Supportive Care in Cancer

https://doi.org/10.1007/s00520-018-4430-8

ORIGINAL ARTICLE

Qigong intervention for breast cancer survivors with complaints

of decreased cognitive function
Jamie S. Myers 1 & Melissa Mitchell 2 & Susan Krigel 3 & Andreanna Steinhoff 1 & Alyssa Boyce-White 4 &
Karla Van Goethem 3 & Mary Valla 5 & Junqiang Dai 6 & Jianghua He 6 & Wen Liu 7 & Susan M Sereika 8 & Catherine M Bender 9

Received: 20 April 2018 / Accepted: 15 August 2018

# Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract
Purpose The purpose of this pilot study was to evaluate the feasibility of an 8-week Qigong intervention to improve objectively
and subjectively assessed cognitive function in breast cancer survivors who were 2 months to 8 years post completion of
chemotherapy and radiation therapy.
Methods A randomized, single-blind, three-arm intervention pilot was conducted to compare Qigong to gentle exercise and
survivorship support. Feasibility was measured by recruitment, group session attendance, and adherence to home practice for the
two exercise groups. Changes in self-report and objectively measured cognitive function were compared between the three
groups from baseline (T1) to completion of the intervention (T2) and 4 weeks post intervention (T3).
Results Fifty participants consented (83% of desired sample) with an overall attrition rate of 28%. Attrition was highest for the
gentle exercise group (50%). Group attendance adherence ranged from 44 to 67%. The a priori established rate of 75% weekly
attendance was not achieved, nor was the goal of 75% adherence to home practice for the two exercise groups (7 to 41%). Self-
report of cognitive function improved most for the Qigong group (p = .01). Improvement was demonstrated for the Trail Making
A (gentle exercise, p = .007) and F-A-S verbal fluency (support group, p = .02) tests. Qigong participants reported the most
reduction of distress (p = .02).
Conclusions The study results suggest that mindfulness-based exercise may be superior to gentle exercise alone or survivorship
support for improving self-report of cognitive function and distress after treatment for breast cancer. The mindfulness component
may enhance the positive impact of exercise on cognitive function.

Keywords Breast cancer . Cognitive function . Qigong . Mindfulness-based exercise . Intervention

Introduction and purpose concern to breast cancer survivors and can be exacerbated
by fatigue, sleep disturbance, and psychosocial distress (rang-
Cancer and cancer treatment-related declines in subjective and ing from sadness/fear to depression/anxiety) [1–4]. The
objective cognitive functions continue to be of significant Oncology Nursing Society Practice Guidelines acknowledges

* Jamie S. Myers 5
North Kansas City Hospital, 2750 Clay Edwards Drive, North
[email protected] Kansas City, MO 64116, USA
6
Department of Biostatistics, University of Kansas Medical Center,
1
Office of Grants and Research, University of Kansas School of 3901 Rainbow Blvd, Kansas City, KS 66160, USA
Nursing, MS 4043, 3901 Rainbow Blvd, Kansas City, KS 66160,
7
USA Physical Therapy and Rehabilitation Science, University of Kansas
2
Medical Center, 3901 Rainbow Blvd, Kansas City, KS 66160, USA
Department of Radiation Oncology, University of Kansas Medical
Center, 3901 Rainbow Blvd, Kansas City, KS 66160, USA 8
Center for Research and Evaluation, University of Pittsburgh School
3
Midwest Cancer Alliance, University of Kansas Medical Center, of Nursing, 3500 Victoria St, Pittsburgh, PA 15213, USA
4350 Shawnee Mission Parkway, Fairway, KS 66205, USA 9
Nancy Glunt Hoffman Endowed Chair of Oncology Nursing,
4
University of Kansas Medical Center, 3901 Rainbow Blvd, Kansas University of Pittsburgh School of Nursing, 3500 Victoria St,
City, KS 66160, USA Pittsburgh, PA 15213, USA
 Support Care Cancer

this important area of concern for survivors and oncology 19]. Results from a number of studies (including our own
nurses [4]. However, sufficient evidence is lacking to recom- work) have demonstrated that Qigong has a positive impact
mend specific interventions as standard care. Interventional on fatigue, sleep disturbance, and distress, for a variety of
studies are needed to provide the necessary evidence to inform different patient populations [16, 20–24]. Studies designed
health care professionals’ practice for meeting the cognitive to investigate the use of Qigong for cancer-related symptoms
survivorship needs of women with breast cancer. such as fatigue and quality of life have yielded positive results
Various forms of physical activity and meditation have [13, 16, 20, 22, 23].
been studied in association with the reduction of symptoms One study has been conducted to date to investigate the
related to cancer and cancer therapy [5, 6]. The body of inter- effect of Qigong on cognitive function for cancer survivors
vention work conducted in the area of cancer and cancer [13]. This study included 81 survivors who had received or
treatment-related declines in cognitive function is growing, were receiving chemotherapy. Participants (n = 37) who re-
but published works lack consistency for randomization and ceived the Qigong intervention demonstrated a significant im-
control groups, and for the measurement of subjective and provement in subjective cognitive function compared to con-
objective cognitive function [7]. However, some positive pre- trols (n = 44) (p < .05) and were adherent to the intervention
liminary results have been demonstrated for the use of aerobic (70% attended 80% of ten weekly sessions). Objective cogni-
exercise, resistance training, mindfulness-based exercise, and tive function was not measured.
meditation in improving cognitive function [8–15]. No published research to date has included objective mea-
Adherence to aerobic and resistance training exercises can sures of cognitive function to evaluate Qigong as an interven-
be challenging due to persistent fatigue following cancer ther- tion to improve cognitive function for breast cancer survivors
apy [16]. Mindfulness-based exercise (combining the modal- after primary treatment. Previous research conducted with
ities of meditation and low-intensity physical activity, such as cancer survivors has not been designed to differentiate be-
qigong, tai chi, and yoga) can be tailored to less strenuous tween the individual effects of gentle exercise and the
formats, and therefore may be less susceptible to the adher- mindfulness-based component of breathing techniques and
ence challenges due to persistent fatigue or reduced perfor- meditation employed in interventions such as Qigong on cog-
mance status [16]. Preliminary evidence suggests that physical nitive function. Neither has previous study designs controlled
activity and meditation may influence cognitive function by for the potential effect of the group participation component
different mechanisms [10]. This hypothesis makes the combi- inherent in any facilitated group activity.
nation of physical activity and meditation very appealing as a The purpose and primary aim of this pilot study was to
potential intervention for decreased cognitive function. evaluate the feasibility of conducting a three-arm, random-
However, since low-intensity physical activity (or gentle ex- ized, single-blind, controlled intervention trial to improve
ercise) may positively affect cognitive function, studies must objectively and subjectively assessed cognitive function in
be carefully designed to discern the effect of the mindfulness- breast cancer survivors who were 2 months to 8 years post
based exercise versus that of gentle exercise alone by keeping completion of chemotherapy and radiation therapy.
the intensity of exercise constant between groups [16]. Study Interventions included 8-weekly group sessions of
design experts have suggested that usual care control groups Qigong, gentle exercise, or attention control (survivorship
may not be sufficient for interventions with a behavioral focus support). The exploratory aims were to compare changes
and that comparison groups should be included to create con- in: (1) objective or subjective cognitive function; (2) other
ditions needed to discern placebo effects, such as expectancy, cancer and cancer treatment-related symptoms (fatigue,
social support, and attention [17]. Thus, a three-arm study sleep disturbance, and distress); and (3) quality of life be-
design to explore the effect of Qigong, gentle exercise, and tween the three groups following the 8-week intervention
attention control was a logical step to begin to investigate the period.
contributions of gentle exercise, mindfulness, and the combi-
nation of the two to improve cognitive function in breast can-
cer survivors.
Qigong (of which Tai Chi is one form) is a mindfulness- Methods
based form of exercise originating from Chinese medicine
combining physical postures or movements with breathing Setting
techniques and meditation. Qigong sometimes is referred to
as a meditative-movement therapy and differs from Tai Chi in Participants were recruited from the University of Kansas
that the movements are simpler, more repetitive, and easier to Cancer Centers, including one academic and five community
learn [16, 18]. practices, and from participating Midwest Cancer Alliance
Fatigue, sleep disturbance, and distress are known to con- (MCA) sites including North Kansas City Hospital and
tribute to deficits in cognitive function and quality of life [2, Olathe Medical Center.
Support Care Cancer

Eligibility criteria shoulder-width apart. Qigong group participants were

instructed to calm their minds into a state of emptiness, not
Eligible participants included adult women (over age 18) di- to feel bad if thoughts or feelings intrude, and to synchronize
agnosed with stage I–III breast cancer who completed chemo- slow breathing with smooth arm movements. Concentration is
therapy (and radiation if received) 2 months to 8 years prior to focused on the feeling of the lower abdominal contraction and
enrollment, and who reported clinically meaningful decreased expansion. Qigong participants were instructed in the pronun-
cognitive function (a score < 59 on the Perceived Cognitive ciation of the six healing sounds and performance of accom-
Impairment subscale of the Functional Assessment for Cancer panying postures. The gentle exercise participants were
Therapy-Cognition, FACT-Cog) [25]. Ongoing anti-estrogen instructed only in the gentle arm movements and postures
therapy and anti-HER-2 therapy were allowed. Women were without the diaphragmatic breathing, clearing of the mind, or
excluded for Alzheimer’s disease, dementia, or other psycho- chanting of the six healing sounds. A very detailed description
logical diagnoses that would impact cognitive function, of the movements, postures, and sounds has been published
current/planned participation in mindfulness-based exercise previously [26].
programs, and physical limitations that would preclude gentle Training and certification of the four Qigong and gentle
(low-intensity) exercise. exercise group leaders and intervention fidelity monitoring
was carried out by a co-investigator (WL), a 29-year practi-
Procedures tioner of the six healing sounds form of Qigong. The support
group (attention control) consisted of sessions facilitated by a
At the time of consent, all participants were informed of the clinical psychologist (SK). The focus of these sessions was
invitation to attend the group of their choice following com- issues related to cancer survivorship. Participants in the sup-
pletion of data collection. The baseline assessment (T1) was port group were encouraged to share their concerns and dis-
conducted by the primary investigator (PI, JM). Height, cuss problem-solving strategies.
weight, and waist/hip circumference were measured for body
mass index calculation (BMI). Participants completed the Instruments
study questionnaire and neuropsychological testing.
Participants subsequently were randomized to one of the three The study questionnaire included demographic information
groups and informed of the result by a research assistant not collected at baseline (T1) and a series of psychometrically
involved with group leadership, neuropsychological testing, sound self-report instruments collected at all three time points
or data collection. Each group met for eight weekly 60-min (Table 1) [27–30]. The neuropsychological tests (administered
sessions. Participants randomized to either of the exercise at baseline and T3) were selected to be consistent with the core
groups (Qigong or gentle exercise) were instructed to practice tests recommended by the International Cancer and Cognition
for 15 min at home twice a day and keep a log, including any Task Force (See Table 1) [31–33].
barriers to home practice. The home practice logs were col-
lected weekly at the group sessions. At the completion of Data analyses
8 weeks (T2), group leaders administered the second study
questionnaire and a satisfaction survey. Four weeks later Descriptive statistics were used to summarize the distribution
(T3), neuropsychological testing was repeated by the PI and of all variables (SPSS Statistics 24). A preliminary analysis
the final study questionnaire was administered. Once partici- was used to assess accuracy of inputted data, potential outliers
pants completed the T3 assessment, they could attend a dif- and influential points, the amount and pattern of missing data,
ferent group if desired, without further data collection. and potential violations of assumptions necessary for planned
analyses. Fisher’s exact tests and analyses of variance (based
Intervention Fidelity on whether variables were categorical or continuous) were
used to assess for group differences at baseline and for differ-
The form of Qigong selected for the study was the six healing ences for participants who dropped out of the study compared
sounds. This form of Qigong is comprised of the synchronous to those who completed the study.
combination of controlled diaphragmatic breathing, quiet
chanting of the six healing sounds (to enhance meditative Primary aim
focus), and specific gentle movement of the arms.
Diaphragmatic breathing minimizes the role of the upper chest The percentage of the desired sample achievement was used to
and neck muscles, relying primarily on the contraction and define the success of recruitment for the three-arm study de-
relaxation of the muscles in the lower abdomen including sign. Retention was measured by the number, characteristics,
the diaphragm. Participants were instructed to assume a pos- and reasons for attrition for each group. Group means for
ture with the knees slightly bent and feet approximately weekly session attendance and documentation of twice-daily
 Support Care Cancer

Table 1 Instruments exploratory nature of these analyses. The significance level for
Self-report Neuropsychological tests all tests was 0.05.

Functional assessment of cancer therapy- Rey auditory verbal learning

cognition test (RAVLT; memory) Results
(FACT-Cog, version 3) subscales:
Perceived cognitive impairment (PCI) RAVLT 1–5 total
Perceived cognitive abilities (PCA) RAVLT interference
Demographics
Quality of life (QOL) RAVLT post interference
Fifty participants primarily were comprised of non-Hispanic,
RAVLT delay
white, post-menopausal, married women diagnosed with early
RAVLT recognition A
stage disease (Table 2). Most participants were educated at the
RAVLT recognition B
college level or above (84%) and employed full-time (56%).
Patient reported outcomes measurement F-A-S test of verbal fluency
system (PROMIS) applied cognition Most had received, or were currently receiving, anti-estrogen
short forms 8a (version 1): therapy (78%). About half had received radiation therapy.
General concerns Only a third received anti-HER2 therapy. The mean age was
Abilities 53.68 and the mean time since chemotherapy was slightly
MD Anderson Cancer Symptom Trail Making Test A and B more than 2 years. The mean BMI was 30 with a mean
Inventory (MDASI) items: (processing speed, waist/hip ratio of .86. Participants’ activity levels ranged from
executive function)
3.5 to 80 total METs (metabolic equivalent of task) per week
Fatigue
with a mean of 21.2 (range of 8–16 recommended by the
Sleep disturbance
Physical Activity Guidelines Advisory Committee, Office of
Distress
Disease Prevention and Health Promotion), although 50%
Women’s health initiative brief physical
were below 18. Nine study cohorts were held over a 20-
activity questionnaire (WHI PAQ)
(participants’ activity levels) month timeframe. Group sizes ranged from two to seven.
No differences were noted between groups at baseline ex-
cept that participants in the gentle exercise group reported
home practice were used to assess adherence. Successful fea- lower quality of life (QOL) on the FACT-Cog QOL subscale
sibility was defined as achieving a total of 45 evaluable par- (p = .004), and worse perceived cognitive function on the two
ticipants (15 in each group), 75% or greater adherence to PROMIS Cognition short forms (general concerns, higher
weekly attendance for all 3 groups and twice-daily home prac- scores = worse function, p = .046) (abilities, lower scores =
tice for the Qigong and gentle exercise groups, with 25% or worse function, p = .049) (Table 2). Baseline differences for
less attrition. participants who dropped out were significant for two vari-
ables, e.g., higher stage of disease (3 of 3 participants with
stage III disease, p = .006) and lower QOL scores (mean dif-
Exploratory aims ference = 3.1, F = 4.82, p = .033).

Mean changes in scores (and 95% confidence intervals) for Primary aim
the self-report instruments (T1 to T3 and T2 to T3) and
domain-specific neuropsychological tests (memory, process- The recruitment feasibility goal was achieved (50 participants,
ing speed, and executive function from T1 to T3) were calcu- 83% of desired sample); however, we did not achieve 15
lated for each group and used to describe change in cognitive evaluable participants in each group. Attrition was unequal
function within and among Qigong, gentle exercise, and at- across groups. Four participants dropped out of the gentle
tention control groups. Independent sample t tests or exercise group prior to attending any sessions and were ex-
Wilcoxon rank sum tests were used based on whether or not cluded from further analyses. At the T3 assessment, attrition
the normality assumption was satisfied. Mean scores for the was 21% for Qigong, 50% for gentle exercise, and 0% for the
self-report instruments and domain-specific neuropsycholog- support group (Table 3).
ical tests at T2 and T3 were used to calculate effect sizes (as Group attendance adherence of 75% was not achieved.
standardized mean difference of change scores, d) for the po- Mean weekly group session attendance was 5.2 for Qigong
tential impact of Qigong, compared to gentle exercise, and/or (52%), 4.4 for gentle exercise (44%), and 6.7 for the support
attention control conditions. Additionally, we calculated effect group (67%). The twice-daily home practice adherence goal
sizes (i.e., d) at T2 and T3 for the effect of each of the three of 75% also was not achieved for the two exercise groups:
groups on fatigue, sleep disturbance, distress, and quality of Qigong (31% overall, 41% T1 to T2 and 13% T2 to T3) and
life. No control for multiple tests was considered due to the gentle exercise (21% overall, 28% T1 to T2 and 7% T2 to T3).
Support Care Cancer

Table 2 Demographics for full sample and group differences at baseline (n = 50)

Characteristic (categorical) Freq (%) MD (%) Qigong (n = 19) Gentle exercise (n = 20) Support group (n = 11) p (Fisher’s exact)
Stage
I 21 (42) 10 9 2 .081
II 26 (52) 9 8 9
III 3 (6) 0 3 0
Ethnicity 2 (4)
Hispanic or Latino 3 (6) 1 1 1 1.0
Not Hispanic or Latino 45 (90) 17 18 10
Race
Asian 2 (4) 0 1 1 .755
Black/African American 3 (6) 1 2 0
Caucasian 45 (90) 18 17 10
Menopausal status
Pre- 2 (4) 1 0 1 .148
Peri- 4 (8) 3 0 1
Post- 44 (88) 15 20 9
Education level 1 (2)
High school 7 (14) 2 4 1 .832
College 32 (64) 12 12 8
Graduate school 10 (20) 5 3 2
Employment 1 (2)
Full time 28 (56) 10 9 9 .184
Part time 6 (12) 1 4 1
Medical leave 1 (2) 0 0 1
Retired 9 (10) 5 4 0
Unemployed 5 (10) 3 2 0
Marital status
Single 4 (8) 2 2 0 .439
Married 33 (66) 14 12 7
Divorced 6 (12) 0 3 3
Widowed 4 (8) 2 1 1
In relationship 3 (6) 1 2 0
Radiation therapy 26 (52) 9 12 5 .672
Anti-HER-2 therapy 17 (34) 10 5 17 .120
Anti-estrogen therapy 38 (78) 15 15 9 1.0
Previous smoker 21 (42) 8 9 4 .932
Diabetic 6 (12) 4 2 0 .243
Characteristic (continuous) Min, max Mn (SD) Qigong (n = 19) Gentle exercise (n = 20) Support group (n = 11) p (ANOVA)
Age (years) 29, 76 53.68 (11.19) 52.89 (11.96) 53.05 (10.74) 56.18 (11.30) .744
Time since chemo (years) .25, 8 2.3 (1.65) 2.34 (1.67) 2.429 (1.89) 1.98 (1.18) .768
BMI 22.2, 45.0 30.2 (5.33) 31.46 (6.30) 20.08 (4.31) 28.28 (5.01) .291
Waist/hip ratio .76, 1.02 .86 (.06) .88 (.068) .85 (.60) .85 (.03) .273
Total METs per week 3.5, 80.42 21.2 (15.31) 23.32 (13.52) 19.57 (18.15) 20.52 (13.38) .744
PCI 9, 57 32.32 (12.85) 33.26 (10.57) 28.55 (14.54) 37.55 (12.06) .163
PCA 4, 26 13.4 (5.47) 14.79 (5.54) 4.90 (1.10) 5.72 (.77) .118
QOL 0, 16 8.88 (4.29) 10.74 (4.13) 6.50 (3.74) 10.0 (3.69) .004**
PROMISCogTS 30.9, 62.7 42.64 (7.02) 40.21 (5.13) 45.56 (8.27) 41.52 (5.85) .046*
PROMISATS 27, 54.9 40.71 (6.88) 43.51 (6.0) 38.17 (7.37) 40.49 (5.60) .049*
Fatigue 0, 10 4.4 (2.7) 3.74 (2.96) 5.25 (2.67) 4.0 (2.0) .188
Sleep 0, 10 3.72 (2.97) 2.89 (3.11) 4.95 (2.74) 2.91 (2.55) .054
 Support Care Cancer

Table 2 (continued)
Distress 0, 9 2.9 (2.85) 2.58 (3.04) 3.55 (3.02) 2.27 (2.10) .413

*p < 0.05, **p < .01

BMI, body mass index; Freq, frequency; Mn, mean; Max, maximum; Min, minimum; MD, missing data; MET, metabolic equivalent of task; PCI,
perceived cognitive impairment; PCA, perceived cognitive ability; PROMISAbility, patient reported outcomes measurement system applied cognition,
abilities; PROMISCog, patient reported outcomes measurement system applied cognition, general concerns; QOL, quality of life; SD, standard deviation;
Sig, significance

Exploratory aims for subjective cognitive function/symptoms No significant improvements in fatigue or sleep distur-
bance scores were noted (lower scores = less severity). A trend
FACT-Cog PCI subscale scores (higher scores, better per- toward improvement was noted for gentle exercise compared
ceived cognitive function) improved for the Qigong and gen- to Qigong (p = .05, d = 0.81).
tle exercise groups and worsened for the support group Improvement in distress scores (lower scores = less sever-
(Table 4). PCI improvement for the Qigong was significantly ity) was significantly better for Qigong compared to support
better than the support group between T1 and T2 (p = .01, d = group (p = .02, d = −0.91). A trend for improvement was not-
1.14). A trend for higher improvement with gentle exercise ed between Qigong and gentle exercise (p = .06, d = −0.76).
compared to support group was noted (p = .08, d = −0.83).
Similar results were seen for improvement in FACT-Cog Exploratory aims for objective tests of cognitive function
PCA subscale scores with significantly higher scores in the
Qigong group compared to support group (p = .04, d = 0.75). The Qigong group improved more than gentle exercise on the
A trend for improvement was noted between gentle exercise Trail Making A test between T1 and T3 (lower scores = better
and support group (p = .08, d = −0.83). No significant differ- cognitive function; p = .007, d = 1.21). The gentle exercise
ence for PCI or PCA was noted between groups from T2 to group trended toward more improvement than the support
T3. group (d = 0.76). Only the Qigong group improved on the
All three group scores improved between T1 to T2 for the Trail Making B test. While not significant, moderately large
PROMIS cognitive general concerns scores (lower scores = effect sizes were noted in comparison to gentle exercise (d =
better perceived cognitive function). No significant differ- −0.43) and support groups (d = 0.56). The support group im-
ences between groups were noted. However, effect sizes for proved more than gentle exercise on the test of verbal fluency
the Qigong group were large compared to gentle exercise (d = (p = .02, d = 1.14). No correlation was seen between Bdose^
−0.65) and support group (d = −0.72). No significant differ- (total group or home practice sessions), BMI, or activity level
ences were noted between T2 and T3. The effect sizes were and subjective or objective cognitive function.
greatest for the gentle exercise group compared to Qigong
(d = 1.04) and support groups (d = 0.88). No significant group
differences were noted for PROMIS cognitive abilities short Discussion
form scores.
QOL improved for all three groups between T1 and T2. The study was faced with retention challenges, particularly in
Significant improvement was noted for gentle exercise be- the gentle exercise group. Recruitment and consent language
tween T2 and T3 when compared to support group (p = .03, were neutral when describing the three intervention groups as
d = −1.02), and the effect size compared to Qigong was large participants could not be blinded to the randomization selec-
(d = − 0.8). tion. The opportunity to attend a different group following

Table 3 Study assessment completion by group

Assessment completion

Group T1 (baseline) T2 (8 weeks) T3 (12 weeks Dropped (%)

Qigong 19 16 15 4 (21)
Gentle exercise 20 11 10 10 (50)
Support 11 11 11 0 (0)
Total dropped 0 12 2 14 (28)
Total completed 50 38 36
Support Care Cancer

Table 4 Exploratory aims results

Q T1 to T3 GE T1 to T3 SS T1 to T3 Q:GE Q:SS GE:SS

Mean change SD Mean change SD Mean change SD p d p d p d
PCI 10.94 10.47 10.27 14.16 − 0.73 9.87 .89 0.06 .01* 1.14 .05 − 0.9
PCA 2.38 4.27 2.911 4.68 −1 4.77 .76 − 0.12 .04* 0.75 .08 − 0.83
PROMISCog − 4.48 4.47 − 1.75 3.78 − 1.42 3.98 .11 − 0.65 .08 − 0.72 .84 0.09
PROMISAbl 3.37 3.38 4.17 6.51 2.15 1.68 .71 − 0.16 .23 0.43 .34 − 0.43
QOL 2 3.52 1.36 4.11 1.82 2.96 .90 0.17 1 0.05 .77 0.13
Fatigue 0.94 2.67 − 0.36 2.42 − 0.73 3 .14 0.51 .14 0.59 .64 − 0.13
Sleep 0.69 2.73 − 1.36 2.25 − 0.36 3.04 .05 0.81 .36 0.37 .39 0.37
Upset 0.13 3.56 0.18 2.48 − 0.55 2.21 .96 − 0.02 .58 0.22 .48 − 0.31
Q T2 to T3 GE T2 to T3 SS T2 to T3 Q:GE Q:SS GE:SS
Mean change SD Mean change SD Mean change SD p d p d p d
PCI 3.47 10.2 5.2 11.48 2.36 11.3 65 − 0.16 .81 0.1 .58 − 0.25
PCA 0.6 4.84 0.9 3.18 0.18 4.9 1 − 0.07 .79 0.09 .70 − 0.18
PROMISCog − 0.02 1.94 − 3.12 4.15 0.65 4.39 06 1.04 .65 − 0.21 .07 0.88
PROMISAbl − 0.79 2.86 1.55 2.3 0.49 2.61 .06 − 0.88 .32 − 0.46 .34 − 0.43
QOL 0.07 2.05 1.6 1.71 − 0.64 2.62 .06 − 0.8 .45 0.31 .03* − 1.02
Fatigue − 0.87 2.64 − 0.7 2.36 0.64 2.66 .87 − 0.07 .17 − 0.57 .24 0.53
Sleep − 0.13 2.33 0.3 1.83 − 0.73 1.74 .91 − 0.2 .22 0.28 .28 − 0.58
Upset − 0.93 2.02 0.4 1.26 0.82 1.78 .06 − 0.76 .02* − 0.91 .55 0.27
Q T1 to T3 GE T1 to T3 SS T1 to T3 Q:GE Q:SS GE:SS
Mean change SD Mean change SD Mean change SD p d p d p d
F-A-S 4.67 5.73 0.6 4.62 6.64 5.99 .07 0.76 .40 − 0.34 .02* 1.14
RAVLT 1–5 3.07 8.51 1.4 6.85 − 1.64 4.80 .61 0.21 .11 0.65 .25 − 0.51
RAVLT int − 0.27 2.34 − 0.5 2.01 − 0.18 1.33 .80 0.11 .92 − 0.04 .67 0.19
RAVLT post int 0.73 2.28 1.1 1.60 − 0.45 2.62 .66 − 0.18 .23 0.49 .12 − 0.73
RAVLT delay 0.2 2.37 1.7 1.34 − 0.91 2.26 .08 − 0.74 .24 0.48 .01* − 1.43
RAVLT rec A 0 1.46 0.8 1.62 − 0.27 1.35 .21 − 0.52 .63 0.19 .11 − 0.72
RAVLT rec B 0.33 3.27 1.2 3.77 2.09 3.62 .55 − 0.25 .21 − 0.51 .59 0.24
TMT A − 0.25 4.37 − 7.4 7.75 − 2.30 5.26 .01* 1.21 .29 0.43 .09 0.77
TMT B − 9.11 29.08 2.374 22.59 5.9 23.87 .94 − 0.43 022 − 0.56 .73 0.15

Independent sample t tests or Wilcoxon rank sum tests were used based on whether or not the normality assumption was satisfied
*p < .05
d, Cohen’s d effect size; GE, gentle exercise; F-A-S, test of verbal fluency, int, interference; PCI, perceived cognitive impairment; PCA, perceived
cognitive abilities; post int, post interference; PROMISAbl, patient reported outcomes measurement information system, cognition abilities;
PROMISCog, patient reported outcomes measurement information system, cognition general concerns; Q, qigong; QOL, quality of life; RAVLT, rey
auditory verbal learning test; RAVLT 1–5, trials 1–5 total; Rec, recognition; SD, standard deviation; SS, survivorship support; TMT, trail making test

completion of data collection was a study design strategy to We were disappointed in the higher than expected attrition
minimize attrition (in the event participants were not ran- rate and poor adherence to home practice. These results con-
domized to their preferred group or wanted to try another trast with our earlier work. One co-investigator (WL) previ-
group at the end of their formal study participation). The ously explored the use of the six healing sounds of Qigong, in
chief barriers to overall study retention were described as two pilot studies. Fourteen participants with fibromyalgia
time commitment for 8 weekly sessions, driving long dis- were randomized to an 8-week Qigong intervention group
tances to the research center, and dealing with rush hour (also with home practice twice daily) versus gentle exercise
traffic to get to the sessions. A few participants expressed [21]. Adherence rates for the control group (75% home prac-
disappointment in the small group size. Several participants tice and 77% group attendance) and the Qigong intervention
(and potential recruits who decided not to participate) group (85% home practice and 79% group attendance) were
expressed interest in a virtual attendance option. The most higher than the current study. The Qigong group in the fibro-
common barriers to home practice documented in the logs myalgia study reported significant reduction in fatigue and
included: Btoo busy, too tired, forgot.^ pain and improvement in sleep quality (p < .05) compared to
 Support Care Cancer

the gentle exercise group. A subsequent small single-arm pilot breathing techniques and silently thinking the six healing
study was conducted for eight breast cancer survivors at least sounds (so as not to disturb her spouse) to help her go to sleep
3 months post completion of primary cancer treatment [24]. at night.
Participants received the 8-week Qigong intervention. Good
adherence was demonstrated (78.5% home practice and
89.6% group attendance). Significant reduction in fatigue Study limitations
and improvement for sleep quality, insomnia, and quality of
life was reported (p < .01). Study strengths included the three-arm randomized design
Attrition rates for the current study were not unlike other with attention control and sample homogeneity. The power
published works. Oh and colleagues reported results for a 10- of the study to investigate the exploratory aims was limited
week medical Qigong intervention versus standard of care for by small sample size and high attrition rates. Adherence may
162 participants [23]. Their dropout rate was 32% for the have been affected by small group sizes.
intervention group and 35% for controls. Mean session atten-
dance for those who completed the study was 8 of 10.
Adherence to keeping the home practice diary was only
50%. They subsequently published results of a substudy in Conclusions
which self-report of cognitive function was added as an end-
point [13]. The attrition for the 81 substudy participants was This three-arm pilot study involving eight weekly intervention
33%, and adherence to the home practice diary was reported sessions proved to be challenging to conduct. Initial recruit-
to be less than 50%. ment met the study objective. However, significant attrition
Results for the current study demonstrated significant im- occurred, particularly in the gentle exercise group. Despite the
provement in the FACT-Cog subscales for PCI and PCA and reduction in power caused by high attrition rates, the study
distress (MDASI) for participants in the Qigong group. Trends results suggest that mindfulness-based exercise may be supe-
for improvement (large effect sizes) were seen for improve- rior to gentle exercise alone or group support for improvement
ment in the PCI and PCA subscales for the gentle exercise in self-report of cognitive function and distress after treatment
group. The gentle exercise group reported the highest im- for breast cancer. The mindfulness component may enhance
provement in the FACT-Cog subscale for QOL. However, this the positive impact of exercise on cognitive function. Further
group reported lower QOL scores at baseline. Due to the sig- investigation that includes a virtual participation option (after
nificant attrition (and potential attrition bias) in the gentle ex- mastery of the movements/postures/sounds to the satisfaction
ercise group, the results must be interpreted cautiously. The of the group leader) to improve adherence to group attendance
potential exists for dilution of the effects of gentle exercise on may be warranted.
cognitive function due to the loss of power from attrition and
may bias the results for the improvement in QOL scores. Funding information This research was funded by the Oncology Nursing
Society Foundation through an unrestricted grant from the Sigma Theta
Only one secondary analysis of a randomized, controlled, Tau International Foundation. A portion of Dr. Myers’ time was support-
trial (12-week mindfulness-based exercise intervention, yoga) ed by the National Institution of Nursing Research (NINR) T32:
has reported results for objective cognitive function. Interdisciplinary Training of Nurse Scientists in Cancer Survivorship
Neuropsychological tests conducted with 100 breast cancer while a postdoctoral scholar with the University of Pittsburgh School of
Nursing.
survivors and 100 wait-list controls demonstrated a dose effect
for frequency of yoga practice at 3 months (p < .001) [10]. In
our study, improvement was seen for three of the neuropsy- Compliance with ethical standards
chological tests: Trail Making A, RAVLT delay, and F-A-S.
Approval from the University of Kansas Human Subjects Committee was
However, this improvement may have been due to practice obtained and all procedures were performed in accordance with institu-
effect as the neuropsychological testing was conducted ap- tional ethical standards and the 1964 Helsinki declaration and its later
proximately 12 weeks apart. amendments.
Overall, satisfaction surveys yielded excellent ratings for
Conflict of interest The authors have no financial relationship with the
study participation and group leadership. One cohort in par-
organization that funded the research (the Oncology Nursing Society
ticular wanted to have the support group continue. A few Foundation through an unrestricted grant from Sigma Theta Tau
participants in the exercise groups indicated they wished time International Foundation). The first author and the primary investigator
had been allowed for more interaction with other participants for the study have full control of all primary data. However, data sharing
agreement execution would be necessary if the journal requests data
and the opportunity to share experiences. Several participants
review.
indicated they intended to maintain Qigong as part of their
ongoing lifestyle. One participant described significant im- Informed consent Informed consent was obtained from all individual
provement in sleep disturbance. She described using the participants included in the study.
Support Care Cancer

References 17. Lindquist R, Wyman JF, Talley KM, Findorff MJ, Gross CR (2007)
Design of control-group conditions in clinical trials of behavioral
interventions. J Nurs Scholarsh 39:214–221
1. Carlson LE, Bultz BD (2003) Cancer, distress screening: needs,
18. Kelley GA, Kelley KS (2015) Meditative movement therapies and
models and methods. J Psychosom Res 55:403–409
health-related quality-of-life in adults: a systematic review of meta-
2. Hampson JP, Zick SM, Khabir T, Wright BD, Harris RE (2015)
analyses. PLoS One 10(6):e0129181. https://doi.org/10.1371/
Altered resting brain connectivity in persistent cancer related fa-
journal.pone.0129181
tigue. Neuroimage Clin 8:305–313
19. Ahles TA, Saykin A (2002) Breast cancer chemotherapy-related
3. Myers JS, Wick J, Klemp JR (2015) Potential factors associated
cognitive dysfunction. Clin Breast Cancer 3(Suppl 3):S84–S90
with perceived cognitive impairment in breast cancer survivors.
Support Care Cancer 23:3219–3228. https://doi.org/10.1007/ 20. Campo RA, Agarwal N, LaStayo PC, O'Connor K, Pappas L,
s00520-015-2708-7 Boucher KM, Gardner J, Smith S, Light KC, Kinney AY (2014)
4. Von AD, Jansen C, Allen DH (2014) Evidence-based interventions Levels of fatigue and distress in senior prostate cancer survivors
for cancer- and treatment-related cognitive impairment. Clin J enrolled in a 12-week randomized controlled trial of Qigong. J
Oncol Nurs 18:17–25 Cancer Surviv 8:60–69
5. Asher A, Myers JS (2015) The effect of cancer treatment on cog- 21. Liu W, Zahner L, Cornell M, Le T, Ratner J, Wang Y, Pasnoor M,
nitive function. Clin Adv Hematol Oncol 13(7):441–450 Dimachkie M, Barohn R (2012) Benefit of Qigong exercise in
6. Myers JS (2015) Complementary and integrative interventions for patients with fibromyalgia: a pilot study. Int J Neurosci 122(11):
cancer-related cognitive changes. Asia Pac J Oncol Nurs 657–664. https://doi.org/10.3109/00207454.2012.707713
7. Myers JS, Erickson KI, Sereika SM, Bender CM (2017) Exercise as 22. Oh B, Butow P, Mullan B, Clarke S (2008) Medical Qigong for
an intervention to mitigate decreased cognitive function from can- cancer patients: pilot study of impact on quality of life, side effects
cer and cancer treatment: an integrative review. Cancer Nurs:1. of treatment and inflammation. Am J Chin Med 36:459–472
https://doi.org/10.1097/ncc.0000000000000549 23. Oh B, Butow P, Mullan B, Clarke S, Beale P, Pavlakis N, Kothe E,
8. Baumann FT, Drosselmeyer N, Leskaroski A, Knicker A, Lam L, Rosenthal D (2010) Impact of medical Qigong on quality of
Krakowski-Roosen H, Zopf EM, Bloch W (2011) 12-week resis- life, fatigue, mood and inflammation in cancer patients: a random-
tance training with breast cancer patients during chemotherapy: ized controlled trial. Ann Oncol 21(3):608–614. https://doi.org/10.
effects on cognitive abilities. Breast Care 6:142–143 1093/annonc/mdp479
9. Crowgey T, Peters KB, Hornsby WE, Lane A, McSherry F, 24. Liu W, Schaffer L, Herrs N, Choller C, Taylor S (2015) Improved
Herndon JE, West MJ, Williams CL, Jones LW (2014) sleep after Qigong exercise in breast cancer survivors, a pilot study.
Relationship between exercise behavior, cardiorespiratory fitness, Asia Pac J Oncol Nurs submitted
and cognitive function in early breast cancer patients treated with 25. Sanford SD, Beaumont JL, Butt Z, Sweet J, Cella D, Wagner L
doxorubicin-containing chemotherapy: a pilot study. Appl Physiol (2014) Prospective longitudinal evaluation of a symptom cluster
Nutr Metab 39:724–729 in breast cancer. J Pain Symptom Manag 47:721–730
10. Derry HM, Jaremka LM, Bennett JM, Peng J, Andridge R, Shapiro 26. Moon S, Schmidt M, Smirnova IV, Colgrove Y, Liu W (2017)
CL, Malarkey WB, Emery CF, Layman R, Mrozek EE, Glaser R, Qigong exercise may reduce serum TNF-alpha levels and improve
Kiecolt-Glaser JK (2014) Yoga and self-reported cognitive prob- sleep in people with Parkinson’s disease: a pilot study. Medicines
lems in breast cancer survivors: a randomized controlled trial. (Basel, Switzerland) 4(2). https://doi.org/10.3390/
Psychooncology 24:958–966. https://doi.org/10.1002/pon.3707 medicines4020023
11. Knobf MT, Thompson A, Fennie K, Erdos D (2013) The effect of a 27. Cleeland CS, Mendoza TR, Wang XS, Chou C, Harle MT,
community-based exercise intervention on symptoms and quality of Morrissey M, Engstrom MC (2000) Assessing symptom distress
life. Cancer Nurs 37:E43–E50 in cancer patients: the M.D. Anderson symptom inventory. Cancer
12. Milbury K, Chaol A, Biegler KA, Wangyal T, Spelmen A, Meyers 89:1634–1646
C, Arun B, Palmer JL, Taylor JA, Cohen L (2013) Tibetan sound 28. Johnson-Kozlow M, Rock CL, Gilpin EA, Hollenbach KA, Pierce
meditation for cognitive dysfunction: results of a randomized con- JP (2007) Validation of the WHI brief physical activity question-
trolled pilot trial. Psychooncology 22:2354–2363 naire among women diagnosed with breast cancer. Am J Health
13. Oh B, Butow PN, Mullan BA, Clarke SJ, Beale PJ, Pavlakis N, Lee Behav 31:193–202
MS, Rosenthal DS, Larkey L, Vardy J (2012) Effect of medical 29. Lai J, Butt Z, Wagner L, Sweet J, Beaumont JL, Vardy J, Jacobsen
Qigong on cognitive function, quality of life, and a biomarker of PB, Jacobs SR, Shapiro CL, Cella D (2009) Evaluating the dimen-
inflammation in cancer patients: a randomized controlled trial. sionality of perceived cognitive function. J Pain Symptom Manag
Support Care Cancer 20:1235–1242. https://doi.org/10.1007/ 37:982–995
s00520-011-1209-6 30. Lai J, Wagner L, Jacobsen PB, Cella D (2014) Self-reported cogni-
14. Reid-Arndt AA, Matsuda S, Cox CR (2012) Tai chi effects on tive concerns and abilities: two sides of one coin? Psychooncology
neuropsychological, emotional, and physical functioning following 23:1133–1141. https://doi.org/10.1002/pon.3522
cancer treatment: a pilot study. Complement Ther Clin Pract 18:26– 31. Wefel J, Vardy J, Ahles TA, Schagen SB (2011) International cog-
30 nition and cancer task force recommendations to harmonise studies
15. Sprod LK, Mohile SG, Demark-Wahnefried W, Janelsins MC, of cognitive function in patients with cancer. Lancet Oncol 12:703–
Peppone LJ, Morrow GR, Lord R, Gross H, Mustian KM (2012) 708
Exercise and cancer treatment symptoms in 408 newly diagnosed 32. Luo L, Luk G, Bialystok E (2010) Effect of language proficiency
older cancer patients. J Geriatr Oncol 3:90–97 and executive control on verbal fluency performance in bilinguals.
16. Larkey LK, Roe DJ, Weihs KL, jahnke R, Lopez AM, Rogers CE, Cognition 114:29–41
OH B, Guillen-Rodriguez (2015) Randomized controlled trial of 33. Strauss E, Sherman EM, Spreen O (2006) A compendium of neu-
qigong/tai chi easy on cancer-related fatigue in breast cancer survi- ropsychological tests: Administration, norms, and commentary, vol
vors. Ann Behav Med 49:165–176 3rd edn. Oxford University Press, New York