COLUMN ONE : Judging the Risks of Infection : The AIDS epidemic is making patients fearful of being infected by health-care workers. But some say the odds are far less that one will be hit by lightning.
A 9-year-old girl materialized on Stephen Kindland’s television screen one day last month. She had just learned that her dentist was dying of AIDS. She was sobbing, terrified that she, too, would come down with the disease.
The sight disturbed Kindland, a Florida health official. The girl’s risk of having been infected by the ailing dentist is estimated at one in 263,000 to 2.6 million--far less than her risk of dying in a car crash or of being killed by lightning.
“It’s like the Florida lottery,” Kindland said recently of the public’s fear of infection by health-care workers. “People believe when they pick those numbers, they can win. So it stands to reason: If they can be that lucky, they can be that unlucky.”
The extraordinary wave of fear that has swept the country in recent weeks over the possibility of catching the human immunodeficiency virus from a doctor is shedding light on the complex and often baffling ways Americans perceive risks.
Much of what is known about so-called “risk perception” has emerged from research that has used the tools of psychology, sociology and anthropology to explore why people panic at lesser risks while seeming to ignore much greater ones.
Their findings include the following:
* Unusual and unknown risks are more terrifying than familiar ones, though everyday risks claim more lives. A jetliner crash killing 200 people triggers widespread anxiety, while Americans have become blase about 50,000 deaths every year on the roads.
* Risks undertaken voluntarily seem more tolerable and controllable than lesser risks imposed from outside. People are more likely to accept risks that benefit them or give them pleasure, such as the risks of alcohol or fatty foods.
* Many people have difficulty understanding probability. They do not appreciate the difference between a risk of one in 1 million and one in 100,000. So they worry more about dying in a tornado than dying from a stroke.
* The social implications of hazards may shape how they are perceived. An accident in a familiar setting may go barely noticed, while a small accident in an unfamiliar setting--say, a genetic engineering lab--may provoke panic if it seems to set a precedent.
* Experts and lay people value risks differently. Experts count lives lost and therefore emphasize risks such as smoking. The general public looks at many other factors, such as controllability, and worries about terrorism, contaminated Tylenol and nuclear power.
* Once people make a decision about the size of a risk, their minds are difficult to change. People tend to overlook conflicting evidence. If the degree of risk is unclear, they take the uncertainty as vindication of their own position.
For all those reasons, the public’s sense of the severity of a risk may be based less on how many people might be killed--and more on factors such as who wins or loses, who calls the shots and whether future generations might be hurt.
Baruch Fischhoff, a psychologist at Carnegie Mellon University in Pittsburgh and a researcher in the field, believes that the risk itself often becomes peripheral to other issues such as fairness and who holds power in a society.
For example, people who believe that a new technology has been foisted upon them may be inclined to overstate its risks, researchers suggest. People who stand to benefit from the same technology tend to do the opposite: They understate the hazard.
“People judge risks according to a lot of things other than how likely they are to kill them,” said Peter Sandman, director of the environmental communication research program at Rutgers University. “One thing people care a lot about is dread.”
The risk of HIV infection by a health-care worker surfaced publicly 13 months ago when the U.S. Centers for Disease Control reported that a Florida dentist, Dr. David Acer, appeared to have infected a young woman, Kimberly Bergalis, during a tooth extraction.
Since that time, the CDC has identified four other patients apparently infected in Acer’s office. It is unclear how the patients were infected--whether directly by Acer, who was infected, or by instruments contaminated with blood from HIV-infected patients.
The Acer case is the only instance in which a health-care worker is believed to have spread HIV to patients--compared to as many as 40 reported instances in which the infection may have gone the other way.
The risk of HIV infection by a health-care worker is ranked well below most hazards that patients face--including the risk of a hepatitis B infection, which could prove fatal, and the risk of death from a reaction to penicillin or anesthesia.
Yet the public reaction has been fierce.
A CDC hot line is fielding 1,000 calls a week on the AIDS issue. Americans now tell pollsters that they want mandatory AIDS testing for doctors. The U.S. Senate voted last month in favor of imprisoning certain infected workers, though the House of Representatives has not taken up such a proposal.
The Senate bill, sponsored by Sen. Jesse Helms (R-N.C.), would impose prison terms of at least 10 years and fines of up to $10,000 on all infected health workers who perform invasive procedures without revealing their condition to patients.
Another bill, also passed by the Senate, calls for testing of all health professionals doing invasive procedures. It would ban infected workers from doing such procedures without the approval of a panel of experts and without informing their patients.
“When it comes to bizarre or mysterious things, particularly abominable things, zero seems to be the only acceptable level of risk,” said Fred Kroger, director of the national AIDS information and education program at the CDC.
The health-care worker issue has become one of the most controversial and troubling in the field of public health, seen by some as pitting the rights of patients to information about their doctors against health-care workers’ privacy and civil rights.
Proponents of the Senate bills argue that patients have a right to make an informed decision about whether to entrust themselves to an HIV-infected health-care worker. No matter how small the risk, they say they should have the right to choose.
Many critics acknowledge that there may be some high-risk procedures that infected physicians should agree not to do. But they say mandatory testing is unworkable and expensive, and the virus is best controlled in health care through rigorous hygiene.
“Each test will cost $50 to $100,” said Dr. M. Roy Schwarz of the American Medical Assn. “We’ve got 600,000 physicians . . . and how often would you have to test to assure the public? The answer, as far as I’m concerned, is every day.”
Meanwhile, Acer has died of AIDS. Bergalis is close to death. But she has waged a powerful, national campaign to raise awareness of the risk of infection by a health-care worker in hopes of ensuring that no other patient will ever face her fate.
“Today, there is no spokesman on the AIDS issue that’s been listened to more carefully than Bergalis and her family,” said one official who asked not to be named. “It’s the way the spotlight sometimes shines in odd corners, distorting the importance of the story.
“It’s an incredibly compelling story, everybody’s favorite daughter, the daughter you would love to have,” the official said. “And now she’s dying this tragic, terrible death through no fault of her own. Who do you blame? She blamed the entire health-care profession.”
It is impossible to calculate the precise risk of infection by a health-care worker. The CDC has developed estimates--based on what is known about the transmissibility of HIV and on the number of health professionals believed to be infected.
The CDC calculates that the risk of a patient being infected through contact with an infected surgeon falls somewhere between one in 41,667 and one in 416,667. The risk of infection by an infected dentist falls between one in 263,000 and one in 2.6 million.
By comparison, the risk of death in a car accident has been estimated at one in 4,000 for a person who drives 10,000 miles a year. The risk of death by lightning is one in 30,000 in a person’s lifetime, one in 2 million in a single year.
Even in health care, most risks are greater than that of AIDS, the CDC contends.
The hepatitis B virus is more common among health-care workers. It is also more easily transmitted than HIV. Studies show that people exposed to blood tainted with the hepatitis B virus are more likely to become infected than people exposed to HIV-infected blood.
For those reasons, the CDC estimates that the risk of an infected surgeon spreading hepatitis B to a patient at 100 to 1,000 times greater than the risk of infection with HIV.
Similarly, the risk of death from anesthesia is four to 40 times greater than the risk of HIV infection. The risk of death from a reaction to penicillin is at least equal to, and perhaps four times greater than, the HIV risk.
“HIV is way down on the list,” said Elizabeth Cooper, staff counsel to the American Civil Liberties Union’s AIDS project. “(And CDC officials) don’t talk about (health-care worker) negligence, incompetence, fatigue and drug and alcohol addiction problems.”
A number of factors, experts say, conspire to make the threat of infection by a health-care worker seem so potent--the unusual nature of the hazard, the intense publicity, the nature of the doctor-patient relationship, and the fact that risk assessment is an imperfect science.
Frank Farley, a professor of psychology at the University of Wisconsin, said a major factor in how a risk is perceived is how it stands out against the background of everyday risk. If it is unusual, people tend to overrate it.
For that reason, the rare plane crash that kills 200 people stands in sharp relief against the tens of thousands of uneventful flights--and against the many dozens of fatal motor vehicle accidents that occur every day and go barely noticed.
“Often, you can take one case and people will use it as a rallying cry, when it may not reflect the truth,” said Farley, an authority on risk-taking behavior. “It’s hard to know when the generalization really should stop.”
In the case of infected health-care workers, the risk stands out in part because of media attention. Psychologist Paul Slovic calls the phenomenon an “availability bias”: By making a hazard easy to imagine or recall, publicity makes it seem more of a threat than it is.
So the public tends to overestimate the risk of tornadoes while underestimating the risk of asthma and diabetes, Slovic and others say. People think death by fire is more common than death by drowning, though their frequency is about the same.
Of the Bergalis case, Slovic said: “It stands out in one’s mind. You don’t have any knowledge of how many thousands of other patients have been treated by HIV-infected practitioners who have not been infected.”
People are also outraged by secrecy, argues Sandman, the Rutgers University specialist in risk perception and risk communication. He and others say people suspect the motives of health officials who hide their HIV status from their patients.
“There’s something very deep in our psyches, going back to (the practice of equipping) lepers with bells, that says that someone who is sick and knowingly infects others is committing a particularly awful evil,” said Sandman. “That’s very deep in our culture.
“And if you are my physician, whose relationship is one of power in your hands and trust on my part, what more devastating abuse of power could there be than a doctor knowing he might endanger the health of a patient and not tell his patient?” Sandman asked.
Added to all that is the public’s dread of AIDS and the stigma attached, the lack of scientific certainty in assessing AIDS risk and risk in general, and the suspicion, right or wrong, that officials in the past underestimated the risk of infection from tainted blood.
“So there are long lists of outrage factors,” said Sandman. “It means that the public is likely to ignore the data and respond to the outrage. People are going to have a tough time responding rationally to the data because of their response to other things.”
As a result, public outrage may prompt inappropriate policies, where the human and economic costs may exceed what is warranted by the risk. Critics cite as an example the banning of all sales of Chilean fruit in 1989 after one incident involving tainted grapes.
In that case, the U.S. government banned the importation of Chilean fruit after the discovery of two grapes that had been injected with cyanide--only to back down several weeks later after no other poisoned fruit was found and after Chilean officials complained that their country’s economy would be ruined.
How then might appropriate policy-making occur?
Many medical professionals would like to see the health-care worker issue left to the CDC, an agency well versed in epidemiology and the science of risk. The CDC reviewed the issue for many months before coming out last month against mandatory testing.
CDC officials recommended voluntary testing and said infected health-care workers should stop doing invasive procedures unless they have informed their patients and received approval from a panel of doctors.
“We have wasted far too much time and money on attempting to reduce a risk that is infinitesimal to begin with, and have spent not nearly enough resources in trying to stem the epidemic where it really is a problem,” said Cooper of the ACLU.
Sandman argues that “this society lives on information, not secrecy.” He contends that people have a “right to know” a health-care worker’s HIV status, just as communities have a right to know what hazardous materials are stored in their neighborhoods.
Health-care workers who balk “want different standards applied to (them) than to the chemical industry,” Sandman said. “What they say is the same (thing) the chemical industry says--that people would misunderstand and the data aren’t foolproof.”
Only full disclosure will dissipate public outrage, Sandman believes. The medical profession, he argues, must be open about how many health-care professionals are infected and must determine with greater certainty the risk of transmission to patients.
He would also like to see other information made public, such as doctors’ hepatitis B infection status and the patient death rates of individual surgeons doing specific operations--”which is a . . . lot more serious information than HIV status.”
“What they need to do is take the outrage seriously,” said Sandman. “The public will then learn to assess the hazard rationally. Can we do that with AIDS? Not next week. Particular technologies and risks become martyred, stigmatized. It becomes a long road back.”
