. Author manuscript; available in PMC: 2021 Jan 1.

Published in final edited form as: J Am Geriatr Soc. 2019 Oct 14;68(1):103–111. doi: 10.1111/jgs.16182

Table 3.

Performance of Restricted Model Using Different Decision Rules

Risk cut-off percentile^*)	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)

≥ 99^th	6.5	99.4	23.7	97.2
≥ 95^th	22.5	96.4	15.7	97.6
≥ 90^th	36.2	91.6	11.5	97.9
≥ 85^th	47.1	87.2	10.0	98.2
≥ 80^th	59.4	82.9	9.5	98.5
≥ 75^th	65.9	78.0	8.3	98.7
≥ 70^th	72.5	72.9	7.5	98.9
≥ 65^th	79.0	67.5	6.8	99.1
≥ 60^th	83.3	62.6	6.3	99.2
≥ 55^th	87.7	56.9	5.8	99.4
≥ 50^th	91.3	51.9	5.4	99.5

NPV, negative predictive value; PPV, positive predictive value

Percentile of the predicted risk distribution observed in the training sample. Performance characteristics are then shown in the test sample when using the given risk cut-off for classifying as unrecognized dementia. Values for the model that considered the full set of predictors are included in Supplementary Table S3.