Abstract
Background:
With multimorbidity increasing among older people, polypharmacy and the use of potentially inappropriate medications (PIMs) are assuming a prominent role in the life of the geriatric population.
Objective:
To investigate the association of polypharmacy and PIM use with a wide range of factors in older people with mild cognitive impairment (MCI) to mild dementia.
Methods:
The study population comprised 160 outpatients with a Clinical Dementia Rating of 0.5–1 and a Mini-Mental State Examination score of 20–30. Patients were classified as receiving polypharmacy when they took ≥5 different medications at the same time. PIMs were identified using the STOPP-J criteria. Cognitive, neuropsychological, nutritional, and physical function tests were performed and body measurements taken. Quality of life (QOL) was assessed using both components of the EQ-5D scale, the index score and the visual analogue scale (QOL VAS). A comorbidity index was calculated for all participants.
Results:
PIM use was significantly associated with lower scores on the verbal fluency (initial letters) test and QOL index. Participants receiving polypharmacy showed an increased likelihood of worse frailty status and lower QOL VAS score. The number of medications was significantly associated with a worse frailty status.
Conclusion:
In a geriatric population with MCI to mild dementia, PIM use was associated with lower verbal fluency (initial letters) score and lower QOL, while the presence of polypharmacy was correlated with a worse frailty status and lower QOL. The number of medicines, instead, was correlated with a worse frailty status only.
INTRODUCTION
In recent decades, thanks to medical advances and sociodemographic changes, most diseases have shifted from being incurable to being chronic, with consequent longer life expectancy [1] and more older persons in the population. Various diseases tend to coexist now and the resulting multimorbidity requires the use of multiple medications.
Polypharmacy, defined as the use of various medications simultaneously [2], increases the risk of receiving potentially inappropriate medications (PIMs) [3]. PIMs are drugs with risks that exceed their clinical benefits, particularly when a more effective and less risky treatment is possible [4]. Previous research has shown that polypharmacy and use of PIMs are widespread in the geriatric population, with each having a prevalence of more than 50% [5, 6]. The effect of polypharmacy and PIM use on the lives of older people becomes even more relevant if we consider that, with aging, older people are affected by physiological changes that lead to alterations in drug pharmacokinetics and pharmacodynamics [7]. As such, polypharmacy and PIM use may increase the risk of adverse drug reactions, adverse drug events, and drug–drug interactions [8, 9] and could cause serious medical problems such as increased hospitalizations, increased costs, falls, and death [10]. In addition, how PIM use and polypharmacy relate to the concept of quality of life (QOL) in older people is extremely important to consider, because all of the physiological age-related changes experienced by older people [11] make them more sensitive to mental problems and physical decline, which can in turn affect their QOL.
Nowadays, adverse drug reactions in older persons represent a serious and escalating problem in public health [12]. Several criteria have therefore been developed to help estimate the appropriateness of some drugs over others. Two approaches are the Beers criteria and the more recently developed screening measure STOPP (Screening Tool of Older Person’s Potentially Inappropriate Prescriptions). The STOPP criteria include drug–drug and drug–disease interactions, appropriate clinical indications, and correct treatment dose and duration [13]. Some studies have determined that, compared with the Beers criteria, STOPP better identifies PIM use in older patients [14]. However, because the classification system for medications and their use differ by country [15], the Beers and STOPP criteria have been used to develop country-specific PIM criteria. Japan uses the Screening Tool for Older Person’s Appropriate Prescription for Japanese (STOPP-J), developed by the Japan Geriatrics Society [16].
Many studies have tried to analyze the effect of polypharmacy and PIM use in a geriatric population. Polypharmacy was found to be associated with frailty [17], lower QOL [1], and cognitive impairment and a lower score on the Mini-Mental State Examination (MMSE) [18]. Meanwhile, PIMs were found to be associated with low cognitive performance and functional impairment [19] and, in some studies, with lower QOL [20].
It is particularly important to consider the effects of polypharmacy and PIM use on cognition in patients with dementia and mild cognitive impairment (MCI). Both patient groups often experience worsening of their QOL [21, 22] and, therefore, polypharmacy and PIMs could place even more burden on them in everyday life.
The aim of this research was to study the association of polypharmacy and PIM use with different factors in older patients, focusing on people in the early stages of cognitive decline, MCI to mild dementia. This population was selected because these patients are particularly vulnerable, due to the fact that the presence of various factors or diseases could increase their risk of cognitive function decline and promote progression of dementia. Furthermore, to take as global an approach as possible, this study took into account a wide range of factors to examine the association of polypharmacy and PIM use with them.
METHODS
Study design and participants
This cross-sectional study involved 160 older Japanese adults (≥64 years) treated as outpatients at the Memory Clinic, Nagoya University Hospital, Japan. The participants had a Clinical Dementia Rating [23] of 0.5–1 and MMSE score of 20–30, indicating MCI to mild dementia. Patients who had severe medical or psychiatric disorder, symptomatic cerebrovascular disease, or worse cognitive impairment were excluded. For each participant, the Charlson Comorbidity Index (CCI) [24] was calculated. This comorbidity index is obtained from the sum of the weighted scores of all of the patient’s comorbidities. Each comorbidity category is adjusted based on the mortality risk and has a weight that ranges from 1 to 6, with a zero score indicating no comorbidities present. Some of the comorbidities considered are ischemic heart disease, congestive heart failure, cerebrovascular disease, diabetes mellitus, dementia, cancer, and neurodegenerative disorders, including Parkinson’s disease.
Informed consent was obtained from all participants included in the study. The study protocol was approved by the Ethics Committee of the Graduate School of Medicine, Nagoya University.
Measurements
Neuropsychological assessments
Several neuropsychological assessments were carried out in the context of this study. Memory was evaluated using the Logical Memory I and II subtests and Visual Reproduction I and II subtests of the Wechsler Memory Scale-Revised [25], the MMSE [26], and the Alzheimer’s Disease Assessment Scale [27]. Working memory was evaluated using the Digit Span subtests (forward and backward) and the Visual Memory Span subtests (forward and backward) of the Wechsler Memory Scale-Revised [28]. Verbal fluency was measured using the letter fluency test and category fluency (animal naming) test. Visuospatial ability was assessed using the quantitative and qualitative clock-drawing tests [29]. Processing speed was evaluated using the Digit Symbol subtest of the Wechsler Adult Intelligence Scale-III [30]. Executive function was assessed using the Trail Making Test (TMT) parts A and B [31] and the Stroop test [32]. Depressive mood was evaluated with the Geriatric Depression Scale 15 (GDS-15) [33].
Health-related QOL
The EQ-5D is a generic and non-disease-specific instrument used to evaluate the current health-related QOL of participants. It consists of two components: a self-administered health index and a visual analogue scale (QOL VAS) [34]. The health index component consists of 5 dimensions: mobility, self-care, pain/discomfort, usual activities, and anxiety/depression. Each of these dimensions has 5 levels of severity: no problems, slight problems, moderate problems, severe problems, and extreme problems. The scores of these 5 dimensions can be combined to obtain up to 3125 possible health states, from which a single index score can be computed.
The QOL VAS is a 100-point scale where the participants score their health-related QOL using a single number ranging from 0 to 100, with 100 representing the best imaginable health state and 0 representing the worst imaginable health state [35].
Body composition and other variables
A bioelectrical impedance measurement system (Inbody 430; Biospace, Seoul, Korea) was used to determine the amount of extracellular and intracellular water in the body and, consequently, each participant’s appendicular skeletal muscle mass. Then, the skeletal muscle mass index (SMI) was calculated by dividing the participant’s muscle mass by his or her height squared (kg/m2).
Frailty was assessed using the Fried criteria [36]: unintentional weight loss, exhaustion, low physical activity, low grip strength, and reduced gait speed. Reduced gait speed was defined as <1 m/s and low handgrip strength as <26 kg for men and <18 kg for women. Participants with at least 3 of the 5 criteria were defined as “frail”, with 1 or 2 criteria as “pre-frail”, and with none of the 5 criteria as “robust”. Here, we use the term “frail phenotype” (FP) to refer to a score ranging from 0 to 5, which amounts to the total number of frailty criteria met by the participant.
Body mass index was calculated and the Mini Nutritional Assessment short-form (MNA-SF) was used to quickly evaluate nutritional status.
Assessments were also made of basic activities of daily living (BADL) using the Barthel index and instrumental activities of daily living (IADL) using the Lawton index. The IADL has 8 items for women and 5 items for men. Each item is scored as 0 (dependent) or 1 (independent), and then the total score is divided by the number of evaluated items to obtain a score ranging from 0 (completely dependent) to 1 (completely independent).
Medication
All medication records were collected. Topical and occasional medications were excluded and only long-term medications (>2 weeks) were considered. Patients were classified as being subject to polypharmacy when they took ≥5 different medications at the same time [37]. PIMs were identified using the STOPP-J criteria [16], and the accuracy of the medication categories determined was independently checked by two pharmacists. The medication categories in the STOPP-J are as follows: antidepressants, antiparkinson drugs, antipsychotics (first- and second-generation), hypnotics (barbiturates, benzodiazepines, non-benzodiazepine receptor agonists), sulpiride, steroids, diuretics, antithrombotic drugs (antiplatelet drugs, anticoagulants), beta-blockers, alpha-blockers, digitalis, first-generation H1 receptor antagonists, H2 receptor antagonists, antiemetic drugs, laxatives, insulin, oral antidiabetic drugs, overactive bladder medications (antimuscarinic receptors), and NSAIDs.
Data analysis
Continuous variables are expressed as the mean and standard deviation, and categorical variables as percentages.
The independent t-test or nonparametric Mann–Whitney U test was used to examine the differences in variables between the participants who took PIMs (PIMs group) and those who did not (no PIMs group) and between participants with polypharmacy (POL group) and without polypharmacy (no POL group). In addition, to evaluate differences in terms of sex, chi-square test was used for the PIMs and no PIMs groups and the POL and no POL groups.
Two multiple logistic regression analyses were performed to identify the independent factors correlated with PIM use and polypharmacy. The forward method was used for the regression of polypharmacy. Moreover, unadjusted and then adjusted linear regression were used to analyze the relation between the number of medicines and the factors found to be associated with polypharmacy in the previous analysis. Variables that showed a p-value less than 0.1 after the independent t-test (or Mann–Whitney U test) were included in these regressions.
All statistical analysis was performed using SPSS software (ver. 25.0). Data are reported as mean±standard deviation. Two-tailed significance tests (α= 0.05) were performed.
RESULTS
Study population characteristics
Participants’ characteristics are shown in Table 1. The 160 participants (73 men, 87 women) had a mean age of 78.3±5.8 years, mean MMSE score of 25.4±2.7, and mean SMI of 6.4±1.0 kg/m2. 20.5% of the participants were frail, 60.3% were pre-frail, and 19.2% were robust. Mean EQ-5D index score was 0.8±0.1 and mean QOL VAS score was 76.7±17.3. PIM use was found in 33.8% of the participants and polypharmacy (≥5 medications) in 41.9% of the participants. Other variables that were taken into account are summarized in Table 1.
Sample characteristics
PIMs, potentially inappropriate medications; FP, frailty phenotype; MMSE, Mini-Mental State Examination; ADAS-J cog, Alzheimer’s Disease Assessment Scale cognitive subscale (Japanese version); WAIS-R, Wechsler Adult Intelligence Scale Revised; GDS-15, Geriatric Depression Scale-15; TMT, Trail Making Test; QOL, quality of life; BADL, basic activities of daily living; IADL, instrumental activities of daily living; BMI, body mass index; SMI, skeletal muscle mass index; MNA-SF, Mini Nutritional Assessment short-form, CCI, Charlson Comorbidity Index.
Table 2 shows the number of medications and PIMs used by the participants. The medications that they used more frequently were: calcium channel blocker (40.0%), angiotensin-converting enzyme inhibitors/angiotensin II receptor blockers (31.8%), statins (28.7%), and proton pump inhibitors (22.5%). The most frequent PIMs, instead, were: antithrombotic drugs (24.3%), hypnotics (21.2%), and oral antidiabetic drugs (13.7%). Among the hypnotics, the percentage of benzodiazepines was 13.7% and that of non-benzodiazepines was 7.5%.
Frequency of used medications and PIMs
PIMs, potentially inappropriate medications; ACEI, angiotensin-converting enzyme inhibitors; ARB, angiotensin II receptor blockers; PPI, proton pump inhibitor; NSAIDs, non-steroidal anti-inflammatory drugs; SSRI, selective serotonin reuptake inhibitors; OAB, overactive bladder.
Group comparisons
Differences between the PIMs and POL groups are shown in Table 3. Compared with the no PIMs group, the PIMs group had a higher FP (p = 0.035) and a higher CCI score (p = 0.009).
Differences, in terms of variables, between the PIMs and polypharmacy groups
PIMs, potentially inappropriate medications; MMSE, Mini-Mental State Examination; ADAS-J cog, Alzheimer’s Disease Assessment Scale cognitive subscale (Japanese version); WAIS-R, Wechsler Adult Intelligence Scale Revised; GDS-15, Geriatric Depression Scale-15; QOL, quality of life; TMT, Trail Making Test; BADL, basic activities of daily living; IADL, instrumental activities of daily living; MNA-SF, Mini Nutritional Assessment short-form; BMI, body mass index; FP, frailty phenotype; SMI, skeletal muscle mass index; CCI, Charlson Comorbidity Index. PIMs group, participants taking PIMs. PIMs group, participants not taking PIMs. No PIMs group, participants without polypharmacy. POL group: participants with polypharmacy. *p < 0.05; **p < 0.01.
The POL group when compared with the no POL group was older (p = 0.004), had a higher FP (p = 0.004), and had higher scores for the GDS-15 (p = 0.007), TMT part B (p = 0.002), and CCI (p = 0.001) as well as lower scores for the WAIS-R Digit Symbol substitution (p = 0.013), QOL VAS (p = 0.013), BADL (p = 0.016), and IADL (p = 0.012). No differences were found in terms of sex between the PIMs group and no PIMs group (p = 0.831) or between the POL group and no POL group (p = 0.855).
Relationships of PIMs and polypharmacy with participants’ characteristics
Tables 4–6 show the results of the two logistic multiple regression analyses and the linear regression performed using only variables with p < 0.1 in Table 3. As shown in Table 4, the use of PIMs was significantly associated with a lower score on the verbal fluency (initial letters) test (odds ratio [OR] 0.801; 95% confidence interval [CI] 0.654–0.981; p = 0.032) and a lower QOL index score (OR 0.001; 95% CI 0.000–0.441; p = 0.025). In the logistic regression (Table 5), the POL group showed a significantly higher FP (OR 2.222; 95% CI 1.296–3.810; p = 0.004) and a lower QOL VAS (OR 0.967; 95% CI 0.936–0.999; p = 0.045). The number of medications was used for the linear regression shown in Table 6, instead of the presence of polypharmacy. This value was found to be significantly associated with FP in both the unadjusted model (p = 0.001) and the adjusted one (β= 0.287, p = 0.022), and with QOL VAS (p = 0.000) only in the unadjusted model.
Logistic regression between PIMs and variables
WAIS-R, Wechsler Adult Intelligence Scale Revised; QOL, quality of life; FP, frailty phenotype; SMI, skeletal muscle mass index; CCI, Charlson Comorbidity Index. *p < 0.05; **p < 0.01.
Logistic regression between polypharmacy and variables
FP, frailty phenotype; WAIS-R, Wechsler Adult Intelligence Scale Revised; GDS-15, Geriatric Depression Scale-15; TMT, Trail Making Test; QOL, quality of life; BADL, basic activities of daily living; IADL, instrumental activities of daily living; CCI, Charlson Comorbidity Index. *p < 0.05; **p < 0.01.
Linear regression between number of medicines and variables
FP, frailty phenotype; WAIS-R, Wechsler Adult Intelligence Scale Revised; GDS-15, Geriatric Depression Scale-15; TMT, Trail Making Test; QOL, quality of life; BADL, basic activities of daily living; IADL, instrumental activities of daily living; CCI, Charlson Comorbidity Index. *p < 0.05; **p < 0.01.
DISCUSSION
The main findings of this study are the significant negative relationships between PIM use and both QOL and verbal fluency (initial letters) scores and between the presence of polypharmacy and QOL. Furthermore, a significant positive relationship was found between polypharmacy (or number of medications) and frailty level. The distinctive characteristic of this study is that we focused on analyzing PIM use and polypharmacy in a population with early cognitive decline, namely, MCI to mild dementia.
In terms of PIMs, the inverse relationship found between PIM use and QOL in this study is in line with that found in other studies involving populations with different ages [20], cognitive statuses [20, 38], or environments [39]. Due to the design of the present study, it is not possible to determine if PIM use is the cause of the decreased QOL or if worsened QOL triggers PIM use. Previous studies have shown that low QOL is associated with an increased risk of institutionalization and death within 1 year for older people [40], but further research is needed to deepen our understanding of this relationship.
In previous studies, the use of PIMs was associated with a risk of low cognitive performance [19]. This is consistent with the results of the present study, from which the previous research differs in terms of the degree of cognitive impairment and the methods used to assess PIMs. Here, PIM use was associated with a lower score on the verbal fluency (initial letters) test, a neuropsychological assessment used to help detect people that might develop neurodegenerative disorders such as Alzheimer’s disease [41]. In previous studies, many of the authors speculated that some degree of cognitive deficit could be attributed to PIM use due to the ability of some medications to decrease cholinergic function, thereby impairing memory and attention and increasing cognitive problems [42]. Thus, in people with cognitive impairment, the adverse effects can exceed the therapeutic benefit of some medications, because cognitive performance worsens. For this reason, limiting the prescription of PIMs becomes a necessary countermeasure.
Furthermore, among the PIMs, the class of most frequently used medicines was antithrombotic drugs, followed by hypnotics (barbiturates, benzodiazepines, non-benzodiazepine receptor agonists). Previous studies found that the use of benzodiazepines can be associated with a decrease of QOL [43] and that long-term benzodiazepine users were more cognitively impaired than non-users [44].
The existence of a relationship between polypharmacy and frailty status was proven by many studies [17, 45]. The underlying explanation of this finding is somewhat complex. Previous reports describe this correlation as bidirectional. Frailty is linked to many chronic diseases [46], which can in turn increase drug usage. At the same time, the number of medications is linked to several elements that can be considered typical of frailty, such as impaired balance, poor nutritional status, and functional impairment [47]. Furthermore, the risk of increased frailty status increases together with the increase in the number of medications [48]. The present study shows that this correlation can also be found in a geriatric population with MCI to mild dementia. Thus, measures to reduce the number of medications could effectively prevent frailty in older people [46], and this is even more important due to the frequent association of cognitive impairment with frailty [49].
Also, the present study found an inverse association between the presence of polypharmacy and QOL, measured with a VAS. This association was found with logistic regression, but not with linear regression. This result is in agreement with the results of previous studies [1, 20], which differ from the present study in terms of the participants’ cognitive level. A possible explanation for this association could be that polypharmacy increases the risk of adverse drug reactions, falls, hospitalization, and mortality [10], which are all factors that affect QOL in older people.
Some limitations of this study should be mentioned. First, because this is a cross-sectional study, it is not possible to determine causal relationships between the use of PIMs and polypharmacy and the other variables. Furthermore, the effects of polypharmacy and PIMs on people with a severer dementia have not been evaluated and remain to be investigated. Furthermore, the occurrence of measurement errors during the assessment of cognitive tests cannot be excluded and, due to the small sample size, it was not possible to analyze differences in the results in relation to the extent of cognitive impairment. Finally, due to the sample size and the specific selection of people with MCI to mild dementia, the applicability and generalizability of this study is limited.
Notwithstanding, because this study examines a group of people with an early cognitive decline that ranges from MCI to mild dementia, it allows to focus on patients on which an early intervention is still possible, before their cognitive performance declines further and progresses to more severe dementia and an overall more compromised general health condition. About the other merits of this study, the instrument used to assess QOL was the EQ-5D scale, which is widely used in international studies with similar purposes and is applicable to a wide range of health conditions. Above all, the EQ-5D is suitable to evaluate QOL in a population with cognitive impairment [50].
Conclusion
In a geriatric population with cognitive performance ranging from MCI to mild dementia, the use of PIMs was associated with a lower QOL score and a lower score in the verbal fluency (initial letters) test than no use of PIMs. The presence of polypharmacy was correlated with a worse frailty status and lower QOL if compared to the absence of polypharmacy. The number of medicines, instead, was correlated with a worse frailty status only.
As for the clinical implication of this study, the results suggest that the use of PIMs and polypharmacy assume a prominent role in the life of a geriatric population with MCI and mild dementia, since they have been found to be associated with important factors like QOL, cognition and frailty. This is even more important if we consider that, at present, guidelines and goals have been defined to treat isolated diseases, but there is no well-defined procedure to follow when several diseases coexist together, which is often the case of old people. Strategies to guide or reduce the use of PIMs and polypharmacy should be implemented in order to diminish their impact on the life of geriatric patients.
