our story · 02. Feb 2022
what is N-of-1?
an interview with our in-house health psychologist and N-of-1 expert.
*Dr. Nicki O’Brien is an associate professor of psychology at Northumbria University and our in-house health psychologist here, as part of the sofi team. Her research focus is on complex interventions for population and individual health using both mixed-methodologies and N-of-1.
what exactly is N-of-1?
“N-of-1,” sometimes known as single-subject analysis, is a method used within scientific research that considers an individual as the sole unit of observation in a study, rather than averaging over a group of people. It involves the repeated measurement of whatever is of interest (symptoms, feelings, thoughts, behaviours, etc.) in one individual over a period of time.
The ultimate goal of an N-of-1 trial is to determine the optimal or best treatment or intervention for a specific individual using the experience and data collected on that specific person.
what are some of the benefits of N-of-1, compared to other research methodologies?
The goal of N-of-1 is ultimately to try and better understand processes within an individual.
From my perspective as a health psychologist, we have a lot of ideas and many different theories about why people do or do not exhibit certain behaviours. We generally look at people as this high level “group” and make conclusions on differences between group A and group B, for example. We try to find out why group A and group B might be feeling, thinking or doing something different to the other group, and our theories make broad stroke predictions and conclusions based on these data.
A key limitation with this group-based approach is that our conclusions don't necessarily hold true for all individuals in that group — or maybe even any individual in any group.
Particularly, with N-of-1 we are trying to better understand individuals. Generally, our theories that make claims about differences between people use data which is a one-off occurrence or snapshot in time when we happened to study them. What we need to ask is whether these theories are helpful to better understand an individual, when we use data on the individual that was collected on repeated occasions over a period of time.
N-of-1 is thinking about it at an individual personalised level and it’s recognising that we aren’t all the same. In fact, we are not the same person now as we were yesterday, or that we will be tomorrow, and even at a finer detail of that: from hour to hour, from minute to minute…who knows!
N-of-1 is about specifically trying to elicit that information.
As a result, the benefits of N-of-1 are the ability to truly target investigation and personalise support, treatment, medicine — whatever it might be for the person that they are going to benefit from the most at that time.
where did N-of-1 study first come from, and how has it been impacted by new technologies?
Originally, the concept of N-of-1 was seen in education settings and - within psychology — in behavioural and neuropsychological rehabilitation. We would see N-of-1 trials with a very specific population or those with a rare condition etc., and the treatment needs to be tested, but it’s not feasible or desirable to recruit a larger number of people.
In the world of clinical medicine, N-of-1 methods have long been used in drug trials for pain. A drug is given to an individual, pain is monitored to see if there are any improvements and then the drug is taken away. If the improvement goes away too, then we have some level of confidence that the treatment is effective.
In other fields of psychology and health, N-of-1 studies have increasingly become more popular over the last 20 years or so, which is fantastic.I also think there’s a stronger need and demand, now, for more personalised medicine. There’s a huge — and there has been for a while — advocacy for personalising and individualising medicine, and taking more of an idiosyncratic approach to who we are. N-of-1 is ideal for this.
To answer the second part of your question, technology now makes it much easier to do the studies and much easier to view people at an individual level and treat them as one too. Because we have the automaticity and ease of collecting information via our phones, and other devices and activity monitors, it’s much simpler to have that moment by moment data collection.
With new technology, the data collection can be done automatically, taking the burden off of the individual that previously would’ve had to somehow calculate and provide that information every step of the way.
what are some of the negatives of N-of-1? Are there instances where it could never be used?
In research, the challenge of an N-of-1 research study is that you need to collect a large amount of data to be able to make strong conclusions that have statistical evidence behind them. The statistics for N-of-1 are still developing, such as how we can combine data from different individuals to identify similarities and differences. People are also doing a lot of work to better understand the size of treatment effects in N-of-1s in the same way that group-based studies commonly report.
In clinical practice, I think you can’t provide an entirely individualised approach in certain circumstances. For example, a health professional might not have the time to collect data on an individual over a period of time in order to determine which treatment may be best. Sometimes there is an urgency to treat someone so the health professional has to give the treatment that they think will be the best, from their clinical experience of treating others and the scientific evidence.
In the same way, policy decisions about whether to implement a new healthcare service are more likely to be based on population-level data that indicate that the large majority of people within a given population will benefit from the new service. Therefore there will always be a need for large group-based studies.
But, ideally, even within a randomised control trial - which is said to be the gold standard study design to determine the effectiveness of a treatment by randomly assigning people to either receive the treatment or to receive something else - you could embed a series of N-of-1s. It would be an excellent way to understand what’s happening on a deeper, individual level, giving you data that you wouldn’t otherwise have had and that may really help to explain some of the findings.
N-of-1 at sofi
Typically, when a product in human health is created, it is made for the herd. It’s not specifically for any one of us. One hundred people are put in a study group and another hundred in a control group. One group is given the product, and the other placebo to see what happens. The goal is to find a product that works for some people, most of the time.
What sofi does differently is that we break that single study of 100 people into one hundred studies of one person. Our sample size is at the “N-of-1” level – you.
sofi’s “statistical brain” learns to predict how you feel and how you sleep based on your specific experience with a particular plant. In other words, instead of relying on randomised controlled trials to tell us what might work for an average person (which may not be that representative of you), we look specifically at you — and the data points we collect via our app — to make the main expert of supporting your health with plant-based solutions, ultimately you.