NUR 603 Module 6 Screening Analysis Example

Seventy-Four Reactive Results, Twenty-Nine Infections: A Screening Analysis of Hepatitis C Antibody Testing in a Low-Prevalence Primary Care Clinic

[Student Name]

Southern New Hampshire University

NUR 603: Epidemiology

Module Six Screening Analysis

[Instructor Name]

[Date]

The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.

Seventy-Four Reactive Results, Twenty-Nine Infections: A Screening Analysis of Hepatitis C Antibody Testing in a Low-Prevalence Primary Care Clinic

Screening tests are judged in the laboratory by sensitivity and specificity, but patients experience them through predictive value: the chance that a positive result means disease and that a negative result means none. Predictive value depends on how common the condition is in the people tested, which is why a highly accurate test can still produce many misleading results when applied to a whole population. This paper analyzes universal hepatitis C antibody screening in a composite primary care clinic. It argues that the antibody test performs well as a screen, that a large share of reactive results nonetheless do not mean current infection and that the clinic's protocol should be built around that fact.

Background and Recommendations

An estimated 1.7% of US adults, about 4.1 million people, had hepatitis C antibodies in 2013 through 2016, indicating past or current infection, and 1.0%, about 2.4 million, had detectable virus, indicating current infection (Hofmeister et al., 2019). Because many people with chronic infection have no symptoms and curative treatment is now available, the US Preventive Services Task Force gives a B grade to one-time testing of every adult from 18 through 79 (US Preventive Services Task Force, 2020), and the CDC recommends screening every adult at least once where antibody prevalence is at least 0.1%, followed by RNA testing of anyone with a reactive antibody result, preferably reflexively from the same sample (Schillie et al., 2020).

Assumptions

The clinic serves a mostly insured suburban population with few people who inject drugs, so antibody prevalence among its adults is assumed to be 1.0%, below the national figure. For the antibody test, a meta-analysis found pooled sensitivity of 98% and specificity near 100% for rapid antibody tests compared with laboratory immunoassay (Tang et al., 2017). This analysis uses a sensitivity of 98% and, to be conservative, a specificity of 99.5%, since even a small shortfall in specificity matters when prevalence is low. Among adults with antibodies, the proportion with current infection is taken from the national ratio of RNA to antibody prevalence, 1.0 divided by 1.7, or about 59% (Hofmeister et al., 2019). The clinic plans to screen 5,000 adults in its first year.

The Two-by-Two Table

With 1.0% prevalence, 50 of the 5,000 adults have antibodies and 4,950 do not. A sensitivity of 98% detects 49 of the 50, leaving 1 false negative. A specificity of 99.5% correctly identifies 4,925 of the 4,950 without antibodies and produces about 25 false positives. Table 1 sets out the results.

Table 1. Hepatitis C Antibody Screening of 5,000 Adults at 1.0% Prevalence

Note. Composite clinic. Sensitivity 98%, specificity 99.5%. False positives rounded to the nearest whole person.

Predictive Values

The positive predictive value is the proportion of reactive results that are true positives: 49 divided by 74, or 66%. One in three people told that their antibody test is reactive does not have antibodies at all. The negative predictive value is 4,925 divided by 4,926, or more than 99.9%, so a nonreactive result is highly reassuring and needs no further testing unless there has been a recent exposure, since antibodies take weeks to appear after infection.

Prevalence drives the difference. If the same test were used at a syringe services program where, say, 30% of those tested had antibodies, then among 1,000 people, 294 of 300 with antibodies would test reactive and about 4 of 700 without would test falsely reactive. The positive predictive value would be 294 divided by 298, or almost 99%. The test has not changed; only the population has. The same logic explains why a clinic should not respond to a low positive predictive value by abandoning screening. The false positives are cheap to resolve with a confirmatory test, while the true positives are people who can be cured. What changes with prevalence is not whether to screen but how much weight a single reactive result can bear before confirmation, and how the result should be explained to the patient.

From Antibodies to Infection

Even a true positive antibody result does not mean current infection. Antibodies usually persist for life, including in people whose immune system cleared the virus without help and in people cured by antiviral drugs. Applying the national ratio, about 59% of the 49 true positives, or roughly 29 people, would have detectable virus, and about 20 would have resolved infection. Of the 74 people with reactive results, therefore, only about 29, or 39%, currently have hepatitis C. The remaining 45 are either false positives or people with past infection, and none of them needs treatment.

This is why the CDC recommends RNA testing after every reactive antibody result, ideally run automatically on the same blood sample (Schillie et al., 2020). Without it, patients may be told they have hepatitis C when they do not, may be referred unnecessarily or may be lost to follow-up while waiting for a second blood draw. Reflex testing collapses two visits into one and gives the clinician a definitive answer.

Recommendations for the Clinic

The clinic should order antibody testing with automatic reflex to RNA for every adult aged 18 to 79 who has not been screened, as recommended. Staff should report a reactive antibody result to the patient only together with the RNA result, using wording such as the screening test found signs of past exposure, and the confirmatory test shows the virus is or is not present now. Patients with detectable RNA should be linked to treatment, and those with a nonreactive result should be told that no further testing is needed unless they have new risk. Tracking the proportion of reactive screens that lead to an RNA result, and of RNA-positive patients who start treatment, will show whether the program works.

Conclusion

In a clinic where 1% of adults carry hepatitis C antibodies, an accurate antibody test yields a positive predictive value of about 66% and, after accounting for resolved infection, only about 39% of reactive results reflect current disease. The negative predictive value is excellent. Screening is worthwhile, but its value depends on reflex RNA testing and on telling patients what a reactive result does and does not mean.

References

Hofmeister, M. G., Rosenthal, E. M., Barker, L. K., Rosenberg, E. S., Barranco, M. A., Hall, E. W., Edlin, B. R., Mermin, J., Ward, J. W., & Ryerson, A. B. (2019). Estimating prevalence of hepatitis C virus infection in the United States, 2013-2016. Hepatology, 69(3), 1020-1031. https://doi.org/10.1002/hep.30297

Schillie, S., Wester, C., Osborne, M., Wesolowski, L., & Ryerson, A. B. (2020). CDC recommendations for hepatitis C screening among adults: United States, 2020. MMWR Recommendations and Reports, 69(2), 1-17. https://doi.org/10.15585/mmwr.rr6902a1

Tang, W., Chen, W., Amini, A., Boeras, D., Falconer, J., Kelly, H., Peeling, R., Varsaneux, O., Tucker, J. D., & Easterbrook, P. (2017). Diagnostic accuracy of tests to detect hepatitis C antibody: A meta-analysis and review of the literature. BMC Infectious Diseases, 17(Suppl 1), Article 695. https://doi.org/10.1186/s12879-017-2773-2

US Preventive Services Task Force. (2020). Screening for hepatitis C virus infection in adolescents and adults: US Preventive Services Task Force recommendation statement. JAMA, 323(10), 970-975. https://doi.org/10.1001/jama.2020.1123