Chicago Tribune (CT) - Sunday November 15, 1987
John Crewdson, Chicago Tribune

One official said the report would "scale down" the widely quoted projection that as many as 1.5 million Americans are infected with HIV, the virus suspected of causing AIDS.

"It's going to open some eyes," the official said of the still uncompleted report that was ordered by President Reagan nearly six months ago. "It's going to be a lower number, and it's going to show that the spread has slowed down drastically."

A second administration official said that while U.S. Public Health Service epidemiologists had talked about "a wide range of possibilities" for adjusting the upper and lower estimates of the AIDS epidemic, "they haven't committed themselves to a particular number at this point."

Sources said the range of infection discussed at a White House meeting last week was "from a low of 350,000 to a high of 1 million," considerably below the range of 1 million to 1.5 million set by government researchers in the summer of 1986.

The second official said Public Health Service administrators had acknowledged privately last month that the infection "is not spreading beyond the existing risk groups and a small range just outside-sex partners of intravenous drug users and so forth."

The official quoted the administrators as saying "there is no rapidexpansion, and they don't expect there to be a rapid expansion."

A key element in the new, lower AIDS estimate, according to those officials and others, will be so far unreleased data showing that the trend of HIV infections among prospective military recruits has remained "essentially flat" during the last six months.

As the increase in both AIDS cases and new HIV infections begins to level off among homosexual men in New York and San Francisco, and among heterosexuals nationwide, a consensus is also building outside of government that the current AIDS estimates are too high.

"I'm sure they'll come out lower this time around," said Andrew Moss, a professor of epidemiology at the University of California Medical Center in San Francisco, where about 10 percent of the nation's AIDS patients reside.

"One-point-five million is too large, and the case predictions for the next few years are also too large. All these big numbers are being scaled down, and the reason they're being scaled down is that the wrong numbers have passed into public circulation.

"It's a very political number, but I think one needs to be realistic about these numbers," Moss said. "People fall in love with big numbers to crank up a problem. The media likes big numbers too. The downside is that, when you get the real numbers and they aren't so big, people minimize it." White House officials have been angry for months with the Public Health Service, and particularly its Atlanta-based component, the Centers for Disease Control (CDC), over what they call a sluggish response to the President's order to sharpen the focus of the administration's estimates of the AIDS epidemic.

"They're spending a lot of time counting AIDS victims, but that's telling us where the infection was five years ago," said one official. "We're missing the most essential information, which is where the disease is now."

The new AIDS report, which under Reagan's order must be finished by the end of this month, will draw upon all the available data from tests administered for the presence of HIV, the virus believed by many researchers to play a key role in causing acquired immune deficiency syndrome.

Among the groups being tested are most potential blood donors, some hospital patients and those who visit venereal disease clinics, as well as millions of active duty military personnel and prospective recruits.

Because they are not randomly selected, none of the groups is representative of the larger population. But even a large nonrandom sample has some validity, and the data collected from more than 1 million recruits over the past two years is thought to be reasonably typical of young people in general.

It is also encouraging. Two sources who have seen the latest Pentagon numbers say the incidence of HIV infection has declined for black and white men, the two largest groups; has remained the same for white women, and has risen only for black women and Hispanic men.

A particular point of contention between the White House and the Centers for Disease Control has been the latter's assertion, which has become a feature of virtually every news article and broadcast about AIDS, that 1.5 million Americans may already be infected with HIV.

Among epidemiologists, that estimate has become known as the "Coolfont number," after the West Virginia resort where the Public Health Service conference that produced it was held in May, 1986.

Some White House staff members, sensitive to charges that the Reagan administration has not done enough to combat AIDS, have long contended that the Coolfont number is too large and that the administration's response has been adequate.

The Public Health Service's reluctance to explain how the number was derived has also been the source of much irritation.

"We finally forced out of them an explanation for Coolfont, and it is sorrier than you might have imagined," one White House official said last week.

According to a two-page Public Health Service memo supplied to the White House, the conference participants estimated the numbers of homosexual men, intravenous drug abusers and hemophiliacs, the three groups among whom the percentage of AIDS patients is greatest, then tried to guess the prevalence of HIV infection in each group.

One administration official dismissed the Coolfont estimate, which virtually ignored all the HIV testing data then available, as "not a serious effort." Another official declared that "the future of American science is in dire peril if they rely on an estimate like this one."

It was because of dissatisfaction with Coolfont that the White House health policy working group last month developed its own estimate of the prevalence of HIV infection.

The seven-page memorandum, a copy of which was obtained by The Tribune, suggests that only about 680,000 Americans are infected with HIV. One White House staff member said that when he shared the estimate with CDC officials they told him: " 'Oh, we agree with you. It's probably real low.' "

According to researchers outside government, the White House estimate is not beyond statistical probability for the lower boundary of the AIDS epidemic. It is also close to the midpoint of the 350,000 to 1 million range being discussed for the HIV "infection pool."

But some researchers think that dueling with computers over the prospective size of an ongoing epidemic whose precise magnitude can never be known is both futile and unnecessary.

"The White House came up with 700,000," says MIT economist Jeff Harris. "Then you've got the CDC people who say 1.5 million. Now you may say, 'Oh, it's a hundred percent off.' But all those numbers are within a factor of two.

"They're still saying that hundreds of thousands of people are infected. What are you asking for, the U.S. Census? What are you trying to do, allocate revenue sharing block by block? Does this change our public policy response?"

"Models are bad ways to do this," agrees UC's Moss. "It's better to look at what we actually know. If you look at what's happening in San Francisco and New York, you see the number of (new) cases about flat. It's been flat for some time, and it will flatten out elsewhere among gay men. In gay men, the thing has essentially stopped."

Jim Wiley, a demographer and social epidemiologist at the University of California's Berkeley campus, agrees.

"We're close to the peak incidence of disease right now in San Francisco," he said. "It looks like we'll get 1,400 or 1,500 new cases per year through 1993."

Among heterosexuals, the AIDS epidemic has never really begun. The number of heterosexual AIDS cases is small by any measure, only about 1,700, and while such cases are continuing to increase, they are doing so at an ever slower rate. Even those numbers overstate the rate of spread.

About half the heterosexual cases recorded by the CDC, as MIT's Jeff Harris points out, "are people who are called heterosexuals because they came here from countries where heterosexual spread is significant. That doesn't mean it's a reflection of heterosexual transmission in the United States."

Of the remaining thousand or so heterosexual AIDS patients, all are described by the CDC as having had "heterosexual contact with a person with AIDS or at risk for AIDS."

Put another way, no one in this country who has fallen ill with AIDS is believed to have acquired the disease from someone who is not a bisexual man, intravenous drug user, hemophiliac or a member of one of the other recognized risk groups.

Because incubation times for AIDS patients vary wildly, counting those who have contracted diseases attributed to HIV infection is less instructive about the future than knowing how many people are infected with the virus-are "seropositive"-though not yet ill.

Blood tests among homosexual men, who make up by far the largest risk group for AIDS, show that the number who are becoming infected with HIV has been dropping sharply, probably because of the more conservative sexual behavior many homosexuals say they are adopting.

"The seroconversion rate in gay men crashed quite a few years ago," says Moss. "San Francisco's doing new estimates, and they're a lot lower than any previous estimates. In San Francisco we've got less than 5,000 cases and maybe 20,000 seropositives. In New York, if it's the same ratio, it's only 40,000 seropositives."

Together, San Francisco and New York account for about 15,000 AIDS cases, more than one-third of the nation's total, and the two cities are believed to be home to the bulk of the nation's gay men.

That the rate HIV infection among heterosexuals is also low is a source of encouragement among epidemiologists.

"Most people in the business think it isn't very big at all," Moss says. "You can't find it when you look for it."

But the news appears to have surprised some within the Reagan administration.

Two administration officials mentioned a recent exchange between Education Secretary William Bennett and James Mason, the Centers for Disease Control director, at a meeting of the White House Domestic Policy Council to discuss the federal government's response to the epidemic.

"Bill Bennett said, 'You mean this thing is not exploding into the heterosexual community?,' " one White House official recalled. "And Dr. Mason said, 'No, it's not.' And Bennett said, 'Well, why have you been telling everybody that it is?'

"We asked them what they were going to do to change their approach to AIDS awareness month," the official said. "To this date there's been no answer, and that was in September."

AIDS awareness month, which took place in October, featured a public service campaign built around the theme that AIDS is "the killer that doesn't discriminate."

The evidence for the low rate of HIV infection among heterosexuals comes from several sources, including the Pentagon test data.

Military researchers believe that homosexual men and heroin users, the two groups most likely to test positive for HIV, have been applying for military service in smaller numbers since the widely publicized Pentagon testing program began two years ago.

While such deferrals would make the recruit applicant pool less representative of those most likely to get AIDS, they would presumably leave it weighted in favor of heterosexuals with no special risk, a bias that would make them more typical of the general population.

If HIV ever does cross into the heterosexual population in a major way, it is most likely to do so in a place where there is already a high incidence of AIDS and to show up first among those residents who are most sexually active.

Such a place is the San Francisco Bay Area, but even there not much HIV infection has been found among heterosexual patients at clinics that treat sexually transmitted diseases.

"They're doing blind testing in the VD clinics here, and they're getting absolutely flat HIV positivity with respect to heterosexual infection," says Berkeley's Wiley.

"It's encouraging to me, because it suggests that in the highest risk groups, which one presumes are the people going to VD clinics, we're seeing very little growth of transmission. Right now, heterosexuals are at relatively low risk."

The growth rate of an epidemic is measured by the number of new cases produced by any given case. When the number is greater than one, the result is epidemic growth. When it equals one, there is an endemic persistence of the disease. When it is less, there are only isolated outbreaks.

"What we're getting are isolated cases of transmission from IV drug users to their partners, and from bisexual men," Wiley says. "Part of that is a function of the fact that a disease like this feeds on itself. If you can keep it below a certain threshold, it doesn't produce a lot of new infections, and I think we are below the threshold for reproduction of the epidemic."

Wiley, who has studied the transmission rates of HIV between men and women infected with the virus and their regular sex partners, concludes that "the chances of acquiring the infection are much less than for syphilis or gonorrhea, even herpes-less than almost any other venereal disease you can name, on the order of 1 in 1,000 contacts."

Like other epidemiologists, Wiley points out the potential dangers of what might be called a hydraulic effect-the possibility that a relatively tiny change in some variable, like sexual behavior or the virulence of the virus, can produce an increase in infection that is larger by several orders of magnitude.

"We have a window here where we don't have a lot of heterosexual infection and it doesn't seem to be spreading very fast," he says. "But it's still possible to have a heterosexual epidemic, and I think people ought to be relatively conservative until we find out more."

Among the most important things to be learned is whether those heterosexuals who are becoming infected with HIV are acquiring the virus from people in the general population.

"It's one of the central issues now," says MIT's Jeff Harris. "Are we beginning to see heterosexuals who don't have any history of contact with a risk group member? By and large, the observation is that the infected heterosexuals are either people who are in a high-risk group or who had contact with a high risk group."

The principal shortcoming of predictions about the future of shape of AIDS in America is that the predictors have barely five years' worth of data with which to feed their computer models. The biostatisticians, one epidemiologist says, "have taken time-series forecasting about as far as it can go, and that isn't very far."

"I've seen various estimates of the number of AIDS cases and I think they're all very shaky," says Wiley. "We're in a situation here where we need very, very refined estimates. The way to really know this is to do really good seroprevalence surveys in the general population."

The Centers for Disease Control-most reluctantly, according to some in the White House-has lately agreed to put together a prospectus for an AIDS survey of randomly selected households across the country.

Such a survey, which would use the same techniques that public opinion polltakers employ when asking citizens how they plan to vote, could provide a near definitive answer to the question of how widespread the AIDS virus is.

But the CDC has offered several objections to proposals for a random, nationwide AIDS test that would essentially involve knocking on doors and asking whoever answers for a sample of their blood.

One is that, whether because of privacy concerns or fear of hypodermic needles, not enough Americans are likely to cooperate with such a survey to lend the results the needed statistical significance. Another objection has been that, since AIDS appears to be confined to the recognized risk groups, the incidence of HIV infection among the general population would be close to zero anyway.

"The President ordered an epidemiological survey to see how widespread HIV infection is, and they have used every means possible to avoid that," one angry White House staff member said.

"First they were trying to get out of doing the survey by pointing out that probably nobody will turn up positive. Then they said they decided it couldn't be done, or that it if it could it would take 2 1/2 years. For some reason, they don't want to test for this disease."

Epidemiologists outside the government are not so pessimistic, however. A random HIV prevalence study, said Moss, "is surprisingly easy, much easier to do than people think. People didn't think you could do it with gay men either, but in fact you could."

Moss referred to the San Francisco Men's Health Study, in which hundreds of men selected at random from city census tracts where the incidence of AIDS is highest agreed to have their blood drawn and tested for HIV. That study, which is continuing, has provided much of the best current data about the progression of AIDS and HIV infection among gay men. Now the UC researchers who helped design the men's health study have begun a random survey of the city's general population.

"We're doing one in San Francisco for heterosexuals," Wiley said. "There will be about 1,000 people in the first study, and we have another one that's being planned for 1,500."

Judging from the first 120 interviews, the randomized technique appears to work.

"We're getting a high participation rate," Wiley said. "It's about 70 percent, the same return rate as we get in ordinary survey studies.

"I was surprised, because we were going into some fairly rough areas of town-the Tenderloin and South-of-Market area, the Western Addition and the Inner Mission areas, all traditionally tough areas to do research on any topic."

Despite the initial success, the CDC remains skeptical. "They regard San Francisco as kind of a strange place," Wiley said. "They may concede that I can do it in San Francisco, but their argument is that it won't fly in Peoria."

Nonetheless, a White House official said that the CDC was now going ahead with plans for a nationwide survey, which it hoped to complete within 18 months.

"Somebody over there at CDC has got religion," he said. "It's been explained to them that they do think it's possible, and at last they have seen the light."

CAPTION: Photo: UPI photo. Dr. Beny Primm, a member of the AIDS task force, takes necessary precautions at Mt. Sinai Medical Center in Miami Beach.

CHART: The AIDS epidemic. Risk groups are remaining about the same. In percent of new cases classifiable by group

1986

-Heterosexuals ................. 6.6% (822)

-Transfusions & hemophiliacs ... 2.8% (346)

-Homosexuals & intravenous

drug users ................... 90.2% (10,767)

1987

-Heterosexuals ................. 4.3% (70)

-Transfusions & hemophiliacs ... 3.8% (62)

-Homosexuals & intravenous

drug users ................... 91.7% (14,693)

Rate of increase is leveling off (increase in new cases per year in thousands, 1983 thru '87); changing ratio of dead to living shows epidemic entering a later stage (1982 thru '87).

Note: Figures are as of late September of each year; percentages do not add to 100 due to rounding. Chicago Tribune Graphic; Source: Centers for Disease Control.

871115
CT871101

Always watch for outdated information. This article first appeared in 1987. This material is designed to support, not replace, the relationship that exists between you and your doctor.

AEGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.

Copyright ©1980, 1987. AEGiS. All materials appearing on AEGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of AEGiS, or the party credited as the provider of the content. .