A personal Homepage

The Coming Collision:

Medical Insurance and the Genome

Written Spring 2000


Each day new genes are discovered around the world in a coordinated international program to sequence the entire human genome.  This program officially began in 1990 and was inspired in part by some of the very men who discovered the structure of DNA in the 1950’s.  This project, formally known as the Human Genome Project, has been gaining recently even more momentum and will near completion in 2003. The Project is an enormous undertaking in that close to 3 billion pairs of nucleotides need to be chemically sequenced.  Its goals are fivefold.  The Project first seeks to identify all of the approximately 100,000 genes coded in the human DNA and the sequences of the 3 billion chemical bases that code these genes as well as comprise the remaining portion of the DNA.  The Project seeks to also store the information, develop tools for analysis, and address the legal, ethical, and social issues which might arise as a consequence of the research (ELSI of Genome Project).  It is this last goal that has for good reason garnered much contemplation recently.  Contrary to the general workings of a large government program, this project will actually be completed ahead of schedule.  Initial plans outlined a fifteen-year effort, but rapid technological advances have accelerated the expected completion date to 2003 (ELSI of Genome Project).  Scientific advancement, as of late, seems to be racing ahead with no signs of relenting in the near future.  Accordingly, examining consequences of future hypothetical research and discovery should be started now.  The future is closer than we think.

            This oncoming revolution in genetic understanding is removing many levels of ignorance about an individual’s susceptibility to future illness.  An individual’s genetic blueprint will in many cases provide a clear representation of future health.  This information may be a welcome relief to some or upsetting for others, but regardless, its coming reality is unavoidable.  What is also unavoidable is that genetic information will influence the health insurance industry.  As will soon become evident, in this industry, ignorance is bliss.  It is an often-overlooked fact the insurance industry in its entirety is based on ignorance.  This is not limited to health insurance.  If someone were to look into a crystal ball and find that they would never be involved in a car accident, they would have no need for car insurance.  A person’s genetic makeup could be considered a biological crystal ball composed of the microscopic fibers of DNA.  The picture of future health will become progressively clearer as we wipe away the layers that have occluded our view.

            The system of healthcare financing that has evolved in United States is unique in the western world.  The word “evolution” is a good term to describe how our present insurance system has come to be.  World War II laws, which limited wage increases, but did not regulate benefits, were left in place immediately following the war.  To attract workers, employers offered health insurance, which at that time was relatively inexpensive.  Due to rapid technological advance and hence increased costs in medicine, health benefits have become an increasingly larger portion of an employee’s compensation.  Because healthcare does cost so much, insurance is essentially required to gain access to a hospital.  Thus, although employers cringe at the costs of providing health insurance to their employees, it is increasingly important to their employees that the insurance is provided.  Thus the conflict between providers and those covered is growing more tense.  It just might be the case that expanding amounts of genetic information will dislodge this conflict and move health insurance in a new direction. 

The current health insurance landscape is comprised of a patchwork of plans.  These plans rely on both public and private funds to accomplish the goal of insuring Americans.  Approximately 150 million people are covered by employer-provided private group insurance.  About 56 million are covered by the government’s Medicare and Medicaid, and 15 million rely on self-purchased policies.  As one quick to compute the figures realizes, this leaves over 40 million uninsured (Greely 256).  This significant number of uninsured raises doubts as to the current method employed to accomplish the goal of providing health insurance to Americans.  As progress is made in the art and science of medical prediction, this patchwork of plans comprising the current system of healthcare financing needs to undergo significant change.

Scrutinizing effects of genetic screening on self-purchased plans offers a clear representation of the imminent upheaval of the health insurance industry.  Typically, people who independently seek to purchase health insurance are self-employed.  These persons need to submit applications, which are carefully analyzed by the prospective insurers.  The method of insurer review of applications is based on a process of medical underwriting (Hudson et al. 392).  Underwriting uses medical history as well as individual risk factors, such as smoking, to determine whether to provide coverage and under what terms.  Sometimes visits to an insurance company doctor may be required.  Currently about 30 percent of all applicants are denied coverage (Greely 256).  Small percentages of applicants are accepted but with higher premiums or exemptions of certain pre-existing conditions.  Genetic information on the part of the insurance company would have obvious consequences here.  People known to be susceptible to certain illnesses due to genetic information will be denied coverage or have coverage which might term the susceptible disease a pre-existing condition.  This then defeats the purpose of health insurance. 

            A number of states have enacted laws, which restrict insurance companies in the obtaining, and using genetic information.  The state laws specifically “prohibit insurers from denying coverage on the basis of genetic test results, and prohibit the use of this information to establish premiums, charge differential rates, or limit benefits” (Hudson et al. 393).  Although the laws are politically popular, several factors exist which will limit their effectiveness.  First, the laws do not apply to self-funded plans due to the Employee Retirement Income Security Act (ERISA).  Second, the state laws do not prevent insurance companies from using genetic information generated by a family history, physical exam, or medical record.  To properly protect the consumer against genetic discrimination, it is necessary that the laws be broadened to include information about genes, gene products, or any inherited characteristics.  But one must ask where the line is to be drawn—everything has its basis in genetics.  Even if insurance companies are limited in obtaining genetic information, correlating phenotypic characteristics to genotype is becoming more feasible.  Consequently, the blurring of the line between genetic information and genetic manifestation will make targeting legislation difficult and enforcing it impossible.

Although medical underwriting superficially seems a rather discriminatory mechanism devised by insurance companies to increase profits, in actuality it is a necessity for their survival.  Underwriting is a reaction to a problem known as adverse selection (Hudson et al. 392).  Adverse selection can be explained, “all other things being equal, people who know they are at high risk will be more likely to seek insurance” (Greely 266).  Individuals may not elect to purchase health insurance until they are already ill or predict needing future health care.  With the advent of genetic screening, the anticipated need of healthcare will be become a virtual certainty.  The Health Insurance Association of America agues that “it is unfair to insurance providers if individuals are privy to information about their health conditions that might cause them to buy added coverage.”  This situation “‘could create a great deal of havoc in the industry, causing costs to go up and fewer people ultimately being able to afford coverage’” (Shulman 17).    If use of a genetic test became widespread and an insurer did not have access to, or use these results to exclude applicants, the insurer would end up paying large sums of money due to coverage of individuals with the condition.  If this could be applied to most medical conditions, the consequence would be the collapse of the entire insurance market.  Only individuals who know they are at risk for disease and whose costs would be higher than the premiums, would purchase health insurance.  This of course would raise premiums across the board.  Again customers would reevaluate premium cost to predicted healthcare costs and some would discontinue coverage.  This viscous cycle would repeat until those left would be paying premiums equaling the cost of their care.  What then is the benefit in owning a health insurance policy?

The picture grows more complex and worrisome when private group health insurance is considered.  The United States is the only country in the world that relies on individual employers to be responsible for providing health insurance to the majority of its citizens (Greely 267).  Employers and insurance companies have accomplished this through several slightly different varieties of group health insurance.  Employers can either buy a group policy from a health insurance company or can assume employee healthcare risk themselves by providing coverage.  The latter of the two is the leading method for providing group coverage, while initially the former approach was more common.  A gradual refinement of cost analyses or rating methods employed by insurance companies is responsible for this change.  In its purest form, community-rating insurers charge employers premiums that correspond to standard employee health costs in the region (Insurance Market Reforms).  One might be quick to point out that health costs are by no means equal for every company.  This has led to experience-rating which charges different rates to different employers (Insurance Market Reforms).  Healthcare costs from previous years are figured when calculating a premium.  As more employer healthcare cost information is gathered, competition will favor experience-rating.  Much like adverse selection mentioned above, those companies which have higher than average costs will flock to the community-rating firms.  These firms will then be forced to raise rates to be more consistent with actual paid benefits, beginning to resemble experience-rating until experience-rating becomes the only plausible alternative.

            One might wonder why, as experience-rating becomes more exact and premiums more closely match actual health costs, employers do not self-insure?  This has in fact happened and was given a significant boost from the ERISA.  This Act exempted self-insured employers from state insurance laws, freeing them from insurance premiums and state laws mandating specific coverage (Greely 269).  Since employers are financing coverage for their employees, every dollar spent on healthcare by an employee would be coming directly from company profits.  In effect, employers have become small-scale insurance companies.  When applying for positions in these companies, applicants are in actuality applying for both employment and health insurance coverage.  When making hiring decisions, employers realize that healthcare costs will affect future employee compensation.  The principle of adverse selection, which was applied to analyzing insurance companies, can be applied here.  If a certain company provides full coverage with little prior applicant health screening, a higher percentage of individuals in poor health might apply there.  This company would bear disproportionately greater healthcare costs and become less competitive.  To remain competitive and profitable, employers must consider an employee’s health and associated costs.  If genetic testing becomes available and widely used, it is evident that those with high susceptibilities will have trouble in both buying insurance and securing employment.  A possibility exists that these conditions might be excluded in a coverage agreement with an employer, still leaving large future medical bills unpaid.   

A general trend is evident when considering the evolution of health insurance.  This trend involves the continual refinement in the art of projecting future health costs.  Projection has moved from regional cost averages, to employer-based estimates, and now encroaches on individual employee future cost analyses.  There is little reason the refinement process will not seek to include genetic information, as it becomes available.  Currently there has been little evidence besides a few anecdotes that employers are using genetic testing in discrimination of future employees.  However, in a survey performed of 330 Fortune 500 companies, half found the idea of genetic testing acceptable (Wingerson 272).  Employees are more than anything else, fearful of the prospect of increased genetic testing.  In a 1997 national survey conducted by National Center for Genome Resources, two-thirds of respondents would not take a genetic test if employers and health insurers could see the results.  And, an understandable majority felt that these groups should be prohibited from obtaining information about an individual’s genetic condition (Miller 190).    

            There is currently a wide spectrum covering the degree to which genetic knowledge can be integrated into disease prediction.  On one end of the spectrum are diseases like Huntington’s chorea.  In these diseases, genotype is the primary factor influencing disease expression.  Having genetic traits of this category leads to virtually certain expression of the disease.  Fortunately there are currently few known genetic factors which correlate so consistently with disease expression.  This may be due to the fact that either there are in actuality few genetic diseases of this type, or that sufficient knowledge is yet available to make specific correlations.  It is probably the case that both these assertions are to some extent correct.  Evolution will tend to allow deleterious genes to be excluded from the genome (Huntington’s has escaped this fate because it is expressed after prime reproductive age).  However, most traits are expressed by a concert of multiple genes.  Evolutionary eradication of deleterious combinations is a more lengthy process.  It is also the case that determining harmful combinations of genes is a more intricate puzzle for geneticists to solve.  Discovering that a single dominant allele is responsible for Huntington’s would not compare to the elucidation of those genetic factors coding a predisposition to contracting heart disease.  After the entire human genome is sequenced, it will no doubt be easier for geneticists to correlate genetic data to disease—the blueprint will be in place.  All one needs to do is look for unique modifications to an individual’s genetic sequence. 

Diseases or conditions, which are believed to have little or no genetic link, fall in the middle of predictability spectrum.  As was previously noted, as more genetic information is gained, these diseases or conditions will make a slow slide to the extreme end.  There are also some conditions that in all likelihood will remain near the genetically unpredictable end of the spectrum.  These include viruses like HIV and polio, routine Streptococcus infections, or broken bones and pulled muscles.  Even at this far end, however, genetic contributions cannot be discounted.  Immune defense to pathogens affects if one will contract the common cold or a strep infection upon exposure.  Bone density and muscle flexibility are also genetically influenced factors, preventing even the complete exclusion of a broken bone from a genetic susceptibility screening.

Although the threat of genetic testing and its implications in the insurance market seem very real, some might argue that methods today are available which offer similar opportunities for predicting future health.  Moreover, these determining factors have not led to the collapse of the health insurance market.  As previously noted, insurance applications now ask a variety of questions in an attempt to predict healthcare costs.  Smoking is a good example of a non-genetic factor that can profoundly influence future health.  All the most common diseases resulting in death have smoking as a risk factor, including heart disease and cancer.  Although smokers do pay somewhat higher rates, they are not uniformly excluded from health insurance plans or employment.  Under close examination, a person’s smoking habit is categorically different from genetic factors.  Smoking a cigarette is an active decision made by an individual, which (for purposes of argument) can be discontinued at any time. Cigarette smoking is one of many possible environmental factors contributing to a related disease.  Genetic makeup is an uncontrollable factor determined during conception. Environmental factors are in constant interplay with a person’s genes leading to the manifestation of a unique phenotype.  Future health predictions always weigh environmental factors but generalize their specific influence due to lack of genetic information.  For example, drug advertisements always conclude by stating “individual results may vary.”  Reasons why specific results occur in any given individual are not known.  Knowledge of genetic factors will remove any remaining mystery.  Current diagnostic and environmental information do offer insight in future health.  This current degree of insight will seem only a glimpse, however, after the integration of environmental with genetic information is complete.

If determining genetic information of humans is going to lead to several severe disruptions, one might begin to wonder why it is being pursued to begin with.  Over the course of recent human history, the pursuit of knowledge often does not beforehand begin with contemplation of consequences.  A classic example is the development of nuclear technologies.  Much like the Human Genome Project is progressing today, the development of atomic weapons was rushed ahead because of competition (yes, there are currently competing parties sequencing the genome).  Each included in its ultimate goal the improvement of human life.  Atomic research had as one of its goals, among other things, the development of an inexpensive and powerful energy source.  The Genome Project hopes to enable more effective medical therapies.  The promise of these treatments is the ability to correct nascent flaws at the level of the genome.  Current medical treatment focuses on correcting flawed genetic manifestation.  If medical science were able to make the correction before manifestation, patient suffering and possible death could be avoided.  Focusing on correcting genetic problems is the next logical step for modern medicine to make.  Superficial treatments have come a long way, but in order to progress further, many diseases will need a more intimate cure.  Several examples include cystic fibrosis, Huntington’s, and Alzheimer’s disease (Kitcher 107).  And certainly many more that still are not known to have their basis in genetics exist.

Because of greater knowledge of the human genome, one might argue that better treatments would be developed to prevent onset of many diseases.  In this sense medical costs would decrease, and health insurance might actually be helped by the genetic revolution.  It is true that a single genetically targeted treatment to prevent a disease like cystic fibrosis from occurring would be far less costly than treating its symptoms.  The problem is that a defective gene has to be first identified before a treatment can be developed.  The gene for Huntington’s has been known for quite some time, yet a treatment is not available.  This is where the problem lies.  A significant gap in time has existed between discovery of a defective gene and treatment.  For Huntington’s and most other genetic diseases, this gap is still waiting to be closed. 

An additional gap exists which confronts a different argument. Since it is estimated that every person contains from five to seven lethal recessive genes, a leveling effect will smooth disparities between individuals’ perceived genetic problems (Capron 506).  With prospective future health looking similar for most persons, insurance will still function.  A time gap here however will present another major hurdle to overcome.  After the genome is completely mapped, not all genetic diseases will spring out from the code of T’s, C’s, G’s, and A’s.  The map will only provide a framework to assist in isolating deleterious sequences which will require many more years.  In these years, some unfortunate individuals will find their genes as being suspect and used to limit their insurability.  Thus the time-intervals between discovery and effective treatment and the discovery of large number of genetic diseases will present severe disruptions to healthcare financing.  It is in these two cavernous gaps of time where the health insurance industry will falter. Although problems may be encountered, these are gaps that need to be weathered.  The promise of what genetic treatment holds is too great to let go even if that were an option.

            The picture drawn thus far looks rather bleak for the future of health insurance in this country.  This bleak picture is meant to prove the fact that fundamental changes outlining a new system of healthcare financing will occur.  The alternative is not a possibility.  Several reasons govern this conclusion.  The perceived benefits of allowing improved medical prediction to continue unchecked do not merit its substantial costs.  Benefits would be exclusive to insurance companies and those at low risk.  These people could buy insurance and would have an easier time finding a good job.  Costs however would be widely spread over the large remaining portion of society.  These people would most likely be uninsured.  Even today a large percentage of Americans do not have health insurance; consequently, it is not a stretch to say this number would be significantly increased.  Those at higher risk would have more trouble finding good jobs because their employment would entail increased compensation for the company in question.  Being one of their only sources for care, the uninsured would then flock to emergency rooms.  Due to the large amount of uncompensated care provided, these emergency room services would have to be discontinued at many hospitals.  Even the elite, low-risk, insured individuals would have a hard time finding an emergency room. 

Elite is what thee individuals would become.  This status would be conferred on the basis of genetic traits—they were born that way.  American society carries a strong sentiment that people should not be discriminated on the basis of personally uncontrollable factors.  In fact, equal rights legislation has been enacted as proof of these sentiments.  As Francis Collins stated at a gathering of the National Bioethics Advisory Commission, “’You didn’t get to pick your DNA.  I didn’t get to pick mine.  Genetic information about you should not be allowed to deny insurance or a job.  It’s a civil rights issue’” (Wingerson 274).  Title VII of the Civil Rights Act of 1964, which prevents discrimination on the basis of race, color, religion, or national origin, may provide protection against some forms of genetic discrimination.  For example, an employer may violate the Title VII if genetic discrimination centers on some group that disproportionately carries the genetic trait in question, such as sickle cell anemia, or Tay-Sachs (Miller 193).  The impact of this interpretation of Title VII in affecting genetic discrimination will be somewhat limited however.  Most genetic diseases are not unique to the protected classes under Title VII.

Existing federal laws, which are designed to prevent discrimination in employment based on disabilities, provide another avenue to explore protection against genetic discrimination.  The Americans with Disabilities Act (ADA) was originally passed with little foresight as to genetic screening.  In drafting the legislation however, broad language which prohibits “discrimination against a ‘qualified individual with a disability’ in hiring, promotion, discharge, compensation, and other terms of employment” was used (Miller 191).  It is clear that this would apply to those individuals expressing a disease or condition as a result of a genetic trait; however, it is not clear if this would apply to an asymptomatic genetic condition.  In 1995 the U.S. Equal Employment Opportunity Commission (EEOC) clarified the debate by stating that the language of the ADA is meant to include prevention of discrimination on the basis of genetic makeup (Miller 192).  Federal law seemingly has resolved this issue.  What must also be considered is that employers need to be brought to court in order for there to be enforcement of these laws.  A denied applicant would need to bring suit against a prospective employer and prove that genetic discrimination was the reason he or she was not hired.  This is a long and tedious process that would require a great expenditure of energy by both parties to resolve.  Thus, the ADA will not curb the coming collision.

Although past legislation has not yet curtailed the imminent crisis with the health insurance system, governmental legislation is the necessary method.  A legislative approach is necessary because free-market, competition-based resolutions are not forthcoming—they are responsible for the current situation.  What has been attempted legislatively thus far deals with either preventing discrimination based on genetic information, or preventing the transfer of information in the first place.  Neither of these solutions has entirely solved the problem.  Although politically popular, they do not address both sides of the issue.  Americans are now concerned that their personal genetic information may be used against them.  Insurance companies fear the opposite: they fear genetic information will be used to their disadvantage.  Insurance is based on ignorance; if one party is no longer ignorant, the system falters.  Solutions need to be devised that not only deal with genetic discrimination, but target the underlying system itself.

            A clear solution to the problem would involve placing all Americans in one health insurance program.  This of course means instituting a national health insurance program.  Problems of both adverse selection and genetic discrimination would be avoided.  Health risk would be spread evenly across the entire population.  Predicting future health costs would become a meaningless effort—everyone would be covered.  The system could institute one mandatory community-rate and avoid experience-rating altogether.  Instituting a national healthcare policy would also have the double effect of providing insurance for the disgracefully large number of uninsured persons in the U.S.  This clear solution does not come without significant problems and tradeoffs.  Instituting national health insurance would mean sacrificing the benefits of cost control due to competition.  There is no doubt a large government health insurance program would be less efficient than a private provider is.  The healthcare sector already comprises nearly 14% of GDP. After national health insurance was put in place, one would notice a sharp increase in this percentage.  It would also mean losing choice between insurance plans.  But as the influence of adverse selection increases, this choice will become more of a hindrance than a benefit.

            Less-revolutionary solutions also exist.  One possibility would involve expanding the existing framework of employer-provided health coverage by requiring employers to cover all employees without exclusion.  For those employees with demonstrated high-risk, government subsidies would be available for reimbursement of excessive medical costs.  Providing these “insurance vouchers” would lessen the incentive to hire a low-risk applicant (Greely 279).  These vouchers would also be made available to insurance companies to help offset costs from their high-risk clients.  This proposal is not altogether without flaws either.  Initially, this system would be well suited for filling the previously mentioned gap between genetic knowledge and viable treatments.  However, when will the plan end, and who will be included in the high-risk group?  This proposal presents a good possible approach if these questions can be answered in advance.

            Another approach would be to mandate a return to community-rated insurance billing methods.  This proposal would call for employers to once again purchase health insurance from a provider and end their group policies.  The insurance providers would be required to offer set premiums at community-rated levels (Greely 280).  Future health compensation costs would then not be an issue worth considering in hiring decisions.  Other companies would offset disproportionately high healthcare expenditures by any one company in the community-rated plan.  This proposal clearly illustrates the paradoxical corner into which the health insurance industry has backed itself.  For years experience-rating reigned as the mantra for controlling and streamlining costs.  Now, experience-rating is going to meet its demise as it seeks to stray into the land of complete predictability.  Costs could potentially be streamlined out of existence, along with any profits.  Putting that ironic possibility aside, any viable solution must return to a system where the risks are shared across a broader group in order to distribute evenly costs and benefits.

           Increased predictably will result in a serious problem for the current system of financing health care.  Modern genetics and the Genome Project will make significant contributions to this threat.  The overall alarmist tone of this paper is meant only to describe the upcoming need for corrective actions to be taken in order to prevent the threat of the current collision course from continuing.  This collision course exists between the system of health insurance and those insured.  Each has valid competing claims to genetic information.  To avoid the coming collision, genetic information and the increased predictability it provides need to be separated from the health insurance system.  Promising solutions hope to accomplish this goal by making health predictability irrelevant to the providers.  Only then will genetic technology be freed from insurance fears and the benefits of medical technology enjoyed.  Then too, more ethically important implications of the genetics revolution can be considered, unhindered.