Clinical trials had previously reported higher rates of success, but Dr. Deeks and his colleagues insist that trials often do not reflect the real world. Most clinical trials, they said bluntly, "tend to enroll patients who are healthy, who haven't been on much therapy in the past, and who are highly motivated." In other words, he said point blank, "they are not the typical patient."

"This was a 'real world' study," insists Deeks.

He said that those participating in the UCSF study were more typical of patients seen regularly in such places as general hospitals.

People with advanced HIV were seen in this study as particularly vulnerable to experiencing a failure of protease inhibitor therapy. They had one or more risk factors, including their having gone into therapy with T-cells (infection-fighters wiped out by HIV) at low averages and high HIV viral loads in their blood.

Having previously taken other antiretroviral drugs also complicated matters for the test subjects. The new therapy requires that patients follow a strict drug-consumption schedule. Those straying from this schedule were likely to show evidence of therapy-failure. Human nature, Dr. Deeks seemed to be saying, is closer to the reality of his test-subjects—not clinical volunteers—but people who walk into doctors' offices from the street.

Clinical trial volunteers, he believes, are of more specialized personality types.

He also critiqued studies announced by pharmaceutical companies. These studies show success rates in protease inhibitor therapy. Patients—28 in number-- under a long-range study by Dr. Roy Gulick in New York have enjoyed a 79% success rate that is touted by pharmaceutical businesses.

The famed researcher, Dr. David Ho would not be quick, he says, to give easy explanations for such failures as the UCSF study highlights. Ho, of the Aaron Diamond AIDS Research Center in New York City is sure the reasons are multiple, he indicated.

Ho particularly emphasized failures due to a patient's faulty scheduling in a drug regimen, saying that the therapy is most likely to fail in those who have not taken other drugs because they fail to take clock-watching seriously.

This means consuming a total of 20 scheduled pills throughout each day.

Guidelines already established for protease inhibitor therapy were supported by the new study, however, and indicate that patients begin triple drug therapy using all three drugs from the start rather than adding protease inhibitors later.