To the Editor.
—The notion that tension or anxiety is a predictor of hypertension has been assumed intuitively and has been studied by many over decades. Yet convincing evidence of such a relationship has not appeared. The recent article by Dr Markovitz and colleagues1 purportedly demonstrates that anxiety levels predict development of hypertension in middle-aged men. However, the findings are open to question on several grounds.First, there was no evidence of a relationship between tension and development of hypertension in women or in older men. Thus, in three of four groups tested, no relationship was seen. Further, the odds of a type I error are substantial. Given the number of subgroups, the number of predictors tested, and the stated P value of.04, there is a 39% likelihood of a chance association between some predictor and hypertension. For the .02 value reported with multivariate analysis, there is a 20% likelihood
Mann SJ. Is There Tension in Hypertension?. JAMA. 1994;271(13):979. doi:10.1001/jama.1994.03510370031013