Personalization is everywhere: what benefits for the health sector?

The world is becoming faster-paced and more digital. Many industries (entertainment, shopping) have recently made great advances in using personalization in their digital interactions with customers. Healthcare systems must adapt to keep up and meet patient expectations, especially those of patients with chronic conditions, who require repeated, sustained care and often need to change their everyday lifestyle and habits.

Most of us are exposed to personalization every day through targeted advertising. With this type of advertising, a system shows consumers products or services they are likely to want in a way that is likely to engage them. Netflix is a champion of these methods. Netflix, with a complicated algorithm, learns what a user likes to watch and makes suggestions, all the time improving these suggestions by tracking watching habits. 

Health care systems need to learn from these industries and adapt their methods to best serve patient interests while using available resources (such as medical personnel) in the most efficient way possible. This will involve incorporating complex algorithms and the terabytes of available data on patients' behaviors, habits, needs and wishes to customize the user experience in a simple, streamlined way, while continuously collecting more behavioral data to improve the algorithms.

Note that healthcare differs in two very important ways from these other industries:

1. Whereas Netflix users want to consume content, patients with chronic conditions would rather not have to consume medication or undergo treatments. Any e-health interface must take this into account.
2. Healthcare is a highly regulated sector, and health data are particularly sensitive.

The trend towards personalization in the health sector has already begun with personalized or “precision” medicine, which uses an individual’s genetic profile to diagnose certain conditions and guide treatment decisions⁴.

• Personalized medicine considers the individual’s genetic makeup as well as the genes or other biomarkers characteristic of the condition to guide diagnosis and treatment decision-making.  Cancer care is one of the first medical specialties to apply personalized medicine.¹Even in tumors originating in the same organ, there are many different subtypes. Genetic testing can reveal both bio-specificities of cancer cells and indicate the most appropriate medication, or personalized care. For example, someone with HER2-positive breast cancer is likely to respond to the drugs that target that protein. ¹. In women who have surgically removed all traces of this subtype of tumor, a therapy regimen using anti-HER2 drugs roughly halves the risk that the cancer will recur.2

• Predictive medicine is one facet of personalized medicine enabled by genetic screening. Returning to the example of breast cancer, women from families with a history of breast cancer are at higher risk of developing the disease, particularly if the women are carriers of the BRCA1 or BRCA2 gene mutations. These mutations prevent cells from effectively repairing DNA. The ability to detect the presence of the BRCA mutations using genetic screening can determine the appropriate cancer-screening frequency for these women, thereby helping to detect breast tumors at an early stage and improving patient outcomes and quality of life.²

• Another key example of personalization in care is the CART-cell therapy to treat certain blood cancers ³. This treatment consists in extracting from a given patient a type of immune cells – called T cells – and genetically modifying the genome of these cells to make them produce a chimeric receptor that will efficiently bind a protein expressed at the cancer cell surface. After growing these Chimeric Antigen Receptor T cells in laboratory, they are transfused to the same patient and enhance the immune response against the cancer cells.

This technology is revolutionary and allows patients to avoid compatibility issues that could arise with donor cell therapies.

In parallel to diagnosis and care’s personalization, tailored e-health solutions are also emerging to help patients self-manage their disease. These are personalized, digital interfaces that adapt to the needs of the patients to help them successfully navigate their treatment, care pathway, paperwork, and other aspects of their daily life.

Personalization of this sort has already shown itself to be useful in trials on small groups.

• A study on self-management in patients compared individually-tailored interventions to a “one size fits all” approach and found that the group who received personalized intervention plans were twice as likely to show the desired behavior.⁴
• Real-time digital monitoring of behavior and physiological measures can improve health outcomes in cases of diabetes type 1 or 2 – but only if this monitoring is unobtrusive and aligns with patients’ life habits.⁵
• Personalized coaching for schizophrenia patients increases treatment adherence and social interactions while decreasing hallucinations.⁶

All these examples highlight the benefits of collecting patient data (genetic, clinical, behavioral…) to conceive personalized care solutions and improve the patient’s quality of life.

Before personalization of e-health solutions is possible, it is necessary to understand the behaviors, needs, and wants of patients in many different situations and with different backgrounds. Once this data is available, effective solutions can be tailored. Though massive progress has clearly been made in this domain, there remains a long way to go. Observia has taken recent developments into account and is responding to the obvious need for continued refinement, having spent the last ten years creating d.tells™, an intelligent engine for optimizing and personalizing e-health solutions.

Incorporating SPUR™, a powerful behavioral diagnostic tool that is based on theoretical approaches and evidence from numerous scientific domains, d.tells™ combines this model with artificial intelligence to provide patient support that constantly evolves and adapts to the patient's needs.

Complex algorithms and artificial intelligence underlying this process allow evaluation of each patient's individual needs, leading to adaptation of the content the patient sees, the frequency at which he/she sees it, the medium or channel used to communicate it, and even the type of language used to address the patient. The end-result is optimization of patient experience and care pathway at the same time. A key to this optimization is that, though the algorithm is complex, it all happens in the background, leaving the patient (and their health care team) with a simple, smooth user experience; d.tells™ provides a seamless patient experience.

In this way, Observia is moving toward providing patients and health care workers with intelligent, whole-person care that is fully automated and quickly adapts to individual needs.

Behavioral scientists are constantly improving techniques for data collection and analysis while computer and data scientists do the same for machine learning and AI. With these scientists working together [at Observia{and elsewhere}], there is no doubt that machine learning powered by big data will continue to help healthcare service providers make rapid progress in personalization.

Download the article

References : 

1. “Precision Medicine for Breast Cancer.” Mayo Clinic, Mayo Foundation for Medical Education and Research, 10 Oct. 2020, www.mayoclinic.org/tests-procedures/precision-medicine-breast-cancer/about/pac-20385240#:~:text=Precision%20medicine%20for%20breast%20cancer%20is%20an%20approach%20to%20diagnosis,collected%20for%20analysis%2C%20often%20genetic.

2. Esmo. “Personalised Medicine at a Glance: Breast Cancer.” ESMO, Feb. 2017, www.esmo.org/for-patients/personalised-medicine-explained/Breast-Cancer.

3. Michael Kalos, Bruce L. Levine, David L. Porter, Sharyn Katz, Stephan A. Grupp, Adam Bagg and Carl H. June. T Cells with Chimeric Antigen Receptors Have Potent Antitumor Effects and Can Establish Memory in Patients with Advanced Leukemia. Sci Transl Med. 2011 Aug 10; 3(95): 95ra73. doi: 10.1126/scitranslmed.3002842

4. Eikelenboom N, Van Lieshout J, Jacobs A, et al. Effectiveness of personalised support for self-management in primary care: A cluster randomised controlled trial. Br J Gen Pract. 2016;66(646):e354-e361. doi:10.3399/bjgp16X684985

5. Oikonomidi T, Ravaud P, Cosson E, Montori V, Tran VT. Evaluation of Patient Willingness to Adopt Remote Digital Monitoring for Diabetes Management. JAMA Netw open. 2021;4(1):e2033115. doi:10.1001/jamanetworkopen.2020.33115

6. Granholm E, Ben-Zeev D, Link PC, Bradshaw KR, Holden JL. Mobile assessment and treatment for schizophrenia (MATS): A pilot trial of an interactive text-messaging intervention for medication adherence, socialization, and auditory hallucinations. Schizophr Bull. 2012;38(3):414-425. doi:10.1093/schbul/sbr155