Physical Function and Residential Relocation of Older Adults Living With Chronic Conditions

Abstract

Environmental Gerontology has shown that mobility difficulty can be a barrier to aging in place when the home environment is not physically accessible. However, difficulty may vary by underlying chronic conditions. Using data from a cohort of community-dwelling adults age 65 or older in the 2011-2019 National Health and Aging Trends Study, we found that mobility difficulty was associated with moving to housing with fewer environmental barriers. Mobility difficulty was not associated with moving to other destinations, such as housing with the same or more barriers, or nursing home. Those with dementia or cancer were more likely to move to a nursing home, and those with cancer were less likely to move to housing with the same or more barriers. However, having these chronic conditions did not change the relationship between mobility difficulty and relocation. These findings suggest that addressing mobility and reducing environmental barriers may promote aging in place.

Keywords

person-environment fit aging in place mobility housing

What This Paper Adds

• This paper explores whether chronic conditions like dementia and cancer influence residential relocation among U.S. older adults who have difficulty standing and walking.

• Our study is unique in evaluating the role of mobility difficulty in relocation separately from other activity limitations and in considering how it interacts with chronic conditions.

• Our use of a large, nationally representative dataset with annual follow up creates findings that can be broadly applied to various settings in the U.S.

Applications of Study Findings

• Screening for difficulty standing and walking in the home is a simple way to identify older adults who may need more support to age in place—regardless of chronic conditions.

• Future research can expand on our findings by separately evaluating desired and undesired relocation and by focusing on older adults with incident, rather than prevalent chronic conditions.

Introduction

The majority of older Americans prefer to age in their current homes, however, chronic health conditions may make this more difficult, particularly when these influence ability to complete day-to-day tasks (Brim et al., 2021; Davis, 2021). This can be understood as poor fit between the person’s capacity and their environment, as proposed by the Ecological Theory of Aging (Granbom et al., 2014; Lawton & Nahemow, 1973). We previously showed that difficulty with mobility-related physical function (e.g., standing, walking, here referred to as mobility difficulty) was associated with relocation that reduces environmental demands (Jensen-Battaglia et al., 2024). Better person-environment fit is associated with aging in place, and our finding adds to the evidence that older adults may undertake health-related moves to improve their person-environment fit (Granbom et al., 2016; Stineman et al., 2012). However, the experience of mobility difficulty—and by extension, its relationship with relocation—can differ greatly by chronic condition (Lin et al., 2016). Given that declining health is the most common reason for relocation among older adults, understanding how mobility difficulty and health conditions relate to relocation in combination may inform development of targeted interventions to promote aging in place (Pope & Kang, 2010).

To understand relocation as a response to poor person-environment fit (and loss of function), we draw on the Selective Optimization with Compensation theory of aging, which has often been used to examine adaptation to loss. We use it in this paper to understand the way individuals compensate for mobility difficulty in order to accomplish valued activities. In this context, we propose relocation as a type of compensation to maintain or improve function at home (Baltes & Baltes, 1990; Friedman et al., 2016). However, diagnosis of health conditions with distinct prognoses may also drive condition-specific differences in the adaptation process because they embody different patterns of functional change. For example, an older adult who has experienced a stroke may expect to recover ability, while one with dementia may anticipate progressive decline. To explore this, we compared three chronic conditions which impact physical function in distinct ways.

Cancer, myocardial infarction (MI) or stroke, and dementia are chronic conditions which present with varied trajectories of physical function. For instance, those with cancer are at increased risk for intermittent decline related to disease progression and the effects of treatment (e.g., recovery after surgical resection, chemotherapy-induced peripheral neuropathy) (Mohile et al., 2009; Presley et al., 2019). Both MI and stroke are associated with declines in physical function, often marked by sudden worsening following the event followed by some degree of recovery, rather than intermittent decline (Dhamoon et al., 2017; Hajduk et al., 2019). Finally, those with cognitive impairment and dementia experience monotonic declines in their physical function (Cohen et al., 2016). Existing evidence shows that all of these conditions are associated with relocation to a nursing home (Granbom et al., 2019; Smedegaard et al., 2018; Stineman et al., 2012), and potentially with relocation within the community, although the strength of the association varies (Friedman et al., 2016; Granbom et al., 2019; Lovasi et al., 2014). However, to date there are no analyses of chronic conditions in conjunction with mobility-related function.

To address this gap, we analyzed Medicare beneficiaries participating in the National Health and Aging Trends (NHATS) study. Our aims were (1) to describe residential relocation for those with cancer, dementia, and MI/stroke, and (2) to analyze whether these chronic conditions modified the association of mobility difficulty and residential relocation. We hypothesized that progressive conditions such as dementia and cancer may strengthen the association as older adults selectively deprioritize aging in place in favor of making compensatory moves to housing with better person-environment fit and potential to improve their function. Conversely, those with conditions where recovery is expected (like MI and stroke) may anticipate returning to better person-environment fit in their current housing and be less likely to move in response to mobility difficulty. Mobility difficulty and chronic conditions are highly prevalent among older Americans and understanding this association can inform use of screening to target interventions which support aging in place.

Materials and Methods

Data Source and Structure

The NHATS is a nationally representative cohort study of U.S. older adults drawn from a random sample of Medicare Beneficiaries (Freedman & Kasper, 2019). Annual data collection includes an in-person interview, physical performance measurements, and assessment of the home environment. Proxy respondents of deceased participants complete a last month of life interview. We included participants present in the starting round (2011) and limited our analytic sample to those with ≥2 consecutive rounds of follow up (e.g., time t and t + 1) from 2011 (Round 1) to 2019 (Round 9) who were community dwelling in the time t round. We focused on moves by community-dwelling older adults, as moves from residential care settings are very different, and affected by the policies of the residential facility (Ashcraft et al., 2006). Data from 2020 and 2021 (Rounds 10 and 11) were excluded due to the influence of the COVID pandemic on U.S. housing markets, nursing home admissions, and NHATS data collection procedures. Characteristics of each participant from Rounds 1-8 are paired with their relocation status from the subsequent t + 1 round (Rounds 2-9). Thus, individual participants contribute up to 8 observations of t and t + 1 pairs in our analytic sample (Round 1 characteristics and Round 2 relocation, Round 2 characteristics and Round 3 relocation, etc.).

Measures

Exposure: Score of Mobility-Related Physical Function

NHATS assesses three mobility activities: going outside the home/building, getting around inside the home/building, and getting out of bed. Participants reported whether they “ever did” each activity in the prior month and if so, whether they had any difficulty. Participants reporting they never did an activity, or never did the activity independently are asked if this was because they did not have sufficient help or the task was too difficult (indicating a possible unmet need for assistance). In our prior work we adapted an existing scale for these activities ranging from 0 (indicating no difficulty or assistance with any activity), to 9 points (unmet need for assistance with all activities) (Gill & Williams, 2018; Jensen-Battaglia et al., 2024). Activities at round t were scored 0 if the person reported being fully able, using accommodations (mobility devices, doing task less frequently), 1 if they reported any difficulty with or without accommodations, 2 if they reported assistance or not doing the activity, and 3 if they had unmet need for assistance.

Outcome: Residential Relocation, Environmental Barriers

Our outcome was residential relocation to a new permanent address, by the t + 1 round. Mobility difficulty is dynamic and older adults may experience periods of recovery or decline (Gill et al., 2006). Therefore we evaluated four types of moves in the year immediately following the assessment of mobility: (1) no move, (2) moving to a home with ≥ the number of home environment barriers, (3) moving to a home with fewer home environment barriers, and (4) moving to a nursing home or residential care (e.g., multi-unit buildings or formal retirement communities which offer assistance, group home, board and care home, or assisted living facilities). Although older adults may use other means to compensate for poor person-environment fit (e.g., increased assistance from informal caregivers), we chose to focus on environmental barriers as caregivers often address care needs which are not directly related to in-home mobility, such as shopping or transportation (Wolff et al., 2016). These home environment barriers included participant reported presence of steps to enter the home without a ramp, absence of shower/tub seat or grab bars around the tub or toilet, and >1 floor with bed, bath, and kitchen on different floors and no elevator or stair lift/glide between them.

Effect Modifiers: Cancer, Myocardial Infarction or Stroke, Dementia

Non-skin cancer and myocardial infarction or stroke, were measured as self-report of a doctor’s diagnosis. For dementia, we used the NHATS-provided measure of possible or probable dementia that incorporates self-report, cognitive function tests, and presence of a proxy respondent (Kasper et al., 2013). In the baseline round (2011), participants reported whether a doctor had ever told them they have these conditions. In future rounds, participants are asked if a doctor has told them they have this condition since the last interview. However, participants are not asked to distinguish a recurrence from a new diagnosis (e.g., different cancer type). For this reason, we classified conditions as prevalent, defined as participant report of the condition in round t or any previous round. Details regarding these measures are provided in the supplemental methods and Supplemental Table S.1.

Covariates

We chose priori covariates from round t to include in our analytic models previously shown to be associated with residential relocation and/or self-reported mobility, but not a product of either. These covariates were included to produce unbiased estimates of our main exposure (mobility difficulty). Therefore we did not report all associations to avoid overinterpretation of covariate effects (Westreich & Greenland, 2013). We included demographic characteristics (age, gender, race/ethnicity, marital status, education, and non-housing wealth), and variables that indicated health status (number of self-reported comorbid medical conditions, number of overnight hospitalizations in the past year, falls in the past year, ability to complete self-care activities, and the NHATS short physical performance battery, SPPB). Additionally, we incorporated the following housing-related characteristics: home ownership, number of household members, and years living in current housing. Because barriers in the housing environment can both drive difficulty with mobility and residential relocation (Clarke, 2014; Granbom et al., 2019), we also adjusted for environmental barriers at round t. For detailed descriptions of these variables see supplemental methods/Table S.2 and previously published work (Jensen-Battaglia et al., 2024).

Statistical Analysis

Observations with missing data for variables of interest were case-wise deleted, with the exception of those missing one component of their SPPB score, for whom we imputed values using multiple Markov chain Monte Carlo method with 5 datasets. We reported descriptive statistics at the baseline round for all participants and by chronic condition. We estimated adjusted odds ratios (ORs) using multinomial logistic regression with non-movers as the referent category. Our results include fully adjusted results from two timeframes: (1) the baseline round (time t = 2011, t + 1 = 2012 using the NHATS analytic weights) to reflect our analytic sample without loss to follow up, and (2) pooled observations over the entire follow up period (2011-2019) from models with robust standard errors, and a clustering correction at the participant level to correct for within-participant correlation in the outcome. As participants who contributed more rounds of follow up are likely to differ from those lost to follow up, we also applied inverse probability of complete follow up weights to models run on the pooled sample as previously described (Jensen-Battaglia et al., 2024; Schmidt & Woll, 2017).

We included effect modifiers (chronic conditions) as independent variables and as interaction terms with the exposure (mobility difficulty). As older adults commonly experience more than one chronic condition, our models included all three interaction terms while adjusting for total number of comorbidities. We followed the STrengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines (von Elm et al., 2008)—see Supplemental File 2. All p values were from two-sided tests with α = 0.05, and analysis was completed using SAS v.9.4 (SAS Institute Inc., Cary, NC).

Results

Sample Composition

There were 7232 unique participants from the 2011 NHATS cohort contributing 32,051 observations where they were community dwelling in the round t. The majority of missing data on chronic conditions was related to our decision to exclude participants who reported cancer in NHATS Rounds 2 and 3, where skin cancer status was unknown (1227 observations from 281 unique participants). Our final analytic sample had 25,103 observations from 5673 unique participants with complete data for all variables of interest. See Supplemental Table S.3 for detailed participant flow.

Baseline Characteristics

As shown in Table 1, of the 5673 unique participants, 2807 (50%) did not report a diagnosis of cancer, MI, or stroke, or meet criteria for possible or probable dementia during the baseline round (2011). Of the remaining 2866 with at least one chronic condition, 1,083 (38%) had cancer, 1,310 (46%) had MI or stroke, and 1349 (47%) had dementia. Among participants with any of these chronic conditions, nearly three-quarters (2075, 74%) had one, 25% (706) had two, and only 3% (85) reported all three conditions. Most participants with cancer and MI/stroke reported the condition in the baseline round (1083/1,255, 86%; and 1310/1748, 75%, respectively). However, for possible/probable dementia, only 56% of cases (1349/2417) were present in the baseline round, with the remainder spread over the following rounds (see Supplemental Table S.4) Those with chronic conditions were older, had lower educational attainment, and were less likely to own their home. Participants without chronic conditions reported the least mobility difficulty [mean = 0.56 (SD = 0.03) on the 0-9 scale], while those with cancer, myocardial infarction, and dementia reported successively higher amounts [0.99 (0.06), 1.61 (0.06), and 2.14 (0.07), respectively].

Table 1.

First round (2011) characteristics by chronic condition

		No CC		Cancer		MI/stroke		Dementia
Variable	Category	N	%	N	%	N	%	N	%
Age in years	65–69	752	26.79	160	14.77	169	12.90	101	7.49
	70–74	684	24.37	222	20.50	242	18.47	180	13.34
	75–79	570	20.31	242	22.35	258	19.69	239	17.72
	80–84	472	16.82	219	20.22	319	24.35	342	25.35
	85–89	217	7.73	147	13.57	197	15.04	263	19.50
	90+	112	3.99	93	8.59	125	9.54	224	16.60
Gender	Male	1115	39.72	500	46.17	654	49.92	577	42.77
Gender	Female	1692	60.28	583	53.83	656	50.08	772	57.23
Race/ethnicity	White/non-Hispanic	2040	72.68	794	73.31	896	68.40	742	55.00
Race/ethnicity	Other^a	761	27.11	289	26.69	412	31.45	605	44.85
Education	Less than high school	534	19.02	248	22.90	444	33.89	647	47.96
	High school and some college	1367	48.70	523	48.29	628	47.94	507	37.58
	College graduate	906	32.28	312	28.81	238	18.17	195	14.46
Marital status	Unmarried	1315	46.85	515	47.55	681	51.98	834	61.82
Marital status	Married	1492	53.15	568	52.45	629	48.02	515	38.18
Number of household members	Lives alone	876	31.21	317	29.27	374	28.55	400	29.65
	2 household members	1514	53.94	575	53.09	657	50.15	582	43.14
	3 or more household members	417	14.86	191	17.63	279	21.30	367	27.21
NHATS SPPB <10		2113	75.28	917	84.67	1201	91.68	1292	95.77
Number of comorbid conditions	0	387	13.79	.	.	.	.	92	6.82
	1–3	2139	76.20	606	55.96	492	37.56	768	56.93
	4–7	281	10.01	467	43.12	798	60.92	475	35.21
	8-10	.	.	10	0.92	20	1.53	14	1.04
Any fall in the past month		203	7.23	111	10.25	194	14.81	233	17.27
Going outside the home	Fully able, decreased frequency or uses mobility device	2445	87.10	849	78.39	864	65.95	746	55.30
	Difficulty	196	6.98	87	8.03	154	11.76	149	11.05
	Does not do or receives assistance	110	3.92	106	9.79	205	15.65	331	24.54
	Receives assistance and has unmet need	56	2.00	41	3.79	87	6.64	123	9.12
Getting around inside the home	Fully able, decreased frequency or uses mobility device	2431	86.60	858	79.22	900	68.70	817	60.56
	Difficulty	289	10.30	146	13.48	229	17.48	241	17.87
	Does not do or receives assistance	46	1.64	46	4.25	97	7.40	174	12.90
	Receives assistance and has unmet need	41	1.46	33	3.05	84	6.41	117	8.67
Getting in and out of bed	Fully able, decreased frequency or uses mobility device	2388	85.07	850	78.49	900	68.70	848	62.86
	Difficulty	368	13.11	173	15.97	252	19.24	288	21.35
	Receives assistance	36	1.28	45	4.16	113	8.63	158	11.71
	Receives assistance and has unmet need	15	0.53	15	1.39	45	3.44	55	4.08
Bathing	Fully able	2556	91.06	885	81.72	942	71.91	852	63.16
	Difficulty	169	6.02	92	8.49	149	11.37	155	11.49
	Does not do or receives assistance	82	2.92	106	9.79	219	16.72	342	25.35
Dressing	Fully able	2459	87.60	845	78.02	900	68.70	835	61.90
	Difficulty	183	6.52	92	8.49	142	10.84	141	10.45
	Does not do or receives assistance	165	5.88	146	13.48	268	20.46	373	27.65
Eating	Fully able	2730	97.26	993	91.69	1121	85.57	1055	78.21
	Difficulty	45	1.60	29	2.68	57	4.35	81	6.00
	Does not do or receives assistance	32	1.14	61	5.63	132	10.08	213	15.79
Toileting	Fully able	2676	95.33	981	90.58	1098	83.82	1032	76.50
	Difficulty	109	3.88	60	5.54	122	9.31	151	11.19
	Does not do or receives assistance	22	0.78	42	3.88	90	6.87	166	12.31
Housing status	Owns home	2267	80.76	832	76.82	926	70.69	840	62.27
	Rents home	336	11.97	147	13.57	243	18.55	284	21.05
	Other arrangement	204	7.27	104	9.60	141	10.76	225	16.68
Housing tenure (years)	Less than 8 years	742	26.43	273	25.21	400	30.53	396	29.36
	8 to 20 years	727	25.90	267	24.65	306	23.36	286	21.20
	21 to 39 years	682	24.30	256	23.64	298	22.75	276	20.46
	40 years or longer	656	23.37	287	26.50	306	23.36	391	28.98
Multiple floors in building or housing unit without stair lift or elevator		1382	49.23	525	48.48	526	40.15	511	37.88
Entry steps without ramp		2037	72.57	783	72.30	902	68.85	882	65.38
Bed bath and kitchen on different floors		323	11.51	103	9.51	81	6.18	90	6.67
Shower or tub without bath seat		1835	65.37	636	58.73	672	51.30	664	49.22
No grab bars in tub or shower		1601	57.04	570	52.63	618	47.18	628	46.55
No grab bars in toilet		2450	87.28	899	83.01	1023	78.09	1004	74.43
Any permanent relocation between 2011 and 2012		137	4.88	59	5.45	99	7.56	133	9.86
Types of permanent relocation	Moved within community to home with more barriers	34	1.21	9	0.83	23	1.76	27	2.00
	Moved within community to home with fewer barriers	75	2.67	26	2.40	38	2.90	33	2.45
	Moved to nursing home or residential care	28	1.00	24	2.22	38	2.90	73	5.41
Total^b		2807	100.00	1083	100.00	1310	100.00	1349	100.00

^aIncludes Black/non-Hispanic, Other/non-Hispanic, and Hispanic.

^bTotals do not sum to sample N as participants may have >1 chronic condition.

Residential Relocation

From 2011-2019, 1,180 participants relocated a total of 1,438 times, with 196 moving two or more times [multiple moves mean = 2.3 moves (SD = 0.70), range 2, 6]. While 5% of those without chronic conditions relocated in the first year, 6% of those with cancer, 8% of those with MI/stroke, and 10% of those with dementia had a permanent change in residential address between 2011 and 2012. We also found a greater proportion of those with chronic conditions reported living in their current home for less than 2 years at the baseline round vs. those without conditions in the baseline round: no chronic conditions 324 (9%), cancer 159 (11%), MI/stroke 261 (15%), dementia 320 (16%).

Residential Relocation: Adjusted Results

Direct Effects

Figure 1 displays adjusted ORs of relocation associated with mobility difficulty and each chronic condition. For relocation between 2011 and 2012 (left panel), dementia was significantly associated with greater odds of a move to nursing home (OR 2.86, 95% CI 1.77, 4.63), while we did not find evidence of association between cancer or MI/stroke and odds of moving to any destination. Mobility difficulty was positively associated with moving within the community to housing with fewer environmental barriers (OR 1.30, 95% CI 1.10, 1.53), but not the other destinations. This differed slightly from results across the full 2011-2019 follow up period (right panel) where mobility difficulty was no longer independently associated with relocation to any setting, and cancer was negatively associated with moves to homes with more barriers (OR 0.40, 95% CI 0.21, 0.77) but positively associated with moves to nursing home/residential care (OR 1.88, 95% CI 1.09, 3.23). There was no evidence that the other conditions were associated with relocation of any type. See Supplemental Table S.6 for full results.

Figure 1.

Adjusted associations of difficulty with mobility and chronic conditions with relocation destinations

Interaction Effects

Figure 2 shows adjusted ORs of relocation associated with mobility difficulty among older adults with each type of chronic condition. The left panel shows the estimates for the first year of follow up (2011-2012), and the right shows the equivalent associations for the 2011-2019 time period. Mobility difficulty was not statistically significantly associated with relocation to housing with more barriers or to nursing home/residential care among those with any of the chronic conditions (p > 0.05 for all). Although mobility difficulty remained positively associated with relocation to housing with fewer barriers among those with each type of chronic condition (OR range 1.25 to 1.49), the strength of association did not differ from those without chronic conditions. While those with cancer demonstrated numerically greater odds of relocation in response to mobility difficulty than those with MI/stroke or dementia, this difference was not statistically significant (p = 0.234 and p = 0.221, respectively). As shown in the full model results in Supplemental Table S.7a, the betas for interaction terms between chronic conditions and mobility difficulty were small in size and not statistically significant for any level of the outcome. We additionally found the odds of relocation to housing with fewer barriers remained higher for those with greater mobility difficulty (OR 1.34; 95% CI 1.12, 1.62) who did not have dementia, cancer, or MI/stroke. Having dementia was associated with relocation to nursing home/residential care among those who also reported no mobility difficulty (OR 1.79; 95% CI 1.36, 2.36). The ORs from these models are complemented by the top panels (A) of Figures 3 –5 displaying predicted probability of relocation from 2011-2012 across levels of mobility difficulty by chronic condition. Across all levels of difficulty the average probability of relocation to each setting is low (<0.01) and similar for those with and without chronic conditions. The probability of relocation rises as mobility difficulty increases for moves to housing with fewer barriers while other destinations do not exhibit a clear directional relationship.

Figure 2.

Adjusted association of mobility difficulty with relocation among those with each type of chronic condition

Figure 3.

Predicted probability of relocation to housing with ≥barriers 2011-2012 (A) and 2011-2019 (B) associated with mobility difficulty and chronic condition

Figure 4.

Predicted probability of relocation to housing with <barriers 2011-2012 (A) and 2011-2019 (B) associated with mobility difficulty and chronic condition

Figure 5.

Predicted probability of relocation to nursing home or residential care 2011-2012 (A) and 2011-2019 (B) associated with mobility difficulty and chronic condition

In the full follow up period (2011-2019, right panel of Figure 2), mobility difficulty was not statistically significantly associated with relocation to any destination for those with any of the chronic conditions (p > 0.05 for all). Although the point estimates differed numerically for the association with relocation to housing with fewer environment barriers (Dementia OR: 1.15, MI/stroke OR: 0.92, Cancer OR: 1.39), we do not find evidence of interaction between these chronic conditions and mobility difficulty (p for interaction >0.05 for all levels of the outcome). As shown in the full results in Supplemental Table S.7b, we find those who have cancer and report no mobility difficulty were less likely to move to housing with more barriers (OR 0.59; 95% CI 0.41, 0.84). Patterns of predicted probability of relocation at each level of mobility difficulty for 2011-2019 (Panel B of Figures 3 –5) are similar to the 2011-2012 results, with all estimates showing significant overlap for those with vs. without chronic conditions.

Discussion

Our study of community-dwelling U.S. Medicare beneficiaries shows that while mobility difficulty is associated with relocation to housing with fewer environmental barriers, the strength of this association does not appear to differ across three exemplar chronic conditions. Consistent with prior literature, we find that dementia is positively associated with moves to nursing home or residential care even after adjusting for confounding demographic and health characteristics. We also show that cancer is associated with relocation to a nursing home in the following year, although we only observed this over the longer 8-year period. However, there was no evidence that chronic conditions were associated with relocation within the community, or that these conditions modified the effect of mobility difficulty on relocation. This suggests that older Americans prioritize aging in place and optimize their housing environments primarily in response to their functional status, rather than medical diagnoses. Although this is contrary to our hypotheses that motivated this project, these results add to the evidence that screening for mobility difficulty among all older adults can help identify those who may require more accessible living environments.

Chronic Conditions and Relocation

Previous analysis of community-dwelling older adults linked dementia with moves to a nursing home over timeframes ranging from 6 months to 12 years, even after adjusting for physical function (Bharucha et al., 2004; Granbom et al., 2019; Stineman et al., 2012). Although we only observed this association in the first year of follow up, we focused on permanent changes in residence. It is possible older adults who were most likely to relocate in response to their dementia did so, while others made proactive decisions to stay in their housing for the remaining follow up period (Auriemma et al., 2024). We further found that cancer was associated with relocation to nursing home over the full follow up period, while MI and stroke were not associated with any type of relocation. This contrasts with evidence showing U.S. older adults with incident cancer had an increased probability of moving within the community over the next 10 years, and another study finding those who experienced a cardiovascular event had twice the odds of moving in the following year (Friedman et al., 2016; Lovasi et al., 2014). However, the majority of participants in our sample had prevalent cancer and MI/stroke in the baseline round. This suggests that relocation is more closely tied to new changes in health status, rather than stable conditions which can be managed in current housing. In addition, those with longstanding chronic conditions may have already made health-related moves—as suggested by the greater proportion in our sample who reported moving in the 2 years prior to baseline.

Chronic Conditions as Modifiers of the Association Between Mobility Difficulty and Relocation

We did not find evidence that chronic conditions modified the association between mobility difficulty and relocation. This suggests that it is mobility difficulty itself that drives older adults to compensate for poor person-environment fit by relocating to housing with lower environmental demand. In short, differences in expectations for future health were not distinct enough across these conditions to change how participants selectively prioritized aging in place vs. improving person-environment fit by moving. It may be that chronic conditions drive relocation to housing with higher levels of care (e.g., nursing home) through their influence on trajectories of health status such as symptom burden, physical and cognitive function, which we did not incorporate into our measurement of these conditions (Wang et al., 2013). Alternatively, chronic conditions may modify the association between mobility difficulty and relocation primarily during their active medical management (e.g., post-hospitalization for MI, during surgical or chemotherapy treatment for cancer). This is supported by work showing acute events like falls and hospitalizations are associated with plans to relocate (Samuel et al., 2015; Stoeckel & Porell, 2010). Older adults living with chronic conditions including dementia learn to leverage selection, optimization, and compensation techniques to maintain function in their current environments, and time since diagnosis may be an important factor in developing skills to compensate for condition-related changes in function (Zhang & Radhakrishnan, 2018). Future work with detailed data on treatment, disease progression and prognosis is needed to evaluate these interactions further.

Strengths

Our analysis has several important strengths. First, we utilize a nationally representative and contemporary data source which has substantial detail on housing characteristics and clear temporality between exposure and outcome. Second, we attempt to reduce bias from health-related loss to follow up by employing inverse probability weighting, reporting results from the first round of follow up where all participants are present, and using proxy-reported information for deceased participants. Finally, by evaluating modification of the association between mobility difficulty and relocation by chronic conditions, rather than the chronic conditions themselves, we add to the literature understanding the mechanism by which chronic conditions influence outcomes important to older adults.

Limitations

This work should also be understood in the context of its limitations. Most significantly, we characterized chronic conditions as prevalent and adjusted for other health characteristics. However, we were lacking detailed information on disease and treatment trajectory in our current dataset, and it may be that those older adults experiencing active manifestations and management of their disease represent a meaningfully different subgroup. Second, while our study is motivated by the importance of aging in place, health-related relocation is a complex decision influenced by a variety of external factors including future health expectations/prognosis, access to informal care/social support, available resources, the housing choice set, and many other factors (Roy et al., 2018). It may be that some older adults who moved did so for positive reasons, while others who did not move were forced to stay in place due to lack of resources (Granbom et al., 2021). Conversely, some who did not move may have compensated for their mobility difficulty (or optimized for their future mobility needs) in other ways—such as home modifications to their current housing, or obtaining additional informal care (Campbell et al., 2021; Kelly et al., 2014). Although evaluating these other means of optimization and compensation was beyond the scope of our analysis, they remain an important area for future work. While our analysis controlled for many potential confounders, there remain unmeasured factors which may modify the impact of mobility difficulty on relocation to a greater extent than or in conjunction with chronic conditions. Future qualitative research is needed to contextualize how older adults select and optimize their housing when experiencing mobility difficulty in conjunction with specific diagnoses.

Conclusions

Our study provides preliminary evidence that functioning, rather than medical diagnosis, is an important indicator of older adults who are struggling to age in place. Older adults are resourceful and electively select to continue doing mobility tasks independently by optimizing their approach and compensating with adaptive equipment (Remillard et al., 2019). So when older adults report mobility difficulty, those of us involved in their care should pay attention. Unfortunately, we do not routinely screen for mobility in ambulatory practice settings (Shuman et al., 2023). Although there are many measures to assess mobility limitation (Chung et al., 2015), some may be less pragmatic for busy clinical environments. Our study suggests there is value in simply asking about mobility activities. This may be accomplished as part of a brief geriatric assessment screening or Medicare Annual Wellness visit, which includes questions about self-care activities and falls (eCFR :: 42 CFR 410.15, 2025; Tatum et al., 2018). Findings from these assessments offer an opportunity to consider both the person and the fit of their home environment, and can inform referral for simple home modifications to help older adults avoid relocation (Hwang et al., 2011). As older Americans live longer with chronic conditions, asking about mobility difficulty can serve as an important first step toward optimizing home environments and supporting selected aging-related goals.

Supplemental Material

Supplemental material - Physical Function and Residential Relocation of Older Adults Living With Chronic Conditions

Supplemental material for Physical Function and Residential Relocation of Older Adults Living With Chronic Conditions by Marielle Jensen-Battaglia, Kah Poh Loh, Emily M. Agree, Robert C. Block, Ying Wang, Supriya G. Mohile and Christopher L. Seplaki in Journal of Applied Gerontology

Supplemental Material

Supplemental material - Physical Function and Residential Relocation of Older Adults Living With Chronic Conditions

Footnotes

ORCID iDs

Marielle Jensen-Battaglia

Ying Wang

Ethical Considerations

This study uses publicly available and sensitive use data provided by the National Health and Aging Trends (NHATS) which obtains informed consent from participants as described in the Data Collection Procedures documentation at .

Consent to Participate

The data for this study were previously collected by NHATS, which obtains informed consent from each participant. The University of Rochester Research Subjects Review Board approved the current secondary analysis of this existing data as exempt from requiring ethical review (#STUDY00008378).

Author Contributions

MJB contributed to conceptualization, methodology, formal analysis, funding acquisition, investigation, validation, and writing of the manuscript. KPL contributed to study concept, funding acquisition (as a sponsor), and supervision. EMG contributed to study concept, data curation, methodology, and supervision. RCB contributed to study concept, manuscript writing, and supervision. YW contributed to data curation, investigation, and validation. SGM contributed to study concept, funding acquisition (as a sponsor), and supervision. CLS contributed to study concept, data curation, funding acquisition (as a sponsor), methodology, and supervision. All authors contributed to interpretation of results and edits to the manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Cancer Institute at the National Institutes of Health (grant number F99CA284180) to M. Jensen-Battaglia. The funder had no role in study design, analysis, manuscript preparation or decision to publish. The National Health and Aging Trends Study is produced and distributed by with funding from the National Institute on Aging (grant number U01AG032947).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Data for this study come from years 2011-2019 of the National Health and Aging Trends Study (NHATS) public use and sensitive sample person files. The NHATS is produced and distributed by https://www.nhats.org with funding from the National Institute on Aging (grant number U01AG032947). Data are available to approved researchers by completing the NHATS Sensitive Data Investigator Form, NHATS Sensitive Data Use Agreement, and NHATS Sensitive Data Supplemental Agreement with Research Staff (if applicable) as described by: .

Supplemental Material

Supplemental material for this article is available online.

References

Ashcraft

A. S.

Owen

D. C.

Feng

(2006). A comparison of cognitive and functional care differences in four long-term care settings. Journal of the American Medical Directors Association, 7(2), 96–101. https://doi.org/10.1016/j.jamda.2005.10.005

Auriemma

C. L.

Butt

M. I.

McMillan

Silvestri

J. A.

Chow

Bahti

Klaiman

Harkins

Karlawish

Halpern

S. D.

(2024). “What choice do we have?” reactive and proactive decision-making for aging in place with dementia. Journal of the American Geriatrics Society, 1(11), 3398–3412. https://doi.org/10.1111/JGS.19140

Baltes

P. B.

Baltes

M. M.

(1990). Psychological perspectives on successful aging: The model of selective optimization with compensation. In: Successful aging: Perspectives from the behavioral sciences/the european science foundation: University of Cambrige.

Bharucha

A. J.

Pandav

Shen

Dodge

H. H.

Ganguli

(2004). Predictors of nursing facility admission: A 12-Year epidemiological study in the United States. Journal of the American Geriatrics Society, 52(3), 434–439. https://doi.org/10.1111/j.1532-5415.2004.52118.x

Brim

Fromhold

Blaney

(2021). Older adults’ self-reported barriers to aging in place. Journal of Applied Gerontology, 40(12), 1678–1686. https://doi.org/10.1177/0733464820988800

Campbell

Stewart

Brunkert

Campbell-Enns

Gruneir

Halas

Hoben

Scott

Wagg

Doupe

(2021). Prioritizing supports and services to help older adults age in place: A Delphi study comparing the perspectives of family/friend care partners and healthcare stakeholders. PLoS ONE, 16(11), e0259387. https://doi.org/10.1371/JOURNAL.PONE.0259387

Chung

Demiris

Thompson

H. J.

(2015). Instruments to assess mobility limitation in community-dwelling older adults: A systematic review. Journal of Aging and Physical Activity, 23(2), 298–313. https://doi.org/10.1123/JAPA.2013-0181

Clarke

P. J.

(2014). The role of the built environment and assistive devices for outdoor mobility in later life. Journals of Gerontology - Series B Psychological Sciences and Social Sciences, 69(Suppl 1), S8–S15. https://doi.org/10.1093/geronb/gbu121

Cohen

J. A.

Verghese

Zwerling

J. L.

(2016). Cognition and gait in older people. Maturitas, 93(1), 73–77. https://doi.org/10.1016/j.maturitas.2016.05.005

10.

Davis

M. R.

(2021). New AARP survey reveals older adults want to age in place. https://www.aarp.org/home-family/your-home/info-2021/home-and-community-preferences-survey.html

11.

Dhamoon

M. S.

Longstreth

W. T.

Bartz

T. M.

Kaplan

R. C.

Elkind

M. S. V.

(2017). Disability trajectories before and after stroke and myocardial infarction: The cardiovascular health study. JAMA Neurology, 74(12), 1439–1445. https://doi.org/10.1001/JAMANEUROL.2017.2802

12.

eCFR :: 42 CFR 410.15 . (2025). Annual wellness visits providing Personalized Prevention Plan Services: Conditions for and limitations on coverage. Retrieved January 14, 2025, from. https://www.ecfr.gov/current/title-42/chapter-IV/subchapter-B/part-410/subpart-B/section-410_15

13.

Freedman

V. A.

Kasper

J. D.

(2019). Cohort profile: The national health and aging trends study (NHATS). International Journal of Epidemiology, 48(4), 1044–1045g. https://doi.org/10.1093/ije/dyz109

14.

Friedman

E. M.

Weden

M. M.

Shih

R. A.

Kovalchik

Singh

Escarce

(2016). Functioning, forgetting, or failing health: Which factors are associated with a community-based move among older adults? Journals of Gerontology - Series B Psychological Sciences and Social Sciences, 71(6), 1120–1130. https://doi.org/10.1093/geronb/gbv075

15.

Gill

T. M.

Allore

H. G.

Hardy

S. E.

Guo

(2006). The dynamic nature of mobility disability in older persons. Journal of the American Geriatrics Society, 54(2), 248–254. https://doi.org/10.1111/J.1532-5415.2005.00586.X

16.

Gill

T. M.

Williams

C. S.

(2018). Development and validation of a functional outcome measure in the national health and aging trends study. The Journals of Gerontology: Series A, 73(8), 1111–1118. https://doi.org/10.1093/gerona/gly091

17.

Granbom

Löfqvist

Horstmann

Haak

Iwarsson

(2014). Relocation to ordinary or special housing in very old age: Aspects of housing and health. European Journal of Ageing, 11(1), 55–65. https://doi.org/10.1007/s10433-013-0287-3

18.

Granbom

Nkimbeng

Roberts

L. C.

Gitlin

L. N.

Taylor

J. L.

Szanton

S. L.

(2021). “so I am stuck, but it´s OK”: Residential reasoning and housing decision-making of low-income older adults with disabilities in Baltimore, Maryland. Housing and Society, 48(1), 43–59. https://doi.org/10.1080/08882746.2020.1816782

19.

Granbom

Perrin

Szanton

Cudjoe

Gitlin

L. N.

(2019). Household accessibility and residential relocation in older adults. The Journals of Gerontology: Series B, 74(7), e72–e83. https://doi.org/10.1093/geronb/gby131.K.M.

20.

Granbom

Slaug

Löfqvist

Oswald

Iwarsson

(2016). Community relocation in very old age: Changes in housing accessibility. The American Journal of Occupational Therapy, 70(2), 7002270020p1–7002270020p9. https://doi.org/10.5014/ajot.2016.016147

21.

Hajduk

A. M.

Murphy

T. E.

Geda

M. E.

Dodson

J. A.

Tsang

Haghighat

Tinetti

M. E.

Gill

T. M.

Chaudhry

S. I.

(2019). Association between mobility measured during hospitalization and functional outcomes in older adults with acute myocardial infarction in the SILVER-AMI study. JAMA Internal Medicine, 179(12), 1669–1677. https://doi.org/10.1001/JAMAINTERNMED.2019.4114

22.

Hwang

Cummings

Sixsmith

(2011). Impacts of home modifications on aging-in-place. Journal of Housing for the Elderly, 25(3), 246–257. https://doi.org/10.1080/02763893.2011.595611

23.

Jensen-Battaglia

Loh

K. P.

Wang

Block

R. C.

Mohile

S. G.

Agree

E. M.

Seplaki

C. L.

(2024). Residential relocation of community-dwelling older adults: The role of physical function and the housing environment. Journal of Aging and Health, 37(10), 654–668. https://doi.org/10.1177/08982643241293585

24.

Kasper

Freedman

Spillman

(2013). Classification of persons by dementia status in the national health and aging trends study. Technical Paper #5 . https://www.NHATS.org

25.

Kelly

A. J.

Fausset

C. B.

Rogers

Fisk

A. D.

(2014). Responding to home maintenance challenge scenarios: The role of selection, optimization, and compensation in aging-in-place. Journal of Applied Gerontology : The Official Journal of the Southern Gerontological Society, 33(8), 1018–1042. https://doi.org/10.1177/0733464812456631

26.

Lawton

M. P.

Nahemow

(1973). Ecology and the aging process. In: Eisdorfer

Lawton

(Eds.), The psychology of adult development and aging (pp. 619–674). American Psychological Association. https://doi.org/10.1037/10044-020

27.

Lin

S. F.

Beck

A. N.

Finch

B. K.

(2016). The dynamic contribution of chronic conditions to temporal trends in disability among U.S. adults. Disability and Health Journal, 9(2), 332–340. https://doi.org/10.1016/J.DHJO.2015.11.006

28.

Lovasi

G. S.

Richardson

J. M.

Rodriguez

C. J.

Kop

W. J.

Ahmed

Brown

A. F.

Greenlee

Siscovick

D. S.

(2014). Residential relocation by older adults in response to incident cardiovascular health events: A case-crossover analysis. Journal of Environmental and Public Health, 2014(1), 951971. https://doi.org/10.1155/2014/951971

29.

Mohile

S. G.

Xian

Dale

Fisher

S. G.

Rodin

Morrow

G. R.

Neugut

Hall

(2009). Association of a cancer diagnosis with vulnerability and frailty in older medicare beneficiaries. Journal of the National Cancer Institute, 101(17), 1206–1215. https://doi.org/10.1093/jnci/djp239

30.

Pope

N. D.

Kang

(2010). Residential relocation in later life: A comparison of proactive and reactive moves. Journal of Housing for the Elderly, 24(2), 193–207. https://doi.org/10.1080/02763891003757122

31.

Presley

C. J.

Han

Leo-Summers

Hurria

Gross

C. P.

Davidoff

A. J.

Allore

H. G.

Gill

T. M.

(2019). Functional trajectories before and after a new cancer diagnosis among community-dwelling older adults. Journal of Geriatric Oncology, 10(1), 60–67. https://doi.org/10.1016/j.jgo.2018.05.017

32.

Remillard

E. T.

Fausset

C. B.

Fain

W. B.

Bowers

B. J.

(2019). Aging with long-term mobility impairment: Maintaining activities of daily living via selection, optimization, and compensation. The Gerontologist, 59(3), 559–569. https://doi.org/10.1093/geront/gnx186

33.

Roy

Dubé

Després

Freitas

Légaré

(2018). Choosing between staying at home or moving: A systematic review of factors influencing housing decisions among frail older adults. PLOS ONE, 13(1), Article e0189266. https://doi.org/10.1371/journal.pone.0189266

34.

Samuel

L. J.

Glass

T. A.

Thorpe

R. J. J.

Szanton

S. L.

Roth

D. L.

(2015). Household and neighborhood conditions partially account for associations between education and physical capacity in the national health and aging trends study. Social Science & Medicine, 128(1), 67–75. https://doi.org/10.1016/j.socscimed.2015.01.009

35.

Schmidt

S. C. E.

Woll

(2017). Longitudinal drop-out and weighting against its bias. BMC Medical Research Methodology, 17(1), 164. https://doi.org/10.1186/s12874-017-0446-x

36.

Shuman

Brach

J. S.

Bean

J. F.

Freburger

J. K.

(2023). Prevalence and predictors of ambulatory care physicians’ documentation of mobility limitations in older adults. Archives of Physical Medicine and Rehabilitation, 104(5), 719–727. https://doi.org/10.1016/J.APMR.2022.11.018

37.

Smedegaard

Kragholm

Numé

A. K.

Charlot

M. G.

Gislason

G. H.

Hansen

P. R.

(2018). Nursing home admission after myocardial infarction in the elderly: A nationwide cohort study. PloS One, 13(8), e0202177. https://doi.org/10.1371/JOURNAL.PONE.0202177

38.

Stineman

M. G.

Xie

Streim

J. E.

Pan

Kurichi

J. E.

Henry-Sánchez

J. T.

Zhang

Saliba

(2012). Home accessibility, living circumstances, stage of activity limitation, and nursing home use. Archives of Physical Medicine and Rehabilitation, 93(9), 1609–1616. https://doi.org/10.1016/j.apmr.2012.03.027

39.

Stoeckel

K. J.

Porell

(2010). Do older adults anticipate relocating? The relationship between housing relocation expectations and falls. Journal of Applied Gerontology, 29(2), 231–250. https://doi.org/10.1177/0733464809335595

40.

Tatum

P. E.

Talebreza

Ross

J. S.

(2018). Geriatric assessment: An office-based approach. American Family Physician, 97(12). https://pubmed.ncbi.nlm.nih.gov/30216018/

41.

von Elm

Altman

D. G.

Egger

Pocock

S. J.

Gøtzsche

P. C.

Vandenbroucke

J. P.

STROBE Initiative . (2008). The strengthening the reporting of observational studies in epidemiology (STROBE) statement: Guidelines for reporting observational studies. Journal of Clinical Epidemiology, 61(4), 344–349. https://doi.org/10.1016/J.JCLINEPI.2007.11.008

42.

Wang

S. Y.

Shamliyan

T. A.

Talley

K. M. C.

Ramakrishnan

Kane

R. L.

(2013). Not just specific diseases: Systematic review of the association of geriatric syndromes with hospitalization or nursing home admission. Archives of Gerontology and Geriatrics, 57(1), 16–26. https://doi.org/10.1016/J.ARCHGER.2013.03.007

43.

Westreich

Greenland

(2013). The table 2 fallacy: Presenting and interpreting confounder and modifier coefficients. American Journal of Epidemiology, 177(4), 292–298. https://doi.org/10.1093/aje/kws412

44.

Wolff

J. L.

Spillman

B. C.

Freedman

V. A.

Kasper

J. D.

(2016). A national profile of family and unpaid caregivers who assist older adults with health care activities. JAMA Internal Medicine, 176(3), 372–379. https://doi.org/10.1001/jamainternmed.2015.7664

45.

Zhang

Radhakrishnan

(2018). Evidence on selection, optimization, and compensation strategies to optimize aging with multiple chronic conditions: A literature review. Geriatric Nursing, 39(5), 534–542. https://doi.org/10.1016/j.gerinurse.2018.02.013

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.13 MB

0.83 MB