Identifying Fall Risk Factors: A 7-Year Home Study Using Innovative Marginal Predictions Strategy

Identifying Fall Risk Factors: A 7-Year Home Study Using Innovative Marginal Predictions Strategy | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Identifying Fall Risk Factors: A 7-Year Home Study Using Innovative Marginal Predictions Strategy Amadou M. Djiogomaye Ndiaye, Michel Harel, Laurent Billonnet, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8132697/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Background Falls can occur unpredictably or follow patterns linked to modifiable risk factors and adverse outcomes. Identifying fall trajectories and their key predictors can help clinicians implement targeted prevention strategies. We hypothesize that distinct clinical fall trajectories exist, each with identifiable baseline predictors. Methods This seven-year prospective study followed 1,648 community-dwelling older adults (≥ 60 years). Participants were assessed at home. Data collection included cardiovascular risk factors, fall occurrences, socio-environmental characteristics and a comprehensive geriatric assessment summary score. Fall trajectories were identified using a Gaussian Mixture Model (GMM) and Multinomial Logistic Regression (MLR) determined the predictors of each trajectory. Marginal prediction allowed us to refine predictor analysis by identifying the category within each feature that contributes the most to the steadiest trajectory. Results Four distinct fall trajectories were identified during the 7 years follow up: Cluster Falls and No Falls (65.5%), Increasing Falls (6.7%), Chronic Recurring Falls (15.7%) and Low-Rate Chronic Falls (12.1%). At baseline the steadiest trajectory is Increasing Falls. Clustered Falls and No Falls trajectory is characterized by lack of leisure activities, functional impairment (Instrumental Activities of Daily Living [IADL] < 8) and pathological performance on the Single Leg Balance (SLB) test. The Chronic Recurring Falls trajectory was primarily composed of women with obesity. The Low-Rate Chronic Falls group also consisted mainly of obese women with IADL < 8 and pathological SLB. We also investigated predictors at the 18-month follow-up. Conclusions Falls in older individuals may occur at discrete intervals or follow recurrent patterns, including chronic recurrence, all of which are associated with increased risks. Women, obesity, impairment in activities of daily living, reduced physical performance and depressive symptoms should be prioritized for intervention. Health sciences/Diseases Health sciences/Health care Health sciences/Medical research Health sciences/Risk factors Falls Older Adults Fall Trajectories Predictors Prevention Strategies Figures Figure 1 Figure 2 Figure 3 Brief summary This 7-year home-based study identified four fall trajectories. Baseline predictors included female sex, obesity, low IADL, poor balance, low activity and other factors at 18 months. 1. Introduction According to the World Health Organization (WHO), older individuals are those aged ≥ 60 years. 1 The National Institute of Statistics and Economic Studies (INSEE) estimates that one in three individuals in France will be aged ≥ 60 years by 2060, compared to one in four individuals in 2021. 2 Older age is characterized by a progressive cycle of aging, which manifests as gradual degeneration of functional capacities and an increased risk of falling. 3 , 4 Falls are the leading cause of mortality in this age group and can result in a loss of independence, with significant social, physical and physiological consequences. In France, more than 2 million falls occurred among individuals aged > 65 years in 2022, leading to over 100,000 hospitalizations and 10,000 deaths. 5 The fear of falling often causes older individuals to limit activities such as walking and climbing stairs. While this fear may reduce immediate fall risk, it promotes sedentary behavior, which in turn contributes to progressive frailty over time. 6 Recent studies have aimed to provide concrete evidence on the causes and mechanisms of falls, highlighting fall prevention as a critical concern in geriatrics and public health. 7 – 9 While some falls occur as random, unpredictable events, others follow distinct patterns associated with modifiable risk factors and adverse outcomes. 10 , 11 The study by Ellingsen et al. (2018) identified falls as a cause of death in 88% of reported cases. 12 Understanding these fall trajectories and their underlying risk factors can help clinicians design more effective and targeted prevention strategies. 3 Many studies have used MLR to identify predictive factors for falls, often based on data collected in hospital settings. These data are accessible through patient admissions; however, time constraints frequently limit the assessment of all dimensions of patient health. In our study, data collected at the patient’s home encompassed three key dimensions of health: physical/organic, thymic/cognitive, and social-environmental. Given that the order of variable categories matters in MLR, we used marginal predictions to evaluate the contribution of each category and to explore additional predictive factors without disregarding those already identified, before performing a new MLR. We hypothesized that: (1) distinct clinical patterns of fall trajectories exist, ranging from no falls to recurrent falls, progressively increasing fall frequency, and chronic falls; and (2) baseline predictive factors for falls already reported in the literature such as physical characteristics, female sex, and cognitive impairment can be identified to encourage healthcare professionals to perform a holistic assessment whenever possible. 2. Materials and Methods 2.1. Study participants To contribute to the growing body of research, we analyzed a longitudinal database from the Unit for Prevention, Monitoring and Analysis of Ageing (UPSAV – Unité de Prévention, de Suivi et d'Analyse du Vieillissement ) which systematically collected fall data from older adults living at home through multiple home visits over a seven-year period. The UPSAV team consists of nurses, geriatricians and other healthcare practitioners. Each patient underwent an initial visit, followed by a second visit six months later. Annual follow-ups were then conducted for up to six years, provided the patient remained in their home. 13 The study includes men and women aged 60 and older. To be eligible, participants had to meet the following criteria: Provide written informed consent, either personally or through a legal representative. Not be enrolled in a clinical trial that modifies their standard medical management. Not have progressive pathologies that could significantly affect short-term prognosis. Not reside in a long-term care unit or a nursing home. Be covered by social security at 100%. The data-based study was registered in the data studies registry of the University Hospital of Limoges, in accordance with the General Data Protection Regulation (GDPR) and MR-004 (registration number: 87RI24_0006). In compliance with French regulations, non-opposition to participation was obtained from each participant. Ethical approval was granted by the Ethics Committee of the University Hospital of Limoges, Espace de Réflexion Éthique based in Nouvelle-Aquitaine (ERENA), under approval number 28-2025-05. All methods were performed in accordance with the relevant guidelines and regulations. 2.2. Falls and Clinical Outcomes Assessments During the Follow-up, a fall was defined as unintentionally coming to rest on the ground or other lower level not as a result of a major intrinsic event (e.g., myocardial infarction, stroke, or seizure) or an overwhelming external hazard (e.g., hit by a vehicle). 14 Each patient underwent a Comprehensive Geriatric Assessment (CGA) and received a personalized care plan. According to the patient’s health needs, interventions by an occupational therapist, a psychomotor therapist, or a social worker can be carried out at the patient’s home to implement a personalized care plan, in addition to the interventions provided by nurses and geriatricians. The CGA is a multidimensional and standardized approach designed to enhance clinical practices in the care of older adults through a comprehensive health assessment. 15 2.3. Covariates Covariates included cardiovascular risk factors, fall occurrences, socio-environmental characteristics and the CGA summary. Socio-environmental characteristics assessed in the home included gender, age, previous profession, education level, family situation, lifestyle, number of children, housing conditions, presence of an elevator, long-term illness status, health insurance coverage, leisure activities, social activity, human assistance and pet ownership. Cardiovascular risk factors considered were hypertension, diabetes, dyslipidemia, obesity and tobacco use. The CGA summary encompassed multiple functional and cognitive assessments, including: Verbal fluency test, 16 Single Leg Balance (SLB) test, scored 0–60 seconds, 17 Clock-Drawing Test (CDT), scored 0–5, 18 Activities of Daily Living (ADL), scored 0–6, 19 Instrumental Activities of Daily Living (IADL), scored 0–8, 20 Mini-Mental State Examination (MMSE), scored 0–30, 21 Mini Nutritional Assessment (MNA), scored 0–30, 22 Short Physical Performance Battery (SPPB), scored 0–12, 23 Geriatric Depression Scale (GDS), scored 0-30. 24 For consistency, in the rest of the document, we added 'Pathological' to the feature names SLB test, CDT, Verbal Fluency and GDS to indicate whether the test result is positive or not. By leveraging this database, we identified subgroups with distinct fall trajectories and examined their baseline characteristics to gain a deeper understanding of the factors influencing fall risk. Our study examines a seven-year dataset collected through multiple home visits by the UPSAV unit. 2.4. Data analysis A descriptive analysis was conducted to provide an overview of the recorded study variables. Additionally, a figure illustrates the proportions of fallers and non-fallers. Fallers were classified into two categories based on the number of falls reported at each visit: Faller: a participant who experienced a single fall within a year. Recurrent faller: a participant who experienced two or more falls within a year. Long-term observation of fall events over multiple years allowed us to define chronic faller as a participant who experienced at most one fall per year over several consecutive years, indicating a persistent but non-acute fall pattern over time. To identify clinically distinct trajectories of falls, we applied a GMM, 25 analyses were performed using Python, version 3.12 (Python Software Foundation, Wilmington, DE). The final model was selected using a combination of the Bayesian Information Criterion (BIC)-which balances model complexity and fit quality-and by ensuring that each estimated trajectory group represented at least 5% of the study population. 26 The metric used to define trajectories was the number of months participants remained in the study. A MLR was then applied to identify the best predictors for each trajectory, analyses were performed using R, version 4.3.3 (R Foundation for Statistical Computing, Vienna, Austria). 27 When identifying predictors, the order of categories within each feature is crucial. This study utilizes marginal predictions to quantify the isolated effect of each category within its feature, allowing us to determine which category contributes the most to maintaining the steadiest trajectory, setting it as the reference within its feature. 28 With this new ordering, we conducted a new MLR to identify additional predictors. 3. Results 3.1. Participants A total of 1,648 individuals met the study inclusion criteria. Table 1 presents the socio-environmental and health characteristics of the study sample at baseline. Among the included older adults, 1,113 (68%) were female and 535 (32%) were male. Additionally, 73% had hypertension and only 288 (17%) participated in social activities. The mean age of participants was 83 ± 6 years. Regarding fall occurrences, 823 participants (nearly 50%) had experienced a fall in the past year. In terms of housing conditions, 991 (60%) were homeowners. Furthermore, 449 participants (27%) were classified as depressive patients. Table 1 Overview of Baseline Characteristics According to falls of the study. Falls of the study Features of the study Total sample (N = 1,648) n (%) No falls (n = 794, 48.2%) Falls (n = 854, 51.8%) p-value* Woman 1,113 (68%) 500 (63%) 613 (72%) < 0.001 Age, mean ± SD, years 83 ± 6 82 ± 6 83 ± 6 0.001 Hypertension 1,209 (73%) 575 (72%) 634 (74%) 0.40 Diabetes 339 (21%) 146 (18%) 193 (23%) 0.035 Dyslipidemia 742 (45%) 360 (45%) 382 (45%) 0.80 Tobacco 178 (11%) 85 (11%) 93 (11%) 0.90 Obesity 405 (25%) 179 (23%) 226 (26%) 0.065 Previous profession 0.10 Business owner / Executive 165 (10%) 87 (11%) 78 (9.1%) Company employee 500 (30%) 253 (32%) 247 (29%) Farmer 106 (6.4%) 53 (6.7%) 53 (6.2%) Housewife/Househusband 163 (9.9%) 64 (8.1%) 99 (12%) Intermediate/liberal profession 71 (4.3%) 31 (3.9%) 40 (4.7%) Merchant, craftsman, or independent service provider 170 (10%) 73 (9.2%) 97 (11%) Public sector employee 331 (20%) 169 (21%) 162 (19%) Worker / Other 142 (8.6%) 64 (8.1%) 78 (9.1%) Education level 0.15 Can read, write, count 300 (18%) 141 (18%) 159 (19%) Higher education 136 (8.3%) 75 (9.4%) 61 (7.1%) Middle school diploma 241 (15%) 106 (13%) 135 (16%) Primary school certificate 617 (37%) 289 (36%) 328 (38%) Secondary education 354 (21%) 183 (23%) 171 (20%) Family situation 0.27 Divorced / Free union / Unknown 129 (7.8%) 63 (7.9%) 66 (7.7%) Married 700 (42%) 356 (45%) 344 (40%) Single 68 (4.1%) 32 (4.0%) 36 (4.2%) Widowed 751 (46%) 343 (43%) 408 (48%) Lifestyle 0.078 Alone 848 (51%) 398 (50%) 450 (53%) With a partner 685 (42%) 349 (44%) 336 (39%) With family members 115 (7.0%) 47 (5.9%) 68 (8.0%) Number of children 2 ± 2 2 ± 2 2 ± 2 0.30 Housing (Owner) 991 (60%) 482 (61%) 509 (60%) 0.65 Elevator 389 (24%) 173 (22%) 216 (25%) 0.094 Long-term illness 1,209 (73%) 565 (71%) 644 (75%) 0.051 Health insurance 1,573 (95%) 751 (95%) 822 (96%) 0.10 Leisure 1,377 (84%) 689 (87%) 688 (81%) < 0.001 Social activity 288 (17%) 162 (20%) 126 (15%) 0.003 Human assistance 1,402 (85%) 644 (81%) 758 (89%) < 0.001 Pet 410 (25%) 204 (26%) 206 (24%) 0.46 ADL, mean ± SD 5 ± 1 5 ± 1 5 ± 1 < 0.001 IADL, mean ± SD 6 ± 2 6 ± 2 5 ± 2 < 0.001 MMSE, mean ± SD 23 ± 7 24 ± 7 23 ± 7 0.006 Pathological CDT 585 (35%) 244 (31%) 341 (40%) < 0.001 Pathological verbal fluency 672 (41%) 269 (34%) 403 (47%) < 0.001 MNA, mean ± SD 24 ± 4 24 ± 4 23 ± 4 < 0.001 SPPB, mean ± SD 7 ± 4 7 ± 4 6 ± 4 < 0.001 Pathological GDS 449 (27%) 176 (22%) 273 (32%) < 0.001 Pathological SLB 708 (43%) 261 (33%) 447 (52%) < 0.001 * Pearson's Chi-squared test; Wilcoxon rank sum test. Statistically significance (p-value < .05). SD, Standard deviation; SLB, Single leg balance; CDT, Clock-drawing test; ADL, Activities of Daily Living; IADL, Instrumental Activities of Daily Living; MMSE, Mini-Mental State Examination; MNA, Mini Nutritional Assessment; SPPB, Short Physical Performance Battery; GDS, Geriatric Depression Scale. Data are shown as the number (percentage) or mean ± SD unless otherwise indicated. Figure S1 shows that 40% of the study population experienced a single fall, 14% had recurrent falls (at least twice a year) and 46% did not fall during the study period. The distribution of falls in our study aligns with findings from the literature, which indicate that among individuals aged 80 years and older, one in two experiences a fall. 29 Based on these different fall categories, we analyze how each feature is distributed between fallers and non-fallers. Fallers include individuals who experience recurrent falls as well as those who fall only once. 3.2. Features of the study according to the fallers and non-fallers Table 1 presents a comparison of study characteristics between patients who have fallen and those who have not. Within the faller group, several features exhibited significant differences (p < 0.05) compared to non-fallers: on average: Age, ADL, IADL, MMSE, MNA and SPPB. by proportion: Gender, Diabetes, Leisure, Social activity, Human assistance, Pathological CDT, Pathological verbal fluency, Pathological GDS and Pathological SLB. After highlighting significant differences between fallers and non-fallers, we will group patients with similar characteristics into clusters. Each cluster will be represented in fall trajectories by plotting the mean number of falls per month using the GMM. 3.3. Determining Falls Trajectories over 7 years To identify participants’ fall trajectories, we applied clustering methods that group individuals with similar profiles within the study population. One such method is the GMM, which we implemented using Python’s Scikit-learn library. 30 GMM is a probabilistic clustering approach that models the data as a mixture of multiple Gaussian distributions. Unlike K-means clustering, which assigns each data point to a single cluster, GMM assigns probabilities to multiple clusters, enabling more flexible classifications. The model is trained using the Expectation-Maximization (EM) algorithm, which iteratively estimates cluster probabilities and updates distribution parameters to maximize likelihood. This process improves the accuracy of trajectory classification. 31 Figure 1 depicts the mean number of falls over time (in months) for a model of four different groups performed well, each of the trajectories is constituted by at least five percent of the population study according to the Nagin criteria. 32 The distinct trajectories were Cluster Falls and No Falls (n = 1079, 65.5%), Increasing Falls (n = 111, 6.7%), Chronic Recurring Falls (n = 258, 15.7%) and Low-Rate Chronic Falls (n = 200, 12.1%). Among all the features in our study, we will analyze their distribution within each trajectory. This will allow us to observe how specific characteristics vary across different fall patterns and identify potential associations with fall risk. 3.4. Features of the study according to each trajectory Table S1 demonstrates that the distribution of several characteristics differs significantly across the four fall-trajectory groups (p < 0.05): Gender, Age, Obesity, Long-term illness, Leisure, Social activity, IADL, MMSE, Pathological verbal fluency, MNA, SPPB and Pathological SLB. In sections 3.5 and 3.6, we will identify the strongest predictors for each trajectory using MLR. When examining all trajectories across the entire study period, Clustered Falls and No Falls emerges as the steadiest trajectory. At baseline, the trajectory associated with no falls was Increasing Falls, whereas at the 18-month follow-up, it was Clustered Falls and No Falls (see Fig. 1 ). We investigated predictive factors at these two key time points baseline and 18 months when a trajectory reached a mean of zero falls. 3.5. Trajectories at baseline 3.5.1. Predictors of Falls at baseline MLR revealed that specific baseline characteristics predicted membership within each of the three fallers trajectory groups as compared to the Increasing Falls trajectory group are shown in Fig. 2 (see Table S2 ) . Predictors of Clustered Falls and No Falls were Leisure (No), IADL < 8 and pathological SLB. Predictors of Chronic Recurring Falls trajectory were woman and obesity. Predictors of Low-Rate Chronic Falls trajectory were woman, obesity, IADL < 8 and pathological SLB. We reordered the features using marginal predictions and conducted a new MLR to evaluate whether more or fewer predictors would be identified. 3.5.2. Marginal predictions at baseline We observe that, out of all the features in our study (see 3.5), only some remained after the MLR analysis (see Table S2 or Fig. 2 ). However, not all features in Fig. 2 served as predictors across different trajectories. In MLR, the order of categories within each feature is crucial. To establish this order, we set the steadiest trajectory, Increasing Falls as the reference. Then, for each feature, we identify the category that contributes the most, on average, to this steadiest trajectory and designate it as the reference within its feature. This process is carried out using marginal predictions. According to table S3 (see Fig. S2), the probability of being a non-faller is highest on categories man, age under 80 years old, obesity (No), hypertension (Yes), leisure (Yes), social activity (No), IADL = 8, MMSE ≥ 24, Pathological verbal fluency (No), Pathological SLB (No). We conducted a new MLR using the same features (see Fig. 2 ) but redefined the reference category within each feature based on marginal predictions (see 3.5.2). 3.5.3. Predictors of Falls Trajectories after Rearrangement of Categories in each Feature at baseline Similar to the MLR performed in section 3.5.1, we identified specific baseline characteristics that predicted membership within each of the three faller trajectories compared to the Increasing Falls group. These results are presented in table S4 (see Fig. S3). We observe that the same predictors from table S2 (see Fig. 2 ) reappear in the results of the new MLR, along with additional predictors only in Chronic Recurring Falls trajectory namely hypertension (No). 3.6. Trajectories at 18 months 3.6.1. Predictors of Falls at 18 months MLR revealed that specific baseline characteristics predicted membership within each of the three fallers trajectory groups as compared to the Clustered Falls and No Falls trajectory group are shown in Fig. 3 (see Table S5). Predictors of Increasing Falls were SPPB (between 6 and 9), SPPB < 5 and pathological GDS. Predictors of Chronic Recurring Falls trajectory were woman, ADL < 4, SPPB < 5 and pathological GDS. Predictors of Low-Rate Chronic Falls trajectory were obesity, SPPB < 5 and pathological GDS. We reordered the features using marginal predictions and conducted a new MLR to evaluate whether more or fewer predictors would be identified. 3.6.2. Marginal predictions at 18 months We apply the same procedure as 3.5.2. To establish this order, we set the steadiest trajectory, Clustered Falls and No Falls as the reference. According to table S6 (see Fig. S4), the probability of being a non-faller is highest on categories, man, hypertension (Yes), diabetes (No), dyslipidemia (No), obesity (No), ADL (between 4 and 6), MMSE < 24, SPPB (between 10 and 12), Pathological GDS (No), Pathological SLB (Yes). We conducted a new MLR using the same features (see Fig. S3) but redefined the reference category within each feature based on marginal predictions (see 3.6.2). 3.6.3. Predictors of Falls Trajectories after Rearrangement of Categories in each Feature at 18 months Similar to the MLR performed in section 3.6.1, we identified specific baseline characteristics that predicted membership within each of the three faller trajectories compared to the Clustered Falls and No Falls group. These results are presented in table S7 (see Fig. S5). We observe that the same predictors from table S5 reappear in the results of the new MLR, along with additional predictors for each trajectory: Increasing Falls: Pathological SLB (No). Chronic Recurring Falls: Hypertension (No) and Pathological SLB (No). Low-Rate Chronic Falls: Hypertension (No), Dyslipidemia (No) and MMSE ≥ 24. 4. Discussion In this study, we explore the different trajectories of falls among home-dwelling patients over a seven-year period. We found four clinically distinct trajectories of falls, which we have labeled: Cluster Falls and No Falls, Increasing Falls, Chronic Recurring Falls and Low-Rate Chronic Falls. Cluster Falls and No Falls show minimal risk after an initial decrease. Increasing Falls have a delayed peak (around 30 months) before a rapid decline, suggesting episodic high-risk periods rather than ongoing vulnerability. Chronic Recurring Falls experience a sharp peak and drop early on, indicating a temporary high fall risk. Low-Rate Chronic Falls are the most persistent group, showing a continuous risk of falls throughout the study period. Table S1 provides a clear distinction between fall trajectories, showing that Chronic Recurring Falls tend to be frailer and less socially engaged; while Increasing and Low-Rate Chronic Falls maintain more mobility and social interaction but still experience frequent falls. The configuration of our trajectories prompted us to investigate predictive factors both at baseline and after 18 months. The factors identified at these two time points were female sex and obesity. This finding underscores the importance of assessing predictive factors not only at baseline but also during follow-up. Previous studies have similarly reported that female sex and obesity are consistent predictors at both key stages. Furthermore, in the recent study by Nicklett et al. (2017), female sex was identified as a risk factor for fall events. 33 This finding is consistent with previous studies by Deandrea et al. (2010), Shumway-Cook et al. (2009), Stahl and Albert (2015), and de Rekeneire et al. (2003), which have also confirmed this association. 34 – 37 A recent study of Abdo et al. (2022) find obesity as one of the major predictive factors for falls, that align with the study of Erlandson et al. (2019) and Máximo et al. (2019). Among the additional predictors identified at baseline were the absence of leisure activities, pathological SLB, and an IADL score below 8. SLB and leisure are both related to physical activity, and leisure in particular may influence the psychological (thymic) dimension of a patient’s health. SLB and leisure have been identified as predictive factors for falls in recent studies by Blodgett et al. (2022) and Kwok et al. (2024), respectively. 38 , 39 In the study by Kwok et al. (2024), the population consisted exclusively of women, which highlights the continuing importance of exploring the relationship between female sex and fall risk. The last predictive factor identified at baseline, namely IADL, is related to patient autonomy and can provide insight into mental health status. The study by Brown et al. (2014) used this variable as a standalone tool for screening fall risk when stratified by stage. 40 This article considered IADL to be one of the strongest predictive factors for falls. Furthermore, Britton et al. (2019) confirmed through their study that IADL is not only a predictor of falls but also a predictor of recurrent falls. 41 Recurrent falls reflect a lack or complete absence of effective prevention strategies, which can lead to more severe consequences. While many studies have focused on preventing the first fall, others have sought to identify strategies to prevent subsequent or recurrent falls. 33 , 41 – 46 As is well known, the ADL identified at 18 months in our study also serves to assess patient autonomy, similarly to the IADL. In the study by Henry et al. (2012), it was demonstrated that individuals with intermediate levels of ADL limitation who remain partially active have a higher risk of falling. 47 The identification of SLB confirms that falls cannot be dissociated from balance, while the SPPB identified at 18 months provides complementary information, indicating that falls are also closely associated with walking ability (gait). Several studies have demonstrated that physical frailty must be taken into account in fall prevention strategies. Recent studies by Choi et al. (2023), Ge et al. (2023), and Abdo et al. (2022) have confirmed that the SPPB is a predictive factor for falls, as previously demonstrated in the studies by Erlandson et al. (2019), Souza et al. (2019), and Moreland et al. (2018). 42 , 43 , 46 , 48 – 50 One of the recurrent predictive factors is related to the cognitive dimension of patient health, namely the GDS. Several studies have included this variable in their assessment of fall risk and have identified it as a significant predictive factor. Initially identified in the studies by Nicklett et al. (2017), Britton et al. (2019), and Erlandson et al. (2019), the GDS has since been confirmed as one of the strongest predictive factors for falls in more recent research by Jo et al. (2020), Fallaci et al. (2022), Ge et al. (2023), and Pesonen et al. (2025). 33 , 41 , 43 – 45 , 48 , 51 In particular, Pesonen et al. (2025) examined the combined role of impaired vision and cognitive decline in falls and fall-related risk factors. Moreover, the relationship between depression and physical frailty was demonstrated by Chang et al. (2011). 52 All the variables identified as predictive factors were also reported in the literature. All these factors are related to patients’ health impairments. In our MLR results, we observed that other features were present but did not emerge as significant predictors of falls in any of the trajectories. The use of marginal predictions to assess the contribution of each category within each feature allowed us to redefine the reference category within each feature before performing a new MLR. This MLR led us to find more predictors, namely, hypertension (No) at baseline and pathological SLB (No), hypertension (No) and pathological SLB (No), dyslipidemia (No) and MMSE ≥ 24 at 18 months. This result confirms that people who fall are not always those with multiple comorbidities. This aligns with the unintentional and therefore unpredictable nature of falls in older adults. 53 Marginal prediction analysis revealed that being male or having a normal weight status increased the likelihood of belonging to the most stable fall trajectory. This result confirms the impact of sex, with female gender remaining a significant predictive factor for falls. The study by Renner et al. (2021), in which the entire study population consisted of older men, demonstrated that higher levels of fatigue can increase their risk of falling. 54 These findings suggest that it would be valuable to conduct separate analyses within each dataset: one including both sexes and one for each sex individually. Such an approach could reveal sex-specific factors that should be taken into account when developing fall prevention strategies. In our study, age was not identified as a predictive factor; however, the difference in mean age between fallers and non-fallers was significant (Table 1 ). In contrast, the studies by Hollinghurst et al. (2022), Mata et al. (2017) and Nicklett et al. (2017) demonstrated that age is a predictive factor for falls. 9 , 10 , 33 It may appear that individuals who move less experience fewer falls due to their health condition. However, while reducing movement might seem like a strategy to prevent falls, it is not an effective solution. Older individuals who limit their mobility enter a cycle of progressive loss of autonomy. In fact, individuals who have already experienced a fall may develop a fear of falling, leading to sedentary behavior and reduced physical activity, further exacerbating frailty and loss of independence. Across the different trajectories, two of the most consistent predictors were pathological SLB and impaired physical performance, underscoring the critical role of balance and mobility in fall risk. We observe in our study that a higher proportion of individuals in the faller group require human assistance (see 3.2). In each trajectory, at least one patient out of two lived alone, which could contribute to higher fall risk and its consequences (see Table S1 ). 55 This finding highlights the crucial role of caregivers in fall prevention and post-fall assistance, as individuals who fall may remain on the ground for an extended period, a phenomenon known as long lie. 56 In such situations, the lack of assistance can lead to serious complications, including death. The study by Leggett et al. (2018) show that caregivers’ greater nighttime awakenings were associated with caring for care recipients with higher fall risk. 57 In hospital settings, the thymic (mood-related) and cognitive dimensions of falls are often not a primary focus during patient evaluations. However, for older adults living at home, teams like the UPSAV unit conduct CGA, enabling early detection of depressive symptoms or cognitive decline. Previous studies have emphasized the importance of considering these factors when designing effective fall prevention strategies. 58 , 59 In summary, all the variables identified in our study as predictive factors for falls have also been reported in the literature published between 2019 and 2025. Our strategy of examining fall predictors among trajectories with a mean number of falls equal to zero (both at baseline and at 18 months) allowed us to identify additional predictors, including one of the most significant, GDS, which reflects the depressive state of the patient. Moreover, the use of marginal predictions highlighted that a patient may still experience falls despite having good scores on certain tests, such as the MMSE or SLB, or in the absence of cardiovascular issues. This may be explained by the interventions provided by UPSAV between the two visits. The quality of these interventions may have improved several dimensions of patients’ health yet not prevented the occurrence of falls due to other fall-related characteristics, precisely those identified in our study and consistently described in the literature. Our study has several limitations. (1) We used baseline clinical characteristics obtained between visits, which limited our ability to identify risk factors at the exact time of the falls. Consequently, we could not determine whether acute illnesses, medication use, or other transient factors influenced specific fall trajectories or whether certain patterns reflected the progression of chronic diseases. (2) Some individuals classified as Cluster Falls and No Falls may have had prior fall episodes that were not accounted for, resulting in left-censored data in our analysis. This limitation could have influenced the classification of fall trajectories. Additionally, implementing a Randomized Controlled Trial (RCT) design could enhance the evidence for intervention effectiveness and help establish causal relationships between fall prevention strategies and outcomes. (3) The study population was restricted to older adults residing in France, which may limit the generalizability of our findings to other geographic or cultural contexts. Future research should aim to include more diverse populations to enhance the applicability of the results. 5. Conclusion In summary, our study identifies four clinically distinct fall trajectories over seven years among community-dwelling older adults, each associated with a set of unique risk factors. Low-Rate Chronic Falls exhibit a persistent fall risk, while Increasing Falls experience episodic high-risk periods. Interestingly, age category did not emerge as a significant predictor for any of the identified fall trajectories. Key predictors across trajectories include being female, obesity, impairment in activities of daily living (ADL/IADL), reduced physical performance and depressive symptoms all of which should be prioritized in fall prevention interventions. Additionally, older individuals who live alone are also at risk of falling. Given the significant impact of recurrent falls on health outcomes, targeted interventions should prioritize those at the highest risk to prevent further decline and injury. Declarations Acknowledgments: The authors thank the Unit for Prevention, Monitoring and Analysis of Ageing (UPSAV – Unité de Prévention, de Suivi et d'Analyse du Vieillissement ) and the Clinical Research and Innovation Unit in Gerontology (URCI – Unité de Recherche Clinique et innovation en Gérontologie ). Sponsor's role: Data grant by “CARSAT Centre-Ouest”, France were collected by UPSAV and URCI. The analysis and manuscript preparation were funded by a grant from chair of excellence in Artificial Intelligence and Healthy Ageing in the Nouvelle-Aquitaine Region (IABV-NA – Intelligence Artificielle et Bien-Vieillir en Nouvelle-Aquitaine ). Conflict-of-interest statement: No conflicts of interest Funding sources : FUNDING INFORMATION This research was funded by a grant from chair of excellence in Artificial Intelligence and Healthy Ageing in the Nouvelle-Aquitaine Region (IABV-NA – Intelligence Artificielle et Bien-Vieillir en Nouvelle-Aquitaine ) and “CARSAT Centre-Ouest” . References WHO. Ageing and health. Accessed March 28, 2023. (2022). https://www.who.int/fr/news-room/fact-sheets/detail/ageing-and-health Insee Population by age - Tableaux de l’Économie Française | Insee. Accessed March 28, 2023. (2016). https://www.insee.fr/fr/statistiques/1906664?sommaire=1906743 Moylan, K. C. & Binder, E. F. Falls in older adults: risk assessment, management and prevention. Am. J. Med. 120 (6), 493e1–493e6. 10.1016/j.amjmed.2006.07.022 (2007). Thierry, X. Accidents and physical assaults among the elderly: less frequent than among the young, but more severe. Popul. Sociétés . 468 (6), 1–4. 10.3917/popsoc.468.0001 (2010). Health.gouv. Accessed March 29. (2023). https://sante.gouv.fr/IMG/pdf/dp_plan-antichute-accessible28-02-2022.pdf Park, S. H. Tools for assessing fall risk in the elderly: a systematic review and meta-analysis. Aging Clin. Exp. Res. 30 (1), 1–16. 10.1007/s40520-017-0749-0 (2018). Lim, C. et al. Associations Between Multiple Falls and Mental Health Indices in Korean Older Adults: A Cross-sectional Study. J. Prev. Med. Public. Health . 58 (3), 269–277. 10.3961/jpmph.24.482 (2025). May, P., Roe, L., McGarrigle, C. A., Kenny, R. A. & Normand, C. End-of-life experience for older adults in Ireland: results from the Irish longitudinal study on ageing (TILDA). BMC Health Serv. Res. 20 (1), 118. 10.1186/s12913-020-4978-0 (2020). da Mata, L. R. F., Azevedo, C., Policarpo, A. G. & Moraes, J. T. Factors associated with the risk of fall in adults in the postoperative period: a cross-sectional study. Rev. Lat Am. Enfermagem . 25 , e2904. 10.1590/1518-8345.1775.2904 (2017). Hollinghurst, R. et al. Annual risk of falls resulting in emergency department and hospital attendances for older people: an observational study of 781,081 individuals living in Wales (United Kingdom) including deprivation, frailty and dementia diagnoses between 2010 and 2020. Age Ageing . 51 (8), afac176. 10.1093/ageing/afac176 (2022). Jayakody, O., Blumen, H. M., Ayers, E. & Verghese, J. Longitudinal Associations Between Falls and Risk of Gait Decline: Results From the Central Control of Mobility and Aging Study. Arch. Phys. Med. Rehabil . 104 (2), 245–250. 10.1016/j.apmr.2022.08.975 (2023). Ellingsen, C. L., Ebbing, M., Alfsen, G. C. & Vollset, S. E. Injury death certificates without specification of the circumstances leading to the fatal injury - the Norwegian Cause of Death Registry 2005–2014. Popul. Health Metr. 16 (1), 20. 10.1186/s12963-018-0176-2 (2018). Tchalla, A. et al. Patterns, predictors, and outcomes of frailty trajectories in community-dwelling older adults: Results from the FREEDOM Cohort Study. Arch. Gerontol. Geriatr. 114 , 105101. 10.1016/j.archger.2023.105101 (2023). Kellogg The prevention of falls in later life. A report of the Kellogg International Work Group on the Prevention of Falls by the Elderly. Dan. Med. Bull. 34 (Suppl 4), 1–24 (1987). Welsh, T. J., Gordon, A. L. & Gladman, J. R. Comprehensive geriatric assessment – a guide for the non-specialist. Int. J. Clin. Pract. 68 (3), 290–293. 10.1111/ijcp.12313 (2014). Benton, A. L. Development of a multilingual aphasia battery: Progress and problems. J. Neurol. Sci. 9 (1), 39–48. 10.1016/0022-510X(69)90057-4 (1969). Trojian, T. H. & McKeag, D. B. Single leg balance test to identify risk of ankle sprains. Published online July . 1 10.1136/bjsm.2005.024356 (2006). Freedman, M. et al. and D. Delis. Clock drawing: a neuropsychological analysis. New York: Oxford, 1994. | Request PDF. ResearchGate . Published online 1994. 10.1017/S0714980800013398 Katz, S., Ford, A. B., Moskowitz, R. W., Jackson, B. A. & Jaffe, M. W. Studies of Illness in the Aged: The Index of ADL: A Standardized Measure of Biological and Psychosocial Function. JAMA 185 (12), 914–919. 10.1001/jama.1963.03060120024016 (1963). Lawton, M. P. & Brody, E. M. Assessment of Older People: Self-Maintaining and Instrumental Activities of Daily Living1. Gerontologist 9 (3_Part_1), 179–186. 10.1093/geront/9.3_Part_1.179 (1969). Escobar, J. I. et al. Use of the Mini-Mental State Examination (MMSE) in a community population of mixed ethnicity. Cultural and linguistic artifacts. J. Nerv. Ment Dis. 174 (10), 607–614. 10.1097/00005053-198610000-00005 (1986). Vellas, B. et al. The Mini Nutritional Assessment (MNA) and its use in grading the nutritional state of elderly patients. Nutrition 15 (2), 116–122. 10.1016/s0899-9007(98)00171-3 (1999). Guralnik, J. M. et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J. Gerontol. 49 (2), M85–94. 10.1093/geronj/49.2.m85 (1994). Yesavage, J. A. et al. Development and validation of a geriatric depression screening scale: A preliminary report. J. Psychiatr. Res. 17 (1), 37–49. 10.1016/0022-3956(82)90033-4 (1982). Yang, M. S., Lai, C. Y. & Lin, C. Y. A robust EM clustering algorithm for Gaussian mixture models. Pattern Recogn. 45 (11), 3950–3961. 10.1016/j.patcog.2012.04.031 (2012). Nagin, D. S. & Odgers, C. L. Group-based trajectory modeling in clinical research. Annu. Rev. Clin. Psychol. 6 , 109–138. 10.1146/annurev.clinpsy.121208.131413 (2010). Böhning, D. Multinomial logistic regression algorithm. Ann. Inst. Stat. Math. 44 (1), 197–200. 10.1007/BF00048682 (1992). Arel-Bundock, V., Greifer, N. & Heiss, A. How to Interpret Statistical Models Using marginaleffects for R and Python. J. Stat. Softw. 111 , 1–32. 10.18637/jss.v111.i09 (2024). Grundstrom, A. C., Guse, C. E. & Layde, P. M. Risk Factors for Falls and Fall-Related Injuries in Adults 85 Years of Age and Older. Arch. Gerontol. Geriatr. 54 (3), 421–428. 10.1016/j.archger.2011.06.008 (2012). Scikit-learn developers. GaussianMixture. scikit-learn. Accessed February 23. (2025). https://scikit-learn.org/stable/modules/mixture.html Moon, T. K. The expectation-maximization algorithm. IEEE. Signal. Process. Mag. 13 (6), 47–60. 10.1109/79.543975 (1996). Nagin, D. S. Group-Based Trajectory Modeling: An Overview. ANM 65 (2–3), 205–210. 10.1159/000360229 (2014). Nicklett, E. J., Taylor, R. J., Rostant, O., Johnson, K. E. & Evans, L. Biopsychosocial Predictors of Fall Events Among Older African Americans. Res. Aging . 39 (4), 501–525. 10.1177/0164027516651974 (2017). Deandrea, S. et al. Risk factors for falls in community-dwelling older people: a systematic review and meta-analysis. Epidemiology 21 (5), 658–668. 10.1097/EDE.0b013e3181e89905 (2010). Shumway-Cook, A. et al. Falls in the Medicare population: incidence, associated factors, and impact on health care. Phys. Ther. 89 (4), 324–332. 10.2522/ptj.20070107 (2009). Stahl, S. T. & Albert, S. M. Gender differences in physical activity patterns among older adults who fall. Prev. Med. 71 , 94–100. 10.1016/j.ypmed.2014.12.016 (2015). De Rekeneire, N. et al. Is a fall just a fall: correlates of falling in healthy older persons. The Health, Aging and Body Composition Study. J. Am. Geriatr. Soc. 51 (6), 841–846. 10.1046/j.1365-2389.2003.51267.x (2003). Blodgett, J. M. et al. One-Legged Balance Performance and Fall Risk in Mid and Later Life: Longitudinal Evidence From a British Birth Cohort. Am. J. Prev. Med. 63 (6), 997–1006. 10.1016/j.amepre.2022.07.002 (2022). Kwok, W. S. et al. Leisure-Time Physical Activity and Falls With and Without Injuries Among Older Adult Women. JAMA Netw. Open. 7 (1), e2354036. 10.1001/jamanetworkopen.2023.54036 (2024). Brown, J., Kurichi, J. E., Xie, D., Pan, Q. & Stineman, M. G. Instrumental activities of daily living staging as a possible clinical tool for falls risk assessment in physical medicine and rehabilitation. PM R . 6 (4), 316–323. 10.1016/j.pmrj.2013.10.007 (2014). quiz 323. Britton, G. B. et al. Inflammatory Biomarkers, Depressive Symptoms and Falls Among the elderly in Panama. Curr. Aging Sci. 11 (4), 236–241. 10.2174/1874609812666190215125104 (2019). Choi, N. G., Marti, C. N., Choi, B. Y. & Kunik, M. M. Recurrent Falls over Three Years among Older Adults Age 70+: Associations with Physical and Mental Health Status, Exercise, and Hospital Stay. J. Appl. Gerontol. 42 (5), 1089–1100. 10.1177/07334648221150884 (2023). Erlandson, K. M. et al. Risk Factors for Falls, Falls With Injury, and Falls With Fracture Among Older Men With or at Risk of HIV Infection. J. Acquir. Immune Defic. Syndr. 81 (4), e117–e126. 10.1097/QAI.0000000000002074 (2019). Fallaci, I. V., Fabrício, D., de Alexandre, M. & da Chagas, T. Association between falls and cognitive performance among community-dwelling older people: a cross-sectional study. Sao Paulo Med. J. 140 (3), 422–429. 10.1590/1516-3180.2021.0180.R1.15092021 (2022). Jo, K. H., Park, J. & Ryu, S. Y. The effects of mental health on recurrent falls among elderly adults, based on Korean Community Health Survey data. Epidemiol. Health . 42 , e2020005. 10.4178/epih.e2020005 (2020). de Souza, A. Q., Pegorari, M. S., Nascimento, J. S., de Oliveira, P. B. & Tavares, D. M. D. S. Incidence and predictive factors of falls in community-dwelling elderly: a longitudinal study. Cien Saude Colet . 24 (9), 3507–3516. 10.1590/1413-81232018249.30512017 (2019). Henry-Sánchez, J. T., Kurichi, J. E., Xie, D., Pan, Q. & Stineman, M. G. Do elderly people at more severe activity of daily living limitation stages fall more? Am. J. Phys. Med. Rehabil . 91 (7), 601–610. 10.1097/PHM.0b013e31825596af (2012). Ge, M. L., Chu, N. M., Simonsick, E. M., Kasper, J. D. & Xue, Q. L. Order of Onset of Physical Frailty and Cognitive Impairment and Risk of Repeated Falls in Community-Dwelling Older Adults. J. Am. Med. Dir. Assoc. 24 (4), 482–488e4. 10.1016/j.jamda.2023.01.020 (2023). Abdo, M. et al. Regional Differences in Risk of Recurrent Falls Among Older U.S. Women and Men with HIV in the HIV Infection, Aging, and Immune Function Long-Term Observational Study. AIDS Res. Hum. Retroviruses . 38 (7), 530–537. 10.1089/AID.2021.0162 (2022). Moreland, B. L., Durbin, L. L., Kasper, J. D. & Mielenz, T. J. Rehabilitation Utilization for Falls Among Community-Dwelling Older Adults in the United States in the National Health and Aging Trends Study. Arch. Phys. Med. Rehabil . 99 (8), 1568–1575. 10.1016/j.apmr.2018.02.009 (2018). Pesonen, T., Siira, H., Väisänen, V., Edgren, J. & Elo, S. The role of impaired vision and declined cognition in falls and fall-related risk factors among older people receiving home care in Finland-a cross-sectional register study. Eur. J. Ageing . 22 (1), 22. 10.1007/s10433-025-00860-2 (2025). Chang, C. I., Chan, D. C., Kuo, K. N., Hsiung, C. A. & Chen, C. Y. Prevalence and correlates of geriatric frailty in a northern Taiwan community. J. Formos. Med. Assoc. 110 (4), 247–257. 10.1016/S0929-6646(11)60037-5 (2011). Ambrose, A. F., Paul, G. & Hausdorff, J. M. Risk factors for falls among older adults: A review of the literature. Maturitas 75 (1), 51–61. 10.1016/j.maturitas.2013.02.009 (2013). Renner, S. W. et al. Higher Fatigue Prospectively Increases the Risk of Falls in Older Men. Innov. Aging . 5 (1), igaa061. 10.1093/geroni/igaa061 (2021). Choi, N. G., Choi, B. Y., DiNitto, D. M., Marti, C. N. & Kunik, M. E. Fall-related emergency department visits and hospitalizations among community-dwelling older adults: examination of health problems and injury characteristics. BMC Geriatr. 19 (1), 303. 10.1186/s12877-019-1329-2 (2019). Thakur, N. & Han, C. Y. A Study of Fall Detection in Assisted Living: Identifying and Improving the Optimal Machine Learning Method. J. Sens. Actuator Networks . 10 (3), 39. 10.3390/jsan10030039 (2021). Leggett, A. N., Polenick, C. A., Maust, D. T. & Kales, H. C. Falls and Hospitalizations Among Persons With Dementia and Associated Caregiver Emotional Difficulties. Gerontologist 58 (2), e78–e86. 10.1093/geront/gnx202 (2018). Gambaro, E. et al. The complex associations between late life depression, fear of falling and risk of falls. A systematic review and meta-analysis. Ageing Res. Rev. 73 , 101532. 10.1016/j.arr.2021.101532 (2022). Tchalla, A. E. et al. Patterns, Predictors, and Outcomes of Falls Trajectories in Older Adults: The MOBILIZE Boston Study with 5 Years of Follow-Up. PLOS ONE . 9 (9), e106363. 10.1371/journal.pone.0106363 (2014). Additional Declarations No competing interests reported. Supplementary Files SUPPLEMENTARYMATERIAL.doc Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 18 May, 2026 Reviews received at journal 09 Feb, 2026 Reviewers agreed at journal 19 Jan, 2026 Reviewers agreed at journal 14 Dec, 2025 Reviewers invited by journal 08 Dec, 2025 Editor assigned by journal 08 Dec, 2025 Editor invited by journal 24 Nov, 2025 Submission checks completed at journal 21 Nov, 2025 First submitted to journal 21 Nov, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8132697","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":558070330,"identity":"75a729c7-616b-4add-bf96-09d5bd971d3a","order_by":0,"name":"Amadou M. Djiogomaye Ndiaye","email":"data:image/png;base64,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","orcid":"","institution":"Laboratoire Vie-Santé, UR 24 134, Faculté de Médecine","correspondingAuthor":true,"prefix":"","firstName":"Amadou","middleName":"M. Djiogomaye","lastName":"Ndiaye","suffix":""},{"id":558070331,"identity":"14e9b293-be43-45df-a686-2f3fe89ecd14","order_by":1,"name":"Michel Harel","email":"","orcid":"","institution":"Laboratoire Vie-Santé, UR 24 134, Faculté de Médecine","correspondingAuthor":false,"prefix":"","firstName":"Michel","middleName":"","lastName":"Harel","suffix":""},{"id":558070332,"identity":"ba6aceee-4341-4e20-882a-85a33b5663c2","order_by":2,"name":"Laurent Billonnet","email":"","orcid":"","institution":"XLIM-SRI - Systèmes et Réseaux Intelligents","correspondingAuthor":false,"prefix":"","firstName":"Laurent","middleName":"","lastName":"Billonnet","suffix":""},{"id":558070337,"identity":"0bd2ad07-47e4-4f16-b048-0407be657cff","order_by":3,"name":"Achille Tchalla","email":"","orcid":"","institution":"Laboratoire Vie-Santé, UR 24 134, Faculté de Médecine","correspondingAuthor":false,"prefix":"","firstName":"Achille","middleName":"","lastName":"Tchalla","suffix":""}],"badges":[],"createdAt":"2025-11-17 08:23:56","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8132697/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8132697/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":97990547,"identity":"7e749882-bb2d-4840-9dbb-8c8299445e9c","added_by":"auto","created_at":"2025-12-11 14:30:07","extension":"doc","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1326080,"visible":true,"origin":"","legend":"","description":"","filename":"ArticleFallsTrajectoriesBMC.doc","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/5ceb707471f5765c0fe78dd1.doc"},{"id":98423618,"identity":"b1542d43-7e4c-41d0-8475-9abb82f28af7","added_by":"auto","created_at":"2025-12-17 16:32:26","extension":"json","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8224,"visible":true,"origin":"","legend":"","description":"","filename":"9443e2c5f4474a80bb7d702e7f59d053.json","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/60f4fe779cca2081036a4624.json"},{"id":98425484,"identity":"2d27899f-66bd-4199-8208-98de9292b962","added_by":"auto","created_at":"2025-12-17 16:34:50","extension":"doc","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1219072,"visible":true,"origin":"","legend":"","description":"","filename":"SUPPLEMENTARYMATERIAL.doc","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/fe065b3949a31bd4b63d688f.doc"},{"id":97990555,"identity":"7963f517-2555-457a-883a-d7c1e387ddcf","added_by":"auto","created_at":"2025-12-11 14:30:08","extension":"xml","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":178755,"visible":true,"origin":"","legend":"","description":"","filename":"9443e2c5f4474a80bb7d702e7f59d0531enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/d8f02d3a46f98b137c4f7bf6.xml"},{"id":98424167,"identity":"618aa83f-ef0a-4b2d-9f85-e7fc8ac93cd0","added_by":"auto","created_at":"2025-12-17 16:33:01","extension":"pdf","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":94403,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/5f64ebeed2f52c515a331c7d.pdf"},{"id":97990552,"identity":"d6f6dcf2-c346-486d-932f-0c83dc55b329","added_by":"auto","created_at":"2025-12-11 14:30:08","extension":"pdf","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":332247,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/540d000a4757e8fcefc57e8d.pdf"},{"id":97990563,"identity":"b74effed-749b-4036-8cff-aa7f297e695a","added_by":"auto","created_at":"2025-12-11 14:30:08","extension":"pdf","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":320065,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/f0f6fac339926d27c90f854e.pdf"},{"id":98425690,"identity":"007d804d-6df7-4924-a4cc-aa1b8952e24a","added_by":"auto","created_at":"2025-12-17 16:35:05","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34671,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/40a382b64eeb2f05a7cf23bc.png"},{"id":98425034,"identity":"a440e55c-5dd8-4d3f-a15f-c7337b0f0d10","added_by":"auto","created_at":"2025-12-17 16:34:11","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":278246,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/57b49966e0434f4af58370c0.png"},{"id":97990562,"identity":"a5f1eb4d-2ee4-4083-9122-ae8ccd736cb0","added_by":"auto","created_at":"2025-12-11 14:30:08","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":153421,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/c864c646cbd18e93641aba5d.png"},{"id":98425497,"identity":"04cf1932-0a38-4b9c-bb5f-3e38a27f5e23","added_by":"auto","created_at":"2025-12-17 16:34:51","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":290296,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/2ed4923b84e4ed9e0122bff3.png"},{"id":98425040,"identity":"27498d1e-49b3-4f80-9320-e3cd0db90ea8","added_by":"auto","created_at":"2025-12-17 16:34:12","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":145389,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/cfa57c3ee42169ee331145ca.png"},{"id":98424777,"identity":"06f43e20-46f9-4bf9-9f22-997a741b7fe2","added_by":"auto","created_at":"2025-12-17 16:33:49","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13392,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/dfb6e0163cb8b13c1e1a2348.png"},{"id":98424224,"identity":"c88782d7-c01d-465a-89e4-1b04440f065a","added_by":"auto","created_at":"2025-12-17 16:33:05","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":58248,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/adc6f3a061e16bf2d3ea15a9.png"},{"id":97990559,"identity":"c4de7238-9847-4cd3-865b-468670eb7bd1","added_by":"auto","created_at":"2025-12-11 14:30:08","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":26908,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/43e648746c5ca90cdceda093.png"},{"id":98424629,"identity":"288a6857-a79e-4cf4-a261-67ee6a56af69","added_by":"auto","created_at":"2025-12-17 16:33:35","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":58816,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/6c3aba2087d0e4ff4fa7d83c.png"},{"id":98425563,"identity":"3e4fd89a-6981-401e-88b5-3748256c8301","added_by":"auto","created_at":"2025-12-17 16:34:56","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":24628,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/6e2e3013078412bb3bc6efc1.png"},{"id":98423714,"identity":"64b90757-d626-447f-9757-4abe5a5a74cf","added_by":"auto","created_at":"2025-12-17 16:32:33","extension":"xml","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":172989,"visible":true,"origin":"","legend":"","description":"","filename":"9443e2c5f4474a80bb7d702e7f59d0531structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/e49e41b828f83d962ca62698.xml"},{"id":97990565,"identity":"a80aef4e-fa24-4d09-aa46-fa312f2c9a02","added_by":"auto","created_at":"2025-12-11 14:30:08","extension":"html","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":195500,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/ff6152b0b6cca8d7d5fdc0a5.html"},{"id":97990548,"identity":"daa7cc42-68a8-46cf-a2a1-aeef00e1b183","added_by":"auto","created_at":"2025-12-11 14:30:08","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":177976,"visible":true,"origin":"","legend":"\u003cp\u003eillustrates the longitudinal trajectories of mean fall frequency across four distinct fall pattern trajectories over a 78-month period. The Cluster Falls and No Falls trajectory (blue line; 65.5% of participants) experienced a brief period of falls followed by sustained periods without falls. The Increasing Falls trajectory (red line; 6.7%) showed a progressive rise in fall frequency, peaking around 30 months before declining. The Chronic Recurring Falls trajectory (purple line; 12.1%) had consistently high fall rates in the early months, which sharply declined after 24 months. The Low-Rate Chronic Falls trajectory (green line; 15.7%) exhibited a steady but moderate fall rate over time, with a gradual decline after month 30. The y-axis represents the mean number of falls and the x-axis indicates time in months.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/e2d3f3c9de343f166f51ff7f.png"},{"id":97990544,"identity":"f45a7bc1-1694-4644-a970-7702948083d0","added_by":"auto","created_at":"2025-12-11 14:30:07","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":124143,"visible":true,"origin":"","legend":"\u003cp\u003epresents odds ratios (OR) and 95% confidence intervals (CI) for various characteristics measured at baseline associated with three fall trajectories: Cluster Falls and No Falls, Chronic Recurring Falls and Low-Rate Chronic Falls. Dot colors represent different variables and the categories in each of them and dot outlines indicate statistical significance (solid dots: p-value (p)\u003cem\u003e \u003c/em\u003e≤ 0.05; open dots: p \u0026gt; 0.05). The x-axis is presented on a logarithmic scale centered at OR = 1, indicating no effect. An OR greater than 1, represented by a solid dot, indicates that the characteristic is a predictive factor of the respective fall trajectory group.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/cc9040764853c988fa4f2c09.png"},{"id":97990545,"identity":"af0d7239-0177-4644-a05e-fd537498d189","added_by":"auto","created_at":"2025-12-11 14:30:07","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":116991,"visible":true,"origin":"","legend":"\u003cp\u003epresents odds ratios (OR) and 95% confidence intervals (CI) for various characteristics measured at 18 months associated with three fall trajectories: Increasing Falls\u003cem\u003e, \u003c/em\u003eChronic Recurring Falls\u003cem\u003e and \u003c/em\u003eLow-Rate Chronic Falls\u003cem\u003e.\u003c/em\u003e Dot colors represent different variables and their categories, while dot outlines indicate statistical significance (solid dots\u003cem\u003e: \u003c/em\u003ep\u003cem\u003e \u003c/em\u003e≤ 0.05;\u003cem\u003e \u003c/em\u003eopen dots\u003cem\u003e: \u003c/em\u003ep\u003cem\u003e \u003c/em\u003e\u0026gt; 0.05\u003cem\u003e).\u003c/em\u003e The x-axis is presented on a logarithmic scale centered at OR = 1, indicating no effect. An OR greater than 1, represented by a solid dot, indicates that the characteristic is a predictive factor of the respective fall trajectory group.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/d112d60662b139d51c8ea780.png"},{"id":98622079,"identity":"b4a7585a-969d-475c-82a6-786e473a36b4","added_by":"auto","created_at":"2025-12-19 16:44:04","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1713991,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/88d4aecf-0a3c-47ae-a269-3fcd03ac5e03.pdf"},{"id":98425248,"identity":"4391afd7-e39d-440e-ad9a-5875bb6cab9b","added_by":"auto","created_at":"2025-12-17 16:34:33","extension":"doc","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1219072,"visible":true,"origin":"","legend":"","description":"","filename":"SUPPLEMENTARYMATERIAL.doc","url":"https://assets-eu.researchsquare.com/files/rs-8132697/v1/324d9bed9d281e8d67d870e0.doc"}],"financialInterests":"No competing interests reported.","formattedTitle":"Identifying Fall Risk Factors: A 7-Year Home Study Using Innovative Marginal Predictions Strategy","fulltext":[{"header":"Brief summary","content":"\u003cp\u003eThis 7-year home-based study identified four fall trajectories. Baseline predictors included female sex, obesity, low IADL, poor balance, low activity and other factors at 18 months.\u003c/p\u003e"},{"header":"1. Introduction","content":"\u003cp\u003eAccording to the World Health Organization (WHO), older individuals are those aged\u0026thinsp;\u0026ge;\u0026thinsp;60 years.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e The National Institute of Statistics and Economic Studies (INSEE) estimates that one in three individuals in France will be aged\u0026thinsp;\u0026ge;\u0026thinsp;60 years by 2060, compared to one in four individuals in 2021.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eOlder age is characterized by a progressive cycle of aging, which manifests as gradual degeneration of functional capacities and an increased risk of falling.\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e Falls are the leading cause of mortality in this age group and can result in a loss of independence, with significant social, physical and physiological consequences. In France, more than 2\u0026nbsp;million falls occurred among individuals aged\u0026thinsp;\u0026gt;\u0026thinsp;65 years in 2022, leading to over 100,000 hospitalizations and 10,000 deaths.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e The fear of falling often causes older individuals to limit activities such as walking and climbing stairs. While this fear may reduce immediate fall risk, it promotes sedentary behavior, which in turn contributes to progressive frailty over time.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eRecent studies have aimed to provide concrete evidence on the causes and mechanisms of falls, highlighting fall prevention as a critical concern in geriatrics and public health.\u003csup\u003e\u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e While some falls occur as random, unpredictable events, others follow distinct patterns associated with modifiable risk factors and adverse outcomes.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e The study by Ellingsen et al. (2018) identified falls as a cause of death in 88% of reported cases.\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e Understanding these fall trajectories and their underlying risk factors can help clinicians design more effective and targeted prevention strategies.\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e Many studies have used MLR to identify predictive factors for falls, often based on data collected in hospital settings. These data are accessible through patient admissions; however, time constraints frequently limit the assessment of all dimensions of patient health. In our study, data collected at the patient\u0026rsquo;s home encompassed three key dimensions of health: physical/organic, thymic/cognitive, and social-environmental. Given that the order of variable categories matters in MLR, we used marginal predictions to evaluate the contribution of each category and to explore additional predictive factors without disregarding those already identified, before performing a new MLR.\u003c/p\u003e\u003cp\u003eWe hypothesized that: (1) distinct clinical patterns of fall trajectories exist, ranging from no falls to recurrent falls, progressively increasing fall frequency, and chronic falls; and (2) baseline predictive factors for falls already reported in the literature such as physical characteristics, female sex, and cognitive impairment can be identified to encourage healthcare professionals to perform a holistic assessment whenever possible.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1. Study participants\u003c/h2\u003e\u003cp\u003eTo contribute to the growing body of research, we analyzed a longitudinal database from the Unit for Prevention, Monitoring and Analysis of Ageing (UPSAV \u0026ndash; \u003cem\u003eUnit\u0026eacute; de Pr\u0026eacute;vention, de Suivi et d'Analyse du Vieillissement\u003c/em\u003e) which systematically collected fall data from older adults living at home through multiple home visits over a seven-year period. The UPSAV team consists of nurses, geriatricians and other healthcare practitioners. Each patient underwent an initial visit, followed by a second visit six months later. Annual follow-ups were then conducted for up to six years, provided the patient remained in their home.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eThe study includes men and women aged 60 and older. To be eligible, participants had to meet the following criteria:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eProvide written informed consent, either personally or through a legal representative.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eNot be enrolled in a clinical trial that modifies their standard medical management.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eNot have progressive pathologies that could significantly affect short-term prognosis.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eNot reside in a long-term care unit or a nursing home.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eBe covered by social security at 100%.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003e The data-based study was registered in the data studies registry of the University Hospital of Limoges, in accordance with the General Data Protection Regulation (GDPR) and MR-004 (registration number: 87RI24_0006). In compliance with French regulations, non-opposition to participation was obtained from each participant. Ethical approval was granted by the Ethics Committee of the University Hospital of Limoges, \u003cem\u003eEspace de R\u0026eacute;flexion \u0026Eacute;thique\u003c/em\u003e based in Nouvelle-Aquitaine (ERENA), under approval number 28-2025-05. All methods were performed in accordance with the relevant guidelines and regulations.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2. Falls and Clinical Outcomes Assessments\u003c/h2\u003e\u003cp\u003eDuring the Follow-up, a fall was defined as unintentionally coming to rest on the ground or other lower level not as a result of a major intrinsic event (e.g., myocardial infarction, stroke, or seizure) or an overwhelming external hazard (e.g., hit by a vehicle).\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e Each patient underwent a Comprehensive Geriatric Assessment (CGA) and received a personalized care plan. According to the patient\u0026rsquo;s health needs, interventions by an occupational therapist, a psychomotor therapist, or a social worker can be carried out at the patient\u0026rsquo;s home to implement a personalized care plan, in addition to the interventions provided by nurses and geriatricians. The CGA is a multidimensional and standardized approach designed to enhance clinical practices in the care of older adults through a comprehensive health assessment.\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3. Covariates\u003c/h2\u003e\u003cp\u003eCovariates included cardiovascular risk factors, fall occurrences, socio-environmental characteristics and the CGA summary.\u003c/p\u003e\u003cp\u003eSocio-environmental characteristics assessed in the home included gender, age, previous profession, education level, family situation, lifestyle, number of children, housing conditions, presence of an elevator, long-term illness status, health insurance coverage, leisure activities, social activity, human assistance and pet ownership.\u003c/p\u003e\u003cp\u003eCardiovascular risk factors considered were hypertension, diabetes, dyslipidemia, obesity and tobacco use.\u003c/p\u003e\u003cp\u003eThe CGA summary encompassed multiple functional and cognitive assessments, including:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eVerbal fluency test,\u003csup\u003e16\u003c/sup\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eSingle Leg Balance (SLB) test, scored 0\u0026ndash;60 seconds,\u003csup\u003e17\u003c/sup\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eClock-Drawing Test (CDT), scored 0\u0026ndash;5,\u003csup\u003e18\u003c/sup\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eActivities of Daily Living (ADL), scored 0\u0026ndash;6,\u003csup\u003e19\u003c/sup\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eInstrumental Activities of Daily Living (IADL), scored 0\u0026ndash;8,\u003csup\u003e20\u003c/sup\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eMini-Mental State Examination (MMSE), scored 0\u0026ndash;30,\u003csup\u003e21\u003c/sup\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eMini Nutritional Assessment (MNA), scored 0\u0026ndash;30,\u003csup\u003e22\u003c/sup\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eShort Physical Performance Battery (SPPB), scored 0\u0026ndash;12,\u003csup\u003e23\u003c/sup\u003e\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eGeriatric Depression Scale (GDS), scored 0-30.\u003csup\u003e24\u003c/sup\u003e\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eFor consistency, in the rest of the document, we added 'Pathological' to the feature names SLB test, CDT, Verbal Fluency and GDS to indicate whether the test result is positive or not.\u003c/p\u003e\u003cp\u003eBy leveraging this database, we identified subgroups with distinct fall trajectories and examined their baseline characteristics to gain a deeper understanding of the factors influencing fall risk. Our study examines a seven-year dataset collected through multiple home visits by the UPSAV unit.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4. Data analysis\u003c/h2\u003e\u003cp\u003eA descriptive analysis was conducted to provide an overview of the recorded study variables. Additionally, a figure illustrates the proportions of fallers and non-fallers. Fallers were classified into two categories based on the number of falls reported at each visit:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eFaller: a participant who experienced a single fall within a year.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eRecurrent faller: a participant who experienced two or more falls within a year.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eLong-term observation of fall events over multiple years allowed us to define chronic faller as a participant who experienced at most one fall per year over several consecutive years, indicating a persistent but non-acute fall pattern over time.\u003c/p\u003e\u003cp\u003eTo identify clinically distinct trajectories of falls, we applied a GMM,\u003csup\u003e25\u003c/sup\u003e analyses were performed using Python, version 3.12 (Python Software Foundation, Wilmington, DE). The final model was selected using a combination of the Bayesian Information Criterion (BIC)-which balances model complexity and fit quality-and by ensuring that each estimated trajectory group represented at least 5% of the study population.\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e The metric used to define trajectories was the number of months participants remained in the study. A MLR was then applied to identify the best predictors for each trajectory, analyses were performed using R, version 4.3.3 (R Foundation for Statistical Computing, Vienna, Austria).\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eWhen identifying predictors, the order of categories within each feature is crucial. This study utilizes marginal predictions to quantify the isolated effect of each category within its feature, allowing us to determine which category contributes the most to maintaining the steadiest trajectory, setting it as the reference within its feature.\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e With this new ordering, we conducted a new MLR to identify additional predictors.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e3.1. Participants\u003c/h2\u003e\u003cp\u003eA total of 1,648 individuals met the study inclusion criteria. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents the socio-environmental and health characteristics of the study sample at baseline. Among the included older adults, 1,113 (68%) were female and 535 (32%) were male. Additionally, 73% had hypertension and only 288 (17%) participated in social activities. The mean age of participants was 83\u0026thinsp;\u0026plusmn;\u0026thinsp;6 years. Regarding fall occurrences, 823 participants (nearly 50%) had experienced a fall in the past year. In terms of housing conditions, 991 (60%) were homeowners. Furthermore, 449 participants (27%) were classified as depressive patients.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eOverview of Baseline Characteristics According to falls of the study.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003eFalls of the study\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFeatures of the study\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTotal sample\u003c/p\u003e\u003cp\u003e(N\u0026thinsp;=\u0026thinsp;1,648)\u003c/p\u003e\u003cp\u003en (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eNo falls\u003c/p\u003e\u003cp\u003e(n\u0026thinsp;=\u0026thinsp;794, 48.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eFalls\u003c/p\u003e\u003cp\u003e(n\u0026thinsp;=\u0026thinsp;854, 51.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ep-value*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eWoman\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e1,113 (68%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e500 (63%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e613 (72%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eAge, mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, years\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e83\u0026thinsp;\u0026plusmn;\u0026thinsp;6\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e82\u0026thinsp;\u0026plusmn;\u0026thinsp;6\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e83\u0026thinsp;\u0026plusmn;\u0026thinsp;6\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHypertension\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1,209 (73%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e575 (72%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e634 (74%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.40\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eDiabetes\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e339 (21%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e146 (18%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e193 (23%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.035\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDyslipidemia\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e742 (45%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e360 (45%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e382 (45%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.80\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTobacco\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e178 (11%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e85 (11%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e93 (11%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.90\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eObesity\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e405 (25%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e179 (23%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e226 (26%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.065\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePrevious profession\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.10\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBusiness owner / Executive\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e165 (10%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e87 (11%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e78 (9.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCompany employee\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e500 (30%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e253 (32%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e247 (29%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFarmer\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e106 (6.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e53 (6.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e53 (6.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHousewife/Househusband\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e163 (9.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e64 (8.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e99 (12%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIntermediate/liberal profession\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e71 (4.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e31 (3.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e40 (4.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMerchant, craftsman, or independent service provider\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e170 (10%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e73 (9.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e97 (11%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePublic sector employee\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e331 (20%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e169 (21%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e162 (19%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWorker / Other\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e142 (8.6%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e64 (8.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e78 (9.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEducation level\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.15\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCan read, write, count\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e300 (18%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e141 (18%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e159 (19%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHigher education\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e136 (8.3%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e75 (9.4%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e61 (7.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMiddle school diploma\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e241 (15%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e106 (13%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e135 (16%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePrimary school certificate\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e617 (37%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e289 (36%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e328 (38%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSecondary education\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e354 (21%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e183 (23%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e171 (20%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFamily situation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.27\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDivorced / Free union / Unknown\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e129 (7.8%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e63 (7.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e66 (7.7%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMarried\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e700 (42%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e356 (45%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e344 (40%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSingle\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e68 (4.1%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e32 (4.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e36 (4.2%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWidowed\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e751 (46%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e343 (43%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e408 (48%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLifestyle\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.078\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAlone\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e848 (51%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e398 (50%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e450 (53%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWith a partner\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e685 (42%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e349 (44%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e336 (39%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWith family members\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e115 (7.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e47 (5.9%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e68 (8.0%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNumber of children\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2\u0026thinsp;\u0026plusmn;\u0026thinsp;2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2\u0026thinsp;\u0026plusmn;\u0026thinsp;2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2\u0026thinsp;\u0026plusmn;\u0026thinsp;2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.30\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHousing (Owner)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e991 (60%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e482 (61%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e509 (60%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.65\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eElevator\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e389 (24%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e173 (22%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e216 (25%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.094\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLong-term illness\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1,209 (73%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e565 (71%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e644 (75%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.051\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHealth insurance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1,573 (95%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e751 (95%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e822 (96%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.10\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eLeisure\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e1,377 (84%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e689 (87%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e688 (81%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eSocial activity\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e288 (17%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e162 (20%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e126 (15%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.003\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eHuman assistance\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e1,402 (85%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e644 (81%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e758 (89%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePet\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e410 (25%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e204 (26%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e206 (24%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.46\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eADL, mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e5\u0026thinsp;\u0026plusmn;\u0026thinsp;1\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e5\u0026thinsp;\u0026plusmn;\u0026thinsp;1\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e5\u0026thinsp;\u0026plusmn;\u0026thinsp;1\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eIADL, mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e6\u0026thinsp;\u0026plusmn;\u0026thinsp;2\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e6\u0026thinsp;\u0026plusmn;\u0026thinsp;2\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e5\u0026thinsp;\u0026plusmn;\u0026thinsp;2\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eMMSE, mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e23\u0026thinsp;\u0026plusmn;\u0026thinsp;7\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e24\u0026thinsp;\u0026plusmn;\u0026thinsp;7\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e23\u0026thinsp;\u0026plusmn;\u0026thinsp;7\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.006\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003ePathological CDT\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e585 (35%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e244 (31%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e341 (40%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003ePathological verbal fluency\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e672 (41%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e269 (34%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e403 (47%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eMNA, mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e24\u0026thinsp;\u0026plusmn;\u0026thinsp;4\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e24\u0026thinsp;\u0026plusmn;\u0026thinsp;4\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e23\u0026thinsp;\u0026plusmn;\u0026thinsp;4\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eSPPB, mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e7\u0026thinsp;\u0026plusmn;\u0026thinsp;4\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e7\u0026thinsp;\u0026plusmn;\u0026thinsp;4\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e6\u0026thinsp;\u0026plusmn;\u0026thinsp;4\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003ePathological GDS\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e449 (27%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e176 (22%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e273 (32%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003ePathological SLB\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003e708 (43%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003e261 (33%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e447 (52%)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e\u003cp\u003e\u003csup\u003e\u003cem\u003e*\u003c/em\u003e\u003c/sup\u003ePearson's Chi-squared test; Wilcoxon rank sum test. Statistically significance (p-value\u0026thinsp;\u0026lt;\u0026thinsp;.05).\u003c/p\u003e\u003cp\u003eSD, Standard deviation; SLB, Single leg balance; CDT, Clock-drawing test; ADL, Activities of Daily Living; IADL, Instrumental Activities of Daily Living; MMSE, Mini-Mental State Examination; MNA, Mini Nutritional Assessment; SPPB, Short Physical Performance Battery; GDS, Geriatric Depression Scale.\u003c/p\u003e\u003cp\u003eData are shown as the number (percentage) or mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD unless otherwise indicated.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e shows that 40% of the study population experienced a single fall, 14% had recurrent falls (at least twice a year) and 46% did not fall during the study period. The distribution of falls in our study aligns with findings from the literature, which indicate that among individuals aged 80 years and older, one in two experiences a fall.\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eBased on these different fall categories, we analyze how each feature is distributed between fallers and non-fallers. Fallers include individuals who experience recurrent falls as well as those who fall only once.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e3.2. Features of the study according to the fallers and non-fallers\u003c/h2\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents a comparison of study characteristics between patients who have fallen and those who have not. Within the faller group, several features exhibited significant differences (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) compared to non-fallers:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eon average: Age, ADL, IADL, MMSE, MNA and SPPB.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eby proportion: Gender, Diabetes, Leisure, Social activity, Human assistance, Pathological CDT, Pathological verbal fluency, Pathological GDS and Pathological SLB.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eAfter highlighting significant differences between fallers and non-fallers, we will group patients with similar characteristics into clusters. Each cluster will be represented in fall trajectories by plotting the mean number of falls per month using the GMM.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e3.3. Determining Falls Trajectories over 7 years\u003c/h2\u003e\u003cp\u003eTo identify participants\u0026rsquo; fall trajectories, we applied clustering methods that group individuals with similar profiles within the study population. One such method is the GMM, which we implemented using Python\u0026rsquo;s Scikit-learn library.\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e GMM is a probabilistic clustering approach that models the data as a mixture of multiple Gaussian distributions. Unlike K-means clustering, which assigns each data point to a single cluster, GMM assigns probabilities to multiple clusters, enabling more flexible classifications. The model is trained using the Expectation-Maximization (EM) algorithm, which iteratively estimates cluster probabilities and updates distribution parameters to maximize likelihood. This process improves the accuracy of trajectory classification.\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e depicts the mean number of falls over time (in months) for a model of four different groups performed well, each of the trajectories is constituted by at least five percent of the population study according to the Nagin criteria.\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e The distinct trajectories were Cluster Falls and No Falls (n\u0026thinsp;=\u0026thinsp;1079, 65.5%), Increasing Falls (n\u0026thinsp;=\u0026thinsp;111, 6.7%), Chronic Recurring Falls (n\u0026thinsp;=\u0026thinsp;258, 15.7%) and Low-Rate Chronic Falls (n\u0026thinsp;=\u0026thinsp;200, 12.1%).\u003c/p\u003e\u003cp\u003eAmong all the features in our study, we will analyze their distribution within each trajectory. This will allow us to observe how specific characteristics vary across different fall patterns and identify potential associations with fall risk.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e3.4. Features of the study according to each trajectory\u003c/h2\u003e\u003cp\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e demonstrates that the distribution of several characteristics differs significantly across the four fall-trajectory groups (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05): Gender, Age, Obesity, Long-term illness, Leisure, Social activity, IADL, MMSE, Pathological verbal fluency, MNA, SPPB and Pathological SLB.\u003c/p\u003e\u003cp\u003eIn sections 3.5 and 3.6, we will identify the strongest predictors for each trajectory using MLR. When examining all trajectories across the entire study period, Clustered Falls and No Falls emerges as the steadiest trajectory. At baseline, the trajectory associated with no falls was Increasing Falls, whereas at the 18-month follow-up, it was Clustered Falls and No Falls (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). We investigated predictive factors at these two key time points baseline and 18 months when a trajectory reached a mean of zero falls.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e3.5. Trajectories at baseline\u003c/h2\u003e\u003cdiv id=\"Sec13\" class=\"Section3\"\u003e\u003ch2\u003e3.5.1. Predictors of Falls at baseline\u003c/h2\u003e\u003cp\u003eMLR revealed that specific baseline characteristics predicted membership within each of the three fallers trajectory groups as compared to the Increasing Falls trajectory group are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e (see Table S2\u003cb\u003e)\u003c/b\u003e. Predictors of Clustered Falls and No Falls were Leisure (No), IADL\u0026thinsp;\u0026lt;\u0026thinsp;8 and pathological SLB. Predictors of Chronic Recurring Falls trajectory were woman and obesity. Predictors of Low-Rate Chronic Falls trajectory were woman, obesity, IADL\u0026thinsp;\u0026lt;\u0026thinsp;8 and pathological SLB.\u003c/p\u003e\u003cp\u003eWe reordered the features using marginal predictions and conducted a new MLR to evaluate whether more or fewer predictors would be identified.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section3\"\u003e\u003ch2\u003e3.5.2. Marginal predictions at baseline\u003c/h2\u003e\u003cp\u003eWe observe that, out of all the features in our study (see 3.5), only some remained after the MLR analysis (see Table S2 or Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). However, not all features in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e served as predictors across different trajectories. In MLR, the order of categories within each feature is crucial. To establish this order, we set the steadiest trajectory, Increasing Falls as the reference. Then, for each feature, we identify the category that contributes the most, on average, to this steadiest trajectory and designate it as the reference within its feature. This process is carried out using marginal predictions. According to table S3 (see Fig. S2), the probability of being a non-faller is highest on categories man, age under 80 years old, obesity (No), hypertension (Yes), leisure (Yes), social activity (No), IADL\u0026thinsp;=\u0026thinsp;8, MMSE\u0026thinsp;\u0026ge;\u0026thinsp;24, Pathological verbal fluency (No), Pathological SLB (No).\u003c/p\u003e\u003cp\u003eWe conducted a new MLR using the same features (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) but redefined the reference category within each feature based on marginal predictions (see 3.5.2).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section3\"\u003e\u003ch2\u003e3.5.3. Predictors of Falls Trajectories after Rearrangement of Categories in each Feature at baseline\u003c/h2\u003e\u003cp\u003eSimilar to the MLR performed in section 3.5.1, we identified specific baseline characteristics that predicted membership within each of the three faller trajectories compared to the Increasing Falls group. These results are presented in table S4 (see Fig. S3). We observe that the same predictors from table S2 (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) reappear in the results of the new MLR, along with additional predictors only in Chronic Recurring Falls trajectory namely hypertension (No).\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e3.6. Trajectories at 18 months\u003c/h2\u003e\u003cdiv id=\"Sec17\" class=\"Section3\"\u003e\u003ch2\u003e3.6.1. Predictors of Falls at 18 months\u003c/h2\u003e\u003cp\u003eMLR revealed that specific baseline characteristics predicted membership within each of the three fallers trajectory groups as compared to the Clustered Falls and No Falls trajectory group are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e (see Table S5). Predictors of Increasing Falls were SPPB (between 6 and 9), SPPB\u0026thinsp;\u0026lt;\u0026thinsp;5 and pathological GDS. Predictors of Chronic Recurring Falls trajectory were woman, ADL\u0026thinsp;\u0026lt;\u0026thinsp;4, SPPB\u0026thinsp;\u0026lt;\u0026thinsp;5 and pathological GDS. Predictors of Low-Rate Chronic Falls trajectory were obesity, SPPB\u0026thinsp;\u0026lt;\u0026thinsp;5 and pathological GDS.\u003c/p\u003e\u003cp\u003eWe reordered the features using marginal predictions and conducted a new MLR to evaluate whether more or fewer predictors would be identified.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section3\"\u003e\u003ch2\u003e3.6.2. Marginal predictions at 18 months\u003c/h2\u003e\u003cp\u003eWe apply the same procedure as 3.5.2. To establish this order, we set the steadiest trajectory, Clustered Falls and No Falls as the reference. According to table S6 (see Fig. S4), the probability of being a non-faller is highest on categories, man, hypertension (Yes), diabetes (No), dyslipidemia (No), obesity (No), ADL (between 4 and 6), MMSE\u0026thinsp;\u0026lt;\u0026thinsp;24, SPPB (between 10 and 12), Pathological GDS (No), Pathological SLB (Yes).\u003c/p\u003e\u003cp\u003eWe conducted a new MLR using the same features (see Fig. S3) but redefined the reference category within each feature based on marginal predictions (see 3.6.2).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section3\"\u003e\u003ch2\u003e3.6.3. Predictors of Falls Trajectories after Rearrangement of Categories in each Feature at 18 months\u003c/h2\u003e\u003cp\u003eSimilar to the MLR performed in section 3.6.1, we identified specific baseline characteristics that predicted membership within each of the three faller trajectories compared to the Clustered Falls and No Falls group. These results are presented in table S7 (see Fig. S5). We observe that the same predictors from table S5 reappear in the results of the new MLR, along with additional predictors for each trajectory:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eIncreasing Falls: Pathological SLB (No).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eChronic Recurring Falls: Hypertension (No) and Pathological SLB (No).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eLow-Rate Chronic Falls: Hypertension (No), Dyslipidemia (No) and MMSE\u0026thinsp;\u0026ge;\u0026thinsp;24.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eIn this study, we explore the different trajectories of falls among home-dwelling patients over a seven-year period. We found four clinically distinct trajectories of falls, which we have labeled: Cluster Falls and No Falls, Increasing Falls, Chronic Recurring Falls and Low-Rate Chronic Falls. Cluster Falls and No Falls show minimal risk after an initial decrease. Increasing Falls have a delayed peak (around 30 months) before a rapid decline, suggesting episodic high-risk periods rather than ongoing vulnerability. Chronic Recurring Falls experience a sharp peak and drop early on, indicating a temporary high fall risk. Low-Rate Chronic Falls are the most persistent group, showing a continuous risk of falls throughout the study period. Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e provides a clear distinction between fall trajectories, showing that Chronic Recurring Falls tend to be frailer and less socially engaged; while Increasing and Low-Rate Chronic Falls maintain more mobility and social interaction but still experience frequent falls.\u003c/p\u003e\u003cp\u003eThe configuration of our trajectories prompted us to investigate predictive factors both at baseline and after 18 months. The factors identified at these two time points were female sex and obesity. This finding underscores the importance of assessing predictive factors not only at baseline but also during follow-up. Previous studies have similarly reported that female sex and obesity are consistent predictors at both key stages. Furthermore, in the recent study by Nicklett et al. (2017), female sex was identified as a risk factor for fall events.\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e This finding is consistent with previous studies by Deandrea et al. (2010), Shumway-Cook et al. (2009), Stahl and Albert (2015), and de Rekeneire et al. (2003), which have also confirmed this association.\u003csup\u003e\u003cspan additionalcitationids=\"CR35 CR36\" citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e A recent study of Abdo et al. (2022) find obesity as one of the major predictive factors for falls, that align with the study of Erlandson et al. (2019) and M\u0026aacute;ximo et al. (2019). Among the additional predictors identified at baseline were the absence of leisure activities, pathological SLB, and an IADL score below 8. SLB and leisure are both related to physical activity, and leisure in particular may influence the psychological (thymic) dimension of a patient\u0026rsquo;s health. SLB and leisure have been identified as predictive factors for falls in recent studies by Blodgett et al. (2022) and Kwok et al. (2024), respectively.\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e In the study by Kwok et al. (2024), the population consisted exclusively of women, which highlights the continuing importance of exploring the relationship between female sex and fall risk. The last predictive factor identified at baseline, namely IADL, is related to patient autonomy and can provide insight into mental health status. The study by Brown et al. (2014) used this variable as a standalone tool for screening fall risk when stratified by stage.\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e This article considered IADL to be one of the strongest predictive factors for falls. Furthermore, Britton et al. (2019) confirmed through their study that IADL is not only a predictor of falls but also a predictor of recurrent falls.\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e Recurrent falls reflect a lack or complete absence of effective prevention strategies, which can lead to more severe consequences. While many studies have focused on preventing the first fall, others have sought to identify strategies to prevent subsequent or recurrent falls.\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan additionalcitationids=\"CR42 CR43 CR44 CR45\" citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e As is well known, the ADL identified at 18 months in our study also serves to assess patient autonomy, similarly to the IADL. In the study by Henry et al. (2012), it was demonstrated that individuals with intermediate levels of ADL limitation who remain partially active have a higher risk of falling.\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e The identification of SLB confirms that falls cannot be dissociated from balance, while the SPPB identified at 18 months provides complementary information, indicating that falls are also closely associated with walking ability (gait). Several studies have demonstrated that physical frailty must be taken into account in fall prevention strategies. Recent studies by Choi et al. (2023), Ge et al. (2023), and Abdo et al. (2022) have confirmed that the SPPB is a predictive factor for falls, as previously demonstrated in the studies by Erlandson et al. (2019), Souza et al. (2019), and Moreland et al. (2018).\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e,\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e,\u003cspan additionalcitationids=\"CR49\" citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e One of the recurrent predictive factors is related to the cognitive dimension of patient health, namely the GDS. Several studies have included this variable in their assessment of fall risk and have identified it as a significant predictive factor. Initially identified in the studies by Nicklett et al. (2017), Britton et al. (2019), and Erlandson et al. (2019), the GDS has since been confirmed as one of the strongest predictive factors for falls in more recent research by Jo et al. (2020), Fallaci et al. (2022), Ge et al. (2023), and Pesonen et al. (2025).\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e,\u003cspan additionalcitationids=\"CR44\" citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e,\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e,\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e In particular, Pesonen et al. (2025) examined the combined role of impaired vision and cognitive decline in falls and fall-related risk factors. Moreover, the relationship between depression and physical frailty was demonstrated by Chang et al. (2011).\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eAll the variables identified as predictive factors were also reported in the literature. All these factors are related to patients\u0026rsquo; health impairments. In our MLR results, we observed that other features were present but did not emerge as significant predictors of falls in any of the trajectories. The use of marginal predictions to assess the contribution of each category within each feature allowed us to redefine the reference category within each feature before performing a new MLR. This MLR led us to find more predictors, namely, hypertension (No) at baseline and pathological SLB (No), hypertension (No) and pathological SLB (No), dyslipidemia (No) and MMSE\u0026thinsp;\u0026ge;\u0026thinsp;24 at 18 months. This result confirms that people who fall are not always those with multiple comorbidities. This aligns with the unintentional and therefore unpredictable nature of falls in older adults.\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eMarginal prediction analysis revealed that being male or having a normal weight status increased the likelihood of belonging to the most stable fall trajectory. This result confirms the impact of sex, with female gender remaining a significant predictive factor for falls. The study by Renner et al. (2021), in which the entire study population consisted of older men, demonstrated that higher levels of fatigue can increase their risk of falling.\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e These findings suggest that it would be valuable to conduct separate analyses within each dataset: one including both sexes and one for each sex individually. Such an approach could reveal sex-specific factors that should be taken into account when developing fall prevention strategies.\u003c/p\u003e\u003cp\u003eIn our study, age was not identified as a predictive factor; however, the difference in mean age between fallers and non-fallers was significant (Table \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). In contrast, the studies by Hollinghurst et al. (2022), Mata et al. (2017) and Nicklett et al. (2017) demonstrated that age is a predictive factor for falls.\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eIt may appear that individuals who move less experience fewer falls due to their health condition. However, while reducing movement might seem like a strategy to prevent falls, it is not an effective solution. Older individuals who limit their mobility enter a cycle of progressive loss of autonomy. In fact, individuals who have already experienced a fall may develop a fear of falling, leading to sedentary behavior and reduced physical activity, further exacerbating frailty and loss of independence. Across the different trajectories, two of the most consistent predictors were pathological SLB and impaired physical performance, underscoring the critical role of balance and mobility in fall risk.\u003c/p\u003e\u003cp\u003eWe observe in our study that a higher proportion of individuals in the faller group require human assistance (see 3.2). In each trajectory, at least one patient out of two lived alone, which could contribute to higher fall risk and its consequences (see Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e This finding highlights the crucial role of caregivers in fall prevention and post-fall assistance, as individuals who fall may remain on the ground for an extended period, a phenomenon known as long lie.\u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e In such situations, the lack of assistance can lead to serious complications, including death. The study by Leggett et al. (2018) show that caregivers\u0026rsquo; greater nighttime awakenings were associated with caring for care recipients with higher fall risk.\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eIn hospital settings, the thymic (mood-related) and cognitive dimensions of falls are often not a primary focus during patient evaluations. However, for older adults living at home, teams like the UPSAV unit conduct CGA, enabling early detection of depressive symptoms or cognitive decline. Previous studies have emphasized the importance of considering these factors when designing effective fall prevention strategies.\u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e,\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eIn summary, all the variables identified in our study as predictive factors for falls have also been reported in the literature published between 2019 and 2025. Our strategy of examining fall predictors among trajectories with a mean number of falls equal to zero (both at baseline and at 18 months) allowed us to identify additional predictors, including one of the most significant, GDS, which reflects the depressive state of the patient. Moreover, the use of marginal predictions highlighted that a patient may still experience falls despite having good scores on certain tests, such as the MMSE or SLB, or in the absence of cardiovascular issues. This may be explained by the interventions provided by UPSAV between the two visits. The quality of these interventions may have improved several dimensions of patients\u0026rsquo; health yet not prevented the occurrence of falls due to other fall-related characteristics, precisely those identified in our study and consistently described in the literature.\u003c/p\u003e\u003cp\u003eOur study has several limitations. (1) We used baseline clinical characteristics obtained between visits, which limited our ability to identify risk factors at the exact time of the falls. Consequently, we could not determine whether acute illnesses, medication use, or other transient factors influenced specific fall trajectories or whether certain patterns reflected the progression of chronic diseases. (2) Some individuals classified as Cluster Falls and No Falls may have had prior fall episodes that were not accounted for, resulting in left-censored data in our analysis. This limitation could have influenced the classification of fall trajectories. Additionally, implementing a Randomized Controlled Trial (RCT) design could enhance the evidence for intervention effectiveness and help establish causal relationships between fall prevention strategies and outcomes. (3) The study population was restricted to older adults residing in France, which may limit the generalizability of our findings to other geographic or cultural contexts. Future research should aim to include more diverse populations to enhance the applicability of the results.\u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eIn summary, our study identifies four clinically distinct fall trajectories over seven years among community-dwelling older adults, each associated with a set of unique risk factors. Low-Rate Chronic Falls exhibit a persistent fall risk, while Increasing Falls experience episodic high-risk periods. Interestingly, age category did not emerge as a significant predictor for any of the identified fall trajectories. Key predictors across trajectories include being female, obesity, impairment in activities of daily living (ADL/IADL), reduced physical performance and depressive symptoms all of which should be prioritized in fall prevention interventions. Additionally, older individuals who live alone are also at risk of falling. Given the significant impact of recurrent falls on health outcomes, targeted interventions should prioritize those at the highest risk to prevent further decline and injury.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments:\u0026nbsp;\u003c/strong\u003eThe authors thank the Unit for Prevention, Monitoring and Analysis of Ageing (UPSAV \u0026ndash;\u0026nbsp;\u003cem\u003eUnit\u0026eacute; de Pr\u0026eacute;vention, de Suivi et d\u0026apos;Analyse du Vieillissement\u003c/em\u003e) and the Clinical Research and Innovation Unit in Gerontology (URCI \u0026ndash;\u0026nbsp;\u003cem\u003eUnit\u0026eacute; de Recherche Clinique et innovation en G\u0026eacute;rontologie\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSponsor\u0026apos;s role:\u0026nbsp;\u003c/strong\u003eData grant by \u003cem\u003e\u0026ldquo;CARSAT Centre-Ouest\u0026rdquo;,\u003c/em\u003e France were collected by UPSAV and URCI. The analysis and manuscript preparation were funded by a grant from chair of excellence in Artificial Intelligence and Healthy Ageing in the Nouvelle-Aquitaine Region (IABV-NA \u0026ndash;\u0026nbsp;\u003cem\u003eIntelligence Artificielle et Bien-Vieillir en Nouvelle-Aquitaine\u003c/em\u003e\u003cem\u003e).\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict-of-interest statement:\u0026nbsp;\u003c/strong\u003eNo conflicts of interest\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding sources\u003c/strong\u003e: \u003cstrong\u003eFUNDING INFORMATION\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was funded by a grant from chair of excellence in Artificial Intelligence and Healthy Ageing in the Nouvelle-Aquitaine Region (IABV-NA \u0026ndash;\u0026nbsp;\u003cem\u003eIntelligence Artificielle et Bien-Vieillir en Nouvelle-Aquitaine\u003c/em\u003e\u003cem\u003e)\u0026nbsp;\u003c/em\u003eand \u003cem\u003e\u0026ldquo;CARSAT Centre-Ouest\u0026rdquo;\u003c/em\u003e.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eWHO. Ageing and health. Accessed March 28, 2023. (2022). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.who.int/fr/news-room/fact-sheets/detail/ageing-and-health\u003c/span\u003e\u003cspan address=\"https://www.who.int/fr/news-room/fact-sheets/detail/ageing-and-health\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eInsee Population by age - Tableaux de l\u0026rsquo;\u0026Eacute;conomie Fran\u0026ccedil;aise | Insee. Accessed March 28, 2023. (2016). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.insee.fr/fr/statistiques/1906664?sommaire=1906743\u003c/span\u003e\u003cspan address=\"https://www.insee.fr/fr/statistiques/1906664?sommaire=1906743\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMoylan, K. C. \u0026amp; Binder, E. F. Falls in older adults: risk assessment, management and prevention. \u003cem\u003eAm. J. Med.\u003c/em\u003e \u003cb\u003e120\u003c/b\u003e (6), 493e1\u0026ndash;493e6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.amjmed.2006.07.022\u003c/span\u003e\u003cspan address=\"10.1016/j.amjmed.2006.07.022\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2007).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eThierry, X. Accidents and physical assaults among the elderly: less frequent than among the young, but more severe. \u003cem\u003ePopul. Soci\u0026eacute;t\u0026eacute;s\u003c/em\u003e. \u003cb\u003e468\u003c/b\u003e (6), 1\u0026ndash;4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3917/popsoc.468.0001\u003c/span\u003e\u003cspan address=\"10.3917/popsoc.468.0001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHealth.gouv. Accessed March 29. (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://sante.gouv.fr/IMG/pdf/dp_plan-antichute-accessible28-02-2022.pdf\u003c/span\u003e\u003cspan address=\"https://sante.gouv.fr/IMG/pdf/dp_plan-antichute-accessible28-02-2022.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePark, S. H. Tools for assessing fall risk in the elderly: a systematic review and meta-analysis. \u003cem\u003eAging Clin. Exp. Res.\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e (1), 1\u0026ndash;16. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s40520-017-0749-0\u003c/span\u003e\u003cspan address=\"10.1007/s40520-017-0749-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLim, C. et al. Associations Between Multiple Falls and Mental Health Indices in Korean Older Adults: A Cross-sectional Study. \u003cem\u003eJ. Prev. Med. Public. Health\u003c/em\u003e. \u003cb\u003e58\u003c/b\u003e (3), 269\u0026ndash;277. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3961/jpmph.24.482\u003c/span\u003e\u003cspan address=\"10.3961/jpmph.24.482\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMay, P., Roe, L., McGarrigle, C. A., Kenny, R. A. \u0026amp; Normand, C. End-of-life experience for older adults in Ireland: results from the Irish longitudinal study on ageing (TILDA). \u003cem\u003eBMC Health Serv. Res.\u003c/em\u003e \u003cb\u003e20\u003c/b\u003e (1), 118. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12913-020-4978-0\u003c/span\u003e\u003cspan address=\"10.1186/s12913-020-4978-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eda Mata, L. R. F., Azevedo, C., Policarpo, A. G. \u0026amp; Moraes, J. T. Factors associated with the risk of fall in adults in the postoperative period: a cross-sectional study. \u003cem\u003eRev. Lat Am. Enfermagem\u003c/em\u003e. \u003cb\u003e25\u003c/b\u003e, e2904. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1590/1518-8345.1775.2904\u003c/span\u003e\u003cspan address=\"10.1590/1518-8345.1775.2904\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHollinghurst, R. et al. Annual risk of falls resulting in emergency department and hospital attendances for older people: an observational study of 781,081 individuals living in Wales (United Kingdom) including deprivation, frailty and dementia diagnoses between 2010 and 2020. \u003cem\u003eAge Ageing\u003c/em\u003e. \u003cb\u003e51\u003c/b\u003e (8), afac176. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/ageing/afac176\u003c/span\u003e\u003cspan address=\"10.1093/ageing/afac176\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJayakody, O., Blumen, H. M., Ayers, E. \u0026amp; Verghese, J. Longitudinal Associations Between Falls and Risk of Gait Decline: Results From the Central Control of Mobility and Aging Study. \u003cem\u003eArch. Phys. Med. Rehabil\u003c/em\u003e. \u003cb\u003e104\u003c/b\u003e (2), 245\u0026ndash;250. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.apmr.2022.08.975\u003c/span\u003e\u003cspan address=\"10.1016/j.apmr.2022.08.975\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEllingsen, C. L., Ebbing, M., Alfsen, G. C. \u0026amp; Vollset, S. E. Injury death certificates without specification of the circumstances leading to the fatal injury - the Norwegian Cause of Death Registry 2005\u0026ndash;2014. \u003cem\u003ePopul. Health Metr.\u003c/em\u003e \u003cb\u003e16\u003c/b\u003e (1), 20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12963-018-0176-2\u003c/span\u003e\u003cspan address=\"10.1186/s12963-018-0176-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTchalla, A. et al. Patterns, predictors, and outcomes of frailty trajectories in community-dwelling older adults: Results from the FREEDOM Cohort Study. \u003cem\u003eArch. Gerontol. Geriatr.\u003c/em\u003e \u003cb\u003e114\u003c/b\u003e, 105101. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.archger.2023.105101\u003c/span\u003e\u003cspan address=\"10.1016/j.archger.2023.105101\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKellogg The prevention of falls in later life. A report of the Kellogg International Work Group on the Prevention of Falls by the Elderly. \u003cem\u003eDan. Med. Bull.\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e (Suppl 4), 1\u0026ndash;24 (1987).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWelsh, T. J., Gordon, A. L. \u0026amp; Gladman, J. R. Comprehensive geriatric assessment \u0026ndash; a guide for the non-specialist. \u003cem\u003eInt. J. Clin. Pract.\u003c/em\u003e \u003cb\u003e68\u003c/b\u003e (3), 290\u0026ndash;293. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/ijcp.12313\u003c/span\u003e\u003cspan address=\"10.1111/ijcp.12313\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBenton, A. L. Development of a multilingual aphasia battery: Progress and problems. \u003cem\u003eJ. Neurol. Sci.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e (1), 39\u0026ndash;48. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/0022-510X(69)90057-4\u003c/span\u003e\u003cspan address=\"10.1016/0022-510X(69)90057-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1969).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTrojian, T. H. \u0026amp; McKeag, D. B. Single leg balance test to identify risk of ankle sprains. \u003cem\u003ePublished online July\u003c/em\u003e. \u003cb\u003e1\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1136/bjsm.2005.024356\u003c/span\u003e\u003cspan address=\"10.1136/bjsm.2005.024356\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2006).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFreedman, M. et al. and D. Delis. Clock drawing: a neuropsychological analysis. New York: Oxford, 1994. | Request PDF. \u003cem\u003eResearchGate\u003c/em\u003e. Published online 1994. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1017/S0714980800013398\u003c/span\u003e\u003cspan address=\"10.1017/S0714980800013398\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKatz, S., Ford, A. B., Moskowitz, R. W., Jackson, B. A. \u0026amp; Jaffe, M. W. Studies of Illness in the Aged: The Index of ADL: A Standardized Measure of Biological and Psychosocial Function. \u003cem\u003eJAMA\u003c/em\u003e \u003cb\u003e185\u003c/b\u003e (12), 914\u0026ndash;919. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jama.1963.03060120024016\u003c/span\u003e\u003cspan address=\"10.1001/jama.1963.03060120024016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1963).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLawton, M. P. \u0026amp; Brody, E. M. Assessment of Older People: Self-Maintaining and Instrumental Activities of Daily Living1. \u003cem\u003eGerontologist\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e (3_Part_1), 179\u0026ndash;186. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/geront/9.3_Part_1.179\u003c/span\u003e\u003cspan address=\"10.1093/geront/9.3_Part_1.179\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1969).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEscobar, J. I. et al. Use of the Mini-Mental State Examination (MMSE) in a community population of mixed ethnicity. Cultural and linguistic artifacts. \u003cem\u003eJ. Nerv. Ment Dis.\u003c/em\u003e \u003cb\u003e174\u003c/b\u003e (10), 607\u0026ndash;614. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/00005053-198610000-00005\u003c/span\u003e\u003cspan address=\"10.1097/00005053-198610000-00005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1986).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVellas, B. et al. The Mini Nutritional Assessment (MNA) and its use in grading the nutritional state of elderly patients. \u003cem\u003eNutrition\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e (2), 116\u0026ndash;122. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/s0899-9007(98)00171-3\u003c/span\u003e\u003cspan address=\"10.1016/s0899-9007(98)00171-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1999).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuralnik, J. M. et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. \u003cem\u003eJ. Gerontol.\u003c/em\u003e \u003cb\u003e49\u003c/b\u003e (2), M85\u0026ndash;94. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/geronj/49.2.m85\u003c/span\u003e\u003cspan address=\"10.1093/geronj/49.2.m85\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1994).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYesavage, J. A. et al. Development and validation of a geriatric depression screening scale: A preliminary report. \u003cem\u003eJ. Psychiatr. Res.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e (1), 37\u0026ndash;49. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/0022-3956(82)90033-4\u003c/span\u003e\u003cspan address=\"10.1016/0022-3956(82)90033-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1982).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang, M. S., Lai, C. Y. \u0026amp; Lin, C. Y. A robust EM clustering algorithm for Gaussian mixture models. \u003cem\u003ePattern Recogn.\u003c/em\u003e \u003cb\u003e45\u003c/b\u003e (11), 3950\u0026ndash;3961. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.patcog.2012.04.031\u003c/span\u003e\u003cspan address=\"10.1016/j.patcog.2012.04.031\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNagin, D. S. \u0026amp; Odgers, C. L. Group-based trajectory modeling in clinical research. \u003cem\u003eAnnu. Rev. Clin. Psychol.\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, 109\u0026ndash;138. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1146/annurev.clinpsy.121208.131413\u003c/span\u003e\u003cspan address=\"10.1146/annurev.clinpsy.121208.131413\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eB\u0026ouml;hning, D. Multinomial logistic regression algorithm. \u003cem\u003eAnn. Inst. Stat. Math.\u003c/em\u003e \u003cb\u003e44\u003c/b\u003e (1), 197\u0026ndash;200. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/BF00048682\u003c/span\u003e\u003cspan address=\"10.1007/BF00048682\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1992).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eArel-Bundock, V., Greifer, N. \u0026amp; Heiss, A. How to Interpret Statistical Models Using marginaleffects for R and Python. \u003cem\u003eJ. Stat. Softw.\u003c/em\u003e \u003cb\u003e111\u003c/b\u003e, 1\u0026ndash;32. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.18637/jss.v111.i09\u003c/span\u003e\u003cspan address=\"10.18637/jss.v111.i09\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGrundstrom, A. C., Guse, C. E. \u0026amp; Layde, P. M. Risk Factors for Falls and Fall-Related Injuries in Adults 85 Years of Age and Older. \u003cem\u003eArch. Gerontol. Geriatr.\u003c/em\u003e \u003cb\u003e54\u003c/b\u003e (3), 421\u0026ndash;428. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.archger.2011.06.008\u003c/span\u003e\u003cspan address=\"10.1016/j.archger.2011.06.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eScikit-learn developers. GaussianMixture. scikit-learn. Accessed February 23. (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://scikit-learn.org/stable/modules/mixture.html\u003c/span\u003e\u003cspan address=\"https://scikit-learn.org/stable/modules/mixture.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMoon, T. K. The expectation-maximization algorithm. \u003cem\u003eIEEE. Signal. Process. Mag.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e (6), 47\u0026ndash;60. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/79.543975\u003c/span\u003e\u003cspan address=\"10.1109/79.543975\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1996).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNagin, D. S. Group-Based Trajectory Modeling: An Overview. \u003cem\u003eANM\u003c/em\u003e \u003cb\u003e65\u003c/b\u003e (2\u0026ndash;3), 205\u0026ndash;210. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1159/000360229\u003c/span\u003e\u003cspan address=\"10.1159/000360229\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNicklett, E. J., Taylor, R. J., Rostant, O., Johnson, K. E. \u0026amp; Evans, L. Biopsychosocial Predictors of Fall Events Among Older African Americans. \u003cem\u003eRes. Aging\u003c/em\u003e. \u003cb\u003e39\u003c/b\u003e (4), 501\u0026ndash;525. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1177/0164027516651974\u003c/span\u003e\u003cspan address=\"10.1177/0164027516651974\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDeandrea, S. et al. Risk factors for falls in community-dwelling older people: a systematic review and meta-analysis. \u003cem\u003eEpidemiology\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e (5), 658\u0026ndash;668. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/EDE.0b013e3181e89905\u003c/span\u003e\u003cspan address=\"10.1097/EDE.0b013e3181e89905\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShumway-Cook, A. et al. Falls in the Medicare population: incidence, associated factors, and impact on health care. \u003cem\u003ePhys. Ther.\u003c/em\u003e \u003cb\u003e89\u003c/b\u003e (4), 324\u0026ndash;332. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2522/ptj.20070107\u003c/span\u003e\u003cspan address=\"10.2522/ptj.20070107\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStahl, S. T. \u0026amp; Albert, S. M. Gender differences in physical activity patterns among older adults who fall. \u003cem\u003ePrev. Med.\u003c/em\u003e \u003cb\u003e71\u003c/b\u003e, 94\u0026ndash;100. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ypmed.2014.12.016\u003c/span\u003e\u003cspan address=\"10.1016/j.ypmed.2014.12.016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDe Rekeneire, N. et al. Is a fall just a fall: correlates of falling in healthy older persons. The Health, Aging and Body Composition Study. \u003cem\u003eJ. Am. Geriatr. Soc.\u003c/em\u003e \u003cb\u003e51\u003c/b\u003e (6), 841\u0026ndash;846. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1046/j.1365-2389.2003.51267.x\u003c/span\u003e\u003cspan address=\"10.1046/j.1365-2389.2003.51267.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2003).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBlodgett, J. M. et al. One-Legged Balance Performance and Fall Risk in Mid and Later Life: Longitudinal Evidence From a British Birth Cohort. \u003cem\u003eAm. J. Prev. Med.\u003c/em\u003e \u003cb\u003e63\u003c/b\u003e (6), 997\u0026ndash;1006. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.amepre.2022.07.002\u003c/span\u003e\u003cspan address=\"10.1016/j.amepre.2022.07.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKwok, W. S. et al. Leisure-Time Physical Activity and Falls With and Without Injuries Among Older Adult Women. \u003cem\u003eJAMA Netw. Open.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e (1), e2354036. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jamanetworkopen.2023.54036\u003c/span\u003e\u003cspan address=\"10.1001/jamanetworkopen.2023.54036\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBrown, J., Kurichi, J. E., Xie, D., Pan, Q. \u0026amp; Stineman, M. G. Instrumental activities of daily living staging as a possible clinical tool for falls risk assessment in physical medicine and rehabilitation. \u003cem\u003ePM R\u003c/em\u003e. \u003cb\u003e6\u003c/b\u003e (4), 316\u0026ndash;323. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.pmrj.2013.10.007\u003c/span\u003e\u003cspan address=\"10.1016/j.pmrj.2013.10.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2014). quiz 323.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBritton, G. B. et al. Inflammatory Biomarkers, Depressive Symptoms and Falls Among the elderly in Panama. \u003cem\u003eCurr. Aging Sci.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e (4), 236\u0026ndash;241. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2174/1874609812666190215125104\u003c/span\u003e\u003cspan address=\"10.2174/1874609812666190215125104\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChoi, N. G., Marti, C. N., Choi, B. Y. \u0026amp; Kunik, M. M. Recurrent Falls over Three Years among Older Adults Age 70+: Associations with Physical and Mental Health Status, Exercise, and Hospital Stay. \u003cem\u003eJ. Appl. Gerontol.\u003c/em\u003e \u003cb\u003e42\u003c/b\u003e (5), 1089\u0026ndash;1100. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1177/07334648221150884\u003c/span\u003e\u003cspan address=\"10.1177/07334648221150884\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eErlandson, K. M. et al. Risk Factors for Falls, Falls With Injury, and Falls With Fracture Among Older Men With or at Risk of HIV Infection. \u003cem\u003eJ. Acquir. Immune Defic. Syndr.\u003c/em\u003e \u003cb\u003e81\u003c/b\u003e (4), e117\u0026ndash;e126. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/QAI.0000000000002074\u003c/span\u003e\u003cspan address=\"10.1097/QAI.0000000000002074\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFallaci, I. V., Fabr\u0026iacute;cio, D., de Alexandre, M. \u0026amp; da Chagas, T. Association between falls and cognitive performance among community-dwelling older people: a cross-sectional study. \u003cem\u003eSao Paulo Med. J.\u003c/em\u003e \u003cb\u003e140\u003c/b\u003e (3), 422\u0026ndash;429. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1590/1516-3180.2021.0180.R1.15092021\u003c/span\u003e\u003cspan address=\"10.1590/1516-3180.2021.0180.R1.15092021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJo, K. H., Park, J. \u0026amp; Ryu, S. Y. The effects of mental health on recurrent falls among elderly adults, based on Korean Community Health Survey data. \u003cem\u003eEpidemiol. Health\u003c/em\u003e. \u003cb\u003e42\u003c/b\u003e, e2020005. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.4178/epih.e2020005\u003c/span\u003e\u003cspan address=\"10.4178/epih.e2020005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ede Souza, A. Q., Pegorari, M. S., Nascimento, J. S., de Oliveira, P. B. \u0026amp; Tavares, D. M. D. S. Incidence and predictive factors of falls in community-dwelling elderly: a longitudinal study. \u003cem\u003eCien Saude Colet\u003c/em\u003e. \u003cb\u003e24\u003c/b\u003e (9), 3507\u0026ndash;3516. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1590/1413-81232018249.30512017\u003c/span\u003e\u003cspan address=\"10.1590/1413-81232018249.30512017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHenry-S\u0026aacute;nchez, J. T., Kurichi, J. E., Xie, D., Pan, Q. \u0026amp; Stineman, M. G. Do elderly people at more severe activity of daily living limitation stages fall more? \u003cem\u003eAm. J. Phys. Med. Rehabil\u003c/em\u003e. \u003cb\u003e91\u003c/b\u003e (7), 601\u0026ndash;610. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/PHM.0b013e31825596af\u003c/span\u003e\u003cspan address=\"10.1097/PHM.0b013e31825596af\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGe, M. L., Chu, N. M., Simonsick, E. M., Kasper, J. D. \u0026amp; Xue, Q. L. Order of Onset of Physical Frailty and Cognitive Impairment and Risk of Repeated Falls in Community-Dwelling Older Adults. \u003cem\u003eJ. Am. Med. Dir. Assoc.\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e (4), 482\u0026ndash;488e4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jamda.2023.01.020\u003c/span\u003e\u003cspan address=\"10.1016/j.jamda.2023.01.020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAbdo, M. et al. Regional Differences in Risk of Recurrent Falls Among Older U.S. Women and Men with HIV in the HIV Infection, Aging, and Immune Function Long-Term Observational Study. \u003cem\u003eAIDS Res. Hum. Retroviruses\u003c/em\u003e. \u003cb\u003e38\u003c/b\u003e (7), 530\u0026ndash;537. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1089/AID.2021.0162\u003c/span\u003e\u003cspan address=\"10.1089/AID.2021.0162\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMoreland, B. L., Durbin, L. L., Kasper, J. D. \u0026amp; Mielenz, T. J. Rehabilitation Utilization for Falls Among Community-Dwelling Older Adults in the United States in the National Health and Aging Trends Study. \u003cem\u003eArch. Phys. Med. Rehabil\u003c/em\u003e. \u003cb\u003e99\u003c/b\u003e (8), 1568\u0026ndash;1575. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.apmr.2018.02.009\u003c/span\u003e\u003cspan address=\"10.1016/j.apmr.2018.02.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePesonen, T., Siira, H., V\u0026auml;is\u0026auml;nen, V., Edgren, J. \u0026amp; Elo, S. The role of impaired vision and declined cognition in falls and fall-related risk factors among older people receiving home care in Finland-a cross-sectional register study. \u003cem\u003eEur. J. Ageing\u003c/em\u003e. \u003cb\u003e22\u003c/b\u003e (1), 22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s10433-025-00860-2\u003c/span\u003e\u003cspan address=\"10.1007/s10433-025-00860-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChang, C. I., Chan, D. C., Kuo, K. N., Hsiung, C. A. \u0026amp; Chen, C. Y. Prevalence and correlates of geriatric frailty in a northern Taiwan community. \u003cem\u003eJ. Formos. Med. Assoc.\u003c/em\u003e \u003cb\u003e110\u003c/b\u003e (4), 247\u0026ndash;257. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/S0929-6646(11)60037-5\u003c/span\u003e\u003cspan address=\"10.1016/S0929-6646(11)60037-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAmbrose, A. F., Paul, G. \u0026amp; Hausdorff, J. M. Risk factors for falls among older adults: A review of the literature. \u003cem\u003eMaturitas\u003c/em\u003e \u003cb\u003e75\u003c/b\u003e (1), 51\u0026ndash;61. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.maturitas.2013.02.009\u003c/span\u003e\u003cspan address=\"10.1016/j.maturitas.2013.02.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRenner, S. W. et al. Higher Fatigue Prospectively Increases the Risk of Falls in Older Men. \u003cem\u003eInnov. Aging\u003c/em\u003e. \u003cb\u003e5\u003c/b\u003e (1), igaa061. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/geroni/igaa061\u003c/span\u003e\u003cspan address=\"10.1093/geroni/igaa061\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChoi, N. G., Choi, B. Y., DiNitto, D. M., Marti, C. N. \u0026amp; Kunik, M. E. Fall-related emergency department visits and hospitalizations among community-dwelling older adults: examination of health problems and injury characteristics. \u003cem\u003eBMC Geriatr.\u003c/em\u003e \u003cb\u003e19\u003c/b\u003e (1), 303. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12877-019-1329-2\u003c/span\u003e\u003cspan address=\"10.1186/s12877-019-1329-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eThakur, N. \u0026amp; Han, C. Y. A Study of Fall Detection in Assisted Living: Identifying and Improving the Optimal Machine Learning Method. \u003cem\u003eJ. Sens. Actuator Networks\u003c/em\u003e. \u003cb\u003e10\u003c/b\u003e (3), 39. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/jsan10030039\u003c/span\u003e\u003cspan address=\"10.3390/jsan10030039\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLeggett, A. N., Polenick, C. A., Maust, D. T. \u0026amp; Kales, H. C. Falls and Hospitalizations Among Persons With Dementia and Associated Caregiver Emotional Difficulties. \u003cem\u003eGerontologist\u003c/em\u003e \u003cb\u003e58\u003c/b\u003e (2), e78\u0026ndash;e86. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/geront/gnx202\u003c/span\u003e\u003cspan address=\"10.1093/geront/gnx202\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGambaro, E. et al. The complex associations between late life depression, fear of falling and risk of falls. A systematic review and meta-analysis. \u003cem\u003eAgeing Res. Rev.\u003c/em\u003e \u003cb\u003e73\u003c/b\u003e, 101532. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.arr.2021.101532\u003c/span\u003e\u003cspan address=\"10.1016/j.arr.2021.101532\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTchalla, A. E. et al. Patterns, Predictors, and Outcomes of Falls Trajectories in Older Adults: The MOBILIZE Boston Study with 5 Years of Follow-Up. \u003cem\u003ePLOS ONE\u003c/em\u003e. \u003cb\u003e9\u003c/b\u003e (9), e106363. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pone.0106363\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0106363\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2014).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Falls, Older Adults, Fall Trajectories, Predictors, Prevention Strategies","lastPublishedDoi":"10.21203/rs.3.rs-8132697/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8132697/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eFalls can occur unpredictably or follow patterns linked to modifiable risk factors and adverse outcomes. Identifying fall trajectories and their key predictors can help clinicians implement targeted prevention strategies. We hypothesize that distinct clinical fall trajectories exist, each with identifiable baseline predictors.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eThis seven-year prospective study followed 1,648 community-dwelling older adults (\u0026ge;\u0026thinsp;60 years). Participants were assessed at home. Data collection included cardiovascular risk factors, fall occurrences, socio-environmental characteristics and a comprehensive geriatric assessment summary score. Fall trajectories were identified using a Gaussian Mixture Model (GMM) and Multinomial Logistic Regression (MLR) determined the predictors of each trajectory. Marginal prediction allowed us to refine predictor analysis by identifying the category within each feature that contributes the most to the steadiest trajectory.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eFour distinct fall trajectories were identified during the 7 years follow up: Cluster Falls and No Falls (65.5%), Increasing Falls (6.7%), Chronic Recurring Falls (15.7%) and Low-Rate Chronic Falls (12.1%). At baseline the steadiest trajectory is Increasing Falls. Clustered Falls and No Falls trajectory is characterized by lack of leisure activities, functional impairment (Instrumental Activities of Daily Living [IADL]\u0026thinsp;\u0026lt;\u0026thinsp;8) and pathological performance on the Single Leg Balance (SLB) test. The Chronic Recurring Falls trajectory was primarily composed of women with obesity. The Low-Rate Chronic Falls group also consisted mainly of obese women with IADL\u0026thinsp;\u0026lt;\u0026thinsp;8 and pathological SLB. We also investigated predictors at the 18-month follow-up.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eFalls in older individuals may occur at discrete intervals or follow recurrent patterns, including chronic recurrence, all of which are associated with increased risks. Women, obesity, impairment in activities of daily living, reduced physical performance and depressive symptoms should be prioritized for intervention.\u003c/p\u003e","manuscriptTitle":"Identifying Fall Risk Factors: A 7-Year Home Study Using Innovative Marginal Predictions Strategy","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-11 14:30:03","doi":"10.21203/rs.3.rs-8132697/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"52285277744365831524452378931119752078","date":"2026-05-18T11:40:01+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-10T01:55:58+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"147510019555902318776984031925476374122","date":"2026-01-19T12:08:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"200275651352154755871621058344308199279","date":"2025-12-14T23:04:40+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-08T16:17:47+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-12-08T16:13:28+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-11-24T19:13:20+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-11-21T18:49:52+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-11-21T18:46:21+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"fe58fe6b-dd45-4d07-98ef-54178513c6fe","owner":[],"postedDate":"December 11th, 2025","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"52285277744365831524452378931119752078","date":"2026-05-18T11:40:01+00:00","index":191,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":59402374,"name":"Health sciences/Diseases"},{"id":59402375,"name":"Health sciences/Health care"},{"id":59402376,"name":"Health sciences/Medical research"},{"id":59402377,"name":"Health sciences/Risk factors"}],"tags":[],"updatedAt":"2025-12-11T14:30:03+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-11 14:30:03","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8132697","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8132697","identity":"rs-8132697","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}