The Impact of Smart Pump Technology in the Healthcare System: A Scope Review

Smart infusion pump technology prevents errors caused by parenteral therapy. This paper aims to review the recent literature about smart pump uses, cases and adverse events, and strategies to minimize these disadvantages. Literature was explored from January 2000 to November 2021 using Google Scholar, PubMed, and ScienceDirect. There were assessments of the advantages and adverse effects of using smart pumps and strategies to overcome the adverse effects of smart pumps. The advantage of using smart pumps is that they decrease errors like incorrect rate and dose. Other benefits include a decrease in medication event rates and the ability to connect smart pumps to home health providers. However, compliance rates were negatively influenced by improper smart pumps and the overriding of soft alerts, which can cause alert fatigue and drug library update delays. Recent studies have tried to address the negative issues by improving drug library compliance and decreasing alerts to avoid alert desensitization. The investigations revealed that the smart pumps reduced errors but would only prevent some programming errors. Compliance with utilizing smart pump technology is critical in stopping medication errors. Opportunities for future improvement are broad, including integrating a smart pump infusion with the hospital system, implementing auto programming, and designing smart pump devices to be lighter, smaller, and more portable instead of the heavy, large smart pump used by most hospitals today.

Detecting Unusual Intravenous Infusion Alerting Patterns with Machine Learning Algorithms

OBJECTIVE

To detect unusual infusion alerting patterns using machine learning (ML) algorithms as a first step to advance safer inpatient intravenous administration of high-alert medications.

MATERIALS AND METHODS

We used one year of detailed propofol infusion data from a hospital. Interpretable and clinically relevant variables were feature engineered, and data points were aggregated per calendar day. A univariate (maximum times-limit) moving range (mr) control chart was used to simulate clinicians' common approach to identifying unusual infusion alerting patterns. Three different unsupervised multivariate ML-based anomaly detection algorithms (Local Outlier Factor, Isolation Forest, and k-Nearest Neighbors) were used for the same purpose. Results from the control chart and ML algorithms were compared.

RESULTS

The propofol data had 3,300 infusion alerts, 92% of which were generated during the day shift and seven of which had a times-limit greater than 10. The mr-chart identified 15 alert pattern anomalies. Different thresholds were set to include the top 15 anomalies from each ML algorithm. A total of 31 unique ML anomalies were grouped and ranked by agreeability. All algorithms agreed on 10% of the anomalies, and at least two algorithms agreed on 36%. Each algorithm detected one specific anomaly that the mr-chart did not detect. The anomaly represented a day with 71 propofol alerts (half of which were overridden) generated at an average rate of 1.06 per infusion, whereas the moving alert rate for the week was 0.35 per infusion.

DISCUSSION

These findings show that ML-based algorithms are more robust than control charts in detecting unusual alerting patterns. However, we recommend using a combination of algorithms, as multiple algorithms serve a benchmarking function and allow researchers to focus on data points with the highest algorithm agreeability.

CONCLUSION

Unsupervised ML algorithms can assist clinicians in identifying unusual alert patterns as a first step toward achieving safer infusion practices.

Detecting Unusual Intravenous Infusion Alerting Patterns with Machine Learning Algorithms

OBJECTIVE

To detect unusual infusion alerting patterns using machine learning (ML) algorithms as a first step to advance safer inpatient intravenous administration of high-alert medications.

MATERIALS AND METHODS

We used one year of detailed propofol infusion data from a hospital. Interpretable and clinically relevant variables were feature engineered, and data points were aggregated per calendar day. A univariate (maximum times-limit) moving range (mr) control chart was used to simulate clinicians' common approach to identifying unusual infusion alerting patterns. Three different unsupervised multivariate ML-based anomaly detection algorithms (Local Outlier Factor, Isolation Forest, and k-Nearest Neighbors) were used for the same purpose. Results from the control chart and ML algorithms were compared.

RESULTS

The propofol data had 3,300 infusion alerts, 92% of which were generated during the day shift and seven of which had a times-limit greater than 10. The mr-chart identified 15 alert pattern anomalies. Different thresholds were set to include the top 15 anomalies from each ML algorithm. A total of 31 unique ML anomalies were grouped and ranked by agreeability. All algorithms agreed on 10% of the anomalies, and at least two algorithms agreed on 36%. Each algorithm detected one specific anomaly that the mr-chart did not detect. The anomaly represented a day with 71 propofol alerts (half of which were overridden) generated at an average rate of 1.06 per infusion, whereas the moving alert rate for the week was 0.35 per infusion.

DISCUSSION

These findings show that ML-based algorithms are more robust than control charts in detecting unusual alerting patterns. However, we recommend using a combination of algorithms, as multiple algorithms serve a benchmarking function and allow researchers to focus on data points with the highest algorithm agreeability.

CONCLUSION

Unsupervised ML algorithms can assist clinicians in identifying unusual alert patterns as a first step toward achieving safer infusion practices.

Characterization of Medication Errors in a Medical Intensive Care Unit of a University Teaching Hospital in South Korea

OBJECTIVES

The objective of this study was to characterize the current status of medication errors (MEs) throughout the medication therapy process from prescribing to use and monitoring in a medical intensive care unit (MICU) in Korea.

METHODS

Four trained research pharmacists collected data through retrospectively reviewing electronic medical records for adults hospitalized in the MICU in 2017. The occurrence of MEs was determined through interprofessional team discussion led by an academic faculty pharmacist and a medical intensivist based on the medication administration records (MARs). The type of MEs and the consequent ME-related outcome severity were categorized according to the Pharmaceutical Care Network Europe and the National Coordinating Council for Medication Error Reporting and Prevention, respectively.

RESULTS

Overall, electronic medical records for 293 patients with 78,761 MARs were reviewed in this study. At least one type of ME occurred in 271 patients (92.5%) in association with 16,203 MARs (21%), primarily caused by inappropriate dose (35.5%), drug (27.8%), and treatment duration (25.1%). Clinically significant harmful events occurred in 24 patients (8%), including life-threatening (n = 5) and death (n = 2) cases. The 2 patients died of enoxaparin-induced fatal hemorrhage and neutropenia associated with ganciclovir and cefepime. Antibiotics were the most common culprit medications leading to clinically significant harmful events.

CONCLUSIONS

In conclusion, MEs are prevalent in the MICU in Korea, most commonly prescribing errors. Although mostly benign, harmful events including deaths may occur due to MEs, mainly associated with antibiotics. Systematic strategies to minimize these potentially fatal MEs are urgently needed.

Face-to-face instruction combined with online resources improves retention of clinical skills among undergraduate nursing students

BACKGROUND

There is growing evidence that online resources used to develop clinical skills among students in the healthcare professions can produce equivalent learning outcomes to traditional face-to-face training methods. Whether clinical competence is retained equally well for online and face-to-face training methods is not yet established.

OBJECTIVES

The objective of the study was to compare retention of competence in using an IV infusion pump among nursing students trained in its use using three different protocols.

DESIGN

A quasi-experimental design was used.

SETTING

The study was conducted in the School of Nursing and Midwifery at a regional university in Queensland, Australia.

PARTICIPANTS

Participants were 102 first year nursing students (female=89, male=13) enrolled in a medications course, ranging in age from 18 to 44years.

METHODS

Three groups of participants were trained in the use of an IV infusion pump and competence was assessed following a 26-week period of no access to the pump. Group 1 participants (ONL; n=34) were trained online using an Intravenous Pump Emulator (IVPE); Group 2 participants (ONC; n=38) were trained on campus using an actual IV pump in a traditional face-to-face setting; Group 3 participants (ONL+ONC; n=30) were trained both on campus using the actual IV pump and online using the IVPE.

RESULTS

As hypothesised, no significant differences in learning outcomes, measured by assessment scores out of 80 points, were found between the ONL (M=68.7±5.9) and ONC (M=65.5±11.5; p>0.05) groups. The ONL+ONC group recorded the highest mean assessment score (M=70.0±5.0) and completed the assessment task significantly faster (p<0.001) than the other two groups.

CONCLUSIONS

This study suggests that nursing students retained clinical competence in preparing and administrating IV infusions better when face-to-face and online learning were combined.

Clinical Practice Guideline: Safe Medication Use in the ICU

OBJECTIVE

To provide ICU clinicians with evidence-based guidance on safe medication use practices for the critically ill.

DATA SOURCES

PubMed, Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials, CINAHL, Scopus, and ISI Web of Science for relevant material to December 2015.

STUDY SELECTION

Based on three key components: 1) environment and patients, 2) the medication use process, and 3) the patient safety surveillance system. The committee collectively developed Population, Intervention, Comparator, Outcome questions and quality of evidence statements pertaining to medication errors and adverse drug events addressing the key components. A total of 34 Population, Intervention, Comparator, Outcome questions, five quality of evidence statements, and one commentary on disclosure was developed.

DATA EXTRACTION

Subcommittee members were assigned selected Population, Intervention, Comparator, Outcome questions or quality of evidence statements. Subcommittee members completed their Grading of Recommendations Assessment, Development, and Evaluation of the question with his/her quality of evidence assessment and proposed strength of recommendation, then the draft was reviewed by the relevant subcommittee. The subcommittee collectively reviewed the evidence profiles for each question they developed. After the draft was discussed and approved by the entire committee, then the document was circulated among all members for voting on the quality of evidence and strength of recommendation.

DATA SYNTHESIS

The committee followed the principles of the Grading of Recommendations Assessment, Development, and Evaluation system to determine quality of evidence and strength of recommendations.

CONCLUSIONS

This guideline evaluates the ICU environment as a risk for medication-related events and the environmental changes that are possible to improve safe medication use. Prevention strategies for medication-related events are reviewed by medication use process node (prescribing, distribution, administration, monitoring). Detailed considerations to an active surveillance system that includes reporting, identification, and evaluation are discussed. Also, highlighted is the need for future research for safe medication practices that is specific to critically ill patients.

Accidental alfaxalone overdose in a mature cat undergoing anaesthesia for magnetic resonance imaging

Case summary This case report describes the clinical signs and treatment of an alfaxalone 10 times overdose in a 12-year-old cat undergoing anaesthesia for MRI. The cat was discharged from hospital following a prolonged recovery including obtunded mentation and cardiorespiratory depression for several hours following cessation of anaesthesia. The cat received supportive therapy that included supplemental oxygen via a face mask, intravenous crystalloid fluids and active rewarming. The benefits of using alfaxalone for maintenance of anaesthesia, its pharmacokinetics and previously reported lethal doses are discussed. Strategies for reducing the incidence of medication errors are presented. Relevance and novel information An unintentional overdose of alfaxalone by continuous rate infusion has not been reported previously in a cat. Treatment is supportive and directed towards maintenance of the cardiorespiratory systems. Whenever possible, smart pumps that have been designed to reduce human error should be used to help prevent medication errors associated with continuous rate infusions.

Online intravenous pump emulator: As effective as face-to-face simulation for training nursing students

BACKGROUND

The expansion of online education for nursing students has created the challenge of finding innovative ways to teach clinical skills. An online intravenous pump emulator (IVPE) modelled on actual IV pumps used in clinical healthcare settings was developed to facilitate online education delivery.

OBJECTIVES

The objectives of the study were to implement the online IVPE and evaluate student learning outcomes and perceptions of device use.

DESIGN

A mixed method, quasi-experimental design was used.

SETTING

The study was conducted in the School of Nursing and Midwifery at a regional university in Queensland, Australia.

PARTICIPANTS

Participants were 179 first year nursing students enrolled in a medications course, ranging in age from 18 to 44years, of whom 150 were female.

METHODS

Participants were assigned to one of three groups and trained in the use of IV infusion pumps. Group 1 (n=57) were trained online using the IVPE (ONL); Group 2 (n=73) were trained on-campus using an actual IV pump (ONC); Group 3 (n=49) were trained both on-campus using the actual IV pump and online using the IVPE (ONL+ONC). Competence in using the actual IV pump was assessed for all participants at the conclusion of the training period.

RESULTS

No significant differences in learning outcomes, measured by assessment scores out of 80 points, were found between the ONL (M=65.5±9.2) and ONC (M=62.0±14.8; p>.05) groups. Significantly better learning outcomes were evident for the ONL+ONC group (M=68.7±4.9) compared to the ONC group (p<.01).

CONCLUSIONS

This study highlights that the nursing students became more competent in the skill of preparing and administrating IV infusions when face-to-face and online learning were combined.

Benefits and risks of using smart pumps to reduce medication error rates: a systematic review

BACKGROUND

Smart infusion pumps have been introduced to prevent medication errors and have been widely adopted nationally in the USA, though they are not always used in Europe or other regions. Despite widespread usage of smart pumps, intravenous medication errors have not been fully eliminated.

OBJECTIVE

Through a systematic review of recent studies and reports regarding smart pump implementation and use, we aimed to identify the impact of smart pumps on error reduction and on the complex process of medication administration, and strategies to maximize the benefits of smart pumps.

METHODS

The medical literature related to the effects of smart pumps for improving patient safety was searched in PUBMED, EMBASE, and the Cochrane Central Register of Controlled Trials (CENTRAL) (2000-2014) and relevant papers were selected by two researchers.

RESULTS

After the literature search, 231 papers were identified and the full texts of 138 articles were assessed for eligibility. Of these, 22 were included after removal of papers that did not meet the inclusion criteria. We assessed both the benefits and negative effects of smart pumps from these studies. One of the benefits of using smart pumps was intercepting errors such as the wrong rate, wrong dose, and pump setting errors. Other benefits include reduction of adverse drug event rates, practice improvements, and cost effectiveness. Meanwhile, the current issues or negative effects related to using smart pumps were lower compliance rates of using smart pumps, the overriding of soft alerts, non-intercepted errors, or the possibility of using the wrong drug library.

CONCLUSION

The literature suggests that smart pumps reduce but do not eliminate programming errors. Although the hard limits of a drug library play a main role in intercepting medication errors, soft limits were still not as effective as hard limits because of high override rates. Compliance in using smart pumps is key towards effectively preventing errors. Opportunities for improvement include upgrading drug libraries, developing standardized drug libraries, decreasing the number of unnecessary warnings, and developing stronger approaches to minimize workarounds. Also, as with other clinical information systems, smart pumps should be implemented with the idea of using continuous quality improvement processes to iteratively improve their use.

Reducing the risk of harm from intravenous potassium: a multi-factorial approach in the haematology setting

AIMS

To describe the implementation of safety systems for the use of intravenous potassium chloride in haematology patients.

METHODS

We assessed the use of intravenous potassium in a haematology ward at a tertiary hospital. Initially, we prospectively analysed the prescribing and administration of intravenous potassium to all patients over a two-week period. To complement this data, we retrospectively analysed all clinical incidents involving intravenous potassium and the dispensing patterns of potassium ampoules for the past 12 months. Drawing on evidence and recommendations from international safety literature, gaps in the safe use of potassium were identified, and a multi-factorial approach to system change was implemented.

RESULTS

A total of 18 patients were analysed with 90 intravenous bags of potassium prepared on the ward using 624 ampoules. We identified multiple opportunities for error and a lack of standardisation of therapy. The following safety systems were introduced: (i) a new prescribing and monitoring form that included dose calculation, prescriber support and pre-printed orders; (ii) removal of potassium ampoules and introduction of premixed bags; (iii) independent double checking by nursing staff at point of administration; (iv) dedicated labelling of intravenous lines; (v) extensive clinician training supported by guidelines; and (vi) introduction of 'smart pump' infusion software. The number of incidents significantly reduced from 23 to 9 (p < 0.001), and the number of ampoules dispensed reduced from 10,100 to 0.

CONCLUSIONS

A multi-factorial approach to the safe prescribing, dispensing and administration of intravenous potassium has reduced the potential for patient harm in the haematology setting.