π Health and biotech
Personalised medicine: custom healthcare on a national scale?

“Let’s personalise healthcare pathways, too”

Agnès Vernet, Science journalist
On February 2nd, 2021 |
3 min reading time
Etienne Minvielle
Etienne Minvielle
CNRS Research Director and Professor of Health Management at Ecole Polytechnique (IP Paris)
Key takeaways
  • Personalised medicine doesn’t just involve treatments, it could also incorporate the entire healthcare setting and patient experience.
  • Étienne Minvielle explains that better patient care means extra-medical assistance must also be personalised.
  • Patients have different needs, desires, personalities, and lifestyles. Taking these differences into account could improve treatment outcomes.
  • Personalised healthcare pathways could be developed with the aid of digital tools, similar to those used in e-commerce.
  • The title of this Braincamp is inspired by Étienne Minvielle's thesis at École polytechnique: "Managing singularity on a large scale".

Just as the prac­tice of med­i­cine is more than pre­scrib­ing med­ica­tion, per­son­alised med­i­cine more than tar­get­ed treat­ments. Éti­enne Min­vielle, pro­fes­sor at École Poly­tech­nique is work­ing on more per­son­alised health­care path­ways. His aim is to con­sid­er patients as a whole, tak­ing into account their social con­text, habits and even beliefs.

“We’re talk­ing about a per­son­alised approach through­out the the entire jour­ney of a patient through the health­care sys­tem. From hos­pi­tal to home, includ­ing prac­ti­tion­ers such as inde­pen­dent nurs­es and phar­ma­cists,” Min­vielle explains. 

Offer patients the best guidance

The idea is to draw on tools devel­oped in the ser­vice and e‑commerce sec­tors to tai­lor cus­tomer expe­ri­ence. How­ev­er, instead of offer­ing dis­counts based on shop­ping habits, per­son­alised care means tak­ing into account the patient’s social deter­mi­nants and psy­choso­cial needs.

“Once diag­nosed with a chron­ic ill­ness, a patient will require ser­vices and non-med­ical inter­ven­tions to improve their qual­i­ty of life. This could involve get­ting a pet-sit­ter dur­ing treat­ment ses­sions, help with paper­work to report their long-term con­di­tion, or organ­is­ing trans­port,” he adds.

A patient’s behav­iour and per­son­al­i­ty also affect the way they relate to health­care. They might, for exam­ple, have a habit of research­ing health-relat­ed infor­ma­tion online, tend to engage in risky behav­iour, have a pos­i­tive or neg­a­tive out­look, be intro­vert­ed, extro­vert­ed, etc. “All of these fac­tors influ­ence the patient’s response,” Min­vielle says. In order to take them into account, he is devel­op­ing patient-cen­tred man­age­ment models.

These tools require data, and AI can facil­i­tate data col­lec­tion and help build rel­e­vant pro­files. “But in health, much of the data is high­ly com­plex,” he says. “We’ll need new pro­fes­sion­als, such as coor­di­nat­ing nurs­es, to make sense if it.” Con­nect­ed objects, such as pill dis­pensers, could also be used; blood tests could be pre­scribed to check for metabo­lites so as to mon­i­tor patient com­pli­ance. “These solu­tions must go hand in hand with eth­i­cal con­sid­er­a­tions. How far can we go with­out intrud­ing on patients’ lives?”

Pre­serv­ing patient autonomy

When it comes to ethics, vig­i­lance is required on sev­er­al fronts. “Per­son­alised med­i­cine must not restrict access to care,” Min­vielle stress­es. “There are risks per­tain­ing to access to per­son­al­i­sa­tion, as well as minor­i­ty rep­re­sen­ta­tion in our mod­els.” In a sys­tem that adapts patient care path­ways to their spe­cif­ic pro­file, dig­i­tal mod­els could be used to organ­ise care options into broad cat­e­gories. This would stream­line care path­ways, restrict­ing them in a way that may omit less pop­u­lar pref­er­ences with­in the patient population.

Last­ly, per­son­al­i­sa­tion must not come at the expense of patient auton­o­my. Min­vielle believes that “patients should be able to opt out of care path­way per­son­al­i­sa­tion, and choose their own per­son­al­i­sa­tion path­way, even if it is not optimal.” 

Human and arti­fi­cial intelligence

Giv­en what’s at stake, such a pro­gram can­not sole­ly rely on dig­i­tal tools. Human intel­li­gence is required. Min­vielle reminds us that “AI can­not yet detect its own errors, where­as a human will imme­di­ate­ly notice if a patient has been giv­en the wrong advice.”

Will this sys­tem save mon­ey? “It remains to be seen. On the one hand, it should reduce waste; on the oth­er, we would be cre­at­ing a more com­plex sys­tem of care.” Min­vielle agrees that more research is need­ed, and tru­ly per­son­alised care path­ways are still some time off.

“The main dif­fi­cul­ty is bring­ing every­thing togeth­er”, he says. The indi­vid­u­al­i­sa­tion mod­el is based on research from a num­ber of fields: pub­lic health (inte­grat­ing socioe­co­nom­ic deter­mi­nants); psy­chol­o­gy (analysing patient behav­iour and demands); and med­ical research (exam­in­ing the effi­ca­cy of treat­ments on var­i­ous patient pro­files). Yet “all of these groups are work­ing in iso­la­tion. If we are to devel­op a cul­ture of per­son­al­i­sa­tion with­in the health sys­tem, we have to demon­strate its feasibility.”

Our world explained with science. Every week, in your inbox.

Get the newsletter