Research Profile - The power of one

While random clinical studies with thousands of participants are regarded as best, an alternative medicine expert points out that one size does not fit all

[ Back to main article ]

When it comes to testing new medications, there is strength – or at least faith – in numbers. Physicians tend to trust therapies that have been proven to work in randomized control trials (RCTs) involving hundreds, and sometimes thousands, of participants.

But health care isn't a one-size-fits-all solution, says Dr. Sunita Vohra, a professor at the University of Alberta and a practising pediatrician.

"Real life patients often have more than one condition and often take more than one therapy at the same time," says Dr. Vohra. "If that isn't how a particular drug was studied in the RCT, then it's hard to say, 'Well, we actually have the evidence we need to guide your therapy.'"

Dr. Vohra leads a Canadian Institutes of Health Research (CIHR) project to investigate an increasingly popular evaluation option: N-of-1 trials.

"N-of-1 means a single patient is involved and the therapy is provided on a random on/off, on/off basis," says Dr. Vohra, a specialist in complementary and alternative medicine (CAM) for children. "The patient doesn't actually know when they are on the therapy and when they may be off it and on the placebo. You have an ongoing period of formal evaluation where the patient gets to see if they're getting better."

RCTs, in contrast, are studies in which participants are randomly assigned to groups. One is the "control" group of the trial. Participants in this group may receive a placebo, the standard practice for treating the disease or no treatment at all. Another group receives the new therapy being tested. At the end of the trial, researchers can measure the effects of the therapy against the placebo effects, the results of standard care or no care.

While RCTs are widely regarded as the gold standard for health research, N-of-1 trials are especially useful in evaluating CAM therapies, says Dr. Vohra.

"One of the reasons N-of-1 trials are such a good fit is that many times the patient is given an individualized approach to treatment. For example, a diagnosis of pneumonia in an otherwise well person means one standard treatment in Western medicine. In contrast, the CAM approach would vary based on symptoms – the presence or absence of fever, cough, appetite changes and energy changes. It would also take into consideration what the patient was like pre-illness. The therapy would be individualized based on the predominant symptoms in the context of the person's constitution."

There is considerable need for more rigorous evaluation of CAM therapies. The Fraser Institute has estimated that Canadians spend almost $8 billion a year on visits to providers of alternative medicine, as well as books, medical equipment, herbs, vitamins, and special diet programs. Many of these products and services lack scientific evidence to support their safety and effectiveness.

That's where Dr. Vohra sees a stronger role for N-of-1 trials: "An RCT is great unless your health needs are specific and different from the next person's. An individualized therapy can be more challenging to evaluate. N-of-1 gives you answers for the individual, which are often more important to the patient than answers for a group."

One of the side benefits of N-of-1 trials is that knowledge translation – turning research findings into improved treatment – can be immediate. "They call it the know/do gap," says Dr. Vohra. "N-of-1 trials can close that gap. They bring the research to the individual patient."

Dr. Vohra's team is working in collaboration with the CONSORT Group, an international organization dedicated to improving and standardizing how clinical trials are reported.

"Dr. Vohra's work is groundbreaking and it will be very useful for researchers who often say they can't do big trials because CAM is all about providing care at the individual level," says Dr. David Moher, a senior scientist at the Ottawa Hospital Research Institute and a CONSORT executive member. "What N-of-1 manages to do is to take the superiority of randomization and apply it to individual patient questions about treatment effectiveness."

The Study

Dr. Vohra's study is called the CONSORT Extension for N-of-1 Trials (CENT). Her team began its work with three CIHR-funded systematic reviews to summarize what is known about N-of-1 trial methods. They took their findings to the scientific community through a consensus-building process and invited international leaders in the field to participate through an electronic survey. That was followed by a meeting in Banff in the spring of 2009. Based on those discussions, Dr. Vohra's team is developing a checklist and guidelines for reporting N-of-1 trials.

How to do an N-of-1 right

An N-of-1 isn't simply trying a therapy, seeing if it works and then writing up the results. That's a case study and not a clinical trial, says Dr. Vohra.

"An N-of-1 trial is prospectively planned – you don't just come to it after the fact and then look back in hindsight. You plan the randomization for whether the therapy or the placebo will be given, so you can get a true measure of how the patient is doing."

N-of-1 trials are used to evaluate therapies for chronic conditions such as asthma or inflammatory bowel disease. Dr. Vohra explains: "If you have a condition that's here today and gone tomorrow, you wouldn't know if the patient got better because of the therapy or because the condition got better on its own."

The trials also require considerable buy-in from the patients.

"Patients need to be enthusiastic partners – because it's work for them. We give them outcome measures they have to fill out. They are active participants in their therapy and its evaluation. It's appealing because people want to be active in their own health care. This is an opportunity to do that."

"When my kids were little, they played these games where they put shapes in a box: if they don't fit, you put the piece aside. An RCT is like asking the patient if they fit the box– if they don't fit, they're put aside. In an N-of-1 trial, the patient remains the patient and the study is wrapped around the patient. We evaluate them as they are, in real life."
– Dr. Sunita Vohra