Working in health, it’s common to hear many stories of people’s experiences of care. Indeed, a participatory and user centred design approach actively seeks out these stories, to make sense of experiences and design better services. In a way, health stories are a basic currency of care. An echo, I guess, of the clinical practice of taking a patient’s history.
When designing health services, people readily bring their personal health stories to the fore. In some cases, a person may have had direct experience of the problem we are tackling. In other cases, it may be just a part of the problem, such as a small section of post-operative care. These stories are often a passionate re-telling of an experience (sometimes good, sometimes bad). It’s hardly surprising. We see this in research all the time, participants replay health journeys that have made a massive impact on their lives.
I want to examine two facets of these health stories and see how they contribute to a bias that affects the way services are designed and made. As a counterpoint, it’s worth looking at this bias from a patient’s perspective, and see how someone’s health story can affect the way health services are received.
Product biases
This isn’t just a designer bias. It involves anyone in health who makes decisions about product and services. It falls into the simple premise that when a person has had a health condition, this can affect their decision making process (for good and bad) in a number of ways.
Recalling a health journeys is problematic.
With so many actors involved and so many variables for each actor, a health journey is always unique to the time and place it occurred. Even though it may feel typical, and there may be many aspects that are typical, the sequence of events is always unique.
The understanding of that journey has gone through the brain’s interpretive processes. The ability to understand, process and then recall that journey is very likely to be influenced by what was going at the time, feelings, mental state and most importantly, how long since the experience occurred.
This is well known in psychology and epidemiology as a recall bias, defined as a “a systematic error caused by differences in the accuracy or completeness of the recollections retrieved (“recalled”) by study participants regarding events or experiences from the past.” In short, people are poor at recalling their stories.
For any symptom or condition, the ideal research respondents are those who are currently going through this journey. If not, their health experience should be as close as possible to now. This may reduce the bias that occurs during recall of someone’s health story. Furthermore, speaking to a range of people who have had the condition will complete the story of an experience from as many viewpoints as possible.
Over representativeness
Health experiences can be vivid. There is a risk that people may over-represent that experience. This is known as the representative heuristic: A fault in judgement based on the presumption that once people or events are categorised, they share all the features of others members in that category. In the case of a personal health experience, you wrongly believe that your experience is the same as others.
One of the problem of over-representativeness is that it’s also difficult to understand which parts (if any) are representative of the general patient experience.
There is a further worry that over reliance on a personal health experiences devalues opinions from outside of that experience. That is, over representativeness becomes a mechanism of exclusion. This removes objectivity from research as your personal story now trumps others.
Another form of this is (thankfully) less common. The opinion that unless you have a condition then you can’t contribute. This is nonsensical.
I have heard about some bizarre examples of this. One organisation eschewed user research, believing that the development team (as users of the NHS) were representative of the broad range of health seekers in the UK. Their position was “we don’t need to do research, because our developers use the NHS. So we are testing as we are building”. Other times I have seen the HIPPO exert pressure on product direction (against prevailing messages from user research) because of a personal health experience.
Confirmation bias
What drives this is confirmation bias. A bias which occurs when people seek out or evaluate information in a way that fits with their existing thinking and preconceptions. Within health and design, it’s typified by over-valuing personal experiences too much and undervaluing the benefits of broad, holistic qualitative research that informs design decisions.
The remedy for this is simple but hard to execute. On a personal level Steve Portigal (Interviewing users, 2015) advice to counter this is to “check your worldview at the door”. To change this within an organisation…well that’s a lot harder.
Patients as experts
There is another side to this. One I have seen in interviews I am doing with people who suffer from Type 1 diabetes although I suspect is frequently seen with people who are suffering from a long term condition (LTC).
It is the situation where, over time, the patients become experts in their own LTC and management. With this growing knowledge of their personal health story, they then interact with a clinician who is perceived as being dismissive and uninterested in their understanding of their illness and how it applies to their specific situation.
I’ve heard this most often when type 1 diabetics move from paediatric to adult care. It’s a period of transition that is fraught with many issues. Characterised (from the stories I have heard) by close and intimate paediatric care, that contrasts with brief and unengaged care in the adult system.
Many late adolescents are shocked and somewhat disheartened by the perfunctory care they receive from the clinicians.From their perspective, they’ve had the lived experience of this condition for years and find it odd that their opinions are not listened it. They say ‘I know me, I’ve had this condition for 15 years..but they don’t seem to want to listen’.
Clinically this can result in the delivery of poor medical care. A recent lancet article (Saini et al., 2017), highlighted a number of drivers for poor medical care. Specifically, the thinking frameworks that influence decision making. Amongst them were the clinician’s response to “evidence that contradicts training or practice experience”. It’s easy to see where there may be fractures in the patient-clinician relationship when clinical diagnosis and treatment does not align with the patient’s expectation of care.
Conclusion
Biases in research are unavoidable. For a design researcher in health, they can be particularly problematic, especially since your chief goal is to try and understand the patient experience as authentically as possible. In a work environment, the people you and interact with, the organisational culture, that too, could bring unwarranted pressure to bear. Pushing design work towards desired outcomes.
Increasing your research sample size, widening or modifying your research methods, as well as understanding your own health story and recognising potential biases, can help guard against your own (mis) interpretation of the needs we’re are designing for here.
…and finally
This is a great run-down on cognitive biases. Well worth a read.
https://betterhumans.coach.me/cognitive-bias-cheat-sheet-55a472476b18
References
Portigal, S. (2013). Interviewing Users. 1st ed. Brooklyn, NY: Rosenfeld Media.
Saini, V., Garcia-Armesto, S., Klemperer, D., Paris, V., Elshaug, A., Brownlee, S., Ioannidis, J. and Fisher, E. (2017). Drivers of poor medical care. The Lancet.
