Current issues of ACP Journal Club are published in Annals of Internal Medicine


Therapeutics

n of 1 trials led to less theophylline use with no adverse effects

ACP J Club. 1996 Sept-Oct;125:32. doi:10.7326/ACPJC-1996-125-2-032


Source Citation

Mahon J, Laupacis A, Donner A, Wood T. Randomised study of n of 1 trials versus standard practice. BMJ. 1996 Apr 27;312:1069-74.


Abstract

Objective

To compare the effects of doing single-patient (n of 1) randomized trials of theophylline with giving theophylline by standard practice in patients with irreversible chronic airflow limitation.

Design

Randomized controlled trial with 6-month follow-up.

Setting

Outpatient practices in Ontario, Canada.

Patients

31 patients (mean age 70 y, 62% men) who had chronic airflow limitation and had taken theophylline for 1 to 5 years but were unsure whether theophylline was helpful. 26 patients (84%) completed the study.

Intervention

14 patients were allocated to n of 1 trials (single-patient, randomized, multiple-crossover comparisons of theophylline with placebo). 12 patients were allocated to standard practice, in which theophylline was stopped and resumed if dyspnea worsened; if dyspnea improved, theophylline therapy was continued.

Main outcome measures

Proportion of patients not receiving theophylline at 6 months, exercise capacity measured by 6-minute walking distance, and quality of life measured by the chronic respiratory disease questionnaire.

Main results

9 patients (64%) allocated to n of 1 trials were not receiving theophylline at 6 months compared with 2 patients (17%) allocated to standard practice { P = 0.02. This absolute risk improvement of 47% means that clinicians would have to conduct n of 1 trials in 2 patients receiving theophylline (rather than giving theophylline by standard practice) to have 1 additional patient discontinue therapy with the drug by 6 months, 95% CI 1 to 10; the relative risk improvement was 286%, CI 26% to 1326%*}. The difference between change in 6-minute walking distance in the n of 1 group compared with the standard practice group was small (9 m, CI -23 to 41 m; the smallest important difference in an individual patient was judged to be 50 m), as were the differences in physical function {-0.08, CI -1.4 to 1.2}* and emotional function {0.3, CI -0.6 to 1.2}* (the smallest important difference in both physical and emotional function was judged to be 0.5).

Conclusion

n of 1 trials led to less theophylline use at 6 months, with no decrease in exercise capacity or quality of life in patients with chronic airflow limitation.

Sources of funding: Ontario Ministry of Health and Astra Pharma Incorporated (theophylline).

For article reprint: Dr. J. Mahon, Room 60F-11, University Hospital, P.O. Box 5339, London, Ontario, N6A 5A5 Canada. FAX 519-663-3232.

*Numbers calculated from data in article.


Commentary

The recent clinical literature contains numerous examples of the feasibility and value of n of 1 randomized trials. These blinded, randomized, crossover trials in a single patient can aid in the evaluation of drugs, such as inhaled bronchodilators, inhaled steroids, and oral theophylline for chronic airflow limitation; nonsteroidal anti-inflammatory drugs and acetaminophen for osteoarthritis; histamine-receptor antagonists for nonulcer dyspepsia; antihistamines for atopic dermatitis or allergic rhinitis; and tricyclic antidepressants for fibromyalgia. Despite the promise of avoiding unnecessary treatment that shows no benefit to individual patients, clinicians rarely use n of 1 randomized trials.

This carefully designed, well-done study shows that n of 1 trials can reduce the burden of unnecessary drug use in chronic disease management. The results also suggest that clinicians' and patients' overenthusiasm about therapy can result in long-term administration of treatments that are of little or no benefit to patients with chronic diseases, such as irreversible chronic airflow limitation. This study's small sample size and short follow-up (6 mo) limit further conclusions.

We still do not know whether the benefits of improved therapeutic precision are sufficient to justify the extra cost of n of 1 trials or the extent to which patients might receive lasting benefit from results of an n of 1 trial. At present, one reason that n of 1 trials are infrequently used is that clinicians lack access to convenient n of 1 trial services. I suspect that n of 1 trials will not achieve more widespread use unless their cost-effectiveness can be convincingly shown in larger clinical trials. If n of 1 trials were generally considered cost-effective, health-care organizations might be willing to assume the added overhead costs of funding an n of 1 trial service.

This study is a powerful reminder of the biases of open or conventional trials of medicines—natural history, placebo effects, patient and physician expectations, and patients' desire to please—all of which can lead to a false inference that a treatment is producing an important benefit.

Eric B. Larson, MD
University of WashingtonSeattle, Washington, USA