Nobutaka DOBA^*1 Shigeaki HINOHARA^*2

　

　Recent prominent increase of elderly population in Japan urges us to rethink about the assessment of health status in this age group. It has been often claimed that the elderly is not an extension of the usual adult, and numerous studies have disclosed the specific features of physical, psychological and social aspects for the elderly. Age related declines in functions have been extensively studied in gerontology even until recently. Actually, however, most of these populations are still active and productive. On the other hand, 4 to 8% of the elderly, especially in the female, have been identified as the frailty or vulnerability ¹⁾ , and its clinical entity as a phenotype has been proposed with its tentative diagnostic criteria²⁾. If this is indeed the case, frailty or vulnerability in the elderly must be preventable, and another strategy for the health maintenance and promotion in the elderly could be developed with an expectedly reasonable success ³⁾ .

　The definition of the elderly has traditionally been settled based on its chronological view point and the age of 65 years or more has been used to be the standard without biological and /or sociological evidences. In the era of extended longevity, the renewal of the standard for anthropometric measurements and their related functions must be necessary for the evaluation of the reasonably defined healthiness of the elderly. For the young generation mainly consisting of healthy people, the concept of mean values obtained from these groups might be acceptable as the standard for the healthy status. However, the elderly mostly consisting of populations with age associated declines of functions or with several co-morbidities and disabilities, mean values could not be the representative of healthiness. In the elderly, the physical and psychosocial standards must expectedly be desirable goals rather than mealy average status from the view point of health maintenance and promotion.

　In the present study, we will demonstrate the standard for the anthropometric measurements and related functions based on the data obtained from the special population consisting of voluntarily participated males and females, whose ages were more than 70 years, health conditions have been stabilized without moderate to severe comorbidities and whose social activities have been still reasonably active and productive.

　Such approaches as anthropometric measurements are indeed less expensive, and are more easily, repeatedly and reliably performed in any situation than other sophisticated measurements used to be utilized in many institutions. Beside that, these data could be quite useful for dieticians, community nurses, and other comedical staffs as physical and occupational therapists in an assessment of nutritional and physical conditions, particularly for the frail or vulnerable elderly.

　

　

　The study was conducted at the Life Planning Center, Tokyo, where, since October 2000, the new elder citizen project aiming to promote activities and productivities of the healthy elderly has been developed. The details of the project has already been described elsewhere⁴⁾, and numbers of the member had exceeded 3,000 by the end of July 2003. The present study was conducted on the subjects among those members mainly living in Tokyo metropolitan area, who voluntarily joined to the program with written informed consents after appropriate specification of the study purposes. Two hundreds and nine healthy subjects, aged 70-94 years, 107 males (mean age: 78.5±4.6 years) and 102 females (mean age: 77.8±36 years) were selected for the present study. All subjects lived at their own home supporting themselves and fulfilled the normal daily activity levels based on ADL and IADL criteria. The exclusion criteria for the present study were as follows; active infective, inflammatory, neoplastic and other moderate to severe chronic diseases. Subjects whose medical conditions had been stabilized with such usual cares as dietary regimens and/or antihypertensive, oral antihyperglycemic and antilipidemic medications were included to the study. Their weight also had remained stable in the previous 6 month before the study. All health information including physical, psychological and social aspects was obtained from the data based on the structured interview performed by trained nurses.

　

　All subjects were invited to the Life Planning Clinic in Tokyo from November 1, 2002 to June 30, 2003, where anthropometric measurements were performed in a standing position with the automated measurement system (TBF-210, Tanita Co.) that instantaneously gave such values as height, weight, BMI and % body fat mass (bioelectric impedance method by TBF-110/TBF-210, Tanita). The following limb circumferences were measured in each subject by means of a flexible tape: (a) mid-arm circumference (MAC), at the mid point between the tip of the acromion and the olecranon process, (b) mid-thigh circumference (MTC) at the mid point between the inguinal fold and the proximal margin of the patella and (c) calf circumference (CC) at its largest level. Skin folds were measured at the left side of the dorsal aspect of the triceps with a Holtain scaffold caliper and the triceps muscular areas (TMA) was derived from the following formula: TMA =(MAC) ²/4π . The average value of triplicate measurements was used in the statistical analysis. All these processes were conducted by a trained technician.

　

　Handgrip test was performed on the dominant hand with a dynamometer (Grip Strength Dynamometer, Takei Co) and the result thus obtained was expressed as kg unit. Walking speed was expressed as sec/5m, and resting metabolic rate (RMR) was tested with METAVINE-N (Vine Bio-Dynamic System, Inc.) and was expressed as kcal/min. Bone mineral density (BMD) was measured on the lumbar spines (2nd, 3rd, 4th spine) with dual-energy x-ray absorptiometry (DEXA) and was expressed as g/cm².

　

　Several laboratory variables including blood cell counts (RBC, WBC, platelets, lymphocyte), hemoglobin, creatinine. BUN, Na, K, total cholesterol, total protein, albumin, hemoglobin Alc, serum osmolarity, luteinizing hormone (LH: RIA) , cortisol (RIA), free testostereone (RIA), dehydro-epiandrosterone sulfate(DEAS: RIA) , and interleukin-6 (IL-6: EIA) were obtained from the blood drawn at non-fast state and all measurements were performed at BML Co. (Tokyo) with above mentioned individually specified method.

　

　Statistical analyses were performed using Statistical Analysis System (SPSS Base 11.0J, 2001, SPSS Inc) . All anthropometric variables were expressed as mean ± SD and percentile expressions (15^th, 50^th, 90^th). Other laboratory variables were expressed only with mean±SD. Sex specific mean values with SDs for each anthropometric variable were calculated separately for two age groups; the septuagenarians and octo-/nonagenarians. Sex and age group differences between corresponding values were analyzed by Student-t test and the level of statistical significance was fixed at 0.05.

　

　

　Table 1 shows demographic findings obtained from structured interview and, as a whole, over-all prevalence of selected co-morbidities except for hypercholesterolemia was definitely lower in this particular elderly group than in ordinary Japanese elderly (the data had already described elsewhere) ⁵⁾. Concerning gender difference, cerebrovascular accident was more prevalent in male than in female group, while hypercholesterolemia was higher in female than in male group. Other items did not show significant differences between two genders.

　Table 2 and 3 present sex specific mean values and standard deviations for the measured anthropometric and blood variables of all ages. The male group was taller and heavier than the female group with concomitant significant differences in BFR, Sf, TC, TMA, MTC and CC. BMI in itself was larger in the male than in female group. Concerning blood variables, hemoglobin, creatinine, BUN, free testosterone and DEAS were higher in male than in female group. On the other hand, total cholesterol, Na, and LH were higher in female than in male group. There were no significant differences observed in K, total protein, albumin. CRP, IL-6, lymphocyte, hemoglobin Alc, osmolarity and cortisol.

　Selected functions possibly related to anthropometric characteristics measured in the present study were shown in Table 4, in which each function was compared between two genders. A11 items were more superior in the male than in the female subject.

　Age specific mean values and standard deviations in each gender were shown in Table 5 and 6, in which age-dependent differences were observed for weight, TC, MCT in the male and only height in the female group.

　Table 7 and 8 show percentile values for each anthropometric measurement and selected functions in both genders and those can be used as references for these age groups.