MétaCan
Menu
Back to cohort
Record W2567269832 · doi:10.1373/clinchem.2016.260349

Direct-to-Consumer Testing

2016· article· en· W2567269832 on OpenAlexaff
Michelle Li, Eleftherios P. Diamandis, David G. Grenache, Michael J. Joyner, Daniel T. Holmes, Rodger Seccombe

Bibliographic record

VenueClinical Chemistry · 2016
Typearticle
Languageen
FieldMedicine
TopicBiotechnology and Related Fields
Canadian institutionsSt. Paul's HospitalUniversity Health NetworkMount Sinai HospitalUniversity of Toronto
Fundersnot available
KeywordsGenetic testingDiseasePersonalized medicineInternet privacyMedicineGeneticsBiologyComputer sciencePathology

Abstract

fetched live from OpenAlex

Due to technological advancements, self-testing has become widely accessible to the public. Individuals can opt to have their genome sequenced or their blood tested for markers at a relatively cheap price. These direct-to-consumer services are essentially a commercialization of technologies being marketed to the general masses. Some genomic giants in the industry include 23andMe and Gene by Gene. Their test kits can be delivered internationally and sampling is performed by the user and sent back for laboratory analysis, thus establishing an accessible and flexible service model. Users can opt to test for specific genes that correspond to a potential disease or learn about disease predisposition, drug responses, or genetic characteristics. Other companies offer to quantify a range of biomarkers that can potentially predict the early onset of a disease or condition. Their kiosks and laboratories are situated within pharmacies and the blood tests can be performed without a physician's consent. The results are then electronically delivered to a physician or directly to the consumer, and are subject to self-interpretation. The underlying notion is that such testing may uncover abnormalities that could potentially serve as an early marker of disease. By identifying this pathogenic link at an early, asymptomatic stage, the consumer can possibly take steps to prevent disease later on. However, it is important to keep in mind that due to epigenetics, environmental and other factors a gene sequence is not always reflective of a phenotype. The sequencing only provides minimal information about a possible genetic foundation, yet ambiguous gene expression deems results inconclusive. Likewise, testing for biomarker concentrations in the blood is not necessarily a reflection of a patient's condition. Due to the large variability in individual physiology, there can be ambiguity with self-interpretation. Despite having access to reference ranges/intervals from online sources, many patients are in a poor position to judge their own health. Factors such as lifestyle, time of day, eating patterns, and ethnicity heavily affect results; for these reasons, a physician may be best positioned to analyze results. Here, we ask 4 individuals about their opinions on various aspects of direct-to-consumer testing. Do you think whole genome sequencing in 2016 is beneficial in some shape or form to the majority of consumers? David Grenache: It depends on what one considers beneficial. For some individuals, simply having access to their unique genetic information is itself the benefit, regardless of whether or not it is actionable. For others it's the knowledge of ancestry that accompanies direct-to-consumer genetic testing services that's of value. As with any service or product, the consumer makes the determination regarding value. Why should genetic information, or any laboratory test for that matter, be any different? That said, consumers of sequencing tests must be fully aware of the limitations of whole genome sequencing results and need to be educated about what these data can and cannot reveal about health and health risks. Michael J. Joyner: No, I believe this is a niche technology best used for research purposes and to investigate rare diseases that are difficult to diagnose or that are “one-offs.” To do the latter it will need to be part of a comprehensive rare diseases or clinical genetics evaluation. This sort of testing is already providing causal insights in some but not all cases. Unfortunately, in many cases it does not lead to a beneficial therapeutic intervention but can help families think about future reproductive options and testing. In cases like channelopathies and sudden death in the young, a sort of “genetic purgatory” has emerged where an apparently causal DNA variant in the case is also seen in healthy relatives. Then the question is literally about who does or does not get an implantable defibrillator based on what criteria. For most issues in this area there are no easy answers and the potential for less than ideal unintended consequences and ambiguity is substantial. Daniel T. Holmes: I think the key phrase in this question is “to the majority of consumers.” My position is that presently it is generally not medically beneficial to the majority of consumers but may be useful in specific contexts. Consumers can receive various products from these companies. For example, they can purchase a single nucleotide polymorphism analysis but the clinical inferences one can make from single nucleotide polymorphisms are frequently of little value. How should a consumer respond to, “You have a 7.3% increased risk of disease X”? Accordingly, this type of report has been subject to regulatory limitation in North America due to the ambiguity of available evidence. Consumers can also receive ancestry reports, primarily for personal interest and carrier-status and drug-sensitivity reports, which may have utility in select circumstances. Finally, they can receive reports about certain well defined but medically banal Mendelian traits such as whether their earwax is crusty or slimy (no, I am not kidding). Rodger Seccombe: While there remains a substantial gap between our current ability to complete vs effectively interpret a whole human genome sequence, we are already seeing the benefits of whole genome sequencing as a tool to help diagnose and treat rare diseases in symptomatic individuals, such as in England's 100000 Genomes Project. The average curious consumer might also benefit in some ways. However, for most individuals, the insights gained would not justify the expense of completing a whole sequence. Given the limitations of our current genomic knowledge base, the majority of whole genome sequencing consumers would not receive much “actionable” information beyond the basic pharmacogenomic and carrier status reporting that can be obtained from less costly genotyping services. That said, however, new discoveries are being made by comparing entire genomes. So on some level we are all benefiting from those individuals who are completing and agreeing to share their whole genome sequences for research. How reflective of a patient's health are sequencing and blood testing results? David Grenache: It depends on the gene associated with the sequence or the particular analyte being evaluated in the blood test. It's well known that many DNA polymorphisms have no effect on one's health while others are very predictive of disease. The result of any blood test represents a small snapshot in time captured at the moment of phlebotomy. No single test is capable of being reflective of “health” any more than a single word can reflect the plot story of a novel. Michael J. Joyner: For the average person the results are likely to be of marginal or no benefit and difficult to put into context. A large issue here is the so-called incidentalome, which would reveal the presence of “pathologic” variants in individuals who are phenotypically normal. Daniel T. Holmes: With respect to sequencing, in certain contexts, for example where there is a yet undiagnosed or subclinical monogenic disorder, the data could be very useful and reflective of the patient's current or future health. But this scenario would certainly represent the exception and not the rule. The same is true of blood testing. Untargeted use of blood tests is more likely to lead to confusion and waste because of the Bayesian principle that as the disease prevalence approaches 0, the positive predictive value of the test approaches 0. This means that performing a nonindicated test leads to diagnostic confusion, further testing, and referrals. If we factor in the principle that most reference intervals are based on the central 95%, then the probability that at least one test falls outside its reference interval by chance alone is 1–0.95n, when n tests are performed. This probability exceeds 50% for 14 or more tests. There is also an implicit assumption that results inside the reference interval imply health and results outside the reference interval imply pathology. However, results outside the reference interval don't always imply pathology and there can be “inappropriately normal” results that are highly pathological, e.g., a TSH (thyroid-stimulating hormone) of 2.5 mIU/L that looks ostensibly normal would be indicative of secondary hypothyroidism if the free thyroxine were 0.60 ng/dL (6.4 pmol/L). Rodger Seccombe: We know that for certain disease states, some genetic mutations and serum biomarkers (particularly once confirmed) can be highly reflective of a patient's current health or even diagnostic in and of themselves. This is possible, in part, due to extensive research and standardization of laboratory testing for certain analytes, such as hemoglobin A1c, creatinine/eGFR (estimated glomerular filtration rate), and lipids. Other results and genetic findings are not as well known, standardized, or uniquely predictive. However, they can still be important indicators when evaluated as a group and in the context of a patient's history and symptoms. As technology enables more health- and lifestyle-related data points to be made accessible and analyzed in aggregate over time, the current and likely future state of a patient's health will become more certain. Until then, genomic sequencing and blood test results will reveal only part of the story. What is your opinion on commercializing genome/exome sequencing to the masses? Is the motivation mainly for profit? David Grenache: What motivates any individual to launch a company or provide a service? In an economic system based on capitalism, profiting from the sale of a product is necessary but that doesn't mean it's the main motivator. One could ask the same question of hospitals and healthcare delivery systems that often make huge profits from laboratory services because their charges are marked-up greatly, vary widely, and are not transparent. I believe we are at the threshold of a new era in medicine, one in which patients are growing much more engaged in how decisions are made regarding their care and are demanding access to information that, until recently, has not been easily available to them. Now that information is becoming increasingly available and at prices that are simultaneously transparent and affordable. Michael J. Joyner: I believe some of the motivation is for profit. I also believe that this is a bit of a Trojan horse effort to get health information and DNA from a large sample in an effort to generate “big data” and look for rare variants that might give insights into potentially “druggable” or therapeutic targets. The goal at some level is to find the next proprotein convertase subtilisin/kexin type 9–like target. Daniel T. Holmes: This appears a profit-motivated business directed at the well heeled, worried, and curious well. If this were about the good of humankind, we would expect philanthropic organizations working to bring these services to the underprivileged of the world, which they are not doing. Rodger Seccombe: For-profit companies have been making significant investments to provide genomic services to the public, ultimately to create and return value to their shareholders. While there's a clear profit motive, these efforts also help fuel advancements in the underlying technology and enable new scientific discoveries. As demand grows for these services, we will need to ensure the necessary safeguards are in place to protect the public's genetic information. For example, enacting new legislation, such as proposed Bill S-201 in Canada, to prohibit and prevent discrimination from insurance companies, employers, or others based on genetic characteristics. What are the pros and cons of making self-testing accessible at pharmacies without a physician's oversight? David Grenache: I see several benefits. Patients are becoming active consumers of medical information and are becoming more educated about their health. As such, they rightfully demand more control and access to personal health information. There are known benefits of patient self-testing, particularly when it comes to the management of chronic diseases like diabetes. Cost is always a concern, even more so for patients without health insurance or for those with high deductible insurance plans. Direct-to-consumer laboratory testing companies have price transparency and offer their services at a cost that is usually considerably lower than what would be charged in a hospital or reference clinical laboratory. One disadvantage is that easy and affordable access to laboratory tests could prompt some patients to over indulge, with the unintended consequence of initiating a cascade of events due to the inevitable false positive result. The disadvantage I hear voiced most frequently is that patients won't understand the results or will misinterpret them. I don't completely disagree with this but I do think that patient's who are engaged enough to use these services are motivated to learn what tests to consider and understand what test results mean. There is a great opportunity here for clinical laboratory professionals to be leaders in this patient education, which would have the added advantage of raising public awareness of the role we play in healthcare delivery. Michael J. Joyner: I see no pros. I see only issues related to a reduction in quality and things being interpreted out of context. Daniel T. Holmes: Are there pros? I suppose if someone checked their lipids or fasting blood glucose and obtained increased results, this might spur them to see their physician. But they could get this self-same testing (essentially) free-of-charge though the medical system in Canada and their care in the I see a of the being consumers have a means to or the of any or The is evaluated not on and but on its The consequence is that a can be without consumer or for an by the regulatory and reports on and its are part of the in the laboratory and clinical care contexts, they can the data and the and cost value. to is testing for some tests on the We have one of these at our While some to others have very The to regulatory and quality Are direct-to-consumer laboratories to to Are they to to an quality Other issues What do direct-to-consumer laboratories do with is medically and if an result and there are medical The to If the patient an result by chance alone and then to their who the cost of the it is the Rodger Seccombe: The most important benefit we have seen with is the ability to individuals in their own particularly in the case of chronic disease and are often more accessible than and can patients to the out by their or potential testing can also provide an where can the of a healthy at a time when patients are more the often the But that most of how healthy we So for person who is by a test there are more who are in of their health and should be with their direct-to-consumer services a David Grenache: I would to consumer testing has been in some form or for glucose and I don't see that it has more is simply that the test available to patients has beyond that by tests. It's easy to see how a patient could to a with a laboratory report of results from tests they performed and ask about what the result might mean. If that their then it's time to find more physician. This issue is not to with time, there will be increasingly in which patients will in has a and and who are and patients are to find they have a of them. Michael J. Joyner: in a there are many of information that are already this and this is one of technology that can be or Daniel T. Holmes: patient care is patient in laboratory medicine, we are so on that we we are for not testing to be that can be but laboratory testing does not look at the patient's and their It does not ask about or the for the It does not the or the It does not for a It does not for or masses. it does not are things with your or time you were about your How are you with of it does not and interpret the tests in a Rodger Seccombe: is the on its This has been the of the patients these (particularly the will the seeing their So the physician is no the of health information but an access testing is an of this providing patients with more information and insights about their own and health. not without this is ultimately positive because it means patients are more and engaged when they their or any other for that It more of a Do you believe that the will become the of our health information for direct-to-consumer How could a become a medical David Grenache: In that it's well on its to becoming We already an of personal and data on and in the health information is one more The is at the of health and patients are already their personal health data which can then be analyzed and on important is the and aspects that the small with the and of the makes as medical a future several and public companies are already such Michael J. Joyner: to and at least so many and have The have been by and and the so-called gene testing, it is how they are in the to get to By as to take more might be But this is a very use of Daniel T. Holmes: I think it is that we would all our medical on a in the future but I don't see this as an that would much interest in the possibly as a for a of early However, the is in a with for of This makes the an tool to which one could a of any or The of a exceeds what is for data reduction for blood testing and this is an area of active there are associated with the use of whole blood and a small sample that would need to be but I that this of technology is inevitable and could be beneficial in Rodger Seccombe: The with medical and health will be the most for and the as the will also the many and medical and patient health to of As we learn more from data and information, there will be new to enable as well as and available directly from the in technology should also more data to be and analyzed in time without the need to a particularly for less tests such as glucose or lipids. consumer testing become a of David Grenache: I don't believe testing has been available for many over time, not only have technologies but have One only to look at the test of the many laboratories that offer direct-to-consumer testing to see that the are not out of for most and are certainly not affordable only by the Michael J. Joyner: It will become a of the well will get the testing, get to interpret risk demand and the consequences of over testing. This is of a great Daniel T. Holmes: I think testing is already the In Canada, the that medically necessary healthcare is by the services have with laboratories to this same this is the of testing and its is to to the However, the business principle of the industry is to create relatively that offer health those as a for The of an to the does not a notion and this may the industry and prevent it from becoming a of the Rodger Seccombe: no that the early of new consumer testing technologies will the with the help of seen how prices can for the underlying ultimately making these once products and services accessible to the masses. For example, one of the direct-to-consumer whole genome sequencing services at in and likely less than by In certain a direct-to-consumer testing could result in significant healthcare and should be as the new For example, providing testing in pharmacies for individuals or at risk for disease or diabetes. By the and the we know that we can to and physician In from what do you see in of technology to or David Grenache: technologies are to a in The of the as a of health information, will to in that patients to diagnose disease. like health that to type or and will then be to provide likely the of and to as the of to and share health data can be used not only to chronic disease but diagnose in early in will the of laboratories that will with a to analyze and interpret results. This it's If you to the a to the that can an health that can diagnose patients or to a you Michael J. Joyner: The is blood glucose and of This has been the I in medical in It will be to see if they get Daniel T. Holmes: I think there will be a of or online for the of and data that will provide a like this and the technology and information and are all in to medical of these to an individual for to see a physician is or This of tool could also represent a to But the and so to medical I find the at the to in part of the world, is an active area of research in the of and the This will be to and will many regulatory In of in a of companies are working to glucose in as part of the management of but of this type have been to and should see other have been made for and other targets. Rodger Seccombe: I think technology will become increasingly and in will bring technology to the consumer, where it would ultimately be with likely also see the next of capable of and much more such as and other markers see even with health such as or implantable glucose The most advancements, however, will from decisions and being made with the help of data and For example, is already being used to make care decisions in and this type of will become accessible and help become educated and engaged in their own health.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.008
metaresearch head score (Gemma)0.031
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesInsufficient payload (model declined to judge)
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Not applicable · Consensus signal: Not applicable
GenreCandidate signal: Commentary · Consensus signal: none
Teacher disagreement score0.327
Threshold uncertainty score0.960

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0080.031
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0020.001
Science and technology studies0.0010.001
Scholarly communication0.0040.003
Open science0.0030.004
Research integrity0.0030.003
Insufficient payload (model declined to judge)0.3270.198

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.055
GPT teacher head0.353
Teacher spread0.299 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

Study designNot applicable
Domainnot available
GenreCommentary

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations12
Published2016
Admission routes1
Has abstractyes

Explore more

Same venueClinical ChemistrySame topicBiotechnology and Related FieldsFrench-language works237,207