Factors Associated with Performance-based Physical Function of Older Veterans of the PLAAF: A Pilot Study.

[Purpose] This study investigated the factors associated with performance-based physical function of older veterans of the People's Liberation Army Air Force of China (PLAAF).
[Subjects and Methods] A cross-sectional survey of 146 older veterans of the PLAAF was carried out. Their physical function was measured using the Chinese Mini-Physical Performance Testing (CM-PPT). The demographics and health status (including physical measures, blood chemical tests, chronic diseases, and number of morbidities) were collected from health examination reports and computer records of case history. Cognition was measured using the Mini-Mental Status Examination (MMSE).
[Results] In multiple linear regressions, age, MMSE, Parkinsonism, and chronic obstructive pulmonary disease were independently associated with CM-PPT, while previous stroke and albumin level reached borderline statistical significance. The association between the number of morbidities and CM-PPT was significant after adjustment for MMSE and demographics. The CM-PPT of low (0 or 1), medium (2 to 4) and high count (5 or more) morbidities were 11.3±3.9, 10.2±4.1, 6.1±3.8 respectively, and the difference among these three groups was significant.
[Conclusion] Some modified conditions and the number of chronic diseases might be associated with the physical function of older veterans of the PLAAF.

INTRODUCTION

The quality of life of older persons is more determined by their level of physical function and ability to maintain independence than by the specific diseases diagnosed by their physicians1). When physical function declines to the point where older persons cannot fully care for themselves, the burden on their families and the medical care system becomes substantial. With increasing age, there is a noticeable rise in the proportion of the population residing in nursing homes or living at home and needing the help of other people2). In addition, it has been consistently demonstrated that measures of physical function of the elderly can predict major health-related events, such as falls, disability, hospitalization, institutionalization, and mortality3,4,5,6). Therefore, it is imperative to understand and realize the determinants of physical functional decline. These determinants can not only be used as an early warning system for identifying persons at great risk of subsequent decline of physical function, but also to suggest therapy strategies to delay or even ameliorate physical function impairment, in order to prevent and postpone adverse health-related events.

A series of studies have reported that the age, cognitive function, life-style, psychosocial factors, and chronic medical conditions, are associated with the physical function of older persons. However, these studies have inconsistent results which may be related to differences in race and geography, in addition to different designs and methods1, 2, 7,8,9,10,11). As such, it is necessary to investigate these associations in populations of different races and countries, but this has not been carried out in China.

Physical function can be assessed by self-report or performance-based measures. Different from the self-report measures, physical performance tests (PPT) provide a multidimensional, objective and standardized assessment of elderly persons, and are less confounded by culture, language, or educational level1, 7, 12). In addition, PPT have been proved to be able to identify more limitations in physical function than self-report measures, and consequently they are more sensitive to changes leading to disability13,14,15). PPT have more recently been considered as the ‘gold standard’ for the measurement and follow-up of physical function.

Recently, we modified the Mini-Physical Performance Test of Wilkins et al.16) by replacing the 50-ft timed walk with a 20-ft timed walk, and adding one-leg standing to construct the Chinese version of the Mini-PPT (CM-PPT)17). CM-PPT has an unchanged score for the timed walk with the shortened distance, and has a more sensitive assessment of balance. Fitzpatrick et al. found that the tasks requiring greater functional reserve appear to be more sensitive at assessing the physical function than the usual tasks, and CM-PPT is required to be completed as quickly as possible18). A pilot study showed that CM-PPT is reliabe, and is more sensitive than the Barthel index (BI) at evaluating and monitoring the physical function of elderly Chinese17).

In the present cross-sectional study, we investigated the association of CM-PPT with demographics, cognition and health status in older veterans of The People’s Liberation Army Air Force of China (PLAAF). We found that aging, cognition, some chronic conditions, and a number of morbidities were associated with the physical function of older veterans of the PLAAF.

SUBJECTS AND METHODS

The data used in this study were collected in our previous study, in which 170 Veterans taking part in the 2010 health examination at the General Hospital of Air Force PLA were recruited for a study of the reliability and validity of CM-PPT17). Eligibility criteria were an age of 60 years or older, and the ability to follow a one-step command. Participants were excluded if they were enrolled in a hospice program with anticipated survival of less than 6 months, enrolled in a sub-acute or acute care program, or were unable to follow directions due to severe visual and hearing impairment, or profound dementia. In the present study, we selected only those persons with complete health examination and computer records of case history. Subjects with dementia were also excluded in order to make sure that the participants had completely comprehended their instructions. A cross-sectional survey was carried out of the remaining 146 persons. Information about their demographics, health status, cognitive and physical functions was collected. All procedures were approved by the Institutional Ethical Review Committee, and all participants provided their written informed consent to participation in this study.

The CM-PPT contains 4 physical performance tests: standing static balance, 20-ft timed walk (time of 10-ft forward and back), chair-stand time (time to sit in and rise from a chair with arms folded across the chest 5 times) and picking up a penny from the floor (a penny was placed approximately 12 inches from the subject’s foot on the dominant side, and the time taken to pick up the penny from the floor and stand up). Standing static balance also contains 4 items: feet together (standing still with the feet together), semi-tandem (standing with the heel of one foot placed to the side of the first toe of the opposite foot), tandem (standing with the heel of one foot directly in front of the other foot), and one-leg standing (standing unassisted on one leg with eyes open). Each test is scored from 0 to 4 points and the sum of the scores from the 4 physical tests ranges from 0 to 16, with higher scores indicating better physical performance. Standing static balance is scored by the ability to maintain the following positions: feet together for 10 seconds, semi-tandem for 10 seconds, tandem for 10 seconds, and one-leg for 5 seconds. Each position is scored. The 20-ft timed walk was scored as follows: a score of 1 for >11 seconds; a score of 2 for >9 and ≤11 seconds; a score of 3 for >7 and ≤9 seconds; a score of 4 for ≤7 seconds. The chair-stand time was scored as follows: a score of 1 for >15 seconds; a score of 2 for >12 and ≤15 seconds; a score of 3 for >9 and ≤12 seconds; a score of 4 for ≤9 seconds. Picking up a penny was scored as follows: a score of 1 for >4 seconds; a score of 2 for >3 and ≤4 seconds; a score of 3 for >2 and ≤3 seconds; a score of 4 for ≤2 seconds. A score of 0 indicates the subject is unable to do the test.

Demographics measures included age, sex, marital status, and years of formal education. Health status included physical measures, blood chemical levels, and chronic diseases. The conditions selected to be investigated in this study were based on the fact that they are common or have been reported in the literature as related to the physical function of older persons. Physical measures included systolic and diastolic blood pressures, body mass index (BMI), self-rated visual and hearing scales. BMI is weight (kg)/height (m2), and underweight was defined as BMI<18.5 and overweight as BMI>25. Visual assessment was divided into normal reading and poor reading (only large fonts), while the hearing assessment was divided into normal hearing and poor hearing (only loud talking). Blood measures included fasting blood sugar, total cholesterol, triglyceride, low and high density lipoproteins, blood creatine and urea nitrogen, blood uric acid, blood haematin, total protein, albumin. Chronic diseases included coronary disease, chronic heart failure (CHF), hypertension, hyperlipidemia, diabetes, sleep apnea syndrome (SAS), chronic obstructive pulmonary disease (COPD), neoplasm, arthritis, chronic renal failure, anemia, previous stroke, and parkinsonism. The number of chronic diseases was counted and categorized as low (0 or 1), medium (2 to 4) and high (5 or more). Cognitive function was measured using the Mini-Mental State Examination (MMSE)19).

Chronic diseases were collected from computer records of case history. The MMSE and self-rated visual and hearing assessment were completed by trained nurses. Other information was obtained from health examination reports.

The data of continuous variables is presented as mean values±SD, and the data of categorical variables as percentage. The associations of CM-PPT with demographics measures, all data of health status, and MMSE were examined using multiple linear regression. In Model 1, only demographic measures were included. In Model 2, all covariates of health status and MMSE were examined in separate demographics-adjusted regression models. In Model 3, the significant variables of Model 2 were simultaneously entered into a summary demographics-adjusted regression model. In Model 4, the number of morbidities was examined in the regression model after adjustment for MMSE and demographics. The correlation between CM-PPT and the number of morbidities is expressed by Pearson’s correlation coefficient. The difference of CM-PPT scores among the low, medium and high counts of morbidities was examined by analysis of variance. Statistical significance was accepted for values of p≤0.05, and all statistical analyses were performed using SPSS version 16.0.

RESULTS

The main characteristics of CM-PPT, demographics measures, health status and MMSE of 146 subjects are shown in Table 1. Except for BMI, there were complete clinical data for all subjects. Most of the participants were male in this study because our sample was comprised of PLAAF veterans. In Model 1, age was significantly associated with CM-PPT (Table 2): CM-PPT declined with age. In separate demographic-adjusted linear regression models, MMSE, albumin level, SAS, COPD, previous stroke, and parkinsonism were significantly associated with CM-PPT, in addition to age (Table 2). In the summary demographics-adjusted linear regression model, age, MMSE, COPD and parkinsonism were significantly associated with CM-PPT, and previous stroke and albumin level reached borderline statistical significance (Table 2). The number of morbidities was negatively associated with CM-PPT (r=0.337, p<0.001). This association still existed after adjusted for MMSE and demographics (p=0.007, Table 2). The CM-PPT of the low, medium and high count morbidities were 11.3±3.9, 10.2±4.1, and 6.1±3.8 respectively, and the difference among the three groups was significant (p=0.002).

Table 1.

Demographic measures, health status, MMSE and CM-PPT of the 146 particpants

Characteristics	Mean±SD	Range	Number (%)
Demographic measures
Age (year)	80.8±6.8	60–94
Sex (male)			136 (93.2)
Marital Status (married)			129 (88.4)
Educational levels
Illiteracy			1 (0.7)
≤6 years			10 (6.8)
≤12 years			63 (43.2)
>12 years			72 (49.3)
Physical measures
Systolic BP (mmHg)	130.1±15.4	90–180
Diastolic BP (mmHg)	73.3±9.3	58–110
BMI (kg/m2)
<18.5			3 (2.1)
18.5–25			70 (47.9)
>25			60 (41.1)
Unknown			13 (8.9)
Hearing (normal)			127 (87.0)
Vision (normal)			81 (55.5)
Blood chemistry
Fasting blood sugar (mmol/L)	5.5±1.3	3.6–11.9
Total cholesterol (mmol/L)	4.0±0.8	2.0–8.1
Triglyceride (mmol/L)	1.4±0.8	0.5–6.8
LDL (mmol/L)	2.6±0.7	0.4–6.2
HDL (mmol/L)	1.1±0.3	0.6–1.9
Creatine (umol/L)	108.2±102.2	48.0–795.0
Urea nitrogen (mmol/L)	6.6±3.2	2.0–23.4
Uric acid (umol/L)	344.6±95.1	129.0–667.0
Haematin (g/L)	124.0±19.0	69.0–163.0
Albumin (g/L)	40.5±3.2	33.0–49.0
Chronic diseases
Coronary disease			22 (15.1)
Chronic heart failure			13 (8.9)
Hypertension			92 (63.0)
Hyperlipidemia			20 (13.7)
Diabetes			44 (30 .1)
SAS			3 (2.1)
COPD			9 (6.2)
Neoplasm			33 (22.6)
Arthritis			7 (4.8)
Chronic renal failure			15 (10.3)
Anemia			61 (41.8)
Previous stroke			21 (14.4)
Parkinsonism			10 (6.8)
Number of morbidities
≤1			38 (26.0)
2–4			98 (67.2)
≥5			10 (6.8)
MMSE	27.0±2.6	20–30
CMPPT	10.2±4.2	0–16

SD = standard deviation; BP = blood pressure; BMI = body mass index; LDL = low density lipoproteins; HDL = high density lipoproteins; SAS = sleep apnea syndrome; COPD = chronic obstructive pulmonary disease; MMSE = Mini-Mental State Examination;CM-PPT = Chinese version Mini-Physical Performance Test

Table 2.

Associations of CM-PPT with demographics, physical health state, and MMSE by multiple linear regressions

	Model 1			Model 2			Model 3			Model 4
	B	β	P	B	β	P	B	β	P	B	β	P
Age	−0.352	−0.576	<0.001				−0.283	−0.463	<0.001	−0.281	−0.460	<0.001
Sex	0.055	0.003	0.963				0.300	0.018	0.777	0.301	0.018	0.786
Marital Status	−0.282	−0.044	0.547				−0.320	−0.050	0.450	−0.143	−0.022	0.745
Educational levels	−0.343	−0.054	0.484				−0.575	−0.090	0.194	−0.552	−0.086	0.231
SBP (mmHg)				0.010	0.036	0.614
DBP (mmHg)				0.010	0.023	0.748
BMI (kg/m2)
<18.5 vs 18.5~25.0				−0.053	−0.002	0.979
>25.0 vs 18.5~25.0				−0.080	−0.010	0.891
Hearing				−0.561	−0.046	0.543
Vision				−0.821	−0.099	0.162
Fasting blood sugar (mmol/L)				−0.325	−0.098	0.166
Total cholesterol (mmol/L)				0.243	0.048	0.497
Triglyceride (mmol/L)				−0.545	−0.099	0.157
LDL (mmol/L)				0.243	0.040	0.568
HDL (mmol/L)				0.634	0.041	0.562
Creatine (umol/L)				−0.003	−0.068	0.339
Urea nitrogen (mmol/L)				−0.059	−0.045	0.539
Uric acid (umol/L)				−0.002	−0.044	0.531
Haematin (g/L)				0.022	0.102	0.166
Albumin (g/L)				0.263	0.204	0.004	0.162	0.125	0.062*
Previous stroke				−1.743	−0.148	0.036	−1.353	−0.115	0.077*
Coronary disease				−0.959	−0.083	0.237
Chronic heart failure				−1.033	−0.071	0.321
Hypertension				−0.597	−0.070	0.326
Hyperlipidemia				−0.134	−0.011	0.875
Diabetes				−0.930	−0.103	0.149
SAS				−4.067	−0.139	0.045	−3.073	−0.105	0.098
COPD				−3.129	−0.182	0.010	−3.121	−0.181	0.005
Neoplasm				−0.010	−0.001	0.988
Arthritis				−1.039	−0.054	0.449
Parkinsonism				−3.634	−0.222	0.001	−3.004	−0.183	0.005
Chronic renal failure				−1.307	−0.096	0.183
Anemia				−0.277	−0.033	0.662
MMSE				0.437	0.273	<0.001	0.293	0.183	0.010	0.387	0.242	<0.001
Number of morbidities										−0.568	−0.188	0.007

SBP = systolic blood pressure; DBP = diastolic blood pressure; BMI = body mass index; LDL = low density lipoproteins; HDL = high density lipoproteins; SAS = sleep apnea syndrome; COPD = chronic obstructive pulmonary disease; MMSE = Mini-Mental State Examination; CM-PPT = Chinese version Mini-Physical Performance Test. Model 1: analysis of demographics; Model 2: analysis of physical health state and MMSE after adjustment for demographics; Model 3: analysis using the significant variables of Model 2 after adjustment for demographics; Model 4: analysis of the number of morbidities after adjustment for both MMSE and demographics. Bold and italic p <0.05; *p close to 0.05.

DISCUSSION

In the present study, we explored the association of a performance-based physical function assessment with demographics, cognition, and health status in a sample of non-demented and non-institutionalized veterans of the PLAAF who were older than 60 years. Our results show that age, cognitive function, and some medical conditions (such as parkinsonism and COPD) might independently affect physical function, and that previous stroke and low albumin level showed associations that were close to statistical significance. In addition, the number of morbidities was also associated with physical function, and persons with more diseases had poorer physical performance.

Our cross-sectional study show that age was negatively associated with physical function, which is consistent with the results of other studies7, 9). The longitudinal studies of Miller et al.6) and Ishizaki et al.20) also showed that age is associated with physical function decline. In addition, Ostir et al. found that 65–74-year-olds had the highest initial PPT scores and the most moderate slope of decline, followed by 75–84-year-olds, and ≥85 year-olds7). Therefore, age is the main risk factor of physical function decline. The mechanism of decline may be that muscle loss due to aging results in declines in motor ability and physical function21, 22).

Our results also showed that the cognitive function was associated with physical performance, similar to many other studies18, 23), even though we had excluded dementia patients in order to eliminate the confounder of instructions of CM-PPT being not comprehended well. This suggests that cognition is another factor of physical function decline, as well as age. Physical frailty and cognitive impairment may share a common underlying etiopathogenesis (e.g., vascular pathology, inflammation, stress)24); However, the accumulation of Alzherimer’s disease pathology in brain regions that subserve cognition may affect components of frailty through the impairment of neural systems involved in the planning and monitoring of simple movement24).

In our study, some medical conditions affected physical performance, a result which is consistent with those of previous studies1, 2, 7,8,9,10,11). Neurological diseases (such as parkinsonism and previous stroke) have an negative effect on physical function1, 2, 7, 8) since these diseases can cause muscle weakness, abnormal gait, balance impairment, and other movement disorders25, 26). A recent study showed that sustained or unsustained decline in physical performance is also common after an ischemic stroke or transient ischemic attack, even in the absence of a recurrent neurological event27). Some systemic conditions (such as COPD and low albumin level) can also contribute to poor physical function28,29,30,31). COPD is being increasingly viewed as a multiple-system disease associated with depleted lean mass that results in muscle weakness and physical fraityl28). Besides a sedentary lifestyle, the factors that lead to skeletal muscle atrophy include systemic or muscular hypoxia, systemic inflammation, and oxidative stress32, 33). A albumin is a negative acute phase protein that decreases with systemic inflammation and is conventionally viewed as a marker of chronic inflammation30, 31). Since pro-inflammatory cytokines may cause muscle atrophy and are associated with physical frailty, it is very interesting that the common mechanism for the observed association of these two medical conditions with physical function may be chronic inflammantion33).

Some other chronic medical conditions (e.g., chronic heart failure, obesity, diabetes, coronary vascular disease, impaired vision, hypertension, cancer, arthritis, depression) have also been reported in other studies to affect physical function1, 2, 7,8,9,10,11), but these were not verified in our study. The reason for this inconsistency may be that different chronic diseases play a role in the decline of physical function in different populations. Some studies have shown that chronic heart failure (CHF), but not COPD, influences physical function in American and European populations1, 6,7,8). In the past decade, it has become clear that both COPD and CHF have important systemic consequences. Cachexia and muscle weakness are two clearly visible expressions of the systemic consequences associated with both diseases33). Therefore, COPD and CHF may affect the physical function of Chinese and Western populations, respectively. Miller and Troosters found that obesity (BMI≥30 kg/m2), another condition associated with inflammation, is strongly associated with disability in Western populations1, 6), but this was not demonstrated in our study. Since the prevalence of obesity (only 4.4% in our study) is lower in China than in Western countries34), it’s no surprise that a low albumin level, but not obesity, may be a biomarker for the decline of physical function in the Chinese population. Finally, many studies have shown that diabetes is associated with physical function1, 6,7,8), but the reason for the lack of this relationship in our population is not clear.

The inconsistencies between our study results and those of other studies may also arise from chronic conditions exerting an effect on physical function proportionate to their severity11). Therefore, it is understandable that the mere presence or absence of most chronic conditions is not related to physical performance measures in our population. Moreover, lifestyle can influence the effect of chronic diseases on physical performance and function, but such associations were not confirmed in our sample due to the absence of information on physical activity and training programs33, 35, 36). In addition, the limited data of some conditions in our study (such as arthritis) reduced its power to confirm their association with physical function. Lastly, the single effect of these medical conditions on physical function was low, but the combined effect may be significant, since our results show that multi-morbidity (two or more morbidities in one individual) rather than single or zero morbidity, was associated with lower physical function. This association between the number of morbidities (multi-morbidity) and physical function has also been demonstrated in other studies1, 11).

Our study had several important strengths. First, chronic diseases were collected from computer records of case history, not self-report by the participants, so the accuracy of the information is higher. Second, CM-PPT has the characteristics of being brief and easy to implement and is well-suited to the clinical environment, so the results can be used to guide clinical practice17).

This study also had a few limitations. The sample size was not large enough, and the study design was cross-sectional, which reduces its statistical power. However, this study was a pilot one. The good reliability and validity of CM-PPT as well as its ease of implementation will ensure a wide range of surveys evaluating and monitorinf the physical function of elderly Chinese will be undertaken. A further limitation is that most participants were male, because our sample was of retired veterans, which limits the generalization of our results. However, our results still provide some important information about male subjects.

In summary, our study showed that the age, cognitive impairment, parkinsonism, COPD, and the number of morbidities might be independently associated with the physical function of elderly Chinese and that previous stroke and low albumin level also showed association that were close to statistical significance. In addition, the number of chronic diseases was also associated with the physical function. In the future, we need to study the relationship between these conditions and physical function in large samples or in a longitudinal study. Our findings also suggest that the prevention and intervention of some specific conditions or the comprehensive management of multi-morbidity might retard the decline of physical function.