What clinicians need to know
How to get patients to tell you what they won’t tell you
“Can’t see the wood for the trees” is one of my favorite phrases that sums up perfectly a common mindset in modern science.
“BFW” is the Navy equivalent.
In my previous life as a young navigation officer, I was often tested, as are so many others to this day, by the captain of the ship by being asked, “What tools do you have on the bridge to help navigate the ship?”
Panic usually ensued as we rattled off the tools we could think of. “The chart, Ma’am, the sonar Sir, the GPS, the depth gauge...” We would all receive a nod of approval from the captain, but were always asked for more until we had run out, puzzled and feeling depleted.
“What about the BFW?” the captain would point forward…
“The Big F*****g Window!”
Many of us in science get so weighed down in the details and end-points that we often forget to start with the basics, and the basics are more often than not all the patient has to go on.
Delivering understandable information to patients is only one half of the story when evaluating a medical history in any clinical setting. Asking a patient the wrong question about their symptoms or lifestyle will, of course, deliver the wrong answers and that can have considerable consequences in clinical and scientific work.
Following from Dr. Nadine’s post entitled What patients need to know, let’s consider the reverse: how do we facilitate communication that originates on the patient’s side that’s understandable to the clinician?
For us at Medicus AI, we always start by thinking from the patient’s perspective. So how do we ask questions in a way that patients understand, with answers they understand, and free from bias from either side?
This is even more challenging when implementing medical AI, as you only have a set number of questions and responses and no way to look at the patient, and read their language for extra clues.
The dangers of “normal”: The patient is their own point of reference
Schistosoma haematobium is one of the most prevalent parasitic infections in humans and is found across sub-Saharan Africa and parts of the Arabian peninsula. Spread through infected water sources, S.haematobium burrows through pores in human skin and later in their incredibly complex life-cycle, releases eggs through the wall of the bladder, to be urinated back into the environment. Releasing eggs through the walls of the bladder inevitably cause bleeding into the bladder and produces characteristic blood-red urine.
To evaluate the incidence of S.haematobium infections in east Africa, a group of scientists secured funding to begin a widespread questionnaire of the population with several questions aimed at remotely diagnosing infections based on responses, and in turn, use that information to target prevention plans in regions most affected.
The problem was, that even though S. haematobium infections were known to be fairly widespread, once the questionnaires came back almost nobody could be said to have an infection. Why?
The answer is, of course, the wrong question was being asked.
Textbooks, lectures, and research mean that clinicians have a vast overview of medicine, but often forget that the patient has only one reference; their own body.
The killer question, in this case, was “Is your urine normal-colored?”, to which many, if not most, replied with a “yes.” It was only until face-to-face interviews elicited the further response “Yes my urine is normal colored, it has always been red,” that the penny dropped.
The power of personal narrative: Let patients choose the boxes they want to be put in
Another classic symptom of forcing a patient to give you misleading information is the practice of splitting off groups into neat categories. Categorization is, of course, useful when studying populations, but the names of these can be misleading, offensive, or based on the bias of the researchers. We need to give patients enough room to be able to put themselves into categories that match their own experiences and feelings.
During the height of the AIDS epidemic in the west, extensive coverage and attention were given to certain high-risk groups, homosexual men, intravenous drug users, and certain ethnic minorities. The problem again was that demographic data from questionnaires and HIV incidence were not matching up, and once again it was due to clinicians and scientists not putting what we call their UX hats on when designing their interventions.
The term Men who have Sex with Men (MSM) was invented late into the epidemic to finally plug a gap in patient communication. Male, unprotected, homosexual sex is a risk factor for HIV transmission; being gay is not. In sexual health interviews and on questionnaires, many men did not identify as being homosexual, even though they were engaged in what other people would consider homosexual intercourse.
The point here is that it is not about what other people think; it is about what the patient thinks and feels.
Trying to put our own biases into a patient’s mouth means they will either take the wrong decisions or spit them out.
With the birth of tele- and digital- medicine it is important for us as scientists to get as much relevant information from a patient as possible and to minimize noise from, and to, the patient. This is a subtle skill that requires a massive amount of empathy and understanding of people and their situations.
Science communication is often pushed and implemented as a way to essentially turn patients into lay-scientists: teaching them the technical words so they can communicate with clinicians. So in that case, is it not right that scientists should meet the patient halfway, and learn to understand their point-of-view, perhaps that way we might all gain better insights?
