How effective are electronic medication systems in reducing medication error rates and associated harm among hospital inpatients? A systematic review and meta-analysis

ObjectiveTo conduct a systematic review and meta-analysis to assess: 1) changes in medication error rates and associated patient harm following electronic medication system (EMS) implementation; and 2) evidence of system-related medication errors facilitated by the use of an EMS.Materials and MethodsWe searched Medline, Scopus, Embase, and CINAHL for studies published between January 2005 and March 2019, comparing medication errors rates with or without assessments of related harm (actual or potential) before and after EMS implementation. EMS was defined as a computer-based system enabling the prescribing, supply, and/or administration of medicines. Study quality was assessed.ResultsThere was substantial heterogeneity in outcomes of the 18 included studies. Only 2 were strong quality. Meta-analysis of 5 studies reporting change in actual harm post-EMS showed no reduced risk (RR: 1.22, 95% CI: 0.18–8.38, P = .8) and meta-analysis of 3 studies reporting change in administration errors found a significant reduction in error rates (RR: 0.77, 95% CI: 0.72–0.83, P = .004). Of 10 studies of prescribing error rates, 9 reported a reduction but variable denominators precluded meta-analysis. Twelve studies provided specific examples of system-related medication errors; 5 quantified their occurrence.Discussion and ConclusionDespite the wide-scale adoption of EMS in hospitals around the world, the quality of evidence about their effectiveness in medication error and associated harm reduction is variable. Some confidence can be placed in the ability of systems to reduce prescribing error rates. However, much is still unknown about mechanisms which may be most effective in improving medication safety and design features which facilitate new error risks.     

Association between limiting the number of open records in a tele-critical care setting and retract–reorder errors

BackgroundWrong patient selection errors may be tracked by retract–reorder (RAR) events. The aim of this quality improvement study was to assess the impact of reducing the number of concurrently open electronic health records from 4 to 2 on RAR errors generated by a tele-critical care service.MethodsThe study encompassed 32 months before and 21 months after restriction. Chi-Square test of proportions and T statistical process control chart for rare events were used.ResultsThere were 156 318 orders with 57 RAR errors (36.5/100 000 orders) before restriction, and 122 587 orders with 34 errors (27.7/100 000 orders) after. Rates were not statistically different (P = .20), but analysis was underpowered. When plotted on a T control chart, random variation was detected between RAR errors.ConclusionWe found no significant difference in RAR errors in the tele-critical care setting after open record limitation. Other strategies should be studied to reduce wrong patient selection errors.     

Clinical decision support system,using expert consensus-derived logic and natural language processing,decreased sedation-type order errors for patients undergoing endoscopy

ObjectiveDetermination of appropriate endoscopy sedation strategy is an important preprocedural consideration. To address manual workflow gaps that lead to sedation-type order errors at our institution, we designed and implemented a clinical decision support system (CDSS) to review orders for patients undergoing outpatient endoscopy.Materials and MethodsThe CDSS was developed and implemented by an expert panel using an agile approach. The CDSS queried patient-specific historical endoscopy records and applied expert consensus-derived logic and natural language processing to identify possible sedation order errors for human review. A retrospective analysis was conducted to evaluate impact, comparing 4-month pre-pilot and 12-month pilot periods.Results22 755 endoscopy cases were included (pre-pilot 6434 cases, pilot 16 321 cases). The CDSS decreased the sedation-type order error rate on day of endoscopy (pre-pilot 0.39%, pilot 0.037%, Odds Ratio = 0.094, P-value < 1e-8). There was no difference in background prevalence of erroneous orders (pre-pilot 0.39%, pilot 0.34%, P = .54).DiscussionAt our institution, low prevalence and high volume of cases prevented routine manual review to verify sedation order appropriateness. Using a cohort-enrichment strategy, a CDSS was able to reduce number of chart reviews needed per sedation-order error from 296.7 to 3.5, allowing for integration into the existing workflow to intercept rare but important ordering errors.ConclusionA workflow-integrated CDSS with expert consensus-derived logic rules and natural language processing significantly reduced endoscopy sedation-type order errors on day of endoscopy at our institution.     

Validation of an electronic trigger to measure missed diagnosis of stroke in emergency departments

ObjectiveDiagnostic errors are major contributors to preventable patient harm. We validated the use of an electronic health record (EHR)-based trigger (e-trigger) to measure missed opportunities in stroke diagnosis in emergency departments (EDs).MethodsUsing two frameworks, the Safer Dx Trigger Tools Framework and the Symptom-disease Pair Analysis of Diagnostic Error Framework, we applied a symptom–disease pair-based e-trigger to identify patients hospitalized for stroke who, in the preceding 30 days, were discharged from the ED with benign headache or dizziness diagnoses. The algorithm was applied to Veteran Affairs National Corporate Data Warehouse on patients seen between 1/1/2016 and 12/31/2017. Trained reviewers evaluated medical records for presence/absence of missed opportunities in stroke diagnosis and stroke-related red-flags, risk factors, neurological examination, and clinical interventions. Reviewers also estimated quality of clinical documentation at the index ED visit.ResultsWe applied the e-trigger to 7,752,326 unique patients and identified 46,931 stroke-related admissions, of which 398 records were flagged as trigger-positive and reviewed. Of these, 124 had missed opportunities (positive predictive value for “missed” = 31.2%), 93 (23.4%) had no missed opportunity (non-missed), 162 (40.7%) were miscoded, and 19 (4.7%) were inconclusive. Reviewer agreement was high (87.3%, Cohen’s kappa = 0.81). Compared to the non-missed group, the missed group had more stroke risk factors (mean 3.2 vs 2.6), red flags (mean 0.5 vs 0.2), and a higher rate of inadequate documentation (66.9% vs 28.0%).ConclusionIn a large national EHR repository, a symptom–disease pair-based e-trigger identified missed diagnoses of stroke with a modest positive predictive value, underscoring the need for chart review validation procedures to identify diagnostic errors in large data sets.     

Bias and fairness assessment of a natural language processing opioid misuse classifier: detection and mitigation of electronic health record data disadvantages across racial subgroups

ObjectivesTo assess fairness and bias of a previously validated machine learning opioid misuse classifier.Materials & MethodsTwo experiments were conducted with the classifier’s original (n = 1000) and external validation (n = 53 974) datasets from 2 health systems. Bias was assessed via testing for differences in type II error rates across racial/ethnic subgroups (Black, Hispanic/Latinx, White, Other) using bootstrapped 95% confidence intervals. A local surrogate model was estimated to interpret the classifier’s predictions by race and averaged globally from the datasets. Subgroup analyses and post-hoc recalibrations were conducted to attempt to mitigate biased metrics.ResultsWe identified bias in the false negative rate (FNR = 0.32) of the Black subgroup compared to the FNR (0.17) of the White subgroup. Top features included “heroin” and “substance abuse” across subgroups. Post-hoc recalibrations eliminated bias in FNR with minimal changes in other subgroup error metrics. The Black FNR subgroup had higher risk scores for readmission and mortality than the White FNR subgroup, and a higher mortality risk score than the Black true positive subgroup (P < .05).DiscussionThe Black FNR subgroup had the greatest severity of disease and risk for poor outcomes. Similar features were present between subgroups for predicting opioid misuse, but inequities were present. Post-hoc mitigation techniques mitigated bias in type II error rate without creating substantial type I error rates. From model design through deployment, bias and data disadvantages should be systematically addressed.ConclusionStandardized, transparent bias assessments are needed to improve trustworthiness in clinical machine learning models.     

Renal medication-related clinical decision support (CDS) alerts and overrides in the inpatient setting following implementation of a commercial electronic health record: implications for designing more effective alerts

ObjectiveTo assess the appropriateness of medication-related clinical decision support (CDS) alerts associated with renal insufficiency and the potential/actual harm from overriding the alerts.Materials and MethodsOverride rate frequency was recorded for all inpatients who had a renal CDS alert trigger between 05/2017 and 04/2018. Two random samples of 300 for each of 2 types of medication-related CDS alerts associated with renal insufficiency—“dose change” and “avoid medication”—were evaluated by 2 independent reviewers using predetermined criteria for appropriateness of alert trigger, appropriateness of override, and patient harm.ResultsWe identified 37 100 “dose change” and 5095 “avoid medication” alerts in the population evaluated, and 100% of each were overridden. Dose change triggers were classified as 12.5% appropriate and overrides of these alerts classified as 90.5% appropriate. Avoid medication triggers were classified as 29.6% appropriate and overrides 76.5% appropriate. We identified 5 adverse drug events, and, of these, 4 of the 5 were due to inappropriately overridden alerts.ConclusionAlerts were nearly always presented inappropriately and were all overridden during the 1-year period studied. Alert fatigue resulting from receiving too many poor-quality alerts may result in failure to recognize errors that could lead to patient harm. Although medication-related CDS alerts associated with renal insufficiency had previously been found to be the most clinically beneficial alerts in a legacy system, in this system they were ineffective. These findings underscore the need for improvements in alert design, implementation, and monitoring of alert performance to make alerts more patient-specific and clinically appropriate.     

Reducing electronic health record-related burnout in providers through a personalized efficiency improvement program

ObjectiveTo give providers a better understanding of how to use the electronic health record (EHR), improve efficiency, and reduce burnout.Materials and MethodsAll ambulatory providers were offered at least 1 one-on-one session with an “optimizer” focusing on filling gaps in EHR knowledge and lack of customization. Success was measured using pre- and post-surveys that consisted of validated tools and homegrown questions. Only participants who returned both surveys were included in our calculations.ResultsOut of 1155 eligible providers, 1010 participated in optimization sessions. Pre-survey return rate was 90% (1034/1155) and post-survey was 54% (541/1010). 451 participants completed both surveys. After completing their optimization sessions, respondents reported a 26% improvement in mean knowledge of EHR functionality (P < .01), a 19% increase in the mean efficiency in the EHR (P < .01), and a 17% decrease in mean after-hours EHR usage (P < .01). Of the 401 providers asked to rate their burnout, 32% reported feelings of burnout in the pre-survey compared to 23% in the post-survey (P < .01). Providers were also likely to recommend colleagues participate in the program, with a Net Promoter Score of 41.DiscussionIt is possible to improve provider efficiency and feelings of burnout with a personalized optimization program. We ascribe these improvements to the one-on-one nature of our program which provides both training as well as addressing the feeling of isolation many providers feel after implementation.ConclusionIt is possible to reduce burnout in ambulatory providers with personalized retraining designed to improve efficiency and knowledge of the EHR.     

The impact of time spent on the electronic health record after work and of clerical work on burnout among clinical faculty

ObjectiveTo identify specific thresholds of daily electronic health record (EHR) time after work and daily clerical time burden associated with burnout in clinical faculty.Materials and MethodsWe administered an institution-wide survey to faculty in all departments at Mount Sinai Health System from November 2018 to February 2019. The Maslach Burnout Inventory and Mayo Well-Being Index assessed burnout. Demographics, possible confounding variables, and time spent on EHR work/clerical burden were assessed.ResultsOf 4156 eligible faculty members, 1781(42.9%) participated in the survey. After adjustment for background factors, EHR frustration (odds ratio [OR]=1.64–1.66), spending >90 minutes on EHR-outside the workday by self-report (OR = 1.41–1.90) and >1 hour of self-reported clerical work/day (OR = 1.39) were associated with burnout. Reporting that one’s practice unloads clerical burden (OR = 0.50–0.66) and higher resilience scores (OR = 0.77–0.84) were negatively associated with burnout.Spending >90 minutes/day on EHR-outside work (OR = 0.66–0.67) and >60 minutes/day on clerical work (OR = 0.54–0.58) was associated with decreased likelihood of satisfactory work–life integration (WLI) and professional satisfaction (PS). Greater meaning in work was associated with an increasedlikelihoodof achieving WLI (OR = 2.51) and PS (OR = 21.67).ConclusionResults suggest there are thresholds of excessive time on the EHR-outside the workday (>90 minutes) and overall clerical tasks (>60 minutes), above which clinical faculty may be at increased risk for burnout, as well as reduced WLI and PS, independent of demographic characteristics and clinical work hours. These thresholds of EHR and clerical burden may inform interventions aimed at mitigating this burden to reduce physician burnout.     

Factors contributing to an increase in duplicate medication order errors after CPOE implementation

Objective

To evaluate the incidence of duplicate medication orders before and after computerized provider order entry (CPOE) with clinical decision support (CDS) implementation and identify contributing factors.

Design

CPOE with duplicate medication order alerts was implemented in a 400-bed Northeastern US community tertiary care teaching hospital. In a pre-implementation post-implementation design, trained nurses used chart review, computer-generated reports of medication orders, provider alerts, and staff reports to identify medication errors in two intensive care units (ICUs).

Measurement

Medication error data were adjudicated by a physician and a human factors engineer for error stage and type. A qualitative analysis of duplicate medication ordering errors was performed to identify contributing factors.

Results

Data were collected for 4147 patient-days pre-implementation and 4013 patient-days post-implementation. Duplicate medication ordering errors increased after CPOE implementation (pre: 48 errors, 2.6% total; post: 167 errors, 8.1% total; p<0.0001). Most post-implementation duplicate orders were either for the identical order or the same medication. Contributing factors included: (1) provider ordering practices and computer availability, for example, two orders placed within minutes by different providers on rounds; (2) communication and hand-offs, for example, duplicate orders around shift change; (3) CDS and medication database design, for example confusing alert content, high false-positive alert rate, and CDS algorithms missing true duplicates; (4) CPOE data display, for example, difficulty reviewing existing orders; and (5) local CDS design, for example, medications in order sets defaulted as ordered.

Conclusions

Duplicate medication order errors increased with CPOE and CDS implementation. Many work system factors, including the CPOE, CDS, and medication database design, contributed to their occurrence.     

Calcium channel blockers improve prognosis of patients with coronavirus disease 2019 and hypertension

Background:Hypertension is considered an important risk factor for the coronavirus disease 2019 (COVID-19). The commonly anti-hypertensive drugs are the renin-angiotensin-aldosterone system (RAAS) inhibitors, calcium channel blockers (CCBs), and beta-blockers. The association between commonly used anti-hypertensive medications and the clinical outcome of COVID-19 patients with hypertension has not been well studied.Methods:We conducted a retrospective cohort study that included all patients admitted with COVID-19 to Huo Shen Shan Hospital and Guanggu District of the Maternal and Child Health Hospital of Hubei Province, Wuhan, China. Clinical and laboratory characteristics were extracted from electronic medical records. Hypertension and anti-hypertensive treatment were confirmed by medical history and clinical records. The primary clinical endpoint was all-cause mortality. Secondary endpoints included the rates of patients in common wards transferred to the intensive care unit and hospital stay duration. Logistic regression was used to explore the risk factors associated with mortality and prognosis. Propensity score matching was used to balance the confounders between different anti-hypertensive treatments. Kaplan-Meier curves were used to compare the cumulative recovery rate. Log-rank tests were performed to test for differences in Kaplan-Meier curves between different groups.Results:Among 4569 hospitalized patients with COVID-19, 31.7% (1449/4569) had a history of hypertension. There were significant differences in mortality rates between hypertensive patients with CCBs (7/359) and those without (21/359) (1.95% vs. 5.85%, risk ratio [RR]: 0.32, 95% confidence interval [CI]: 0.13–0.76, χ² = 7.61, P = 0.0058). After matching for confounders, the mortality rates were similar between the RAAS inhibitor (4/236) and non-RAAS inhibitor (9/236) cohorts (1.69% vs. 3.81%, RR: 0.43, 95% CI: 0.13–1.43, χ² = 1.98, P = 0.1596). Hypertensive patients with beta-blockers (13/340) showed no statistical difference in mortality compared with those without (11/340) (3.82% vs. 3.24%, RR: 1.19, 95% CI: 0.53–2.69, χ² = 0.17, P = 0.6777).Conclusions:In our study, we did not find any positive or negative effects of RAAS inhibitors or beta-blockers in COVID-19 patients with hypertension, while CCBs could improve prognosis.     

Impact of event notification services on timely follow-up and rehospitalization among primary care patients at two Veterans Affairs Medical Centers

ObjectiveTo examine the effectiveness of event notification service (ENS) alerts on health care delivery processes and outcomes for older adults.Materials and methodsWe deployed ENS alerts in 2 Veterans Affairs (VA) medical centers using regional health information exchange (HIE) networks from March 2016 to December 2019. Alerts targeted VA-based primary care teams when older patients (aged 65+ years) were hospitalized or attended emergency departments (ED) outside the VA system. We employed a concurrent cohort study to compare postdischarge outcomes between patients whose providers received ENS alerts and those that did not (usual care). Outcome measures included: timely follow-up postdischarge (actual phone call within 7 days or an in-person primary care visit within 30 days) and all-cause inpatient or ED readmission within 30 days. Generalized linear mixed models, accounting for clustering by primary care team, were used to compare outcomes between groups.ResultsCompared to usual care, veterans whose primary care team received notification of non-VA acute care encounters were 4 times more likely to have phone contact within 7 days (AOR = 4.10, P < .001) and 2 times more likely to have an in-person visit within 30 days (AOR = 1.98, P = .007). There were no significant differences between groups in hospital or ED utilization within 30 days of index discharge (P = .057).DiscussionENS was associated with increased timely follow-up following non-VA acute care events, but there was no associated change in 30-day readmission rates. Optimization of ENS processes may be required to scale use and impact across health systems.ConclusionGiven the importance of ENS to the VA and other health systems, this study provides guidance for future research on ENS for improving care coordination and population outcomes.Trial RegistrationClinicalTrials.gov {"type":"clinical-trial","attrs":{"text":"NCT02689076","term_id":"NCT02689076"}}NCT02689076. “Regional Data Exchange to Improve Care for Veterans After Non-VA Hospitalization.” Registered February 23, 2016.     

Cost-effectiveness analysis of a hospital electronic medication management system

Objective To conduct a cost–effectiveness analysis of a hospital electronic medication management system (eMMS).Methods We compared costs and benefits of paper-based prescribing with a commercial eMMS (CSC MedChart) on one cardiology ward in a major 326-bed teaching hospital, assuming a 15-year time horizon and a health system perspective. The eMMS implementation and operating costs were obtained from the study site. We used data on eMMS effectiveness in reducing potential adverse drug events (ADEs), and potential ADEs intercepted, based on review of 1 202 patient charts before (n = 801) and after (n = 401) eMMS. These were combined with published estimates of actual ADEs and their costs.Results The rate of potential ADEs following eMMS fell from 0.17 per admission to 0.05; a reduction of 71%. The annualized eMMS implementation, maintenance, and operating costs for the cardiology ward were A$61 741 (US$55 296). The estimated reduction in ADEs post eMMS was approximately 80 actual ADEs per year. The reduced costs associated with these ADEs were more than sufficient to offset the costs of the eMMS. Estimated savings resulting from eMMS implementation were A$63–66 (US$56–59) per admission (A$97 740–$102 000 per annum for this ward). Sensitivity analyses demonstrated results were robust when both eMMS effectiveness and costs of actual ADEs were varied substantially.Conclusion The eMMS within this setting was more effective and less expensive than paper-based prescribing. Comparison with the few previous full economic evaluations available suggests a marked improvement in the cost–effectiveness of eMMS, largely driven by increased effectiveness of contemporary eMMs in reducing medication errors.     

Inviting patients to identify diagnostic concerns through structured evaluation of their online visit notes

BackgroundThe 21st Century Cures Act mandates patients’ access to their electronic health record (EHR) notes. To our knowledge, no previous work has systematically invited patients to proactively report diagnostic concerns while documenting and tracking their diagnostic experiences through EHR-based clinician note review.ObjectiveTo test if patients can identify concerns about their diagnosis through structured evaluation of their online visit notes.MethodsIn a large integrated health system, patients aged 18–85 years actively using the patient portal and seen between October 2019 and February 2020 were invited to respond to an online questionnaire if an EHR algorithm detected any recent unexpected return visit following an initial primary care consultation (“at-risk” visit). We developed and tested an instrument (Safer Dx Patient Instrument) to help patients identify concerns related to several dimensions of the diagnostic process based on notes review and recall of recent “at-risk” visits. Additional questions assessed patients’ trust in their providers and their general feelings about the visit. The primary outcome was a self-reported diagnostic concern. Multivariate logistic regression tested whether the primary outcome was predicted by instrument variables.ResultsOf 293 566 visits, the algorithm identified 1282 eligible patients, of whom 486 responded. After applying exclusion criteria, 418 patients were included in the analysis. Fifty-one patients (12.2%) identified a diagnostic concern. Patients were more likely to report a concern if they disagreed with statements “the care plan the provider developed for me addressed all my medical concerns” [odds ratio (OR), 2.65; 95% confidence interval [CI], 1.45–4.87) and “I trust the provider that I saw during my visit” (OR, 2.10; 95% CI, 1.19–3.71) and agreed with the statement “I did not have a good feeling about my visit” (OR, 1.48; 95% CI, 1.09–2.01).ConclusionPatients can identify diagnostic concerns based on a proactive online structured evaluation of visit notes. This surveillance strategy could potentially improve transparency in the diagnostic process.     

Medication errors—an enduring problem for children and elderly patients

Objective

To analyze the types and reasons of medication errors, committed by health care professionals, which led to toxicological consultations at the Czech Toxicological Information Centre (TIC).

Methods

Inquiries arising from medication errors for 2000–2010 were extracted and evaluated from the database of the TIC, recording the consultations of poisonings due to drugs, household products, plants, and mushrooms.

Results

From a total of 44,344 calls concerning pharmaceuticals, 215 (0.5%) were denoted by the caller as medication errors; 130 involved children (90 below 5 years of age) and 85 involved adults (30–60 years of age). The most common errors were: improper dosage (60.9%), wrong medication (19.3%), or erroneous route of administration (12.9%). The most frequent medication errors appeared using drugs affecting the nervous system (psycholeptics and antiepileptics), antibiotics, and drugs affecting the respiratory system. Nurses administering the drugs were responsible for 43.0%, physicians prescribing the drugs for 36.8%, and pharmacists dispensing the drugs for 20.2% of the errors. Of 25 patients with severe drug intoxications, 60.0% were children under 5 years of age treated with pharmaceuticals affecting the CNS, and 28.0% patients over 60 years of age with chronic application of theophylline, digoxin, or lithium.

Conclusions

The trend in medication errors has remained relatively stable over the past 11 years. The analysis of medication errors shows two high-risk categories: children of less than 5 years of age, in whom the dose was not correctly adjusted, and elderly people with chronic medication and insufficient control of their medication level. Therefore, the measures for risk reduction should focus primarily on them.     

Family history of esophageal cancer modifies the association of serum lipids and malignant esophageal lesions: a nested case-control study from the “Endoscopic Screening for Esophageal Cancer in China” trial

Background:The association of lipids and cancer has varied greatly among different cancer types, lipid components and study populations. This study is aimed to investigate the association of serum lipids and the risk of malignant lesions in esophageal squamous epithelium.Methods:In the “Endoscopic Screening for Esophageal Cancer in China” (ESECC) trial, serum samples were collected and tested for total cholesterol (TC), triglycerides, low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol at the time of subject enrollment. Cases were defined as malignant esophageal lesions identified by baseline endoscopic examination or by follow-up to May 31, 2018. Controls were randomly selected using incidence density sampling in the same cohort. Conditional logistic models were applied to identify the association of serum lipids and the risk of malignant esophageal lesions. Effect modification was evaluated by testing interaction terms of the factor under assessment and these serum lipid indicators.Results:No consistent association between serum lipid levels and esophageal malignant lesions were found in a pooled analysis of 211 cases and 2101 controls. For individuals with a family history of esophageal cancer (EC), high TC, and LDL-C were associated with a significantly increased risk of having malignant lesions (odds ratio [OR]_{High vs. Low TC} = 2.22, 95% confidence interval [CI]: 1.14–4.35; OR_{High vs. Low LDL-C} = 1.93, 95% CI: 1.01–3.65). However, a negative association was observed in participants without an EC family history (OR_{High vs. Low TC} = 0.69, 95% CI: 0.48–0.98, P_interaction = 0.002; OR_{High vs. Low LDL-C} = 0.50, 95% CI: 0.34–0.76, P_interaction < 0.001).Conclusions:In this study, we found that the association of serum lipids and malignant esophageal lesions might be modified by EC family history. The stratified analysis would be crucial for population-based studies investigating the association of serum lipids and cancer. The mechanism by which a family history of EC modifies this association warrants further investigation.