Why hair treatment studies can give different answers

One review ranks a treatment near the top. Another gives a different estimate. Both may draw on published trials, yet measure or combine the evidence differently.

That difference matters when a study result becomes a product claim or treatment recommendation. The useful task is to establish what was compared and how closely it matches the decision in front of you.

Start with the outcome

A count of all hairs differs from a count of thicker terminal hairs. A result measured over one scalp area also needs the correct conversion before comparison with another area.

Follow-up duration matters. A 12-week result and a 24-week result can describe different stages of the response. Research methods should make those distinctions visible.^1,2

Our poster at the World Hair Research Conference, 2024, Dallas, USA, compared three important meta-analyses from highly regarded medical journals. It identified differences in methods, eligible studies and data extraction that affected the results. The chart shows why a ranking deserves examination before it becomes a recommendation.¹

Give the error bars their correct meaning

A standard deviation, or SD, describes the spread of observed results. A confidence interval describes uncertainty around an estimate, such as an average effect. A Bayesian credible interval expresses uncertainty under that analysis model.^3,4

A study can estimate an average precisely while people have quite different outcomes. The interval describes the precision of the estimated average; individual outcomes can vary widely.

A treatment ranking is useful only after its outcome, comparison and uncertainty are understood.

Separate appearance from growth

A serum can improve the appearance of fullness or contain an active treatment. A shampoo may reduce scale or help hair handling. A study of hair counts answers another question.

The product format alone cannot establish the outcome. US regulatory categories also depend on intended use and claims.⁵ The finished preparation, amount used and population studied need to match the proposed claim.

Small studies can be informative. Their design, measurements, consistency and follow-up determine how much they establish. A large percentage improvement still needs its starting number and comparison.

Use the evidence in a personal decision

A systematic review organizes studies. A meta-analysis combines results statistically. The quality of the underlying studies and the populations included shape the result.

Before choosing a treatment, identify the condition studied, the outcome achieved and the effort required. Evidence from the original trials helps resolve questions left by a summary.

Bring the claim or paper when seeking advice. I use it with the examination and treatment history to recommend where the option fits and how its benefit should be assessed.

Open full-size figure (opens in a new tab)