---
title: "Psychotropic drug prescribing trends in Dutch nursing home residents with dementia, 2016–2023"
id: "plos-one-17-trends-in-psychotropic-drug-use-in-nursing-home-residents-with-dementia-and"
canonical_url: "https://medichelpline.com/clinical-feed/plos-one-17-trends-in-psychotropic-drug-use-in-nursing-home-residents-with-dementia-and"
content_type: "clinical_feed_article"
specialty: "Neurology"
source_name: "PLOS ONE (Medicine)"
source_url: "https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0361903"
published_at: "2026-11-04T14:00:00.000Z"
evidence_level: "Journal Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Psychotropic drug prescribing trends in Dutch nursing home residents with dementia, 2016–2023
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/plos-one-17-trends-in-psychotropic-drug-use-in-nursing-home-residents-with-dementia-and
- **Specialty:** [Neurology](https://medichelpline.com/clinical-feed/neurology.md)
- **Primary Source:** PLOS ONE (Medicine)
- **Source URL:** [Original Journal Publication](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0361903)
- **Published At:** 2026-11-04T14:00:00.000Z
- **Evidence Rating:** Journal Feed
## Executive GIST (TL;DR)
- Routine care data from electronic health records and prescription systems were analyzed for residents in dementia special care units across 11 Dutch nursing home organizations. The dataset included prescription records and demographic and comorbidity information extracted for the period January 1, 2016 to March 31, 2023. - Investigators calculated quarterly point prevalences for any psychotropic drug prescription and for five ATC-defined categories: **antipsychotics**, anxiolytics, hypnotics, **antidepressants**, and anti-dementia drugs, distinguishing regular versus PRN use. - The analytic approach combined descriptive quarterly prevalences with Wilson score 95% confidence intervals and segmented interrupted time series (ITS) regression to test for trend changes across three periods tied to regulatory events. - The cohort comprised 17,738 residents. Data indicated an overall decline in psychotropic prescriptions from 48.3% in 2016 to 42.7% in 2023 when considering regular prescriptions. Antidepressants and antipsychotics were the most frequently prescribed classes. - Antidepressants showed the largest absolute decrease in prevalence over time; anti-dementia drugs exhibited the greatest relative decline. PRN prescription prevalence was reported separately by quarter. - ITS analysis revealed a steeper declining trend after the combined introduction of the **Care and Coercion Act** (CCA) and the Dementia Care Standard in Q2 2020 compared with the trend after the 2018 guideline revision. Investigators therefore observed an apparent temporal association between regulatory changes and reduced prescribing. - The authors noted persistent, substantial psychotropic use despite declines, and stressed that routine data cannot alone assess appropriateness of prescribing. Data access is restricted; some residents may lack documented dementia diagnoses, and anticonvulsants were excluded because of mixed indications. - Key limitations documented: exclusion of rows with missing admission details (about 40% of rows), potential under-recording of diagnoses in EHR free text, and that data are pseudonymized and stored in a secured institutional environment and not publicly available.
## Clinical Analysis & Structured Key Points
Trends in psychotropic drug use in nursing home residents with dementia and their association with regulatory changes: A routine care data analysis from 2016 to 2023 | PLOS One Browse Subject Areas ? Click through the PLOS taxonomy to find articles in your field. For more information about PLOS Subject Areas, click here . Article Authors Metrics Comments Media Coverage Reader Comments Figures Figures Abstract Introduction Non-pharmacological interventions are the first-choice treatment for managing neuropsychiatric symptoms in nursing home (NH) residents with dementia, yet psychotropic drugs (PDs) are frequently prescribed despite their limited effectiveness. In the current century, policies in various countries have increasingly emphasized the need for restraint in prescribing PDs, partly through amendments to guidelines and legislation. In the Netherlands, regulatory changes included the 2018 revision of the guideline for challenging behavior, the introduction of the Care and Coercion Act (CCA), and a new Dementia Care Standard in 2020. This study examined the prevalence of PD prescriptions (PDPs) from 2016 to 2023 among residents with dementia in Dutch NHs, and the course of the PDP trend in the context of regulatory changes, using routine care data. Methods We analyzed routine care data from 11 NH organizations. Quarterly point prevalences were calculated for any PDP and five PD categories separately (antipsychotics, anxiolytics, hypnotics, antidepressants, anti-dementia drugs). We calculated confidence intervals to indicate temporal changes in PDP prevalence and assess the relationship with different reforms. We used interrupted time series (ITS) analysis to explore the association between regulatory changes and the PDP trend. Results Data from 17,738 residents showed an overall decline in PDP prevalence from 48.3% in 2016 to 42.7% in 2023. Antidepressants and antipsychotics were the most frequently prescribed PDs. Antidepressants showed the largest absolute prevalence decline over time. Anti-dementia drugs showed the greatest relative decline. ITS showed a steeper declining trend after the combined introduction of the CCA and Dementia Care Standard in the second quarter of 2020, compared with the period after the guideline revision in 2018. Discussion The PDP rate decreased over time but was still substantial. Regulatory changes, including new legislation and (clinical) guidelines, appear to have contributed to a further decline from Q2 2020 onwards. This decline suggests that policy and guideline interventions may meaningfully influence prescribing behavior, offering a promising lever for reducing potentially inappropriate psychotropic use in dementia care. Insight into the appropriateness of prescribing practices is essential to identify areas for further improvement. Citation: Albers CAW, Wouters F, de Bruijne MC, Smalbrugge M, Joling KJ (2026) Trends in psychotropic drug use in nursing home residents with dementia and their association with regulatory changes: A routine care data analysis from 2016 to 2023. PLoS One 21(11): e0361903. https://doi.org/10.1371/journal.pone.0361903 Editor: Ioannis Liampas, University of Thessaly Faculty of Medicine: Panepistemio Thessalias Tmema Iatrikes, GREECE Received: June 26, 2026; Accepted: October 16, 2026; Published: November 4, 2026 Copyright: © 2026 Albers et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: The data that support the findings of this study are not publicly available due to privacy and confidentiality restrictions. Although the data are pseudonymized, they contain information that could potentially identify individual participants. In addition, the participating nursing home organizations consented to the use of their data only within specific scientific organizations in the Netherlands, as laid down in the participation agreements. These restrictions were imposed by Gerimedica, the participating organizations, and the Amsterdam UMC. Therefore, the data cannot be deposited in a public repository. Data can be requested from the ‘Learning from Data’ program at Amsterdam UMC, Department of Medicine of Older People (email: lerenvandata@amsterdamumc.nl ). This program is an institutional point of contact that operates independently of the individual authors; requests are reviewed by the program leader, in accordance with the conditions set by the participation agreements. Data can only be accessed and analyzed within the secured research environment of Amsterdam UMC. To ensure long-term availability, the data are stored on the secure institutional research storage infrastructure of Amsterdam UMC, in accordance with the institution’s data management and security policies. The data will be retained for at least 15 years after completion of the study, in line with the Dutch legal retention period for research data involving human participants. Funding: This research is part of the national program ‘Leren van Data’ (‘Learning from Data’), launched in the Netherlands in 2020, which aims to promote the reuse of routine care data from NHs for scientific research, quality improvement, and policy recommendations. The program is funded by the Ministry of Health, Welfare, and Sports in the Netherlands (Leren van Data II, grant number: 90002320). The funder had no role in the study design, methods, data collection, analysis, or preparation of this article. Competing interests: The authors have declared that no competing interests exist. 1. Background People with dementia and other neurocognitive disorders frequently exhibit neuropsychiatric symptoms (NPSs), such as agitation, apathy, or sleep disturbances [ 1 ]. These symptoms significantly impact individuals with dementia, their relatives, and care professionals [ 2 ]. Therefore, effective NPS management is essential. Although most guidelines recommend non-pharmacological interventions as the first-line treatment and new research is being performed to find suitable interventions for NPSs [ 3 ], psychotropic drugs (PDs) are still often prescribed to manage challenging behaviors [ 4 ]. However, the evidence supporting the efficacy of PDs in treating NPSs in dementia is limited, and their use can result in severe adverse effects, such as an increased risk of falls or fractures [ 5 – 10 ]. Besides, people with dementia are more susceptible to polypharmacy and drug-drug interactions, and adverse drug events can decrease the patients’ quality of life and worsen their prognosis [ 11 ]. Therefore, it is not only important to reduce PD prescriptions (PDPs) overall, but also to reduce the number of different PDs a patient receives. In the Netherlands, about 25% of people with dementia live in nursing homes (NHs) [ 12 ]. Several studies have assessed the prevalence of PDPs in NH residents with dementia. Across Western European NHs, PDs, especially antipsychotics and antidepressants, are widely prescribed, with antipsychotic use ranging from 12% to 59% and antidepressant use ranging from 19% to 68% [ 13 ]. In the Netherlands, Zuidema et al. [ 14 ] reported a stable PDP prevalence of approximately 60% from 2003 to 2011. However, a more recent analysis using baseline data from cohort studies and cluster-randomized controlled trials (RCTs) from the Netherlands noted a decrease from 63% in 2003 to 40% in 2018, with antipsychotics and antidepressants remaining the most commonly prescribed PDs [ 15 ]. In the current century, policies in various countries have increasingly emphasized the need for restraint in prescribing PDs, partly through amendments to guidelines and legislation. Already in 2005, the United States Food and Drug Administration (FDA) issued a public health advisory warning of fatal adverse events in patients with dementia treated with atypical antipsychotic drugs [ 16 ]. The United Kingdom developed a national dementia strategy in 2009 [ 17 ]. In Australia, the Royal Commission into Aged Care Quality and Safety, which ran from 2018 to 2021, introduced several aged care reforms, including targeting the inappropriate use of chemical restraint and the mandatory reporting on antipsychotic use [ 18 , 19 ]. More recently, the FDA wrote a report to address the inappropriate use of antipsychotic drugs in NHs [ 20 ]. In the Netherlands, long-term care (LTC) has also undergone substantial changes in recent years. First, in 2015, the Dutch government introduced a major LTC reform with the new Long-Term Care Act (Dutch: Wet Langdurige Zorg, WLZ). Most importantly, they initiated the reform to control spending growth and ensure the long term financial viability of LTC, as well as to improve its quality by making it more client-tailored. One pillar of the reform was shifting from residential to non-residential care [ 21 ], with residents admitted to NHs at more advanced stages of dementia, exhibiting more severe challenging behavior [ 22 ]. Second, the Dutch Association of Elderly Care Physicians (Dutch: Verenso) revised its guideline on challenging behavior in 2018 [ 23 ], providing explicit recommendations on PDPs for NH residents with dementia by placing more emphasis on non-pharmacological interventions and increased restraint in PDPs. Third, a new law concerning involuntary care for people with psychogeriatric syndromes was enacted in 2020, called the Care and Coercion Act (CCA) [ 24 ]. This legislation equates PDs to involuntary care if they are not prescribed in line with guideline recommendations for persons unable to consent, theoretically raising the threshold for prescribing PDs. Finally, in 2020, the Dementia Care Standard 2020 was launched as a revised version of the Dementia Care Standard 2013, alongside national campaigns by Alzheimer Nederland promoting minimal use of PDs and greater use of psychosocial interventions [ 25 , 26 ]. Given these changes over the last decade, it is important to know whether these initiatives reduced the prevalence of PDPs. Recent attention has focused on reducing PDPs. However, the relationship between regulatory changes and PDP rates remains unclear. To address this gap, research using a large, unselected sample over an extended period is needed. Most previous studies on PDP prevalence in residents with dementia relied on clinical trial data [ 14 , 15 ] or cross-sectional designs, which limit insight into PDP patterns over time [ 27 – 32 ]. Studies using routine care data over extended periods were conducted less recently (2000–2013) [ 33 , 34 ]. Therefore, this study aims to assess the prevalence of PDPs among residents admitted to dementia special care units in Dutch NHs and the course of the PDP trend in the context of regulatory changes, using electronic prescription system (EPS) data from 2016 to 2023. Using routine care data allows us to evaluate real-world prescribing practices over time, reflecting usual clinical practice rather than research-driven data collection, across a large, unselected population that would be difficult to capture through prospective data collection alone. To reach this aim, we have formulated the following research question: What is the prevalence of PDPs among NH residents with dementia between 2016 and 2023, and are observed prevalences of PDPs associated with regulatory changes? 2. Materials and methods This study used an observational design over a long period, using routine care data from electronic health records (EHRs) and an EPS for people residing in dementia special care units in NHs in the Netherlands. 2.1. Data sources We conducted this research within the ‘Learning from Data’ (Dutch: Leren van Data) research program, which promotes the (re)use of routine care data in NHs [ 35 ]. Within this research program, we use pseudonymized data from the EHR Ysis and EPS Medimo for scientific research. Of the 27 organizations that granted permission to share data from both EHR Ysis and EPS Medimo for this research, 11 used both systems at the start and throughout the study period; therefore, we included data from NH residents in these 11 organizations in the study sample. 2.2. Study sample We included NH residents residing in a dementia special care unit for at least one day between 1 January 2016 and 31 March 2023. Most people with dementia or similar psychogeriatric syndromes residing in NHs are admitted to dementia special care units [ 36 ]. Therefore, we included only residents of dementia special care units and excluded those who received geriatric rehabilitation or somatic care. The dementia special care units included residents of four types of care: chronic psychogeriatric care, gerontopsychiatric care, Huntington care, and Korsakoff care. Residents in gerontopsychiatric, Huntington, and Korsakoff care units were included alongside people with chronic psychogeriatric care needs, as these groups share a comparable profile of chronic cognitive and behavioral impairment and require the same intensity of long-term, 24-hour supervised care. Some of the residents therefore do not have a dementia diagnosis. In addition, some dementia diagnoses may not have been recorded in the EHR. We excluded data from residents with an unexpected age ( 110 years) (n = 1) or a negative length of stay (n = 22) (<0.01%). 2.3. Data collection We extracted data from the EPS and EHR for analysis between January 2016 and March 2023. We accessed the data on 14 February 2024. We retrieved data on age, sex, type of care, date of admission, (when applicable) date of deregistration or death, and organizational ID from the EHR. Furthermore, the presence of various medical conditions was assessed, including COPD, heart failure, other cardiovascular diseases without heart failure, stroke, diabetes, reduced kidney function, liver disease, dementia, Parkinson’s disease, Korsakoff, Huntington’s disease, other chronic progressive neurological diseases, delirium, cancer or leukemia, and indication of palliative care or terminal/end-stage disease. We derived the conditions from the medical history free-text field in the EHR using a query with the most common disease names and synonyms to extract the mentioned diagnoses [ 37 ]. A researcher and elderly care physician developed the query, and a medical informatician refined it. It is based on the comorbidities that are used to calculate the Charlson Comorbidity Index [ 38 ]. To improve the query, elderly care physicians (in training) tested it using an iterative process. Residents whose medical history consisted completely of conditions that we did not include in our comorbidity list were classified as having ‘other conditions only’ [ 39 ]. Information on medication prescriptions, such as drug ‘Anatomical Therapeutic Chemical’ (ATC) code, start and stop dates, administration frequency, dosing, and whether it was a regular or ‘as needed’ (pro re nata, PRN) prescription, was extracted from the EPS. Based on the first four alphanumeric characters of the drug ATC code, PDPs can be divided into main categories: antidepressants (N06A), anxiolytics (N05B), hypnotics (N05C), antipsychotics (N05A), and anti-dementia drugs (N06D) (World Health Organization, 2024). Anticonvulsants (N03A) were excluded from this study because they are frequently prescribed for reasons other than NPS, such as epilepsy and neuropathic pain [ 40 ]. Baseline variables such as year of birth and sex had no missing values, just as variables on medication prescriptions. When information to define the type of care and admission period was missing, we excluded these residents (almost 40% of the data rows). From previous work, we know that people without an admission goal are mostly those residing in the community or receiving short-term care and were therefore probably not part of our study population. 2.4. Prevalence of psychotropic drugs The point prevalence of PDPs was calculated by dividing the number of residents with at least one PDP on the first day of each quarter of the year by the number of residents residing in a participating NH in a dementia special care unit on that day, from January 1, 2016, to March 31, 2023. Quarterly point prevalence of PDPs was calculated overall (i.e., a prescription in at least one of the PD categories) and for the five main ATC categories separately. In the analysis, we distinguished between regular and PRN prescriptions. To assess psychotropic polypharmacy, we calculated the number of different categories from which a resident had a prescription. Based on the five main ATC categories, residents with a PDP could have prescriptions in at least one and up to five different categories. To assess inter-organizational differences, we also calculated the overall quarterly point prevalence of regular PDPs per organization. To assess whether the exclusion of anticonvulsants may have obscured substitution effects, we calculated the quarterly prevalence of anticonvulsant use separately, using the same methods and reported the outcomes in the supporting information ( S5 Table ). 2.5. Data analysis We used IBM SPSS Statistics 28 [ 41 ], R and RStudio version 4.4.3 [ 42 ] to analyze the data, and figures were produced using the package ggplot2 [ 43 ] and GraphPad Prism version 10.6.0 [ 44 ]. We used descriptive statistics to assess the quarterly point prevalence of PDPs among NH residents admitted to dementia special care units from 2016 to the first quarter of 2023. To assess temporal changes in prevalence and course of the PDP trend in the context of regulatory changes, we calculated Wilson score 95% confidence intervals (CIs) for each quarter. We compared them to three quarters matching the regulatory changes: Q1 2016 for the first available quarter after the introduction of the new Long-Term Care Act, Q1 2018 for the revision of the guideline on challenging behavior, and Q2 2020 for the combined introduction of the CCA and the Dementia Care Standard 2020. When looking only at regular PDPs, the reference values are Q1 2016: 0.483 (95% CI, 0.467 to 0.499); Q1 2018: 0.476 (95% CI, 0.460 to 0.491); and Q2 2020: 0.452 (95% CI, 0.437 to 0.467). When including PRN PDPs, the reference values are Q1 2016 (0.193; 95% CI, 0.181 to 0.206), Q1 2018 (0.215; 95% CI, 0.203 to 0.228), and Q2 2020 (0.215; 95% CI, 0.202 to 0.227). We used non-overlapping 95% CIs as a conservative indicator of differences between quarters. As these comparisons were exploratory, no adjustment for multiple comparisons was applied. We used interrupted time series (ITS) analysis to explore whether changes in the prevalence trend of PDPs coincided with regulatory changes. The 2015 Long-Term Care Act preceded the study period and was not modeled. We combined the introduction of the CCA (Q1 2020) and the Dementia Care Standard 2020 (Q2 2020) into a single period, resulting in three periods: baseline (2016 Q1-2017 Q4), post-guideline revision (2018 Q1-2020 Q1), and post-dementia care standard/CCA (2020 Q2 onward). We first compared mean prevalence between periods. We then used segmented linear regression, including a dummy-coded period variable, a continuous quarter variable, and their interaction, to test whether the linear trend across quarters differed between periods, allowing effects to be evaluated relative to both baseline and the immediately preceding period. 2.6. Ethics The Medical Ethics Committee of the VU University Medical Center determined that the Medical Research Involving Human Subjects Act (Dutch: WMO) did not apply to this study and granted a waiver for ethics approval (reference number: METc VUmc 2016.516 and 2022.0076). The authors did not have access to information that could identify individual participants during or after data collection. We used a no-objection (opt-out) procedure, in which residents of NHs were informed by their healthcare orga
## Related Clinical Research

- [Revolutionizing Female Heart Health: New Life Stage Framework by Medical Societies](https://medichelpline.com/clinical-feed/medical-news-today-1-female-heart-health-top-medical-groups-issue-female-life-stage-framework.md)
- [Controlling Hypertension Could Avert Over 500,000 Deaths Annually, Analysis Finds](https://medichelpline.com/clinical-feed/medical-news-today-1-over-500-000-deaths-a-year-could-be-prevented-by-managing-blood-pressure.md)
- [Pre-diagnostic 1-year prevalence of polypharmacy and PIMcog in Norwegian outpatients with MCI or d](https://medichelpline.com/clinical-feed/plos-one-14-one-year-pre-diagnostic-prevalence-and-associated-factors-of-polypharmacy-and.md)
- [High-resolution map of the basal ganglia reveals cell types tied to Parkinson’s and schizophrenia](https://medichelpline.com/clinical-feed/stat-news-1-stat-scientists-debut-map-of-basal-ganglia-a-key-brain-region-involved-in-an.md)
- [SniffCell: Cell-type-resolved somatic variant discovery from bulk long-read sequencing](https://medichelpline.com/clinical-feed/medrxiv-11-cell-type-resolved-somatic-variant-discovery-from-bulk-long-read-sequencing.md)

## Navigation
- [← Back to Neurology Feed](https://medichelpline.com/clinical-feed/neurology.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.