Clinical Crash Course

Tests done on humans are called clinical trials. People in the tests are called subjects.   Some tests are better than others. The better tests tell you more about what a product can do for you. Below are some tough questions to ask when reviewing clinical literature.

Who was it tested on? Be sure test subjects represent you. Men and women often respond differently to treatment. So do the young and the old. Also, about 5% of hair loss is due to rare and often temporary conditions. Studies done on people in this 5% tell you nothing about androgenetic alopecia, 95% of all hair loss. Don't be surprised, though, if test subjects have lesser degrees of hair loss. More can be done for dying follicles than dead ones. 
Terminology:
Alopecia Areata- hair loss in patches believed to be an immunilogic disorder
Alopecia Totalis- hair loss over the entire scalp
Alopecia Universalis- loss of all hair, all over the body
Anagen Effluvium-  sudden loss of growing hairs- often caused by chemotherapy
Telogen Effluvium- sudden hair loss when large numbers of follicles enter the resting phase- usually temporary and may be caused by severe stress, medications and others
Hamilton Scale (modified by Norwood)- a scale of hair loss pattern and degree in men
Ludwig Scale - similar scale for women
Indications- symptoms or conditions for which a product is intended
Contraindications- symptoms or conditions for which a product is not intended

Where exactly does it work? Some products work only in specific areas of the scalp. If you're thin up front, for example, you may not care about a product that works only in the back. 
Terminology:
Anterior Region- forward or front of scalp
Temporal Region- flat areas on either side of forehead
Superior Region, Vertex or Crown- highest point on scalp
Posterior Region- rearward or back of scalp
Occipital Region- lower, back area of scalp
Lateral- on the sides of the scalp

What was measured? Hair count and controlled photography are the best ways to measure effectiveness. They should be done at regular intervals by the researcher and not by wishful test subjects. Hair count should be a census in a precisely located, measured area. Be wary of less meaningful measurements like hairs in the comb or shower strainer. After all, "40% less hair loss" doesn't mean that a product will make you look like Elvis. Also, look for shifty photographic techniques. Photos should be consistent, with identical camera settings, lighting, angle, distance, hair length and styling.
Terminology:
Terminal Hairs- thick, fully-pigmented hairs- the ones we want.
Velus Hairs- fine, minimally pigmented hairs with limited growth potential
Clinical Significance- the importance of the test findings 
Efficacy- product's effectiveness within the conditions of the test

How long was the test? The longer the better. Hair grows and sheds in natural cycles. These cycles could be mistakenly attributed to a test product in the short term. Also, over time side effects may emerge or a product may lose effectiveness. A test conducted over a year is certainly better that one done in a few months.
Terminology:
Baseline- beginning of the test
End Point- end of test
Anagen Phase- 2 to 6 year hair growth phase- normally about 85% of all hairs
Catagen Phase- 1 to 2 week transitional phase- normally about 5% of hairs
Telogen Phase- 5 to 6 week resting phase- normally about 15% of hairs
Normal or Physiological Shedding- loss of about 100 hairs daily replaced by growing hairs- this is what your hair does if you don't have "hair loss"

How many people were studied? The more the merrier. Five hundred is better than five.  The number of people studied is the sample size. The sample represents the population.  A large enough sample is needed to give results that those of us in the population can count on.         
Terminology:
N- this is the sample size, N=100 means 100 people in the sample
P Value- if you see P<.05, that means there is less than 5 percent chance that test results are due to chance and are not from the product- these are results you can count on Statistical Significance- test's ability to show results that will hold true for the population

Did the test have controls? Controls are essential. Control groups are subjects who get nothing at all or a placebo. A placebo is a fake. Test groups are subjects who get the real product. The control and test groups are both measured. Results from the two groups are compared. Comparison to a control group isolates a product's true effect. When people are assigned to either test or control groups with a known probability, the test is randomized.  Randomization ensures that no favorites get picked for either group. When no one (including the researcher) knows who's getting a placebo or the real product, it's a double-blind test.  Blinding helps eliminate cheating when reporting results.
Terminology:
Phase I study- initial trial to establish indications, dosage and toxicity
Phase II study- small scale trials to see if phase III is warranted
Phase III study- full blown, randomized, clinical trials

Has the study been published?  Publication in a medical journal gives you some assurance of study quality. Scientific journals have standards. The most prestigious journals require every detail from every aspect of the study. They open up a study for peer review. Other scientists and researchers can pore over the details of a study and rip it to shreds if something doesn't look right. This is important, because it helps keep research accurate and honest. Most research- good and bad -doesn't get published, but published is better.

Industrial Strength Reading:
Studying a Study and Testing a Test  (3rd Edition)
How to Read the Health Science Literature
Richard K. Riegelman, MD, PhD and Robert P. Hirsch, PhD
Little, Brown and Company

An Introduction to Clinical Research
Catherine DeAngelis, MD
Oxford University Press