Objective: We conducted this systematic review to support the U.S. Preventive Services Task Force in updating its 2014 recommendation on screening for cognitive impairment in older adults. Our review addressed the direct evidence on the benefits and harms of screening for cognitive impairment versus no screening, the test accuracy of screening instruments to detect mild cognitive impairment (MCI) and dementia, and the benefits and harms of treatment for MCI and mild to moderate dementia among community-dwelling older adults age 65 years and older.
Data Sources: We performed an updated search of MEDLINE, PubMed Publisher-Supplied, PsycINFO, and the Cochrane Central Register of Controlled Trials for studies published through January 2019. We supplemented searches by examining reference lists from related articles and expert recommendations and searched federal and international trial registries for ongoing trials.
Study Selection: Two researchers reviewed 11,644 titles and abstracts and 966 full-text articles against prespecified inclusion criteria. We included test accuracy studies that included screening instruments that could be delivered in primary care in 10 minutes or less by a clinician or self-administered in 20 minutes or less compared with a reference standard. We included trials of major pharmacologic and nonpharmacologic interventions in persons with MCI or mild to moderate dementia and large, observational studies examining adverse effects of these interventions. We conducted dual, independent critical appraisal of all provisionally included studies and abstracted all important study details and results from all studies rated fair or good quality. Data were abstracted by one reviewer and confirmed by another.
Data Analysis: We synthesized data separately for each key question and within subcategories of screening instruments and treatments. For diagnostic accuracy studies, we focused on sensitivity and specificity of instruments that were evaluated in more than one study. We conducted a qualitative synthesis of results using summary tables and figures to capture key study characteristics, sources of clinical heterogeneity, and overall results of each study. Quantitative synthesis was limited to test performance of the Mini Mental State Examination (MMSE) (due to insufficient number of homogeneous studies for other instruments) and U.S. Food and Drug Administration (FDA)–approved medications to treat Alzheimer’s Disease on global cognitive outcomes, global function, and harms; nonpharmacologic interventions aimed at the patient on global cognitive outcomes; and caregiver and caregiver-patient dyad interventions on caregiver burden and depression outcomes. We ran random-effects meta-analyses using the DerSimonian and Laird method and sensitivity analyses using a Restricted Likelihood Estimation Model with the Knapp-Hartung correction to calculate the pooled differences in mean changes (for continuous data) and pooled risk ratio (for binary data). We used meta-regression to explore potential effect modification by various study, population, and intervention characteristics in cases where 10 or more studies were pooled. We generated funnel plots and conducted tests for small-study effects for all pooled analyses. Using established methods, we assessed the strength of evidence for each question.
Results: Screening (Key Questions 1–3): Only one trial was identified that examined the direct effect of screening for cognitive impairment on important patient outcomes, including potential harms. In that trial, at 12 months, there was no difference in health-related quality of life between those who were screened vs. not screened. Symptoms of depression and anxiety were also similar between groups at 1, 6, and 12 months as was health care utilization and advance care planning. We identified 59 studies that addressed the diagnostic accuracy of 49 screening instruments to detect cognitive impairment. Most instruments were only studied in a handful of well-designed diagnostic accuracy studies in primary care–relevant populations. The MMSE, a brief test taking 7 to 10 minutes to complete, remains the most thoroughly studied instrument. The pooled estimate across 15 studies (n=12,796) resulted in 89 percent sensitivity (95% CI, 0.85 to 0.92) and 89 percent specificity (95% CI, 0.85 to 0.93) to detect dementia at a cutoff of 23 or less or 24 or less. Other screening instruments evaluated in more than one study included the very brief instruments (≤5 minutes) of the CDT, MIS, MSQ, Mini-Cog, Lawton IADL, VF, AD8, and FAQ and the brief instruments (6 to 10 minutes) of the 7MS, AMT, MoCA, SLUMS, and TICS with sensitivity to detect dementia usually at 0.75 or higher and specificity at 0.80 or higher for all instruments. For self-administered, longer tests (>10 minutes), only the IQCODE was assessed in more than one study, with sensitivity to detect dementia ranging from 0.80 to 0.88 and specificity ranging from 0.51 to 0.91. Across all instruments, test performance was generally higher in the detection of dementia vs. mild cognitive impairment, although confidence intervals overlapped. No studies directly addressed the adverse psychological effects of screening or adverse effects from false-positive or false-negative testing. Treatment (Key Questions 4 and 5): We identified 224 trials and 3 observational studies representing more than 240,000 patients and/or caregivers that addressed the treatment or management of MCI or mild to moderate dementia. None of the treatment trials were linked with a screening program; in all cases, trial participants were persons with known MCI or dementia. Pharmacologic Interventions: Based on 45 trials (n=22,431) and three observational studies (n=190,076) that evaluated acetylcholinesterase inhibitors (AChEIs) (i.e., donepezil, galantamine, rivastigmine) and memantine, these medications may improve measures of global cognitive function in the short term, but the magnitude of change is small. In meta-analyses, the differences in changes between those on AChEIs or memantine compared with those on placebo ranged from approximately 1 to 2.5 points on the ADAS-Cog-11 and 0.5 to 1 point on the MMSE over 3 months to 3 years of followup. AChEIs and memantine appeared to increase the likelihood of improving or maintaining patients’ global function by 15 percent (for memantine) to 50 percent (for rivastigmine) in the short term (pooled 95% confidence interval range, 0.49 to 2.69). Other outcome measures were inconsistently reported. Total adverse events and discontinuation due to adverse events were more common with AChEIs, but not memantine, compared with placebo. Rates of serious adverse events overall were not higher among those taking medications vs. placebo, but individual studies noted increased rates of serious adverse events. Trials evaluating other medications or dietary supplements (k=29; n=6,489), including discontinuing antihypertensives, 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) reductase inhibitors (atorvastatin and simvastatin), nonsteroidal anti-inflammatory drugs (ibuprofen, naproxen, indomethacin, and celecoxib), gonadal steroids (estrogen [plus or minus progesterone] and testosterone), and dietary supplements and vitamins (multivitamins, B vitamins, vitamin E, and omega-3 fatty acids) showed no benefit on global cognitive or physical function in persons with mild to moderate dementia or MCI. Nonpharmacologic Interventions: We identified 61 trials (n=7,847) that evaluated nonpharmacologic patient-level interventions, including cognitive-focused, exercise, and multicomponent and other interventions. Among all interventions, there was no clear benefit on global or domain-specific measures of cognitive function compared with control conditions at 3 months to 2 years followup. Effect estimates generally favored the intervention groups over control groups, but the magnitude of effect was inconsistent across trials and represented very wide confidence intervals (ranging from no effect to a large effect). Physical function outcomes, including change in activities of daily living and independent activities of daily living, as well as quality of life and mental and neuropsychiatric symptoms, were inconsistently reported. There was, however, a pattern of effect for exercise interventions, with small improvements seen in measures of physical function and symptoms for intervention groups, whereas control groups reported worsening function. Caregiver and caregiver-patient dyad interventions including psychoeducation for the caregiver and care and case management interventions, reported in 88 trials (n=14,880), resulted in a consistent benefit on caregiver burden and depression outcomes. Effect sizes were mostly small, however, and were of unclear clinical significance. Little harm was evident in the few nonpharmacologic intervention trials that reported harms.
Limitations: There is a lack of evidence around how screening for and treating MCI and early-stage dementia affects decision making outcomes. Furthermore, there has been little reproducibility in testing specific screening instruments in primary care populations. The treatment literature is limited by a lack of consistency in the specific outcomes reported and short followup duration. It is difficult to interpret the clinical importance of the small average effects seen among treatment trials, and many measures likely have limited responsiveness for patients with less pronounced cognitive impairment. Consistent and standardized reporting of results according to meaningful thresholds of clinical significance would be helpful in interpreting the small average effects on continuous outcome measures. Other important measures such as quality of life, physical function, and institutionalization, were inconsistently reported.
Conclusions: Several brief screening instruments can adequately detect cognitive impairment, especially in populations with a higher prevalence of underlying dementia. There is no empiric evidence, however, that screening for cognitive impairment or early diagnosis of cognitive impairment improves patient, caregiver, family, or clinician decision making or other important outcomes nor causes harm. In general, there is support that AChEIs and memantine and interventions that support caregivers, including those that help coordinate care for patients and caregivers, can result in small improvements in the short term. Unfortunately, the average effects of these benefits are quite small and likely not of clinical significance. Any benefits are further limited by the commonly experienced side effects of medications and the limited availability of complex caregiver interventions. Cognitive stimulation and training, exercise interventions, and other medications and supplements showed some favorable effects on patients’ cognitive and physical function, but trial evidence lacked consistency and the estimates of benefit were imprecise. There is less evidence related to screening for and treating MCI. The test performance of the few instruments evaluated to detect MCI was lower than the sensitivity and specificity to detect dementia and there is little evidence for any pharmacologic or nonpharmacologic interventions to preserve or improve patient functioning in persons with MCI.