Corresponding Author: Mary Oakley Strasser, [email protected]

Conflicts of interest: The authors have no conflicts of interest to disclose.

Historically, car safety testing has relied on data collected from crash dummies that uniformly assumed 170-lb male drivers. Studies now show that this approach fails to identify key injury patterns in individuals who deviate from this standard.[1] When exposed to the traumatic forces of a motor vehicle accident (MVA), a person’s body composition, including age, sex, and body mass index (BMI), can be as informative as the speed at impact or rate of deceleration in predicting the severity and pattern of accident-related injuries.[2] Stewart Wang, MD, PhD, an endowed professor of trauma and burn surgery at the University of Michigan, analyzes human anatomy to guide improvements in motor vehicle safety that focus on the individual rather than the standard. By analyzing computed tomography (CT) scans of individuals with MVA-related injuries, Dr. Wang has identified patterns in measurements such as bone density, muscle mass, and body fat distribution that reliably predicted both the severity and pattern of injury.[3] Automotive engineering initiatives and designs have since started to embrace this idea that injury is a factor of both the traumatic insult and inherent patient characteristics.[4] This use of measurements of an individual’s anatomy with established prognostic value to predict outcomes is known as morphomics and has started to expand beyond automotive safety research. The following review will examine potential additional applications of morphomics as well as consider potential ethical challenges raised by broader use of this technology.

In addition to redefining individual parameters for automobile safety, morphomics can characterize meaningful markers of patient health or frailty. It demonstrates promising utility in improving personalized outcome prognostication and treatment-related decision-making for patients with life-threatening disease. For example, the cross-sectional area of psoas muscle significantly stratifies mortality risk for patients in the surgical intensive care unit, thereby stratifying outcome predictions.[5] Compiling this and other similar types of data can contribute to a calculated morphomic age, an estimate that weighs many prognostic anatomic parameters in order to individualize outcome predictions for patients. Morphomic data has been studied as a predictor of survivability in many disease processes and proved to be an accurate measure of mortality risk in many of the most common malignancies. This greater ability to predict outcomes may prove to be a powerful tool for physicians and patients to better understand a patient’s individual risks. Such data-driven risk stratification can be applied to inform improved shared decision-making for patients and decision support tools for physicians.

In fact, morphomics has already being applied in this way in pediatric care. Current research is underway to investigate the use of morphomics data to delineate “normal” and reconstruct traditional growth charts in pediatric patient populations. Variables including muscle, bone, and fat, as opposed to the traditional variables weight and height, were used to create the Pediatric Reference Analytic Morphomic Population data.[6] This tool will be used to better define what is abnormal, improve screening, and allow for personalized pediatric clinical care.

The next logical step from risk stratification is risk modification and prevention. Researchers at the University of Michigan have taken this step using morphomics data to improve outcomes. For example, in patients undergoing abdominal surgery, sarcopenia has been shown to increase costs.[7] Implementation of a preoperative training program including exercise has resulted in decreased hospital length of stay and costs.[8]

Although morphomics has the ability to create positive change in healthcare, as with any new technology, it also carries risk. In its current form, morphomics uses only CT scans obtained for other medical indications, but its benefits may drive physicians to request patients obtain CT scans prior to certain procedures or interventions. If incorporated into medical decision-making, morphomic data could drive the need for more frequent CT scans, leading to increased incidentalomas, radiation exposure, and cost to the patient and the healthcare system. In order to secure insurance coverage for these scans, CT-driven costs would need to be offset by overall healthcare savings; even if this were successfully demonstrated, it would still present a barrier to care for low-income individuals if required prior to treatment. Access to CT scanning is limited in low-income communities, and this existing disparity could further increase disparities in care.[9]

Morphomics could also be problematic for insurance companies, as high-risk morphomics markers could be considered a preexisting condition. Although preexisting conditions are currently protected under the Affordable Care Act (ACA), it is unclear if the future attempts to repeal the ACA will eventually target nondiscrimination clauses. If this legislation is repealed, morphomic data useful for predicting outcomes and healthcare costs could contribute to health and life insurance companies’ decision-making. Similar factors could promote discrimination against employees, potentially identifying individuals who have susceptibility to conditions that may limit the longevity of their career or ability to perform physical labor. This presents an ethical risk for physicians choosing to use this technology and for broadening its use to standard clinical care.

The application of morphomics data has the potential to personalize healthcare and predict outcomes better than can current demographics, medical history, and family history. Morphomics fills a void in the ability to personalize large-scale health efforts to the individual, from automobile safety to operative risk stratification to pediatric screening tools, but potential ethical implications must be considered moving forward in order to apply this new technology responsibly.

References

1. Boulanger BR, Milzman D, Mitchell K, Rodriguez A. Body habitus as a predictor of injury pattern after blunt trauma. J Trauma. 1992;33(2):228–232.

2. Wang SC, Rupp JD. Alterations in body composition and injury patterns with aging. Presented: CIREN Public Presentations, National Highway Traffic Safety Administration; 2006. https://one.nhtsa.gov/DOT/NHTSA/NVS/CIREN/2006 Presentations/MI_0306b.pdf.

3. Morphomic Analysis Group. Reference Analytic Morphomic Population (RAMP). Michigan Medicine website. http://www.med.umich.edu/surgery/morphomics/ramp. Accessed April 18, 2018.

4. Crandall J. Simulating the road forward: the role of computational modeling in realizing future opportunities in traffic safety. Lecture presented: IRCOBI Conference; September 2009; York, UK. http://www.ircobi.org/wordpress/downloads/irc0111/2009/BertilAldmanLecture/bal.pdf.

5. Englesbe, MJ, Patel SP, He K, et al. Sarcopenia and mortality after liver transplantation. J Am Coll  Surg. 2010;211(2):271–278.

6. Harbaugh CM, Zhang P, Henderson B, et al. Personalized medicine: enhancing our understanding of pediatric growth with analytic morphomics. J Pediatr Surg. 2017;52(5):837–842.

7. Gani F, Buettner S, Margonis GA, et al. Sarcopenia predicts costs among patients undergoing major abdominal operations. Surgery. 2016;160(5):1162–1171.

8. Englesbe MJ, Grenda DR, Sullivan JA, et al. The Michigan Surgical Home and Optimization Program is a scalable model to improve care and reduce costs. Surgery. 2017;161(6):1659–1666.

9. Glover M, Daye D., Khalilzadeh O, et al. Socioeconomic and demographic predictors of missed opportunities to provide advanced imaging services. J Am Coll Radiol. 2017;14(11):1403–1411.