Accepted on 05 Apr 2022
Submitted on 13 Jan 2022
With the increasing use of infusion pumps for medication administration in health care services, smart pumps have been accepted as a standard of care for reducing infusion-related medication errors [1]. Features of smart pumps include medication libraries, profiles of care areas, built-in soft and hard limits for administration rates, doses and concentrations, and the option to integrate with Electronic Medical Record (EMR) systems [2].
The use of advanced medication technologies is an increasingly important part of health-system pharmacy practice [1]. Interoperability of systems refers to the automatic pairing of an EMR infusion order with a smart pump entry and subsequent upload of infusion-related information from the smart pump to the EMR [3]. Bi-directional communication between smart pumps and EMR can lead to improved patient safety [1]. These features play an important role in facilitating appropriate medication selection and reducing user errors in the medication administration process [2]. Administration is the final step of the medication management pathway and is therefore the last opportunity to check for any discrepancies between what is prescribed and what is intended for administration before the patient receives the medication [4]. Automation of the ‘administration of medicine’ step in the medication management pathway creates a closed-loop medication management pathway which reduces human errors and optimises patient safety [4, 5].
Well-documented medication safety initiatives to improve quality of care of patients in health settings that are cited in the literature involve the introduction of system-based interventions and elimination of human-based processes [5]. These include integration of various interfaces together, introduction of forcing functions in electronic medication management systems, as well as maximising opportunities for automation and computerisation of any relevant processes in the medication pathway [6]. These interventions have demonstrated reduction of medication errors across the medication management pathway which have the potential to drive excellence in healthcare [7, 8].
Auto-programming smart pumps via EMR integration is a medication safety initiative that offers an opportunity to introduce a closed-loop medication management system in hospital settings and improve patient safety [5]. This is achieved via a bi-directional wireless connection through which information from a Computerised Provider/Physician Order Entry (CPOE) order is transmitted from the EMR and pre-populates on the infusion pump device [9]. Auto-programming reduces keystrokes and opportunities for manual data input error [6]. Additionally, smart pump integration provides the ability for infusion-related actions performed on the pump to be automatically documented on the EMR. Auto-programming and auto-documentation enhances patient safety as a result of streamlined clinician workflows [6, 9].
The uptake of digital technology to integrate processes and functionalities to support automated documentation, and delivery of near real-time data between medical devices is becoming the new standard of care for infusion therapy that provides quality of care [1, 2]. To date, there is limited published data on interoperability of smart pumps with EMR in Australia [10]. This article outlines an Australian experience in mapping the steps required to integrate the BD™ (United States (US)) Alaris with Guardrails dose error reduction software (DERS) smart pump medication/fluid library with Cerner© (US) EMR CPOE systems across a 160-bed specialist health service.
To ensure successful alignment and interoperability of the systems, a hospital-wide, multidisciplinary, systematic approach is required to standardise pump medication/fluid library, EMR infusion orders and clinical workflows [6, 11]. This methodology and implementation study used the Triple C method of consultation, collaboration and consolidation to develop and implement this project. The principles of this model allow for sustainability of the project implementation at all stages, through process mapping, defining key stakeholders’ roles and responsibilities and standardising workflows within an organisation [12].
The project started with a kick-off meeting to define the scope of the project, timeline, key milestones and key stakeholders. The main goal was to ensure the smart pump and the EMR systems were capable of interoperability from a technical and clinical workflow standpoint. In line with recommendations from the Institute for Safe Medication Practices (ISMP) Guidelines for Optimising Safe Implementation and Use of Smart Infusion Pumps, an expression of interest was sent out to form our team [3]. This phase included 2 steps as follows:
Our team included key stakeholders from external Becton, Dickinson and Company (BD™) smart pump vendors, Cerner© EMR vendors, pharmacists, subject matter experts in digital health integration, clinical nursing and nursing education, medical personnel, information technology (IT) experts, biomedical engineering and risk and project management [6, 13, 14]. All members of the group had extensive experience in their professional fields and were leaders in their practice area.
Integrated workgroups were formed following the identification of key stakeholders. A working group of staff were identified and started the initiative. Their roles and responsibilities are defined in Table 1. The following steps were undertaken by the working group:
Table 1
Smart Pump and EMR Integration Implementation Project Team Roles and Primary responsibilities.
| TEAM | ROLES | PRIMARY RESPONSIBILITIES |
|---|---|---|
| Executive sponsors | Hospital
|
|
| Project management team | Hospital
|
|
| Clinical management team | Hospital
|
|
| Pharmacy management team | Hospital
|
|
| Medical | Hospital
|
|
| Information systems | Hospital
|
|
| Biomedical engineers | Hospital
|
|
SME*: Subject matter expert (clinical nurse specialists or medical specialist).
CNE**: Clinical nurse educator.
* EMR: Electronic medical records.
IT: Information technology.
The collaboration process of this study required the team members to complete each of their allocated tasks to ensure successful alignment of medication orders in the EMR and the smart pump. This is to ensure readiness of pump-EMR integration as outlined in similar studies [13].
To facilitate alignment, the EMR pharmacist and the medication safety and lead smart pump pharmacist, worked simultaneously to achieve this goal. Reviewing and alignment of all medication data sets were required for initial testing. An exact match was required of the medication names, standardised concentrations and units, dosing units and rates, diluent volume and dosing limits [3, 13].
Medicines and fluids included for EMR-smart pump interoperability were classified into three infusion categories: intermittent infusions, continuous infusions and fluids for interoperability validation testing. A total of 194 unique medicines and 23 fluids, corresponding to 484 medicines and fluid concentration entries, were required to be built in the hospital dataset and integrated with the EMR system. In this dataset, the same medicine has multiple concentrations built, accounting for the various doses and indications that it can be used for.
Dataset alignment started with simpler infusions, such as medications built in the smart pump dataset as intermittent infusions. This was followed by fluids, then continuous infusions. This approach was used to assist in understanding the integration process for simpler infusions, and to apply principles of integration on the more complicated, titratable medicines and fluids.
A test server profile for the smart pump dataset was created by IT services from the systems vendors and hospital. This was built to assess interoperability. Each medication and fluid in the dataset was built and assigned a corresponding catalog CD or Alias (for multi-additive infusions) by the EMR medication management team and entered by the smart pump pharmacist. A catalog CD or Alias number is an EMR identifier used to associate an EMR CPOE order with a smart pump entry. An Alias is an organisation defined number that only exists to create a match with a smart pump entry. Conversely, a catalog CD is provided by the EMR system.
Any changes were made prior to validation testing in a systems test server to minimise disruption to daily clinical practice for the rest of the organisation.
During the process of alignment between EMR formulary and smart pump dataset some of the challenges identified were:
To overcome some of the above-mentioned challenges and after consultation with relevant stakeholders, changes to medication build in the EMR and/or smart pumps medication/fluid library, as well as clinician’s workflows were made. The following changes made included:
Once alignment had been established and instances where failed interoperability identified, an updated workflow was mapped out and documented for end users by the working group. This was completed according to the ISMP Guidelines for Optimising Safe Implementation and Use of Smart Infusion Pumps [3, 13, 14].
The consolidation phase required extensive validation testing to verify that each catalog CD or Alias had been correctly assigned to the appropriate medication in the smart pump dataset. Additionally, it required designing clinicians’ workflows and setting up governance of the new process.
Validation testing of integration of the dataset was conducted as a live event. The lead smart pump pharmacist conducted testing and troubleshooting prior to the live event that included project managers, pharmacy, nursing, IT consultants and external pump and EMR vendors. This involved validating that each EMR infusion order had successfully auto-programmed to a corresponding medication on the smart pump [15]. Furthermore, this stage included mapping clinician workflows to ensure sustainability and successful integration into every day practice.
A total of 217 unique medications and fluids that were built in the smart pump dataset were tested. Of these, a total of 87% (188/217) passed full record testing. The remaining 13% (29/217) were excluded from interoperability. Refer to Tables 2, 3 and 4 for a breakdown of medications and fluids entries. The 13% of medication and fluids that were excluded were due to the medication being prescribed on paper (i.e. medicines with desensitisation regimen), the medication unable to be prescribed as a medicine in the EMR (i.e. blood products), the smart pump limitations to build safety limits around only one medicine (i.e. multi-additive infusions) or excluded due to specialist hospital formulary.
Table 2
The classification of medications and fluids included for interoperability.
| INTERMITTENT* | CONTINUOUS** | FLUIDS*** | ||||||
|---|---|---|---|---|---|---|---|---|
| MEDICATION CLASS | NO.# | % | MEDICATION CLASS | NO.# | % | FLUIDS | NO.# | % |
| Antiarrhythmic | 4 | 3.4 | Anaesthetic | 5 | 8.3 | Contrast | 1 | 10.0 |
| Antibiotic | 39 | 33.1 | Antiarrhythmic | 4 | 6.7 | Glucose | 3 | 30.0 |
| Antidiuretic hormone analogue | 2 | 1.7 | Anticoagulant | 2 | 3.3 | Glucose/sodium chloride combination | 2 | 20.0 |
| Antidote | 3 | 2.5 | Antidote | 2 | 3.3 | Hartmanns | 1 | 10.0 |
| Antiemetic | 1 | 0.9 | Antihypertensive | 3 | 5.0 | Sodium chloride | 2 | 20.0 |
| Antiepileptic | 4 | 3.4 | Antiplatelet | 1 | 1.7 | Parenteral nutrition | 1 | 10.0 |
| Antifungal | 10 | 8.5 | Beta2 agonist | 1 | 1.7 | |||
| Antihypertensive | 1 | 0.9 | Diuretic | 1 | 1.7 | |||
| Antineoplastic | 1 | 0.9 | Electrolyte | 6 | 10.0 | |||
| Antipsychotic | 1 | 0.9 | Inotrope/vasopressor | 10 | 16.7 | |||
| Antiviral | 4 | 3.4 | Insulin | 1 | 1.7 | |||
| Bisphosphonate | 3 | 2.5 | Iron | 1 | 1.7 | |||
| Corticosteroid | 2 | 1.7 | Monoclonal antibody | 3 | 5.0 | |||
| Diuretic | 1 | 0.9 | Neuromuscular blocker | 4 | 6.7 | |||
| Electrolyte | 7 | 5.9 | Nitrate | 1 | 1.7 | |||
| H2 antagonist | 1 | 0.9 | Opioid | 4 | 6.7 | |||
| Immunosuppressant | 5 | 4.2 | Other | 7 | 11.7 | |||
| Insulin | 1 | 0.9 | Other endocrine | 2 | 3.3 | |||
| Iron | 2 | 1.7 | Proton pump inhibitor | 2 | 3.3 | |||
| Monoclonal antibody | 4 | 3.4 | ||||||
| Non-opioid analgesic | 3 | 2.5 | ||||||
| Other | 11 | 9.3 | ||||||
| Other endocrine | 1 | 0.9 | ||||||
| Proton pump inhibitor | 2 | 1.7 | ||||||
| Thrombolytic | 1 | 0.9 | ||||||
| Vitamin | 4 | 3.4 | ||||||
| Total | 118 | 60 | 10 | |||||
* Intermittent medicine: A medication or fluid infusion that is delivered over a specified time at prescribed intervals.
** Continuous medicine: A medication or fluid that is prescribed with a dose-rate (e.g., 10 mg/kg/min). The infusion continues until therapy is no longer required or when the solution container is depleted. Dose-rate programming changes may occur during the infusion.
*** Fluids: A fluid that is prescribed with an infusion rate (e.g., mL/hour). The infusion continues until the therapy is no longer required or when the solution container is depleted. Rate programming changes may occur during the infusion.
#No.: The number of unique medicines and fluid in each class.
Table 3
The classification of medications and fluids excluded for interoperability.
| INTERMITTENT* | CONTINUOUS** | FLUIDS*** | ||||||
|---|---|---|---|---|---|---|---|---|
| MEDICATION CLASS | NO.# | % | MEDICATION CLASS | NO.# | % | MEDICATION CLASS | NO.# | % |
| Antivenom/Antidotes | 5 | 83.3 | Medication Challenge | 9 | 90.0 | Blood product | 11 | 84.6 |
| Other | 1 | 16.7 | Electrolyte | 1 | 10.0 | Palliative Care 24hr subcutaneous infusions | 1 | 7.7 |
| Clinical trial | 1 | 7.7 | ||||||
| Total | 6 | 10 | 13 | |||||
* Intermittent medicine: A medication or fluid infusion that is delivered over a specified time at prescribed intervals.
** Continuous medicine: A medication or fluid that is prescribed with a dose-rate (e.g., 10 mg/kg/min). The infusion continues until therapy is no longer required or when the solution container is depleted. Dose-rate programming changes may occur during the infusion.
*** Fluids: A fluid that is prescribed with an infusion rate (e.g., mL/hour). The infusion continues until the therapy is no longer required or when the solution container is depleted. Rate programming changes may occur during the infusion.
#No.: The number of unique medicines and fluid in each class.
Table 4
Number and percentage of medication and fluids tested for smart pump and EMR interoperability.
| MEDICATIONS | FLUIDS | TOTAL NUMBER OF MEDICATIONS AND FLUIDS | PERCENTAGE | |
|---|---|---|---|---|
| Passed | 178 | 10 | 188 | 87% (188/217) |
| Excluded | 16 | 13 | 29 | 13% (29/217) |
| Total | 194 | 23 | 217 | |
Passed: The number and percentage of medication and fluids that passed interoperability testing of smart pumps and EMR.
Excluded: The number and percentage of medication and fluids that were excluded from interoperability testing of smart pumps and EMR.
All stakeholders were involved in identifying emerging risks, discrepancies and refining workflows. In line with best practice and published literature, procedures and processes were documented to outline and guide workflows of medications [14]. These included:
All procedures and workflows were reviewed and endorsed by the hospital Medication Safety Committee to ensure they are embedded into the hospital governance process.
At the conclusion of this process, all 178 medications were successfully integrated with the EMR system and have undergone bi-directional integration with smart pumps. Clinician workflows were mapped out to ensure sustainability and integration into daily clinical practice.
This study used a consultation, collaboration and consolidation model to successfully develop and implement a complex intervention across a healthcare organisation to improve patient safety (refer to Figure 1). This process ensures adaptability and sustainability of the intervention discussed to enable continuous improvement processes in health organisations [12].
The Triple C Methodology for Implementation of Smart Pump and EMR Interoperability.
Auto-programming and integration between various systems interfaces removes human involvement with manual selection and transcription error. Studies have demonstrated that automation reduced keystrokes needed to program an infusion pump from 15 to 2 strokes (an 86% decrease), eliminating human errors [1, 6, 11]. Pump alerts, alert overrides, and reprogrammed or cancelled infusions also decreased [11], and studies demonstrated the use of DERS in smart pumps is guaranteed almost 100% of the time [3, 4, 6, 11]. EMR alerting systems also contribute to further improvements in patient safety [11]. More recently, automation in healthcare overall has been an area of considerable interest to reduce human errors, improve efficiency and deliver better care to patients [16].
This study demonstrated that applying an evidence-based methodology such as the Triple C method has resulted in the successful execution of the integration of smart pumps with EMR, and utilised a well-documented contemporary medication safety intervention to develop a closed-loop medication management system [6, 7, 17]. This intervention has the potential to significantly reduce human errors in the medication administration process, potentially improving patient care. Additionally, several studies have demonstrated similar contemporary initiatives when implemented successfully, reduce the overall incidence of medication errors in hospitals and improve patient safety [17].
Using a strategy of process mapping by clearly outlining the roles and responsibilities of each member of the stakeholder group in the project contributed to successful completion of the initiative. Similar to other studies, using a systemised approach enabled the successful development of a framework to guide integration of smart pumps and EMR interoperability [18]. In addition, as per other studies detailing interoperability of smart pumps with EMR, collaborative partnerships with key stakeholders of clinicians, informaticists, researchers, and engineers has led to successful implementation and development of a framework for interoperability to optimise patient safety [19, 20].
Although, the main strength of this methodology study is the detailed outline of implementation of an integrated medication and fluid dataset in smart pumps with EMR in an Australian setting, it has some limitations. This includes, the lack of quantitative data on the impact of the intervention on the frequency and patterns of intravenous medication administration errors involving the smart pumps. Additionally, qualitative data and economic cost for implementation was not evaluated.
Future work will focus on assessing user acceptability with interoperability, and the impact on workflow and organisational cost. This will be achieved by evaluating the DERS compliance, number of keystrokes and pump alerts, and the impact on medication errors to optimise patient safety. Additionally, this will include development of relevant policies and procedures to govern staff education and training, as well as, formalising quality improvement activities related to maintenance and updates of the datasets and their integration process.
Finally, this model of framework can also be used to map interoperability and automation of similar digital health projects such as integration of automated dispensing cabinets with EMR and dispensing software.
The study process described and identified key stages and phases for integrating smart pump infusions with EMR in an Australia hospital setting. They included collaboration and engagement of key stakeholders, consultation and support for technology infrastructure (such as the presence of pump wireless capability), and consolidation. Further work will include assessment of end user satisfaction as well as reviewing impact of both systems’ interoperability on frequency and patterns of administration errors and patient safety.
This present work has not been published previously, is not under consideration for publication elsewhere, and will not be published elsewhere in the same form. All authors approved publication of this article.
The datasets for all medications that were tested during the current study are available from the corresponding author on reasonable request.
This initiative was a methodology project of an improvement for a current practice and as it did not involve any participants, it received an exemption from the hospital ethics office. Study Human Research Ethics Committee (HREC) reference number: QA/82162/MonH-2021-294730(v1).
The authors would like to thank members of the working group for their assistance in the project development and roll out as well as members of the Medication Safety committee for their review and endorsement of the relevant guidelines involved in the project.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
The authors have no competing interests to declare.
All authors were involved in conceptualising and conducting all steps in the study. VK and MD drafted the manuscript. AR, UH, JN and YY reviewed the manuscript.
Ohashi K, Dalleur O, Dykes PC, Bates DW. Benefits and risks of using smart pumps to reduce medication error rates: a systematic review. Drug Saf. 2014 Dec; 37(12): 1011–20. DOI: https://doi.org/10.1007/s40264-014-0232-1
Giuliano KK, Su WT, Degnan DD, Fitzgerald K, Zink RJ, DeLaurentis P. Intravenous Smart Pump Drug Library Compliance: A Descriptive Study of 44 Hospitals. J Patient Saf. 2018 Dec; 14(4): e76–e82. DOI: https://doi.org/10.1097/PTS.0000000000000383
Vanderveen T, Husch M. Smart Pumps, BCMA, and EMRs: Lessons Learned About Interoperability. Patient Safety & Quality Healthcare. Analysis published online 2015, June 1st [Accessed 25/11/2021]. https://www.psqh.com/analysis/smart-pumps-bcma-and-emrs-lessons-learned-about-interoperability/.
Khalil V. Evaluating the impact of various medication safety risk reduction strategies on medication errors in an Australian Health Service. Int J Clin Pharm. 2020 Dec; 42(6): 1515–1520. DOI: https://doi.org/10.1007/s11096-020-01142-w
Westermark A, Elliott RA, Hellings J, Fairbourn M, Dean J. Optimising the medication pathway: Meeting the WHO 2022 target of 50% fewer medication errors and clearing patient backlogs through automation. BMJ International Forum on Quality & Safety in Healthcare. 2021; 9–11 June. Virtual Europe Conference.
AAMI Foundation Healthcare Technology Safety Institute (HTSI). Best Practice Recommendations for Infusion Pump-Information Network Integration, 2012. https://www.aami.org/docs/default-source/uploadedfiles/filedownloads/foundation/safetyinnovation/2012-si-pump-integration.pdf.
Khalil V, Bates A. Assessing the impact of a mixed intervention model on the reduction of medication administration errors in an Australian hospital. Ir J Med Sci; 2021 Dec 2. Epub ahead of print. PMID: 34859334. DOI: https://doi.org/10.1007/s11845-021-02872-0
Khalil V. Evaluating the impact of various medication safety risk reduction strategies on medication errors in an Australian Health Service. Int J Clin Pharm. 2020; 42: 1515–1520. DOI: https://doi.org/10.1007/s11096-020-01142-w
Biltoft J, Finneman L. Clinical and financial effects of smart pump-electronic medical record interoperability at a hospital in a regional health system. Am J Health Syst Pharm. 2018 Jul 15; 75(14): 1064–1068. DOI: https://doi.org/10.2146/ajhp161058
Hambleton SJ, Aloizos Am J. Australia’s digital health journey. Med J Aust. 2019 Apr; 210(Suppl 6): S5–S6. PMID: 30927473. DOI: https://doi.org/10.5694/mja2.50039
Chaturvedi RR, Etchegaray JM, Raaen L, Jackson J, Friedberg MW. Technology Isn’t the Half of It: Integrating Electronic Health Records and Infusion Pumps in a Large Hospital. Jt Comm J Qual Patient Saf. 2019 Oct; 45(10): 649–661. DOI: https://doi.org/10.1016/j.jcjq.2019.07.006
Khalil H, Kynoch K. Implementation of sustainable complex interventions in health care services: the triple C model. BMC Health Serv Res. 2021 Feb 15; 21(1): 143. DOI: https://doi.org/10.1186/s12913-021-06115-x
Institute for Safe Medication Practices (ISMP). Building a Smart Infusion System Drug Library; 2017, Jan 30th [Accessed September 2021]. https://www.ismp.org/sites/default/files/attachments/2017-11/Building-Smart-Infusion-System-Drug-Library.pdf.
Institute for Safe Medication Practices (ISMP). ISMP Guidelines for Optimizing Safe Implementation and Use of Smart Infusion Pumps; 2020. https://www.ismp.org/node/972.
ECRI Institute. Top 10 Health Technology Hazards for 2013. Health Devices. 2012 Nov; 41(11).
Zayas-Cabán T, Haque SN, Kemper N. Identifying Opportunities for Workflow Automation in Health Care: Lessons Learned from Other Industries. Appl Clin Inform. 2021 May; 12(3): 686–697. Epub 2021 Jul 28. PMID: 34320683; PMCID: PMC8318703. DOI: https://doi.org/10.1055/s-0041-1731744
Lapkin S, Levett-Jones T, Chenoweth L, et al. The effectiveness of interventions designed to reduce medication administration errors: a synthesis of findings from systematic reviews. Journal of nursing management. 2016; 24(7): 845–858. DOI: https://doi.org/10.1111/jonm.12390
Antonacci G, Lennox L, Barlow J, Evans L, Reed J. Process mapping in healthcare: a systematic review. BMC Health Serv Res. 2021 Apr 14; 21(1): 342. PMID: 33853610; PMCID: PMC8048073. DOI: https://doi.org/10.1186/s12913-021-06254-1
Marwitz KK, Fritschle AC, Trivedi V, Covert ML, Walroth TA, DeLaurentis P, Saunders T, Walleser N, Fuller J, Degnan D. Investigating multiple sources of data for smart infusion pump and electronic health record interoperability. Am J Health Syst Pharm. 2020 Aug 20; 77(17): 1417–1423. PMID: 32462189. DOI: https://doi.org/10.1093/ajhp/zxaa115
Joseph R, Lee SW, Anderson SV, Morrisette MJ. Impact of interoperability of smart infusion pumps and an electronic medical record in critical care. Am J Health Syst Pharm. 2020 Jul 23; 77(15): 1231–1236. PMID: 32620966. DOI: https://doi.org/10.1093/ajhp/zxaa164