Nonparametric approach for the computation of tolerance intervals for circadian blood pressure variability

The construction of time-specified reference limits requires systematic sampling in clinical health, in particular for those variables that are characterized by circadian rhythms of large amplitude, as it is the case for blood pressure. For the detection of false negatives, the use of tolerance intervals is indicated. In the case of hybrid data (time series of data collected from a group of subjects), such a tolerance interval could be very difficult to determine by following a parametric approach which is similar to the procedure used for the computation of prediction intervals, if one wanted to consider both intra- and inter-subject variance. Accordingly, we have developed a nonparametric method for the computation of such tolerance intervals on the basis of bootstrap techniques that does not need to assume normality or symmetry in the data. We also used this method to establish time-qualified reference limits for a series of blood pressures and heart rates monitored automatically in healthy individuals of both genders. By the use of these tolerance intervals, many false positive and false negative diagnoses can be eliminated. These limits serve as a reference for comparisons of a given subject´s blood pressure series over time, yielding nonparametric measures of the extent and timing of any blood pressure excess or deficit. Such indices can then be used for an objective and positive definition of health, for the screening and diagnosis of disease, and for gauging the subject´s response to treatment