Department store magnate John Wanamaker once famously said that he knew that half of the money he spent on advertising was wasted -- he just didn't know which half.

His wry comment helps describe the serious dilemma now faced by those who create advertising for new drugs or organize public service health campaigns, according to researchers at Indiana University. It seems that many drug ads and public service messages may have the unintended consequence of dissuading people from obtaining potentially life-saving drugs or taking preventive steps such as getting vaccinated, according to studies conducted by Anthony and Dena Cox, professors of marketing in the university's Kelley School of Business, and Gregory D. Zimet, a professor of pediatrics and clinical psychology at the Indiana University School of Medicine.

"People who design real campaigns may -- with the best intentions -- actually end up designing programs that are ineffective or even harmful," Anthony Cox said. And what's more, he added, it is exceedingly difficult to predict whether a particular ad message will help or hurt.

In a study funded by the National Institutes of Health, the researchers tested different print messages to see whether they affected the willingness of people to get vaccinated against hepatitis B, a potentially lethal liver disease transmitted by sexual contact and drug abuse. Their test subjects were 213 randomly selected clients at three public clinics offering treatment for sexually transmitted diseases -- precisely the group that should get vaccinated.

One group saw messages emphasizing that the vaccine would greatly reduce the chance of getting hepatitis B or spreading the disease to uninfected partners. Advertisers call this a "gain-framed" message, because it highlights the benefits that would be obtained from getting vaccinated.

The others read a "loss-framed" message that stressed the awful consequences -- death or infecting a loved one -- of not getting vaccinated. (Only a few words were changed to alter the messages. The positive message included this statement: "People who get the hepatitis B shot are gaining a chance to protect themselves and the ones they love," while the negative message read, "People who don't get the hepatitis B shot are losing a chance to protect themselves and the ones they love.")

These subtle but opposing approaches produced dramatically different results. Test subjects who read the positive pitch were inclined toward vaccination, and more likely to discount so-called "nuisance risks" ("the shot will be painful"). The subjects who read the negative appeal were more likely to say they didn't want to get vaccinated and to worry about the nuisance risks, the researchers report in a forthcoming issue of the Journal of Marketing.

The solution seems simple enough to me -- design ads that emphasize the benefits and avoid messages that attempt to scare people into action.

Not so fast, says Anthony Cox. An earlier study produced what he called "a very different" finding.

In that study by the Coxes, published four years ago, women were shown different print ads that encouraged them to get a mammogram. One ad emphasized the benefits of getting the test, including a greatly reduced risk of dying from breast cancer. (Significantly, Cox noted that similar upbeat messages are almost universally employed in campaigns encouraging women to get regular breast exams.)

The other ad emphasized what would happen if they didn't get a mammogram, including the fact that they ran a much higher risk of dying from cancer. A third group saw no ad. Then the women in the three groups were asked whether they intended to get a mammogram and their overall views of breast cancer, including whether they expected to develop the disease sometime in their lives.

Those who saw the upbeat ads said they were less likely to get a mammogram than those who saw the negatively framed ad. And if that weren't enough, they also were more likely to believe that they would not get breast cancer than either the group that saw the negative ad or the women who saw nothing.

So, Professor, what's your prescription for deciding whether to go positive or negative in a health campaign?

"Pretest these campaigns to see what the real-world effects may be," Cox advised.

And cross your fingers.

The Great Katrina Giveaway

Forget about rebuilding most of New Orleans, says Harvard University economics professor Edward L. Glaeser. Instead, he suggests, Katrina's victims might be better off if the federal government gave each of them $200,000 -- or some similarly significant sum -- so they can move to a higher, drier, more economically vibrant city.

Glaeser says that some of the city should be rebuilt -- at least the parts on higher ground, as well as the port and oil pipelines -- but the rest should be left to rest in soggy peace. He points out that the city was declining long before it went through nature's soak and rinse cycle, with many of its residents mired in poverty for decades. Its port and pipeline operations cannot employ a large workforce and spending $200 billion is unlikely to change that, he asserts in the latest issue of the Economists' Voice, an online journal.

Instead of rebuilding everything, the federal government should give permanently displaced residents a grubstake and let them start anew, Glaeser proposes.

"Giving checks to impacted residents, who then will move to Houston or Atlanta or Las Vegas, will actually reduce the negative spillovers from dysfunctional neighborhoods -- not increase them," he wrote. "Perhaps, if significant funds are given to New Orleans residents to help them start life anew in some more vibrant city, then there will be a silver lining to Katrina after all."

Call It Hand-Brain Coordination

Switch hitters in baseball must have very good memories.

At least that's what your Unconventional Wiz is guessing after reading an article in the latest issue of Cognition and Memory by a research team headed by psychologist Ruth E. Propper of Merrimack College.

The researchers tested 98 undergraduates to see how often they used either hand, even if they considered themselves to be left- or right-handed. They also tested the students to see how well they remembered lists of words.

People who showed greater "mixed handedness" performed better on memory tests than those who used one hand exclusively. The reason: Being nimble with both hands is a sign of increased interaction between the left and right hemispheres of the brain, which also aids recall, these researchers claimed.

morinr@washpost.com