If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Institute of Clinical Medicine, Campus Ahus, University of Oslo, Oslo, NorwayResearch Centre for age related functional decline and diseases, Innlandet Hospital Trust, Ottestad, NorwayHealth Services Research Unit, Akershus University Hospital, Lørenskog, Norway
Faculty of Medicine, University of Oslo, Oslo, NorwayNorwegian National Advisory Unit on Ageing and Health, Vestfold Hospital Trust, Tønsberg, NorwayFaculty of Medicine, Department of Geriatric Medicine, Oslo University Hospital, Oslo, Norway
Research Centre for age related functional decline and diseases, Innlandet Hospital Trust, Ottestad, NorwayNorwegian National Advisory Unit on Ageing and Health, Vestfold Hospital Trust, Tønsberg, Norway
To explore the course of psychotropic drug (PTD) prescription from admission (BL) to 6-month follow-up (6m) in Norwegian nursing homes (NHs). To examine how clinical variables, such as neuropsychiatric symptoms (NPS), cognition, physical health, and NH characteristics at BL are associated with prescription rates at 6 months.
Design
An observational longitudinal cohort study (data from the Resource Use and Disease Course in Dementia–Nursing Home study) designed to examine the course of dementia, psychiatric and somatic diseases, and drug prescriptions in NH patients during the first 6 months after admission.
Setting and Participants
We included 696 patients at admission to 47 representative Norwegian NHs.
Methods
Demographic and clinical characteristics at BL and 6m are presented. Dementia severity was assessed by the Clinical Dementia Rating scale and the Functional Assessment Staging of Alzheimer's Disease scale. Final diagnosis was made by 2 of the authors (G.S. and S.B.) according to ICD-10 criteria. Prevalence, incidence, and persistence rates of PTD prescriptions for people with dementia are presented. Generalized mixed models were used to identify possible predictors for the course of PTD prescription from BL to 6m.
Results
Prescription rates of antidepressants, antipsychotics, anxiolytics, sedatives, and hypnotics increased in people with dementia from BL (67.5% received at least 1 PTD) to 6m (74.0% received at least 1 PTD). Younger age and higher Neuropsychiatric Inventory–affective subsyndrome score at BL were associated with higher odds of antidepressant prescription, whereas patients with higher comorbidity at BL had lower odds of receiving antidepressants, both at BL and 6m. Higher Neuropsychiatric Inventory-affective subsyndrome scores at BL were associated with higher odds of sedative and hypnotic prescription at both assessment points.
Conclusions and implications
PTD prescription rates increase from BL to 6m. Medication appropriateness should be frequently evaluated after admission to optimize PTD prescriptions.
Antidepressants are still widely used to treat depression in Alzheimer's disease, but few studies are available to make conclusions about their effectiveness.
Psychotropic drug prescription in nursing home patients with dementia: Influence of environmental correlates and staff distress on physicians' prescription behavior.
Association between behavioral and psychological symptoms and psychotropic drug use among old people with cognitive impairment living in geriatric care settings.
Furthermore, differences in the methodological approach and choice of study population make it challenging to compare results from different studies. This longitudinal aspect is particularly important to understand which clinical and environmental factors might be associated with PTD prescription over time, to identify possible risk groups at admission, and to promptly target specific pharmacologic or nonpharmacologic interventions.
The aim of this article was to present the PTD prescription rates in older individuals with dementia at admission to Norwegian NHs and 6 months after admission. We examine whether NPS, cognition, and psychological symptoms at admission were associated with prescription rates at 6 months’ follow-up (6m), and we describe the differences in prescription rates between people with and without dementia.
Methods
Data were collected through the Resource Use and Disease Course in Dementia–Nursing Home (REDIC-NH), a longitudinal cohort study designed to examine the course of dementia and other psychiatric and somatic diseases in NH patients from admission until death.
Resourse Use and Disease Couse in dementia–Nursing Home (REDIC-NH), a longitudinal cohort study; design and patient characteristics at admission to Norwegian nursing homes.
We included 696 patients at admission to 47 representative Norwegian NHs. Inclusion criteria were as follows: (1) 65 years or older; (2) younger than 65 years could participate if established dementia; (3) expected stay at the NH > 4 weeks; and (4) life expectancy >6 weeks.
REDIC-NH Data Collection and Selection
Baseline (BL) data were collected between March 2012 and November 2014, within a month of admission to the NH. Follow-up data were collected every 6 months.
Figure 1 shows the flow chart. We excluded all patients with no medication registered at BL (n = 18) or at 6m (n = 8) because we could not differentiate participants who were not prescribed medication from those with missing data on this variable. Seven patients had no data regarding dementia diagnosis and were excluded. Of the remaining patients, 179 patients dropped out before 6m, and 484 patients were included in the analysis.
Fig. 1Flow chart: Selection of patients for analysis.
Seventy-four percent of data collectors were nurses who completed a 2-day training program. They were supervised by 10 research nurses, who completed a 5-day standardized training program. Information was collected using structured interviews with the patient, their next of kin and other caregivers, clinical examinations, and medical records.
Assessments Included in This Study
Age, gender, marital status, the type of NH unit, the number of patients living in each unit, and the number of health care providers working dayshift on a weekday per unit were collected. Cognitive function, NPS, medication, physical health status and comorbidity, pain, functioning in daily living, and quality of life (QoL) were collected using validated instruments as reported in Table 1.
Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale (MOBID): Development and validation of a nurse-administered pain assessment tool for use in dementia.
Stability of the factor structure of the Neuropsychiatric Inventory in a 31-month follow-up study of a large sample of nursing-home patients with dementia.
Based on all collected information, mild cognitive impairment, dementia, and its etiological subtypes were diagnosed by the authors G.S. and S.B. according to International Classification of Diseases, 10th Revision (ICD-10) criteria.
Resourse Use and Disease Couse in dementia–Nursing Home (REDIC-NH), a longitudinal cohort study; design and patient characteristics at admission to Norwegian nursing homes.
A previous principal component analysis identified the NPI-NH subsyndromes: NPI-NH agitation (agitation/aggression, disinhibition, and irritability), NPI-NH psychosis (delusions and hallucinations), and NPI-NH affective (depression and anxiety).19,30
0–12
Item score was calculated by multiplying severity(score 1–3) by frequency (score 1–4). An NPI-NH item score of 4 and above was considered clinically significant (CS-NPS).
For each patient, we analyzed the exposure to a PTD group, but we did not take into consideration if a patient was prescribed 2 or more PTDs in the same group, except for antipsychotics.
Physical health status and pain
The General Medical Health Rating (GMHR) scale
Excellent, good, fair, poor
Used to assess the general medical health status of each participant.
Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale (MOBID): Development and validation of a nurse-administered pain assessment tool for use in dementia.
∗ A previous principal component analysis identified the NPI-NH subsyndromes: NPI-NH agitation (agitation/aggression, disinhibition, and irritability), NPI-NH psychosis (delusions and hallucinations), and NPI-NH affective (depression and anxiety).
Resourse Use and Disease Couse in dementia–Nursing Home (REDIC-NH), a longitudinal cohort study; design and patient characteristics at admission to Norwegian nursing homes.
Stability of the factor structure of the Neuropsychiatric Inventory in a 31-month follow-up study of a large sample of nursing-home patients with dementia.
† For each patient, we analyzed the exposure to a PTD group, but we did not take into consideration if a patient was prescribed 2 or more PTDs in the same group, except for antipsychotics.
IBM SPSS Statistics version 25 (IBM Corp., Armonk, NY) and SAS Institute Inc. (Cary, NC) SAS version 9.4 statistical software were used for the analyses.
BL characteristics were analyzed in the whole cohort as well as stratified by dementia status. Continuous variables were presented as means and standard deviations (SD), and categorical variables as frequencies and percentages. Differences between dementia groups and between patients excluded and included in further analyses were assessed by a linear mixed model for continuous variables and generalized linear mixed model for categorical variables with random effects for NH units. For each PTD category, prevalence at BL and 6m, and incidence and persistence between BL and 6m were calculated. Prevalence was defined as the proportion of patients prescribed a particular PTD category. Incidence was defined as the proportion of patients prescribed a particular PTD category at 6m relative to the number of patients not prescribed the same PTD category at BL. Persistence was defined as the proportion of patients prescribed a particular PTD category at 6m relative to the number of patients prescribed the same PTD category at BL.
Change in the dichotomous outcome “PTD category use” (yes/no) was assessed with generalized linear models with fixed effects for time dummy (with BL as reference), dementia status at BL (with dementia as reference), and interaction between those 2 variables. The model contained random intercepts for patients nested within NH units. Predefined covariates, age, gender, marital status, physical health, functioning in daily living, cognitive function, depression, NPS (apathy and agitation, psychosis, and affective subsyndrome), Charlson Comorbidity Index, QoL, pain, type of unit, and number of staff members per unit working dayshift were then included in the models as fixed effects together with the interactions between each covariate and dementia status at BL. Akaike's Information Criteria (AIC, smaller values indicate a better model) was applied to reduce the multiple models for excessive interactions and covariates. Results were tabulated only for the interactions and covariates retained in the model.
Most of the covariates had missing values. For Mini-Mental State Examination, Physical Self-Maintenance Scale, QoL, and NPS scores, the missing values were imputed for each item separately by drawing a random number from its empirical distribution. Missing values for Charlson comorbidity index were substituted by zero. Only cases with fewer than 50% missing item values were imputed. The regression models were estimated for the cases with no missing values on covariates (n = 402). Those included in the regression analysis were compared with those not included (n = 82) by appropriate tests.
The results of the generalized linear mixed models were presented as odds ratios (ORs) with the corresponding 95% confidence intervals (CIs). The level of significance was set at 5%.
Ethical and Legal Considerations
The patients' capacity to consent to participation in the study was evaluated by the NH personnel. A written consent for participation was signed. The participants’ next of kin gave consent on behalf of those patients lacking the capacity to consent and for providing information about themselves. The Regional Ethics Committee for Medical Research in South-Eastern Norway approved the study (2011/1738).
Results
Table 2 presents the demographics and clinical data of the patients at admission to the NH. The mean (SD) age of the total sample (N = 671) was 84.4 (7.5) years, 64.4% were women, and 30.7% were married/in a partnership. Patients with dementia (83.9%) were younger (P = .021), scored lower on the Charlson comorbidity index (P = .01), and had less pain (P < .001) compared with patients without dementia, although there was no difference in the general physical health between the 2 groups assessed with the General Medical Health Rating scale. Patients with dementia had more severe NPS (NPI total score; P = .001) and had a higher score in the NPI-agitation subsyndrome (P = .002) and NPI-psychosis subsyndrome (P = .014). The patients were prescribed on average 6.1 drugs a day, but participants with dementia received fewer drugs (mean 5.9) compared with patients without dementia (mean 7.5, P < .001).
Table 2Demographic and Clinical Data of the Patients at Admission to the Nursing Homes (N = 671)
Comparison between patients with dementia and patients with no dementia; linear mixed model is estimated for continuous variables and generalized linear mixed model for categorical variables. The models contain random effect for unit nested within NHs.
NPI-subsyndromes are calculated as the sum of the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
NPI-subsyndromes are calculated as the sum of the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
NPI-subsyndromes are calculated as the sum of the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
N = 661
n = 554
n = 107
Mean (SD)
3.7 (5.8)
3.9 (5.9)
2.7 (4.5)
.052
NPI-Apathy
N = 659
n = 552
n = 107
Mean (SD)
1.3 (2.8)
1.4 (2.8)
1.2 (2.8)
.434
QUALID
N = 667
n = 560
n = 107
Mean (SD)
19.9 (7.2)
20.0 (7.2)
19.4 (7.0)
.325
Total prescribed drugs
Mean (SD)
6.1 (3.1)
5.9 (3.0)
7.5 (3.5)
<.001
MOBID 2
N = 643
n = 538
n = 105
Mean (SD)
2.1 (2.2)
2.0 (2.1)
2.9 (2.4)
<.001
Type of unit
Regular unit
370 (55.1)
292 (51.9)
78 (72.2)
.985
Special care unit
216 (32.2)
207 (36.8)
9 (8.3)
Respite and rehabilitation unit
85 (12.7)
64 (11.4)
21 (19.4)
Number of patients per unit
N = 669
n = 561
n = 108
Mean (SD)
12.0 (6.2)
11.5 (5.9)
14.7 (7.2)
.069
Number of staff members per unit working dayshift
N = 670
n = 562
n = 108
Mean (SD)
3.7 (2.0)
3.6 (1.9)
4.2 (2.2)
.237
CSDD, Cornell scale for depression in dementia; GMHR, General Medical Health Rating Scale; MMSE, Mini-Mental Status Evaluation; MOBID-2 Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale; NPI, The Neuropsychiatric Inventory; PSMS, Physical Self-Maintenance Scale; QUALID, Quality of Life in Late-Stage Dementia.
P values < .05 (statistical significance) are reported in bold.
∗ Comparison between patients with dementia and patients with no dementia; linear mixed model is estimated for continuous variables and generalized linear mixed model for categorical variables. The models contain random effect for unit nested within NHs.
† NPI-subsyndromes are calculated as the sum of the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
Patients excluded or dropped out (n = 187) had poorer general physical health, a higher level of depression, a lower QoL, and a lower level of functioning compared with those who remained in the study (Supplementary Material, Supplementary Table 1).
Table 3 presents the prevalence, incidence and persistence of PTD prescription in patients with dementia. We found an overall increase in the prescription of any PTD (BL: 67.5%, 6m: 74.0%; P = .008). There was a significant increase in the prescription rates of antidepressants (BL: 31.0%, 6m: 40.1%; P < .001), antipsychotics (BL: 13.5%, 6m: 19.0%; P < .001), anxiolytics (BL: 17.1%, 6m: 21.4%; P = .004), and sedatives/hypnotics (BL: 22.6%, 6m: 30.3%; P < .001) between BL and 6m. The persistence of prescription 6 months after NH admission was more than 60%, except for typical antipsychotics, in which the prescription was less persistent compared with atypical antipsychotics (41.7%/72.2%). The incidence of PTD prescription was highest for antidepressants (19.5%) and sedatives/hypnotics (16.8%).
Table 3Prevalence of Psychotropic Drug Prescription in Patients With Dementia at Baseline and 6 Months’ Follow-Up; Persistence and Incidence of Psychotropic Drug Prescription in Patients With Dementia From BL to 6m, N = 416
Persistence is the proportion of patients exposed to a PTD category at 6m relative to the number of patients who were prescribed the same PTD category at BL.
Incidence is the proportion of patients exposed to a PTD category at 6m relative to the number of patients who were not prescribed the same PTD category at BL.
Comparison between the prescription of a psychotropic drug at baseline and 6 months follow-up; linear mixed model is estimated for continuous variable and generalized linear mixed model for categorical variables.
Any type of medication registered with ATC-number.
5.8 (3.0)
6.5 (2.8)
<.001
P values < .05 (statistical significance) are reported in bold.
∗ Comparison between the prescription of a psychotropic drug at baseline and 6 months follow-up; linear mixed model is estimated for continuous variable and generalized linear mixed model for categorical variables.
† Persistence is the proportion of patients exposed to a PTD category at 6m relative to the number of patients who were prescribed the same PTD category at BL.
‡ Incidence is the proportion of patients exposed to a PTD category at 6m relative to the number of patients who were not prescribed the same PTD category at BL.
§ Any type of medication registered with ATC-number.
Table 4 presents the results from the bivariate and multiple generalized linear mixed models assessing changes in prescription between BL and 6m, for all the major categories of PTDs.
Table 4Bivariate and Multiple Models of the Logistic Regression Analysis Regarding Changes in Prescription Between BL and 6m, for all the Major Categories of Psychotropic Drugs
NPI, The Neuropsychiatric Inventory; NPI-subsyndromes are calculated as the sum the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
NPI, The Neuropsychiatric Inventory; NPI-subsyndromes are calculated as the sum the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
NPI, The Neuropsychiatric Inventory; NPI-subsyndromes are calculated as the sum the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
1.4 (1.2–1.7)
<.001
1.1 (1.0–1.3)
.083
1.2 (1.1–1.5)
.013
P values < .05 (statistical significance) are reported in bold.
∗ NPI, The Neuropsychiatric Inventory; NPI-subsyndromes are calculated as the sum the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
In the bivariate model, we found that antidepressants were more likely to be prescribed at 6 months compared with BL in people both with (OR 7.3; 95% CI 2.8–18.9; P < .001) and without dementia (OR 10.5; 95% CI 1.3–86.1; P = .028) with no significant difference between the groups. In the multiple model, we found that younger patients and those with higher NPS-affective subsyndrome score had higher odds of receiving antidepressants (OR 1.4; 95% CI 1.2–1.7; P < .001).
Three interactions were identified in the multiple model. Men with dementia had higher odds of being prescribed antidepressants than men without dementia at both assessment points. There were no significant differences among women. The association between dementia and prescription of antidepressants was significantly different between men and women (P = .003 for interaction). Overall, the association between the prescription of antidepressants and NPI-psychosis was significantly different between patients with and without dementia (P = .048 for interaction), whereas the difference was not significant for Charlson comorbidity index. However, with increasing comorbidity, patients with dementia were prescribed significantly fewer antidepressants than patients without dementia at both BL and 6m (for Charlson comorbidity index ≥2).
In the bivariate model, antipsychotics were more likely to be prescribed at 6m than at BL among patients with dementia (OR 3.4; 95% CI 1.3–8.7; P = .012), with no significant change among those without dementia. This association, however, did not differ significantly between diagnosis groups. The prevalence of antipsychotic prescription among people without dementia was too low to allow for multiple model estimation.
In the bivariate model, we found no difference in anxiolytic prescription rates between BL and 6m in patients with dementia, whereas patients without dementia were prescribed slightly more anxiolytics at 6m compared with BL (OR 14.5; 95% CI 1.1–193.7; P = .043) with no significant differences between diagnosis groups regarding this association. In the multiple model, we found no significant associations.
In the bivariate model, we found that sedatives were more likely to be prescribed at 6m compared with BL in patients with dementia than in patients without dementia (OR 4.6; 95% CI 1.7–12.8; P = .003), although overall, this association did not differ significantly between the diagnosis groups. In the multiple model, more severe affective symptoms were associated with a higher sedative prescription rate (OR 1.2; 95% CI 1.1–1.5; P = .013). Even though not significant, 2 interactions were retained in the model. People with dementia were prescribed more sedatives at 6m than BL (OR 3.9; 95% CI 1.6–9.1; P = .002), with no gender differences. Further, among people with dementia, a higher NPI-agitation score was associated with an increased odds of sedatives prescription at 6m compared with BL, whereas among patients without dementia, there was no difference between the 2 assessment points. We found that the differences between patients with and without dementia regarding the odds of sedative prescription were significant only for NPI-agitation subsyndrome scores below 5 at BL and below 4 at 6m.
Compared with the excluded cases from the regression analysis, the included cases had a significantly higher level of functioning (P = .004); lived more often in regular units (P = .001) and in wards with a higher number of patients per unit (P = .001); and had a significantly lower NPI total score (P = .009), NPI-agitation subsyndrome score (P = .013), and QoL (P = .005) (Supplementary Material, Supplementary Table 2).
Discussion
Prescription rates of antidepressants, antipsychotics, anxiolytics, sedatives, and hypnotics increased in people with dementia during the first 6 months after admission, with a very high persistence of prescription of any PTD category.
Antidepressants
Thirty-one percent of the patients with dementia were prescribed antidepressants at admission, increasing to 40.1% 6 months later. Comparable results can be found in a Norwegian study,
Prevalence of depression among recently admitted long-term care patients in Norwegian nursing homes: Associations with diagnostic workup and use of antidepressants.
Higher NPI-affective subsyndrome scores and younger age at admission were associated with higher odds of antidepressant prescription at both assessment points, comparable to previous international studies,
Psychotropic drug prescription in nursing home patients with dementia: Influence of environmental correlates and staff distress on physicians' prescription behavior.
Association between behavioral and psychological symptoms and psychotropic drug use among old people with cognitive impairment living in geriatric care settings.
although their cross-sectional nature makes it difficult to compare with the results from our study.
Compared with people without dementia, patients with dementia had higher odds of being prescribed antidepressants with increasing NPI-psychosis subsyndrome score. This finding is difficult to explain, considering there is little evidence of effectiveness in the use of antidepressants to treat psychosis in dementia.
On the other hand, antidepressants in this case might be prescribed to control concurrent symptoms to psychosis, such anxiety or agitation.
Higher comorbidity at admission had lower odds of antidepressant prescription at both assessment points. Physicians might avoid prescribing antidepressants due to adverse effects and interactions in patients with several illnesses. When depression is comorbid to other diseases, the physical health has a greater decline.
It is therefore important to identify patients with a higher comorbidity and depression and treat them correctly. Undetected depression in older institutionalized people is still common,
Prevalence of depression among recently admitted long-term care patients in Norwegian nursing homes: Associations with diagnostic workup and use of antidepressants.
and national guidelines suggest the use of antidepressants as first-line treatment only for moderate to severe depressive symptoms in patients with dementia.
It is alarming that there is a high prevalence of antidepressants prescription among older people in NHs, considering their potentially increased risk of adverse effects.
At admission, 13.5% of the patients with dementia received antipsychotics, increasing to 19% 6 months later. Most were atypical antipsychotics. International studies showed higher prevalence rates of antipsychotics use at NH admission: in Canada 27.2%, in Belgium 28.5%, and in Australia 27.1%.
The lower prevalence reported in Norway might be explained by the increasing awareness of the adverse effects antipsychotics may cause in people with dementia, and the national and local campaigns to find alternative nonpharmacological approaches.
Antipsychotics are often used in NHs to treat NPS. National guidelines warn that antipsychotics should be prescribed for short periods of time, and only in severe cases of agitation and aggression.
Unfortunately, antipsychotic use is still persistent over time in institutionalized older patients, although the prevalence results we found are considerably lower than another Norwegian study conducted in NHs
; however, a recent Norwegian study showed a considerable decrease in the prescription of antipsychotic drugs between 2004 and 2011 (more than a 30% reduction in use), with only minor changes for the other PTD.
A Norwegian study showed that the prevalence of NPS is high from admission to a NH over time, and the mean NPI-agitation subsyndrome score tends to increase.
This might explain the increase of antipsychotic prescription during the first 6 months in our study, as physicians tend to treat NPS with antipsychotics. It is still alarming that antipsychotics are used in patients with dementia, considering that antipsychotic review and reduction in antipsychotic use, together with the implementation of nonpharmacological treatments such as social interaction, decreases the risk of mortality.
Impact of antipsychotic review and nonpharmacological intervention on antipsychotic use, neuropsychiatric symptoms, and mortality in people with dementia living in nursing homes: A factorial cluster-randomized controlled trial by the Well-Being and Health for People With Dementia (WHELD) program.
The prevalence of anxiolytic prescription was 17.1% at admission and 21.4% 6 months later, but with no significant differences between the 2 assessment points. In patients with dementia, 22.6% were prescribed sedatives at admission and 30.3% 6 months later. Comparable results for sedatives prescription ranged from 23.8% to 35.2% in Norwegian studies.
The prescription of sedatives and hypnotics increased significantly during the first 6 months after NH admission; however, other studies showed differing results: in Belgium rates decreased from 43.0% to 41.5% 2 years after admission,
These differences should be interpreted cautiously due to the different research populations and study methods.
We found that patients with more affective symptoms at admission were at a higher risk of sedative and hypnotic prescription at both assessment points. This differs from results presented by Zuidema et al.,
Psychotropic drug prescription in nursing home patients with dementia: Influence of environmental correlates and staff distress on physicians' prescription behavior.
Our findings might be explained by the fact that sedatives and hypnotics are usually prescribed to treat depressive symptoms such as sleep disturbances or anxiety at night. Unfortunately, sedatives or hypnotics are still among the most commonly used drugs in people with dementia,
despite the large consensus on the fact that these drugs should be used for as short a time as possible as an adjunct to other nonpharmacological treatments.
The vast and persistent use of hypnotics in NHs and its increase should raise concern considering the amount of adverse effects hypnotics have on older individuals.
This study is one of a few studies reporting data about PTD prescription in patients newly admitted to a NH and following up after admission together with changes in clinical factors. Previous studies reporting PTD use have been either cross-sectional or longitudinal with the registration of PTD use every 12 to 18 months. Assessing patients more frequently than every 12 to 18 months better identifies changes in a patient's medication over time.
This study has some limitations. The data collection was conducted by personnel in all the participating NHs, and that might give variability in the quality of the collected data; however, all data collectors had extensive training and supervision throughout the study period. We were not able to analyze data concerning daily dosage of PTD due to the high imprecision of the reported dosages. Even small changes in the PTD dosage might significantly influence a patient's clinical symptoms. Further, it was not possible to get information about patients who were prescribed no drugs versus patients in whom medication data were missing due to lack of precision in data collection, but not many patients were excluded for this reason. In addition, several reasons might have reduced the representativeness of the patients at admission to NHs: respite patients were excluded, and many patients who were eligible for inclusion did not participate in the study.
Resourse Use and Disease Couse in dementia–Nursing Home (REDIC-NH), a longitudinal cohort study; design and patient characteristics at admission to Norwegian nursing homes.
In addition, two-fifths of the included patients had to be excluded from the multiple model analysis due to missing values on covariates.
Conclusion and Implications
The prevalence and persistence of prescription of PTD at admission to Norwegian NHs is high, especially for antidepressants and sedatives/hypnotics. PTD prescription rates increased for all the major classes of PTDs from admission to 6 months. Although some treatment might be justified by the severity of NPS, medication appropriateness should be carefully evaluated right after NH admission to avoid unnecessary prescriptions. Frequent NPS evaluation might be useful to target deprescribing or continuation strategies and optimize PTD prescriptions.
Acknowledgments
The REDIC-NH study was administrated by the Research Centre for Age-related Functional Decline and Disease, Innlandet Hospital Trust, Norway, and was initiated by the Norwegian Directorate of Health, which also provided funding for the data collection. The first author's PhD study was funded by the DAM Foundation, Norway.
We thank the patients and their next of kin for participating in the study and giving us their information. We also thank the nursing home managers for their cooperation, the staff members in the nursing homes who filled out the questionnaires, and the research nurses who collected the data.
Supplementary Material
Supplementary Table 1Comparison Between 484 Patients Who Were Included in the Analysis and 187 Patients Who Were Excluded/Dropped Out (N = 671)
Comparison between patients included and excluded from the study; linear mixed model is estimated for continuous variables and generalized linear mixed model for categorical variables. The models contain random effect for unit nested within NHs.
NPI-subsyndromes are calculated as the sum of the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
NPI-subsyndromes are calculated as the sum of the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
NPI-subsyndromes are calculated as the sum of the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
n = 180
n = 481
Mean (SD)
3.9 (6.0)
3.6 (5.7)
.411
NPI-Apathy
n = 180
n = 479
Mean (SD)
1.6 (3.0)
1.3 (2.7)
.140
QUALID
n = 184
n = 483
Mean (SD)
21.0 (7.6)
19.5 (6.9)
.010
Total prescribed drugs
Mean (SD)
6.4 (3.2)
6.0 (3.1)
.149
MOBID 2
n = 170
n = 473
Mean (SD)
2.4 (2.3)
2.0 (2.1)
.107
Type of unit
Regular unit
98 (52.4)
272 (56.2)
.871
Special care unit
61 (32.6)
155 (32.0)
Respite and rehabilitation unit
28 (15.0)
57 (11.8)
Number of patients per unit
n = 482
Mean (SD)
12.6 (6.7)
11.8 (6.0)
.281
Number of staff members per unit working dayshift
n = 483
Mean (SD)
3.8 (2.1)
3.7 (1.9)
.534
CSDD, Cornell scale for depression in dementia; GMHR, General Medical Health Rating Scale; MMSE, Mini-Mental Status Evaluation; MOBID-2 Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale; NPI, The Neuropsychiatric Inventory; PSMS, Physical Self-Maintenance Scale; QUALID, Quality of Life in Late-Stage Dementia.
P values < .05 (statistical significance) are reported in bold.
∗ Comparison between patients included and excluded from the study; linear mixed model is estimated for continuous variables and generalized linear mixed model for categorical variables. The models contain random effect for unit nested within NHs.
† NPI-subsyndromes are calculated as the sum of the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
Comparison between patients included and excluded from the study; linear mixed model is estimated for continuous variables and generalized linear mixed model for categorical variables. The models contain random effect for unit nested within NHs.
NPI-subsyndromes are calculated as the sum of the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
NPI-subsyndromes are calculated as the sum of the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
NPI-subsyndromes are calculated as the sum of the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
n = 402
n = 81
Mean (SD)
3.5 (5.5)
4.3 (6.4)
.246
NPI-Apathy
n = 402
n = 78
Mean (SD)
1.3 (2.7)
1.2 (2.5)
.814
QUALID
n = 402
n = 81
Mean (SD)
19.1 (6.6)
21.8 (8.0)
.005
Total prescribed drugs
n = 402
n = 82
Mean (SD)
6.1 (3.1)
5.7 (2.8)
.263
MOBID 2
n = 402
n = 82
Mean (SD)
2.0 (2.1)
2.3 (2.2)
.309
Type of unit
n = 402
n = 82
Regular unit
235 (58.5)
37 (45.1)
.001
Special care unit
115 (28.6)
40 (48.8)
Respite and rehabilitation unit
52 (12.9)
5 (6.1)
Number of patients per unit
n = 400
n = 82
Mean (SD)
12.1 (6.1)
10.1 (4.9)
.001
Number of staff members per unit working dayshift
n = 402
n = 81
Mean (SD)
3.7 (1.9)
3.4 (2.0)
.156
CSDD, Cornell scale for depression in dementia; GMHR, General Medical Health Rating Scale; MMSE, Mini-Mental Status Evaluation; MOBID-2 Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale; NPI, The Neuropsychiatric Inventory; PSMS, Physical Self-Maintenance Scale; QUALID, Quality of Life in Late-Stage Dementia.
P values < .05 (statistical significance) are reported in bold.
∗ Comparison between patients included and excluded from the study; linear mixed model is estimated for continuous variables and generalized linear mixed model for categorical variables. The models contain random effect for unit nested within NHs.
† NPI-subsyndromes are calculated as the sum of the following items: NPI-Agitation = Agitation + Disinhibition + Irritability, NPI-Psychosis = Delusions + Hallucinations, NPI-Affective = Depression + Anxiety.
Psychotropic drug prescription in nursing home patients with dementia: Influence of environmental correlates and staff distress on physicians' prescription behavior.
Association between behavioral and psychological symptoms and psychotropic drug use among old people with cognitive impairment living in geriatric care settings.
Resourse Use and Disease Couse in dementia–Nursing Home (REDIC-NH), a longitudinal cohort study; design and patient characteristics at admission to Norwegian nursing homes.
Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale (MOBID): Development and validation of a nurse-administered pain assessment tool for use in dementia.
Stability of the factor structure of the Neuropsychiatric Inventory in a 31-month follow-up study of a large sample of nursing-home patients with dementia.
Prevalence of depression among recently admitted long-term care patients in Norwegian nursing homes: Associations with diagnostic workup and use of antidepressants.
Impact of antipsychotic review and nonpharmacological intervention on antipsychotic use, neuropsychiatric symptoms, and mortality in people with dementia living in nursing homes: A factorial cluster-randomized controlled trial by the Well-Being and Health for People With Dementia (WHELD) program.
30740-4&id=gr1.jpg){kind=link}