Older adult cancer survivors fall significantly more often than those without cancer,1 with an increased falls incidence up to 6 months postdiagnosis.2 Those having a history of falling in the 12 months prior to cancer diagnosis are almost 4 times more likely to fall shortly after cancer treatment decision.3 Falls in cancer survivors are especially problematic due to disease-specific sequelae from the disease or treatments provided such as weakness, fatigue, decreased physical activity, and impaired cognitive function, of which each contributes to falls risk.4–9
In cancer survivors aged 65+ years, baseline cognitive function is believed to be influenced by type of cancer treatment10–12 and by symptoms such as cancer-related fatigue.13 Mild to moderate cognitive impairments have been reported in 16% to 75% of cancer survivors, specifically in those who received chemotherapy.13–17 These cognitive impairments can persist 2 to 10 years post–chemotherapy treatment18–20 and are believed to contribute to long-term cognitive changes such as the development of dementia.21 Deficits in several cognitive domains have been reported in the older survivor including visual memory, spatial function, executive function, attention, memory, and concentration,22–24 of which these same processes are often used in mobility and gait and contribute to falls risk.
Falls are multifactorial and include both internal and external risk factors. The presence of cognitive impairment has been associated with an increased risk of falls and falls-related injuries in older cancer survivors.21 However, declines in cognition that are associated with changes in gait, mobility, and falls go beyond just global cognitive deficits but have been reported in specific cognitive domains.25–27 A number of studies have examined falls risk factors in older cancer survivors, but few have described how specific cognitive processes contribute to balance, mobility, and falls. Therefore, the purpose of this study was to determine how cognitive changes contribute to balance, mobility, and falls in older cancer survivors.
The sample for this cross-sectional study comes from the 2010 wave of the Health and Retirement Study (HRS), a prospective observational study of community-dwelling people 50 years or older in the United States. The HRS is a longitudinal study that surveys a representative sample of approximately 20 000 Americans every 2 years.28 A detailed description of the study can be found at http://hrsonline.isr.umich.edu. This secondary analysis of the HRS data was approved by the Institutional Review Board of the University of Michigan-Flint. All respondents provided informed consent, and their anonymity was preserved in the process.
Data from this interview took place in 2010 through information provided by the respondent or their proxy. The sample comprised 22 034 individuals, of which 3078 were identified as having cancer. From that sample, respondents who had incomplete demographics, disease information, or physical function or cognitive information were excluded from further analyses.
HRS respondent data were included in this study if respondents were older than 65 years; had a self-reported diagnosis of cancer (excluding skin cancer) ever or since the previous HRS wave; did not have Alzheimer disease, dementia, or history of stroke; were able to stand unassisted for at least 1 minute; and did not have a recent surgery, injury, or other health condition that prevented the ability to walk.
Demographic data extracted from the data set included age (years), marital status, gender, and education. Cancer-related variables included the diagnosis of any type of cancer except skin cancer (yes/no), which was by self-report or proxy report, and time in years since cancer diagnosis. Other health-related variables included the reported frequency of engaging in mild-, moderate-, or vigorous-intensity physical activity at least once per week and anthropometric measures for calculation of body mass index ([BMI]; kg/m2). Falls data accounted for falls that occurred in the 2 years prior to the survey interview and included incidence (yes/no), number of falls, and whether an injury (yes/no) had occurred from the fall.
Physical performance data extracted included measures of strength, gait speed, and balance. To characterize the sample, dominant hand grip strength measurement data were a general measure of overall strength. Respondents first identified their hand dominance and then 2 measures of grip strength were performed for each hand using a hand-held dynamometer, alternating hands between trials. The average of the measures, in kilograms, was recorded for each hand. Gait speed was measured over 2 trials. The time in seconds needed to cover an 8.2-ft distance (98.5 in) was recorded, which was later converted to meters per second for analyses. Respondents were instructed to stand with their toes at the starting line and to walk at their normal pace past the finish line marked on the floor. The interviewer told the respondent to “begin,” and timing began when the foot passed the starting line and was in full contact with the floor and ended when the same foot fully cleared the finish line and was in full contact with the floor. The use of an assistive device was permitted during this measure if the respondent normally used the device to walk. Considered to be a necessary component of balance in community-dwelling older adults, the ability to perform tandem stance has been associated with falls risk.29 Tandem balance was assessed by recording the total amount of time, in seconds; respondents were able to stand in each position while wearing appropriate footwear (shoes with low or no heel) and while standing on a level surface. The position was held for 30 seconds, and the time ended if the respondent held the position for the full duration or if the respondent stepped out of place or grabbed the interviewer's arm.28
Cognitive data extracted included assessments of memory (immediate [IWR] and delayed word recall [DWR]), orientation, and executive function, of which each of these has been reported to decline in older cancer survivors.22–24 For IWR, the interviewer read a randomly assigned list of 10 nouns to the respondent and asked the respondent to recall as many words as possible from the list in any order.28 The number of correctly recalled words was recorded as the score for this variable. DWR required respondents to name as many of those original 10 words after a period of 5 minutes had passed and during which the respondent was engaged in other tasks.28 DWR was scored in the same fashion as IWR. Orientation was assessed by the respondent being able to accurately give the correct day of the week, month, day, and year.28 Score range from “0” indicating that no correct responses were given to “4” indicating that respondents were able to recall all items for orientation and were considered to have intact orientation. Executive function was assessed using a semantic verbal fluency test that required respondents to name as many animals as one could in 60 seconds, with the total number of named animals to comprise the score.28 The verbal fluency task measures executive function, as it requires strategic search, set-shifting, and semantic memory.30
Group assignment was based on a history of falls (yes/no) within the 2 years prior to the survey (Falls group; No Falls group). All respondent characteristics were first described for the whole sample and for each group. Then characteristics were compared by group assignment using analysis of variance for continuous variables and χ2 for categorical variables. Linear regression analyses were performed to examine associations between cognition and physical mobility and falls. Initially, this was done in the full sample to describe these associations within older adults with cancer. They were then performed on the basis of a falls history to determine whether associations between cognitive function and balance and mobility differed for each group. Covariates controlled for in regression models included age, education, gender, grip strength, and physical activity level (mild intensity).31 The cognitive predictor variables included executive function, IWR, DWR, and orientation. The outcome variables were falls incidence (number of falls), gait speed, and tandem stance time. Standardized β values of the dependent variables were provided, as they indicate the number of standard deviations (SDs) that the dependent variable will change as a result of 1 SD change in the predictor. Predicted values for each of the dependent variables were calculated using the standardized β values from the regression if P < .05. All data were analyzed using Statistical Package for the Social Sciences (SPSS, version 24), and a significance level was set at P < .05.
Overall, 573 individuals met inclusion criteria and comprised the study population. The mean age was 77.14 years (SD = 5.86), 62.1% were married, 47.7% were female, and, on average, 3.75 years (SD = 4.06) had passed since the diagnosis of cancer.
Within the sample, 38.7% (N = 222) reported falling in the 2 years prior to the survey and were assigned to the Falls group. The average number of falls was 2.94 (SD = 5.11), with a range of 1 to 50, and 27.9% reported experiencing an injurious fall. Groups did not differ in age, years since cancer diagnosis, education, marital status, or BMI. Significant differences existed between groups across multiple variables including gender, grip strength, and engagement in mild- or moderate-intensity physical activity. Differences in mobility were also found, as those in the Falls group walked slower and were not able to hold a tandem stance position as long as those in the No Falls group.
Cognitive performance differed between groups. The Falls group had more impaired recall for both IWR (4.90 words, SD = 1.71; P = .05) and DWR (3.66 words, SD = 1.95; P = 0.03) than the No Falls group who had better recall for both measures (IWR: 5.16 words, SD = 1.46; DWR: 4.01 words, SD = 1.77). Orientation and executive function were not significantly different by group assignment. Full demographic, mobility, and cognitive data can be found in Table 1.
TABLE 1 -
Demographic, Mobility, and Cognitive Data of the Sample and by Group Assignment
||Total Sample (N = 573)
||Falls Group (N = 222)
||No Falls Group (N = 351)
|Mean (SD) or Percentage
||Mean (SD) or Percentage
||Mean (SD) or Percentage
|Years since cancer diagnosis
|Gender, % female
|Marital status, % married
|Physical activity once per week
|Mild intensity, %
|Moderate intensity, %
|Vigorous intensity, %
|Physical and cognitive measures
|Body mass index, kg/m2
|Grip strength, kg
|Gait speed, m/s
|Tandem stance time, s
Abbreviations: DWR, delayed word recall; IWR, immediate word recall.
aP < .05.
Linear regression analyses of the whole sample of cancer survivors revealed significant associations between cognition and balance, mobility, and falls. A significant inverse relationship between executive function and number of falls (β = −0.18, P < .05) was found, indicating that as executive function declined, the number of falls increased in the older cancer survivor. Specifically, for every SD decline in executive function (SD = 6.28) as measured with the verbal fluency task, falls in the older cancer survivor increase by 1. The other cognitive functions of immediate and delayed recall or orientation were not associated with the number of falls.
All cognitive processes were significantly associated with gait speed: IWR (β = 0.19, P < .05), DWR (β = 0.19, P < .05), orientation (β = 0.11, P < .05), executive function (β = 0.13, P < .05). This linear association between each of these cognitive processes and gait speed indicates that as each one improves or declines by 1 SD, gait speed will also improve or decline by a rate of 0.04 m/s for changes in recall, by a rate of 0.02 m/s for changes in orientation, and by a rate of 0.03 m/s for changes in executive function.
Relationships between static balance performance and each of the cognitive processes were different after controlling for age, education, gender, and physical activity. Tandem stance was only significantly associated with DWR (β = 0.11, P < .05) and not with the other cognitive functions. This value indicates that as delayed recall improves by 1 SD, which is 1.85 words, tandem stance time would increase by 1.13 seconds. Regression results for the full sample can be found in Table 2.
TABLE 2 -
Associations Between Cognitive Function and Balance
, Mobility, and Falls
in Older Adult Cancer
||Immediate Word Recall
||Delayed Word Recall
|Number of falls
aAdjusted for age, education, gender, and physical activity level.
bP < .05.
Similar to analyses performed in the full sample, significant linear relationships between recall and gait speed were present in both groups: Falls group: IWR (β = 0.15, P < .05), DWR (β = 0.17, P < .05); No Falls group: IWR (β = 0.22, P < .05), DWR (β = 0.19, P < .05). These linear associations indicate that after controlling for age, education, gender, and physical activity, gait speed would change by a rate of 0.03 m/s for a 1 SD change in IWR and by 0.04 m/s for a 1 SD change in DWR for cancer survivors with a history of falls. In those without a history of falls, these changes would be greater: 0.05 m/s for a 1 SD change in IWR and by 0.04 m/s for a 1 SD change in DWR. Associations between gait speed and executive function differed by group, with significant associations found only in those with a history of falls (β = 0.11, P < .05), indicating that as executive function increases or decreases, a similarly directed change in gait speed by 0.02 m/s would be expected. Conversely, gait speed was only significantly associated with orientation in those who had not fallen (β = 0.11, P < .05). This relationship indicates that as orientation (to day of the week, month, day, and year) changes by 1 unit, a corresponding change in gait speed of 0.05 m/s would be expected.
Associations between cognition and balance did not differ on the basis of a falls history. No significant associations were found between tandem stance and any of the cognitive functions in either group after adjusting for age, education, gender, and physical activity level. Regression data by group can be found in Table 3.
TABLE 3 -
Associations Between Cognitive Function and Balance
, Mobility, and Falls
History in Older Adult Cancer
||Immediate Word Recall
||Delayed Word Recall
|No Falls group
aAdjusted for age, education, gender, and physical activity level.
bP < .05.
This study describes relationships between cognitive function, physical mobility, and falls in a national sample of older cancer survivors. As the cognitive processes of recall, orientation, and executive function become more impaired in this population, falls incidence increases, gait speed decreases, and balance becomes more impaired. This study adds to the body of literature identifying how cognitive processes influence mobility and falls and extends these findings to include older cancer survivors. The simple cognitive screening measures used in this study highlight these relationships and direct clinicians and researchers to consider the influence of cognitive function on physical mobility and falls in older adults with a history of cancer.
In the sample, those who reported falling had more impaired balance, slower gait speed, engaged less often in mild- or moderate-intensity physical activity, and had decreased overall general body strength. Although these characteristics have been reported to contribute to falls in older adults25 and are not necessarily unique to cancer survivors, falls history was able to detect differences in cognitive function that have not been previously reported. Delayed recall was significantly more impaired in fallers than in nonfallers and was also identified in regression analyses to be associated with mobility. Similarly, immediate recall was more impaired and associated with mobility in fallers, yet differences in IWR by group approached but did not reach statistical significance (P = .05). The effect that underlying cognitive dysfunction has on mobility emphasizes the need to screen cognitive function within falls risk management in older cancer survivors, especially in those with a history of falls.
Gait speed has been reported to be a significant predictor of disability, falls, hospitalization, and mortality in older adults.32,33 Significant associations between each of the cognitive processes and gait speed were found in the whole sample that highlight the influence that these cognitive processes have on walking speed. However, when falls history was considered, these relationships differed. Associations between both immediate and delayed recall measures and gait speed were stronger in those without a falls history (β = 0.19-0.22) than in those in the Falls group and within the total sample. Furthermore, falls history was able to discriminate that orientation was linearly associated with gait speed in the No Falls group, whereas executive function was only associated with gait speed in the Falls group. Collectively, these results indicate that different cognitive processes, beyond just executive function, are associated with gait speed and that the strength of the associations differ by cognitive function. This information is valuable in clinical settings where tracking of cognitive and physical functions should be included in the comprehensive management of falls risk. For example, if recall is screened every 6 months, a decline in remembering 2 words on the DWR test may correspond with a slowed gait speed of 0.05 m/s or greater, thus increasing falls risk in the older cancer survivor.34 However, as previous literature has reported stronger associations between cognition (specifically executive function) and gait speed in older adults without diagnosed cognitive impairment,26,27,32,33,35 additional research is needed using more comprehensive cognitive measures to further examine these relationships.
The inability to complete tandem stance has been identified as a significant predictor of falls in older adults29 and in older cancer survivors.36 Although tandem stance was significantly different between groups, it was only associated with DWR in the full sample. When relationships were examined by falls history, tandem stance was no longer associated with delayed recall. The lack of associations between this measure of balance and the other cognitive processes reported in this study (immediate recall, orientation, and executive function) suggests that they may not be used to complete tandem stance; however, further study is indicated. Furthermore, the use of one measure of balance to describe associations between cognitive function and balance in older cancer survivors is a study limitation and other studies should be conducted to determine how other measures of balance are associated with cognitive function in this population.
Orientation is reportedly the most commonly screened cognitive function completed as a part of a falls risk assessment.34 Orientation was not found to be significantly different by falls history, nor found to be associated with falls incidence in regression analyses in this sample of older cancer survivors. Given the paucity of literature describing direct or indirect associations between orientation and impaired balance or falls, these results suggest that clinicians should consider screening more than orientation as a part of the assessment of falls risk in older cancer survivors. Other measures of cognitive function, such as the FACT-Cog,37 which have been developed to detect cancer-specific changes in cognition, should be considered in practice and within studies that describe the influence of cognitive function on falls in cancer survivors.
Although HRS has data on treatment for cancer (radiation therapy, surgery, chemotherapy, or hormonal therapy), initial review of the data revealed that more than 75% of the sample of older cancer survivors were missing data for these variables and therefore were not included in the demographics or analyses. A growing body of literature has reported that cancer treatment–related variables are associated with falls risks and that the falls risk profile of an older cancer survivor is different by cancer type.7,8 Yet, the influence of cognitive function on falls risk by cancer type has not been examined and these risk profiles should consider the effect that specific cognitive processes have on falls and falls risk.
Strengths of the study include the large national sample of older cancer survivors and the use of simple cognitive and physical mobility screening measures that can be easily implemented in clinical settings. However, limitations should be considered. First, respondent or proxy data regarding having a diagnosis of cancer were gathered as a “yes/no” response and verification of medical history was not performed. In addition, the sample did not include those with skin cancer, nor those with other types of cancer but who had incomplete physical or cognitive data reported, which may limit the generalizability of results. Group assignment was completed by falls history recall over a period of 2 years, and it may be that some in the No Falls group may not have remembered falling when asked by the interviewers, especially those with impaired recall. The use of a large data set for a secondary analysis allows for the comparison of different variables, but problems with missing data and data provided by proxy respondents may result in inaccurate responses. Therefore, the results of this study may underestimate the magnitude of the relationship between cognitive function and physical mobility and falls in older adults with cancer. The presence of other factors that contribute to falls in older cancer survivors such as sensory impairments from chemotherapeutic agents was unable to be assessed and not controlled for in analyses. Finally, this was a cross-sectional study that prevents determining the causality relationship between cognitive performance and mobility and falls in cancer.
Results of this study indicate that in older cancer survivors, specific areas of cognitive function are associated with falls and with mobility, of which these relationships differ by a falls history. Furthermore, as premorbid status and cancer-related therapies are likely to contribute to the presence of impaired cognition across different cognitive domains, screening of cognitive function within a falls risk assessment should include more than just 1 measure, as impairments may exist in other areas of cognitive function in the older cancer survivor.
The author thanks Michelle Baumgart for assistance with the literature review for the manuscript.
1. Spoelstra S, Given B, Schutte D, Sikorskii A, You M, Given C. Do older adults with cancer
fall more often? A comparative analysis of falls
in those with and without cancer
. Oncol Nurs Forum. 2013;40(2):E69–E78.
2. Puts MT, Monette J, Girre V, et al. The fall rate of older community-dwelling cancer
patients. Support Care Cancer
3. Vande Walle N, Kenis C, Milisen K. Fall predictors in older cancer
patients: a multicenter prospective study. BMC Geriatr. 2014;14(135):1–13.
4. Poloni P, Omodei M, Nahas-Neto J, Uemura G, Véspoli Hde L, Nahas EA. Prevalence of low bone mineral density in postmenopausal breast cancer
survivors. Rev Bras Ginecol Obstet. 2015;37(1):30–35.
5. Rao AV, Cohen HJ. Fatigue in older cancer
patients: etiology, assessment, and treatment. Semin Oncol. 2008;35:633–642.
6. Luciani A, Jacobsen PB, Extermann M, et al. Fatigue and functional dependence in older cancer
patients. Am J Clin Oncol. 2008;31(5):424–430.
7. Huang M, Blackwood J, Johnson-Lawrence V, Godoshian M, Pfalzer L. Prevalence of self-reported falls
or walking problems in older cancer
survivors from Surveillance Epidemiology and End Results-Medicare Health Outcomes Survey. J Geriatr Oncol. 2017;8(4):255–261.
8. Bird M, Cheney MJ, Williams AD. Accidental fall rates in community-dwelling adults compared to cancer
survivors during and post-treatment: a systematic review with meta-analysis. Oncol Nurs Forum. 2016;43:E64.
9. Wildes TM, Depp B, Colditz G, Stark S. Fall-risk prediction in older adults with cancer
: an unmet need. Support Care Cancer
10. Fitch MI, Armstrong J, Tsang S. Patients' experiences with cognitive changes after chemotherapy. Can Oncol Nurs J. 2008;18(4):180–192.
11. Hede K. Chemobrain is real but may need new name. J Natl Cancer
Inst. 2008;100(3):162–163, 169.
12. Hurria A, Rosen C, Hudis C, et al. Cognitive function of older patients receiving adjuvant chemotherapy for breast cancer
: a pilot prospective longitudinal study. J Am Geriatr Soc. 2006;54(6):925–931.
13. Johns SA, Von Ah D, Brown LF, et al. Randomized controlled pilot trial of mindfulness-based stress reduction for breast and colorectal cancer
survivors: effects on cancer
-related cognitive impairment. J Cancer
14. Van Dam FS, Schagen SB, Muller MJ, et al. Impairment of cognitive function in women receiving adjuvant treatment for high-risk breast cancer
: high-dose versus standard-dose chemotherapy. J Natl Cancer
15. Schagen SB, van Dam FS, Muller MJ, Boogerd W, Lindeboom J, Bruning PF. Cognitive deficits after postoperative adjuvant chemotherapy for breast carcinoma. Cancer
16. Phillips KA, Bernhard J. Adjuvant breast cancer
treatment and cognitive function: current knowledge and research directions. J Natl Cancer
17. Tannock IF, Ahles TA, Ganz PA, Van Dam FS. Cognitive impairment associated with chemotherapy for cancer
: report of a workshop. J Clin Oncol. 2004;22(11):2233–2239.
18. Brezden CB, Phillips KA, Abdolell M, Bunston T, Tannock IF. Cognitive function in breast cancer
patients receiving adjuvant chemotherapy. J Clin Oncol. 2000;18(14):2695–2701.
19. Ahles TA, Saykin A. Cognitive effects of standard-dose chemotherapy in patients with cancer
20. Koppelmans V, Breteler MM, Boogerd W, Seynaeve C, Gundy C, Schagen SB. Neuropsychological performance in survivors of breast cancer
more than 20 years after adjuvant chemotherapy. J Clin Oncol. 2012;30(10):1080–1086.
21. Heck JE, Albert SM, Franco R, Gorin SS. Patterns of dementia diagnosis in Surveillance, Epidemiology, and End Results breast cancer
survivors who use chemotherapy. J Am Geriatr Soc. 2008;56(9):1687–1692.
22. Wieneke M, Dienst E. Neuropsychological assessment of cognitive functioning following chemotherapy for breast cancer
. Psycho-Oncology. 1995;4:61–66.
23. Williams GR, Deal AM, Nyrop KA, et al. Geriatric assessment as an aide to understanding falls
in older adults with cancer
. Support Care Cancer
24. Minisini AM, De Faccio S, Ermacora P, et al. Cognitive functions and elderly cancer
patients receiving anticancer treatment: a prospective study. Crit Rev Oncol Hematol. 2008;67(1):71–79.
25. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls
among elderly persons living in the community. N Engl J Med. 1988;319(26):1701–1707.
26. Muir SW, Gopaul K, Montero Odasso MM. The role of cognitive impairment in fall risk among older adults: a systematic review and meta-analysis. Age Ageing. 2012;41:299–308.
27. Blackwood J, Shubert T, Chase C, Fogarty K. Relationships between performance on assessments of executive function and fall risk in community dwelling older adults. J Geriatr Phys Ther. 2016;39(2):89–96.
28. Health and Retirement Study. http://hrsonline.isr.umich.edu
. Updated 2017. Accessed November 14, 2017.
29. Pajala S, Era P, Koskenvuo M, Kaprio J, Törmäkangas T, Rantanen T. Force platform balance
measures as predictors of indoor and outdoor falls
in community-dwelling women aged 63-76 years. J Gerontol A Biol Sci Med Sci. 2008;63(2):171–178.
30. Savva GM, Maty SC, Setti A, Feeney J. Cognitive and physical health of the older populations of England, the United States, and Ireland: international comparability of the Irish Longitudinal Study on Ageing. J Am Geriatr Soc. 2013;61(suppl 2):S291–S298.
31. Smith BD, Smith GL, Hurria A, Hortobagyi GN, Buchholz TA. Future of cancer
incidence in the United States: burdens upon an aging, changing nation. J Clin Oncol. 2009;27(17):2758–2765.
32. Guralnik JM, Ferrucci L, Pieper CF, et al. Lower extremity function and subsequent disability: consistency across studies, predictive models, and value of gait speed
alone compared with the short physical performance battery. J Gerontol A Biol Sci Med Sci. 2000;55(4):M221–M231.
33. Cesari M, Kritchevsky SB, Penninx BW, et al. Prognostic value of usual gait speed
in well-functioning older people—results from the Health, Aging, and Body Composition Study. J Am Geriatr Soc. 2005;53(10):1675–1680.
34. Blackwood J, Martin A. Screening for cognitive impairment as a part of fall risk management in physical therapist practice. J Geriatr Phys Ther. 2017;40(4):197–203.
35. Watson NL, Rosano C, Boudreau RM, et al. Executive function, memory, and gait speed
decline in well-functioning older adults. J Gerontol A Biol Sci Med Sci. 2010;65(10):1093–1100.
36. Chen TY, Janke MC. Predictors of falls
among community-dwelling older adults with cancer
: results from the Health and Retirement Study. Support Care Cancer
37. Wagner L, Sweet J, Butt Z, Lai JS, Cella D. Measuring patient self-reported cognitive function: development of the functional assessment of cancer
therapy-cognitive function instrument. J Support Oncol. 2009;7(6):W32–W39.