Towards digitally-enhanced psychiatry

Digit­al tech­no­logy is already trans­form­ing the world of health­care, open­ing up unpre­ced­en­ted pro­spects. From per­son­al­ising care to redu­cing hos­pit­al over­crowding, the expect­a­tions are aston­ish­ing. But for this revolu­tion to live up to its prom­ise, health­care pro­fes­sion­als need to both adopt these tools, while rethink­ing how they are used. The i3-CRG labor­at­ory, headed by Étienne Min­vi­elle at École Poly­tech­nique (IP Par­is), has launched a series of sem­inars on the integ­ra­tion of digit­al tech­no­logy in health­care. One of the most eagerly awaited top­ics has been men­tal health, an area where tech­no­lo­gic­al advances could really change the game. Pro­fess­ors Pierre-Alex­is Geof­froy and Jean-Bap­tiste Mas­son take a look back at the sem­in­ar ses­sion devoted to psy­chi­atry, which was also atten­ded by Guil­laume Couil­lard, Dir­ect­or Gen­er­al of the Par­is Psy­chi­atry & Neur­os­ciences GHU, and Raphaël Gail­lard, Pro­fess­or of Psy­chi­atry at Uni­versité Paris-Cité.

“Giv­en the eco­nom­ic con­straints on the sys­tem, it is impossible to ima­gine that every­one will be going to hos­pit­al in the future,” explains Pro­fess­or Pierre-Alex­is Geof­froy, a psy­chi­at­rist at GHU Par­is. “One per­son told me that her son had developed sleep dis­orders quite early on, and that she had tried to have him mon­itored by a child psy­chi­at­rist, to no avail. In the end, six years later, he was dia­gnosed with schizo­phrenia. But child psy­chi­atry will nev­er be able to treat all chil­dren with sleep dis­orders or anxi­ety dis­orders. And so these digit­al solu­tions, which are also less expens­ive, will be there to adapt to the level of intens­ity of care that needs to be put in place, but also to the level of inter­ven­tion that we can offer.”

“We talk about digit­al health­care as if it were sci­ence fic­tion,” admits Pierre-Alex­is Geof­froy. “But the truth is that digit­al tech­no­logy is already here, and we all use it on a daily basis.” Today, 95% of prac­ti­tion­ers com­plete their patients’ files using digit­al tools. “When we write pre­scrip­tions, for example, AI can already tell us about pos­sible drug inter­ac­tions,” adds the pro­fess­or. “All of this is based on sci­entif­ic lit­er­at­ure, which is updated in real time.” The point, then, is not so much to take stock of the tools already avail­able and in use in the world of psy­chi­atry. Rather, it’s about look­ing ahead to the pos­sible changes that this world will under­go as a res­ult of the tech­no­lo­gic­al advances that could be made.

“For a long time, we were try­ing to trans­pose vari­ous scales and assess­ments that we used in real life into the digit­al world. But that did­n’t work because adapt­ing to new digit­al tools requires us to rethink everything,” explains the psy­chi­at­rist. Exist­ing health applic­a­tions are an example. Although few­er in num­ber than well­ness apps, only 15% of them fol­low a sci­entif­ic approach, mean­ing they are based on a study with proof of effect­ive­ness. “When we look at using this type of solu­tion, we also real­ise that there is a prob­lem with com­pli­ance,” he con­tin­ues. “Only 30% of people com­plete their pro­grammes. So, the ques­tion is: how do we devel­op this type of solu­tion properly?”

“In men­tal health, we are for­tu­nate to have very robust mod­els,” points out Pierre-Alex­is Geof­froy. “To devel­op this type of solu­tion in addict­o­logy, for example, we need to adapt mod­els with a known sci­entif­ic approach to digit­al tech­no­logy.” For example, an applic­a­tion to encour­age people to give up an addic­tion, such as smoking, needs to be developed on the basis of an exist­ing mod­el, such as Prochaska and Di Clem­en­te’s mod­el of pre­par­a­tion for change. 

“If my solu­tion fol­lows the logic of ‘one size fits all’ we will estab­lish a bal­ance of the pros and cons of what stop­ping smoking will bring to the patient. How­ever, if the patient is already in a relapse phase, this will not speak to him. They will need much more tan­gible pro­pos­als. Prochaska and Di Clem­en­te’s mod­el is there­fore import­ant for determ­in­ing what phase the patient is in, and there­fore what type of fol­low-up they will need. If they are at the con­tem­pla­tion stage, they will need motiv­a­tion­al inter­views to try and cla­ri­fy with them what they want and what they are pre­pared to do. If they are already in the action stage, we need to organ­ise with­draw­al with them. And if they are at the relapse stage, we need to ask them about what they have already done, to determ­ine what has worked well and what has not. This is essen­tial to ensure that patients are com­mit­ted to the solu­tion and that they don’t stop everything after 5 minutes because the applic­a­tion does­n’t meet their needs.”

Accord­ing to the pro­fess­or, the primary interest of digit­al tech­no­logy lies in provid­ing an addi­tion­al ser­vice to the doc­tor’s prac­tice. Offer­ing patients more reg­u­lar mon­it­or­ing does not mean mak­ing them inde­pend­ent in their approach, or even ask­ing them to be too involved in this mon­it­or­ing. “Just under 10 years ago, the Mon­arca I¹ study had an inter­est­ing idea. How­ever, today, this solu­tion already seems ‘has-been’,” he argues. “The idea was to self-mon­it­or patients with bipolar dis­order, in order to pre­dict when they might relapse. To do this, the 61 patients were asked to record their symp­toms on depres­sion scales.” The authors of this study showed that the more depressed patients were, the less they inter­ac­ted with the med­ic­al team. Con­versely, the more man­ic the patient, the great­er the num­ber and dur­a­tion of calls.

“The res­ults were suf­fi­ciently clear for the patient’s con­di­tion to be eas­ily clas­si­fied,” con­firms Pierre-Alex­is Geof­froy. “This led the authors to con­clude that smart­phone applic­a­tions were val­id for real-time patient mon­it­or­ing.” How­ever, a second study fol­lowed this one, with far less favour­able con­clu­sions. “In this second study, the authors decided to keep all the patients, even those who had stopped using the pro­posed solu­tion,” explains the pro­fess­or. “The res­ult was that there was no sig­ni­fic­ant effect of self-mon­it­or­ing, and the authors even observed that not­ing down depress­ive symp­toms every day worsened the patient’s men­tal state.” 

“This type of mon­it­or­ing is not inten­ded to replace the doc­tor, but rather to offer an addi­tion­al ser­vice that he or she can pre­scribe.” So, it’s a new tool avail­able to doc­tors to ensure that their treat­ment is effect­ive. Because, in addi­tion to the pos­sib­il­ity of remote patient mon­it­or­ing, digit­al tech­no­logy also offers thera­peut­ic solu­tions. “I often use the example of ther­apy using aug­men­ted vir­tu­al real­ity. I’m a psy­cho­ther­ap­ist and I have a patient with a pho­bia of cock­roaches. I can work with him on expos­ure to insects by pro­ject­ing insects around his hand using vir­tu­al reality.”

“Accept­ance of this type of tool is a major issue. When we ask stu­dents wheth­er they would prefer to be treated using a digit­al solu­tion or by a real per­son, only 9% of them choose the digit­al option,” notes Pierre-Alex­is Geof­froy. “We there­fore need to provide evid­ence of the effect­ive­ness of these meth­ods to unlock these bar­ri­ers. Digit­al tech­no­logy is not yet present in my prac­tice, because digit­al solu­tions for provid­ing real-time patient data are not yet avail­able. I dream of one day, in my prac­tice as a psy­chi­at­rist, in addi­tion to my tra­di­tion­al prac­tice, hav­ing digit­al argu­ments to help me make decisions. So, I think that sup­port will be de facto great­er when such solu­tions, proven to be effect­ive, are available.”

This type of mon­it­or­ing is not inten­ded to replace the doc­tor, but rather to offer an addi­tion­al ser­vice that he or she can prescribe

Pro­fess­or Jean-Bap­tiste Mas­son, a research­er at Insti­tut Pas­teur, ques­tions the meth­od­o­lo­gies used to assess the effect­ive­ness of digit­al tech­no­logy in med­ic­al con­texts. “At some point, if we want to prove that some­thing is effect­ive, we’ll have to carry out stat­ist­ic­al tests,” he says. “We’ll have to com­pare one group with anoth­er, where­as the human mind is not eas­ily put into cat­egor­ies. It is true that digit­al tech­no­logy provides an enorm­ous quant­ity of data and enlarges the con­trol group. In psy­chi­atry, the lar­ger the group, the more het­ero­gen­eous it will be. Sub-groups will emerge, and com­par­is­ons will become less spe­cif­ic. So, one dif­fi­culty will be in trans­pos­ing con­vin­cing res­ults from a small sample of people to a lar­ger sample for which the res­ults will be less reli­able,” he adds. This lim­it­a­tion is also reflec­ted in the num­ber of para­met­ers that can be meas­ured digit­ally: the more para­met­ers and data that are stud­ied, the great­er the chance of cor­rel­a­tions, due to ran­dom­ness, without them being sig­ni­fic­ant.” These meth­od­o­lo­gic­al lim­it­a­tions are still hold­ing back the val­id­a­tion and adop­tion of these digit­al solu­tions by practitioners.

The mar­riage between digit­al tech­no­logy and men­tal health there­fore opens up fas­cin­at­ing pro­spects. But for these solu­tions to win the con­fid­ence of prac­ti­tion­ers and patients, they must be accom­pan­ied by sol­id proof of their effect­ive­ness. And that’s not easy to do. As Pro­fess­or Geof­froy sums up, “digit­al tech­no­logy is not inten­ded to replace the doc­tor, but it can become a valu­able ally in our prac­tice.” With rig­or­ous clin­ic­al research and gradu­al adop­tion, digit­al tech­no­logy could well redefine the approach to psy­chi­at­ric care, mak­ing treat­ments more access­ible, per­son­al­ised and effect­ive. How­ever, the final hurdles to the suc­cess­ful devel­op­ment of these solu­tions will have to be overcome.

Pablo Andres