Abstract
Objectives
To examine prevalence of mobility device use in US community-dwelling older adults, including older adults with cancer history (‘survivors’), and to estimate mobility disability noting variation by cancer history, cancer site and other factors to improve early detection of mobility limitations.
Design
Cross-sectional analysis from the 2011 National Health and Aging Trends Study
Setting
In-person interviews in the homes of study participants.
Participants
Nationally representative sample of community-dwelling Medicare beneficiaries, ages 65+ (n=6,080, including 1,203 survivors).
Measurements
Participants were asked about: cancer history, pain that limited activity, mobility device use (e.g., canes, walkers, wheelchairs and scooters), history of falls and medical conditions plus assessed for approximate mobility disability using a 3-Meter gait speed test. The results were scored on a scale of 0–4 (0=lowest, 4=highest) using criteria from the National Institute on Aging.
Results
Nineteen percent of older adults and 23% of survivors reported using ≥1 mobility device, most commonly a single-point cane. Approximately 10% of breast, 6% of prostate, and 3% of colorectal cancer survivors reported using ≥2 devices in the past month. Survivors had lower mean gait speed scores (2.27) than adults without cancer history (2.39). In regression models, survivors were 18% less likely than adults without cancer history to score high on the gait speed test (OR=0.82, p<.05). Prior mobility device use, history of multiple falls, unhealthy weight, Black race, multimorbidity and pain that limited activity were associated with lower gait speed scores in all participants (all p<.05).
Conclusions
A greater proportion of older survivors used mobility devices than adults without cancer history. Mobility device use varied by cancer site and was highest in breast, colorectal and gynecologic cancer survivors. Survivors were also more likely to show signs of mobility disability, based on gait speed, compared to adults without cancer history. These indications, though modest, suggest that older survivors may require special attention to functional changes in survivorship.
Keywords: Cancer survivors, mobility device, gait, physical performance, NHATS
INTRODUCTION
More than 6 million adults in the United States use mobility devices and approximately two-thirds of these users are adults older than 65 years1. Concurrently, the aging of the US population has increased the cancer burden for older adults. In 2016, 62% of cancer survivors were 65 years or older and roughly one-third also experienced comorbidities which may impede cancer recovery 2. Common side effects of cancer treatment (e.g., pain, neuropathy, bone loss) along with age-related conditions (e.g., sarcopenia, arthritis), may hinder mobility, activities of daily living and quality of life 3. Despite these established challenges, the precise proportion of older cancer survivors that have mobility limitations and may benefit from mobility devices to facilitate daily function is unknown 4, 5. Further, there is a paucity of data currently available on prevalence of mobility device use and mobility disability risk factors specifically among older cancer survivors, a high-risk population, to inform clinical application 6, 7.
Physical performance assessment is increasingly recognized as an evidence-based way to determine overall functional status and to disentangle functional aging from chronological aging 8–10. However, it is not routinely used in assessing older adults that have been diagnosed with cancer (‘survivors’)11. Some studies have shown that pre-diagnosis physical performance assessment may be predictive of subsequent disability in cancer survivors 12 and may also be a potent prognostic biomarker 13. Gait speed assessment, in particular, offers a simple and efficient way to clinically assess potential mobility impairments as well as illness-related gait changes over time, including cardiovascular disease14, cognitive impairment 15 and mortality trajectory16. Short walking tests, including the 3-meter gait test, have been demonstrated as valid and meaningful predictors of mobility disability in several longitudinal cohort studies in research with adults 65 years and older 17. Assessing and monitoring gait speed provides a reliable way to detect changes in a timely, and precise manner during cancer recovery. Having a clearer understanding of both mobility device use and mobility disability risk factors among survivors (including survivors of different cancer sites) will provide essential data to inform interventions to reduce mobility impairments and optimize function in this vulnerable older adult population.
The goals of this study were to 1) describe the prevalence of mobility device use among older survivors compared to adults without cancer history; 2) compare mobility device use and mobility disability (approximated through gait speed scores) by cancer site and; 3) assess the association of cancer history, sociodemographic and common health conditions on mobility disability in older adults. These data may provide a method for identification of increased risk of mobility disability in patients with cancer history and guide clinicians in supporting patients in cancer recovery.
METHODS
Data
Data from the 2011 National Health and Aging Trends Study (NHATS) were analyzed (http://www.nhats.org). The NHATS is funded by the U.S. National Institute on Aging to examine late life trends in disability and to advance understanding of functional changes in U.S adults who are Medicare beneficiaries, age 65 and older. The study protocol was implemented by investigators at Johns Hopkins University. All study participants provided written informed consent. This study utilized a multistage sampling design with oversampling of Black, non-Hispanic persons and the oldest-old (≥85 years). Data collection consisted of in-home standardized interviews. Special permission was obtained from the NHATS Data Confidentiality Committee to access sensitive information, including site of cancer diagnosis.
Study population
A nationally representative sample of community-dwelling adults age 65 and older (n=8245) was enrolled in the NHATS using the Medicare enrollment database as the sampling frame (71% survey response rate). Participants living in nursing homes that were not expected to return to their previous residence (5.7%) and those who did not complete the in-person interview (2%) were excluded from this analysis. Participants with proxy respondents, in circumstances of dementia, significant cognitive impairment, speech impairment, or severe illness (6.7%) were also excluded from the analysis. Owing to incomplete data on skin cancer subtype, skin cancer survivors (7.9%), were also excluded from the analysis. This resulted in a final analytic sample of 6,080 community-dwelling older adults (4,877 adults without cancer history and 1,203 survivors).
Measures
Participant Characteristics
Demographic variables from the baseline interview included age, gender, self-identified race/ethnicity and education. Age was further categorized into three categories (65–74, 75–84, 85+) and race/ethnicity was categorized into four groups (non-Hispanic white, non-Hispanic Black, Hispanic, and other). Education was categorized into five groups (<9 years, 9–11 years, high school graduate, some college/vocational school, and ≥ college degree). Participants were asked about their marital status and categories were created into two groups (ever married/never married).
Clinical Characteristics – Cancer Diagnosis, Medical Conditions
Participants were asked if they had been bothered by pain in the last month (Yes/No), a common concern among older adults18. Participants were also asked if they had ever been diagnosed with cancer and if so, which site (i.e., prostate, breast, cervical/ovarian/uterine, colorectal, bladder, kidney, skin or other cancer). Participants were asked if, in the last month, they took any medicines prescribed by a doctor (Yes/No). Body mass index (BMI) was calculated from measured height and weight with obesity defined as BMI ≥ 30.0 kg/m2. Participants were asked how many times they fell within the last 12 months. We categorized those with fewer than 2 falls as ‘low fall risk’ and 2 or more falls as ‘high risk’ or positive for history of multiple falls.
In addition to cancer, participants were asked (Yes/No) about occurrence of ten common age-related conditions (i.e., arthritis, heart disease, diabetes, osteoporosis, hip fracture, visual impairment, stroke, depression and dementia). We further categorized number of medical conditions with 0–2 conditions (low co-morbidity burden) and 3 or more conditions (i.e., high burden or multimorbidity).
Mobility Device Use
Participants were asked to report any mobility device use in the month prior to the baseline interview, including use of a cane, walker, wheelchair (manual, power, electric, or motorized), and/or scooter. Categories of device use were created based on number of devices used in the last month (0, 1, ≥2).
Gait Speed and Mobility Disability
Mobility disability was estimated using the gait speed test, which is one component of the Short Physical Performance Battery 19. Similar walking tests have been demonstrated as valid predictors of mobility disability in multiple longitudinal cohorts studies in a variety of older adult populations 10,17. The NHATS gait speed test is timed on a 3-meter course at a person’s usual pace. Participants start from a standing position and time was marked when the last foot crossed over the 3-meter mark. Participants were allowed to attempt the 3-meter walk two times. The final score reflects the average of the two attempts or the score of one if only one attempt was completed. The scoring categories (range 0 to 4, 0=lowest, 4=highest) for the NHATS measures represent quartiles of the weighted distribution for non-missing, non-zero values. The approximate cut points in meters per second for NHATS gait speed categories were: 0=no attempt; 1=≤.579; 2=.580-.748; 3=.749-.904; 4=≥.905 20. It is estimated that approximately 5% of participants or less were not eligible to attempt the test for safety reasons or scored a 0 on the test 20. As demonstrated by Freedman, et al. and others, individual physical performance items have reasonable test-retest reliability 21, 22.
Statistical Analysis
To understand the discrete differences between cancer survivors and adults without cancer history, we first analyzed demographic variables for adults without cancer history and survivors and for the total analytic sample. Differences in demographic characteristics, number of medical conditions, pain limitations, prescribed medications and gait speed scores were evaluated using the adjusted Wald statistic.
Prevalence of mobility device use among all adults was calculated and compared to survivors using weighted proportions. Prevalence of mobility device use was also estimated by cancer site. Differences in mobility device use by cancer history and cancer site were evaluated using the adjusted Wald statistic. Total mean and individual gait speed scores were estimated among adults without cancer history, cancer survivors and by cancer site.
Two multivariable logistic regression models were used to estimate the association of sociodemographic (e.g., gender, age, education and race/ethnicity,) and clinical characteristics (cancer history, medical conditions, pain that limited activity, history of falls and body mass index) with 1) mobility device use and also with 2) mobility disability, which we estimated using the gait speed score. For the second model, we categorized gait speed as binary variable, such that regression coefficients assessed likelihood of achieving the highest score on the gait speed test v. not. Prior history of mobility device use was included as a covariate in the model estimating mobility disability. Logistic regression coefficients were exponentiated and expressed as odds ratios (with 95% confidence intervals) for ease of interpretation. All data were analyzed using Stata SE v. 14 (Stata Corporation, College Station, TX). To account for non-response, oversampling, and incomplete interviews, analytic weights were applied.
Following completion of these analyses, we also stratified the two main models by cancer history to help identify any unique contribution of cancer history in older adults of the same age to mobility device use or gait speed, adjusting for covariates as before.
RESULTS
In our analytic sample of community-dwelling adults aged 65 and older in the United States, there were a total of 1,203 (20%) survivors and 4,877 (80%) adults without cancer history. A greater proportion of participants were female (57.2%) than male. Compared to adults without cancer history, survivors overall were older (mean age=75.4; CI: 75.0–75.9%, p<0.05) and mostly white (83.4%; CI: 80.7–85.8%, p<0.05). Prescription medication use (94.4%; CI: 92.4–95.9%) and limitations of activity due to pain (58.5%, CI: 53.7–63.2%) were also higher in cancer survivors (Table 1).
Table 1.
Characteristics of Participants by Cancer History
| Variables | No Cancer History (n=4,877) | Cancer survivors (n=1,203) | Total (n=6,080) |
|---|---|---|---|
| Age (Years) | |||
| Overall Mean [95% CI]; SE | 74.2 [74.0,74.4], 0.10* | 75.4 [75.0,75.9], 0.23* | 74.4 [74.2,74.6], 0.09* |
| 65–74 | 58.6 [57.2,60.0] * | 50.0 [46.5,53.5] * | 57.1 [55.9,58.3] * |
| 75–84 | 32.1 [30.9,33.4] * | 38.9 [35.7,42.1] * | 33.3 [32.3,34.4] * |
| 85+ | 9.3 [8.56,10.06] * | 11.2 [9.72,12.76] * | 9.62 [8.94,10.34] |
|
Gender Proportion, [95% CI] |
|||
| Female | 57.4 [55.9,59.0] | 55.9 [52.9,58.9] | 57.2 [55.6,58.7] |
|
Race/ethnicity Proportion, [95% CI] |
|||
| Non-Hispanic White | 79.5 [77.5,81.5] * | 83.4 [80.7,85.8] * | 80.2 [78.3,82.0] * |
| Non-Hispanic Black | 8.98 [8.07,9.99] * | 8.99 [7.52,10.7] * | 8.98 [8.1,9.96] * |
| Hispanic | 8.08 [6.86,9.49] * | 4.5 [3.17,6.37] * | 7.44 [6.38,8.66] * |
| Other | 3.4 [2.49,4.63] * | 3.07 [2.12,4.43] * | 3.34 [2.54,4.39] * |
|
Education Proportion, [95% CI] |
|||
| < 9 years | 9.99 [8.8,11.3] | 8.41 [6.74,10.4] | 9.71 [8.57,11] |
| 9–11 years | 12 [10.9,13.2] | 10.7 [8.91,12.9] | 11.8 [10.8,12.9] |
| High school graduate | 27.8 [26.3,29.3] | 27.7 [24.8,30.7] | 27.7 [26.3,29.2] |
| Some college/vocational school | 26.1 [24.6,27.5] | 28.7 [26,31.6] | 26.5 [25.2,27.9] |
| ≥ College degree | 24.2 [22.1,26.4] | 24.5 [21.3,28] | 24.2 [22.1,26.5] |
|
Marital status Proportion, [95% CI] |
|||
| Ever married | 93.9 [92.9,94.7] | 94.2 [92.2,95.7] | 93.9 [93,94.8] |
|
Body Mass Index kg/m2 (mean, [95% CI]) |
27.8 [27.6,28.0]; 0.11 | 27.7[27.4,28.0]; 0.14 | 27.8 [27.6,28.0]; 0.09 |
|
Body Mass Index, Proportion by category, [95% CI] |
|||
| <18.5 | 1.74 [1.45,2.1] | 2.35 [1.6,3.43] | 1.85 [1.55,2.22] |
| 18.5–24.9 | 31.1 [29.6,32.6] | 30.9 [28.2,33.8] | 31.1 [29.6,32.5] |
| 25–29 | 38 [36.5,39.5] | 39.1 [36.3,42.1] | 38.2 [36.8,39.6] |
| ≥ 30 | 29.2 [27.6,30.9] | 27.6 [25.3,30] | 28.9 [27.4,30.4] |
|
Medical Conditions (excl cancer), (Proportion, [95% CI]) |
|||
| Arthritis | 51.9 [50.3,53.6] | 55.4 [52.2,58.6] | 52.6 [51.1,54] |
| Osteoporosis | 19.9 [18.6,21.2] | 21.7 [19.2,24.4] | 20.2 [19.1,21.4] |
| Hip Fracture | 27.4 [25.7,29.2] | 29.7 [25.6,34] | 27.8 [26.3,29.4] |
| Heart Disease | 16.2 [15.1,17.3] | 16.6 [14.4,19] | 16.3 [15.3,17.2] |
| Diabetes Mellitus | 23.2 [21.9,24.6] | 23.7 [21,26.7] | 23.3 [22,24.7] |
| Stroke | 8.61 [7.85,9.44] | 9.1 [7.48,11] | 8.7 [7.92,9.54] |
| Dementia | 1.96 [1.67,2.29] | 2.52 [1.63,3.88] | 2.06 [1.75,2.41] |
| Visual Impairment | 55.1 [53.4,56.7] | 56 [52.4,59.6] | 55.2 [53.9,56.5] |
| Depression | 26.6 [25.2,28.1] | 28.7 [26.2,31.3] | 27 [25.7,28.3] |
| Number of Medical Conditions | |||
| Overall Mean [95% CI]; SE | 2.21 [2.17,2.26]; 0.023 | 2.33 [2.21,2.44]; 0.057 | 2.24 [2.19,2.28]; 0.021 |
| 0–2 (proportion low comorbidity burden, [95% CI]) | 52.64 [50.8,54.48] | 48.6 [44.86,52.36] | 51.9 [50.36,53.44] |
| 3 or more (high burden/multimorbidity), [95% CI]) | 47.36 [45.52,49.2] | 51.4 [47.64,55.14] | 48.1 [46.56,49.64] |
|
Bothered by Pain Proportion, [95% CI] |
|||
| Yes | 50.7 [49.2,52.2] * | 55.5 [52.3,58.5] * | 51.5 [50.1,52.9] * |
|
Pain ever limits activity Proportion, [95% CI] |
|||
| Yes | 53.7 [51.3,56.1] | 58.5 [53.7,63.2] | 54.6 [52.5,56.7] |
|
History of Multiple Falls Proportion, [95% CI] |
11.67 [10.7, 12.73] | 14.06 [11.96, 16.47] | 12.1 [11.19, 13.07] |
|
Takes prescribed meds Proportion, [95% CI] |
|||
| Yes | 90 [87.8,90.1] * | 94.4 [92.4,95.9] * | 90 [89,90.9] * |
|
3-meter gait speed score (0–4, 0=lowest;4=highest) Overall Mean [95% CI]; SE |
2.39 [2.34,2.45], 0.027 | 2.27 [2.18,2.35], 0.043 | 2.37 [2.32,2.42], 0.025 |
| 3-meter gait speed score by cancer site (Mean [95% CI]); SE | |||
| Breast Cancer | -- | 2.19 [2.03, 2.35], 0.08 | -- |
| Prostate | -- | 2.52 [2.38, 2.67], 0.07 | -- |
| Colorectal | -- | 2.26 [1.97, 2.55], 0.14 | -- |
| Gynecologic | -- | 2.26 [2.01, 2.51], 0.12 | -- |
p<0.05
CI=Confidence Interval
SE=standard error
Among survivors, the most commonly diagnosed cancer site was prostate (n=344, 29%), followed by breast (n=314, 26.0%), colorectal (n=116, 9.6%), gynecologic (including cervical/ovarian/uterine cases) (n=109, 9%), bladder (n=50, 4%), kidney cancer (n=41, 3.4%), or other cancer diagnoses (n=229, 19%). Overall, the distribution of medical conditions was similar in adults with and without cancer history. Visual impairment was the most commonly reported condition overall (55.2%), followed by arthritis (52.6%), hip fracture (27.8%), depression (27%) and diabetes (23%). History of multiple falls was slightly higher in survivors (14%). The prevalence of arthritis, hip fractures and dementia were slighter higher in cancer survivors compared to adults without cancer history, but by and large, the health profile was comparable (Table 1).
Mobility Device Use
Prevalence of Mobility Device Use by Cancer Status and Cancer Site
Twenty percent (95% CI: 19–21.2%) of participants reported using any mobility device in the last month (Figure 1). Among mobility device users, 36.6% (7.35%; 95% CI: 6.67–8.08) reported using two or more devices in the last month. The most commonly reported mobility device was the single-point cane (15.3%; 95% CI: 14.4, 16.2) and the least commonly reported device was the scooter (2.1%; 95% CI: 1.7, 2.59).
Figure 1.
Prevalence of Mobility Device Use in Older Adults by Cancer History (Cancer History v. No Cancer History) (n≈6080)
Compared to adults without cancer history, mobility device use among survivors was higher (23.4%; 95% CI: 21.2, 25.8, p<0.05). Ten percent (95% CI: 7.37–13.8%, p<0.05) of breast cancer survivors reported the use of two or more mobility devices in the past month (Figure 2). Additionally, 5.7% of prostate cancer survivors (95% CI: 3.67, 8.7) and 3.4% of colorectal cancer survivors (95% CI: 1.67, 6.9) reported use of two or more mobility devices in the past month.
Figure 2.
Mobility Device Use among Cancer Survivors by Cancer Site (n≈1203)
Associations with Sociodemographic Characteristics
Older age was a significant negative correlate for mobility device use; adults 75–84 were twice as likely (OR=2.25, 95% CI 1.81–2.79, p<.001) and adults older than 85 years were nine times as likely (OR=9.4, 95% CI 6.71–13.2, p<.001) to use any mobility device, compared to adults 65–74 years. Non-Hispanic Black participants were also twice as likely (OR=2.13, 95% CI 1.72–2.64, p<.001) to use a mobility device compared to non-Hispanic white participants (Table 2).
Table 2.
Logistic Regression Results: Association of Any Mobility Device use by Cancer History, Sociodemographic and Clinical Characteristics
| Variable | Mobility Device Use | ||
|---|---|---|---|
| Odds Ratio | 95% Confidence Interval | p-value | |
| Cancer History | |||
| No | 1.00 (ref) | ||
| Yes | 1.23 | .96, 1.57 | 0.10 |
| Age | |||
| 65–74 | 1.00 (ref) | ||
| 75–84 | 2.25 | 1.81, 2.79 | <.001* |
| 85+ | 9.40 | 6.71, 13.19 | <.001* |
| Gender | |||
| Male | 1.00 (ref) | ||
| Female | 1.15 | .94, 1.40 | 0.16 |
| Race/Ethnicity | |||
| Non-Hispanic White | 1.00 (ref) | ||
| Non-Hispanic Black | 2.13 | 1.72, 2.64 | <.001* |
| Hispanic | 1.20 | .67, 2.17 | 0.54 |
| Other | .96 | .64, 1.44 | 0.84 |
| Education | |||
| <High school | 1.00 (ref) | ||
| High school | .82 | .58, 1.16 | 0.26 |
| Some college | .67 | .48,.92 | <.02* |
| ≥College grad | .72 | .52,.99 | <.04* |
| Body Mass Index kg/m2 | |||
| 18.5–24.9 | 1.00 (ref) | ||
| <18.5 | 1.26 | .60, 2.64 | 0.53 |
| 25–29 | 1.21 | .93,1.57 | 0.15 |
| ≥ 30 | 2.08 | 1.59, 2.71 | <.001* |
| Number of Medical conditions | |||
| <2 | 1.00 (ref) | ||
| 3+ | 2.13 | 1.71, 2.67 | <.001* |
| Activity Limited by Pain | |||
| No | 1.00 (ref) | ||
| Yes | 2.67 | 2.20, 3.23 | <.001* |
| History of Multiple Falls | |||
| <2 falls in last12 months | 1.00 (ref) | ||
| 2+ falls in last 12 months | 2.97 | 2.4, 3.68 | <.001* |
p<0.05
Ref=reference category
Associations with Clinical Characteristics
Individuals with three or more medical conditions were more than twice as likely as individuals with two or fewer conditions to use a mobility device (OR=2.13, 95% CI 1.71–2.67, p<.001). Adults with a BMI greater than 30 (i.e., obese) were also twice as likely (OR=2.08, 95% CI 1.59–2.71, p<.001) to report using any mobility device (Table 2). Adults that were positive for history of falls were three times more likely than those at low risk of falls to use any mobility device (OR=2.97, 95% CI 2.4, 3.68, p<.001). Additionally, adults that reported that pain limited their activities were nearly three times as likely as those who did not report that pain limited activities to use any mobility device (OR=2.67, 95% CI 2.4–3.68, p<.001).
In regression models, survivors were more likely than adults without cancer history to use any mobility device, but this was not statistically significant (Table 2).
In stratified analysis, we noted that survivors were mostly similar to adults without cancer history for factors associated with mobility device use considering covariates, except in cases of education, race and history of multiple falls (see Supplementary T1).
Mobility Disability Estimates Based on Gait Speed Scores
Survivors exhibited lower mean gait speed score than adults without cancer history. The mean score for all survivors was 2.27 (95% CI 2.18–2.35), compared to 2.39 (95% CI 2.34–2.45) in adults without cancer history. Prostate cancer survivors had the highest mean gait speed score of 2.52 (95% CI 2.38–2.67). Breast cancer survivors had the lowest mean gait speed score of all cancer sites at 2.19 (95% CI 2.03–2.35) (Table 1).
Associations with Sociodemographic Characteristics
On the gait speed test, we noted an age-related decline in speed, such that adults 75–84 years were more than 50% less likely (OR=0.5, 95% CI 0.43–0.59, p<.001) and adults 85 years and older were about 70% less likely (OR=0.31, 95% CI 0.25–0.39, p<.001) to achieve the fastest gait speed score compared to adults 65–74 years (Table 3). Female participants were significantly less likely than male participants to perform well on the gait speed test (OR=0.75, 95% CI 0.65–0.87, p<.001). Non-Hispanic Black participants were about 60% less likely to perform well on the gait speed test compared to non-Hispanic white adults (OR=0.42, 95% CI 0.33–0.53, p<0.001).
Table 3.
Logistic Regression Results: Association of Gait Speed Score by Cancer History, Sociodemographic variables and Clinical Risk Factors
| Characteristic | Gait Speed (likelihood of achieving highest score) | ||
|---|---|---|---|
| Odds Ratios | 95% Confidence Interval | p-value | |
| Cancer History | |||
| No | 1.00 (ref) | ||
| Yes | 0.82 | 0.67,1.00 | <.05* |
| Age (in years) | |||
| 65–74 | 1.00 (ref) | ||
| 75–84 | 0.5 | 0.43, 0.59 | <.001* |
| 85+ | 0.31 | 0.25, 0.39 | <.001* |
| Gender | |||
| Male | 1.00 (ref) | ||
| Female | 0.75 | 0.65, 0.87 | <.001* |
| Ethnicity | |||
| Non-Hispanic White | 1.00 (ref) | ||
| Non-Hispanic Black | 0.42 | 0.33, 0.53 | <.001* |
| Hispanic | 0.68 | 0.32, 1.47 | 0.325 |
| Other | 0.48 | 0.3, 0.75 | <.002* |
| Education | |||
| <High school | 1.00 (ref) | ||
| High school | 1.2 | 0.84,1.72 | 0.3 |
| Some college | 1.59 | 1.22, 2.07 | <.001* |
| ≥College grad | 2.66 | 1.91, 3.71 | <.001* |
| Body Mass Index kg/m2 | |||
| 18.5–24.9 | 1.00 (ref) | ||
| <18.5 | 0.53 | 0.32, 0.89 | <.02* |
| 25–29 | 1.04 | 0.82, 1.32 | 0.77 |
| ≥ 30 | 0.85 | 0.68, 1.05 | 0.14 |
| Number of Medical conditions | |||
| <2 | 1.00 (ref) | ||
| 3+ | 0.63 | 0.52,0.75 | <.001* |
| Activity Limited by Pain | |||
| No | 1.00 (ref) | ||
| Yes | 0.71 | 0.58, 0.89 | <.002* |
| Mobility Device Use | |||
| No | 1.00 (ref) | ||
| Yes | 0.17 | 0.14, 0.2 | <.001* |
| History of Multiple Falls | |||
| <2 falls in last12 months | 1.00 (ref) | ||
| 2+ falls in last 12 months | 0.67 | 0.57,0.83 | <.001* |
p<0.05
Ref=reference category
Associations with Clinical Characteristics
From our regression model, we observed that survivors were 18% less likely to score high (i.e., a 4) on the gait speed test compared to all adults without cancer history (OR=0.82, 95% CI 0.64–0.93, p<.05), controlling for covariates (Table 3). Participants that reported history of mobility device use were more than 83% less likely to achieve the fastest score on the gait speed test than non-users of mobility devices (OR=0.17, 95% CI 14–20, p<.001) (Table 3). Adults who were underweight (<.18.5 BMI) were half as likely (OR=0.53, 95% CI 0.32–0.89, p<.05) to achieve the highest score compared to adults with a healthy weight. Both chronic disease burden and pain that limited activities were associated with lower scores on the gait speed test. Adults with 3 or more medical conditions were half as likely as adults with two or fewer conditions to perform well on the test (OR=0.63, 95% CI 0.52–0.75, p<.001). Similarly, adults that reported pain that limited activities were nearly half as likely as adults without pain that limited activity to perform well on the test (OR=0.71, 95% CI 0.58–0.89, p<.001).
Although cancer history overall (using aggregated estimates of survivors) was a significant correlate for mobility disability in our regression model, we did not observe significant differences on gait speed performance when we compared this data by specific cancer sites in adjusted models.
In stratified analysis, we noted that being underweight had a strong, significant negative correlation with gait speed for cancer survivors, controlling for all other covariates (Supplementary T2). Survivors who had a BMI <18.5 were 67% less likely (OR=0.33, 95%CI 0.16–0.7, p<.001) than survivors with a healthy weight to achieve the highest score on the gait speed test.
DISCUSSION
More survivors reported using at least one mobility device compared to adults without cancer history. Similarly, survivors were less likely to score high on the gait speed test, suggesting that survivors may present with greater mobility disability compared to other adults. Further, while the relationship between cancer history and mobility device use was attenuated by adjustment for age, the relationship between cancer history and lower gait speed score persisted even after controlling for age. Consistent with other studies, greater levels of multimorbidity, history of multiple falls and older age (i.e., 85+ years) appeared to exacerbate mobility limitations in this older adult population, especially for survivors 23,24. We observed a recurring, yet nuanced influence, of cancer history on mobility measures for these older participants that could be clinically meaningful and has not been fully described. Aging and aging with cancer may require unique consideration in designing an individualized care plan.
Our study extends the current literature by providing national benchmarks to understand extent of mobility device use among US older adults, especially older survivors, an understudied population. To our knowledge, we are also the first to provide prevalence estimates of specific mobility devices used by older cancer survivors delineated by cancer site. Additionally, we used a clinically valid objective measure (i.e., 3-meter gait speed test) shown to be predictive of mobility disability in other studies to characterize the experience of older survivors. These objective tests are easily performed in clinical settings, require little training and no specialized equipment, and thus are feasible to use in a clinical, outpatient or community setting. As pain that interfered with activity was reported by more than half of participants, appropriate pain management may assist in improving physical function and potentially in reducing reliance on mobility devices. Physical activity programs for older adults and survivors to improve movement and function may be especially important to support this goal 25,26 Interventions that target healthy weight management, especially in breast cancer survivors, may have particular benefit as they had the greatest difficulty with the gait speed test compared to other survivors. Comparing physical performance test results with other frailty indicators may also be helpful in making recommendations for patients with mobility challenges. For survivors with impairments such as neuropathy and decreased balance, rehabilitation interventions (e.g., physical or occupational therapy) may also be beneficial.27 Physical therapy is chronically under-prescribed for survivors and other individuals who could greatly benefit. Several key papers report persistent mobility problems among community-dwelling cancer survivors who are not in active treatment and do not receive assessment of functional problems nor a referral to rehabilitation.5, 28,29
Given the wide variability of functional status in older adults in the US, it is not surprising that one-fifth of community-dwelling older adults rely on mobility devices to live their lives. It is also not surprising that almost one-quarter of older survivors said they rely on at least one mobility device to function. However, while understanding the extent of mobility device use in the US older adult and cancer survivor population is essential, quantifying use of mobility devices does not provide the full context for why an older adult may use a device, how they chose their device(s) and upon whose recommendation (i.e., a provider, a family member or on one’s own). Further, it does not indicate whether this is selected for temporary use or as part of a therapeutic effort (e.g., a rehabilitation program) to restore a higher level of function. These are all questions that warrant further exploration.
Our study was strong in several ways. First, we used a large sample of robust, nationally representative data to explore and characterize a growing population of US adults that are significantly changing the US healthcare landscape. We also believe we have contributed valuable information on older cancer survivors, a vastly heterogeneous population, for which there is little data on which to base critical clinical interventions. Recognizing that breast, prostate and colorectal cancer survivors reported greater use of multiple mobility devices provides a basis on which to prioritize mobility interventions for these groups. Additionally, understanding that Black older adults may be at higher risk of mobility disability presents an opportunity to engage this community in preventive and supportive care.
We faced some limitations in our study. As a dataset administered by the National Institute on Aging, we benefited from many rigorous age-related measures. However, NHATS does not include a few key variables that would be relevant for a cancer population, such as time since cancer diagnosis, stage at diagnosis, time since completion of cancer treatment and type of cancer treatment received. The sensitive files associated with the dataset includes records of cancer diagnoses, but lung cancer and subtypes of skin cancer were not available. The fact that we did not find significant differences in gait speed test scores by cancer site in our regression model may be party explained by our limited representation of cancer sites. Lung cancer survivors, for example, are known to have greater multimorbidity than survivors of other cancer sites26, so would likely have had lower gait speed scores had they been included in the study. The NHATS interview asked whether the participant had arthritis, but did ask about severity or specific areas of the body affected. We analyzed as a cross-sectional study, which does not provide an opportunity to view events prospectively. Future investigations would benefit from including these additional cancer variables and using a longitudinal and/or experimental design.
Like most adults later in life, aging for cancer survivors often occurs on a spectrum. Tools such as the gait speed test and broader geriatric assessment 30, 31 provide important opportunities to identify mobility disability risk and tailor interventions to mitigate deficits and fortify the health and quality of life of all older adults, including cancer survivors.
Supplementary Material
Supplementary T1. Association of Any Mobility Device Use by Cancer History, Sociodemographic and Clinical Characteristics
Supplementary T2. Association of Gait Speed Scores by Cancer History, Sociodemographic and Clinical Characteristics
ACKNOWLEDGEMENTS
FUNDING
NHATS is sponsored by National Institute on Aging Grant U01AG32947 and was conducted by the Johns Hopkins University.
Dr. Bluethmann is supported by a Mentored Research Scholar Grant in Applied and Clinical Research, MSRG-18–136-01-CPPB, from the American Cancer Society. Dr. Bluethmann is also funded by a Junior Faculty Development Support Grant through the Penn State CTSI (NIH UL1 TR002014).
Ms. Flores is supported by a graduate assistantship in the Department of Public Health Sciences at the Penn State College of Medicine.
Dr. Klepin is supported by the National Institute on Aging of the National Institutes of Health under Award Number R33AG059206.
The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the American Cancer Society.
Footnotes
The authors declare no conflicts of interest.
REFERENCES
- 1.Gell NM, Wallace RB, Lacroix AZ, Mroz TM, Patel KV. Mobility device use in older adults and incidence of falls and worry about falling: Findings from the 2011–2012 National Health and Aging Trends Study. J Am Geriatr Soc. 2015;63(5):853–859. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bluethmann SM, Mariotto AB, Rowland JH. Anticipating the “Silver Tsunami”: Prevalence Trajectories and Comorbidity Burden among Older Cancer Survivors in the United States. Cancer Epidemiology Biomarkers & Prevention. 2016;25(7):1029–1036. doi: 10.1158/1055-9965.EPI-16-0133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Winters-Stone KM, Horak F, Jacobs PG, et al. Falls, functioning, and disability among women with persistent symptoms of chemotherapy-induced peripheral neuropathy. Journal of Clinical Oncology. 2017;35(23):2604–2612. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Jette AM, Balcazar F, Ball L. The Promise of Assistive Technology to Enhance Activity and Work Participation [Assistive Technology; ]. 2017. [DOI] [PubMed] [Google Scholar]
- 5.Stubblefield MD. The Underutilization of Rehabilitation to Treat Physical Impairments in Breast Cancer Survivors. PM&R. 2017;9(9):S317–S323. [DOI] [PubMed] [Google Scholar]
- 6.Bradley SM, Hernandez CR. Geriatric assistive devices. Am Fam Physician. 2011;84(4):405–411. doi: d9027 [pii]. [PubMed] [Google Scholar]
- 7.Salminen A, Brandt Å, Samuelsson K, Töytäri O, Malmivaara A. Mobility devices to promote activity and participation: a systematic review. J Rehabil Med. 2009;41(9):697–706. [DOI] [PubMed] [Google Scholar]
- 8.Rejeski WJ, Brubaker PH, Goff DC, et al. Translating weight loss and physical activity programs into the community to preserve mobility in older, obese adults in poor cardiovascular health. Arch Intern Med. 2011;171(10):880–886. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hardy R, Cooper R, Sayer AA, et al. Body mass index, muscle strength and physical performance in older adults from eight cohort studies: the HALCyon programme. PloS one. 2013;8(2):e56483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Perera S, Studenski S, Newman A, et al. Are estimates of meaningful decline in mobility performance consistent among clinically important subgroups?(Health ABC study). Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences. 2014;69(10):1260–1268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Verweij NM, Schiphorst AHW, Pronk A, van den Bos F, Hamaker ME. Physical performance measures for predicting outcome in cancer patients: a systematic review. Acta Oncol. 2016;55(12):1386–1391. doi: 10.1080/0284186X.2016.1219047 [DOI] [PubMed] [Google Scholar]
- 12.Klepin HD, Geiger AM, Tooze JA, et al. Physical performance and subsequent disability and survival in older adults with malignancy: results from the health, aging and body composition study. J Am Geriatr Soc. 2010;58(1):76–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Brown J, Harhay M, Harhay M. Physical function as a prognostic biomarker among cancer survivors. Br J Cancer. 2015;112(1):194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lo AX, Donnelly JP, McGwin G Jr, Bittner V, Ahmed A, Brown CJ. Impact of gait speed and instrumental activities of daily living on all-cause mortality in adults≥ 65 years with heart failure. Am J Cardiol. 2015;115(6):797–801. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Mielke MM, Roberts RO, Savica R, et al. Assessing the temporal relationship between cognition and gait: slow gait predicts cognitive decline in the Mayo Clinic Study of Aging. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences. 2013;68(8):929–937. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.White DK, Neogi T, Nevitt MC, et al. Trajectories of gait speed predict mortality in well-functioning older adults: the Health, Aging and Body Composition study. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences. 2013;68(4):456–464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Miller ME, Magaziner J, Marsh AP, et al. Gait speed and mobility disability: Revisiting meaningful levels in diverse clinical populations. J Am Geriatr Soc. 2018;66(5):954–961. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Patel KV, Guralnik JM, Dansie EJ, Turk DC. Prevalence and impact of pain among older adults in the United States: findings from the 2011 National Health and Aging Trends Study. Pain®. 2013;154(12):2649–2657. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Guralnik JM, Simonsick EM, Ferrucci L, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol. 1994;49(2):M85–M94. [DOI] [PubMed] [Google Scholar]
- 20.Kasper JD, Freedman VA, Niefeld MR. Construction of performance-based summary measures of physical capacity in the National Health and Aging Trends Study. NHATS technical paper. 2012;4. [Google Scholar]
- 21.Cesari M, Kritchevsky SB, Newman AB, et al. Added value of physical performance measures in predicting adverse health-related events: results from the Health, Aging and Body Composition Study. J Am Geriatr Soc. 2009;57(2):251–259. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Freedman VA, Kasper JD, Cornman JC, et al. Validation of new measures of disability and functioning in the National Health and Aging Trends Study. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences. 2011;66(9):1013–1021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lucas AR, Bitting RL, Fanning J, Isom S, Rejeski WJ, Klepin HD, Kritchevsky SB. Trajectories in muscular strength and physical function among men with and without prostate cancer in the health aging and body composition study. PloS one. 2020. February 13;15(2):e0228773. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Weaver KE, Leach CR, Leng X, Danhauer SC, Klepin HD, Vaughan L, Naughton M, Chlebowski RT, Vitolins MZ, Paskett E. Physical functioning among women 80 years of age and older with and without a cancer history. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences. 2016. March 1;71(Suppl_1):S23–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Bluethmann SM, Sciamanna CN, Winkels RM, Sturgeon KM, Schmitz KH. Healthy Living After Cancer Treatment: Considerations for Clinical and Community Practice. American Journal of Lifestyle Medicine. 2018:1559827618755681. doi: 10.1177/1559827618755681. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Bluethmann S, Foo W, Winkels R, Mama S, Schmitz K. Physical Activity in Older Cancer Survivors: What Role Do Multimorbidity and Perceived Disability Play?. Journal of Aging and Physical Activity. 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Pergolotti M, Deal AM, Lavery J, Reeve BB, Muss HB. The prevalence of potentially modifiable functional deficits and the subsequent use of occupation and physical therapy by older adults with cancer. Journal of Geriatric Oncology, 6(3), 194–201. 10.1016/j.jgo.2015.01.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Cheville AL, Beck LA, Peterson TL, Marks RS, Gamble GL. The detection and treatment of cancer-related functional problems in an outpatient setting. Supportive Care in Cancer; 2009; 17:61–67. DOI 10.1007/s00520-008-0461-x. [DOI] [PubMed] [Google Scholar]
- 29.Campbell G, Hagan T, Gilbertson-White S, Houze M, Donovan H. Cancer and treatment-related symptoms are associated with mobility disability in women with ovarian cancer: A cross-sectional study. Gynecologic oncology. 2016. December 1;143(3):578–83 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Puts MTE, Santos B, Hardt J, et al. An update on a systematic review of the use of geriatric assessment for older adults in oncology. Annals of Oncology. 2014;25(2):307–315. doi: 10.1093/annonc/mdt386. [DOI] [PubMed] [Google Scholar]
- 31.Wildiers H, Heeren P, Puts M, et al. International Society of Geriatric Oncology Consensus on Geriatric Assessment in Older Patients With Cancer. Journal of Clinical Oncology. 2014;32(24):2595–2603. doi: 10.1200/JCO.2013.54.8347. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary T1. Association of Any Mobility Device Use by Cancer History, Sociodemographic and Clinical Characteristics
Supplementary T2. Association of Gait Speed Scores by Cancer History, Sociodemographic and Clinical Characteristics