Get Newsletter
AlzRisk Risk Factor Discussion
Back

Risk Factor:
Risk Factor Type: Behavior
Current Understanding:
The tables below summarize results from a series of observational studies of cognitive activity in relation to Alzheimer's disease dementia (AD) risk. Overall, these data are consistent with an association between participation in cognitive activities and lower risk of both clinical AD and all-cause dementia. Confounding and reverse causation could have biased some of these results, meaning that they should be interpreted with caution. Nonetheless, our bias sensitivity analysis, shown in an expanded paper, suggests that confounding per se is unlikely to explain the entirety of the findings, although reverse causation remains a concern (Sajeev G, et al. Epidemiology. 2016). In addition, the lack of an agreed-upon definition of cognitive activity as a construct and of its relevant dimensions, along with differences in its operationalization across studies also complicate comparison and synthesis of results. Thus far, the only large randomized trials of cognitive training interventions have shown modest, typically transient benefits on targeted cognitive domains, and it is unclear to what extent these relate to long-term pursuit of personally selected leisure activities. Further, whereas recommendations to remain cognitively engaged (“use it or lose it!”) are unlikely to be harmful, little is known about dose-response relationships, and how issues such as type, duration, intensity and timing of activity influence AD risk. Overall, engaging in activities enjoyable for their own sake seems a safe approach while further information is gathered. For a review of the putative mechanisms by which cognitive activity may influence AD risk and detailed commentary on interpreting these findings in a broader context, please view the Discussion. An expanded review, which includes studies reporting only on all-cause dementia as well, and bias sensitivity analyses, see our published paper, Sajeev G, Weuve J, Jackson JW, VanderWeele TJ, Bennett DA, Grodstein F, Blacker D. Epidemiology. 2016 Sep;27(5):732-42.
Literature Extraction: Search strategy  * New *
Last Search Completed: 16 June 2016

Risk Factor Overview

Cite as:

Sajeev G, Weuve J, Jackson JW, Gillis JC, Blacker D. "Cognitive activity." The AlzRisk Database. Alzheimer Research Forum. Available at: http://www.alzrisk.org. Accessed [date of access]*.

***

INTRODUCTION

The tables in the Risk Factor Overview summarize results from observational studies of the relationship between cognitive activity and clinical Alzheimer disease (AD) and total dementia. Overall, these data are consistent with an association between participation in cognitive activities and lower risk of AD and dementia. As discussed below, however, these results should be interpreted with caution because reverse causation or confounding could explain much of the inverse association. The lack of an agreed-upon definition of cognitive activity as a construct (or delineation of its relevant dimensions) and differences in its operationalization also complicate comparison and synthesis of results across studies. This wide variation also makes meta-analysis of the data in the tables impossible.

POTENTIAL MECHANISMS OF ACTION

The term, cognitive activity, broadly refers to behaviors whose core feature is information processing. Cognitive activities are thought to contribute to building and maintaining brain structure and function, analogous to the role of physical exercise in the building and maintenance of muscle and motor function. These effects, in turn, are thought to contribute to brain reserve, or more broadly, cognitive reserve, although these terms and others, e.g., neural reserve, are used in a variety of ways. The brain changes associated with cognitive activity are thought to occur over the lifespan. Evidence from animal studies indicates that exposure to an “enriched environment” leads to greater neurogenesis and angiogenesis in specific brain regions, a higher rate of synapse formation, and reorganization of cortical networks, all of which could lead to a form of brain reserve that helps to preserve cognition in old age [1,2]. However, there is evidence that brain plasticity continues even into late life,[3] so in principle late-life activities could also have benefits. In one study, aged rats showed improved cognition even with short-term stimulation, and displayed greater improvements with longer exposure to an enriched environment.4 However, the animal studies need to be interpreted with caution as the comparison group is often a deprived rather than neutral environment.

It has also been suggested that cognitive activity may directly affect amyloid-beta (Abeta) accumulation, viewed as a key initiating event in AD pathogenesis[5]. Cognitively active individuals may have more efficient neural processing, and the reduced degree of brain activation needed to accomplish a task may result in lower Abeta deposition in certain key brain areas [6]. However, mechanistic studies of effects on amyloid pathology in animals have yielded conflicting results. Studies of transgenic mice have found, along with reduced cognitive impairment, no changes [7], decreases [8-10] and even increases[11] in brain amyloid levels among mice exposed to “enriched” instead of standard environments. Finally, a recent study found no direct association between cognitive activities during early- or late-life and any common neuropathology including Abeta, tau, macroscopic infarctions or Lewy bodies, suggesting that its beneficial effects are relatively independent of the accumulation of brain pathology.

METHODOLOGICAL ISSUES

Exposure

Cognitive activity is a complex construct encompassing a wide variety of activities occurring in different contexts, at different intensities, for different durations, and at different times in the life span. In the reviewed studies, popular leisure activities considered to require information seeking and processing [12] have been characterized as “cognitive,” with measures of frequency of participation generally being the focus of assessment. However, other characteristics, such as the type of activity, duration per instance of participation, intensity of cognitive exertion required, the stage of life of participation, and the leisure/occupational context of the activity may also be relevant to AD incidence. While these inter-related dimensions all contribute to the overall amount of cognitive activity, more targeted investigations may additionally provide unique mechanistic insights, help delineate dose-response relationships, and also contribute to developing more specific activity guidelines [13,14] (as has been done in physical activity epidemiology [15,16]).

Type and context of cognitive activity. Most studies assessed a relatively narrow range of primarily leisure activities. Reading books, newspapers, or magazines; playing games like crosswords or cards; and watching television and listening to the radio were the most commonly assessed items. However, other leisure activities as well as many, often common, non-leisure activities (e.g., occupational responsibilities, household duties) were typically not assessed [17]. Failure to capture these activities would therefore result in an overall underestimate of late-life cognitive activity for many individuals. If the differences between individuals’ cognitive activity from such unmeasured sources are related to differences in education, occupation, socioeconomic status, or other variables that are also related to AD, the estimated association between overall late-life cognitive activity and AD will be biased unless these variables are adequately accounted for, as discussed below.

Many studies of cognitive activity focus on activities that are primarily or exclusively cognitive, in order to examine the effect of purely cognitive activities. If social and/or physical cognitive activities, such as card games or bowling, are included, any beneficial effect of social or physical activity could play a role in the observed effect. If such activities are not included, on the other hand, the level of cognitive activity will be underestimated.

Intensities of Cognitive Activity: “Total” cognitive activity encompasses activities that differ greatly in complexity and cognitive domains exercised. Even a relatively narrowly-defined activity, “watching television” for example, may consist of a range of practices with varying cognitive demand. While typically relatively passive, the intensity of this activity may vary depending on the nature of the content and the individual's level of interest and engagement. Similar differences exist for other activities like reading and playing games. Thus far, only a few investigators have attempted to distinguish the degree to which different activities involve cognitive stimulation. In one study, each leisure activity assessed was assigned a score on a four-point scale, corresponding to investigators’ judgment of its relative cognitive, social and physical involvement [18], while another distinguished more broadly between “passive” and “stimulating” cognitive activities19. In another cohort, derived measures of cognitive intensity were found to be highly correlated with frequency measures, and, consequently, were not separately assessed in relation to AD incidence [20,21]. In general, these few studies also showed that participation in more cognitively intensive activities was associated with better cognitive outcomes. However, more detailed explorations of intensity in relation to cognitive outcomes would help build a broader evidence base for more specific guidance on maintaining late-life cognition.

Frequency and Duration of Cognitive Activity: In general, the reviewed studies summarized reports of participation frequency across collections of leisure activities deemed to be cognitively stimulating. However, even indices of participation frequency and methods of summarization across activities varied widely. Some studies used solely the total number of leisure activities, while others averaged participation frequency indices across activities, or used activity-frequency composites such as “activity-days”. This heterogeneity therefore makes meta-analyses of these studies impossible.

Self-reporting of Cognitive Activity: In all the reviewed studies, participation in cognitive activities was self-reported, usually at an assessment late in life. Self-reported assessments were not validated against diary records or direct observation in any studies, and only one study verified self-reported participation against informant reports. It is possible that individuals with incipient AD may be less able to accurately report cognitive activity, which could cause results to be biased in an indeterminate direction. For instance, if participation frequency is underreported among those with incipient AD because of difficulty in recall, lower participation frequency would be spuriously linked with greater AD risk. Alternatively, if participation frequency is overstated by those with incipient AD because of decline-induced embarrassment, or failure to accurately recognize or recall that activity has declined from lifetime levels, this might attenuate any protective association between cognitive activity and AD.

Timing of Cognitive Activity: In almost all the reviewed studies, specifically late-life cognitive activity was the focus. Given the decades-long pathogenesis and insidious onset of AD, assessments of earlier life or cumulative cognitive activity might be more relevant to AD risk. In two small nested case-control studies, prospectively assessed mid-life cognitive activity was linked with lower AD risk [22,23]. With respect to late-life activity, a larger number of studies have found lower AD incidence among more cognitively active individuals. It is less clear whether this association can be attributed specifically to greater late-life cognitive activity; there is good reason to believe that late-life cognitive activity represents in part life-long habits, and late-life decreases in cognitive activity might be associated with the early manifestations of AD (for instance, if such activities became less pleasurable as they became more difficult). Evidence of a favorable impact of late-life cognitive activity independent of earlier life cognitive activity would suggest that even cognitively 'sedentary' individuals might benefit from increasing their cognitive activity late in life, but this has been addressed in only one study, which found that late-life cognitive activity was more strongly associated with lower AD incidence than early-life cognitive activity when both were included in the same model [24]. However, this should be interpreted with caution given the concerns about bias in the estimation of direct effects on dichotomous outcomes (e.g., effect of early-life cognitive activity on AD risk above and beyond its influence via late-life cognitive activity) [25,26], and concerns about reverse causation noted below.

Design and Analysis

Reverse Causation: While the overall type, intensity, and frequency of cognitive activity are likely to be correlated over the lifespan, changes might occur with illness and a variety of other evolving life circumstances associated with age. In particular, individuals with the subtle cognitive impairments that occur in the long preclinical stages of AD might be more likely to eschew participation in cognitively stimulating activities than their healthy counterparts [24,27]. Although this could result in a spurious association between reduced late-life cognitive activity and greater AD risk, one study found that engagement in cognitive activity predicted level of global cognitive function over the following year, but global cognitive function did not predict subsequent degree of cognitive activity. On the other hand, it should be noted that ability in two specific cognitive domains, working memory and perceptual speed, were associated with cognitive activity over the following year [29].

Furthermore, results have been largely unchanged in sensitivity analyses that exclude individuals more likely to have had prodromal AD at baseline (e.g., those with poorer performance on screening tests, or those who developed AD early in the follow-up period). However, these exclusions may not be sufficient given the increasing recognition that detectable cognitive decline occurs as early as 5-8 years prior to AD diagnosis [28,29], and even more subtle impairments, usually noticeable only to affected individuals themselves, occur still earlier.30 The average follow-up time in most studies reviewed here was between 2 and 5 years, and all studies had less than 7 years of follow-up. If even a small fraction of those who became demented over the follow-up period in these studies were already experiencing a great enough decline to affect their cognitive activity but not their likelihood of being screened out as demented at baseline, it is likely that the observed relationship is at least partially due to reverse causation. Longer follow-up studies will be needed to more conclusively eliminate this source of bias. One study found that both early- and late-life cognitive activity were associated with cognitive decline in analyses that adjusted for measures of neuropathology [24]. While this makes it unlikely that change in cognitive activity was a consequence of neuropathology, it cannot exclude the possibility of reverse causality from other unmeasured causes of cognitive decline.

Adjustment for Confounding: As indices of socioeconomic status (SES) like education and occupation may be associated with both cognitive activity and with AD incidence independently of cognitive activity, careful adjustment for confounding by these variables is important. While all studies performed some form of adjustment for education, residual confounding is likely in studies that adjusted for education in broad categories, and may be present as well in studies that used finer groupings or that modeled years of education as a continuous variable, given the strong relationship between education/SES and AD risk [31,32]. Vascular risk factors like diabetes and smoking are correlated with SES-related variables, and also have moderate associations with AD; further adjustment for SES would likely reduce any residual confounding by these factors. A few studies did adjust for factors like occupation and income, and the results remained largely unchanged, suggesting that confounding by components of SES is unlikely to explain the results entirely.

Another potential confounder is longstanding intellectual ability, which is likely correlated with cognitive activity, as well as education and a variety of other factors that may affect AD incidence. In a study done in the Lothian Birth Cohort of 1936 [33], adjustment for intelligence quotient (IQ) at age 11 mostly eliminated the positive associations observed between late-life cognitive activity and cognitive performance [34]. While measures of childhood intellectual ability were not available in most reports, at least two studies adjusted their analyses for baseline cognitive test score by including it as in independent variable in the statistical model. Adjustment for baseline cognition has some appeal because, intuitively, it adjusts for a composite of longstanding intellectual function, the impact of education, and pre-existing cognitive decline. However, recent methodological work has shown that this approach can lead to biased effect estimates due to regression to the mean if cognitive activity is strongly associated with baseline cognitive score [35]. These baseline-adjusted analyses therefore may generate overestimated or entirely spurious protective effect estimates of cognitive activity on AD outcomes. (By contrast, typical mixed linear models of cognitive change account for within-individual correlations between repeated cognitive measurements , including baseline score, but are mathematically distinct from regression models of change that include baseline score as an independent variable.)

Because at least some of the cognitive activities assessed in these studies also involve social engagement (e.g., playing cards/board games) or the ability to travel outside the home (e.g., attending cultural events), there could be confounding by general health status, which is associated with incident dementia [36]. Failure to account for such confounding would likely lead to an overestimated influence of cognitive activity on AD risk.

RESULTS FROM OTHER LINES OF RESEARCH

Related Exposures: The observed associations between markers of higher educational [37,38] and occupational achievement [31] with lower AD/dementia incidence have been cited as evidence that cognitive activity may be involved in the development of AD. These factors, however, might lower AD risk via other, particularly cerebrovascular, mechanisms (e.g., by reducing rates of diabetes). Other cognitive activity-like factors thought to contribute to lower risk of cognitive decline and dementia are occupational complexity [39,40] and social engagement [41]. However, these associations could be due to longstanding intellectual ability, the association of education with lower cardiovascular risk, and other factors. With respect to other attributes of cognitive activity, a prospective cohort study of older women found participation in a larger variety of activities, regardless of their cognitive intensity, to be associated with better cognitive performance [42]. However, this may simply reflect greater capacity to engage in these activities for health or social reasons.

Related Outcomes: Studies in some of the same populations reviewed here have also found late-life cognitive activity to be linked to better performance in particular cognitive domains [43], a slower rate of cognitive decline [24,44], and lower incidence of mild cognitive impairment [24,45]. While these studies are subject to many of the same limitations and challenges as studies of incident dementia, they may be less susceptible to reverse causation if follow-up begins earlier in the disease process.

Cognitive Reserve Markers and AD Biomarkers/Pathology: Recent work has also started to examine the link between markers of cognitive activity and biomarkers of AD pathology, with mixed results. While two small clinical studies linked lifetime cognitive activity with less amyloid deposition [46] and hippocampal atrophy [47], these findings were not replicated in a recent larger, population-based study [48]. Similarly, another study found no association between level of cognitive activity and autopsy-based measures of AD pathology [24].

RCTs of Cognitive Training Interventions: Randomized trials have begun to examine the effects of behavioral and computerized cognitive training interventions on neuropsychological test performance and functional outcomes, both among healthy individuals [49,50] and in those with mild cognitive impairment [51]. Although randomized trials have the potential to overcome many of the previously described limitations of observational studies, most early trials in this area have suffered from important methodological limitations, including absent or suboptimal control groups, relatively small sample sizes, and inadequate follow-up. Most trials have shown small to moderate immediate benefits on the specifically targeted cognitive domains, but consistent and convincing evidence of “transfer” – i.e., more generalized cognitive improvement -- and sustained effects on functional and quality of life outcomes is presently lacking49.

One key trial was the Improvement in Memory with Plasticity-based Adaptive Cognitive Training (IMPACT) study, which examined the effect of a 'plasticity-based' training intervention on cognition. The intervention consisted of a series of computerized exercises designed to improve central sensory system functioning, and, thereby, cognitive performance. In this study of almost 500 adults aged 65 and older, those assigned to the plasticity-based program outperformed an active control group who received a training-time matched intervention (consisting of computer-based educational videos on history, art, and literature) on both targeted and untargeted measures of memory and attention immediately after training [52]. However, these training effects generally waned over a 3-month no-contact period following the intervention [53].

The largest prevention trial was the Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) study, which compared, against a no-contact control group, three distinct group-based behavioral interventions individually targeting memory, reasoning and processing speed in a sample of nearly 3,000 cognitively normal, independently living adults aged 65-94 [54]. The interventions each consisted of ten training sessions of about 60-75 minutes delivered over a six-week period, and also incorporated pre-planned booster doses of four 75-minute sessions at 11 and 35 months following the initial training sessions. All three training interventions appeared to improve performance on their respective target domains at both 2 and 5 years follow-up [55,56]. Self-reported difficulty in carrying out instrumental activities of daily living (IADLs; e.g., shopping, preparing meals, taking medications, managing finances) was comparable across the training and control groups from baseline until year 2, and began to increase in all groups, most markedly among controls, thereafter. At the end of the 5-year period, all groups had worsened from their own baseline level of difficulty, but the training groups all reported less difficulty carrying out IADLs than the control group [56]. A secondary analysis of the ACTIVE study found no difference between trained and untrained individuals on the rate of dementia occurrence over the 5-year follow-up period [57]. Expanding training to simultaneously target multiple cognitive domains and increasing intervention time have been suggested as modifications that may yield larger effects [57].

The stimulation arising from cognitive training interventions may differ from that attained via performance of more general/everyday/conventional cognitive leisure activities. Exposures more in line with a way of life that more fully incorporates cognitive activity, as practiced in observational studies, are difficult to define and operationalize in randomized trials because of the many dimensions and contexts of cognitive activity. Consequently, evidence of lack of efficacy of specific cognitive training interventions in randomized trials does not rule out potential benefits of cognitively active lifestyles. Evidence for any salutary effects of this more general notion of “engagement” (social, cognitive and otherwise) on late-life cognition may be better assessed from trials investigating more holistic interventions. One such example is an ongoing trial among 702 older adults aged 60 and older of a volunteer service program that incorporates cognitive, physical and social activity into the proposed intervention and examines its effects on a series of cognitive and functional outcomes58.

DISCUSSION AND RECOMMENDATIONS

Overall, this body of work suggests that being cognitively active is associated with a lower risk of cognitive decline, AD and all-cause dementia. However, these results could result from lifetime cognitive activity or baseline cognitive function, or could be simply the result of disease-induced reductions in cognitive activity among those with incipient dementia. Moreover, the extent to which these findings are confounded by socioeconomic advantage or conflated with a broader level of engagement remains unclear.

Regardless, largely on the basis of this literature, older adults have been encouraged to stay mentally active (“use it or lose it!”) in order to protect their brain health. While it is still unclear whether this is an effective intervention for preventing or delaying AD, following such advice is unlikely to cause harm and may result in a variety of other social and physical benefits. Yet the notion that being cognitively active may protect against cognitive decline and AD/dementia has also spurred a growing industry of commercial cognitive training products that promise, to varying degrees, improvement in late-life cognition. While “brain training” programs may be potentially useful, training interventions have thus far not been shown to be effective in preventing dementia. The available evidence indicates that the effects of cognitive training interventions are generally modest, mostly do not generalize beyond the specific cognitive domains targeted, and, depending on the endpoint, may need to be sustained over many years. Overall, engaging in cognitive activities that are enjoyable is a safe recommendation.


REFERENCES

1. Valenzuela MJ, Breakspear M, Sachdev P. Complex mental activity and the aging brain: molecular, cellular and cortical network mechanisms. Brain Res Rev. Nov 2007;56(1):198-213.
2. Nithianantharajah J, Hannan AJ. The neurobiology of brain and cognitive reserve: mental and physical activity as modulators of brain disorders. Prog Neurobiol. Dec 2009;89(4):369-382.
3. Greenwood PM, Parasuraman R. Neuronal and cognitive plasticity: a neurocognitive framework for ameliorating cognitive aging. Frontiers in aging neuroscience. 2010;2:150.
4. Kobayashi S, Ohashi Y, Ando S. Effects of enriched environments with different durations and starting times on learning capacity during aging in rats assessed by a refined procedure of the Hebb-Williams maze task. Journal of neuroscience research. Nov 1 2002;70(3):340-346.
5. Reitz C. Alzheimer's disease and the amyloid cascade hypothesis: a critical review. Int J Alzheimers Dis. 2012;2012:369808.
6. Jagust WJ, Mormino EC. Lifespan brain activity, beta-amyloid, and Alzheimer's disease. Trends Cogn Sci. Nov 2011;15(11):520-526.
7. Arendash GW, Garcia MF, Costa DA, Cracchiolo JR, Wefes IM, Potter H. Environmental enrichment improves cognition in aged Alzheimer's transgenic mice despite stable beta-amyloid deposition. NeuroReport. Aug 6 2004;15(11):1751-1754.
8. Lazarov O, Robinson J, Tang YP, et al. Environmental enrichment reduces Abeta levels and amyloid deposition in transgenic mice. Cell. Mar 11 2005;120(5):701-713.
9. Cracchiolo JR, Mori T, Nazian SJ, Tan J, Potter H, Arendash GW. Enhanced cognitive activity--over and above social or physical activity--is required to protect Alzheimer's mice against cognitive impairment, reduce Abeta deposition, and increase synaptic immunoreactivity. Neurobiol Learn Mem. Oct 2007;88(3):277-294.
10. Costa DA, Cracchiolo JR, Bachstetter AD, et al. Enrichment improves cognition in AD mice by amyloid-related and unrelated mechanisms. Neurobiol Aging. Jun 2007;28(6):831-844.
11. Jankowsky JL, Xu G, Fromholt D, Gonzales V, Borchelt DR. Environmental enrichment exacerbates amyloid plaque formation in a transgenic mouse model of Alzheimer disease. J Neuropathol Exp Neurol. Dec 2003;62(12):1220-1227.
12. Wilson RS, Bennett DA, Beckett LA, et al. Cognitive activity in older persons from a geographically defined population. J Gerontol B Psychol Sci Soc Sci. May 1999;54(3):P155-160.
13. La Rue A. Healthy brain aging: role of cognitive reserve, cognitive stimulation, and cognitive exercises. Clin Geriatr Med. Feb 2010;26(1):99-111.
14. Bielak AA. How can we not 'lose it' if we still don't understand how to 'use it'? Unanswered questions about the influence of activity participation on cognitive performance in older age--a mini-review. Gerontology. 2010;56(5):507-519.
15. National Physical Activity Plan Alliance. National physical activity plan. Available at: http://www.physicalactivityplan.org/index.php.
16. healthypeople.gov. Physical activity. Available at: http://www.healthypeople.gov/2020/topicsobjectives2020/objectiveslist.aspx?topicId=33.
17. Schinka JA, McBride A, Vanderploeg RD, Tennyson K, Borenstein AR, Mortimer JA. Florida Cognitive Activities Scale: initial development and validation. Journal of the International Neuropsychological Society : JINS. Jan 2005;11(1):108-116.
18. Karp A, Paillard-Borg S, Wang HX, Silverstein M, Winblad B, Fratiglioni L. Mental, physical and social components in leisure activities equally contribute to decrease dementia risk. Dement Geriatr Cogn Disord. 2006;21(2):65-73.
19. Akbaraly TN, Portet F, Fustinoni S, et al. Leisure activities and the risk of dementia in the elderly: results from the Three-City Study. Neurology. Sep 15 2009;73(11):854-861.
20. Wilson RS, Bennett DA, Bienias JL, et al. Cognitive activity and incident AD in a population-based sample of older persons. Neurology. Dec 24 2002;59(12):1910-1914.
21. Wilson RS, Mendes De Leon CF, Barnes LL, et al. Participation in cognitively stimulating activities and risk of incident Alzheimer disease. JAMA : the journal of the American Medical Association. Feb 13 2002;287(6):742-748.
22. Carlson MC, Helms MJ, Steffens DC, Burke JR, Potter GG, Plassman BL. Midlife activity predicts risk of dementia in older male twin pairs. Alzheimer's & dementia : the journal of the Alzheimer's Association. Sep 2008;4(5):324-331.
23. Crowe M, Andel R, Pedersen NL, Johansson B, Gatz M. Does participation in leisure activities lead to reduced risk of Alzheimer's disease? A prospective study of Swedish twins. The journals of gerontology. Series B, Psychological sciences and social sciences. Sep 2003;58(5):P249-255.
24. Wilson RS, Scherr PA, Schneider JA, Tang Y, Bennett DA. Relation of cognitive activity to risk of developing Alzheimer disease. Neurology. Nov 13 2007;69(20):1911-1920.
25. Hernan MA, Hernandez-Diaz S, Robins JM. A structural approach to selection bias. Epidemiology (Cambridge, Mass.). Sep 2004;15(5):615-625.
26. Kaufman JS, Maclehose RF, Kaufman S. A further critique of the analytic strategy of adjusting for covariates to identify biologic mediation. Epidemiologic perspectives & innovations : EP+I. Oct 8 2004;1(1):4.
27. Hultsch DF, Hertzog C, Small BJ, Dixon RA. Use it or lose it: engaged lifestyle as a buffer of cognitive decline in aging? Psychology and aging. Jun 1999;14(2):245-263.
28. Hall CB, Lipton RB, Sliwinski M, Stewart WF. A change point model for estimating the onset of cognitive decline in preclinical Alzheimer's disease. Stat Med. Jun 15-30 2000;19(11-12):1555-1566.
29. Wilson RS, Segawa E, Boyle PA, Anagnos SE, Hizel LP, Bennett DA. The natural history of cognitive decline in Alzheimer's disease. Psychol Aging. Dec 2012;27(4):1008-1017.
30. Reisberg B, Prichep L, Mosconi L, et al. The pre-mild cognitive impairment, subjective cognitive impairment stage of Alzheimer's disease. Alzheimer's & dementia : the journal of the Alzheimer's Association. Jan 2008;4(1 Suppl 1):S98-S108.
31. Stern Y, Gurland B, Tatemichi TK, Tang MX, Wilder D, Mayeux R. Influence of education and occupation on the incidence of Alzheimer's disease. JAMA : the journal of the American Medical Association. Apr 6 1994;271(13):1004-1010.
32. Meng X, D'Arcy C. Education and dementia in the context of the cognitive reserve hypothesis: a systematic review with meta-analyses and qualitative analyses. PloS one. 2012;7(6):e38268.
33. Deary IJ, Gow AJ, Pattie A, Starr JM. Cohort profile: the Lothian Birth Cohorts of 1921 and 1936. International journal of epidemiology. Dec 2012;41(6):1576-1584.
34. Gow AJ, Corley J, Starr JM, Deary IJ. Reverse causation in activity-cognitive ability associations: the Lothian Birth Cohort 1936. Psychology and aging. Mar 2012;27(1):250-255.
35. Glymour MM, Weuve J, Berkman LF, Kawachi I, Robins JM. When is baseline adjustment useful in analyses of change? An example with education and cognitive change. Am J Epidemiol. Aug 1 2005;162(3):267-278.
36. Montlahuc C, Soumare A, Dufouil C, et al. Self-rated health and risk of incident dementia: a community-based elderly cohort, the 3C study. Neurology. Oct 11 2011;77(15):1457-1464.
37. Sharp ES, Gatz M. Relationship between education and dementia: an updated systematic review. Alzheimer Dis Assoc Disord. Oct 2011;25(4):289-304.
38. Valenzuela MJ, Sachdev P. Brain reserve and dementia: a systematic review. Psychol Med. Apr 2006;36(4):441-454.
39. Andel R, Crowe M, Pedersen NL, et al. Complexity of work and risk of Alzheimer's disease: a population-based study of Swedish twins. J Gerontol B Psychol Sci Soc Sci. Sep 2005;60(5):P251-258.
40. Karp A, Andel R, Parker MG, Wang HX, Winblad B, Fratiglioni L. Mentally stimulating activities at work during midlife and dementia risk after age 75: follow-up study from the Kungsholmen Project. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry. Mar 2009;17(3):227-236.
41. Barnes LL, Mendes de Leon CF, Wilson RS, Bienias JL, Evans DA. Social resources and cognitive decline in a population of older African Americans and whites. Neurology. Dec 28 2004;63(12):2322-2326.
42. Carlson MC, Parisi JM, Xia J, et al. Lifestyle activities and memory: variety may be the spice of life. The women's health and aging study II. Journal of the International Neuropsychological Society : JINS. Mar 2012;18(2):286-294.
43. Wilson R, Barnes L, Bennett D. Assessment of lifetime participation in cognitively stimulating activities. J Clin Exp Neuropsychol. Aug 2003;25(5):634-642.
44. Verghese J, Lipton RB, Katz MJ, et al. Leisure activities and the risk of dementia in the elderly. The New England journal of medicine. Jun 19 2003;348(25):2508-2516.
45. Verghese J, LeValley A, Derby C, et al. Leisure activities and the risk of amnestic mild cognitive impairment in the elderly. Neurology. Mar 28 2006;66(6):821-827.
46. Landau SM, Marks SM, Mormino EC, et al. Association of lifetime cognitive engagement and low beta-amyloid deposition. Arch Neurol. May 2012;69(5):623-629.
47. Valenzuela MJ, Sachdev P, Wen W, Chen X, Brodaty H. Lifespan mental activity predicts diminished rate of hippocampal atrophy. PLoS ONE. 2008;3(7):e2598.
48. Vemuri P, Lesnick TG, Przybelski SA, et al. Effect of lifestyle activities on Alzheimer disease biomarkers and cognition. Ann Neurol. Nov 2012;72(5):730-738.
49. Papp KV, Walsh SJ, Snyder PJ. Immediate and delayed effects of cognitive interventions in healthy elderly: a review of current literature and future directions. Alzheimer's & dementia : the journal of the Alzheimer's Association. Jan 2009;5(1):50-60.
50. Tardif S, Simard M. Cognitive stimulation programs in healthy elderly: a review. Int J Alzheimers Dis. 2011;2011:378934.
51. Gates NJ, Sachdev PS, Fiatarone Singh MA, Valenzuela M. Cognitive and memory training in adults at risk of dementia: a systematic review. BMC Geriatr. 2011;11:55.
52. Smith GE, Housen P, Yaffe K, et al. A cognitive training program based on principles of brain plasticity: results from the Improvement in Memory with Plasticity-based Adaptive Cognitive Training (IMPACT) study. J Am Geriatr Soc. Apr 2009;57(4):594-603.
53. Zelinski EM, Spina LM, Yaffe K, et al. Improvement in memory with plasticity-based adaptive cognitive training: results of the 3-month follow-up. J Am Geriatr Soc. Feb 2011;59(2):258-265.
54. Jobe JB, Smith DM, Ball K, et al. ACTIVE: a cognitive intervention trial to promote independence in older adults. Control Clin Trials. Aug 2001;22(4):453-479.
55. Ball K, Berch DB, Helmers KF, et al. Effects of cognitive training interventions with older adults: a randomized controlled trial. JAMA : the journal of the American Medical Association. Nov 13 2002;288(18):2271-2281.
56. Willis SL, Tennstedt SL, Marsiske M, et al. Long-term effects of cognitive training on everyday functional outcomes in older adults. JAMA : the journal of the American Medical Association. Dec 20 2006;296(23):2805-2814.
57. Unverzagt FW, Guey LT, Jones RN, et al. ACTIVE cognitive training and rates of incident dementia. Journal of the International Neuropsychological Society : JINS. Jul 2012;18(4):669-677.
58. Fried LP, Carlson MC, McGill S, et al. Experience Corps: A dual trial to promote the health of older adults and children's academic success. Contemp Clin Trials. May 13 2013;36(1):1-13.