Women With Clinically Significant Fatigue After Breast Cancer Treatment Report Increased Falls and Perform Poorly on Objective Measures of Physical Fitness and Function

Abstract

Purpose: 

To determine whether there were significant differences in self-report and objective measures of physical fitness and function between women post–breast cancer (BC) treatment with and without clinically significant fatigue.

Methods: 

We performed a secondary analysis of baseline data collected from 281 postmenopausal women post–BC treatment (62 ± 10 years; time since diagnosis = 5.6 ± 3.9 years) who participated in prior exercise trials. Scores on the SF-36 (36-Item Short Form Health Survey) vitality subscale (0-100) were used to identify women with clinically significant fatigue, where a score of 45 or less represents fatigue and a score of 70 or more represents no fatigue. Objectively measured function included maximal leg press strength (kg), physical function battery, timed chair stand, Unipedal Stance Test, and gait characteristics. Self-reported outcomes included physical function, pain, mental health, sensory symptoms of chemotherapy-induced peripheral neuropathy (CIPN), habitual physical activity level, and fall history.

Results: 

Twenty percent of women (n = 57) reported clinically significant fatigue 6 years postdiagnosis, whereas 34% (n = 100) reported no fatigue. Except for leg strength, unipedal stance, step length, and base of support while walking, fatigued women had significantly worse objectively measured function than women without fatigue. Fatigued women also reported poorer physical function, greater bodily pain, and poorer mental health and were twice as likely as nonfatigued women to report symptoms of CIPN and having fallen.

Conclusions: 

Women with a history of BC, and in particular those with CIPN, should be routinely screened for clinically significant fatigue in the survivorship period. Clinical and survivorship care plans aimed at limiting functional decline and reducing falls in survivors of BC should include fatigue-reduction strategies.

Breast cancer (BC) is the most common female cancer and is the second leading cause of cancer death in the United States, with more than 250 000 new cases diagnosed each year.¹ Advances in BC diagnosis and treatment have greatly improved BC survival rates,² and there are currently more than 3 million women who have been treated for BC in the United States.¹ Despite these improvements in survival, long-term effects of cancer treatment pose a considerable problem for many cancer survivors. Female cancer survivors often report functional limitations during treatment that can persist for months or years after treatment has ended. They are more likely to report difficulties performing heavy household work, walking ½ mile, or walking up and downstairs.³ These functional limitations are more prevalent closer to diagnosis and treatment, but even long-term survivors (5 years+) remain significantly more functionally limited than women without a history of cancer.⁴ Functional declines are likely due to the effect of cancer treatment on multiple physiologic systems that influence functional status such as the neurologic, musculoskeletal, and cardiovascular systems. For example, taxanes cause sensory neuropathy in the feet,^5,6 which impairs balance and gait and increases the risk of falls and disability.⁷

Moreover, peripheral neuropathy impacts performance on objectively measured functional tests, including chair stand time, and gait speed.⁷ Cancer treatment–related declines in physical function may exacerbate age-related declines in function that accompany aging, further increasing the risk of falls and disability in older women post–BC treatment.^4,8 In addition to sensory neuropathy, other treatment-related symptoms may impact physical function in women post–BC treatment. Cancer-related fatigue (CRF), one of the most common and distressing symptoms experienced by patients with BC undergoing cancer treatment,^9,10 has consistently been shown to correlate negatively with self-report physical function.¹¹ Few studies, however, have examined the effect of clinically significant fatigue on objective measures of physical fitness and function and none have determined whether women post–BC treatment with clinically relevant fatigue report increased falls compared with those without fatigue. The purpose of the present study was to determine whether women post–BC treatment with clinically significant fatigue perform worse on objective measures of physical fitness and function and report more falls than their nonfatigued counterparts. We performed a secondary data analysis of baseline data collected from 3 independent exercise trials. Using a cross-sectional design, we compared self-reported and objective measures of physical fitness and function in postmenopausal, posttreatment women with and without clinically significant fatigue.

METHODS

Design

Baseline data collected from 291 postmenopausal women who have completed treatment of BC and who participated in one of 3 separate randomized controlled trials that tested the effects of resistance training in women with a history of BC at risk of poor musculoskeletal health and functional decline related to cancer treatment (NCT00659906, NCT00591747, and NCT00665080). Eligibility for the original trials included diagnosis of stage 0-IIIc BC, postmenopausal status, less than twice-weekly resistance training within the last month, 1 or more years past chemotherapy and/or radiation therapy, and physician clearance to participate in moderate-intensity exercise. Documented self-report cancer treatment type and a baseline 36-Item Short Form Health Survey (SF-36) vitality score were additional inclusion criteria for this secondary data analysis. The Institutional Review Board approved all parent studies.

Demographic, Behavioral, and Clinical Data

Demographic data, clinical history, and use of antiestrogen hormone therapy for BC (ie, selective estrogen receptor modulator or aromatase inhibitor therapy) were obtained by self-report at baseline. Comorbidities were assessed using the Age-Adjusted Charlson Comorbidity Index, with higher scores indicating worse health.¹² The presence or absence of symptoms associated with chemotherapy-induced peripheral neuropathy (CIPN) was based on the sensory neuropathy items in the Functional Assessment of Cancer Therapy/Gynecologic Oncology Group-Neurotoxicity.¹³ Participants were asked whether they experienced numbness, tingling, or discomfort in their feet in the past week and were then categorized into one of 2 groups: CIPN+ or CIPN−. Self-report mental health and bodily pain were assessed using the SF-36 mental health and bodily pain subscales, respectively.¹⁴ Self-report physical activity in the last 4 weeks was assessed using the Community Health Activity Model Program for Seniors (CHAMPS) questionnaire for older adults.¹⁵ The CHAMPS questionnaire asks about sedentary, low, moderate, and vigorous activities during the last 4 weeks and calculates weekly energy expended in each activity (kcal per day).

Classifying Fatigued and Nonfatigued Groups

Scores on the SF-36 v1.0 vitality subscale (VS) were used to categorize women into fatigued and nonfatigued groups. The VS is commonly used and validated in studies of both diseased and general populations¹⁶ and is the criterion standard tool to support the development of other fatigue assessment instruments.¹⁷ It has well-established internal consistency, reliability, content validity, construct validity, and criterion-related validity, having been tested in a variety of population samples.¹⁸ The 4-item VS assesses how much of the time during the past 4 weeks participants have felt full of life, full of energy, felt worn out, or felt tired. Responses are scored on a 0 to 100 scale, with higher scores corresponding to less fatigue. Scores of 45 or less represent clinically significant fatigue, and scores of 70 or more represent no fatigue.^19,20 Women who scored 45 or more but 70 or less did not meet criteria for fatigue or no fatigue and consequently were removed from the analysis.

Body Composition

Height and weight were measured with footwear, outerwear, and headwear removed using a standard stadiometer. Body composition was assessed by dual-energy x-ray absorptiometry (Hologic QDR Discovery Wi; APEX software v4.0, Hologic, Marlborough, Massachusetts), which allows for calculation of body mass, fat mass, and lean mass in the total body or specific compartments (ie, leg and trunk). All scans were performed and analyzed by personnel trained in densitometry and blinded to group assignment. Body mass index (BMI), lean mass index (LMI), fat mass index (FMI), leg LMI, and trunk FMI were derived from body weight (kg), whole-body lean mass (kg), whole-body fat mass (kg), leg lean mass (kg), and trunk fat mass (kg), respectively, divided by height (m) squared (kg/m²). Like BMI, LMI and FMI allow the comparison of body composition data between participants with different heights.^21,22

Objectively Measured and Self-reported Functional Outcomes

Objective physical function was measured using the Short Physical Performance Battery, which consists of the following 3 timed tests: 5-time chair stand, standing balance, and 4-m usual walk speed.²³ Performance on each test is scored on a 0 to 4 scale, with a total score range of 0 to 12, where a lower score means more severe functional limitations than at higher scores. We also evaluated changes in chair time stand separately because performance on this test independently predicts poor outcomes in older adults.^24–26 We also administered an additional balance test: the one-leg stance test, which involves participants standing on 1 leg under 2 conditions. In the first condition, eyes are open, while in the second, the eyes are closed. The time each position was maintained was recorded in sections up to a maximum of 30 seconds. Finally, gait characteristics (speed, cadence, step length, swing and stance time, base of support, and double support time) were analyzed by having participants walk on an electronic walkway (GAITRite; CIR System, Sparta, New Jersey). Self-reported physical function was measured using the SF-36 physical function subscale.^27–30 The physical function subscale is scored from 0 to 100, with a higher score indicating better physical function than lower scores. Fall history was determined by asking women whether they had fallen in the prior 6 months and the circumstances surrounding the fall.³¹

Analysis

We examined frequency distributions and descriptive statistics for all study variables. t tests, Mann-Whitney U tests, one-way analysis of variance, and χ² analyses were conducted to determine statistically significant group differences (fatigue vs no fatigue) in outcomes. Variables that differed significantly between the 2 groups were controlled for in subsequent analyses where appropriate. Logistic and linear regression models were used to examine the association between fatigue status and functional outcomes. Linear regression coefficients and odds ratios (with 95% confidence intervals) were used to quantify effect size for the linear and logistic regression models, respectively. All statistical tests were performed using SPSS version 24.0 (IBM Corporation, Armonk, NY, USA).

RESULTS

Study Participant Characteristics

Of the 291 women who participated in the parent studies, 281 had a documented cancer treatment type and SF-36 VS score. Of these 281 women, 20% (n = 57) met the criteria for clinically significant fatigue based on a VS score of 45 or less while 36% women (n = 100) scored 70 or more on the VS and were considered nonfatigued. Only women who were deemed to be fatigued or nonfatigued were included in subsequent analyses. Compared with nonfatigued women, those in the fatigued group (n = 57) were significantly younger, were closer to their cancer diagnosis, were more likely to report sensory symptoms of CIPN in their feet, and report poorer mental health and greater bodily pain (Table 1). Both groups reported similar levels of low-intensity physical activity. However, women in the fatigued group engaged in less moderate- to vigorous-intensity activities than their nonfatigued counterparts. Groups were similar on marital status, education, ethnicity, daily nutritional intake, and age-adjusted comorbidity index. Compared with nonfatigued women, those with fatigue had a significantly higher BMI, FMI, and trunk FMI, while there was no difference in LMI and leg LMI between the 2 groups (Table 2). Group differences in BMI, FMI, and trunk FMI were lost after controlling for level of moderate to vigorous physical activity.

TABLE 1 - Participant Demographic and Clinical Characteristics^a

Characteristic	Fatigue+ (n = 57), Mean (SD) or %	Fatigue− (n = 100), Mean (SD) or %	Pb
Age, y	59.6 (10.7)	64.2 (9.5)	.01
Race, %			.49
White	94.7	96.0
Black	3.5	1.0
Other	1.8	3.0
Non-Hispanic	98.2	100	.36
Employment, %			.07
Retired	43.9	48.0
Full- or part-time	38.6	46.0
Homemaker/unemployed	17.5	6.0
Marital status, %			.54
Married/partnered	66.7	62.0
Divorced/separated	22.8	21.0
Widowed/single	10.5	17.0
Highest education achieved, %			.14
High school or less	36.8	26.3
Associate's/technical degree	17.5	11.6
College education or higher	45.6	62.1
Breast cancer stage, %			.12
Stage 0	3.6	7.2
Stage I	36.4	48.5
Stage II	45.5	39.2
Stage III	14.6	5.2
Time since diagnosis, mo	53.2 (35.7)	75.1 (50.3)	<.01
Treatment type			.06
Radiation only	26.3	42.0
Chemotherapy only	12.3	16.0
Radiation and chemotherapy	61.4	42.0
Still on antiestrogens	53.6	45.0	.30
Age-adjusted CCI	3.8 (2.2)	3.6 (1.8)	.53
Sensory symptoms CIPN, %
Feet	56.1	18.0	<.01
Hands	50.0	23.0	<.01
Physical activity, kcal/d
Low activity	1.49 (1.09)	1.46 (0.95)	.89
Moderate to vigorous activity	0.88 (1.07)	1.64 (1.5)	<.01
Caloric intake, kcal/d	1426 (569)	1350 (434)	.36
SF-36 mental health	60.9 (17.3)	87.5 (9.2)	<.01
SF-36 bodily pain	56.4 (25.9)	76.0 (20.0)	<.01

Abbreviations: CCI, Charlson Comorbidity Index; CI, confidence intervals; CIPN, chemotherapy-induced peripheral neuropathy; SF-36, 36-Item Short Form Health Survey.

^aData represent means (SD) or percentage of the sample.

^bGroup comparisons (fatigued+ vs fatigue−) were performed using the t test or χ² test.

TABLE 2 - Group Differences in Body Composition^a

Characteristic	Unadjusted			Adjustedb			r
	Fatigued+, Mean (SD)	Fatigue−, Mean (SD)	P	Fatigue+, Mean (95% CI)	Fatigue−, Mean (95% CI)	P
BMI, kg/m2	30.2 (6.8)	28.0 (4.7)	.04	29.3 (27.9 to 30.6)	28.2 (27.2 to 29.2)	.24	0.09
FMI, kg/m2	12.2 (4.2)	10.9 (3.3)	.03	11.9 (10.9 to 12.8)	11.2 (10.5 to 11.9)	.24	0.09
Trunk FMI, kg/m2	6.2 (2.4)	5.9 (2.0)	.08	5.9 (5.4 to 6.5)	5.6 (5.2 to 6.0)	.36	0.07
LMI, kg/m2	17.0 (1.7)	16.6 (1.7)	.26	16.9 (16.4 to 17.4)	16.6 (16.2 to 17.0)	.24	0.07
Leg LMI, kg/m2	5.5 (0.9)	5.4 (0.6)	.28	5.5 (5.3 to 5.7)	5.4 (5.2 to 5.5)	.44	0.08

Abbreviations: BMI, body mass index; CI, confidence intervals; FMI, fat mass index; LMI, lean mass index.

^aData are presented as unadjusted mean (SD) or adjusted means with 95% CI for adjusted tests.

^bGroup comparisons were performed using t tests or analysis of covariance.

Objectively Measured and Self-reported Functional Outcomes

Associations between fatigue and functional outcomes were examined using logistic and linear regression models while serially adjusting for group differences identified in Tables 1 and 2. In unadjusted models (Table 3 and 4; model 1), fatigued women performed the chair stand significantly slower than women without fatigue, had slower walking speeds, took fewer steps, and spent more time in the stance and double support phase of the gait cycle. Significant differences remained after adjusting for age, time since diagnosis, presence of sensory symptoms of CIPN in the feet, and FMI (model 2) except for time in the stance and double support phase of the gait cycle. In model 3, we further adjusted for bodily pain and all but chair stand time lost significance. The group difference in chair stand time remained in fully adjusted models in which we also controlled for level of habitual moderate to vigorous physical activity. For self-report functional outcomes, fatigued women reported poorer physical function in unadjusted and fully adjusted models and were almost twice as likely as nonfatigued women to report a fall in the past 6 months (28.2%, = 0.4 vs 51.8%, = 1.5; P < .001). Significance differences in falls weakened after adjusting for age, time since treatment, presence of CIPN, and FMI (Table 4; model 2, P = .052), and group differences in past falls were lost after further adjusting for bodily pain (Table 4; model 3).

TABLE 3 - Fatigue Group Differences in Objectively Measured and Self-report Functional Outcomes^a

Outcome	Fatigue+	Fatigue−	Pb
Objectively measured
Physical performance battery	10.6 (1.6)	11.1 (1.2)	.041
Chair stand, s	13.0 (4.4)	11.1 (2.2)	.003
UST− eyes open, s	18.1 (11.6)	20.6 (10.8)	.172
UST eyes closed, s	5.3 (5.1)	5.5 (5.7)	.800
Max leg press, kg	170.9 (50.7)	169.5 (45.4)	.861
Gait characteristics
Speed, m/s	1.23 (0.22)	1.31 (0.20)	.017
Cadence, steps/min	110.1 (7.3)	115.1 (8.3)	<.001
Step length, cm	65.8 (9.0)	67.5 (7.6)	.221
Swing time, % gait cycle	35.5 (2.5)	36.6 (1.8)	.005
Stance time, % gait cycle	64.6 (2.5)	63.5 (1.8)	.002
Base of support, cm	9.3 (3.6)	8.9 (3.2)	.518
Double support time, % gait cycle	29.5 (4.9)	27.4 (3.5)	.005
Self-reported
Physical function	68.2 (24.7)	86.9 (16.5)	<.001
Fallen, %	51.8	28.2	.004

Abbreviation: UST, Unipedal Stance Test.

^aData are presented as unadjusted mean (SD) unless indicated otherwise.

^bGroup comparisons were performed using t, χ², and Mann-Whitney U tests.

TABLE 4 - Unstandardized Regression Coefficients (95% Confidence Intervals) Comparing Breast Cancer Survivors With and Without Clinically Significant Fatigue (N = 157)^a

Outcome	Model 1	Model 2	Model 3	Model 4
Objectively measured
Physical performance battery	−0.42 (−0.86 to 0.01)b	−0.48 (−0.86 to −0.9)c	−0.35 (0.76 to 0.06)	−0.31 (−0.73 to 0.11)
Chair stand, s	1.76 (0.72 to 2.81)d	1.51 (0.53 to 2.50)d	1.13 (0.09 to 2.18)c	1.09 (0.02 to 2.16)c
UST− eyes open, s	−2.23 (−5.87 to 1.42)	−3.68 (−6.76 to −0.60)c	−1.74 (−4.97 to 1.41)	−1.58 (−4.83 to 1.65)
UST eyes closed (s)	−0.16 (−2.00 to 1.67)	−1.06 (−2.73 to 0.61)	−0.98 (−2.77 to 0.80)	−1.09 (−2.90 to 0.74)
Max leg press, kg	3.64 (−11.73 to 19.0)	−9.17 (−23.45 to 5.12)	−6.28 (−21.59 to 9.02)	−5.50 (−21.09 to 10.10)
Gait characteristics
Speed, m/s	−0.08 (−0.14 to −0.01)c	−0.07 (−0.13 to −0.004)d	−0.03 (−0.10 to 0.04)	−0.20 (−0.09 to 0.04)
Cadence, steps/min	−4.83 (−7.48 to −2.17)d	−3.86 (−6.62 to −1.09)d	−1.86 (−4.70 to 0.99)	−1.71 (−4.60 to 1.19)
Step length, cm	−1.26 (−3.91 to 1.38)	−1.48 (−3.90 to 0.94)	−6.6 (−3.24 to 1.93)	−0.34 (−2.95 to 2.27)
Swing time,% gait cycle	−0.97 (−1.63 to −0.30)d	−0.52 (−1.09 to 0.06)	−0.34 (−0.95 to −0.25)	−0.30 (−0.93 to 0.33)
Stance time,% gait cycle	0.97 (0.31 to 1.64)d	0.52 (−0.53 to 1.09)	0.34 (−0.27 to 0.96)	0.31 (−0.32 to 0.93)
Base of support, cm	0.28 (−0.83 to 1.39)	−0.19 (−1.37 to 0.99)	−0.08 (−1.35 to 1.20)	−0.06 (−1.35 to 1.24)
Double support time, % gait cycle	1.88 (0.58 to 3.18)d	1.00 (−0.12 to 2.12)	0.69 (−0.51 to 1.89)	0.61 (−0.61 to 1.83)
Self-reported
Physical function	−17.57 (−24.0 to −11.17)d	−15.71 (−21.90 to −9.53)d	−8.49 (−14.17 to −2.81)d	−7.51 (−13.10 to −1.97)d
Fallen, %	0.38 (0.19 to 0.76)d	0.47 (0.22 to 1.01)e	0.59 (0.26 to 1.33)	0.645 (0.28 to 1.49)

Abbreviation: UST, Unipedal Stance Test.

^aModel 1 is not adjusted; model 2 is adjusted for age, time since diagnosis, presence of sensory symptoms of chemotherapy-induced peripheral neuropathy in the feet, and fat mass index; model 3 is further adjusted for bodily pain; and model 4 is fully adjusted including all prior covariates and self-reported participation in moderate-vigorous physical activity.

^bP = .056.

^cP < .05.

^dP < .01.

^eP = .052.

DISCUSSION

Here, we compare objective measures of physical fitness and function in postmenopausal women following treatment of BC with and without clinically significant fatigue, while controlling for treatment-related factors that impact functional outcomes. Using a cutoff score of 45 or less on the SF-36 VS, we found that 20% of women who enrolled in the parent exercise studies met criteria for clinically significant fatigue, a finding that is consistent with prior reports.³² Why a large percentage of women go onto experience persistent fatigue posttreatment is not clear, although several potential mechanisms have been proposed.³³ Fatigue is a mutifactorial symptom, the severity of which is influenced by other symptoms that frequently co-occur in cancer survivors, including pain, sleep disruption, and mood disturbance.^11,34 Thus, it is not surprising that fatigued women in the present study were more likely to report greater pain and poorer mental health than their nonfatigued counterparts. An unexpected, and to our knowledge, a novel finding was that fatigued women were twice as likely as women without fatigue to report sensory symptoms of CIPN in their feet. This finding could suggest that fatigue and CIPN are prominent symptoms of neurotoxicity as has been proposed previously.³⁵ Alternatively, more severe fatigue in women with CIPN may reflect the greater attentional demands needed to maintain balance while carrying out activities of daily living. Motor function (eg, walking or getting out of a chair) is an attentionally demanding process requiring the integration of visual, vestibular, and somatosensory inputs from changes in task and environment. Attentional load is increased when the motor task becomes more complex^36,37 (eg, walking while navigating obstacles), when the individual has poor balance^38,39 (eg, due to sensory symptoms of CIPN), or when a motor task is performed at the same time as a cognitive task (dual-tasking).⁴⁰ Attentional fatigue, which is the decreased capacity to direct attention at another process (ie, walking),⁴¹ increases in patients undergoing cancer treatment and is associated with increases in anxiety, depressive symptoms, and fatigue.^42,43 Although attentional fatigue in cancer survivors has proven to be a barrier to return to work,^44,45 the effect of attentional fatigue on functional outcomes and fall risk is not known and is worthy of investigation.

Fatigued women took an average of 13 seconds to complete the sit-to-stand test and walked, on average, 0.1 m/s slower than women without fatigue. The effect of fatigue status on these functional outcomes is clinically significant since a minimal clinically important difference is 0.1 m/s for gait speed,⁴⁶ while a time of 12 or more seconds predicts a 2.4-fold increased risk of falls in older adults.⁴⁷ Slower chair stand times and gait speed in fatigued women did not appear to be related to poorer lower extremity strength or gross balance impairments as evidenced by similar leg strength and performance on the Unipedal Stance Test, respectively. Poorer performance on these tests could be related to the increased rate of CIPN in fatigued women, who would have diminished sensorimotor feedback resulting in slower chair stand times and gait speed. However, group differences remained after controlling for the presence of CIPN, which suggests that factors other than sensorimotor deficits contribute to performance on this test in fatigued women. Bodily pain has been shown to contribute to functional decline, mobility limitations, and increased fall risk in older community-dwelling adults.⁴⁸ Bodily pain, especially in the joints of the lower extremities, could also contribute to the functional limitations and increased falls observed in fatigued women. While group differences in gait characteristics and balance were lost after controlling for level of bodily pain, group differences in chair stand time remained. Thus, our findings suggest that while bodily pain influences performance on some functional tests, factors other than pain contribute to chair stand time in fatigued women.

Although the five times sit-to-stand test and gait speed are commonly used objective tests of physical function that independently predict poor outcomes in older adults,^24–26 the chair stand test is more physically demanding. Moving from a seated to a standing position is a complex movement requiring not only muscle strength but also muscle power, balance, sensorimotor, and psychological factors.^49,50 To date, diminished lower extremity skeletal muscle power has not been examined as a contributor to functional decline in women with a history of BC with or without fatigue. Yet, it is well recognized that skeletal muscle power (force × velocity of muscle contraction) declines earlier and more rapidly than skeletal muscle strength during aging.⁵¹ When compared with traditional measures of health such as aerobic capacity or BMI, muscle power is a greater predictor of mobility and associated with higher scores on the Short Physical Performance Battery.⁵² Loss of lower extremity power is also associated with decreased function in older adults without a history of cancer.^53,54 In longitudinal studies of lower extremity power in older adults, lower power was associated with poorer physical function and increased relative risk of falls in older women.^54,55 Slower chair stand times in fatigued women may reflect diminished muscle power or a more significant decline in muscle strength and power during repeated sit to stands. Fatigability is the rate of decline in muscle strength and power during muscular exercise. Few studies have examined fatigability in cancer survivors,^56–60 and only one, to our knowledge, compared measures of fatigability in cancer survivors with and without CRF.⁶¹ Prior studies examined fatigability using 2 main strategies: (1) evaluation of voluntary and involuntary muscle activation using electromyography (EMG), a procedure where the electrical activity of muscle tissue is measured using electrodes attached to the skin; or (2) examination of skeletal muscle strength by measuring the force generated during a static isometric contraction of an isolated muscle group.⁶⁰ EMG findings must be interpreted with caution because muscle force cannot be inferred from muscle activation due to well-established limits of EMG (signal cancellation, motor unit size determination, inability to capture passive forces unrelated to voluntary muscle activation, influence of adipose tissue on signal amplitude).⁶² EMG studies in cancer survivors to date are also limited by their focus on isolated small muscle groups (ie, forearm flexors or ankle plantar flexor muscles) under isometric conditions (ie, static position), which is not how muscles typically contract when a person goes about his or her normal activities of daily living such as walking upstairs and getting out of a chair. Nor did they examine the fatigability of larger muscle groups, particularly of the lower extremities, which is most likely to decline in physical functioning.⁶³ Further work is needed to determine whether increased fatigability contributes to the functional limitations observed in fatigued cancer survivors.

There are limitations to this study, including the cross-sectional design, which does not allow us to determine whether fatigue was a cause or consequence of functional decline in female survivors of BC. Also, CRF level was not an eligibility criterion or primary outcome of the parent studies. As such, it is unclear whether fatigue existed before or was explicitly related to cancer treatment.

In summary, fatigued women performed more poorly on functional tests, reported poorer physical function, and reported almost twice the rate of falls compared with their nonfatigued counterparts. Our findings suggest that women with a history of BC treatment should be routinely screened for clinically significant fatigue long into the survivorship period, and fatigue-reduction strategies should be included in clinical and survivorship care plans aimed at limiting functional decline and reducing falls in survivors of BC.