SUMMARY The variation of the amplitude and frequency of finger tremor was studied in 190 subjects ranging in age from 7 to 77 years. Tremor was measured in various postures with an accelerometer and a transduced signal was analysed by Fourier analysis. The variation of tremor frequency and amplitude with age depended upon the posture adopted for measurement and in certain postures some of the variation was related to changes in stature with age.The variation in tremor between individuals has prompted surveys of the way in which the variation is related to age. Marshall and Walsh' measured tremor in children and adults using a double-diode accelerometer at various sites of the body and concluded that generally children showed lower dominant frequency than adults. Marshall2-3 used visual analysis of the waveform from an accelerometer, held in the hand with the arms outstretched in front of the body, to estimate tremor frequency. Based on records from over 600 subjects, randing in age from 2-96 years, he concluded that dominant frequency increased from 5-6 Hz in children to 8-12 Hz in adults with a gradual decrease after 40 years of age to 6-7 Hz at the age of 70. Buskirk and Fink4 using optical methods of measurement and visual analysis found a similar trend in 140 subjects.Marsden, Meadows, Lange and Watson' used spectral analysis of recordings from an accelerometer on the index finger in 88 subjects from 3 to 85 years of age and found no difference in dominant frequency between children and adults; frequencies were in the range of 8-9 Hz, but over the age of 60 dominant frequency decreased to 7-7 Hz.One study has shown a higher dominant frequency in children than in adults,6 but the methods of analysis were subjective and measurement techniques less sensitive than those used in the other studies quoted.There is no close agreement between the few tremor studies on age variation so far reported. The measurement techniques, the postures adopted dur- ing measurement and the methods of analysis varied widely, which may account for the disagreement. We now report a survey of finger tremor using spectral analysis of accelerometer recordings in three different postures to try to resolve some of the disagreement. Methods SubjectsThe subjects were randomly selected from the following populations: ( 1) 7-11 years olds from a local junior school;(2) 11-18 years olds from a local comprehensive school; (3) 19-64 years olds from the students and staff at the Medical school (4) over 65-year-old males attending the Medical School for other physiological measurements; (5) over 65-year-old females attending a local keep fit class. Table 1 shows the age group classifications with the mean age, height and weight for each group. Of the 190 subjects 102 were female and 88 were male; 10 subjects were left handed. I remor measurementMeasurements of finger tremor of both hands were made using a small accelerometer (mass 13g) (Bruel and Kjaer type 4367) attached by a "Perspex" ring (mass 2g) to the terminal phalanx of the middle finger. R...

No abstract