This registered report protocol outlines a retrospective cohort study that will estimate the 12-month pre-diagnostic prevalence of polypharmacy and potentially inappropriate medications for cognitive impairment (PIMcog) among community-dwelling adults assessed in Norwegian specialist outpatient clinics for cognitive symptoms between 2014 and 2024. The study links the Norwegian Registry of Persons Assessed for Cognitive Symptoms (NorCog) to dispensing records from the Norwegian Prescribed Drug Registry (NorPD) and comorbidity data from the Norwegian Patient Registry (NPR). The objectives are to quantify the prevalence of polypharmacy and PIMcog use in the year preceding diagnosis of mild cognitive impairment (MCI) or dementia, compare characteristics of PIMcog users and non-users, and identify factors associated with PIMcog exposure prior to diagnosis.
The incidence of MCI and dementia rises sharply with age, and demographic shifts in Norway indicate a growing older population. NorCog, established in 2013 as a national quality registry, collects standardized diagnostic and treatment data from specialist clinics assessing cognitive symptoms. By 2024, NorCog included data from 46 clinics and roughly 27,500 patients (mean age ~74 years), with about 85% diagnosed with MCI or dementia.
Multiple comorbidities are common in older adults with cognitive disorders and increase with dementia severity. These conditions—such as depression, diabetes, coronary heart disease, stroke, hypertension, heart failure, and incontinence—are commonly treated pharmacologically, contributing to high medication counts. Prior studies indicate that approximately half of individuals with dementia are exposed to polypharmacy and that average medication counts are higher than in cognitively healthy controls; one Norwegian population-based comparison reported mean medication counts of 5.1 ± 3.6 in Alzheimer’s dementia versus 2.9 ± 2.4 in controls.
Prescription tools used in geriatric medicine, including the Screening Tool of Older Persons’ Prescriptions (STOPP) and the Beers criteria, identify certain drug classes as potentially inappropriate for persons with cognitive disorders. These classes include central nervous system depressants (for example, benzodiazepines and related drugs), certain antipsychotics, and agents with strong anticholinergic properties. Such medications can interact adversely with cognitive disorders by worsening cognition or increasing functional impairment and risk of adverse drug events. Reported prevalence of PIMcog varies by setting and country; studies in memory clinics and community settings have found wide ranges (for example, 20–30% in some community samples and higher proportions in specialist referrals).
Prescribing patterns and PIMcog prevalence vary across healthcare systems, and no prevalence study from Nordic specialist outpatient clinics has been described previously. Trends in Norway show a reduction in antipsychotic prescribing in nursing homes and among community-dwelling individuals with dementia over recent decades, but only limited changes for other psychotropic drugs.
Design and data sources
This is a retrospective cohort study linking three national registries: NorCog (diagnostic and clinical data from specialist outpatient assessments), NorPD (individual-level pharmacy dispensing records for all prescriptions dispensed at Norwegian pharmacies), and NPR (hospital- and specialist-care comorbidity information). Linkage will use the unique national personal identifying number recorded in the registries.
Study population
All patients recorded in NorCog and diagnosed with mild cognitive impairment (MCI) or dementia in specialist outpatient clinics from 2014 through 2024 will be eligible. NorCog’s coverage in 2024 included 46 clinics and approximately 27,500 registered patients with a mean age of about 74 years; roughly 85% have diagnoses of MCI or dementia.
Outcomes and definitions
Primary outcomes are the twelve-month prevalence prior to diagnosis of (1) polypharmacy and (2) PIMcog use. PIMcog will be identified based on medication classes flagged by established geriatric prescribing tools (for example, STOPP and Beers criteria) and will include central nervous system depressants such as benzodiazepines and related agents, certain antipsychotics, and drugs with strong anticholinergic effects. Polypharmacy is considered in the context of the total number of medications dispensed during the 12-month pre-diagnostic window.
Covariates and comorbidity
Comorbidity data will be extracted from NPR and summarized using established measures such as the Charlson Comorbidity Index where appropriate. Sociodemographic and clinical characteristics from NorCog (for example, age, sex, diagnostic category, assessment clinic) will be compared between PIMcog users and non-users.
Analyses
The protocol specifies estimating the twelve-month prevalence of polypharmacy and PIMcog use in the year prior to MCI or dementia diagnosis. Comparative analyses will examine differences in sociodemographic and clinical characteristics between PIMcog users and non-users. Regression models will be used to identify factors associated with PIMcog exposure, incorporating comorbidity measures and other relevant covariates. Exact analytic methods, variable coding, and statistical modeling details are prespecified in the registered protocol (supporting materials to be made available via the Open Science Framework).
Data access and sharing
Individual-level registry data cannot be publicly released due to legal and ethical constraints. Qualified researchers may apply for access to NorPD, NPR, and NorCog through the respective data custodians and ethics approvals. Study materials and supporting documentation will be shared on OSF consistent with the registered report format.
The study aims explicitly to: (1) estimate the 12-month prevalence of polypharmacy and PIMcog use prior to diagnosis of MCI or dementia in Norwegian specialist outpatient clinics; (2) describe prescribing patterns in the year before diagnosis; and (3) identify patient-level factors associated with PIMcog use to inform targeted deprescribing and medication optimization strategies.
By providing population-level estimates of pre-diagnostic polypharmacy and PIMcog exposure in a national outpatient registry cohort, the study intends to inform clinical efforts to reduce medication-related harms in persons with cognitive impairment. Characterizing prescribing patterns and associated patient factors may guide targeted screening, deprescribing interventions, and safer pharmacotherapy for this vulnerable group.
The protocol notes that linkage of NorCog, NorPD, and NPR is conducted under applicable legal and ethical approvals. Individual-level data cannot be deposited publicly because this would violate conditions set by ethics committees and data protection regulations. Access for external researchers requires formal applications to the relevant data custodians and approval from ethics and regulatory bodies.
Older adults assessed for cognitive symptoms often have multiple comorbidities and higher medication use compared with cognitively healthy peers. The protocol highlights gaps in Nordic data on pre-diagnostic polypharmacy and PIMcog prevalence in specialist outpatient settings and the need to understand local prescribing patterns. International literature indicates wide variation in PIMcog prevalence and frequent use of psychotropic and anticholinergic medications in some referral populations; this study will provide Norway-specific estimates using comprehensive dispensing and clinical registries.
Limitations inherent to registry linkage are acknowledged implicitly: registry data capture dispensed prescriptions rather than confirmed ingestion, and registry-based comorbidity coding depends on healthcare contact and documentation. The protocol addresses transparency by registering methods and making supporting materials available via OSF while respecting data protection constraints.
This registered report protocol describes a planned registry-linked study to estimate the one-year pre-diagnostic prevalence of polypharmacy and PIMcog use among community-dwelling adults diagnosed with MCI or dementia in Norwegian specialist outpatient clinics between 2014 and 2024. Results are intended to inform deprescribing initiatives and safer prescribing practices for individuals with cognitive impairment. Specific analytic details, timeline, and supporting documents are provided in the registered materials, and individual-level registry data remain available to qualified researchers through formal application channels.