Journal: I&D Week 1

Today I want to note some definitions that I will be requiring as I proceed with this topic.

Screening
"A good screening test should be able to discriminate clearly between those who are at high risk and those who are not.  It should be safe, have a reasonably defined cut off level, and be both valid and reliable." (Tracy, 2010. p721)

"Validity is the ability of a test to measure what it sets out to measure, usually differentiating between those with and without the disease." (Tracy, 2010. p721)

 "... reliability indicates repeatability (Grimes&Schulz 2002, p 882) ... a method of measurement that consistently gives the same results (NICE 2008)." (Tracy, 2010. p721))

Assessing the effectiveness of a test
Sensitivity = detection rate
Specificity = ability to identify those without the condition (Tracy, 2010. p721)

Positive test = true positive (a) +false positive (b)
Negative test = false negative (c) + true negative (d)

Test validity
Sensitivity = a/(a+c)
Specificity = d/(d+b)
Positive predictive value = a/(a+b)
Negative predictive value = d/(c+d)
(Grimes&Schulz 2002)

Hypothetical trade-off between sensitivity and specificity
"For any continuous outcome measure (eg blood glucose level), the sensitivity and specificity of a test will be inversely related. ... Selecting a cut-off point is a compromise, mislabeling some healthy people and some people with diabetes.  Where the cut-off should be depends on the implications of the test." (Tracy, 2010. p722)

That's all I'll have time for today.

Thankyou for your comments