Road Exposure After ICD Implantation: Implications for Crash Risk Assessment

Recipients of implantable cardioverter-defibrillators sharply reduce driving after implantation — a behavioral shift that can mask true crash risk if analyses ignore exposure.

A new population-based analysis documents substantial, transient reductions in road exposure in the weeks after implantation that materially lower crude crash counts. Exposure-adjusted measures are therefore required to avoid underestimating crash risk and to guide counseling and post-implant surveillance.

This population-based cohort linked health and driving records from 2003–2018 to compare drivers who underwent implantable cardioverter-defibrillator (ICD) implantation with matched controls. Motor vehicle crashes were the primary outcome and estimated road exposure the principal exposure metric. The cohort pooled primary- and secondary-prevention recipients (3,454 primary and 3,070 secondary ICD recipients), matched to controls and followed for 1–6 months depending on indication.

The core finding: measured driving fell markedly after implantation, so observed reductions in crude crash counts reflect decreased driving time rather than lower per-hour crash risk.

Quantitatively, the study reports a mean first-month road exposure relative to baseline: the mean RERB for primary-prevention recipients in month one was 0.29. Adjusted crash-rate comparisons versus controls in that interval produced an aIRR of 2.22 (95% CI, 0.72–6.87), not statistically significant. Using the study’s estimated road exposure relative to baseline (RERB) as an offset in quasi-Poisson models, the authors show that the magnitude of the RERB reduction is large enough to bias unadjusted crash comparisons downward; raw post‑implant crash counts can therefore be misleading without exposure adjustment.

Sensitivity analyses change the interpretation: when RERB was fixed at plausible lower-bound values, the exposure-adjusted crash rate in month one for primary-prevention recipients rose to roughly a fivefold elevation versus controls, while secondary-prevention sensitivity tests showed about a twofold increase in the first three months.

Although primary analyses that adjusted for estimated RERB did not find statistically significant differences overall, these sensitivity-model results reveal a hidden risk that crude or non–exposure-adjusted models can obscure. Exposure-adjusted, person‑time approaches materially alter risk estimates and should shape interpretation of registry- or claims-based crash statistics.

Key Takeaways:

• Post‑implant driving reductions can mask per‑hour crash‑risk signals in the weeks after ICD implantation.
• Patients in the first month after ICD implantation (primary‑prevention recipients in particular) and clinicians counseling them about driving.
• Incorporate exposure metrics into counseling and surveillance; require exposure‑adjusted analyses when evaluating post‑implant crash risk.