Hospitalization after Oral Antibiotic Initiation in Finnish Community Dwellers with and without Alzheimer's Disease: Retrospective Register-Based Cohort Study.

Background: Persons with Alzheimer's disease (AD) are frequently hospitalized from infection-related causes. There are no previous studies investigating hospitalization associated with antibiotic initiation in persons with AD.

Objective: To investigate the frequency and risk of hospitalization associated with oral antibiotic initiation among community dwellers with and without AD.

Methods: We performed a retrospective register-based study utilizing register-based Medication Use and Alzheimer's disease (MEDALZ) cohort. It includes all community dwellers diagnosed with AD during 2005-2011 in Finland and their matched comparison persons without AD. Antibiotic use was initiated by 34,785 persons with and 36,428 without AD. Drug use data were collected from Prescription Register and comorbidities from Special Reimbursement and Hospital Care Registers. Infection diagnoses were collected from the Hospital Care Register. Factors associated with hospitalization were estimated utilizing logistic regression models.

Results: Risk of hospitalization following antibiotic initiation was higher among antibiotic initiators with AD than without AD (adjusted odds ratio, aOR, 1.37, 95% Cl 1.28-1.46).Strongest association with hospitalization was found for oral glucocorticoid use, aOR 1.41 (1.25-1.59); epilepsy, aOR 1.33 (1.10-1.63); and active cancer, aOR 1.30 (1.14-1.49). Among initiators of cephalexin, pivmecillinam, amoxicillin/amoxicillin, and enzyme inhibitor and doxycycline, persons with AD were more frequently hospitalized than persons without AD. A quarter of hospitalized antibiotic initiators had infection diagnosis in their hospital care records.

Conclusions: Persons with AD initiating an antibiotic had a higher risk for hospitalization than antibiotic initiators without AD. Further research is needed to determine whether infection-related hospitalization could be reduced.