- Diagnoses are made by signs and symptoms, physical exams and often times LABORATORY WORK
- lab work includes: blood, urine, stool, saliva
- Not all tests (physical or lab) are perfect: this is where sensitivity and specificity come in play
- Sensitivity (SnOut) - helps us rule out disease
- Specificity (SpIn) - helps us rule in disease
Testing for Rheumatoid Arthritis (or any autoimmune disease) can be a difficult, confusing process. We have certain tests that measure autoantibodies. Autoantibodies are the part of our immune system, which could attach our own tissue and cause debilitating autoimmune disease, such as Rheumatoid Arthritis.
The presence of autoantibodies in a patient DOES NOTassure the diagnosis of an autoimmune disease. In fact, most healthy individuals have some autoantibodies . Additionally, the testing methods themselves are not perfect.
Most people going to doctors (MDs and NDs) are familiar with blood work and lab testing. We do testing for a number of reasons, but one concept many people don’t understand is how we assess their accuracy.
Not every test is 100% accurate. Therefore diagnosis must be made by comparing medical history with signs and symptoms as well as lab or blood work. Most importantly, lab work needs to be properly interpreted and used according to sensitivity and specificity. More on lab sensitivity and specificity later on.
Why We Test
We run all sorts of lab work, we test blood, urine, stool, saliva, hair and do physical exams. There are a few reasons we do these tests:
- Screening: we performs some tests to screen for disease 
- Example: mammogram for breast cancer
- Example: Bone mineral density scans for osteoporosis
- Example: Fasting lipid profiles to assess coronary artery risk
- Monitoring: we run certain tests to monitor how disease is progressing 
- Example: Hemoglobin A1C to evaluate long term blood sugar control
- Example: INR (international normalized ratio) to monitor warfarin (drug) therapy
- Example: Liver enzymes to monitor drug therapies used in many autoimmune disease
- Confirmatory tests: some tests can be used to rule in or rule out a disease 
- Example: Anti-smith and double-stranded DNA antibodies to confirm SLE
- Example: Urine test to confirm UTI
Tests are Not Perfect
This is where a lot of confusion lies. We know tests aren’t perfect, but sometimes this concept isn’t explained fully to patients.
A Few Terms To Know
Terms describing results for people WITH a disease
- True Positive (TP)
- Number of people with a specific disease who have a positive test result
- You go for an x-ray and actually have a fracture. The x-ray shows a broken bone.
- False Negative (FN)
- Number of people with a specific disease who have a NEGATIVE test result
- You go for an x-ray and actually have a fracture. The x-ray shows you do not have a broken bone.
Terms describing results for people WITHOUT a disease
- True Negative (TN)
- Number of people without disease who have a negative test result
- You go for an x-ray because your arm hurts, but isn’t actually broken. The x-ray shows no fracture.
- False Positive (FP)
- Number of people without disease, who have a positive test result.
- You go for an x-ray because your arm hurt, but isn’t actually broken. The x-ray shows a fracture (or someone interprets it that way).
We have two stats which help us to calculate the likelihood of getting the correct diagnosis. In other words, it helps us determine how accurate these tests really are.
If a patient is communicated a diagnosis, it’s important everyone understands the limitations and nuances of testing. I often see patients in clinical practice who were told one thing based on their lab testing, only to find out a misdiagnosis was made because lab testing sensitivity and specificity weren’t considered.
Sensitivity and Specificity
Sensitivity and specific help determine accuracy of a diagnostic test . They are presented as a percentage value and determined by research .
If a test has 100% sensitivity:
If 100 people are tested and all have a positive test result, it means all 100 do in fact have the disease. In other words, no false positives.
If a test has 80% sensitivity:
If 100 people are tested, and all have a positive test result, it means 80 people do in fact have the disease but 20 do not actually have the disease. In other words, 20 people were diagnosed inappropriately (20 false positives).
Always try to remember the mnemonic SpIn..this too will make sense as we go!
Test specificity is represented as a percentage.
It is obtained by performing the test on people without a specific disease for which the test is intended , .
Test specificity represents the likelihood that a person without a disease will have a negative test result , .
SnOut and SpIN
Sensitivity helps us RULE OUT a disease (SnOut) 
If we have a test with 100% sensitivity, and the test comes back negative, we can rule the diagnosis out with full confidence.
Specificity helps us RULE IN a disease (SpIn) 
If we have a test with 100% specificity, and the test comes back positive we can rule the diagnosis in with full confidence.
Examples to Explain Sensitivity and Specificity
Let’s look at a commonly used test for diagnosing a condition called Systemic Lupus Erythematosus (SLE), an autoimmune disease.
The blood test used is called the Serum Anti-Nuclear Antibody (ANA) test.
The sensitivity of serum ANA for SLE = 100% (SnOut)
The specificity of serum ANA for SLE = 80%. (SpIn)
The prevalence of SLE = 1%.
The population we’ll analyze is 1000 people.
Why Does This Matter?
Testing can be confusing for patients and doctors alike. I often find patients will say “I’ve been tested and don’t have _____________”. I will often ask what specific tests were done. It’s not that I don’t trust the patient, I sometimes find doctors don’t do all the appropriate testing when these complicated cases present. A test may be done, but perhaps not ALL the testing that should have been done was requisitioned.
Sensitivity and specificity are proof that not every test is perfect. Some are better at ruling disease in, others are better at ruling things out. Either way, we need to appreciate the shortcomings of certain tests and make sure we do the appreciate tests when investigating symptoms and coming up with a diagnosis.