News your connection to The Boston Globe

Lapses cited in Harvard clinical trials

Federal health officials recently found ethical lapses in eight Harvard Medical School experiments involving human subjects and have asked the school to enact measures that would heighten protections for patients involved in clinical trials there.

The lapses involved errors by the medical school's institutional review board, or IRB, a 12-member ethics panel that reviews all planned experiments to make sure human subjects are protected and understand the risks involved.

No patients were hurt in the Harvard experiments cited. Two of the experiments are taking place abroad, in China and Tanzania.

The lapses were found in a random government audit, which reviewed 25 experiments conducted over the last four years.

According to a summary of the audit obtained by the Globe, the Harvard IRB mistakenly approved six experiments in which patients were given inadequate or confusing consent forms, the documents that spell out the risks of participating in a medical experiment. In another case, the IRB failed to conduct a second review of an experiment that had been modified by researchers after an initial review. In one case, a scientist launched an experiment involving human subjects without any review, a serious lapse that the IRB caught but failed to report to federal health officials.

The federal government also determined that the Harvard IRB panel was too dominated by white males, lacking the racial and gender diversity reflective of patients that participate in clinical trials.

Harvard has admitted to the lapses. In a recently submitted letter to the federal government, provided to the Globe, medical school officials ''acknowledge the need to improve and enhance our existing program and to address the concerns you raised with appropriate corrective action plans."

The letter outlined plans for more aggressive IRB reviews at Harvard. It also noted that the medical school plans to add black and Asian members to the IRB ''within the next several months."

The government audit was part of an effort by federal regulators to improve ethical standards at the nation's medical schools and hospitals, said federal officials. They have investigated two other comparable medical schools in recent months, the University of Alabama at Birmingham and the University of Texas at Austin, finding a relatively similar set of IRB lapses.

Harvard and federal officials described the audit process as amicable. The case will remain open until federal officials are satisfied with Harvard's response. A spokeswoman for the Office of Human Research Protections, which conducted the audit, refused to comment on the Harvard case.

Dr. Raphael Dolin, Harvard Medical School's dean for academic and clinical programs said: ''There's no real controversy involved here. The recommendations they've made we think will strengthen our program."

The lapses described appear to be mild in comparison to recent ethical scandals at other institutions, most notably at the University of Pennsylvania, where researchers did not fully disclose the risks of a gene therapy experiment that led to a teenage boy's death in 1999. That case prompted much of the federal government's effort to police clinical trials.

Medical research at Harvard is widely considered guided by a model of ethical conduct, and the lapses at Harvard would probably turn up at any medical school examined, said medical ethics specialists. Nonetheless, they illustrate the complexities involved in protecting human subjects.

''The larger message is the prevention of any kind of risk to subjects involved in human research," said Dr. Carolyn M. Connelly, Harvard Medical School's director of research subject protection. ''What we need to do, and continue to do, is educate researchers on how to do research and protect subjects."

The audit was conducted by the Office of Human Research Protections, a branch of the US Department of Health and Human Services. The office performs ''several" random audits annually, according to a spokeswoman.

Four agency officials arrived at Harvard's medical school campus in Boston on March 17. Harvard gave them a list of about 400 IRB-reviewed experiments run by the medical school. The auditors selected 25 and spent nearly three days reviewing paperwork and interviewing Harvard scientists and administrators. Their prime concern was the IRB's performance.

Every hospital and medical school that conducts research on humans must have an institutional review board. At Harvard, the 12-member panel is split between scientists and laypeople, including a lawyer, a retired nurse, and a rabbi. All but two have Harvard affiliations, working for one of the many schools or hospitals connected to the university. All are unpaid, and meet monthly to review cases. The federal investigators generally praised their work.

Neither the federal government nor Harvard would reveal the scientists involved in the experiments with ethical lapses.

One experiment used anonymous survey data about gambling habits taken in another state, according to Harvard officials. The researcher began working with the data without IRB approval, believing anonymous data didn't require it. The IRB flagged the researcher, but failed to report its finding to the federal health officials, drawing a rebuke.

Another experiment involved a pilot study in Shanghai on improving treatment of patients with depression. During an initial review, the IRB asked for more information about how researchers would deal with potentially suicidal subjects, pregnant women, and those stigmatized by their families and communities. The IRB gave the go-ahead to the project before getting those answers, according to the audit summary. Harvard has promised to follow the rules more closely in these types of cases.

Two years ago, three Harvard-affiliated institutions were criticized by the Office of Human Research Protections for lax patient protections during genetic research experiments in China. That case, which federal officials consider far more serious than the recent Harvard audit, was launched after a Harvard whistleblower made specific allegations.

Another experiment cited in the recent audit was a study of how villages in Tanzania cope with HIV. The consent form should have been written at a fourth-grade reading level, but contained such words as ''participation," ''organized," and ''assessment," which federal officials deemed possibly confusing for the test subjects signing it.

Medical ethics requires that patients fully comprehend the risks of any experiment before they agree to participate. Harvard said it will retain a local grade-school teacher to review consent forms intended for patients with limited education.

Two other experiments were cited for overly complex consent forms -- a small immune-system study and a dental experiment. Another experiment compared chiropractic treatment to acupuncture and to standard therapies for lower back pain. The consent form read: ''We anticipate no harm associated with this survey." But the government determined there were risks and that this was not an ''adequate description."

The government also found other minor problems in the consent forms for two other experiments involving a survey on medical school curriculums and heart disease.

But the government assertion that Harvard's IRB lacked diversity caused the most concern at the medical school.

''We are looking to increase our racial diversity," said Margaret L. Dale, Harvard's associate dean for faculty affairs. At the time of the audit, three of 12 IRB members were women, and one member was a minority. Federal officials said increased diversity would improve the IRB's ability to monitor experiments, which typically involve patients from a broad array of social groups. Harvard has since added a female member. The school is looking to add two black and one Asian member in the coming months, said medical school officials.

Harvard officials said they have been working closely with federal officials to enact the suggested improvements. In addition, government ethics specialists plan seminars on patient protections at Harvard. The federal Office of Human Research Protections recently received Harvard's reply and will respond soon, though the case could remain open for months.

''We don't usually set a time frame," said an agency spokeswoman, Patricia El-Hinnawy.

Boston University medical ethicist George Annas said: ''You could go to any IRB in the country and find this stuff."

Annas faulted one aspect of the audit. ''What you'd like to see them do is go talk to some research subjects," he said. ''How were they taken care of? Did they know what they were getting into?"

Raja Mishra can be reached at 

Globe Archives
Today (free)
Yesterday (free)
Past 30 days
Last 12 months