HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bermúdez, O. I Articles by Tucker, K. L Search for Related Content PubMed PubMed Citation Articles by Bermúdez, O. I Articles by Tucker, K. L Agricola Articles by Bermúdez, O. I Articles by Tucker, K. L

Development of sex-specific equations for estimating stature of frail elderly Hispanics living in the northeastern United States^1,2,3,4

	ABSTRACT

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Background: The accurate measurement of stature is not possible in many frail elderly persons because of problems affecting their ability to stand straight. In such cases, knee height may be used to estimate stature.

Objective: This study was designed to explore the applicability of published regression equations to estimate stature of Puerto Rican and other Hispanic elderly persons living in the northeastern United States and to formulate ethnicity-specific equations for these persons.

Design: The study subjects (60–92 y of age) included 569 Hispanics and a comparison group of 153 non-Hispanic whites. Equations to estimate stature of Hispanics and Puerto Ricans living in the northeastern United States were developed with regression models in a randomly selected subgroup of the Hispanics. These equations were tested with the remaining Hispanic subgroup.

Results: The published equations significantly overestimated stature of our Hispanic subjects. Equations developed for Massachusetts Hispanics and Puerto Ricans provided estimates of stature that did not differ significantly from measured stature. We found further that equations for non-Hispanic whites published in 1985 predicted statures of our relatively low-income, non-Hispanic white subjects better than did newer 1998 equations developed from a national sample.

Conclusions: The stature of elderly Hispanics from the northeastern United States can be estimated by using equations derived from the same population. These, or similar equations, should be used to estimate stature of frail elderly persons for whom standing height cannot be taken accurately. Socioeconomic status as well as ethnicity may affect results when knee height equations are used.

Key Words: Anthropometry • knee height • stature • elderly • Hispanics • Puerto Ricans • northeastern United States • HHANES • Hispanic Health and Nutrition Examination Survey • third National Health and Nutrition Examination Survey • NHANES III • Mexicans • Massachusetts Hispanic Elders Study

	INTRODUCTION

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
The rapid rise in the number of elderly persons in all US subpopulations (1) increases the need for ethnicity-specific interpretations of results from anthropometric indicators to evaluate the health and nutritional status of minority populations. Because stature is a component of several nutritional status indicators, measuring an individual's stature with accuracy is important. However, obtaining accurate measurements in some elderly subjects can be difficult because of postural problems, spinal deformities, or confinement to a wheelchair or bed. Research on loss of stature has been consistent in showing age and generational effects on the stature of elderly individuals (2–5), as well as the effects of diseases such as osteoporosis (6). There is also some indication that stature loss may differ significantly by economic status (7).

Knee height has been proposed to estimate the stature of elderly (8–10) and disabled (11, 12) subjects. The aging process does not affect the length of long bones, such as those in arms and legs, as it does vertebral height (13–15). Furthermore, predicted stature is considered a valid surrogate to be used in the construction of anthropometric indexes (10) and adjusting and interpreting of body-composition measurements (14, 16, 17).

Chumlea et al (8) developed equations to estimate the stature of elderly subjects using knee height. More recently, these authors proposed new equations to estimate stature from knee height using the nationally representative sample from the third National Health and Nutrition Examination Survey (NHANES III; 10). These new equations are specific for the 3 ethnic groups studied in NHANES III: non-Hispanic whites, non-Hispanic blacks, and Mexican Americans.

Because of the recency of publication of these 1998 ethnicity-specific equations, there is no information available about the application of such equations to other elderly populations. However, the ability of such equations to successfully predict stature in other ethnic groups seems questionable. Myers et al (18) showed that the 1985 Chumlea et al (8) equations led to systematic errors when applied to elderly Japanese Americans. Prothro and Rosenbloom (19) showed that ethnicity-specific equations were needed to predict the stature of elderly African Americans.

The study reported here originated from an epidemiologic survey, the Massachusetts Hispanic Elders Study (MAHES), which included primarily Puerto Ricans. The purpose of this analysis was to explore the applicability of recommended regression equations (8, 10, 20) to estimate the stature of an elderly Hispanic group residing in Massachusetts and, if necessary, to formulate ethnicity-specific equations that will allow valid estimations of the stature of those subjects who are difficult to measure because of postural problems.

	SUBJECTS AND METHODS

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Subjects
MAHES was a statewide survey conducted between 1993 and 1997, which included a representative sample of elderly Hispanics residing in Massachusetts and a neighborhood control group of non-Hispanic whites. One thousand thirty elderly subjects were examined. For the current study a subset of 722 subjects aged who were 60–92 y free of postural problems was used. This study group was composed of 153 non-Hispanic whites and 569 Hispanics (58% Puerto Ricans and 42% "other Hispanics"). The other Hispanics were a heterogeneous mix including Dominicans, Cubans, and Central and South Americans. There were only 5 Mexican Americans in this group.

Information on posture was collected during the anthropometric survey, which was approved by the Human Investigation Review Committee of Tufts/New England Medical Center. Subjects with kyphosis or other physical problems that prevented them from standing fully straight for their stature measurements (n = 207) or with leg amputations (n = 8) were excluded from analysis, as were 88 for whom there were incomplete anthropometric data. In addition, the ratio of knee height to total height was estimated. Those subjects outside the mean ± 3 SDs were considered outliers and were also excluded from this study (n = 5). In total, 308 subjects (210 Hispanics and 98 non-Hispanic whites) were excluded, leaving a working sample of 569 Hispanics and 153 non-Hispanic whites. The Hispanic subjects selected for this study were further randomly divided into 2 groups: a development group (n = 294), in whom regression equations were developed to estimate stature, and a validation group (n = 275), in whom the equations were tested.

Measurements
The anthropometric measures used in this study were stature (cm) and knee height (cm). Information about whether the subject had straight or stooped posture or other problems affecting posture was also recorded. Measures were taken in duplicate at the subjects' homes. A Harpenden pocket stadiometer (Holtain Ltd, Crosswell, United Kingdom) that included a base plate with an attached steel tape centimeter ruler was used to measure stature and sitting knee height. Measurements were recorded to the nearest 0.1 cm. Knee height was defined as the distance from the anterior surface of the thigh, just proximal to the patella, to the sole of the foot when the knee and ankle were flexed at a 90° angle (8).

Statistical methods
Descriptive analyses of the characteristics of the sample and cross-group comparisons were conducted by using the general linear models (GLM) procedure and 2 independent sample t tests. Pearson correlations were used to measure associations between stature, knee height, and age within each of the ethnic and sex groups. SPSS for WINDOWS (version 7.5; SPSS Inc, Chicago) was used for all analyses.

Previously specified equations are presented in Table 1. They include those developed in a group of 269 non-Hispanic whites (65–90 y of age) and originally recommended for elderly Americans by Chumlea et al (8), and the newly published equations for elderly non-Hispanic whites and Mexican Americans. These new equations were developed with 1369 non-Hispanic white men (aged 60–102.8 y), 1472 non-Hispanic white women (aged 60–98.5 y), 497 Mexican American men (aged 60–96.7 y), and 457 Mexican American women (aged 60–98.5 y) (10). The equations in Table 1 were tested for applicability to our study group. The differences between the measured stature of the study subjects and stature estimated with those equations were assessed by using paired t tests to evaluate the applicability of these formulas to estimate stature of MAHES subjects.

(1)

where a represents the intercept and b₁ and b₂ the regression coefficients. Regression equations were also generated to predict stature of the Puerto Rican segment of the development group because of the high proportion of this ethnic group in our Hispanic sample.

The regression equations were used to predict stature of MAHES subjects in the validation group. Paired t tests were performed to determine the magnitude and significance of the differences between measured stature and estimated stature produced with each of the Hispanic and Puerto Rican regression equations.

	RESULTS

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
There were no significant differences in age (based on t tests), stature, or knee height between the 2 Hispanic subgroups based on sex-specific results obtained with the GLM procedure with controls for age. In all ethnic groups men were taller and had a larger knee height than women (P 0.001). No significant sex differences in age were detected. The means and SDs for age, stature, and knee height, by sex and ethnic group from MAHES are shown in Table 2. In both sexes, MAHES elderly Hispanics were shorter in both stature and knee height than were the neighborhood comparison group of non-Hispanic whites. Differences between Hispanics and non-Hispanic whites in stature (P 0.001) and knee height (P 0.001) remained significant after GLM adjustment for age in both sexes. Puerto Ricans also differed from other Hispanics in age, stature, and knee height; these differences in stature (P 0.001) and knee height (men: P 0.001; women: P 0.01) remained significant after adjustment for age.

Pearson correlations between stature and knee height (Table 3) were strong and positive for each of the ethnic and sex subgroups of the study sample. Age was significantly and negatively associated with stature in each group of women and for non-Hispanic white men. Age was not significantly associated with stature in Hispanic men in total or for the Puerto Rican or other Hispanic subgroups. Knee height and age correlations were also negative but were only significant for non-Hispanic white and Puerto Rican women.

We also estimated non-Hispanic white stature using the 1985 and 1998 Chumlea et al equations (8, 10) (Table 4). The 1985 equation estimated stature of non-Hispanic white men accurately, with only 0.1 cm of difference between measured and estimated stature. In the case of the non-Hispanic white women, the 1985 equation overestimated stature significantly by 1.7 cm (P < 0.001). The newer 1998 non-Hispanic white equations (10) predicted stature of MAHES non-Hispanic white men with less accuracy, overestimating stature by 2.8 cm (P < 0.001). For women, the 1998 equation overestimated stature by >2.1 cm (P < 0.001), slightly more than did the 1985 equation.

Clearly, more specific predictors of stature for northeastern Hispanic elderly persons are needed. Therefore, we used the data from the development subset of the study group to formulate regression models to estimate stature from knee height and age of elderly MAHES Hispanics (Table 5). Knee height and age resulted in strong models for the explanation of the variance in stature of Hispanic women (R² = 0.73) and Puerto Rican women (R² = 0.70). Knee height alone made up the final model for stature of men in the Hispanic (R² = 0.72) and Puerto Rican (R² = 0.77) groups. The addition of age did not improve the regression models in either group of men.

	DISCUSSION

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
We used anthropometric data from 569 elderly Hispanics living in the Northeast to develop ethnicity-specific equations for estimating stature from knee height measurements. In addition, anthropometric data from a neighborhood comparison group of 153 elderly non-Hispanic whites were also presented and contrasted with the results from Hispanics. For each ethnic group, men were taller and heavier and had larger knee heights than women. Sex differences in stature or knee height of elderly subjects were documented previously (3, 5, 8, 19, 22).

After adjustment for age, elderly Hispanic men and women were shorter and had shorter knee heights than their respective non-Hispanic white comparison group. Racial and ethnic differences in stature of elderly subjects have also been reported in the literature (4, 5, 10). Ethnic differences in length of body segments have also been documented. For example, Chumlea et al (10), working with the NHANES III data reported that non-Hispanic whites had significantly larger knee height and sitting height than did their Mexican American counterparts.

When compared with the HHANES elderly Puerto Rican subjects, our MAHES Puerto Rican subjects were shorter, which may be due to differences in age ranges in the 2 samples. On the other hand, NHANES Mexican Americans, with a larger age range, were taller but had smaller knee heights than the MAHES Hispanics. Differences in stature between MAHES Hispanics and the nationally representative samples from HHANES and NHANES reinforce previous reports indicating differences among Hispanic subgroups (21).

Changes in stature have been estimated at 3.6–5.6 cm between the ages of 25 and 74 y by using the samples from the National Health Examination Survey (NHES, 1960–1962), NHANES II (1976–1980), and HHANES (1982–1984) (4). These were cross-sectional samples, in which changes in stature with age or with secular trends cannot be separated. Declines in stature with age, estimated with longitudinal studies, have been documented (22–25). For example, in the SENECA Study, a longitudinal study of elderly Europeans, there was a decrease of 1–2 cm in stature over 4 y from 70–75 to 74–79 y of age (22). Differences in stature between the youngest and oldest groups (22–70 y of age) in a longitudinal study of non-Hispanic white men, the Normative Aging Study, was estimated at 7.3 cm (24). The investigators in this study were able to estimate, over a 10-y follow-up period, the amount of the difference in height due to aging (4.3 cm) and to secular trends (3.00 cm) (24).

The significant, positive correlation between stature and recumbent knee height reported previously by other researchers (8, 18, 19) for elderly non-Hispanic whites was also observed in the elderly Hispanic and non-Hispanic white subjects studied here using sitting knee height. Analysis of reliability, when using recumbent or sitting position to measure knee height, has produced high reliability coefficients (8). Knee height is considered a reliable measure because it is easy to measure in either a sitting or a recumbent position, regardless of the individual's mobility (10).

Both Hispanic and non-Hispanic white MAHES women had significantly smaller knee heights with increasing age, a finding that has also been reported by others (8, 11). This could be explained by cohort differences; however, the cross-sectional sample used here does not allow further exploration of this issue. Age was not significantly associated with knee height for the Hispanic men included in the study sample. Similar results were reported for men in some previous studies (8, 18), although significant differences in knee height with age have been reported in other groups of men (10, 24).

The recently published equations for Mexican Americans (10) did not to predict accurately the stature of the MAHES subjects. This clearly shows that these equations should not be used for non-Mexican American Hispanic groups in the northeastern United States. The elderly Hispanics in this study were from different ethnic backgrounds—Puerto Rican, Dominican, Cuban, and Central and South American—and included only a few Mexican Americans. These MAHES Hispanics had shorter stature and larger knee height than the NHANES Mexican Americans, indicating a different proportional distribution of body segments. Further testing of these equations for Mexican Americans is needed in the southwestern United States, where they may be more appropriately used.

With the MAHES sample of northeastern non-Hispanic whites, the 1985 non-Hispanic white equations (8) predicted stature of men more accurately than did the new 1998 equations developed by Chumlea et al (10). The differences between measured and predicted stature of MAHES non-Hispanic white women were significantly different when using the 1985 and 1998 non-Hispanic white equations, although the difference obtained with the 1985 equation (1.7 cm) was slightly lower than the one obtained with the 1998 equation (2.1 cm). These findings, particularly the lack of precision of the 1998 equations, were unanticipated. The 1985 study was based on a small convenience sample of non-Hispanic whites who were free of postural problems, whereas the 1998 equations were based on the total sample from NHANES III without restrictions based on posture (10). The MAHES non-Hispanic whites included in the study reported here were also free of obvious postural problems, which may make them more similar to the 1985 group.

Cohort differences among the MAHES and the other non-Hispanic white groups could also be due to socioeconomic and ecologic conditions. The non-Hispanic white group of the MAHES study is a comparison group obtained from the same neighborhood in which the Hispanic subjects resided. In general, this sample had a lower-than-average socioeconomic level, making them different from the national population. A significant association between socioeconomic status and loss of stature has been reported, with a slower rate of decline for women in the high socioeconomic status categories compared with those in the lower category (7). Despite the inclusion of subjects with postural problems, the NHANES non-Hispanic whites were taller than the MAHES subjects, with mean stature of 174 cm compared with 170 cm for men and 159 cm compared with 157 cm for women, respectively. Mean ages were similar between the 2 groups at 71 y for men and women.

Puerto Ricans are the largest Hispanic subgroup living in the northeastern United States and they made up >50% of the MAHES sample of Hispanics. Therefore, we conducted specific analyses for this group and compared them with the rest of the elderly Hispanics living in Massachusetts. The differences observed between the Puerto Rican–specific and the more general Hispanic equations were not large. We therefore concluded that the use of only one set of equations for this group of Hispanics, including Puerto Ricans, was acceptable. For more specific work with Puerto Ricans or other Hispanic groups it is possible that more accurate results may be obtained with the use of subgroup-specific equations.

In conclusion, equations were developed to estimate the stature of elderly Hispanics residing in the northeastern United States. These equations will be used to estimate stature in subjects for whom standing height was not measured or in whom postural problems were indicative of unreliable measures of stature. Existing equations for elderly Mexican Americans are not appropriate for use with Puerto Ricans or for other heterogeneous groups of elderly Hispanics. The equations developed here for northeastern Hispanics and Puerto Ricans require further testing to determine their applicability in independent samples.

	FOOTNOTES

 
¹ From the Jean Mayer US Department of Agriculture Human Nutrition Research Center on Aging at Tufts University, Boston.

² The contents of this publication do not necessarily reflect the views or policies of the US Department of Agriculture, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government.

³ Supported in part by NIA grant AG10425-05 and US Department of Agriculture contract 53-3K06-01.

⁴ Address reprint requests to OI Bermúdez, Dietary Assessment Research Program, US Department of Agriculture/HNRCA, 711 Washington Street, Boston, MA 02111. E-mail: Bermudez_he{at}hnrc.tufts.edu.

	REFERENCES

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. US Bureau of the Census. Current population reports series. Population profile of the United States: 1995. Washington, DC: US Bureau of the Census, July 1995. (Report no.: series P23-189.)
2. Friedlaender JS, Costa PT Jr, Bosse R, Ellis E, Rhoads JG, Stoudt HW. Longitudinal physique changes among healthy white veterans in Boston. Hum Biol 1977;49:541–58.[Medline]
3. Chumlea WC, Garry PJ, Hunt WC, Rhyne RL. Distributions of serial changes in stature and weight in a healthy elderly population. Hum Biol 1988;60:917–25.[Medline]
4. Kuczmarski RJ. Need for body composition information in elderly subjects. Am J Clin Nutr 1989;50(suppl):1150S–7S.[Abstract]
5. Launer LJ, Harris T. Weight, height and body mass index distributions in geographically and ethnically diverse samples of older persons. Age Ageing 1996;25:300–6.[Abstract/Free Full Text]
6. Ensrud KE, Black DM, Harris F, Ettinger B, Cummings S. Correlates of kyphosis in older women. J Am Geriatr Soc 1997;45:682–7.[Medline]
7. Himes JH, Mueller WH. Age-associated statural loss and socioeconomic status. J Am Geriatr Soc 1977;25:171–4.[Medline]
8. Chumlea WC, Roche AF, Steinbaugh ML. Estimating stature from knee height for persons 60 to 90 years of age. J Am Geriatr Soc 1985;33:116–9.[Medline]
9. Chumlea WC. Anthropometric assessment of nutritional status in the elderly. In: Himes JH, ed. Anthropometric assessment of nutritional status. New York: Wiley-Liss, 1991:399–418.
10. Chumlea WMC, Guo SS, Wholihan K, Cockram D, Kuczmarski RJ, Johnson CL. Stature prediction equations for elderly non-Hispanic white, non-Hispanic black, and Mexican-American persons developed from NHANES III data. J Am Diet Assoc 1998;98:137–42.[Medline]
11. Chumlea WC, Guo SS, Steinbaugh ML. Prediction of stature from knee height for black and white adults and children with application to mobility-impaired or handicapped persons. J Am Diet Assoc 1994;94:1385–8.[Medline]
12. Johnson RK, Ferrara MS. Estimating stature from knee height for persons with cerebral palsy: an evaluation of estimation equations. J Am Diet Assoc 1991;91:1283–4.[Medline]
13. Parker JM, Dillard TA, Phillips YY. Arm span-height relationships in patients referred for spirometry. Am J Respir Crit Care Med 1996;154:533–6.[Abstract]
14. Hurley RS, Bartlett BJ, Witt DD, Thomas A, Taylor EZ. Comparative evaluation of body composition in medically stable elderly. J Am Diet Assoc 1997;97:1105–9.[Medline]
15. Mitchell CO, Lipschitz DA. The effect of age and sex on the routinely used measurements to assess the nutritional status of hospitalized patients. Am J Clin Nutr 1982;36:340–9.[Free Full Text]
16. Roubenoff R, Wilson PWF. Advantage of knee height over height as an index of stature in expression of body composition in adults. Am J Clin Nutr 1993;57:609–13.[Abstract/Free Full Text]
17. Han T, Lean M. Lower leg length as an index of stature in adults. Int J Obes Relat Metab Disord 1996;20:21–7.[Medline]
18. Myers SA, Takiguchi S, Yu M. Stature estimated from knee height in elderly Japanese Americans. J Am Geriatr Soc 1994;42:157–60.[Medline]
19. Prothro J, Rosenbloom CA. Physical measurements in an elderly black population: knee height as the dominant indicator of stature. J Gerontol 1993;48:M15–8.[Abstract]
20. Chumlea WC. Methods of nutritional anthropometric assessment for special groups. In: Lohman TG, Roche AF, Martorell R, eds. Anthropometric standardization reference manual. Champaign, IL: Human Kinetics Books, 1988:93–5.
21. Najjar MF, Kuczmarski RJ. Anthropometric data and prevalence of overweight for Hispanics: 1982–1984. Hyattsville, MD: National Center for Health Statistics, 1989.
22. SENECA Investigators. Longitudinal changes in anthropometric characteristics of elderly Europeans. Eur J Clin Nutr 1996;50(suppl):S9–15.
23. Noppa H, Andersson M, Bengtsson C, Bruce A, Isaksson B. Longitudinal studies of anthropometric data and body composition. The population study of women in Gotenberg, Sweden. Am J Clin Nutr 1980;33:155–62.[Abstract/Free Full Text]
24. Borkan GA, Hults DE, Glynn RJ. Role of longitudinal change and secular trend in age differences in male body dimensions. Hum Biol 1983;55:629–41.[Medline]
25. Hallfrisch J, Muller D, Drinkwater D, Tobin J, Andres R. Continuing diet trends in men: The Baltimore Longitudinal Study of Aging (1961–1987). J Gerontol 1990;45:M186–91.[Abstract]

This article has been cited by other articles:

				H. Lin, O. I. Bermudez, and K. L. Tucker Dietary Patterns of Hispanic Elders Are Associated with Acculturation and Obesity J. Nutr., November 1, 2003; 133(11): 3651 - 3657. [Abstract] [Full Text] [PDF]

				O. I Bermudez, W. Velez-Carrasco, E. J Schaefer, and K. L Tucker Dietary and plasma lipid, lipoprotein, and apolipoprotein profiles among elderly Hispanics and non-Hispanics and their association with diabetes Am. J. Clinical Nutrition, December 1, 2002; 76(6): 1214 - 1221. [Abstract] [Full Text] [PDF]

				H. Chen, O. I. Bermudez, and K. L. Tucker Waist Circumference and Weight Change Are Associated With Disability Among Elderly Hispanics J. Gerontol. A Biol. Sci. Med. Sci., January 1, 2002; 57(1): M19 - 25. [Abstract] [Full Text] [PDF]

				K. L. Tucker and S. Buranapin Nutrition and Aging in Developing Countries J. Nutr., September 1, 2001; 131(9): 2417S - 2423. [Abstract] [Full Text] [PDF]

				C. Castaneda, O. I Bermudez, and K. L Tucker Protein nutritional status and function are associated with type 2 diabetes in Hispanic elders Am. J. Clinical Nutrition, July 1, 2000; 72(1): 89 - 95. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bermúdez, O. I Articles by Tucker, K. L Search for Related Content PubMed PubMed Citation Articles by Bermúdez, O. I Articles by Tucker, K. L Agricola Articles by Bermúdez, O. I Articles by Tucker, K. L