Says who? Laws are murky when it comes to surrogate consent

University of California sets model consent policy

Researchers in California have a bit less to worry about when it comes to consent. As of Jan. 1, 2003, there is a specific law to cover who can provide surrogate consent for participation in medical research. The law, AB2328, places some restrictions on the types of sites and the kinds of participants for which surrogate consent is acceptable.

Hot on the heels of the law’s enactment, the University of California adopted a new policy on surrogate consent that may become a model for all institutions conducting human research.

California isn’t alone. Illinois, too, has a law regarding using surrogates to consent to research. Unlike the California law, however, in Illinois, friends or guardians of a prospective subject’s estate can provide consent.

The law there also limits the cases in which those who can’t consent for themselves can be part of a medical research program. Specifically, legal surrogates can consent on behalf of a subject only if these conditions are met:

  • The subject can benefit by participating in the research based on scientific information available and getting consent doesn’t do anything to infringe on the proposed subject’s rights.
  • A protocol has been approved for emergency use, and the subject may benefit from the research.
  • The subject has a terminal condition for which all other treatments and therapies have been unsuccessful, and an experimental program may or may not help them.
  • The protocol is a last resort to save the life of the subject, and there is no advance directive. Any such directive takes precedence over the surrogate’s decision.

At Winthrop University Hospital in Mineola, NY, Helen Panageas, IRB director, has been working on its policy for surrogate consent for the last year or so.

"We started by asking other institutions here in New York," she says, "but we found that almost everyone was lacking a developed policy."

Panageas says she spent a lot of time surfing the web and asking questions based on what she read. One problem she ran into was a state case: T.D., et al vs the New York State Office of Mental Health [91 N.Y.2d 860, 690 N.E.2d 1259, 668 N.Y.S.2d 153 (1997)].

In February 1995, the New York Supreme Court found that the office of mental health’s research regulations were "invalid and unenforceable" and that the commissioner in charge of the office did not have the authority to promulgate regulations concerning research using human subjects.

"In our state, that means that the state wouldn’t say what groups could provide surrogate consent for those who couldn’t consent for themselves," she says. At the time, most research institutions were using health care proxies or the DNR (do not resuscitate) advance directive hierarchy of spouse, adult child, parent, or adult sibling.

The case showed there was a problem, and although the state’s department of health put together an advisory group to look at the issue of surrogate consent, Panageas says the state has never implemented it.

Taking a wild guess

So, unlike in California and Illinois, institutions in New York are forced to guess what might be good enough, she says.

"I talked to one institution that said they aren’t doing any research using populations that can’t give their own consent, but I find that really hard to believe. Most, though, just cobble together a policy."

Through her months of research, Panageas found that in many cases, that cobbling involved developing policies that required close IRB scrutiny of any studies that would use these populations.

"They try to determine if the study could be done using another group of subjects," she says. "If so, then the investigators have to use that other group."

If a study must use those who can’t provide their own consent, then the IRB will allow the study to move forward, providing there is minimal risk, or greater than minimal risk but with potential direct benefit.

Next, the IRBs would determine whether the principal investigators should get the consent or whether that should fall to someone outside the study, says Panageas.

"The reason for that is they may have a really vested interest in the research and may be less stringent about getting consent," she says. "Not in a malicious way, but subconsciously."

In the end, Panageas formulated a policy that was based in part on the State University of New York (SUNY) Stony Brook policy. It can be viewed on-line in the university’s IRB policy handbook at

It enumerates the individuals who can provide surrogate consent as those who have legal authority to make decisions specifically about participating in research; family members in order of priority; and those named in health care proxies, although those are only for research protocols recognized by the medical community as offering the best treatment choice.

The policy gives the example of the latter as those with rare or aggressive cancers. In those cases, respected medical entities such as the National Cancer Institute recommend being part of research protocols as the best option.

If a patient has the ability to consent but may lose it, as with a dementia patient, the IRB at SUNY Stony Brook recommends that a formal surrogate be discussed with the subject early in the process.

The policy notes that the surrogates have to be told explicitly that they should make their decisions based not on their own beliefs, but reflecting what the subjects would think if they could make their own decisions. If that’s not possible, then the surrogates must consider only the best interests of the patients.

Even if the subject can’t consent initially, the policy states that the patient should be reassessed regularly throughout the study. If the capacity to consent is regained, then the patient should be given the information about the study as in an initial consent and be given the opportunity to quit the protocol.

Problems come from the top

One of the reasons the issue of surrogate consent is something of a hot topic is that there is little guidance from federal regulators. All the federal regulations state is that there must be a decision made by a legally authorized representative, "but what that is has to be defined at the state level. If it’s not, or if what the state decides is too prohibitive, then what do you do?" asks Panageas.

"What investigators have to remember is that they need to get consent from someone who can give it," says J. Mark Waxman, JD, general counsel at Care Group Healthcare System in Boston.

"But what federal law doesn’t specify is how you determine whether a specific adult is capable of being informed enough to give consent."

With vulnerable populations — children, pregnant women, prisoners, etc. — there are special provisions. "But when you have diminished capacity due to a disorder like Alzheimer’s, drug abuse, or some degenerative disease, they may have diminished capacity that is still good enough to make a decision," he adds.

Determining capability to give consent

The responsibility must fall on the investigator and the IRB to make that determination.

Waxman gives two examples from his system’s own experience. "Suppose you want to do a trial on a new drug for Alzheimer’s or dementia patients. The question comes up when the study is being presented to the IRB about how you determine whether people can give consent or not. There may be some standards that the IRB can impose that say a subject has good enough understanding," he says.

"There are tests that you can give. And you may find people who can determine if they wish to be operated on or not, but they can’t balance their checkbook any more. The IRB can determine what level of score potential subjects must achieve on these tests to participate, and at what level surrogate consent must be sought," Waxman explains.

Some samples of these tests that determine decision-making capacity are available at

Another example, he says, involves a study that only can be performed on people having heart attacks.

"If you say to a heart attack patient, Be in my study,’ would that person be able to make a rational decision? My gut instinct when this came up was no. But this kind of study can only be done on this class of people. In addition, people give consent to all sorts of procedures when they are in that condition.

"They say yes to procedures, medications, and even surgery. How is being in a study any different? If someone is informed enough to consent to having an operation, maybe that’s good enough," he adds.

Waxman says some areas of the federal government are making progress in coming up with hard-and-fast guidelines. The Veterans Health Administration came out with a new handbook on human subjects research in July that devotes an entire section to surrogate consent.

The new policy states that surrogate consent can come from a health care agent appointed by the patient in a durable power of attorney for health care; court-appointed guardians; or next of kin in this order: spouse, adult child, parent, adult sibling, grandparent, or adult sibling. Those are the only surrogate entities that can provide consent for research purposes in the VA. "I think it’s a pretty good list," Waxman adds.

Laws must not be pre-empted

The policy also outlines when surrogate consent can be accepted, and — just to muddy the waters — it notes that other federal, state, or local laws can’t be pre-empted by the policy.

There will always be murk in this topic, though, says Waxman. For example, if you are doing research with children, do you need consent from one parent or two? The federal regulations say one, but some states may require two. And in this day and age, what defines parent? If a grandparent has been raising her grandson as her own for a dozen years and is the only parent that child knows or has contact with, can the grandparent make decisions on whether the child should be in a research study? he points out.

Waxman says the best thing an IRB can do is bone up on the state and federal regulations. Reading something like the Belmont Report (available on-line at might help.

"You really just have to make sure you pay special attention when you are dealing with any vulnerable population," he says, whether they are those with diminished capacity, children, prisoners, or those who are economically and educationally disadvantaged.

Further Resources

Sample forms and policies:

Suggested readings:

• National Library of Medicine. Current Bibliographies in Medicine 99-3: Ethical Issues in Research Involving Human Participants and covers articles and books through 1998.

• Bramstedt KA. Questioning the decision-making capacity of surrogates. Intern Med J 2003; 33(5-6):257-259.

• Luce JM. California’s new law allowing surrogate consent for clinical research involving subjects with impaired decision-making capacity. Intensive Care Med 2003; 29(6):1,024-1,025.

• Luce JM. Is the concept of informed consent applicable to clinical research involving critically ill patients? Crit Care Med 2003; 31(3 Suppl):S153-160.

• Chen DT, Miller FG, Rosenstein DL. Enrolling decisionally impaired adults in clinical research. Med Care 2002; 40(9 Suppl):V20-9.

• Rich BA. The ethics of surrogate decision making. West J Med 2002; 176(2):127-129.

• Koppelman ER. Dementia and dignity: Toward a new method of surrogate decision making. J Med Philos 2002; 27(1):65-85.

• Stineman MG, Musick DW. Protection of human subjects with disability: Guidelines for research. Arch Phys Med Rehabil 2001; 82(12 Suppl 2):S9-14.

• Sugarman J, Cain C, Wallace R, et al. How proxies make decisions about research for patients with Alzheimer’s disease. J Am Geriatr Soc 2001; 49(8):1,110-1,119.