Table 2.
Impact of practice effects
| A. Progression from cognitively normal to MCI | ||||
|---|---|---|---|---|
| # of cases, based on PE-unadjusted cognitive scores | # of cases, based on PE-adjusted cognitive scores | Difference (%) # of cases | χ2; p | |
| MCI diagnosis | 90 | 113 | +23 (+26%) | 20.0; <.001 |
| Amnestic domain impaired | 74 | 88 | +14 (+19%) | 12.1; .005 |
| Attenti on/Executive domain impaired | 15 | 26 | +11 (+73%) | 9.01; .003 |
| Language domain impaired | 11 | 14 | +3 (+27%) | 1.3; .25 |
| B. Concordance of MCI diagnosis and biomarker-positivity | ||||
| Converters to MCI | # of returnees who are biomarker-positive and MCI (PE-unadjusted) | # of returnees who are biomarker-positive and MCI (PE-adjusted) | Difference (%) # of cases | p |
| Aβ+ | 44 | 53 | +9 (+20%) | .007 |
| p-tau+ | 47 | 56 | +9 (+19%) | .031 |
| t-tau+ | 40 | 46 | +6 (+15%) | <.001 |
| Stable CN | # of returnees who are biomarker-positive and CN (PE-unadjusted) | # of returnees who are biomarker-positive and CN (PE-adjusted) | Difference (%) # of cases | p |
| Aβ+ | 157 | 148 | −9 (−6%) | .007 |
| p-tau+ | 175 | 166 | −9 (−5%) | .031 |
| t-tau+ | 122 | 116 | −6 (−5%) | <.001 |
Follow-up diagnoses were made with practice effect-unadjusted (PE-unadjusted) or practice effect-adjusted (PE-adjusted) scores. The difference in the number of cases is calculated by subtracting the number of cases, based on PE-unadjusted scores, from the number of cases based on PE-adjusted scores. The percent difference (%) in number of cases is the differences in number of cases divided by the number of cases based on PE-unadjusted cognitive scores (e.g., 26%=23/90). χ2 is McNemar χ2. Individuals could be impaired in more than one domain. Consequently, the sum of impaired individuals within each domain is greater than the total number of MCI cases. The MCI diagnosis row counts an individual only once, even if they are impaired in more than one domain.