Methods: Prospective data from the National Alzheimer's Coordinating Center Uniform Data Set. Subjects for quantitative analyses had baseline visit and at least one follow-up visit from September 1, 2005 to December 31, 2011 (n=16,245). Measures include prevalence of pacemakers and implantable implantable cardioverter-defibrillator (ICDs) at baseline, point prevalence of incident devices by year and cognitive status, and adjusted models of incident pacemakers, by cognitive status. Co-variates include age, sex, race, functional status, Hachinski ischemic score, cardiac co-morbidities and risk factors for cardiac disease, and the 2006 rate of new pacemakers per 1000 Medicare beneficiaries within each of the 33 ADCs hospital referral region. A sub-sample of 7 patients and 15 family members completed surveys on decisional involvement, conflict, and regret and participated in semi-structured interviews.
Results: Prevalence of pacemakers was 2.1% for those with normal cognition (NC), 2.9% with mild cognitive impairment (MCI), and 3.1% for dementia. Prevalence of ICDs at baseline was <1% for all groups. Over the 7 year study period, unadjusted incidence of pacemakers averaged 4 per 1000 person years (PY) for NC subjects, 4.7 per 1000 PY for MCI subjects, and 6.5 per 1000 PY for dementia subjects. In adjusted models, subjects with dementia were 1.85 times more likely (p<0.001) to get a pacemaker than NC subjects. Preliminary results from surveys and interviews demonstrate that more impaired people were more likely to have others involved in the decision about the device; many patients and caregivers did not feel informed at the time of the decision, yet few reported regret about the decision.
Conclusion: Older adults with dementia are significantly more likely to get a pacemaker than older adults without dementia. Hypotheses to explain these findings include: (1) symptoms of bradycardia may mirror dementia symptoms; (2) dementia patients’ less accurate reporting of history and symptoms may result in conservative device implantation; (3) when the decision is framed as a gamble with certain loss (“without this, s/he could die?”) caregivers are risk averse; (4) confounding by unmeasured heart disease severity/indications for implantation.