This study examines the results of therapeutic interventions performed by pharmacists at the Veterans Affairs Pittsburgh Healthcare System and their relevance to electronic medical records.
Published Online: April 15, 2014
Megan E. Sartore, PharmD; Kimberly M. Ehman, PharmD; and Chester B. Good, MD, MPH
Objectives: In this study we sought to defi ne, describe, and categorize the nature of pharmacy-initiated therapeutic interventions performed in a comprehensive VA healthcare system containing acute inpatient care, specialty and primary care outpatient clinics, surgeries, long-term care, and community-based outpatient clinics. Previous investigators have explored the effects of pharmacy interventions but are either dated or limited in scope. This article provides an update on types of pharmacy interventions in the era of electronic order entry, as well as the magnitude and number of interventions in a comprehensive VA healthcare system.
Study Design and Methods: This is an observational, descriptive study of existing monthly aggregate Drug Use Evaluation “Therapeutic Intervention” reports over 1 year’s time (September 1, 2009, through August 31, 2010) at the VA Pittsburgh Healthcare system. These reports quantify the type, number, and signifi cance of each pharmacist initiated intervention made during the study period, including “notable” or “signifi cant” interventions. “Significant” is defined as an action by the reporting pharmacist that potentially prevents serious harm and possibly death from occurring and is documented as a “close call” in our computer system.
Results: Out of 1,488,869 prescriptions documented between September 2009 and August 2010, 9649 interventions (0.65%) were reported. Errors inherent with electronic order entry include: wrong drug, wrong patient, and those related to alert fatigue (ie, overriding drug allergies, interactions). Wrong drug and wrong patient accounted for 2.5% of total interventions while interventions related to wrong dose and interactions were 5.3% and 4.5% of interventions, respectively.
Conclusions: We believe this study will add substantially to the medical literature by presenting new information about the types of interventions seen relating to an electronic medical record system.
Am J Pharm Benefits. 2014;6(2):e24-e30
Our study highlights the importance of pharmacists in improving the fidelity of the prescription process and improving patient safety and prescribing efficiency by performing pharmacy interventions. Despite potential benefits of electronic ordering in the context of an electronic medical record, we found that pharmacists continue to play a critical role in the review of prescriptions. These reports should enhance understanding of vulnerabilities in the medication ordering process and document the ongoing value of pharmacy interventions at the point of prescribing.
According to the American Society of Health-System Pharmacists (ASHP), the mission of a pharmacist is to provide pharmaceutical care, which is defined as “the direct, responsible provision of medication-related care for the purpose of achieving definite outcomes that improve a patient’s quality of life.”1 In the provision of pharmaceutical care, pharmacists use their unique perspective and knowledge of medication therapy to evaluate patients’ actual and potential medication-related problems.1 One of the main duties of a pharmacist is to review prescriptions and, when indicated, suggest a therapeutic intervention to improve the safety, efficacy, or cost-effectiveness of medications.
Pharmacy-initiated interventions address a wide array of potential medication-related issues. Most interventions directly involve patient safety concerns, although some interventions may be more administrative (ie, order clarification) or related to cost (ie, encouraging the use of formulary items or less costly alternatives). These interventions illustrate the significance of the pharmacist in every aspect of patient care.
In this article, we seek to define, describe, and categorize the nature of pharmacy interventions performed in a large teaching healthcare system. Previous investigators have explored the effects of pharmacy interventions, but their studies either predate the electronic medical record (EMR) era or have a narrower study population, at least at the time this study was conducted. Reports have focused on community pharmacies,2 acute care inpatient pharmacies,3 a specific type of teaching hospital,4 or documented economic effects of clinical pharmacy interventions.5 Our study provides an update on types of pharmacy interventions, given the unfolding era of electronic order entry (EOE), as well as assess the magnitude and number of interventions over a 1-year period at a large, comprehensive healthcare system—the Veterans Affairs Pittsburgh Healthcare System (VAPHS), containing acute inpatient care, specialty- and primary-care outpatient clinics, long-term care, community-based outpatient clinics, and a comprehensive surgical program including transplantation.
Reporting Pharmacy Interventions
This study was approved by the VAPHS Institutional Review Board. During the study period, VAPHS consisted of 3 divisions, including an acute care facility, an intermediate care facility, and an acute and intermediate psychiatric care facility. As an affiliate of the University of Pittsburgh, many trainees participate in the inpatient and outpatient settings. In the inpatient and outpatient pharmacies at all 3 divisions of VAPHS, pharmacists are encouraged to record all interventions using a standardized tracking form. These interventions are categorized according to the type of intervention, and reported each month to the Drug Utilization Evaluation (DUE) subcommittee (Table 1). A significant intervention was defi ned as a pharmacist intervention that potentially prevented serious harm and possibly death from occurring. Typically only those interventions which were accepted by the clinician were recorded.
We reviewed de-identified information from pharmacy intervention reports that were presented at the monthly DUE meetings for 1 year (September 2009 through August 2010). We condensed the categories of interventions from Table 1 into 3 broad categories: (1) patient safety, defi ned as any intervention which directly affects the well-being of the patient and may prevent harm (eg, wrong dose prescribed, potential allergy, drug contraindicated/precaution, duplicate therapy, drug level monitoring/pharmacokinetics, drug-drug/disease/ diet/test interaction, more appropriate dosing schedule, discontinue therapy, wrong drug prescribed, more appropriate dose, and adverse drug reaction); (2) administrative/ documentation, defi ned as any intervention requiring further documentation or a failure to adhere to hospital policy (eg, order clarifi cation, drug indicated but not prescribed, no indication for drug, and drug information; and (3) cost avoidance, defi ned as any intervention recommending a more cost-effective therapy or more effi cient medication administration schedule (eg, wrong length of therapy, change from parenteral to oral therapy, alternative route recommended, change to a formulary agent/nonrestricted drug, and alternative therapy recommended). The rate of interventions (by category) was calculated as a proportion of the overall interventions, and by total number of prescriptions entered during the 1-year time frame.
VAPHS pharmacists processed 1,488,869 prescriptions (1,165,163 outpatient, 323,706 inpatient) from September 2009 to August 2010. Of these, 21,481 inpatient prescriptions were from the acute/intermediate psychiatric site, 41,440 were from the intermediate care site, and 260,785 were from the acute care site. In all, 9649 interventions were made (0.65% of total prescriptions). Home-based primary care (HBPC) was responsible for 1294 interventions, the acute and intermediate psychiatric care facility 848 interventions, the intermediate care facility 2472 interventions, the acute care inpatient facility 3528 interventions, and the acute care outpatient facility 1507 interventions. From the 9649 interventions, 1535 (15.9%) were categorized as signifi cant or 0.1% of all prescriptions.
The patient safety category had the largest number of interventions, followed by the administrative and cost avoidance categories (Table 2). Within the patient safety subcategories, drug level monitoring/pharmacokinetics, more appropriate dose, and duplicate therapy were the 3 most common interventions to occur. Combined, these 3 intervention types accounted for 54.7% of all patient safety interventions and 31.6% of all interventions. Order clarification was the most common intervention for administrative reasons, while change to a formulary or nonrestricted drug was the most common type of intervention related to cost avoidance.
Within the 1535 significant interventions made, prevention of a critical interaction with warfarin was most frequent, accounting for 47% of the significant interventions made. Identification of a contraindicated drug (for example, based on renal failure or other clinical conditions) was the second-most common and accounted for 20% of significant interventions. Other significant interventions included identification of previous adverse events (AEs) caused by prescribed medication (including several cases of prior anaphylaxis), wrong drug prescribed (look-alike, sound-alike), and wrong patient (Table 3).
To our knowledge, this is the first comprehensive report to evaluate pharmacy interventions for medications in an integrated healthcare system, including inpatient, outpatient, long-term care, and acute care, in the age of EMRs. Our report illustrates the importance of the pharmacy review of medication orders, which can potentially prevent adverse drug events and help highlight costsaving measures and formulary management strategies. While other investigators have highlighted problems with computerized EOE systems,6-8 we focused on broad categories of interventions, some that are relevant to EOE, as well as others that are not.
The rate of pharmacy interventions we report is similar to that reported in other studies. Hawksworth reported 1503 interventions in 201,000 prescriptions (0.75%).2 The PROMISe (Pharmacy Recording of Medications Incidents and Services) project performed in Australia examined 2396 interventions in 435,520 prescriptions over 8 weeks in 52 community pharmacies, for a rate of 0.55%.9 Dodd reviewed interventions in a hospital setting with acute and specialty services, of which 6.3% to 6.8% of interventions were classified as major.10 Hawkey identified 182 interventions in 6170 prescriptions (2.9%) in 2 teaching hospitals providing both inpatient and outpatient services.11
Others have investigated the benefits and weaknesses of EOE in reducing medication errors. Bates reported a 55% decrease in medication errors within 1 hospital that started using EOE; most of the errors had been originating during the ordering phase.6 Bobb reported that 65% of prescribing errors were likely preventable with EOE.7 Schnipper reported that an electronic medication reconciliation application resulted in a “28% relative risk reduction in unintentional medication discrepancies with potential for harm.”8
Although EOE may offer improvements in the safety of prescribing, others have identified potential weaknesses— most notably, communication issues when entering orders and the ability to customize orders in some fashion. As a result of EOE, Singh found that 0.95% of prescriptions had an error related to communication/ordering problems.12 Koppel identified 22 scenarios in which prescribing errors were likely increased because of electronic prescribing.13 Others note that most electronic alerts regarding drug allergies and potential interactions are ignored.14 Thus, EOE provides a new set of unique potential errors or problems in the ordering of medications.
PDF is available on the last page.