A NOVEL META-ANALYTIC APPROACH WITH APPLICATIONS TO ALZHEIMER’S DISEASE RESEARCH

Abstract

Quantitative literature review has evolved from the time of Glass’ (1976) publication, a methodology he called “meta-analysis,” and Tran (1985) published the first meta-analysis in a medical journal (JAMA). Glass advocated for quantitative reviews to include all the scientific evidence. Since then, meta-analyses have, as a priori decisions, excluded relevant evidence. Many analyzed only a handful of trials (Tran's 1985 publication included 95 trials). Excluding relevant evidence does not advance medical science. Ease of access and volume of data available from keyword searches (e.g., PubMed) require a system to enable investigators to comprehensively review, extract, and aggregate all the evidence from medical literature. “Systematic literature review,” followed by analysis of extracted data paints a myopic view of the evidence. Keyword-based searches miss much of the literature. To address this quandary, MedAware Systems, Inc. has developed a process where data extraction is rapid and nearly 100% accurate. Two scientists, blinded to each other, extract data from the same study. Intelligent software guides the initial extraction comparing each data field for matches. For mismatches, a senior scientist reviews and reconciles the data. The database is built and maintained using a process that continuously identifies and downloads the available body of published and unpublished clinical trials literature. Datasets derived from this database support any topic, providing complete coverage of patient, treatment, and outcome domains. A subset of Alzheimer's disease (AD) literature was reviewed to determine whether the literature supports a differential effect of pharmacologic and non-pharmacologic treatments for AD. 795 clinical studies reporting AD interventions and outcomes were processed. 280 and 84 publications investigated pharmacologic and non-pharmacologic treatments, respectively. Overall, analysis showed no significant reduction in cognitive decline or improvement in QOL/ADL scores with either pharmacologic or non-pharmacologic treatments (Ivashchenko 2018). AD literature is vast and difficult to interpret. This integrated process reviewed and analyzed studies in a timely and efficient manner, providing results that traditional approaches cannot duplicate. MedAware Systems’ literature database and meta-analytic methodology readily profiled and demonstrated the diversity, poor efficacy, and ADL/QOL effects of leading pharmacologic and non-pharmacologic AD treatments.