- The Washington Times - Tuesday, August 5, 2008

EXCLUSIVE:

An investigation of research conducted at an Arkansas veterans hospital has uncovered rampant violations in its human experiments program, including missing consent forms, secret HIV testing and failure to report more than 100 deaths of subjects participating in studies.

The Office of the Inspector General of the Department of Veterans Affairs (VA) on Tuesday will release its findings in a report on human subject protection violations at the Central Arkansas Veterans Healthcare System in Little Rock. The studies involved thousands of veterans who had volunteered for behavioral and drug experiments.

The investigation, which began last August, reviewed more than a half-dozen human experiments - including studies of colon, breast and prostate cancer - that had been conducted since 2006.

It found that entire consent forms were missing, signatures were missing from consent forms, HIV testing was conducted without documented consent, and research officials failed to obtain witness signatures in a study involving patients with dementia.

Additionally, the investigation found that researchers had failed to report “serious adverse events” during the experiments, including the deaths of 105 veterans. The researchers were required to report such events, regardless of whether they were accidental or linked to the experiments, to the Internal Review Board.

The board, which conducted oversight of the experiments, had been implemented and operated by the University of Arkansas for Medical Sciences but was transferred to the VA after the investigation began. The VA created a review board and halted all new experiments involving human subjects.

“The issues at the VA medical center in Little Rock were detected by VA employees, revealed through investigations by [the Food and Drug Administration] and VA’s Office of Research Oversight, and ultimately referred to VA’s Office of the Inspector General,” said VA spokesman Matt Smith. “This is an example of VA detecting and fixing its own problems.

“The issues cited in VA’s Inspector General report are being addressed through an aggressive action plan. The Little Rock research program is under heightened scrutiny to ensure no recurrence,” Mr. Smith said.

“VA strives to provide world-class health care to its patients and that includes observing the highest professional standards in protecting people who agree to participate in medical research,” Mr. Smith said.

The University of Arkansas for Medical Sciences did not respond to calls for comment.

The VA employs strict guidelines for consent forms and witness signatures for experiments involving veterans to ensure that volunteers are fully informed of all risks and possible side effects associated with the experiments.

When a veteran signs a consent form, the principal research investigator is supposed to sign the form as well. However, the inspector general (IG) found that in one test the researcher did not sign nearly 200 forms until two months later. In four studies reviewed by the IG, times, dates and witness signatures differed on 103 consent forms.

In a review of several cancer studies involving 1,400 veterans, investigators randomly sampled the files of 105 patients and could locate only 20 consent forms.

The findings are being released on the heels of July 9 IG report prompted by a Washington Times/ABC News joint investigation that the VA failed to pass on new drug warnings and risks in a timely manner to more than 200 participants in a smoking-cessation study who were also taking the drug Chantix. The eventual warnings of hallucinations and possible suicidal behavior were issued too late for James Elliott, who was subdued by police with a stun gun during a psychotic breakdown.

The IG said researchers in the smoking study “did not ensure that patients involved in the smoking cessation study were notified of the risk of suicidal thoughts or behavior in a timely manner.”

The IG also said it could not confirm whether warning letters reached the intended study participants, and more than half have not signed and returned amended consent forms advising that the drug may cause hallucinations or psychotic or suicidal behavior.

In Arkansas, the IG report found fault with the IRB operated by the University of Arkansas for Medical Sciences, saying it “failed to identify and address severe and continuing noncompliance.”

“Every protocol [study] discussed in this report was audited and deficiencies were identified and communicated to the IRB,” the IG said. “The IRB did not suspend or terminate any of them prior to outside inspections.”

In addition, the IG said there was a failure to ensure researchers had required skills and training to conduct the clinical trials, and that unlicensed research personnel performed some medical procedures it described as “minimally invasive.”

One study to determine the success rate of heart surgeries required a cardiac catheterization one year later to see how open the arteries remained, but the procedure was performed on only one out of 70 patients.

In another study reviewed by the IG, the principal researcher told the IRB that no subjects were enrolled in the experiment, but told the Office of Research and Development that 47 veterans had been recruited for the study. The IG found that $138,000 had been spent on study-related expenses.

All research by one physician was suspended after it was discovered he was reviewing the medical records of 678 veterans for a study that never received approval.

As of March 18, the hospital was conducting nearly 300 active research projects, 200 of which used human participants and 55 that were evaluating investigational drugs. The IG recommended that the VA determine whether human subject research should continue at the hospital and to take appropriate administrative action.

In addition to creating the internal review board, the VA has since hired new officials to take over as chief of staff, assistant chief of staff for research and development and a new medical center director.