Medication errors during a pandemic: what have we learnt?

KEYWORDS:

• Medication errors
• COVID-19
• pandemic
• mitigation strategies
• patient safety

1. Introduction

Medication errors (MEs) are preventable misadventures caused by inappropriate prescribing, transcribing, dispensing, or administration of medications [1]. The global prevalence of MEs was reported to be 3%, with the highest rates reported in elderly settings (11%) and intensive care units (7%) [2]. Of all MEs reported in a large systematic review and meta-analysis, 26% were potentially life-threatening, 58% occurred at prescribing stage, and 21% related to central nervous system medications [2]. While the Coronavirus disease 2019 (COVID-19) pandemic has invaded nearly all the world’s countries forcing governments to adopt extreme measures, the quality and utilization of healthcare have been substantially impacted [3,4]. During such disasters, unplanned organizational changes in healthcare resulting from interconnected factors have taken place [5]. For example, expanding the roles of certain healthcare professionals without proper training and facilitating the licensing process for certain specialties without fluffing training requirements. Furthermore, adopting remote healthcare services with uncompleted infrastructure [6,7]. These changes may lead to the use of untested drugs or technologies, deploying a large number of poorly skilled healthcare workers or untrained medical students, and increased workload, stress, and clinical complexity of cases [8]. These error-producing conditions could impact medication knowledge, prescriber training, and knowledge of the patient; hence, they were significantly associated with increased rates of errors in clinical decision-making [7,8]. Consequently, the risk for adverse drug events (ADEs) including medication errors (MEs) may be more likely during times of pandemic. It is of utmost importance to take lessons from the current situation and investigate whether there are effective strategies to mitigate MEs during a pandemic. Therefore, this editorial takes in lessons from the literature concerning medication safety during the COVID-19 pandemic and investigates the available evidence about the rates and nature of MEs during the COVID-19 pandemic to understand its underlying causes and recommend mitigation strategies.

2. Medication errors (MEs) during the COVID-19 pandemic

There is a small body of evidence reporting the incidence and types of MEs during the COVID-19 pandemic. Nonetheless, MEs could occur at high rates for several reasons. First, the conflicting reports received by health professionals about the disease’s manifestations and treatment, especially in the first wave of the pandemic. This could lead to diagnostic errors, which can trigger MEs, particularly prescribing errors [9]. Second, sleep quality, is another factor that could support the claim that MEs occur at high rates during COVID-19. A systematic review and meta-analysis study, carried out by Salari et al, found that the rates of sleep disturbance among nurses and physicians during COVID-19 were 34.8% and 41.6%, respectively [10]. Simone et al found a significant association between poor sleep quality and the risk of medication errors among nurses [11]. Third, we expect that the overwhelming attention to COVID-19 patients during the first wave of the pandemic could leave non-COVID-19 patients exposed to risks of misdiagnosis and MEs, especially when almost all resources and staff were reallocated to acute care settings. The impact of measures taken in response to COVID-19 on medication safety of non-COVID-19 cases should be investigated.

Medication safety during the COVID-19 pandemic was addressed in six studies from the United States (US) [12], Saudi Arabia [11], Netherlands [13], Jordan [14], and France [15]. Other data reports were released from the US [16] (Table 1). These studies were extracted from the literature based on several keywords including: ‘medication errors,’ ‘COVID-19,’ ‘prescribing errors,’ ‘administration errors,’ ‘dispensing errors,’ ‘transcribing errors,’ ‘drug-related problems,’ ‘medication safety,’ ‘SARS-CoV-2,’ ‘medication safety,’ and ‘adverse events.’ Scopus, PubMed, and Google scholar databases were used in the data search. Some of these studies reported a high rate of medication errors during the COVID-19 pandemic [12,13]. The US study aimed to examine medication safety practices in the context of hydroxychloroquine use during the COVID-19 pandemic [17]. The study followed a multicentre observational point prevalence methodology and included 284 patients, of which 167 (59%) received hydroxychloroquine; a quarter of them encountered QTc prolongation and 1.1% Ventricular arrhythmia. In a Saudi, retrospective, pre-post study, 19 MEs per 100 medication orders were reported during the first wave of COVID-19 compared with 1.5 MEs per 100 medication errors during the pre-COVID-19 period. Roughly one-fifth (21.5%) of the errors reported during the COVID-19 pandemic reached patients; however, no significant patient harm was reported. Physicians were reported to be the source of 87.5% of errors reported [12]. On the other hand, the types of MEs reported before and after the pandemic were similar. A prospective observational study was conducted in a teaching hospital in the Netherlands, which assessed the prevalence, predictors, and severity of PEs in post-COVID-19 patients [14]. PEs were identified in 92% of recovered COVID-19 patients, of which 8% experienced serious errors. Admission to ICU (OR 6.08) and history of asthma (OR 5.36) were reported significant predictors for PEs. In a 3-month, disguised, direct observational study, medication administration errors were observed among Jordanian nurses in the internal medicine ward [15]. Of the 1012 administration processes observed, 910 (89.9%) administration errors were detected, of which 35.9% were adherence errors, and 24.4% were incorrect drug preparation. Among errors, 15% needed further monitoring to ensure no patient harm resulted. A systematic review carried out by Conti et al examined the impact of online checkers on the occurrence of drug-drug interactions (DDIs)-related adverse events in patients with COVID-19 [16]. The review comprised 20 studies with 575 DDIs, of which most were related to Lopinavir and Ritonavir, and concluded that using online tools to check for DDIs can be efficient in identifying DDIs-related adverse events, including life-threatening incidents. Several studies reported that antimicrobials and vitamins are the most common medications associated with safety issues during the pandemic [13,16,18,19]. This can be interpreted in the context of the claims that antibiotics are prescribed for 70% of COVID-19 patients without clinical evidence and the 6.7% increase in the number of antibiotic prescriptions between April 1 and 3 August 2020 in the United Kingdom [17].

Table 1. Summary of studies in the literature.

Author	Location	Study type	Objective	Sample size	Outcomes	Main findings
D. Almazrou et al, [11]	Saudi Arabia	Retrospective, pre-post study	To assess MEs among hospitalized patients with COVID-19	343 patients	Incidence, types, and severity of MEs	The incidence of MEs was 19% (19/100). Seventy-five MEs (21.5%) reached the patient but did not cause any harm. Wrong dose (n = 101, 29.4%) was the most common type of MEs. Physicians were the most common source of MEs (87.5%). Antibiotics (32%) and antineoplastics (25%) were the most common drug categories involved in MEs and ADRs, respectively.
Rashudy F. Mahomedradja et al, [13]	Netherlands	Prospective observational	To study PEs in post – COVID-19 patients	98 patients	The rate, seriousness, and predictors for PEs	The study findings indicated that 92% of the study sample had ≥1 PE and 8% experienced medication-related harm requiring an immediate change in medication therapy to prevent detoriation. Overall, 68% of all identified PEs were made during or after the COVID-19 related hospitalization. Multivariate analyses identified ICU admission (OR 6.08, 95% CI 2.16–17.09) and a medical history of COPD/asthma (OR 5.36, 95% CI 1.34–21.5) as independent risk factors for PEs.
Steven M Smoke et al, [12]	USA	Prospective observational cohort study	To assess medication safety in the context of COVID-19 and the use of repurposed drugs	284 patients	The percentage of patients receiving appropriate QTc monitoring	A total of 59% (167/284) of patients treated with hydroxychloroquine received appropriate QTc monitoring. QTc prolongation occurred in 25%.
Alaa M.Yousef et al, [14]	Jordan	Prospective observational study	To examine MAEs in times of COVID-19	NM	The rate, seriousness, and predictors for MAEs	A total of 1,012 opportunities of errors were observed. During the observation, 910 MAEs were identified. The adherence errors were the most frequent type (n = 364, 35.97%), which were followed by incorrect drug preparation (n = 247, 24.4%).
M Chappe et al, [15]	France	Retrospective Cohort study	To study medication safety in a geriatric acute care unit prior to and during the COVID-19 pandemic.	355 patients	Rate and types of DRPs	Clinical pharmacists detected a higher number of DRPs within prescriptions during the first wave of the COVID-19 pandemic than beforehand.
Steffanie A Strathdee et al, [16]	USA	Report	To examine COVID-19 related antimicrobial resistance challenges and opportunities	NM	Inappropriate Antibiotic prescribing and dispensing	Increase in preemptive antibiotic prescribing to prevent secondary bacterial infections

MEs: medication errors. PEs: prescribing errors. USA: United States of America. COVID-19: Coronavirus disease 2019. MAEs: medication administration errors. NM: not mentioned. DRPs: drug related problems

3. Expert opinion

The results indicate the necessity for further investigation to examine and keep track of changes in medication safety during pandemics. However, a relatively small and nascent body of research has addressed MEs during COVID-19. The available literature is limited in scope as it focuses on a single factor in a complex process. Our study demonstrates the significance of reaching an international consensus for ME definitions, classification, and severity. This could facilitate the interpretation of the studies’ findings, and thus provide a better understanding of medication safety practices in times of the pandemic. This global collaboration could lay the groundwork for active medication surveillance to be a key element of preparedness plans for pandemics and disasters. Information technology tools have been widely adopted and used for predicting clinical outcomes and severity of specific diseases, which has helped in preventing inpatient death, reducing hospitalization, and saving resources [20]. Each utilizes a set of clinical and laboratory markers, meticulously chosen, to predict the prognosis and outcomes of diseases [20]. However, our study shows that computer-based tools have been not adequately adopted for identifying and correcting MEs.

While providers’ stress and burnout are potential causes of MEs during COVID-19, the impact of organizational changes, inappropriate use of technology, and social factors on medication safety during a pandemic should be further studied in the literature. Specifically, three key elements should be considered in MEs-reduction strategies during times of pandemic:

3.1. Organizational factors

Organizational factor is vital to ensure functional infrastructure, optimized workflow, and up-to-date guidelines and policies. Staff should be allowed to work in an environment that encourages MEs reporting and continual learning on best safety practices. Further, clinicians should be fully equipped to mitigate errors. This includes access to online tools such as interaction checkers of high-quality video consultation software. Additionally, the system should be equipped with algorithms that use guidelines to decide which patients can be most suitably handled by face-to-face, video, phone calls, or just messages. To avoid patient harm, reducing MEs should be prioritized during pandemic surges and ensuring medication safety should be an integral component in preparedness plans for pandemics.

3.2. Human factors

Because many MEs are induced due to burnout and stress of healthcare personnel, the human factor, which focuses on the mental health and competency of clinicians during a pandemic is crucial [13]. In this regard, coaching clinicians on using calming techniques and interpersonal skills would reduce burnout and stress, and thus may contribute to minimizing MEs [21]. In some cases, psychological interventions are needed to treat clinicians’ distress caused by the pandemic and exacerbated by self-blame, self-isolation, drug abuse, and denial. Because clinicians with deteriorated mental health are at high risk for MEs, the above strategies would be a crucial step in reducing MEs during a pandemic. Human factor also concerns with the continuous improvement of core competencies for clinicians [22–24].

3.3. Social factors

Building a positive and effective patient-provider relationship during times of pandemic has been challenging given the absence of physical examination and difficulties in access to healthcare in some places in the world [20,21]. Hence, we believe that the integration of social factors into preparedness plans for future pandemics may sustain the patient-provider relationship during changing circumstances, which allows for early detection of MEs. In this case, a ME is called ‘a near miss,’ because it occurs, but is detected and avoided before it causes any harm. Furthermore, this relationship allows for harmful MEs to be easily disclosed to patients, which improves patient awareness of MEs and enhance safe practices. However, such a relationship is challenging in times of pandemic, given technological barriers and social distance situation. It is quite difficult to build a positive relationship with someone whom you do not see, especially if this relationship involves medical decisions that concern your health and safety. From my perspective, social media websites can be an effective tool to develop social ties between clinicians and their patients during a pandemic. During a pandemic, not only patient-provider relationship can be undermined, but also the provider-provider relationship or interprofessional collaboration is significantly influenced. Such a relationship between providers helps in sharing information and experiences about daily events, which may include key hints for safe practice and ME prevention. Therefore, maintaining socialization among clinicians during a pandemic even on a small scale would contribute to reducing MEs. A potential tool to bridge the relationship between patients and their providers is using social media websites; nonetheless, ethical issues including patient confidentiality should be thoroughly considered [22].

4. Conclusion

Although the risk for MEs during a pandemic is expected to be increased due to flaws in the system, inefficient preparedness plans, stress and burnout of personnel, and new clinical complexities, there is a small body of evidence reporting the rates, nature, and causes of MEs during the COVID-19 pandemic. This could be attributed to the plethora of error definitions and classifications, and absence of an international consensus on the methodologies of medication safety surveillance and intervention strategies. Consequently, there is a necessity for collaboration on a global scale to investigate medication issues and understand its epidemiology and etiology. This will help in developing a multifactorial mitigation strategy that ensures medication safety is integrated into organizational, human, and social factors in any preparedness plan for a pandemic.

Declaration of interests

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.