Soru

Zorluk: ZorDiagnostic Testing Performance

A 58-year-old man presents with a 2-month history of progressive dysphagia to solid foods and unintentional weight loss. The physician suspects esophageal cancer. Based on his age, symptoms, and risk factors (chronic smoking and alcohol use), the estimated pretest probability of esophageal cancer is 30%30\%. A barium swallow study is ordered as an initial evaluation. In this clinical setting, the barium swallow has a sensitivity of 70%70\% and a specificity of 90%90\% for detecting esophageal malignancy. The barium swallow result is reported as normal (negative).

What is the patient's post-test probability of having esophageal cancer despite the negative barium swallow? (Express your answer as a percentage).

Cevap: 12.5 %

Cevap

12.5
The post-test probability of having the disease despite a negative test (also known as the false omission rate) evaluates the proportion of false negatives among all negative test results. Using a hypothetical cohort of 10001000 patients with a 30%30\% pretest probability, there are 300300 patients with the disease and 700700 without. A sensitivity of 70%70\% means that 30%30\% of the diseased patients will falsely test negative (300×0.30=90300 \times 0.30 = 90 false negatives). A specificity of 90%90\% means 90%90\% of the non-diseased patients will truly test negative (700×0.90=630700 \times 0.90 = 630 true negatives). The total number of negative tests is the sum of false negatives and true negatives (90+630=72090 + 630 = 720). The probability of actually having the disease if the test is negative is the ratio of false negatives to total negatives: 90/720=0.12590 / 720 = 0.125, or 12.5%12.5\%.

Alternatively, this can be solved using Likelihood Ratios (LR). The Negative Likelihood Ratio (LR−LR-) is calculated as (1−Sensitivity)/Specificity=(1−0.70)/0.90=0.30/0.90=1/3(1 - \text{Sensitivity}) / \text{Specificity} = (1 - 0.70) / 0.90 = 0.30 / 0.90 = 1/3. The pretest odds are calculated from the pretest probability: 0.30/0.70=3/70.30 / 0.70 = 3/7. The post-test odds equal the pretest odds multiplied by the LR−LR-: (3/7)×(1/3)=1/7(3/7) \times (1/3) = 1/7. Converting the post-test odds back to a probability yields (1/7)/(1+1/7)=(1/7)/(8/7)=1/8(1/7) / (1 + 1/7) = (1/7) / (8/7) = 1/8, which equals 12.5%12.5\%.

Adım Adım Çözüm

1
Calculate diseased and non-diseased populations in a theoretical 1000-person cohort
Diseased = 300, Non-diseased = 700
Establishes the base populations for determining true and false test results based on the 30%30\% pretest probability.
2
Calculate false negative (FN) cases
FN = 90
FN is needed to find how many diseased patients falsely test negative (300×(1−0.70)=90300 \times (1 - 0.70) = 90).
3
Calculate true negative (TN) cases
TN = 630
TN is needed to determine the total number of negative results from the non-diseased population (700×0.90=630700 \times 0.90 = 630).
4
Calculate post-test probability of disease given a negative test
90 / (90 + 630) = 90 / 720 = 12.5%
The post-test probability after a negative test (1 - NPV) equals the number of diseased patients who tested negative divided by all patients who tested negative.

Anahtar Kavram

Calculating the post-test probability of a negative test result (False Omission Rate) by synthesizing pretest probability, sensitivity, and specificity.
Bu soruyu puanla