1. Introduction
Transfer of care is a vulnerable process that increases the risk of drug-related problems, especially in the elderly. The World Health Organisation (WHO) launched the third Global Patient Safety Challenge with aim of improving the medication safety, “Medication Without Harm”, where the key areas of the challenge represent the transfer of care, polypharmacy, and high-risk situations. High-risk situations include hospital treatment, and drug management in elderly patients, as well as those with concomitant renal impairment. The WHO’s goal is to reduce severe avoidable harm related to medications by 50% [
1].
More than 20% of the European Union (EU) population are the elderly, and by 2050, this figure will double at the global level [
2,
3]. Population aging, higher incidence of chronic diseases, and the application of multiple guidelines are some of the reasons for the growing trend of prescribing drugs. Polypharmacy, usually defined as the use of ≥5 medications, is rising, especially in the elderly [
4]. Polypharmacy is the main risk factor for adverse drug events (ADE)-related hospitalization in older adults and has been linked to the high risk of potentially inappropriate medication (PIMs) use and drug-drug interactions (DDIs) [
5,
6]. PIMs are defined as medications whose adverse risks outreach their positive therapeutic effects when compared to alternative therapies [
7]. The use of PIMs is prevalent among the elderly and is affiliated with an increased risk of adverse health outcomes [
8]. Apart from PIMs, DDIs represent a major problem for therapy management, especially in the elderly, as they can compromise patient safety [
6]. It is estimated that DDIs cause up to 5% of hospitalizations in elderly patients [
9].
Treatment of chronic kidney disease (CKD) represents a global health burden with a significant share of health care costs [
10]. Older adults with CKD are highly vulnerable to polypharmacy [
11]. The aging process and renal impairment modify the clinical drug profile, which results in a 3–10 times higher incidence of adverse drug reactions in older adults with CKD than in those without it [
12]. In order to avoid the occurrence of ADEs, optimal drug selection and dosing modification should be carried out for renally cleared and potentially nephrotoxic drugs in the elderly. The recent systematic review reported the use of a contraindicated medication or inappropriately high dose according to kidney function ranging from 9.4% to 81.1% [
13].
With the length of hospital stay the number of medications increases, which complicates pharmacotherapy management and jeopardizes patients’ safety [
14]. The detection of pharmacotherapy problems upon admission to the hospital is important for further pharmacotherapy optimization, especially when new medications are introduced into therapy.
Medication reconciliation is a relevant safety procedure in medication management during the transition of care which has also been proven as a cost-saving measure [
15]. Medication reconciliation is the process of creating the Best Possible Medication History (BPMH) and comparing the list with orders written at each transition of care. BPMH is an accurate and thorough list of medications that a patient is taking, including dose, frequency, and administration route, which may vary from other healthcare records [
16]. Guidance states that BPMH in an acute setting should be completed as soon as possible, within 24 hours, upon transfer of care [
17]. BPMH is a useful instrument for detecting unintentional pharmacotherapy discrepancies (drug omission, addition, substitution, incorrect dose, frequency, and route of administration) [
18,
19]. However, the use of BPMH should be expanded for the detection of a wider spectrum of the above-mentioned pharmacotherapy problems (polypharmacy, PIMs, DDIs, renal risk drugs-RRDs) in order to increase patient safety.
The aim of the study was to evaluate polypharmacy, PIMs, DDIs, and RRDs by using the BPMH on hospital admission and to determine their mutual relationship and association with patient characteristics.
4. Discussion
Transfer of care is a sensitive process that increases the risk of ADEs, especially in the elderly. In our study, a clinical pharmacist, using multiple sources of information, obtained and evaluated BPMH for 383 hospitalized older patients admitted to the Internal Medicine Clinic. This research determined a very high incidence of polypharmacy, PIMs, DDIs, and inappropriately prescribed RRDs detected in BPMH. To the best of our knowledge, this is the first study to explore the incidence and types of polypharmacy, PIMs, DDIs, and inappropriately prescribed RRDs by using BPMH, and to determine their mutual relationship and association with patient characteristics.
A high level of polypharmacy was identified in in this research. According to a recent review by Khezrian et al., the prevalence of polypharmacy varied between 10% and 90% [
26]. Polypharmacy represents one of the major challenges for the healthcare system. It has increased markedly in recent years and is still increasing as more people suffer from chronic diseases [
26,
27]. In this study, a higher number of medications was determined in the BPMH upon admission compared to the study conducted in 2016 in the same clinical setting but included the general population (8 vs. 6 per patient) [
18]. Our regression analysis showed the association of excessive polypharmacy with the number of diagnoses in addition to impaired renal function and recent hospitalization. Patients with renal impairment often experience polypharmacy especially in the later stages of CKD when patients develop numerous metabolic complications that require the prescription of multiple drugs according to guidelines [
11,
28]. The association between recent hospitalizations and polypharmacy in elderly patients can be explained by the fact that patients with a weaker health status have more complex pharmacotherapy and experience hospitalizations more frequently [
29]. In our research, polypharmacy positively correlated with detected pharmacotherapy problems: PIMs, DDIs, and inappropriately prescribed RRDs and had the highest Spearman coefficient.
The BPMH is a valuable source of information for deprescription. The most important instruments for deprescription are PIM tools. The EU(7)-PIM list employed in this study presents the most comprehensive and up-to-date tool for the evaluation of PIM prescribing in Europe. It is specifically designed to cover the European drug market more appropriately than the other existing PIM criteria [
7]. The prevalence of PIM use was 80% in our study which is higher than the prevalence reported in European studies in the non-hospital environment [
30,
31], and also higher than the prevalence determined in the study conducted in 2017 in the Clinical Hospital Dubrava [
32]. A recently published study that enrolled hospitalized older patients at an internal medicine ward in Portugal detected a similar prevalence of EU(7)-PIMs (79.7%) [
14]. Furthermore, analysis of this study showed that multiple medication use was the strongest predictor for PIMs, which is in line with the findings of Guillot et al. [
33] and others [
34,
35,
36]. Our results showed that recent hospitalization was also a risk factor for PIMs use. Hospitalizations increase drug use which also increases the risk of PIM prescribing. Regular evaluation of pharmacotherapy after hospitalization is necessary.
The most frequently detected EU(7)-PIM drugs were proton pump inhibitors (PPIs) (40%), previously detected in numerous studies [
32,
33,
37]. EU(7)-PIM list consider PPI use for more than eight weeks as inappropriate for prescribing in the elderly. Long-term use of PPIs is associated with an increased risk of Clostridium difficile colitis, parietal cell hyperplasia, myopathy caused by hypomagnesemia, respiratory infections, osteoporosis-related fractures, and tubulointerstitial nephritis [
7,
38,
39]. Elderly patients with CKD are considered to be at even higher risk of adverse effects from PPIs [
11]. A study that included 2.6 million subjects outlines that PPI use was associated with a significantly increased risk of developing CKD [
40]. PPIs have been highlighted as one of the three specific targets for medication optimization and deprescribing in older adults with CKD [
11]. Indication for PPI use is not always clear, and its dosage and duration of use should be regularly reevaluated, especially in the elderly with renal impairment [
11].
Our research found a high prevalence of potential clinically significant (C, D, X) DDIs upon admission (90.6%). Regression analysis showed that women have a higher risk for D interactions and a lower risk for X interactions as opposed to men which could be explained by the fact that drugs that depress the CNS, most often represented in D interactions, are more commonly used by women [
41]. On the other hand, a lower risk for X interactions found in women can be explained by the fact that the most common interactants in X interactions were drugs indicated for chronic obstructive pulmonary disease (COPD), a condition more prevalent in men [
42]. The analysis also showed a lower risk of X interactions in patients with impaired renal function, which would indicate the fact that these drugs are prescribed cautiously in this vulnerable group of patients. Cox and Snell R2 and Nagelkerke R2 measures were used to fit models in logistic regression. In terms of models, the best model was overall DDI and the weakest model was PIMs.
The most common potential consequence of the identified X interactions was an increased anticholinergic effect, with its side effects particularly high in elderly patients. Additionally, the most commonly identified clinically significant interaction between perindopril and indapamide carried an increased risk of nephrotoxicity. This result is of particular importance considering the fact that more than 40% of patients upon admission had impaired renal function (eGFR < 60 mL/min/1.73 m
2). Recent research conducted in Croatia that included 1211 patients also found this interaction to be the most common [
43]. The risk of acute renal impairment is higher when nonsteroidal anti-inflammatory drugs (NSAIDs) are added to therapy [
44]. By obtaining the BPMH, a high prevalence of NSAIDs, most commonly involved in D interactions, was found. BPMH is also a key tool for OTC detection as they are not usually registered in medical documentation.
Only 11% of the study sample had normal renal function (KDIGO G1) implying the need to reconsider RRD use in the BPMH already on hospital admission. A lower level of renal function for certain drugs may require therapy adjustment and increase the risk of adverse drug events [
45]. Our results showed that 64.7% of elderly patients with stages of renal impairment G3–G5 had inappropriately prescribed RRD. The prevalence was higher than the prevalence reported in an American study of elderly patients with CKD stages 3–5, but it was lower than the prevalence reported in a French study of patients aged ≥75 with eGFR < 20 mL/min/1.73 m
2 [
28,
46].
This study compared the prescribing of inappropriately prescribed RRDs in patients with impaired renal function according to gender. Regression analysis showed that women were at higher risk of having inappropriately prescribed RRDs, which puts them at higher risk of ADEs. The epidemiology of CKD differs by gender; it reports a higher prevalence of CKD in women compared to men [
47]. Faster renal function decline in men compared to women and longer life expectancy in women can partially explain the gender difference in CKD epidemiology [
48]. Another risk factor contributing to the inappropriate RRD use detected in this study was the number of medications, as shown in previous studies [
46,
49,
50].
ACEIs were the most common RRDs in this study. Inappropriate prescription of ACEIs in elderly patients with renal impairment has been noted in other studies [
51,
52]. ACEIs are considered superior to ARBs and other antihypertensive agents in reducing adverse renal events in non-dialyzed CKD 3–5 patients, however, the prerequisite is that they are used appropriately [
53,
54].
This research had certain limitations. The study included one hospital, one clinic, and was observed at only one point of care transition. Further research should also evaluate the scope of polypharmacy, PIMs, DDIs, and RRDs upon hospital discharge. The study did not include surgical patients who are considered as patients requiring more complicated therapeutic management and future research should broaden the research focus.
Despite limitations, our study suggests that the BPMH is a useful tool for detecting a wider spectrum of pharmacotherapy problems. High incidence of PIMs, DDIs, and inappropriately prescribed RRDs indicate the need for their early detection. Detection of pharmacotherapy problems upon admission is one of the crucial steps for therapy optimization during a hospital stay. The clinical pharmacist has specific pharmacotherapy knowledge and therefore, can significantly contribute to pharmacotherapy rationalization and patient safety. Although there are different decision support systems for detecting pharmacotherapy problems, they cannot adequately replace medication reconciliation and clinical pharmacists’ professional interpretation of data [
55]. Decision support systems especially cannot replace a clinical pharmacist when evaluating a wider spectrum of pharmacotherapy problems, which are all positively intercorrelated and will probably occur simultaneously. We should strive for the BPMH and clinical pharmacists’ evaluation of BPMH upon admission to become the standard of health care in order to prevent the transfer and circulation of pharmacotherapy problems and to increase patient safety.