EJINME-02935; No of Pages 8 European Journal of Internal Medicine xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
European Journal of Internal Medicine journal homepage: www.elsevier.com/locate/ejim
Original Article
Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist Bertrand Guignard a,b,⁎, Pascal Bonnabry a,c, Arnaud Perrier d, Pierre Dayer b,c, Jules Desmeules b,c, Caroline Flora Samer b,e a
Pharmacy, Geneva University Hospitals, Geneva, Switzerland Service of Clinical Pharmacology and Toxicology, Geneva University Hospitals, Geneva, Switzerland School of Pharmaceutical Sciences, University of Geneva, University of Lausanne, Geneva, Switzerland d Service of General Internal Medicine, Geneva University Hospitals, Geneva, Switzerland e Swiss Centre for Applied Human Toxicology, University of Geneva, Geneva, Switzerland b c
a r t i c l e
i n f o
Article history: Received 27 December 2014 Received in revised form 18 May 2015 Accepted 19 May 2015 Available online xxxx Keywords: Drug-related problems Internal medicine Hospital
a b s t r a c t Background: Patients admitted to general internal medicine wards might receive a large number of drugs and be at risk for drug-related problems (DRPs) associated with increased morbidity and mortality. This study aimed to detect suboptimal drug use in internal medicine by a pharmacotherapy evaluation, to suggest treatment optimizations and to assess the acceptance and satisfaction of the prescribers. Methods: This was a 6-month prospective study conducted in two internal medicine wards. Physician rounds were attended by a pharmacist and a pharmacologist. An assessment grid was used to detect the DRPs in electronic prescriptions 24 h in advance. One of the following interventions was selected, depending on the relevance and complexity of the DRPs: no intervention, verbal advice of treatment optimization, or written consultation. The acceptance rate and satisfaction of prescribers were measured. Results: In total, 145 patients were included, and 383 DRPs were identified (mean: 2.6 DRPs per patient). The most frequent DRPs were drug interactions (21%), untreated indications (18%), overdosages (16%) and drugs used without a valid indication (10%). The drugs or drug classes most frequently involved were tramadol, antidepressants, acenocoumarol, calcium–vitamin D, statins, aspirin, proton pump inhibitors and paracetamol. The following interventions were selected: no intervention (51%), verbal advice of treatment optimization (42%), and written consultation (7%). The acceptance rate of prescribers was 84% and their satisfaction was high. Conclusion: Pharmacotherapy expertise during medical rounds was useful and well accepted by prescribers. Because of the modest allocation of pharmacists and pharmacologists in Swiss hospitals, complementary strategies would be required. © 2015 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
1. Introduction Pharmacotherapy is becoming increasingly complex, and inappropriate drug prescription might be associated with increased healthcare costs and hospital admissions as well as prolonged hospital stays, reduced quality of life, and increased morbidity or mortality [1–4]. A drug-related problem (DRP) is defined as an event or circumstance involving drug therapy that actually or potentially interferes with the desired health outcomes [5]. The majority of DRPs are predictable
⁎ Corresponding author at: Pharmacy, Geneva University Hospitals, Rue GabriellePerret-Gentil 4, CH-1211 Geneva 14, Switzerland. Tel.: +41 22 382 3963; fax: +41 22 382 3990. E-mail address: [email protected] (B. Guignard).
and potentially avoidable [6]. DRP frequency might be reduced by pharmacotherapy optimization, such as medication reviews led by pharmacists or clinical pharmacologists [7–9]. The positive effects of these medication reviews on costs were reported [10,11]. Moreover, these interventions were associated with reduced durations of hospital stay or decreased frequency of re-admissions as well as with better control of certain biomarkers (lipid levels, anticoagulation levels or blood pressure) and disease events such as decompensated heart failure or thromboembolic events [12]. Inappropriate drug use has been largely studied in the elderly, a population characterized by frailty, polymorbidity and polymedication; fewer studies have addressed the question in internal medicine wards, in which younger patients are admitted. It has been shown that polymorbidity and polymedication were independent risk factors of DRPs, whereas age and gender were not [13,14]. Elderly and middle-
http://dx.doi.org/10.1016/j.ejim.2015.05.012 0953-6205/© 2015 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
2
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
aged patients admitted to internal medicine wards suffer from multiple diseases or have several risk factors and receive a large number of drugs. They are thus at high risk for DRPs. In a French study, the in-hospital incidence rate of adverse drug reactions (ADRs) in internal medicine was 10.1 per 1000 patient-days, and 80% of the ADRs were considered preventable [15]. Our study was conducted in two internal medicine wards at the Geneva University Hospitals, a 2000-bed university health center in Switzerland. The Geneva University Hospitals have the following two complementary information centers for caregivers regarding medication use: the Clinical Pharmacology and Toxicology medical service for pharmacotherapeutic questions (such as drug–drug interactions, ADRs, selection of drugs and dosage, use of drugs in pregnancy, extreme age populations or organ dysfunction) and a Pharmacy hotline for pharmacotechnical questions (such as intravenous drug dilutions, intravenous drugs compatibilities, or tablet crushability for administration through a nasogastric feeding tube). This study is a joint collaboration between these two entities, through a clinical pharmacist and a clinical pharmacologist tandem, working together, addressing the question of DRPs in internal medicine by screening medical records, attending medical rounds, and providing advice for pharmacotherapy optimization. Whereas most studies evaluating DRPs are retrospective, our study offers the advantage of being prospective. We sought to optimize the impact of our intervention by prioritizing DRPs to be reported to prescribers during rounds. The aims of the study were as follows: (1) to detect all of the DRPs in the included patients and to identify the drugs or drug classes most frequently causing the problems; (2) to assess which intervention (none, verbal advice and written consultation) was required for each DRP according to its clinical relevance or complexity; (3) to measure the acceptance rate of the prescribers and the actual impact on the prescription; and (4) to assess the satisfaction of the prescribers regarding the suggestions for treatment optimization. 2. Methods 2.1. Patients and design This prospective interventional study was conducted during a consecutive 6-month period. From September 2011 to February 2012, a clinical pharmacist and a clinical pharmacologist (i.e. a physician specializing in clinical pharmacology and internal medicine) both attended medical rounds approximately once a week on two internal medicine wards. All the patients admitted in the visited internal medicine wards were considered eligible and there were no exclusion criteria. The following data and comorbidities as well as the known risk factors for DRPs were systematically recorded for each patient: sex, age, and the number of drugs prescribed; the occurrence of hypertension, coronary heart disease, heart failure, atrial fibrillation, diabetes, and reduced renal function (a creatinine clearance ≤50 ml/min according to the Cockroft–Gault formula); previously reported liver dysfunction; chronic obstructive pulmonary disease or asthma; a previous history of a transient ischemic attack or stroke, malignant disease, alcohol abuse, tobacco use; and polypharmacy (≥5 drugs).
pharmacist screened the medical records of the patients on the ward and performed a structured medication review with an assessment grid (Table 1) to detect the DRPs. The assessment grid was developed by compiling the clinical decision supports classically used for medication reviews (e.g., drug interaction screening tools [17–19], drug databases [20–22] or textbooks [23–26] and published tools for the medication review [27]). All of the identified DRPs were discussed between the clinical pharmacist and the clinical pharmacologist before the medical round to determine their clinical relevance and to prioritize interventions. DRPs might be dependent on each other, and only primary DRPs are presented in the results. An unwanted drug interaction might require a dose reduction (overdosage is a secondary DRP) or an ADR might render a drug inappropriate for a patient (improper drug selection is a secondary DRP). Because drug interactions involve, by definition, at least two drugs, only the drug/drug class causing the DRP and associated with the highest patient risk (because of the efficacy/toxicity ratio modulation) is presented in the results. 2.3. Decision on the intervention Based on the potential clinical relevance and complexity of the detected DRPs, an attitude was determined, as follows: (1) no intervention and/or continuation of usual follow-up by caregivers, (2) verbal advice to the physicians and/or nurses during medical rounds (treatment optimization or recommendation to introduce monitoring) or (3) verbal advice during medical rounds followed by a written specialized clinical pharmacology consultation. No intervention was conducted for non-clinically relevant DRPs or when appropriate monitoring was already provided (e.g., monitoring of blood pressure, heart rate, kalemia and serum creatinine). Clinical relevance of DRPs has previously been defined by Dooley et al. and Blix et al. DRPs were thus considered non-clinically relevant if they were non major, i.e. the chance of noticed effect was lower than 20%, the chance of harmful effect was lower than 5%, or they would not require intervention due to the lack of detrimental effects [13,28]. Written specialized consultations were reserved for complex clinical pharmacological situations, when specifically requested by prescribers to induce a change in current clinical practice, or when ADRs had to be reported to the Swiss national pharmacovigilance center (Swissmedic). 2.4. Acceptance and application rate by prescribers The acceptance rates of treatment optimization recommendations made to prescribers during the medical round were measured. In case of recommendation acceptance by the physicians during medical rounds, the prescription medical records were systematically rescreened after five days to assess whether the treatments had actually been changed according to the advice (application rate). 2.5. Prescribers' satisfaction survey The satisfaction level of prescribers during medical rounds was rated with an anonymous questionnaire assessing the following factors:
2.2. Detection of DRPs and the drugs or drug classes involved The actual and potential DRPs were classified into the following seven categories, as previously described by Hepler and Strand when they first described the process of pharmaceutical care: drug interaction, subtherapeutic dosage, overdosage, drug use without indication, untreated indication, improper drug selection, and adverse drug reaction [16]. The treatment adherence problems were not addressed in the study. The day before the medical round, a clinical
(1) The physician education level (a categorical choice); (2) The global opinion on the participation of the pharmacotherapy experts in medical round: general usefulness of pharmacotherapy experts, time for verbal advices, and optimal frequency of attending (a categorical choice); (3) The usefulness of the advice for preventing specific DRPs (a fourpoint Likert scale). An average score was attributed to each type of DRP by attributing different points according to the level of
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
3
Table 1 Assessment grid to detect the drug-related problems. Drug-related problems
Detection criteria
Used tools and references
Drug interaction
- Interaction described by Lexi Interact OR - Pharmacokinetic interaction between a substrate and an inhibitor/inducer of a common metabolic/elimination pathway (cytochrome P450 or P-glycoprotein) OR - Pharmacodynamic interaction between QT prolongating drugs or between serotoninergic drugs AND - Interaction described by Drugdex as associated with a risk of a decreased efficacy and/or increased toxicity Subtherapeutic - Dosage lower than the minimal recommended dosage for a given indication (published in the Swiss drug Compendium or in other dosage references, for drugs unlicensed in Switzerland) OR - Improved renal function with drugs whose dosage has been previously adjusted to renal failure and becoming underdosed with renal function improvement. Overdosage - Dosage higher than the maximal recommended dosage for a given indication (published in the Swiss drug Compendium or in other references, for drugs unlicensed in Switzerland) OR - Drug dosage not adjusted to renal function in renal impairment (creatinine clearance estimated with the Cockroft–Gault formula), following recommendations from Swiss drug Compendium or other references OR - Hepatic impairment reported in medical record and prescription of drugs whose dosage has to be adjusted to hepatic impairment, following recommendations from Swiss drug Compendium or other references. - Drugs belonging to the same group of Anatomical Therapeutic Chemical classification level 4 or 5 (therapeutic duplication) OR Drug use - Drug with no clear indication in the medical record OR without - Criterion in Screening Tool of Older Persons' Prescriptions (STOPP) OR indication - One of the following prescribing cascade: - Nonsteroidal anti-inflammatory drug + antihypertensive drug - Thiazide + gout treatment - Neuroleptics + antiparkinsonian drug - Metoclopramide + antiparkinsonian drug - Cholinesterase inhibitor + anticholinergic drug - Anticholinergic drug + treatment of benign prostate hyperplasia - Angiotensin-converting enzyme inhibitor + antitussive - Antihypertensive drug + vestibular system suppressor - Opioid + vestibular system suppressor - Sedative + vestibular system suppressor Untreated - Criterion in Screening Tool to Alert doctors to Right Treatment (START) OR indication - Drugs recommended for prophylaxis but not prescribed: - Laxative and opioids - Calcium and vitamin D in elderly - Prophylactic anticoagulation after surgery or venous thromboembolism risk factors - Vitamin B6 and isoniazid Improper drug - Drug contraindicated according to Swiss drug Compendium OR selection - More effective or safer alternative OR - Less expensive alternative (with equivalent effectiveness and safety) Adverse drug - ADR suspected by the physician and reported in the medical record OR reaction (ADR) - Correlation between one of the following symptoms and ADR reported in the reference literature (Swiss drug Compendium, Drugdex or Meyler's Sides Effects of Drugs): - Abnormal liver biochemical tests - Hematologic abnormalities (neutropenia, agranulocytosis, thrombopenia, hemolytic anemia) - Hemorrhage - Cutaneous rash - Anaphylaxis - Acute renal failure - Digestive symptoms (dysphagia, dyspepsia, constipation, diarrhea) - Cough - Hypoglycemia or hyperglycemia - Neuromuscular symptoms (vertigo, seizure, muscle weakness/pain, extrapyramidal symptoms) - Cardiovascular toxicity (arrhythmia, heart failure, hypotension, hypertension) Others - Parenteral drug prescription when an oral formulation is available in a hemodynamically stable patient, able to swallow oral medications and with normal gastrointestinal tract function: - Intravenous antibiotic and antifungal agents with high bioavailability (e.g., fluoroquinolones, macrolides, cotrimoxazole, metronidazole, rifampin, azoles) - Other intravenous antibiotic agents after clinical improvement ≥ 48 h of intravenous treatment and apyrexia ≥ 24 h - Intravenous paracetamol - Intramuscular vitamin B12 for long term replacement - Inappropriate intravenous drug reconstitution, dilution or administration flow according to Swiss drug Compendium, Lexi Drug and AHFS - Intravenous drugs incompatibilities described in Trissel's Handbook on Injectable Drugs - Prescription of medicines that cannot be crushed (e.g., sustained release tablets, gastroresistant tablets) for a patient with a nasogastric feeding tube - Suboptimal drug administration timing (e.g., regarding to meals, regarding to circadian rhythm) according to Swiss drug Compendium, Lexi Drug, and AHFS.
usefulness (useless: 1 point, average usefulness: 2 points, useful: 3 points, very useful: 4 points) and by dividing the total score by the total number of opinions; (4) The types of DRPs for which the physicians wanted to acquire new skills (multiple choice).
[17–19,22]
[20–22]
[20–23]
[27]
[27]
[20]
[20–22,24]
[20,21,25,26]
2.6. Data analysis The data analysis was descriptive. The categorical variables were expressed as the percentage. The continuous data were reported as the mean ± standard deviation (SD).
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
4
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
Table 2 Characteristics of the patients included in the study with frequency of chronic conditions. Male (n) Age (mean ± SD), range Number of drugs prescribed (mean ± SD), range
75 (52%) 68 ± 16, 21–99 10.6 ± 4.0, 0–21
Chronic conditions Hypertension (n) Coronary heart disease (n) Heart failure (n) Atrial fibrillation (n) Diabetes mellitus (n) Reduced renal functiona (n) Previously reported liver dysfunction (n) Chronic obstructive pulmonary disease or asthma (n) Previous history of transient ischemic attack or stroke (n) Previous history of malignant disease (n) Previous or current alcohol abuse (n) Previous or current tobacco use (n) Polypharmacy (≥5 drugs) (n)
66 (46%) 30 (21%) 37 (26%) 42 (29%) 38 (26%) 27 (19%) 8 (6%) 33 (23%) 11 (8%) 30 (21%) 28 (19%) 22 (15%) 136 (94%)
a
Creatinine clearance ≤ 50 ml/min according to Cockroft–Gault formula.
Table 3 Frequency and types of detected drug-related problems and number of patients exposed. Drug-related problems (DRPs)
Number (frequency)
Number of patients exposed (prevalence)
Drug interaction Untreated indication Overdosage No valid indication Adverse drug reaction Subtherapeutic dosage Improper drug selection Others All DRPs
82 (21%) 69 (18%) 60 (16%) 39 (10%) 36 (9%) 31 (8%) 29 (8%) 37 (10%) 383 (100%)
58 (40%) 48 (33%) 52 (36%) 33 (23%) 29 (20%) 25 (17%) 21 (14%) 31 (21%) 117 (81%)
3. Results 3.1. Patient characteristics The clinical pharmacist and the clinical pharmacologist attended 16 medical rounds and met 22 prescribers. A total of 145 patients were included, and 52% were male, with a mean age of 68 years (range 21–99 years, SD 16 years). The mean time for the screening of medical records and for the structured medication review was 1 h per patient. A total of 1523 prescriptions were recorded, with a mean of 10.6 prescriptions per patient (range 0–21, SD 4.0). The characteristics of the patients are detailed in Table 2. 3.2. DRPs and the drugs or drug classes involved A total of 383 DRPs were detected and had not been previously identified by the caregivers in charge (e.g. physicians, nurses), which
was a mean of 2.7 DRPs per patient (0–12, SD 2.4). At least one DRP was detected in 81% of the assessed medical records. The most frequently detected DRPs were drug interaction (21%), untreated indication (18%) and overdosage (16%), including drug dosage not adjusted to the renal function (Table 3). This distribution concerned all DRPs, independently of their clinical relevance or the decision to report them to prescribers or not. The prevalence of patients exposed to each type of DRP is also reported in Table 3. The drugs or drug classes involved in each type of DRP are described in Table 4. The drugs or drug classes most frequently involved in unwanted drug interactions were substrates of cytochrome P450 (CYP) isoenzymes (predominantly CYP 2D6, 2C9, 2C19 and 3A4) and/or, more rarely, the drug transporter, P-glycoprotein (P-gp). Drug interactions concerned mostly tramadol (11%), antidepressants (11%), anticoagulants (9%) and beta-blockers (9%). Interactions involving tramadol were either associated with a risk of decreased opioid therapeutic effect (4/9 interactions) by CYP 2D6 inhibition (e.g. by antidepressants), or increased toxicity (6/9 interactions) by CYP 3A4 inhibition (e.g. by clarithromycine, diltiazem) and/or a pharmacodynamic interaction through cumulation of serotoninergic effects with other serotoninergic drugs (e.g. antidepressants, antipsychotics). Risk of decreased opioid effect and increased toxicity was concomitant in one interaction with tramadol (i.e. through 2D6 and 3A4 inhibition by amiodarone). Antidepressants are substrates of CYP and their safety/efficacy ratio is modulated by pharmacokinetic interactions via CYP inhibitors (increased toxicity) or CYP inducers (reduced efficacy). Their toxicity might also increase via pharmacodynamic interactions with serotoninergic drugs or drugs known to prolong QT interval. Vitamin K antagonists are substrates of CYP, whereas direct thrombin or Xa inhibitors are substrates of CYP and/or P-gp. Inhibitors or inducers of these elimination pathways/transporters could lead to increased toxicity or decreased efficacy, respectively. The most frequent drugs or drug classes lacking in untreated indications were (a) calcium and vitamin D in elderly patients with osteoporosis and/or in those patients receiving or having received long-term systemic corticosteroids, associated with an increased risk of fracture, (b) statins and/or aspirin in ischemic heart disease patients, (c) angiotensin-converting enzyme (ACE) inhibitors, angiotensin II receptor blockers or beta-blockers withdrawn in decompensated heart failure patients and not reintroduced after resolution of the acute episode, and (d) laxatives in patients receiving opioids. Drugs or drug classes most frequently involved in overdosages were proton pump inhibitors (e.g. esomeprazole prescribed at 40 mg daily for peptic ulcer prophylaxis or for the treatment of gastroesophageal reflux) and paracetamol (dose of/or above 4 g per day despite alcoholism, cachexia or coprescription of enzymatic inducers), as well as drugs whose dosage was not adjusted for the occurrence of renal failure (e.g., enoxaparin, digoxin, ciprofloxacin and pregabalin). In contrast, there were drugs at a subtherapeutic dosage, like beta-blockers and ACE inhibitors for heart failure, which were not in their target dose, although blood pressure, heart rate and renal function would have permitted a dose titration. Paracetamol was also underdosed (e.g. 500 mg for pain, with insufficient efficacy).
Table 4 Drugs or drug classes involved in each type of drug-related problem. Drug interactiona
Untreated indication
Overdosage
No valid indication
Adverse drug reaction
Subtherapeutic dosage
Improper drug selection
Tramadol Antidepressants Acenocoumarol Metoprolol Clopidogrel Neuroleptics Digoxin
Calcium/vitamin D Statins Aspirin ACEI/ARB Beta-blockers Laxatives
PPI Paracetamol
PPI Aspirin
CNS drugs Anticoagulants Chemotherapy
Paracetamol Opioids Beta-blockers ACEI/ARB
Antihypertensives Opioids
ACEI: angiotensin-converting enzyme inhibitors. ARB: angiotensin II receptor blockers. PPI: proton pump inhibitors. CNS: central nervous system. a Because drug interactions involve, by definition, at least two drugs, only the drug/drug class causing the DRP and associated with the highest patient risk (because of the efficacy/ toxicity ratio modulation) is presented here.
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
5
3.3. Decision on the intervention One-half of identified DRPs (196/383, 51%) were not reported to the prescribers because they were not considered clinically relevant or required monitoring already provided by caregivers (e.g., monitoring of blood pressure, heart rate, kalemia and serum creatinine). The majority of the other DRPs (161/383, 42%) resulted in a verbal intervention during medical rounds, with a suggestion for treatment modification or introduction of monitoring. In 7% of the cases (26/383), the DRPs were considered too complex to be solved by a single verbal intervention (e.g. interactions involving more than two drugs and/or presence of drug inhibitors and inducers, therapeutic drug monitoring requiring a dose adaptation for several days after the medical round). Such consultations were also driven by specific requests of prescribers to induce a change in current clinical practice, or when ADRs had to be reported to the Swiss national pharmacovigilance center (Swissmedic). The types of intervention according to each type of DRP are detailed in Fig. 1. Among the 82 detected unwanted drug interactions, 52 (63%) were not reported to the prescribers because they were considered non-clinically relevant, concerned drugs with a large therapeutic margin, or required monitoring already provided by caregivers (e.g., monitoring of blood pressure, heart rate, kalemia and serum creatinine). Twenty-seven unwanted drug interactions (33%) warranted verbal advice during rounds. The recommendations were dosage modifications or molecule changes because of a pharmacokinetic and/or pharmacodynamic interaction (simvastatin/clarithromycin, raltegravir/rifampin, tramadol replacement by morphine with CYP2D6 or 3A4 inhibitors); therapeutic drug monitoring (e.g., digoxin, cyclosporine); or monitoring not yet conducted (e.g., control of the QTc interval). Three complex unwanted drug interaction cases (e.g., those involving more than two drugs) (4%) gave rise to written consultations. Among 60 overdosages, 38 (61%) were reported to the physicians in charge and a reduction of the dose was suggested. The most common overdosage cases were related to esomeprazole (a dose reduction to 20 mg daily for peptic ulcer prophylaxis or gastroesophageal reflux treatment), paracetamol (1 g twice daily as the maximum in alcoholism, cachexia or with concomitant enzymatic inducers), or drugs adjusted to renal function. One-half of the detected adverse drug effects (20/38, 55%) were considered clinically relevant and were reported to the physician and to the national pharmacovigilance center (a written consultation).
Fig. 2. Flow chart of detected drug-related problems (DRPs), choice of intervention and advice acceptance rates with or without actual prescription change.
accepted by the physicians (e.g. change of tramadol to morphine or buprenorphine, control of the QTc interval, introduction of calcium and vitamin D, reduction or stop of proton pump inhibitors, increase of beta-blocker or ACE inhibitor dose). The suggestions that were not accepted for specific patients included changing one antihypertensive to another more adapted to the patient comorbidities, or adding aspirin or a statin in ischemic conditions (coronary heart disease, previous stroke or peripheral artery disease). Among 135 accepted verbal suggestions, 42 (31%) were ultimately not followed, and the prescriptions were unchanged when the medical records were screened again five days after the medical round. 3.5. Prescribers' satisfaction survey The survey response rate was 68% (15 of 22 prescribers), with eleven residents and four chief residents. The presence of the pharmacotherapy experts was rated useful for 100% of the prescribers, and the medical rounds were considered the best time for advice on therapy optimization. The optimal frequency of this participation varied from once a week for seven prescribers (46.7%) to twice a month for seven prescribers (46.7%) to once a month for one prescriber (6.7%). The usefulness of the pharmacotherapy experts during the medical rounds was assessed for each type of DRP. The scores are reported in Table 5, as is the number of prescribers willing to acquire new skills for each type of DRP.
3.4. Acceptance and application rate by the prescribers 4. Discussion Fig. 2 presents the number of detected DRPs that were reported to the prescribers, the number of advisory suggestions that were accepted and the number of suggestions that were actually applied. Among the 161 verbal suggestions offered during rounds, most (135, 84%) were
This work was an interventional prospective study that aimed to detect suboptimal drug use in internal medicine by a specialized pharmacotherapy evaluation in the ward, to suggest previously prioritized treatment optimizations, and to assess the acceptance and satisfaction Table 5 Usefulness of the pharmacotherapy expert and number of prescribers willing to acquire new skills by type of drug-related problem (DRP). An average usefulness score was attributed to each type of DRP by attributing different points according to the level of usefulness (useless: 1 point, average usefulness: 2 points, useful: 3 points, very useful: 4 points) and by dividing by the total score by the total number of opinions.
Fig. 1. Attitude chosen according to each type of drug-related problem: no intervention (white), verbal advice during medical rounds (grey), and verbal advice during medical rounds followed by a written specialized clinical pharmacology consultation (black).
Drug interaction Adverse drug reaction No valid indication Overdosage Subtherapeutic dosage Improper drug selection Untreated indication
Average usefulness score
Number of physicians willing to acquire new skills
3.87 3.47 3.07 3.00 2.87 2.80 2.33
14 11 3 6 8 4 3
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
6
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
of the prescribers. A mean of 2.7 DRPs per patient was observed in the patients included, with 81% of the patients having at least one DRP. The majority of the patients was polymorbid and was polymedicated, as 94% received at least five drugs. These observations are consistent with studies concerning DRP detection in internal medicine wards, predominantly involving polymorbid and polymedicated patients [13,29,30]. Our study showed that the most represented types of DRPs in internal medicine wards were unwanted drug interactions, untreated indications, overdosages and drugs with no valid indication. In our study, ADRs were less represented (9% of the detected DRPs) than in other studies (e.g., 31%) [31]. One reason for this difference might be that ADRs are traditionally more investigated than other DRPs. ADRs might have been underdetected in our study because the detection of DRPs was made by the clinical pharmacist and not by the clinical pharmacology specialist (a physician). Pharmacists are not educated to diagnose, and a diagnosis approach is warranted to detect ADRs by ruling out medical causes for the observed problem. Unwanted drug interactions, ADRs and overdosages are the DRPs traditionally described in the literature [31–33]. Other DRPs such as untreated indications merit being recognized as well. The untreated indication DRP was the second in frequency in our study. This DRP is referenced in the literature as “underuse” or “therapeutic inertia” and has been described in chronic conditions such as hypertension, diabetes, coronary heart disease, heart failure, or secondary stroke prevention [34,35]. In our study, one-half of the DRPs detected were not considered clinically relevant and were not reported to the prescribers. The clinical relevance of the detected DRPs has rarely been assessed in comparable studies. This assessment is of value for prioritizing the interventions during busy medical rounds, during which advice on treatment optimization might be time-consuming. Reporting non-clinically important DRPs might reduce the effect of pharmacotherapy interventions. Blix et al. classified detected DRPs into the following four groups, according to their clinical significance: extremely important, major, moderate and minor [13]. The DRPs considered to be extremely important and major were defined by Blix et al. as requiring interventions to prevent death or severe or major detrimental effects, or lack of therapy when evidence-based options were available. Extremely important and major DRPs represented 49.6% of the detected DRPs in the study of Blix et al., which is similar to the clinically relevant DRPs reported in our study (49%). The rate of acceptance of advice for treatment optimization in the prescribers was high (84%) and consistent with the acceptance rates described in the literature (50% to 98%) [8,29,36]. Prioritization of the most clinically relevant DRPs to be reported to the prescribers might partly explain the high acceptance rate. Another contributing factor might be the presence of the pharmacotherapy experts at medical rounds intervening directly during the prescribing process. Other studies found lower acceptance rates (e.g., 40.9%) when treatment optimization suggestions were written [37]. The isolated suggestions that were not accepted by physicians were related to a different assessment of the clinical situation by the physician in charge of the patient (e.g., the therapeutic objectives, life expectancy) or the non-willingness to modify chronic condition treatments introduced by general practitioners in the community, not to futile DRPs. In total, 31% of the accepted suggestions were not followed, and the prescriptions were unchanged when the medical records were rescreened after five days. Some treatment optimization suggestions might have been forgotten by the medical resident or a validation by their supervisors might have been required if the supervisor were not present during medical rounds. To our knowledge, similar studies only assessed the acceptance rate among prescribers and did not assess the actual prescription changes. The satisfaction rate of prescribers was maximal (100%) regarding the usefulness of the pharmacotherapy experts during rounds.
Prescribers were primarily willing to deepen their knowledge of drug interactions, which is consistent with drug interactions being the most frequently detected DRP in the study. Untreated indication DRP (the second DRP in frequency) ranked last in the DRPs of which prescribers wanted to acquire new skills. According to our feeling, untreated indication DRPs seemed not to be related to a lack of knowledge; they seamed to result from an unintentional omission, particularly when the untreated pathology was not acute and did not cause the hospitalization. This impression would merit further investigation. In our observation, prescribers tend to prioritize acute problems and leave chronic condition therapy improvement to the general practitioner. Unintentional drug omission for a long period might however lead to a phenomenon called “underuse” or “therapeutic inertia” [38]. Detection of this type of underuse might be challenging for physicians, particularly if the patient is admitted for another medical problem. Systematic medication reviews led by a professional who prioritizes drug efficacy and safety, such as a clinical pharmacist or a clinical pharmacologist, is a way to detect this type of underuse [12]. Screening tools that are easy to use and time-efficient are another detection method [27,39]. They were primarily developed in geriatrics and are useful in an aged population; tools adapted to middle-aged patients are not available. This study has some limitations. First, this study was a before-andafter study without a control group. Patients were their own controls before the interventions of the pharmacotherapy experts. The impact of these interventions on prescription quality was assessed by the number of accepted and applied suggestions of prescription change (93 suggestions accepted and followed). A control group would have prevented the main confounding factor, being detection and resolution of DRPs by residents themselves, without the intervention of the pharmacotherapy experts. Furthermore, the number of patients included was smaller than in other studies assessing types of DRPs in wards. A bigger size study would be necessary to confirm the observed trends although our results are consistent with those of other studies. Our findings are of interest as a description of the types of DRPs most often detected in internal medicine and would help to define future specific projects focused on these DRPs. The number of prescribers met during medical rounds was small, and the satisfaction survey does not necessarily reflect the opinion of all of the prescribers in internal medicine. These results show a trend that should be confirmed on a larger scale; however, they are encouraging concerning the collaboration of prescribers with pharmacotherapy experts. To strengthen these findings, a multicentric future study could involve other internal medicine wards from other hospitals. Finally, the included patients were seen once during rounds, and the available information was from medical charts exclusively. A more sustained follow-up and better knowledge of the settings of the patients would have permitted us to offer more relevant suggestions of treatment optimization and to address the failure to receive drugs DRP (identified by Hepler and Strand). This DRP relies on the psychosocial and economical setting of the patient and is linked to treatment adherence [16]. This investigation would have been possible only by time-consuming structured and exhaustive interviews of the patients as well as with close collaboration with community pharmacists and general practitioners. This study showed that the presence of a pharmacotherapy expert in internal medicine wards could be useful to prevent DRPs. The pool of clinical pharmacologists or trained clinical pharmacists is frequently too small to cover all of the hospital wards in most European countries. Thus, simple DRP identification interventions might be “dematerialized” to allow these experts to focus on more complex DRPs and intervention prioritization. This dematerialization could be carried out with medication review tools (on a paper format or as Smartphone applications) or could be directly integrated in computerized physician order entry (CPOE) systems as drug safety alerts. The DRPs most frequently detected in this study must be prioritized in future developments. Regarding unwanted drug interactions, various interaction detection programs/tools are commercially available, some
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
of which could be implemented in CPOE systems to generate drug safety alerts [40]. One-third of the unwanted interactions detected in our study were considered clinically relevant, indicating that a drug safety alert on drug interactions would not be useful in all cases. The reported interactions were selected according to the condition of the patient, and this type of subtle judgment is difficult if not impossible to dematerialize in a CPOE system. Research regarding additional criteria that could help to refine drug interaction detection by CPOE systems (e.g., age, laboratory values and renal function) would be of value to improve the specificity of generated interaction alerts. Regarding the untreated indication DRP and drug use without indication DRP, tools are described in the literature that might help prescribers to review their own prescription in a systematic manner. The most commonly tools cited are STOPP/START, the Beers' criteria, the Medication Appropriateness Index (MAI) and the Assessing Care of Vulnerable Elders (ACOVE) [27]. These tools were primarily developed for older people and focus on contraindicated drugs or drugs without a valid indication. The untreated indication DRP has received relatively little attention in the literature, and STOPP/START is one of the rare tools able to detect it. This DRP was the second most frequently occurring in our study; few assessment grids for medication review of younger adults are available, and the development of such a tool would be of added value. In conclusion, a specialized pharmacotherapy expertise during medical rounds was useful and well accepted by internal medicine ward prescribers in polymorbid and polymedicated patients. The high acceptance and satisfaction rate was favored by stringent prioritization of reported DRPs because treatment optimization advice is timeconsuming during rounds. Because of the modest allocation of clinical pharmacists and pharmacologists in Swiss hospitals, complementary strategies would be required, such as dematerialization of interventions through computerized clinical decision support and tools of medication reviews for prescribers. Prospective evaluations will be required to confirm the benefits of these strategies. Learning points • Patients admitted to general internal medicine wards might receive a large number of drugs and be at risk for drug-related problems (DRPs) associated with increased morbidity and mortality. • The most frequent DRPs detected in the internal medicine wards were drug interactions (21%), untreated indications (18%), overdosages (16%) and drugs used without a valid indication (10%). The drugs or drug classes most frequently involved were tramadol, antidepressants, acenocoumarol, calcium–vitamin D, statins, aspirin, proton pump inhibitors and paracetamol. • The satisfaction survey showed that prescribers were primarily willing to deepen their knowledge of drug interaction DRP (the first detected in frequency), in contrast to the untreated indication DRP (the second in frequency) that ranked last in the DRPs of which they wanted to acquire new skills. Unintentional drug omission for a long period might however lead to a phenomenon called “underuse” or “therapeutic inertia”. • Because of the modest allocation of pharmacists and pharmacologists in Swiss hospitals, complementary strategies would be required (e.g., dematerialization of interventions through computerized clinical decision support). Conflict of interest The authors declare that they have no conflict of interest. References [1] Bates DW, Spell N, Cullen DJ, Burdick E, Laird N, Petersen LA, et al. The costs of adverse drug events in hospitalized patients. Adverse Drug Events Prevention Study Group. JAMA 1997;277:307–11.
7
[2] Pirmohamed M, James S, Meakin S, Green C, Scott AK, Walley TJ, et al. Adverse drug reactions as cause of admission to hospital: prospective analysis of 18 820 patients. BMJ 2004;329:15–9. [3] Classen DC, Pestotnik SL, Evans RS, Lloyd JF, Burke JP. Adverse drug events in hospitalized patients. Excess length of stay, extra costs, and attributable mortality. JAMA 1997;277:301–6. [4] Ernst ME, Iyer SS, Doucette WR. Drug-related problems and quality of life in arthritis and low back pain sufferers. Value Health 2003;6:51–8. [5] Strand LM, Morley PC, Cipolle RJ, Ramsey R, Lamsam GD. Drug-related problems: their structure and function. DICP 1990;24:1093–7. [6] Brennan TA, Leape LL, Laird NM, Hebert L, Localio AR, Lawthers AG, et al. Incidence of adverse events and negligence in hospitalized patients. Results of the Harvard Medical Practice Study I. N Engl J Med 1991;324:370–6. [7] Hanlon JT, Weinberger M, Samsa GP, Schmader KE, Uttech KM, Lewis IK, et al. A randomized, controlled trial of a clinical pharmacist intervention to improve inappropriate prescribing in elderly outpatients with polypharmacy. Am J Med 1996;100: 428–37. [8] Kucukarslan SN, Peters M, Mlynarek M, Nafziger DA. Pharmacists on rounding teams reduce preventable adverse drug events in hospital general medicine units. Arch Intern Med 2003;163:2014–8. [9] Kaur S, Mitchell G, Vitetta L, Roberts MS. Interventions that can reduce inappropriate prescribing in the elderly: a systematic review. Drugs Aging 2009;26:1013–28. [10] Boyko Jr WL, Yurkowski PJ, Ivey MF, Armitstead JA, Roberts BL. Pharmacist influence on economic and morbidity outcomes in a tertiary care teaching hospital. Am J Health Syst Pharm 1997;54:1591–5. [11] McMullin ST, Hennenfent JA, Ritchie DJ, Huey WY, Lonergan TP, Schaiff RA, et al. A prospective, randomized trial to assess the cost impact of pharmacist-initiated interventions. Arch Intern Med 1999;159:2306–9. [12] Viktil KK, Blix HS. The impact of clinical pharmacists on drug-related problems and clinical outcomes. Basic Clin Pharmacol Toxicol 2008;102:275–80. [13] Blix HS, Viktil KK, Reikvam A, Moger TA, Hjemaas BJ, Pretsch P, et al. The majority of hospitalised patients have drug-related problems: results from a prospective study in general hospitals. Eur J Clin Pharmacol 2004;60:651–8. [14] Viktil KK, Blix HS, Moger TA, Reikvam A. Polypharmacy as commonly defined is an indicator of limited value in the assessment of drug-related problems. Br J Clin Pharmacol 2007;63:187–95. [15] Lagnaoui R, Moore N, Fach J, Longy-Boursier M, Begaud B. Adverse drug reactions in a department of systemic diseases-oriented internal medicine: prevalence, incidence, direct costs and avoidability. Eur J Clin Pharmacol 2000;56:181–6. [16] Hepler CD, Strand LM. Opportunities and responsibilities in pharmaceutical care. Am J Hosp Pharm 1990;47:533–43. [17] Lexi-Comp Online, Lexi Interact Online. Hudson, Ohio: Lexi-Comp, Inc. [18] Interactions médicamenteuses, cytochromes P450 et P-glycoprotéine (Pgp). Genève: Service de pharmacologie et toxicologie cliniques. Hôpitaux Universitaires de Genève; 2011. [19] Credible Meds, QTDrugs Lists. Arizona: AZCERT, Inc. [20] Swissmedic, Information sur le médicament. Bern: Swissmedic. [21] Lexi-Comp Online, Lexi Drug Online. Hudson, Ohio: Lexi-Comp, Inc. [22] DRUGDEX System (Micromedex 2.0). Greenwood Village, CO: Truven Health Analytics. [23] Aronoff GR, Berns JS, Brier ME, Golper TA, Morrison G, Singer I, et al. Drug prescribing in renal failure: dosing guidelines for adults. 4th Revised ed. Philadelphia, PA: American College of Physicians; 1999. [24] Aronson JK, Dukes MNG. Meyler's side effects of drugs. 15th revised ed. Amsterdam, Netherlands: Elsevier; 2006. [25] McEvoy GK, Snow EK. AHFS: drug information. Bethesda, MD: American Society of Health-System Pharmacists; 2012. [26] Trissel LA. Handbook on injectable drugs. 17th ed. Bethesda, MA: American Society of Health-System Pharmacists; 2013. [27] Gallagher P, Ryan C, Byrne S, Kennedy J, O'Mahony D. STOPP (Screening Tool of Older Person's Prescriptions) and START (Screening Tool to Alert Doctors to Right Treatment). Consensus validation. Int J Clin Pharmacol Ther 2008;46:72–83. [28] Dooley MJ, Allen KM, Doecke CJ, Galbraith KJ, Taylor GR, Bright J, et al. A prospective multicentre study of pharmacist initiated changes to drug therapy and patient management in acute care government funded hospitals. Br J Clin Pharmacol 2004;57: 513–21. [29] Blix HS, Viktil KK, Moger TA, Reikvam A. Characteristics of drug-related problems discussed by hospital pharmacists in multidisciplinary teams. Pharm World Sci 2006;28:152–8. [30] Bondesson A, Eriksson T, Kragh A, Holmdahl L, Midlov P, Hoglund P. In-hospital medication reviews reduce unidentified drug-related problems. Eur J Clin Pharmacol 2013;69:647–55. [31] Mannheimer B, Ulfvarson J, Eklof S, Bergqvist M, Andersen-Karlsson E, Pettersson H, et al. Drug-related problems and pharmacotherapeutic advisory intervention at a medicine clinic. Eur J Clin Pharmacol 2006;62:1075–81. [32] Blix HS, Viktil KK, Moger TA, Reikvam A. Identification of drug interactions in hospitals—computerized screening vs. bedside recording. J Clin Pharm Ther 2008; 33:131–9. [33] McDonnell PJ, Jacobs MR. Hospital admissions resulting from preventable adverse drug reactions. Ann Pharmacother 2002;36:1331–6. [34] Guthrie B, Inkster M, Fahey T. Tackling therapeutic inertia: role of treatment data in quality indicators. BMJ 2007;335:542–4. [35] Touze E, Coste J, Voicu M, Kansao J, Masmoudi R, Doumenc B, et al. Importance of inhospital initiation of therapies and therapeutic inertia in secondary stroke prevention: IMplementation of Prevention After a Cerebrovascular evenT (IMPACT) Study. Stroke 2008;39:1834–43.
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
8
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
[36] Galindo C, Olive M, Lacasa C, Martinez J, Roure C, Llado M, et al. Pharmaceutical care: pharmacy involvement in prescribing in an acute-care hospital. Pharm World Sci 2003;25:56–64. [37] Patel HR, Pruchnicki MC, Hall LE. Assessment for chronic kidney disease service in high-risk patients at community health clinics. Ann Pharmacother 2005;39:22–7. [38] Miedema MD, Cohn JN, Garberich RF, Knickelbine T, Graham KJ, Henry TD. Underuse of cardiovascular preventive pharmacotherapy in patients presenting with ST-elevation myocardial infarction. Am Heart J 2012;164:259–67.
[39] Shekelle PG, MacLean CH, Morton SC, Wenger NS. Acove quality indicators. Ann Intern Med 2001;135:653–67. [40] Ing Lorenzini K, Reutemann B, Samer CF, Guignard B, Bonnabry P, Dayer P, et al. Comparison of four drug interaction screening programs. Rev Med Suisse 2012;8: 1978–82.
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
Contents lists available at ScienceDirect
European Journal of Internal Medicine journal homepage: www.elsevier.com/locate/ejim
Original Article
Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist Bertrand Guignard a,b,⁎, Pascal Bonnabry a,c, Arnaud Perrier d, Pierre Dayer b,c, Jules Desmeules b,c, Caroline Flora Samer b,e a
Pharmacy, Geneva University Hospitals, Geneva, Switzerland Service of Clinical Pharmacology and Toxicology, Geneva University Hospitals, Geneva, Switzerland School of Pharmaceutical Sciences, University of Geneva, University of Lausanne, Geneva, Switzerland d Service of General Internal Medicine, Geneva University Hospitals, Geneva, Switzerland e Swiss Centre for Applied Human Toxicology, University of Geneva, Geneva, Switzerland b c
a r t i c l e
i n f o
Article history: Received 27 December 2014 Received in revised form 18 May 2015 Accepted 19 May 2015 Available online xxxx Keywords: Drug-related problems Internal medicine Hospital
a b s t r a c t Background: Patients admitted to general internal medicine wards might receive a large number of drugs and be at risk for drug-related problems (DRPs) associated with increased morbidity and mortality. This study aimed to detect suboptimal drug use in internal medicine by a pharmacotherapy evaluation, to suggest treatment optimizations and to assess the acceptance and satisfaction of the prescribers. Methods: This was a 6-month prospective study conducted in two internal medicine wards. Physician rounds were attended by a pharmacist and a pharmacologist. An assessment grid was used to detect the DRPs in electronic prescriptions 24 h in advance. One of the following interventions was selected, depending on the relevance and complexity of the DRPs: no intervention, verbal advice of treatment optimization, or written consultation. The acceptance rate and satisfaction of prescribers were measured. Results: In total, 145 patients were included, and 383 DRPs were identified (mean: 2.6 DRPs per patient). The most frequent DRPs were drug interactions (21%), untreated indications (18%), overdosages (16%) and drugs used without a valid indication (10%). The drugs or drug classes most frequently involved were tramadol, antidepressants, acenocoumarol, calcium–vitamin D, statins, aspirin, proton pump inhibitors and paracetamol. The following interventions were selected: no intervention (51%), verbal advice of treatment optimization (42%), and written consultation (7%). The acceptance rate of prescribers was 84% and their satisfaction was high. Conclusion: Pharmacotherapy expertise during medical rounds was useful and well accepted by prescribers. Because of the modest allocation of pharmacists and pharmacologists in Swiss hospitals, complementary strategies would be required. © 2015 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
1. Introduction Pharmacotherapy is becoming increasingly complex, and inappropriate drug prescription might be associated with increased healthcare costs and hospital admissions as well as prolonged hospital stays, reduced quality of life, and increased morbidity or mortality [1–4]. A drug-related problem (DRP) is defined as an event or circumstance involving drug therapy that actually or potentially interferes with the desired health outcomes [5]. The majority of DRPs are predictable
⁎ Corresponding author at: Pharmacy, Geneva University Hospitals, Rue GabriellePerret-Gentil 4, CH-1211 Geneva 14, Switzerland. Tel.: +41 22 382 3963; fax: +41 22 382 3990. E-mail address: [email protected] (B. Guignard).
and potentially avoidable [6]. DRP frequency might be reduced by pharmacotherapy optimization, such as medication reviews led by pharmacists or clinical pharmacologists [7–9]. The positive effects of these medication reviews on costs were reported [10,11]. Moreover, these interventions were associated with reduced durations of hospital stay or decreased frequency of re-admissions as well as with better control of certain biomarkers (lipid levels, anticoagulation levels or blood pressure) and disease events such as decompensated heart failure or thromboembolic events [12]. Inappropriate drug use has been largely studied in the elderly, a population characterized by frailty, polymorbidity and polymedication; fewer studies have addressed the question in internal medicine wards, in which younger patients are admitted. It has been shown that polymorbidity and polymedication were independent risk factors of DRPs, whereas age and gender were not [13,14]. Elderly and middle-
http://dx.doi.org/10.1016/j.ejim.2015.05.012 0953-6205/© 2015 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
2
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
aged patients admitted to internal medicine wards suffer from multiple diseases or have several risk factors and receive a large number of drugs. They are thus at high risk for DRPs. In a French study, the in-hospital incidence rate of adverse drug reactions (ADRs) in internal medicine was 10.1 per 1000 patient-days, and 80% of the ADRs were considered preventable [15]. Our study was conducted in two internal medicine wards at the Geneva University Hospitals, a 2000-bed university health center in Switzerland. The Geneva University Hospitals have the following two complementary information centers for caregivers regarding medication use: the Clinical Pharmacology and Toxicology medical service for pharmacotherapeutic questions (such as drug–drug interactions, ADRs, selection of drugs and dosage, use of drugs in pregnancy, extreme age populations or organ dysfunction) and a Pharmacy hotline for pharmacotechnical questions (such as intravenous drug dilutions, intravenous drugs compatibilities, or tablet crushability for administration through a nasogastric feeding tube). This study is a joint collaboration between these two entities, through a clinical pharmacist and a clinical pharmacologist tandem, working together, addressing the question of DRPs in internal medicine by screening medical records, attending medical rounds, and providing advice for pharmacotherapy optimization. Whereas most studies evaluating DRPs are retrospective, our study offers the advantage of being prospective. We sought to optimize the impact of our intervention by prioritizing DRPs to be reported to prescribers during rounds. The aims of the study were as follows: (1) to detect all of the DRPs in the included patients and to identify the drugs or drug classes most frequently causing the problems; (2) to assess which intervention (none, verbal advice and written consultation) was required for each DRP according to its clinical relevance or complexity; (3) to measure the acceptance rate of the prescribers and the actual impact on the prescription; and (4) to assess the satisfaction of the prescribers regarding the suggestions for treatment optimization. 2. Methods 2.1. Patients and design This prospective interventional study was conducted during a consecutive 6-month period. From September 2011 to February 2012, a clinical pharmacist and a clinical pharmacologist (i.e. a physician specializing in clinical pharmacology and internal medicine) both attended medical rounds approximately once a week on two internal medicine wards. All the patients admitted in the visited internal medicine wards were considered eligible and there were no exclusion criteria. The following data and comorbidities as well as the known risk factors for DRPs were systematically recorded for each patient: sex, age, and the number of drugs prescribed; the occurrence of hypertension, coronary heart disease, heart failure, atrial fibrillation, diabetes, and reduced renal function (a creatinine clearance ≤50 ml/min according to the Cockroft–Gault formula); previously reported liver dysfunction; chronic obstructive pulmonary disease or asthma; a previous history of a transient ischemic attack or stroke, malignant disease, alcohol abuse, tobacco use; and polypharmacy (≥5 drugs).
pharmacist screened the medical records of the patients on the ward and performed a structured medication review with an assessment grid (Table 1) to detect the DRPs. The assessment grid was developed by compiling the clinical decision supports classically used for medication reviews (e.g., drug interaction screening tools [17–19], drug databases [20–22] or textbooks [23–26] and published tools for the medication review [27]). All of the identified DRPs were discussed between the clinical pharmacist and the clinical pharmacologist before the medical round to determine their clinical relevance and to prioritize interventions. DRPs might be dependent on each other, and only primary DRPs are presented in the results. An unwanted drug interaction might require a dose reduction (overdosage is a secondary DRP) or an ADR might render a drug inappropriate for a patient (improper drug selection is a secondary DRP). Because drug interactions involve, by definition, at least two drugs, only the drug/drug class causing the DRP and associated with the highest patient risk (because of the efficacy/toxicity ratio modulation) is presented in the results. 2.3. Decision on the intervention Based on the potential clinical relevance and complexity of the detected DRPs, an attitude was determined, as follows: (1) no intervention and/or continuation of usual follow-up by caregivers, (2) verbal advice to the physicians and/or nurses during medical rounds (treatment optimization or recommendation to introduce monitoring) or (3) verbal advice during medical rounds followed by a written specialized clinical pharmacology consultation. No intervention was conducted for non-clinically relevant DRPs or when appropriate monitoring was already provided (e.g., monitoring of blood pressure, heart rate, kalemia and serum creatinine). Clinical relevance of DRPs has previously been defined by Dooley et al. and Blix et al. DRPs were thus considered non-clinically relevant if they were non major, i.e. the chance of noticed effect was lower than 20%, the chance of harmful effect was lower than 5%, or they would not require intervention due to the lack of detrimental effects [13,28]. Written specialized consultations were reserved for complex clinical pharmacological situations, when specifically requested by prescribers to induce a change in current clinical practice, or when ADRs had to be reported to the Swiss national pharmacovigilance center (Swissmedic). 2.4. Acceptance and application rate by prescribers The acceptance rates of treatment optimization recommendations made to prescribers during the medical round were measured. In case of recommendation acceptance by the physicians during medical rounds, the prescription medical records were systematically rescreened after five days to assess whether the treatments had actually been changed according to the advice (application rate). 2.5. Prescribers' satisfaction survey The satisfaction level of prescribers during medical rounds was rated with an anonymous questionnaire assessing the following factors:
2.2. Detection of DRPs and the drugs or drug classes involved The actual and potential DRPs were classified into the following seven categories, as previously described by Hepler and Strand when they first described the process of pharmaceutical care: drug interaction, subtherapeutic dosage, overdosage, drug use without indication, untreated indication, improper drug selection, and adverse drug reaction [16]. The treatment adherence problems were not addressed in the study. The day before the medical round, a clinical
(1) The physician education level (a categorical choice); (2) The global opinion on the participation of the pharmacotherapy experts in medical round: general usefulness of pharmacotherapy experts, time for verbal advices, and optimal frequency of attending (a categorical choice); (3) The usefulness of the advice for preventing specific DRPs (a fourpoint Likert scale). An average score was attributed to each type of DRP by attributing different points according to the level of
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
3
Table 1 Assessment grid to detect the drug-related problems. Drug-related problems
Detection criteria
Used tools and references
Drug interaction
- Interaction described by Lexi Interact OR - Pharmacokinetic interaction between a substrate and an inhibitor/inducer of a common metabolic/elimination pathway (cytochrome P450 or P-glycoprotein) OR - Pharmacodynamic interaction between QT prolongating drugs or between serotoninergic drugs AND - Interaction described by Drugdex as associated with a risk of a decreased efficacy and/or increased toxicity Subtherapeutic - Dosage lower than the minimal recommended dosage for a given indication (published in the Swiss drug Compendium or in other dosage references, for drugs unlicensed in Switzerland) OR - Improved renal function with drugs whose dosage has been previously adjusted to renal failure and becoming underdosed with renal function improvement. Overdosage - Dosage higher than the maximal recommended dosage for a given indication (published in the Swiss drug Compendium or in other references, for drugs unlicensed in Switzerland) OR - Drug dosage not adjusted to renal function in renal impairment (creatinine clearance estimated with the Cockroft–Gault formula), following recommendations from Swiss drug Compendium or other references OR - Hepatic impairment reported in medical record and prescription of drugs whose dosage has to be adjusted to hepatic impairment, following recommendations from Swiss drug Compendium or other references. - Drugs belonging to the same group of Anatomical Therapeutic Chemical classification level 4 or 5 (therapeutic duplication) OR Drug use - Drug with no clear indication in the medical record OR without - Criterion in Screening Tool of Older Persons' Prescriptions (STOPP) OR indication - One of the following prescribing cascade: - Nonsteroidal anti-inflammatory drug + antihypertensive drug - Thiazide + gout treatment - Neuroleptics + antiparkinsonian drug - Metoclopramide + antiparkinsonian drug - Cholinesterase inhibitor + anticholinergic drug - Anticholinergic drug + treatment of benign prostate hyperplasia - Angiotensin-converting enzyme inhibitor + antitussive - Antihypertensive drug + vestibular system suppressor - Opioid + vestibular system suppressor - Sedative + vestibular system suppressor Untreated - Criterion in Screening Tool to Alert doctors to Right Treatment (START) OR indication - Drugs recommended for prophylaxis but not prescribed: - Laxative and opioids - Calcium and vitamin D in elderly - Prophylactic anticoagulation after surgery or venous thromboembolism risk factors - Vitamin B6 and isoniazid Improper drug - Drug contraindicated according to Swiss drug Compendium OR selection - More effective or safer alternative OR - Less expensive alternative (with equivalent effectiveness and safety) Adverse drug - ADR suspected by the physician and reported in the medical record OR reaction (ADR) - Correlation between one of the following symptoms and ADR reported in the reference literature (Swiss drug Compendium, Drugdex or Meyler's Sides Effects of Drugs): - Abnormal liver biochemical tests - Hematologic abnormalities (neutropenia, agranulocytosis, thrombopenia, hemolytic anemia) - Hemorrhage - Cutaneous rash - Anaphylaxis - Acute renal failure - Digestive symptoms (dysphagia, dyspepsia, constipation, diarrhea) - Cough - Hypoglycemia or hyperglycemia - Neuromuscular symptoms (vertigo, seizure, muscle weakness/pain, extrapyramidal symptoms) - Cardiovascular toxicity (arrhythmia, heart failure, hypotension, hypertension) Others - Parenteral drug prescription when an oral formulation is available in a hemodynamically stable patient, able to swallow oral medications and with normal gastrointestinal tract function: - Intravenous antibiotic and antifungal agents with high bioavailability (e.g., fluoroquinolones, macrolides, cotrimoxazole, metronidazole, rifampin, azoles) - Other intravenous antibiotic agents after clinical improvement ≥ 48 h of intravenous treatment and apyrexia ≥ 24 h - Intravenous paracetamol - Intramuscular vitamin B12 for long term replacement - Inappropriate intravenous drug reconstitution, dilution or administration flow according to Swiss drug Compendium, Lexi Drug and AHFS - Intravenous drugs incompatibilities described in Trissel's Handbook on Injectable Drugs - Prescription of medicines that cannot be crushed (e.g., sustained release tablets, gastroresistant tablets) for a patient with a nasogastric feeding tube - Suboptimal drug administration timing (e.g., regarding to meals, regarding to circadian rhythm) according to Swiss drug Compendium, Lexi Drug, and AHFS.
usefulness (useless: 1 point, average usefulness: 2 points, useful: 3 points, very useful: 4 points) and by dividing the total score by the total number of opinions; (4) The types of DRPs for which the physicians wanted to acquire new skills (multiple choice).
[17–19,22]
[20–22]
[20–23]
[27]
[27]
[20]
[20–22,24]
[20,21,25,26]
2.6. Data analysis The data analysis was descriptive. The categorical variables were expressed as the percentage. The continuous data were reported as the mean ± standard deviation (SD).
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
4
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
Table 2 Characteristics of the patients included in the study with frequency of chronic conditions. Male (n) Age (mean ± SD), range Number of drugs prescribed (mean ± SD), range
75 (52%) 68 ± 16, 21–99 10.6 ± 4.0, 0–21
Chronic conditions Hypertension (n) Coronary heart disease (n) Heart failure (n) Atrial fibrillation (n) Diabetes mellitus (n) Reduced renal functiona (n) Previously reported liver dysfunction (n) Chronic obstructive pulmonary disease or asthma (n) Previous history of transient ischemic attack or stroke (n) Previous history of malignant disease (n) Previous or current alcohol abuse (n) Previous or current tobacco use (n) Polypharmacy (≥5 drugs) (n)
66 (46%) 30 (21%) 37 (26%) 42 (29%) 38 (26%) 27 (19%) 8 (6%) 33 (23%) 11 (8%) 30 (21%) 28 (19%) 22 (15%) 136 (94%)
a
Creatinine clearance ≤ 50 ml/min according to Cockroft–Gault formula.
Table 3 Frequency and types of detected drug-related problems and number of patients exposed. Drug-related problems (DRPs)
Number (frequency)
Number of patients exposed (prevalence)
Drug interaction Untreated indication Overdosage No valid indication Adverse drug reaction Subtherapeutic dosage Improper drug selection Others All DRPs
82 (21%) 69 (18%) 60 (16%) 39 (10%) 36 (9%) 31 (8%) 29 (8%) 37 (10%) 383 (100%)
58 (40%) 48 (33%) 52 (36%) 33 (23%) 29 (20%) 25 (17%) 21 (14%) 31 (21%) 117 (81%)
3. Results 3.1. Patient characteristics The clinical pharmacist and the clinical pharmacologist attended 16 medical rounds and met 22 prescribers. A total of 145 patients were included, and 52% were male, with a mean age of 68 years (range 21–99 years, SD 16 years). The mean time for the screening of medical records and for the structured medication review was 1 h per patient. A total of 1523 prescriptions were recorded, with a mean of 10.6 prescriptions per patient (range 0–21, SD 4.0). The characteristics of the patients are detailed in Table 2. 3.2. DRPs and the drugs or drug classes involved A total of 383 DRPs were detected and had not been previously identified by the caregivers in charge (e.g. physicians, nurses), which
was a mean of 2.7 DRPs per patient (0–12, SD 2.4). At least one DRP was detected in 81% of the assessed medical records. The most frequently detected DRPs were drug interaction (21%), untreated indication (18%) and overdosage (16%), including drug dosage not adjusted to the renal function (Table 3). This distribution concerned all DRPs, independently of their clinical relevance or the decision to report them to prescribers or not. The prevalence of patients exposed to each type of DRP is also reported in Table 3. The drugs or drug classes involved in each type of DRP are described in Table 4. The drugs or drug classes most frequently involved in unwanted drug interactions were substrates of cytochrome P450 (CYP) isoenzymes (predominantly CYP 2D6, 2C9, 2C19 and 3A4) and/or, more rarely, the drug transporter, P-glycoprotein (P-gp). Drug interactions concerned mostly tramadol (11%), antidepressants (11%), anticoagulants (9%) and beta-blockers (9%). Interactions involving tramadol were either associated with a risk of decreased opioid therapeutic effect (4/9 interactions) by CYP 2D6 inhibition (e.g. by antidepressants), or increased toxicity (6/9 interactions) by CYP 3A4 inhibition (e.g. by clarithromycine, diltiazem) and/or a pharmacodynamic interaction through cumulation of serotoninergic effects with other serotoninergic drugs (e.g. antidepressants, antipsychotics). Risk of decreased opioid effect and increased toxicity was concomitant in one interaction with tramadol (i.e. through 2D6 and 3A4 inhibition by amiodarone). Antidepressants are substrates of CYP and their safety/efficacy ratio is modulated by pharmacokinetic interactions via CYP inhibitors (increased toxicity) or CYP inducers (reduced efficacy). Their toxicity might also increase via pharmacodynamic interactions with serotoninergic drugs or drugs known to prolong QT interval. Vitamin K antagonists are substrates of CYP, whereas direct thrombin or Xa inhibitors are substrates of CYP and/or P-gp. Inhibitors or inducers of these elimination pathways/transporters could lead to increased toxicity or decreased efficacy, respectively. The most frequent drugs or drug classes lacking in untreated indications were (a) calcium and vitamin D in elderly patients with osteoporosis and/or in those patients receiving or having received long-term systemic corticosteroids, associated with an increased risk of fracture, (b) statins and/or aspirin in ischemic heart disease patients, (c) angiotensin-converting enzyme (ACE) inhibitors, angiotensin II receptor blockers or beta-blockers withdrawn in decompensated heart failure patients and not reintroduced after resolution of the acute episode, and (d) laxatives in patients receiving opioids. Drugs or drug classes most frequently involved in overdosages were proton pump inhibitors (e.g. esomeprazole prescribed at 40 mg daily for peptic ulcer prophylaxis or for the treatment of gastroesophageal reflux) and paracetamol (dose of/or above 4 g per day despite alcoholism, cachexia or coprescription of enzymatic inducers), as well as drugs whose dosage was not adjusted for the occurrence of renal failure (e.g., enoxaparin, digoxin, ciprofloxacin and pregabalin). In contrast, there were drugs at a subtherapeutic dosage, like beta-blockers and ACE inhibitors for heart failure, which were not in their target dose, although blood pressure, heart rate and renal function would have permitted a dose titration. Paracetamol was also underdosed (e.g. 500 mg for pain, with insufficient efficacy).
Table 4 Drugs or drug classes involved in each type of drug-related problem. Drug interactiona
Untreated indication
Overdosage
No valid indication
Adverse drug reaction
Subtherapeutic dosage
Improper drug selection
Tramadol Antidepressants Acenocoumarol Metoprolol Clopidogrel Neuroleptics Digoxin
Calcium/vitamin D Statins Aspirin ACEI/ARB Beta-blockers Laxatives
PPI Paracetamol
PPI Aspirin
CNS drugs Anticoagulants Chemotherapy
Paracetamol Opioids Beta-blockers ACEI/ARB
Antihypertensives Opioids
ACEI: angiotensin-converting enzyme inhibitors. ARB: angiotensin II receptor blockers. PPI: proton pump inhibitors. CNS: central nervous system. a Because drug interactions involve, by definition, at least two drugs, only the drug/drug class causing the DRP and associated with the highest patient risk (because of the efficacy/ toxicity ratio modulation) is presented here.
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
5
3.3. Decision on the intervention One-half of identified DRPs (196/383, 51%) were not reported to the prescribers because they were not considered clinically relevant or required monitoring already provided by caregivers (e.g., monitoring of blood pressure, heart rate, kalemia and serum creatinine). The majority of the other DRPs (161/383, 42%) resulted in a verbal intervention during medical rounds, with a suggestion for treatment modification or introduction of monitoring. In 7% of the cases (26/383), the DRPs were considered too complex to be solved by a single verbal intervention (e.g. interactions involving more than two drugs and/or presence of drug inhibitors and inducers, therapeutic drug monitoring requiring a dose adaptation for several days after the medical round). Such consultations were also driven by specific requests of prescribers to induce a change in current clinical practice, or when ADRs had to be reported to the Swiss national pharmacovigilance center (Swissmedic). The types of intervention according to each type of DRP are detailed in Fig. 1. Among the 82 detected unwanted drug interactions, 52 (63%) were not reported to the prescribers because they were considered non-clinically relevant, concerned drugs with a large therapeutic margin, or required monitoring already provided by caregivers (e.g., monitoring of blood pressure, heart rate, kalemia and serum creatinine). Twenty-seven unwanted drug interactions (33%) warranted verbal advice during rounds. The recommendations were dosage modifications or molecule changes because of a pharmacokinetic and/or pharmacodynamic interaction (simvastatin/clarithromycin, raltegravir/rifampin, tramadol replacement by morphine with CYP2D6 or 3A4 inhibitors); therapeutic drug monitoring (e.g., digoxin, cyclosporine); or monitoring not yet conducted (e.g., control of the QTc interval). Three complex unwanted drug interaction cases (e.g., those involving more than two drugs) (4%) gave rise to written consultations. Among 60 overdosages, 38 (61%) were reported to the physicians in charge and a reduction of the dose was suggested. The most common overdosage cases were related to esomeprazole (a dose reduction to 20 mg daily for peptic ulcer prophylaxis or gastroesophageal reflux treatment), paracetamol (1 g twice daily as the maximum in alcoholism, cachexia or with concomitant enzymatic inducers), or drugs adjusted to renal function. One-half of the detected adverse drug effects (20/38, 55%) were considered clinically relevant and were reported to the physician and to the national pharmacovigilance center (a written consultation).
Fig. 2. Flow chart of detected drug-related problems (DRPs), choice of intervention and advice acceptance rates with or without actual prescription change.
accepted by the physicians (e.g. change of tramadol to morphine or buprenorphine, control of the QTc interval, introduction of calcium and vitamin D, reduction or stop of proton pump inhibitors, increase of beta-blocker or ACE inhibitor dose). The suggestions that were not accepted for specific patients included changing one antihypertensive to another more adapted to the patient comorbidities, or adding aspirin or a statin in ischemic conditions (coronary heart disease, previous stroke or peripheral artery disease). Among 135 accepted verbal suggestions, 42 (31%) were ultimately not followed, and the prescriptions were unchanged when the medical records were screened again five days after the medical round. 3.5. Prescribers' satisfaction survey The survey response rate was 68% (15 of 22 prescribers), with eleven residents and four chief residents. The presence of the pharmacotherapy experts was rated useful for 100% of the prescribers, and the medical rounds were considered the best time for advice on therapy optimization. The optimal frequency of this participation varied from once a week for seven prescribers (46.7%) to twice a month for seven prescribers (46.7%) to once a month for one prescriber (6.7%). The usefulness of the pharmacotherapy experts during the medical rounds was assessed for each type of DRP. The scores are reported in Table 5, as is the number of prescribers willing to acquire new skills for each type of DRP.
3.4. Acceptance and application rate by the prescribers 4. Discussion Fig. 2 presents the number of detected DRPs that were reported to the prescribers, the number of advisory suggestions that were accepted and the number of suggestions that were actually applied. Among the 161 verbal suggestions offered during rounds, most (135, 84%) were
This work was an interventional prospective study that aimed to detect suboptimal drug use in internal medicine by a specialized pharmacotherapy evaluation in the ward, to suggest previously prioritized treatment optimizations, and to assess the acceptance and satisfaction Table 5 Usefulness of the pharmacotherapy expert and number of prescribers willing to acquire new skills by type of drug-related problem (DRP). An average usefulness score was attributed to each type of DRP by attributing different points according to the level of usefulness (useless: 1 point, average usefulness: 2 points, useful: 3 points, very useful: 4 points) and by dividing by the total score by the total number of opinions.
Fig. 1. Attitude chosen according to each type of drug-related problem: no intervention (white), verbal advice during medical rounds (grey), and verbal advice during medical rounds followed by a written specialized clinical pharmacology consultation (black).
Drug interaction Adverse drug reaction No valid indication Overdosage Subtherapeutic dosage Improper drug selection Untreated indication
Average usefulness score
Number of physicians willing to acquire new skills
3.87 3.47 3.07 3.00 2.87 2.80 2.33
14 11 3 6 8 4 3
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
6
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
of the prescribers. A mean of 2.7 DRPs per patient was observed in the patients included, with 81% of the patients having at least one DRP. The majority of the patients was polymorbid and was polymedicated, as 94% received at least five drugs. These observations are consistent with studies concerning DRP detection in internal medicine wards, predominantly involving polymorbid and polymedicated patients [13,29,30]. Our study showed that the most represented types of DRPs in internal medicine wards were unwanted drug interactions, untreated indications, overdosages and drugs with no valid indication. In our study, ADRs were less represented (9% of the detected DRPs) than in other studies (e.g., 31%) [31]. One reason for this difference might be that ADRs are traditionally more investigated than other DRPs. ADRs might have been underdetected in our study because the detection of DRPs was made by the clinical pharmacist and not by the clinical pharmacology specialist (a physician). Pharmacists are not educated to diagnose, and a diagnosis approach is warranted to detect ADRs by ruling out medical causes for the observed problem. Unwanted drug interactions, ADRs and overdosages are the DRPs traditionally described in the literature [31–33]. Other DRPs such as untreated indications merit being recognized as well. The untreated indication DRP was the second in frequency in our study. This DRP is referenced in the literature as “underuse” or “therapeutic inertia” and has been described in chronic conditions such as hypertension, diabetes, coronary heart disease, heart failure, or secondary stroke prevention [34,35]. In our study, one-half of the DRPs detected were not considered clinically relevant and were not reported to the prescribers. The clinical relevance of the detected DRPs has rarely been assessed in comparable studies. This assessment is of value for prioritizing the interventions during busy medical rounds, during which advice on treatment optimization might be time-consuming. Reporting non-clinically important DRPs might reduce the effect of pharmacotherapy interventions. Blix et al. classified detected DRPs into the following four groups, according to their clinical significance: extremely important, major, moderate and minor [13]. The DRPs considered to be extremely important and major were defined by Blix et al. as requiring interventions to prevent death or severe or major detrimental effects, or lack of therapy when evidence-based options were available. Extremely important and major DRPs represented 49.6% of the detected DRPs in the study of Blix et al., which is similar to the clinically relevant DRPs reported in our study (49%). The rate of acceptance of advice for treatment optimization in the prescribers was high (84%) and consistent with the acceptance rates described in the literature (50% to 98%) [8,29,36]. Prioritization of the most clinically relevant DRPs to be reported to the prescribers might partly explain the high acceptance rate. Another contributing factor might be the presence of the pharmacotherapy experts at medical rounds intervening directly during the prescribing process. Other studies found lower acceptance rates (e.g., 40.9%) when treatment optimization suggestions were written [37]. The isolated suggestions that were not accepted by physicians were related to a different assessment of the clinical situation by the physician in charge of the patient (e.g., the therapeutic objectives, life expectancy) or the non-willingness to modify chronic condition treatments introduced by general practitioners in the community, not to futile DRPs. In total, 31% of the accepted suggestions were not followed, and the prescriptions were unchanged when the medical records were rescreened after five days. Some treatment optimization suggestions might have been forgotten by the medical resident or a validation by their supervisors might have been required if the supervisor were not present during medical rounds. To our knowledge, similar studies only assessed the acceptance rate among prescribers and did not assess the actual prescription changes. The satisfaction rate of prescribers was maximal (100%) regarding the usefulness of the pharmacotherapy experts during rounds.
Prescribers were primarily willing to deepen their knowledge of drug interactions, which is consistent with drug interactions being the most frequently detected DRP in the study. Untreated indication DRP (the second DRP in frequency) ranked last in the DRPs of which prescribers wanted to acquire new skills. According to our feeling, untreated indication DRPs seemed not to be related to a lack of knowledge; they seamed to result from an unintentional omission, particularly when the untreated pathology was not acute and did not cause the hospitalization. This impression would merit further investigation. In our observation, prescribers tend to prioritize acute problems and leave chronic condition therapy improvement to the general practitioner. Unintentional drug omission for a long period might however lead to a phenomenon called “underuse” or “therapeutic inertia” [38]. Detection of this type of underuse might be challenging for physicians, particularly if the patient is admitted for another medical problem. Systematic medication reviews led by a professional who prioritizes drug efficacy and safety, such as a clinical pharmacist or a clinical pharmacologist, is a way to detect this type of underuse [12]. Screening tools that are easy to use and time-efficient are another detection method [27,39]. They were primarily developed in geriatrics and are useful in an aged population; tools adapted to middle-aged patients are not available. This study has some limitations. First, this study was a before-andafter study without a control group. Patients were their own controls before the interventions of the pharmacotherapy experts. The impact of these interventions on prescription quality was assessed by the number of accepted and applied suggestions of prescription change (93 suggestions accepted and followed). A control group would have prevented the main confounding factor, being detection and resolution of DRPs by residents themselves, without the intervention of the pharmacotherapy experts. Furthermore, the number of patients included was smaller than in other studies assessing types of DRPs in wards. A bigger size study would be necessary to confirm the observed trends although our results are consistent with those of other studies. Our findings are of interest as a description of the types of DRPs most often detected in internal medicine and would help to define future specific projects focused on these DRPs. The number of prescribers met during medical rounds was small, and the satisfaction survey does not necessarily reflect the opinion of all of the prescribers in internal medicine. These results show a trend that should be confirmed on a larger scale; however, they are encouraging concerning the collaboration of prescribers with pharmacotherapy experts. To strengthen these findings, a multicentric future study could involve other internal medicine wards from other hospitals. Finally, the included patients were seen once during rounds, and the available information was from medical charts exclusively. A more sustained follow-up and better knowledge of the settings of the patients would have permitted us to offer more relevant suggestions of treatment optimization and to address the failure to receive drugs DRP (identified by Hepler and Strand). This DRP relies on the psychosocial and economical setting of the patient and is linked to treatment adherence [16]. This investigation would have been possible only by time-consuming structured and exhaustive interviews of the patients as well as with close collaboration with community pharmacists and general practitioners. This study showed that the presence of a pharmacotherapy expert in internal medicine wards could be useful to prevent DRPs. The pool of clinical pharmacologists or trained clinical pharmacists is frequently too small to cover all of the hospital wards in most European countries. Thus, simple DRP identification interventions might be “dematerialized” to allow these experts to focus on more complex DRPs and intervention prioritization. This dematerialization could be carried out with medication review tools (on a paper format or as Smartphone applications) or could be directly integrated in computerized physician order entry (CPOE) systems as drug safety alerts. The DRPs most frequently detected in this study must be prioritized in future developments. Regarding unwanted drug interactions, various interaction detection programs/tools are commercially available, some
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
of which could be implemented in CPOE systems to generate drug safety alerts [40]. One-third of the unwanted interactions detected in our study were considered clinically relevant, indicating that a drug safety alert on drug interactions would not be useful in all cases. The reported interactions were selected according to the condition of the patient, and this type of subtle judgment is difficult if not impossible to dematerialize in a CPOE system. Research regarding additional criteria that could help to refine drug interaction detection by CPOE systems (e.g., age, laboratory values and renal function) would be of value to improve the specificity of generated interaction alerts. Regarding the untreated indication DRP and drug use without indication DRP, tools are described in the literature that might help prescribers to review their own prescription in a systematic manner. The most commonly tools cited are STOPP/START, the Beers' criteria, the Medication Appropriateness Index (MAI) and the Assessing Care of Vulnerable Elders (ACOVE) [27]. These tools were primarily developed for older people and focus on contraindicated drugs or drugs without a valid indication. The untreated indication DRP has received relatively little attention in the literature, and STOPP/START is one of the rare tools able to detect it. This DRP was the second most frequently occurring in our study; few assessment grids for medication review of younger adults are available, and the development of such a tool would be of added value. In conclusion, a specialized pharmacotherapy expertise during medical rounds was useful and well accepted by internal medicine ward prescribers in polymorbid and polymedicated patients. The high acceptance and satisfaction rate was favored by stringent prioritization of reported DRPs because treatment optimization advice is timeconsuming during rounds. Because of the modest allocation of clinical pharmacists and pharmacologists in Swiss hospitals, complementary strategies would be required, such as dematerialization of interventions through computerized clinical decision support and tools of medication reviews for prescribers. Prospective evaluations will be required to confirm the benefits of these strategies. Learning points • Patients admitted to general internal medicine wards might receive a large number of drugs and be at risk for drug-related problems (DRPs) associated with increased morbidity and mortality. • The most frequent DRPs detected in the internal medicine wards were drug interactions (21%), untreated indications (18%), overdosages (16%) and drugs used without a valid indication (10%). The drugs or drug classes most frequently involved were tramadol, antidepressants, acenocoumarol, calcium–vitamin D, statins, aspirin, proton pump inhibitors and paracetamol. • The satisfaction survey showed that prescribers were primarily willing to deepen their knowledge of drug interaction DRP (the first detected in frequency), in contrast to the untreated indication DRP (the second in frequency) that ranked last in the DRPs of which they wanted to acquire new skills. Unintentional drug omission for a long period might however lead to a phenomenon called “underuse” or “therapeutic inertia”. • Because of the modest allocation of pharmacists and pharmacologists in Swiss hospitals, complementary strategies would be required (e.g., dematerialization of interventions through computerized clinical decision support). Conflict of interest The authors declare that they have no conflict of interest. References [1] Bates DW, Spell N, Cullen DJ, Burdick E, Laird N, Petersen LA, et al. The costs of adverse drug events in hospitalized patients. Adverse Drug Events Prevention Study Group. JAMA 1997;277:307–11.
7
[2] Pirmohamed M, James S, Meakin S, Green C, Scott AK, Walley TJ, et al. Adverse drug reactions as cause of admission to hospital: prospective analysis of 18 820 patients. BMJ 2004;329:15–9. [3] Classen DC, Pestotnik SL, Evans RS, Lloyd JF, Burke JP. Adverse drug events in hospitalized patients. Excess length of stay, extra costs, and attributable mortality. JAMA 1997;277:301–6. [4] Ernst ME, Iyer SS, Doucette WR. Drug-related problems and quality of life in arthritis and low back pain sufferers. Value Health 2003;6:51–8. [5] Strand LM, Morley PC, Cipolle RJ, Ramsey R, Lamsam GD. Drug-related problems: their structure and function. DICP 1990;24:1093–7. [6] Brennan TA, Leape LL, Laird NM, Hebert L, Localio AR, Lawthers AG, et al. Incidence of adverse events and negligence in hospitalized patients. Results of the Harvard Medical Practice Study I. N Engl J Med 1991;324:370–6. [7] Hanlon JT, Weinberger M, Samsa GP, Schmader KE, Uttech KM, Lewis IK, et al. A randomized, controlled trial of a clinical pharmacist intervention to improve inappropriate prescribing in elderly outpatients with polypharmacy. Am J Med 1996;100: 428–37. [8] Kucukarslan SN, Peters M, Mlynarek M, Nafziger DA. Pharmacists on rounding teams reduce preventable adverse drug events in hospital general medicine units. Arch Intern Med 2003;163:2014–8. [9] Kaur S, Mitchell G, Vitetta L, Roberts MS. Interventions that can reduce inappropriate prescribing in the elderly: a systematic review. Drugs Aging 2009;26:1013–28. [10] Boyko Jr WL, Yurkowski PJ, Ivey MF, Armitstead JA, Roberts BL. Pharmacist influence on economic and morbidity outcomes in a tertiary care teaching hospital. Am J Health Syst Pharm 1997;54:1591–5. [11] McMullin ST, Hennenfent JA, Ritchie DJ, Huey WY, Lonergan TP, Schaiff RA, et al. A prospective, randomized trial to assess the cost impact of pharmacist-initiated interventions. Arch Intern Med 1999;159:2306–9. [12] Viktil KK, Blix HS. The impact of clinical pharmacists on drug-related problems and clinical outcomes. Basic Clin Pharmacol Toxicol 2008;102:275–80. [13] Blix HS, Viktil KK, Reikvam A, Moger TA, Hjemaas BJ, Pretsch P, et al. The majority of hospitalised patients have drug-related problems: results from a prospective study in general hospitals. Eur J Clin Pharmacol 2004;60:651–8. [14] Viktil KK, Blix HS, Moger TA, Reikvam A. Polypharmacy as commonly defined is an indicator of limited value in the assessment of drug-related problems. Br J Clin Pharmacol 2007;63:187–95. [15] Lagnaoui R, Moore N, Fach J, Longy-Boursier M, Begaud B. Adverse drug reactions in a department of systemic diseases-oriented internal medicine: prevalence, incidence, direct costs and avoidability. Eur J Clin Pharmacol 2000;56:181–6. [16] Hepler CD, Strand LM. Opportunities and responsibilities in pharmaceutical care. Am J Hosp Pharm 1990;47:533–43. [17] Lexi-Comp Online, Lexi Interact Online. Hudson, Ohio: Lexi-Comp, Inc. [18] Interactions médicamenteuses, cytochromes P450 et P-glycoprotéine (Pgp). Genève: Service de pharmacologie et toxicologie cliniques. Hôpitaux Universitaires de Genève; 2011. [19] Credible Meds, QTDrugs Lists. Arizona: AZCERT, Inc. [20] Swissmedic, Information sur le médicament. Bern: Swissmedic. [21] Lexi-Comp Online, Lexi Drug Online. Hudson, Ohio: Lexi-Comp, Inc. [22] DRUGDEX System (Micromedex 2.0). Greenwood Village, CO: Truven Health Analytics. [23] Aronoff GR, Berns JS, Brier ME, Golper TA, Morrison G, Singer I, et al. Drug prescribing in renal failure: dosing guidelines for adults. 4th Revised ed. Philadelphia, PA: American College of Physicians; 1999. [24] Aronson JK, Dukes MNG. Meyler's side effects of drugs. 15th revised ed. Amsterdam, Netherlands: Elsevier; 2006. [25] McEvoy GK, Snow EK. AHFS: drug information. Bethesda, MD: American Society of Health-System Pharmacists; 2012. [26] Trissel LA. Handbook on injectable drugs. 17th ed. Bethesda, MA: American Society of Health-System Pharmacists; 2013. [27] Gallagher P, Ryan C, Byrne S, Kennedy J, O'Mahony D. STOPP (Screening Tool of Older Person's Prescriptions) and START (Screening Tool to Alert Doctors to Right Treatment). Consensus validation. Int J Clin Pharmacol Ther 2008;46:72–83. [28] Dooley MJ, Allen KM, Doecke CJ, Galbraith KJ, Taylor GR, Bright J, et al. A prospective multicentre study of pharmacist initiated changes to drug therapy and patient management in acute care government funded hospitals. Br J Clin Pharmacol 2004;57: 513–21. [29] Blix HS, Viktil KK, Moger TA, Reikvam A. Characteristics of drug-related problems discussed by hospital pharmacists in multidisciplinary teams. Pharm World Sci 2006;28:152–8. [30] Bondesson A, Eriksson T, Kragh A, Holmdahl L, Midlov P, Hoglund P. In-hospital medication reviews reduce unidentified drug-related problems. Eur J Clin Pharmacol 2013;69:647–55. [31] Mannheimer B, Ulfvarson J, Eklof S, Bergqvist M, Andersen-Karlsson E, Pettersson H, et al. Drug-related problems and pharmacotherapeutic advisory intervention at a medicine clinic. Eur J Clin Pharmacol 2006;62:1075–81. [32] Blix HS, Viktil KK, Moger TA, Reikvam A. Identification of drug interactions in hospitals—computerized screening vs. bedside recording. J Clin Pharm Ther 2008; 33:131–9. [33] McDonnell PJ, Jacobs MR. Hospital admissions resulting from preventable adverse drug reactions. Ann Pharmacother 2002;36:1331–6. [34] Guthrie B, Inkster M, Fahey T. Tackling therapeutic inertia: role of treatment data in quality indicators. BMJ 2007;335:542–4. [35] Touze E, Coste J, Voicu M, Kansao J, Masmoudi R, Doumenc B, et al. Importance of inhospital initiation of therapies and therapeutic inertia in secondary stroke prevention: IMplementation of Prevention After a Cerebrovascular evenT (IMPACT) Study. Stroke 2008;39:1834–43.
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
8
B. Guignard et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
[36] Galindo C, Olive M, Lacasa C, Martinez J, Roure C, Llado M, et al. Pharmaceutical care: pharmacy involvement in prescribing in an acute-care hospital. Pharm World Sci 2003;25:56–64. [37] Patel HR, Pruchnicki MC, Hall LE. Assessment for chronic kidney disease service in high-risk patients at community health clinics. Ann Pharmacother 2005;39:22–7. [38] Miedema MD, Cohn JN, Garberich RF, Knickelbine T, Graham KJ, Henry TD. Underuse of cardiovascular preventive pharmacotherapy in patients presenting with ST-elevation myocardial infarction. Am Heart J 2012;164:259–67.
[39] Shekelle PG, MacLean CH, Morton SC, Wenger NS. Acove quality indicators. Ann Intern Med 2001;135:653–67. [40] Ing Lorenzini K, Reutemann B, Samer CF, Guignard B, Bonnabry P, Dayer P, et al. Comparison of four drug interaction screening programs. Rev Med Suisse 2012;8: 1978–82.
Please cite this article as: Guignard B, et al, Drug-related problems identification in general internal medicine: The impact and role of the clinical pharmacist and pharmacologist, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.05.012
