Effectiveness and safety of pattern identification-based herbal medicine for Alzheimer's disease: A systematic review and meta-analysis

Objective: Alzheimer's disease (AD), the most common cause of dementia, has only symptomatic treatments in conventional Western medicine (WM). Disease-modifying drugs are still under development. This study evaluated the efficacy and safety of herbal medicine (HM) based on pattern identification (PI) as a whole system practice for treating AD.

Methods: Thirteen databases were searched from inception to August 31, 2021. Twenty-seven randomized controlled trials (RCTs) with 2069 patients were included in the evidence synthesis.

Results: The meta-analysis showed that, compared with WM, HM prescription based on PI, either alone or in combination with WM, could significantly improve the cognitive functions of AD patients (Mini-Mental State Examination [MMSE]-HM vs. WM: mean difference [MD] = 1.96, 95% confidence intervals [CIs]: 0.28–3.64, N = 981, I² = 96%; HM+WM vs. WM: MD = 1.33, 95% CI: 0.57–2.09, N = 695, I² = 68%) and their ability to perform activities of daily living (ADL–HM vs. WM: standardized mean difference [SMD] = 0.71, 95% CI: 0.04–1.38, N = 639, I² = 94%; HM+WM vs. WM: SMD = 0.60, 95% CI: 0.27–0.93, N = 669, I² = 76%). Duration-wise, 12 weeks of HM+WM were superior to 12 weeks of WM and 24 weeks of HM were superior to 24 weeks of WM. None of the included studies found any severe safety concerns. The odds of mild-to-moderate adverse events were marginally lower in HM than in WM (odds ratio = 0.34, 95% CI: 0.11–1.02, N = 689, I² = 55%).

Conclusion: Hence, prescribing PI-based HM is a safe and effective therapeutic option for AD, either as first-line therapy or adjuvant treatment. However, most of the included studies have a high or uncertain risk of bias. Thus, well-designed RCTs with proper blinding and placebo controls are needed.