# Prescribing cascades in older adults with and without dementia

> **NIH NIH K24** · UNIVERSITY OF CALIFORNIA, SAN FRANCISCO · 2022 · $178,374

## Abstract

PROJECT SUMMARY / ABSTRACT
Background: In older adults, individual prescribing decisions may appear reasonable in isolation yet can lead
to a tangled web of complications. A key exemplar of this is prescribing cascades. These occur when a
medication causes an adverse effect, and rather than changing the offending drug, a second medication is
given to manage the symptoms. This in turn may cause more adverse effects that are managed with a third
medication, and so forth. While relevant for all older adults, cascades may be particularly common and harmful
in people with dementia, in part because communication challenges may increase the risk that clinicians will
not recognize that an adverse drug effect has occurred, and instead interpret new signs or symptoms as the
manifestation of an underlying disease. Although prescribing cascades are likely common and can confer
substantial morbidity, there has been remarkably little scholarship on this topic. In this proposal we will explore
two potential cascades that are models of this problem: (A) gabapentin and pregabalin  peripheral edema 
loop diuretics, and (B) loop diuretics  urinary symptoms  medications used to treat urinary symptoms.
Aims: (1) To determine the prevalence and risk factors for a prescribing cascade involving gabapentinoids and
loop diuretics, and how these differ between older adults with and without dementia; (2) To determine the
prevalence and risk factors for a prescribing cascade involving loop diuretics and urinary symptom
medications, and how these differ between older adults with and without dementia; (3) To characterize the
pathways through which the gabapentinoid-loop diuretic cascade occurs in these populations.
Methods: Using national data from VA and Medicare, for Aims 1 and 2 we will assemble a retrospective cohort
of older adults who were non-users of both the index medication (the one that starts the cascade) and the
outcome medication (the one used to treat adverse effects caused by the index medication). We will then
identify which subjects subsequently started the index medication and which did not start the index medication
but instead started another type of medication (i.e., a control group). We will compare use of the outcome
medication between these groups using multivariable time-to-event approaches that account for competing
risks. Interaction terms and stratified analyses will be used to compare outcome rates and risk factors among
people with and without dementia. For Aim 3, we will conduct chart review on 500 subjects (250 with dementia
and 250 without) with the gabapentinoid-loop diuretic cascade and evaluate elements of decision-making that
contributed to the cascade. The proposed research will provide tremendous opportunities for mentoring junior
investigators and supporting the mentoring aims of this K24 award.
Relevance / public health significance: Understanding prescribing cascades and the patterns of decision-
making that lead to them will info...

## Key facts

- **NIH application ID:** 10376243
- **Project number:** 5K24AG049057-07
- **Recipient organization:** UNIVERSITY OF CALIFORNIA, SAN FRANCISCO
- **Principal Investigator:** MICHAEL A. STEINMAN
- **Activity code:** K24 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2022
- **Award amount:** $178,374
- **Award type:** 5
- **Project period:** 2015-04-01 → 2026-01-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10376243

## Citation

> US National Institutes of Health, RePORTER application 10376243, Prescribing cascades in older adults with and without dementia (5K24AG049057-07). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10376243. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*