(BSPOT45) High-throughput sequence symmetry analysis to detect Beta-Blocker induced prescribing cascades

Background: Beta-blocker (BB) use is ubiquitous in patients with hypertension, particularly with underlying cardiac disease, such as heart failure. Due to the widespread use of BBs, adverse effects (AEs) are common and may go undetected, prompting initiation of a new treatment indicated for such BB-related AEs – a so-called prescribing cascade.

Objectives: To screen for potential BB-induced prescribing cascades via high throughput sequence symmetry analysis (SSA).

Methods: Using claims from 5% (2011-15) and 15% (2016-20) national samples of Medicare FFS beneficiaries, we identified new oral BB users aged ≥66 y with continuous enrollment for ≥360 days pre- and ≥180 days post-BB initiation. We screened for initiation of 446 other ‘marker’ drug classes (based on WHO Anatomical Therapeutic Classification level 4 codes) within ±90 days of BB initiation, generating sequence ratios (SRs) representing the proportions of BB initiators starting the marker class after, vs. before, the BB. We calculated adjusted SRs (aSRs), accounting for prescribing trends over time, with 95% CIs >1 considered statistically significant. We then calculated naturalistic numbers needed to harm (NNTH), i.e., the inverse of the number of marker class initiators after starting a BB minus the number initiating before BB use, divided by the number of BB initiators. Independent reviewers classified signals as potential prescribing cascades or not based on biologic plausibility.

Results: We identified 357,727 BB initiators (mean ± SD age, 76.6 ± 7.6 year; 59% women; 85.3% hypertension). Of 446 marker classes studied, excluding other antiHTN classes, we identified 123 statistically significant signals (aSR >1). After the first round of clinical review, 24 signals (19.5%) were classified as potential prescribing cascades. The top 3 potential prescribing cascades (ranked by aSR), along with their potential BB AE/marker class indication, were: Iron trivalent, oral preparations (possible AE/indication, anemia; aSR 4.75 [95% CI 1.04-21.68]); glycogenolytic hormones (hypoglycemia; aSR 2.59 [2.11-3.18]); and direct factor Xa inhibitors (thrombosis; aSR 2.47 [2.39-2.55]). The top 3 potential prescribing cascades (ranked by NNTH) were: direct factor Xa inhibitors (NNTH 43 [95% CI, 42-44]), electrolyte solutions (NNTH 81 [77-85]), and selective serotonin reuptake inhibitors (NNTH 265 [233-310]).

Conclusions: High-throughput SSA screening is a novel approach to screen potential prescribing cascades. Using this approach, we identified signals reflecting known and possibly underrecognized BB AEs in this Medicare cohort. Further studies are warranted to determine the impact of these cascades on patient outcomes.