Polypharmacy and Potentially Inappropriate Medication in People with Dementia: A Nationwide Study

Abstract

Background:

Polypharmacy (use of ≥5 different medications) and potentially inappropriate medication (PIM) are well-known risk factors for numerous negative health outcomes. However, the use of polypharmacy and PIM in people with dementia is not well-described.

Objective:

To examine the prevalence of polypharmacy and PIM in older people with and without dementia in a nationwide population.

Methods:

Cross-sectional study of the Danish population aged ≥65 in 2014 (n = 1,032,120) based on register data, including information on diagnoses and dispensed prescriptions. Polypharmacy and PIM use among people with (n = 35,476) and without dementia (n = 994,231) were compared, stratified by living situation and adjusted for age, sex, and comorbidity. The red-yellow-green list from the Danish Institute for Rational Pharmacotherapy and the German PRISCUS list were used to define PIM.

Results:

People with dementia were more frequently exposed to polypharmacy (dementia: 62.6% versus no-dementia: 35.1%, p < 0.001) and likewise PIM (red-yellow-green: 45.0% versus 29.7%, p < 0.001; PRISCUS: 24.4% versus 13.2%, p < 0.001). After adjustments for age, sex, and comorbidity, the likelihood of polypharmacy and PIM was higher for community-dwelling people with dementia than without dementia (odds ratio (OR); [95% confidence interval (CI)] polypharmacy: 1.50 [1.45–1.55]; red-yellow-green: 1.27 [1.23–1.31]; PRISCUS: 1.25 [1.20–1.30]). In contrast, dementia slightly decreased the odds of polypharmacy and PIM in nursing home residents.

Conclusion:

Use of polypharmacy and PIM were widespread in the older population and more so in people with dementia. This could have negative implications for patient-safety and demonstrates the need for interventions to improve drug therapy in people with dementia.

Keywords

Dementia inappropriate prescribing pharmacoepidemiology polypharmacy potentially inappropriate medication list

INTRODUCTION

Polypharmacy—although sometimes necessary—has been linked to an increased risk of a number of negative health outcomes in older people; including adverse effects, drug-drug interactions, hospitalizations, frailty, and even mortality [1 –3]. Consequently, prescribing for older people is a complex process where benefits of treatment have to be weighed against the risks. In people with dementia, prescribing is further complicated by difficulties with communication, changing goals of care, and a high prevalence of multimorbidity [4]. Older people and the multimorbid are seldom included in drug trials; therefore, our knowledge on drug effects in older people is limited [5].

Due to alterations in pharmacokinetics and pharmacodynamics occurring with age, some drugs are associated with higher risks of adverse outcomes and drug-drug interactions in older people [5]. Several explicit quality indicators have been developed based on expert consensus and literature review to screen for such ‘potentially inappropriate medication’to avoid in many or most older individuals, most notably the American Beers list and European lists such as the STOPP-START criteria and the PRISCUS list [6 –8]. Different national indicators have been developed: The Danish Institute for Rational Pharmacotherapy, an institute under the Danish Health and Medicines Agency, has developed the red-yellow-green list geared toward use in Denmark [9].

Previous studies have shown a high prevalence of polypharmacy and potentially inappropriate medication among people with dementia [4 , 11]. However, only few studies have compared the prevalence of polypharmacy and potentially inappropriate medication in older people with and without dementia, and these have been limited in size or focused only on people with dementia either in a specific setting or at a specific disease stage, thereby limiting generalizability. The following hypothesis was examined in both older community-dwelling people and nursing home residents: people with dementia were more often exposed to polypharmacy and potentially inappropriate medication than people without dementia. Recent research has demonstrated widespread use of psychotropic medication in people with dementia but, to our knowledge, psychotropic medication has not been studied as a contributing factor to the high prevalence of polypharmacy in people with dementia. Therefore, we also investigated whether a potential disparity between the prevalence of polypharmacy in people with and without dementia was due to the use of psychotropic medication.

MATERIALS AND METHODS

The Danish registers

In Denmark, a personal civil registration number is assigned to all permanent residents making it possible to retrieve and link data from nationwide registers on an individual level. This study was based on linked data from the Danish Civil Registration System, the Danish National Patient Register, the Danish Psychiatric Central Research Register, and the Danish National Prescription Registry [12 –14]. The Danish National Patient Register contains data on all hospital contacts, admissions, discharges, and invasive procedures since 1977 and additionally hospital-based outpatient clinic contacts and emergency department contacts since 1995. Diagnosis codes are registered by use of International Classification of Diseases (ICD) codes as classified by WHO [15]. The Danish Psychiatric Central Research Register contains data on all admissions to psychiatric hospitals since 1969 and all outpatient contacts since 1995, including data on discharge diagnoses [13]. The Danish National Prescription Registry holds information regarding dispensed prescription medication since 1995 registered according to the Anatomical Therapeutic Chemical (ATC) classification system, including information on date of dispensing, strength, and package size. Demographic data were provided by Statistics Denmark.

Study population

The study population, identified through the Danish Civil Registration System, included all Danish residents aged 65 or above and alive on January 1, 2014 (baseline). People with dementia were identified as individuals registered with a dementia diagnosis in the Danish National Patient Registry or the Danish Psychiatric Central Research Register or as individuals who had filled at least one prescription for anti-dementia medication (N06D) before baseline (specific diagnosis codes and drug names available in Supplementary Table 1). Individuals registered with a dementia diagnosis or having filled their first prescription for anti-dementia medication before the age of 60 were excluded due to the questionable validity of register-based dementia diagnoses in younger age groups [16]. The Danish registers offer a unique opportunity to examine the drug use of community-dwelling people and nursing home residents alike. Thus, the study population was further stratified according to living situation.

Polypharmacy

Polypharmacy was defined as filling prescriptions for ≥5 different drugs on the substance level (ATC level 5) and excessive polypharmacy as filling prescriptions for ≥10 different drugs within a pre-specified window of three months [2]. The number of different drugs used within three months has previously been validated as an estimate of the number of drugs used concurrently [17]. Polypharmacy was examined in the first quarter of 2014. Potential irregularities due to the Danish reimbursement system were examined by comparing results from the first and second quarter, which showed that the difference was negligible. Polypharmacy was examined both with and without anti-dementia medication (N06D). To enable comparisons between people with and without dementia, polypharmacy not counting anti-dementia medication is presented in this article unless otherwise specified. In a sensitivity analysis, psychotropic medication (antipsychotics (N05A), hypnotics (N05C), anxiolytics (N05B), and antidepressants (N06A)) was additionally excluded to investigate the widespread use of psychotropic medication in people with dementia as a possible explanation for a potential discrepancy between the prevalence of polypharmacy in people with and without dementia.

Potentially inappropriate medication

Two lists of explicit quality indicators were used separately to define potentially inappropriate medication: 1) the Danish red-yellow-green list and 2) the German PRISCUS list [8, 9]. The original 2011-version of the red-yellow-green list was used as this was the version available in 2014. The red-yellow-green list is divided into three categories: the red category lists drugs that should be avoided in older people except for special circumstances; the yellow category lists drugs where the correct indication should be reevaluated; the green category lists drugs where little evidence exists for long-term effects or adverse effects. Only the red category, which lists 28 drugs and drug classes to be avoided in all older people except for special circumstances, was used to define potentially inappropriate medication in this study. One drug, acetylsalicylic acid, was excluded as it was only defined as inappropriate above a certain dose (detailed information available in Supplementary Table 2).

The PRISCUS list, published in Germany in 2010, contains recommendations on 82 drugs and drug classes to avoid in older people. Of these, 36 drugs/drug classes were excluded as they were not available in Denmark. Additionally, eight of the remaining drugs were excluded as recommendations were dose-dependent (haloperidol, olanzapine, oxazepam, lorazepam, lormetazepam, zopiclone, zolpidem), or dependent on indication (terazosin). There is an overlap between the two lists: Of the remaining 38 drugs/drug classes on the PRISCUS list, 13 are also included in the red category of the red-yellow-green list. The PRISCUS list is not entirely compatible with the Danish setting but was used to facilitate international comparisons.

Previous studies have shown an association between the number of drugs used and the use of potentially inappropriate medication [18, 19]. Therefore, adjustments for the number of different drugs were made in a sensitivity analysis.

Comorbidity

Comorbidity was assessed by calculating the Charlson Comorbidity Index score, which comprises 19 chronic somatic diseases [20]. The conditions used to calculate the score were defined by registered discharge hospital diagnoses codes prior to baseline.

Statistics

Descriptive statistics were used to examine baseline characteristics of the study population. Continuous variables were summarized by medians and 25–75 percentiles as data were not normally distributed. Nominal variables were expressed by proportions with 95% confidence intervals. Pearson’s chi-squared test and the non-parametric Mann-Whitney U-test were used to compare characteristics of the population and prevalence of polypharmacy and potentially inappropriate medication. Logistic regression analysis was performed to evaluate the effects of covariates independently (crude analysis) and in a multivariate analysis (adjusted). Adjustments were made stepwise: female sex and higher age were included in model 1; and sex, age, and the Charlson Comorbidity Index score in model 2. For potentially inappropriate medication, the number of different drugs (excluding anti-dementia medication (N06D)) was added in a third model as a sensitivity analysis. Odds ratios were accompanied by 95% confidence intervals. A p-value of <0.05 was considered statistically significant. The data analysis was performed using SAS statistical software, version 9.4 (SAS Institute Inc., Cary, NC, USA).

RESULTS

Study population

On January 1_, 2014, 1,032,120 residents in Denmark were aged 65 or above. Of these, 37,889 had dementia, but 2,413 were excluded because they had been diagnosed with dementia or filled their first prescription for anti-dementia medication before the age of 60. Of those with dementia, 12.9% were identified by the use of anti-dementia drugs rather than a registered diagnosis. People identified with dementia based on drug use were more likely than those identified by a registered diagnosis to be male, younger, and to have lower comorbidity scores. Characteristics of the population are presented in Table 1. People with dementia were more likely to be female irrespective of living situation. Community-dwelling people with dementia were older and had higher comorbidity scores than community-dwelling people without dementia, whereas the opposite was the case for nursing home residents.

Table 1

Characteristics of the study population stratified by living situation and dementia diagnosis

	Community-dwelling		Nursing home residents		Total
Characteristic	Dementia (n = 16,228)	No-dementia (n = 963,599)	Dementia (n = 19,248)	No-dementia (n = 30,632)	Dementia (n = 35,476)	No-dementia (n = 994,231)
Sex (Female)	9,252 (57.0)	519,005 (53.9)^*	13,438 (69.8)	20,430 (66.7)^*	22,690 (64.0)	539,435 (54.3)^*
Age	80.9 (75.3–86.0)	72.3 (68.4–78.3)^*	85.1 (79.7–89.5)	86.0 (79.3–91.2)^*	83.2 (77.5–88.2)	72.5 (68.5–78.8)^*
Married^†	7,737 (47.7)	563,024 (58.8)^*	4,502 (23.4)	5,376 (17.6)^*	12,239 (34.5)	568,400 (57.5)^*
Time since dementia diagnosis	2.3 (0.9–4.6)	NA	3.4 (1.7–5.8)	NA	3.0 (1.3–5.3)	NA
Charlson Comorbidity Index score (excluding dementia)
0	5,577 (34.4)	499,370 (51.8)	6,393 (33.2)	7,508 (24.5)	11,970 (33.7)	506,878 (51.0)
1	3,652 (22.5)	172,620 (17.9)	4,684 (24.3)	7,067 (23.1)	8,336 (23.5)	179,687 (18.1)
2	2,861 (17.6)	147,106 (15.3)	3,507 (18.2)	5,704 (18.6)	6,368 (18.0)	152,810 (15.4)
≥3	4,138 (25.5)	144,503 (15.0)^*	4,664 (24.2)	10,353 (33.8)^*	8,802 (24.8)	154,856 (15.6)^*
Cardiovascular disease	6,546 (40.3)	213,320 (22.1)^*	8,172 (42.5)	15,921 (52.0)^*	14,718 (41.5)	229,241 (23.1)^*
Pulmonary disease	2,098 (12.9)	91,062 (9.5)^*	2,359 (12.3)	4,955 (16.2)^*	4,457 (12.6)	96,017 (9.7)^*
Diabetes	2,068 (12.7)	77,712 (8.1)^*	2,351 (12.2)	4,583 (15.0)^*	4,419 (12.5)	82,295 (8.3)^*
Any tumor	3,325 (20.5)	169,924 (17.6)^*	3,749 (19.5)	7,045 (23.0)^*	7,074 (19.9)	176,969 (17.8)^*

NA, not applicable. Numbers are given as n (%) and median (25–75% interquartile range) as appropriate. ^*p < 0.001. ^†Information missing on civil status for n = 5,520.

People with versus without dementia

Polypharmacy

Figure 1 shows the prevalence of polypharmacy, excessive polypharmacy, and potentially inappropriate medication, stratified by dementia status and living situation. Comparing people with and without dementia, polypharmacy was more frequent in people with dementia (dementia: 62.6% [95% confidence interval: 62.1–63.1] versus no-dementia: 35.1% [35.0–35.2], p < 0.001) which was also the case for excessive polypharmacy (17.3% [17.0–17.7] versus 7.2% [7.1–7.2], p < 0.001).

Fig.1

Polypharmacy (use of ≥ 5 different drugs), excessive polypharmacy (use of ≥ 10 different drugs), and potentially inappropriate medication in people with and without dementia stratified by living situation. Potentially inappropriate medication as defined by the red category of the red-yellow-green list or the PRISCUS list. Error bars represent 95% confidence intervals.

Potentially inappropriate medication

A larger proportion of people with dementia than people without dementia had filled at least one prescription for potentially inappropriate medication defined as the red category of the red-yellow-green list or the PRISCUS list in 2014 (red-yellow-green: 45.0% [44.4–45.5] versus 29.7% [29.7–29.8], p < 0.001; PRISCUS: 24.4% [23.9–24.8] versus 13.2% [13.1–13.3], p < 0.001). Table 2 presents the most frequent drugs from the red category of the red-yellow-green list and the PRISCUS list. Despite differences between the lists, there was a considerable overlap of the most common drugs (Table 2). The most frequent potentially inappropriate drugs in people with dementia included NSAIDs (dementia: 11.4% versus no-dementia: 18.3%), the 2nd generation antipsychotic quetiapine (8.1% versus 0.7%), and the antibiotic nitrofurantoin (7.2% versus 2.0%).

Table 2

Prevalence of the most commonly used potentially inappropriate drugs from the red-yellow-green list (red category) and the PRISCUS list in older people with dementia

Active substance	ATC	PRISCUS list	Red-yellow-green list	Dementia (n = 35,476)
NSAIDs	M01A	x^*	x	11.4
Quetiapine	N05AH04		x	8.1
Bisacodyl	A06AB02, A06AG02		x	7.9
Nitrofurantoin	J01XE01	x		7.2
Digoxin	C10AA05	x	x	6.6
Sodium picosulfate	A06AB08		x	6.5
Diazepam	N05BA01	x	x	5.5
Quinine	P01BC01		x	2.9
Olanzapine	N05AH03		x	2.9
Haloperidol	N05AD01		x	2.1

Frequencies presented as percentages. ^*only some NSAIDs are included on the PRISCUS list (details available in Supplementary Table 2).

Community-dwelling people with versus without dementia

Polypharmacy

Among community-dwelling people, polypharmacy was more frequent in people with dementia than without dementia (54.5% [53.7–55.3] versus 33.8% [33.8–33.9], p < 0.001), and excessive polypharmacy was twice as frequent among community-dwelling people with dementia (13.8% [13.3–14.4] versus 6.5% [6.4–6.5], p < 0.001). Table 3 presents the results of the logistic regression analyses investigating the odds of polypharmacy, excessive polypharmacy, and potentially inappropriate medication in people with dementia compared to people without dementia (detailed results available in Supplementary Table 3). After adjustments for age and sex, the odds for polypharmacy were still higher for people with dementia compared to people without dementia (polypharmacy: 1.64 [1.59–1.69]; excessive polypharmacy: 1.68 [1.61–1.76]). With further adjustments for comorbidity, this was still the case. However, the odds ratio decreased (polypharmacy: 1.50 [1.45–1.55]; excessive polypharmacy: 1.51 [1.44–1.58]). In the sensitivity analysis where psychotropic medication was excluded, the crude odds of polypharmacy and excessive polypharmacy were still almost twice as high for community-dwelling people with dementia compared to community-dwelling people without dementia (polypharmacy: 1.96 [1.90–2.02]; excessive polypharmacy: 1.89 [1.79–2.00]); the difference remained significant after adjustments for age and sex (polypharmacy: 1.37 [1.33–1.41]; excessive polypharmacy: 1.36 [1.29–1.44]) and after further adjustments for comorbidity (polypharmacy: 1.23 [1.19–1.27]; excessive polypharmacy: 1.18 [1.12–1.25]).

Table 3

Polypharmacy, excessive polypharmacy, and potentially inappropriate medication among people with dementia versus people without dementia according to living situation

	Community-dwelling			Nursing home residents
	Crude OR	Model 1	Model 2	Crude OR	Model 1	Model 2
Polypharmacy^*	2.34 [2.27–2.42]	1.64 [1.59–1.69]	1.50 [1.45–1.55]	0.79 [0.76–0.82]	0.78 [0.75–0.81]	0.88 [0.84–0.92]
Excessive polypharmacy^†	2.33 [2.22–2.43]	1.68 [1.61–1.76]	1.51 [1.44–1.58]	0.62 [0.59–0.64]	0.61 [0.58–0.63]	0.68 [0.65–0.71]
Red-yellow-green^‡	1.51 [1.46–1.56]	1.32 [1.28–1.36]	1.27 [1.23–1.31]	0.87 [0.84–0.90]	0.86 [0.83–0.89]	0.88 [0.85–0.91]
PRISCUS	1.77 [1.70–1.84]	1.32 [1.27–1.38]	1.25 [1.20–1.30]	0.76 [0.73–0.79]	0.75 [0.72–0.78]	0.78 [0.75–0.81]

OR, odds ratio. NOTE: This table shows the results of the logistic regression analysis comparing people with dementia to people without dementia according to living situation. Results are given as odds ratios (OR) accompanied by 95% confidence intervals. Model 1 includes age and sex. Model 2 includes age, sex, and comorbidity. ^*Polypharmacy: the use of ≥5 different drugs. ^†Excessive polypharmacy: the use of ≥10 different drugs. ^‡Red-yellow-green: at least one drug from the red category of the red-yellow-green list.

Potentially inappropriate medication

Among community-dwelling people, potentially inappropriate medication was more frequent in those with dementia (red-yellow-green: 38.1% [37.4–38.9] versus 29.0% [28.9–29.1], p < 0.001; PRISCUS: 20.4% [19.8–21.0] versus 12.5% [12.5–12.6], p < 0.001). Following adjustments for age and sex in the logistic regression (Table 3) the odds of having filled at least one prescription for potentially inappropriate medication were still higher for community-dwelling people with dementia compared to community-dwelling people without dementia (red-yellow-green: 1.32 [1.28–1.36]; PRISCUS: 1.33 [1.28–1.38]). Further adjustments for comorbidity decreased the odds ratios slightly to 1.27 [1.23–1.31] and 1.25 [1.20–1.30] when defined by the red category of the red-yellow-green list and the PRISCUS list, respectively.

In a sensitivity analysis, after additional adjustments for the number of different drugs, the odds ratios decreased. However, the odds of potentially inappropriate medication were still higher for community-dwelling people with dementia than without dementia (red-yellow-green: 1.27 [1.23–1.31]; PRISCUS: 1.16 [1.11–1.21]).

Overlap

Figure 2 presents the overlap between polypharmacy and the use of potentially inappropriate medication in people with and without dementia stratified by living-situation: Among community-dwelling people with dementia, 12.9% [12.4–13.4] were exposed to polypharmacy and potentially inappropriate medication as defined by the red-yellow-green list and the PRISCUS list. Among community-dwelling people without dementia, 6.6% [6.5–6.6] were exposed to polypharmacy and potentially inappropriate medication as defined by both lists. Coincident polypharmacy and use of potentially inappropriate medication as defined by the red-yellow-green list was found among 27.5% [26.8–28.1] of people with dementia versus 15.5% [15.4–15.6] of people without dementia.

Fig.2

Frequency of overlap between polypharmacy (≥5 different drugs) and potentially inappropriate medication defined as drugs from the red category of the red-yellow-green list and the PRISCUS list. Error bars indicate 95% confidence intervals.

Nursing home residents with versus without dementia

Polypharmacy

Figure 3 presents the prevalence of polypharmacy in people with and without dementia stratified by living situation and age group. Polypharmacy and excessive polypharmacy were less frequent among nursing home residents with dementia compared to nursing home residents without dementia (dementia: 69.5% [68.8–70.1] versus no-dementia: 74.3% [73.8–74.8], p < 0.001) and likewise this was the case for excessive polypharmacy (20.3% [19.8–20.9] versus 29.2% [28.7–29.8], p < 0.001). It should also be noted that in a sensitivity analysis where anti-dementia medication was not excluded, no significant difference was observed between the prevalence of polypharmacy in nursing home residents with and without dementia. The logistic regression analysis with adjustments for age and sex showed that the odds of polypharmacy and especially excessive polypharmacy were lower for nursing home residents with dementia than nursing home residents without (polypharmacy: 0.78 [0.75–0.81]; excessive polypharmacy: 0.61 [0.58–0.63]). Further adjustments for comorbidity decreased the relative difference (polypharmacy: 0.88 [0.84–0.92]; excessive polypharmacy: 0.68 [0.65–0.71]).

Fig.3

Percentage filling prescriptions for ≥ 5 different drugs within the first quarter of 2014 stratified by dementia, age group, and living situation. Error bars indicate 95% confidence intervals.

Potentially inappropriate medication

Among nursing home residents, potentially inappropriate medication was less frequent in nursing home residents with dementia compared to nursing home residents without dementia (red-yellow-green: 50.7% [50.0–51.4] versus 54.2% [53.6–54.7], p < 0.001; PRISCUS: 27.7% [27.1–28.3] versus 33.7% [33.1–34.2], p < 0.001). In the logistic regression with adjustments for age and sex, potentially inappropriate medication was less likely in nursing home residents with dementia than in nursing home residents without dementia (red-yellow-green: 0.86 [0.83–0.89]; PRISCUS: 0.75 [0.72.0.79]). Additional adjustments for comorbidity score had minimal effects (red-yellow-green: 0.88 [0.85–0.91]; PRISCUS: 0.78 [0.75–0.81]).

Overlap

Among nursing home residents, coincident exposure to polypharmacy and potentially inappropriate medication was widespread: In nursing home residents 24.3% [23.8–24.8] of those without dementia and 18.6% [18.0–19.1] of those with dementia were exposed to polypharmacy as well as potentially inappropriate medication as defined by the red-yellow-green list and the PRISCUS list (Fig. 2).

DISCUSSION

To our knowledge, this is the first nationwide study to examine polypharmacy and potentially inappropriate medication among people with dementia. Both were widespread in the older population, but significantly more so in people with dementia where almost two-thirds were exposed to polypharmacy and half to potentially inappropriate medication as defined by the Danish red-yellow-green list. Interestingly, the excess polypharmacy among people with dementia could not be explained by the use of psychotropic medication alone. Additionally, while polypharmacy and potentially inappropriate medication were, overall, more widespread in people with dementia, the difference was modified by living situation.

In our study, polypharmacy was more frequent among community-dwelling people with dementia than in community-dwelling people without dementia (54.5% versus 33.8%). Other studies using the same definition of polypharmacy have found a comparable, albeit slightly lower, prevalence of 33.5–52.0% in community-dwelling people with dementia [10 , 22]. However, these studies focused on specific dementia types and/or disease stages and therefore, unlike our study, might not represent the entire population of people with dementia. A study by Clague et al. found, in line with our results, that community-dwelling people with dementia were more likely to be exposed to polypharmacy even after adjusting for differences in age, sex, and comorbidity [4], which is comparable to our findings. However, they did not investigate whether the difference was caused by anti-dementia medication or psychotropic medication.

One of our major findings was that use of potentially inappropriate medication was more frequent in community-dwelling people with dementia than community-dwelling people without dementia. Several studies have reported a high prevalence of potentially inappropriate medication among community-dwelling people with dementia [11 , 23]. However, only few studies have compared the prevalence among community-dwelling people with and without dementia and these have had varying results [24 –26]; two studies found dementia to increase the likelihood of potentially inappropriate medication as defined by the 2012 Beers list [25, 26]. Meanwhile, Lau et al. found potentially inappropriate medication, as defined by the Beers 2003 list, to be less frequent among community-dwelling people with dementia than community-dwelling people without dementia [24]. However, when excluding oral estrogens (which are not a part of the red category of the red-yellow-green list or the PRISCUS list) they found no significant difference between the two groups. Therefore, the discrepancy is likely due to the use of different lists and different settings.

Earlier studies have found a lower frequency of polypharmacy in nursing home residents with dementia compared to nursing home residents without dementia; however, in our study, this was only the case after anti-dementia medication had been excluded [27]. A possible explanation could be that dementia is underdiagnosed and likely more so among nursing home residents which could lessen the difference between the prevalence of polypharmacy in nursing residents with and without dementia.

A systematic review looking at the use of potentially inappropriate medication as defined by various explicit quality indicators found that approximately half of nursing home residents were exposed to at least one potentially inappropriate drug which is comparable to our results. Furthermore, they also concluded that dementia was associated with a decrease in the likelihood of potentially inappropriate medication, which is in line with our findings [28]. Comparisons, however, are restricted by the heterogeneity of the various lists and changing treatment guidelines as well as international differences in available medication and prescribing practice. In our study, the prevalence of potentially inappropriate medication was higher as measured using the red category of the Danish red-yellow-green list than the German PRISCUS list. A plausible explanation is that the PRISCUS list is not entirely relevant in a Danish context as demonstrated by the fact that almost half of the drugs on the list are not available in Denmark. Another possible explanation is our exclusion of dose-dependent recommendations on the PRISCUS list.

In our study, neither differences in age, sex, or comorbidity score could explain the difference between the prevalence of polypharmacy and potentially inappropriate medication in people with dementia compared to people without dementia. Furthermore, the excess polypharmacy among people with dementia could not be explained by the use of anti-dementia medication or psychotropic medication prescribed for symptoms associated with dementia such as depression, anxiety, behavioral symptoms, and sleep disturbances. To our knowledge, no other studies have investigated this aspect. Previous research has reported a strong association between polypharmacy and potentially inappropriate medication [18, 19]. However, we found that the disparity in the prevalence of potentially inappropriate medication in people with and without dementia could not be explained solely by the higher drug use among people with dementia.

In our study, we found that while community-dwelling people with dementia were more likely than community-dwelling people without dementia to be exposed to polypharmacy and potentially inappropriate medication, the opposite was found among nursing home residents. A possible explanation could, as mentioned earlier, be undiagnosed and/or unregistered dementia cases in nursing home residents. Furthermore, people placed in nursing homes for reasons other than problems associated with dementia are often multimorbid and thus more likely to be taking multiple drugs, whereas this may not, to the same extent, be the case for patients with dementia. Cautious prescribing in the later stages of dementia has also been proposed as a likely explanation [27, 29]. However, the prevalence of polypharmacy and potentially inappropriate medication is notably higher among nursing home residents compared to community-dwelling older people, indicating that cautious prescribing is not the case.

We have previously found that the use of psychotropic medication and opioids is common in nursing home residents and people with dementia [30 –33]. Additionally, a recent study has reported increasing polypharmacy in the last year of life partly due to the use of medication of questionable benefit [34]. Our findings further add to the concern about prescribing practices in these vulnerable groups. Furthermore, we found that a significant portion of people with dementia and nursing home residents were exposed to both polypharmacy and potentially inappropriate medication coincidently. Current knowledge on drug effects in people with dementia is limited as this group is seldom included in clinical drug trials, but studies have shown a high rate of adverse drug reactions and hospitalizations due to drug-related problems [35 –37]. Consequently, the prevalence of polypharmacy and potentially inappropriate medication, both risk factors for drug-related problems, is alarming and accentuates the need for research on drug effects in people with dementia to support prescribing.

It is important to acknowledge that medication on the various lists of potentially inappropriate medication is sometimes the appropriate choice for the individual. Additionally, polypharmacy can be necessary in the multimorbid. However, due to the risks associated with polypharmacy and potentially inappropriate medication, our study raises the question of whether the prevalence described in this study is appropriate. Furthermore, this study demonstrates the need for thorough monitoring of people with dementia, especially considering that the ability to self-report can be affected in this patient group. An analysis from the Danish National Board of Health showed that people with dementia 75 years and older had fewer visits to GP and ambulatory hospital visits than people without dementia [38]. After the age of 85, the same was true for somatic hospitalizations. Meanwhile, the number of consultations per phone or e-mail, most likely through a care-giver, increased with age for people with dementia [38]. Therefore, the level of monitoring does not seem to follow the widespread use of polypharmacy and potentially inappropriate medication which potentially challenges patient-safety.

Our study has several limitations: First, dementia is underdiagnosed in Denmark as well as in other countries. According to previous research, only 50% to 70% of dementia cases are recorded in the Danish registers [39]. As one of the most frequent reasons for nursing home placement is severe dementia, underdiagnosing could have influenced our results in this setting. Furthermore, neither the Danish National Patient Register nor the Danish Psychiatric Central Research Register covers the primary sector [12, 13]: Consequently, comorbidity assessment and identification of people with dementia were limited to information available from the secondary sector. Not all comorbidity can be expected to be registered in the hospital setting. Therefore, we cannot eliminate residual comorbidity as a possible explanation for the association found between dementia and polypharmacy as well as potentially inappropriate medication. Additionally, people diagnosed with dementia before the age of 60 were excluded due to the possibility of misdiagnosis and thus our study does not cover cases of early-onset dementia. Second, the Danish National Prescription Registry contains information on all filled prescriptions, but it is unknown whether the medication was consumed. Additionally, we do not have information on over-the-counter sales; therefore, our estimate of polypharmacy might be on the conservative side. Medication on the PRISCUS list only deemed inappropriate above a certain dosage was excluded as we could not estimate the daily dosage reliably. Consequently, the use of potentially inappropriate medication is likely more widespread than reported. Furthermore, neither the red-yellow-green list nor the PRISCUS list is targeted at this specific population and additional drugs not included on these lists are likely inappropriate in the context of dementia.

A major strength of our study is that it is based on reliable data from multiple registers linked by a unique identification number, allowing us to examine real-life prescription habits in the entire older population of Denmark. Furthermore, linking data from nationwide registers permitted adjustments for age, sex, and comorbidity and allowed us to stratify according to living situation which, as shown by the difference found between these groups, has proven to be a great strength of our study. Additionally, the validity of dementia diagnoses in the Danish nationwide hospital registers has previously been investigated and was accurate in 85.8% of cases [40].

In this nationwide register-based study, polypharmacy and potentially inappropriate medication were considerably more frequent among older people with dementia than older people without dementia. The disparity could neither be explained by confounding from age and sex nor by the larger comorbidity burden observed among people with dementia. Future research should explore factors that could contribute to the discrepancy found in this study, including caregiver burden and frequency and quality of contact with the healthcare system. Furthermore, the excess polypharmacy among people with dementia could not be explained solely by the widespread use of psychotropic medication and the disparity in the use of potentially inappropriate medication was not only due to the higher number of different drugs used among people with dementia. Due to the size of our population and the avoidance of selection bias, we believe our results are of interest internationally. Further research is needed on the impact of polypharmacy and potentially inappropriate medication on the general health of people with dementia and whether this differs from the effect on older people without dementia to guide clinicians in reviewing medication.

Footnotes

ACKNOWLEDGMENTS

The Danish Dementia Research Centre is supported by grants from the Danish Ministry of Health (file no. 2007-12143-112/59506, file no. 1604063, and file no. 0901110/34501). All researchers were independent of the funding source. All authors declare no support from any organization for the submitted work.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Maher

, Hanlon

, Hajjar

(2014) Clinical consequences of polypharmacy in elderly. Expert Opin Drug Saf 13, 57–65.

Gnjidic

, Hilmer

, Blyth

, Naganathan

, Waite

, Seibel

, McLachlan

, Cumming

, Handelsman

, Le Couteur

(2012) Polypharmacy cutoff and outcomes: Five or more medicines were used to identify community-dwelling older men at risk of different adverse outcomes. J Clin Epidemiol 65, 989–995.

Veronese

, Stubbs

, Noale

, Solmi

, Pilotto

, Vaona

, Demurtas

, Mueller

, Huntley

, Crepaldi

, Maggi

(2017) Polypharmacy is associated with higher frailty risk in older people: An 8-year longitudinal cohort study. J Am Med Dir Assoc 18, 624–628.

Clague

, Mercer

, McLean

, Reynish

, Guthrie

(2017) Comorbidity and polypharmacy in people with dementia: Insights from a large, population-based cross-sectional analysis of primary care data. Age Ageing 46, 33–39.

Shi

, Morike

, Klotz

(2008) The clinical implications of ageing for rational drug therapy. Eur J Clin Pharmacol 64, 183–199.

(2015) American Geriatrics Society 2015 Updated Beers Criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc 63, 2227–2246.

O’Mahony

, O’Sullivan

, Byrne

, O’Connor

, Ryan

, Gallagher

(2015) STOPP/START criteria for potentially inappropriate prescribing in older people: Version 2. Age Ageing 44, 213–218.

Holt

, Schmiedl

, Thurmann

(2010) Potentially inappropriate medications in the elderly: The PRISCUS list. Dtsch Arztebl Int 107, 543–551.

Rød-gul-grøn listen (the Red-yellow-green list), Last updated 2011, Accessed on 2011.

10.

Andersen

, Viitanen

, Halvorsen

, Straume

, Engstad

(2011) Co-morbidity and drug treatment in Alzheimer’s disease. A cross sectional study of participants in the dementia study in northern Norway. BMC Geriatr 11, 58.

11.

Barry

, Cooper

, Ryan

, Passmore

, Robinson

, Molloy

, Darcy

, Buchanan

, Hughes

(2016) Potentially inappropriate prescribing among people with dementia in primary care: A retrospective cross-sectional study using the enhanced prescribing database. J Alzheimers Dis 52, 1503–1513.

12.

Lynge

, Sandegaard

, Rebolj

(2011) The Danish National Patient Register. Scand J Public Health 39, 30–33.

13.

Mors

, Perto

, Mortensen

(2011) The Danish Psychiatric Central Research Register. Scand J Public Health 39, 54–57.

14.

Kildemoes

, Sorensen

, Hallas

(2011) The Danish National Prescription Registry. Scand J Public Heal. 39, 38–41.

15.

WHO (2016) ICD-10 Version:2016.

16.

Salem

, Andersen

, Nielsen

, Stokholm

, Jorgensen

, Rasmussen

, Waldemar

(2012) Overdiagnosis of dementia in young patients - a nationwide register-based study. Dement Geriatr Cogn Disord 34, 292–299.

17.

Bjerrum

, Rosholm

, Hallas

, Kragstrup

(1997) Methods for estimating the occurrence of polypharmacy by means of a prescription database. Eur J Clin Pharmacol 53, 7–11.

18.

Johnell

, Klarin

(2007) The relationship between number of drugs and potential drug-drug interactions in the elderly: A study of over 600,000 elderly patients from the Swedish Prescribed Drug Register. Drug Saf 30, 911–918.

19.

Montastruc

, Gardette

, Cantet

, Piau

, Lapeyre-Mestre

, Vellas

, Montastruc

, Andrieu

(2013) Potentially inappropriate medication use among patients with Alzheimer disease in the REAL.FR cohort: Be aware of atropinic and benzodiazepine drugs!. Eur J Clin Pharmacol 69, 1589–1597.

20.

Sundararajan

, Henderson

, Perry

, Muggivan

, Quan

, Ghali

(2004) New ICD-10 version of the Charlson comorbidity index predicted in-hospital mortality. J Clin Epidemiol 57, 1288–1294.

21.

Fereshtehnejad

, Johnell

, Eriksdotter

(2014) Anti-dementia drugs and co-medication among patients with Alzheimer’s disease: Investigating real-world drug use in clinical practice using the Swedish Dementia Quality Registry (SveDem). Drugs Aging 31, 215–224.

22.

Oesterhus

, Aarsland

, Soennesyn

, Rongve

, Selbaek

, Kjosavik

(2017) Potentially inappropriate medications and drug-drug interactions in home-dwelling people with mild dementia. Int J Geriatr Psychiatry 32, 183–192.

23.

Wucherer

, Eichler

, Hertel

, Kilimann

, Richter

, Michalowsky

, Thyrian

, Teipel

, Hoffmann

(2017) Potentially inappropriate medication in community-dwelling primary care patients who were screened positive for dementia. J Alzheimers Dis 55, 691–701.

24.

Lau

, Mercaldo

, Harris

, Trittschuh

, Shega

, Weintraub

(2010) Polypharmacy and potentially inappropriate medication use among community-dwelling elders with dementia. Alzheimer Dis Assoc Disord 24, 56–63.

25.

Alhmoud

, Khalifa

, Bahi

(2015) Prevalence and predictors of potentially inappropriate medications among home care elderly patients in Qatar. Int J Clin Pharm 37, 815–821.

26.

Zeenny

, Wakim

, Kuyumjian

(2017) Potentially inappropriate medications use in community-based aged patients: A cross-sectional study using 2012 Beers criteria. Clin Interv Aging 12, 65–73.

27.

Jokanovic

, Tan

, Dooley

, Kirkpatrick

, Bell

(2015) Prevalence and factors associated with polypharmacy in long-term care facilities: A systematic review. J Am Med Dir Assoc 16, 535.e1–12.

28.

Morin

, Laroche

, Texier

, Johnell

(2016) Prevalence of potentially inappropriate medication use in older adults living in nursing homes: A systematic review. J Am Med Dir Assoc 17, 862.e1–9.

29.

Vetrano

, Tosato

, Colloca

, Topinkova

, Fialova

, Gindin

, van der Roest

, Landi

, Liperoti

, Bernabei

, Onder

(2013) Polypharmacy in nursing home residents with severe cognitive impairment: Results from the SHELTER Study. Alzheimers Dement 9, 587–593.

30.

Zakarias

, Jensen-Dahm

, Norgaard

, Stevnsborg

, Gasse

, Andersen

, Soren

, Waldorff

, Moos

, Waldemar

(2016) Geographical variation in antipsychotic drug use in elderly patients with dementia: A nationwide study. J Alzheimers Dis 54, 1183–1192.

31.

Norgaard

, Jensen-Dahm

, Gasse

, Hansen

, Waldemar

(2017) Psychotropic polypharmacy in patients with dementia: Prevalence and predictors. J Alzheimers Dis 56, 707–716.

32.

Norgaard

, Jensen-Dahm

, Gasse

, Hansen

, Waldemar

(2015) Time trends in antipsychotic drug use in patients with dementia: A nationwide study. J Alzheimers Dis 49, 211–220.

33.

Jensen-Dahm

, Gasse

, Astrup

, Mortensen

, Waldemar

(2015) Frequent use of opioids in patients with dementia and nursing home residents: A study of the entire elderly population of Denmark. Alzheimers Dement 11, 691–699.

34.

Morin

, Vetrano

, Rizzuto

, Calderon-Larranaga

, Fastbom

, Johnell

(2017) Choosing wisely? Measuring the burden of medications in older adults near the end of life: Nationwide, Longitudinal Cohort Study. Am J Med 130, 927–936.e9.

35.

Melis

, Marengoni

, Rizzuto

, Teerenstra

, Kivipelto

, Angleman

, Fratiglioni

(2013) The influence of multimorbidity on clinical progression of dementia in a population-based cohort. PLoS One 8, e84014.

36.

Gustafsson

, Sjolander

, Pfister

, Jonsson

, Schneede

, Lovheim

(2016) Drug-related hospital admissions among old people with dementia. Eur J Clin Pharmacol 72, 1143–1153.

37.

Ganjavi

, Herrmann

, Rochon

, Sharma

, Lee

, Cassel

, Freedman

, Black

, Lanctot

(2007) Adverse drug events in cognitively impaired elderly patients. Dement Geriatr Cogn Disord 23, 395–400.

38.

Sundheds- og ældreministeriet. (2016). Statusrapport på demensområdet I Danmark. http://www.sum.dk/Aktuelt/Publikationer/~/media/Filer%20-%20Publikationer_i_pdf/2016/Statusrapport-demens-2016/Statusrapport-paa-demensomraadet-i-dk.ashx

39.

Phung

, Waltoft

, Kessing

, Mortensen

, Waldemar

(2010) Time trend in diagnosing dementia in secondary care. Dement Geriatr Cogn Disord 29, 146–153.

40.

Phung

, Andersen

, Hogh

, Kessing

, Mortensen

, Waldemar

(2007) Validity of dementia diagnoses in the Danish hospital registers. Dement Geriatr Cogn Disord 24, 220–228.