Abstract
Background
Intrinsic capacity (IC), defined as the composite of physical and mental capacities, is a dynamic indicator of healthy ageing. Understanding its longitudinal patterns may improve risk stratification and inform care planning among home care recipients.
Methods
This population-based retrospective cohort study used routinely collected interRAI Home Care data in Canada. IC was derived across five domains and modelled using group-based trajectory modelling to identify distinct patterns over time. Associations between trajectory membership and key outcomes, including mortality and long-term care (LTC) admission, were examined.
Results
The study included 302 467 home care recipients; 62.2% were female, with a median age of 81 years (IQR 14). Five distinct IC trajectories were identified, demonstrating substantial heterogeneity in functional ageing. Most individuals followed declining trajectories, while smaller groups remained stable or showed improvement. Cognitive impairment was strongly associated with adverse trajectories, whereas gains were most evident in locomotion and psychological domains. Importantly, baseline IC level alone did not fully capture subsequent outcome heterogeneity. For example, individuals with moderate baseline IC had higher LTC admission than those with low IC who improved (48.4% vs 38.1%), underscoring the importance of longitudinal assessment.
Conclusion
IC trajectories capture clinically meaningful differences in ageing pathways that are not apparent from single assessments. These findings support the use of routinely collected interRAI data to enable scalable, trajectory-informed care, with implications for earlier intervention, targeted service allocation and improved planning across the continuum of care. Further work is required to translate these findings into practical clinical tools to support early risk stratification at admission and guide future care planning.
Keywords: intrinsic capacity, home care, interRAI, ageing trajectory, ICOPE, older people
Key Points
Intrinsic capacity trajectories among home care recipients cluster into five distinct longitudinal patterns.
interRAI Home Care assessments enable routine, multidomain measurement of intrinsic capacity in real-world settings.
Declining intrinsic capacity trajectories are associated with increased mortality risk in home care recipients.
Adverse intrinsic capacity trajectories are associated with higher likelihood of long-term care admission.
Longitudinal intrinsic capacity profiling supports population-level care planning and health system decision-making.
Introduction
In the WHO Healthy Ageing framework [1, 2] intrinsic capacity (IC) is defined as an individual’s physical and mental abilities across five domains: locomotion, cognition, vitality, sensory function and psychological well-being [3]. IC offers a holistic, person-centred view of ageing that shifts focus from deficit accumulation to functional resilience. Changes in IC may offer substantial predictive value for early identification of health decline. Trajectory-based analyses have demonstrated strong prognostic value in related domains of functional decline, including among nursing home residents, where distinct functional trajectories were shown to meaningfully predict adverse outcomes [4]. Building on this evidence, intrinsic capacity trajectories may offer an even more holistic representation of functional resilience. To date, however, most research on IC has focused on community-dwelling adults, leaving a critical gap in evidence for home health care recipients, a clinically more complex population at elevated risk of health deterioration, emergency department use and institutionalisation.
Home care generally refers to health and supportive services that help individuals maintain independence at home, aligning with most older adults’ preference to age in place [5–7]. Its scale is substantial: ~1.4 million Canadians [8], 5.9 million Europeans [9] and up to 12 million people in the US receiving formal home care services. Given the strong link between baseline health, functional trajectories and downstream outcomes in this population [10–13], characterising their changes in relevant health indices is essential.
Frailty and multimorbidity measures are commonly used to describe health status among home care recipients [14–17]. While widely applied, these measures are primarily based on deficit accumulation and may be limited in their ability to capture positive aspects of health and functional reserve [1, 2]. In contrast, health models centred on individual strengths such as intrinsic capacity are central to the WHO Healthy Ageing framework [2]. Emerging evidence suggests that IC has value in predicting adverse outcomes and informing care planning [18–21]. Understanding longitudinal changes in IC may support more person-centred care planning and contribute to the promotion of healthy ageing [22].
Prior studies describe heterogeneous trajectories of functional decline in home care populations, but these reflect downstream performance rather than underlying reserve. Because IC captures the multidomain reserves that precede functional loss, tracking these trajectories may identify risk earlier. While research examining IC trajectories is expanding [23–31], it has not been examined in home care populations using group-based trajectory modelling.
Scalable implementation of IC as a routine health measure in home care requires data that is systematically collected without imposing additional burden on care providers, an infrastructure that is lacking in many health systems. The interRAI assessment system, used internationally for routine assessment of home and community health care recipients, offers a practical solution [32–36], as it includes standardized suite of instruments and validated items [33, 37, 38]. In particular, the interRAI Home Care instrument contains validated items across all five IC domains, enabling standardized home health care level IC measurement.
This study operationalizes intrinsic capacity among home health care recipients in Canada and examines its longitudinal evolution using routinely collected interRAI Home Care data. It evaluates the association between IC trajectories and key outcomes, including mortality and long-term care admission, generating new population-based evidence to inform care planning and health system decision-making.
Methods
Study design and population
We conducted a retrospective longitudinal cohort study of Canadian home care recipients aged ≥18 years assessed between 2010 and 2019 using the interRAI Home Care (interRAI-HC) instrument. Eligible individuals required at least three repeated assessments (completed at intake, semi-annually or upon significant health changes), consistent with latent class growth modelling requirements. Data were sourced from the Canadian Institute for Health Information’s Home Care Reporting System. Ethics approval was granted by the University of Waterloo Office of Research Ethics (ORE# 30173).
In Canada, publicly funded home care supports individuals requiring assistance due to functional impairment, chronic illness, disability or recovery from acute events. Recipients are heterogeneous, spanning younger adults with disabilities to older adults with frailty or multimorbidity. In Ontario, all clients are assessed using the interRAI Home Care (interRAI-HC) at intake and at least every 6 months or following significant change. These repeated assessments form the longitudinal data used in this study. Services may be short-term or ongoing, resulting in variable follow-up.
Outcome of interest—IC
The primary outcome was IC, operationalised as the unweighted sum of five WHO domains [3] mapped to interRAI-HC items (locomotion, vitality, cognition, psychological, sensory; range 0–15; higher scores indicate greater capacity; see Appendix 1).
Independent variables
Baseline predictors of trajectory membership included sociodemographic factors (age, sex, marital status and living arrangement) and clinical conditions (e.g. diabetes, hypertension, heart failure, dementia, Parkinson’s, chronic obstructive pulmonary disease (COPD), hip fracture, schizophrenia, as well as clinical summary scales such as the Changes in Health, End-stage disease, Signs and Symptoms (CHESS) health instability scale [39]. See full list in Appendix 2.
Statistical analysis
Analyses were performed in SAS v9.4 (SAS Institute, Cary, NC). Baseline characteristics were summarized using frequencies (%) and medians (IQR) and compared using chi-square and Wilcoxon rank-sum tests.
Group-based trajectory modelling [40–44] identified IC trajectories using a censored normal model to account for floor and ceiling effects. Model selection relied on Bayesian Information Criterion (ΔBIC >2) [45], average posterior probabilities ≥0.70 and odds of correct classification >5.0. Attrition (death, institutionalization, or discharge) was handled via the DROPOUT extension in PROC TRAJ [46]. Multivariable logistic regression identified predictors of trajectory membership.
Five-year mortality was estimated using Kaplan–Meier curves (log-rank test). Long-term care admission was evaluated using cumulative incidence functions with death as a competing risk (Grey’s test), supplemented by multivariable Cox and Fine-Grey models.
Sensitivity analysis
Two sensitivity analyses were conducted: (i) comparing baseline characteristics and outcomes of included individuals against those with fewer than three assessments and (ii) re-estimating trajectories using a Percentage of Maximum Possible (POMP)-standardised IC score (0–100).
Results
The study sample comprised 302 467 home care recipients. Overall, 62.2% were female, and the median age was 81 years (IQR 14). The median baseline intrinsic capacity (IC) score was 11 (IQR 2.0), similar between male and females. Mean baseline IC was broadly similar across age groups, ranging from 10.4 among adults aged 90+ to 10.8 among those aged 60–69 years. Baseline characteristics are shown in Table 1.
Table 1.
Baseline characteristics of study participants.
| Variable | All n (%) |
Group 1 Low initial rapid decline n (9.1%) |
Group 2 Moderate rapid initial declined n (24.7%) |
Group 3 Low initial improved n (12.4%) |
Group 4 High initial rapid decline n (34.7%) |
Group 5 High initial stable n (19.1%) |
P value Chi-square test |
|---|---|---|---|---|---|---|---|
| Age group | |||||||
| <60 | 25 294 (8.4%) | 2090 (7.7%) | 3509 (10.2%) | 7765 (7.0%) | 3855 (5.2%) | 8075 (14.8%) | <.0001 |
| 60–69 | 31 322 (10.4%) | 2133 (7.8%) | 4079 (11.8%) | 10 959 (9.8%) | 5897 (7.9%) | 8254 (15.2%) | |
| 70–79 | 74 132 (24.5%) | 6175 (22.6%) | 8598 (24.9%) | 27 634 (24.7%) | 17 198 (23.1%) | 14 527 (26.7%) | |
| 80–89 | 132 188 (43.7%) | 12 233 (44.8%) | 14 153 (41%) | 50 829 (45.5%) | 35 201 (47.2%) | 19 772 (36.3%) | |
| 90+ | 39 531 (13.1%) | 4656 (17.1%) | 4216 (12.2%) | 14 475 (13%) | 12 388 (16.6%) | 3796 (7.0%) | |
| Sex | |||||||
| F | 188 124 (62.2%) | 16 685 (61.1%) | 21 371 (61.8%) | 70 328 (63%) | 45 443 (61%) | 34 297 (63%) | <.0001 |
| M | 114 322 (37.8%) | 10 600 (38.8%) | 13 181 (38.1%) | 41 327 (37%) | 29 092 (39%) | 20 122 (37%) | |
| Living arrangement | |||||||
| Alone | 3014 (1%) | 132 (0.5%) | 180 (0.5%) | 1083 (1%) | 487 (0.7%) | 1132 (2.1%) | <.0001 |
| With spouse only | 108 833 (36%) | 5386 (19.7%) | 11 292 (32.7%) | 44 759 (40.1%) | 22 777 (30.6%) | 24 619 (45.2%) | |
| With spouse + others | 94 561 (31.3%) | 9505 (34.8%) | 10 737 (31.1%) | 33 764 (30.2%) | 26 191 (35.1%) | 14 364 (26.4%) | |
| With child | 21 637 (7.2%) | 3094 (11.3%) | 2831 (8.2%) | 6714 (6%) | 5871 (7.9%) | 3127 (5.7%) | |
| With others | 38 756 (12.8%) | 5310 (19.5%) | 5064 (14.7%) | 12 912 (11.6%) | 10 917 (14.6%) | 4553 (8.4%) | |
| Group setting | 19 776 (6.5%) | 2082 (7.6%) | 2595 (7.5%) | 6768 (6.1%) | 4393 (5.9%) | 3938 (7.2%) | |
| Marital status | |||||||
| Single | 22 047 (7.3%) | 1722 (6.3%) | 2632 (7.6%) | 7783 (7.0%) | 3630 (4.9%) | 6280 (11.5%) | <.0001 |
| Married | 111 453 (36.8%) | 12 335 (45.2%) | 1353 (38.6%) | 38 831 (34.8%) | 30 785 (41.3%) | 16 149 (29.7%) | |
| Widowed | 115 105 (38.1%) | 10 066 (36.9%) | 12 814 (37.1%) | 44 719 (40%) | 29 464 (39.5%) | 18 042 (33.2%) | |
| Separated | 7942 (2.6%) | 534 (2.0%) | 1033 (3%) | 2874 (2.6%) | 1755 (2.4%) | 1746 (3.2%) | |
| Divorced | 19 240 (6.4%) | 1096 (4.0%) | 2429 (7%) | 7409 (6.6%) | 3729 (5.0%) | 4577 (8.4%) | |
| Other | 3961 (1.3%) | 332 (1.2%) | 528 (1.5%) | 1443 (1.3%) | 817 (1.1%) | 841 (1.5%) | |
| (For all IC domains, higher score is better) | |||||||
| Cognition (IC domain) | |||||||
| 0 | 7539 (2.5%) | 3729 (13.7%) | 1504 (4.4%) | 625 (0.6%) | 1598 (2.1%) | 83 (0.2%) | <.0001 |
| 1 | 24 980 (8.3%) | 6360 (23.3%) | 4982 (14.4%) | 5217 (4.7%) | 7485 (10%) | 936 (1.7%) | |
| 2 | 170 969 (56.5%) | 15 187 (55.7%) | 23 363 (67.6%) | 64 167 (57.5%) | 48 406 (64.9%) | 19 846 (36.5%) | |
| 3 | 98 979 (32.7%) | 2011 (7.4%) | 4706 (13.6%) | 41 653 (37.3%) | 17 050 (22.9%) | 33 559 (61.7%) | |
| Depression (IC domain) | |||||||
| 0 | 19 688 (6.5%) | 6346 (23.3%) | 6745 (19.5%) | 2021 (1.8%) | 3734 (5.0%) | 842 (1.5%) | <.0001 |
| 1 | 40 359 (13.3%) | 7652 (28.0%) | 9939 (28.8%) | 8984 (8.0%) | 10 685 (14.3%) | 3099 (5.7%) | |
| 2 | 75 151 (24.8%) | 7035 (25.8%) | 10 218 (29.6%) | 26 457 (23.7%) | 21 864 (29.3%) | 9577 (17.6%) | |
| 3 | 167 268 (55.3%) | 6254 (22.9%) | 7652 (22.1%) | 74 200 (66.5%) | 38 256 (51.3%) | 40 906 (75.2%) | |
| Sensory (IC domain) | |||||||
| 0 | 343 (0.1%) | 197 (0.7%) | 63 (0.2%) | 20 (0%) | 62 (0.1%) | 1 (0.0%) | <.0001 |
| 1 | 57 120 (18.9%) | 13 286 (48.7%) | 11 213 (32.4%) | 12 173 (10.9%) | 19 092 (25.6%) | 1356 (2.5%) | |
| 2 | 101 063 (33.4%) | 9457 (34.7%) | 14 145 (40.9%) | 38 417 (34.4%) | 29 392 (39.4%) | 9652 (17.7%) | |
| 3 | 143 941 (47.6%) | 4347 (15.9%) | 9134 (26.4%) | 61 052 (54.7%) | 25 993 (34.9%) | 43 415 (79.8%) | |
| Vitality (IC domain) | |||||||
| 0 | 10 446 (3.5%) | 2681 (9.8%) | 3507 (10.1%) | 1658 (1.5%) | 2050 (2.8%) | 550 (1.0%) | <.0001 |
| 1 | 71 018 (23.5%) | 9696 (35.5%) | 13 651 (39.5%) | 21 652 (19.4%) | 18 804 (25.2%) | 7215 (13.3%) | |
| 2 | 184 395 (61.0%) | 13 921 (51.0%) | 16 352 (47.3%) | 72 958 (65.3%) | 46 646 (62.6%) | 34 518 (63.4%) | |
| 3 | 36 607 (12.1%) | 989 (3.6%) | 1044 (3.0%) | 15 394 (13.8%) | 7039 (9.4%) | 12 141 (22.3%) | |
| Locomotion (IC domain) | |||||||
| 0 | 4226 (1.4%) | 2195 (8.0%) | 1024 (3.0%) | 299 (0.3%) | 568 (0.8%) | 140 (0.3%) | <.0001 |
| 1 | 9102 (3.0%) | 2612 (9.6%) | 2343 (6.8%) | 1580 (1.4%) | 1949 (2.6%) | 618 (1.1%) | |
| 2 | 66 212 (21.9%) | 11 635 (42.6%) | 13 260 (38.4%) | 17 151 (15.4%) | 18 700 (25.1%) | 5466 (10%) | |
| 3 | 222 926 (73.7%) | 10 845 (39.7%) | 17 927 (51.9%) | 92 632 (83.0%) | 53 322 (71.5%) | 48 200 (88.6%) | |
| CHESS | |||||||
| 0 | 68 232 (22.6%) | 3128 (11.5%) | 3531 (10.2%) | 28 608 (25.6%) | 14 944 (20%) | 18 021 (33.1%) | <.0001 |
| 1–2 | 181 616 (60.0%) | 14 664 (53.7%) | 19 611 (56.8%) | 69 249 (62.0%) | 46 034 (61.8%) | 32 058 (58.9%) | |
| 3+ | 52 618 (17.4%) | 9495 (34.8%) | 11 412 (33.0%) | 13 805 (12.4%) | 13 561 (18.2%) | 4345 (8.0%) | |
| IADL | |||||||
| 0 | 6106 (2.0%) | 77 (0.3%) | 146 (0.4%) | 2184 (2.0%) | 764 (1.0%) | 2935 (5.4%) | <.0001 |
| 1 | 8990 (3.0%) | 101 (0.4%) | 303 (0.9%) | 3744 (3.4%) | 1222 (1.6%) | 3620 (6.7%) | |
| 2 | 41 770 (13.8%) | 709 (2.6%) | 2301 (6.7%) | 18 355 (16.4%) | 7081 (9.5%) | 13 324 (24.5%) | |
| 3 | 34 790 (11.5%) | 1410 (5.2%) | 3103 (9.0%) | 15 387 (13.8%) | 8372 (11.2%) | 6518 (12.0%) | |
| 4 | 46 310 (15.3%) | 1958 (7.2%) | 4491 (13.0%) | 19 444 (17.4%) | 10 686 (14.3%) | 9731 (17.9%) | |
| 5 | 101 940 (33.7%) | 8847 (32.4%) | 13 882 (40.2%) | 37 110 (33.2%) | 27 364 (36.7%) | 14 737 (27.1%) | |
| 6 | 62 561 (20.7%) | 14 185 (52.0%) | 10 329 (29.9%) | 15 438 (13.8%) | 19 050 (25.6%) | 3559 (6.5%) | |
| Health/Disease condition | |||||||
| Stroke | 46 090 (15.2%) | 5446 (20.0%) | 6480 (18.8%) | 15 874 (14.2%) | 11 565 (15.5%) | 6725 (12.4%) | <.0001 |
| Heart failure | 29 562 (9.8%) | 3141 (11.5%) | 4005 (11.6%) | 10 463 (9.4%) | 7842 (10.5%) | 4111 (7.6%) | <.0001 |
| Hypertension | 176 772 (58.4%) | 15 748 (57.7%) | 20 721 (60.0%) | 65 799 (58.9%) | 43 764 (58.7%) | 30 740 (56.5%) | <.0001 |
| Alzheimer’s Dx | 22 450 (7.4%) | 3194 (11.7%) | 2107 (6.1%) | 7929 (7.1%) | 7263 (9.7%) | 1957 (3.6%) | <.0001 |
| Dementia (other than ALZ) | 56 279 (18.6%) | 8270 (30.3%) | 6624 (19.2%) | 19 073 (17.1%) | 17 381 (23.3%) | 4931 (9.1%) | <.0001 |
| Hemiplegia/Hemiparesis | 6140 (2.0%) | 830 (3.0%) | 1007 (2.9%) | 1759 (1.6%) | 1159 (1.6%) | 1385 (2.5%) | <.0001 |
| Parkinsonism | 13 770 (4.6%) | 1807 (6.6%) | 1550 (4.5%) | 4823 (4.3%) | 4137 (5.6%) | 1453 (2.7%) | <.0001 |
| Hip Fracture | 12 267 (4.1%) | 1189 (4.4%) | 1464 (4.2%) | 4565 (4.1%) | 2881 (3.9%) | 2168 (4.0%) | <.0001 |
| Osteoporosis | 54 388 (18.0%) | 5115 (18.7%) | 6487 (18.8%) | 20 305 (18.2%) | 13 600 (18.2%) | 8881 (16.3%) | <.0001 |
| Cataract | 33 809 (11.2%) | 3571 (13.1%) | 4363 (12.6%) | 11 916 (10.7%) | 8657 (11.6%) | 5302 (9.7%) | <.0001 |
| Glaucoma | 19 772 (6.5%) | 2260 (8.3%) | 2393 (6.9%) | 7079 (6.3%) | 5497 (7.4%) | 2543 (4.7%) | <.0001 |
| HIV infection | 411 (0.1%) | 23 (0.1%) | 74 (0.2%) | 138 (0.1%) | 75 (0.1%) | 101 (0.2%) | <.0001 |
| Diabetes | 78 258 (25.9%) | 7133 (26.1%) | 9654 (27.9%) | 28 518 (25.5%) | 18 884 (25.3%) | 14 069 (25.9%) | <.0001 |
| Emphysema/COPD/Asthma | 48 220 (15.9%) | 4856 (17.8%) | 6825 (19.8%) | 17 181 (15.4%) | 12 319 (16.5%) | 7039 (12.9%) | <.0001 |
| Renal failure | 20 265 (6.7%) | 2092 (7.7%) | 2773 (8.0%) | 7088 (6.3%) | 5026 (6.7%) | 3286 (6.0%) | <.0001 |
| Thyroid disease | 48 239 (15.9%) | 4274 (15.7%) | 5469 (15.8%) | 18 153 (16.3%) | 11 896 (16.0%) | 8447 (15.5%) | <.0001 |
Baseline IC distribution
Baseline IC demonstrated a smooth, non-linear relationship with age (Figure 1). Overall, mean baseline IC rose from 8.31 at age 18, increasing steadily through early and middle adulthood before reaching a peak of 10.82 at age 63, followed by a gradual decline into older age. Sex-stratified analyses revealed modest but consistent differences in the shape of this trajectory. Males exhibited slightly higher IC scores in younger adulthood and reached their peak earlier (age 56; mean IC = 10.74), with decline commencing in the sixth decade. Females demonstrated a later and more sustained peak (age 65; mean IC = 10.90), maintaining marginally higher IC scores than males from approximately age 60 onward.
Figure 1.
Line graph showing intrinsic capacity (IC) scores by age at admission, presented separately for females, males, and the overall population. IC scores rise rapidly in early adulthood, level off during midlife, and gradually decline with increasing age. The trajectories for females and males are broadly similar, with no major sex-related differences in the overall pattern.
IC trajectory groups
Group-based trajectory modelling identified five distinct IC trajectories (Figure 2). Model adequacy evaluation showed that all groups had an AVePP >0.7 and OCC > 5. The model selection step with their associated BIC, the adequacy metrics and PROC Traj model output for the best model are presented in Appendix 3.
Figure 2.
Line graph showing five distinct trajectories of intrinsic capacity over 72 months among home care recipients. The trajectories demonstrate heterogeneous patterns of change over time, including stable, declining, and improving patterns. The proportion of participants in each trajectory group is indicated in the figure legend.
Mean baseline and final IC scores and trajectory patterns are summarized below.
Group 1: Low initial rapid decline (9.1%): Baseline IC 8.0; steep decline in the first year, then slower decline; final IC 7.0 (−14.3%). High proportion with baseline impairment across vitality (96.4%), cognition (92.4%), and sensory (84.2%) domains.
Group 2: Moderate initial rapid decline (24.7%): Baseline IC 10.0; decline over 24–36 months; final IC 8.0 (−25.0%). Vitality (97.0%) and Cognitive impairment (86.4%) were common.
Group 3: Low initial improved (12.4%): Baseline IC 9.0; improvement over ~36 months; final IC 10.0 (+10.0%). High baseline impairment in vitality domain and mild impairment in cognition domain are features of this group.
Group 4: High initial rapid decline (34.7%): Baseline IC 11.0; steady decline; final IC 10.0 (−10.0%). Locomotion domain function is largely intact for this group with moderate impairments in vitality and cognition.
Group 5: High initial Stable (19.1%): Baseline IC 12.0; minimal change over time; final IC 13.0 (+7.7%). Moderate cognition domain impairment with minimal impairments in other domains.
Over time, the proportion of recipients with impairments increased across all domains for all trajectory groups, except for the Low initial improved group, where the proportion with impairment decreased in the locomotion and psychological domains, Figure 3.
Figure 3.
Grouped bar chart showing the percentage of individuals with no impairment in each intrinsic capacity domain at baseline and at the last available assessment, stratified by five trajectory groups. Higher percentages indicate better preservation of intrinsic capacity. Across trajectory groups, the percentage without impairment varies by domain and changes between baseline and the last assessment. Percentages in parentheses indicate the proportion of individuals belonging to each trajectory group.
Trajectory groups and outcomes
Five-year mortality (12.2%, 8.5%, 8.2%, 7.2% and 6.4%; Figure 4) and competing-risk-adjusted LTC admission rates (50.0%, 48.4%, 38.1%, 38.6% and 22.7%; Appendix 4) differed significantly across Groups 1–5 (all P < .0001), Appendix 5.
Figure 4.
Kaplan–Meier survival curves showing survival probability over follow-up time among five intrinsic capacity trajectory groups. The curves are clearly separated, with trajectory group 5 showing the highest survival probability and group 1 showing the lowest. Differences between the groups are statistically significant (log-rank P < 0.0001). Shaded areas around the curves represent confidence intervals.
After adjusting for baseline covariates and clinical diagnoses, associations remained independent and strong. Relative to the High Initial Stable group, mortality hazards were significantly higher for Low Declined (HR 2.53, 95% CI 2.41–2.66), Moderate Declined (HR 1.69, 1.62–1.76), Low Improved (HR 1.50, 1.43–1.58), and High Declined (HR 1.28, 1.24–1.33). LTC admission sub-distribution hazards followed a matching pattern: Low Declined (SHR 2.43, 2.36–2.50), Moderate Declined (SHR 2.27, 2.22–2.32), Low Improved (SHR 1.84, 1.79–1.89), and High Declined (SHR 1.73, 1.69–1.76).
Predictors of trajectory membership
Sociodemographic and clinical characteristics were associated with trajectory assignment, Appendix 6. Older age (≥90) was associated with the least likelihood of improving trajectory (Group 3), while younger age (<60) and female sex were associated with the high initial Stable group (Group 5). Alzheimer’s disease, Parkinsonism, cataract, and glaucoma were associated with low initial decline trajectories.
Sensitivity analysis
Of 489 367 assessed individuals, 302,467 (61.9%) with ≥3 assessments were included. Excluded individuals (38.1%) had significantly higher 1-year (5.54% vs 1.21%) and 5-year (10.91% vs 6.91%) mortality, but lower LTC admission rates (32.19% vs 39.18%; all P < 0.0001). Excluded participants were older with greater baseline health frailty and multimorbidity (Appendix 7).
IC trajectories based on standardized scores closely resembled those derived from raw IC scores, with similar numbers of groups and trajectory shapes (Appendices 8 & 9).
Discussion
We identified five distinct IC trajectories, indicating heterogeneous functional ageing. Trajectory membership was associated with socio-demographic and clinical characteristics and independently predicted mortality and LTC admission. These findings support longitudinal IC profiles for risk stratification, care planning and intervention targeting.
IC can be operationalised on a scale using routinely collected health data and its derivation from interRAI assessments provides a feasible mechanism for embedding it into routine care practice. The interRAI system provides an existing infrastructure for its implementation without additional clinical data collection burden, particularly in home care settings where repeated standardized assessments are already performed.
The IC trajectories observed in this study were broadly consistent with functional ageing patterns proposed by WHO [2]. IC domains were derived from validated interRAI items but differed in measurement scale: vitality items were binary variables, whereas cognition, sensory, psychological well-being, and locomotion were captured using multi-level clinical scales (CPS, vision, hearing, DRS, and ADL hierarchy). These differences reflect measurement properties of the instruments rather than conceptual differences in domain importance. Sensitivity analyses using POMP-standardized domains (0–100 scaling across all domains) reproduced identical trajectory structures and temporal patterns, supporting robustness of the IC construct. Raw IC scores were retained for primary analyses because they preserve clinical interpretability and allow domain-level changes to be understood in clinically meaningful units.
Individuals generally commence homecare in one of three broad IC spectrums: low, medium or high, and from this, cluster into different trajectories. Decline was the predominant pattern, with limited improvement and no evidence of rapid recovery across the cohort. Importantly, baseline IC categories alone did not fully capture subsequent heterogeneity in functional evolution, as individuals within each starting level diverged into distinct trajectories over time based on rate and pattern of change.
Among individuals with high baseline IC, two distinct trajectories were identified. One group exhibited a relatively rapid decline in IC over time (group 4), while the other remained stable or showed slight improvement (group 5). These differences were not explained by baseline IC alone but by underlying clinical and demographic differences. Individuals in the stable trajectory were younger and had lower prevalence of neurodegenerative conditions, including Alzheimer’s disease, other dementias, and Parkinsonism. This suggests that even among initially high-functioning individuals, underlying neurological disease burden and age structure strongly influence the trajectory of functional decline.
Among individuals with low baseline IC, two additional trajectories were identified, representing divergent functional pathways. The declining trajectory was characterized by higher baseline impairment across locomotion, cognition, and sensory domains. The Low Improved group had near-universal baseline vitality impairment but showed overall IC gains over time. This improvement may reflect psychological recovery and minor locomotion gains. Severe baseline vitality likely reflects acute decompensation at home care intake rather than permanent deficits; psychological and physical recovery can proceed as the acute phase stabilizes. Because interRAI-HC vitality items (stamina, dyspnea and appetite) track acute symptoms, they may linger during overall functional improvement. This suggests domain relationships are not strictly hierarchical in clinical populations when recovery is predominantly psychosocial.
Further, cognitive impairment was consistently associated with declining trajectories. In contrast, improvements observed in the low-IC improving group appeared to be driven primarily by changes in locomotion and psychological well-being domains. This suggests that certain IC domains may be more responsive to intervention or environmental modification than others. Specifically, mobility and psychological function may be more modifiable in home care settings, whereas cognition and sensory impairment appear less reversible once established. The implication for these is that while improving overall IC is the clinical objective, targeting modifiable domains may yield more immediate functional gains, particularly in individuals with preserved cognition. Conversely, where cognitive or sensory impairment predominates, interventions focused on environmental adaptation, support systems, and compensatory strategies may be more effective than attempts to directly modify these domains.
Trajectory-based classification provided more nuanced prognostic information than baseline IC alone where individuals with moderate baseline IC had slightly higher 5-year mortality and substantially higher LTC admission compared to low IC individuals who improved. Individuals with similar baseline IC levels could follow very different trajectories with substantially different outcomes. Most prior studies have relied on baseline IC measures [18], which may be appropriate in cross-sectional analyses but do not capture dynamic change over time. Given that IC is conceptualized as a multidomain construct that evolves in response to ageing, disease progression, and interventions, reliance on baseline values alone risks misclassification of future risk. Our findings align with emerging evidence showing that longitudinal IC transitions improve prediction of adverse outcomes compared with static measures [47]. For individuals receiving ongoing home care, trajectory-based assessment therefore provides more clinically relevant prognostic information and may support more precise targeting of services and interventions.
The identification of five IC trajectories differs from previous studies reporting three or four groups [24–31, 48, 49]. This likely reflects differences in population characteristics and case-mix. Homecare recipients represent a more clinically complex population than community-dwelling cohorts, with higher levels of multimorbidity, functional impairment, and health instability. This increased heterogeneity likely contributes to greater diversity in observed trajectory patterns. In addition, this study included adults aged 18 years and older, whereas most prior studies have focused on older adults aged ≥60 years. This broader inclusion aligns with a life-course perspective on intrinsic capacity recommended by WHO [19, 50, 51] and allows examination of how IC evolves across adulthood into older age. Notably, participants aged 18–35 in this cohort are clinically selected into home care by severe conditions such as acquired brain injury, progressive neurological disease, physical disability, or serious mental illness, rather than age-related decline, likely explaining their lower IC relative to mid-life adults. While IC has been suggested to peak in early adulthood, no large empirical study has established a universal peak age in general or clinical population. Our finding of peak IC at age 63 is therefore not inconsistent with existing evidence but reflects the clinical selection underlying this cohort. IC trajectory patterns in service-use populations should be interpreted relative to the referral mechanism, not compared directly to assumed population-level norms.
Clinical and implementation implications of IC trajectories
An important consideration is that trajectory membership was derived retrospectively based on observed longitudinal patterns and therefore cannot be assigned at baseline. However, early clinical characteristics associated with each trajectory suggest the potential for prospective classification. Future work should focus on developing predictive models that identify likely trajectory membership early in the care episode using baseline and short-term follow-up data. Such models would enable proactive risk stratification and trajectory-informed care planning rather than retrospective classification. This represents a key step toward translation of IC trajectories into routine clinical decision-making.
IC can be operationalized using existing interRAI systems without additional data collection, enabling implementation of WHO’s IC framework in routine home care where interRAI assessments are already embedded. This supports scalable population-level use.
Trajectories provide clinically meaningful stratification and distinguish functional courses linked to different care needs. For instance, rapid decline may require urgent intervention, closer monitoring, and early LTC planning, while stable or improving trajectories may support maintenance care and targeted rehabilitation.
Domain patterns further suggest tailored strategies, with cognition requiring compensatory approaches and locomotion and psychological domains being more amenable to intervention. Overall, the identified IC trajectories support a shift toward actionable, trajectory-informed care planning.
This study extends the WHO life-course IC framework to home care settings. Identifying five distinct trajectories shows that functional ageing is non-uniform and unfixed at clinical presentation. The Low Improved pathway proves that home care entry does not signal irreversible decline, expanding the timeline for capacity-preserving interventions.
Limitations
Individuals with fewer than three assessments were excluded as required for group-based trajectory modelling. This may underrepresent individuals with short care duration or rapid decline, who are often among the most clinically vulnerable. Sensitivity analyses showed that excluded individuals were older, more medically complex, and had higher mortality. Trajectory group membership was derived retrospectively from observed longitudinal patterns and cannot be assigned prospectively at the point of admission. Prospective validation of trajectory-based risk stratification is required before these findings can be translated into routine clinical decision-making. These limitations should be considered when generalizing findings to the full home care population.
Conclusion
This study provides the first population-based evidence of IC trajectories among home care recipients, demonstrating that IC can be operationalised at scale using routinely collected interRAI data. Five distinct trajectories revealed substantial heterogeneity in functional ageing, with longitudinal patterns offering stronger prognostic value for mortality and long-term care admission than baseline IC alone. Improvement in IC was observed even among individuals with low initial capacity, driven largely by modifiable domains, while cognitive impairment consistently predicted decline. These findings highlight the value of monitoring IC over time to support earlier risk stratification, guide targeted interventions, and strengthen person-centred care. Operationally, IC trajectories could be used to inform tiered service models, where trajectory group membership guides service intensity, monitoring frequency, and early transition planning. As health systems pursue WHO-aligned approaches to healthy ageing, integrating IC trajectory monitoring into routine home care practice offers a scalable pathway toward more anticipatory and resilience-focused care.
Supplementary Material
Acknowledgements
We thank all interRAI colleagues who contributed ideas to the project. We are grateful to Miceala Jantzi and Jonathan Chen for assistance with data access.
Contributor Information
Bonaventure Amandi Egbujie, University of Waterloo Faculty Health - School of Public Health Sciences, 200 University Av West, Waterloo, Ontario N2L 3G1, Canada.
Hein van Hout, Amsterdam Public Health Research Institute - Aging and Later Life, Amsterdam, Netherlands; Department of General Practice and Elderly Care Medicine, Amsterdam UMC Locatie AMC, Amsterdam, Netherlands.
George Heckman, Lawson Health Research Institute, London, Ontario, Canada; Western University - Faculty of Applied Health Sciences, School of Public Health and Health Systems, London, Ontario, Canada.
Leon Geffen, University of Cape Town - The Albertina and Walter Sisulu Institute of Ageing in Africa, Rondebosch, WC, South Africa.
John N Morris, Hebrew Senior Life - Institute for Aging Research, Boston, MA, USA.
John P Hirdes, University of Waterloo Faculty Health - School of Public Health Sciences, 200 University Av West, Waterloo, Ontario N2L 3G1, Canada.
Declaration of Conflicts of Interest
None declared.
Declaration of Sources of Funding
None declared.
References
- 1. Beard JR, Officer A, De Carvalho IA et al. The world report on ageing and health: a policy framework for healthy ageing. The Lancet 2016;387no. 10033. Preprint:2145. 10.1016/S0140-6736(15)00516-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. World Health Organization (WHO) . World Report on Ageing and Health. World Health Organization: World Health Organization, 2015. [Google Scholar]
- 3. Cesari M, De CIA, Thiyagarajan JA et al. Evidence for the domains supporting the construct of intrinsic capacity. J Gerontol Series A 2018;73:1653–60. 10.1093/gerona/gly011. [DOI] [PubMed] [Google Scholar]
- 4. Egbujie BA, Turcotte LA, Heckman G et al. Trajectories of functional decline and predictors in long-term care settings: a retrospective cohort analysis of Canadian nursing home residents. Age Ageing 2024;53:afae264. 10.1093/AGEING/AFAE264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Binette J, Vasold K. 2018 Home and Community Preferences: A National Survey of Adults Age 18-Plus. Washington, DC: AARP Research; published online August 31, 2018. 10.26419/RES.00231.001. [DOI] [Google Scholar]
- 6.March of Dimes Canada. Transforming Lives through Home Modification: A March of Dimes Canada National Survey. Toronto, ON: March of Dimes Canada; n.d. Available at: March of Dimes Canada. https://www.marchofdimes.ca/en-ca/aboutus/newsroom/pr/prarchive/Pages/MODC-Home-Modification-Survey.aspx (11 December 2025, date last accessed).
- 7. Fernández-Carro C. Ageing at home, co-residence or institutionalisation? Preferred care and residential arrangements of older adults in Spain. Ageing Soc 2016;36:586–612. 10.1017/S0144686X1400138X. [DOI] [Google Scholar]
- 8. Access to health care . n.d.. https://www150.statcan.gc.ca/n1/pub/82-570-x/2024001/section4-eng.htm (6 December 2025, date last accessed).
- 9. Yu Y, Petrovic M, Zhang WH. Older European Adults and Access to Healthcare during the COVID-19 Pandemic. China CDC Wkly. Beijing, China: Chinese Center for Disease Control and Prevention; 2022;4:879–884. 10.46234/CCDCW2022.182. [DOI] [Google Scholar]
- 10. Badia JG, Santos AB, Segura JCC et al. Predictors of mortality among elderly dependent home care patients. BMC Health Serv Res 2013;13:316. 10.1186/1472-6963-13-316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Sørbye LW, Hamran T, Henriksen N et al. Home care patients in four Nordic capitals—predictors of nursing home admission during one-year followup. J Multidiscip Healthc. 2010;3:11–18. 10.2147/JMDH.S8979. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Morris JN, Howard EP, Steel K et al. Predicting risk of hospital and emergency department use for home care elderly persons through a secondary analysis of cross-national data. BMC Health Serv Res 2014;14:519. 10.1186/S12913-014-0519-Z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Kristinsdottir IV, Jonsson PV, Hjaltadottir I et al. Scandinavian Journal of caring sciences prevalence and predictors of caregiver distress in six European Communities: data from the IBenC study, using interRAI-home care assessments. Scand J Caring Sci 2025;39:e70005. 10.1111/scs.70005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Stratidaki E, Mechili EA, Ouzouni C et al. An investigation of the risk factors related to frailty in older adults receiving home care services. Nutrients. 2024;16:3982. 10.3390/nu16233982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Campitelli MA, Bronskill SE, Hogan DB et al. The prevalence and health consequences of frailty in a population-based older home care cohort: a comparison of different measures. BMC Geriatr 2016;16:133. 10.1186/S12877-016-0309-Z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Ouellette-Kuntz H, Martin L, McKenzie K. Rate of deficit accumulation in home care users with intellectual and developmental disabilities. Ann Epidemiol 2018;28:220–4. 10.1016/j.annepidem.2018.01.010. [DOI] [PubMed] [Google Scholar]
- 17. McKenzie K, Ouellette-Kuntz H, Martin L. Using an accumulation of deficits approach to measure frailty in a population of home care users with intellectual and developmental disabilities: an analytical descriptive study. BMC Geriatr 2015;15:170. 10.1186/S12877-015-0170-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Stolz E, Mayerl H, Freidl W et al. Intrinsic capacity predicts negative health outcomes in older adults. J Gerontol A Biol Sci Med Sci 2021;77:101–5. 10.1093/GERONA/GLAB279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Tan LF, Le HA, Lee XX et al. Intrinsic capacity deficits across the lifespan in a nationally representative community cohort: findings from the Queenstown study. Arch Gerontol Geriatr 2026;140:106031. 10.1016/J.ARCHGER.2025.106031. [DOI] [PubMed] [Google Scholar]
- 20. Locquet M, Sanchez-Rodriguez D, Bruyère O et al. Intrinsic capacity defined using four domains and mortality risk: a 5-year follow-up of the SarcoPhAge cohort. J Nutr Health Aging 2021;26:23–9. 10.1007/S12603-021-1702-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Campbell CL, Cadar D, McMunn A et al. Operationalization of intrinsic capacity in older people and its association with subsequent disability, hospital admission and mortality: results from the English longitudinal study of ageing. J Gerontol A Biol Sci Med Sci 2022;78:698–703. 10.1093/GERONA/GLAC250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Zhang S, Tange C, Huang ST et al. Multi-trajectories of intrinsic capacity and their effect on higher-level functional capacity, life satisfaction, and self-esteem in community-dwelling older adults: the NILS-LSA. J Nutr Health Aging 2025;29:100432. 10.1016/j.jnha.2024.100432. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Wang Y, Tang N, Shao M et al. Transitions and trajectories in intrinsic capacity states over time: a systematic review. BMC Geriatr 2025;25:407. 10.1186/S12877-025-05938-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Liu S, Kang L, Liu XH et al. Trajectory and correlation of intrinsic capacity and frailty in a Beijing elderly community. Front Med (Lausanne) 2021;8:751586. 10.3389/fmed.2021.751586. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Salinas-Rodríguez A, González-Bautista E, Rivera-Almaraz A et al. Longitudinal trajectories of intrinsic capacity and their association with quality of life and disability. Maturitas 2022;161:49–54. 10.1016/J.MATURITAS.2022.02.005. [DOI] [PubMed] [Google Scholar]
- 26. Salinas-Rodríguez A, Fernández-Niño JA, Rivera-Almaraz A et al. Intrinsic capacity trajectories and socioeconomic inequalities in health: the contributions of wealth, education, gender, and ethnicity. Int J Equity Health 2024 23:1 2024;23:48. 10.1186/S12939-024-02136-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Chen JJ, Liu LF, Chang SM. Approaching person-centered long-term care: the trajectories of intrinsic capacity and functional decline in Taiwan. Geriatr Gerontol Int (Japan) 2022;22:516–22. 10.1111/ggi.14391. [DOI] [PubMed] [Google Scholar]
- 28. Zhou Y, Wang G, Li J et al. Trajectory of intrinsic capacity among community-dwelling older adults in China: the China health and retirement longitudinal study. Arch Gerontol Geriatr 2024;124:105452. 10.1016/j.archger.2024.105452. [DOI] [PubMed] [Google Scholar]
- 29. Cui Y, Lei S, Xia Y et al. The trajectory of intrinsic capacity and its related factors among elderly Chinese patients with cardiovascular disease: a prospective cohort study. Front Endocrinol (Lausanne) 2025;16:1539982. 10.3389/fendo.2025.1539982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Zhang S, Wu S, Guo R et al. Patterns of intrinsic capacity trajectory and onset of activities of daily living disability among community-dwelling older adults. J Glob Health 2024;14:04159. 10.7189/JOGH.14.04159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Zhao Y, Chen Y, Xiao LD et al. Intrinsic capacity trajectories, predictors and associations with care dependence in community-dwelling older adults: a social determinant of health perspective. Geriatr Nurs (Minneap) 2024;56:46–54. 10.1016/J.GERINURSE.2023.12.022. [DOI] [PubMed] [Google Scholar]
- 32. Heckman GA, Gray LC, Hirdes JP. Addressing health care needs for frail seniors in Canada: the role of interRAI instruments. CGS J CME. 2013;3:8–16. https://uwspace.uwaterloo.ca/handle/10012/11701 (23 May 2023, date last accessed). [Google Scholar]
- 33. Hirdes JP, Ljunggren G, Morris JN et al. Reliability of the interRAI suite of assessment instruments: a 12-country study of an integrated health information system. BMC Health Serv Res 2008;8:1–11. 10.1186/1472-6963-8-277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Fries BE, Heckman G, Hirdes JP et al. Born from crisis: the evolution of interRAI and its relevance to Today’s healthcare challenges. Healthc Manage Forum 2026;39:75–84. 10.1177/08404704251389190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Gray LC, Berg K, Fries BE et al. Sharing clinical information across care settings: the birth of an integrated assessment system. BMC Health Serv Res 2009;9:71. 10.1186/1472-6963-9-71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Heckman GA, Jantzi M, Hirdes JP et al. Building a rational clinical information system for older adults in acute care: the role of the interRAI acute care suite. Healthc Manage Forum 2026;39:141–8. 10.1177/08404704251389449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Boscart V, Taucar LS, Heyer M et al. InterRAI acute care instrument for seniors in Canadian hospitals: findings of an inter-rater reliability pilot study. Can J Nurs Res 2021;53:155–61. 10.1177/0844562120920513. [DOI] [PubMed] [Google Scholar]
- 38. Kim H, Il JY, Sung M et al. Reliability of the interRAI long term care facilities (LTCF) and interRAI home care (HC). Geriatr Gerontol Int 2015;15:220–8. 10.1111/GGI.12330. [DOI] [PubMed] [Google Scholar]
- 39. Hirdes JP, Frijters DH, Teare GF. The MDS-CHESS scale: a new measure to predict mortality in institutionalized older people. J Am Geriatr Soc 2003;51:96–100. 10.1034/J.1601-5215.2002.51017.X. [DOI] [PubMed] [Google Scholar]
- 40. Nagin D. Group-Based Modeling of Development. Cambridge, MA: Harvard University Press; 2005. 10.4159/9780674041318. [DOI] [Google Scholar]
- 41. Jones BL, Nagin DS. Advances in group-based trajectory modeling and an SAS procedure for estimating them. Sociol Methods Res 2007;35:542–571. Thousand Oaks, CA: SAGE Publications. 10.1177/0049124106292364. [DOI] [Google Scholar]
- 42. Jones BL, Nagin DS. Proc TRAJ: a SAS procedure for group-based modeling of longitudinal data. Annual Meeting Home 2007;1:585. [Google Scholar]
- 43. Nagin DS. Group-based trajectory modeling: an overview. Ann Nutr Metab 2014;65:205–10. 10.1159/000360229. [DOI] [PubMed] [Google Scholar]
- 44. Nagin DS, Odgers CL. Group-based trajectory modeling in clinical research. In: Annu Rev Clin Psychol. Preprint 2010;6:109–38. 10.1146/annurev.clinpsy.121208.131413. [DOI] [PubMed] [Google Scholar]
- 45. Jones BL, Nagin DS, Roeder K. A SAS procedure based on mixture models for estimating developmental trajectories. Sociol Methods Res 2001;29:374–93. 10.1177/0049124101029003005. [DOI] [Google Scholar]
- 46. Haviland AM, Jones BL, Nagin DS. Group-based trajectory modeling extended to account for nonrandom participant attrition. Sociol Methods Res 2011;40:367–90. 10.1177/0049124111400041. [DOI] [Google Scholar]
- 47. Hwang AC, Chen LY, Tseng SH et al. Intrinsic capacity transitions predict overall and cause-specific mortality, incident disability, and healthcare utilization. J Nutr Health Aging 2024;28:100359. 10.1016/J.JNHA.2024.100359. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Meng LC, Chuang HM, Lu WH et al. Multi-trajectories of intrinsic capacity decline and their impact on age-related outcomes: a 20-year National Longitudinal Cohort Study. Aging Dis 2024;15:2697–709. 10.14336/AD.2023.1115-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Lu WH, Guyonnet S, Martinez LO et al. Association between aging-related biomarkers and longitudinal trajectories of intrinsic capacity in older adults. Geroscience 2023;45:3409–18. 10.1007/S11357-023-00906-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Lu WH, Rolland Y, Guyonnet S et al. Reference centiles for intrinsic capacity throughout adulthood and their association with clinical outcomes: a cross-sectional analysis from the INSPIRE-T cohort. Nat Aging 2023;3:1521–8. 10.1038/S43587-023-00522-X. [DOI] [PubMed] [Google Scholar]
- 51.World Health Organization. Framework to implement a life course approach in practice. Geneva, Switzerland: World Health Organization; 2025. Available at: https://www.who.int/publications/i/item/9789240112575 (14 December 2025, date last accessed).
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.