Osteoporosis is an age-related disorder characterized by decreased bone mass and increased susceptibility to fractures.1 Common fracture sites include the spinal vetebrae, hip and wrist.2 The condition contributes to as many as 50% of all adult fractures,3 or an annual incidence in the US of over 1 million.1 It is particularly prevalent among elderly Caucasian women. The acute suffering which occurs as a result of osteoporotic fractures is certainly sufficient to warrant our concern as gerontologica! nurses. Even more importantly, however, is the deleterious effect these fractures have on the functional ability of the older adult. In this population, fractures, particularly hip fractures, are strongly associated with loss of independence and nursing home placement.5
With regard to osteoporosis treatment, there is both good and bad news. The good news is that treatment of osteoporosis has advanced significantly in the past decade and a number of new therapeutic regimens have been developed which appear promising in slowing bone loss. The bad news is that it is difficult to detect the condition early in the disease process so that treatment can be initiated. The most desirable way of dealing with osteoporosis is through primary prevention, which means detecting and treating individuals who are likely to develop the condition before it occurs. Counseling individuals regarding calcium intake, exercise, and risk factors for osteoporosis is one strategy for primary prevention. Unfortunately there is no way at present to predict with sufficient accuracy who will get osteoporosis so as to more effectively target these preventive interventions. Secondary prevention means identifying individuals who have the condition earlier in the disease process. Radiologie techniques have been developed to accomplish this end. However, the tests are expensive and have not been approved by major third-party payers for screening because of their marginal efficiency in identifying the osteoporotic individual.6 As a result, their use is confined primarily to the management of patients who have been previously diagnosed with osteoporosis on the basis of a fracture. Thus a cost-effective, readily applicable method is needed to identify the osteoporotic patient in order to prevent further collapse of the spine and hip fracture.
One method of identifying osteoporotic individuals, which has been postulated by several researchers, is to detect a significant loss of stature.7-8 Loss of height is frequently the first clinical sign of osteoporosis. This height loss occurs in the trunk as a result of vertebral fracture and collapse. Clinically, the diagnosis of a vertebral fracture may go undetected since these fractures sometimes occur without pain. Even when pain is present, the individual may not seek medical evaluation because it may be temporary and relieved by simple analgesic medications and bed rest. Although a small amount of height is lost through normal aging, if it can be established that the magnitude of height loss exceeds that expected through aging, then osteoporosis is likely to be present. It should be noted that damage to the spinal vertebrae usually precedes bone loss in the hip by an average of 10 years.9 Individuals whose osteoporosis is identified and treated at this comparatively early stage of the illness might avoid hip fracture with its disabling consequences.
In order to examine height loss as a marker of clinical osteoporosis we designed a study to answer the following questions:
1 . Does the determination of height loss accurately identify individuals with osteoporosis?
2. How much height must be lost before osteoporosis is suspected?
3. What is the best method of determining height loss?
Sixty-five female subjects were recruited from a clinic specializing in the identification and treatment of osteoporosis. Subjects ranged in age from 3 1 to 85 with a median age of 63 . Height loss was determined using two different methods. The first technique, referred to as the Arm Span Method, is based on the fact that arm span closely approximates mature adult height10·11 and does not change significantly with aging. Height loss is calculated by subtracting the current height from the arm span. The second technique, referred to as the Recall Method, utilizes the subject's recall of her maximum adult height. For this method, height loss is calculated by subtracting the current height from the subject's recall of her maximum height.
Arm span was measured by having the subject face the wall with the arms spread laterally and palms against the wall. As the subject stretched her arms as far as possible, a pencil mark was placed at the end of both middle fingertips and the distance between the marks was measured in inches with a tape measure. Height was measured by having the subject stand barefooted with her back against a rigid measuring standard attached to the wall. A flat carpenter's level was placed on the head with the subject looking straight ahead and the height was read at the measuring standard.
Spine x-rays were available on all subjects. Osteoporosis was defined by the presence of one or more compression fractures of the spine on the x-ray.
Threshold values of height loss were selected by convenience and examined as predictors of osteoporosis. For example, a 1-in threshold means that individuals who have lost 1 or more inches in height are predicted to have osteoporosis and those losing less than 1 in are predicted not to have osteoporosis. We then compared these predictions to the actual diagnosis based on x-ray. In this way we were able to determine which amount of height loss best discriminated subjects with actual osteoporosis from those without osteoporosis. We examined, in half-inch increments, height losses of .5 in to 3 in.
For the Recall Method we found that the most accurate predictor of osteoporosis was a height loss of 2 in. In our sample, 75% of individuals who had lost 2 or more inches in height were found to have osteoporosis on the x-ray. Only 21% of those who lost less than 2 in had osteoporosis on the x-ray. These results were statistically significant at p<.001.
In contrast, the Arm Span Method produced results which were of much less value in predicting osteoporosis. With this technique, the 2-in level of height loss again produced the best correlation with osteoporosis. Only 55% of those testing positive, however, actually had osteoporosis on the x-ray, while 28% of those testing negative had osteoporosis on the x-ray. These results were significant at p < .05.
The results of this study indicate that women who have undiagnosed compression fractures can be identified earlier in the course of their illness by using this screening method. Specific recommendations for practice are summarized in the Figure. The technique requires no expensive or bulky equipment, takes but a few seconds to do, and can be easily implemented in a variety of nursing practice settings. An important limitation of this screening technique is that a negative test does not rule out osteopenia or early bone loss. This must be conveyed to women who test negative in order to prevent the mistaken impression that they are at reduced risk of osteoporosis.
RECOMMENDATIONS FOR PRACTICE
The degree of disparity between the two methods of determining height loss was a surprising finding in this study. In measuring arm span we observed that there seemed to be considerable random variation between arm span and height. For example it was not uncommon to find subjects whose stature significantly exceeded their arm span. This would not be expected if arm span closely approximated mature adult height as defined in the study. This variability greatly reduces the sensitivity of the Arm Span Method in predicting osteoporosis. We therefore recommend using the Recall Method to determine height loss.
As mentioned previously, there is some loss of height which results from normal aging. Some causes of this height loss include thinning of weightbearing cartilage including the intervertebral discs, and flattening of the arches. We postulate, therefore, that the opti mal threshold of height loss for predicting osteoporosis may vary with age. A loss of stature of 2.5 in may be normal for an 80-year-old woman. In contrast, a loss of 1 .5 in in a 55 year old may signal osteoporosis. This hypothesis was supported by our data; however, the size of the sample was not large enough to allow meaningful statistical analysis of specific age groups. Future research should be conducted to delineate optimal thresholds of height loss for specific age groups.
- 1. National Institutes of Health. Osteoporosis (Publication Number 1984-121-132:4652). Washington, DC, US Government Printing Office, 1984.
- 2. AIoia JF, Cohn SH, Vaswani A, et al: Risk factors for postmenopausal osteoporosis. Am J Med 1985;78(1):95-100.
- 3. Melton LJ, Riggs BL: Epidemiology of agerelated fractures, in LV Avioli (ed): The Osleoporotic Syndrome: Detection, Prevention and Treatment New York, Gruñe & Stratton, 1983. pp 45-72.
- 4. Rowe JW, Kahn RL: Human aging: Usual and successful. Science 1987; 237:143-149.
- 5. Perry BC: Falls among the elderly: A review of the methods and conclusions of epidemiologie studies. J Am Geriatr Soc 1982; 30:367-371.
- 6. Hall FM, Davis MA, Baran DT: Bone mineral screening for osteoporosis. N Engl J Med 1987; 316(4):212-214.
- 7. Albarese AA: Bone loss: Causes, detection, and therapy. New York, Alan R. Liss, 1977, pp 101-103.
- 8. Dequeker JV, Baeyens JP, Claessens J: The significance of stature as a clinical measurement of aging. J Am Geriatr Soc 1969; 17(2):169-179.
- 9. Riggs BL, Melton LJ: Involutional osteoporosis. N Engl J Med 1986; 314:1676-1684.
- 10. Brown OT, Wigzell FW: The significance of span as a clinical measurement, in Anderson WF, Isaacs B (eds): Current Achievements in Geriatrics. London, Cassell, 1964, pp 246-251.
- 11 . Harris JA, Jackson CM, Patterson DG, et al: The Measurement of Man. Minneapolis. University of Minnesota Press, 1930.
RECOMMENDATIONS FOR PRACTICE