Exacerbation of Existing Disparities in Healthcare
In addition to the goals already mentioned, leaders in this field have hoped that personalized medicine will contribute to the elimination of health disparities. One proposed mechanism for attaining this goal is that the over-representation of particular phenotypes in certain racial or ethnic groups may serve as an indicator of underlying genotype–phenotype associations, which might then allow for the development of targeted therapies. Similarly, there is hope that pharmacogenomics researchers may identify the genetic variants that contribute to recognized differences in drug responses among racial and ethnic groups.
This account of the potential for personalized medicine to address health disparities has raised controversy. In particular, a number of critics have argued that work to explain race-based health disparities within the framework of genomics has tended to reinforce the mistaken belief that racial categories can be mapped directly onto biological realities. Others have argued that by highlighting genetics as an important avenue for addressing health disparities, we may obscure the importance of social, cultural and economic factors in perpetuating disparities.
Even though disagreement remains, the debate on these issues at least makes it clear that the elimination of health disparities is 'on the radar' within the personalized medicine movement. This is fortunate, since efforts to apply personalized medicine in routine clinical care have the potential not only to alleviate health disparities, but also to exacerbate them. In fact, the challenge of translating personalized medicine insights in a way that does not worsen health disparities should be a top priority of leaders in this area. In this section, we explore three dimensions of personalized medicine that could contribute to the problem of health disparities: the input–output problem, cost and access to healthcare and access to information technologies.
The Input–Output Problem
The clinical utility of personalized medicine depends on earlier scientific work focused on identifying genotype–phenotype associations within population groups. However, racial and ethnic minorities have been significantly under-represented in the studies that serve as the 'inputs' for translational efforts.[25–27] In a 2011 study of publications included in the National Human Genome Research Institute (NHGRI) Catalog of Genome-Wide Association Studies, nearly 75% of studies involved only populations of European descent. Fewer than 10% focused exclusively on non-European populations, and these primarily focused on populations from China, Japan and other Asian countries. The proportion of genome-wide association studies conducted with members of racial and ethnic groups that have suffered from health disparities in the USA remains vanishingly small. The causes of this disparity are complex, but an important contributing factor is suspicion of the research enterprise among potential research participants.
If disparities in the scientific work that informs personalized medicine continue, any benefits that personalized medicine will be able to deliver are likely to be distributed unevenly among population groups. This is because the diagnostic and therapeutic approaches that inform personalized medicine practice are developed using data from this type of research. This research reveals, for example, which genetic variants are relevant to disease risk and what "effect size" each variant has on disease risk or response to therapies. The input–output problem arises because allele frequencies and environmental exposures tend to vary among population groups. Because of this, the assumptions that inform personalized medicine practice in well-studied populations are not necessarily generalizable to poorly studied populations. Genetic test panels designed using data from one population group may not capture the genetic variants relevant to disease risk or treatment response in another group. In addition, pharmacogenomic algorithms that guide drug dosing or selection may lead to suboptimal outcomes in patients whose 'background' genetic variants and environmental exposures are significantly different from well-studied populations. This is especially concerning because the groups that have been under-represented in genomic research are also the groups that are already receiving suboptimal benefits from existing healthcare services.
Cost & Access to Healthcare
Another set of challenges that threaten to exacerbate health disparities in the coming decade are economic barriers that both limit access to healthcare and reduce the benefit patients are able to derive from that care. This characteristic is common to many new healthcare technologies: if patients are unable to access a new technology, then they are also unable to enjoy the benefits of that technology.
In the case of personalized medicine, the laboratory tests that inform personalization, such as next-generation sequencing, are likely to be quite expensive at first, despite optimism that the US$1000 genome has arrived. Such milestones do not account for labor costs, analytical costs or commercial mark-up.[31,32] An additional source of increased cost will be the interventions that are recommended in light of these laboratory test results. For example, pharmacogenomic testing may have the potential to decrease overall costs at the level of the healthcare system. At the level of individual patients, however, many are still likely to end up taking medications with higher direct costs compared with the standard therapy.[33,34]
The costs of personalized therapeutics and the tests that inform their use are unlikely to cause difficulty for patients who are already well served in the healthcare system. In countries with private insurance systems, patients with comprehensive health insurance coverage or the ability to cover such costs out-of-pocket will be able to undergo new tests and receive the benefits of individualized treatments despite their cost. Patients with no insurance, as well as patients with insurance designed to provide only urgent care, are unlikely to benefit from these advances.
Patients living in countries with nationalized health insurance systems are likely to fare somewhat better. Nevertheless, these systems generally limit coverage to treatments with established efficacy. Since studies related to clinical applications of personalized medicine could require larger study samples compared with conventional approaches, it may take longer for an evidence base to emerge for such applications.[35,36] If this is the case, nationalized health insurance systems may be slow to adopt personalized medicine approaches. If this occurs, even interventions that eventually prove efficacious might be available only on the private market for an extended period of time.
Regardless of the nature of the health insurance system, most patients in developed nations will eventually receive benefits from personalized medicine. In comparison, improvements in the care received by patients in developing nations are likely to be limited. The cost of new diagnostic tests and alternative treatments are likely to limit their availability in these parts of the world for the foreseeable future. Even more importantly, perhaps, the medical problems that cause the most morbidity and mortality in developing nations are comparatively rare in the developed world. If personalized medicine is to be efficacious for these patients, then research efforts focused on personalized medicine will need to expand to include work on the medical conditions endemic to these areas. Furthermore, since 'background' genetic variants and environmental exposures are so important to personalized treatments, this work will need to be performed with the populations of these developing nations, a group that has previously been under-represented in personalized medicine research.
In both developed and developing nations, the costs associated with health behavior changes are another source of disparity in the benefits personalized medicine will deliver. There is already ample evidence that medical problems influenced by health behaviors disproportionately affect patients in lower socioeconomic strata.[38,39] Among other insights, this disparity reflects the degree to which personal finances affect patient access to healthy foods, exercise facilities and other resources related to healthy behaviors. Many advocates for personalized medicine have argued that genomic tests might improve health by helping patients identify their health risks and undertake health behavior changes that could help mitigate these risks.[20,40–43] However, the likelihood that risk information will help individual patients make meaningful changes in health behaviors will be influenced by a range of factors beyond the control of the healthcare system; one of these will undoubtedly be the financial resources that patients have to support these changes.
Access to Information Technologies
Information technologies that allow patients to access their own health records play a central role in many visions of personalized medicine. In part, this is due to the value of such technologies for addressing the challenge of information overload. However, the centrality of these technologies to personalized medicine also reflects growing interest in empowering patients to monitor their own health, perform research on their own health problems and make positive health behavior changes. In this way, personalized medicine reflects more general trends in healthcare to encourage patients to use information technologies in order to take responsibility for their own health needs.[45,46]
This movement towards increased patient responsibility for health is reflected not only in cultural trends, but also in public policy. In the USA in particular, a range of recent public policy changes are intended to encourage healthcare institutions to provide patients with direct electronic access to their health records. The 2009 Health Information Technology for Economic and Clinical Health (HITECH) Act created financial incentives for institutions capable of demonstrating 'meaningful use' of an EHR. Patient portal functionality is one application that can help an institution prove it has attained meaningful use of an EHR. Similarly, a recent amendment to regulations promulgated under the Clinical Laboratory Improvement Amendments of 1988 allows laboratories to give patients or their designated representative direct access to laboratory test reports; in the past, only healthcare providers were authorized to receive such results.
While efforts to empower patients are laudable, they raise significant challenges related to health disparities. A patient can only benefit from an electronic patient portal if he or she has access to internet services and an internet-capable device, as well as the necessary computer literacy to navigate to and within the portal website. Going further, the information presented on an electronic patient portal is only useful to those patients with adequate health literacy. This is especially problematic for 'omics'-based laboratory results, which can prove especially difficult to understand. In short, the patients who are most likely to have the resources needed to make productive use of a patient portal for personalized medicine are those patients who are already well served by the healthcare system.
Personalized medicine should account not only for the genetic individuality of patients, but also for individual environmental exposures and the unique social situations that influence patient abilities to utilize healthcare. In the current vision of personalized medicine, electronic patient portals are portrayed as one-size-fits-all tools for patient empowerment. If personalized medicine is to be successful, more targeted approaches will be required. Without such alternatives, it is possible that electronic patient portals may create an illusion that all patients have the resources they need to improve their health. Such an illusion could be counterproductive, since the more patients are perceived to have the power to improve their health, the more likely they are to be seen as responsible for their health outcomes.[45,50,51]
Personalized Medicine. 2015;12(1):43-51. © 2015 Future Medicine Ltd.