Picturing DNA
Chapter 3:
Genes and Privacy

Introduction

Chapter 1

Chapter 2

Chapter 4

Chapter 5

Chapter 6

Chapter 7

Epilogue

Family Sues Police; Son's Genetic "Propensity for Violence" is No Justification for Surveillance

******************************

HMO Announces Two-Track Pricing; Patients With Genes Linked to Reproductive and Colon Cancers Among Those to Pay More

******************************

State Requires Applicants for Drivers' Licenses to Submit DNA Sample; Test Using Swab of Tissue From Inside Mouth Called "Vital Aid" by Law Enforcement

You haven't seen such headlines yet, but it may not be long before you do. Researchers working with the completed map of the human genome are linking more and more sequences of human DNA to a wide variety of medical conditions and behavioral patterns. These links affect not only the person whose particular DNA is involved but also raise issues of legal and financial concern to society at large. Many people are understandably worried about what will happen when the interests of the state conflict with issues of personal privacy.

Questions are already beginning to arise as so-called DNA fingerprinting becomes an important tool for solving crime. The use of DNA databanks, in which DNA samples are taken not only from apprehended criminals but also from wide swaths of the law-abiding population, was pioneered by the police in Great Britain. Their first notable successes came in 1986 with exoneration of a suspected serial killer and the following year with the conviction of a rapist-killer; in both cases, DNA evidence collected at the crime scene and compared with tissue samples from the accused was decisive.

American law-enforcement agencies are eager to catch up. Many states have passed laws that require the taking of DNA samples, not only from convicted killers and sex offenders, but from several additional classes of convicted felons, as well as burglars,
drug dealers, arsonists and hit-and-run drivers. On the principle that recidivists commit a disproportionate amount of crimes, it someday should be much easier for police to identify perpetrators by comparing a sample of DNA evidence taken from the crime scene with data stored in the registry. New York City Police Commissioner Howard Safir has said, "DNA testing is a tool we are going to use not only to catch murderers and rapists, but also burglars and car thieves."

But the definitions of civil liberties and issues of privacy are different in Great Britain and the United States. In England, it has not been uncommon for the entire adult population of small towns and villages to agree to give samples for the national registry; few people complain that such data collection is an invasion of their privacy. However, in the United States, such cooperation would be highly unlikely. Alarms have already been sounded. Widespread DNA sampling would probably not be considered a violation of the Fifth Amendment on the analogy of fingerprinting, which is legal, but it might be challenged by lawyers citing the Thirteenth Amendment, which guarantees due process under the law.

Fingerprinting is an interesting antecedent. Sir Francis Galton, considered the founder
of the eugenics movement, was also the founder of our current system of fingerprint identification. In the 1890s, officials of the Bengal police in India devised and implemented the first feasible method for the regular use of prints to identify criminals, and their work was adopted by law enforcement agencies in England and the United States in the early twentieth century. The FBI has for many years maintained the world's largest collection of fingerprint records, which have been computerized for easy access by cooperating law enforcement agencies throughout the United States and elsewhere around the world. But even with skilled experts in charge, fingerprints are hard to find at many crime scenes. DNA samples, although often equally minute and tricky to capture, are much more decisive.

Thus the call for a national DNA registry to supplement the fingerprint records. However, reasonable voices have been raised to question its downside. They worry that once DNA samples have been entered into a national databank dedicated to the sole use of law-enforcement agencies, they might be accessed by anyone interested in knowing more about someone. As the head of the New York Civil Liberties Union, Norman Siegel, has warned, "We have to be vigilant. Social Security numbers were only to be used to identify pension accounts. Today they are a chief form of identification."

And what if DNA profiling were to be used in attempts to prevent, rather than solve a crime? Some psychobiologists claim that the tendency toward violence may be transmitted in our genes. Such researchers may attempt to tone down the forcefulness of their assertions and further studies may debunk their initial claim, but the public often continues to believe the first reports it heard. A team of Dutch and American scientists has reported on a Dutch family in which five generations of males had been unusually prone to aggressive outbursts, rape and arson. These men carried a genetic defect that made them deficient in an enzyme that regulates levels of serotonin in their brains. Serotonin is a neurotransmitter, a brain chemical that has a lot to do with mood, personality and behavior. Several genes are linked to the production of serotonin and to the way that brain cells utilize it, and it is possible to identify versions of the genes that produce less serotonin than other versions do. A geneticist in the Netherlands who studied the family warned that their genetic history could not be generalized to the population at large, but his warning was ignored. Stories in both the popular press and specialized journals reported the discovery of an "aggression gene."

The notion of a genetic propensity for violence has a long and lurid history. In the early 1950s, Maxwell Anderson's popular play The Bad Seed, which was also made into a movie, presented the melodramatic tale of a young girl with an inherited lack of moral sensitivity who causes a series of deaths among her close family and friends. Ironically, the play was written just at the time that James Watson and Francis Crick were teasing out the mysteries of the double helix. It prematurely supported the implication that biology, not environment, is the root of evil.

The idea that violence is rooted in biology stretches deep into the past. In the early twentieth century some biologists and psychologists made their names by using observations from the new science of genetics to explain the destructive behavioral traits of the masses of immigrants flooding into the nation's cities. They accepted without question the "truth" that poverty, alcoholism, prostitution and violent criminality arose primarily from a trait called feeblemindedness.

In an era in which Mendel's laws of inheritance were being rediscovered, it was all too easy to misapply the implications of his very careful analysis of the inheritance of single-gene traits. While Mendel meticulously limited his experiments and observations to data concerning four specifically defined traits of pea plants, many eugenicists claimed to trace the genetic transmission of human personality and behavior with the same certainty that they could track the inheritance of brown rather than blue eyes.

Henry Goddard, the leading American authority on the subject, believed that the feebleminded were a lesser form of humanity. "A vigorous animal organism of low intellect but strong physique-the wild man of today," was how he described them in 1914. Goddard and his peers thought that feeblemindedness occurred with unusually high frequency among the poor and especially among recent immigrants to the United States from eastern and southern Europe. Biologist Charles Davenport, one of the country's most prominent eugenicists, predicted that the "great influx of blood from Southeastern Europe" would rapidly make the American population "darker in pigmentation, smaller in stature, more mercurial. . . more given to crimes of larceny, kidnapping, assault, murder, rape, and sex-immorality."

These men and others like them generally dismissed the role of environment in shaping human behavior. The science of genetics was still in its infancy, and they did not know that an inherited tendency to any behavior, if it existed at all, would probably be the product of multiple genes acting together and triggered by environmental conditions. Although their simplistic predictions seem ridiculous to us today, blaming violence on biology has never gone completely out of fashion.

In the mid-1960s, a team of British researchers reported that a disproportionate number of men imprisoned in a Scottish hospital for violent prisoners had an extra Y chromosome in addition to the normal male complement of one X and one Y. Further research indicated that the double Y was irrelevant to violent behavior and, in fact, may have been linked to lowered intelligence, which led to a higher chance of running afoul of the law. Later studies showed that the incidence of imprisoned men with an extra Y was virtually the same as found in men on the street, but not before the so-called XYY syndrome was taken up by the press and by defense lawyers in some high-profile criminal cases. By the early 1970s, several sensational crime novels and at least two movies featured killers whose behavior was explained by an extra Y chromosome. As late as the 1990s, long after the original assertions had been disproved, references to XYY killers continued to show up in the popular press.

Today's attempts today to link violence to genes are more sophisticated and are applied to cases of individual behavior rather than to entire ethnic groups. Researchers link atypical levels of brain chemicals like serotonin and dopamine to impulsive violent behaviors, as well as to conditions of the brain that have no relationship to violence, such as Parkinson's disease. Abnormal levels of these neurotransmitters may be caused by genetically encoded traits, by response to some environmental stimulus or by the interaction of the two. Scientists have known for some time that people whose behavior is excessively violent and antisocial tend to have lower cholesterol than is the norm. Males whose cholesterol count ranks in the lowest quarter of those tested are four times as likely to commit violent acts against themselves or others than those in the highest quartile. The link between low cholesterol and violence probably has something to do with serotonin. Monkeys who are fed low-cholesterol diets show correspondingly low levels of serotonin. And there is evidence that the same correlation can be found in people.

But does evidence that people who have brain-chemistry irregularities are "bad seeds?" Is it reasonable to consider giving them serotonin or cholesterol supplements or to mark them as violence-prone and therefore subject to preventive surveillance?

In addition to ethical and social qualms, we now have scientific evidence that suggests the answer is "no." It turns out that although serotonin production is dependent on genetic factors, it is also responsive to social cues. Social status and stress, in particular, have a strong influence on how much serotonin is present in the human brain, regardless of genes. It seems that while built-in genetic propensities may influence violent behavior, social and environment factors are just as important. The kind of plant that grows from a bad seed or a good one is influenced by the soil, the weather, fertilizer and other nongenetic conditions. And in human behavior, as in plants, nature and nurture each play an important role. "It's not the genes that cause violence in our society," says Paul Billings, a clinical geneticist at Stanford University. "It's our social system." There is far more evidence that violent individuals are produced by violent families and violent neighborhoods than by any genetic inheritance uncovered to date.

Nonetheless, scientists are hopeful that the new behavioral genetics will lead to the kinds of therapies and cures for aberrant psychosocial behaviors that medical geneticists hope to discover for physical diseases. Huge pitfalls impede research, however. It is almost impossible to eliminate or control for environmental influences. Organizing a study in which a control group of families can be compared to the subjects of the research is tricky. Also, many research projects have defined the traits they wish to study with astonishing imprecision. Studies have been conducted that claim to have found genes that create the propensity to crave thrills, to have leadership qualities, to be unhappy, to divorce, and to wear a lot of rings.

If it does become possible to use genetic information to predict an individual's tendency to violence, there are grave questions concerning how such information should be handled in a democratic society. Does labeling a child as prone to violence influence that child to live up to those expectations? Is the labeled child's behavior affected by the way families, neighbors, teachers and law-enforcement officials treat him or her? And most of all, defenders of privacy wonder, should government agencies or the communities we live in have any right to know anything about our genetic propensity for any behavior?

DNA typing can be developed to tell individuals about many of their propensities and potentialities. That information would be available to any institution or individual with access to the testing agency's database. Schools, employers, insurance agencies and health maintenance organizations might insist on their right to know your DNA profile.
So might the motor vehicle bureau, the police, sports teams, the military and any other organization that would profit from this knowledge.

Q & A with Dr. David Baltimore

Q: Would your DNA ever be sampled without you knowing it?

Baltimore: No. In principal no. But even if you give blood, you're giving DNA.

Q: Right. Does anybody ever do anything with it?

Baltimore: Well, no. Your doctor orders tests. If your doctor doesn't order any genetic tests, then no genetic test will be done. They can't do a test on you, on your blood sample that leads to you, which the doctor doesn't order. Now, there are situations where people get samples of blood from somewhere, as just numbers: number one, number two, number three, number four. And they are looking, for instance, at a population based survey of a gene or something. But if there is no linkage to the individuals who gave blood, then there is no danger of compromise of privacy. So you can do that. But if you need to know the names of the people who gave the sample, then there are potential privacy issues. For instance, if you want to know if they had breast cancer or didn't have breast cancer, that's not a problem. But if you do need to know the diagnosis, you need a medical work-up for the individual. You need very careful controls, and then the people have to give permission to be part of the study.

The most compelling claims for access to genetic information come from researchers doing broad-based longitudinal health studies. Surely in those cases, it would be possible to mask the identity of individuals whose DNA data is included. The model they point to is census data, which is used in thousands of statistical records without the privacy of individual citizens being invaded.

But many people who are protective of their privacy feel that their medical records in particular should not be shared with anyone other than their immediate health-care providers; they worry about the risks that identifiable intimate information can be passed on to others without their approval. Genetic information about an individual's susceptibility to certain diseases could lead health insurers to charge differential rates or cause HMOs to deny membership to those likely to consume more care.

Between 5 percent and 10 percent of the two hundred thousand women who will be diagnosed with breast cancer each year have hereditary forms of the disease, the most common of which are linked to mutations in two genes, called BRCA1 and

BRCA2. These mutations also put women at an increased risk of ovarian cancer and link men to increased risk of breast and prostate cancers.

In families with a cluster of breast cancer cases-particularly those with an Eastern-European Jewish heritage, among whom the incidence of BRCA1 and BRCA2 is higher than that of the general population-it is possible for doctors to determine if the victims share an altered form of one of the genes. If they do, they are said to have a "known mutation." Other women in the family can then elect to be tested to see if they have the mutation as well. If so, their risk of developing breast or ovarian cancer is certainly higher than the national average, though it is by no means certain that the gene will lead to cancer in any one given case.

Many women whose family members have a high incidence of breast or ovarian cancer would welcome being tested. If they discover that they do not carry the mutation, their chances of developing reproductive cancers drop to the national average. If they do have the mutations, they may modify their lifestyle by cutting back on alcohol consumption, for example, in hopes of improving their odds. They may have more frequent mammograms in hopes of finding the cancer at a very early stage, or they may decide to have radical prophylactic surgery.

Or they may decide that they can best protect their peace of mind and greater well being by not being tested at all. As the National Cancer Institute concedes, "The inclusion of genetic test results in a patient's records may have serious implications." The patient may be asked to sign forms that give access to her medical records to her insurance company or HMO; if so, she may be denied medical insurance or forced to pay a higher rate. If her doctor agrees to keep the results of a test out of her medical records, she can miss out on insurance coverage for prophylactic surgery, for example, or conceivably be accused of fraud should cancer ever develop. Patients and doctors alike are torn about how to handle both the practical and the emotion implications of this dilemma.

A different ethical consideration faces doctors who must decided how to advise a high-risk patient if there is no treatment or preventive therapy available. Some people know that the propensity for Huntington's disease (a debilitating and fatal neurological condition often associated with songwriter and singer Woody Guthrie, its most famous victim) or some other terrible affliction lurks in their family tree. Nonetheless, they refuse to be tested for the genes that put them at risk; they'd rather live with uncertainty than hear a death sentence from which present-day science offers no reprieve. Nancy Wexler, who organized much of the research into a cluster of Venezuelan victims, and her sister, Alice Wexler, who has written extensively about the disease, are the daughters of a woman who died of Huntington's. The sisters therefore have a 50-50 chance of coming down with the disease. Yet both have decided not to be tested.

Nancy has spent more than twenty years studying and befriending the high-risk Venezuelans, all of them descendants of an Amerindian woman who lived in the early 1800s. Some of them come from families that have intermarried so extensively with
people of various heritages that their risk of developing Huntington's is low. But others have intermarried to the degree that their chances are much higher. Genetic testing can determine if a person carries the dreaded mutation on chromosome 4 that predicts their fate.

But what are the ethical considerations for doctors who can offer the test but cannot, at the moment, promise anything other than palliative treatment to a victim of the disease? Imagine the burden of someone who discovers through testing that he or she carries the mutation. Sooner or later he or she will become ill, placing a terrible burden on the rest of the family. If already a parent, he or she also has the grievous burden of knowing the chances are high that one or more of the children in the family have risk of illness as well.

Wexler and her colleagues recognized that significant genetic counseling must be made available to anyone being tested for the telltale genetic markers for Huntington's and that psychiatric counseling is necessary as well. They have also decided that the test should only be given to people over the age of 18 who have signed consent forms. But while well-intentioned, as Wexler is the first to admit, these precautions are only stopgap measures that do not begin to solve all the ethical dilemmas posed by testing for serious hereditary illness without a cure in sight.

The growing number of tests that predict a person's risk of contracting serious diseases is of interest to employers. Today, they commonly ask prospective employees to take personality or intelligence tests; in the future they might well ask for DNA samples. Some employers already use genetic screening to identify, and therefore not hire, people who might be especially sensitive to chemicals used in the workplace. They insist this screening is justified because it protects the person at risk from serious illness; of course it also eliminates the company's future liability for insurance costs or compensation claims. It puts the burden on the worker and lessens the imperative for the company to pay the costs of making the workplace safer for everyone, if that is even possible.

In the future, companies might decide to enlarge the list of genetic predispositions that they consider undesirable in their work force. They might identify conditions that could drive up the company's health insurance premiums. They might also decide that certain genes that indicate potential personality patterns might make people more or less useful employees. In response to this problem, President Clinton issued an order in February 2000 prohibiting federal agencies from using genetic information in any decision to hire, promote or dismiss workers. The order protects the nearly three million people who work for the federal government. And the president has asked Congress to pass legislation expanding some protection to the private sector. Such legislation has been introduced in Congress and in several state legislatures, but the debate over its adoption will be bitter and protracted.

The Americans with Disabilities Act, in effect since 1992, says that employers are breaking the law if they disqualify anyone from a job unless the person is actually unable to perform its duties. No one expects that a paraplegic has the right to be a firefighter or a windowwasher. But if otherwise qualified, that same paraplegic cannot currently be denied a job as an office worker, and the employer has the responsibility to adapt the workplace to accommodate special needs. The same legislation would presumably modify the worst-case scenario of an employer's refusing to hire someone of the basis of a DNA test. However, there's nothing in the law that prevents insurance companies from limiting workplace coverage.

Q & A with Dr. David Baltimore

Baltimore: If insurance companies insist on your having a given kind of test as a requirement for getting insurance-perhaps they do that with HIV testing-you have to prove you're HIV negative in order to get certain insurance coverages under certain situations. I don't know where that is today. In fact, probably it's just done routinely, but I don't know.

Q. But it wouldn't be done by just a computer search of hospital records?

Baltimore: But there the insurance company just makes it a requirement. Insurance companies can do the same thing for the BRAC 1 and 2 testing. The problem is only if an insurance company makes it a given that you have such a test in order to get insurance. There won't be insurance unless they're assured that you don't carry genes that give you a predilection to breast cancer or a bunch of other things. That's what people are worried about, and rightly so.

Saying that someone has a genetic predisposition for something, except in the very rare case of single-gene conditions, is actually a way of indicating the odds of something happening. Yet the calculation of these odds is imperfect as best, especially when the factors of environment, which increase or decrease those odds, cannot be measured. But many people lose sight of this important fact. Even scientists, in their zeal to gain attention for the results of their work, sometimes underestimate the consequences of this general misunderstanding. The burden falls on people who carry the stigma of "scientific" evidence; they risk being barred from education or work.

Most Americans give a high priority to privacy. Some hang up on telephone public-opinion polls and refuse to fill out census forms, thinking that by doing so they are striking a blow for privacy. In the early 1990, a survey by the March of Dimes revealed that most people once believed that genetic information was somehow different. They divided the right to know into categories, and said that information should be available to prospective spouses, other family members, insurers and employers. But that was before the revolution in medicine that put people into large HMOs whose priority was to eliminate all possible risk. The public's willingness to chuck privacy in favor of what was seen as the public good is changing. More recent polls suggest that public debate has begun to shape the issue as the private citizen versus the combined forces of government, big business and neighborhood gossips. While not everyone understands completely the distinction between the predictability of risk and the inevitability of result, many people are more protective of their privacy than ever before. In the age of the Internet, of anonymous credit checks, of private companies that can provide anyone who pays a fee with almost any information they want about you, that may not be a bad idea.

Return to Table of Contents Return to Previous Chapter Go to Next Chapter

To see an interview with davidkremers, click here

Picturing DNA by Bettyann Holtzmann Kevles & Marilyn Nissenson
Copyright © 2000 Bettyann Holtzmann Kevles & Marilyn Nissenson
All Rights Reserved