The Benefits and Challenges of Computerized Physician Order Entry (CPOE): How are physicians affecting implementation?
Anita Marban
University of Maryland

The Benefits and Challenges of Computerized Physician Order Entry (CPOE): How are physicians affecting implementation? Executive Summary: Physicians have always been the driving force behind the success of new technologies and their effects on healthcare. They are interested in new ways of providing care by utilizing medications or trying new procedures and medical devices. Through research they seek to understand the human body and find cures for the diseases that attack it. They spend years in school and incur debt in upwards of $200,000 dollars before they graduate and have the opportunity to practice independently. Confronted with technology that can improve patient safety by up to 95% and save billions of dollars, they resist change. In the U.S. less than 10% of hospitals and less that 25% of physician offices have fully functioning CPOE systems. Cedars Mt Sinai pulled the plug on their multi-million dollar CPOE system, as did 6 other hospitals because physicians refused to use them. A review of the literature shows that approximately 68% of physicians surveyed identify CPOE as the solution to preventing many medical errors, adopting best-clinical-practices and reducing healthcare costs by billions of dollars annually. It also shows that physicians have issues and concerns with adopting CPOE and until hospitals and government regulatory bodies are ready to, not only address, but solve them, CPOE has little chance of being successful any time soon. CPOE is one of solutions to resolve the issues plaguing the U.S. healthcare system. We have the knowledge and capability to address the issues and concerns of physicians around the cost of purchasing and maintaining a system; their perceived value of adopting a system; for their loss of autonomy and the errors being facilitated by CPOE. Additionally, by addressing and developing solutions to interoperability issues we can move forward with having and effectively utilizing CPOE to its fullest capacity. Introduction: Health information technology (HIT) has demonstrated the ability to improve the quality, delivery, continuity and affordability of care. Surveys show that healthcare professionals strongly believe that HIT would have a positive impact on healthcare and reduce medical errors (HIMSS Leadership Survey, 2009). The healthcare field however, has prolonged adopting modern information technology even as they incorporate advanced technology to treat patients (Levey). Numerous studies show that information technology has the potential to improve patient safety by decreasing the number of medical errors in a system that is plagued by error and inefficiency. In To Err is Human: Building a Safer Health System, the Institute of Medicine brought forth factors contributing to as many as 98,000 deaths each year in U.S. hospitals due to medical errors, especially around medication and order transcription. The Leapfrog Group reported in 2008 that as many as one million serious medication errors occur every year because of the administration of the wrong drug or the wrong dose, when drug interactions or allergies are missed, and when illegible handwriting is transcribed incorrectly (Leapfrog Group, 2008). With medication errors costing on average of $2,000 per occurrence, that translates to $2 billion dollars per year nationwide (Leapfrog, 2008). One specialized software application that has been identified by national organizations such as the Institute of Medicine, Joint Commission, President’s Information Technology Advisory Committee, and Leapfrog Group as the silver bullet to prevent these errors is computerized physician order entry (CPOE) (Cutler, Feldman, & Horwitz, 2005). CPOE is an electronic system that allows orders to be entered directly into the computer. The advantages of CPOE over handwritten orders include the ability to decrease the number of transcription errors related to illegible handwriting, to organize the ordering process to help physicians enter complete orders, to provide clinical decision support (CDS) through diagnosis order sets, and to automatically check orders for potential drug-allergy, drug-drug and drug-food interactions (Lindenauer et al., 2006). Other benefits include: • Prompts that warn against the possibility of drug allergy, or overdose; • Accurate, current information that helps physicians keep up with new drugs as they are introduced into the market; • Drug-specific information that eliminates confusion among drug names that sound alike; • Improved communication between physicians and pharmacists; and • Reduced healthcare costs due to improved efficiencies (Leapfrog Group, 2009).

Even though the benefits of CPOE are widely recognized, few hospitals in the U.S. have successfully implemented this system. According to 2008 HIMSS analysis of 5166 hospitals that are following the EMR Adoption Model, only 2.5% of these hospitals have reached stage 4: CPOE, CDS out of the 7 stages to a fully functional EMR system. The facilities that have successfully implemented CPOE have depended on house staff to enter the majority of orders (Kaushal, Shojania, & Bates, 2003) and on developing the applications internally through their highly experienced IT department. It is important to note that most patients are cared for by attending physicians who do not have the luxury of being assisted by house staff, and that most hospitals don’t have highly functioning IT departments that are capable of developing applications internally (www.ahrq.gov, 2004). The literature suggests that cost may be a primary reason for low implementation of CPOE. Estimates vary widely for set up alone ranging from $3 million to $10 million dollars. A start up of $8 million and $1 million for annual maintenance translates to a per bed expense of $32,000 initially and $4,000 annually ongoing in a rural area. Cutler, Feldman and Horwitz identified that provider investment is determined primarily by uncommitted money to invest in projects. In this setting, one would expect positive net income or system membership to matter for investment. What the authors identified is a role for hospital ownership status in explaining CPOE investment; “Neither net income nor system membership is positively associated with CPOE investment” (Cutler, Feldman, & Horwitz, 2005 pg. 5). They identified that Government hospitals, community not federal, are the most likely to invest in CPOE systems and for-profits the least likely; nonprofits were in the middle. In a case study of five sites the authors found that administrators who invested in CPOE were not expecting a return on investment or financial benefits, but appeared to place more value on the quality and safety improvements that the system provided. The authors speculated that government hospital administrators may be more motivated by public interest in patient safety and therefore more willing to invest in these systems. They also noted that each hospital ownership form may have different sources of decision making authority and that “physicians are sufficiently powerful to prevent CPOE adoption at private hospitals, but not sufficiently powerful to delay adoption at public institutions” (Cutler, Feldman, & Horwitz, 2005 pg. 6). In an article by Cohn et al, the authors’ state, “Healthcare information technology (HIT) is one of the most expensive capital investments for any healthcare organization. HIT adoption is a complex process because adoption and implementation depend on buy-in from physicians, most of whom are not employed by the organization and whose thinking varies widely” (Cohn et al, 2009 pg 291). The fact that one of the most expensive investments is dependent on physician buy-in says a lot about the influence of the medical profession on health care. Physicians have the power to influence what happens or doesn’t happen at hospitals. Physicians perceive greater trust between themselves and hospital management when they have power to control hospital decision making and greater capacity to ensure that the resulting decisions benefit their interests as well as the hospitals (Mintzberg 1983; Pfeffer 1992; Alexander and Morlock 1994). In January 2003 physicians pulled the plug on Cedars Mt Sinai Hospital’s CPOE system stating that it slowed down the process of caring for patients. Cedars had established a 40 physician medical executive committee to assist with the design and implementation of their new system; however rank and file physicians said it didn’t represent their interests. According to Dr. David Clausen of First Consulting Group, 6 other hospitals pulled paperless systems when physicians resisted using the systems (amednews.com, Febuary 17, 2003). Hospitals and government initiatives cannot succeed without physician buy-in and addressing the issues around their ability to adopt CPOE into their practices. Physicians and technology: Physicians do not inherently resist new technology and they are not technophobes. They were early adopters of cell phones, home computers, PDAs and manufacturers have capitalized on their love of surgical devices, robotics and endoscopic instruments (Lowenhaupt, 2004). However, they have not embraced clinical information systems even though more than 60% of hospitals CEOs are planning to invest in a system, CPOE reduces medical errors up to 80% (Bates, 2005) and the Government offers billions of dollars in incentives to entice them to purchase systems. What are the concerns and issues physicians have with CPOE and how will hospitals and providers overcome physicians’ resistance? The literature suggests that the top four issues physicians’ have with CPOE systems are cost, perceived value versus the time commitment, loss of autonomy to practice medicine, and the issue that CPOE may actually facilitate more serious medical errors (Koppel et al., 2005 and Bia II , 2006). Cost: The current cost of a CPOE system is staggering. Independent physicians can plan to spend $25,000 for the first physician in a practice and $10,000 for each additional doctor, with annual maintenance and support estimated at $4,000 to $6,500 a year. Smaller practices lack the access to financial recourses in order to equip their offices (Jaspen, 2009) and 75% of the nation’s physicians work in a small office of 10 doctors or less (Lohr). Doctors are also extremely reluctant to invest in technology that they view is related to non-core functions and is not directly linked to billable revenue. Physician offices manage on low profit margins and with little evidence of a return on investment, they will resist any increase in their overhead (Lowenhaupt, 2004). In a report from the Center for Information Technology Leadership, 89% of the cost benefit from CPOE accrues to the holder of the financial risk for health care such as the insurance companies (Bria II, 2006), not the physician. Even with President Obama’s stimulus package offering $40,000 over a 5 year period to physicians, only about 25% of doctors’ offices have adopted CPOE in the U.S. (Brooks & Grotz, 2010). Physicians need assistance with the paperwork and time involved with selecting and implementing a CPOE system. If 75% of the physician offices are considered small practices, they don’t have additional resources with the time it takes to initiate this process. Financial incentives from the government are great, but if the provider doesn’t see them as attainable, they are of no benefit. Value: The physician’s perceived value of CPOE is that it is great for the hospital and hopefully provides patients with better care, but the time commitment from the physician to develop, implement and utilize this system leaves him with the larger share of the time cost, and time is money (Poggio, 2010). According to a study published in JAMA 2001, physicians on average took 25 seconds to 4 minutes longer to document and enter orders into a CPOE system versus handwritten progress notes and orders, but the proper use of order sets could result in a 37% decrease in ordering time. However, this did not translate into dollars for the physician. The hospital, patient and insurer are the beneficiaries because of improved efficiency and reduction in costly errors or repeated tests. If you look at where and why CPOE has resulted in value to the physician, Mayo, Kaisers and Cleveland Clinic, it is related to the fact that the physicians are part owners of the hospital and are paid a salary and bonus based on the performance of their practice. They understand that using less support staff to enter orders for them saves the hospital money and improves care, which translates into more patient referrals and more revenues for them. Still, the fact remains that the majority of physicians are independent practitioners and if CPOE adds an additional 4 minutes of time to the care they provide, seeing 30 patients adds 2 hours of effort. This can equal 8-10 patient visits, costing them $800-$1,000 dollars in billings (Poggio, 2010). As far as any cost savings in malpractice premiums related to a decrease in medical errors, physicians have yet to experience this benefit (Sidorov, 2006). If hospitals, patients and insurers reap the benefits of CPOE while physicians see it as costing them time and money, it will definitely not succeed. Consideration should be given to the fact that the beneficiaries should bear some of the burden of the cost. Physicians are not paid for their time on committees at hospitals and, under normal circumstances, most hospital committees are a professional obligation. However, committees for projects involving a significant amount of time, like CPOE design and implementation, translate into significant financial loss to the physician. Incentives or reimbursement should be given deliberation by organizations and regulatory bodies in order to improve physician involvement in such important initiatives and to ensure the future success of CPOE. Loss of autonomy: Many studies, “although mostly anecdotal, have presented that CPOE has become a source of professional conflict because it changed physicians’ workflow has undermined their autonomy” (Aarts and Koppel, 2009 pg. 2). Physicians have concerns over being forced to practice via clinical pathways or with clinical decision support systems (CDSS) that tell them what to do (Morgenstern, 2008). In the early 1990’s order sets were developed in order to streamline the physicians’ work by reducing the time it took to write standard admission or procedure orders. Physicians viewed this as “cookbook” medicine because they feared that interns and residents would not develop the essential critical thinking skills needed to care for their patients and would eventually not be able to function without them. These concerns have turned out to be unfounded (Bria II, 2006). Today CPOE offers embedded knowledge as a tool for providers to give the care they know is right, not to force them to do something with which they do not agree (Morganstern, 2008). Jamal, McKenzie and Clark identified that the greatest realized advantage of CPOE with CDSS to physicians is around medication orders and reminders for follow up tests or procedures. Physicians have come to appreciate drug dose calculations, especially with antibiotics in children with specific weight dose requirements and in patients with impaired liver and renal functions, leading to a significant reduction in the number of medications ordered outside of dosing ranges (Mahoney et al, 2007). It has become evident that a properly designed information system can become an important tool for preventing medical errors by “enforcing” clinical adherence to evidenced based guidelines (Bates et al., 1998). However, physicians do not like anyone “telling” them what to do for their patients. The over use of alert systems and extra steps to guide physicians through best practices to complete an order eventually leads to desensitization of the provider (Menke et al, 2005). They are more likely to adopt CPOE with CDSS when they are presented with options, rather than warnings and when they have input into the system design (Aarts & Koppel, 2009). Physician resistance to the concept of CPOE with CDSS is becoming less salient as technology and its integration with other systems is improving. As they gain IT sophistication, however, their dissatisfaction with inadequately designed interfaces and CPOE-required repetitive/clerical tasks has increased (Massaro and Ash et al., 2004). CPOE systems must be improved with faster response times and to be easier to use. According to reports in USA Today, approximately 34% of physicians are age 55 or older and age plays a role in acceptance and use of CPOE systems (Nir Menachemi et al and Lindenauer et al, 2006). Hospitals need to consider that older physicians often do not type because they have relied on dictation and voice recognition software currently does not format to the structure of CPOE systems (Brooks and Grotz, 2010). Possibly using support staff such as, nurse practitioners and physician assistants would reduce some of the physicians’ time, but hospitals will need to evaluate workflow processes around clinical hand-offs to ensure quality care and communication between providers (Poggio, 2010) and identify ways to pay for the additional staff. CPOE facilitating more serious medical errors: Physicians have raised concerns that even though CPOE reduces a significant number of medical errors, studies have shown that it may also facilitate more serious, unintended errors (Sidorov, 2006). A study published in the Journal of Pediatrics in 2005 looked at the effects of CPOE to reduce mortality rates. Their findings were just the opposite. They suggest that the implementation of CPOE altered the prior patterns of work, communication, and relationships among clinicians, therefore causing delays and reduced clinician interaction that may have contributed to higher mortality rates in children. In the past 3 years several studies have suggested that CPOE contributed to medication errors because of issues such as ignored false alarms, computer crashes and orders in the wrong medical records (Ash et al., 2004 and Kaushal et al., 2003). The Journal of the American Medical Association (JAMA) published a study in 2005 on The Role of Computerized Physician Order Entry Systems in Facilitating Medication Errors. The study revealed that a widely used leading CPOE system facilitated 22 types of medication error risks attributed to fragmentation and systems integration failure and human-machine interface flaws (Koppel et al., 2005). The studies mentioned above still promote the use of CPOE as an effective way to reduce medical errors. Their conclusions state that providers must be aware of these potential issues and take actions to address them immediately. “The belief that simply buying and installing health information technology will automatically lead to safer and better care is a myth. Hospitals and vendors must continue to work together over time to ensure the effectiveness and efficiency of CPOE” (Binder, 2010). Additionally, Leapfrog Group introduced a CPOE evaluation tool that simulates patients and medical orders, allowing the system to run a report that checks for medication and potentially fatal errors. Providers using this tool corrected the issues that were identified and showed significant correction of errors after running the tool a second time. Their report in June 2010 called up federal officials to build oversight requirements into current regulations and made recommendations that hospitals and CPOE technology companies should take to assure the quality and safety of these systems (Hagland).

Conclusion and the Future of CPOE: The adoption of interoperable electronic medical record (EMR) systems could produce efficiency and safety savings of $142 to $371 billion dollars (Hillestad et al., 2005). A CPOE , when designed and implemented correctly, is an important part of the EMR that will reduce medical errors and save billions of dollars by aiding in the management of chronic diseases. Achieving the benefits that CPOE can have on prevention and disease management requires substantial participation from physicians (Hillestad et al., 2005). Strengthening relationships with doctors by giving them power through input on the design and implementation of CPOE may improve their acceptance and use of the systems; identifying ways to assist small physician practices adopt and pay for CPOE systems; creating new incentives, like reimbursement of physicians for time spent on IT committees and training for CPOE utilization, and for their efforts that lead to improved patient quality and reduced costs; and continued efforts from hospitals and government regulatory bodies to address and find solutions for the issues of interoperability, clinical workflow and competing agendas in American health care, would help the U.S. move closer to the realization of true EMRs and personal health records (PHR) for patients and would help to ensure the success of CPOE. We stand on the verge of beginning a new era of interoperable health information technology that would significantly change the face of healthcare as we know it. Medical care can be provided in a quality, patient centered and cost effective way that allows physicians to focus on their patients and to develop preventative solutions and cures for life threatening diseases. For the past 30 years the U.S. has been too slow to correct the issues plaguing the implementation and utilization of health-care information technology in an integrated and cost effective manner. We have the ability to make CPOE a reality that improves the quality of care and saves money that can be used for other important initiatives. Hospitals, insurers, government and physicians need to resolve the small issues that are blocking the progress of a country that needs to set the example for healthcare around the world.

References
American Academy of Pediatrics (2003) Policy Statement: Prevention of Pediatric Errors in the Pediatric Inpatient Setting. www.aap.org (Last accessed April 19 2010.)
American Journal of Health-System Pharmacy, 64, 1969-1977. Retrieved from http://www.ajhp.org/cgi/content/abstract/64/18/1969
. Health Affairs, 25(4). doi: 10.1377/hlthaff.25.4.1079
Aarts, J., & Koppel, R. (2009). Soldiering on: Implementation of computerized physician order entry in U.S., European, and Australian Hospitals. Health Affairs.
Alexander, J. A., & Morlock, L. L. (1994). Power and politics in health services organizations. Albany, N.Y.: Delmar Publishers, Inc.
Ash et al., J. S. (2004). Computerized physician order entry in U.S. hospitals: Results of a 2002 survey. Journal of the American Medical Informatics Association, 11(2).
Bates et al., D. W. (1998). Effects of computerized physician order entry and a team intervention on prevention of serious medication errors. Journal of the American Medical Association, 280, 1311-1316.
Bates, D. W. (2005). Computerized physician order entry and medication errors: Finding a balance. Journal of Biomedical Informatics, 38(4), 260.
Bria II, W. F. (2006). Applied medical informatics for the chest physician. CHEST, 129, 1057-1060. doi: 10.1378/chest.129.4.1057
Brooks, R., & Grotz, C. (2010). Implementation of electronic medical records: How healthcare providers are managing the challenges of going digital. Journal of Business and Economics Research, 8(6), 73-84.
Cash, J. (2008). Technology can make or break the hospital-physician relationship. hfm (Healthcare Financial Management), 62(12), 104-109. Retrieved from CINAHL database.
Cohn, K., Berman, J., Chaiken, B., Green, D., Green, M., Morrison, D., et al. (2009). Engaging physicians to adopt healthcare information technology. Journal of Healthcare Management, 54(5), 291-300. Retrieved from CINAHL database.
Cutler, D. M., Feldman, N. E., & Horwitz, J. R. (2005). Physician order entry systems. Health Affairs, 24, 1654-1663. doi: 10.1377/hlthaff.24.6.1654
Hagland, M. (, August 1, 2010). Leapfrog Group releases new CPOE study. Healthcare Informatics. Retrieved from http://www.allbusiness.com/technology/software-services-applications-information/15062529-1.html
Hillestad, R., Bigelow, J., Bower, A., Girosi, F., Meili, R., Scoville, R., & Taylor, R. (2005). Can electronic medical record systems transform health care? Potential health benefits, savings, and costs.. Health Affairs, 24, 1103-1117. doi: 10.1377/hlthaff.24.5.1103
HIMSS Survey April 6, 2009; HIMSS; HIMSS Foundation; HIMSS Analytics.
Jamal, A., McKenzie, K., & Clark, M. (2009). The impact of health information technology on the quality of medical and health care: a systematic review. health Information Management Journal, 38.
Jaspen, B. (2009). Doctors in smaller practices need infrastructure, administrative help and access to latest technology data. Retrieved from www.chicagotribune.com/business
Kaushal, R., Shojania, K. G., & Bates, D. W. (2003). Effects of computerized physician order entry and clinical decission support systems on medication safety: a systematic review. Archieves of Internal Medicine, 163, 1409-1416.
Koppel et al., R. (2005). Role of computerized physician order entry systems in facilitating medication errors. Journal of the American Medical Association, 293, 1197-1203.
Koppel, R. (2005). Role of computerized physician order entry systems in facilitating medication errors. Journal of the American Medical Association, 293, 1197-1203.
Levey, N. N. (, June 8, 2009). Shot in the arm for digital health. Chicago Tribune.
Lindenauer et al., P. K. (2006). Physician chacteristics, attitudes and use on computer order entry. Society of Hospital Medicine, 221-230. doi: 10.1002/jhm.106
Lohr, S. (, March 11, 2009). Wal-Mart to digitalize health data. New York Times, p. B 1.
Lohr, S. (2009). Electronic health records: How to spend the money wisely. Retrieved from http://bits.blogs.nytimes.com/2009/02/10/electronic-health-records-how-to-spend-the-money-wisely/
Lowenhaupt, M. (2004). Removing barriers to technology. Health Publications. Retrieved from http://findarticles.com/p/articles/mi_m0843/is_2_30/ai_n609436/?tag=content;co11
Massaro, T. A. (2004). Introducing physician order entry. Journal of American Medical Informatics Association, 11(3).
Menke, J. A. (2005). Physician Response to CPOE Allergy Information Alert. American Medical Informatics Association. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1560761/
Mintzberg, H. (1983). Power in and around organizations. Englewood Cliffs NJ: prentice-Hall, Inc.
Morgenstern, D. (2008). A CPOE Primer for Nursing (PowerPoint Slides). Massachusetts: Massachusetts Technology Collaborative and New England Healthcare Institute.
Nir Menachemi, (2006). Managed care penetration and other factors affecting computerized physician order entry in the ambulatory setting. The American Journal of Managed Care, 738-744.
Pfeffer, J. (1992). Managing with power. Harvard Business Review.
Poggio, F. (2010). CPOE and the doc dilemma. Retrieved from http://www.healthcare-informatics.com/ME2/dirmod.asp?sid=&nm=&type=Publishing&mod=Publications%3A%3AArticle&mid=8F3A7027421841978F18BE895F87F791&tier=4&id=9D02B41A45A64124B6E773C5A1DE4346 amednews.com. (Febuary 17, 2003). http://www.ama-assn.org/amednews/2003/02/17/bil20217.htm www.ahrq.gov. (2004). http://www.ahrq.gov/data/hcup/hcupnet.htm