Download
1 / 12
120 likes | 236 Views
The maternal serum alpha-fetoprotein (MSAFP) test is critical for screening fetal abnormalities, using pretest probability, sensitivity, and specificity to revise probability. With a pretest probability of 0.20, sensitivity of 0.34, and specificity of 0.86, we calculate the positive predictive value (PV+) as 0.378 and the negative predictive value (PV-) as 0.839. Various methods such as the 2x2 table, decision trees, Bayesian approaches, and likelihood ratios are explored to refine these probabilities for better clinical decision-making.
E N D
Diagnosis Computation of Post Test Probability
Using Sen. and Spec. to Revise Probability We often use maternal serum α-fetoprotein (MSAFP) test to screen the fetal abnormalities. Pretest Probability = P(D+) = 0.20 Sensitivity = P(T+|D+)=0.34 Specificity = P(T-|D-)=0.86 →PV+=? →PV-=?
Methods to Revise Probability • 2 by 2 Table Methods • Decision Tree Methods • Bayesian Methods • Likelihood Ratio Methods
The 2 by 2 Method (3) PV+=P(|D+| T+)=68/180=0.378 PV-= P(D-|T-)=688/820=0.839
T＋= 0.34 6.8 % D＋= 0.2 T－= 0.66 13.2 % 11.2 % T＋= 0.14 D－= 0.8 68.8 % T－= 0.86 Decision Tree Method (1)
D＋ 6.8 % T＋= 18 % D－ 11.2 % 13.2 % D＋ T－= 82 % 68.8 % D－ Decision Tree Method (2)
