日本財団図書館（電子図書館）　高齢者ケア国際シンポジウムの開催

1st(1990) Toward on Aging Society without Anxiety

Special Report
MUSCLE STRENGTH, LIFESTYLE AND FALL RATES IN JAPANESE AND AMERICAN NURSING HOME RESIDENTS

Assistant Prof. Harvard Medical School

Lewis A. Lipsitz. M.D.

A cross-cultural study of muscle strength and falls in elderly Japanese and American nursing home residents: medications are the message. LA Lipsitz. T Hirayama, I Nakajima, M Kelley, R Ruthazer, and H Izumo. Hebrew Rehab. Ctr for Aged, Boston, MA and Hirayama Nursing Home, Chiba. Japan.
Recent studies suggest a relationship between falls and quadriceps muscle weakness in elderly residents of U.S. nursing homes (NHs). Little is known about falls in Japanese NHs where elderly residents with the custom of squatting to toilet or sleep on the floor, may have increased quad strength and fewer falls. To examine this hypothesis, we evaluated disease histories, and current medications, quad strength, and mobility of comparable ambulatory American (N-50) and Japanese (N-60) NH residents, then followed these residents prospectively for the development of falls. Project staff from both sites trained together to assure uniform data ascertainment.
Japanese NH residents were similar in age, sex distribution, and functional status compared to their American counterparts. The Japanese took fewer medications (2+/-1 vs 4+/-2, p <.001), particularly diuretics, digoxin, antidepressants, antipsychotics, and nitrates. Despite preadmission squatting behaviours in 3/4 of Japanese, quad strength and gait characteristics were nearly identical in the two populations. However, the Japanese had fewer falls than Americans during 4 months follow up: 6/60 (10%) vs 14/38 (37%), respectively (p = .001). Logistic regression analyses showed that falls were strongly related to antidepressant or antipsychotic medication use (adjusted odds ratio = 69, p <.05), and gait speed (A.O.R. = .0002, p <.05) in the U.S. cohort, but not in Japan. Thus, falls are less frequent among Japanese NH residents, probably because of factors other than muscle strength. Psychotropic medications which were rarely used in Japan, may increase falls risk among institutionalized Americans.

Investigators:
Hebrew Rehabilitation Center for Aged Hirayama Nursing Home
Boston, Massachusetts
Lewis Lipsitz, MD
Hiroko lzumo, MD
Maria Fiatarone, MD
Susan Greenspan, MD
Peggy Kelley, RN
Robin Ruthazer, MPH
Douglas Levine
Chiba. Japan
Toshio Hirayama, MD
Ikumasa Nakajima, MD
Satoshi Tanaka, MD
Dr. Mochizuki
Nrs. Amagata
Nrs. Uchida
Mr. Kato

Falls are a common cause of morbidity and mortality among elderly people in all nations. In the United States falls are a major reason for admission to nursing homes. They may lead to serious fractures. with subsequent immobility; deconditioning and ultimately death. Only recently has the research community been focusing on the problem of falls in order to better understand their causes and prevention.
Falls are common among elderly people in the United States. Thirty-five to forty percent of community-dwelling elderly experience at least 1 fall each year. Up to 50% of institutionalized elderly experience falls annually. Approximately 6% of falls in the United States result in fractures.
Over the past few years, I have been conducting research into the pathophysiology and prevention of falls at the Hebrew Rehabilitation Center for Aged (HRCA). The HRCA is a 725 bed academic nursing home, closely affiliated with Beth Israel Hospital and Harvard Medical School in Boston. Elderly people living at the HRCA are divided into three levels of care from the most independent to the most dependent in their activities of daily living. The average age of residents at the Center is 88 years. Falls are quite common, occurring annually in 50% of the residents. Approximately one-third of the population experiences recurrent falls. Falls are less common among the independent residents, 11 % of whom experience recurrent falls. One-third of the most dependent residents fall multiple times in the course of a year. Sixteen percent of falters experience fractures and 6% fracture the hip.
At the HRCA we recently completed a study of causes of falls in the nursing home population (Lipsitz et al, J Gerontology 1991, in press). In this study we carefully evaluated 70 ambulatory nursing home residents with recurrent falls and compared their clinical characteristics to 56 controls with no history of falling over the preceding 2 years. A logistic regression analysis of clinical variables distinguishing falters from non-fallers revealed 4 significant factors that were highly associated with falling. These included the use of antidepressant medications, abnormal position sense in the great toe, taking more than 12 steps to turn a circle, and inability to stand from a chair without using the arms to push off. These results are similar to those from other studies of risk factors for falls in the nursing home. One variable which appears repeatedly in such studies is muscle weakness. The inability to stand from a chair without using one's arms to push off, suggests that there is weakness in the quadriceps muscles of the leg. Fallers have been found to have more lower extremity weakness than non-fallers. CT scans of the thigh reveal atrophy of the large muscles of the leg and an increase in the fat to muscle ratio in elderly subjects compared to young.
Much of the leg muscle weakness and atrophy observed in elderly individuals may be due to a sedentary lifestyle rather than aging per se. Thus, if aging is associated with a more sedentary lifestyle, the decline in activity may lead to leg muscle weakness and resultant falls and fractures.
Very little is known about fall rates or the relationship between muscle strength and falling in the Japanese population. When I last visited Japan 2 years ago. I was struck by the fact that elderly Japanese have lower bone mineral density than elderly Americans. yet also have a lower prevalence of hip fracture. On review of the literature I found very little data regarding falls and fractures in Japan. In 1966 Norton published his observations on population differences in vertebral osteoporosis from a random review of spine films obtained from hospital x-ray departments in Tokyo. Los Angeles. and Washington D.C. (Clin Ortho. 1966). He found an increased prevalence of osteoporosis in elderly Japanese women compared to women living in the United States. Despite this high prevalence of osteoporosis. Ross et al recently reported in abstract form that hip fracture rates are lower in Japan than the United States (JBMR S (2). August. 1990). This could be due to a lower rate of falling among elderly Japanese.
I am aware of only one study that examines annual fall rates in Japan. This study was performed by Haga et al al the Tokyo Metropolitan Home for the Elderly in 1982 (Arch Gerontol Geriatr, 1986). The investigators examined independently living males and females of average age 76 years. They reported a 15% annual rate of recurrent falls and 9% rate of fractures. However, this study was retrospective in design and depended on self report for the determination of falls and fractures. The study examined only recurrent falters. To compare these findings to those of a comparable population in the United States, I examined the rate of recurrent falls and fractures in independent nursing home residents living at the HRCA. There, the rate of recurrent falls is similar to that in Tokyo, 11 % in Boston versus 15% in Tokyo. However, in Tokyo only 9% of falters experienced fractures while 16% had fractures as a result of their fall in Boston. This supports the notion that fracture rates are lower in Japan.
If, indeed, fracture rates are lower in Japan, despite an increased prevalence of osteoporosis, elderly Japanese may fall less than elderly Americans. This might be due to lifestyle characteristics that increase quadriceps muscle strength in the Japanese. The lifelong custom of squatting to sleep on tatami mats or use Japanese-style toilets might increase quadriceps muscle strength and protect the elderly Japanese from the development of falls.
With funding from the Sasakawa Health Science Foundation and Japan Shipbuilding Industry Foundation, and in collaboration with Dr. Toshio Hirayama, I have been conducting a cross-cultural study of fall rates, muscle strength, and lifestyle characteristics in comparable elderly Japanese and American nursing home residents. This study has been facilitated by a team of investigators at the Hebrew Rehabilitation Center for Aged in Boston and the Hirayama Nursing Home in Chiba. Japan. In March of this year. Dr. Ikumasa Nakajima from the Yamanashi Medical College traveled to the HRCA in order to learn our data collection techniques. He then returned to Japan where he worked with Dr. Toshio Hirayama to institute the study in the Hirayama Nursing Home.
Our hypotheses at the outset of the study were three-fold:
(1) Fall rates are lower among elderly Japanese than American nursing home residents.
(2) The lifelong Japanese custom of squatting to use tatami mats or Japanese-style toilets results in stronger leg quadriceps muscle strength in elderly Japanese than American nursing home residents.
(3) Quadriceps muscle strength is inversely correlated with fall rates in elderly nursing home residents.
Residents from the Hirayama Nursing Home living in the Ro-ken nursing units were compared to pattern 1 and 2 residents of the HRCA. Both groups were ambulatory and less than 90 years of age. Their minimum length of stay was six months. The HRCA is a life care institution where residents stay for an average of 4.4 years. The mean age at the Hirayama Nursing Home is 80 while it is 88 at the HRCA. The male/female ratio is the same at both sites. The study was conducted in two phases. First, baseline data were obtained from residents and their medical records at both sites. These included medical history information, lifestyle characteristics, and formal gait, mobility, and quadriceps muscle strength measurements. Muscle strength was determined by the maximum weight that could be lifted at the ankle by a single leg extension at the knee (the 1 repetition maximum).
During a four month follow-up period (Phase 2) all residents were surveyed for the occurrence of falls. When a fall occurred information about the circumstances of the fall was obtained. A clinical assessment was performed on all subjects at the end of the follow-up period.
Fifty subjects from Boston were compared to 60 subjects from Chiba. The age was slightly younger in Chiba. Two-thirds of the subjects were female. previous falls were reported more frequently in Boston than in Chiba. A comparison of baseline subject characteristics showed very similar impairments in activities of daily living. Bed-rail restraints were used more commonly to prevent falls in the Boston population. Over half of the residents in each group maintained a relatively high level of physical activity. climbing stairs and occasionally leaving the institution. Fewer residents in Chiba were oriented to month. year and place than in Boston.
The Boston population had a higher prevalence of cancer, deforming arthritis, chronic lung disease, heart disease, diabetes, and hypertension. Also, many more medications were used in Boston. These included diuretics, nitrates, calcium channel blockers, digoxin, antidepressants, and antipsychotics. HRCA residents took an average of 4 types of mediations while Hirayama Nursing Home residents took an average of two. The Japanese nursing home residents were smaller in height, weight, and body mass index. They also had higher diastolic blood pressures.
Most importantly, quadriceps muscle strength was nearly identical in the two populations. When corrected for body mass index, right and left quadriceps muscle strength was the same in Boston and Japan. Other functional gait characteristics were also similar in the two populations, with the exception of the chair stand which was more frequently impaired in Chiba than Boston.
During the four month follow-up period, 14 out of 38 HRCA residents experienced falls (37%). In striking contrast, only 6 of 60 nursing home residents in Chiba experienced falls (10%, p=.001).
To determine the clinical factors associated with falls in Boston, we compared the characteristics of fallers to non-falters at the HRCA. A logistic regression analysis showed a highly significant independent relationship between the development of falls and the taking of psychotropic medications (antidepressants or antipsychotics) and slow walking speed (adjusted odds ratio for psychotropic modicactions=68.58, AOR for walking velocity=.0002, both significant at p <.05). These two variables correctly predicted 84.2% of the Boston fallers and none of the Chiba fallers. Only one of the nursing home residents in Chiba was taking antidepressant medications. None were taking antipsychotics.
When the Japanese nursing home residents were divided according to whether or not they used tatami mats and Japanese toilets prior to their nursing home admission, mean corrected quadriceps muscle strength was found the be higher among those with previous squatting behaviors. Thus, squatting may increase muscle strength. However, rapid de conditioning probably occurs once a resident becomes sedentary in the nursing home.
In conclusion. we have observed that fall rates are lower among Japanese nursing home residents than functionally comparable residents in an American nursing home. The Japanese nursing home residents take fewer antidepressant and antipsychotic medications than their American counterparts. In the United States these medications are strongly associated with the development of falls. Finally. Japanese and American nursing home residents have similar quadriceps muscle strength and functional gait characteristics. Life-long squatting behaviors may increase muscle strength among the Japanese.
This study has several important messages regarding the prevention of falls in the elderly. First. we in the United States can learn from the Japanese experience and strive to avoid psychotropic medications to prevent falls associated with their use. Perhaps we can learn alternative ways of managing depression and behavior problems from our Japanese colleagues. Also. it is important to recognize that many residents of nursing homes in both countries have marked muscle weakness that places them at increased risk of mobility impairment. falls and fractures. Therefore. we should continue to encourage muscle strengthening exercises in an attempt to maintain an active lifestyle and hopefully prevent the development of falls among nursing home residents.

CONCLUSIONS
(1) Fall rates are lower among Japanese NH residents than among comparable American NH residents.
(2) Japanese NH residents take fewer antidepressant and antipsychotic medications than their American counterparts. In America. these medications are strongly associated with the development of falls.
(3) Japanese and American NH residents have similar quadriceps muscle strength and functional gait characteristics. Lifelong squatting behaviors may increase muscle strength among the Japanese.