Table 2.

Determination that an N-of-1 trial is feasible.

Question	Determination	Example
Is treatment efficacy questionable?	If it is clear to the patient and provider that a treatment does or does not work, then an N-of-1 trial is unnecessary.	After dramatic decrease in seizure frequency following the initiation of a new medication, an N-of-1 trial would be unnecessary to determine if therapy should be continued.
Is treatment for a chronic condition?	Given the resources needed to conduct an N-of-1 trial, they are only recommended for chronic conditions.	An N-of-1 trial for a single case of status epilepticus is not feasible.
Is there a rapid treatment onset and offset?	N-of-1 trials are only reasonable for therapies with quick onset and offset.	An N-of-1 trial with carbamazepine may not be feasible. Levetiracetam is an appropriate therapy for an N-of-1 trial.
Can the outcome be measured in a reasonable amount of time?	When taking the duration of a therapeutic trial and the number of trials, it is suggested to try to keep the N-of-1 trial to less than 12 weeks. However, patient willingness to complete the trial will factor into how long the trial can last.	For a patient who averages one seizure per week, a minimum of three weeks is needed to measure therapy efficacy. If an additional week is needed to washout and transfer to a new therapy, this N-of-1 trial would take upwards of 6 months to complete. This may not be considered reasonable by many patients.
Can the pharmacy department assist?	Pharmacy departments can often serve as the removed nonblinded individual.	Pharmacy departments can perform the randomization and prepare the therapy and placebo or control to maintain patient and provider blinding.
Is the patient interested in an N-of-1 trial?	Shared-decision making with a cognitively intact patient is essential for patient selection in an N-of-1 trial.	It is unethical to give a placebo or control therapy to a patient who has not undergone informed consent. Additionally, if a patient is uninterested in the N-of-1 trial, they will be unlikely to complete the required data collection.