A Prescribing Cascade of Proton Pump Inhibitors Following Anticholinergic Medications in Older Adults With Dementia

Abstract

Introduction: Prescribing cascade refers to use of a medication to treat a drug-related adverse event. Prescribing cascades increase medication use, cost, and risk of adverse events.Objective: Our objective was to use administrative health data to identify whether use of medications from the anticholinergic cognitive burden scale was associated with proton pump inhibitor (PPI) prescribing consistent with a prescribing cascade in older adults with dementia.Method: The cohort was comprised of Nova Scotia Seniors’ Pharmacare beneficiaries identified to have dementia and medication dispensation data recorded between 1 April 2010, or cohort entry and 31 March 2015. Anticholinergic medications from the anticholinergic cognitive burden scale (ACB) were abstracted. A look back period of 365 days identified if a PPI had been dispensed preceding anticholinergic dispensation. PPI initiation within 30, 60, 90, or 180 days of the anticholinergic medication was assessed. Demographic description of those dispensed anticholinergic medications or PPIs were reported. Risk factors for the prescribing cascade were investigated with logistic regression and Cox proportional hazards modelling including a sex-stratified analysis.Results: We identified 28,952 Nova Scotia Seniors’ Pharmacare beneficiaries with dementia and prescription dispensation data. Anticholinergic medications were frequently dispensed with 63.4% of the cohort dispensed at least one prescription for an anticholinergic medication. The prescribing cascade defined as up to 180-days between anticholinergic medication inititation and PPI dispensation, occurred in 1,845 Nova Scotia Seniors’ Pharmacare beneficiaries with dementia (incidence 6.4%). Multivariate regression showed those experiencing the prescribing cascade after initiating any anticholinergic were younger (OR 0.98, 95%CI [0.97–0.98]), less likely to live in an urban location (OR 0.82, 95%CI [0.74–0.91]), or to be men (OR 0.74, 95%CI [0.67–0.82]). Cox regression demonstrated an increased risk of starting a PPI within 180 days when initiating any medication from the ACB (HR 1.38, 95%CI [1.29–1.58]).Discussion: Regression modelling suggested that anticholinergic medications increased the risk of PPI dispensation consistent with a prescribing cascade in the cohort. The identification of the prescribing cascade in this population of older Nova Scotia Seniors’ Pharmacare Program beneficiaries with dementia using administrative health data highlights how routinely collected health data can be used to identify prescribing cascades.

Investigation of geriatric syndromes associated with medication in Japan using insurance claims data

“…These results suggested that GSAM may be a risk factor for prescribing cascades. It has been suggested that dementia may be caused by a prescribing cascade of therapeutic medications 17,18 . A higher prevalence of dementia (12.1%) was observed in participants subjected to a prescription cascade than in those prescribed medications specifically for GS.…”

“…It has been suggested that dementia may be caused by a prescribing cascade of therapeutic medications. 17,18 A higher prevalence of dementia (12.1%) was observed in participants subjected to a prescription cascade than in those prescribed medications specifically for GS. Our result supports those of previous studies.…”

An investigation of psychoactive polypharmacy and related gender-differences in older adults with dementia: a retrospective cohort study

“…Use of medications from the following four classes; antidepressants (tricyclic antidepressants, selective serotonin reuptake inhibitors (SSRIs), bupropion and trazodone), cholinesterase inhibitors, antipsychotics (first- and second-generation), benzodiazepines, and nonbenzodiazepine benzodiazepine receptor agonist hypnotics (z-drugs)) was reported. Psychoactive medications from these classes were chosen as those more commonly used to manage behavioural symptoms of dementia and this data were available for analysis from another cohort study [ 29 – 31 ]. Data were not available for some other medications that could have been relevant to the analysis, including the atypical antidepressant mirtazapine, the selective norepinephrine reuptake inhibitor duloxetine, and anticonvulsants.…”

An investigation of psychoactive polypharmacy and related gender-differences in older adults with dementia: A retrospective cohort study

“…Psychoactive medications from these classes were chosen as those more commonly used to manage behavioural symptoms of dementia and this data were available for analysis from another cohort study. [28][29][30] Data were not available for some other medications that could have been relevant to the analysis, including the atypical antidepressant mirtazapine, the selective norepinephrine reuptake inhibitor duloxetine, and anticonvulsants. Medication use was approximated from the dispensing data recorded in the Seniors' Pharmacare database (PHARM).…”