A clinical data warehouse-based process for refining medication orders alerts

What You Need to Know Before Implementing a Clinical Research Data Warehouse: Comparative Review of Integrated Data Repositories in Health Care Institutions

Background

Integrated data repositories (IDRs), also referred to as clinical data warehouses, are platforms used for the integration of several data sources through specialized analytical tools that facilitate data processing and analysis. IDRs offer several opportunities for clinical data reuse, and the number of institutions implementing an IDR has grown steadily in the past decade.

Objective

The architectural choices of major IDRs are highly diverse and determining their differences can be overwhelming. This review aims to explore the underlying models and common features of IDRs, provide a high-level overview for those entering the field, and propose a set of guiding principles for small- to medium-sized health institutions embarking on IDR implementation.

Methods

We reviewed manuscripts published in peer-reviewed scientific literature between 2008 and 2020, and selected those that specifically describe IDR architectures. Of 255 shortlisted articles, we found 34 articles describing 29 different architectures. The different IDRs were analyzed for common features and classified according to their data processing and integration solution choices.

Results

Despite common trends in the selection of standard terminologies and data models, the IDRs examined showed heterogeneity in the underlying architecture design. We identified 4 common architecture models that use different approaches for data processing and integration. These different approaches were driven by a variety of features such as data sources, whether the IDR was for a single institution or a collaborative project, the intended primary data user, and purpose (research-only or including clinical or operational decision making).

Conclusions

IDR implementations are diverse and complex undertakings, which benefit from being preceded by an evaluation of requirements and definition of scope in the early planning stage. Factors such as data source diversity and intended users of the IDR influence data flow and synchronization, both of which are crucial factors in IDR architecture planning.

"Right-to-Try" experimental drugs: an overview

The “Right-to-Try” experimental drugs act passed by Donald Trump in 2018 provides an opportunity of early access to experimental drugs for the treatment of life-threatening diseases and a potential boon to many young and under-capitalized biotechnology or pharmaceutical companies. The pros and cons of experimental drugs, including a number of “cutting edge” scientific, clinical, and a number of synergistic approaches such as artificial intelligence, machine learning, big data, data refineries, electronic health records, data driven clinical decisions and risk mitigation are reviewed.

Improving medication-related clinical decision support

PURPOSE

Current uses of medication-related clinical decision support (CDS) and recommendations for improving these systems are reviewed.

SUMMARY

Using a systematic approach, articles published from 2007 through 2014 were identified in MEDLINE and EMBASE using MeSH terms and keywords relating to the 5 basic medication-related CDS functionalities. A total of 156 full-text articles and 28 conference abstracts were reviewed across each of the 5 areas: drug-drug interaction (DDI) checks (n = 78), drug allergy checks (n = 20), drug dose support (n = 55), drug duplication checks (n = 11), and drug formulary support (n = 20). The success of medication-related CDS depends on users finding the alerts valuable and acting on the information received. Improving alert specificity and sensitivity is important for all domains. Tiering is important for improving the acceptance of DDI alerts. The ability to perform appropriate cross-sensitivity checks is key to producing appropriate drug allergy checks. Drug dosage alerts should be individualized and deliver practical recommendations. How the system is configured to identify certain drug duplications is important to prevent possible patient toxicity. Accurate knowledge databases are needed to produce relevant drug formulary alerts and encourage formulary adherence. Medication-related CDS is still relatively immature in some organizations and has substantial room for improvement. For example, decision support should consider more patient-specific factors, human factors principles should always be considered, and alert specificity must be improved in order to reduce alert fatigue.

CONCLUSION

Standardization, integration of patient-specific parameters, and consideration of human factors design principles are central to realizing the potential benefits of medication-related CDS.

Data Integration for Future Medicine (DIFUTURE)

INTRODUCTION

This article is part of the Focus Theme of Methods of Information in Medicine on the German Medical Informatics Initiative. Future medicine will be predictive, preventive, personalized, participatory and digital. Data and knowledge at comprehensive depth and breadth need to be available for research and at the point of care as a basis for targeted diagnosis and therapy. Data integration and data sharing will be essential to achieve these goals. For this purpose, the consortium Data Integration for Future Medicine (DIFUTURE) will establish Data Integration Centers (DICs) at university medical centers.

OBJECTIVES

The infrastructure envisioned by DIFUTURE will provide researchers with cross-site access to data and support physicians by innovative views on integrated data as well as by decision support components for personalized treatments. The aim of our use cases is to show that this accelerates innovation, improves health care processes and results in tangible benefits for our patients. To realize our vision, numerous challenges have to be addressed. The objective of this article is to describe our concepts and solutions on the technical and the organizational level with a specific focus on data integration and sharing.

GOVERNANCE AND POLICIES

Data sharing implies significant security and privacy challenges. Therefore, state-of-the-art data protection, modern IT security concepts and patient trust play a central role in our approach. We have established governance structures and policies safeguarding data use and sharing by technical and organizational measures providing highest levels of data protection. One of our central policies is that adequate methods of data sharing for each use case and project will be selected based on rigorous risk and threat analyses. Interdisciplinary groups have been installed in order to manage change.

ARCHITECTURAL FRAMEWORK AND METHODOLOGY

The DIFUTURE Data Integration Centers will implement a three-step approach to integrating, harmonizing and sharing structured, unstructured and omics data as well as images from clinical and research environments. First, data is imported and technically harmonized using common data and interface standards (including various IHE profiles, DICOM and HL7 FHIR). Second, data is preprocessed, transformed, harmonized and enriched within a staging and working environment. Third, data is imported into common analytics platforms and data models (including i2b2 and tranSMART) and made accessible in a form compliant with the interoperability requirements defined on the national level. Secure data access and sharing will be implemented with innovative combinations of privacy-enhancing technologies (safe data, safe settings, safe outputs) and methods of distributed computing.

USE CASES

From the perspective of health care and medical research, our approach is disease-oriented and use-case driven, i.e. following the needs of physicians and researchers and aiming at measurable benefits for our patients. We will work on early diagnosis, tailored therapies and therapy decision tools with focuses on neurology, oncology and further disease entities. Our early uses cases will serve as blueprints for the following ones, verifying that the infrastructure developed by DIFUTURE is able to support a variety of application scenarios.

DISCUSSION

Own previous work, the use of internationally successful open source systems and a state-of-the-art software architecture are cornerstones of our approach. In the conceptual phase of the initiative, we have already prototypically implemented and tested the most important components of our architecture.

A novel data-driven workflow combining literature and electronic health records to estimate comorbidities burden for a specific disease: a case study on autoimmune comorbidities in patients with celiac disease

BACKGROUND

Data collected in EHRs have been widely used to identifying specific conditions; however there is still a need for methods to define comorbidities and sources to identify comorbidities burden. We propose an approach to assess comorbidities burden for a specific disease using the literature and EHR data sources in the case of autoimmune diseases in celiac disease (CD).

METHODS

We generated a restricted set of comorbidities using the literature (via the MeSH® co-occurrence file). We extracted the 15 most co-occurring autoimmune diseases of the CD. We used mappings of the comorbidities to EHR terminologies: ICD-10 (billing codes), ATC (drugs) and UMLS (clinical reports). Finally, we extracted the concepts from the different data sources. We evaluated our approach using the correlation between prevalence estimates in our cohort and co-occurrence ranking in the literature.

RESULTS

We retrieved the comorbidities for 741 patients with CD. 18.1% of patients had at least one of the 15 studied autoimmune disorders. Overall, 79.3% of the mapped concepts were detected only in text, 5.3% only in ICD codes and/or drugs prescriptions, and 15.4% could be found in both sources. Prevalence in our cohort were correlated with literature (Spearman's coefficient 0.789, p = 0.0005). The three most prevalent comorbidities were thyroiditis 12.6% (95% CI 10.1-14.9), type 1 diabetes 2.3% (95% CI 1.2-3.4) and dermatitis herpetiformis 2.0% (95% CI 1.0-3.0).

CONCLUSION

We introduced a process that leveraged the MeSH terminology to identify relevant autoimmune comorbidities of the CD and several data sources from EHRs to phenotype a large population of CD patients. We achieved prevalence estimates comparable to the literature.

A Fast Healthcare Interoperability Resources (FHIR) layer implemented over i2b2

BACKGROUND

Standards and technical specifications have been developed to define how the information contained in Electronic Health Records (EHRs) should be structured, semantically described, and communicated. Current trends rely on differentiating the representation of data instances from the definition of clinical information models. The dual model approach, which combines a reference model (RM) and a clinical information model (CIM), sets in practice this software design pattern. The most recent initiative, proposed by HL7, is called Fast Health Interoperability Resources (FHIR). The aim of our study was to investigate the feasibility of applying the FHIR standard to modeling and exposing EHR data of the Georges Pompidou European Hospital (HEGP) integrating biology and the bedside (i2b2) clinical data warehouse (CDW).

RESULTS

We implemented a FHIR server over i2b2 to expose EHR data in relation with five FHIR resources: DiagnosisReport, MedicationOrder, Patient, Encounter, and Medication. The architecture of the server combines a Data Access Object design pattern and FHIR resource providers, implemented using the Java HAPI FHIR API. Two types of queries were tested: query type #1 requests the server to display DiagnosticReport resources, for which the diagnosis code is equal to a given ICD-10 code. A total of 80 DiagnosticReport resources, corresponding to 36 patients, were displayed. Query type #2, requests the server to display MedicationOrder, for which the FHIR Medication identification code is equal to a given code expressed in a French coding system. A total of 503 MedicationOrder resources, corresponding to 290 patients, were displayed. Results were validated by manually comparing the results of each request to the results displayed by an ad-hoc SQL query.

CONCLUSION

We showed the feasibility of implementing a Java layer over the i2b2 database model to expose data of the CDW as a set of FHIR resources. An important part of this work was the structural and semantic mapping between the i2b2 model and the FHIR RM. To accomplish this, developers must manually browse the specifications of the FHIR standard. Our source code is freely available and can be adapted for use in other i2b2 sites.

Leveraging the EHR4CR platform to support patient inclusion in academic studies: challenges and lessons learned

Background

The development of Electronic Health Records (EHRs) in hospitals offers the ability to reuse data from patient care activities for clinical research. EHR4CR is a European public-private partnership aiming to develop a computerized platform that enables the re-use of data collected from EHRs over its network. However, the reproducibility of queries may depend on attributes of the local data. Our objective was 1/ to describe the different steps that were achieved in order to use the EHR4CR platform and 2/ to identify the specific issues that could impact the final performance of the platform.

Methods

We selected three institutional studies covering various medical domains. The studies included a total of 67 inclusion and exclusion criteria and ran in two University Hospitals. We described the steps required to use the EHR4CR platform for a feasibility study. We also defined metrics to assess each of the steps (including criteria complexity, normalization quality, and data completeness of EHRs).

Results

We identified 114 distinct medical concepts from a total of 67 eligibility criteria Among the 114 concepts: 23 (20.2%) corresponded to non-structured data (i.e. for which transformation is needed before analysis), 92 (81%) could be mapped to terminologies used in EHR4CR, and 86 (75%) could be mapped to local terminologies. We identified 51 computable criteria following the normalization process. The normalization was considered by experts to be satisfactory or higher for 64.2% (43/67) of the computable criteria. All of the computable criteria could be expressed using the EHR4CR platform.

Conclusions

We identified a set of issues that could affect the future results of the platform: (a) the normalization of free-text criteria, (b) the translation into computer-friendly criteria and (c) issues related to the execution of the query to clinical data warehouses. We developed and evaluated metrics to better describe the platforms and their result. These metrics could be used for future reports of Clinical Trial Recruitment Support Systems assessment studies, and provide experts and readers with tools to insure the quality of constructed dataset.

Electronic supplementary material

The online version of this article (doi:10.1186/s12874-017-0299-3) contains supplementary material, which is available to authorized users.

The Georges Pompidou University Hospital Clinical Data Warehouse: A 8-years follow-up experience

BACKGROUND

When developed jointly with clinical information systems, clinical data warehouses (CDWs) facilitate the reuse of healthcare data and leverage clinical research.

OBJECTIVE

To describe both data access and use for clinical research, epidemiology and health service research of the "Hôpital Européen Georges Pompidou" (HEGP) CDW.

METHODS

The CDW has been developed since 2008 using an i2b2 platform. It was made available to health professionals and researchers in October 2010. Procedures to access data have been implemented and different access levels have been distinguished according to the nature of queries.

RESULTS

As of July 2016, the CDW contained the consolidated data of over 860,000 patients followed since the opening of the HEGP hospital in July 2000. These data correspond to more than 122 million clinical item values, 124 million biological item values, and 3.7 million free text reports. The ethics committee of the hospital evaluates all CDW projects that generate secondary data marts. Characteristics of the 74 research projects validated between January 2011 and December 2015 are described.

CONCLUSION

The use of HEGP CDWs is a key facilitator for clinical research studies. It required however important methodological and organizational support efforts from a biomedical informatics department.

Rule-Mining for the Early Prediction of Chronic Kidney Disease Based on Metabolomics and Multi-Source Data

¹H Nuclear Magnetic Resonance (NMR)-based metabolic profiling is very promising for the diagnostic of the stages of chronic kidney disease (CKD). Because of the high dimension of NMR spectra datasets and the complex mixture of metabolites in biological samples, the identification of discriminant biomarkers of a disease is challenging. None of the widely used chemometric methods in NMR metabolomics performs a local exhaustive exploration of the data. We developed a descriptive and easily understandable approach that searches for discriminant local phenomena using an original exhaustive rule-mining algorithm in order to predict two groups of patients: 1) patients having low to mild CKD stages with no renal failure and 2) patients having moderate to established CKD stages with renal failure. Our predictive algorithm explores the m-dimensional variable space to capture the local overdensities of the two groups of patients under the form of easily interpretable rules. Afterwards, a L2-penalized logistic regression on the discriminant rules was used to build predictive models of the CKD stages. We explored a complex multi-source dataset that included the clinical, demographic, clinical chemistry, renal pathology and urine metabolomic data of a cohort of 110 patients. Given this multi-source dataset and the complex nature of metabolomic data, we analyzed 1- and 2-dimensional rules in order to integrate the information carried by the interactions between the variables. The results indicated that our local algorithm is a valuable analytical method for the precise characterization of multivariate CKD stage profiles and as efficient as the classical global model using chi2 variable section with an approximately 70% of good classification level. The resulting predictive models predominantly identify urinary metabolites (such as 3-hydroxyisovalerate, carnitine, citrate, dimethylsulfone, creatinine and N-methylnicotinamide) as relevant variables indicating that CKD significantly affects the urinary metabolome. In addition, the simple knowledge of the concentration of urinary metabolites classifies the CKD stage of the patients correctly.

The Virtuous Circles of Clinical Information Systems: a Modern Utopia

CONTEXT

Clinical information systems (CIS) are developed with the aim of improving both the efficiency and the quality of care.

OBJECTIVE

This position paper is based on the hypothesis that such vision is partly a utopian view of the emerging eSociety.

METHODS

Examples are drawn from 15 years of experience with the fully integrated Georges Pompidou University Hospital (HEGP) CIS and temporal data series extracted from the data warehouses of Assistance Publique - Hôpitaux de Paris (AP-HP) acute care hospitals which share the same administrative organization as HEGP. Three main virtuous circles are considered: user satisfaction vs. system use, system use vs. cost efficiency, and system use vs quality of care.

RESULTS

In structural equation models (SEM), the positive bidirectional relationship between user satisfaction and use was only observed in the early HEGP CIS deployment phase (first four years) but disappeared in late post-adoption (≥8 years). From 2009 to 2013, financial efficiency of 20 AP-HP hospitals evaluated with stochastic frontier analysis (SFA) models diminished by 0.5% per year. The lower decrease of efficiency observed between the three hospitals equipped with a more mature CIS and the 17 other hospitals was of the same order of magnitude than the difference observed between pediatric and non-pediatric hospitals. Outcome quality benefits that would bring evidence to the system use vs. quality loop are unlikely to be obtained in a near future since they require integration with population-based outcome measures including mortality, morbidity, and quality of life that may not be easily available.

CONCLUSION

Barriers to making the transformation of the utopian part of the CIS virtuous circles happen should be overcome to actually benefit the emerging eSociety.

Prescription reporting with immediate medication utilization mapping (PRIMUM): development of an alert to improve narcotic prescribing

BACKGROUND

Prescription narcotic overdoses and abuse have reached alarming numbers. To address this epidemic, integrated clinical decision support within the electronic medical record (EMR) to impact prescribing behavior was developed and tested.

METHODS

A multidisciplinary Expert Panel identified risk factors for misuse, abuse, or diversion of opioids or benzodiazepines through literature reviews and consensus building for inclusion in a rule within the EMR. We ran the rule "silently" to test the rule and collect baseline data.

RESULTS

Five criteria were programmed to trigger the alert; based on data collected during a "silent" phase, thresholds for triggers were modified. The alert would have fired in 21.75 % of prescribing encounters (1.30 % of all encounters; n = 9998), suggesting the alert will have a low prescriber burden yet capture a significant number of at-risk patients.

CONCLUSIONS

While the use of the EMR to provide clinical decision support is not new, utilizing it to develop and test an intervention is novel. We successfully built an alert system to address narcotic prescribing by providing critical, objective information at the point of care. The silent phase data were useful to appropriately tune the alert and obtain support for widespread implementation. Future healthcare initiatives can utilize similar methodology to collect data prospectively via the electronic medical record to inform the development, delivery, and evaluation of interventions.

Detection of Drug-Drug Interactions Inducing Acute Kidney Injury by Electronic Health Records Mining

BACKGROUND AND OBJECTIVE

While risk of acute kidney injury (AKI) is a well documented adverse effect of some drugs, few studies have assessed the relationship between drug-drug interactions (DDIs) and AKI. Our objective was to develop an algorithm capable of detecting potential signals on this relationship by retrospectively mining data from electronic health records.

MATERIAL AND METHODS

Data were extracted from the clinical data warehouse (CDW) of the Hôpital Européen Georges Pompidou (HEGP). AKI was defined as the first level of the RIFLE criteria, that is, an increase ≥50 % of creatinine basis. Algorithm accuracy was tested on 20 single drugs, 10 nephrotoxic and 10 non-nephrotoxic. We then tested 45 pairs of non-nephrotoxic drugs, among the most prescribed at our hospital and representing distinct pharmacological classes for DDIs.

RESULTS

Sensitivity and specificity were 50 % [95 % confidence interval (CI) 23.66-76.34] and 90 % (95 % CI 59.58-98.21), respectively, for single drugs. Our algorithm confirmed a previously identified signal concerning clarithromycin and calcium-channel blockers (unadjusted odds ratio (ORu) 2.92; 95 % CI 1.11-7.69, p = 0.04). Among the 45 drug pairs investigated, we identified a signal concerning 55 patients in association with bromazepam and hydroxyzine (ORu 1.66; 95 % CI 1.23-2.23). This signal was not confirmed after a chart review. Even so, AKI and co-prescription were confirmed for 96 % (95 % CI 88-99) and 88 % (95 % CI 76-94) of these patients, respectively.

CONCLUSION

Data mining techniques on CDW can foster the detection of adverse drug reactions when drugs are used alone or in combination.

[Bariatric surgery, stomas and other digestive tract reductions: Insufficient data and recommendations to adapt medicines regimens in therapeutic practice]

Surgery modifying digestive tract may alter drugs pharmacokinetics. To maintain concentrations of active substance in their therapeutic ranges, a dosage adjustment or change of drug may be necessary. This is particularly important when no pharmacological or pharmacodynamic parameter reflecting the medication effectiveness is easily measurable. Our objective was to gather the information and documentary tools that can guide prescription in these patients with rearranged digestive tract. We searched information on the documentary portals of French agencies, on gray literature, on MEDLINE and in the summaries product characteristics. No information was found on the website of French agencies, sparse data were identified in gray literature. Some document are discordant, most are imprecise. One hundred and ten studies or case reports referenced on MEDLINE describe 79 medications pharmacokinetics after gastrointestinal surgery. Four are not available in France. Six literature reviews were found. Four summaries of product characteristics provided information related to drug absorption. No documentary tool adapted to clinical routine exists. This unsatisfactory situation is a barrier to optimal patients care. Information is available. It is however necessary to gather under an ergonomic shape adapted to clinical routine, bringing the surgery type, pharmacokinetic changes induced and what to do about the dose adjustment.

Factors associated with medication warning acceptance for hospitalized adults

BACKGROUND

Computerized provider order entry (CPOE) systems can warn clinicians ordering medications about potential allergic or adverse reactions, duplicate therapy, and interactions with other medications. Clinicians frequently override these warnings. Understanding the factors associated with warning acceptance should guide revisions to these systems.

OBJECTIVE

Increase understanding of the factors associated with medication warning acceptance.

DESIGN

Retrospective study of all single-medication warnings generated in a CPOE system from October 2009 through April 2010.

SETTING

Academic medical center.

PATIENTS

All adult non-intensive care unit patients hospitalized during the study period.

RESULTS

A total of 40,391 medication orders generated a single-medication warning during the 7-month study period. Of these warnings, 47% were duplicate warnings, 47% interaction warnings, 6% allergy warnings, 0.1% adverse reaction warnings, and 9.8% were repeated for the same patient, medication, and provider. Only 4% of warnings were accepted. In multivariate analysis, warning acceptance was positively associated with male patient gender, admission to a service other than internal medicine, caregiver status other than resident, parenteral medications, lower numbers of warnings, and allergy or adverse reaction warning types. Older patient age, longer length of stay, inclusion on the Institute for Safe Medication Practice's List of High Alert Medications, and interaction warning type were all negatively associated with warning acceptance.

CONCLUSIONS

Medication warnings are rarely accepted. Acceptance is more likely when the warning is infrequently encountered, and least likely when it is potentially most important. Warning systems should be redesigned to increase their effectiveness for the sickest patients, the least experienced physicians, and the medications with the greatest potential to cause harm.

A survey of nursing home physicians to determine laboratory monitoring adverse drug event alert preferences

OBJECTIVE

We conducted a survey of nursing home physicians to learn about (1) the laboratory value thresholds that clinical event monitors should use to generate alerts about potential adverse drug events (ADEs); (2) the specific information to be included in the alerts; and (3) the communication modality that should be used for communicating them.

METHODS

Nursing home physician attendees of the 2010 Conference of AMDA: The Society for Post-Acute and Long-Term Care Medicine.

RESULTS

A total of 800 surveys were distributed; 565 completed surveys were returned and seven surveys were excluded due to inability to verify that the respondents were physicians (a 70% net valid response rate). Alerting threshold preferences were identified for eight laboratory tests. For example, the majority of respondents selected thresholds of ≥5.5 mEq/L for hyperkalemia (63%) and ≤3.5 without symptoms for hypokalemia (54%). The majority of surveyed physicians thought alerts should include the complete active medication list, current vital signs, previous value of the triggering lab, medication change in the past 30 days, and medication allergies. Most surveyed physicians felt the best way to communicate an ADE alert was by direct phone/voice communication (64%), followed by email to a mobile device (59%).

CONCLUSIONS

This survey of nursing home physicians suggests that the majority prefer alerting thresholds that would generally lead to fewer alerts than if widely accepted standardized laboratory ranges were used. It also suggests a subset of information items to include in alerts, and the physicians' preferred communication modalities. This information might improve the acceptance of clinical event monitoring systems to detect ADEs in the nursing home setting.

Big data and clinicians: a review on the state of the science

BACKGROUND

In the past few decades, medically related data collection saw a huge increase, referred to as big data. These huge datasets bring challenges in storage, processing, and analysis. In clinical medicine, big data is expected to play an important role in identifying causality of patient symptoms, in predicting hazards of disease incidence or reoccurrence, and in improving primary-care quality.

OBJECTIVE

The objective of this review was to provide an overview of the features of clinical big data, describe a few commonly employed computational algorithms, statistical methods, and software toolkits for data manipulation and analysis, and discuss the challenges and limitations in this realm.

METHODS

We conducted a literature review to identify studies on big data in medicine, especially clinical medicine. We used different combinations of keywords to search PubMed, Science Direct, Web of Knowledge, and Google Scholar for literature of interest from the past 10 years.

RESULTS

This paper reviewed studies that analyzed clinical big data and discussed issues related to storage and analysis of this type of data.

CONCLUSIONS

Big data is becoming a common feature of biological and clinical studies. Researchers who use clinical big data face multiple challenges, and the data itself has limitations. It is imperative that methodologies for data analysis keep pace with our ability to collect and store data.

Validity of a clinical decision rule-based alert system for drug dose adjustment in patients with renal failure intended to improve pharmacists' analysis of medication orders in hospitals

OBJECTIVE

The main objective of this study was to assess the diagnostic performances of an alert system integrated into the CPOE/EMR system for renally cleared drug dosing control. The generated alerts were compared with the daily routine practice of pharmacists as part of the analysis of medication orders.

MATERIALS AND METHODS

The pharmacists performed their analysis of medication orders as usual and were not aware of the alert system interventions that were not displayed for the purpose of the study neither to the physician nor to the pharmacist but kept with associate recommendations in a log file. A senior pharmacist analyzed the results of medication order analysis with and without the alert system. The unit of analysis was the drug prescription line. The primary study endpoints were the detection of drug dose prescription errors and inter-rater reliability (Kappa coefficient) between the alert system and the pharmacists in the detection of drug dose error.

RESULTS

The alert system fired alerts in 8.41% (421/5006) of cases: 5.65% (283/5006) "exceeds max daily dose" alerts and 2.76% (138/5006) "under-dose" alerts. The alert system and the pharmacists showed a relatively poor concordance: 0.106 (CI 95% [0.068-0.144]). According to the senior pharmacist review, the alert system fired more appropriate alerts than pharmacists, and made fewer errors than pharmacists in analyzing drug dose prescriptions: 143 for the alert system and 261 for the pharmacists. Unlike the alert system, most diagnostic errors made by the pharmacists were 'false negatives'. The pharmacists were not able to analyze a significant number (2097; 25.42%) of drug prescription lines because understaffing.

CONCLUSION

This study strongly suggests that an alert system would be complementary to the pharmacists' activity and contribute to drug prescription safety.