EHR design isn't a 'wicked' problem
Why EHR design isn’t a wicked problem but a collaborative challenge
Dave Pao
3 min read
November 2023
'Once more he turned to that which could not be fixed.'
- Wallace Stevens (1879 - 1955)
EHR interface design has been called a wicked problem (Rajabiyazdi et al., 2021)—a criticism never levelled at paper records, for all their deficiencies. In their original description, Rittel & Webber (1973) contrasted wicked problems with relatively tame, soluble problems such as mathematics or chess. They pointed out the inadequacy of a sequential, structured methodologies for understanding complex design problems, which were ‘bound to fail.'
A wicked problem defies any standard attempt to find a solution because it is a symptom of multiple, contingent and conflicting issues. An isolated design solution, or that of any one discipline, arrived at through an established process, will by definition exacerbate a wicked problem (Marshall, 2008).
Understanding whether EHR interface design is truly a wicked problem or merely a design challenge is a distinction worth making. My view is that, even if there is an element of wickedness, it is probably—as Conklin (2003) suggests—more predominantly a failure in communication between systemically siloed disciplines.
Conklin, three decades after Rittel & Webber, tackled wicked problems with a focus on collective intelligence, a natural property of which is collaboration. Conversely, he describes the forces challenging collective intelligence as forces of fragmentation.
Fragmentation, as Conklin describes it, 'pulls apart something which is potentially whole', a condition in which the people (or disciplines) involved see themselves as more separate than united. Knowledge is chaotic and scattered, and the fragmented pieces are the 'perspectives, understandings, and intentions of the collaborators.'
Fragmentation can be hidden, with stakeholders unaware that there are incompatible tacit assumptions about a problem, not shared by all. Conklin posits that fragmentation is induced by:
1. The fragmenting force of wicked problems, compounded by social and technical complexity
2. The confusion, chaos and blame in failing to distinguish wicked problems from complexity
3. The general lack of tools, techniques and processes for defragmenting projects
Conklin asserts that the real challenge is to divert from the blame that inevitably emerges from the suffering and away from ‘problem-solving’ through easy technical fixes—and focus instead on building capacity to collaborate effectively.
The architect Stanford Anderson (1966) expressed scepticism towards conventional problem-solving techniques, deeming them unsuitable for the evolving and complex field of architecture. Problem-solving was 'neither descriptive of the traditional behaviour of the best architects nor applicable to the current problem situation of architecture.’
Instead, Anderson promotes 'problem-worrying' as a more dynamic, rigorous and professionally engaged alternative to problem-solving. Taking this approach, combined with careful documentation, allows for future research to be built robustly forwards from it. Inappropriate problem-solving, on the other hand, can lead to unstable products resembling the game Jenga®, which mis-inform future research such that the only guaranteed outcome is the catastrophic end game.
Anderson's ideas about architecture can be applied to the design of EHR interfaces. In this context, problem-worrying resonates because the nature of clinical practice is also one where traditional problem-solving techniques that rely on explicit problem definitions or distinct goals are not always appropriate and are often counter-productive.
Can EHR interface design be framed less as a wicked problem and more as a talented design collective waiting for more authentic, strategic collaboration? A problem that merely needs defragmenting, through the ‘shared understanding and shared commitment’ (Conklin, 2005) of the contributing disciplines? As more of a tame problem, like chess?
Anderson, S. (1966). Problem-worrying. [Online]. Available at: http://web.mit.edu/soa/www/downloads/1963-69/TH_AALond-Lect_66.pdf.
Conklin, J. (2003). Wicked problems and social complexity. [Online]. Available at: https://qi.elft.nhs.uk/wp-content/uploads/2015/08/cn-wickedproblems.pdf.
Marshall, T. (2008). Wicked Problems. In: Erlhoff, M. and Marshall, T. (Eds). Design Dictionary: Perspectives on Design Terminology. Basel, Switzerland: Birkhäuser Basel. pp.447–447.
Rajabiyazdi, F. et al. (2021). Communicating Patient Health Data: A Wicked Problem. IEEE Computer Graphics and Applications, 41 (6), pp.179–186.
Rittel, H. & Webber, M. (1973). Dilemmas in a general theory of planning. Policy Sciences, 4 (2), pp.155–169.
