3D printing in hospitals : open innovation during the COVID-19 crisis

When COVID-19 hit, Grea­ter Paris Uni­ver­si­ty Hos­pi­tals (AP‑HP) faced dire shor­tages of per­so­nal pro­tec­tive equip­ment (PPE) and medi­cal sup­plies. In March 2020, an AP-HP sur­geon and the CEO of a 3D prin­ting medi­cal ser­vices start-up, pro­po­sed set­ting up a “3D farm” of 60 prin­ters inside one of network’s hos­pi­tals. Backed by €1.7m in pri­vate funds, the plat­form was up and run­ning within weeks—a tes­ta­ment to what is pos­sible when urgen­cy removes bureau­cra­tic fric­tion. Due to the nature of the chal­lenge of sup­plying such an orga­ni­sa­tion, the plat­form inno­va­ted through know­ledge flows across the esta­blish­ments, making it an inter­es­ting case stu­dy of an “open inno­va­tion” approach.

This orga­ni­sa­tio­nal para­digm dates back to the ear­ly 2000s¹, in the wake of the infor­ma­tion revo­lu­tion. Then, Ber­ke­ley Pro­fes­sor Hen­ry Ches­brough posi­ted that infor­ma­tion had become so wides­pread that inno­va­tion could sprout from anyw­here, and com­pa­nies that wan­ted to stay ahead of the curve nee­ded to open up to exter­nal ideas. Open inno­va­tion has demons­tra­ted its poten­tial in the health sec­tor, par­ti­cu­lar­ly during the rapid deve­lop­ment and tes­ting of a new COVID-19 vac­cine². Howe­ver, a hos­pi­tal set­ting is not a blank slate and requires spe­ci­fic know­ledge flows within a dense net­work of public and pri­vate orga­ni­sa­tions. Unders­tan­ding the flow of infor­ma­tion and inter­ac­tions within such a set­ting could help enhance inno­va­tion within this com­plex ecosystem.

The AP-HP ini­tia­tive pro­vides a rare case stu­dy to ans­wer this cen­tral ques­tion : can open inno­va­tion work within a hospital ?

Invest­ment in the 3D prin­ting plat­form qui­ck­ly paid off. Over nine months (April to Decem­ber 2020), more than 33,000 units rol­led off the prin­ters’ banks. The majo­ri­ty of these (87%) were pro­tec­tive equip­ment, but there were also a sur­pri­sing range of other objects : cus­to­mi­sed sur­gi­cal ins­tru­ments, sup­ports for oxy­gen bot­tles, repla­ce­ment parts for clea­ning machines, and even hea­drests for intensive‑care patients kept prone for long spells.

In a sys­tem where “inno­va­tion” is often seques­te­red in spe­cia­list silos, the plat­form ope­ned the door to all comers. Engi­neers were embed­ded on‑site, which meant they could glean ideas from not just doc­tors, but also nurses, the­ra­pists, tech­ni­cians, and main­te­nance staff, rever­sing the typi­cal hie­rar­chi­cal struc­tures. The close col­la­bo­ra­tion with an exter­nal 3D prin­ting star­tup, nota­bly its engi­neers, its sup­plier, and its aca­de­mic part­ners (for ins­tance, French engi­nee­ring schools), bol­ste­red inno­va­tion in a way that would not have been achie­vable through sole­ly in-house inno­va­tion, as contrac­ting out these engi­neers would have been too expensive.

Proxi­mi­ty allo­wed for tacit know­ledge trans­fer : by wal­king from a work­bench to a ward, engi­neers could create pro­to­types in the space of a few hours rather than months, and these could be ite­ra­ted rapid­ly. This not only met urgent needs, but often unco­ve­red hid­den chal­lenges, and crea­ted moments for spillo­ver inno­va­tion, some­times repur­po­sing devices across departments.

Beyond new pro­ducts, the plat­form brought mana­ge­rial inno­va­tions (on-site inter­dis­ci­pli­na­ry col­la­bo­ra­tion) and pro­cess inno­va­tions (loca­li­sed manu­fac­tu­ring that cut lead times from months to days, redu­ced costs, and avoi­ded sto­ckouts). It also seems to have impro­ved work satis­fac­tion by empo­we­ring staff to solve prac­ti­cal problems.

A tur­ning point came in Sep­tem­ber 2020. Without the focu­sing power of the glo­bal pan­de­mic, and away from the pres­sing need to pro­vide PPE, it was time to ima­gine how the plat­form could work in the long run. Cham­pions ima­gi­ned seve­ral ways that it could create value in the long run, some more finan­cial­ly tan­gible than others. These ran­ged from pro­du­cing regu­la­ted medi­cal devices, to licen­sing inno­va­tive 3D desi­gns to other hos­pi­tals, to desi­gn paten­ting. Per the plan, the models’ licen­cing fees were also to be sha­red with the inven­tor, which could act as a magnet for pro­mi­sing phy­si­cians that could coun­ter­ba­lance AP-HP’s renown admi­nis­tra­tive com­plexi­ty and limi­ted recruit­ment budget.

In-house pro­duc­tion was also fore­cas­ted to be cost-cut­ting, as there were ear­ly signs that it could help reduce some pro­cu­re­ment costs, improve care, retain control of valuable patient data tied to cus­to­mi­sed devices and enhance tea­ching and research. Trai­ning could also be offe­red to other hos­pi­tals to roll out a simi­lar plat­form. One of the most advan­ced plans was the 3D Print for Afri­ca pro­ject, which ima­gi­ned trai­ning staff and hel­ping in the imple­men­ta­tion of 3D prin­ting hos­pi­tals in five west Afri­can hospitals.

Buoyed by its ver­sa­ti­li­ty, per­ma­nent ambi­tions were sket­ched out to increase the platform’s role, while impro­ving its value cap­ture. In theo­ry, these plans met the three­fold mis­sion of a tea­ching hos­pi­tal : patient care, research, and edu­ca­tion. In prac­tice, howe­ver, seve­ral orga­ni­sa­tio­nal and finan­cial bar­riers hin­de­red pro­gress. Phase‑two invest­ment was esti­ma­ted to require €1.4m, with annual run­ning costs valued at €1.2m. While senior lea­ders ack­now­led­ged the patient‑care bene­fits of the pro­po­sed phase 2 plans, it was dif­fi­cult to sup­port a clear revenue‑or‑savings model. The hospital’s pro­cu­re­ment agen­cy, also cla­shed with the pro­ject team over medi­cal device regu­la­tion, nota­bly fea­ring that pro­to­types pro­du­ced by the plat­form would fail to meet com­pliance stan­dards lin­ked to medi­cal equip­ment. There was lit­tle visi­bi­li­ty on the facility’s sca­ling up, as the plat­forms’ ini­tial small size, which had pro­ven cata­ly­tic to inno­va­tion, would also limit development.

Three essen­tial condi­tions have been iden­ti­fied to make such inno­va­tion sus­tai­nable : a culture of inno­va­tion, ali­gn­ment of inter­ests, and cla­ri­ty of expec­ted rewards.

His­to­ri­cal in-figh­ting also pro­ved to be an issue — in prac­tice, during the first phase of the platform’s life, hos­pi­tals clo­sest to the 3D farm were most like­ly to use the ser­vice. As the second phase was being exa­mi­ned, sta­ke­hol­ders expres­sed fears that AP-HP’s repu­ta­tion for inter­nal com­pe­ti­tion, refer­red to by one inter­vie­wee as “turf wars”, could limit access of some of the net­works’ par­ti­ci­pants to the machines. With no consen­sus on objec­tives and regu­la­to­ry doors clo­sed, the contract with the exter­nal part­ner was ended ear­ly in 2021.

Des­pite its limitations—including the excep­tio­nal nature of the COVID-19 pan­de­mic, the study’s small sample size (a case stu­dy), and the spe­ci­fi­ci­ty of this organisation’s culture—this research sheds light on the poten­tial of open inno­va­tion in a hos­pi­tal set­ting. It also demons­trates how a pro­ject can be qui­ck­ly aban­do­ned if the value cap­ture is not qui­ck­ly and clear­ly defined.

Our research iso­lates three vital condi­tions for making such inno­va­tion sus­tai­nable. First is an inno­va­tion culture. Bureau­cra­tic rigi­di­ty, hie­rar­chi­cal deci­sion-making, and finan­cial constraints arres­ted momen­tum until lea­der­ship appoin­ted a dedi­ca­ted direc­tor and pro­ject lead. Without com­mit­ted struc­tures and cham­pions, exter­nal know­ledge flows struggle to take root. Second is ali­gn­ment of inter­ests. All inter­nal sta­ke­hol­ders — not just cli­ni­cians, but also admi­nis­tra­tors and sup­port ser­vices — should be able to see their roles and rewards clear­ly. In AP‑HP’s case, misa­li­gned com­pe­ten­cies and per­cei­ved turf threats under­mi­ned pro­gress. Third is cla­ri­ty on expec­ted rewards. The project’s ori­gin as an emer­gen­cy PPE fac­to­ry blur­red its poten­tial long-term impact. Mana­ge­ment sought finan­cial returns ; cham­pions also valued harder‑to‑measure gains such as effi­cien­cy, morale, and data sove­rei­gn­ty. Absent consen­sus, the ini­tia­tive lost poli­ti­cal and fis­cal support.

The ori­gi­nal “3D farm” did leave a lega­cy. In 2024, AP‑HP laun­ched a new addi­tive manu­fac­tu­ring plat­form. The pro­ject lear­ned from its ear­lier cau­tio­na­ry tale — this time it has for­mal gover­nance, a net­work of thir­teen other public research hos­pi­tals, and a clear mis­sion to pro­duce both medi­cal and main­te­nance equip­ment inside the health system.