Objectives: To assess whether doxycycline and rifampin have a therapeutic role in patients with Alzheimer's disease (AD).

Design: Randomized, triple‐blind, controlled trial.

Setting: Three tertiary care and two community geriatric clinics in Canada.

Participants: One hundred one patients with probable AD and mild to moderate dementia.

Intervention: Oral daily doses of doxycycline 200 mg and rifampin 300 mg for 3 months.

Measurements: The primary outcome was a change in Standardized Alzheimer's Disease Assessment Scale cognitive subscale (SADAScog) at 6 months. Secondary outcomes were changes in the SADAScog at 12 months and tests of dysfunctional behavior, depression, and functional status.

Results: There was significantly less decline in the SADAScog score at 6 months in the antibiotic group than in the placebo group, (−2.75 points, 95% confidence interval (CI)=−5.28 to −0.22, P=.034). At 12 months, the difference between groups in the SADAScog was −4.31 points (95% CI=−9.17–0.56, P=.079). The antibiotic group showed significantly less dysfunctional behavior at 3 months. There was no significant difference in adverse events between groups (P=.34). There were no differences in Chlamydia pneumoniae detection using polymerase chain reaction or antibodies (immunoglobulin (Ig)G or IgA) between groups.

Conclusion: Therapy with doxycycline and rifampin may have a therapeutic role in patients with mild to moderate AD. The mechanism is unlikely to be due to their effect on C. pneumoniae. More research is needed to investigate these agents.