This site is intended for health professionals only
LaDonna S Hale
Department of Physician Assistant
Wichita State University
Accurate, timely and complete medication histories are a crucial part of the initial patient assessment upon admission into hospital. Knowing the patient’s medication history is essential in decisionmaking as well as continuation of appropriate medications through the hospital stay.
Inaccuracies in the medication history cost considerable time and effort and may jeopardise patient safety.(1,2) As many as 60% of medication errors occur during admission, discharge or transfer between units.(1) A streamlined process provided by trained personnel, such as pharmacists, helps to eliminate duplications of time and effort and improves patient safety.(1,3)
The following case scenario represents the frustrations, wasted time and potential for medication errors arising from ambiguous medication orders written from incomplete histories. A 62-year-old female is admitted for elective hip surgery. Her postoperative medications include: isosorbide 20mg twice daily, budesonide inhaler, orders for both captopril and the captopril/hydrochlorothiazide combination product, and several other medications. The surgeon writing these orders is not her usual physician and relied on the medication history from the chart. The nurse obtaining this history was pressed for time due to numerous other patient care duties and relied solely on information from the patient and the patient’s family. The pharmacist is not able to fill these orders as written. First, the isosorbide must be clarified as mononitrate or dinitrate, sustained release or regular release, with the exact timing of the twice-daily dosing. Corticosteroid inhalers should be taken on a scheduled basis for maximal effectiveness. However, hers is ordered “as needed” because “that is how she takes it at home”. The pharmacist must also clarify whether the patient is actually taking both captopril products or whether one should have been stopped when starting the other. The patient also states allergies to penicillin and “cold medications” with no specific medications listed or description of the type of reactions. These clarifications will require multiple communiqués between pharmacy, nursing, the patient and physician to clarify these discrepancies.
Although performing medication histories is an acceptable clinical pharmacy service, only about 3% of US hospitals do so.(3) In the USA it is estimated that by providing this service a hospital may reduce expenditures by US$7m ($1=e. 1) and may prevent 128 deaths annually.(2,4) Pharmacists are especially suited to obtaining medication histories. Studies indicate that pharmacists obtain more accurate medication information than do physicians.(5,6) Since pharmacists are most familiar with the thousands of medication names, dosage forms and dosing regimens, they are able to ask appropriate follow-up questions necessary to clarify inconsistencies between what the patient is reporting and what “makes sense”. As the pharmacist is the person ultimately responsible for filling these orders, he/she is keenly aware of the complete information necessary to accomplish this safely.
Study summary and results
The results of a study conducted at our hospital was recently published in the American Journal of Health-System Pharmacy.(7) We compared time efficiency and clinical impact of pharmacist-obtained medication histories to our usual practice of nurse-obtained histories on 100 adult, nonintensive care admissions. When pharmacists conducted the initial medication history, they were able to identify and correct more discrepancies between the patient’s reported home medications and the hospital admission orders. Pharmacists also intervened on a higher percentage of patients when allowed to conduct that initial medication history (34% vs 16%, p<0.001), with a higher number of mean clinical interventions per patient (0.60±1.07 vs 0.22±0.55, p=0.27). These interventions included identifying route, dose and allergy discrepancies, missing medications and incomplete orders, and avoiding therapeutic duplications and drug interactions. Pharmacists also identified more patients taking nonprescription and herbal medications (98% vs 70%, p<0.001), with a higher mean number identified per patient (5.12±3.14 vs 1.52±1.16, p<0.001). Even if nonprescription products are not continued during hospitalisation or are only used occasionally, they can have significant disease interactions, drug interactions and side-effects that are important in adequately assessing a patient’s health.(8) Complete allergy information is vital to patient safety. In our study, pharmacists identified four of their 50 patients as having incorrect allergy information documented in the chart (4 vs 0, p<0.05). Using pharmacists to interview patients also expedited the entry of allergy information in the hospital’s computer system and prevented delays in drug dispensing. (7)
As in the previous case scenario, when the physician is unfamiliar with the home regimen, an incomplete history may result in missed medications during hospitalisation. This becomes an issue if the inadvertently discontinued medication causes withdrawal symptoms, is required for maintenance therapy or the patient has a prolonged hospital stay. For example, in our study, we identified a patient missing his maintenance epilepsy medications. It is difficult to fully evaluate the patient’s health without a complete medication picture. The patient may be experiencing a side-effect of a medication the medical staff is unaware of, or a new drug may be added that interacts with a drug that will be resumed after discharge.
A major concern with implementing such a programme is the impact it has on pharmacist and nursing time. During our study, pharmacists documented the amount of time required to conduct the history and clarify and process ambiguous admission orders. Similar nursing time surveys were conducted at a later date on 15 randomly selected patients. These data suggest that pharmacist time spent upfront, performing the medication history (13.4±6.7 minutes), reduces the time required during order-entry. Pharmacists spent 6.0±6.5 minutes clarifying orders in the pharmacist- obtained history group, vs 14.4±9.9 minutes when nursing obtained the history. This information is provided as descriptive data only as no statistical analysis was performed and the two time surveys were conducted at different times.(7)
Before implementing a pharmacy medication history service, support must be sought from management, medical staff, nursing and pharmacy. Any redistribution of duties has potential territorial issues. Focusing the goal of this service on relieving nurses to perform direct patient care and improving efficiency for both departments proved successful. We encountered no resistance from physicians or nursing. On the contrary, both strongly supported the prospect of pharmacists taking on this responsibility and recognised their expertise in conducting medication interviews. Physicians recognised the positive effect on patient safety and also admitted frustration with the current process requiring frequent clarifications.
A standardised form should be developed, including questions about prescriptions, nonprescriptions, herbals, tobacco, alcohol and caffeine use, allergy and adverse reaction history, immunisation history and some past medical history (see Figure 1). This is an opportune time for pharmacists to reinforce proper administration instructions and warnings, appropriate selection of OTC products and immunisations, and to check for interactions. Since a common complaint among physicians obtaining patient histories is frequent interruptions, a workflow designed without disturbances is an important implementation goal.(9)
For most institutions, implementing a partial service may be more feasible than a 24-hour hospitalwide service. Several models could be explored. A pharmacist may be located in the emergency room or admissions area. Units with decentralised pharmacy services could add medication histories to the other clinical services already offered. Targeting medically complicated patients makes the service even more feasible from a time-management and patient safety standpoint.
Although the current process was deemed labour- intensive, redundant and fraught with potential medical errors, our hospital has not yet instituted this service. We first attempted it in 2001, but personnel and budget reductions halted our efforts. We continue to face similar restraints. Some of the logistic and budgetary hurdles we face may explain why many other hospital pharmacies cannot spare pharmacists in this role. Further studies investigating the economic impact of this service may be required to convince management and budget persons of the efficiency, cost-effectiveness and patient safety benefits of pharmacist-obtained medication histories.