Highlights
- •Medication errors are multifacceted requring complex intervetion.
- •Causes of errors need to be identified prior implementation of appropriate intervetions.
- •Medication safety education is an integral element of interventions in a bid to reduce administration errors.
- •Interdisciplinary collaboration in the medication process contributes to the reduction of medication administration errors
Abstract
Background
Medication errors are a great concern to health care organisations as they are costly and pose a significant risk to patients. Children are three times more likely to be affected by medication errors than adults with medication administration error rates reported to be over 70%.
Objective
To identify nursing interventions to reduce medication administration errors and perform a meta-analysis.
Methods
Online databases; British Nursing Index (BNI), Cochrane Database of Systematic Reviews, Cumulative Index to Nursing and Allied Health Literature (CINAHL), EMBASE and MEDLINE were searched for relevant studies published between January 2000 to 2020. Studies with clear primary or secondary aims focusing on interventions to reduce medication administration errors in paediatrics, children and or neonates were included in the review.
Results
442 studies were screened and18 studies met the inclusion criteria. Seven interventions were identified from included studies; education programmes, medication information services, clinical pharmacist involvement, double checking, barriers to reduce interruptions during drug calculation and preparation, implementation of smart pumps and improvement strategies. Educational interventional aspects were the most common identified in 13 out of 18 included studies. Meta-analysis demonstrated an associated 64% reduction in medicine administration errors post intervention (pooled OR 0.36 (95% Confidence Interval (CI) 0.21–0.63) P = 0.0003).
Conclusion
Medication safety education is an important element of interventions to reduce administration errors. Medication errors are multifaceted that require a bundle interventional approach to address the complexities and dynamics relevant to the local context. It is imperative that causes of errors need to be identified prior to implementation of appropriate interventions.
Keywords
Introduction
Medication errors, be it prescribing, preparation and dispensing, administration or monitoring are key patient safety concerns and a quality measure of healthcare medication process management. When they occur, medication errors produce a variety of problems for patients, ranging from minor discomfort to substantial morbidity that may lead to increased length of hospital stay or death under certain circumstances (
Stucky, 2003
). An estimated 18.7% - 56% of all adverse events among hospitalised patients result from preventable medication errors (von Laue et al., 2003
), however even more significant to realise is that medication errors can occur in the absence of injury to the patient. The National Coordinating Council for Medication Error Reporting and Prevention (National Coordinating Council for Medication Error Reporting an Prevention (NCC MERP), 2020
) described medication errors as ‘any preventable event that may cause or lead inappropriate medication use or patient harm while medication is in the control of healthcare professional, patient or consumer’, demonstrating that medication errors occur as a result of human mistakes or system flaws (Stucky, 2003
). For these reasons and more, in 2017 the World Health Organisation launched a global initiative aiming at reducing medication errors by 50% within a 5 year time span (Bonner, 2017
).- Bonner L.
WHO initiative to focus on reducing medication errors by 50% in 5 years.
Pharmacy Today. 2017; 23 (availabale online at): 6
https://www.pharmacytoday.org/action/showPdf?pii=S1042-0991%2817%2930769-7
Date accessed: October 19, 2020
The paediatric and neonatal patient population are three times more likely to be affected by medication errors than adults (
Kaushal et al., 2001
; Simpson et al., 2004
), and significantly higher error rates have been reported during prescribing and administration in comparison to dispensing and monitoring (Kaushal et al., 2001
; Krähenbühl-Melcher et al., 2007
; Ross et al., 2000
). Medication prescribing and administration error rates have been reported in various studies ranging from 3 to 37% and 72–78% respectively (Bannan and Tully, 2016
; Elliott et al., 2018
).Elliott, R., Camacho, E., Campbell, F., Jankovic, D., Martyn St James, M., Kaltenthaler, E., et al (2018). Prevalence and Economic Burden of Medication Errors in The NHS in England. Rapid evidence synthesis and economic analysis of the prevalence and burden of medication error in the UK. Policy Research Unit in Economic Evaluation of Health and Care Interventions 2018. Universities of Sheffield and York. Available online at: http://www.eepru.org.uk/article/prevalence-and-economic-burden-of-medication-errors-in-the-nhs-in-england/ (Accessed on 12 May 2021).
Implementation of drug therapy in children and infants can be complex due to one or more of the following; (i) treatment of rare and life-threatening conditions requiring continuous adjustment of dosing and drug therapy, (ii) narrow therapeutic range of some drugs, and (iii) individualised dosing based on factors such as age, weight, renal functions and maturation of enzyme systems (
Bannan and Tully, 2016
; Niemann et al., 2014
). Literature highlights that children are at greater risk of harm than adults when medication errors occur (Chedoe et al., 2012
; Lindell-Osuagwu et al., 2009
).Due to the nature of errors, many studies have looked broadly at interventions that looked at reducing errors across all healthcare professionals' practice. There is a place for such interventions; however nurses are at the forefront of medication administration where errors are highest (reported to range from 72 to 78%) (
Bannan and Tully, 2016
; Elliott et al., 2018
), and it has been realised that administration errors are the hardest to intercept causing direct harmful consequences for patients (Elliott, R., Camacho, E., Campbell, F., Jankovic, D., Martyn St James, M., Kaltenthaler, E., et al (2018). Prevalence and Economic Burden of Medication Errors in The NHS in England. Rapid evidence synthesis and economic analysis of the prevalence and burden of medication error in the UK. Policy Research Unit in Economic Evaluation of Health and Care Interventions 2018. Universities of Sheffield and York. Available online at: http://www.eepru.org.uk/article/prevalence-and-economic-burden-of-medication-errors-in-the-nhs-in-england/ (Accessed on 12 May 2021).
Buckley et al., 2007
). Therefore the purpose of this review was to comprehensively compile specific nursing interventions to reduce drug administration errors and where applicable outline size of effect.Methods
Data sources
Five electronic databases were searched; British Nursing Index (BNI), Cochrane Database of Systematic Reviews, Cumulative Index to Nursing and Allied Health Literature (CINAHL), EMBASE and MEDLINE. To maximise search sensitivity, a combination of various free text key words and Medical Subject Headings (MeSH terms) were used. Search terms included; paediatric, neonates, infants, babies, medication, drug, error. Databases were searched from January 2000 to February 2020.
Study selection
The review considered peer reviewed published studies involved in implementation of an intervention aimed at reducing medication administration errors among nurses in in-patient paediatric clinical settings. Children were defined as individuals between 0 and 18 years of age. Papers published in other languages were considered if an English translation was available. Excluded studies included case studies, epidemiological studies, reviews, editorials and opinion papers.
Search strategy was performed by the hospital librarian with extensive experience in literature searches. Two authors independently identified studies at abstract and full text stages. Disagreements on potential studies for inclusion were resolved by discussion to reach consensus or involvement of the third author.
Data extraction
Data extraction was performed independently by two authors using a pre-piloted standardised form. The form sections included; author and date of publication, study design, country of study, clinical setting type, sample size, intervention implemented, pre and post- intervention outcomes.
Methodology quality assessment
The quality Assessment Tool for Before and After (Pre-Post) studies with No Control Group (BAQA) was used to assess the risk of bias of included studies (
National Heart Lung and Blood Institute, 2018
). This was calculated independently by two authors, any inconsistencies in scoring of items were resolved through consensus.Statistical analysis
Most included studies presented the results in percentages and error rates. Using pre and post intervention total drug administration error numbers, odds ratios (OR) and 95% confidence intervals (95% CI) for the likelihood of medication error reduction after the intervention (exposure) was calculated. A meta-analysis was performed in Rev. Man5 (
Higgins and Green, 2019
) using the random effect method for a pooled size effect of implementing any error reduction intervention. For the studies where OR could not be calculated, a qualitative synthesis is given.Results
Search process
Online searches identified 512 studies. After removal of duplicates 442 articles were screened at title and abstract stages. Nineteen studies were assessed at full-text and 18 studies met the inclusion criteria and were included for data extraction (Table 1 Characteristics of included studies). Fig. 1 provides an overview of the search process.
Table 1Characteristics of included studies.
| Author and year | Country | Type of study | Participants | Sample size (N) | Interventions | Results |
|---|---|---|---|---|---|---|
Yamamoto and Kanemori, 2010 | USA | Prospective cohort | Nurses (Paediatric Emergency Department) | 38 | computer-assisted vs conventional paediatric drug dosing and administration | Mean conventional drug administration and dosing time 1243s vs mean computer program total time of 879 s (P < 0.001), Mean error 1.8 conventional method vs 0.7 computer programme (P < 0.001). OR 0.55 95% CI (0.35–0.86). |
Hebbar et al., 2018 | USA | Prospective cohort | Children's Hospital Nurses (general care units, emergency departments, and intensive care units) | 1434 | simulation training program implementing MAE bundle (The Five Rights, Med-Zone, and Independent Double Check) | Serious MAEs rates decreased from 2.5 events per month during the 12-month pre-intervention period to 1.4 events per month during the 20-month intervention rollout (RR = 1.78, 95% CI = 1.03–3.1, P = 0.029). MAEs further decreased to 0.86 events per month during the 7-month post-intervention period (RR = 2.9, 95% CI = 1.2–8.5, P = 0.014) (63% decrease in MAEs from the baseline period, with a 45% drop in higher classified errors). |
Colligan et al., 2012 | USA | Prospective observational | Paediatric Specialty(nurses) | Single ward (20 nurses) | Barriers to reduce interruptions during drug calculation and preparation | Frequency of interruption: mean interruption rate per minute of occurrence was significantly reduced from 1.4 pre-intervention to 0.27 post-intervention (paired t-test = 5.7, df = 98, p < 0.01). The length of time spent from beginning to end of each occurrence was not significantly reduced (120 s vs 117 s pre-intervention and post-intervention, respectively). |
McSweeney et al., 2019 | USA | Prospective cohort | Paediatric Specialty (nurses) | 19 errors were analysed and deemed preventable by the SERS reporter | improvement strategy included several initiative(policy change, change of process, education programme, and hospital-wide safety initiatives) | The number of therapy errors per 1000 patient days fell from 19.28 in 2009 to 5.95 in 2016 (The CIPT error rate decreased by 69% from 2009 to 2016). Chi2 analysis was used to compare the result for 2009 with that for each subsequent year, with P values of 0.66, 0.35, 0.16, 0.09, 0.03, 0.12, and 0.25 found for 2010 through 2016, respectively. |
Subramanyam et al., 2016 | USA | Prospective study | Children's Hospital (anaesthetists and nurses) | 129 | 2-person verification system for infusion pump programming before medication administration | During the intervention 4 errors were rectified before the medication was administered to the patient. There was no delay in case starts (>90% before and during the project).The rate of 2-person verification of infusion pump programming increased from 0% to 90% and was sustained. |
Kanjia et al., 2019 | USA | Prospective cohort | Paediatric anaesthesia (anaesthetists, physicians and nurses) | 633 checklists and electronic medical records | Education and Checklist tool | The percentage of compliance with the safe administration checklist for acetaminophen in the preoperative period increased to 97%. Use of the paper checklist likely prompted the appropriate increase in compliance with safe administration. Additionally, provider-specific feedback produced a significant increase in compliance with use of the checklist. |
Campino et al., 2009 | Spain | Prospective cohort | NICU (doctors, nurses, nurses' aids and pharmacists) | 94 | Education strategy | Post-intervention prescription error rate and the percentage of registers with one or more incident decreased significantly from 20.7 to 3% (p < 0.001) and from 19.2 to 2.9% (p < 0.001), respectively. OR 0.12 95% CI (0.09–0.16) |
Manrique-Rodríguez et al., 2013 | Spain | Prospective observational interventional | PICU (nurses) | 97 alarms associated with real programming errors | implementing smart infusion pumps | 92 programming errors with potentially harm to the patient were intercepted (84% of errors involved analgesics, anti-infectives, inotropes, and sedatives, 97% of the errors resulted from user programming of doses or infusion rates above the hard limits defined in the smart pump drug library, potential consequences of the intercepted errors were considered to be of moderate, serious, or catastrophic severity in 49% of cases. |
Campino et al., 2009 | Spain | Prospective observational study | NICU (doctors, nurses and pharmacists) | 10 Spanish neonatal intensive care units and one hospital pharmacy service participated in the study | educational and protocol standardisation programme | In NICU, 1.35% of samples registered calculation errors in pre-intervention phase; no calculation errors were registered in hospital pharmacy service (HPS) samples. In post-intervention phase, no calculation errors were registered in either group. Accuracy error rate decreased both in NICU (54.7vs 23% P < 0.001 Chi2) and HPS (38.3 vs 14.6% P < 0.010 Chi2). OR 0.38 95% CI (0.28–0.53) |
Niemann et al., 2014 | Germany | Prospective intervention study | PCCU (nurses) | 1 PCCU Unit (142 participants) | 3 step intervention (training course, information handout, and a 76-page reference book) | Drug error prevalence decreased from 83% (555 errors/668 processes) to 63% (554/883; p < 0.001) after the intervention, OR 0.76 95% CI (0.65–0.88).Number of affected patients remained unchanged (95% vs. 89%, p = 0.370). PO drugs (1.33 errors/ process) were more error-prone than IV drugs (0.64), despite being used less frequently (27% vs. 73% of all processes, p < 0.001). The interventions decreased the prevalence to 0.77 errors/process (p < 0.001) in PO and to 0.52 in IV drugs (p = 0.025). O |
Bertsche et al., 2010
Prospective pilot intervention study to prevent medication errors in drugs administered to children by mouth or gastric tube: A programme for nurses, physicians and parents. Quality & Safety in Health Care. 2010; 19 (Article e26) | Germany | Prospective, two-period cohort intervention | Paediatric neurology ward (doctors, nurses and parents) | 17 nurses and 30 parents | educational program including; implementation instructions for appropriate drug administration, and drug information services | Post-intervention errors reduced from 261/646 tasks (40.4%) to 36/453 (7.9%, p < 0.001) in nurses and from 28/29 (96.6%) to 2/36 (5.6%, p < 0.001) in parents. OR 0.11, 955 CI (0.07–0.16) |
Otero et al., 2008 | Argentina | Cross-sectional | Paediatrics hospital (physicians, nurses and pharmacists) | 4496 medications | educational program including; a) proposed modifications on medication prescription process (2) active interaction with pharmacists during rounds, and (3) implementation of a 10 steps check list on medication error reduction. | Prevalence of medication error rate reduced from 11.4% pre-intervention to 7.3% post-intervention (ARR (−4.1%), 95% CI (−2.3) - (−5.8)), OR (0.61, 95% CI 0.50–0.75). Prescription error rate decreased from 11.4% to 7.3%, ARR ((−8.1%), 95% CI (−4.6) - (−11.6)), OR (0.48, 95% CI, 0.36–0.65). Administration error rate decreased from 17.3% to 9.2% ARR ((−2.5%), 95% CI (−0.5) - (−4.5)), OR (0.68, 95% CI, 0.51–0.91). |
Guérin et al., 2015 | Canada | Retrospective pre–post | Paediatric hospital (nurses, doctors and pharmacists) | All staff in a 500 bed mother-child hospital | implementing smart infusion pumps | A total of 2911 (accidents and incidents) events related to medications, devices, and equipment were self-reported by clinical staff in the pre-phase (Y0), 3523 in the post-phase (Y1), and 2788 in the post-phase (Y2). The total AIIV increased from 1432 in Y0 to 1834 in Y1 and further decreased to 1389. Drug related events OR post phase (Y2) 0.81, 95% CI (0.67–1.0) |
Raja Lope et al., 2009 | Malaysia | Prospective pre and post intervention study | NICU (nurses) | 50 nurses observed during phase 1 51 nurses observed during phase 2 | Education programme | MAEs reduced from 31% (59/88) pre intervention to 15.4% (26/169) post-intervention (P < 0.001) OR 0.40, 95% CI (0.24–0.67) |
Stewart et al., 2010 | Northern Ireland | Prospective cohort | Fourth year medical students and third year nursing students | 193 | education programme (teaching and workshop training) | After intervention, students reported an increase in their knowledge and awareness on paediatric medication safety, pre and post mean scores 53.9 vs 69.8 (mean difference 15.9, 95% CI 10.4–21.4 p < 0.001), shared learning pre and post mean scores 67.9 vs 76.6 (mean difference 8.9, 95% CI 4.3–13.1 p < 0.001) |
Chedoe et al., 2012 | Netherlands | Prospective cohort | NICU nurses | 311/718 (43%) doses observed pre-intervention, 284/1221 (23%) doses observed post-intervention | Multifaceted educational programme | The incidence of errors decreased from 49% to 31%. Pre-intervention rates 0.3% severe errors, 26% moderate and 23% minor errors. Post-intervention rates 0% severe errors, 23% moderate and 8% minor errors. OR 0.49 (0.29–0.84) for period (p = 0.032), route of administration (p = 0.001) |
Simpson et al., 2004 | Scotland, UK | Prospective cohort | NICU(doctors and nurse) | 105 drug errors | Educational program including; implementation of clinical pharmacy, and drug information services. | Post-interventions, monthly medication errors fell from a mean (SD) of 24.1 (1.7) per 1000 neonatal activity days to 5.1 (3.6) per 1000 days (p < 0.001) in the following three months. The subsequent change over of junior medical staff was associated with a significant increase in medication errors to 12.2 (3.6) per 1000 neonatal activity days (p = 0.037). |
Abuelsoud, 2019 | Saudi Arabia | Prospective cohort | Paediatric specialty (physicians, nurses and pharmacists) | 900 medical files reviewed to detect drug related problems | educational program including; implementation of clinical pharmacy, and drug information services | MAE rates during prescribing, administration, and monitoring stages decreased from 47, 60, and 56% to 10, 10 and 15% respectively within 9 months after intervention. OR 0.15, 95% ci (0.13–0.17), prescription error OR 0.13, 95% CI (0.08–0.16), administration error OR 0.12, 95% (0.09–0.15). |
Abbreviations: PCCU, Paediatric Critical Care Unit, ARR, absolute rate ratio, OR, odds ratio, RR, rate ratio, MAEs, medical administration errors.
Fig. 1Flow diagram of included and excluded studies.
Studies were conducted in seven countries. The majority of studies (six) were from the USA (
Colligan et al., 2012
; Hebbar et al., 2018
; Kanjia et al., 2019
; McSweeney et al., 2019
; Subramanyam et al., 2016
; Yamamoto and Kanemori, 2010
), three from Spain (Campino et al., 2009
; Campino et al., 2016
; Manrique-Rodríguez et al., 2013
), two from Germany (Bertsche et al., 2010
; - Bertsche T.
- Bertsche A.
- Krieg E.M.
- Kunz N.
- Bergmann K.
- Hanke G.
- Haefeli W.E.
Prospective pilot intervention study to prevent medication errors in drugs administered to children by mouth or gastric tube: A programme for nurses, physicians and parents.
Quality & Safety in Health Care. 2010; 19 (Article e26)
Niemann et al., 2014
) and one from each of the following nations; Argentina, Canada, Malaysia, Netherlands, Northern Ireland, Scotland (UK) and Saudi Arabia;(Abuelsoud, 2019
; Chedoe et al., 2012
; Guérin et al., 2015
; Otero et al., 2008
; Raja Lope et al., 2009
; Simpson et al., 2004
; Stewart et al., 2010
).Methodology quality assessment
All included studies in this paper clearly stated study objectives, study population and the interventions. All studies were before and after intervention studies. Sixteen studies were prospective cohort studies and one was a retrospective cohort (
Guérin et al., 2015
) and another cross-sectional (Otero et al., 2008
). Two studies (Colligan et al., 2012
; Stewart et al., 2010
) rated moderate on risk of bias (25–75%) while the rest rated low risk of bias. Randomisation and blinding of the assessors was not possible due to the nature of the studies.Outcomes
The studies looked at implementation of interventions to reduce medication administration errors (MAEs) in specific individual paediatric clinical areas or across the children's hospital as a whole. Six studies were conducted in all specialties across children's hospitals (
Colligan et al., 2012
; Guérin et al., 2015
; Hebbar et al., 2018
; McSweeney et al., 2019
; Otero et al., 2008
), seven in NICU or PCCU only (Campino et al., 2009
; Campino et al., 2016
; Chedoe et al., 2012
; Manrique-Rodríguez et al., 2013
; Niemann et al., 2014
; Raja Lope et al., 2009
), two in paediatric theatres (Kanjia et al., 2019
; Subramanyam et al., 2016
) and one in an emergency department (Yamamoto and Kanemori, 2010
) and one in a paediatric neurological ward (Bertsche et al., 2010
). One that looked at interprofessional approach to improving paediatric medication safety through interprofessional workshops to facilitate learning of knowledge was conducted on fourth year medical students and third year nursing students (- Bertsche T.
- Bertsche A.
- Krieg E.M.
- Kunz N.
- Bergmann K.
- Hanke G.
- Haefeli W.E.
Prospective pilot intervention study to prevent medication errors in drugs administered to children by mouth or gastric tube: A programme for nurses, physicians and parents.
Quality & Safety in Health Care. 2010; 19 (Article e26)
Stewart et al., 2010
).Interventions
Seven individual interventions were identified from the included studies (Appendix 1 Table 1 Summary of Interventions). These included; education programs, drug information services (posters, reference books), involvement of clinical pharmacists, double checking, smart pumps, barriers to reduce interruptions during drug calculation and preparation and improvement strategies (checklists, policy and process change).
Some of the studies implemented single interventions only while others used a bundle of interventions. Education program interventions were the most common, observed in thirteen studies. Five studies implemented the specific education programs only (
Campino et al., 2009
; Campino et al., 2016
; I. Chedoe et al., 2012
; Raja Lope et al., 2009
; Stewart et al., 2010
), while the other eight studies included other interventions with education program (Abuelsoud, 2019
; Bertsche et al., 2010
; - Bertsche T.
- Bertsche A.
- Krieg E.M.
- Kunz N.
- Bergmann K.
- Hanke G.
- Haefeli W.E.
Prospective pilot intervention study to prevent medication errors in drugs administered to children by mouth or gastric tube: A programme for nurses, physicians and parents.
Quality & Safety in Health Care. 2010; 19 (Article e26)
Hebbar et al., 2018
; Kanjia et al., 2019
; McSweeney et al., 2019
; Niemann et al., 2014
; Otero et al., 2008
; Simpson et al., 2004
).The following three intervention; education, drug information services and clinical pharmacy involvement programmes are closely associated and in some context interdependent, depending on the delivery strategy. Education intervention sessions were delivered periodically through lectures, workshops with scenarios, drug administration simulations and specific practical training for staff involved in intravenous administration. Staff competency was assessed via a series of practical exercises including dose calculations. Staff education materials or medicines information were printed and made available (training course information, handouts, and information providing clear medication administration instruction and good administration practices). Simulations and workshop practicals consisted of high reliability principles consistent with error prevention such as closed loop communication, team member check in and STAR (Stop, Think, Act, Review) (
Hebbar et al., 2018
), and double checking. An example of a simulation exercise is as follows; the staff nurse reads the drug chart, identifies the correct drug from storage, correct dose calculation counter checked by a colleague, prepares medication aseptically, labels prepared drug, check patient's identification, check drug prescription/medication label again, administer medication to the patient ensuring the right time, rate, route, colleague counterchecks administration (Raja Lope et al., 2009
). Sessions were delivered by various staff groups with additional support from pharmacy staff until competency is achieved.Clinical pharmacy service interventions were not limited to education only. Clinical pharmacist led daily bedside medications reviews focusing on issues relating to prescribing, documentation and administration were implemented (
Simpson et al., 2004
). Senior pharmacists were included in the daily medical team ward rounds with specific intravenous medications mixtures prepared in pharmacy (Abuelsoud, 2019
).The use of smart pumps were evaluated in three studies (
Guérin et al., 2015
; Manrique-Rodríguez et al., 2013
; Subramanyam et al., 2016
), computer assisted drug calculations, barriers to reduce interruptions during drug calculations and double checking before intravenous administration were used in one study each (Colligan et al., 2012
; Subramanyam et al., 2016
; Yamamoto and Kanemori, 2010
).Smart infusion pump interventions involved provision of training for all nursing staff in programing medicine administration pumps. These smart pumps have safety software with built in drug libraries and drug infusion guardrails or limits to reduce probability of error occurrence. Training covered the expected and appropriate operation of the pumps as well as the expected change in practice. The process change involved use of itemised checklist, for compliance to ensure all steps of safe drug administration were followed to reduce drug errors. Pocket cards with steps to prevent errors in the administration of medications were given to all clinicians (
Otero et al., 2008
).Meta-analysis of interventions to reduce MAEs.
Ten studies were included in a meta-analysis of interventions to reduce MAEs (Fig. 2). Three studies (
Bertsche et al., 2010
; - Bertsche T.
- Bertsche A.
- Krieg E.M.
- Kunz N.
- Bergmann K.
- Hanke G.
- Haefeli W.E.
Prospective pilot intervention study to prevent medication errors in drugs administered to children by mouth or gastric tube: A programme for nurses, physicians and parents.
Quality & Safety in Health Care. 2010; 19 (Article e26)
Chedoe et al., 2012
; Otero et al., 2008
) had their results presented in odds ratio (OR) and for the other seven studies, we used pre and post intervention data provided to calculate OR. Calculated OR for individual studies are presented in Table 1. The meta-analysis showed a pooled OR 0.36 (95% Confidence Interval (CI) 0.21–0.63) P = 0.0003) for test of overall effect. The analysis demonstrates an associated 64% reduction in medicines administration errors post intervention. However a high heterogeneity (I2 = 98%) was observed demonstrating high variation in study outcomes between included studies. No subgroup group analysis was performed.
Fig. 2Meta-analysis of interventions to reduce MAEs.
Publication bias
Publication bias was visually assessed using funnel plot developed using Rev. Man5. The plot was asymmetric indicating a possible risk of publication bias (Fig. 3 Funnel Plot).
Fig. 3Funnel Plot.
Narrative synthesis
Studies not included in the meta-analysis presented their results in various forms. Four studies (
Hebbar et al., 2018
; Manrique-Rodríguez et al., 2013
; McSweeney et al., 2019
; Simpson et al., 2004
) showed a significant reduction in MAEs after interventions. Hebbar et al., 2018
reported a decrease in MAE events from 2.5 to 1.4 per month (risk ratio (RR) 1.78, 95% CI 1.03–3.1 p = 0.02) during the 20 months rollout of a targeted simulation training to reduce serious medication events. A further reduction of monthly error events to 0.86 at 7 months post intervention was reported demonstrating a 63% decrease in MAEs from baseline period (RR = 2.9, 95% CI = 1.2–8.5, P = 0.014).In a study by
Manrique-Rodríguez et al., 2013
, the intervention (use of smart infusion pumps) intercepted 92 user programming errors rated as potential harm to the patient. The authors concluded that using smart infusion pump programming such as pump drug library, defined hard limits for infusion rates is associated with a 49% potential reduction or intercepting drug errors of moderate, serious or catastrophically severe cases. One study (Colligan et al., 2012
) implemented the barriers to reduce interruptions during drug calculation and preparation. They measured mean interruption rate per minute. They reported a significant mean interruption reduction from 1.4 pre-intervention to 0.27 post-intervention, p < 0.01.For studies that involved process and policy changes (
Kanjia et al., 2019
; Subramanyam et al., 2016
) uptake of process change as part of the interventions increased to 97% and from 0 to 90% respectively. The Northern Ireland study (Stewart et al., 2010
) targeted fourth year medical students and third year nursing students to deliver an education programme through teaching and workshop training to raise awareness on paediatric medication safety. After the intervention, students reported an increase in their knowledge and awareness of paediatric medication safety; with pre and post mean scores of 53.9 vs 69.8 (mean difference 15.9, 95% CI 10.4–21.4 p < 0.001), and shared learning experience on medication safety pre and post mean scores 67.9 vs 76.6 (mean difference 8.9, 95% CI 4.3–13.1 p < 0.001).Discussion
The current review provides an update on intervention strategies to reduce medication administration errors in paediatric and neonates. Causes of medication errors during administration are multifactorial sometimes requiring multi-faceted interventions to address the root cause. This is reflected in our findings. The seven interventions identified included; education projects, medication information services, clinical pharmacist involvement, double checking, barriers to reduce interruptions during drug calculation and preparation, implementation of smart pumps (new technology) and improvement strategies. Most studies performed bundle as opposite to single interventions. Similarly to what has been observed in previous reviews, all included studies were single arm, pre and post design without a comparative, concurrent control group (
Bannan and Tully, 2016
; Nguyen et al., 2017
). All interventions demonstrated a reduction in medication errors.In comparison to other similar recent reviews (
Bannan and Tully, 2016
; Nguyen et al., 2017
; Rinke et al., 2014
; Santesteban et al., 2015
), in this review we performed a meta-analysis in ten studies with pooled size of effect OR showing a 64% associated reduction in MAEs after intervention. Methodological and study results presentation heterogeneity was observed in included studies. This variation precluded meta-analysis for some of studies.One of the predominant reason children are at greater risk from medication errors is the nature of drug dosages that are calculated based on the child's weight or body surface (
Wong et al., 2009
). However, weight can change during hospitalisation period and recalculation of drug dosages becomes a necessity particularly in neonates. As such, in addressing this key element, all education program interventions observed included elements of dose calculations, with some offering nursing staff further support until competency is achieved. Accurate medicine administration calculations is an essential part of patient safety and nurses competency such that it has become part of the curriculum with nursing schools dedicating time and effort to teach their students how to calculate drug doses and subsequently test them for these skills (Rowe et al., 1998
).It has previously been hypothesised that a bundle interventional approach, that is using more than one intervention with more than one function has an additive advantage on reducing medication errors (
Bannan and Tully, 2016
). This is considerably plausible as individual interventions often target different aspects of the medication management and administration process demonstrating that a combination of interventions is most likely required to achieve a significant reduction in MAEs (Nguyen et al., 2017
). Therefore investigating and categorising different types of medication errors in accordance to the local context is a necessary step as part of designing and before implementing any proposed intervention.Several studies (
Campino et al., 2009
; Campino et al., 2016
; Raja Lope et al., 2009
) have demonstrated that to reduce medication errors, a change in the patient safety culture is of necessity coupled with specific training to promote effective sustainable change. To this effect, the role of structured educational intervention component as part of a MAE preventive strategy is worth noting. Thirteen studies (72%) out of 18 included studies in this review included an education component. Two previous reviews on a similar topic reported 94% (Bannan and Tully, 2016
), and 30% (Nguyen et al., 2017
) of included studies having either a staff or medication education component as part of the intervention strategy. Primarily the education aspects of the interventions are focused on increasing and addressing medication knowledge and understanding. This is often delivered in form of simulations or lectures addressing potential error prone points in the medication administration process (Campino et al., 2009
; Campino et al., 2016
). The use of simulation techniques during teaching sessions has been reported to have a substantial effect in medication administration error reduction (Ford et al., 2010
). A systematic review of 15 studies on educational interventions alone to improve prescribing quality provided no definitive conclusion (Ross and Loke, 2009
), which might suggest that education programmes are impactful only as part of bundle as opposed to stand alone intervention. However studies included in this review using education interventions alone to address medication errors during administration (Campino et al., 2009
; Campino et al., 2016
; Raja Lope et al., 2009
; Stewart et al., 2010
) demonstrated reduction in drug errors post intervention, although one of the studies concluded that the reduction was insufficient to achieve an adequate level of medication safety (Chedoe et al., 2012
).In this review, another commonly observed intervention used as part of bundle is the collaboration of clinical pharmacists on medicine activities with nurses. Involvement of clinical pharmacists has shown to prevent 58% of medication errors and 72% of potentially high risk errors (
Fortescue et al., 2003
). The presence of clinical pharmacists may help nurses to make informed clinical decisions at any point during drug administration process. Pharmacokinetic monitoring, up-to-date information on reconstitution and dilution of drugs, compatibility of intravenous medication and collaboration with other healthcare professionals to optimise therapeutic plans are roles a clinical pharmacist can undertake (Campino et al., 2009
; Prot-Labarthe et al., 2013
).Study limitations.
There are various types of medicine errors; prescribing, preparation and dispensing, monitoring and administration errors and these errors often influence the outcome of each other. Also the type of errors might be influenced more by one healthcare professional group, for example medication prescription is mainly the medical team's responsibility while drug administration is often done by the nursing staff. This review focused on drug administration error reduction interventions only, which might not give a true picture of the challenges and complexities to be considered when implementing such interventions. Some studies that predominantly looked at other drug error types with minimum emphasis on medicines administration errors could have been missed. However we performed a comprehensive search strategy to ensure inclusion of all possible studies.
There are frameworks such as the
Cochrane Effective Practice and Organisation of Care (EPOC), 2015
taxonomy of intervention and the Behaviour Change Wheel (BCW) approach (- Cochrane Effective Practice and Organisation of Care (EPOC)
EPOC taxonomy.
epoc.cochrane.org/epoc-taxonomy
Date: 2015
Michie et al., 2011
), which have categorised interventions to improve the quality of care or reduce errors (Maaskant et al., 2015
; Smith et al., 2016
). The EPOC taxonomy stratifies interventions into four groups; professional, organisational, financial and regulatory depending on the target. On the other hand, the BCW classify interventions based on their function; environmental restructuring, education, persuasion, incentivisation, coercion, training, restriction, modelling and enablement. While such frameworks have been used in previous reviews (Bannan and Tully, 2016
; Nguyen et al., 2017
) in this review we grouped identified interventions from included studies together according to their similarity for clarity to the readers.Implications for practice
Findings from this review broadly highlight the following three aspects for consideration by practicing nurses and their respective health care organisation; a) medication errors are multifaceted and the causes of errors need to be identified prior to implementation of appropriate interventions, b) medication safety education is an integral element of interventions in a bid to reduce administration errors and c) interdisciplinary collaboration in the medication process contributes to the reduction of medication administration errors. In view of the outlined considerations, the following, bundle interventions could be considered most beneficial to practicing nurses. Education interventions (which include but not limited to lectures, workshops, simulations and practical training) and provision of relevant drug information services remain predominantly fundamental in reducing medicines administration errors. Regular education sessions and updates on medicines administration could also be used as part of nursing staff continuous professional development (CPD) to enhance evidence based practice. The multidisciplinary nature of healthcare provision requires the involvement of clinical pharmacy services to ensure delivery of safe medications care. This affords practising nurses easy access to pharmacists during working hours and out of hours if further discussion is required (
Abuelsoud, 2019
). The use of smart pumps with medicines library and guard rails are now prevalent and part of standard care in most clinical areas with provision of equipment training a prerequisite.Conclusion
Administration of medicines to children via any route can sometime be a complex process requiring special attention and multifaceted interventions to reduce and or avoid potential errors. There is no ‘one size fit all’ solution in reducing medication administration errors. Identifying causes of errors within the local context and understanding conditions and mechanisms that exacerbate such practice performances is necessary in designing or choosing potential effective interventions from the list outlined in this review. Continuous monitoring and evaluation of interventions used in clinical practice is paramount for measuring effectiveness and ensuring patient safety.
Funding
None.
Availability of data and material
N/A
Code availability
N/A
Authors' contributions
Marufu and Manning conceptualized and designed the study. Hendron performed the search strategy, Marufu, Bower and Manning performed data extraction. All the authors were involved in data analysis and have contributed to the drafting, critical review and revision of the manuscript. All authors have approved the final manuscript.
Conflicts of interest
Dr. Joseph C. Manning is a current recipient of an NIHR HEE funded ICA Clinical Lectureship [ ICA-CL-2018-04-ST2-009 ]. The views expressed in this article are those of the authors and not necessarily those of the NIHR or Department of Health and Social Care, UK.
Appendix A. Supplementary data
Supplementary material
References
- Pharmacy quality improvement project to enhance the medication management process in pediatric patients.Irish Journal of Medical Science. 2019; 188: 591-600
- Bundle interventions used to reduce prescribing and administration errors in hospitalized children: A systematic review.Journal of Clinical Pharmacy and Therapeutics. 2016; 41: 246-255
- Prospective pilot intervention study to prevent medication errors in drugs administered to children by mouth or gastric tube: A programme for nurses, physicians and parents.Quality & Safety in Health Care. 2010; 19 (Article e26)
- WHO initiative to focus on reducing medication errors by 50% in 5 years.Pharmacy Today. 2017; 23 (availabale online at): 6https://www.pharmacytoday.org/action/showPdf?pii=S1042-0991%2817%2930769-7Date accessed: October 19, 2020
- Direct observation approach for detecting medication errors and adverse drug events in a pediatric intensive care unit.Pediatric Critical Care Medicine. 2007; 8: 145-152
- Educational strategy to reduce medication errors in a neonatal intensive care unit.Acta Paediatrica. 2009; 98: 782-785
- Strategies implementation to reduce medicine preparation error rate in neonatal intensive care units.European Journal of Pediatrics. 2016; 175: 755-765
- The effect of a multifaceted educational intervention on medication preparation and administration errors in neonatal intensive care.Archives of Disease in Childhood. Fetal and Neonatal Edition. 2012; 97: 449-455
- EPOC taxonomy.epoc.cochrane.org/epoc-taxonomy(accessed 09 September 2020)Date: 2015
- Designing for distractions: A human factors approach to decreasing interruptions at a centralised medication station.BMJ Quality and Safety. 2012; 21: 939-947
Elliott, R., Camacho, E., Campbell, F., Jankovic, D., Martyn St James, M., Kaltenthaler, E., et al (2018). Prevalence and Economic Burden of Medication Errors in The NHS in England. Rapid evidence synthesis and economic analysis of the prevalence and burden of medication error in the UK. Policy Research Unit in Economic Evaluation of Health and Care Interventions 2018. Universities of Sheffield and York. Available online at: http://www.eepru.org.uk/article/prevalence-and-economic-burden-of-medication-errors-in-the-nhs-in-england/ (Accessed on 12 May 2021).
- Impact of simulation-based learning on medication error rates in critically ill patients.Intensive Care Medicine. 2010; 36: 1526-1531
- Prioritizing strategies for preventing medication errors and adverse drug events in pediatric inpatients.Pediatrics. 2003; 111: 722-729
- Accidents and incidents related to intravenous drug administration: A pre–post study following implementation of smart pumps in a teaching hospital.Drug Safety. 2015; 38: 729-736
- A quality initiative: A system-wide reduction in serious medication events through targeted simulation training.Simulation in Healthcare. 2018; 13: 324-330
- Cochrane Handbook for Systematic Reviews of Interventions Version 6.0. The Cochrane Collaboration 2019.(Available from) (Accessed on 07 August 2020)
- Increasing compliance of safe medication administration in pediatric anesthesia by use of a standardized checklist.Paediatric Anaesthesia. 2019; 29: 258-264
- Medication errors and adverse drug events in pediatric inpatients.Jama. 2001; 285: 2114-2120
- Drug-related problems in hospitals: A review of the recent literature.Drug Safety. 2007; 30: 379-407
- The epidemiology of preventable adverse drug events: A review of the literature.Wiener Klinische Wochenschrift. 2003; 115: 407-415
- Off-label and unlicensed drug prescribing in three paediatric wards in Finland and review of the international literature.Journal of Clinical Pharmacy and Therapeutics. 2009; 34: 277-287
- Interventions for reducing medication errors in children in hospital.Cochrane Database of Systematic Reviews. 2015; 3
- Impact of implementing smart infusion pumps in a pediatric intensive care unit.American Journal of Health-System Pharmacy. 2013; 70: 1897-1906
- Improving safety of intravenous prostacyclin administration to pediatric patients with pulmonary hypertension.Critical Care Nurse. 2019; 39: e1-e7
- The behaviour change wheel: A new method for characterising and designing behaviour change interventions.Implementation Science. 2011; 6: 42
- About medication errors.(Available online at) (accessed on 14 September 2020)
- Study quality assessment tools.(Available at:) ([Accessed on 24 April 2018])
- Interventions to reduce medication errors in neonatal care: A systematic review.Therapeutic Advances in Drug Safety. 2017; 9: 123-155
- Drug handling in a paediatric intensive care unit--can errors be prevented by a three-step intervention?.Klinische Pädiatrie. 2014; 226: 62-67
- Medication errors in pediatric inpatients: Prevalence and results of a prevention program.Pediatrics. 2008; 122: e737-e743
- Pediatric drug-related problems: A multicenter study in four French-speaking countries.International Journal of Clinical Pharmacy. 2013; 35: 251-259
- A quality assurance study on the administration of medication by nurses in a neonatal intensive care unit.Singapore Medical Journal. 2009; 50: 68-72
- Interventions to reduce pediatric medication errors: A systematic review.Pediatrics. 2014; 134: 338-360
- Medication errors in a paediatric teaching hospital in the UK: Five years operational experience.Archives of Disease in Childhood. 2000; 83: 492-497
- Do educational interventions improve prescribing by medical students and junior doctors? A systematic review.British Journal of Clinical Pharmacology. 2009; 67: 662-670
- Errors by paediatric residents in calculating drug doses.Archives of Disease in Childhood. 1998; 79: 56-58
- Medication errors in neonatal care: A systematic review of types of errors and effectiveness of preventive strategies.Journal of Neonatal Nursing. 2015; 21: 200-208
- Reducing medication errors in the neonatal intensive care unit.Archives of Disease in Childhood. Fetal and Neonatal Edition. 2004; 89: F480-F482
- Interventions for improving outcomes in patients with multimorbidity in primary care and community settings.Cochrane Database of Systematic Reviews. 2016; 3 (Article Cd006560)
- An interprofessional approach to improving paediatric medication safety.BMC Medical Education. 2010; 10: 19
- Prevention of medication errors in the pediatric inpatient setting.Pediatrics. 2003; 112: 431-436
- Infusion medication error reduction by two-person verification: A quality improvement initiative.Pediatrics. 2016; 138
- Minimising medication errors in children.Archives of Disease in Childhood. 2009; 94: 161-164
- Comparing errors in ED computer-assisted vs conventional pediatric drug dosing and administration.The American Journal of Emergency Medicine. 2010; 28: 588-592
Article Info
Publication History
Published online: September 07, 2021
Accepted:
August 27,
2021
Received in revised form:
August 26,
2021
Received:
May 28,
2021
Identification
Copyright
© 2021 Elsevier Inc. All rights reserved.
00257-8&id=gr1.jpg){kind=link}
00257-8&id=gr2.jpg){kind=link}
00257-8&id=gr3.jpg){kind=link}