| Course | HS340 Epidemiology |
|---|---|
| Unit | Unit 8 |
| Paper type | Screening test analysis |
| Length | About 720 words, 5 pages double-spaced |
| Format | APA 7 student paper |
| School | Purdue University Global |
| Program | BS in Health Science |
| Updated | October 2026 |
Free sample paper for HS340 Unit 8
Most Positive Screens Are Wrong at First: A Unit 8 Screening Test Analysis of Pre-Employment Drug Screening With a Confirmation Step
[Student Name]
Purdue University Global
HS340: Epidemiology
Unit 8 Screening Test Analysis
[Instructor Name]
[Date]
The employer, hires and test characteristics are composite values chosen for the exercise; the clinical guide and the methods papers cited are real.
Assumptions
Population tested: [5,000] new hires. Prevalence of recent use of the tested drug class: [2] percent, or [100] people. Screening test: an immunoassay with an assumed sensitivity of [90] percent and specificity of [97] percent. Confirmation test: gas or liquid chromatography with mass spectrometry, assumed sensitivity [98] percent and specificity [99.9] percent, run only on screen-positive samples.
These values are for the exercise. Real figures vary by drug, by assay and by the cutoff a laboratory uses, and immunoassays are known to cross-react with some legal medicines, which is why their results are treated as presumptive (Moeller et al., 2008).
Table 1: The Screen Alone
True positives: 90 percent of [100] users, or 90. False negatives: the remaining 10 users. False positives: 3 percent of [4,900] non-users, or 147. True negatives: 4,753. Total positive screens: 90 plus 147, or 237.
Positive predictive value: 90 divided by 237, or 38.0 percent. Negative predictive value: 4,753 divided by 4,763, or 99.8 percent.
Fewer than two in five positive screens come from people who actually used the drug, even though the test is right about 97 percent of the time for non-users, because there are forty-nine non-users for every user.
Why the Predictive Value Is So Low
A test's sensitivity and specificity belong to the test; the meaning of any single result belongs to the population being tested, and it shifts with prevalence (Trevethan, 2017). Among [4,900] non-users, even a 3 percent error rate produces 147 false alarms, more than the 90 true users the test catches. If the same screen were used in a population where 20 percent were recent users, about 88 percent of positive screens would be true. Nothing about the test would have changed, only the people tested.
Table 2: Adding the Confirmation Step
Only the 237 screen-positive samples go to confirmation.
Among the 90 true users: 98 percent confirm, or about 88. Among the 147 false positives: 0.1 percent confirm, or about 0.15, which rounds to zero in practice. Confirmed positives: about 88, nearly all true.
Positive predictive value after confirmation: about 99.8 percent.
What the Two-Step Approach Costs
Testing in series raises specificity and predictive value, but it lowers overall sensitivity, since a user must test positive on both tests to be counted (Lalkhen & McCluskey, 2008). Net sensitivity is 90 percent times 98 percent, or 88.2 percent. Of the [100] users, about 12 are missed: 10 by the screen and about 2 by confirmation. Net specificity rises to nearly 100 percent.
The trade-off is reasonable here because the cost of a false positive is high. A hire wrongly reported as a drug user could lose a job offer and carry a mark on their record. Missing some users is a smaller harm in this setting, especially since other workplace safety measures exist.
Costs in Dollars
Running confirmation on all [5,000] samples would cost far more than confirming only the 237 screen positives, since confirmation tests are much more expensive than immunoassays. The serial approach confirms fewer than one sample in twenty and still reaches a positive predictive value near 100 percent. That efficiency is why screening and confirmation are standard practice in workplace drug testing.
Points for a Policy
Three points follow for the hospital's human resources office. No hiring decision should rest on an immunoassay result alone. A medical review officer should speak with anyone who confirms positive before a result is reported, since a prescription can explain some results. And if the hospital ever considered dropping confirmation to save money, Table 1 shows that more than three in five reported positives would be wrong.
Summary
At 2 percent prevalence, the screen's positive predictive value is 38 percent and its negative predictive value is 99.8 percent. Adding confirmation raises positive predictive value to nearly 100 percent at the cost of missing about 2 more users. The analysis shows why a sensitive, cheap screen followed by a specific confirmation is the usual design for low-prevalence testing.
References
Lalkhen, A. G., & McCluskey, A. (2008). Clinical tests: Sensitivity and specificity. Continuing Education in Anaesthesia Critical Care & Pain, 8(6), 221-223. https://doi.org/10.1093/bjaceaccp/mkn041
Moeller, K. E., Lee, K. C., & Kissack, J. C. (2008). Urine drug screening: Practical guide for clinicians. Mayo Clinic Proceedings, 83(1), 66-76. https://doi.org/10.4065/83.1.66
Trevethan, R. (2017). Sensitivity, specificity, and predictive values: Foundations, pliabilities, and pitfalls in research and practice. Frontiers in Public Health, 5, 307. https://doi.org/10.3389/fpubh.2017.00307
Reading the HS 340 Unit 8 instructions
Unit 8 usually asks you to apply a screening test's sensitivity and specificity to a population, build a two by two table, calculate predictive values and discuss what happens when a second test is added or prevalence changes. Analyses commonly run three to four pages with two tables. Instructors look for correct cell counts, correct predictive values, an explanation of why predictive value depends on prevalence, and the trade-off between serial and parallel testing. Many sections reward a practical recommendation and a comment on the harms of false results in the setting. State any assumed values clearly. Cite methods sources for the definitions and a clinical source for the test itself.
How this HS340 Unit 8 example is built
The analysis opens with the assumptions: population, prevalence and the characteristics of both tests, with a note that real values vary. Table 1 gives the four cells for the screen alone and both predictive values, and the highlighted sentence explains the low positive predictive value by the ratio of non-users to users. A section shows how the same test would perform at 20 percent prevalence. Table 2 follows the screen positives through confirmation. A section on cost explains the drop in net sensitivity and why it is acceptable when false positives are costly. A dollar comparison explains why only screen positives are confirmed, and three rules for the hiring office and a summary of results close it.
HS340 Unit 8 rubric: what earns full points
Screening analyses tend to earn points for correct table construction, correct predictive values, explanation of prevalence effects, analysis of serial testing and a practical recommendation. Explanation of why predictive value changes with prevalence usually carries heavy weight, since it is the idea the unit is built around. Table construction and calculation earn credit when every cell follows from the stated assumptions. Serial testing points reward stating the gain in specificity and the loss in sensitivity together. The recommendation is rewarded when it reflects the harms of each type of error in the setting. Clear statement of assumptions and correct citations for definitions and the test complete the expectations, and showing working allows partial credit.
HS340 Unit 8 help from the desk
A common mistake is reading sensitivity as the chance that a positive result is correct. Another is computing predictive values from the screen alone and then claiming the program works without mentioning confirmation. Some students forget that serial testing lowers sensitivity. Others present assumed values as if they were universal facts about the test. Ignoring the consequences of a false positive for the person tested weakens any recommendation. If your scenario uses a different test, perhaps a cancer screen or an infection test in pregnancy, send the prevalence and test values and we can help you build both tables and explain what the results mean.
Get an HS340 Unit 8 example written to your instructions
Sensitivity, specificity and prevalence from the HS340 Unit 8 scenario, the questions as set and the rubric are all the desk requires. Each table is built in whole people, every measure computed along its proper margin, and predictive value explained for the person holding a result. The first custom sample is free and comes back in 24-48 hours. The paper above is an original model document written by our desk, not a submitted student paper and not an official Purdue University Global document.
More HS340 papers and BS in Health Science samples
- HS340 Unit 1: Surveillance Discussion Post
- HS340 Unit 2: Rate Calculation Set
- HS340 Unit 3: Study Design Appraisal
- HS340 Unit 4: Two by Two Table
- HS340 Unit 5: Bias and Confounding Memo
- HS340 Unit 6: Outbreak Investigation Writeup
- HS340 Unit 7: Seminar Reflection on an Outbreak
- HS340 Unit 9: Surveillance Data Review
- HS340 Unit 10: Epidemiology Case Report
- HS315 Unit 8: Preparedness Memo
- HS305 Unit 8: Sampling and Measurement Plan
- HS292 Unit 8: Urgent Care Coding Audit Worksheet
- HS450 Unit 8: Change Readiness Assessment
HS340 Unit 8 questions, answered
Where can I find a free HS340 Unit 8 sample paper?
This page holds the complete Unit 8 screening test analysis, following a pre-employment drug screen and its confirmation test through two tables.
Why is a positive screen usually wrong when prevalence is low?
Because there are many more people without the condition, so even a small false positive rate among them produces more false positives than the true positives found among the few with it.
What does serial testing do to sensitivity and specificity?
It raises specificity, since a person must test positive twice, and lowers sensitivity, since anyone missed by either test is missed overall.
Why are immunoassay drug screens called presumptive?
Because they can cross-react with some legal medicines and other substances, so a positive result needs a more specific confirmation test before anyone acts on it.
How much does prevalence change predictive value?
A great deal. In the example, the same screen has a positive predictive value of 38 percent at 2 percent prevalence and about 88 percent at 20 percent prevalence.