The Wrong Rules for Social Science?

Scholars say U.S. human-subjects protections, designed for medical studies, hinder vital work

By JEFFREY BRAINARD

Last August, Sean D. O'Laughlin was eager to wrap up his master's thesis in history at Illinois State University. He'd been at it for a year, and a master's degree would earn him a raise at the high school where he teaches history.

But just before he was to graduate, he found he had a big problem.

His thesis traced how the social-sciences curriculum had evolved between 1890 and 1960 at his high school, in Pekin, Ill. He had interviewed retired administrators and teachers, and some former students. They had signed consent forms agreeing to participate -- a standard practice for oral-history interviews.

But when he submitted his thesis to the graduate-school office, red flags went up. Before doing the interviews, he hadn't sought approval from the university's institutional review board, or I.R.B., which under federal law oversees research involving human participants. As a result, Mr. O'Laughlin was told that his thesis could be rejected if the board decided he hadn't followed proper procedures.

Over the next two weeks, M. Paul Holsinger, his faculty adviser and a professor of history, held a hasty series of talks with I.R.B. members. After reviewing the project, they found it posed minimal risk to the people interviewed, and approved it. But for Mr. O'Laughlin, the wait was harrowing.

"Perhaps it was naive on my part, but I never imagined that [requirement] would apply to oral history," he says. "It's unbelievable, after the hundreds of hours you put into this."

Increasingly, scholars in the social sciences and humanities at other campuses are reporting similar run-ins with I.R.B.'s. The projects in question include oral-history interviews, survey research, anthropological fieldwork, and journalistic interviews.

The review boards typically worry about whether questions asked by researchers might strike the participants as intrusive violations of their privacy, and whether sensitive details they share are kept confidential. But social-science and humanities scholars complain that they are being unnecessarily burdened by a system of supervision that was primarily designed to prevent abuses of patients involved in medical studies -- where a poorly designed survey could result in serious harm or even death.

Oral history "is so outside the realm of what human-subjects protections were originally designed to deal with that a rational person would wonder what is happening," says Alan Lessof, an associate professor of history at Illinois State University.

Such complaints have begun to get a wider hearing. In a draft report issued in December, the National Bioethics Advisory Commission, which advises the president, suggested giving I.R.B.'s more flexibility in how they monitor social-science research. At the same time, however, the government is moving toward a policy that would force encourage scholars who interact with people as study subjects to undergo training in research ethics, a development that alarms some social scientists.

In the past two years, the federal government has ratcheted up pressure on all researchers studying people, urging them to do a better job of protecting the participants' safety and dignity.

The push has followed some widely publicized lapses in review of medical experiments -- especially the September 1999 death of 18-year-old Jesse Gelsinger in a gene-therapy trial at the University of Pennsylvania. Between October 1998 and July of last year, officials of the National Institutes of Health suspended federally financed research at eight institutions.

"After the hysteria that set in, I.R.B. review has turned into the equivalent of practicing defense against medical malpractice," says Michael Silverstein, a professor of anthropology and linguistics at the University of Chicago. He served for four years on an I.R.B. there that reviews only social-science research.

The methods and ethics of social-science and behavioral research were not at issue in the federal suspensions. Nevertheless, such research falls under a set of federal regulations that have evolved over the past 50 years in response to abuses like the Nazi experiments on prisoners during World War II, and the Tuskegee Syphilis Study. In that experiment, which ran from 1932 to 1972, the government deliberately and deceptively withheld treatment from poor black men to study the progression of the disease.

The regulations apply to universities that receive federal research grants to study human subjects. But many universities have voluntarily extended their review to research not financed directly by the government.

To be sure, social scientists have conducted some studies that are viewed today as unethical -- including the work of the late Stanley Milgram at Yale University.

In 1961, Mr. Milgram asked volunteers to administer a quiz to strangers. For "wrong" answers, the volunteers were told to give electric shocks to the quiz-takers that escalated in severity. Even though the quiz-takers yelled in pain, the volunteers complied. In reality, as the distraught participants learned afterward, the shocks were fake, and the quiz-takers were acting. Mr. Milgram used the experiment to draw conclusions about people's susceptibility to commit immoral acts when instructed to do so by authority figures. But critics have said he violated the volunteers' dignity by manipulating them.

Questions about the proper ethical conduct of social-science research continue to arise. Anthropologists are debating accusations, published last fall, that researchers who studied Yanomami Indians in Venezuela beginning in the 1960's mistreated them -- by fomenting warfare in order to film it, for example.

Calls for closer scrutiny of social-science research have also come from outside academe. Universities should more closely monitor academic projects that survey opinion among public-school students on controversial behaviors, like illegal drug use and sexual activity, says Robert H. Knight, senior director for cultural studies at the Family Research Council, a conservative public-policy organization in Washington.

Federal rules require I.R.B.'s to pay especially careful attention to studies involving minors. Nevertheless, Mr. Knight says, some survey questions "seem designed to make students feel like they are dweebs if they haven't tried certain things, and that is dangerous." Universities could do more, he says, "to make sure that there won't be intrusive questions that might be unnecessary to the research."

But over all, social scientists argue that their work typically poses little risk to participants -- and that careful scrutiny by an I.R.B. is thus an unnecessary distraction from their work.

"There's a real difference in the nature of the harm done between not being treated for syphilis as in Tuskegee, and asking someone a blundering, embarrassing question" during an oral-history interview, says Linda Shopes, a historian with the Pennsylvania Historical and Museum Commission.

As with proposals for medical research, I.R.B.'s rarely reject social-science studies outright. Instead, researchers are increasingly voicing complaints that review boards are proposing cumbersome, needless modifications to their work. The reports have been anecdotal, and were elicited by a survey conducted last year by several social-science organizations in cooperation with the American Association of University Professors. The survey covered about 200 social-science faculty members, and the association summarized their worries in a report issued in December. The report can be read online (http://www.aaup.org/IRBdoc.htm).

Some researchers complain that I.R.B.'s are reviewing research projects that are exempt from review, or that they are delaying the approvals process for projects that should be exempt. The federal regulations allow the boards to exempt certain low-risk social-science research, such as opinion surveys that do not name respondents, from review.

However, federal officials and I.R.B. members -- who fear that any perceived lapse in supervision could lead the government to shut down their campus's research -- have discouraged scholars from deciding for themselves whether a project is exempt.

Other scholars say the I.R.B.'s are asking them to keep the names of their interview subjects anonymous, or requiring them to destroy audiotapes from interviews after they have been transcribed. I.R.B.'s also have asked to screen interview questions in advance, and have urged scholars to avoid asking interview subjects sensitive questions that might embarrass them.

Nothing in the federal regulations specifically requires I.R.B.'s to seek the anonymity of sources or the destruction of audiotapes -- the boards have developed these suggestions on their own. But because of the pressure on I.R.B.'s to review all research more carefully, the boards are taking a closer look at research that they might have quickly approved a few years ago. While some scholars say these I.R.B.'s are overzealous, the panels say they are merely doing their job -- protecting research subjects.

Many social scientists are not submitting their projects to the boards, either deliberately or out of ignorance that the rules exist, according to interviews and the A.A.U.P. survey. Some researchers are calling the very existence of the supervision a violation of academic freedom and their First Amendment rights.

When they do deal with the boards, "most historians are ultimately able to resolve differences with I.R.B.'s, but the negotiations can be lengthy and frustrating," says Ms. Shopes of the Pennsylvania commission. The encounters, she adds, create among scholars a "disdain for a narrow bureaucratic process rather than a healthy respect for underlying ethical principles."

Some I.R.B.'s do seem to be erring on the side of caution and need to find "the appropriate middle ground," says E. Greg Koski, director of the federal Office of Human Research Protection, which oversees the boards. He says that one of his goals is to help teach institutions that existing regulations allow them to tailor their level of review to a particular project's risk. "We want to be sure we're not using a 50-pound hammer to drive a thumbtack," he says.

When proposed research is not exempt, federal regulations still allow the review boards to handle it through an expedited evaluation conducted by a single member of the panel, rather than the full board. Most kinds of social-science research -- including surveys and oral-history projects -- are eligible for such a review. But few I.R.B.'s use it, researchers say.

Even with expedited reviews, a single review-board member can place conditions on research. One request that especially worries oral historians is that they mask participants' identities.

"If you keep a person anonymous, much of the context of that person's individual perspective is lost," says Mary A. Larson, assistant director of the oral-history program at the University of Nevada at Reno.

Before she came to Reno in 1998, she was at the University of Alaska at Fairbanks, where she and other colleagues interviewed Alaskans on such topics as the history of the state's national parks and the careers of the small-plane pilots who serve Alaska's vast interior.

At one point, the staff of the university's I.R.B. requested that the historians keep their participants anonymous. But the request was dropped after Ms. Larson laid out a plan for participants in the studies to review and approve the interviews after they were transcribed. That is a standard procedure recommended by the 1,500-member Oral History Association for all interviews.

"I can't think of anyone who, in the eight years I was there, asked for anonymity," Ms. Larson says. "Sometimes, the protections that I.R.B.'s want are more than what the people want."

Social scientists also have become irritated when review boards have asked them to obtain signed forms from a study's participants pledging their informed consent. Such documentation is required by federal regulations.

The formal tone of these documents can put off some people, especially those lacking education or language skills, Ms. Larson says. Participants may be especially unwilling to sign if the study involves questions about criminal behavior. In some studies, where the researchers do not need to identify the subjects, and medical histories are not at issue, the consent form is the only documentation of their identities.

Scholars argue that formal documentation of informed consent is overkill in interviews, where participants can refuse to answer nosy questions or terminate the session altogether.

Consider, for example, students at the University of North Carolina at Chapel Hill's School of Journalism, who interviewed professional journalists as part of their theses on the history of the media in the United States.

"We've had students get a very strange reaction from professional journalists when they stick a permission form under their nose," says Margaret A. Blanchard, a professor of journalism and mass communications. The professionals, she notes, "ought to be able to safeguard themselves."

But not all interview subjects are as savvy, says Gary L. Chadwick, executive director of the review board at the University of Rochester. "What seems obvious to us -- that people can drop out -- may be less obvious to the subjects," he says. "People ought to know what they're getting involved in."

The National Bioethics Advisory Commission considered such arguments, and its draft report suggests a new, more flexible regulation about documenting informed consent in low-risk studies. For example, instead of signing forms, participants could state on an audiotape their consent to participate -- preserving anonymity for those who desire it.

Even with such a change, journalism-school projects have posed a particularly knotty problem for review boards. Ms. Blanchard and some journalism professors at other institutions argue that the First Amendment prohibits any I.R.B. review of such scholarship.

Nevertheless, the North Carolina journalism school last year reached an understanding with the I.R.B. that such projects should be reviewed, with expedited review whenever possible. Student-written news stories are exempt.

Other complaints about review boards appear to stem from a clash of cultures between social scientists and the biomedical researchers that predominate on the boards at some institutions. In particular, medium-sized and small universities and colleges may have only one panel to review all human-subjects research. During the same meeting, such boards may consider oral-history projects and medical studies involving experimental drugs with potentially life-threatening side effects.

Federal regulations do not specifically require I.R.B.'s to include social scientists. Larger universities typically create more than one I.R.B., and divide their responsibilities so that at least one of the boards deals exclusively with social-science and humanities research projects. However, no one has systematically studied the prevalence of such specialized panels among the estimated 3,000 to 5,000 I.R.B.'s that now exist nationwide.

The December report of the American Association of University Professors recommends that universities make a concerted effort to recruit more social scientists to serve on I.R.B.'s that review their work.

"At least if you can get somebody on the board who understands historians and journalism, they can argue with the scientists about what's an appropriate review, and not have a one-size-fits-all approach," says Donald A. Ritchie, assistant historian with the U.S. Senate Historical Office.

The A.A.U.P. and social-science organizations also have sought a voice in developing a federal policy to be issued this year, encouraging human-subjects researchers to be accredited to assure they are familiar with government regulations and ethical conduct.

Researchers receiving federal grants would be urged to obtain accreditation issued by a private, nonprofit group. To be certified, researchers would first have to undergo training, perhaps in an on-campus seminar or a Web-based tutorial, and pass a test showing mastery of the material.

The move toward certification "seems to be a real divergence from the goal of making I.R.B.'s more efficient with respect to humanities and social-science research," warns Jonathan Knight, associate secretary of the American Association of University Professors.

The concerns of social scientists are a priority of a new Health and Human Services advisory committee that met for the first time in December, Dr. Koski says -- even though only one social scientist, Felice J. Levine, executive officer of the American Sociological Association, is on the 17-member panel.

Not all social scientists are unequivocally critical of I.R.B.'s. Some respondents to last year's surveys said that the boards' supervision made them think more seriously about protecting the privacy of their research subjects.

At Illinois State, the angst that arose over Mr. O'Laughlin's thesis last summer has faded. He got his master's degree and his raise, and the history department honored his project as the best master's thesis of the year.

The initial uproar over his failure to obtain review-board approval led to fruitful talks between the department and the board members about what sorts of interviews historians perform, says Mr. Holsinger, the thesis adviser. And the I.R.B. has said that from now on, it will probably give most if not all of the department's oral-history projects expedited review.

"I think what had been an adversarial relationship has become a more collegial one," Mr. Holsinger says. "It forced us to sit down and try to figure what each other was doing."

Easy-to-print version

E-mail this article