E-prescribing and medication safety in community settings: A rapid scoping review

Background Medication prescribing is essential for the treatment, curing, maintenance, and/or prevention of an illness and disease, however, medication errors remain common. Common errors including prescribing and administration, pose significant risk to patients. Electronic prescribing (e-prescribing) is one intervention used to enhance the safety and quality of prescribing by decreasing medication errors and reducing harm. E-prescribing in community-based settings has not been extensively examined. Objective To map and characterize the current evidence on e-prescribing and medication safety in community pharmacy settings. Methods We conducted a rapid scoping review of quantitative, qualitative, and mixed methods studies reporting on e-prescribing and medication safety. MEDLINE All (OVID), Embase (Elsevier), CINAHL Full Text (EBSCOHost), and Scopus (Elsevier) databases were searched December 2022 using keywords and MeSH terms related to e-prescribing, medication safety, efficiency, and uptake. Articles were imported to Covidence and screened by two reviewers. Data were extracted by a single reviewer and verified by a second reviewer using a standardized data extraction form. Findings are reported in accordance with JBI Manual for Evidence Synthesis following thematic analysis to narratively describe results. Results Thirty-five studies were included in this review. Most studies were quantitative (n = 22), non-experimental study designs (n = 16) and were conducted in the United States (n = 18). Half of included studies reported physicians as the prescriber (n = 18), while the remaining reported a mix of nurse practitioners, pharmacists, and physician assistants (n = 6). Studies reported on types of errors, including prescription errors (n = 20), medication safety errors (n = 9), dispensing errors (n = 2), and administration errors (n = 1). Few studies examined patient health outcomes, such as adverse drug events (n = 5). Conclusions Findings indicate that most research is descriptive in nature and focused primarily on rates of prescription errors. Further research, such as experimental, implementation, and evaluation mixed-methods research, is needed to investigate the effects of e-prescribing on reducing error rates and improving patient and health system outcomes.


Introduction
2][3] When prescribing, clinicians must use professional judgment and follow professional standards to ensure patient and medication safety. 1,3It is essential that prescriptions clearly define the right patient, right medication, right dose, right duration, and right route; otherwise, potential harm can be caused to patients, and if serious, can be fatal. 4,5Traditionally, prescriptions were written by hand, or faxed, therefore, illegible prescriptions increased the risk of medication errors. 6n Canada, approximately 7.5% of all hospital admissions in an intensive care unit experience adverse events as a result of medication errors, 7 and approximately 1 in 30 patients are exposed to medication harm globally. 8Despite the prevalence of medication-related harm, most errors could be prevented. 8A systematic review and meta-analysis on preventable medication harm in hospital settings shows a majority of medication errors occur at the prescribing (58%, 42 to 73%, n = 9, I 2 = 94%) and monitoring (47%, 21 to 73%, n = 8, I 2 = 99%) stages of medication use. 8Common medication errors including prescribing and administration, pose significant risk to vulnerable and critically ill patients, particularly for pediatric 9 and elderly 10 patients that have the highest rates of preventable medication harm. 8,11In 2018, researchers analyzed medication errors that reached patients through communitybased pharmacies in Nova Scotia, Canada. 12The most common medication errors found included incorrect dose or frequency (27.4%; n = 254), incorrect strength or concentration (20.2%; n = 187), and incorrect drug (19.9%;n = 185). 12Medication errors in both hospital and community-based settings have a direct impact on the health system such as, increased rehospitalization rates, prolonged length of stay, and increased cost of hospitalization. 13lectronic prescribing (e-prescribing) is one intervention strategy used to enhance the safety and quality of the prescribing process by decreasing medication errors and reducing harm. 14E-prescribing is the electronic transmission of a prescription between an authorized prescriber and pharmacist using an electronic medical record or pharmacy management software. 15The potential benefits of e-prescribing include simplifying and improving the process of submitting a prescription to a pharmacy to facilitate efficient dispensing and refills. 16,17E-prescribing can also facilitate important information sharing between providers and across multiple pharmacies. 16For example, e-prescribing has shown to decrease risk of medication errors (99% relative risk reduction) and adverse drug events (98% relative risk reduction) in hospital settings and outpatient settings. 18,19se of e-prescribing has increased substantially over the past fifteen years. 20,21In hospital settings, the uptake of e-prescribing has increased ten-fold between 2008 and 2014 from 7 to 70%. 21However, in community-based settings such as primary care clinics, uptake of eprescribing is limited. 22Barriers to implementing e-prescribing within community-based clinical settings include cost, lack of provider support, patient privacy, system errors, and legal issues. 14,23Uptake of e-prescribing in community-based clinical settings is more common for young early career physicians, pediatricians, and large clinical practices with more than 9000 patients. 22As more patients have access to virtual care, 16 e-prescribing is an important tool that allows patients to receive their prescriptions at a pharmacy of their choosing close to home. 21ptake of e-prescribing in community-based settings is important for improving medication safety and efficiency, improving patient satisfaction, preventing medication error harms, and reducing health system costs from hospital admission/readmission rates. 14,19,24Despite extensive research on e-prescribing and medication safety in hospital settings; e-prescribing and medication safety in community-based settings has not been systematically examined.

Review objectives
The primary objective of this review was to map and characterize the current evidence on e-prescribing and medication safety in communitybased settings.The secondary objective was to identify outcomes related to prescriber uptake of e-prescribing, efficiencies, and process improvements.

Methods
This rapid scoping review was guided by the Joanna Briggs Institute (JBI) Manual for Evidence Synthesis, with modifications made due to the rapid timeline based on Tricco (2017) guide to rapid reviews. 25,26

Search strategy
A health information specialist (LB) designed the initial search in MEDLINE All (Ovid) using keywords and MeSH terms relating to eprescribing, medication safety, efficiency, and uptake.The search terms were developed after conducting preliminary scoping searches in November 2022, consulting with team members and subject matter experts, and mining appropriate terminology from previously identified literature on the topic.No search hedges, study design filters, language or date limits were applied to the search.A second information specialist (MR) reviewed the MEDLINE search strategy using the Peer Review of Electronic Search Strategies (PRESS) guidelines prior to translation to other databases. 27

Information sources
After peer review, the MEDLINE search strategy was translated to Embase (Elsevier), CINAHL Full Text (EBSCOHost), and Scopus (Elsevier).The search strategy for all databases is included in Supplementary Material File 1.Given the timeline for this rapid review, a separate grey literature search was not conducted, but some grey literature and preprints were captured by the database searches.All searches were executed on December 15, 2022.

Inclusion criteria 2.3.1. Participants
This review considered all literature on physicians, nurse practitioners, or pharmacists as primary users of the e-prescribing system.This review also considered all patient groups.

Concept
The primary concept of interest for this review was medication safety in the context of e-prescribing.E-prescribing was described as a tool that enables prescribers to transmit prescriptions electronically and communicate with pharmacists.Articles that only reported on electronic transmission to fax were excluded.Medication safety included communication/transcribing errors, adverse drug events, or inappropriate prescribing.A secondary concept of interest for this review was health system, prescriber, or patient factors associated with efficiency in e-prescribing or uptake of e-prescribing.The primary concept was required for a study's inclusion in the review, while the secondary concept was optional for inclusion.

Context
This review included articles where e-prescribing took place between clinician prescribers and community pharmacists.This may have included e-prescribing in primary or ambulatory care clinics or during care transitions from hospital to home.Studies were excluded from this review if the pharmacy was in a hospital or institutional setting.

Types of sources
This review considered experimental, non-experimental, quasiexperimental, mixed methods, or qualitative studies for inclusion.Descriptive studies with no evaluation component, commentaries, discursive papers, and other publications that did not present original data were excluded from this review.Evidence syntheses were excluded from this review; however, potentially relevant evidence syntheses were tagged and their included studies were screened for inclusion in this review.Non-English records and conference abstracts were excluded during screening.

Study selection
Records retrieved by the database searches were exported to RIS format.In Scopus, MEDLINE and Embase records were removed prior to export using the search string AND NOT ((INDEX(medline) OR (INDEX (embase))).All remaining citations from each database were imported into Covidence for removal of duplicates and screening.Following scoping review guidelines, 25,26 citations were screened independently by two reviewers at both the title/abstract and full text levels.Any conflicts that arose were resolved by a third reviewer or by consensus.

Data extraction
Data extraction forms were created in Microsoft Excel by the research team in consultation with a panel of content experts (Supplementary Material File 2) and piloted with a sample of included studies by two independent reviewers.Data extracted included date of publication, setting and time covered, study design, description of e-prescribing system, key outcome(s), and summary of key findings in relation to the outcome(s).Data were extracted by a single reviewer and verified independently by a second reviewer.Primary outcome definitions are described in Table 1.

Data analysis
The Canadian Medication Incident Reporting and Prevention System (CMIRPS) classification system was used to better understand the type of quality-related event. 34Quantitative data were summarized in tabular format and qualitative data were summarized narratively.

Knowledge user and patient/public involvement
Members of the review team met biweekly with a panel of content experts including pharmacists, clinicians, and a Patient/Public Partner to ensure alignment with their experiences and to learn from their expertise on e-prescribing.Content experts contributed to designing the review questions, inclusion/exclusion criteria, and data extraction form, and supported interpretation of the study findings.

Characteristics of included studies
Thirty-five studies were included in this review.There were studies identified in the database search; and 2261 duplicates removed prior to screening.A total of 4499 studies were screened at the title/ abstract level and 750 studies screened at the full-text level.The final number of included and excluded articles, with reasons for exclusion, is reported in a PRISMA diagram (Fig. 1).An overview of included study characteristics and a summary of key findings is presented (see Supplementary Material File 3).
Thirty-five included studies (quantitative: n = 22; qualitative: n = 8; mixed methods: n = 5) were published between 2007 and 2022, with studies published since 2018 (29%).Of the 21 quantitative studies, used non-experimental designs and 3 used quasi-experimental designs.Only 2 studies used experimental designs to evaluate the effect of eprescribing on prescription and/or dispensing errors compared to handwritten or printed prescriptions.
Fewer studies examined e-prescribing and patient health outcomes, including adverse drug events (n = 5). 35,36,38,40,43For example, Nanji et al. (2011) examined outpatient computerized prescribing systems across 3 American states to identify incidence of medical errors, potential adverse drug events, and rate of prescribing errors by error type and prescribing system. 40In a sample of 3850 prescriptions, 452 (12%) contained a total of 466 errors.Researchers classified 163 (35%) of these errors as potential adverse drug events, indicating that 4% of prescriptions contained potential adverse drug events.Of the potential adverse drug events, 95 (58%) were significant, 68 (42%) were serious, and none were life-threatening. 4060 Greater details of studies reporting medication safety outcomes can be found in Supplementary Material File 4.

Table 1
Primary outcome definitions.

Prescription errors
Errors associated with the prescribing process and/or the prescription 28

Medication safety errors
Errors with the potential for harm 29

Dispensing errors
Errors associated how the medication is dispensed at the pharmacy 30

Administrative errors
Errors associated with patient insurance and/or affordability 31

Patient safety
Outcomes with implications for broader patient safety 32

Adverse drug events
Harm caused by medication or lack of intended medication 33 C.E. Cassidy et al.

Efficiency, process improvement, and uptake outcomes
Fifteen studies (n = 15) measured outcomes related to efficiency, process improvements, uptake, and health system outcomes (Fig. 3).

Discussion
This review identified 35 studies reporting on e-prescribing and medication safety in community-based settings.Most of the included studies used non-experimental, quantitative research designs with a focus on describing rates of prescription errors.Few studies measured patient and health system outcomes directly.When it comes to e-prescribing, physicians are the most common e-prescriber; however, 17 of the included studies reported on a mix of e-prescribers including nurse practitioners, physician assistants, and pharmacists.Moreover, close to half of the included studies examined efficiency and uptake of e-prescribing and reported a mix of efficiencies and inefficiencies with eprescribing.Some studies examined the impact of e-prescribing on medication safety, indicating positive trends in reducing prescription and dispensing errors and improvements in patient safety and communication.This rapid scoping review highlights the need for more experimental research on the direct impacts of e-prescribing on medication safety.While the majority of studies included in this scoping review were quantitative designs (n = 22), most were non-experimental (n = 16) or quasi-experimental designs (n = 3); very few studies (n = 2) used experimental designs.Given the limitations of observational study designs on evaluating the impact of e-prescribing, future research would benefit from using experimental study design to investigate the effect of e-prescribing on the medication safety outcomes identified in this review.
Efforts are also needed to examine implications of e-prescribing on patient health outcomes beyond prescription errors.Most studies included in this review examined the rate of prescription errors rather than the impact of prescription errors on patient and health system outcomes.Studies reporting on e-prescribing in ambulatory care settings show improved patient safety, improved legibility, reduced time between prescribing and dispensing of medications, and fewer medication errors and adverse drug events. 14,19Despite these improvements, adverse outcomes from e-prescription errors in hospital settings continue to have direct negative impacts on patients and the health system. 19Adverse outcomes include mortality, increased frequency and length of hospitalizations, failure of treatment, and side effects such as increased pain, nausea and vomiting, and risk of infection. 19Further research is needed to understand contexts influencing adverse outcomes   from e-prescribing in comparison to traditional paper prescribing methods and potential strategies to mitigate these negative impacts.Pull (n = 6) and push (n = 7) models were 2 different types of eprescribing models identified in this review.Although half of included studies described either pull or push models, details were limited on the impacts of each model on medication safety, including benefits/challenges to each type of model.Improved reporting of e-prescribing interventions is needed, including descriptions of the type of pull or push e-prescription model, as well as type of e-prescribing technology used, to inform future design, implementation, scale, and spread of e-prescribing interventions.
There was a strong focus on physicians as prescribers in the included studies; however, there is a need for examination of e-prescribing practices with other prescribers.1][72][73] While patients and clinicians report positive outcomes from their expanded scope of practice, 71,72 it is unclear how e-prescribing influences patient care, communication between multiple interprofessional clinicians, and integration across health services with different electronic health record systems.
Findings from this rapid scoping review provide important insight on the range of medication safety outcomes to measure, technological barriers to address, and system process considerations.While there are multiple available e-prescribing systems, those identified in this review include Electronic Prescription Service, 61 Prescription Centre, [52][53][54] National Online Prescription Repository, 57 Integrated electronic prescribing system, 58 PocketScript, 35 SureScripts, 39,50 and Prescription Mediator. 60Although studies described the characteristics of these eprescribing systems, there was a lack of description of how to implement and sustain these systems in practice.It is anticipated that each system and local implementation context may have their own unique barriers and facilitators to implementation and sustainability.Gagnon and colleagues (2013) conducted a systematic review on the barriers and facilitators to implementing e-prescribing in primary care settings. 74arriers to implementing e-prescribing identified software/hardware problems with the system, irrelevant alerts related to medications, and patient medical information not updating as needed (e.g.prescribing history). 23,74Similarly, research has found that technology has been a barrier to e-prescribing implementation among pharmacists (in both community and hospital settings), due to loss of working hours to deal with technological problems. 23,58,75Facilitators associated with implementing e-prescribing include prescribers seeing the value to e-prescribing, technology being familiar, patients' attitudes, and efficiency. 74mproved reporting of e-prescribing interventions in the published literature is needed (including the type of push or pull e-prescription model, type of e-prescribing technology), to inform the future design, implementation, scale, and spread of e-prescribing interventions.Building on this work, additional research designs (e.g., qualitative, mixed methods) could expand our understanding of how and why eprescribing impacts medication safety, patient outcomes, and prescriber behaviours.We recommend embedding implementation science methods into future e-prescribing research to enhance the acceptance, uptake, and sustainability of e-prescribing interventions in communitybased pharmacy settings.

Limitations
Although our approach to searching and screening was comprehensive and systematic by following the discussed scoping review guidelines, we only included English-language studies during screening due to the short timeline and limited non-English language proficiencies of our research team.This may have resulted in the omission of relevant literature published in other languages.Grey literature searching was not undertaken due to time constraints and the low likelihood of retrieving additional relevant studies.

Conclusion
This rapid scoping review mapped and characterized the current evidence on e-prescribing and medication safety in community-based settings, and prescriber uptake and efficiency outcomes related to eprescribing.Findings indicate that most of the reviewed research is nonexperimental in nature and focused primarily on rates of prescribing errors.Further experimental research is needed to investigate the effects of e-prescribing on reducing error rates and improving patient and health system outcomes.This rapid scoping review provides important insight on the range of medication safety outcomes, technological barriers, and system process considerations.Future research would benefit from including detailed descriptions on the processes of implementing eprescribing to enhance the acceptance, uptake, and sustainability of eprescribing interventions in community-based pharmacy settings.