Factors associated with the life-space assessment in community-dwelling older adults requiring support Authors: Kodai Hosaka, Makoto Uchinoura, Hiroshi Otao, Hajime Shibata Published: 2026 ABSTRACT [Purpose] We aimed to identify factors associated with life-space mobility in community-dwelling older adults certified as requiring support. [Participants and Methods] We conducted a cross-sectional study of older adults requiring support who attended a day-service program between August and September 2024. We evaluated life-space mobility using a questionnaire. We assessed physical function using a chair stand test, one-leg standing time with eyes open, and maximum gait speed. We assessed cognitive function using a dementia screening scale. We evaluated pain-related disability and lower-limb pain sites. We examined factors associated with life-space mobility. [Results] Sixty-three participants were analyzed. Life-space mobility was associated with maximum gait speed and overall pain-related disability; however, only maximum gait speed showed a stable association. Exploratory analyses suggested that activity-specific pain-related disability, such as difficulties with housework, visiting friends, using public transportation, and shopping, was negatively associated with life-space mobility. [Conclusion] Maximum gait speed was a stable factor associated with life-space mobility in older adults requiring support. Activity-specific pain-related disability may also relate to life-space restriction and should be interpreted cautiously. INTRODUCTION Globally, the population of adults aged 65 years and older is increasing rapidly, making the extension of healthy life expectancy and the prevention of long-term care dependency urgent international priorities1). In Japan, the Long-Term Care Insurance (LTCI) system was introduced in 2000 to provide appropriate assistance and care services based on the physical and mental status of older adults. Within this system, functional status is classified into seven levels: Support Required Levels 1–2 and Care Required Levels 1–52). Older adults designated as requiring support typically have mild impairments in instrumental activities of daily living (IADL) and need partial assistance or preventive care services, whereas those categorized as requiring long-term care need personal assistance in daily life2). Although older adults requiring support have relatively mild functional limitations compared with those requiring long-term care, they generally have greater potential for functional improvement through appropriate interventions. Therefore, implementing effective preventive care programs and comprehensive community-based support is essential for maintaining independence and preventing progression to long-term care dependency. It is important to understand the extent to which older adults leave their homes and engage with society in their daily lives. From this perspective, the concept of life-space—which reflects the breadth of an individual’s mobility and social participation—has gained increasing attention. Maintaining or expanding life-space is essential for preserving independence and sustaining social engagement in later life. LSA is a validated instrument designed to quantitatively evaluate life-space mobility3). Numerous studies have reported associations between LSA scores and a range of health-related outcomes among community-dwelling older adults. In particular, lower LSA scores have been linked to declines in ADL4), reduced cognitive function5), and increased mortality risk6, 7). Given these findings, evaluating the LSA among older adults requiring support is particularly important from a preventive care perspective. This population is prone to restricted life-space due to declines in physical function and reduced social activity, which may increase the risk of transitioning to long-term care dependency. Although previous studies have examined the relationships among life-space mobility, physical function, and social activity in community-dwelling older adults, research focusing specifically on those certified as requiring support remains limited. This group is especially vulnerable to restricted life-space because of functional decline and fewer opportunities to go out, yet the factors influencing their life-space mobility have not been sufficiently clarified. Furthermore, identifying determinants of life-space mobility in this population is essential for preventing long-term care dependency, but studies addressing this perspective are also scarce. Therefore, this study aimed to identify factors associated with the LSA among community-dwelling older adults requiring support. These findings are expected to provide insights that will contribute to the development of effective preventive care programs. PARTICIPANTS AND METHODS This cross-sectional study included participants who were users of a day-service program implemented as part of a municipal preventive long-term care project in Kurume City, held eight times per month (approximately once or twice per week). All participants had been certified as Support Level 1 or 2 under the Japanese long-term care insurance system, or were designated as eligible individuals under the community-based preventive care program, and had been referred through local Community Comprehensive Support Centers. Community Comprehensive Support Centers are publicly funded institutions responsible for providing preventive care coordination, comprehensive consultation services, and community support for older residents. Individuals certified as requiring support or designated as eligible for preventive services generally maintain independence in basic activities of daily living but demonstrate mild functional decline in instrumental activities of daily living. Therefore, independent ambulation was not a mandatory requirement for participation in the program. Care need levels were determined according to the certification criteria of Japan’s Long-Term Care Insurance system8). The Long-Term Care Insurance system was introduced in 2000 in response to rapid population aging1). Certification is based on a standardized questionnaire administered by trained assessors and a physician’s written opinion, evaluating physical function, cognitive function, health status, family support, and living environment. Based on these assessments, the municipal Long-Term Care Certification Committee classifies individuals into Support Levels 1–2 or Care Need Levels 1–58, 9). Individuals classified as requiring support (Support Levels 1–2) are generally independent in basic activities of daily living but demonstrate mild decline in instrumental activities of daily living (2) and are considered at risk of progressing to a higher level of care need. In contrast, individuals classified as requiring long-term care (Care Need Levels 1–5) require assistance with basic activities of daily living and typically exhibit more pronounced physical and/or cognitive impairments, necessitating ongoing formal care services. This study included only older adults certified as requiring support. Exclusion criteria were (1) age below 65 years and (2) missing data for any assessment item. All participants received detailed verbal and written explanations of the study procedures and provided voluntary informed consent (Fig. 1Fig. 1. Flow diagram of participant selection.). Flow diagram of participant selection. The study protocol was approved by the Ethics Committee of Kurume Rehabilitation Hospital, Kabutoyama Medical Corporation (Approval No. 21-001) and was conducted in accordance with the principles of the Declaration of Helsinki. The variables measured in this study included age, sex, and support certification level (Support Level 1 or 2). Physical function was assessed using the 30-Second Chair Stand Test (CS-30), one-leg standing time with eyes open, and gait speed. Cognitive function was evaluated using the Hasegawa Dementia Scale-Revised (HDS-R). Pain-related variables were assessed using the Pain Disability Assessment Scale (PDAS) and by counting the number of painful sites in the lower limbs. Life-space mobility was measured using the self-administered LSA questionnaire. Lower-limb muscle strength was assessed using the CS-30. Participants sat on a chair without armrests (seat height: 40 cm), crossed their arms over the chest, and repeatedly stand up and sat down as many times as possible within 30 s. The number of repetitions was recorded. The CS-30 has been reported to have high reliability and validity as an indicator of lower-limb strength10). One-leg standing time with eyes open was measured using a digital stopwatch. Participants were instructed to maintain a single-leg stance. The test was terminated if (1) both feet touched the floor, (2) assistance was required to maintain balance, or (3) the raised leg touched the supporting leg. The test was performed once for each leg, and the longer time of the two legs was used for analysis11). Gait speed was measured with a digital stopwatch. Participants were instructed to walk as quickly as possible along a 7-m walkway, and the time required to walk the central 5 m was recorded. Measurement of gait speed using a digital stopwatch has been shown to have high reliability. Participants completed one trial each of usual gait (walking at their self-selected pace) and fast gait (walking as quickly as possible). The latter was defined as maximum gait speed and used for the primary analyses. Participants were permitted to use their customary assistive devices, such as a cane, during the assessment of maximum gait speed. Cognitive function was assessed using the HDS-R, a brief screening tool consisting of nine items with a maximum score of 30 points. A higher HDS-R score indicates better cognitive function12). Pain-related disability was evaluated using the PDAS, a self-administered questionnaire comprising 18 items. Higher PDAS scores indicate greater disability due to chronic pain13). Additionally, the number of painful sites in the lower limbs was determined through face-to-face interviews. Participants were asked to identify areas of pain in their lower limbs, and the total number of reported painful sites was recorded. Life-space mobility was evaluated using the LSA questionnaire, which measures the extent of an individual’s movement over the past 4 weeks. The LSA quantifies mobility based on the spatial range of movement—from within one’s room to beyond the neighborhood—along with the frequency of outings and the level of independence (i.e., use of assistance) during movement. Total scores range from 0 to 120, with higher scores reflecting a wider life-space and greater independence in mobility3). Statistical analysis was performed with LSA as the dependent variable using multiple linear regression with the forced entry (enter) method. Independent variables included the CS-30, open-eye one-leg standing time, maximum gait speed, HDS-R score, PDAS total score, and number of lower limb pain sites. Support level (Level 1 or 2), sex, and age were included as covariates. Sex and support level were entered as dummy variables (sex: 0=men [reference], 1=women; support level: 0=Support Level 1 [reference], 1=Support Level 2). The regression model included nine parameters in total (six predictors: CS-30, open-eye one-leg standing time, maximum gait speed, HDS-R score, PDAS total score, and number of lower-limb pain sites; and three covariates: support level, sex, and age). With a sample size of 63 participants, this corresponded to approximately seven participants per parameter. Given this ratio, a potential risk of overfitting was acknowledged, and bootstrap resampling with 1,000 repetitions was performed to assess the internal stability and robustness of the regression coefficients. Model fit, residual autocorrelation, and multicollinearity were assessed using the coefficient of determination (R2), the Durbin–Watson statistic, and the variance inflation factor (VIF), respectively. Assumptions of linear regression were examined using residual plots, the Shapiro–Wilk test, and Cook’s distance. For the exploratory item-level analyses of the 20 PDAS items, associations with LSA were assessed using Spearman’s rank correlation coefficients. Given the multiple comparisons performed, p-values were adjusted using the Benjamini–Hochberg false discovery rate (FDR) procedure. A two-tailed p-value <0.05 was considered statistically significant. All statistical analyses were conducted using EZR (version 2.9-1). RESULTS A total of 63 participants were included in the analysis. The mean age was 83.3 ± 6.3 years. Thirteen participants (20.6%) were male and 50 (79.4%) were female. Regarding care support levels, 45 participants (71.4%) were classified as Support Level 1 and 18 (28.6%) as Support Level 2 (Table 1Table 1. Physical functions and characteristics of the participantsOverall (n=63)Male (n=13)Female (n=50)Age (years)83.3 ± 6.381.8 ± 7.583.6 ± 6.0 aNeed for long-term care support (support level 1: support level 2) (n)63 (45:18)13 (9:4)50 (36:14) bCS-30 (repetitions)17.7 ± 6.416.5 ± 6.018.0 ± 6.5 aOne-leg standing time (seconds)22.5 ± 24.827.5 ± 31.521.2 ± 23.0 aMaximum gait speed (m/s)1.2 ± 0.31.4 ± 0.41.1 ± 0.3** aHDS-R (points)27.3 ± 3.226.6 ± 3.027.5 ± 3.2 aPDAS (points)24.6 ± 10.922.4 ± 10.325.2 ± 11.0 aNumber of lower limb pain sites (sites)1.0 ± 1.20.8 ± 1.51.0 ± 1.2 aLSA (points)53.3 ± 19.763.6 ± 18.750.7 ± 19.3* aData are presented as mean ± SD, a t-test; b Fisher’s exact test. Support Levels are part of the Japanese Long-Term Care Insurance (LTCI) system, corresponding to “yō-shien” (requiring support). *p<0.05, **p<0.01, Sex (male=0, female=1) and support level (Level 1=0, Level 2=1) were dummy coded for regression analyses. CS-30: 30-second chair stand test; HDS-R: Hasegawa dementia scale-revised; PDAS: pain disability assessment scale; LSA: life-space assessment.). Table 1. Physical functions and characteristics of the participantsOverall (n=63)Male (n=13)Female (n=50)Age (years)83.3 ± 6.381.8 ± 7.583.6 ± 6.0 aNeed for long-term care support (support level 1: support level 2) (n)63 (45:18)13 (9:4)50 (36:14) bCS-30 (repetitions)17.7 ± 6.416.5 ± 6.018.0 ± 6.5 aOne-leg standing time (seconds)22.5 ± 24.827.5 ± 31.521.2 ± 23.0 aMaximum gait speed (m/s)1.2 ± 0.31.4 ± 0.41.1 ± 0.3** aHDS-R (points)27.3 ± 3.226.6 ± 3.027.5 ± 3.2 aPDAS (points)24.6 ± 10.922.4 ± 10.325.2 ± 11.0 aNumber of lower limb pain sites (sites)1.0 ± 1.20.8 ± 1.51.0 ± 1.2 aLSA (points)53.3 ± 19.763.6 ± 18.750.7 ± 19.3* aData are presented as mean ± SD, a t-test; b Fisher’s exact test. Support Levels are part of the Japanese Long-Term Care Insurance (LTCI) system, corresponding to “yō-shien” (requiring support). *p<0.05, **p<0.01, Sex (male=0, female=1) and support level (Level 1=0, Level 2=1) were dummy coded for regression analyses. CS-30: 30-second chair stand test; HDS-R: Hasegawa dementia scale-revised; PDAS: pain disability assessment scale; LSA: life-space assessment. Physical functions and characteristics of the participants Data are presented as mean ± SD, a t-test; b Fisher’s exact test. Support Levels are part of the Japanese Long-Term Care Insurance (LTCI) system, corresponding to “yō-shien” (requiring support). *p<0.05, **p<0.01, Sex (male=0, female=1) and support level (Level 1=0, Level 2=1) were dummy coded for regression analyses. CS-30: 30-second chair stand test; HDS-R: Hasegawa dementia scale-revised; PDAS: pain disability assessment scale; LSA: life-space assessment. Multiple linear regression analysis was performed with LSA as the dependent variable and the measured evaluation items as independent variables. Maximum gait speed (standardized β=0.53, 95% CI: 0.25–0.82, p=0.001) and PDAS total score (standardized β=−0.25, 95% CI: −0.49–−0.02, p=0.037) were significantly associated with LSA. The model showed a coefficient of determination (R2) of 0.41 and a Durbin–Watson statistic of 2.18. Multicollinearity was assessed using the variance inflation factor (VIF), and no variable exceeded a value of 5. Residual diagnostics indicated no substantial violation of model assumptions. The Shapiro–Wilk test showed no significant deviation from normality (p=0.08). Cook’s distance values ranged from 0.02 to 0.15. Although several observations exceeded the conventional 4/n threshold, all values remained well below 1.0, a commonly cited indicator of substantial influence. Sensitivity analyses excluding observations with relatively higher Cook’s distance values did not materially alter the regression coefficients or their statistical significance. Bootstrap resampling (1,000 repetitions) demonstrated that the association between maximum gait speed and LSA remained stable, whereas the association between PDAS total score and LSA was attenuated, with the upper bound of the confidence interval reaching 0.00 (Table 2Table 2. Factors associated with the life-space assessment in community-dwelling older adults requiring supportOriginal analysisBootstrap analysisStandardized coefficientStandard error95% CI (Lower)95% CI (Upper)VIFStandardized coefficientStandard error95% CI (Lower)95% CI (Upper)CS-30−0.140.12−0.370.101.41−0.130.13−0.390.11One-leg standing time−0.020.11−0.250.211.360.010.12−0.240.23Maximum gait speed0.530.140.250.82 **2.110.540.140.270.81 **HDS-R0.180.11−0.040.401.230.190.12−0.040.42PDAS−0.250.12−0.49−0.02 *1.46−0.260.14−0.520.00Number of lower limb pain sites0.080.10−0.130.291.130.080.12−0.120.34Need for long-term care support (reference: support level 1)−0.170.10−0.380.041.13−0.170.10−0.360.04Sex (reference: male)−0.080.11−0.300.141.30−0.080.10−0.280.13Age−0.060.10−0.270.151.12−0.050.11−0.360.04Forced entry multiple regression analysis (dependent variable: life-space assessment), ANOVA p<0.001. R2=0.41, Durbin–Watson ratio=2.18. *p<0.05, **p<0.01. In regression analyses, sex (male=0, female=1) and support level (Level 1=0, Level 2=1) were coded as dummy variables. CI: confidence interval; CS-30: 30-second chair stand test; HDS-R: Hasegawa dementia scale-revised; PDAS: pain disability assessment scale; VIF: variance inflation factor.). Table 2. Factors associated with the life-space assessment in community-dwelling older adults requiring supportOriginal analysisBootstrap analysisStandardized coefficientStandard error95% CI (Lower)95% CI (Upper)VIFStandardized coefficientStandard error95% CI (Lower)95% CI (Upper)CS-30−0.140.12−0.370.101.41−0.130.13−0.390.11One-leg standing time−0.020.11−0.250.211.360.010.12−0.240.23Maximum gait speed0.530.140.250.82 **2.110.540.140.270.81 **HDS-R0.180.11−0.040.401.230.190.12−0.040.42PDAS−0.250.12−0.49−0.02 *1.46−0.260.14−0.520.00Number of lower limb pain sites0.080.10−0.130.291.130.080.12−0.120.34Need for long-term care support (reference: support level 1)−0.170.10−0.380.041.13−0.170.10−0.360.04Sex (reference: male)−0.080.11−0.300.141.30−0.080.10−0.280.13Age−0.060.10−0.270.151.12−0.050.11−0.360.04Forced entry multiple regression analysis (dependent variable: life-space assessment), ANOVA p<0.001. R2=0.41, Durbin–Watson ratio=2.18. *p<0.05, **p<0.01. In regression analyses, sex (male=0, female=1) and support level (Level 1=0, Level 2=1) were coded as dummy variables. CI: confidence interval; CS-30: 30-second chair stand test; HDS-R: Hasegawa dementia scale-revised; PDAS: pain disability assessment scale; VIF: variance inflation factor. Factors associated with the life-space assessment in community-dwelling older adults requiring support Forced entry multiple regression analysis (dependent variable: life-space assessment), ANOVA p<0.001. R2=0.41, Durbin–Watson ratio=2.18. *p<0.05, **p<0.01. In regression analyses, sex (male=0, female=1) and support level (Level 1=0, Level 2=1) were coded as dummy variables. CI: confidence interval; CS-30: 30-second chair stand test; HDS-R: Hasegawa dementia scale-revised; PDAS: pain disability assessment scale; VIF: variance inflation factor. As a secondary exploratory analysis, associations between LSA and each PDAS item were examined using Spearman’s rank correlations. To account for multiplicity across the 18 item-level tests, p-values were adjusted using the Benjamini–Hochberg false discovery rate (FDR) procedure. After FDR adjustment, the items that remained statistically significant (q<0.05) were: vacuuming/yard work/housework (ρ=−0.31), climbing stairs (ρ=−0.29), visiting friends (ρ=−0.32), using public transportation (ρ=−0.38), going to a restaurant or café (ρ=−0.37), carrying heavy objects (ρ=−0.33), standing still (ρ=−0.31), walking on a flat surface (ρ=−0.31), jogging slowly (ρ=−0.42), and shopping (ρ=−0.42) (Table 3Table 3. Correlations between the life-space assessment (LSA) and pain disability assessment scale (PDAS) itemsρVacuuming, yard work, or other housework−0.31*aJogging slowly−0.42***bBending over to pick up an object from the floor−0.06Shopping−0.42***bClimbing up and down stairs−0.29*aVisiting friends−0.32*aUsing public transportation (bus or train)−0.38**aGoing to a restaurant or café−0.37**aCarrying heavy objects−0.33**aCooking or washing dishes−0.06Bending or stretching the body−0.24Reaching up to take down a heavy object (e.g., a bag of sugar) from a shelf−0.23Washing or drying the body0.06Sitting down on or rising from a toilet seat−0.05Getting into or out of bed (or from the floor)−0.08Opening or closing a car door0.20Standing still−0.31*aWalking on a flat surface−0.31*aEngaging in hobbies−0.12Washing hair−0.12Spearman’s rank correlation coefficient. *p<0.05, **p<0.01, ***p<0.001 indicate unadjusted p-values and are shown for reference only. Primary statistical significance is based on the FDR-adjusted thresholds: a FDR-adjusted p<0.05; b FDR-adjusted p<0.01 (Benjamini–Hochberg procedure). FDR: false discovery rate.). Table 3. Correlations between the life-space assessment (LSA) and pain disability assessment scale (PDAS) itemsρVacuuming, yard work, or other housework−0.31*aJogging slowly−0.42***bBending over to pick up an object from the floor−0.06Shopping−0.42***bClimbing up and down stairs−0.29*aVisiting friends−0.32*aUsing public transportation (bus or train)−0.38**aGoing to a restaurant or café−0.37**aCarrying heavy objects−0.33**aCooking or washing dishes−0.06Bending or stretching the body−0.24Reaching up to take down a heavy object (e.g., a bag of sugar) from a shelf−0.23Washing or drying the body0.06Sitting down on or rising from a toilet seat−0.05Getting into or out of bed (or from the floor)−0.08Opening or closing a car door0.20Standing still−0.31*aWalking on a flat surface−0.31*aEngaging in hobbies−0.12Washing hair−0.12Spearman’s rank correlation coefficient. *p<0.05, **p<0.01, ***p<0.001 indicate unadjusted p-values and are shown for reference only. Primary statistical significance is based on the FDR-adjusted thresholds: a FDR-adjusted p<0.05; b FDR-adjusted p<0.01 (Benjamini–Hochberg procedure). FDR: false discovery rate. Correlations between the life-space assessment (LSA) and pain disability assessment scale (PDAS) items Spearman’s rank correlation coefficient. *p<0.05, **p<0.01, ***p<0.001 indicate unadjusted p-values and are shown for reference only. Primary statistical significance is based on the FDR-adjusted thresholds: a FDR-adjusted p<0.05; b FDR-adjusted p<0.01 (Benjamini–Hochberg procedure). FDR: false discovery rate. DISCUSSION In this study, we examined the associations between life-space mobility and physical function among community-dwelling older adults certified as requiring support. Life-space mobility was associated with maximum gait speed and the total Pain Disability Assessment Scale score. However, in bootstrap analyses conducted to assess the internal stability of the estimates (1,000 resamples), only maximum gait speed consistently retained its association with life-space mobility, whereas the association with the total Pain Disability Assessment Scale score was not stably reproduced. These findings suggest that maximum gait speed was the more robust correlate of life-space mobility in this sample. Maximum gait speed remained the only physical function measure consistently associated with LSA after stability assessment. Gait speed is considered an integrative indicator reflecting lower-extremity strength, balance, coordination, and cardiopulmonary function, and has been described as a “sixth vital sign”14). Previous studies have demonstrated that slower gait speed is associated with falls15), frailty16), and mortality16, 17), underscoring its value as a global marker of health status in older adults. In community-dwelling populations, gait speed has also been linked to frequency of outings and extent of mobility18). The participants in this study—older adults requiring support—represent an intermediate stage between full independence and long-term care dependency. At this stage, basic activities of daily living are generally preserved; however, even subtle declines in physical capacity may be associated with reductions in activity range. Reduced maximum gait speed may reflect decreased mobility efficiency and increased physical burden during ambulation, which in turn may be related to more restricted LSA. Thus, maximum gait speed may capture not only locomotor capacity but also broader levels of community engagement in this population. We also observed an association between LSA and the total PDAS score. The PDAS evaluates disability attributable to chronic pain in daily and social activities13). However, this association was attenuated and not consistently maintained in bootstrap analyses, indicating that it should be interpreted with caution. In exploratory item-level analyses, pain-related disability associated with activities such as housework, visiting friends, using public transportation, going to restaurants or cafés, carrying heavy objects, maintaining standing posture, and walking on level ground showed negative correlations with LSA. Relatively stronger associations were observed for jogging slowly and shopping. These findings suggest that activity-specific pain-related disability, particularly in activities involving mobility or physical demand, may be associated with LSA. Because the PDAS includes both indoor and outdoor activities, aggregating heterogeneous contexts into a single total score may not fully capture aspects most directly related to community mobility. Nevertheless, these item-level findings should be considered exploratory and hypothesis-generating. Previous studies have reported that declines in ADL and IADL are associated with reduced life-space mobility19,20,21), and our findings are broadly consistent with this literature. A distinctive feature of this study is its exploratory focus on specific domains of pain-related disability. Future large-scale and longitudinal studies are warranted to further examine these relationships. This study has several limitations. First, participants were recruited from a single day-service preventive care program and were relatively active older adults requiring support who were independent in basic ADL. Selection bias may therefore be present, and the findings should be generalized with caution. Second, due to the cross-sectional design, causal relationships between LSA, maximum gait speed, and pain-related disability cannot be inferred. Third, the relatively small sample size and inclusion of multiple explanatory variables raise the possibility of overfitting. Although bootstrap resampling was used to assess internal stability, the association with the total PDAS score was not consistently reproduced. Fourth, the PDAS item-level analyses were exploratory, and the influence of multiple testing cannot be entirely excluded. Fifth, maximum gait speed was assessed using a single fast-gait trial. This approach may raise concerns regarding measurement reliability in very old adults with a mean age exceeding 80 years; therefore, the results should be interpreted with caution. Sixth, social and environmental factors were not fully evaluated; thus, residual confounding cannot be ruled out. Conflict of interest No potential conflict of interest was reported by the authors.