DRUG MONEY

It's every brand manager or marketing director's dilemma. Ad agency account and creative directors wince and hedge when the question is asked. Will the advertising work? Will it help sell more widgets?

But what if the stakes were greater than selling a few more Ford Explorers or boxes of Tide? What if the product in question is "deterrence," as in deterring kids from trying drugs and deterring parents from taking a passive attitude about "having a talk." And what if the clients footing the bill for these anti-drug messages are now the American taxpayers and a raft of media companies being coerced into handing over hundreds of millions of dollars of free time and space?

The stakes are higher than those of household brand management. Up to now, battling drugs with ads has been chiefly a private-sector enterprise. Now, with the U.S. government putting up hundreds of millions in taxpayer money, and agencies and other vendors ready to collect 10-15% of that in commissions, the ad strategy is quickly becoming institutionalized and may well influence ad funding -- both private and public -- and available time and space for other causes Congress and big business deem worthy.

While most advertisers today thoroughly research their potential consumers' demographics, psychographics, wants, needs, dreams, aspirations, household habits and responsibilities, the Partnership for a Drug-Free America (PDFA) and the White House Office of National Drug Control Policy (ONDCP) have jointly earmarked on a nearly $2 billion anti-drug campaign backed by good intentions and self-interested partisan politics, but also by flimsy research that would hardly justify launching a new stain remover, let alone a program meant to help keep children sober and alive.

At least that's what the authors of the research say.

No once casts doubt on the idea that anti-drug advertising can help deter use of, and especially experimentation with, illicit drugs by America's youth. What is being called into question, though, is whether the PDFA and White House are dealing squarely with the public, media companies and corporations that they now ask to jump on the advertising bandwagon with funding and support. At issue: thin and overly-determined research that can, and has, critics claim, resulted in wrong-headed ads that preach to the choir or even insult the intelligence of the target viewer. In addition, assumptions have been made that strongly influence the messages; that, for example, zero-tolerance of any illicit substance is the only acceptable paradigm.

If there were any doubts that the current anti-drug campaign may be a model for things to come, consider these words: "You can effect any social issue," says Mike Townsend, director of operations at the PDFA. "Anything you can identify that is affected by the attitudes of the American public is subject to this kind of advertising."

Justifying the gargantuan ad effort, the PDFA formally cites three pieces of work -- two can't yet be referred to as journal articles -- to support their ads' efficacy and, in effect, ground their entire enterprise. One emanates from The Johns Hopkins University School of Medicine, one from the Stern School of Business at New York University and the third from the University of Michigan's Institute for Social Research.

Brandweek has learned the following regarding the three research projects:

- The lead author of the Hopkins study, Dr. Evelyn Cohen Reis, while standing behind her 1994 paper, now casts grave doubts upon the research techniques that support it. "I think [her respondents] were telling us what they thought we wanted to hear," Reis said. "You can't tell, based on the paper, that it [the advertising] actually works." She published a press-release-ready claim on the impact of anti-drug advertising: "seventy-five-percent reported that they had decreased, stopped, or been convinced never to initiate drug use." Asked if he figure has "any validity," she said she didn't know, adding, "You can't prove it either way, whether they're just saying the 'right' thing."

- Despite years of effort, the NYU paper has not achieved publication. In fact, the lead author, Lauren Block, assistant professor of marketing at NYU, has recently withdrawn it from consideration. A fourth author, William Putsis, from the London Business School, was recently enlisted to provide further econometric support for the report, It will be resubmitted in about a year, Block said.

- The University of Michigan's Professor Lloyd Johnston, an adviser to the ONDCP, has never published his findings on the efficacy of PDFA advertising, Johnston has shared the data at professional conferences and also plans to publish in about a year.

The PDFA and ONDCP cling steadfastly to all three pieces of research, the only work the organization cites among the hundreds of academic articles extant on teens and drugs. In fact, Block's work is of particular importance since it actually analyzes the PDFA's own data from their first six years of operation. Meanwhile, Block said she has informed the PDFA that she's withdrawn her paper from publishing consideration to revamp it.

Given this trinity of ambiguity, the question arises whether the public and the media are footing the bill for a merely theoretical construct of many fine parapets and mom-and-apple-pie banners, The uncertainty, though, would not shock public health scholars. "There's no solid data that show the media campaigns create meaningful changes in behavior," said Lawrence Wallack, professor of public health at the University of California, Berkeley, who met with ONDCP in March of last year.

"My fondest wish is to get these campaigns rigorously evaluated," said William DeJong, lecturer on health communications at the Harvard School of Public Health, who consulted on the ONDCP campaign.

There is "no nice controlled study that unambiguously points to the ads' effectiveness," said professor Robert Hornik of the University of Pennsylvania, another consultant to the ONDCP.

PDFA's research chief Barbara Delaney declined, when asked to comment about the validity of the research, individually or collectively.

The stakes are not only high for the enterprise itself – fighting the acknowledged, steady rise in drug use among kids – but they have also never been higher for the people fighting the fight, namely the PDFA, having won greater funding and donated media than ever before.

The anti-drug forces can thank former New York governor Mario M. Cuomo for the new money. Talking with the PDFA's chairman Jim Burke a couple of years back about the marked decline in donated time and space for PDFA's ads, the ex-governor determined that PDFA needed government funding. So he phoned President Clinton, and soon the administration's drug "czar," Barry McCaffrey, was sitting down with Burke. McCaffrey, an ex-general turned director of the ONDCP, recalled how the Army's recruiting didn't really take off until it started buying advertising rather than relying on donated time and space.

And so, bypassing Washington's normally tortured route from conception to implementation, a bipartisan $195 million was soon in the kitty for research, expenses, time and space for the first year of the National Youth Anti-Drug Media Campaign. Pending annual Congressional approval, there'll be similar funding for each of the next four years, nearly $1 billion of public money aimed at saving impressionable youth, though fully half the advertising will be directed at adults. What's more, the buys will get dollar-for-dollar matches of time and space from media companies, almost $1 billion more, on top of the almost $3.3 billion the media donated for PDFA ads from 1987 to 1997. Still to be counted is a projected barrage by nonmedia companies set to kick in this fail, from anti-drug messages on fast-food tray liners to trailers on home videos.

As in any advertising enterprise, there is general consensus that a raft of ads backed by tens of millions, if not hundreds of millions of dollars, will raise awareness about a product, a service, a brand, a cause or candidate. But awareness does not equal purchase or even purchase intent. And before a company like General Motors or Colgate-Palmolive goes out and spends $100 million on an advertising campaign, they do massive amounts of state-of-the-art quantitative and qualitative research, producing data that determines how best to communicate to the target audience. But with the PDFA/White House effort, that data is simply gossamer.

The most glaring inherent weakness of the case is self-reporting, or drawing conclusions based on what kids say they react to and say they do, rather than measuring what they actually do and actually react to. An editorial in the April 1988 issue of The American Journal of Psychiatry, addressing a drug-use study involving 74,008 students, cast grave doubts on how definitive any such research might be: "Such size creates considerable power, even if limitations include shallowness, even some narrowness, behavioral focus, blending of different categories (use of all substances was blended into generic 'substance' use), reliance on questionnaires, and reliance on self-reporting in teenagers (in a realm where self-delusion, peer pressure, wishes to conform, wariness of adult truth-seeking, etc., are not unlikely to create systematic distortions)."

THE RESEARCH

A graduate of Harvard Medical School, Dr. Evelyn Reis garnered a post-residency fellowship at Johns Hopkins. Her first and only drug-issue study, entitled "The Impact of Anti-Drug Advertising," was co-authored by her faculty mentor and two other colleagues and was published in the December 1994 issue of The Archives of Pediatrics and Adolescent Medicine, an American Medical Association journal. In-school questionnaires were administered in the fall of 1991 to 6th, 8th, 10th and 12th graders in two middle schools and two high schools in greater Baltimore; one of each in the inner city and the suburbs. Not a random sample, the surveys were just given to everyone who happened to be present that day, a total of 837 students, a factor that might already confound results as drug-users tend to be truant or drop out at a higher rate than non-users.

The process was anonymous and voluntary, with teachers not involved in collecting the surveys. The survey asked about anti-drug advertising in relation to marijuana, cocaine and inhalants; questions on alcohol were included as well. Of the kids exposed to the ads, 75% reported a deterrent effect on their own behavior, the study claims. Broken down further, "71% agreed that anti-drug advertising 'convinced me never to start using drugs," 39% 'made me stop using drugs," and 36% 'made me use drugs less often.'" The last two findings are quite extraordinary, as the ONDCP, the PDFA and public health experts all express little confidence that advertising will have effect on regular, confirmed users.

What's more, though drugs have plagued Baltimore for years, lifetime use of marijuana, cocaine or inhalants was reported to be greater among high school students in the suburban school – 35%, versus the urban high-schoolers' 29%. Throw the younger kids into the mix, and the rates fall to 23% for the suburbs and 19% for the city. Yet, Dr. Reis' paper notes, "Urban students were more likely to report having family problems with alcohol and/or other drugs, The city kids who made it to class that day reported the ads had a greater deterrent affect: 81%, compared to the suburban rate of 70%.

In the paper's discussion section, Dr. Reis remarks on the counter-intuitive findings – city versus suburb – she was publishing: "The greater positive response by the urban group in this study is particularly interesting in light of recently reported conflicting results [published elsewhere]. In a study of urban, African-American teenagers, focus group discussions revealed that, while Partnership advertisements were easily recalled, many found them offensive and preachy. Many teenagers also complained that the advertisements were the products of a mainstream source, irrelevant to their urban youth counterculture."

"Being entirely self-reported is a huge limitation," said Reis, now an assistant professor of pediatrics at Pittsburgh's Children's Hospital, an affiliate of the University of Pittsburgh School of Medicine. "I think [the kids] were telling us what they thought we wanted to hear. It's called 'socially desirable' answers. My concern is the kids thought they were supposed to say the ads work, the younger kids more so."

Despite the PDFA-cited, press-release-ready deterrent claim (75%), Reis is more dubious. "You can't tell based on the paper that [the advertising] actually works," she said.

Oddly enough, Reis' faculty mentor at Hopkins and co- author, Dr. Hoover Adger Jr., is the current deputy director of the ONDCP. "You can't get into cause and effect in any self-reported [behavior] study," Adger said. But, "there is no doubt [anti-drug] advertising works."

He could not, however, cite any research to support that contention.

Reis' caution about relying on self-reporting is well founded. Consider the work of professor John Worden of the University of Vermont College of Medicine. He administered self-reported questionnaires in his study of anti-tobacco advertising. He put cotton balls in his youthful respondents' mouths and warned them he was checking their saliva to see if they'd been smoking. Led to believe lying was futile, the kids' socially desirable answers flew out the window; he got a 30% higher self-reported smoking rate. "Kids do get truthful if they think you're going to test them somehow," said Worden.

Reis ultimately stood by her and Adger's work. "My opinion is that we know children and adults are influenced by advertising, so it makes sense that the PDFA ads work... Is advertising the best way to allocate resources? That's the question. If the ads are the best [program] we have, maybe they're worth a gamble."

But the question remains whether the study is enough to substantially support a vast, multi-billion-dollar communications effort or whether the PDFA just grabbed hold of any piece of corroborative research it could. "The PDFA was very anxious to get its hands on the data and wanted to get it before it was published," Reis said. "We did resist that."

The PDFA also may have jumped the gun with the Block study. "The overall pattern of results suggests that the PDFA anti-drug PSA campaign has been effective in reducing adolescent drug consumption," writes Lauren Block, assistant professor of marketing at the Stern School of Business, New York University. "However, given the nature of the available data, it is not possible to prove a causal relationship between advertising and drug consumption."

That passage comes from the second report the PDFA cites as a foundation of its anti-drug campaign, the report since withdrawn from consideration for publication.

For her paper, Block was lucky enough to get access to PDFA's own data from its first six years of operation (1987-to-1992), involving some 8,000 respondents. Most were approached in malls, a highly questionable sampling technique usually used by low-cost researchers since abandoned by the PDFA. "The generalizability of the results to the [general] population of adolescents could be considered to be questionable," Block cautions in the paper.

Block told Brandweek the PDFA data does not support an evaluation of whether anti-drug ads can cause usage or trial reductions. But given the constraints of the available data, Block retreated to examining whether the ads are "associated with a decrease in adolescents' intentions to use drugs in the future."

Changes in behavior, of course, not intentions, are what really counts; or, at least, what concerns the public most.

Again, the issue of self-reporting rears its head, as does the methodology behind Block's draft and its basis on the PDFA's multi-year agglomeration of data. "With drug abuse, yes, there's a bias towards under-reporting," she said, Her report qualifies: "Since this quasi-experiment has neither a control group, nor random assignment, it is open to selection biases, history effects and other sources of error."

Another doubt arises simply from the samplings the PDFA used from 1987 to 1992. It maintained a disproportionate sample of women (55%) throughout. Since Block herself wrote, analyzing the data, "We also find that reported marijuana consumption is significantly more frequent for males than females," it raises the question of whether the PDFA stacked the numbers to bolster the appearance of its efficacy (again, emphasis added). It's also curious that the PDFA's researcher changed the sample from 81% white in 1987 gradually down to 55% white over the years 1987 to 1992, and, reciprocally, from 9% black to 28% black. It's an enormous swing in the sample to a population cross-section not representative of the nation as a whole. The PDFA's Delaney, though not at the organization when the changes in the sample were made, claims the increase in blacks and Hispanics in the sample was done to get a large enough representation to gather meaningful information on those groups, As for the 55% female sample, she contends that reflects the population at large.

Thirdly, the PDFA invokes professor Johnston of Michigan. A respected, experienced researcher in the field, Johnston's name alone confers a certain authority. His annual survey, supported by public money, is about as comprehensive as one can find. In 1996, he polled 49,000 students in 435 schools, up from around 3,000 in 1986.

Johnston has included questions about PDFA's ads in his surveys since the organization's inception in 1987, asking about recall, exposure, credibility and whether the ads made viewers regard drugs less favorably and made them less inclined to use them. But he's yet to formally publish the results, saying he's "just got more important things to do,"

When he does publish, Johnston's data will indicate that the PDFA ads' deterrent effect on kids who've seen them was stable from 1987 to 1992. In that period, three-quarters of the kids surveyed said the ads had at least a little effect in deterring their drug use and one third said the ads had a lot of effect, By 1995, the three-quarters figure for a "a little" fell to 62% and the one-third figure for "a lot" fell to around 20%.

Johnston is supportive of anti-drug advertising, citing its "counter-normative" effect, and he believes the PDFA advertising to date has lowered the use of both heroin and inhalants. But "it's very difficult to measure the effect of the ads, drugs are such a feature of the culture," he said.

He cautions, though, about "the fallacy of single causes. Advertising is only one, and not the strongest influence."

A steady decline in drug use from 1979 through 1992 was due, in part, to the cocaine-related death of basketball star Len Bias in 1986 and in part from the pervasive puritan ethic that seemed to characterize the Reagan-Bush years. That's no longer the case, Johnston said, which goes a long way towards explaining the increase in drug use of the 1990s. In constant dollars, federal funding for school-based drug education fell by 40% in the early '90s. There was a 93% drop in national network news stories on drugs from 1989 to 1993. Fewer celebrities died. And popular music and fashion, Johnston claims, exploded with pro-drug messages.

There was "a decline in role models [dying from drug use] and less vicarious learning," Johnston said. "Society was speaking with a single negative voice about drugs in the late '80s. I'd like to see a ubiquitous campaign, the voice of society [regain that tenor]."

Dr, Reis' doubts over the validity of her paper, and professor Block's retreat, even if temporary, leave only one leg of PDFA's self-designated three-legged stool in place, But, unpublished, it's currently a leg cobbled from claims made in Professor Johnston's conference presentations, i.e., speeches. Is that enough to set the course for a "voice of society"?

[Continued in Part 2 of 2 at http://www.mapinc.org/drugnews/v98/n000/a04.html ]