The harness for n‑of‑1 medicine
What is n‑of‑1 medicine?
N‑of‑1 medicine is trial-grade medicine created for a single person, where each patient is measured against themselves: every draw is read against where they started, so a change inside the reference range still shows, and every change to their plan answers to their own numbers.
In this answer9
What is personalized medicine?
- Personalized medicine
- A plan written for one person and carried out over years, judged by that person’s results.
The plan is the protocol, which a doctor writes for one patient after reading their labs, imaging and genome and hearing what they want from the next ten years. The clinic carries it out, closely monitoring what the protocol names, and the doctor changes it when the patient’s numbers or symptoms do.
It goes by many names, with longevity, anti-aging, concierge medicine, women’s health and men’s health the largest, and hormone therapy, GLP‑1 and metabolic health, peptide therapy, regenerative medicine, functional medicine and integrative oncology are practiced the same way. Each reads its own markers.
Judging a plan by one person’s results takes a method of its own: n‑of‑1 medicine.
Is this what I already do with my long-term patients?
It is, if your care was never episodic or transactional and you follow your long-term patients with the closeness an oncologist gives a cancer patient, when none of them has cancer.
They move through one protocol after another over a year or more, each step starting from what the last one did, and you ask them to score their symptoms before treatment and at every re-test.
The comparison lives in your head, because the EHR keeps each result as a document to open rather than a value it can follow.
You have been practicing this for years, and n‑of‑1 is the name research gave it, from the observational n‑of‑1 study to the registered n‑of‑1 trial further down this page.
Where does the name n‑of‑1 come from?
From research, where n is the number of people enrolled in a clinical study. A large trial might enroll 10,000 and its answer is an average across all of them, while a study with an n of 1 follows one person and compares them with themselves over time.
Medicine took the idea into care, so a patient’s first draw becomes their own baseline and every later draw is read twice, against the reference range and against that baseline. A value can sit inside the range and still be a real change for this patient, while a value outside it can be where this patient has always been.
A change is real for one person when it is larger than the marker’s own variation across repeat draws.
The live question for the epigenetic clocks is how large that variation is. On a reliability scale that reaches 1 when every repeat matches, the same blood sample run twice mostly scored above 0.9 while the same person drawn again hours to days later scored about 0.4 to 0.7, about the range of some routine labs (Aging Cell (opens in a new tab), August 2026). No clinically important difference has yet been defined for a clock (Nature Medicine (opens in a new tab), August 2026).
So a clock is context, trended on one fixed clock against the patient’s own baseline, while function and conventional markers carry the doctor’s decisions.
What does it look like for one patient?
Take Patient Z, 51, on a cardiometabolic protocol, where the doctor reads three markers at every draw: ApoB and hs‑CRP from the patient’s labs, and VO2 max from their wearable, anchored by a measured exercise test at Month 0 and Year 1.
Their first year runs in three cycles of four months, each opening on a draw and closing on a re-test where the doctor decides what changes.
Four months is the pace their markers set and no rule of the method, because a patient whose dose is still being found is drawn sooner and an n‑of‑1 record is only as good as how often someone reads it.
Patient Z’s wearable VO2 max is read as a trend between their two measured tests, because even the estimates a device takes during exercise agree with a laboratory test only to within about 10 mL/kg/min either way (Sports Medicine, 2022 (opens in a new tab)). VO2 max measured in a laboratory still varies by about 5.6% across repeat tests in the same person (Medicine and Science in Sports and Exercise, 1982 (opens in a new tab)). So a change carries a decision once a test that measures their breathing gases has measured it.
Illustrative
Walk through a year of one of your protocols.
Request a walkthroughWhy is every patient at these clinics a clinical trial of one?
Because the clinic takes the discipline of a trial and carries it into care: a protocol written before anything changes, a baseline taken first, certain biomarker changes and possible side effects closely monitored, and each decision kept with the reading behind it.
All of it runs as care, with the doctor deciding, cycle after cycle, for as long as the patient stays.
Trial-grade medicine created for a single person looks like Patient Z’s year, where the doctor changes one thing at each re-test so the next draw can be read against that change, the way an investigator controls variables.
What is a registered n‑of‑1 trial?
It is the same idea set up as a research design, “a prospectively planned, multiple crossover trial in a single individual” (British Journal of Clinical Pharmacology (opens in a new tab), December 2025), with its protocol written to SPENT (opens in a new tab), put through an ethics committee and registered before the patient starts, and its result reported to CENT (opens in a new tab). The trial answers one question about one treatment for that person, while each cycle of care answers a practical question: what should this patient’s plan be now?
Research has since built a second form for therapies made for one patient, which are single-dose or long-acting and don’t lend themselves to crossover, and a framework in Nature Communications (opens in a new tab) (November 2024) takes the baseline from the patient’s own natural history before treatment.
The idea of a patient measured against their own baseline now sits in a regulator’s text too, with a draft guidance on individualized therapies for specific genetic conditions describing “an externally controlled clinical investigation that assesses a patient’s change following treatment compared to baseline” (FDA (opens in a new tab), February 2026).
The same British Journal of Clinical Pharmacology review (opens in a new tab) says “most clinical settings lack dedicated platforms or workflows tailored to the demands of N‑of‑1 design”.
An open, product-neutral draft guide for n‑of‑1 evidence in longevity medicine, posted as a preprint, says the field “still lacks the quality infrastructure that mature health disciplines take for granted” and proposes that causal claims about one person be “registered, falsifiable and honestly reviewed” (Preprints.org (opens in a new tab), September 2026).
Why can’t a clinic do this today?
Because a pharma trial has trial management, data capture, safety reporting, biostatistics and a team of people around one protocol, while Patient Z’s protocol has their doctor and their nurses, and there’s no payer here to share any of it.
Their ApoB is watched because it counts the particles that drive plaque, their hs‑CRP, read as a trend, because inflammation carries risk of its own beside that count, and their VO2 max because how well they function is an outcome of its own beside those two markers, and the raised training should show in their fitness.
With a lipid-lowering medicine started at Month 12, the side effects it can bring are monitored as closely as the numbers, from the muscle symptoms only the patient can report to the liver enzymes taken before the start and re-tested if symptoms point there, and the glucose it can raise.
So the entire clinical, commercial and regulatory record of their care is falling on their doctor and their nurses, from the draw that comes due and the side effect the patient mentions in a message to the reason their plan changed and whether they stay for the next protocol.
The work a patient like this one leaves behind runs on long after they have left the room, and a clinic carries that for its whole panel at once.
Can a clinic do this for every patient?
Yes, when four things hold for every patient, and iksa is the harness for n‑of‑1 medicine, built to run clinical trial grade medicine for individuals, so that they do. One year of Patient Z is 21 values and three decisions, and a clinic runs that year for every patient at once.
The four are each new draw set beside that patient’s own baseline, each decision kept in the doctor’s note with the reason for it, each re-test kept to the protocol’s timing, and the doctor and nurses seeing what changed before the patient walks in.
iksa is built to keep those four for every patient, and the clinic’s doctors make every change. See the platform.
Where does the record of one patient lead?
Every reading keeps its date and its source and each item on the plan keeps the note it came from, so if you run the protocol properly, the evidence falls out of care.
Our vision is every personalized protocol chosen from evidence of what worked for patients like this one.
Read your patients against themselves
Bring a protocol you run. We’ll walk one patient through it.