Sequential testing means a person is labeled screen-positive only when both the first test and a second, separately performed test are positive. This design deliberately raises the bar for being called positive, which changes every predictive property in a predictable direction.
The first test alone will catch most true cases, but it also flags some people who do not actually have the condition. When a second test is required, those without disease must be positive on two independent occasions to remain in the screen-positive group. Because false positives are less likely to reproduce across two tests, fewer disease-free people are ultimately referred. This is why
specificity rises: the proportion of truly negative people correctly identified as negative increases. At the same time,
positive predictive value (PPV) also rises, because among everyone who is now called positive, a larger share actually has the disease.
The trade-off occurs on the other side of the table. Some people who truly have the disease will be positive on the first test but negative on the second test. Under a sequential rule, they are classified as screen-negative and are not referred for confirmation. These missed true cases mean
sensitivity may fall, because fewer of all diseased individuals are detected. As more true cases slip into the negative group, the
negative predictive value (NPV) may also fall slightly, since a negative result becomes somewhat less reassuring. This is the opposite of parallel testing, where a positive on either test is enough to refer; parallel testing increases sensitivity and NPV but lowers specificity and PPV.
Key point! Sequential testing sacrifices sensitivity to gain specificity and PPV. It is most useful when the goal is to avoid unnecessary confirmatory procedures, such as invasive diagnostic workups, in people who are unlikely to have the disease.
Watch out! Do not assume that adding a second test automatically improves every measure. The direction of change depends on whether the tests are combined in series (both must be positive) or in parallel (either positive is enough).
The same principle appears in two-stage screening programs. For example, newborn hearing screening often uses a two-stage approach in which only infants who fail the first screen proceed to a second test before referral. This structure reduces the number of children sent for diagnostic audiology, but it also means some infants with hearing loss may be missed at the screening stage if they pass the second test despite having the condition . Similarly, serial cardiac troponin testing in suspected acute myocardial infarction uses repeated measurements over time. Requiring a rise or fall on a second sample improves specificity for myocardial injury, but a single early normal value may still miss some evolving infarctions if the second sample is not appropriately timed or interpreted .
The distinction between series and parallel testing is a frequent licensure examination concept because it directly affects clinical decision-making. In a community screening program such as the RHU scenario, the choice to require two positive capillary tests on different days reflects a preference for reducing false referrals. The cost is that some true cases will be missed at the screening stage, which is why programs using sequential testing must have clear follow-up protocols for people who test negative but remain symptomatic or high-risk.
| Parameter | Sequential testing (both positive required) | Parallel testing (either positive is enough) |
|---|
| Sensitivity | May decrease (some true cases missed by second test) | Increases (more true cases detected) |
| Specificity | Increases (fewer false positives) | May decrease (more false positives) |
| PPV | Increases (positive result more trustworthy) | May decrease (positive result less specific) |
| NPV | May decrease slightly (negative result less reassuring) | Increases (negative result more reassuring) |
The reason PPV and NPV move in opposite directions under sequential testing is tied to disease prevalence and the shifting composition of the positive and negative groups. When specificity improves, the positive group contains fewer false positives, so PPV rises. When sensitivity drops, the negative group contains more false negatives, so NPV falls. These two effects occur together and explain why option 2 is the correct answer:
specificity and PPV rise, while sensitivity and NPV may fall.