Altman, Lawrence K. "H.I.V. Study Says Rate 40% Higher Than Estimate," New York Times,


Altman, Lawrence K. "H.I.V. Study Says Rate 40% Higher Than Estimate," New York Times,
3 August 2008, pp. 1, 23.

MEXICO CITY The United States has significantly underreported the number of new H.I.V.
infections occurring nationally each year, with a study released here on Saturday showing that the
annual infection rate is 40 percent higher than previously estimated.

The study, conducted by the Centers for Disease Control and Prevention, found that
56,300 people became newly infected with H.I.V in 2006, compared with the 40,000 figure the
agency has cited as the recent annual incidence of the disease.

The findings confirm that H.I.V., the virus that causes AIDS, has its greatest effect among
gay and bisexual men of all races (53 percent of all new infections) and among African-American
men and women.

The new figures are likely to strongly influence a number of decisions about efforts to
control the epidemic, said the disease centers' director, Dr. Julie L. Gerberding, and other AIDS
experts. Timely data about trends in H.I.V. transmission, they said, is essential for planning and
evaluating prevention efforts and the money spent on them.

Dr. Gerberding said the new findings were "unacceptable," adding that new efforts must
be made to lower the infection rates. "We are not effectively reaching men who have sex with
men and African-Americans to lower their risk," she said.

Dr. Kevin A. Fenton, who directs H.I.V. prevention efforts at the agency, said, "C.D.C.'s
new incidence estimates reveal that the H.I.V. epidemic is and has been worse than previously

A separate historical trend analysis published as part of the study suggests that the
number of new infections was probably never as low as the earlier estimate of 40,000 and that it
has been roughly stable overall since the late 1990s.

C.D.C. officials said the revised figure did not necessarily represent an increase in the
number of new infections but reflected the ability of a new testing method to more precisely
measure H.I.V. incidence and secure a better understanding of the epidemic.

Dr. Philip Alcabes, an epidemiologist at Hunter College in Manhattan, raised questions
about the validity of the findings. If they are true, Dr. Alcabes said in a statement, the agency has
undercounted new H.I.V. infections by about 15,000 per year for about 15 years. "Therefore,
there are roughly 225,000 more people living with H.I.V. in the U.S. than previously suspected,"
he said. "The previous estimate was 1 million to 1.1 million."

A C.D.C. spokeswoman said Dr. Alcabes's estimates were incorrect because the new
figures could not be used to calculate the total number of people with H.I.V. The C.D.C. does not
know the total number but is expected to determine it later in the year.

The C.D.C., the federal agency responsible for tracking the AIDS epidemic in the United
States, said its new monitoring system provided more precise estimates than were previously
possible of new infections in specific populations. Infection rates among blacks were found to be
seven times as high as for whites (83.7 per 100,000 people versus 11.5 per 100,000) and almost
three times as high as for Hispanics (29.3 per 100,000 people), a group that was also
disproportionately affected.

The C.D.C. has known of the new figures since last October, when the authors completed
a manuscript and sent it to the first of three journals. But the agency refused to release the
findings until they were published in a peer-reviewed medical journal. The first two journals
rejected the authors' request for a fast-track review.

The paper is being published in the Aug. 6 issue of The Journal of the American Medical
Association. The journal and the disease centers had planned to release it at a news conference on
Sunday at the opening of the 17th International AIDS Conference here. But the paper was
released on Saturday because the embargo was broken. A number of leading health experts have
criticized the agency for not releasing the information earlier. On Nov. 21, C.D.C. officials told
AIDS advocacy groups and reporters that the data would be released soon.

In an editorial on June 21, The Lancet, an internationally prestigious journal published in
London, severely criticized the disease centers for failing to release the information and said,
"U.S. efforts to prevent H.I.V. have failed dismally."

Dr. Gerberding, in defending the decision not to release the data earlier, said: "This paper
has been scrutinized by some of the best statisticians in the country and is much better now than
when we started this process. It was so complicated that even I, who has some expertise in this
area, could not stand by it without making sure we had gone through the review process."

She added, "This is one of those examples where getting the external review process to
really scrutinize the paper, pick it apart, build it back up, has in my opinion fundamentally
improved it."

The delay, however, has also fueled criticism that the Bush administration, which has
earned plaudits for spending tens of billions to fight AIDS in a number of highly affected
countries, has not done enough to fight the disease at home.

Representative Henry A. Waxman, Democrat of California and chairman of the
Committee on Oversight and Government Reform, was critical of the administration. "H.I.V.
prevention has been underfunded and too often hindered by politics and ideology," Mr. Waxman
said in a statement released Saturday.

He said the administration had reduced domestic spending against H.I.V. "Since fiscal
year 2002, when adjusted for inflation, C.D.C.'s prevention budget has actually shrunk by 19
percent. The president has recently requested decreases in funding for H.I.V. prevention at

Mr. Waxman said he would soon hold hearings on why health officials had had "less and
less money to actually get these programs to the communities that need them."

Dr. Alcabes disagreed with critics who contend that the new numbers point to a failure of
United States policy on AIDS, saying his conclusion was that "it looks like prevention campaigns
make even less difference than anyone thought."

"H.I.V. incidence did not decline as much from the 1980s to the 1990s as we believed,"
he said, "despite the dramatic increase in condom promotion and so-called prevention

The C.D.C. said the findings confirmed sharp declines in the number of new H.I.V.
infections each year, from a peak of about 130,000 in the mid-1980s to a low of roughly 50,000
in the early 1990s. But the findings also indicate that the number of new infections increased in
the late 1990s but has since remained relatively stable, with estimates of 55,000 to 58,500 in the
three most recent time periods analyzed.

Dr. Gerberding said: "If there is any good news here, it is hard to report it, but there is a
little bit in the sense that while incidence rates are certainly too high, they are stable. That is
important because a stable number of new infections in a world that has got more and more
people with H.I.V. and people with AIDS living in it suggests that we are keeping up with that
pressure for transmission."

The revised figures are based on a new laboratory test that can distinguish between recent
and long-standing H.I.V. infection, as well as on statistical measures and extrapolations.

The test is done in the laboratory on left-over serum from the standard H.I.V. test after it
shows that a person is infected. The Food and Drug Administration has approved the test, known
as BED.

Dr. Gerberding said that an earlier version of the new test was too crude to do the same
kind of study earlier this decade. Despite the improvements in the test, she said, it is "not perfect