A 43-year-old woman presents to her primary care physician with complaints of mild shortness of breath and right-sided chest pain for three days. She reports that lately she has had a nagging nonproductive cough and low-grade fevers. On examination, her vital signs are: temperature 99.1 deg F (37.3 deg C), blood pressure is 115/70 mmHg, pulse is 91/min, respirations are 17/min, and oxygen saturation 97% on room air. She is well-appearing, with normal work of breathing, and no leg swelling. She is otherwise healthy, with no prior medical or surgical history, currently taking no medications. The attending has a low suspicion for the most concerning diagnosis and would like to exclude it with a very sensitive though non-specific test. Which of the following should this physician order?

Obtain spiral CT chest with IV contrast

Order a lower extremity ultrasound

Obtain ventilation-perfusion scan

A 6-month-old male presents for a routine visit to his pediatrician. Two months ago, the patient was seen for tachypnea and wheezing, and diagnosed with severe respiratory syncytial virus (RSV) bronchiolitis. After admission to the hospital and supportive care, the patient recovered and currently is not experiencing any trouble breathing. Regarding the possibility of future reactive airway disease, which of the following statements is most accurate?

“Your child has a greater than 20% chance of developing asthma”

“There is no clear relationship between RSV and the development of asthma.”

“Your child’s risk of asthma is less than the general population.”

“Your child has a less than 5% chance of developing asthma”

“Your child’s risk of asthma is the same as the general population.”

A public health campaign increases vaccination rates against human papillomaviruses 16 and 18. Increased vaccination rates would have which of the following effects on the Papanicolaou test?

Decreased positive predictive value

Decreased negative predictive value

Increased positive likelihood ratio

A pharmaceutical company develops a sequential testing protocol for a rare genetic disorder (prevalence 0.01%). Initial screening test has sensitivity 95% and specificity 90%. Positive results undergo confirmatory testing with sensitivity 99% and specificity 99.5%. The company claims this approach achieves PPV >80% for the final positive result. Evaluate this claim and the rationale for sequential testing in this context.

The claim is true; sequential testing increases PPV by enriching the population tested in the second step

The claim is false; sensitivity decreases with sequential testing, reducing PPV

Sequential testing is unnecessary; the first test alone achieves adequate PPV

The claim is false; sequential testing cannot achieve PPV >80% with such low prevalence

The claim is true; the high specificity of the confirmatory test ensures high PPV regardless of prevalence

A hospital system is implementing a sepsis screening algorithm using clinical criteria with sensitivity of 92% and specificity of 75%. False positives result in unnecessary antibiotics, cultures, and ICU evaluations costing $3,000 per case. Missing true sepsis cases (false negatives) results in average increased mortality and morbidity costs of $50,000 per case. Hospital sepsis prevalence is 8%. Evaluate the optimal threshold adjustment strategy.

Decrease threshold to improve sensitivity despite more false positives

Maintain current threshold as it balances sensitivity and specificity equally

Implement risk stratification with different thresholds for different populations

Abandon screening due to unacceptable false positive rate

Increase threshold to improve specificity and reduce costs from false positives

Prevalence effects on predictive values — USMLE Lesson

Core Concepts - The Diagnostic Toolkit

Sensitivity: Rules OUT disease (SNOUT). High sensitivity tests are good for screening.
- $Sensitivity = TP / (TP + FN)$
Specificity: Rules IN disease (SPIN). High specificity tests are good for confirmation.
- $Specificity = TN / (TN + FP)$
Predictive Values: Depend on disease prevalence.
- PPV: Probability of disease if test is positive. Directly related to prevalence.
  - $PPV = TP / (TP + FP)$
- NPV: Probability of no disease if test is negative. Inversely related to prevalence.
  - $NPV = TN / (TN + FN)$

⭐ As prevalence decreases, PPV decreases and NPV increases. Sensitivity and specificity remain unchanged as they are intrinsic properties of the test.

Prevalence Effects - The Value Proposition

Prevalence, the proportion of a population with a disease, directly impacts a test's predictive values. It does not affect sensitivity or specificity.

High Prevalence: More true positives (TP) and false negatives (FN).
Low Prevalence: More true negatives (TN) and false positives (FP).

	Disease Present	Disease Absent
Test Positive	True Positive (TP)	False Positive (FP)
Test Negative	False Negative (FN)	True Negative (TN)

-   $PPV = TP / (TP + FP)$

Negative Predictive Value (NPV): Probability of not having the disease with a negative test.
- $NPV = TN / (TN + FN)$

As prevalence changes, so do the predictive values:

↑ Prevalence → ↑ PPV & ↓ NPV.
↓ Prevalence → ↓ PPV & ↑ NPV.

Prevalence effects on PPV and NPV

⭐ Sensitivity and Specificity are intrinsic to the test and are not affected by disease prevalence. A test is just as good at detecting disease in a high-prevalence or low-prevalence population.

Predictive Value Curves - See the Shift

Prevalence effects on PPV and NPV with varying cut-offs

Positive Predictive Value (PPV): Directly proportional to prevalence.
- As disease prevalence ↑, PPV ↑.
- A positive test in a high-prevalence population is more likely a true positive.
Negative Predictive Value (NPV): Inversely proportional to prevalence.
- As disease prevalence ↑, NPV ↓.
- A negative test is most reliable when the disease is rare.

⭐ When prevalence is low, a positive result has a high chance of being a false positive, even with a good test. This is why we don't screen for rare diseases in the general population.

Prevalence directly influences a test's predictive values but not its sensitivity or specificity.

As disease prevalence increases, the Positive Predictive Value (PPV) also increases.

Conversely, as disease prevalence decreases, the Negative Predictive Value (NPV) increases.

In high-prevalence settings, a positive result is more likely to be a true positive.

In low-prevalence settings, a negative result is more likely to be a true negative.

Unlock the full lesson and continue reading

Signup to continue reading this lesson and unlimited access questions, flashcards, AI notes, and more

Scan to download app

UNLOCK FREE ACCESS