To the Editor.—
I have significant concerns about the design and data analysis in the study by Franks et al1 and their conclusions on oral clonidine and smoking cessation, published in the December 1 issue of JAMA. Franks et al make the point that other studies that excluded subjects who were still smoking (even a few cigarettes per day) may have introduced a bias. Yet Franks et al themselves excluded and failed to report the data of almost 80% of their subjects (71 of 92 subjects excluded in the experimental group and 74 of 93 subjects excluded in the placebo group).The exclusion of these subjects introduces a significant selection bias—those subjects excluded who continued to smoke a few cigarettes per day may have been the most addicted to nicotine, may have had the highest severity of withdrawal symptoms, and may have been the subjects most likely to benefit
Ornish SA. A Trial of Clonidine to Stop Smoking. JAMA. 1990;263(20):2746–2747. doi:10.1001/jama.1990.03440200048012
Customize your JAMA Network experience by selecting one or more topics from the list below.