Methodologic considerations for human chronobiology
Titel:
Methodologic considerations for human chronobiology
Auteur:
Reinberg, A.
Verschenen in:
Biological rhythm research
Paginering:
Jaargang 2 (1971) nr. 1 pagina's 1-15
Jaar:
1971-03
Inhoud:
The study of biological rhythms in human beings, as well as in animal species, requires a special methodology, which has at least three main aspects : methods of measurements and/or determinations to investigate physiologic variable(s). Obviously, high standards in physical measurements, chemical determinations, etc. are desirable to warrant each datum of a time serie. This condition can be considered as a necessity but is not sufficient in itself to validate any rhythm study. Statistical methods must be used to analyse the collected or recorded time series. Conventional methods lead to macroscopic examination of rhythms presented graphically as chronograms. More precision and objectivity can be obtained with inferential statistical methods : F. Halberg's cosinor and related means of microscopic analyses. With their help a biorhythm can be characterized by the statistical estimation of several parameters :the period, the acrophase, the amplitude, the rhythm adjusted level, etc. Each one of these parameters is expressed as a mean with its limits for a desired degree of confidence (usually 95%). For certain (transverse) samples, the synchronisation of the subjects studied has to be known, particularly the clock hours of ' light-on' and of 'light-out' when these indicate the schedule of sleep-wakefulness. The prominent synchronizer in man is of socio-ecological origin. When groups are involved, i.e. in transverse studies, one facilitates the analyses of time series if the subjects' synchronization can be standardized. The duration needed to standardize subjects for rhythmometry depends, among others upon their way of life before the study and the biologic function(s) studied. For example, for a rigorous standardization the week before any measurements can be considered as the minimum. There is a relation between the duration of a sampling span, T, the number of subjects, N, the sampling interval between consecutive samples, ? t, and the period of rhythm, τ. ?t of course should be much smaller than τ. If for any reason T is fixed, it is possible to manipulate N or vice versa. Tables from circadian rhythmometry of healthy adults are now available, on so-called internal acrophases in particular. It may be useful therefore in any rhythm experiment on man to add systematically the study of a reference rhythm such as that of body core temperature, urinary excretion of potassium or of 17-OHCS. Comparison can then be made with tables and the acrophase (peak) of reference rhythm can be taken as phase reference to present the internal timing for the function(s) studied.
Tijdschrift beschrijving