To the Editor.
—I call your attention to the graphic exaggeration of research accomplishment in Fig 2 of the recent JAMA article by the Trials of Hypertension Prevention (TOHP) Collaborative Research Group.1 While the text describes an effect of weight reduction on systolic blood pressure of 2.9 mm Hg, the graph shows 3.8 mm Hg. For diastolic, the text says 2.3 mm Hg and the graph indicates 2.6 mm Hg. For sodium reduction, the text value for systolic change is 1.7 while the graph shows 2.1; for the diastolic, the text says 0.9, and the graph says 1.1. This systematic error in the same direction is statistically significant, with a P value that is embarrassingly small.Whatever the tiny statistical significance for sodium restriction, one wonders whether a wise and fully informed patient would wish to give up an unrestricted, tasty salt diet in order to lower his or
Landau WM. Distress Over the Noneffect of Stress. JAMA. 1992;268(2):198. doi:10.1001/jama.1992.03490020042022