This draft report is available in PDF only (Draft Report [2.2 MB]; Draft Appendixes [10.9 MB]). People using assistive technology may not be able to fully access information in these files. For additional assistance, please contact us.
Purpose of Review
To summarize evidence on: (1) the diagnostic accuracy of cognitive tests for Alzheimer's-type dementia (CATD), and biomarkers for Alzheimer's disease (AD); and (2) the benefits and harms of prescription drugs and supplements for cognition, function, and behavioral and psychological symptoms of dementia (BPSD) in patients with CATD.
- Many brief cognitive tests had high sensitivity and specificity (≥0.8) for distinguishing CATD from normal cognition, but were less accurate distinguishing between groups with smaller differences in cognitive impairment.
- For brain imaging, amyloid PET and FDG-PET may increase diagnostic accuracy for autopsy-confirmed AD brain pathology when added to clinical evaluation.
- Individual CSF biomarkers and ratios were moderately sensitive and specific for autopsy-confirmed AD, but combinations of CSF biomarkers may have the highest combination of sensitivity and specificity, and may increase diagnostic accuracy when added to clinical evaluation.
- Cholinesterase-inhibitors very modestly outperformed placebo for cognition and function, but increased withdrawals due to adverse events, regardless of CATD severity.
- In individuals with mild-to-moderate CATD, memantine did not appear better than placebo for function or global change, and evidence was insufficient about cognition, whereas in those with moderate-to-severe CATD it also did not improve function, but slightly improved global change and inconsistently improved cognition.
- Evidence was mostly insufficient to draw conclusions about the efficacy of prescription drugs and supplements on BPSD in CATD.
Objective. To summarize evidence on: (1) diagnostic accuracy of brief cognitive tests for clinical Alzheimer’s-type dementia (CATD), and biomarkers for Alzheimer’s disease (AD); and (2) benefits and harms of prescription drugs and supplements for cognition, function, and behavioral and psychological symptoms of dementia (BPSD) in patients with CATD.
Data sources. Electronic bibliographic databases to August 2018; ClinicalTrials.gov; systematic review bibliographies.
Review methods. Cognitive test studies used explicit CATD diagnostic criteria and a non-CATD control group. Biomarker studies used neuropathologic criteria to define AD cases and non-AD controls. All treatment trials enrolled participants with CATD; those evaluating BPSD enrolled individuals with CATD and BPSD. Minimum study duration was 2 weeks for agitation, aggression, psychosis, and disinhibited sexual behavior, and 24 weeks for other outcomes. Two reviewers rated risk of bias (ROB) and strength of evidence. One reviewer extracted data; a second checked accuracy. We analyzed English-language studies with low or medium ROB.
Results. We included 66 diagnostic accuracy studies (42 cognitive testing, 15 brain imaging, nine cerebrospinal fluid [CSF] testing) and 60 trials of CATD treatment (49 reporting cognition or function, 11 reporting BPSD). Cognitive screens, brief batteries, memory, and verbal fluency tests were highly sensitive and specific in distinguishing CATD from normal cognition, but less so for distinguishing mild CATD from normal cognition, or CATD from mild cognitive impairment. For distinguishing autopsy-confirmed AD from non-AD dementia, amyloid positron emission tomography (PET) and fluorodeoxyglucose (FDG)-PET may improve accuracy beyond clinical evaluation. Combinations of CSF tests may have a higher combination of sensitivity and specificity than individual CSF tests, and may improve accuracy for distinguishing AD from frontotemporal lobar degeneration beyond clinical evaluation. Regardless of CATD severity, cholinesterase-inhibitors produced small improvements in cognition and function compared with placebo, but increased withdrawals due to adverse events. For mild-to-moderate CATD, memantine and placebo did not differ for function and evidence was insufficient for cognition. However, for moderate-to-severe CATD, memantine inconsistently improved cognition, but not function. Evidence was mostly insufficient about the effects of prescription drugs and supplements on agitation, aggression, psychosis, or disinhibited sexual behavior.
Conclusions. Brief cognitive tests accurately distinguished CATD from normal cognition, but less accurately distinguished smaller clinical differences. Whether biomarkers improve diagnostic accuracy when added to clinical evaluation needs further verification, but potential benefits of testing are limited by lack of effective treatments for AD and non-AD dementias. Cholinesterase-inhibitors slightly outperformed placebo for cognition and function, but evidence of whether drug treatments improved BPSD was largely insufficient.