The Association Between Comorbidity,Frailty,and Outdoor Mobility Loss Among Community-Dwelling Individuals 60 Years of Age and Above in Moscow

Abstract

To determine the prognostic value of frailty and comorbidity for outdoor mobility loss and mortality in the elderly. The retrospective study was conducted among outpatients aged ≥60 years. Patients with ≥3 chronic illnesses were treated by doctors who had undergone a 72-hour geriatric training. The outdoor low-mobility group comprised patients who failed to visit a doctor because of decreased outdoor mobility during the 3-year follow-up period. The outdoor high-mobility group comprised participants with no outdoor mobility loss. 5678 patients with a mean age of 71.0 ± 0.1 years were included in the study. The risk of outdoor mobility loss rose by 4% per year with men developing it 30% more than women. The effect of frailty was of particular importance because it increased the risk of developing outdoor mobility loss by 70%. Comorbidity was not associated with a higher risk of outdoor mobility loss, but the investigators did not take into account all possible illnesses, or the severity of disease. The loss of outdoor mobility was associated with increase in mortality. Early detection of frailty can help predict outdoor mobility loss and could reduce mortality among older people.

Introduction

The prevalence of comorbidity tends to increase with age. The rate in older people is high, ranging from 55% to 98%.¹ Aging and chronic illnesses precipitate loss of function, in the form of a vicious circle.²

However, not only chronic diseases, but geriatric syndromes (GSs) contribute to expected outcomes in the elderly.³ GSs, which are defined as clinical conditions in the elderly that do not fit into discrete disease categories,⁴ are common.^5
–7 The most prevalent GSs include delirium, falls, urinary incontinence, dizziness, syncope, and frailty.⁴ Frailty is a multicomponent syndrome characterized by a decrease in physiological and functional reserves and higher vulnerability to adverse health factors.^8,9 The presence of this syndrome indicates an unfavorable prognosis in many cases.¹⁰ Frailty was shown to be a risk factor for falls,¹¹ loss of autonomy and disability,¹² hospitalization,¹³ and institutionalization.^13,14 Comorbidity, GSs, and frailty are widespread, and their cumulative effect on the elderly is important.^15

–19

In an earlier article⁷ we studied and described the frequency of GSs and frailty among community-dwelling older people in Moscow (Russia). The aim of this study was to determine the prognostic value of the main GSs and comorbidity for outdoor mobility loss in this population.

Methods

The retrospective study was conducted among outpatients aged ≥60 years, who participated in a pilot project named “Chronicles” from 2015 to 2018 that was conducted in a Moscow community clinic. In this project, doctors, who had undergone a 72-hour geriatric training program, treated older patients who had three or more chronic illnesses. One of the conditions for participation in the project was patient's ability to come to the clinic for an appointment with their doctor. The participation of patients who failed to see their doctor due to decreased outdoor mobility during the follow-up period was discontinued and they returned to the care of their primary care physician (who had not taken part in the geriatric training program). These patients comprised the “outdoor low-mobility” (OLM) group. Participants who continued to come to the clinic formed the “outdoor high mobility” (OHM) group.

Patients were excluded from the program if at the start of the project they suffered from an acute illness or an exacerbation of a chronic disease, if they suffered from severe dementia, if they were at the terminal stage of an oncological disease, or if they had severe vision and hearing loss (blindness and/or deafness).

At a routine examination at the beginning of the project sociodemographic data (age and gender), and data on comorbidity and smoking were collected. Smokers were defined as people who smoke at least one cigarette daily or those who had ever smoked in the past.

The “Vozrast ne pomekha” questionnaire that translates into English as “Age is no hindrance” was used to identify frailty and other GSs. The questionnaire includes seven questions related to physical, functional, and psychological aspects and weight loss, restriction in daily life due to decreased vision and/or hearing, traumatic falls, symptoms of depression, impaired memory and attention, urinary incontinence, and mobility problems. The maximum number of points was seven, one point for each affirmative answer. The minimal possible score was 0. A cut point of ≥3 points was used as it corresponded to the highest sensitivity rate at 85.7% and 93.3% in relation to the phenotype model of Fried et al.¹⁸ and the deficits accumulation model of Rockwood and colleagues,²⁰ respectively.²¹

Statistical analysis

Quantitative data are presented as mean and standard error and binary data as frequencies and percentages. The chi-square test was used to compare percentages between the two groups (OHM and OLM). Student's t-test was used to compare quantitative variables between the groups as it is applicable for abnormally distributed variables in large samples sizes. To determine the relative contribution of a particular risk factor to outdoor mobility loss and mortality, logit models (Generalized Linear model with binomial response variable and logit link function) were generated. Predictive variables included age, gender, smoking, GSs, frailty (≥3 GSs according to study questionnaire), as well as specific chronic diseases and comorbidity (≥4 illnesses). Variables were selected using forward and backward stepwise regression. The result of the logit model is presented as odds ratios (ORs) (95% confidence intervals), with p-values for each variable. Each logit model was cross-validated five times. Validation samples were generated from the patients with sequence number in the database with a certain remainder of division by five (thus, each validation sample comprised nearly 20% of the total data). The predictive power of the models was estimated by area under the ROC curve (AUC). The optimal sensitivity and specificity rates were calculated as the point nearest to the upper left corner of the receiver operating characteristic (ROC) graph (the highest sum of sensitivity and specificity). The statistical analysis was performed using the Statistica 10 program.

The Helsinki Committee of the National Research Center for Preventive Medicine, Moscow, Russia, approved the study and enabled the investigators to obtain oral consent from the participants (approval N09-06/14).

Results

Sociodemographic characteristics and comorbidity

The total study population included 5678 patients. The mean age was 71.0 ± 0.1 years, and women accounted for 72.3%. Arterial hypertension was the most common disease among patients with comorbidity (99.2%). The OLM group comprised 1332 patients (23.5%) who lost outdoor mobility. The OHM group comprised 4346 patients (76.5%) who maintained outdoor mobility (76.5%).

The OLM group was characterized as having older patients, a lower percentage of women, more smokers, and a higher total number of illnesses with a higher prevalence of myocardial infarction and chronic obstructive pulmonary disease (COPD) (Table 1).

Table 1.

Sociodemographic Characteristics and Comorbidity of Patients in the Two Study Groups (Outdoor High Mobility and Outdoor Low Mobility)

Characteristics	OHM group	OLM group	p
Characteristics	N = 4346 (76.5%)	N = 1332 (23.5%)	p
Age, mean ± SD	70.6 ± 0.10	72.5 ± 0.21	<0.001
Gender, female, n (%)	3200 (74)	908 (68)	<0.001
Active or ever smoker, n (%)	333 (8)	143 (11)	0.003
Comorbidity, mean ± SD	3.3 ± 0.01	3.4 ± 0.02	0.01
Arterial hypertension, n (%)	4313 (99.2)	1323 (99.3)	0.75
Myocardial infarction, n (%)	246 (5.7)	97 (7.3)	0.03
Diabetes mellitus, type 2, n (%)	1439 (33.1)	451 (33.9)	0.62
Asthma, n (%)	357 (8.2)	112 (8.4)	0.82
COPD, n (%)	177 (4.1)	73 (5.5)	0.028

Active or ever smoker: people who smoked at least one cigarette daily or ever smoked in the past; comorbidity: sum of the chronic diseases.

COPD, chronic obstructive pulmonary disease; OLM, outdoor low mobility; OHM, outdoor high mobility; SD, standard deviation.

GSs in the OHM group and in the OLM group at the start of the project

Compared with patients in the OHM group, patients from the OLM group, at the beginning of the project, complained more often about weight loss, falls, mobility problems, depressed mood, memory and attention problems, based on their responses to the study questionnaire (Table 2). The number of frail patients in the OLM group was almost two times higher than in the OHM group (42% against 26%, respectively; p < 0.0001).

Table 2.

The Prevalence of Geriatric Syndromes According to the “Age Is No Hindrance” (“Vozrast Ne Pomekha”) Questionnaire in the Two Study Groups (Outdoor High Mobility and Outdoor Low Mobility)

Syndrome	OHM group	OLM group	p
Weight loss	414 (12)	160 (16)	0.00011
Hearing and vision loss	1352 (38)	389 (40)	0.41
Falls	645 (18)	228 (23)	0.00033
Depressed mood	1564 (45)	489 (50)	0.0014
Attention and memory problems	1686 (48)	507 (52)	0.022
Urinary incontinence	499 (14)	162 (17)	0.066
Mobility problems	948 (27)	309 (32)	0.003
Frailty	1101 (26)	416 (42)	<0.0001

Values are presented as N (%). Frailty: ≥3 points according to the “Age is no hindrance” (“Vozrast ne pomekha”) questionnaire.

OHM, outdoor high mobility; OLM, outdoor low-mobility.

A logit model was generated to identify effects of GSs, comorbidity, smoking, gender, and age on risks of outdoor mobility loss (Table 3).

Table 3.

The Result of a Logit Model on Factors That Affected the Development of Outdoor Mobility Loss Among All Patients

Risk factor	OR	95% CI (OR)		p
Risk factor	OR	Lower boundary	Upper boundary	p
Frailty	1.707	1.447	2.013	0.001
Gender (male)	1.307	1.094	1.561	0.003
Active or ever smoker	1.282	1.064	1.546	0.009
Hearing and vision loss	0.802	0.695	0.969	0.001
Age	1.040	1.029	1.052	0.001

Frailty: ≥3 points according to the “Age is no hindrance” (“Vozrast ne pomekha”) questionnaire. Active or ever smoker: people who smoked at least one cigarette daily or ever smoked in the past.

CI, confidence interval; OR, odds ratio.

The best model, generated by forward and backward stepwise algorithms (both yielded the same results) included only four factors. The predictive power of the model was small, but it almost equaled that of the full model (AUC = 0.61 and 0.62, respectively), so only the best model was analyzed. The predictive power of the model on validation samples was almost the same (AUC range 0.59–0.64 on different samples). The optimal sensitivity and specificity rates, assessed from the ROC curves, were 65% and 52%, respectively.

According to the model, the risk of transition to the OLM group rose by 4% per year of age (p = 0.001), and men were more vulnerable, with an increased risk for developing outdoor mobility loss of 30% (p = 0.003). The risk for people who were active or ever smokers to develop outdoor mobility loss and enter the OLM group was 28% higher (p = 0.009) than among never-smokers. None of the chronic diseases registered at the beginning of the “Chronicles” project, as well as sum score of comorbidity, increased the risk of transitioning to the OLM group, due to mobility loss over the 3 years of follow-up. At the same time, the effect of frailty was of particular importance because it increased the risk of transitioning to the OLM group by 70% (p = 0.001). In contrast, it is noteworthy that complaints of reduced vision and hearing decreased the risk of transitioning to the OLM group by 20% (Table 3).

Mortality

During the 3-year follow-up period 458 patients died: 423 patients from the OLM group (31.8%) and 35 patients from the OHM group (0.008%). The best model for the prediction of mortality among all 5678 patients included the following risk factors: age (OR 1.07, 95% confidence interval [CI] 1.05–1.08, p = 0.001), male gender (OR 2.12, 95% CI 1.62–2.77, p = 0.001), mobility problems (according to the study questionnaire) (OR 1.85, 95% CI 1.45–2.35, p = 0.001), diabetes mellitus (DM) (OR 1.46, 95% CI 1.15–1.86, p = 0.002), and smoking (OR 1.44, 95% CI 1.09–1.92, p = 0.011).

More participants in the OLM group who died during the follow-up period were smokers (33% vs. 25%, p = 0.003), men (42% vs. 27%, p = 0.0001), had mobility problems (according to the study questionnaire) (40% vs. 28%, p = 0.001), COPD (8% vs. 4%, p = 0.01), and DM (38% vs. 32%, p = 0.03) (Table 4). In the OHM group, mobility problems were significantly more common among those participants who died (94% vs. 27%, p = 0.001) or suffered from COPD (17% vs. 4%, p = 0.001) (Table 5).

Table 4.

The Prevalence of Geriatric Syndromes According to the “Age Is No Hindrance” (“Vozrast Ne Pomekha”) Questionnaire and Comorbidity in the Outdoor Low Mobility Group Among Patients Alive at the End of the Follow-Up and Those Who Died

	Alive (N = 909)^a	Dead (N = 423)^a	p
Weight loss	103 (16)	57 (18)	0.259
Hearing and vision loss	252 (38)	136 (44)	0.078
Falls	165 (25)	63 (20)	0.111
Depressed mood	342 (52)	146 (47)	0.164
Attention and memory failure	337 (51)	170 (55)	0.262
Urinary incontinence	113 (17)	48 (15)	0.529
Mobility problems	186 (28)	123 (40)	0.001
Frailty	282 (42)	134 (42)	0.84
Active or ever smoker	216 (25)	132 (33)	0.003
Arterial hypertension	903 (99)	420 (99)	0.91
Myocardial infarction	59 (6)	38 (9)	0.103
Diabetes mellitus, type 2	290 (32)	161 (38)	0.03
Asthma	76 (8)	36 (9)	0.93
COPD	40 (4)	33 (8)	0.01
Comorbidity 4+	297 (33)	161 (38)	0.054

Values are presented as N (%). Comorbidity: the sum of the chronic diseases. Frailty: ≥3 points according to the “Age is no hindrance” (“Vozrast ne pomekha”) questionnaire. Active or ever smoker: people who smoked at least one cigarette daily or ever smoked in the past

Missing data (0–21%).

Table 5.

The Prevalence of Geriatric Syndromes According to the “Age Is No Hindrance” (“Vozrast Ne Pomekha”) Questionnaire and Comorbidity in the Outdoor High Mobility Group Among Patients Alive at the End of the Follow-Up and Those Who Died

	Alive (N = 4280)^a	Dead (N = 35)^a	p
Weight loss	413 (12)	1 (6)	0.411
Hearing and vision loss	1345 (38)	7 (41)	0.411
Falls	641 (18)	2 (12)	0.480
Depressed mood	1558 (45)	6 (35)	0.440
Attention and memory failure	1679 (48)	7 (41)	0.570
Urinary incontinence	495 (14)	4 (24)	0.270
Mobility problems	932 (27)	16 (94)	0.001
Frailty	1093 (26)	8 (24)	0.740
Active or ever smoker	918 (22)	9 (26)	0.593
Arterial hypertension	4278 (99)	35 (100)	0.603
Myocardial infarction	246 (6)	0 (0)	0.146
Diabetes mellitus, type 2	1430 (33)	9 (26)	0.349
Asthma	354 (8)	3 (9)	0.94
COPD	171 (4)	6 (17)	0.001
Comorbidity 4+	1471 (32)	12 (34)	0.755

Missing data (0–21).

The predictive power of the model was small, but it almost equaled that of the full model (AUC = 0.70; the only differences were at the third decimal place), so only the best model was analyzed. The predictive power of the model on validation samples was almost the same (AUC range 0.66–0.72 on different samples). The optimal sensitivity and specificity rates, assessed from the ROC curve, were 74% and 57%, respectively.

Discussion

This study has several principal findings. First, patients with mobility problems at the start of the study were more likely to become home-ridden for the 3-year follow-up. Second, this development was associated with a significant increase in mortality (31.8% vs. 0.008%). It is important to note that comorbidity was not associated with becoming home-ridden but was associated with increased mortality. DM was associated with increased mortality in the OLM group and in the total study population, whereas COPD was associated with increased mortality in both study groups.

Outdoor mobility loss

Additional risk factors for loss of outdoor mobility were age, male gender, smoking, and frailty. The association between aging and the development of mobility problems has been investigated in many prior studies.^22

–28 In contrast with the findings in some,^23,27,28 but not all the studies, we found that men are more likely than women to develop mobility loss. The results of these studies were also heterogeneous in terms of the association between smoking and mobility loss. An association between frailty and mobility problems has also been reported previously.^29,30

There are possible explanations for the finding that comorbidity did not influence outdoor mobility loss. Owing to the retrospective design of the study we did not have data on all the chronic illnesses since the number of illnesses included in the study questionnaire was not exhaustive. Comorbidity was calculated as the numerical sum of the patient's illnesses and the severity of the illnesses was not assessed.

Frailty and comorbidity, although different, are closely related,³¹ as they both reflect the effects of biomolecular damage accumulated over a lifetime.³²

In our study, we observed an inverse relation between patients' complaints of visual and hearing impairment and outdoor mobility loss. According to previously published studies, vision impairment is associated with a range of pathological conditions, including falls, decreased quality of life, and functional loss.^33,34 However, older people often refuse to seek additional support.^35,36 It is possible that complaints indicating concern about a visual or hearing deficit reflect a more active lifestyle, which protects from outdoor mobility loss. Of course, we had no way to verify the validity of this hypothesis in a retrospective study.

Mortality

It is known that free and safe movement is crucial not only for independent living but also for quality of life, and that loss of autonomy in the elderly is associated with mortality.³⁷ The results of our study show that mortality in the OLM group was higher than in the OHM group. Other significant risk factors for mortality included age, male gender, and smoking.

Many previous studies have shown an association between increased mortality and age,^{38

–46} male gender,^{24,38,39,41,43
–45} and smoking.^{39,41
–43,45,47} Associations between mortality and both DM^40,45,48 and COPD⁴⁶ have also been reported in the past.

The association between mobility and mortality has been well documented in previous studies.^{24,39
–41,43,44,46,48

–54} The reason for this association appears to be the well-recognized relationship between frailty, disability, and comorbidity.^15,17,31 In this respect the findings reported by Saajanaho et al.⁵⁵ that older women who developed mobility loss over time reduced their personal goals not only in terms of leisure time activity, but also in terms of physical activity, health and daily function, are also important.

Our logit models were cross-validated and programmed to reveal the most significant factors affecting risk of loss of mobility and death. Nevertheless, they have low predictive power, as evidenced by the ORs at close to one, and low AUCs. This result should not come as a surprise since the probability of these events depends on many other factors, which were not all taken into account. It is noteworthy that mortality seems to be more predictable than loss of mobility (AUC = 0.70 and 0.61, respectively).

This study has several strengths. It is the first study from Russia to show that early detection of frailty and other GSs can help predict outdoor mobility loss and mortality among older people. Second, it included data on >5500 patients. Third, the follow-up period was 3 years.

However, it also has many limitations. One of the weaknesses is that data on frailty were collected by questionnaire without a comprehensive geriatric assessment to assess frailty. The composition of the database was not comprehensive. As discussed earlier, we did not take all possible illnesses into account. Because of the retrospective study design, we could only calculate comorbidity as the total number of the patient's illnesses without taking their severity into account.

The data that were collected over the course of the study did not enable us to identify exactly what caused the loss of outdoor mobility and, thus, inclusion in the LM group. It could have been the development of GSs such as impaired mobility, the development of dementia or depression, or the development of diseases that could have caused the patients to be home-ridden such as stroke or heart failure. The lack of data prevented us from conducting a more profound analysis and is definitely a significant limitation to this study.

Other important data were also not collected, including sociodemographic information on life habits as well as many other health indices. Examples of these missing variables are the association between income, education, and mobility loss,²⁵ or associations between family status,^41,42 and social support⁴⁸ with mortality. Associations have been reported between alcohol use and mobility loss^47,56 and alcohol use and mortality.^{39,41,43,45,47} Another significant limitation of this study is the absence of important factors linked to frailty syndrome, disability and mobility loss, such as sarcopenia, falls, cognitive status, body mass index (BMI), and previous osteoporotic fractures. For example, the lack of data on BMI: associations have been reported between BMI and mobility loss^24,26,56 as well as mortality.

Another significant limitation is the lack of data on the rate of development of outdoor mobility loss. Did the problem develop slowly or with a sudden onset? Guralnik et al. found that progressive and catastrophic walking disability had a clearly different impact on survival.²⁴ We also have no data on causes of death.

Another limitation of this study is the absence of data on the time when the patients lost outdoor mobility and the time of death. Because of this lack of information we were not able to conduct a survival analysis.

As described earlier, the patients in the OLM group were treated by physicians who did not undergo the 72-hour geriatric training, whereas the patients in the HM group were treated by physicians who did. A weakness of the study is that it is impossible to assess the effect of this on mortality and we have no data on differences in the treatment received by patients in the two groups. In any event, the increase in mortality among patients who were not treated by physicians with geriatric training, should motivate health care leaders to reconsider means of assistance for patients with limited mobility.

In conclusion, the results of this study reiterate the importance of frailty as a risk factor of mobility loss and the importance of mobility loss, in turn, on mortality. A large-scale prospective study is needed to confirm these findings.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received.

References

Marengoni

, Angleman

, Melis

, et al. Aging with multimorbidity: A systematic review of the literature. Ageing Res Rev, 2011; 10:430–439.

Yokota

, Van der Heyden

, Nusselder

, et al. Impact of chronic conditions and multimorbidity on the disability burden in the older population in Belgium. J Gerontol A Biol Sci Med Sci, 2016; 71:903–909.

Koroukian

, Warner

, Owusu

, Given

. Multimorbidity redefined: Prospective health outcomes and the cumulative effect of co-occurring conditions. Prev Chronic Dis, 2015; 12:E55.

Inouye

, Studenski

, Tinetti

, Kuchel

. Geriatric syndromes: Clinical, research, and policy implications of a core geriatric concept. J Am Geriatr Soc, 2007; 55:780–791.

Ates Bulut

, Soysal

, Isik

. Frequency and coincidence of geriatric syndromes according to age groups: Single-center experience in Turkey between 2013 and 2017. Clin Interv Aging, 2018; 13:1899–1905.

Chen

, Hwang

, Liu

, Lee

, Peng

. Frailty is a geriatric syndrome characterized by multiple impairments: A comprehensive approach is needed. J Frailty Aging, 2016; 5:208–213.

Tkacheva

, Runikhina

, Ostapenko

, et al. Prevalence of geriatric syndromes among people aged 65 years and older at four community clinics in Moscow. Clin Interv Aging, 2018; 13:251–259.

Clegg

, Young

, Iliffe

, Rikkert

, Rockwood

. Frailty in elderly people. Lancet, 2013; 381:752–762.

Morley

, Vellas

, van Kan

, et al. Frailty consensus: A call to action. J Am Med Dir Assoc, 2013; 14:392–397.

10.

Gill

, Gahbauer

, Allore

, Han

. Transitions between frailty states among community-living older persons. Arch Intern Med, 2006; 166:418–423.

11.

Kojima

. Frailty as a predictor of future falls among community-dwelling older people: A systematic review and meta-analysis. J Am Med Dir Assoc, 2015; 16:1027–1033.

12.

Kojima

. Frailty as a predictor of disabilities among community-dwelling older people: A systematic review and meta-analysis. Disabil Rehabil, 2017; 39:1897–1908.

13.

Wang

, Shamliyan

, Talley

, Ramakrishnan

, Kane

. Not just specific diseases: Systematic review of the association of geriatric syndromes with hospitalization or nursing home admission. Arch Gerontol Geriatr, 2013; 57:16–26.

14.

Kojima

. Frailty as a predictor of nursing home placement among community-dwelling older adults: A systematic review and meta-analysis. J Geriatr Phys Ther, 2018; 41:42–48.

15.

Boeckxstaens

, Vaes

, Legrand

, Dalleur

, De Sutter

, Degryse

. The relationship of multimorbidity with disability and frailty in the oldest patients: A cross-sectional analysis of three measures of multimorbidity in the BELFRAIL cohort. Eur J Gen Pract, 2015; 21:39–44.

16.

Cesari

, Perez-Zepeda

, Marzetti

. Frailty and multimorbidity: Different ways of thinking about geriatrics. J Am Med Dir Assoc, 2017; 18:361–364.

17.

Fried

, Ferrucci

, Darer

, Williamson

, Anderson

. Untangling the concepts of disability, frailty, and comorbidity: Implications for improved targeting and care. J Gerontol A Biol Sci Med Sci, 2004; 59:255–263.

18.

Fried

, Tangen

, Walston

, et al. Frailty in older adults: Evidence for a phenotype. J Gerontol A Biol Sci Med Sci, 2001; 56:M146–M156.

19.

Weiss

. Frailty and chronic diseases in older adults. Clin Geriatr Med, 2011; 27:39–52.

20.

Mitnitski

, Mogilner

, Rockwood

. Accumulation of deficits as a proxy measure of aging. ScientificWorldJournal, 2001; 1:323–336.

21.

Tkacheva

, Runikhina

, Ostapenko

, et al. Validation of the questionnaire for screening frailty. Adv Gerontol, 2017; 30:236–242.

22.

Ayis

, Gooberman-Hill

, Bowling

, Ebrahim

. Predicting catastrophic decline in mobility among older people. Age Ageing, 2006; 35:382–387.

23.

Deshpande

, Metter

, Guralnik

, Bandinelli

, Ferrucci

. Sensorimotor and psychosocial determinants of 3-year incident mobility disability in middle-aged and older adults. Age Ageing, 2014; 43:64–69.

24.

Guralnik

, Ferrucci

, Balfour

, Volpato

, Di Iorio

. Progressive versus catastrophic loss of the ability to walk: Implications for the prevention of mobility loss. J Am Geriatr Soc, 2001; 49:1463–1470.

25.

, Woo

, Yuen

, Sham

, Chan

. Predictors of mobility decline: The Hong Kong old-old study. J Gerontol A Biol Sci Med Sci, 1997; 52:M356–M362.

26.

Idland

, Rydwik

, Smastuen

, Bergland

. Predictors of mobility in community-dwelling women aged 85 and older. Disabil Rehabil, 2013; 35:881–887.

27.

Inzitari

, Carlo

, Baldereschi

, et al. Risk and predictors of motor-performance decline in a normally functioning population-based sample of elderly subjects: The Italian longitudinal study on aging. J Am Geriatr Soc, 2006; 54:318–324.

28.

Leng

, Wang

. Long term determinants of functional decline of mobility: An 11-year follow-up of 5464 adults of late middle aged and elderly. Arch Gerontol Geriatr, 2013; 57:215–220.

29.

Davis

, Rockwood

, Mitnitski

, Rockwood

. Impairments in mobility and balance in relation to frailty. Arch Gerontol Geriatr, 2011; 53:79–83.

30.

Montero-Odasso

, Muir

, Hall

, et al. Gait variability is associated with frailty in community-dwelling older adults. J Gerontol A Biol Sci Med Sci, 2011; 66:568–576.

31.

Theou

, Rockwood

, Mitnitski

, Rockwood

. Disability and co-morbidity in relation to frailty: How much do they overlap?. Arch Gerontol Geriatr, 2012; 55:e1–e8.

32.

Kirkwood

. Understanding the odd science of aging. Cell, 2005; 120:437–447.

33.

Smeeth

, Iliffe

. Effectiveness of screening older people for impaired vision in community setting: Systematic review of evidence from randomised controlled trials. BMJ, 1998; 316:660–663.

34.

, Liljas

AEM

. The relationship between self-reported sensory impairments and psychosocial health in older adults: A 4-year follow-up study using the English Longitudinal Study of Ageing. Public Health, 2019; 169:140–148.

35.

Fabacher

, Josephson

, Pietruszka

, Linderborn

, Morley

, Rubenstein

. An in-home preventive assessment program for independent older adults: A randomized controlled trial. J Am Geriatr Soc, 1994; 42:630–638.

36.

Wormald

, Wright

, Courtney

, Beaumont

, Haines

. Visual problems in the elderly population and implications for services. BMJ, 1992; 304:1226–1229.

37.

Guralnik

, Fried

, Salive

. Disability as a public health outcome in the aging population. Annu Rev Public Health, 1996; 17:25–46.

38.

Brock

, Lemke

, Branch

, Evans

, Berkman

. Mortality and physical functioning in epidemiologic studies of three older populations. J Aging Soc Policy, 1994; 6:21–37.

39.

Brown

, Harhay

. Walking cadence and mortality among community-dwelling older adults. J Gen Intern Med, 2014; 29:1263–1269.

40.

Cesari

, Onder

, Zamboni

, et al. Physical function and self-rated health status as predictors of mortality: Results from longitudinal analysis in the ilSIRENTE study. BMC Geriatr, 2008; 8:34.

41.

Eekhoff

EMW

, van Schoor

, Biedermann

, et al. Relative importance of four functional measures as predictors of 15-year mortality in the older Dutch population. BMC Geriatr, 2019; 19:92.

42.

Ford

, Spallek

, Dobson

. Self-rated health and a healthy lifestyle are the most important predictors of survival in elderly women. Age Ageing, 2008; 37:194–200.

43.

Keevil

, Luben

, Hayat

, Sayer

, Wareham

, Khaw

. Physical capability predicts mortality in late mid-life as well as in old age: Findings from a large British cohort study. Arch Gerontol Geriatr, 2018; 74:77–82.

44.

Minneci

, Mello

, Mossello

, et al. Comparative study of four physical performance measures as predictors of death, incident disability, and falls in unselected older persons: The insufficienza Cardiaca negli Anziani Residenti a Dicomano Study. J Am Geriatr Soc, 2015; 63:136–141.

45.

Noale

, Minicuci

, Bardage

, et al. Predictors of mortality: An international comparison of socio-demographic and health characteristics from six longitudinal studies on aging: The CLESA project. Exp Gerontol, 2005; 40:89–99.

46.

Veronese

, Stubbs

, Fontana

, et al. A comparison of objective physical performance tests and future mortality in the elderly people. J Gerontol A Biol Sci Med Sci, 2017; 72:362–368.

47.

Nelson

, Lui

, Ensrud

, Cummings

, Cauley

, Hillier

. Associations of smoking, moderate alcohol use, and function: A 20-year cohort study of older women. Gerontol Geriatr Med, 2018; 4:2333721418766127.

48.

Blain

, Carriere

, Sourial

, et al. Balance and walking speed predict subsequent 8-year mortality independently of current and intermediate events in well-functioning women aged 75 years and older. J Nutr Health Aging, 2010; 14:595–600.

49.

Hardy

, Kang

, Studenski

, Degenholtz

. Ability to walk 1/4 mile predicts subsequent disability, mortality, and health care costs. J Gen Intern Med, 2011; 26:130–135.

50.

Mutikainen

, Rantanen

, Alen

, et al. Walking ability and all-cause mortality in older women. Int J Sports Med, 2011; 32:216–222.

51.

Newman

, Simonsick

, Naydeck

, et al. Association of long-distance corridor walk performance with mortality, cardiovascular disease, mobility limitation, and disability. JAMA, 2006; 295:2018–2026.

52.

Nofuji

, Shinkai

, Taniguchi

, et al. Associations of walking speed, grip strength, and standing balance with total and cause-specific mortality in a general population of Japanese elders. J Am Med Dir Assoc, 2016; 17:184.e181–e187.

53.

Rolland

, Lauwers-Cances

, Cesari

, Vellas

, Pahor

, Grandjean

. Physical performance measures as predictors of mortality in a cohort of community-dwelling older French women. Eur J Epidemiol, 2006; 21:113–122.

54.

Yates

, Zaccardi

, Dhalwani

, et al. Association of walking pace and handgrip strength with all-cause, cardiovascular, and cancer mortality: A UK Biobank observational study. Eur Heart J, 2017; 38:3232–3240.

55.

Saajanaho

, Viljanen

, Read

, et al. Mobility limitation and changes in personal goals among older women. J Gerontol B Psychol Sci Soc Sci, 2016; 71:1–10.

56.

Wannamethee

, Ebrahim

, Papacosta

, Shaper

. From a postal questionnaire of older men, healthy lifestyle factors reduced the onset of and may have increased recovery from mobility limitation. J Clin Epidemiol, 2005; 58:831–840.