Critical Evaluation of RFID Applications in Healthcare

Critical Evaluation of RFID Applications in Healthcare Yu-Ju Tu1 , Huihui Chi2 , Wei Zhou2 , Gaurav Kapoor3 , Enes Eryarsoy4 , and Selwyn Piramuthu5(B) 1 Management Information Systems, National Chengchi University, Taipei, Taiwan 2 Information and Operations Management, ESCP Europe, Paris, France 3 Infocomm Technology, Singapore Institute of Technology, Singapore, Singapore 4 Information Systems, Istanbul Sehir University, Istanbul, Turkey 5 ISOM, University of Florida, Gainesville, FL, USA [email protected] Abstract. The past two decades have witnessed the explosive growth of RFID-based applications in healthcare settings. This includes appli- cations that are related to ePedigree for pharmaceuticals, asset tracking and workflow management in hospitals, pervasive healthcare, among oth- ers. We consider some of the benefits that are provided by RFID-based systems in healthcare settings. We then provide a critical evaluation of some of the challenges that are faced by these applications. 1 Introduction and RFID-Based Systems in Healthcare RFID (Radio Frequency IDentification) tags belong to a class of auto- identification technologies that also include barcode (e.g., Zhou 2009). These resource-constrained tags have minimal processing power as well as some data storage capability. Of the three main types of tags that include active tags, semi- passive tags, and passive tags, the latter is the most popular primarily because of cost and form factor reasons. Some of the characteristics of RFID tags include the ability to store and process data, being able to communicate with a reader that may not be in direct line-of-sight, durability in harsh environments, batch readability, storage of item-level information, ability to carry on a two-way con- versation with a reader, among others. The unique identifier in each RFID tag allows for uniquely identifying tagged items and the ability to communicate with a reader allows for process automation. Item-level information has been beneficially used in ePedigree systems in the pharmaceutical industry to track and trace individual items (e.g., Coustasse et al. 2010). Item-level information has also been used in hospitals to manage process flows (e.g., Gonzalez et al. 2006b), manage inventory (e.g., Meiller et al. 2011), provide automated data for electronic medical record systems (e.g., Mongan et al. 2016), facilitate patient management systems (e.g., Chowdhury and Khosla 2007), as well as other applications such as in surgical sponges (e.g., Williams et al. 2014). RFID is one of the core technologies in pervasive healthcare applications (e.g., Thuemmler et al. 2009). c Springer Nature Switzerland AG 2019 R. Doss et al. (Eds.): FNSS 2019, CCIS 1113, pp. 240–248, 2019. https://doi.org/10.1007/978-3-030-34353-8_18  RFID in Healthcare 241 Given the flexibility provided by RFID-based systems, its healthcare appli- cations span a large functional range that include automated data collection, identification, sensing, tracking, as well as managing assets and entities such as blood samples, drugs, hospital equipment, patients, among others. RFID-based systems have also been used to address effectiveness of care, operational efficien- cies, patient safety, and quality of care. At hospitals, RFID-based systems are used to link mother with her newborn by storing the mother’s information in the baby’s tag. This allows for quick identification and matching of the mother with her baby. Moreover, RFID-based devices such as the monitors that are used around the ankle or wrist of babies help ensure the safety of babies in hospitals. For example, when a baby with this tag is moved near a hospital door, the door is programmed to automatically close. This prevents kidnapping incidents in pediatric and maternity rooms. RFID tags that are on patients and hospital staff (e.g., on their identification badges) facilitate automatic provision of data for electronic medical records sys- tems. Similarly, RFID tagged medical equipment and supplies are readily tracked and traced. This results in increased efficiency, thereby lowering the costs and increasing the offered service quality. From a survey of published literature, Coustasse and Tomblin (2013) found that RFID use in healthcare settings lead to several benefits and improvements such as the ability to monitor the ambient conditions of perishable and heat-sensitive items, less shrinkage due to theft or misplacement, improved staff productivity and enhanced quality improvement, labor savings and error reduction such as through the use of RFID-enabled sur- gical sponge (e.g., Williams et al. 2014), better knowledge on equipment avail- ability, improved business processes and workflow, ability to track and manage mobile assets as well as high-cost devices and supplies in real-time, tracking of blood samples, and ensuring compatibility of blood transfusions through RFID tags on the sample as well as the patient. RFID-based systems have therefore been effectively used in healthcare set- tings to reduce costs through improved process efficiency and reduced equip- ment shrinkage that includes misplacement and theft, improve patient safety, simplified patient billing due to linkage with electronic medical record system, and improve supply chain effectiveness. RFID also helps with the automation of processes, thereby removal of human factors from clinician workflow, better inventory management and reduction in stock out situations. Clearly, there are numerous advantages to using RFID-based systems in healthcare settings. On the other hand, these advantages don’t come without related issues. We discuss some of these in the next section. 2 Challenges and Issues with RFID Use in Healthcare Although there has been a steady increase in the number of RFID-based systems that have been implemented in healthcare settings over the last decade, barriers to widespread and rapid adoption include the additional cost of such imple- mentations, unclear return on investment, and competition from other strate- 242 Y.-J. Tu et al. gic initiatives. Cost has been used as an excuse for resistance to adopt RFID- based systems. When invaluable human lives are at stake, the marginal cost of implementing RFID-based systems pales in comparison to the indispensable and expensive equipments that are used in healthcare settings. Moreover, such arguments only look at one side of the coin - the cost side. Rarely do such line- of-thought include the other side of the coin - the benefit side. When both costs and benefits are simultaneously considered, it has been shown (e.g., Piramuthu et al. 2014) that RFID-based systems win out in a large number of cases because of the tremendous amount of benefits that accrue due to the real-time item-level visibility of tagged items. The unclear return on investment is due to the fact that we are at the early stage of RFID adoption. This concern will vanish as more and more healthcare institutions start implementing RFID-based systems. As for competition with other strategic initiatives, it is a matter of priority since real-time visibility is of paramount importance in healthcare settings. There are other issues that are not specific to the healthcare context. For a detailed explanation of these issues, the reader is referred to Kapoor et al. (2009). Specifically, these concerns include ownership transfer issues, privacy/security issues, back-end system bottleneck issues, risk of obsolescence, read rate error, economic disincentives to share item-level information across institutions, and evolving standards. We update this list with a few more concerns that we list and discuss below. 2.1 Data Volume Each RFID tag generates a large amount of data since it is in real-time con- tact with reader(s) and each communication instance generates data. Therefore, generation of large volumes of data in RFID-based systems is not difficult. Such data need to be efficiently stored somewhere (e.g., Fazzinga et al. 2009). It is a useless exercise to efficiently store data, when such data are not processed for some useful purpose. Analysis of such massive amounts of RFID-generated data sets is a significant issue in RFID-based systems (e.g., Gonzalez et al. 2006a). With advanced sensing technologies, RFID is capable of capturing data in a wide range of healthcare-related activities (e.g., Alvarez Lopez et al. 2018). RFID enables autonomous data collection during patient location tracking, patient identification, patient medication and monitoring, as well as drug and medi- cal equipment inventory management processes. For example, RFID in sensor- enabled pill allows for the provision of personalized treatment to individual patients (Mathew et al. 2018). On consumption, such a pill extracts vital organ conditions of the patient and transmits the information to the information sys- tem for analysis. It is understandable that patient health is reliant on taking medications regularly. To this end, RFID sensor-based pill bottles (Mathew et al. 2018) are used to ensure that patients take the right dose of medication at the right time. Similarly, sensor-based RFID tags can be used to automat- ically record vital patient information such as heart rate and blood pressure along with the patient’s location and time. These signify the generation of large  RFID in Healthcare 243 volumes of related data that need to be stored and processed for actionable intelligence. Moreover, the extraction of accurate and relevant healthcare infor- mation takes time. And, timeliness is generally considered to be very important with respect to healthcare decision-making quality, transparency, and integrity. This is especially significant when there is an urgent need to immediately access or exchange information among healthcare stakeholders such as patients, physi- cians, and other medical staff. The management of such data in large volumes to support decision-making in a timely manner is a critical requirement in health- care systems. 2.2 Electromagnetic Interference While there are several concerns with the use of RFID-based systems in health- care settings, none of these concerns is as serious as the one associated with electromagnetic interference. In a hospital setting, RFID-based systems are very useful for tracking healthcare equipment, misplacement prevention, and peri- odic reminders to hospital administration staff as well as triggers when repair or replacement of equipment/devices are necessary. To a certain degree, hospi- tal can be compared to the firm that is composed of a number of divisions or departments. This firm needs to take care of not only people but also thousands of assets. In particular, these assets include movable devices that have to be shared by different divisions across several floors, and include gravity infusion devices (infusion pumps), electrocardiogram (ECG) machines, and ventilators. More uniquely, because life-threatening emergencies are hard to predict, a major- ity of such devices are extremely crucial in terms of their availability and mobility with respect to caring for the seriously ill patients. Thus, attaching RFID tags to these devices to monitor these life-saving assets in real-time is common in hospital to ensure that they can perform well at the right place and the right time. However, preliminary testing by the US Food and Drug Administration (FDA) has revealed that the frequency or energy emitted by some RFID-based systems could potentially affect sensitive devices. This issue of electromagnetic interference, that has the potential to affect normal operation of medical devices, could result in serious harm to patients and put their lives in danger (e.g., Ashar and Ferriter 2007, Seidman et al. 2010, Togt et al. 2008). For example, once they are in the proximity of an RFID tag, these critical healthcare equipment may malfunction. The possible consequences include the switch-off for external peace- maker, the disturbance in the atrial and ventricular electrogram curve generated by the pacemaker, change in set ventilation rate of mechanical ventilator, and complete stoppage of syringe pump or renal replacement device. To minimize such EMI-caused danger, hospitals often consider strictly regulating the signal strength used in RFID-based systems. Similarly, another electromagnetic interference-related challenge in health- care pertains to avoiding the false read of RFID based systems, because the signal strength is one determining factor in the accuracy of RFID reads. It is certain that the reliable communication between RFID tag and reader is subject 244 Y.-J. Tu et al. to several conditions, such as the orientation of the tagged object. Moreover, the presence of any liquid, metal, or even mechanical vibration between the tag and its associated reader is not uncommon in hospital environments, but they all can possibly cause coupling interference effect. This effect in turn may reduce the effective reading range and increase the false read rates in RFID-based sys- tems (Haddara and Staaby 2018, Jebali and Kouki 2018). More critically, all of these conditions would become worse, when there is a degradation of the sig- nal strength or sensitivity in RFID based systems due to the electromagnetic interference concern in hospital environments. Therefore, RFID read-rate accu- racy is a challenging issue in healthcare settings (e.g., Tu et al. 2009, Zhou and Piramuthu 2018). 2.3 Tag Separation It is generally assumed that once an RFID tag is placed in position, it stays there forever or at least stays put until its useful life with that object ends and it is then removed from that tagged object. When an RFID tag is placed on top of the tagged item, it is possible for the tag to become separated due to any number of innocuous reasons. However, it is also possible for an adversary to remove the tag from the tagged item in order to achieve some nefarious purpose. While the former results in the ‘loss’ of the tagged item for all practical purposes in automated systems, the latter could precipitate in serious damage to the patient, healthcare personnel, or healthcare organization. When the former occurs, there has to be a mechanism in place that identifies the event as it occurs and resolves the issue at the earliest possible time before any damage is done. When the latter occurs, it would be nice to have a trigger that goes off immediately so that someone in charge can take necessary action to rectify the situation. The damage caused due to intentional tag separation depends on the motive of the perpetrator and could affect patient(s), hospital personnel, as well as medical equipment and supplies. To our knowledge, the only publication that specifically even considers this eventuality is Tu et al. (2018), who attempt to provide a means to address this situation. RFID tag detachment in the healthcare domain may result in the manage- ment of related supply chain(s) to be compromised. As RFID tags become com- monplace to track medical equipment and supplies stock as well as to manage their inventory, the complete trust and reliance placed on RFID tags for these purposes could be violated as tag separation incidents mount. For example, some medicines as well as vaccines are strictly required to be stored under controlled conditions that include allowed temperature range as well as exposure to light. The shelf-life of some of these items are also very short. In RFID-based health- care systems, the process of checking the stock level and making the appropriate call-offs (orders) to replace anything that is about to run out can be readily automated. The accomplishment of such an automation requires the placement of RFID tags on or inside every medicine bottle, container, tray, etc. in order to allow for ease of access of necessary information such as name, type, vendor, series number, quantity, and expiry date (i.e., medicine profile data). In such  RFID in Healthcare 245 a system that completely relies upon RFID-generated data for a wide range of purposes that include compliance determination to ensuring everyone’s safety in a healthcare environment, RFID tag detachments have the potential to lead to unexpected negative consequences. For example, planning based on existing perturbed inventory is bound to be inaccurate due to false read of item profile data that arise from RFID tag detachments. This in turn could result in rush orders or unnecessary inventory. More importantly, when RFID tag detachment instances exceed a critical threshold, related effects on inventory management could prove to be disastrous. For example, the stock level of an important irre- placeable item that generally takes a long time to be restocked could fall short of its safe inventory level. Such an incident could cascade to horrible outcomes. In a majority of supply chains, the consequences of focal items that are out-of- stock due to tag detachment may be tolerable. However, in a healthcare supply chain in which a life-saving focal item is out-of-stock, RFID tag detachment might result in extraordinary consequences in terms of protection and safety of human lives. 2.4 Relay Attack RFID-based system are prone to several types of attacks such as cloning, denial of service, eavesdropping, impersonation, malicious code injection, replay attack, side-channel attack, and spoofing (e.g., Mitrokotsa et al. 2010). Over the years, researchers have developed means to address most of these attacks. However, relay attacks have proved to be immune to attempts at addressing them. This is primarily because relay attacks simply relay messages between the two parties (here, RFID tag and reader) without any modification to any of the commu- nicated messages. The adversary mounting a relay attack intends to effectively shorten the physical separation of the tag and reader in order to convince the reader to do something. For example, an adversary can relay messages between an RFID-embedded car key and the car to open and start the car even though the car key might be farther away from the car. In a healthcare context, an adversary could potentially mount a relay attack on any of the devices that communicate through wireless means and wreak havoc. To our knowledge, so far, none of the medical devices are protected against relay attacks. Although researchers have attempted to develop solutions against relay attacks, none of them are failsafe. A relay attack tricks the tag and reader into believing that they are in communication with each other. A reader and tag can communicate with each other only when they are in close physical proximity of each other. Communication between a tag and a reader are relayed by the adversary during a relay attack. This essentially signifies that the adversary essentially virtually shortens the distance between tag and reader during the attack. There are two streams of research in this general area. One stream uses distance-bounding method in which the distance between the RFID tag and its reader is estimated by measuring the round-trip time taken by a signal to travel between the two. When the estimated distance is more than the communication 246 Y.-J. Tu et al. range of the tag and reader, this signifies that a relay attack is in progress. However, since radio waves move at the speed of light, the instruments that measure the signal speed needs to be extremely accurate at the nanosecond scale. Even a small measurement distortion or error could render the measured value invalid. Such a distortion can easily happen since the signal goes from the reader to the tag, which then processes it and then sends the signal back to the reader. If the time it takes for the tag to process the signal is an order more than the signal travel time, the travel time can easily wash away. Clearly, distance-bounding methods face an uphill battle to prove their effectiveness. The other stream of research uses ambient condition or context at both reader and tag to determine their physical proximity. The underlying idea behind this stream of research is that the ambient conditions at the reader and tag are bound to be the same when they are in close physical proximity of each other. Researchers have used various ambient condition facets such as temperature, light, sound, among others. The issue with this method is that some of the ambient conditions such as light and sound are directional, and could result in different results even though the reader and tag are physically next to each other. 3 Discussion RFID is a promising technology for healthcare environments where information generated in these tags can be used to replace error-prone human input under some circumstances. Erroneous medical identification, patient misidentification, and such errors in healthcare settings have the potential to do unnecessary harm to patient health. In the United States alone, medical errors have resulted in more than forty million adverse events and have taken away more than two million lives a year (Duroc and Tedjini 2018, Haddara and Staaby 2018). It is therefore not surprising that hospitals and healthcare organizations in general are inter- ested in RFID-based systems to enhance or enable autonomous identification, tracking, monitoring, and management of healthcare procedures and processes that include drug compliance and healthcare supply chain. Technologies associated with RFID-based systems have been evolving. For example, some passive RFID tags are already capable of measuring physical motion, physical displacement, physical deformation, airflow temperature, and humidity (Tu et al. 2018, Duroc and Tedjini 2018). Meanwhile, novel RFID designs have been focused on leveraging the electromagnetic signature of the tag itself to be the tag identifier (Duroc and Tedjini 2018). This suggests that the unit cost of RFID tag may thus be further reduced because of the removal of the ‘chip’ inside. There are also novel designs to make RFID tag bendable and combining this with printed sensor or printed battery through 3D printing technology (Duroc and Tedjini 2018, Haddara and Staaby 2018). All of these technological advances have the potential to make the application of RFID tag more attractive and adaptive to healthcare environments. On the other hand, as discussed earlier in this paper, there are still issues for healthcare organizations to seriously consider when adopting RFID-based systems to improve healthcare efficiency and quality.  RFID in Healthcare 247 The concerns with RFID-based systems in the healthcare domain are similar to those in other domains, except for electromagnetic interference and the fact that when something goes wrong, invaluable human lives might be put at stake. Moreover, RFID data privacy and security are critical in healthcare settings. For example, it is a violation of HIPAA (the Health Insurance Portability and Accountability Act) in the United States to let an RFID tag involve any patient data without necessary protection in place. These differences may provide a reasonable explanation as to why the adoption of RFID in the healthcare sector still lags behind that in other industries, despite the many promised benefits of RFID in the healthcare industry. In addition to the challenges that were considered in Kapoor et al. (2009), we identified a few other key issues from the perspective of healthcare applications. More importantly, while all of these issues are non-trivial, prior literature has failed to provide necessary attention to these topics in healthcare settings. We expect that, because of the concise review and critical evaluation of such key issues in this study, there will be more fruitful investigations to address these issues in the near future. References Alvarez Lopez, Y., Franssen, J., Alvarez Narciandi, G., Pagnozzi, J., Gonzalez-Pinto Arrillaga, I., Las-Heras Andres, F.: RFID technology for management and tracking: e-health applications. Sensors 18(8), 2663, 1–17 (2018) Ashar, B.S., Ferriter, A.: Radio frequency identification technology in health care: benefits and potential risks. J. Am. Med. Assoc. 298, 2305–2307 (2007) Chowdhury, B., Khosla, R.: RFID-based hospital real-time patient management sys- tem. In: Proceedings of 6th IEEE/ACIS International Conference on Computer and Information Science (ICIS), pp. 363–368 (2007) Coustasse, A., Arvidson, C., Rutsohn, P.: Pharmaceutical counterfeiting and the RFID technology intervention. J. Hosp. Mark. Public Relat. 20(2), 100–115 (2010) Coustasse, A., Tomblin, S.: Impact of radio-frequency identification (RFID) technolo- gies on the hospital supply chain: a literature review. Perspect. Health Inf. Manag. 10, 1d (2013) Duroc, Y., Tedjini, S.: RFID: a key technology for humanity. Comptes Rendus Phys. 19(1–2), 64–71 (2018) Fazzinga, B., Flesca, S., Masciari, E., Furfaro, F.: Efficient and effective RFID data warehousing. In: Proceedings of the International Database Engineering & Applica- tions Symposium, pp. 251–258 (2009) Gonzalez, H., Han, J., Li, X., Klabjan, D.: Warehousing and analysis of massive RFID data sets. In: Proceedings of the International Conference on Data Engineering (ICDE) (2006a) Gonzalez, H., Han, J., Li, X.: Mining compressed commodity workflows from mas- sive RFID data sets. In: Proceedings of the 15th ACM International Conference on Information and Knowledge Management, pp. 162–171 (2006b) Haddara, M., Staaby, A.: RFID applications and adoptions in healthcare: a review on patient safety. Proc. Comput. Sci. 138, 80–88 (2018) Jebali, C., Kouki, A.B.: Read range/rate improvement of an LF RFID-based tracking system. IEEE J. Radio Freq. Identif. 2(2), 73–79 (2018) 248 Y.-J. Tu et al. Kapoor, G., Zhou, W., Piramuthu, S.: Challenges associated with RFID implementa- tions in supply chains. Eur. J. Inf. Syst. 18, 526–533 (2009) Mathew, P.S., Pillai, A.S., Palade, V.: Applications of IoT in healthcare. In: Sangaiah, A.K., Thangavelu, A., Meenakshi Sundaram, V. (eds.) Cognitive Computing for Big Data Systems Over IoT. LNDECT, vol. 14, pp. 263–288. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-70688-7 11 Meiller, Y., Bureau, S., Zhou, W., Piramuthu, S.: Adaptive knowledge-based system for health care applications with RFID-generated information. Decis. Support Syst. 51(1), 198–207 (2011) Mitrokotsa, A., Rieback, M.R., Tanenbaum, A.S.: Classifying RFID attacks and defenses. Inf. Syst. Front. 12(5), 491–505 (2010) Mongan, W., et al.: A multi-disciplinary framework for continuous biomedical mon- itoring using low-power passive RFID-based wireless wearable sensors. In: IEEE International Conference on Smart Computing (SMARTCOMP), pp. 1–6 (2016) Piramuthu, S., Wochner, S., Grunow, M.: Should retail stores also RFID-tag the ‘cheap’ item? Eur. J. Oper. Res. 233(1), 281–291 (2014) Seidman, S.J., et al.: In vitro tests reveal sample radio frequency identification readers inducing clinically significant electromagnetic interference to implantable pacemak- ers and implantable cardioverter-defibrillators. Heart Rhythm 7(1), 99–107 (2010) Thuemmler, C., Buchanan, W., Fekri, A.H., Lawson, A.: Radio frequency identification (RFID) in pervasive healthcare. Int. J. Healthc. Technol. Manag. 10(1/2), 119–131 (2009) Togt, R., van Lieshout, E.J., Hensbroek, R., Beinat, E., Binnekade, J.M., Bakker, P.J.: Electromagnetic interference from radio frequency identification inducing potentially hazardous incidents in critical care medical equipment. J. Am. Med. Assoc. 299, 2884–2890 (2008) Tu, Y.-J., Zhou, W., Piramuthu, S.: Identifying RFID-embedded objects in pervasive healthcare applications. Decis. Support Syst. 46(2), 586–593 (2009) Tu, Y.-J., Zhou, W., Piramuthu, S.: A novel means to address RFID tag/item separa- tion in supply chains. Decis. Support Syst. 115, 13–23 (2018) Williams, T.L., Tung, D.K., Steelman, V.M., Chang, P.K., Szekendi, M.K.: Retained surgical sponges: findings from incident reports and a cost-benefit analysis of radiofrequency technology. J. Am. Coll. Surg. 1–16 (2014) Zhou, W.: RFID and item-level information visibility. Eur. J. Oper. Res. 198(1), 252– 258 (2009) Zhou, W., Piramuthu, S.: IoT security perspective of a flexible healthcare supply chain. Inf. Technol. Manage. 19(3), 141–153 (2018)