Hair analysis, touted by proponents as a way of assessing health and detecting nutritional imbalance, is a bald-faced ripoff, warns a new study.

Asked to analyze the mineral composition of identical hair samples, 13 laboratories returned widely varing and misleading results, the study reports in the current Journal of the American Medical Association.

The reported levels of 26 minerals -- from aluminum to zinc -- in the hair samples differed from laboratory to laboratory, and the computerized interpretations accompanying the results were "voluminous, bizarre and potentially frightening."

"Basically, they're getting money from people undeservedly," says Dr. Stephen Barrett, author of several books on health fraud, who conducted the investigation.

His study concluded that commercial use of hair analysis as a means of diagnosing illness and nutritional imbalance is "unscientific, economically wasteful and probably illegal."

One laboratory's information packet warned that hair analysis is "highly misleading and therefore unsuitable as a primary method of detecting a nutritional problem." But others were less cautious.

One claimed that hair analysis could offer "a guide to reversing the aging process." Others suggested that hair analysis could help detect metabolic diseases in early stages and "guide correction of mineral imbalances that supposedly cause degenerative disease and death," Barrett reported.

To conduct the study, Barrett took hair samples from two healthy 17-year-old girls and sent them under assumed names to 13 laboratories. About three weeks later, he repeated the process, so that a total of 52 hair samples were sent out for testing.

"Hair all over the house," laughs Barrett, a psychiatrist in Allentown, Pa. "Hair is very unwieldy. You have to weigh it out, homogenize it, wash it. Preparing 52 hair samples isn't easy. It's a real mess. My wife was furious."

So was Barrett -- after the results came back riddled with inconsistencies. For most of the 26 minerals tested, several laboratories reported concentrations at least 10 times those reported by other laboratories.

"Even if hair analysis were a valuable diagnostic tool," he concluded, "it appears that most if not all of these laboratories are unreliable."

What happens all too often, Barrett says, is that a perfectly healthy person is persuaded -- by a mail-order advertisement, a doctor, a chiropractor or a nutritionist -- to undergo hair analysis as a diagnostic test. The results can be frightening.

"All of a sudden, they are told they have 23 things wrong and need a lot of supplements and tests," Barrett says. "If they believe it, they may spend hundreds of dollars getting unnecessary food supplements and lots of tests. If they don't believe it, they run to their doctor and may spend hundreds of dollars to get reassured."

Hair analysis has many limitations, Barrett says, and is never as scientific as its advertisements claim. Among its shortcomings:

Mineral content of hair is affected by exposure to shampoos, bleaches and hair dyes. No test can reliably tell whether a specific mineral comes from internal or external sources.

Levels of some minerals can vary with the person's age and sex, with the time of year, with geographic location, and with the color, diameter and rate of growth of the person's hair.

Normal ranges for minerals in hair have not been established. In the JAMA study, standards of "normal" varied so much from lab to lab that a given mineral "might have been considered low by some laboratories, normal by others and high by others."

Hair grows slowly -- about one centimeter per month -- which means even hair close to the scalp may not reflect current body conditions for diagnostic purposes.

"The state of health of the body may be entirely unrelated to the physical and chemical condition of the hair," concluded the AMA's Committee on Cutaneous Health and Cosmetics in 1974. "Hair metal levels would rarely help a physician select effective treatment."

Not everyone agrees. "The AMA attacks anything new that comes out," says Ollie Popenoe, president of Yes Inc., which owns Yes Natural Gourmet and Yes Bookshop in Georgetown and provides hair analysis through a laboratory in Chicago. For $45, a customer receives a printout analysis of a hair sample, plus an analysis of his or her diet, based on answers to a questionnaire.

Popenoe acknowledges that "there are some fly-by-night operators in the business," but says hair analysis by a reputable lab may offer some added information to supplement urine and blood tests.

"None of these tests is that exact," he says. Hair analysis is "one of those things which is gaining ground all the time as a diagnostic tool, but it has been misused."

Hair analysis is heavily promoted by some alternative health magazines, health food stores and nutrition consultants. A recent survey by the National Council Against Health Fraud found that hair analysis was "the most widely advertised of the unscientific methods for assessing nutritional status," says council president William Jarvis.

"Their work is extremely sloppy," says Jarvis, chairman of the department of public health science at Loma Linda University in California. "The bottom line is, they're used to make recommendations for buying food supplements sold by the company to the client. That is invalid, and in our opinion, a fraud."

Six of the laboratories surveyed by Barrett included advice on dietary supplements in their test results. All six recommended use of supplements, but in widely varying types and amounts. The items recommended included "bizarre mixtures" of vitamins, minerals, nonessential food substances, enzymes and extracts of animal organs, and the number of daily doses ranged from two to 46.

Hair analysis has some validity as a research tool in forensic medicine and epidemiology, Jarvis says. It can detect poisoning by heavy metals -- such as lead, calcium, arsenic and mercury -- in the body, which could help a coroner pinpoint the cause and time of death.

"If someone had been putting arsenic in your tea," says Jarvis, "it would show up in your hair."

Epidemiologists, who study and compare the health differences among populations, also can use hair analysis to measure variations from group to group. It could be helpful, for example, in comparing lead levels in children who live in the District with those in children from rural Virginia, Jarvis says. But the variability from child to child would be so great as to make individual comparisons meaningless.

Originally, Jarvis says, some proponents of hair analysis claimed they could detect not only minerals but vitamins as well -- even though the only vitamins in hair are in the roots, below the scalp.

The dental profession is concerned about a recent trend in which some laboratories interpret the slightest trace of mercury in hair as a health hazard, attribute it to mercury in dental fillings and even recommend removal of such fillings.

"That's nonsense," Jarvis says. "Mercury is ubiquitous in the natural environment. It's in all air, all rocks, all soil. The mercury in fillings is infinitesimal compared with the amount we get naturally from the food supply."