Effect of frailty status on mortality risk among Chinese community-dwelling older adults: a prospective cohort study

DOI: https://doi.org/10.21203/rs.3.rs-1852817/v1

Abstract

Background: Frailty is associated with mortality in the elderly. We aimed to determine the appropriate time and frailty index (FI) threshold for frailty intervention in Chinese community-dwelling older adults.

Methods: In this prospective cohort study, we used data from the 2011 wave of the Chinese Longitudinal Healthy Longevity Study. Follow-up was performed for 7 years from baseline. Using the FI to evaluate frailty and define frailty status, we explored the best time point and FI score for frailty intervention, by comparing the relationships of FI and frailty status with mortality.

Results: From 2011 to 2018, 8642 participants were included and followed up. A total of 4458 participants died during the study period. After adjusting for variables such as age, sex, marital status, education level, and living conditions, the hazard ratio (HR) of mortality risk based on the FI at baseline was 37.484 (95% confidence interval [CI]: 30.217-46.498; P<0.001); female sex, living in the city, being married, and living with spouse were found to be protective factors, whereas ageing was a risk factor of frailty. The mortality risk was higher in pre-frail than in frail participants (HR: 3.588, 95% CI: 3.212-4.009, P<0.001). Piecewise linear regression analysis revealed an FI score threshold of 0.5. When the FI score was >0.5, the HR of mortality based on the FI was 15.758 (95% CI: 3.656-67.924; P<0.001); when the FI score was 0.5, the HR of mortality based on the FI was 48.944 (95% CI: 36.162-66.244; P<0.001).

Conclusion: The FI is a stronger predictor for mortality than the frailty status. The advancement of early interventions for mortality risk reduction is more beneficial in pre-frail than in frail patients, and an FI score of 0.5 was found to be the threshold for mortality prediction using the FI. 

Background

The rapid ageing of the global population has become a major trend in the global demographic structure owing to reductions in fertility and mortality rates [1, 2]. Frailty is becoming an increasingly obvious and common feature of the elderly with ageing; the decline of various physiological functions related to age increases, thereby increasing vulnerability to stressors. In addition to disease or disability, frailty is associated with a systemic impairment of physical and cognitive functions, including symptoms, diseases, and life-long deficits [3, 4]. People with frailty are more likely to experience a variety of negative health conditions, such as falls, fractures, hospitalization, need for nursing home placement, disability, poor quality of life, and dementia [59].

The frailty index (FI) is one of the most commonly used tools to measure frailty. The FI is evaluated based on the concept that frailty is a state caused by a life-long accumulation of health deficits; the higher the number of health deficits, the greater the tendency for frailty. These health deficits include symptoms, disease, disability, abnormal laboratory findings, and social characteristics [1012]. The FI is predictive for adverse outcomes and is directly related to survival outcomes [1315]. Moreover, compared with chronological age, the FI has a stronger correlation with mortality, especially within short intervals less than 4 years [16].

The FI has been shown to vary with time; thus, the FI evaluated using cross-sectional studies cannot accurately predict mortality risk [17, 18]. Therefore, it is necessary to perform mortality risk reassessment using dynamic FI changes [19, 20]. Moreover, frailty is not only associated with age but is also affected by risk factors such as impairment of activities of daily living, chronic diseases, depression, poor lifestyle habits, and geriatric syndromes [21, 22]. Effective prevention and treatment can reduce frailty occurrence in older adults [23]. Hence, mortality risk prediction and early intervention to treat debilitating conditions can prolong survival time, thereby alleviating the pressure on medical care [25].

We aimed to collect and evaluate longitudinal data at different time points, and to accurately determine the best time point for frailty intervention using a long follow-up duration. Our findings will potentially enhance decision-making regarding frailty intervention and the effective utilization of medical resources.

Methods

Participants

The Chinese Longitudinal Healthy Longevity Survey (CLHLS) is a nationwide longitudinal survey conducted in a randomly selected half of the counties and cities in 22 of the 31 provinces in China. All the participants provided written informed consent [24]. We used data from the 2011 wave of the CLHLS, which was followed up in 2014 and 2018. The medical ethics committee of Tongji University approved this study. Participants were excluded if more than 30% of FI variables were missing or died before the 2014 follow-up. Moreover, we excluded individuals who had 80% missing data on cognitive function and less than 30 variables for FI calculation.

Frailty Index

Health deficits were evaluated using the FI. We selected 42 items on self-related health, physical function, psychological and cognitive function, comorbidity, and social deficits [25, 26]. Cognitive function was measured using the Mini-Mental State Examination (MMSE) scale [27]. Binary variables were encoded as 0 or 1. For ordered and continuous variables, encoding was based on the distribution. A score of 2 was assigned if the respondent had suffered from more than one serious disease in the past two years. The FI score was calculated as the ratio of health deficits present to the total number of deficits considered, with values ranging between 0 and 1. Higher scores indicated a higher degree of frailty; FI scores < 0.25 and ≥ 0.25 were considered to indicate non-frailty and frailty statuses, respectively [28, 29].

Statistical analysis

Cox proportional hazards regression and piecewise linear regression[30] were used to evaluate the relationship between the FI and mortality, and the Kaplan-Meier survival function curve was used to estimate the 7-year survival in relation to the FI and frailty status. The areas under the receiver-operating characteristic (ROC) curves (AUCs) of FI and frailty status were calculated to compare the effects of these parameters on death outcomes during the follow-up period. Statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA), IBM SPSS Statistics version 20 (SPSS Inc., Chicago, IL, USA), and R statistical software version 4.2.0 (R Foundation for Statistical Computing, Vienna, Austria).

Results

A total of 8642 older people participated in the baseline survey in 2011. Table 1 shows the participant characteristics and frailty status at baseline. Participants had a median age of 85.6 ±11.3 years, with a range of 50–114 years. At baseline, 2020 (23.4%), 2802 (32.4%), and 3820 (44.2%) participants were robust (FI score ≤0.1), pre-frail (0.1<FI score<0.25), and frail (FI score ≥0.25), respectively.

 

Table 1. Participant baseline frailty characteristics

 

Robust

Pre-frailty

Frailty

P value

Age, n (%)

 

 

 

 

 

60–75

955 (11.1)

145 (1.7)

918 (10.6)

<0.001

76–85

682 (7.9)

464 (5.4)

1193 (13.8)

 

                      86-94

294 (3.4)

852 (9.9)

1074 (12.4)

 

95-114

89 (1.0)

1341 (15.5)

635 (7.3)

 

Sex, n (%)

 

 

 

 

Male

1219 (14.1)

865 (10.0)

1821 (21.1)

<0.001

Female

801 (9.3)

1937 (22.4)

1999 (23.1)

 

Residence, n (%)

 

 

 

 

City

349 (4.0)

526 (6.1)

568 (6.6)

0.001

Town

642 (7.4)

839 (9.7)

1203 (13.9)

 

Rural

1029 (11.9)

1437 (16.6)

2049 (23.7)

 

Education level, n (%)

 

 

 

 

Illiterate

748 (8.7)

2074 (24.0)

2218 (25.7)

<0.001

Primary

871 (10.1)

552 (6.4)

1209 (14.0)

 

Middle

351 (4.1)

129 (1.5)

340 (3.9)

 

Higher

48 (0.6)

37 (0.4)

48 (0.6)

 

Marital status, n (%)

 

 

 

 

Single

27 (0.3)

18 (0.2)

42 (0.5)

<0.001

Married

1234 (14.3)

552 (6.4)

1548 (18.0)

 

Divorced or widowed

753 (8.7)

2220 (25.8)

2219 (25.8)

 

Economic status, n (%)

 

 

 

 

Poor

147 (1.7)

568 (6.6)

608 (7.1)

<0.001

Rich

497 (5.8)

387 (4.5)

619 (7.2)

 

Middle

1369 (16.0)

1794 (21.0)

2572 (30.0)

 

Total           8642

2020 (23.4)

2802 (32.4)

 

3820 (44.2)

 

In addition, 4458 participants died during the study period, as observed in 2018. The AUC of FI at baseline was 0.768 (95% CI: 0.758-0.778, P<0.001), whereas the AUC of frailty status was 0.537 (95% CI: 0.524-0.549, P<0.001), thereby showing a weaker prediction with mortality (Figure 1).

The hazard ratio (HR) of mortality according to the FI at baseline was 37.484 (95% CI: 30.217-46.498), P<0.001). Female sex (HR: 0.624, 95% CI: 0.584-0.666, P<0.001), living in the city (HR: 0.864, 95% CI: 0.792-0.943, P=0.001), being married and living with spouse (HR: 0.797, 95% CI: 0.736-0.864, P<0.001) were found to be protective factors, whereas ageing (HR: 1.057, 95% CI: 1.053-1.061, P<0.001) was a risk factor for mortality (Table 2)..

 Table 2. Cox regression model analysis of the effect of the frailty index on mortality

 

B

SE

Wald

df

Sig.

Exp(B)

95% CI

Lower

Upper

Age

.056

.002

977.972

1

.000

1.057

1.053

1.061

Sex

-.472

.033

200.532

1

.000

.624

.584

.666

Residence

-.146

.045

10.754

1

.001

.864

.792

.943

Marital status

-.227

.041

30.716

1

.000

.797

.736

.864

FI_11

3.624

.110

1086.390

1

.000

37.484

30.217

46.498

Abbreviations: FI_11, frailty index in 2011; B, Regression coefficients; SE, standard error; df, degree of freedom.

We further classified frailty as non-frailty (FI <0.25) and frailty (FI ≥0.25), and analysed the HR for mortality in different states of frailty. The HR of mortality according to the FI was 2.209 (95% CI: 2.064-2.364, P<0.001) when the frailty status was dichotomized. The female sex, education level, being married, and living with spouse were found to be protective factors, whereas ageing was a risk factor of frailty. The HR for mortality was higher in pre-frail (HR: 3.588, 95% CI: 3.212-4.009, P<0.001) than in frail (HR: 1.820, 95% CI: 1.640-2.021, P<0.001) participants, when the frailty status was triaged as robust, pre-frailty, and frailty. The female sex, being married, and living with spouse were found to be protective factors, whereas ageing was a risk factor of frailty (Table 3).

 Table 3. Cox regression model analysis of the effect of frailty status on mortality

 

B

SE

Wald

Df

Sig.

Exp(B)

95% CI

Lower

Upper

Non-frailty/Frailty

Age

.061

.002

1148.558

1

.000

1.062

1.059

1.066

Sex

-.464

.036

167.825

1

.000

.628

.586

.674

Education level

-.065

.026

6.101

1

.014

.937

.890

.987

Marital status

-.202

.041

24.172

1

.000

.817

.754

.885

Frailty

.792

.035

525.386

1

.000

2.209

2.064

2.364

Robust/Pre-frailty/frailty

Age

.058

.002

1077.900

1

.000

1.060

1.056

1.064

Sex

-.449

.033

183.795

1

.000

.638

.598

.681

Marital status

-.182

.041

19.724

1

.000

.834

.769

.903

Pre-Frailty

1.278

.057

510.629

1

.000

3.588

3.212

4.009

Frailty

.599

.053

126.720

1

.000

1.820

1.640

2.021

Abbreviations: B, Regression coefficients; SE, standard error; df, degree of freedom.

 

Due to the inconsistency of the different frailty status classifications, we reconsidered the FI as a continuous variable. We found that the curves of the FI at baseline and 7-year survival rate could be divided into two segments around an FI score of 0.5 (Figure 2), where the partial regression coefficients were 3.891 and 2.757, respectively. To further explore the effect of a unit increase in FI on the mortality risk, piecewise regression analysis was performed by segment within the FI score ranges of 0-0.5 and 0.5-1. When FI score was >0.5, the HR of mortality based on FI was 15.758 (95% CI: 3.656-67.924, P<0.001); however, when the FI score was 0.5, the HR was 48.944 (95% CI: 36.162-66.244, P<0.001). The female sex, living in the city, being married, and living with spouse were found to be protective factors, whereas ageing was a risk factor of frailty (Table 4).

 

 

Table 4. Piecewise Cox regression model analysis of the effect of frailty on mortality

 

B

SE

Wald

df

Sig.

Exp(B)

95% CI

Lower

Upper

FI_11 ≤0.5

 

 

 

 

 

 

 

 

Age

.057

.002

916.001

1

.000

1.059

1.055

1.063

Sex

-.491

.035

197.162

1

.000

.612

.571

.655

Residence

-.141

.048

8.806

1

.003

.868

.791

.953

Marital status

-.197

.043

21.259

1

.000

.821

.755

.893

FI

3.891

.154

634.752

1

.000

48.944

36.162

66.244

FI_11 >0.5

 

 

 

 

 

 

 

 

Age

.028

.006

21.437

1

.000

1.028

1.016

1.041

Sex

-.242

.115

4.454

1

.035

.785

.627

.983

Marital status

-.382

.146

6.789

1

.009

.683

.512

.910

FI

2.757

.745

13.681

1

.000

15.758

3.656

67.924

Abbreviations: FI_11, frailty index in 2011; B, Regression coefficients; SE, standard error; df, degree of freedom.

Discussion

Previous studies have investigated the relationship between the FI and mortality and predicted the mortality risk based on the static and dynamic FI [20,26]. However, the relationship between the frailty status and mortality risk has not been studied [31]. To examine the relationship of the FI and frailty status with survival time, we used Kaplan-Meier survival curves to determine whether the FI was more strongly associated with mortality than the frailty status, by calculating the AUCs. Previous studies have reported a correlation between the FI and the short-term mortality;  furthermore, our findings demonstrated that the FI can be used to predict the 7-year survival rate [21]. 

Impairment in activities of daily living, chronic diseases, depression, poor lifestyle habits, and geriatric syndromes are known risk factors for frailty [32]. Similarly, our study revealed the following predictive factors of frailty: female sex, living in a city, being married, and living with a spouse. This is probably because the marital status and living conditions of older adults are related to their mental health and access to medical resources [33]. Previous research has shown a relationship between frailty and type of death; hence, we used survival analysis to evaluate the association between the FI and mortality. Our findings provide evidence that clinicians should perform frailty interventions to reduce preventable suffering before death; moreover, these interventions should be performed based on the known risk factors associated with the FI [22]. 

We further explored the relationship between the frailty status and mortality risk at baseline (2011) and during follow-up (2014 and 2018), with the aim of establishing suitable frailty interventions [29]. When examining the frailty-related mortality risk, we adjusted for demographic (gender and age), and sociological (education level, marital status, and living conditions) factors. When the frailty status was divided into non-frailty and frailty, education level was found to be a protective factor for frailty, besides the female sex, being married, and living with spouse, whereas ageing was a risk factor; this finding was probably because more education increases health literacy. Furthermore, we found that the HR for mortality was higher in pre-frail than in frail individuals, which provides evidence for the possibility of early intervention in pre-frail older adults.

Frailty, defined by phenotype or FI, was found to be significantly associated with an increased risk of all-cause mortality in community-dwelling Chinese older adults in previous studies[34,35]. Previous studies showed slightly different results of the relative mortality risk for different frailty levels owing to a lack of a unified frailty classification standard and inconsistencies in frailty status classification [36]. In the present study, we stratified the FI by grade rather than frailty categorization, to perform a more precise risk prediction, and to confirm whether 0.5 was the FI threshold. The mortality risk increased with age, and the female sex and being married were found to be protective factors of frailty, which was consistent with previous study findings [37]. Living in the city was found to be a protective factor of frailty when the FI score was less than 0.5, indicating that lifespan may be prolonged by exposure to advanced medications in the early state of frailty [38]. When the FI was greater than 0.5, the effect of frailty on mortality was relatively small because patients with the highest number of health deficits had the highest all-cause mortality rates [26]. A score of 0.5 was the risk threshold when the IF score was close to it, and the risk of death increased significantly with frailty under a score of 0.5.

Conclusions

In addition to increasing mortality risk, frailty is a predictor of negative health outcomes and all-cause mortality. The advancement of early interventions for mortality risk reduction is more beneficial in pre-frail than in frail patients.  An FI score of approximately 0.5 constitutes an adequate intervention point, which provides suggestions for clinical practice [11]. 

 

Abbreviations

AUC, area under the curve; CI, confidence interval; FI, frailty index; HR, hazard ratio; MMSE, mini-mental status examination; ROC, receiver-operating characteristic; B, Regression coefficients; SE, standard error; df, degree of freedom


Declarations

Ethics approval and consent to participate

The Medical Ethics Committee of Tongji University, Shanghai, People’s Republic of China, approved the present study (approval number: 2022tjdxsy041). Written informed consent was obtained from all participants prior to the study.

Consent for publication

Not applicable.

Availability of data and materials

The CLHLS analyzed during our study are available in the Peking University Open Research Data, [https://opendata.pku.edu.cn/].

Competing interests

The authors declare no conflicts of interest.

Funding

National Key R&D Program of China (No. 2020YFC2008703)

Authors’ contribution 

QC and RZ were involved in the research design and funding support. XZ drafted the manuscript and performed the computational analysis. JH and QC acted as corresponding authors, responsible for reviewing the content of the article. All authors read and approved the final manuscript.

Acknowledgement

We would like to thank Editage (www.editage.cn) for English language editing.

Author information

1. School of Medicine, Tongji University, Shanghai 200092, China 

2. Department of Health Statistics, Navy Medical University, Shanghai 200433, China

References

  1. Zeng Y, Feng Q, Hesketh T, Christensen K, Vaupel JW. Survival, disabilities in activities of daily living, and physical and cognitive functioning among the oldest-old in China: a cohort study. Lancet. 2017;389:1619–29.
  2. Jean-Pierre M, Matilde L, Mike M, Matthew P. WHO's report for the decade of healthy ageing 2021–30 sets the stage for globally comparable data on healthy ageing. Lancet Healthy Longev. 2021;2:E121-2.
  3. Mitnitski AB, Mogilner AJ, Rockwood K. Accumulation of deficits as a proxy measure of aging. ScientificWorldJournal. 2001;1:323–36.
  4. Bortz WM 2nd. A conceptual framework of frailty: a review. J Gerontol A Biol Sci Med Sci. 2002 May;57:M283-8. doi: 10.1093/gerona/57.5.m283. PMID: 11983721.
  5. Kojima G. Frailty as a predictor of future falls Among community-dwelling older people: A systematic review and meta-analysis. J Am Med Dir Assoc. 2015 Dec;16:1027–33. doi: 10.1016/j.jamda.2015.06.018. Epub 2015 Aug 5. PMID: 26255098.
  6. Gilbert T, Neuburger J, Kraindler J, Keeble E, Smith P, Ariti C, et al. Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study. Lancet. 2018 May 5;391:1775–82. doi: 10.1016/S0140-6736(18)30668-8. Epub 2018 Apr 26. PMID: 29706364; PMCID: PMC5946808.
  7. Kojima G. Frailty as a predictor of nursing home placement Among community-dwelling older adults: A systematic review and meta-analysis. J Geriatr Phys Ther. 2018 Jan/Mar;41:42 – 8. doi: 10.1519/JPT.0000000000000097. PMID: 27341327.
  8. Kojima G, Taniguchi Y, Iliffe S, Walters K. Frailty as a predictor of Alzheimer disease, vascular dementia, and all dementia among community-dwelling older people: a systematic review and meta-analysis. J Am Med Dir Assoc. 2016;17:881–8.
  9. Kojima G, Iliffe S, Jivraj S, Walters K. Association between frailty and quality of life among community-dwelling older people: a systematic review and meta-analysis. J Epidemiol Community Health. 2016 Jul;70:716–21. doi: 10.1136/jech-2015-206717. Epub 2016 Jan 18. PMID: 26783304.
  10. Rockwood K, Mitnitski A. Frailty in relation to the accumulation of deficits. J Gerontol A Biol Sci Med Sci. 2007;62:722–7.
  11. Hoogendijk EO, Afilalo J, Ensrud KE, Kowal P, Onder G, Fried LP. Frailty: implications for clinical practice and public health. Lancet. 2019;394:1365–75.
  12. Searle SD, Mitnitski A, Gahbauer EA, Gill TM, Rockwood K. A standard procedure for creating a frailty index. BMC Geriatr. 2008;8:24.
  13. Vermeiren S, Vella-Azzopardi R, Beckwée D, Habbig AK, Scafoglieri A, Jansen B, et al.. Frailty and the Prediction of Negative Health Outcomes: A Meta-Analysis. J Am Med Dir Assoc. 2016;17:1163.e1–1163.e17.
  14. Stenholm S, Ferrucci L, Vahtera J, Hoogendijk EO, Huisman M, Pentti J, et al. Natural course of frailty components in people who develop frailty syndrome: evidence from two cohort studies. J Gerontol A Biol Sci Med Sci. 2019;74:667–74.
  15. Rodriguez-Mañas L, Fried LP. Frailty in the clinical scenario. Lancet. 2015;385:e7-9.
  16. Kulminski A, Yashin A, Arbeev K, Akushevich I, Ukraintseva S, Land K, et al. Cumulative index of health disorders as an indicator of aging-associated processes in the elderly: results from analyses of the National Long Term Care Survey. Mech Ageing Dev. 2007 Mar;128:250–8. doi: 10.1016/j.mad.2006.12.004. Epub 2006 Dec 20. PMID: 17223183; PMCID: PMC1866299.
  17. Kojima G, Taniguchi Y, Iliffe S, Jivraj S, Walters K. Transitions between frailty states among community-dwelling older people: A systematic review and meta-analysis. Ageing Res Rev. 2019 Mar;50:81–8. doi: 10.1016/j.arr.2019.01.010. Epub 2019 Jan 16. PMID: 30659942.
  18. Ofori-Asenso R, Chin KL, Mazidi M, Zomer E, Ilomaki J, Zullo AR, et al. Global incidence of frailty and prefrailty among community-dwelling older adults: a systematic review and meta-analysis. JAMA Netw Open. 2019;2:e198398.
  19. Jenkins DA, Sperrin M, Martin GP, Peek N. Dynamic models to predict health outcomes: current status and methodological challenges. Diagn Progn Res. 2018 Dec 18;2:23. doi: 10.1186/s41512-018-0045-2. PMID: 31093570; PMCID: PMC6460710.
  20. Chen Q, Tang B, Zhai Y, Chen Y, Jin Z, Han H, et al. Dynamic statistical model for predicting the risk of death among older Chinese people, using longitudinal repeated measures of the frailty index: a prospective cohort study. Age Ageing. 2020;49:966–73.
  21. Ma L, Tang Z, Zhang L, Sun F, Li Y, Chan P. Prevalence of frailty and associated factors in the community-dwelling population of China. J Am Geriatr Soc. 2018 Mar;66:559–64. doi: 10.1111/jgs.15214. Epub 2017 Nov 23. PMID: 29168883.
  22. Clegg A, Young J, Iliffe S, Rikkert MO, Rockwood K. Frailty in elderly people. Lancet. 2013 Mar 2;381(9868):752 – 62. doi: 10.1016/S0140-6736(12)62167-9. Epub 2013 Feb 8. Erratum in: Lancet. 2013 Oct 19;382(9901):1328. PMID: 23395245; PMCID: PMC4098658.
  23. Fried LP, Ferrucci L, Darer J, Williamson JD, Anderson G. Untangling the concepts of disability, frailty, and comorbidity: implications for improved targeting and care. J Gerontol A Biol Sci Med Sci. 2004 Mar;59:255–63. doi: 10.1093/gerona/59.3.m255. PMID: 15031310.23.
  24. Gu D. General data quality assessment of the CLHLS. In: Healthy longevity in China: demographic, socioeconomic and psychological dimensions. Zeng Y, Poston DL, Vlosky DA, editors. Dordrecht: Springer; 2008.
  25. Goggins WB, Woo J, Sham A, Ho SC. Frailty index as a measure of biological age in a Chinese population. J Gerontol A Biol Sci Med Sci. 2005 Aug;60:1046-51. doi: 10.1093/gerona/60.8.1046. PMID: 16127111.
  26. Dupre ME, Gu D, Warner DF, Yi Z. Frailty and type of death among older adults in China: prospective cohort study. BMJ. 2009 Apr 9;338:b1175. doi: 10.1136/bmj.b1175. PMID: 19359289; PMCID: PMC2667569.
  27. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–98. https://doi.org/10.1016/0022-3956(75)90026-6.
  28. Fan J, Yu C, Guo Y, Bian Z, Sun Z, Yang L, et al. Frailty index and all-cause and cause-specific mortality in Chinese adults: a prospective cohort study. Lancet Public Health. 2020;5:e650-60.
  29. Bennett S, Song X, Mitnitski A, Rockwood K. A limit to frailty in very old, community-dwelling people: a secondary analysis of the Chinese longitudinal health and longevity study. Age Ageing. 2013 May;42:372-7. doi: 10.1093/ageing/afs180. Epub 2012 Dec 11. PMID: 23232936.
  30. Muggeo, V.M.R. (2003) Estimating regression models with unknown break-points. Statistics in Medicine 22, 3055–3071.
  31. Hanlon P, Nicholl BI, Jani BD, Lee D, McQueenie R, Mair FS. Frailty and pre-frailty in middle-aged and older adults and its association with multimorbidity and mortality: a prospective analysis of 493737 UK Biobank participants. Lancet Public Health. 2018 Jul;3:e323-32. doi: 10.1016/S2468-2667(18)30091-4. Epub 2018 Jun 14. PMID: 29908859; PMCID: PMC6028743.
  32. Davies K, Maharani A, Chandola T, Todd C, Pendleton N. The longitudinal relationship between loneliness, social isolation, and frailty in older adults in England: a prospective analysis. The Lancet Healthy Longev. 2021;2:e70-7.
  33. Su D, Wu XN, Zhang YX, Li HP, Wang WL, Zhang JP, et al. Depression and social support between China’ rural and urban empty-nest elderly. Arch Gerontol Geriatr. 2012;55:564–9. doi: 10.1016/j.archger.2012.06.006.
  34. Shi GP, Ma T, Zhu YS, Wang ZD, Chu XF, Wang Y, et al. Frailty phenotype, frailty index and risk of mortality in Chinese elderly population- Rugao longevity and ageing study. Arch Gerontol Geriatr. 2019 Jan-Feb;80:115–9. doi: 10.1016/j.archger.2018.11.001. Epub 2018 Nov 6. PMID: 30448694.
  35. Ekram ARMS, Woods RL, Britt C, Espinoza S, Ernst ME, Ryan J. Erratum: The Association Between Frailty and All-Cause Mortality in Community-Dwelling Older Individuals: An Umbrella Review. J Frailty AgingJ Frailty Aging. 2022;11:247.
  36. Kojima G, Iliffe S, Walters K. Frailty index as a predictor of mortality: a systematic review and meta-analysis. Age Ageing. 2018 Mar 1;47:193–200. doi: 10.1093/ageing/afx162. PMID: 29040347.
  37. Song X, Mitnitski A, Rockwood K. Prevalence and 10-year outcomes of frailty in older adults in relation to deficit accumulation. J Am Geriatr Soc. 2010;58:681–7.
  38. Zeng Y. Towards deeper research and better policy for healthy aging—using the unique data of Chinese longitudinal healthy longevity survey. China Econ J. 2012;5:131–49. https://doi.org/10.1080/17538963.2013.764677.