PROOF FOR CHECKING PURPOSES ONLY

International Journal for Quality in Health Care, 2020, (0), 1–6

doi:10.1093/intqhc/mzaa103
Advance Access Publication Date: xx xxx xxx
Perspectives on Quality

Perspectives on Quality

Human factors engineering for medical devices:

European regulation and current issues
SYLVIA PELAYO1 , ROMARIC MARCILLY1 AND TOMMASO BELLANDI2
1
Univ. Lille, CHU Lille, ULR 2694—METRICS: Évaluation des technologies de santé et des pratiques médicales,
INSERM-CIC-IT 1403/Evalab, F-59000, Lille, France and 2 Director Patient Safety Unit, Tuscany North West Trust,
Regional Health Service
Address reprint requests to: Sylvia Pelayo, CIC-IT 1403, Université de Lille/CHU de Lille/Inserm, Institut Cœur Poumon, Bd du
Professeur jules Leclercq, 3ième étage Aile Est, CS 70001, 59037 Lille Cedex. Tel./Fax: +33 320 444 883;
Fax: +33 320 446 715; E-mail: [email protected]

Received 25 August 2020; Editorial Decision 20 August 2020; Revised 15 August 2020; Accepted 26 August 2020

Abstract
A large proportion of the patient injuries or deaths attributable to medical device (MD) misuse
can be eliminated and/or mitigated by adopting an effective human factors and ergonomics (HFE)
approach. The implementation of a usability engineering process is now mandatory for MD manu-
facturers seeking to obtain the European Union’s CE Mark. Here, we describe the European Union’s
HFE regulation and highlight the challenges faced by (i) manufacturers implementing this regula-
tion and (ii) regulatory bodies charged with assessing the compliance of usability files. In Europe,
95% of MD manufacturers are small- and medium-sized enterprises; compliance with the CE Mark
regulations is a real challenge to their competitiveness. Levels of knowledge about HFE vary greatly
from one regulatory organization to another, which can sometimes lead to very different expec-
tations. We also present the specific use-related risk management approach required by the HFE
regulation. Lastly, we focus on the limitations of the HFE regulation for MDs and on future HFE chal-
lenges in further reducing and/or eliminating MD use errors. The main challenge is the need to go
beyond technology design and the premarket assessment and to look at the postproduction stage;
the coupling between an MD and a sociotechnical system can lead to consequences that were not
predicted during the design process. This implies the need to consider the emerging properties of
technologies in use by involving all the stakeholders.

Key words: ergonomics, risk, human factors, medical device

Introduction are to (i) describe the genesis of MD HFE regulation in Europe,

(ii) describe current issues in the HFE regulation, (iii) specifically
Over the last few decades, innovative medical devices (MDs) have
address the difficulties raised by the use-related risk management
become increasingly important for better patient care and public
approach to MDs required in the European Union and (iv) discuss
health [1]. Technological innovation is always considered to be a key
future challenges in HFE research on MDs.
factor in healthcare; it promises ever better care and, in most cases,
cost-effectiveness. Surprisingly, the human factors and ergonomics
(HFE) community has accorded little importance to MDs. The num- The genesis of the HFE regulation in the field of
ber of HFE publications referring explicitly to MDs has only started MDs: from user error to use error
to grow over the last 10 or so years. The PubMed database contains The MD market is highly innovative, very large and very diverse;
fewer than 300 publications on this topic, starting in 1983 (Figure 1). it encompasses more than 20 000 types of product, ranging from
HFE studies of MDs are strongly influenced by the European single-use consumables (dressings, compresses, etc.), personal pro-
Union’s CE mark regulations. The objectives of the present article tective equipment (PPE), implants (breast prostheses, pacemakers,

© The Author(s) 2020. Published by Oxford University Press on behalf of International Society for Quality in Health Care. All rights reserved.
For permissions, please e-mail: [email protected] 1
 PROOF FOR CHECKING PURPOSES ONLY
2 Pelayo et al.

Figure 1 In June 2020, the PubMed database (titles and abstracts) was searched with the following keywords: ‘Human Factor*’ OR ‘Ergonomic*’ OR ‘Usability’
AND ‘Medical device*’.

etc.), infusion pumps, medical beds, diagnostic devices, reagents and practices and changes must be anticipated during the MD design
lab robots. The European Union’s Regulation 2017/745 defines an phase. This system-based approach to designing tools is widely
MD as ‘any instrument, apparatus, implement, machine, appliance, acknowledged by the HFE community as being essential. The
implant, reagent for in vitro use, software, material or other similar approach requires an understanding of all the factors related to how
or related article, intended by the manufacturer to be used, alone or an MD is used. However, it is known that some issues go beyond
in combination, for human beings, for one or more of the specific the MD per se and concern (for example) collaboration on shared
medical purpose(s) […]’ [2]. The medical indication claimed for an processes—even when people are not using the MD—or institution’s
MD is thus critical. broader organizational policies.
However, the way an MD is used by its users has an impact on In Europe, EN 62366 [8] is the harmonized HFE standard for
its clinical benefit and, more broadly, on the quality of care or even MDs. It was adapted from the international standard IEC 62366
patient safety. The HFE and patient safety literatures abound with (Figure 2) and came into force in 2007—a few years after the USA
stories of death due to a use error, i.e. ‘a user action or lack of action adopted similar regulations [10, 11]. This European standard was
while using the MD that leads to a different result than that intended then significantly revised and reorganized [3,12] so that it was more
by the manufacturer or expected by the user’ [3]. These problems in line with the 2011 draft guidance on human factors of the FDA
are more common than most people realize and are not limited to a [13]. These HFE standards are standards for the analysis, design,
small number of complex MDs. A well-known illustration is the case verification and validation of safety-related usability through the MD
of Denise Melanson—a Canadian who died after an infusion pump development cycle.
was programed by a nurse to deliver a medication over 4 hours rather It should be noted that the terms ‘human factors engineering’ and
than 4 days [4]. Another example involved 34 German patients who ‘usability engineering’ are considered to be synonyms in the above-
required re-operation because their knee prostheses were implanted mentioned standards and can be used interchangeably [14]; we will
in the absence of bone cement [5]. use the term ‘HFE’ hereafter.
If an MD has a problem, the user was always the first to be blamed
[6]. Indeed, if the device has performed technically as intended, then The human factors engineering regulation for MD
the user is typically accused of lacking training, being inattentive
and the corresponding method: a use-related risk
or incompetent or not reading the instructions. However, there is a
management approach
growing recognition that ‘user error’ (‘human error’) is not indepen-
dent of the broader work system within which users act and interact The HFE process is based on human-centered design for interac-
[7]. Under this view, errors are considered as a consequence of a fail- tive systems [15]. This method is intended to make systems usable
ure of one or several parts of the system, rather than being entirely and useful by (i) focusing on users and their needs and (ii) applying
attributable to the individual involved. The European Union’s regu- existing HFE knowledge and techniques [16]. This approach pro-
latory system for MDs has acknowledged this conceptual change by motes effectiveness and efficiency and improves human well-being,
replacing the term ‘user error’ by the term ‘use error’ [8]. Design- user satisfaction, accessibility and sustainability.
induced use errors are therefore seen as a possible cause of patient The objective of the HFE process (as required by the MD reg-
injury or death. This change in mindset raises the question of the ulations) is to maximize the likelihood that new MDs will be safe
appropriateness of design choices for the use of an MD by specific and effective for the intended users, applications and context of
users and in specific contexts [9]. use [12, 17]. The HFE process is therefore closely intertwined with
The change also highlights the importance of taking account of the risk management process (Figure 3); it evaluates and mitigates
actual work procedures, habits and collective aspects of use, as usability-induced use errors that could result in risks for patients
well as the characteristics of the work environment within which and healthcare staff. With this goal in mind, manufacturers have to
humans and technologies interact. The MD will impact work minimize use-related hazards and risks and then confirm that these
 PROOF FOR CHECKING PURPOSES ONLY
HFE for medical devices • Perspectives on Quality 3

Figure 2 The main HFE standards and guidance applicable to MDs in Europe and in the USA.

efforts are successful and that users can use the device safely and cost–benefit ratio. Educating technical and managerial staff about
effectively. The manufacturer has to perform a usability evaluation HFE provides companies with long-term benefits [20]. Whether a
of the reasonably foreseeable hazard-related use scenarios and eval- company should hire a full-time HFE expert will depend on the com-
uate the residual risk. If the HFE process has been complied with, pany size, and the complexity and types of MDs being developed. In
then the usability of an MD as it relates to safety is presumed to this respect, the learned societies for HFE in Europe created a pro-
be acceptable—unless there is objective evidence to the contrary. fessional certification system in 1992: the Centre for Registration of
The design control process must include an analysis of the risks European Ergonomists is a nonprofit organization that certifies HFE
associated with device use and the measures implemented to reduce professionals as ‘European ergonomists’ (‘Eur.Erg.’) and is endorsed
those risks. by the International Ergonomics Association.
The HFE effort can be tailored with regard to the risks created by
usability problems and the risks’ severity. The MD function is also
important; a technically simple MD might pose greater use-associated
Human factors engineering regulation for MDs
risks and call for much more HFE studies than a more complex one. and current issues
When the development effort modifies an existing MD, a smaller- Like all manufacturing industries, MD manufacturers are subject to
scale HFE effort can be then focused on the modified elements and high costs and time pressure. The MD market is global, dynamic
the latter’s effects on use of the MD. and innovative, with short product life cycles. In Europe, the MD
The use-related risk prevention process is not easy to understand sector comprises around 32 000 companies, 95% of which are small-
and implement. It is well known that the normative documents and and medium-sized enterprises [21]. Maintaining an MD manufac-
guidelines that are supposed to facilitate the implementation of the turer’s competitiveness and capacity for innovation is a real challenge
process are themselves affected by usability problems [18]. Based on when it has to comply with the regulations for CE marking. More-
these documents, manufacturers are not able to imagine the HFE pro- over, new European Union regulation on MDs [2] is set to come into
cess that is truly expected of them: they may think that they already force in the spring of 2021; it will tighten requirements and require
have implemented the HFE process by applying the risk management high levels of clinical evidence. Small- and medium-sized enterprises
process. Some manufacturers do not understand what use-related often have limited internal resources and skills—especially when it
risks are. From the perspective of the notified body (NB), the HFE comes to HFE. Furthermore, the requirement related to MD HFE is
documents do not provide relevant criteria for assessing compliance only one of the regulations with which the MD manufacturers must
with the essential requirements. Implementation of the HFE process comply. The impact of the new MD regulations and the economic
clearly requires more than the application of a standard or guideline. crisis associated with the coronavirus disease 2019 pandemic might
Even though the HFE documents must be usable and helpful to their be fatal for many manufacturers but represents a unique opportunity
users (e.g. manufacturers and NBs), they cannot alone substitute for for others.
in-depth expertise in HFE. From the regulatory bodies’ perspective, the European market is
Manufacturers must therefore improve their skills in the field of significantly fragmented because many countries have their own set
HFE. This raises the question of the company’s initial level of matu- of rules [22]. Even though the regulations are harmonized within
rity in HFE [19] and then the level required to achieve a satisfactory the European Union, their implementation is left to the discretion
 PROOF FOR CHECKING PURPOSES ONLY
4 Pelayo et al.

Figure 3 Steps in the HFE process and their relationships with applicable parts of the risk management process, as described in the IEC 62366–1:2015 usability
standard.

of the Member States. NBs are authorized by a Member State to strategy for patients, caregivers and healthcare professionals who will
determine whether a marketable product complies with regulatory have safe innovations with proven clinical benefit. In this context,
requirements. The level of emphasis on HFE for MDs and the cor- HFE cannot be considered as an isolated process; the overall strategy
responding skills and knowledge vary markedly from one NB to for a MD needs to be well thought through. The various studies must
another. After several scandals [23, 24] and ahead of the imminent be linked so that each contribution is leveraged: for instance, clinical
introduction of new MD regulation [2], the requirements with which studies can be designed so that they also generate useful usage data,
NBs must themselves comply have been radically reinforced. A num- and usage studies can help to avoid methodological biases linked to
ber of NBs have closed down in the last 5 years, and only 20 are MD use and thus refine clinical study protocols. If possible, clinical
currently designated in the European Union’s regulation 2017/745 and usage studies should ideally be combined.
on MDs [25]. On a more general level, an optimized HFE effort must be scoped
In this context, one can legitimately question the actual impact of at the company level by considering all the various MDs that the
these regulatory enhancements (particularly the HFE regulation for company markets. If the MDs marketed by a company are relatively
MDs) on patient safety and quality of care. There has been a clear similar in terms of the indication, context of use and/or target users,
increase in the manufacturers’ knowledge of the HFE process. Most the HFE studies must be built up through the company’s entire MD
of the companies no longer question the value of this approach or the portfolio in order to build and capitalize on a set of benchmarks. The
meaning of ‘use errors’, as was the case when the regulation came into key challenge is to be able to exploit the user research analyses and
force [26]. It is currently imperative for manufacturers to ensure an results and optimize the testing on several MD benchmarks. Once a
effective overall market access in order to access the market as quickly company has developed, understood and formalized HFE skills and
as possible. This requires to (i) be able to tailor the HFE effort to methodologies for some current products, it can leverage them for
the type of MD and the associated use-related risks and (ii) integrate other products and for the commercialization of new products. How-
efficiently the HFE process with all the other necessary processes ever, the greater the diversity of the company’s products, contexts of
(e.g. technology maturation, preclinical evaluation and health eco- use and indications, the more difficult it will be to capitalize on the
nomics assessments). This winning strategy will also be a winning corporate HFE program.
 PROOF FOR CHECKING PURPOSES ONLY
HFE for medical devices • Perspectives on Quality 5

Future challenges: going beyond the premarket provide detailed information about how the MD should be integrated
evaluation into the hospital’s workflow and staffing system, based on an HFE
analysis of how end users act and collaborate. The hospital stake-
The European Union’s MD regulations constitute a powerful means
holders (e.g. the hospital information system staff, human resources
of promote HFE studies in this field and thus preventing certain use-
staff, heads of clinical units, and general management) should then
related risks. However, this regulatory framework is obviously not a
commit to certain human resources, training and specific implemen-
‘silver bullet’ that will ensure that all MDs are perfectly safe to use.
tation procedures to comply with the defined requirements. In the
The HFE regulation also poses major problems especially because of
future, regulatory bodies might require even more things, such as
its coverage of the MD life cycle. Receipt of the CE mark for MDs
periodic retraining and periodic usage inspections. This may seem
requires postmarketing surveillance to be part of the overall risk man-
restrictive from the point of view of the user sites, but some of them
agement process; data related to use error must be analyzed in order
are looking for help, guidelines to support the implementation of
to identify the user interface’s strengths and shortcomings. However,
complex tools, and this is ultimately a winning strategy for all. For
postmarketing surveillance does not feature in the HFE standard IEC
home-based MDs, taking a system-based approach is more difficult
62366–1, which makes it difficult to consider what will happen after
to consider from a regulatory perspective.
the market launch in terms of MD use.
In fact, it is when a device is coupled to the sociotechnical system
that unintended consequences (which could not have been predicted
in the design process) emerge [27]. Whereas HFE methods like proac- References
tive risk assessment can identify and mitigate design vulnerabilities, 1. World Health Organization. Medical device—Full definition. US: world
they may not evidence emergent issues. The concept of ‘dynamic Health Organization; 2020. [updated June 14] Available from: http://www.
safety’ is fundamental for appropriate postmarketing surveillance. who.int/medical_devices/full_deffinition/en/
Dynamic safety is both an emergent property of a resilient system 2. Regulation (EU) 2017/745 of the European Parliament and of the Coun-
and the result of adaptive interactions between humans and tech- cil of 5 April 2017 on medical devices, amending Directive 2001/83/EC,
nologies in a context of use; it relies on workers’ tacit knowledge Regulation (EC) No 178/2002 and Regulation (EC) No 1223/2009 and
repealing Council Directives 90/385/EEC and 93/42/EEC.
and ability to constantly adjust daily practices as the context of use
3. International Electrotechnique Commission. Medical devices—part 1:
and job demands change [28]. Let us take the example of PPE. Even
application of usability engineering to medical devices. Geneva: Inter-
though PPE has been used routinely for decades in various epidemic
national Electrotechnique Commission; 2015. 96p. Report No.: IEC
settings, this equipment proved unsatisfactory during the COVID-19 62366-1:2015.
pandemic when outdated designs caused use problems [29]. 4. Institute for Safe Medication Practices Canada. Fluorouracil Incident Root
Going beyond the MD usability during the development phase Cause Analysis. Toronto: Institute for Safe Medication Practices Canada.
and analyzing the MD long-term, real-life use is essential. This has 2007, 75p.
major implications for device design, including the need to go beyond 5. Fakler JK, Hase F, Böhme J et al. Safety aspects in surgical treatment of
technology design and consider emerging properties of technology in pathological fractures of the proximal femur – modular endoprosthetic
use [30]. Product design should be an ongoing process rather than replacement vs. intramedullary nailing. Patient Saf Surg 2013;7:37.
6. Dekker S. The criminalization of human error in aviation and healthcare:
a separate, upstream activity. In the field of design theory, a grow-
a review. Saf Sci 2011;49:121–7.
ing body of research has highlighting the value of ongoing, holistic
7. Vincent C, Blandford A. Maintaining the standard: challenges in adopting
design processes that are tightly coupled to device use. Thus manufac-
best practice when designing medical devices and systems. Workshop on
turers, users, but also other stakeholders, must be able to collaborate Interactive Systems in Healthcare (WISH) at AMIA; 2011. Washington
on MD continuous design and thus collectively address a range of DC: AMIA, 2011, cited April 13, 2019 Available from https://www.
organizational, technical and functional issues (e.g. at a hospital academia.edu/31100632/Maintaining_the_standard_Challenges_in_ado-
level). This approach should deliver real benefits to the manufacturer pting_best_practice_when_designing_medical_devices_and_systems
(in terms of a product’s long-term viability), institutions, healthcare 8. International Electrotechnique Commission. Medical devices—application
professionals and patients (in terms of safety and quality of care). As of usability engineering to medical devices. Geneva: International Elec-
in a learning health system [31], an iterative, virtuous circle of learn- trotechnique Commission; 2007. 198p. Report No.: IEC 62366:2007.
9. Furniss D, Masci P, Curzon P et al.. 7 Themes for guiding situated
ing for all stakeholders based on a knowledge generation process
ergonomic assessments of medical devices: a case study of an inpatient
embedded in daily practice produces continual improvements. The
glucometer. Appl Ergon 2014; 45: 1668–77.
concept of continuous technology implementation [32] even makes
10. Association for the Advancement of Medical Instrumentation. Human
sense for seemingly simple-to-use devices, such as PPE. factors engineering guidelines and preferred practices for the design of
Regulatory bodies could broaden their perspectives and request medical devices. Arlington: Association for the Advancement of Medical
the implementation of more real-world performance monitoring, Instrumentation; 1993. 131p. Report No.: AAMI HE48:1993.
rather than postmarketing surveillance alone. In order to fit the 11. Association for the Advancement of Medical Instrumentation. Human fac-
system-based approach to designing tools, widely acknowledged tors design process for medical devices. Arlington: Association for the
as being essential, a shift in perspective is essential. This requires Advancement of Medical Instrumentation; 2001. 48p. Report No.: AAMI
moving from a product view to a system view to maximize the HE74:2001.
12. International Electrotechnique Commission. Medical devices—part 2:
safety and efficacy of MDs. This change obviously cannot only be
guidance on the application of usability engineering to medical devices.
the responsibility of manufacturers, in particular for certain com-
Geneva: International Electrotechnique Commission; 2016. 94p. Report
plex MDs. It poses significant challenges for regulatory bodies, used
No.: IEC/TR 62366-2:2016.
to regulating products, not systems. While a full-scale move into 13. Food and Drug Administration. Draft guidance for industry and food
a system approach is currently infeasible, regulatory bodies might and drug administration staff—applying human factors and usability engi-
take further steps in this direction. They could conceivably require neering to optimize medical device design. Rockville: Food and Drug
approval, for instance, at a hospital level. The manufacturer could Administration; 2011. 49p. Report No.: FDA-2011-D-0469.
 PROOF FOR CHECKING PURPOSES ONLY
6 Pelayo et al.

14. Vincent CJ, Li Y, Blandford A. Integration of human factors and 23. Arrêt n◦ 610 du 10 octobre 2018 (15-26.093)—Cour de cassation—
ergonomics during medical device design and development: it’s all about Première chambre civile—ECLI:FR:CCASS:2018: C100610.
communication. Appl Ergon 2014; 45: 413–19. 24. The Irish Times. Implant Files: medical devices may have caused more
15. International Organization for Standardization. Ergonomics of human- than 1,000 health incidents last year. Dublin: the Irish Times; 2018 [cited
system interaction—part 210: human-centred design for interactive sys- June 15, 2020] Available from: https://www.irishtimes.com/news/ireland/
tems. Geneva: International Organization for Standardization; 2010. 33p. irish-news/implant-files-medical-devices-may-have-caused-more-than-1-0-
Report No.: ISO 9241-210:2010. 00-health-incidents-last-year-1.3708071
16. International Ergonomics Association. Human Factors/Ergonomics 25. European Commission. Notified bodies Nando—Legislation. Geneva:
(HF/E)—Definition and Applications. Geneva: International Ergonomics European Commission; 2020 [cited June 28, 2020] Available from: https://
Association; 2018 [cited June 15, 2020] Available from: https://iea.cc/ ec.europa.eu/growth/tools-databases/nando/index.cfm?fuseaction=direc-
what-is-ergonomics/ tive.notifiedbody&dir_id=34
17. Gupta SP, Pidgeon A. An analytical approach to identifying potential use- 26. Bras Da Costa S, Beuscart-Zéphir MC, Bastien JMC et al. Usability and
related issues concerning a medical device under development. J Med Eng safety of software medical devices: need for multidisciplinary expertise to
Technol 2016;40:61–71. apply the IEC 62366: 2007. Stud Health Technol Inform 2015;216:353–7.
18. Thovtrup H, Nielsen J. Assessing the Usability of a User Interface Stan- 27. Ash JS. Some unintended consequences of information technology in health
dard. In: Proc. SIGCHI Conf. Hum. Factors Comput. Syst. Association care: the nature of patient care information system-related errors. J Am
for Computing Machinery, New Orleans, Louisiana, USA: 1991, pp. Medl Inf Assoc 2003;11:104–12.
335–341. 28. Dekker ST, Hollnagel E, Woods D et al. Resilience engineering: new direc-
19. Schaffer E. Institutionnalization of Usability: A Step-by-step Guide. tions for measuring and maintaining safety in complex systems. Sweden:
Addison-Wesley: Boston, USA, 2004. Lund University School of Aviation; 2008. 67p. Report No.: Final Report.
20. Sawyer D. Do It by Design: an Introduction to Human Factors in Medi- 29. Parush A, Wacht O, Gomes R et al. Human factor considerations in
cal Devices. Rockville: U.S. Dept. Of Health and Human Services, Public using personal protective equipment in the COVID-19 pandemic context:
Health Service, Food and Drug Administration, Center for Devices and binational survey study. J Med Internet Res 2020;22:e19947.
Radiological Health. 1996, 55p. 30. Carayon P. Human Factors in Health(care) Informatics: toward continuous
21. MedTech Europe. MedTech Europe’s Facts and Figures 2020. Brus- sociotechnical system design. Stud Health Technol Inform 2019;265:22–7.
sels: MedTech Europe; 2020 [cited June 15, 2020] 44p. Available 31. Greene SM, Reid RJ, Larson EB. Implementing the learning health system:
from: https://www.medtecheurope.org/resource-library/medtech-europes- from concept to action. Ann Intern Med 2012;157:207–10.
facts-and-figures-2020/ 32. Carayon P, Wetterneck TB, Hundt AS et al. Continuous technology
22. Brunotte G, Beuscart R, Pariset A et al. Place of high-risk medical devices implementation in health care: the case of advanced IV infusion pump
in European recommendations with a focus on end-users. In: BIOSTEC technology. In: Zink K (ed.). Corporate Sustainability as a Challenge for
2020-13th Int. Jt. Conf. Biomed. Eng. Syst. Technol., Valleta, Malta, 2020. Comprehensive Management. Springer, NewYork, 2008, 139–51.