Independent and joint association of accelerometer-measured sedentary behaviour and physical activity with mild cognitive impairment and intrinsic capacity decline among Chinese older adults: study protocol for a cross-sectional study

Abstract

Introduction

Numerous studies have indicated that sedentary behaviour (SB) negatively impacts cognitive function, while engaging in physical activity (PA) helps maintain cognitive performance and delay cognitive decline. However, fewer studies have examined the interaction between these two factors on cognitive function, especially among the Chinese older population. In the realm of healthy and active ageing, intrinsic capacity emerges as another crucial determinant for maintaining functional wellness. However, the independent and joint relationship of SB and PA to intrinsic capacity remains unknown. Given the substantial implications of cognitive function and intrinsic capacity decline, this study will be the first to investigate the independent and joint association of SB and PA with cognition and intrinsic capacity in Chinese older adults.

Methods and analysis

This is a single-centre, cross-sectional, observational and exploratory study design targeting the older population from communities in Beijing. Our study population will include 270 older individuals aged 65 years and above. SB and PA levels will be tracked using triaxial accelerometers (GT3X+, ActiGraph, Pensacola, FL, USA), worn on the dominant-side waist by participants for seven consecutive days. A well-designed questionnaire will be used to gather initial data on the sociodemographic and other characteristics (all relevant risk factors for mild cognitive impairment (MCI) and intrinsic capacity decline). These characteristics will later be included as confounders in multivariate regression analyses. Participants will be screened for MCI using the montreal cognitive assessment scale, the clinical dementia rating scale and the activities of daily living scale. The amnestic MCI subtype will undergo additional evaluation using the mini-mental state examination scale and the MemTrax continuous memory recognition test. The integrated care for older people screening tool will evaluate the intrinsic capacity of older adults. To further investigate the interaction between SB and PA, participants will be divided into the following groups: (1) Mildly sedentary+active, (2) Mildly sedentary+inactive, (3) Severely sedentary+active and (4) Severely sedentary+inactive. 270 older participants will be stratified by age: 65–69 years, 70–79 years and ≥80 years. We will compare the prevalence of MCI and amnestic MCI, as well as intrinsic capacity scores, among older adults with different levels of SB and PA across these three age groups, and calculate ORs and 95% CIs.

Ethics and dissemination

Approval for the project was granted by the Sports Science Experiment Ethics Committee of Beijing Sport University on 12 June 2024 (ID: 2024193H). Participants will provide informed consent, guaranteeing voluntary participation. The data collected will be anonymised and securely stored in a database. The results of the study will be disseminated through open-access, peer-reviewed publications and scientific events.

Trial registration number

ChiCTR2400085482.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The study employs accelerometers for objective and valid quantification of sedentary behaviour and physical activity levels, instead of questionnaire-based surveys.

• Another strength of the study is the inclusion of 16 risk factors as multiple confounders in regression analysis.

• Outcome measurements rely on established clinical and functional rating scales due to the lack of more precise measures for a substantial sample size.

• Another limitation is the focus on older adults in Beijing communities, and the sample size is relatively small. Therefore, extrapolation to broader populations should be made with caution.

Introduction

Mild cognitive impairment (MCI) represents the initial stage of cognitive decline, marking a transition between healthy ageing and dementia.¹ Those with MCI may struggle with memory loss, diminished language abilities, impaired judgement and decline in other cognitive functions.² Globally, an estimated 19.7% of adults aged 50 years or older have MCI.³ An extensive cross-sectional study in China reported a 15.5% prevalence of MCI among adults aged ≥60 years, affecting approximately 38.77 million individuals.⁴ A prior study demonstrated that approximately 33.6% of patients diagnosed with MCI progressed to dementia within 3 years.⁵ If timely detection and necessary interventions are implemented, some patients can maintain their cognitive status or even return to normal cognitive function.⁶ Additionally, MCI is classified into two subtypes: amnestic MCI and non-amnestic MCI, based on the affected domains of cognitive function.⁷ Amnestic MCI is primarily characterised by significant memory loss, increasing the risk of progression to Alzheimer’s disease with an annual conversion rate of 28%.⁸ The dual burden of economic strain and deteriorating quality of life in ageing populations necessitates prompt intervention. Early identification and management of MCI, especially its subtype amnestic MCI, is vital for preventive healthcare.

Multiple risk factors may influence MCI. A high-quality cross-sectional study involving 46 011 older adults in China identified 12 risk factors for MCI.⁴ Other studies on MCI also support similar conclusions.9,12 Based on their amenability to intervention, these factors are further classified into modifiable and non-modifiable factors. Modifiable risk factors include fewer years of education, rural residence, living alone, smoking and inadequate prevention and management of chronic diseases such as hypertension, diabetes, hyperlipidaemia, heart disease and cerebrovascular diseases. Non-modifiable risk factors include ageing, female sex and family history of dementia. Additionally, other literature suggests that alcohol abuse, obesity or malnutrition, low income, depressive symptoms and poor sleep quality are also associated with an increased risk of cognitive decline.13,15

Regarding daily behaviour, sedentary behaviour (SB) refers to any waking behaviour with an energy expenditure of 1.5 Metabolic Equivalent of Task (MET) or less while sitting, reclining or lying. In 2020, guidelines from the World Health Organisation (WHO) highlighted the association between increased SB levels and elevated risks of all-cause, cardiovascular and cancer mortality, along with negative health outcomes such as cardiovascular diseases. Conversely, physical activity (PA) exerts a beneficial effect on health outcomes.¹⁶ A rigorous systematic review found that among individuals with daily SB exceeding 8 hours, those engaging in the highest level of PA (>35.5 MET-hour per week) had no significant increase in all-cause mortality risk compared with the reference group (<4 hours of SB daily and the highest PA levels). The study demonstrated that high levels of PA can substantially offset the mortality risk associated with prolonged SB.¹⁷

Animal experiments have demonstrated that prolonged sedentariness elevates the risk of cognitive decline, partly by impairing synaptic plasticity and hippocampal neurogenesis. Conversely, exercise enhances neural stem cell differentiation, counteracting neurodegeneration in Alzheimer’s disease models.18,20 Human evidence also supports this bidirectional relationship. Cross-sectional studies consistently show that older adults with MCI exhibit higher SB and lower PA levels.^{21 22} While the precise SB threshold for cognitive risk remains undefined, high-quality studies have indicated that daily sitting exceeding 8 hours significantly elevates the risk of adverse outcomes.^{17 23} Crucially, dose-response analyses have revealed a non-linear relationship between SB and dementia risk. Compared with a reference level of 9.27 hours/day of SB, dementia risk increases by 8% at 10 hours/day, 63% at 12 hours/day and 221% at 15 hours/day.²⁴ In contrast, regular exercise mitigates cognitive decline in older adults with MCI. Moderate to vigorous physical activity (MVPA) confers superior cognitive benefits than low-intensity physical activity.^{25 26} Further analysis of the interaction between SB and PA has found that in non-sedentary older adults (<4 hours of SB/day), inactivity (<600 MET-min/week of PA) raises the risk of MCI by 63% compared with active adults. Compared to non-sedentary and active reference, among sedentary adults, inactivity intensifies the risk by 385%, while maintaining activity attenuates this risk (resulting in a 90% increase).²⁷ The WHO recommends that older adults should reduce their sedentary time by replacing it with PA of any intensity, particularly emphasising MVPA exceeding 150–300 min/week to achieve greater health benefits.²⁸

For older adults with MCI, intrinsic capacity might serve as a more sensitive indicator of functional decline compared with traditional activities of daily living (ADL).²⁹ Intrinsic capacity encompasses cognitive, psychological, sensory (visual, auditory), vitality and motor domains, representing a comprehensive reflection of the physical and mental capacities of older individuals.³⁰ Notably, the risk factors influencing intrinsic capacity are similar in MCI.^{31 32} A Spanish cohort study found high levels of SB reduced intrinsic capacity, while MVPA enhanced it. Moreover, ≥39 min/day of MVPA improved intrinsic capacity regardless of SB levels.³³ Exercise interventions of a 26-week aerobic exercise/resistance training, with home sessions, significantly improved intrinsic capacity in older adults with subjective memory decline.³⁴ Despite these advances, critical gaps persist in understanding SB/PA associations with intrinsic capacity, particularly within China’s ageing population.

In summary, insufficient prior research exists on the independent and joint associations of SB and PA with MCI, particularly its subtype (amnestic MCI), and intrinsic capacity decline among older adults in China. After adjusting for multiple risk factors, we hypothesise that individuals with severe SB (≥480 min/day) will show a higher prevalence of both MCI and amnestic MCI, along with lower intrinsic capacity, compared with those with mild SB (<480 min/day). Conversely, those meeting active PA thresholds (MVPA ≥150 min/week) are expected to demonstrate a lower prevalence of MCI and amnestic MCI, as well as higher intrinsic capacity scores relative to inactive individuals (MVPA <150 min/week). Furthermore, the interaction between SB and PA levels is hypothesised to influence cognitive function and intrinsic capacity, suggesting that engaging in sufficient PA may mitigate the detrimental effects of prolonged SB on cognitive function and overall health.

Objectives

The primary objective is to evaluate and compare the prevalence of MCI in older adults with different levels of SB and PA, considering both the independent and joint associations of SB and PA. The secondary objective is to compare the prevalence of amnestic MCI and intrinsic capacity levels among older adults.

Methods and analysis

Study design and setting

This is a single-centre, cross-sectional, observational and exploratory study design. The study will be conducted in accordance with the strengthening the reporting of observational studies in epidemiology statement guidelines for cross-sectional studies, ensuring methodological rigour and transparency.

This cross-sectional study began on 18 June 2024, and is anticipated to last until 18 June 2025, in Beijing, China. As the capital of China, Beijing has a permanent resident population of 21.832 million, of which 16.5% are 65 years old and above, according to the Beijing Municipal Civil Affairs Bureau in 2024. The survey’s design aims to ensure broad sample representation by including older individuals from diverse age groups and locations, spanning both urban and rural areas. Recruitment for this single-centre study will be guided by the ‘2023 Beijing Elderly Work Development Report’ from the Beijing Association of the Aged and the 2023 regional gross domestic product data from the Beijing Bureau of Statistics. Participants will be recruited from five Beijing municipal districts (Haidian, Shijingshan, Fengtai, Tongzhou and Miyun), selected to achieve population representativeness based on their varying proportions of older residents and differing economic development levels. Among the five districts, communities with a higher concentration of older adults will be prioritised. All recruitment procedures and data collection will be centrally managed and implemented by Beijing Sport University.

Study participants and recruitment

Convenience sampling will be employed to ensure adequate participant recruitment while accounting for diverse age distributions and varying SB and PA levels. Potential participants will be recruited through interviews, posters and contact information from community health centres, senior activity facilities and other diverse settings. Notably, the study design is a prevalence study, and participant selection will be based on the general characteristics of the older population rather than on their cognitive status. The diagnosis of MCI serves as the primary outcome, determined after enrolment through a standardised multi-step assessment protocol. Researchers will provide participants with a detailed explanation of the study’s purpose, potential benefits, risks and other relevant information prior to participation. All participants will be required to sign an informed consent form at the research site. For illiterate participants, a relative may sign the form on their behalf, provided consent is obtained. Participants will retain the right to withdraw from the trial at any point.

Inclusion and exclusion criteria

The inclusion criteria are as follows:

1. Individuals aged 65 years or older.

2. Those who agree to participate and are able to provide written informed consent.

The exclusion criteria include the following:

1. Significant visual or auditory deficits or impaired communication abilities that could potentially interfere with assessment.

2. Restricted hand movement that would interfere with assessment testing.

3. Restrictions in mobility that result in the requirement to use a wheelchair or other assistive tools.

4. Diagnoses of Alzheimer’s disease, other types of dementia, cerebrovascular disease or other disorders (such as central nervous system infections, Parkinson’s disease, traumatic brain injury, poisoning and metabolic disorders) that directly affect cognitive function.

5. Use of medications that may affect cognitive function, such as cholinesterase inhibitors, within the past 6 months.

6. Diagnoses of schizophrenia, anxiety disorders, depression or other psychological disorders.

Patient and public involvement

None.

Data collection procedure

To mitigate inconsistencies and ensure accurate implementation of the survey, three investigators (including two senior undergraduate students majoring in rehabilitation physiotherapy and one graduate student majoring in rehabilitation medicine and physical therapy) will receive comprehensive standardised training. Before the trial begins, three training sessions will be held, each lasting 1.5–2 hours, to ensure uniformity in survey methods and scoring criteria. During the survey, standard language will be employed for explanations to prevent investigator bias. Afterwards, the three investigators will conduct a pre-survey of 20 to familiarise themselves with the data collection methodology and ensure the high quality of the data collected during the formal survey. In the subsequent trial, demographic and sociological data and related outcome measures will be collected through one-on-one interviews conducted by rigorously trained investigators with older adults in quiet rooms. The investigators will read each question on the questionnaire aloud to the older adults and directly record the participants’ responses to ensure standardisation and minimise missing or obviously unreasonable data. Promptly address any omissions and respond to participant inquiries without delay.

Data collection will comprise exposure variables (SB and PA levels), outcome measurements (cognitive function and intrinsic capacity) and relevant covariates. The assessment of SB and PA levels as exposure variables will be carried out using triaxial accelerometers for objective measurements. The collection of outcome measurements will involve a well-designed printed questionnaire and an electronic online assessment tool. The collection of relevant covariates will encompass ‘identified risk factors’ (sociodemographic and clinical characteristics), ‘behavioural factor’ and ‘other putative risk factors’.

Exposure variables (SB and PA levels)

Unlike most prior studies that used questionnaires, this study will quantify SB and PA levels by employing a triaxial accelerometer (GT3X+, ActiGraph, Pensacola, FL, USA). As a motion-sensing device, the accelerometer delivers accurate, continuous data on daily sedentary time, PA intensity levels and energy expenditure.

To investigate the independent associations of SB and PA with cognitive function and intrinsic capacity, the associations of SB and PA will be examined separately. To investigate the joint associations, participants’ behavioural patterns will be classified into four groups. According to previous studies, specific details about accelerometer wearing, along with SB and PA level classifications, are shown in table 1.

Table 1. Details regarding triaxial accelerometer measurements and specific criteria for grouping.

	Detail	Description	Citation
Accelerometer related details	Device placement	On the dominant-side waist	Freedson et al60
	Sampling frequency	30 Hz
	Epoch length	60 s	Trost et al61
	Wearing time	7 days, remove when sleeping, bathing or engaging in water activities	Tucker et al62
	Non-wear time definition	0 count on the accelerometer for a continuous period ≥60 min
	Valid data	≥4 days of wearing, including 3 weekdays and 1 weekend day, and ≥10 hours/day
	Cut-off values	0< SB <100 cpm 100≤ LPA <1952 cpm 1952≤ MPA <5725 cpm VPA ≥5725 cpm	Freedson et al60
SB/PA level classification (for independent association analysis)	SB level classification	Mild: <480 min/day Severe: ≥480 min/day	Huang et al63
	PA level classification	Active: MVPA ≥150 min/week Inactive: MVPA <150 min/week	Strain et al64
SB+PA (for joint association analysis)	Mildly sedentary+active	SB <480 min/day and MVPA ≥150 min/week	You et al65
	Mildly sedentary+inactive	SB <480 min/day and MVPA <150 min/week
	Severely sedentary+active	SB ≥480 min/day and MVPA ≥150 min/week
	Severely sedentary+inactive	SB ≥480 min/day and MVPA <150 min/week

cpm, counts per minute; LPA, low intensity physical activity; MPA, moderate intensity physical activity; MVPA, moderate to vigorous intensity physical activity; PA, physical activity; SB, sedentary behaviour; VPA, vigorous intensity physical activity.

To ensure accurate data collection on SB and PA levels in older adults, an education and training programme will be implemented. This programme will include: (1) Providing detailed information on the device’s purpose to enhance participant understanding, (2) Instructing participants on proper wearing procedures (including positioning and regular checks). Specifically, participants will be instructed to: wear the device on the dominant side waist, fasten the waistband on the outer side of clothing, ensure it remains flat and stable, wear it daily from morning until bedtime and check its position regularly every 3–4 hours and (3) Emphasising the importance of wearing the device throughout the day, except during sleep, bathing or water activities. Participants will also be instructed not to alter their daily routines while wearing the device.

Data extraction and analysis will be performed using ActiLife V.6.2 software (ActiGraph, Pensacola, FL, USA). If the participant’s accelerometer wear time is found to be non-compliant, they will be required to re-wear the device for an additional period. Data from participants who refuse to re-wear will be excluded from statistical analysis prior to data processing.

Outcome assessment

Outcome measurements will focus primarily on cognitive function and intrinsic capacity. The diagnosis of MCI is established based on widely recognised criteria, with its subtype (amnestic MCI) further identified. Intrinsic capacity will be assessed using screening tools recommended by the WHO.³⁰

MCI assessment tools

Subjective cognitive decline

To detect subjective cognitive decline, participants will be queried with two questions: ‘Have you experienced a recent decline in your ability to memorise new things?’ or ‘Have any of your friends or relatives made remarks about your worsened memory or concentration?’³⁵ An affirmative response to either question indicates the presence of subjective cognitive decline.

Montreal cognitive assessment (MoCA) scale

Participants will undergo an evaluation using the Beijing version of the MoCA scale to screen for MCI, given its excellent reliability and validity, with a notably high specificity.³⁶ The MoCA scale evaluates a wide array of cognitive domains, encompassing visuospatial and executive function, naming, attention, abstraction, language, delayed memory and orientation.³⁷ Scores on the MoCA scale range from 0–30, with higher scores indicating better cognitive performance. Specific cut-off scores for diagnosing MCI are set based on education level: ≤13 for illiterate individuals, ≤19 for those with primary school education and ≤24 for those with junior high school education or higher.³⁸

Activities of daily living (ADL) scale

The ADL scale will be used to assess the overall ability of older individuals to perform essential daily tasks. It encompasses both basic activities of daily living and instrumental activities of daily living, with a total of 14 items. Scores on the ADL scale range from 14–56. A score of ADL ≤26 indicates preserved ADLs.^{39 40}

Clinical dementia rating (CDR) scale

The CDR scale will be used to determine the presence or absence of dementia. A CDR score of 0 or 0.5 indicates the absence of dementia or suspected presence.⁴¹

Amnestic MCI assessment tools

Amnestic MCI is defined as a subtype of MCI. It requires assessment for memory domain impairments using designated tools, alongside fulfilling the standard MCI criteria.

Subjective memory complaint

Older individuals will be asked whether they perceive a decline in their memory abilities using the question: ‘Do you feel like your memory has become worse?’⁴²

Mini-mental state examination (MMSE) scale

General cognitive function will be assessed using the MMSE scale. Normal cognitive function is defined by scores surpassing education-specific cut-offs: >16/17 for illiterate individuals, >19/20 for those with primary school education and >23/24 for those with junior high school education or higher.⁴³

MemTrax continuous memory recognition test

MemTrax is a rapid electronic online assessment tool (available at: https://qy.memtrax.com.cn/) that uses a computerised continuous memory recognition test to assess episodic memory. It is useful for identifying individuals with amnestic MCI, demonstrating high sensitivity and specificity.⁴⁴ In each MemTrax test session, participants are presented with a set of 50 images: 25 unique images (5 from each of 5 categories) and 25 repeated images. Each image is displayed for 3 s or until a behavioural response (pressing the space bar) is recorded. Participants are instructed to press the space bar promptly on seeing a repeated image (an image that is identical to a previously displayed one). The next image follows immediately after the participant’s response. Each test lasts approximately 1.5 to 2 min.⁴⁵ After the test, the programme automatically calculates the percentage of correct responses to the repeated pictures (MTX-%C) and the mean response time (MTX-RT). Better cognitive function is indicated by a high MTX-%C and a reduced MTX-RT. An MTX-%C falling 1.5 SD below that of age-matched and culture-matched control populations indicates objective memory impairment.⁴⁶

Intrinsic capacity assessment tool

This study will evaluate the intrinsic capacity of older adults using the integrated care for older people (ICOPE) screening tool, which includes nine questions across five dimensions of intrinsic capacity. Each dimension is assigned a score of either 0 (representing abnormalities) or 1 (representing normal function), yielding a total score ranging from 0–6.⁴⁷ The ICOPE screening tool has demonstrated efficacy in detecting physical and cognitive impairments in middle-aged and older adults, with a sensitivity of 95% and a specificity of 57.6%.^{47 48}

Assessment tools for cognitive function and intrinsic capacity are detailed in table 2.

Table 2. Assessment tools on cognitive function and intrinsic capacity.

Outcome measurement		Diagnostic criteria	Tools	Citation
Cognition	MCI	Concern regarding a change in cognition reported by a patient or informant or clinician	Two valid questions	Albert et al2
		Impairment in one or more cognitive domains	MoCA scale
		Preservation of independence in functional abilities	ADL scale
		Not demented	CDR scale
	Amnestic MCI	Subjective memory complaint	A valid question	Petersen et al66
		Normal general cognitive function	MMSE scale
		Objective evidence of memory impairment	MemTrax continuous memory recognition test
		Preserved activities of daily living	ADL scale
		Not demented	CDR scale
Intrinsic capacity		The decline of a composite of cognitive, psychological, sensory (visual, auditory), vitality and motor	ICOPE screening tool	WHO Guidelines Approved by the Guidelines Review Committee 30

ADL, Activities of Daily Living; CDR, Clinical Dementia Rating; ICOPE, Integrated Care for Older People; MCI, mild cognitive impairment; MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment; WHO, World Health Organisation.

Covariates

Covariate data will be gathered using a well-designed questionnaire on sociodemographic and clinical characteristics, all of which are risk factors for MCI and intrinsic capacity decline. Risk factors will be categorised as ‘identified’ and ‘putative’ factors based on Chinese population-based risk factors screened in previous studies^49,12and other relevant literature.13,15 Given the interconnected nature of sleep, SB and PA within a 24-hour daily cycle, this study will explore the significance of sleep quality as a pivotal ‘behavioural factor’.⁴⁹ The specific items for the questionnaire can be accessed online in online supplemental file 1.

Defined ‘identified risk factors’ (sociodemographic and clinical characteristics)

The ‘identified risk factors’ in the present questionnaire are derived from extensive cross-sectional studies. Age will be stratified (65–69, 70–79, ≥80 years) based on participants’ self-reports.⁴ Family history will be categorised as either ‘Yes’ or ‘No’ in response to the validated query: ‘Do your biological father or mother have or have had dementia?’⁵⁰ Literacy levels will be categorised as ‘<1’, ‘1–6’ or ‘>6’ based on the duration of education received.⁴ Drawing on the China health and retirement longitudinal study database, this study categorises respondents’ primary residence as either ‘rural’ or ‘urban’, while excluding temporary business trips, travel and brief visits to friends and relatives from consideration. Marital status will be categorised into ‘widowed’, ‘divorced or living alone’ or ‘married or cohabiting’.¹⁴ Participants’ smoking status will be assessed by inquiring if they are current smokers.⁹ Chronic conditions will be assessed by self-reporting the presence of diagnosed hypertension, diabetes, hyperlipidaemia and coronary heart disease, as well as the current use of medication for these conditions.⁵⁰

Behavioural factor

A question from the centre for epidemiological studies depression scale will be used to evaluate sleep quality: ‘How would you rate your recent sleep quality?’ Sleep quality will be rated as ‘very good’, ‘good’, ‘poor’ or ‘very poor’.^{51 52}

Other putative risk factors

Putative risk factors include alcohol consumption, body mass index (BMI), monthly income and depressive symptoms. Alcohol consumption will be categorised as follows: ‘non-drinkers’ (those who had not consumed alcohol in the past year and not regularly in most weeks in the past); ‘ex-drinkers’ (individuals who did not consume alcohol regularly in the past year but had done so previously); ‘occasional drinkers’ (participants who consumed alcohol less than weekly in the past year and not regularly in most weeks in the past) and ‘current drinkers’ (those who consumed alcohol on a weekly or regular basis in the past year).^{53 54} Monthly income will be classified into the categories of ‘RMB ≤2000’, ‘2000< RMB ≤4000’ or ‘RMB >4000’.¹⁴ BMI will be calculated by dividing the self-reported weight in kilograms by the square of height in metres. Participants will then be categorised as ‘underweight’ (BMI <18.5 kg/m²), ‘normal’ (18.5≤ BMI <24 kg/m²), ‘overweight’ (24≤ BMI <28 kg/m²) or ‘obese’ (BMI ≥28 kg/m²).⁵⁵ Depressive symptoms will be assessed using the validated screening question: ‘Have you ever felt sad or depressed much of the time in the past year?’⁵⁶

Pre-survey

To facilitate successful study execution, a pre-survey was conducted to assess the feasibility and acceptability of the study protocol, with a particular focus on recruitment and data collection methods. The pre-survey was conducted in a community where 20 participants were chosen through convenience sampling, and data were collected from these participants. During this process, we encountered and resolved a range of challenges, including determining effective communication strategies with older adults and methods for accurately evaluating multiple scales.

Following the pre-survey, we will proceed with the formal study implementation process, as outlined in figure 1.

Figure 1. Flow chart of research implementation.

Sample size

The sample size was calculated based on the number of events per independent variable (EPV) in logistic regression analysis. It is recommended that the EPV be a minimum of 10 to ensure robust results,⁵⁷ with some studies suggesting a threshold of 15 for increased reliability.⁵⁸ Our study involves 18 variables: 11 identified risk factors (excluding cerebrovascular disease) from extensive cross-sectional studies,49,12 2 exposure variables (SB and PA levels), 1 behavioural factor and 4 other putative risk factors identified in the Chinese population through other studies.13,15 The sample size was calculated as 15 times the number of variables, yielding a requirement of 270 participants.

Statistical analysis

The collected data will be analysed using R V.4.4.0. Continuous variables will be reported as mean±SD or median (IQR), and categorical variables as frequency (%). The baseline characteristics of the two groups (mildly vs severely sedentary group, and active vs inactive group) will use independent t-tests or Mann-Whitney U tests for continuous variables, and χ² tests for categorical variables. The baseline characteristics across the four groups (mildly sedentary+active, mildly sedentary+inactive, severely sedentary+active and severely sedentary+inactive) will be compared using a one-way analysis of variance (one-way ANOVA) or Kruskal-Wallis test for continuous variables, and χ² tests for categorical variables.

Before conducting multivariate regression analyses, specific procedures will ensure result reliability. First, univariate analysis (with a significance level of p<0.10) will identify important variables and exclude those uncorrelated with the dependent variable. Subsequently, remaining variables will be assessed for the severity of collinearity using the tolerance and variance inflation factor (VIF). A tolerance of ≥0.2 and a VIF of <5 indicate the absence of collinearity.⁵⁹ These variables will then be included in the regression analysis.

Multivariate logistic regression analysis will examine the independent and joint associations of SB and PA with cognitive function and intrinsic capacity in Chinese older adults, calculating ORs and 95% CIs. Three models will be constructed to adjust for confounding: the Crude Model (unadjusted), Model 1 (adjusted for identified risk factors), Model 2 (Model 1 variables plus the behavioural factor of sleep quality) and Model 3 (Model 2 variables plus putative risk factors, including alcohol consumption, monthly income, BMI and depressive symptoms).

For independent associations, ORs and 95% CIs will compare the severely sedentary group to the mildly sedentary group (reference) and the active group to the inactive group (reference), respectively. For joint associations, ORs and 95% CIs will compare three groups (mildly sedentary+inactive, severely sedentary+active and severely sedentary+inactive), with the mildly sedentary+active group serving as the reference group.

Study status

The recruitment phase began in June 2024 and is expected to conclude in February 2025. As of the submission of this protocol manuscript, the study is in its intermediate phase. The entire study is scheduled to be completed by June 2025.

Data management

Data will be managed using research electronic data capture tools and will undergo double-checking to mitigate human errors. To protect participant privacy, anonymisation techniques will be implemented, and data access will be restricted to authorised personnel.

Ethics and dissemination

The project received approval from the Sports Science Experiment Ethics Committee of Beijing Sport University on 12 June 2024 (ID: 2024193H). All participants will provide written informed consent prior to joining the study, ensuring voluntary involvement of older individuals. The research will adhere to the principles of the Declaration of Helsinki. Findings of this research will be disseminated through open-access peer-reviewed publications in reputable scientific journals and conference presentations.

Discussion

While SB-PA interactions have been shown to be associated with adverse outcomes such as cardiovascular disease, evidence regarding cognitive function and intrinsic capacity is limited. With the advent of global ageing, the WHO has called for attention to the association between cognition and intrinsic capacity, as well as their dynamic changes. To the best of our knowledge, no study has explored the independent and joint association of SB and PA with MCI (including amnestic MCI subtypes) and intrinsic capacity decline in Chinese older adults.

One of the study’s strengths and novelties lies in its utilisation of accelerometers for the objective quantification of SB and PA levels. Moreover, participants are required to wear accelerometers on their dominant-side waist for seven consecutive days. This extended duration exceeds typical protocols in similar studies to minimise potential inaccuracies associated with short-term measurements. An additional strength is the employment of multi-model logistic regression analysis. Sixteen risk factors are incorporated as covariates in three regression models to accurately examine the independent and joint associations of SB and PA with MCI and its subtype (amnestic MCI), as well as intrinsic capacity decline.

Two main limitations of this study should be noted. First, this study mainly uses traditional clinical or functional rating scales due to the lack of more precise measures. Second, it focuses on older adults in several Beijing communities and may lack nationwide representativeness. Consequently, future research should consider conducting multi-centre studies encompassing diverse regions.

Overall, this study investigates the potential interaction between SB and PA in relation to cognitive function and intrinsic capacity. Currently, there are various classification standards for SB and PA levels. We adopted a widely used standard, defining ‘severely sedentary’ as SB ≥480 min/day and ‘mildly sedentary’ as SB <480 min/day. PA meeting WHO recommendations is considered ‘active’ (MVPA ≥150 min/week), while PA below this threshold is considered ‘inactive’. We hypothesise that sufficient MVPA may mitigate the adverse impacts of prolonged SB on cognitive function and intrinsic capacity. Confirmation of this hypothesis may strengthen the rationale behind the WHO-recommended guidelines concerning SB and PA for older adults. This study will contribute to the current body of evidence that reinforces the significance of PA in preserving cognitive health and overall well-being.