Sensitivity measures the accuracy of the test for identifying who has the disease
(TP and FN) and is calculated as follows:
Sensitivity TP
TP FN
=
+
Specificity measures the accuracy of the test for identifying who does not have
the disease (FP and TN) and is calculated as follows:
Specificity TN
TN FP
=
+
Positive predictive value measures the proportion of positive screening test
results that correctly identifies those patients who actually have the disease
and is calculated as follows:
PositivePredictiveValue TP
TP FP
=
+
So, what does this mean? If the sensitivity of a test is 80%, then the screen-
ing will correctly identify 80% of the people who have the disease. In other
words, 80% of the people who have the disease will test positive, while 20% of
the people who are screened will be missed, which is known as a false-negative
result. If the specificity of a test is 95%, then the screening will correctly identify
95% of the people who do not have the disease and 5% of the people without
the disease will have a false-positive result. A false-positive result means that
individuals are told that they have the disease when in reality they do not.
These principles are evident in the reported estimates about mammography.
The American College of Preventive Medicine (ACPM) and the U.S. Preventive
Services Task Force reported estimates for the sensitivity and specificity for
mammography. For all women, sensitivity estimates range from 75% to 95%
with specificity ranging from 90% to 97% (AHRQ, 2016; Ferrini, Mannino,
Ramsdell, & Hill, 1996). Positive predictive value for mammography was also
distributed by age. For women younger than 50 years of
age, the positive predictive value was about 20% compared
to a range of 60% to 80% in women 50 to 69 years of age
(Ferrini et al., 1996). With the positive predictive value for
mammography being 20% in women younger than age
50 years, then 20% of the women in this age group who
screened positive for breast cancer during mammography
actually have breast cancer. If the positive predictive value
for mammography was 70% in women 50 to 69 years of age,
FYI
Screening involves testing people without
known disease to determine whether they
have a disease. It is not a diagnostic tool but
can be used to reduce morbidity and mortal-
ity in a population. Early detection of disease
allows for early entry into the healthcare
system with the idea that early treatment
leads to more favorable outcomes.
KEY TERMS
positive predictive
value: The prob-
ability that a person
who screens posi-
tive actually has the
disease
false negative:
When a screening
gives a negative
result despite the
presence of the
disease
false positive: When
a screening gives a
positive result even
though the disease
is not present
214 CHAPTER 8 Epidemiologic Designs: Using Data to Understand Populations