II.1. In-Person Evaluation

The ADAMS in-person evaluation was a 3- to 4-hour structured assessment conducted in the subject’s residence by a nurse and neuropsychology technician, both specially trained in data collection for dementia evaluation.  This protocol required participation of both the HRS respondent (subject) and an informant who was familiar with the subject’s daily activities and medical history.  A close family member (e.g., spouse or child) usually served as the informant.  After obtaining consent, the neuropsychology technician administered the neuropsychological test battery to the subject and the nurse met separately with the informant to obtain detailed information about the respondent, including cognitive and functional changes, medical and psychiatric history, current medication use, and current behavioral and psychiatric symptoms.  The nurse measured the subject’s blood pressure and heart rate.  The nurse next performed a standardized neurological examination with the subject and collected a buccal tissue sample for DNA testing of the ApoE genotype, while the neuropsychology technician met with the informant to obtain information on the family history of memory impairment and also asked the informant to complete a dementia severity measure for collection of information on the severity of the subject’s symptoms.  The follow-up assessment protocol was essentially the same as the initial assessment protocol.

In addition, a self-administered paper-and-pencil caregiver questionnaire was completed by informants.  This questionnaire included measures about activities in which the subject is involved, the IQCODE, functional limitations of the subject, out-of-pocket medical expenditures, caregiver wellbeing (including depressive symptoms, caregiving strain, and positive aspects of the caregiving experience), and employment and demographic information for the informant.

The figure below displays the procedure for the ADAMS assessment: 

The ADAMS neuropsychology technicians were from Duke University and were trained by a PhD-level clinical neuropsychologist.  The supervising neuropsychologist accompanied new staff on initial visits to certify level of performance, conducted reviews of audiotapes for quality assurance, and also reviewed the raw neuropsychological test data from each visit.  All testing was scored by the original technician and scored again by a second technician before final review by the supervising neuropsychologist.  For more information, please refer to the Langa et al. (2005) ADAMS design and methods paper.

II.2. Diagnostic Evaluation

A Duke University geropsychiatrist , neurologist, neuropsychologist, and cognitive neuroscientist reviewed all information collected during the in-home assessment, and assigned a preliminary research diagnosis regarding cognitive status.  Subsequently, the study geropsychiatrist reviewed available medical records and revised the preliminary research diagnosis when justified by this additional medical information.  Diagnoses fell within the three general categories of normal cognitive function, cognitive impairment, not demented (CIND), and dementia.  Within the CIND and dementia categories, there were several subcategories to denote etiology of cognitive impairment:

Diagnostic criteria were established prior to the start of the ADAMS and were based on published criteria.  For dementia, criteria were based on the DSM-III-R and DSM-IV.  To capture the likely range of etiologies and clinical presentations of CIND, we defined CIND broadly as functional impairment reported by the ADAMS subject or informant that did not meet criteria for dementia, or performance on neuropsychological measures that was below expectation and > 1.5 standard deviations below published norms on any test within a cognitive domain (e.g., memory, orientation, language, executive function, praxis).  Final diagnoses were assigned by a consensus expert panel made up of neuropsychologists, neurologists, geropsychiatrists, and internists.  The Duke University researchers and the consensus diagnosis panel were blind to the HRS cognitive stratum from which the individual was drawn.  For more information, please see Langa et al. (2005).