Association between standing height and physical disability among U.S. adults aged 60 years and older: findings from NHANES 2015–2018

Background Physical disability is an important cause of affecting the quality of life in the elderly. The association between standing height and physical disability is less studied. Purpose The purpose of this study is to investigate the possible link between standing height and physical disability among U.S. adults aged 60 years and older. Methods The cross-sectional data were obtained from the US National Health and Nutrition Examination Survey (NHANES) 2015–2018. Physical disability was assessed by six questions: “Have serious difficulty hearing (SDH)?”, “Have serious difficulty seeing (SDS)?”, “Have serious difficulty concentrating (SDC)?”, “Have serious difficulty walking (SDW)?”, “Have difficulty dressing or bathing (DDB)?” and “Have difficulty doing errands alone (DDEA)?”. Responses to these questions were “yes” or “no”. Answer yes to one of the above six questions was identified as physical disability. Standing height (cm) was measured with an altimeter. Multivariate logistic regression was performed to examine the possible link between standing height and physical disability after adjustment for all covariates. Results A total of 2624 participants aged ≥ 60 years were included in our study, including 1279 (48.7%) females and 1345 (51.3%) males. The mean age of participants was 69.41 ± 6.82 years. After adjusting for all potential confounders, the inverse relationship between standing height and all physical disability (APD) was statistically significant (OR = 0.976, 95%CI:0.957–0.995). In addition, among six types of physical disability (SDH, SDS, SDC, SDW, DDB, DDEA), standing height was also a protective factor for SDW (OR = 0.961, 95%CI:0.939–0.983) and DDEA (OR = 0.944, 95%CI:0.915–0.975) in the full-adjusted model. Conclusion The cross-sectional population based study demonstrates that standing height is a protective factor for physical disability among U.S. adults aged 60 years and older.


Introduction
Disability is defined as an umbrella term for impairments, activity limitations, and participation restrictions [1].Physical disability is one of the most severe stages of disability, including loss of limbs, motor dysfunction, hearing impairment, visual impairment and etc [2].People with physical disabilities are often at a disadvantage in their daily lives.The onset of physical disability is a dynamic process [3].As health problems accumulate, people will eventually lose their mobility.Compared with the young, the elderly are more prone to physical disability due to the decline of physical function and immunity [4][5][6].The World Health Organization (WHO) estimated the prevalence rate of disability was 10.2% in adults over 60 years around the world [5].Meanwhile, one report said the prevalence rate of disability in mobility, hearing and vision in Americans aged 65 years and older was 26.9%, 14.9% and 6.6%, respectively [7].Furthermore, physical disability is an important risk factor for many diseases in the elderly.Some previous studies have demonstrated strong associations between physical disability and the occurrence of depressive symptoms, diabetes, stroke and heart disease [8][9][10][11].Physical disability has become a severe public health problem worldwide.Different factors may influence physical disability in different ages and in different populations.Therefore, identifying the risk factors related to physical disability in the elderly and implementing effective countermeasures are critical.
Standing height is one of the individual unmodifiable factor, so it is rarely the focus of the studies.Currently, standing height has been shown to have implications on human health and can be used to predict adverse outcomes [12].For example, a previous cohort study showed standing height was inversely associated with gestational diabetes mellitus (GDM) [13].For per 5-cm increase in standing height, the risk of GDM decreased by 19%.In addition, a meta-analysis including twelve cohort studies demonstrated greater height is linked with increased pancreatic cancer risk [14].Each 5-cm height increment had a 7% increased risk of pancreatic cancer.Moreover, a review has also reported that height was an independent risk factor for atrial fibrillation [15].However, to our knowledge, the association between standing height and physical disability in the elderly is unknown and has not been investigated.Unraveling the association will help to determine the prevention and treatment strategies of physical disability.
This study aimed to investigate the association between standing height and physical disability in the elderly.We explored the association by analyzing data from the US National Health and Nutrition Examination Survey (NHANES) 2015-2018.

Study population
The data was derived from NHANES database(2015-2018).The NHANES is a cross-sectional survey conducted by the National Centre for Health Statistics (NCHS).As a large clinical program in the United States, NHANES was designed to investigate the health status of Americans, involving demographic, dietary, health-related questionnaires, physical examination and laboratory data.The NHANES program was approved by the Ethics Review Board of the NCHS.Among 19,225 participants in NHANES 2015-2018, 11,526 participants were excluded due to missing data.Meanwhile, we also excluded 5138 participants aged < 60 years.2624 participants aged ≥ 60 years were eventually included in our study (Fig. 1).Complete details about NHANES can be accessed from https://www.cdc.gov/nchs/nhanes/index. htm.

Physical disability measurement
Physical disability was assessed by six questions: "Have serious difficulty hearing (SDH)?", "Have serious difficulty seeing (SDS)?", "Have serious difficulty concentrating (SDC)?", "Have serious difficulty walking (SDW)?", "Have difficulty dressing or bathing (DDB)?" and "Have difficulty doing errands alone (DDEA)?".Responses to these questions were "yes" or "no".Answer yes to one of the above six questions is identified as physical disability.

Standing height measurement
Standing height (cm) usually was measured with an altimeter.It requires the subject to stand in front of the height meter, the head upright, the back and heel close to the height meter, and then the surveyor read the value on the instrument to obtain the height of the subject.Standing height was categorized into quartiles (< 157.40, 157.40−164.99,165.00−172.20,> 172.20), and the lowest quartile was considered as the reference group.

Covariates
Sociodemographic variables included gender (female, male), age (years) and race (Mexican American, Other Hispanic, Non-Hispanic White, Non-Hispanic Black, Other Race).As a continuous variable, age was also categorized into four following groups: 60-64, 65-69, 70-74 and 75 years or older.Health-related variables included diabetes (yes, no), hypertension (yes, no), high cholesterol level (yes, no), trouble sleeping (yes, no), failing kidneys (yes, no), vigorous work activity (yes, no), hepatitis B (yes, no), hepatitis C (yes, no) and feeling depressed (not at all, several days, more than half the days, nearly every day).These variables were obtained by self-report.Physical examination variables included weight (kg), waist circumference (cm), upper leg length (cm), upper arm length (cm), arm circumference (cm), systolic blood pressure (SBP, mmHg) and diastolic blood pressure (DBP, mmHg).Leg and arm measurements were performed on the right side of the body.Measurement would be taken on the left side if a participant had an amputation or other adverse condition.Laboratory examination variables included albumin (mg/L), creatinine (µmol/L), high density lipoprotein cholesterol (HDL-C, mmol/L), total cholesterol (mmol/L), glycohemoglobin (%) and hypersensitive C-reactive protein (Hs-CRP, mg/L).These laboratory indicators were obtained by measuring the serum samples.

Statistical analyses
We used frequencies, percentages, means and standard deviation to describe variable characteristics.Continuous variables and dichotomous variables were analysed by analysis of variance (ANOVA) and Chi-Square (χ 2 ) test to calculate differences between four standing height groups.We created three logistic regression models to determine associations between standing height and physical disability among U.S. adults aged 60 years and older.Model 1 was unadjusted; Model 2 Adjusted for gender, age and race based on Model 1; Model 3 additionally adjusted for diabetes, hypertension, high cholesterol level, trouble sleeping, failing kidneys, vigorous work activity, hepatitis B, hepatitis C, feeling depressed, waist circumference, upper leg length, upper arm length, arm circumference, systolic blood pressure, diastolic blood pressure, albumin, creatinine, high density lipoprotein cholesterol, total cholesterol, glycohemoglobin and hypersensitive C-reactive protein based on model 2. Because weight existed severe collinearity with other variables, we did not include weight as an adjusted variable in model 3.The association between standing height and physical disability was further explored by subgroup analyses stratified by gender, age and race.Moreover, we used restricted cubic spline (RCS) model to detect the possible nonlinear dose-response relationship between standing height and physical disability.The significant level is P < 0.05.R version 4.1.0and SPSS version 25.0 were used to perform all statistical analyses.

Characteristics of participants
Table 1 showed the characteristics of participants.A total of 2624 participants aged 60 years and older were included in our study, including 1279 (48.7%) females and 1345 (51.3%) males.The mean age of participants was 69.41 ± 6.82 years and 1099 (41.9%) participants reported that they were Non-Hispanic White.For the diseased condition, 722 (27.5%) participants had diabetes mellitus, 1563 (59.6%) participants had hypertension, 1485 (56.6%) participants had high cholesterol level, 169 (6.4%) participants had failing kidneys, 42 (1.6%)participants had hepatitis B, and 52 (2.0%) participants had hepatitis C.   Characteristics of all participants were stratified by quartiles of standing height, from which statistically significant differences were found in gender, age, race, systolic blood pressure, diastolic blood pressure, weight, waist circumference, upper leg length, upper arm length, arm circumference, creatinine, HDL-C, total cholesterol, diabetes, vigorous work activity, hepatitis B, feeling depressed, SDH, SDW and DDEA (all P < 0.05).Participants with SDW and DDEA had a lower standing height.There were no significant difference in albumin, glycohemoglobin, hypersensitive C-reactive protein, hypertension, high cholesterol level, trouble sleeping, failing kidneys, hepatitis C, SDS, SDC and DDB among quartiles of standing height (all P > 0.05).

Association between standing height and physical disability
Standing height was included as a continuous variable for the subsequent analysis.We found standing height was a protective factor for all physical disability (APD) after performing a logistic regression analysis (Table 2).The association was statistically significant in the full-adjusted model (Model 3) (OR= 0.976, 95%CI: 0.957-0.995).The RCS model results showed a negative linear relationship between standing height and the risk of APD (P for nonlinear = 0.082) (Fig. 2).For per 1-cm increase in standing height, the risk of APD decreased by 2.4%.Among the six types of physical disability, standing height was a protective factor for SDW (OR= 0.961, 95%CI:0.939-0.983)and DDEA (OR= 0.944, 95%CI:0.915-0.975).These relationships were significant in Model 3. The RCS model results (Fig. 3) also demonstrated a negative linear relationship between standing height and the risk of SDW (P for nonlinear = 0.649) and a negative nonlinear relationship between standing height and the risk of DDEA (P for nonlinear = 0.034).For every 1-cm increase in standing height, the risk of SDW and DDEA decreased by 3.9% and 5.6%, respectively.There was no significant association between standing height and SDH, SDS, SDC, and DDB after adjusting for all potential confounders.

Subgroup analyses
Subgroup analyses of gender, age and race were conducted to further identify the relationship between standing height and physical disability.Subgroup analyses results were showed in Table 3.After adjusting for all potential confounders, the inverse association between standing height and APD was statistically significant in female (OR = 0.968, 95%CI:0.941-0.997),age ≥ 75 years (OR = 0.950, 95%CI:0.916-0.984)and the race of Non-Hispanic White (OR = 0.967, 95%CI:0.939-0.995).In these populations, we also found standing height was a protective factor for SDW and DDEA in Model 3.Moreover, through subgroup analyses, we observed that the negative relationship between standing height and SDC was statistically significant in the age group of 70-74 years (OR = 0.909, 95%CI:0.839-0.984)and the race of Other Hispanic (OR = 0.889, 95%CI:0.810-0.975) in model 3.

Discussion
Disability is an important cause of affecting the quality of life and causing death among the elderly.As the proportion of the elderly population increases globally, the prevalence of physical disability will increase substantially.In the nationally representative survey, we demonstrated that standing height was a protective factor for APD after adjusting for all potential confounders.For per 1-cm increase in standing height, the risk of APD decreased by 2.4%.In addition, a negative dose-response relationship between standing height and risk of APD was also showed by RCS model.Besides, among six types of physical disability, we also found standing height was a  Our findings were in line with previous similar studies reporting the protective effect of standing height on physical disability.A cohort study including 421 rheumatoid arthritis patients from Northern European origin showed that adult height is inversely linked with impairment of joint function and overall disability [16].In addition, a previous study in Japanese participants found that taller (≥ 170 cm for men and ≥ 160 cm for women) people were 16% less likely to report functional limitation in comparison with shorter (< 155 cm for men and < 145 cm for women) individuals [17].Moreover, it is widely accepted that fracture is one of the important reasons for physical disability in the elderly.A global longitudinal study of osteoporosis in 52,939 postmenopausal women also demonstrated a negative association between height and upper arm/shoulder and clavicle fractures [18].For per 10-cm height increment, the risk of upper arm/shoulder and clavicle fractures decreased by 15% and 27%, respectively.In short, the above studies indicated that standing height may be inversely associated with the risk of physical disability in the elderly.The possible biological mechanisms are as follows: Taller people usually have wider bones and may counteract the adverse effects of cortical structural changes, and shorter people with narrower bones may thus be more likely to have fractures and eventually cause physical disability [18,19].
Our findings are not consistent with several previous studies.A recent cross-sectional investigation among 155 Japanese older adults found that participants with higher height were more likely to report a decline in activities of daily living (ADL) [20].Including seven prospective studies, a meta-analysis from US, Norway and other European countries in 2016 demonstrated positive association between height and risk of hip fracture [21].Meanwhile, a cohort study involving 796,081 postmenopausal women with a follow-up period of 8 years also suggested taller women were at increased risk of fracture [22].These seem to contradict the conclusions of our study.There are several possible reasons for these inconsistent conclusions.First, participants in our study were older than 60 years, while participants in almost all other studies were a mix of younger (under 60 years) and older adults.Second, the role of confounders may contribute to inconsistent results.Previous studies did not adjust some possible confounders, such as SBP, DBP, HDL-C, total cholesterol, glycohemoglobin, etc [23,24].Instead, our study considered more potential confounders (24 factors) than others.Thus our findings are even more convincing.Third, different study populations may be one of the reasons why the conclusions are inconsistent.Currently, few studies investigate the relationship between standing height and physical disability in the elderly.More studies are needed to further clarify this relationship in the future.
Our subgroup analyses showed inverse significant relationship between standing height and APD in females in Model 3, but this relationship was not observed in males.Related studies to explain the sex differences are rare.For females, higher height may reduce the risk of APD.More studies should be conducted to explore the cause of sex differences in the future.For age and race, after adjusting for all confounders, the negative association between standing height and APD was statistically significant in participants aged ≥ 75 years and the race of Non-Hispanic White, while other subgroups had no significant association.These associations may not reach statistical significance due to reduced statistical power.More evidence should be explored in a larger sample population in future studies.Our findings have clinically important implications.With aging populations, the increase in physical disability in older adults is a public health problem worldwide.As this trend develops, more older people will eventually lose their ability to care for themselves.Besides, the increase in physical disability may also lead to an increased risk of many chronic diseases [25].As an individual basic characteristic, standing height should be considered as a predictor of physical disability in adults aged 60 years and older.
Our study has several strengths.On the one hand, our study adjusted more possible confounders than others.Therefore, our results were more credible.On the other hand, we classified all physical disability (APD) into six categories (SDH, SDS, SDC, SDW, DDB, DDEA) and explored the relationship between standing height and six types of physical disability separately.Thus our study was more comprehensive.Meanwhile, there are some limitations in our study.First, NHANES was a cross-sectional study, so we were difficult to know the causal relationship between standing height and physical disability.In the future, interventional or prospective studies with larger sample sizes should be conducted to further confirm our conclusions.Second, data on physical disability were obtained by self-reporting from participants, which inevitably had a certain degree of subjectivity.Third, more than half participants were excluded from our study due to missing information, which might cause bias.

Conclusion
Our study suggests that standing height is a protective factor for APD after adjusting for all potential confounders.In addition, among six types of physical disability, inverse association between standing height and SDW and DDEA is also statistically significant in the fulladjusted model.It is necessary to pay close attention to changes in standing height and intervene in time.Future prospective studies with large sample sizes are still needed to verify our findings.

Fig. 1
Fig. 1 Flowchart showing the selection of study population

Fig. 2
Fig. 2 Restricted cubic spline of association between standing height and all physical disability(APD); Model 1: unadjusted; Model 2: Adjusted for gender, age and race; Model 3: adjusted for gender, age, race, diabetes, hypertension, high cholesterol level, trouble sleeping, failing kidneys, vigorous work activity, hepatitis B, hepatitis C, feeling depressed, waist circumference, upper leg length, upper arm length, arm circumference, systolic blood pressure, diastolic blood pressure, albumin, creatinine, high density lipoprotein cholesterol, total cholesterol, glycohemoglobin and hypersensitive C-reactive protein

Fig. 3
Fig. 3 Restricted cubic spline of association between standing height and different types of physical disability after adjusting all confounders; SDH: Serious difficulty hearing; SDS: Serious difficulty seeing; SDC: Serious difficulty concentrating; SDW: Serious difficulty walking; DDB: Difficulty dressing or bathing; DDEA: Difficulty doing errands alone

Table 1
Characteristics of study population according to levels of standing height (N = 2624)

Table 2
Association between standing height and physical disability: models with standing height as a continuous variable Abbreviations APD: All physical disability; SDH: Serious difficulty hearing; SDS: Serious difficulty seeing; SDC: Serious difficulty concentrating; SDW: Serious difficulty walking; DDB: Difficulty dressing or bathing; DDEA: Difficulty doing errands alone; Model 1: unadjusted; Model 2: Adjusted for gender, age and race; Model 3: adjusted for gender, age, race, diabetes, hypertension, high cholesterol level, trouble sleeping, failing kidneys, vigorous work activity, hepatitis B, hepatitis C, feeling depressed, waist circumference, upper leg length, upper arm length, arm circumference, systolic blood pressure, diastolic blood pressure, albumin, creatinine, high density lipoprotein cholesterol, total cholesterol, glycohemoglobin and hypersensitive C-reactive protein

Table 3
Subgroup analyses of association between standing height and physical disability Abbreviations APD: All physical disability; SDH: Serious difficulty hearing; SDS: Serious difficulty seeing; SDC: Serious difficulty concentrating; SDW: Serious difficulty walking; DDB: Difficulty dressing or bathing; DDEA: Difficulty doing errands alone; Model 1: unadjusted; Model 2: Adjusted for gender, age and race; Model 3: adjusted for gender, age, race, diabetes, hypertension, high cholesterol level, trouble sleeping, failing kidneys, vigorous work activity, hepatitis B, hepatitis C, feeling depressed, waist circumference, upper leg length, upper arm length, arm circumference, systolic blood pressure, diastolic blood pressure, albumin, creatinine, high density lipoprotein cholesterol, total cholesterol, glycohemoglobin and hypersensitive C-reactive protein