TY - JOUR
AU - Taylor,, Pam
AB - Purpose The development of a computerized prescriber order-entry (CPOE) system for chemotherapy in a multisite safety-net health system and the challenges to its successful implementation are described. Summary Before CPOE for chemotherapy was first implemented and embedded in the electronic medical record system of Harris Health System (HHS), pharmacy personnel relied on regimen-specific preprinted order sets. However, due to differences in practice styles and workflow logistics, the paper orders across the 3 facilities were mostly site specific, with varying clinical content. Many of these order sets had not been approved by the oncology subcommittee. In addition, disparities in clinical knowledge and lack of communication contributed to inconsistencies in order set development. Led by medical directors from medical oncology departments at the 3 facilities, pharmacy administrators, and information technology representatives, HHS committed resources to supporting the adoption and use of a CPOE system for chemotherapy. Five practical lessons of broad applicability have been learned: engagement of interprofessional stakeholders, optimization of workflow before CPOE implementation, requirement of verification tool for CPOE, consolidation of protocols, and commitment to ongoing training and support. Evaluation of the CPOE system demonstrated a systemwide reduction in medication errors by 75% (p < 0.05). Satisfaction with the CPOE system varied among sites and was unchanged institutionwide 6 months after the CPOE implementation. Conclusion The development and implementation of CPOE for chemotherapy at a multisite safety-net health system created opportunities to optimize patient care and reduce variations through interprofessional collaborations. Initial evaluation suggested that CPOE reduced the medication-order error rate and improved user satisfaction in 1 of 3 facilities. chemotherapy, CPOE, medication error, safety-net hospital, usability, workflow Safety-net health systems in the United States treat a disproportionate number of vulnerable patients, such as those who are uninsured or underinsured, regardless of their ability to pay.1 Almost 50 million people in the United States do not have medical insurance,2 and safety-net hospitals remain a major source of care for these patients. As the healthcare landscape in the United States continues to evolve, safety-net hospitals are at the frontline for the transformation of healthcare delivery to the country’s most vulnerable patients. In the meantime, computerized health information is increasingly incorporated into hospital businesses. Use of computerized prescriber order entry (CPOE) for medication ordering has been touted as a means to reduce cost and improve the quality, safety, and accessibility of health information.3 This is particularly true for safety-net hospitals, which compete in a high-cost healthcare market with limited financial resources while caring for uninsured patients. This article describes the strategies, challenges, and lessons learned in the development and implementation of systemwide CPOE for chemotherapy within 3 major facilities of a large public safety-net health system in Houston, Texas. This CPOE system was implemented to optimize the ordering, administration, dispensing, and procuring of chemotherapeutic and biological response modifiers across many disciplines (e.g., hematology, medical oncology, gynecological oncology, rheumatology, nephrology, neurology, dermatology) for both malignant and nonmalignant indications. The article also highlights practical issues that may serve as an important guide to other institutions as they conceive and develop their own CPOE system for chemotherapy. Background Harris Health System (HHS), formerly known as Harris County Hospital District, is a community-owned safety-net health system that cares for all residents in Harris County, Texas, the third most populous county in the United States.4 The system includes 23 community health centers; 5 school-based clinics; the multispecialty facility Smith Clinic (SC), which houses many outpatient services (including outpatient infusion services for chemotherapy); and 2 full-service hospitals. These hospitals are Lyndon B. Johnson General Hospital (LBJ), a 235-bed acute care hospital affiliated with the University of Texas Medical School in Houston (the cancer program of LBJ is contracted with the University of Texas MD Anderson Cancer Center), and Ben Taub General Hospital (BT), a 486-bed acute care hospital affiliated with Baylor College of Medicine. Together, SC, LBJ, and BT account for more than 5,000 cancer patient visits each year and administer more than 25,000 doses of chemotherapy agents for a variety of solid and hematologic malignancies. KEY POINTS Workflow processes should be optimized and streamlined before implementing computerized prescriber order entry (CPOE) for chemotherapy in a health system. CPOE requires visible and invisible investments by the health system, and the return on investment may vary. Updating of organizational policy and ongoing fine-tuning of operational processes are necessary before and after CPOE implementation. HHS also houses an investigational drug service (IDS) that coordinates over 30 cancer research protocols for adult patients treated on research protocols for uncommon or relapsed malignancies. Its information technology (IT) department, in association with pharmacy, nursing, and medical directors, is responsible for the implementation of electronic medical records (EMRs). Its fully integrated EMR system for patient care (EPIC Systems Corp., Verona, WI) went into operation in 2010. Problem Before CPOE for chemotherapy was first implemented and embedded in the EMR system of HHS in September 2016, pharmacy personnel relied on regimen-specific preprinted order sets. However, due to differences in practice styles and workflow logistics, the paper orders across SC, LBJ, and BT were mostly site specific, with varying clinical content. Many of these order sets had not been approved by the oncology subcommittee, a governing body under the pharmacy and therapeutics (P&T) committee within HHS. Each practice site had variable levels of staff engagement and prescriber preferences when developing paper chemotherapy order sets. Versions of noncurrent and current orders were in use before CPOE implementation. In addition, disparities in clinical knowledge and lack of communication contributed to inconsistencies in order set development. For instance, in 1 facility, before the implementation of CPOE for chemotherapy, the majority of chemotherapy orders were developed by contracted clinical pharmacists who were affiliated with an academic institution but not employed by HHS. Historically, limited interactions existed between the contracted and HHS-employed clinical pharmacists, the latter of whom are further categorized by their job functions into 3 levels: level 1, level 2, and level 3, with major roles in drug distribution services, clinical services, and formulary management services, respectively. Organizational hierarchy and differences in skills were the perceived barriers to the full utilization of HHS clinical pharmacists for order set development and review. This lack of oversight contributed to discrepancies among these order sets as well as an increased risk of medication errors. Analysis and resolution Development of integrated paper protocols for CPOE. In 2015, led by medical directors from medical oncology departments at LBJ, BT, and SC; pharmacy administrators; and IT representatives, HHS committed resources to supporting the adoption and use of a CPOE system for chemotherapy (Beacon, EPIC Systems Corp., Verona, WI). Since 2 chemotherapy ordering systems existed de facto within HHS—1 at LBJ and the other at BT and SC—it was imperative to delineate each step in the medication-use process, eliminate redundancy, and develop pragmatic expectations and solutions in standardizing paper-based chemotherapy order sets as templates. The goal was to implement a safe and comprehensive CPOE system that encompassed a variety of disciplines that prescribe chemotherapy agents in these 3 major facilities. From April 2015 through August 2016, hematology–oncology clinical pharmacist subject matter experts (SMEs) or level 2 clinical pharmacists at each of the 3 facilities coordinated with the CPOE project manager and met with an IT team. The IT team comprised 2 full-time-equivalent pharmacists dedicated to informatics, a project manager, and a variable number of IT employees engaged either directly or indirectly in CPOE development and implementation. The meetings were called to address questions on how to standardize the clinical content of chemotherapy order sets and to streamline processes in CPOE. The decision was made to develop integrated paper-based protocols (which could be either site specific or not) that would serve as initial templates for future electronic protocols. One or more clinical pharmacist SMEs led the efforts in standardizing and developing these paper protocols for CPOE, which could also serve as backup order sets in the event of the EMR system becoming unavailable. CPOE validation meetings and collaboration. Meeting attendees included providers across a broad range of disciplines (e.g., hematology, gynecological oncology, medical oncology, interventional radiology, dermatology, nephrology, neurology, rheumatology, urology) who prescribe chemotherapy agents (defined as cytotoxic antineoplastics or biological or targeted therapies traditionally reserved for malignant indications per institutional policy) as well as the IDS coordinator and nurse SMEs. Specific responsibilities of these validation meetings included the following: (1) creating a generic template that contained all of the essential elements of a chemotherapy treatment plan or therapy plan (designated for agents not considered as chemotherapy within the institution) through which more than 450 chemotherapy treatment plans (including biological or targeted therapies), 50 chemotherapy treatment plans for patients enrolled in clinical trials, and 20 therapy plans were subsequently developed and implemented, (2) providing a forum for discussing the pros and cons of adopting, modifying, or consolidating a specific version of the chemotherapy protocol, and (3) defining and optimizing clinical content. During validation meetings, differences in opinions among clinical pharmacist SMEs were relayed to physician champions who decided on the final clinical content of all electronic chemotherapy protocols, which were subsequently reviewed and approved by members of the oncology subcommittee and the P&T committee. Lastly, these validation meetings helped establish and standardize the ordering of medications for the following supportive care areas: prophylaxis of acute chemotherapy-induced nausea and vomiting (CINV) and treatment of delayed and breakthrough CINV, use of myeloid growth factors, and use of take-home medications. General rules for hydration, sequence of chemotherapy administration, and dose rounding were also standardized and validated. Identifying key elements for harmonization. Due to the inherent problems with regimen-specific and site-specific preprinted order sets (i.e., paper orders) used extensively within HHS, the CPOE task force focused on the harmonization of the following key elements, which served as the basis for the development of electronic protocols and safe implementation of CPOE for chemotherapy: (1) protocol nomenclature and description, (2) supporting references and diagnosis, (3) physician orders, including chemotherapy drug, dose, route, and schedule, (4) nursing instructions, (5) laboratory orders, (6) supportive therapy (including hydration and premedications used for acute emesis prophylaxis and others), (7) emergency medications for anaphylaxis and hypersensitivity, and (8) take-home medications. Each paper protocol for CPOE, revised or consolidated from the preexisting paper orders, was then harmonized and validated among clinical pharmacist SMEs and nursing SMEs. Key lessons summarized from these validations were generated into training points for physicians using the CPOE. Each validated paper protocol was then built into electronic protocols for chemotherapy by IT pharmacists, who validated these protocols with the physician SMEs. Whenever an electronic protocol for chemotherapy in the repository (CPOE system) is selected, modified, or acted on by an attending physician, a physician trainee, a nurse practitioner, or a clinical pharmacist, it is subsequently termed a treatment plan, which requires the attending physician’s signature or cosignature before its release for verification and dispensing. While it is advantageous from the operational perspective to have integrated, standardized electronic chemotherapy protocols used at all 3 sites, pragmatic consideration must be given in developing site-specific protocols (which share similar clinical content but cannot be fully integrated due to differences in drug, dose, route, schedule, or monitoring) to accommodate patient convenience, cost containment, or physician preferences. Program assessment and modification To evaluate the impact of CPOE implementation on safety and the level of user satisfaction (based on a 5-point Likert scale) among end users (physicians, nurses, pharmacists), data were collected for 60 chemotherapy orders before the September 2016 CPOE implementation. A 6-month postimplementation review was conducted by clinical pharmacist SMEs in March 2017 for 40 orders. The review was approved by the quality and accreditation program within HHS. A systemwide reduction in chemotherapy-related medication-ordering errors by approximately 75% (from 60 errors among 60 orders to 10 errors among 40 orders, p < 0.05) was observed after the implementation of CPOE for chemotherapy (Table 1). However, despite the improved satisfaction among users in 1 facility, the other 2 facilities expressed moderate dissatisfaction (based on a survey) toward the CPOE system, driven by difficulties in adjusting to the new workflow with a limited workforce. However, the survey revealed no difference in terms of satisfaction scores institutionwide before and after CPOE implementation. Table 1 Medication Ordering Errors Before and After Implementation of CPOE for Chemotherapya Information Missing From Order No. (%) Errorsb Before Implementation (n = 60 orders) After Implementation (n = 40 orders) Dose 7 (12) 0 Schedule 20 (33) 2 (5) Required premedication 13 (22) 3 (8) Route 0 0 Diluent volume or drug dilution concentration 10 (17) 5 (13) Required laboratory test orders or parameters for tests 9 (15) 0 Infusion or i.v. push rate 1 (2) 0 Information Missing From Order No. (%) Errorsb Before Implementation (n = 60 orders) After Implementation (n = 40 orders) Dose 7 (12) 0 Schedule 20 (33) 2 (5) Required premedication 13 (22) 3 (8) Route 0 0 Diluent volume or drug dilution concentration 10 (17) 5 (13) Required laboratory test orders or parameters for tests 9 (15) 0 Infusion or i.v. push rate 1 (2) 0 a CPOE = computerized prescriber order entry. b p ≤ 0.05 for all comparisons based on Wilcoxon matched-pairs signed-ranked test. View Large Table 1 Medication Ordering Errors Before and After Implementation of CPOE for Chemotherapya Information Missing From Order No. (%) Errorsb Before Implementation (n = 60 orders) After Implementation (n = 40 orders) Dose 7 (12) 0 Schedule 20 (33) 2 (5) Required premedication 13 (22) 3 (8) Route 0 0 Diluent volume or drug dilution concentration 10 (17) 5 (13) Required laboratory test orders or parameters for tests 9 (15) 0 Infusion or i.v. push rate 1 (2) 0 Information Missing From Order No. (%) Errorsb Before Implementation (n = 60 orders) After Implementation (n = 40 orders) Dose 7 (12) 0 Schedule 20 (33) 2 (5) Required premedication 13 (22) 3 (8) Route 0 0 Diluent volume or drug dilution concentration 10 (17) 5 (13) Required laboratory test orders or parameters for tests 9 (15) 0 Infusion or i.v. push rate 1 (2) 0 a CPOE = computerized prescriber order entry. b p ≤ 0.05 for all comparisons based on Wilcoxon matched-pairs signed-ranked test. View Large Informal interprofessional feedback also suggested that most end users at all 3 sites recognized the ability of the CPOE system to improve clarity compared with paper orders. Physicians benefited from the ability to enter and sign orders remotely through portals, especially when they staffed multiple teaching facilities. The orders can be viewed by all end users simultaneously and at any time. In addition, CPOE improved users’ ability to recognize and resolve issues of incomplete orders, evidenced by the significant reduction in medication-order error rates. CPOE has become the primary means for ordering chemotherapy (except methotrexate for ectopic pregnancy termination and a small number of rarely used protocols that could not be implemented due to access or logistic issues for prescribers) at HHS, including clinical trials, which represent approximately 4% of all chemotherapy protocols implemented. Since the majority of chemotherapy agents at HHS are administered in the outpatient setting, protocols that can be used for both the outpatient and inpatient setting are defaulted to outpatient use for convenience. Lessons learned Several practical lessons of broad applicability have been learned through the development and implementation of CPOE for chemotherapy. Optimization of workflow before CPOE implementation. The vendor of CPOE for chemotherapy allows HHS to customize functionality of the system based on institution-specific practice and workflow. At HHS, fellows (physician trainees in specialties) and clinical pharmacist SMEs are permitted to create chemotherapy treatment plans. Orders must be reviewed by clinical pharmacists on the floor or in clinics before dispensing when these pharmacists are available. However, attending physicians must review and sign (or cosign if a fellow has previously signed the order) each chemotherapy treatment plan before it is released by nursing for order verification and drug distribution by pharmacy staff. Figure 1 outlines the workflow process of inpatient-based chemotherapy administration in HHS. Figure 1 View largeDownload slide Schematic representation of workflow of computerized prescriber order entry for inpatient chemotherapy. MD = physician, I/O = input and output, MAR = medication administration record, CP1 = level 1 clinical pharmacist. Figure 1 View largeDownload slide Schematic representation of workflow of computerized prescriber order entry for inpatient chemotherapy. MD = physician, I/O = input and output, MAR = medication administration record, CP1 = level 1 clinical pharmacist. As mentioned above, the wide range of practice differences and staffing levels among each site create barriers in improving the efficiency and safety of processes involved in the ordering, preparation, administration, and documentation of chemotherapy treatments. The workflow modeled by the CPOE system did not seem to match well with the real-world scenarios encountered at HHS. This discrepancy was evidenced by differences in the results of the satisfaction survey conducted after CPOE implementation, in which BT and SC consistently scored lower than LBJ, possibly reflecting the logistic differences in these facilities. LBJ has an affiliated outpatient infusion center and an outpatient pharmacy on campus and generally has more personnel (3 clinical pharmacists in the clinic and 1 clinical pharmacist in the inpatient unit) and physician trainees (5–6 outpatient, 1–2 inpatient), compared with BT and SC which have no onsite outpatient infusion center, no outpatient pharmacy, and a relatively smaller workforce (2–4 physician trainees, 1–2 clinical pharmacists for SC and BT). Early after postimplementation, incidents of expired orders (i.e., orders that, by default of the HHS rule, have dropped off from the active medication administration record, usually 12 hours after the order is released) in the CPOE system were observed infrequently, partly due to logistics (e.g., late admission of patients for chemotherapy, late placement of central venous access devices) and the early release of chemotherapy orders in anticipation of the time required for chemotherapy drug preparation by the evening and night-shift pharmacy staff. One of the solutions that addressed this usability issue was setting up an agreement between pharmacy and nursing on the cutoff time of order release, so that administration of the chemotherapy agent could start before the CPOE order expired. The importance of institutionwide educational efforts in resolving user workflow problems cannot be emphasized enough. The integration of CPOE into the proper workflow continues to pose challenges after its implementation. Although the American Society of Clinical Oncology EHR Workgroup emphasizes the need for standardization of workflows and chemotherapy regimens before EMR implementation, barriers remain in multisite safety-net health systems such as HHS.5 Consistent with findings reported in some studies, challenges in the successful implementation of CPOE for chemotherapy in HHS included usability issues such as the suitability of alerts and warnings, discrepancies in documentation, and the lack of appropriate safeguards.6,7 These usability issues contributed to some of the workflow problems witnessed in HHS after CPOE implementation ( appendix). Specific recommendations from the CPOE vendor and interprofessional engagement are now underway to address these issues, along with continuous quality improvement and modification of the functionality of CPOE for end users. Future study will help define the effects of the CPOE system on the overall safety and quality of cancer care in HHS. Engagement of all interprofessional stakeholders. Our experience corroborated findings from similar studies that implementation of CPOE for chemotherapy requires extensive resources, clinical experience, and collaboration across multiple service lines.8,9 Identification of key stakeholders and the creation of a partnership with a multidisciplinary group is critical to the successful implementation of CPOE.8–10 Development of more than 450 advanced-function chemotherapy order sets to support operations in a multisite system proved challenging. Moreover, since physician trainees are the primary prescribers (and require cosigning from attending physicians) for CPOE (other end users such as nurse practitioners or clinical pharmacists may generate chemotherapy treatment plans but cannot sign them), and they come from different disciplines (traditional service lines that routinely prescribe chemotherapy agents versus nontraditional service lines [e.g., rheumatology]), implementation of CPOE for chemotherapy requires continuous engagement and communication with the nursing and medical staff and more time, commitment, and persistence than anticipated. Similarly, study protocols, once paper orders drafted by the IDS coordinator, were changed to the CPOE format with detailed information such as physician instructions, premedications, laboratory orders, and take-home medications, which required review or validation by study coordinators, clinical pharmacist SMEs, and principal investigators (attending physicians). While the CPOE process is more robust, significant time must be invested compared to that for paper study protocols. The implementation of CPOE for chemotherapy also requires prescribers to implement practice changes and workflow adjustments to support best practice. Prescribers are required to learn and be familiar with all the nuances of a new CPOE system, an investment of human resources that is often invisible in the budget of an institution. Due to significant financial costs, not all institutions can implement an institutionwide CPOE system for chemotherapy.11 This problem is particularly obvious at HHS, which is affiliated with 2 academic institutions with distinct differences in infrastructure and clinical practices. Further, continuity of care is lacking, in which the consultation services (e.g., hematology, medical oncology) must constantly establish effective communication with providers of the primary team who are responsible for the admission, inhouse management, and discharge of patients receiving chemotherapy, especially when the discharged medications are specific to cancer care. Due to the dynamic workflow of physician trainees, problems resolved in the past often resurface among new trainees, necessitating dedicated resources for ongoing education and training. Still undecided at HHS is whether to grant CPOE access to nononcology providers who rarely or infrequently order chemotherapy. In fact, there is a paucity of supporting policy and processes that address these CPOE issues in HHS. Management of health information, allocation of resources, and security of access continue to demand attention and collaborative efforts for guidance and decision-making. Need for a CPOE verification tool. HHS follows national guidelines for safe administration of chemotherapy and requires signed chemotherapy orders to be verified by 2 independent checks by pharmacists, followed by 2 nurse checks before administration.12,13 The outpatient clinics are also staffed with clinical pharmacists who review treatment plans and assist with or consult on clinical issues that affect the oncology drug–use process. Currently, CPOE at HHS allows pharmacists and nurses to document the verification process by using a department-specific electronic “chemotherapy checklist,” but there is no documentation of the name of the person who provided verification or the date and time that verification occurred. Electronic review of the protocol can be electronically documented only after it has been signed and become an actionable treatment plan, necessitating the development of a verification tool with clarity of authorship and time of verification. A work group to navigate this issue is likely needed. Consolidation of similar or identical chemotherapy protocols for different diagnoses. Each chemotherapy protocol for CPOE at HHS is defined by its diagnosis and emetogenic potential, and clinical pharmacist SMEs expressed the desire that each protocol conform to the National Comprehensive Cancer Network (NCCN) chemotherapy templates during CPOE preimplementation.14 However, individual SMEs recognized the disproportionate time invested in developing these protocols to adhere to these recommendations. On the other hand, safety must be provided to physicians when ordering a particular regimen that can be used for different indications. End users learned that many protocols can be consolidated to facilitate convenience and feasibility of the workflow. Every institution should establish its own rule for consolidating similar or identical chemotherapy protocols for different diagnoses in its CPOE system. Commitment to ongoing training and support. Training was required for all team members (pharmacists, nurses, attending physicians, physician trainees) involved in generating or executing chemotherapy orders before CPOE implementation. The IT department at HHS created interactive computer-based training modules for users to gain understanding through predetermined, role-based scenarios. Before CPOE implementation, clinical pharmacist SMEs and a team of physicians collaborated on the incremental conversion of paper-based chemotherapy orders to electronic protocols for patients who were on active chemotherapy treatments. IT staff members were onsite during the conversion and implementation phases. All team members were encouraged to report errors and problems during the CPOE implementation phase. Frequent meetings were held to identify challenges that arose during implementation. Further improvements and changes continue to evolve as new problems arise in this large safety-net health system. As HHS is affiliated with 2 academic institutions, ongoing training for multiple attending physicians across different service lines is needed, even if they infrequently order the chemotherapy agent in the CPOE system. It is the goal of HHS to have 99% of chemotherapy agents ordered in CPOE. Challenges in training a large workforce with a high turnover rate remain. Discussion An increased number of studies over the past few years have demonstrated that CPOE can (1) decrease costs, (2) reduce medication errors, (3) reduce length of stay, (4) standardize practices, (5) integrate clinical decision support, (6) improve interprofessional communication, and (7) facilitate data mining for research, quality improvement, and health information management.15–24 However, some studies also cited an increased risk of medication errors due to implementation and user training.11,25–27 Our experience suggests that when compared with paper-based chemotherapy orders, CPOE can reduce but does not eliminate errors associated with order information such as schedule, final drug dilution concentration, volume of infusion, and infusion rate. However, some of these problems, such as final drug dilution concentration, may arise from an incorrect default to a certain work rule. They can be rectified by modifying the coding process of the CPOE. In fact, workflow problems showed that alert and error-checking functions of CPOE remain a work in progress. In general, these errors were caused by workflow and logistic problems with the CPOE. Implementation of CPOE at HHS demonstrated the feasibility of establishing an integrated CPOE system for chemotherapy across a large multisite system. Collaboration with multiple users, such as nursing and physicians, was critical for its success. However, challenges remain with integrating current workflow processes into the new CPOE system. Efforts from stakeholders are ongoing to realign and improve workflow processes. In many instances, clinical and informatics pharmacists have leading roles in these efforts. A few caveats should be applied to the application of our experience to other practice sites. First, it involved a multisite safety-net health system that serves mostly indigent populations. It is affiliated with 2 academic teaching institutions, with large numbers of patients and prescribers. Therefore, results of this system cannot be extrapolated to the general population. Second, the CPOE system was still in a state of major improvement when the 6-month postimplementation survey was conducted. However, few changes were noted during this time. Third, the categorization of errors reflected the views of the pharmacists. Error rates due to administration may vary considerably and have not been extended to encompass this process. Nevertheless, administration issues in CPOE were mostly due to workflow problems after implementation. Conclusion The development and implementation of CPOE for chemotherapy at a multisite safety-net health system created opportunities to optimize patient care and reduce variations through interprofessional collaborations. Initial evaluation suggested that CPOE reduced the medication-order error rate and improved user satisfaction in 1 of 3 facilities. Acknowledgments The medical staff, IT staff, nurses, and pharmacists involved in the review and implementation of CPOE for chemotherapy at HHS are acknowledged. Appendix Examples of workflow-related issues that affect usability of the CPOE for chemotherapy after its implementation Ordering of oral chemotherapeutic agent in outpatient (e-prescribe) Workflow problem: Physician trainees are not authorized to prescribe oral chemotherapeutic agent (based on institutional definition and policy) but are authorized to prescribe supportive care medications How is it resolved? Outpatient oral chemotherapeutic agent order defaulted to attending physician cosigning Release of take-home medication orders while patient is admitted for inpatient chemotherapy Workflow problem: CPOE for chemotherapy includes a section of take-home medications, which are signed by specialty physician trainees. But primary team physicians and hospitalists also reconcile and prescribe take-home medications, causing potential duplications or missed take-home medications How is it resolved? Requires ongoing communication and education between different team members; specialty teams (e.g., hematology, medical oncology) continue to sign and release take-home medication orders Intrathecal chemotherapy embedded in inpatient chemotherapy treatment plan Workflow problem: Interventional radiology (IR) personnel must be notified to release intrathecal chemotherapy order. Appointment with IR, however, is subject to change, causing infrequent cancelation of intrathecal chemotherapy administration or resulting in expired order How is it resolved? Requires ongoing communication and education between different team members. IT team created a supportive care order group for intrathecal chemotherapy orders that can be separated from systemic chemotherapy treatment plans. Floor stock in automated dispensing machine versus pharmaceutical product dispensed by central pharmacy Workflow problem: Pharmacy inventory is subject to change which necessitates CPOE protocols be updated to reflect correct dispensed location in pharmacy order-entry system How is it resolved? Requires ongoing IT collaboration with pharmacy operations Continuous fluorouracil infusion through elastomeric infusion device in the ambulatory setting Workflow problem: CPOE system defaulted to “once” for frequency of continuous infusion but clinical decision support triggered false alert for “high dose” How is it resolved? Requires rule change of the clinical decision support (frequency changed to “continuous” as opposed to “once”) Disclosures The authors have declared no potential conflicts of interest. Previous affiliations At the time of writing, Dr. Chung was affiliated with Lyndon B. Johnson General Hospital, Houston, TX; Dr. Patel with Smith Clinic, Houston, TX; and Drs. Fu, Reilly, and George with Ben Taub General Hospital, Houston, TX. References 1 Elmendorf DW . CBO’s analysis of the major health care legislation enacted in March 2010: statement before the Subcommittee on Health, Committee on Energy and Commerce, U.S. House of Representatives . www.cbo.gov/sites/default/files/03-30-healthcarelegislation.pdf (accessed 2017 Apr 7). 2 U.S. Census Bureau . Income, poverty, and health insurance coverage in the United States: 2010 . www.census.gov/prod/2011pubs/p60-239.pdf (accessed 2017 Apr 7). 3 Blumenthal D Tavenner M . The “meaningful use” regulation for electronic health records . N Engl J Med . 2010 ; 363 : 501 – 4 . Google Scholar Crossref Search ADS PubMed 4 Harris County Budget Management . Population study . www.harriscountytx.gov/CmpDocuments/74/Budget/6-Vol.%203-FY%202013-14%20Population%20Study.pdf (accessed 2017 Apr 7). 5 Hoffman JM Baker DK Howard SC et al. Safe and successful implementation of CPOE for chemotherapy at a children’s cancer center . J Natl Compr Canc Netw . 2011 ; 9 ( suppl 3 ): S36 – 50 . Google Scholar Crossref Search ADS PubMed 6 Bellamy L Purcell T . One institution’s experience with implementation of EPIC/Beacon: lessons learned . Oncology . 2014 ; 28 : 105 – 6 , 108 , C3 . Google Scholar PubMed 7 Cooke L Smith-Idell C Dean G et al. Research to practice: a practical program to enhance the use of evidence-based practice at the unit level . Oncol Nurs Forum . 2004 ; 31 : 825 – 32 . Google Scholar Crossref Search ADS PubMed 8 Elsaid KA Garguilo S Collins CM . Chemotherapy e-prescribing: opportunities and challenges . J Integr Pharm Res Pract . 2015 ; 4 : 39 – 48 . Google Scholar Crossref Search ADS 9 Shulman LN Miller RS Ambinder EP et al. Principles of safe practice using an oncology EHR system for chemotherapy ordering, preparation, and administration, part 2 of 2 . J Oncol Pract . 2008 ; 4 : 254 – 7 . Google Scholar Crossref Search ADS PubMed 10 Kruse CS Goetz K . Summary and frequency of barriers to adoption of CPOE in the U.S . J Med Syst . 2015 ; 39 : 15 . Google Scholar Crossref Search ADS PubMed 11 Jeon J Taneva S Kukreti V et al. Toward successful migration to computerized physician order entry for chemotherapy . Curr Oncol . 2014 ; 21 : e221 – 8 . Google Scholar Crossref Search ADS PubMed 12 Goldspiel B Hoffman JM Griffith NL et al. ASHP guidelines on preventing medication errors with chemotherapy and biotherapy . Am J Health-Syst Pharm . 2015 ; 72 : e6 – 35 . Google Scholar Crossref Search ADS PubMed 13 Neuss MN Polovich M McNiff K et al. 2013 updated American Society of Clinical Oncology/Oncology Nursing Society chemotherapy administration safety standards including standards for the safe administration and management of oral chemotherapy . J Oncol Pract . 2013 ; 9 : 5s – 13s . Google Scholar Crossref Search ADS PubMed 14 National Comprehensive Cancer Network . Clinical practice guidelines in oncology: antiemesis. V.2.2017 ( March 28 , 2017 ). www.nccn.org (accessed 2017 Apr 7). 15 Shulman LN Miller RS Ambinder EP et al. Principles of safe practice using an oncology EHR system for chemotherapy ordering, preparation, and administration, part 1 of 2 . J Oncol Pract . 2008 ; 4 : 203 – 6 . Google Scholar Crossref Search ADS PubMed 16 Peshek SC Cubera K Gleespen L . The use of interactive computerized order sets to improve outcomes . Qual Manag Health Care . 2010 ; 19 : 239 – 47 . Google Scholar Crossref Search ADS PubMed 17 Adelson KB Oiu YC Evangelista M et al. Implementation of electronic chemotherapy ordering: an opportunity to improve evidence-based oncology care . J Oncol Pract . 2014 ; 10 : e113 – 9 . Google Scholar Crossref Search ADS PubMed 18 Brockstein B Hensing T Carro GW et al. Effect of an electronic health record on the culture of an outpatient medical oncology practice in a four-hospital integrated health care system: 5-year experience . J Oncol Pract . 2011 ; 7 : e20 – 4 . Google Scholar Crossref Search ADS PubMed 19 Small M Barrett A Price GM . The impact of computerized prescribing on error rate in a department of oncology/hematology . J Oncol Pharm Pract . 2008 ; 14 : 181 – 7 . Google Scholar Crossref Search ADS PubMed 20 Fratino LM Daniel DA Cohen KJ et al. Evaluation of quality improvement initiative in pediatric oncology implementation of aggressive hydration protocol . J Nurs Care Qual . 2009 ; 24 : 153 – 9 . Google Scholar Crossref Search ADS PubMed 21 Huertas Fernández M Baena-Canada JM Martinez Bautista M et al. Impact of computerised chemotherapy prescriptions on the prevention of medication errors . Clin Transl Oncol . 2006 ; 8 : 821 – 5 . Google Scholar Crossref Search ADS PubMed 22 Hammond P Modgil S Wyatt JC . Safety and computer-aided design of chemotherapy plans . Top Health Inf Manage . 2000 ; 20 : 55 – 66 . Google Scholar PubMed 23 Voeffray M Pannatier A Stupp R et al. Effect of computerisation on the quality and safety of chemotherapy prescription . Qual Saf Health Care . 2006 ; 15 : 418 – 21 . Google Scholar Crossref Search ADS PubMed 24 Harshberger CA Harper AJ Carro GW et al. Outcomes of computerized physician order entry in an electronic health record after implementation in an outpatient oncology setting . J Oncol Pract . 2011 ; 7 : 233 – 7 . Google Scholar Crossref Search ADS PubMed 25 Wears RL Berg M . Computer technology and clinical work: still waiting for Godot . JAMA . 2005 ; 293 : 1261 – 3 . Google Scholar Crossref Search ADS PubMed 26 Koppel R Metlay JP Cohen A et al. Role of computerized physician order entry systems in facilitating medication errors . JAMA . 2005 ; 293 : 1197 – 203 . Google Scholar Crossref Search ADS PubMed 27 Aita M Belvedere O De Carlo E et al. Chemotherapy prescribing errors: an observational study on the role of information technology and computerized physician order entry systems . BMC Health Serv Res . 2013 ; 13 : 522 . Google Scholar Crossref Search ADS PubMed Copyright © 2018 by the American Society of Health-System Pharmacists, Inc. All rights reserved.
TI - Implementation of an integrated computerized prescriber order-entry system for chemotherapy in a multisite safety-net health system
JO - American Journal of Health-System Pharmacy
DO - 10.2146/ajhp170251
DA - 2018-03-15
UR - https://www.deepdyve.com/lp/oxford-university-press/implementation-of-an-integrated-computerized-prescriber-order-entry-5Zwh68Pyo7
SP - 398
VL - 75
IS - 6
DP - DeepDyve
ER -