Abstract
IMPORTANCE: Prescribing cascades are an underrecognized driver of polypharmacy among older adults (aged ≥65 years). The clinical decision-making processes underlying cascades and their downstream consequences are poorly understood. OBJECTIVE: To explore clinical reasoning leading to prescribing cascades and downstream outcomes (eg, falls, electrolyte abnormalities) via the exemplar gabapentinoid (gabapentin and pregabalin)-loop diuretic (LD) cascade. DESIGN, SETTING, AND PARTICIPANTS: This cohort study randomly selected medical records from a cohort of US veterans aged 66 years or older between January 1, 2013, and August 31, 2019, who potentially experienced the gabapentinoid-LD prescribing cascade. The medical record review and data analysis were performed between October 24, 2023, and July 22, 2025. EXPOSURES: Initiation of gabapentinoid and LD. MAIN OUTCOMES AND MEASURES: Abstractors evaluated clinical documentation in the 30 days prior to and 60 days after LD initiation to evaluate decision-making processes and potential downstream outcomes of the gabapentinoid-LD cascade. Secondary analyses examined whether a dementia diagnosis was associated with clinician decision-making and patient outcomes. RESULTS: The analytic cohort comprised 120 patients (mean [SD] age, 73.9 [7.1] years; 116 male [96.7%]; 106 [88.3%] taking ≥5 long-term medications). Documentation of a differential diagnosis for edema was noted in 73 patients (60.8%), most commonly referencing congestive heart failure (n = 47 [39.2%]) and/or venous stasis (n = 16 [13.3%]). Gabapentinoids were rarely noted in the differential (n = 4 [3.3%]). The majority of clinicians documented the indication for LD (n = 116 [96.7%]), most commonly for lower-extremity edema (n = 104 [86.7%]), congestive heart failure (n = 16 [13.3%]), and/or dyspnea (n = 15 [12.5%]). In the 60 days following LD initiation, 28 patients (23.3%) experienced 37 events potentially attributable to LD initiation. The most common downstream events were worsening kidney function (n = 9 [7.5%]), orthostasis (n = 7 [5.8%]), electrolyte abnormalities (n = 6 [5.0%]), and falls (n = 5 [4.2%]). Six patients (5.0%) were evaluated in the emergency department and/or hospital for potential downstream events. Documentation of differential diagnoses, indications, actions taken regarding gabapentinoids, and downstream events generally did not vary between patients with and without dementia. CONCLUSIONS AND RELEVANCE: This cohort study found that among older veterans who received LD following gabapentinoid initiation and experienced a potential gabapentinoid-LD prescribing cascade, clinicians almost never explicitly considered gabapentinoid adverse drug effects in their treatment of edema. These findings suggest that potential downstream harms of this overlooked prescribing cascade are common, underscoring the importance of addressing prescribing cascades in clinical practice.