Antiepileptic Drugs and Accumulation of Hospital Days Among Persons With Alzheimer's Disease.

OBJECTIVES

To compare the accumulation of hospital days between initiators and noninitiators of antiepileptic drugs (AEDs) among persons with Alzheimer's disease (AD).

DESIGN

Exposure-matched cohort study.

SETTING AND PARTICIPANTS

Persons newly diagnosed with AD in 2005-2011 (n = 70,718) and initiating AED use identified from Finnish health care registers. For each AED initiator, 1 noninitiator matched on age, sex, and time since AD diagnosis was selected. Persons with epilepsy were excluded from the study.

METHODS

Association between AED initiation or use of individual AEDs and accumulation of hospital days during a 2-year follow-up was assessed using negative binomial model.

RESULTS

AED initiators (n = 4432) were hospitalized on average for 43.7 (SD: 88.3) days and matched noninitiators for 32.2 (SD: 71.3) days during the 2-year follow-up. Altogether, 27.3% of the AED initiators and 35.6% of the noninitiators had no hospital days during the study period. Number of accumulated hospital days during the follow-up was 31% higher [adjusted incidence rate ratio (aIRR): 1.31, 95% confidence interval (CI): 1.19-1.43] among AED initiators than the noninitiators. Hospital days due to diseases of the nervous system excluding dementia (aIRR: 2.72, 95% CI: 1.72-4.31), musculoskeletal system (aIRR: 2.49, 95% CI: 1.73-3.58), respiratory system (aIRR: 1.89, 95% CI: 1.47-2.43), and mental and behavioral disorders excluding dementia (aIRR: 1.96, 95% CI: 1.02-3.79) were more common among the AED initiators than noninitiators. Among pregabalin (aIRR: 0.65, 95% CI: 0.56-0.77), gabapentin (aIRR: 0.66, 95% CI: 0.49-0.88), and clonazepam (aIRR: 0.73, 95% CI: 0.55-0.96) initiators, the number of accumulated hospital days was 27% to 35% lower than the days accumulated among the initiators of valproic acid.

CONCLUSIONS AND IMPLICATIONS

AED initiators had more hospital days than noninitiators. Pregabalin and gabapentin were associated with a lower number of hospital days than valproic acid. Further research is needed on the reasons for these findings.