Study design

This study will be performed and reported according to the Strengthening the Reporting of Observational Studies in Epidemiology Criteria (STROBE) [29].

This is a two-phase, observational study:

Phase 1: A descriptive, cross-sectional study on the prevalence of BDs in individuals aged 65–72.

Phase 2: A longitudinal, population-based cohort study to determine the occurrence of postural BDs and their consequences over an 18-month follow-up period in individuals aged 65–72. The potential consequences include: a) Falls: Defined as involuntary events in which an individual loses their balance and ends up on the floor or another hard surface. b) Fall-related consequences: Including contusions, fractures, hospitalization, and death.

Based on the above-mentioned cohort data, the external validation of the HAT scale will be performed by analyzing its predictive capacity for BDs and the use of health services after 18 months of follow-up (hospital admissions, 30-day readmissions, days spent in the hospital, primary care visits, and specialist care visits).

Setting

The study is set in primary care, targeting people aged 65–75 living in Mataró (Barcelona, Spain).

Participants

The target population is an age-based cohort of people aged 65–75 living in Mataró. This age range was selected to find individuals who do not yet have BDs, enabling the calculation of the incidence. According to the 2017 municipal register, and the population’s age structure, there are 2,215 people in this age range. A total of 1,316 people will be randomly selected from the Primary Care Information System (SIAP) database, which includes health cardholders and covers nearly the entire population. The selection process will be conducted using a computer-generated random sequence to form a sample stratified by gender and age. Our research team will contact the selected participants using the phone numbers registered in the SIAP and invite them to participate. To ensure robustness and minimize non—response bias, a protocol of up to six call attempts on different days and at different times will be implemented for each candidate before considering them unreachable.

Eligibility criteria: 1) Aged 65–75 years at the time of recruitment. 2) Resident of Mataró (Barcelona, Spain) and registered in the Primary Care Information System (SIAP). 3) Able to walk independently, defined as the ability to walk at least 10 meters without the assistance of another person, although the use of technical aids (e.g., cane, walker) is permitted.

Exclusion criteria: 1) Permanent inability to walk or requirement of human assistance for mobility. 2) Terminal illness, as a document in the primary care electronic medical record. 3) Severe cognitive decline, as assessed by the Pfeiffer questionnaire or Mini-Mental State Examination (MMSE) during the initial evaluation, or a known diagnosis of advanced dementia. 4) Significant language barrier that prevents the completion of clinical scales or understanding of study procedures. 5) Lack of phone access for follow-up calls.

Sample size

We expect 33% of the target population (2,215 subjects) to present a BD at recruitment. In Spain, between 14% and 46% of adults aged 65 and over experience one fall per year [30]. In the worst-case scenario, if the cohort with BDs has a fall incidence of 50%, to determine a risk ratio of 1.3 compared to the group without BDs, the latter group’s incidence would need to be around 38.5%. Accepting an alpha risk of 5% and a beta risk lower than 20%, with a dropout rate of 20% at follow-up, 329 subjects would be needed in the BD group and 987 in the non-BD group to detect a minimum risk ratio of 1.3 (objective 2). This total sample of 1,316 subjects allows for determining a prevalence of falls of 50% with ±2.7% precision (objective 1).

Regarding the retinal A/V ratio, a previous study on hypertensive patients observed a prevalence of pathological A/V ratio of 15% [21]. Given that at least 75% of the Spanish population presents with hypertension. Therefore, of the total cohort of 1,316 individuals, we expect 148 to show a pathological A/V ratio. If 33% of the population is expected to have BDs, in this sample of 148 individuals, a prevalence of 12.5% in patients with a normal A/V ratio will represent a significant difference (aim 2).

Functional capacity and postural balance

• Tinetti Test: balance and gait assessment: The Tinetti Test (TT) assesses balance (sitting and standing) with 14 items (score out of 24) and gait with 10 items (score out of 16), resulting in a total score out of 40. Higher scores indicate better performance. The test showed good interrater reliability and concurrent validity [31,32].
• Unipedal Stance Test: The Unipedal Stance Test (UPST) measures static balance ability [33]. It can be conducted with eyes open and closed. Participants must remain in unipedal for 10 seconds on each side, first on the dominant side and then on the non-dominant side. After 30 seconds of recovery, the test is repeated on the opposite side.
• Time Up and Go test: The Time Up and Go test (TUG) measures the time it takes for a subject to stand from a chair, walk three meters, turn around, and sit down again. Times begin when the subject stands up and stops when seated [34,35].
• Balance Assessment Using the Nintendo Wii Console: This includes the one leg balance test and center of gravity assessment, as well as Romberg Test variants: Eyes Open (REO), Eyes Closed (REC), and on Foam Pad (RuFP), as performed in the posturography test. The Nintendo Wii provides accurate measures of body center of pressure (COP), an important metric for balance stability assessment approximating the body’s center of mass. The Romberg test assesses gait disturbance caused by abnormal proprioception, disequilibrium from central vertigo, and peripheral vertigo [36]. Patients are asked to stand with feet together, arms next to the body, first with eyes open and then closed. The patient tries to maintain his balance. The test is scored by counting the seconds the patient can stand with eyes closed.
• Posturography: Static posturography is performed in four different situations using the Romberg test and stability limits with the NedSVE/IBV^® platform [37]. Postural stability is quantified using the Dinascan/IBV force platform (600 × 370 mm of active area, 100 mm in height and 25 kg in weight) with the NedSVE/IBV^® system. Participants find the most stable position while barefoot, with relaxed arms and feet at a 30° angle. Four static conditions are measured, each lasting 30 seconds. Conditions: Romberg test with eyes open (ROA), Romberg test with eyes closed (ROC), Romberg test with eyes open on a foam pad (RGA), and Romberg test with eyes closed on a foam pad (RGC).

Stability limits are assessed by maximum anteroposterior (AP) and mediolateral (ML) displacement (mm), representing the furthest point reached by the centers of pressure in the anteroposterior and mediolateral axes during the recording time. Posturography evaluates whether there is a BD and in which component it occurs (visual (VIS), vestibular (VES), or somatosensory (SOM) [37].

The balance assessment is based on the comparison of the parameters that best discriminate the pathology of the general population with those obtained from patterns of normality segmented by age (database of the Institute of Biomechanics of Valencia). Ratings are displayed in percentages, so that results other than 100% reflect discrepancy with respect to normal values. It is considered pathological when the results are less than 95%, indicating that these subjects are more likely to present a disorder in the studied system. Pathological values for the limits of stability are considered to be below 85% [37].

• Short Physical Performance Battery (SPPB) [38]: The Short Physical Performance Battery is an objective tool for measuring functional capacity, balance, and lower limb strength in adults over 65. The test includes three domains to assess functional mobility: gait, sit-to-stand, and balance. A score below 8 indicates mobility and physical exercise limitations and is associated with a higher risk of mobility disability and, consequently, a higher risk of falls. A score of 8 or higher is considered within the normal range.
• Muscle Strength of Lower Limbs: Maximal isometric muscle strength of knee extension following the protocol described by Gandevia [39].
• Gait Speed: Measured over 6 meters or 2.44 meters if the participant reports walking slowly.
• Functional capacity was assessed using the Lawton-Brody IADL scale [40]. This scale assesses instrumental activities of daily living (IADLs), which measure a person’s ability to live independently. It includes tasks such as managing finances, shopping, using the telephone, and handling transportation. The Barthel ADL index is a scale that measures basic activities of daily living (ADLs), assessing a person’s ability to perform essential self-care tasks such as bathing, dressing, toileting, transferring, and eating.
• Physical activity measured using the abbreviated Spanish version of the Minnesota Leisure-Time Physical Activity Questionnaire (VREM) [41]: The VREM is a short Spanish version of the Minnesota Leisure Time Physical Activity Questionnaire (CAFM). It is a tool designed to assess the quality and quantity of physical activity during leisure time. It consists of six items that evaluate physical activity carried out in the past month, considering only activities performed during leisure time.

Health assessment tool

1. Physical function is assessed by measuring gait speed.
2. Functional cognitive status is determined using the Mini-Mental State Examination (MMSE), with a score ranging from 0 (the worst outcome) to 30 (the best).
3. Chronic morbidity is evaluated by counting the number of diagnosed chronic conditions. Diagnoses, based on the ICD-10 system, are gathered from the patient’s electronic medical record.
4. Mild disability is measured using the Lawton-Brody scale, assessing instrumental activities of daily living (IADLs) that the individual can not perform independently. Four activities will be included in the analyses: shopping, managing finances, using the telephone, and using public transportation. Household tasks are excluded to avoid gender-related biases.
5. Severe disability is defined as the number of basic activities of daily living (ADLs) that the person can not perform independently. ADLs, assessed using the Barthel Index, will include five essential self-care activities: bathing, toileting, transferring, and eating.

Covariables

• Short Falls Efficacy Scale International: The Short Falls Efficacy Scale International (Short FES-I) is the abbreviated version of the FES-I, measuring fear of falling. It includes seven of the sixteen FES-I items, assessing how likely the participant believes they might fall in seven daily situations [42]. Responses range from 1 to 4: not at all concerned, somewhat concerned, fairly concerned, or very concerned. The total score ranges from 7 to 28, with higher scores indicating worse fear of falling [43].
• Record of falls and consequences during the year prior to study inclusion.
• Charlson Comorbidity Index, number of chronic conditions, and specific chronic comorbidities (ICD-10).
• Cognitive capacity assessed using the Pfeiffer short portable mental status questionnaire or Mini Mental State Examination.
• Standard eye acuity test.
• Subjective assessment of sense organs (visual/auditory): normal, mild impairment, or severe impairment.
• Need for technical assistance when walking (yes/no), and type of assistance used.
• Record of rehabilitation gait treatment in the past year.
• Loneliness: measurement of loneliness using the UCLA-3 Loneliness Scale and the Jong Gierveld Loneliness Scale.
• Blood sample for a future study of frailty/sarcopenia biomarkers. Aging is associated with increased proinflammatory cytokines. Elevated cytokines levels are also observed in sarcopenia, obesity, and osteoporosis (increased levels of IL6 and low levels of IGF-1), as well as other endocrine biomarkers (low dehydroepiandrosterone sulfate, vitamin D and leptin).
• Other variables: Affiliation data, numerical ID code, anthropometric measurements, regular drug treatment, socioeconomic status (education, financial level, civil status), and lifestyle factors (smoking habits, alcohol consumption.

Retinography

Using a non-mydriatic camera, retinal images were obtained from each eye without pupillary dilation, capturing both central and nasal views. These images were then analyzed with the IMEDOS Vessel Map software (IMEDOS Systems UG, Germany), while the Tortuosity Index and A/V ratio were measured with the retinal analysis platform of the VARPA group from the University of A Coruña.

Cohort follow-up

• Record of falls (indicating place and activity at the time) and consequences: contusions, fractures, hospitalization, or death within 30 days.
• Formal and informal social care (self-reported, hours/month).
• Hospital admissions, 30-day readmission, days spent in the hospital, primary care visits, and specialist care visits.

The same variables collected at baseline will be gathered again at the final follow-up visit.

Data collection

A standardized data collection protocol and logbook will be established following a pilot test on 50 cases. All information and data collection will be conducted using an electronic form designed with RedCAP. This pilot test aims to detect design errors, train recruiters, and verify consistency in data collection to ensure a standardized process. A random sample of candidate subjects will be selected and contacted (up to six attempts on different days and times). With prior knowledge of their referring doctor or nurse, these candidates will be invited to participate in the study. An informative visit will be scheduled to review inclusion and exclusion criteria and to sign the informed consent form.

If included in the study, the complete baseline visit will be conducted at this time, except for the blood test, which will be scheduled within one month from inclusion. From this baseline visit, subjects will be followed up to detect falls by using: a) personal calendar for recording falls, and b) bimonthly phone calls.

The final visit will be conducted at 18 months, during which balance will be assessed again through clinical tests, retinography, and posturography. The detailed schedule for enrollment, interventions, and assessments is presented in the SPIRIT diagram (Fig 1). A flowchart illustrating the overall study protocol and participant flow is provided in Fig 2.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. SPIRIT diagram illustrating the schedule for enrolment, interventions, and assessments.

https://doi.org/10.1371/journal.pone.0346698.g001

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Flowchart of the study protocol and participant assessments.

https://doi.org/10.1371/journal.pone.0346698.g002

Statistical analysis

Qualitative variables will be summarized as absolute and relative frequencies. Continuous variables will be described using mean and standard deviation or median and interquartile range, as appropriate. Distributional assumptions will be assessed using graphical methods, and the Shapiro-Wilk test will be used to check for data normality. Baseline characteristics between participants with and without balance disorders (BDs) will be compared using the Chi-square or Fisher’s exact test for categorical variables, and the Student’s t-test or Mann-Whitney U test for continuous variables, based on the distributional properties of the data.

BD prevalence and cumulative incidence at 18 months will be estimated with 95% confidence intervals (CIs). The association between BDs and the occurrence of at least one fall during the 18-month follow-up will be analyzed using a modified Poisson regression with robust variance to estimate relative risks (RRs) and 95% CIs. Additionally, a survival analysis for the time until the first fall will be performed using Kaplan-Meier survival curves and Cox regression.

To evaluate the predictive capacity of knee extensor strength and the composite physical-cognitive score for BDs and fall risk, separate multivariable modified Poisson regression models with robust variance will be constructed. Continuous predictors will be modeled as such, and adjusted RRs with 95% CIs will be reported. The diagnostic performance of the Wii Balance Board™ will be evaluated against posturography (gold standard) by calculating sensitivity, specificity, predictive values (PPV, NPV), likelihood ratios, and the area under the ROC curve (AUC). For the validation of the Health Assessment Tool (HAT), we will compare baseline values with those obtained at the 18-month follow-up. The predictive capacity of the HAT scale for BDs and health services utilization (e.g., hospital admissions, primary care visits) will be estimated by computing the AUC.

The analysis regarding retinal microvasculature (Aim 2) will be treated as an exploratory analysis. While we investigate the link between retinal vascular health and BDs using modified Poisson regression adjusted for predefined clinical confounders, this component primarily aims to identify potential new non-invasive biomarkers. Therefore, findings related to these parameters will be interpreted as hypothesis-generating for future research rather than primary confirmatory outcomes.

A two-sided p-value <0.05 will be considered statistically significant. All analyses will be performed using SPSS version 23.0.

Risks and disadvantages for patients

There are minimal risks or disadvantages for the subjects arising from their participation in this study. The only anticipated disadvantages are the following: 1) Blood collection: the extraction of a blood sample for the future study of frailty and sarcopenia biomarkers may be accompanied by minor discomfort or temporary swelling or bruising at the puncture site. All standard clinical measures will be taken by trained health professionals to minimize these effects. 2) Retinal photography: While this involves taking images of the ocular fundus, the procedure is performed using a non-mydriatic camera. This method is quick, safe, and non-invasive, as it does not require pupillary dilation or the use of any medication. 3) Clinical and balance assessments: Procedures such as static posturography, knee extensor strength assessment, and balance tests using Wii Balance Board™ or standard clinical scales (Tinetti, TUG, SPPB) are completely non-invasive and do not involve radiation or painful maneuvers. 4) Follow-up: Some participants may experience a slight inconvenience due to the time required for the baseline and final visits, as well as the bimonthly phone calls used to track the occurrence of falls.

Overall, the benefits of participating in this study – including the early detection of balance disorders and the identification of potential fall risks – outweigh theses minimal risks.

Ethical considerations

The study is being conducted in accordance with the Declaration of Helsinki, agreed at the 64th General Assembly in 2013. Confidentiality and anonymity of the data will be ensured in accordance with current state laws (Organic Law 3/2018, of 5 December, on Personal Data Protection and guarantee of digital rights), both during the execution phase and in any presentations or publications arising from the study. All data were anonymized and the confidentiality of Health Electronic Record (HER) was maintained at all times, in accordance with national and international law.

The research protocol has been reviewed and approved by the Ethics and Clinical Research Committee of the IDIAP Jordi Gol, Barcelona, Spain (protocol code P17/239, first date of approval 20/12/2017; date of approval of amendments 16/11/2022). For studies involving humans, it will be under its supervision during its development and will incorporate its recommendation. An informative sheet was distributed to the participants, and their informed consent was requested in writing.

Participants and their primary care doctors have access to the results of all clinical tests performed during the study through their electronic medical records. During the second visit, subjects are provided with their individual results from the baseline assessment. Upon completion of the study, the final report and any resulting publications will be sent to all participants via email or post to ensure they are informed of the study’s outcomes.

Study status

The first participant in the study was enrolled in April 2022, and recruitment is expected to be completed by December 2025. Given that the data collection phase is still ongoing, no data analysis has been initiated, and therefore, the study has not yet generated any preliminary or final results.

Limitations of the study

The study has several limitations. An estimated loss to follow-up of 10% per year is expected. Although the sample size has been calculated to compensate for this loss, efforts will be made to encourage continued participation through a communication line with the healthcare professionals responsible for each patient and by providing participants with their test results. There is also a risk of underreporting falls with minor consequences. To address this, a telephone interview protocol will be implemented to explore all types of falls. Additionally, variations in data collection may occur, but these will be minimized through thorough training of the study personnel. Furthermore, as the study protocol involves using a non-mydriatic camera without pupillary dilation, we anticipate challenges in acquiring high-quality retinal images from all participants. This may be particularly true in the presence of age-related factors like cataracts or small pupils, potentially reducing the number of samples available for analysis. Another potential limitation is a high non-response rate during the initial recruitment phase. To mitigate this, recruitment will be exhaustive and follow a strict protocol, including making calls on different days and at different times. In cases where incorrect phone numbers are encountered, referring healthcare providers will be contacted and informed. Furthermore, basic information on the reasons for non-participation will be collected to assess potential selection biases.

We acknowledge that some traditional clinical assessments, such as the Tinetti Test, may present a ceiling effect when applied to relatively healthy community-dwelling older adults. To mitigate this limitation and ensure the technical soundness of our protocol, we have established static posturography as our primary objective measure, as it provides quantitative data without floor or ceiling effects. Additionally, the Health Assessment Tool (HAT) has been included specifically because it lacks these limitations, allowing for a more sensitive evaluation of functional decline.

Expected results and relevance

We’re addressing a prevalent health problem that has a clear social impact in the context of a progressively aging population. There is very little literature on the percentage of BD in the population, and a tendency not to report falls with minor consequences. This project is aimed at determining the magnitude of the problem so that it can later be addressed with greater knowledge. In this way, we will be able to respond to primary care doctors’ need for markers that make it possible to identify BD. By identifying subjects at the greatest risk of falling, we will be able to design more efficient preventive interventions (on the community and individual level).

The expected results of this study include a better understanding of the prevalence of BD in the population over 65 years old, which will allow us to extrapolate the percentage of affected individuals within this age group. We also expect to identify simple and accessible measures in primary care, such as gait speed or knee extensor strength, which can serve as indicators of BD. Furthermore, the study aims to determine whether retinography images are related to balance, contributing to the development of new diagnostic tools.

Moreover, the HAT index in our population would help the scientific community to have a simple health index that is a powerful predictor of disease and consumption of health services and resources. This validation could optimize the identification of individuals at greater risk, aiding in planning more effective public health programs.

Additionally, the findings are expected to raise awareness about BD, which are currently underdiagnosed and undertreated, through active dissemination in local media and social networks. Ultimately, these results could support the development of fall prevention and rehabilitation programs, helping to reduce the loss of autonomy in the elderly population.