b'CHAPTER 2 | THE AGING BRAINimpairment (MCI) are evolving as we learn morefunctioning. Measures of cognitive functioningTesting Cognitive Functioning about dementia.included in most waves of the HRS since 1992 Estimating the prevalence of dementia andinclude: 10-word immediate and delayed recall toImmediate Word Recallrelated disorders depends not only on how diag- assess memory (see example at right); a serial sev-nostic criteria are defined but also on how theyens subtraction test of working memory; countingIll read a set of 10 words and ask you are translated into questions and assessmentsbackwards to assess attention and processingto recall as many as you can. We have for research purposes. While several measuresspeed; object naming test to assess language; andpurposely made the list long so that it of cognitive functioning are widely used, differ- recall of the date, president and vice president towill be difficult for anyone to recall all ent epidemiologic studies have used differentassess orientation. the wordsmost people recall just a measures and different cut-off scores to indicateInformation from these survey measures isfew. Please listen carefully as I read the disease threshold along the continuum. Scientistsoften summarized as a composite score rangingset of words because I cannot repeat must weigh the tradeoffs involved in setting afrom zero to 35 with a higher score indicatingthem. When I finish, I will ask you to particular threshold for diagnosis. Those set verybetter cognitive functioning. Early researchrecall aloud as many of the words as high ensure that only true cases of dementia aresought to define ranges using this measure thatyou can, in any order.counted yet risk increasing false negatives, thuswould correspond to clinical states. For example, underestimating the prevalence of disease. OnLanga et al. (2008) used a score of 11 or higher to1.Hotel 6.Marketthe other hand, by setting a lower threshold sci- define normal cognitive functioning, eight to 102.River 7.Paperentists risk including false positives and inflatingto indicate MCI and seven or lower as moderate to3.Tree 8.Childdisease estimates. Against this backdrop, HRSsevere impairment. This composite measure has4.Skin 9.Kingresearchers have sought to create public data thatbeen widely used to study trajectories of cognitive5.Gold 10.Bookcould be useful in addressing these and otherfunctioning.important issues. However, without a clinical diagnosis a gold standardto compare to, there was no HRS embraced a broadway to be sure that severe dementia correspond-definition of health to includeed to a typical clinical diagnosis of dementia aspects of mental health andthat a patient could receive during a diagnostic work-up in the health care system. Yet conduct-cognitive functioning. ing a clinical interview with the entire HRS sample was unrealistic and probably undesirable When the HRS began in 1992, measures offor younger participants. The solution to this cognitive functioning were uncommon in largeproblem was to select a subsample of the HRS epidemiologic studies. Notwithstanding the factpopulation aged 71 and older, at higher risk for that the HRS was launched with a focus on healthcognitive impairment, and administer a full and aging in the transition from work to retire- neuropsychological assessment. ADAMS collect-ment, HRS embraced a broad definition of healthed extensive information designed to provide a to include aspects of mental health and cognitivediagnostic determination of dementia, cognitive'