Skip to main content

Table 1 Summary of a typical validation study result

From: Use of diagnostic likelihood ratio of outcome to evaluate misclassification bias in the planning of database studies

Outcome based on claims data algorithm

(phenotype algorithm)

“True” disease status based on gold standard

(phenotype)

With disease (D+)

Without disease (D−)

Positive (O+)

NA

(true positive)

NB

(false positive)

Negative (O−)

NC

(false negative)

ND

(true negative)

  1. N = total number of subjects in the validation study sample (N = NA + NB + NC + ND)
  2. NA, NB, NC, ND = number of subjects in each cell
  3. Sensitivity = NA/(NA + NC)
  4. Specificity = ND/(NB + ND)
  5. Disease prevalence = (NA + NC)/N
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us