Patterns of Medication Use among Romanian Nursing Home Residents ()
1. Introduction
The elderly are frequently exposed to drug-related problems (DRP) and the concepts of potential inappropriate prescribing (PIP) and of potential inappropriate medications (PIM) are used in the analysis of their medicationrelated data; these terms refer to the recommendation of drugs for which the risk of adverse events is higher than the clinical benefit, in a context where safer, more effective or more cost-effective therapies exist [1] -[3] . Three main subtypes of PIM or PIP have been suggested: the misuse of medications (misuse-PIP) referring to the choice of drug, dose, drug interactions, duration of therapy, duplication, or follow-up, the underuse of a therapy (underuse-PIP) proved to be effective in preventing or treating a condition that the elderly patient might have, and the overuse of medications (overuse-PIP) that lack valid indications or proof of effectiveness [4] [5] . Several sets of explicit criteria representing PIP instances have been developed and applied in studies conducted in diverse environments of elderly care: the Beers criteria, the STOPP and START tools, the PRISCUS list etc. [3] [5] -[7] , serving a double intent: to identify DRP (retrospective approach) or to guide the use of medication in this population (prospective approach).
The risk for PIP is emphasized in the elderly living in nursing homes and the clinical relevance of such PIP practices was confirmed for both continuous and intermittent use of PIM, being associated with an increased risk of hospitalization and emergency department visits, an increased mortality and a higher risk of adverse drug events [8] -[10] .
Published information describing the medication use patterns in the Romanian nursing-homes is not available. Life expectancy in the Romanian population is among the lowest in the European Union and the elderly hospitalization rate has a 17.82% annual increase [11] [12] . Referring to the nursing-home environment, present legislation states that one of its main objectives is to supervise and to facilitate the necessary medical care, but there is a lack of explicit recommendations meant to ensure the appropriateness of medications used [13] . On the other hand, studies conducted in different environments of care and applying various quality assessment criteria, suggested that significant differences could exist between the European countries concerning the prevalence of PIP in the elderly, as a consequence of variable prescribing patterns and socioeconomic factors [14] . Consequently, similar investigations in other European countries are encouraged [15] .
The purpose of this observational, retrospective study was to identify the medication use patterns in a sample of Romanian nursing home residents, including prevalence and subtypes of PIP.
2. Methodology
2.1. Study Population and Data Collection
The information analyzed in this retrospective, cohort, and pilot study was the one available for the elderly residents (≥65 years old) admitted to one private nursing home in an urban area of the Cluj County, Romania during a three months interval (March to May 2011).
A general practitioner daily attended the facility and a registered nurse was available 24 hours a day. One clinical pharmacist reviewed the residents’ files, their available laboratory tests or the nurses’ notes recording the drug therapy changes recommended by the physician and also filled a data collection-chart for each resident. Laxative use, vitamins and complementary medicines, dressings and topical medications were occasionally used, but they were not constantly mentioned in the residents’ files or nurses’ notes, so they were not considered for the analysis. The reason for the admission in the facility and the date of admission were not explicitly mentioned in the chart. The cost of the medications administered to the patients was included in the monthly contribution of each resident. The physician obtained the medications from a local community pharmacy and filled the patients’ pill organizers. The nurses and formal care-givers administered the medications. The non-interventional nature of this study and the measures taken to ensure the confidentiality of the patients medical information, were presented to the nursing home representatives who formally accepted to participate to the study. The Ethics Committee approval was not required, considering the retrospective and the non-interventional nature of this investigation and the actions taken to preserve the confidentiality of the analyzed information.
2.2. Data Analysis
For the analysis of the residents’ data, the published available literature covering elderly medication use was considered, including four explicit criteria of misuse, underuse and overuse of medications in this population [5] -[7] [16] [17] . This approach was considered necessary for the identification of the local particularities of the medication use among the elderly, as prior analysis of the medication use patterns in this population was not available to guide a more specific approach. A coding system was developed to facilitate the analysis of the PIM identified, grouping the PIP instances in four large categories: misuse-PIP, underuse-PIP, lack of monitoringPIP, overuse-PIP. Medication use was described using the Anatomic Therapeutic Classification [18] . The residents’ degree of functional independence was assessed considering information pertaining to his/her ability to ambulate and the presence of a diagnosis of severe dementia with a MMSE (Mini Mental State Examination) score of 3 - 10; validated functional assessment tests could not be used as the residents’ charts did not contain the necessary information. Descriptive statistics were used (Microsoft® Office Excel 2007, Microsoft Corporation).
3. Results
91 elderly residents were included in this study. Their demographic characteristics are summarized in Table 1. 6 residents had 85 years old and 22 had >85 years old. 19 residents of this oldest old subgroup (≥85 years old) (67.85%) were functionally dependent, while 21 (75%) had dementia of different degrees of severity. The most frequently prescribed medications were the cardiovascular ones, with nitrates as the most prescribed category (52 residents, 59.09% of total) (Table 2); the medication use patterns inside each age group were similar, the sole exception being digoxin which was more frequently administered to the oldest subgroup.
yo years old; a. estimated with the Cockroft Gault Equation using the last creatinine assessment available; 73 residents had this information mentioned in their chart; b. the resident was considered dependent it he/she was unable to ambulate or if he/she had a diagnosis of severe dementia with a MMSE (Mini Mental State Examination) score of 3 - 10; otherwise, the resident was considered functionally independent. c. Dementia mainly Alzheimer, vascular or mixed type; if present, then status = 1.
Table 2. The most frequently prescribed medications (to ≥ 5% of residents) inside each age group.
yo, years old, The Anatomical Therapeutic Chemical (ATC) classification system; a. adjuvants = vitamin B complex and thioctic acid were locally available, with an indication for the management of neuropathic pain.
39 residents (42.86%) received a psychotropic medication (a benzodiazepine, an antipsychotic or a drug with anticholinergic properties). 18 (19.78%) of all residents and 17 (29.31%) of those with a dementia diagnosis (n = 58) received an antipsychotic medication on a chronic basis. Among the dementia subgroup, 2 residents (3.45%) used medications with anticholinergic properties. Antidementia drugs (memantine, donepezil, rivastigmine) were prescribed to 63.79% of the residents diagnosed with dementia (37 out of 58 residents). Nootropics (Ginkgo biloba standardized extract, nicergoline, piracetam, vinpocetine) were used by 44.82% of the residents with dementia (26 out of 58 residents); 25.86% of this subpopulation was managed with both antidementia drugs and nootropics (15 out of 58 residents); nootropics were still prescribed to 26.98 % (n = 17) of the functionally dependent residents. Nonsteroidal anti-inflammatory drugs (NSAIDs) were chronically used as analgesics by 40% of those with a CrCl < 30 ml/min (4 out of 10 residents).
253 PIP were identified in the entire sample of 91 residents, with an estimated mean (±SD) of 2 (±1.41) PIP per resident, distributed as 117 misuse-PIP (46.24%), 60 underuse-PIP (23.72%), 41 lack of monitoring-PIP (16.21%) and 35 overuse-PIP (13.83%). PIP prevalence increased with age and all residents > 85 years old had at least one PIP instance. The distribution of the most frequent PIP instances inside each age group is presented in Table 3.
The underuse of osteoporosis and fracture prevention therapies (calcium and vitamin D) was a PIP instance recorded for all nursing home residents; excluding this PIP category, 87 residents (95.60%) had at least one PIP instance. 63 residents (69.23%) had at least one misuse-PIP. 27 residents (29.67%) received NSAIDs as chronic analgesic treatment, while 21 elderly (23.08%) had a pain associated condition and no analgesic treatment. Lack of an annual assessment of the renal function was noted for 28 residents (30.77%), respectively for 3 (27.27%) of the residents with an estimated CrCl < 30 ml/min and for 7 (14.89%) of those with a CrCl of 30 - 59ml/min.
Table 3. Five most frequent potential inappropriate prescribing (PIP) category inside each age group.
Among the sub-group of 28 functionally independent residents, 5 of them (17.86%) had analgesics under-prescribed and 3 of them (10.71%) lacked control of arterial hypertension (registered values were constantly over 160/100 mmHg).
4. Discussion
This is the first Romanian study aiming to identify the medication use patterns and the most frequent PIP instances, in a sample of elderly nursing-home residents. Compared to other relevant similar research, we found a higher use of cardiovascular medications, with 96.70% of patients receiving them, compared to 63% reported in a recent British study [19] or 71.1% presented in a Malaysian investigation [20] . Moreover, dementia seemed intensively treated, with a 51.65% use of antidementia molecules and a 41.76% rate of administration of nootropic drugs, compared with a 5.91% antidementia drug use in a Swedish cohort [21] . NSAIDs were prescribed to 29.67% of the residents, while Australian or British investigations suggested different preferences for analgesic management, with a 1.2% or 3.8% frequency of NSAIDs chronic use [22] [23] . On the other hand, we noted an apparent lower use of psychotropic medications (42.86% compared to a 65% prevalence estimated in an US nursing home sample [24] ), a generalized lack of use of calcium and vitamin D as osteoporosis and fall prevention strategies (compared to a 28.5% recommendation in a British nursing home-based investigation) and a limited rate of annual creatinine assessment (30.77%).
Previous studies suggested a variable European prevalence (12.5% to 66%) for misuse-PIM in different settings of care and our results seem to be congruent with this European estimation (69.23% of residents had at least one misuse-PIM) [25] . The PIM number increased with age, so that every older resident ≥85 years old included in this study presented at least one PIM. This contrasts with findings from other studies that identified a lower PIM use among the “oldest old”, independently of the assessment criteria used [21] [23] [25] . A more cautious therapeutic approach could be necessary in our sample, considering life expectancy and functionality, as among the residents ≥85 years old, 67.85% were functionally dependent and 75% had various stages of dementia.
Almost 20% of the residents included in our study received antipsychotic medications, on a long term basis, with the youngest subgroup being the most exposed. Among the subgroup with dementia, close to a third was exposed to antipsychotic medications. Haloperidol was the most frequent antipsychotic prescribed, probably because it was the least expensive available. In other nursing home-based, cross-national studies, the reported prevalence of antipsychotics use, varied from 11% to 38% and it was associated with various instances of inappropriate use: long-term prescribing, use despite risk of falling, combined use with other psychotropics, duplicate use [26] [27] . Considering local patterns of medication use and published similar research, it is possible to suspect the use of antipsychotics for the management of the behavioral and psychological symptoms of dementia and also as sedating and hypnotic drugs [28] [29] . Increasing age favored an increased use of benzodiazepines with long half-lives, zolpidem or zopiclone (23.08% of all residents), but the identified prevalence is lower than the one reported in other European analyses (>50%) [30] [31] .
A particularity of medication use in our study was the relative low use of anti cholinergic drugs, with variable anti cholinergic strength (trihexyphenidyl, oxybutynin, mirtazapine): only 8.79% of the patients were exposed, with the youngest subjects and those taking more than 11 drugs daily, as the most predisposed subgroups [32] . It confirms a similar European pattern of lower anti cholinergic use in the elderly (20.7%), compared with available North American reports (73.62%) and explained through a lower availability for use [21] [33] . Trihexyphenidyl was the most frequently prescribed anticholinergic drug, mainly used to correct the Parkinsonian side-effects of the antipsychotics frequently prescribed to this subgroup.
Pain management was associated with an increased risk for side-effects and with potential underuse. Firstly, 29.67% of all patients received NSAIDs (mostly diclofenac and ketoprofen) on a constant basis, for the management of osteoarthritis pain. This high chronic use of NSAIDs contrasts with findings from other studies that reported a more guideline endorsing approach, with paracetamol or carefully selected opioids as the first-choice chronically used analgesics and with a low use (7%) of NSAIDs [17] [34] [35] . The residents (≥85 years) were the most exposed to the use of NSAIDs (31.82% of them), which were the single form of analgesic therapy recommended to this subgroup, although it is the age-group the most susceptible to their side-effects. On the other hand, 23% of patients had a diagnosis suggesting a subtype of chronic pain, mostly osteoarthritis, but no treatment mentioned in the chart. Pain is undertreated and frequent among the nursing home residents, with a reported prevalence of 48.4% in a recent European study and being estimated to vary word-wide from 3.7% to 79.5%, depending on the research methods used [36] [37] .
Compared to a 4% cumulative antidementia medication use during a 2 years Dutch study follow-up, the prescription of antidementia medication in our study, was appreciably higher, as 64% of the demented patients were receiving this therapy [38] . This pattern of prescribing could suggest overuse, as the magnitude of the clinical effect of antidementia drugs, is currently uncertain, although recommended by the local dementia guidelines available at the time of the data collection. Furthermore, there is little evidence to support the use of cholinesterase inhibitors as monotherapy of behavioral or psychological symptoms in patients with dementia, so questions arise concerning the efficacy of these therapies in the studied population, admitting the fact that a stratified analysis considering specific indication was not conducted [39] .
The residents received an average number of 8 daily administered medications, a degree of medication use that can be associated with the concept of polypharmacy considered to be highly prevalent in the nursing home environment [40] . The risk for polypharmacy is increased by the overuse of medications with debatable efficacy, especially in those with a reduced life expectancy [41] and in this context, nootropics were recommended to 25.86% of the demented subgroup that already received an antidemential specific drug and to 26.98% of the functionally dependent individuals. Advanced dementia has been associated with a reduced survival and although our patients were not stratified according to dementia severity, the therapeutic benefit of nootropics, can be questioned as proved by their inclusion in the PRISCUS criteria considered for this analysis [8] [42] .
To positively influence the outcomes of the identified PIP, several strategies have been suggested by previous published experiences: various educational approaches for the members of the staff caring for the elderly, computerized clinical decision-support systems or multidisciplinary case-conferencing [43] [44] . The clinical pharmacist has been involved in such interventions, as a member of a multidisciplinary team and has succeeded in the improvement of the appropriateness of the medication used, through regular medication reviews or collaboration with the prescribing physicians and the facility nurses [45] . For the moment, the Romanian pharmacist role as a member of the health-care team is theoretically acknowledged, but not routinely accepted or implemented. To increase his contribution to a safer medication use among the elderly, the adoption of a more geriatric-oriented approach on pharmacotherapy could be implemented at a local level, as it already proved to be a successful strategy in this direction [46] .
Our pilot study has several limits that suggest caution in the generalization of the results. Firstly, the information analyzed was obtained from a sample of 91 elderly residing in a single nursing home and does not allow for the appropriate investigation of the possible correlations between various factors and the PIP instances identified. Secondly, the gathered data was interpreted by a single clinical pharmacist. The general and the medical care received by the patients might not be representative for all Romanian nursing-homes, as the facility was attended daily by a general practitioner. The treatments prescribed in the patients files or in the nurses charts were analyzed and not the medication that was actually taken by the patient, who might, for example, refuse to accept the administered therapies. The information available for each resident was not stored in a uniform manner, further limiting the interpretation of the data. Subsequent interventions in this environment of care should consider these methodological aspects requiring improvement. A larger sample of elderly nursing-home residents will be needed and perhaps, an active participation of the pharmacist along the medical-team involved in the elderly’s care.
5. Conclusion
The studied sample of elderly nursing-home residents may not be representative for all similar Romanian facilities, but it offers a first image of the existing medication use patterns. Local availability of medications included in explicit instruments of medication appropriateness evaluation, locally adopted guidelines or prescribing practices could explain a lower use of psychotropic medication, fall preventive and analgesic therapies with a higher use of cardiovascular, dementia medications and nootropics. Prevalence of potentially inappropriate instances of medication use was similar to other European estimates, while polypharmacy was a frequent prescribing practice. Adjustments to the medication use patterns seem necessary in this environment of care, but larger-scale studies are needed to confirm this hypothesis.
Acknowledgements
The study was conducted during an AUF (Agenceuniversitaire de la Francophonie) scholarship for Ph.D. students (2010-11). The authors are also grateful to the nursing-home staff, providing the data included in this analysis.
Abbreviations and Acronyms
DRP, drug-related problems;
NSAIDs, Nonsteroidal anti-inflammatory drugs;
PIM, potential inappropriate medications;
PIP, potential inappropriate prescribing;
PRISCUS, Latin for “old and venerable”;
START, Screening Tool to Alert doctors to Right Treatment;
STOPP, Screening Tool of Older Person’s Prescriptions.