In the Biohacking world, the reasoning behind n=1 is that universal solutions to complex health problems have limited effectiveness, as we each have unique histories, genetic profiles, environments, and patterns of responding.
“Every discipline, profession, and field of endeavor has benefited from breakthrough insights generated by in-depth single cases.” ~Michael Quinn Patton,
For example, Dr Terry Wahls invented the Wahls Protocol and used it to reverse the disabling symptoms of Multiple Sclerosis (MS).
After she got up out of her tilt-recline wheelchair, she started clinical trials to research the impacts of nutrition and lifestyle hacks on other people with MS.
But she started with an n=1.
An n=1 isn’t always focused on oneself. I help design & document the n=1 experiments of my husband Matthew as he works to manage and reverse his autoimmune conditions.
I also run my own n=1.
He’s going for healing: I’m going for optimization.
They’re on the same continuum.
So, I’ve got two n=1 research studies on the go, and I document them both here~.
The following are 2 (of many) approaches to conducting n=1 biohacking experiments.
Unlike experimental designs which have subject and control groups, and compare outcomes based on different treatments in each group, in an ABAB research design, an individual can compare self to self over time.
To do this effectively, an A-B-A-B time series is used:
- A: Take a measurement before the intervention;
- B: Measure again during (or after) the intervention;
- A: Cease the intervention for an appropriate time period: measure again.
- B: Yup: Measure again during (or after) the intervention is reinstated.
Repeat until you are as sure as you need to be.
Quantitative researchers will tell you that to be effective, the data gathered through an ABAB research design needs to be quantifiable (taking the form of numbers). But all they really mean is that it needs to be quantifiable if you want to put it on a graph.
Graphs are fun. But they’re not necessary.
However they are useful if want to be able to quickly & easily track change over time. To get some ideas about using graphs (& what kind of quantitative data you might want to track) check out Chart Myself.
ABAB is a research design that works really well for n=1.
Autoethnography is a methodology that is inherently n=1. Here’s a super-brief introduction:
According to Michael Quinn Patton “autoethnographers struggle to find distinct voice by documenting their own experiences in an increasingly all- encompassing and commercialized global culture.”
In other words, we blog!
Blogs aren’t all autoethnographic, of course, but the blogging phenomenon is an example of the popularization of autoethnography.
Autoethnographic documents can take almost any written form. They are usually written in the 1st-person.
Carolyn Ellis describes her approach to autoethnographic research: “I pay attention to my physical feelings, thoughts and emotions. I use what I call systematic sociological introspection and emotional recall to try to understand an experience I’ve lived through.”
Here are some criteria for judging the quality of completed autoethnographic research (adapted from Laurel Richardson):
- Contribution: Does this research contribute to my understanding?;
- Aesthetics: Is the resulting text “artistically shaped, satisfyingly complex, and not boring?”
- Impact: Has this process generated new questions? Inspired new research? Resulted in action? Deepened self-knowledge?
- Expression of a reality: Does the text include self-awareness and self-exposure? Could an unfamiliar reader enter into this experience?
Autoethnography is controversial in research circles, of course.
Some condemn it for its ‘rampant subjectivism’. But ultimately, the criteria by which to evaluate any methodology is it’s usefulness, given the purposes of the research.
If you find an autoethographic approach useful, then it is.
An N=1 example
Barry Marshall was an internist with a theory: that stomach ulcers were caused by Helicobacter pylori and could be treated with anibiotics.
Marshall couldn’t prove his hypothesis with mice, because Helicobacter pylori is only active in primates. And he wasn’t allowed to experiment on human subjects.
So he decided to use an n=1 approach.
He took some of the bacteria from the gut of an infected person, put it in broth & drank it.
He then “developed gastritis, the precursor to an ulcer: He started vomiting, his breath began to stink, and he felt sick and exhausted. Back in the lab, he biopsied his own gut, culturing H. pylori and proving unequivocally that bacteria were the underlying cause of ulcers.”
His finding was generalizable to the population at large.
Obviously, I’m not recommending that we all undertake risky and experimental n=1 experiments.
What I am suggesting, is that there is no reason why a n=1 should be dismissed as lacking validity (or generalizability).
Depending on your circumstances, there may be nothing more useful than a methodical inquiry into the way that you, a distinct organism with a unique history, genetic profile and environment, respond to a particular intervention.