A 33-year-old woman comes to the clinic for a routine health maintenance examination. She has no lower abdominal pain, vaginal bleeding, or vaginal discharge. Her past medical history is remarkable for 1 episode of uncomplicated urinary tract infection treated with a short course of antibiotics. She is currently not taking any medications. She has had 5 sexual partners in her life and has used barrier contraception the majority of the time. Her latest sexual relationship began 3 months ago. Her last cervical cancer screening examination was a Pap smear performed at the age of 23 that showed no abnormalities. She does not use tobacco, alcohol, or illicit drugs. During her visit, she requests screening for cervical cancer and appears extremely concerned as her older sister was recently diagnosed with it. A national government agency is reviewing data from this paper to decide which screening method to recommend and fund. However, agency officials realize that the prevalence of cervical cancer varies throughout the country. For this reason, a reliable epidemiological parameter is needed to compare the significance of negative and positive results obtained in individual patients, irrespective of prevalence. Which of the following epidemiological parameters would be the most useful in comparing these screening tests when applied to individual patients?

Question

A 33-year-old woman comes to the clinic for a routine health maintenance examination.  She has no lower abdominal pain, vaginal bleeding, or vaginal discharge.  Her past medical history is remarkable for 1 episode of uncomplicated urinary tract infection treated with a short course of antibiotics.  She is currently not taking any medications.  She has had 5 sexual partners in her life and has used barrier contraception the majority of the time.  Her latest sexual relationship began 3 months ago.  Her last cervical cancer screening examination was a Pap smear performed at the age of 23 that showed no abnormalities.  She does not use tobacco, alcohol, or illicit drugs.  During her visit, she requests screening for cervical cancer and appears extremely concerned as her older sister was recently diagnosed with it. A national government agency is reviewing data from this paper to decide which screening method to recommend and fund.  However, agency officials realize that the prevalence of cervical cancer varies throughout the country.  For this reason, a reliable epidemiological parameter is needed to compare the significance of negative and positive results obtained in individual patients, irrespective of prevalence.  Which of the following epidemiological parameters would be the most useful in comparing these screening tests when applied to individual patients?

Quizplus · Accepted Answer

Explanation: The agency officials are looking for an epidemiological parameter that does not vary with disease prevalence and provides clinically useful information for individual patients. The likelihood ratio (LR) is defined as the probability of a given test result occurring in a patient with a disorder compared to the probability of the same result occurring in a patient without the disorder. LRs are calculated from sensitivity and specificity: For a positive test result: Positive LR = sensitivity / (1 - specificity). For a negative test result: Negative LR = (1 - sensitivity) / specificity. Just as sensitivity and specificity are independent of disease prevalence, LRs do not change with the prevalence of the disease. In addition, LRs can be used to calculate post-test odds (post-test odds = pre-test odds * LR), providing clinically relevant information for individual patients based on their pre-test odds of disease. Other advantages of LRs are that they can be used with tests that have >2 possible test results and they can be used to combine the results of multiple diagnostic tests. For all these reasons, LRs would be the most appealing choice for the government agency officials in this scenario. General guidelines for LR interpretation are shown in the Exhibit. Exhibit (Choice A) False positive rate = (1 - specificity) and false negative rate = (1 - sensitivity). These do not vary with changing prevalence but are not as informative as LRs for individual patients. (Choice C) An odds ratio is calculated in case-control studies to compare the exposure of participants with the disease (cases) to the exposure of those without the disease (controls). (Choice D) Positive predictive value (PPV) refers to the probability that a positive test result correctly identifies an individual who actually has the disease. Negative predictive value (NPV) is the probability that a negative test result correctly identifies an individual without the disease. Although useful, both PPV and NPV depend on disease prevalence. (Choice E) Relative risk is calculated in cohort studies; participants are followed over time to assess a risk factor for developing a given disease. Relative risk is the ratio of the probability of an outcome occurring in the exposed group compared to the probability of it occurring in the non-exposed group. Educational objective: The likelihood ratio (LR) has direct clinical significance and does not change as the disease prevalence changes. LRs can also be used to grade the clinical significance of various results when >2 different test results are possible. References: Likelihood ratios: a real improvement for clinical decision making? (http://www.ncbi.nlm.nih.gov/pubmed/7957786)

A 33-Year-Old Woman Comes to the Clinic for a Routine