). This study allows the ontology to be evaluated by simply plugging it into an application and then evaluating the results of the application. Hence, in our case we used the ontology to describe specific items and then measure the performance of the ontology in providing a faithful description of the real item. This stage was also part of the design of the ontology, following which the ontology was further refined. This detailed study should be done for each type of item described in the ontology. Here we report one case study relating to medical gloves. Table 1 shows the workflow followed in order to conduct this task (which should also be followed for specifically studying other items). The first phase consists of obtaining the characteristics of the items included in the defined ontology. It aims to answer the third competency question formulated in Section 3.1. The main difference between the tasks conducted during the first stage (see Fig. 1) and the second stage lies in the degree of detail used for describing each item and the number of specific items considered. Working with information currently being used in hospital management, different medical gloves were modeled by means of the ontology following the workflow explained in Table 1. This task was performed by using the Protégé-OWL v.4.2 ontology editor. Table 2 shows the text description regarding two different

344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368

types of gloves. The ontology was validated by modeling these proposed types of items and using this information. Furthermore, their location and authorization of use was also specified.

369

3.3. Prototype development

372

The third stage consisted of a proof-of-concept on of the software agent required to hold the ontology and facilitate interaction with it. Some means of graphical visualization is required for the understanding of ontologies (Swaminathan & Sivakumar, 2012). The aim of this stage was to develop an interface that allows interaction with the ontology and its exploitation in a real scenario. Java technologies were used to develop this tool. Specifically, the Jena framework (version 2.8.7) (Jena Framework, 2009) was used to process the ontology, create new instances and execute SPARQL (SPARQL Protocol and RDF Query Language) queries (SPARQL Query Language for RDF, 2008) to search in the ontology.

373

3.4. Expert evaluation

384

Using the software developed in the previous stage, the ontology and the system itself were evaluated by a panel of experts (administrative and clinical staff). The aim of this last stage was to assess both the usage (for training and sharing knowledge) and usability of the software application and the developed ontology itself. The evaluation was conducted in order to obtain proofs that may improve their design. In addition, the correctness, coverage and clarity of the ontology were measured. This expert evaluation aimed to answer the following research questions: (1) is the application easy to use? (2) would it be useful in a real hospital management environment? (3) could it be used to solve the health catalog problem? (4) are the language and the concepts expressed in the ontology adequate? The quality of the ontology was thus measured in this stage by involving experts in this task. As suggested by Brank, Grobelnik, and Mladenic´ (2005), different levels of an ontology should be evaluated. Specifically, the correct usage of the language and the thoroughness of the taxonomy were measured. Following the phases proposed in Kushniruk and Patel (2004) for evaluation design, a sample of 8 evaluators with different profiles took part in our evaluation. Five of them work in the health care domain and three of them in the health management area. Specifically, the evaluators had the following roles: (1) health care head, (2) health care assistant manager (3) surgeon, (4) nurse, (5)

385

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023

370 371

374 375 376 377 378 379 380 381 382 383

386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 N. Lasierra et al. / Expert Systems with Applications xxx (2014) xxx–xxx

5

Table 1 Workflow for modeling new items with the ontology (used for stages 1 and 2). Workflow for modeling new items with the ontology Phase one: deriving item characteristics 1. Identify and list all types of available items 2. Classify items into different families (previously identified) - Add new classes if required 3. List all generic characteristics of each type of item and usual reported considerations 4. Determine classes and properties to be included in the ontology in order to be able to model identified characteristics Phase two: verifying the item modeling 5. List specific types of items and detail their characteristics (see Table 2) from different manufacturers 6. Using previous information, include the item as a new instance of the ontology 7. Determine if the item can be successfully modeled - Positive answer: verification completed - Negative answer: identify additional required classes or properties (datatype and object properties) to model the item and evaluate their usefulness for characterizing an item before including them in the model 8. Repeat steps 4, 5 and 6 considering special considerations that can be reported regarding that item 9. After executing steps 4, 5, 6 and 7 for all the identified items, extract (if possible) general classes and properties to cover the new identified characteristics and usual reported considerations

Table 2 Information about different types of items. Health items description Glove 1: Nitrile glove, powder-free, small size. Waterproof and chemicalresistant. Protect against microorganism. Can be used for both hands. Beaded cuff and micro-textured surface. Not sterile and presented in dispenser Box Glove 2: Synthetic glove. Dry rubber, powder-free, sterile, size 7. Elastic, resistant to chemicals, chemotherapeutic and puncture. Protect against microorganism. Anatomical and easy to use. Sterile

409 410 411 412 413 414 415 416 417 418 419 420 421 422

423 424 425 426 427 428

nursing manager, (6) management head, (7) purchasing and contracting manager and (8) maintenance manager. They belong to three different hospitals in the region of Aragón, Spain. All of them are potential final users of the ontology-based system in a real scenario. A combination of common usability techniques (inspection, inquiry and testing) was applied for the evaluation process (González, Lorés, & Granollers, 2008). The evaluation included three tasks: (1) presentation and background interview, (2) cognitive walkthrough and (3) questionnaire evaluation. Using the Think Aloud usability testing method, users were required to verbalize their thoughts as they interacted with the system. The entire evaluation session was recorded using HyperCam screen recording software (Jaspers, 2009; Kushniruk & Borycki, 2006).

3.4.1. Presentation and background evaluation The aim of this first task was to obtain some background information about the user’s knowledge and to discuss the utility of the system for learning purposes. By means of a computer and a personal interview, the topics shown in Table 2 were presented and discussed:

Table 3 Evaluation topics for the personal interview. Topics for discussion Concept of a health catalog and structured items Current situation and problems of the health catalog - Absence of detailed information regarding the use of items - Multiple catalogs - Misuse of medical items - Lack of common knowledge Concept of ontology Idea of ontology for the health catalog

3.4.2. Walkthrough evaluation The purpose was to evaluate the usability and usefulness of the software application. In addition, the language and vocabulary were assessed. Using the prototype developed in Java, users were required to complete a set of actions described in Table 3. This task provided information to assess the quality criteria of correctness, clarity and coverage of the ontology.

429

3.4.3. Questionnaire evaluation Using a Likert scale (ranging from 1 for completely disagree to 5 for completely agree), users were required to scale a list of assertions regarding the language and vocabulary of the ontology and the usage of the software application (see Table 3).

436

4. Results

441

4.1. OntoHealthCatalog: design and implementation

442

The designed ontology has been named OntoHealthCatalog. As shown in Fig 2, the main class of the ontology is the Item. This is actually the core of the health catalog. The ontology formalizes related knowledge and integrates information so as to provide a response to the four competency questions formulated in Section 3.1. First, it integrates information about the hierarchy of items and the relationships among them (‘‘Health Item Classification’’ section in Fig. 2). Second, the ontology includes information about the structure of the organization in terms of services where can be used (authorized services for their usage) and the location where the item is available (‘‘Health Organization Structure’’ section in Fig. 2). Third, it provides information about properties that characterize both technical and functional aspects of items (‘‘Properties section’’ where the main class has been designated as Item Property). Finally, the process class has been included in the ontology to deal with issues relating to the management of the items (‘‘Health Organization Structure’’ section). This part of the ontology includes the process model for unifying purchases, authorizations for using items from other services and the design of new protocols for clinical trials. All of them will be carried out by hospital personnel according to their role. This is a complex and interesting aspect which will be described in the future development of the ontology. As shown in Fig. 2, each item will be identified by its code, its name and an additional short description. By reporting recommendations and suggesting guidelines for good practices, users will contribute to enhancing knowledge included in the health catalog. Furthermore, problems or events can be reported. As shown in Fig. 2, these events can be reported by different members of the

443

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023

430 431 432 433 434 435

437 438 439 440

444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 6

N. Lasierra et al. / Expert Systems with Applications xxx (2014) xxx–xxx

Fig. 2. Main core of the ontology. 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499

hospital organization and according to their role these events can be classified as more or less reliable. The main activity that causes the event and its duration will be indicated as well as the consequence and the effect that the activity has over the properties (item properties) of the materials (see Fig. 2). This model contributes to the sharing of knowledge based on users’ experience and provides improved advice regarding the usage of products for specific purposes. 4.1.1. Item classification (hierarchy of families) The ontology contains three hierarchy levels for classifying items (see Fig. 3). This classification corresponds with the group, sub-group, and family classification currently used in hospitals. Blood-related-item, Pharma-item and Surgery-care-treatment-item, among others, are the main classes at the family level. Items belonging to the Surgery-care-treatment-item class are classified into the following groups: Catheter-item, Sterilization-item, Hygiene-protection-item, Puncture-incision-item and Bandage-plaster-item. Finally, e.g. the Hygiene-protection-item includes the Glove class, the Cover-surgery-item class, the Disposable clothes class and the Spool-sanitaryWipe-sponge class. Thus, the Glove class has been classified as a subclass of the Hygiene-protection-item class which is included as a type of the Surgery-care-treatment-item class. Properties associated to the high-level item classes are inherited by the child-items, that is the subclasses. Additionally, an annotation property was included to incorporate common information about each class. 4.1.2. Health organization structure One of the main parts of the catalog is related to the structure of the health organization in terms of areas that are authorized

for its management or physical spaces where the items are available. The structure of the health organization comprises warehouses, stocked-items, services, centres, divisions, societies and management groups (see Fig. 3). This knowledge of the organization structure is vital for clearly describing management procedures. Each type of item can be located in different warehouses. Hence, each warehouse keeps a stock of each item and associated information regarding its management inside the warehouse (see Warehouse and Stock-items class in Fig. 3). These items will be used or managed by authorized services (Service class). Cardiology, Surgery, Paediatrics or Traumatology are examples of defined services. These Services belong to a Centre (which may have more than one service). Hospitals and primary care centres are examples of types of centres. Each centre has one or more warehouses. These centres are grouped within Divisions. These divisions are used to manage centres. In the public sector, the divisions are normally associated to areas with a main hospital surrounded by small medical centres. Specifically, each division is managed by a management group associated to a certain area (e.g. pharmacies?). They are in charge of purchasing a certain item for a specific centre. Finally, the top entity is the Society. Societies comprise several divisions and are the main pillars supporting the health catalog management. In the public sector in Spain, the society is the Spanish National Health Service.

500

4.1.3. Properties of items Beyond the taxonomy described above, the ontology provides knowledge about the properties of items and the relationships about the entities. Each item has a wide range of characteristics that can be described. A general description of item properties

525

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023

501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524

526 527 528 529

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 N. Lasierra et al. / Expert Systems with Applications xxx (2014) xxx–xxx

7

Fig. 3. Hierarchy classification and health structure organization sections.

530 531 532 533 534 535 536 537 538 539

has been included in the ontology so that a wide range of items can be covered and new items included in the catalog. These item properties were identified following the first phase of the workflow depicted in Table 1. When describing each item, the appropriate characteristics should be linked to it. As shown in Fig. 4, these item properties have been classified in different classes. Specifically, 13 classes have been identified. These classes enumerate possible values of the item properties. Additionally, each item can be associated with incoming information provided by users. Table 5 summarizes highlighted relations described in the ontology.

4.1.4. Management process Purchasing procedures, the control of in and out movements of items in a warehouse, the control of authorization over the usage of items or the management of users and roles are examples of tasks that may be gathered under the process class included in the ontology. This workflow should be modeled and related to authorized users and areas of the health organization structure. Only the main classes and related information regarding item management for the above-mentioned purposes have been included in this version of the ontology.

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023

540 541 542 543 544 545 546 547 548 549

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 8

N. Lasierra et al. / Expert Systems with Applications xxx (2014) xxx–xxx

550

4.2. Case study: gloves

551

At the end of this study, twenty-seven elements were successfully described according to the ontology. Only minor corrections were made to the model. These modifications related to the inclusion of additional examples of item properties (modeled as instances of certain item properties). The main core was not modified Including knowledge about one type of glove involves creating an instance of the ‘‘glove’’ class and completing all the information related to its characteristics (identification and properties), reported events, availability and management. Fig. 5 shows an example of an instance of a glove modeled with the ontology created following the workflow depicted in Table 6. These descriptions correspond to the first example included in Table 2.

552 553 554 555 556 557 558 559 560 561 562

563

4.3. Proof of concept: prototype development

564

This tool enables two main tasks: (1) visualizing the knowledge modeled in the ontology (both classes and included instances) and (2) managing OWL individuals, which are in fact items. It allows

565 566

browsing the items described in the ontology, editing and deleting the depicted information and including new knowledge. This tool makes the main dynamic development and maintenance of the ontology easier. Two snapshots of the tool are shown in Figs. 6 and 7. The main interface (Fig. 6) is divided into two areas. The left area depicts the hierarchy of health items described in the ontology (and thus in the health catalog) by means of a tree structure. Although the purpose was to design an interface close to those currently used in hospitals, it was proposed to include some elements used for visualizing ontologies so as to take maximum advantage of the taxonomy and to make browsing easier (Katifori, Halatsis, Lepouras, Vassilakis, & Giannopoulou, 2007; Swaminathan & Sivakumar, 2012). The right area allows the visualization of individuals classified as members of the selected item class in the tree. This distribution, together with the buttons depicted in the image below, enables us to configure a zoomable tool which shows the information of the items in different phases and in an organized manner. The characteristics (item properties) of the item are visualized by clicking the ‘‘Visualize’’ button.

Fig. 4. Item properties and process related classes.

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023

567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 N. Lasierra et al. / Expert Systems with Applications xxx (2014) xxx–xxx Table 5 Questions for the final evaluation class. Evaluation assertions 1. The model properly includes all the concepts used in regular health catalogs 2. The language selected for the ontology is accurate according to the language used in the health area 3. The structure of the health system organization is like the one that I know 4. The structure of the health items is appropriate 5. The structure of the characteristics of the items is appropriate 6. The software application is intuitive 7. The application shows the knowledge from the health catalog 8. I would like to have this information in my organization 9. The application (and thus the model) allows any type of health item to be represented 10. This work provides a good solution to a real problem

587 588 589 590 591 592 593 594 595 596 597 598 599 600

The screenshots depicted in Fig. 7 allow the user to introduce a new item guided by the steps explained in Table 6. Specifically, these images are related to the first and second steps (item identification and characterization). One of the main advantages of the tool is its usage for searches. The searcher included in the tool makes it possible to filter health items with specific characteristics or item properties. This is done by means of codified SPARQL queries that are executed over the ontology. Other actions related to creating, editing and deleting individuals are done using Jena methods. An example is the SPARQL query depicted in Fig. 8. The first part of the query is to select the best glove (class = glove) of nitrile material (material = nitrile) appropriate for blood extraction surgery (associatedToHealthprocedure) authorized for use in the surgery

9

service (service = surgery). In addition, the second part of the query is to select the best glove with the same features that is available in the warehouses associated to the service centre, filtering the results to those with a price under 3,5 € (the box of gloves). This query will show the appropriate gloves to use according to the restrictions. Additional functionalities will be given to this tool if the ontology is extended for describing purchasing procedures.

601

4.4. Results of the expert evaluation

608

As shown in Fig. 9 the results of the evaluation were, in general, very positive. In the presentation and interview stage, the experts were consulted about problem and our proposed solution. The extent of their concerns about the health catalog problem and their expectations were established. The experts agreed that having dispersed health catalogs was a problem and that the usage of material was inadequate. They expressed their interest in sharing this knowledge and stressed that this formalization would be positive for the hospital from an economic point of view and that the correct usage of materials would lead to safety improvements. Although the ontology concept was unknown to them, they positively valued the idea behind the usage of the health-catalog ontology in a real environment. Conclusions extracted from the interviews (see Table 3 in Section 3.4) can be gathered within A and E conclusions from the results shown in Fig. 9. The walkthrough evaluation gave the experts a much better understanding of the usage and application of the ontology for this specific case. The evaluation allowed us to detect misinterpretations of some words in the ontology which in some cases hindered the performance of the task. For example, the word (and class) Item

609

Fig. 5. Nitrile glove instance (case study).

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023

602 603 604 605 606 607

610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 10

N. Lasierra et al. / Expert Systems with Applications xxx (2014) xxx–xxx

Table 6 Properties (item properties) for describing health-items. Object-property

Description

hasMaterialProperty hasMechanicalProperty

Refers to the material of the health-item. Latex, Nitrile, Vinyl are examples of materials Refers to the mechanical properties of the item such as elasticity, permeability or resistance. Useful for choosing the appropriate item according to the requirements of a clinical test Describes the surface of the health item such as color, transparency or any special texture Refers to factors against which the health item offers protection, e.g. chemicals Indicates the health procedures that can be executed using the selected health item. E.g. for gloves blood extraction, catering-trade or patients’ examination Indicates the period of time for which the health item can safely be used Describes the capacity of the health item. Useful for syringes. Provides valuable information for needles and stitches Indicates the size of the item e.g. large, small Indicates the measurement used to measure the quantity of the health item e.g. L for liters or ‘‘DBxut’’ for items presented in dispenser boxes of x units. This information is relevant for purchasing Indicates the property that can be achieved for the health item after appropriate treatment e.g. making it sterile Indicates the color of the health item. Useful for distinguishing clothes for surgeons, nurses or admin personnel. This property may be useful for identifying the diameter of different needles Used to model particular characteristics of the health item that cannot be classified in any other group. For instance, ambidextrous gloves

hasSurfaceProperty isProtectedAgainst associatedToHealthProcedure hasDuration hasVolume hasSection hasSize hasUnits achieveQuality hasColour hasParticularProperty

Fig. 6. Software application I: general health catalog screen.

629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644

Property was introduced instead of Item Quality after this evaluation. Hence, this evaluation provided useful information for measuring the correctness and clarity of the ontology. In general the application was found to be intuitive and all the participants were able to finish all the tasks after some slight guidance in some cases. They recognized that the language, vocabulary and the organizational structure of the items was the same as that used in their institutions. They expressed the hierarchy classification of the items was not different from the one provided by their current applications, though. However, they stressed their positive opinion regarding the usage of item properties for the description of the health items which led to the correct usage of the materials according to the desired health technique. This walkthrough evaluation may be seen as part of the application-based study. Not only was the interface design evaluated, but also how the ontology could be used for the purpose it was designed for. It means, for

making easier and clear looking for new health items and contribute to share knowledge. The results obtained after conducting the tasks specified in the walkthrough evaluation (see Table 4 in Section 3.4) can be gathered within C and D conclusions from the results shown in Fig. 6. Finally, all the evaluation assertions were scored between 4 and 5 (see Table 3 Section 3.4). The topics discussed were related to B, C, D and E from the results shown in Fig. 3. No additional information was extracted from this last evaluation. During the previous task experts were required to Think Aloud. Thus they discussed these issues while executing the required tasks.

645

5. Discussion

656

Our proposed ontology-based system constitutes a first step towards improving usage and management of items in health care

657

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023

646 647 648 649 650 651 652 653 654 655

658

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 N. Lasierra et al. / Expert Systems with Applications xxx (2014) xxx–xxx

11

Fig. 7. Software application II: new item screen.

Fig. 8. SPARQL query.

Q4 659 660 661 662 663 664 665 666

Fig. 9. Results of the evaluation.

centres. It offers a solution to align and unify dispersed health catalog modeling items and the structure of the organization related to their management. Furthermore, by describing in detail the features of each product and its potential usage, the proposed system can be used as a rich source of shared knowledge to assess clinical personnel regarding the selection of items. Decisions are expected to be made based on multiple-user experience rather than individual experience. Correct management and usage of materials will

have a positive impact on health care costs and, most important, on clinician and patient safety. Indeed, this education functionality of the health catalog is an extra feature and one of the key contributions of our work. Due to the widespread and extensive use of a health catalog, it was our purpose to offer the description of the main pillars that should be included in the ontology and to offer through a four stage methodology the materials and methods used for the design,

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023

667 668 669 670 671 672 673 674

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 12

N. Lasierra et al. / Expert Systems with Applications xxx (2014) xxx–xxx

Table 4 Evaluation tasks of the walk through phase. Tasks for the walkthrough evaluation 1. Search for a specific health item within the family structure and visualize its characteristics 2. Introduce a new health item 3. Search for a specific health item according to selected characteristics

675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730

development and evaluation of the health catalog. This will help researchers to extend our ontology and verify its usage for other products as we studied with the gloves case in the applicationbased study (stage 2). This stage was performed to validate the usage of our model to correctly describe gloves. The third stage showed us how the ontology could be exploited in a real environment and its potential usage. There is no doubt that this tool was essential for the experts involved during the fourth stage to assess the ontology and the application. Note that although the experts directly evaluated the usability of the software, they also indirectly evaluated the knowledge encoded in the ontology, taxonomy and vocabulary involved. This stage led us to demonstrate how the proposed model and the usage of semantic technologies could be used in practice to support the application of the health catalog. Finally, it was concluded from the expert evaluation that sharing knowledge about the correct usage of materials and their adequacy for each medical practicum was the most interesting and innovative application of our ontology based-system. Most of the information extracted during the evaluation was obtained during the Walkthrough evaluation. We intend to continue using this method as a means of assessing the suitability of the ontology for modeling more products and including further refinements (stage 2). Given the educational functionality proposed in our work, it was necessary to provide a solution for clearly sharing knowledge. The representation of semantics offered by the ontology was the key to selecting this technology for its implementation. However, one issue that should be carefully studied in the future is the storage of all the items modeled as instances of the health catalog. Given that a health catalog may contain more than 20,000 references, there is no doubt that big databases are required to store such a large quantity of data. The combination of the ontology with a database should be studied given the usage of the health catalog for managing and storing a considerable quantity of multiple items. Indeed, available tools for dealing with ontologies allow their persistent storage using powerful relational databases (Martinez-Cruz, Blanco, and Vila (2012)). Their efficiency may lead to an interesting point of discussion and invite further research for its improvement. Apart from taxonomy representation, reasoning and inference mechanisms are additional features of ontologies that can be exploited in depth for this application in the future. For instance, its combination with rules provides flexibility for the configuration of alarms that can be used for management purposes or item usage recommendations, among others. Future directions to improve our proposed solution involves policies and permissions regarding the authority to modify and manage the content of the catalog. The ontology presented in this paper enables to register the author (type of user and role) of reported events regarding the usage of an item and consequently to trace the source of information that leads to different levels of reliability of the reported information. Using the presented solution, this knowledge-sharing can be done by different actors such as clinical or management personnel. However, is required to implement policies for improving the permissions and controls and include within the model the reliability of the included information. For this, rules can be used to implement authorizing policies within the management (creation, edition and general

modification) of the data contained in the system. In fact, different types of personnel should be allowed to edit different types of the item characteristics (e.g. safety, functional or economical). In order to derive this users-categorization, a knowledge audit process is needed. Published literature defines knowledge audit process as an important and initial activity conducted within a knowledge management activity (Rahman and Shukor, 2011; Wu & Li, 2008). This process is conducted within the analysis of requirements stage involving different areas that includes a set of steps for knowledge identification, data gathering techniques, analysis methods and tools to conduct the activity. In our work, an initial phase of analysis was conducted where domain experts were interviewed and related documents used at the hospital to manage supplies were analyzed that led to the first definition of the ontology. While different techniques including domain experts interviews (from different related areas) and literature review were taken into account, complex methods are expected to be used in order to formalize and identify the permissions and thus reliability for the instantiation of the developed ontology regarding the adequate utilization of materials. To elaborate the authorizing model in terms of ontology engineering and rules for the implementation of security measures in the system is a future task to improve the proposed solution. Note that this does not require significant changes in the structure of the proposed model but to include additional properties that link with this information and the items descriptions. However, in order to implement these functionalities, additional work is required in terms of knowledge auditing and collaboration with domain experts in order to identify such organizational, authorizing and reliability issues. This may require a classification more complex than a role-based or competence-based one thus involving social and even personal capabilities.

731

6. Conclusions

762

Sharing and learning knowledge from colleagues’ experiences can be a powerful tool for improving personnel safety and resource management in hospitals and medical centres. As shown in our work, semantic technologies can play an important role by providing clear descriptions and flexible models to gather knowledge without misunderstandings. While the benefits of this system can be measured in terms of education, safety and probably savings in purchases, the introduction of IT (information technology) systems and final implementation in the clinical setting may bring other problems that should be addressed in each specific scenario. The implementation and usage of our tool within a hospital organization will require a test pilot phase (to evaluate and correct detected errors in its daily usage) and later an integration phase with the available IT resources and ERP used in hospitals. Eventually, additional work is required in order to implement policies in the system that enable to assess the reliability of the information included reported by different types of users.

763

Acknowledgments

780

This research work has been supported by project TIN-201123792 from the Spanish Ministry of Economy and Competitiveness (MINECO), the European Regional Development Fund (ERDF), the Regional Government of Aragon (DGA) and the European Social Fund (ESF).

781

References

786

Becker, M., Heine, C., Herrler, R., & Krempels, K. H. (2003). OntHoS-an ontology for hospital scenarios. Applications of Software Agent Technology in the Health Care Domain, 87–103. Berners-Lee, T., Hendler, J., & Lassila, O. (2001). The semantic web. Scientific American, pp. 35–43.

787 788 789 790 791

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023

732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761

764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779

782 783 784 785

ESWA 9292

No. of Pages 13, Model 5G

30 April 2014 N. Lasierra et al. / Expert Systems with Applications xxx (2014) xxx–xxx 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857

Q2

Bouamrane, M. M., Rector, A., & Hurrell, M. (2009). Development of an ontology for a preoperative risk assessment clinical decision support system. In 22nd IEEE international symposium on computer-based medical systems, CBMS (pp. 1–6). IEEE. Brank, J., Grobelnik, M., & Mladenic´, D. (2005). A survey of ontology evaluation techniques. In Proceedings of the conference on data mining and data warehouses, SiKDD. Bright, T. J., Yoko Furuya, E., Kuperman, G. J., Cimino, J. J., & Bakken, S. (2012). Development and evaluation of an ontology for guiding appropriate antibiotic prescribing. Journal of Biomedical Informatics, 45(1), 120–128. Burstein, F., & Carlsson, S. (2008). Decision support through knowledge management. In F. Burstein & C. W. Holsapple (Eds.), Handbook on decision support systems 1: Basic themes (pp. 103–120). Berlin Germany: Springer-Verlag. Campana, F., Moreno, A., Riaño, D., & Varga, L. Z. (2008). K4care: Knowledge-based homecare e-services for an ageing Europe. Agent Technology and E-Health, 95–115. Chandra, C., & Kachhal, S. K. (2004). Managing health care supply chain: trends, issues, and solutions from a logistics perspective. In Proceedings of the sixteenth annual society of health systems management engineering forum (pp. 20–21). Chandra, C., & Tumanyan, A. (2007). Organization and problem ontology for supply chain information support system. Data & Knowledge Engineering, 61, 263–280. Corcho, O., Fernández-López, M., & Gómez-Pérez, A. (2003). Methodologies, tools and languages for building ontologies. Where is their meeting point? Data & Knowledge Engineering, 46(1), 41–64. ECDA, European Chronic Disease Alliance Policy Report (2010). A unified prevention approach. [Available online]: . Giménez, D. M., Vegetti, M., Leone, H. P., & Henning, G. P. (2008). PRoduct ONTOlogy: Defining product-related concepts for logistics planning activities. Computers in Industry, 59(2), 231–241. González, M. P., Lorés, J., & Granollers, A. (2008). Enhancing usability testing through datamining techniques: A novel approach to detecting usability problem patterns for a context of use. Information and Software Technology, 50(6), 547–568. Haghighi, P. D., Burstein, F., Zaslavsky, A., & Arbon, P. (2012). Development and evaluation of ontology for intelligent decision support in medical emergency management for mass gatherings. Decision Support Systems. Heinrich, S., Luppa, M., Matschinger, H., Angermeyer, M. C., Riedel-Heller, S. G., & König, H. H. (2008). Service utilization and health-care costs in the advanced elderly. Value in Health, 11(4), 611–620. Jara, A. J., Belchi, F. J., Alcolea, A. F., Santa, J., Zamora-Izquierdo, M. A., & GómezSkarmeta, A. F. (2010). A pharmaceutical intelligent information system to detect allergies and adverse drugs reactions based on internet of things. In 8th IEEE international conference on pervasive computing and communications workshops (PERCOM workshops) (pp. 809–812). IEEE. Jaspers, M. W. (2009). A comparison of usability methods for testing interactive health technologies: Methodological aspects and empirical evidence. International Journal of Medical Informatics, 78(5), 340. Jena Framework (2009). [Available online]: . Katifori, A., Halatsis, C., Lepouras, G., Vassilakis, C., & Giannopoulou, E. (2007). Ontology visualization methods – A survey. ACM Computing Surveys (CSUR), 39(4), 10. Kerr, L. N., Chaput, M. P., Cash, L. D., O’Malley, L. G., Sarhrani, E. M., Teixeira, J. C., et al. (2004). Assessment of the durability of medical examination gloves. Journal of Occupational and Environmental Hygiene, 1(9), 607–612. Kushniruk, A. W., & Borycki, E. M. (2006). Low-cost rapid usability engineering: Designing and customizing usable healthcare information systems. Healthc Q, 9(4), 98–100. Kushniruk, A. W., & Patel, V. L. (2004). Cognitive and usability engineering methods for the evaluation of clinical information systems. Journal of Biomedical Informatics, 37(1), 56. Kuziemsky, C. E., & Lau, F. (2010). A four stage approach for ontology-based health information system design. Artificial Intelligence in Medicine, 50(3), 133–148. Lasierra, N., Alesanco, A., Guillén, S., & García, J. (2013). A three stage methodology approach for personalized chronic care at home. Journal of Biomedical Informatics, 46, 516–529.

13

Lee, J., Chae, H., Kim, C. H., & Kim, K. (2009). Design of product ontology architecture for collaborative enterprises. Expert Systems with Applications, 36(2), 2300–2309. Martinez-Cruz, C., Blanco, I. J., & Vila, M. A. (2012). Ontologies versus relational databases: are they so different? A comparison. Artificial Intelligence Review, 38(4), 271–290. Monteagudo, J. L., & Moreno, O. (2007). e-Health for patient empowerment in Europe. Instituto de Salud Carlos III, Ministerio de Sanidad y Consumo, España. Moore, G., Dunnill, C. W., Peter, A., & Wilson, R. (2013). The effect of glove material upon the transfer of methicillin-resistant staphylococcus aureus to and from a gloved hand. American Journal of Infection Control, 41(1), 19–23. Noy, N. F., & McGuinness, D. L. (2001). Ontology development 101: A guide to creating your first ontology. Technical Report KSL-01-05, Stanford Knowledge Systems Laboratory first. OWL 1 Web Ontology Language Guide. W3C Recommendation (2004). [Available online]: . OWL 2 Web ontology language (2009). [Available online]: . Panetto, H., Dassisti, M., & Tursi, A. (2012). ONTO-PDM: Product-driven ONTOlogy for product data management interoperability within manufacturing process environment. Advanced Engineering Informatics. Pinto, H. S., & Martins, J. P. (2004). Ontologies: How can they be built? Knowledge and Information Systems, 6(4), 441–464. Protégé-OWL Editor (2004). [Available online]: . Rahman, A. A., & Shukor, N. S. A. (2011). Knowledge audit process-tales of two organizations. In International conference on research and innovation in information systems (ICRIIS) (pp. 1–5). Saa, R., Garcia, A., Gomez, C., Carretero, J., & Garcia-Carballeira, F. (2012). An ontology-driven decision support system for high-performance and costoptimized design of complex railway portal frames. Expert Systems with Applications. Scheuermann, A., & Hoxha, J. (2012). Ontologies for intelligent provision of logistics services. In ICIW 2012, the seventh international conference on internet and web applications and services (pp. 106–111). Songthung, P., Sripanidkulchai, K., Luangruangrong, P., Sakulbumrungsil, R. C., Udomaksorn, S., Kessomboon, N., et al. (2012). An innovative decision support service for improving pharmaceutical acquisition capabilities. In SRII global conference (SRII), 2012 annual (pp. 628–636). IEEE. SPARQL Query Language for RDF (2008). W3C Recommendation, [Available online]: . Stoessel, & Smith, S. M. (2008), Does the glove fit? Critical Considerations for the selection of medical gloves, the clinical issue, issue 4, Kimberly-Clark Healthcare education. Studer, R., Benjamins, V. R., & Fensel, D. (1998). Knowledge engineering: Principles and methods. Data & Knowledge Engineering, 25(1), 161–197. Swaminathan, V., & Sivakumar, R. (2012). Comprehensive ontology cognitive assisted visualization tools – A survey. Journal of Theoretical and Applied Information Technology, 41(1). Valls, A., Gibert, K., Sánchez, D., & Batet, M. (2010). Using ontologies for structuring organizational knowledge in home care assistance. International Journal of Medical Informatics, 79(5), 370. Vegting, I. L., van Beneden, M., Kramer, M. H., Thijs, A., Kostense, P. J., & Nanayakkara, P. W. (2012). How to save costs by reducing unnecessary testing: Lean thinking in clinical practice. European Journal of Internal Medicine, 23(1), 70–75. World Health Organization 1, WHO Guidelines on Hand Hygiene in Health care global patient (2006). [Available online] . World Health Organization 2 (2009). Glove use information leaflet [available online] . Wickstrøm, J., Serup-Hansen, N., & Kristiansen, I. S. (2002). Future health care costs – do health care costs during the last year of life matter? Health Policy, 62(2), 161–172. Wu, Y. L., & Li, Y. H. (2008). Research on the model of knowledge audit. In 4th international conference on wireless communications, networking and mobile computing, WiCOM’08 (pp. 1–4). IEEE.

Q3

858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924

Please cite this article in press as: Lasierra, N., et al. Towards improving usage and management of supplies in healthcare: An ontology-based solution for sharing knowledge. Expert Systems with Applications (2014), http://dx.doi.org/10.1016/j.eswa.2014.04.023