Questionable baldness remedies have been peddled since the beginning of medicine. According to Pliny (23-79 A.D.), ashes of seahorse could cure baldness. Almost 2000 years later, the British Medical Association warned the public of the increasing “number of preparations put forward for the cure of baldness,” particularly those which “are not applied locally but taken internally.” The purported active ingredient? “[H]aemoglobin.” (see Secret Remedies (1909), page 114).
While the medicinal use of a seahorse or dried blood matter may sound fanciful to modern ears, one has to wonder whether today’s public is any less credulous: Worldwide, consumers have spent over $400 million per year on a modern baldness remedy known by the trade name Propecia (finasteride). Has science finally triumphed over a medical condition that has persisted through millennia? Today’s consumers might rationally believe that its has, given that Propecia is FDA-approved for the treatment of alopecia (baldness). FDA-approved remedies must, according to federal law (21 U.S.C. § 355(d)), prove their efficacy in well-controlled, clinical investigations.
Yet one need only walk through a crowded street to see that, if a baldness cure has truly arrived, a surprising number of people have not availed themselves of it. Is Propecia, then, not effective? Let us take a look at the official data.
Propecia’s most recent FDA labeling reports three double-blind, randomized, placebo-controlled studies of 12-month duration, involving a total of 1879 men with mild to moderate (but not complete) hair loss. As required by law for approval, the studies did demonstrate a “significant” difference (by which the law means a “statistically significant,” but not necessarily a large or important, difference) between placebo and active groups. “At 12 months there was a 107-hair difference from placebo (p<0.001, PROPECIA [n=679] vs placebo [n=672]) within a 1-inch diameter circle (5.1 cm2),” a 138 hair difference after two years, and a 277 hair difference after five years. All of this may sound much more promising than seahorse ashes or dried hemoglobin, if a bit confusing, and the FDA label goes on to describe many more numbers and outcome measures which interested readers can find in Section 14 of the FDA labeling.
Despite Propecia’s “proven” efficacy, a few caveats are in order as always. While 277 more hairs per square inch may seem like a lot, a cosmetology text notes that the average number of hairs per square inch is 2,200 (Milady’s Standard Cosmetology, p210), though this figure varies by hair color and some other sources provide different numbers. Using 2,200 as a rough number, however, it can be calculated that after 5 years of faithful use of Propecia, those threatened with baldness can expect somewhat more than a 10% improvement over what they would have experienced without any treatment. More importantly, the average gain in hair count over baseline (as opposed to over placebo) was about 50 hairs per square inch (see Figure 1), or just over 2% using the 2,200 figure.
These numbers could be made more accurate and informative had the FDA labeling included a figure for the average number of hairs per square inch in the study participants, but no such figure is presented. (The FDA’s statistical review, at Appendix Table 1A, however, notes for one of the three studies a baseline hair count of 856.1 hairs per square inch, and an average hair count at month 12 of 944.9, for an average increase of 88.8 hairs, or about 10%.)
The men were 88% Caucasian, so the study can tell us little about men of other races. Women do not appear to benefit at all from Propecia: A study about women is presented in the labeling at section 14.2, which briefly concludes: “There was no improvement in hair counts, patient self-assessment, investigator assessment, or ratings of standardized photographs in the women treated with PROPECIA when compared with the placebo group.” The study did not include men over age 41, so the study can tell us little about Propecia’s effect in older men, though one might speculate that long-quiescent follicles might be less amenable to pharmacological stimulation, a plausible reason why older men were excluded from the study. The large majority of participants had “vertex”, i.e. top of the head, hair loss, so the study tells us little about whether Propecia is effective in treating receding hairlines (the studies specifically excluded “the area of bitemporal recession”). Although the number of male subjects, 1879, may seem high and therefore robust, many participants did not continue through the sequential “extension” studies, which brought the total study duration to five years. In one of the study arms, only 26 patients were on placebo by the end of the 5th year; in another, only 15 remained. The 277-hair improvement figure was based on only 219 people remaining in the Propecia group and the scant 15 people just mentioned. Moreover, those more likely to drop out of the study (as with clinical studies generally) are those who are seeing no improvement. This tendency could potentially distort the outcome measures to make them appear greater than they are.
Is Propecia, then, effective? It depends what is meant by “effective.” Based on Figure 1 in Section 14 of the FDA labeling (and recalling the caveats above), Propecia can slow the rate of hair loss and perhaps even increase hair density somewhat (about 100 hairs per square inch over baseline after 12 months, which diminishes to 50 hairs over baseline after year 5). Can it re-grow hair where there is none? The FDA-labeling cannot tell us, since those with “complete” hair loss were not included in the studies. Additional information is available in the FDA’s statistical review.
Regardless of efficacy, there is little reason for anyone ever to buy or consume Propecia (finasteride), or any doctor ever to prescribe it, since a much cheaper and identical chemical sold under the trade name Proscar (finasteride), is available. More on that soon.