Article Text
PDF
Abstract
Introduction The rapidly expanding population of ageing and older adults with disability has been a continuing public health priority in recent decades. The first step towards solving this issue is to assess disability accurately and identify high-risk factors and individuals for early prevention. We aim to establish a prospective cohort, the Beijing Longitudinal Disability Survey in Community Elderly (BLINDSCE), using multidimensional disability assessments and to develop multifactorial models for disability prediction among community-dwelling older adults.
Methods and analysis The BLINDSCE is a prospective cohort study that includes community-dwelling older adults aged ≥65 years with or without disability from urban and rural areas in Beijing. Participants complete structured questionnaires and undergo assessments of disability, cognition and disability-related factors and outcomes. Disability is assessed using the WHO’s Disability Assessment Schedule 2.0, activities of daily living, the Barthel index, locomotor function and physical function. Based on baseline cross-sectional information, the relationships between multiple factors and disability can be initially screened using logistic regression. Every 1–1.5 years, participants will receive a follow-up survey to remeasure disability, cognitive function and other disability-related factors and outcomes. At least three follow-ups are scheduled. The primary outcome of this study is disability. The secondary outcomes include cognition and many disability-related conditions, such as falls, pain, poor health, decreased intrinsic capacity, frailty, sarcopenia, hospitalisation and death. Cox proportional hazards or logistic regression will be used to analyse follow-up data and construct prediction models, which will be validated internally and externally.
Ethics and dissemination The Ethics Committee of Xuanwu Hospital, Capital Medical University, approved this study (No: [2023]129). The results will be published in peer-reviewed journals focusing on geriatric medicine and presented at related scientific conferences.
Trials registration number NCT06863727. Stage of study: recruiting.
- Disabled Persons
- Aging
- Cognition
- Risk Factors
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
The BLINDSCE (Beijing Longitudinal Disability Survey in Community Elderly) study includes highly representative community-based older adults from urban and rural areas in Beijing.
Multidimensional disability assessments include the WHO Disability Assessment Schedule 2.0, activities of daily living, the Barthel index, the short-physical performance battery, grip strength and geriatric conditions, including mobility, cognition, depression, pain, falls, nutrition, frailty, intrinsic capacity and sarcopenia.
The comprehensive risk factors are collected, including individual and social determinants, physical, clinical and laboratory indices, provide a better understanding of the cause and development of disability.
Although the survey sites include different communities and nursing homes, this single-centre study in Beijing considers only older adults aged ≥65 years, which may limit the extrapolation.
Introduction
Background and rationale
Population ageing has become a global public health challenge due to declining birth rates and increasing life expectancy. The global number of adults aged 65 and older has exceeded 700 million, and its proportion is expected to increase to 16.5% by 2050.1 Older adults often experience long-term dependency and reduced quality of life due to disability and functional decline. According to the WHO International Classification of Functioning, Disability and Health (ICF), disability is an umbrella term covering impairments, activity limitations and participation restrictions,2 which is a major health issue associated with hospitalisation and mortality.3 While factors related to disability are profound, it is challenging to predict disability in substantial community-dwelling older adults. Therefore, it is essential to assess disability accurately in the community, identify powerful predictors and develop simple and feasible prediction models.
Researches on risk factors for disability have consistently revealed that disability is a multifactorial problem and is influenced by vascular and metabolic factors, psychosocial status and lifestyle.4–7 However, these previously identified risk factors were generally captured from retrospective or cross-sectional studies and investigated separately, which may not lead to reliable evidence of causation between risk factors and disability. Moreover, prior prediction models for disability in community-dwelling older adults are mainly based on pooled data from published cohorts instead of prospective cohorts purposely designed for disability prediction.8–10 The use of existing data from different cohorts inevitably has several limitations, including inconsistency in assessment tools for disability and functions across different cohorts. A cohort study conducted by Den11 focused on a predictive model for the development of disability in activities of daily living (ADL) and identified the number of chronic diseases, muscle strength, age, sex and socioeconomic status as predictors, but with a relatively small sample size and a narrow set of involved factors. Hence, longitudinal cohort studies on disability prediction models based on comprehensive and accurate disability assessments and multifactorial considerations are lacking.
The ICF model denotes the negative features of the interaction between an individual (with a health condition) and their contextual impacts (environmental and personal factors) on disability,2 encompassing a multidimensional assessment from the individual to the environmental level rather than limiting the assessment to functional dependency and disease. Therefore, the WHO Disability Assessment Schedule (WHODAS) 2.0 was developed as a new standardised tool to measure disability across cultures in multiple life domains, including mobility, life activities and participation.12 The WHODAS 2.0 may help monitor general disability and health status in the enormous community-based population, as it is not limited to a single disease or functional dimension.12 Although several studies have applied the WHODAS 2.0 to measure disability following critical illness,13 trauma14 and neurological disease,15 studies that use the WHODAS 2.0 for disability evaluation in community-dwelling older adults are scarce.
Objectives
The aim is to establish a prospective cohort, the Beijing Longitudinal Disability Survey in Community Elderly (BLINDSCE), to assess disability thoroughly with multidimensional assessments, construct a multifactorial prediction model and identify the developmental trajectory of disability among community-dwelling older adults. The main purpose of this cohort study is to identify older adults at high risk for disability and direct pertinent preventive interventions to ultimately reduce the occurrence and delay the progression of disability.
Methods and analysis
Study design and sites
The BLINDSCE cohort is a longitudinal prospective cohort study in a representative sample of community-dwelling older adults residing in Beijing, China. The major study population is being recruited from Qingta Community Healthcare Centre, an urban primary healthcare facility located in Qingta Street, Fengtai District and Mafang Community Healthcare Centre, which provides primary health services for rural residents located in Mafang Town, Pinggu District, an exurban area of Beijing. Older adults with disability are also being recruited to preanalyse possible risk factors at baseline and to conduct trajectory analysis of disability progression. Participants with disabilities are being recruited mainly from nursing homes, gathering places for individuals with disabilities, which cover public, private and public‒private partnerships, as well as from community healthcare centres. The overall framework of our proposed project is shown in figure 1.
Flowchart of the BLINDSCE cohort.
Sample criteria
The study eligibility criteria are community-dwelling adults aged 65 years or older who reside in Beijing. Participants (or proxy) should sign an institutionally approved informed consent. The exclusion criteria are as follows: severe mental disorders; serious medical conditions preventing study investigation; and long-term professional treatment for physical function and cognitive functioning, such as hospitalisation or rehabilitation.
Sample size calculations
The required sample size is at least 1429 participants without disability at baseline. The sample size for clinical prediction modelling is typically determined and justified by the rule of thumb of 10 outcome events per variable.16 Given that 10 candidate predictors are considered, the expected number of events for the study should be 100. According to the China Health and Retirement Longitudinal Study, the disability incidence during a 2-year follow-up period among Chinese people aged ≥65 years ranged from 12.7% to 19.0%,17 18 so an incidence of events of 12% was used to ensure adequate power. The recommended sample size was estimated as follows: n=10×10/12%×1.2=1000, considering a 20% maximal loss of follow-up as acceptable.19 The validation data set should include up to 429 individuals because it usually requires 3/7 of the number of individuals in the modelling data set. Together, the 1000-person modelling data set and the 429-person validation data set result in the sample size of 1429. If possible, a larger number of samples will be included to increase the accuracy of the prediction model.
The required sample size is at least 460 participants with disability at baseline. To initially screen for disability risk factors, we include participants with disability at baseline, and a generalised multifactorial analysis will be performed. The sample size for generalised multifactorial analysis is recommended using Kendall theory,20 which means that the sample size should be 10–20 times the number of independent variables. We envisage including up to 40 variables; therefore, the sample size was estimated as follows: n=40×10×1.15=460, allowing for a potential non-response rate of 15%.21
Study enrolment and baseline assessment
Different recruitment approaches are used mainly according to the urban‒rural differences in lifestyles and typical means of receiving notifications. Villages are randomly selected in Mafang Town, and the community healthcare staff are helping recruit older residents mainly through village-notification broadcasting, supplemented with door-to-door or telephone-based recruitment. Interested participants are being asked to come to the investigation site to provide written consent and complete the survey performed by the research team. At the Qingta Street site, recruitment information is first disseminated online or posted at the community healthcare centre by community healthcare staff. Then, active telephone or message-based recruitment and survey appointments for interested residents are performed by the research team. Participants are being asked to visit the Qingta Community Healthcare Centre to provide written consent and participate in face-to-face surveys. In nursing homes, in-person recruitment, consent and surveys are being conducted by the research team.
Data collection is being conducted by trained research staff (table 1). First, baseline characteristics and data on multiple risk factors include demographic information, lifestyle, nutritional status, activities, sleep, psychological health, sensory status, chronic health conditions and medication use.
Data collection of the BLINDSCE cohort
The research staff then perform a multidimensional disability assessment, including general disability and health status, functional dependency, locomotor function and physical function. General disability is measured by the WHODAS 2.0, a standardised assessment tool, with six domains: cognition, mobility, self-care, getting along, life activities and participation.12 The WHODAS 2.0 consists of 12 items, each scored from 0 to 4. Functional dependency is assessed by the participant’s performance of basic ADL (BADL) according to the Katz index,22 the Barthel index (BI)23 and instrumental ADL (IADL) designed by Lawton and Brody.24 Locomotor function is defined as individual-reported difficulty with getting up from a chair, climbing one flight (10 steps) of stairs, bending, kneeling or crouching, stretching arms beyond the shoulders, lifting or carrying weights greater than 10 pounds, or walking 100 m.25 Grip strength is assessed for each hand two times with a dynamometer (Camry electronic hand dynamometer EH101). The largest value of the four grip strength measurements is used to represent upper limb function.26 The short physical performance battery, including 4-metre walking speed, chair stands and standing balance tests, is applied to reflect the overall performance of the lower limbs. Each item is scored from 0 (worst) to 4 (best), resulting in a total score of 0–12.26 The timed up-and-go test is also administered to evaluate older adults’ mobility and fall risk.27
In addition, cognition is accessed through the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment-Basic version (MoCA-B), the clock drawing test and the subjective cognitive decline (SCD) questionnaire-9. Notably, considering the impact of educational attainment and the feasibility of community-based surveys, the MoCA-B is suitable for screening mild cognitive impairment (MCI) across different education levels in community-dwelling elderly individuals, and it also has higher sensitivity and accuracy than the MoCA and MMSE.28 29 SCD refers to the self-reported experience of cognitive decline in individuals who perform normally in objective cognitive tests, which may suggest unhealthy cognitive changes. Gifford KA and other scholars developed a 9-question SCD screening questionnaire (SCD-9) through scientific item response theory and exploratory factor analysis.30 The Chinese version of the SCD-9 was translated by Han Ying and may be appropriate for detecting SCD in MCI patients early.31
In addition, a range of disability-related outcomes are being measured, including falls, pain, self-reported health, intrinsic capacity by integrated care for older people,32 frailty by the Fried frailty scale,33 sarcopenia based on Asian Working Group for Sarcopenia 2019 consensus,34 hospitalisation and death.
Finally, clinical data, including physical parameters, biochemical data from regular physical examinations and a blood sample, are collected. The physical parameters of the participants, including height, weight, waist circumference, hip circumference, calf circumference, blood pressure, body fat and bone density, are being measured onsite by research staff. Biochemical data, including data from routine blood tests, liver and kidney function, serum lipids and fasting glucose, are being extracted from the latest check-up records. Some of the participants (<1000) are having their venous blood sampled voluntarily for glucose metabolism-related measures, telomeres and omics analysis.
Follow-up visits
This study includes follow-up visits every 1–1.5 years after baseline at local community healthcare centres or nursing homes and involves assessments of disability and related outcomes, cognitive assessments and clinical measures. At least three follow-ups are now scheduled. A multifactorial questionnaire will be collected every two visits (table 1). Repeated collection of blood samples will be conducted at the third follow-up. Participants lost to follow-up will be supplemented by new recruits in the same community to keep the cohort size relatively stable over time. To encourage study retention and reduce loss to follow-up, we are including residents who are regularly served by community healthcare centres at baseline as much as possible. Moreover, research staff will keep in contact with participants throughout the study period to establish rapport and remind them of their following survey before the follow-up visit. At the same time, participants will receive a gift after each visit to facilitate retention. If needed, telephone-based follow-up will be used for participants who are unable to visit in person.
Study outcomes
The primary outcome of this study is disability. The secondary outcomes include cognition and many conditions related to disability, such as falls, pain, poor health, decreased intrinsic capacity, frailty, sarcopenia, hospitalisation and death.
Data collection and quality control
Data collection using an electronic database system
Data are being collected through face-to-face assessments and supplemented with a review of electronic medical records. We are using an electronic database system (https://h6world.cn/website/index.html) for data collection, storage and management, including three-level identity accounts: survey staff, monitors and principal investigators. The staff accounts are used by trained research staff to assign a profile and a unique identification number for each participant and record real-time case report form (CRF) information online. If necessary, a paper-based CRF can also be used, in which data input must be completed on the survey day. The monitor account is assigned to subinvestigators, who review the collected data and check its completeness, accuracy and authenticity daily. If there is any query, the research staff is requested to provide an explanation or revision promptly. The lead researcher in charge of the entire research process uses the principal investigator account. The database and all original files with searchable catalogues are retained for the required period.
Storage and primary analyses of blood samples
A 10 mL venous blood sample is planned to be drawn from each voluntary participant (a predicted number of <1000) by professional certified nurses. Blood tests, including fasting glucose, insulin and C-peptide tests, are conducted without charge to the participants. After centrifugation, the serum, plasma and mononuclear cells from the peripheral blood are isolated, aliquoted and stored at −80°C in the clinical sample centre, Xuanwu Hospital. These samples will be used for glucose metabolism-related measurements, and the residual blood sample will be preserved for further use, including telomere and omics analyses when needed.
Quality control and staff training
Although many survey sites are involved, the study is managed by a single research centre, Xuanwu Hospital. Detailed guidelines for the survey were developed. To maintain consistency, each research staff member must receive face-to-face professional training. Specifically, the training consists of an introduction to the study aims and procedure, questionnaire administration and scoring instructions, quality control methods and electronic data collection. Each research staff member must complete the training and the simulation survey process before the formal survey starts. At the same time, a major researcher takes responsibility for onsite supervision, and regular meetings are being held throughout the study.
Statistical methods
Cross-sectional analysis of baseline information
Baseline information will be described using summary statistics to demonstrate the characteristics of the participants. Categorical variables will be presented as numbers and percentages, and continuous variables will be reported as the means (SD) or medians (IQRs). Group comparisons will be conducted using χ² tests for categorical variables, Student’s t tests or analysis of variance for normally distributed continuous variables and Wilcoxon rank or Kruskal-Wallis rank tests otherwise. For prescreening the risk variables, multivariable logistic regression will be used to evaluate the associations between disability and potential risk factors based on baseline cross-sectional data, such as sleep, lifestyle, physical activity, chronic health conditions and biochemical measures, including glucose metabolism, telomeres and omics analysis.
Data analysis of longitudinal information
Based on the modelling data set, we plan to use Cox proportional hazards (defining the outcome as the time-event variable) or logistic regression (defining the outcome as the binary variable) to build a prediction model and draw a column-line graph. Then, bootstrapping is used to validate the model performance internally. This prediction model will be externally validated using a validation data set. Latent class growth analysis (LCGA) or growth mixture modelling (GMM) will be conducted to describe the changes in disability and other functions and explore trajectory analyses among community-dwelling older adults. Given the complementary strengths of LCGA and GMM, LCGA will serve as the initial exploratory tool to rapidly delineate different latent classes, and GMM will be used for a more nuanced analysis, accounting for individual variations within each class to yield deeper insights.
Patient and public involvement
We originally conducted a pilot survey covering questionnaires and measurements in several non-medical persons before the formal study, mainly to assess the structure of the CRF and the data collection time. In the future, this study will establish a group of public volunteers, including elderly individuals from different survey sites, to help interpret and disseminate the results in a meaningful way. The contact information of the research team is being provided to all participants and the public.
Ethics and dissemination
Ethical approval for the BLINDSCE was provided by the Ethics Committee of Xuanwu Hospital, Capital Medical University (No: [2023]129). Title of the approved project: Beijing Longitudinal Disability Survey in Community Elderly. Date of approval: 1 August 2023. This study will be performed according to the World Medical Association Declaration of Helsinki Ethical Principles for Medical Research Involving Human Subjects 1964 (as amended) and the International Ethics Guide for Human Biomedical Research (third edition, 2002). Written informed consent will be obtained from all participants (or proxy) who meet the inclusion criteria before data collection. Any amendments to the protocol will be submitted to the Ethics Committee of Xuanwu Hospital, Capital Medical University, for review and approval before implementation. This study was funded by Capital’s Funds for Health Improvement and Research (CFH, No: 2024–2 G-20112). The study results will be published in peer-reviewed academic journals and presented at national and international professional conferences focusing on geriatric medicine and public health, particularly community healthcare communication meetings, to increase the visibility of the findings and aid in the translation and transfer of knowledge.
Discussion
Recent decades have witnessed a rapid and irreversible trend towards an ageing society in China, which has resulted in remarkable growth of the population with disability. Under such circumstances, the government and the public health community have prioritised efforts to address this issue. For the initial step in preventing disability, accurate screening and identification of high-risk factors are essential for the early prevention of disability. This study will establish a longitudinal cohort of community-dwelling older adults with standardised multidimensional disability assessments in Beijing and examine the predictive role of multiple factors.
The strengths of this cohort include the massive number of community-based elderly adults comprising nearly 2000 individuals from urban and rural areas in Beijing, which may add to the stability and external validity of the study findings for other elderly individuals living in this city. Second, this study reports disability using WHODAS 2.0, which measures disability across cultures in multiple life domains, including mobility, life activities and participation, coupled with traditional disability assessments, including BADL, IADL and BI, and disability-related outcomes. Multidimensional disability assessments will improve our capacity to describe the disability and health of community-dwelling elderly individuals and facilitate comparisons across studies. Another strength of this study is that a wide range of modifiable factors are surveyed using easily and readily available questionnaires. Hence, prediction models incorporating these factors can also be applied in primary care to identify high-risk elderly individuals and develop precise interventions, which are essential for reducing the incidence of disability. With a set of community-based, disability-related measures, including clinical, laboratory, physical, individual and social determinants, the BLINDSCE cohort study will provide deep insight into the cause and course of disability and its risk factors for the community ageing population. These advances will direct disability prevention and intervention and eventually achieve health benefits for community-dwelling elderly individuals.
This study also has several limitations. First, this is a single-centre study in Beijing and not a nationwide study in China, which may limit the generalisability of the research results. Second, this cohort included only elderly individuals aged ≥65 years. Given that many disability-related diseases occur at younger ages, the age ranges of participants will be expanded in our future study. Third, the current follow-up time of this study is relatively limited. However, this is merely the start of establishing a more dedicated longitudinal community-dwelling elderly cohort. Ideally, this cohort plans to long-term follow-up targeting multiple disability-related outcomes.
Study status
This study began enrolment in August 2023, and the first collection of baseline data occurred from August 2023 to January 2024. Follow-up visits are ongoing.
Ethics statements
Patient consent for publication
Acknowledgments
The authors thank all the community-dwelling elderly individuals in this study for their willingness to participate. The authors also thank partners in the community healthcare centres of Mafang Town and Qingta Street and nursing homes for their valuable help in study enrolment.
References
Footnotes
Contributors YG acted as guarantor. YG and XY proposed the conception of this study and obtained funding. YG and SC contributed to the design and quality control of this study. SC, XD and HL collaborated in the study conduct. SC, XD and YW were involved in the acquisition of the data. SC, XD, C-BH and SL contributed to the analysis and interpretation of the data. SC wrote the draft of this protocol, and YG revised it. All authors reviewed and approved this manuscript.
Funding This study was funded by Capital’s Funds for Health Improvement and Research (CFH, grant numbers: 2024-2G-20112) and the Project for Innovation and Development of Beijing Municipal Geriatric Medical Research Center (No: 11000025T000003320658). The funding body had no role in the design or writing of this protocol.
Competing interests None declared.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.