Journal of Geriatric Oncology 10 (2019) 210–215

Contents lists available at ScienceDirect

Journal of Geriatric Oncology

Short communication

Oncologists' perceptions on the usefulness of geriatric assessment

measures and the CARG toxicity score when prescribing chemotherapy
for older patients with cancer
Erin B. Moth a,b,⁎, Belinda E. Kiely a,b, Natalie Stefanic b, Vasikaran Naganathan b,c, Andrew Martin b,
Peter Grimison d, Martin R. Stockler a,b, Philip Beale a,b, Prunella Blinman a,b
a
Concord Cancer Centre, Concord Repatriation General Hospital, Sydney, Australia
b
University of Sydney, Sydney, Australia
c
Centre for Education and Research on Ageing, Concord Repatriation General Hospital, Sydney, Australia
d
The Chris O'Brien Lifehouse, Sydney, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Background: The use of geriatric assessment (GA) and the Cancer and Aging Research Group (CARG) Toxicity
Received 28 August 2018 Score by Australian oncologists is low. We sought oncologists' views about the value of GA and the CARG Score
Received in revised form 5 November 2018 when making decisions about chemotherapy for their older patients.
Accepted 15 November 2018 Methods: Patients aged ≥65 yrs. with a plan to start chemotherapy for a solid organ cancer underwent a GA and
Available online 28 November 2018
had their CARG Score calculated. Results of the GA and CARG Score were provided to treating oncologists who
then completed a questionnaire on the value of these measures for each patient.
Keywords:
Decision-making
Results: We enrolled 30 patients from eight oncologists. Patients had a median age of 76 years and most (77%)
Chemotherapy toxicity were ECOG performance status 0 or 1. Risk category for severe chemotherapy toxicity by CARG Score was low
Older adult in 7 patients (23%), intermediate in 18 (60%), and high in 5 (17%). The GA provided oncologists new information
Elderly for 12 patients (40%), most frequently in the domains of function and nutrition. Knowledge of the GA prompted
supportive interventions for 7 patients (23%). Oncologists considered modifications to recommended chemo-
therapy based on the CARG Score for 2 patients (7%) (one more intensive and one less intensive), and based
on GA for no patients. Oncologists judged the GA and CARG Score as useful in 26 (87%) and 25 (83%) patients,
respectively.
Conclusion: Although oncologists valued the GA and CARG Score, they rarely used them to modify chemotherapy.
The GA provided new information that prompted supportive interventions in one quarter of patients.
© 2018 Elsevier Ltd. All rights reserved.

1. Introduction of severe chemotherapy toxicity. [7,8] Despite their potential benefits

Determining the suitability of an older adult with cancer for chemo-
therapy ideally involves geriatric assessment (GA) and the use of clinical
risk prediction tools, both now recommended in international guide-
lines. [1,2] GA identifies co-existent geriatric problems that can affect
tolerance of anti-cancer therapies and prompt supportive interventions,
[1,3] may alter treatment decisions, [4,5] and improve treatment toler-
ance and completion. [5,6] The Cancer and Aging Research Group's
(CARG) Toxicity Score [7,8] is a clinical risk prediction tool that may
aid decision-making about chemotherapy by estimating the likelihood
of severe chemotherapy toxicity in older adults. Eleven clinical and GA
variables are used to classify patients as low, intermediate, or high risk
⁎ Corresponding author at: Concord Cancer Centre, Building 76, Concord Repatriation
General Hospital, Hospital Rd, Concord, NSW 2139, Australia.
E-mail address: Erin.Moth@health.nsw.gov.au (E.B. Moth).
to patient care, use of the GA and CARG Score by Australian oncologists
is low. [9,10]
Barriers to the implementation of clinical tools are multifaceted.
Accessibility, time burden, and resource availability were cited as the
most frequent barriers to the use of a GA in a recent cross-sectional sur-
vey of 69 Australian oncologists, [10] with most of these oncologists
agreeing that a GA would add to their clinical assessment (71%) and
would influence their clinical decision-making (65%). The CARG Score
has been externally validated [7] and tested in a small number of exter-
nal cohorts, [11–14] but there are no published studies on barriers to its
use or its perceived value to oncologists and its potential to influence
decision-making.
We performed a prospective observational study evaluating the
CARG Score and comparing it to oncologists' clinical judgement in
predicting for severe chemotherapy toxicity. [12] This parent study pro-
vided the opportunity to prospectively determine the value of the GA

https://doi.org/10.1016/j.jgo.2018.11.004
1879-4068/© 2018 Elsevier Ltd. All rights reserved.
 E.B. Moth et al. / Journal of Geriatric Oncology 10 (2019) 210–215 211

and CARG Score to oncologists prescribing chemotherapy for older standard treatment here were considered to have been made based
adults. Specific objectives of this substudy were to determine on usual clinical judgement. Oncologists were then asked how their rec-
(i) oncologists' views on the usefulness of the GA and CARG Score, (ii) ommendation would have been modified based on the (i) GA and (ii)
new information gained from the GA, and (iii) the potential impact of CARG Score had they known these results prior to making a treatment
the GA and CARG Score on chemotherapy prescribing and patient decision. A post-hoc brief written survey asked oncologists to comment
management. on the barriers to implementation of the GA and CARG Score in clinical
practice.
2. Methods
2.4. Analysis
2.1. Design and Participants
Questionnaire responses were described using frequencies and pro-
Participants of this substudy were the oncologists of a subset of older portions (%). Proportions of responses within answer categories using
adults with cancer participating in the parent study described and refer- Likert scales were presented using stacked bar charts. A post-hoc analy-
enced above. [12] Eligibility criteria for the parent study included sis explored the relationship between a patient's CARG Score and their
age ≥ 65 years, diagnosis of a solid organ malignancy (any type or oncologist-rated CSHA Clinical Frailty Rating using a 2 × 3 contingency
stage), and plan to start an initial or new line of systemic cytotoxic table and Fisher's Exact Test. We aimed to enrol 50 patients to provide
chemotherapy. 95% confidence intervals of estimated proportions of within +/− 15%.
Ethics approval for this study was provided by the Sydney Local
Health District Human Research Ethics Committee of Concord Repatria- 3. Results
tion General Hospital (HREC/15/CRGH/102) and the study was open at
two cancer centres in Sydney, Australia. Between December 2016 and March 2017, eight oncologists pro-
vided survey responses for 30 patients with a response rate of 100%.
2.2. Procedures
Table 1
Oncologists determined plans for chemotherapy for each patient as Patient (n = 30) characteristics.
per usual clinical practice. A trained study researcher (NS) or clinician
Characteristic Number (%)
researcher (EM) not involved in clinical care then completed a GA and
calculated the CARG Score for each patient. The GA was performed in Sex Male 19 (63)
the outpatient clinic at an agreed time to minimise additional visits, Female 11 (37)
Cancer centre Concord Cancer Centre 14 (47)
and took b30 min. Geriatric health domains assessed were: functional The Chris O'Brien Lifehouse 16 (53)
status by the Timed Up and Go, [15,16] Katz Index of Activities of Age 65 to 69 years 8 (27)
Daily Living, [17,18] OARS Multidimensional Functional Assessment In- 70 to 74 years 4 (13)
strumental Activities of Daily Living, [19] the Medical Outcomes Study 75 to 79 years 12 (40)
≥80 years 6 (20)
(MOS) Physical Functioning Scale, [20] and a self-reported history of
Median (years) 75.5 years
falls; comorbidity by the Cumulative Illness Rating Scale in Geriatrics Employment status Retired or not working 28 (93)
(CIRS-G); [21] cognition by the Short Blessed (Orientation Memory Working 2 (7)
Concentration) Test; [22] psychological health by the Geriatric Depres- Marital status Married/Common law (De-facto) 18 (60)
sion Scale 5-Item Short Form (GDS-5); [23] social supports by the mod- Widowed 1 (3)
Divorced/separated 7 (23)
ified MOS Social Support Survey; [24] and nutrition by unintentional Single 4 (13)
weight loss and the Mini Nutritional Assessment Short Form (MNA- Living arrangements Lives with others 7 (23)
SF). [25] Lives alone 22 (73)
For the first 126 patients, oncologists were blinded to the results of Care facility 1 (3)
Language spoken at home English 22 (73)
the GA and CARG Score and independently estimated the risk of severe
Non-English 8 (27)
chemotherapy toxicity for each of their participating patients. Results of Receiving community services Yes 5 (17)
this first part have been published. [12] For the final 30 patients re- No 25 (83)
ported here, treating oncologists were provided with results of the GA Cancer type Colorectal 14 (47)
and CARG Score as a pro forma written report (Supplementary 1) and Ovarian 4 (13)
Upper gastrointestinal⁎ 3 (10)
then invited to complete a study-specific questionnaire (Supplemen- Lung/pleura 3 (10)
tary 2). For all 156 patients, a chemotherapy treatment recommenda- Prostate 3 (10)
tion had been made prior to the GA and CARG Score being performed. Bladder 1 (3)
The GA and CARG Score results were presented to oncologists for this Other 2 (7)
Stage of cancer I 0 (0)
substudy prior to the commencement of chemotherapy or shortly
II 1 (3)
after starting chemotherapy, as it was apparent from Part 1 that GAs III 12 (40)
were most feasibly performed on day 1 of treatment. [12] IV 17 (57)
Line of treatment Neoadjuvant 2 (7)
2.3. Oncologist Questionnaire Adjuvant 8 (27)
1st line palliative 13 (43)
Subsequent line palliative 7 (23)
The substudy questionnaire addressed themes of (i) new informa- Chemotherapy regimen Single agent 13 (43)
tion gained from the GA; (ii) the impact of the GA and CARG Score on Combination chemotherapy 17 (57)
chemotherapy recommendations; (iii) GA-prompted interventions; Primary G-CSF Yes 0 (0)
No 30 (100)
and (iv) the usefulness and ease of interpretation of the GA and CARG
Initial dose plan for cycle 1⁎⁎ Dose reduced 10 (33)
Score (Supplementary 2). Of note, the impact of the GA and CARG Standard dose 20 (67)
Score on chemotherapy recommendations was evaluated retrospec-
⁎ Upper gastrointestinal includes pancreaticobiliary, gastric, and oesophageal cancers.
tively. Oncologists were first asked how their chemotherapy recom- ⁎⁎ Initial dose plan for cycle 1 was defined as per the item of the Cancer and Aging Re-
mendation compared to standard treatment for a younger, fitter search Group's (CARG) Toxicity Score, as standard or reduced dose for that regimen ac-
patient with the same type and stage of cancer. Deviations from cording to NCCN guidelines.
212 E.B. Moth et al. / Journal of Geriatric Oncology 10 (2019) 210–215

The sample size was smaller than planned due to resource availability
and presentation of results for Part 1, in which the CARG Score did not
predict severe chemotherapy toxicity in our local population. [12]
Given that this may have influenced oncologists' responses to the ques-
tionnaire for the substudy, recruitment was stopped at 30 patients.
Of the eight oncologists, the median years in practice was 14 (range
7–25 years), most (six of eight) worked mainly in the public setting, and
nearly all (seven of eight) never used GA tools in their routine practice.
Most (five of eight) reported patients ≥70 years comprising between
50% and 75% of their practice.
Table 1 outlines patient characteristics. Table 2 outlines GA results.
Oncologists rated most patients as fit or well (21, 70%), and a minority
as vulnerable or frail (9, 30%). Seven patients (23%) were classified as
low, 18 (60%) as intermediate, and 5 (17%) as high risk of severe chemo-
therapy toxicity by CARG Score. The relationship between patients' CARG
Score Risk Group and CSHA Clinical Frailty Scale is explored in Supple-
mentary 3, with these measures being independent (p-value for Fisher's
exact test of 0.46). Treatment plans compared to standard treatment for a
younger, fitter patient with the same type and stage of cancer were: ‘no
different’ for 17 patients (57%); ‘same regimen at reduced dose’ for 4 pa-
tients (13%); ‘less intensive regimen at standard dose’ for 5 patients
(17%); and ‘less intensive regimen at reduced dose’ for 4 patients (13%).
The median time from clinic consultation to: (i) GA was 5.5 days
(range 0–23 days); and (ii) start of chemotherapy was 6 days (range

Table 2
Baseline geriatric assessment results (N = 30).

Characteristic Category N (%) Median Range Range of scores

Self-rated health Excellent or very good 13 (43)

Good, fair or poor 17 (57)
CSHA Clinical Frailty Rating [30] Fit, or well 21 (70)
Vulnerable or frail 9 (30)
Performance status
ECOG Performance Status [31] 0 or 1 23 (77)
Karnofsky Performance Rating Scale [32] ≥2 7 (23)
90–100 11 (37)
80 12 (40)
≤70 7 (23)
Functional status Independent (score 6) 29 (97) 6 5–6 0–6
Katz Activities of Daily Living [18]
OARS Instrumental ADLs [19] Dependent ≥1 task 1 (3)
MOS Physical Functioning [20] Independent (score 14) 19 (63) 14 4–14 0–14
Timed Up and Go [15] Dependent ≥1 task 11 (37)
No limitation 1 (3) 26 15–30 10–30
Falls in last 6 months Some limitation 29 (97)
≥14 s 4 (13) 11.7 s 7.3–21.5
b14 s 21 (70)
Yes 6 (20)
Comorbidities
CIRS-G Total Score [21] 2 (7) 3.5 0–10
CIRS-G Index 2 1–3
Number with category 3 (severe) 2 (7)
Number with category 4 (life threatening)
Polypharmacy 2 0–11
Number of medications
Social Supports Complete social supports 19 (63) 20 4–20 0–20
Social Support Survey [24] 11 (37)
Some deficit in support
Mood and Cognition Score 0 or 1 27 0 0–3 0–5
5-Item Geriatric Depression Scale [23] Score ≥ 2 (abnormal) (90) 2 0–14 0–30
Orientation-Memory-Concentration Test [22] Score 0 to 4 (normal) 3 (10)
Score ≥ 5 25 (83)
5 (17)
Nutrition
Weight loss in last 6 months Yes 22
Mini-Nutritional Assessment Short Form [25] Normal nutrition (score 12 to 14) (73) 11 6–14 0–14
At risk (score 8 to 11) 13 (43)
Malnourished (score 0 to 7) 16 (53)
1 (3)
G8 [33] N14 11 (37)
≤14 (at risk) 19 (63)
Geriatric assessment score* 0 or 1 15 (50)
2 or 3 14 (47)
≥4 1 (3)
CARG Toxicity Score [8] Low risk 7 (23) 8 4–12 0–23
Intermediate risk 18 (60)
High risk 5 (17)

Abbreviations: CSHA- Canadian Study of Health and Aging; ECOG- Eastern Cooperative Oncology Group; OARS- Older Americans Resources and Services; ADL- Activities of Daily Living;
MOS- Medical Outcomes Study; CIRS-G- Cumulative Illness Rating Scale in Geriatrics; G8- Geriatric 8; CARG Toxicity Score- Cancer and Aging Research Group's Toxicity Score.
*Geriatric assessment score (range 0 to 7) is a summary score for the geriatric assessment performed, where a point is scored for a deficit in a geriatric health domain as follows:
- performance status ECOG 2 or more
- functional status: TUG N/= 14 s, any dependency in ADLs
- nutrition: MNA at risk or malnourished
- cognition: at risk or likely consistent with dementia
- social supports: b18 (lowest quartile)
- psychological state: GDS N/= 2
- comorbidity: CIRS G score N 6 (highest quartile)
 E.B. Moth et al. / Journal of Geriatric Oncology 10 (2019) 210–215 213

Did the geriatric assessment provide you with any new

information about your patient?
60 40

Was the information contained in the geriatric assessment

consistent with your clinical impression?
20 80

Would you have modified your existing chemotherapy

recommendation on the basis of any of the information 100
gained from the geriatric assessment?
Would you have modified your existing chemotherapy
recommendation on the basis of the patient's CARG 93 7
Toxicity Score?

0% 50% 100%
No Yes % of patients

Fig. 1. Perceived clinical value and impact on chemotherapy prescribing of the Cancer and Aging Research Group's (CARG) Score and Geriatric Assessment. For the 30 enrolled patients,
their treating oncologist was asked to complete a questionnaire regarding the GA and CARG Score in that patient. The proportion of responses in each answer category reflect the
proportion of patients for whom their treating oncologist answered ‘yes’ or ‘no’ to the presented questions.

0–23 days). The median time from GA to the start date for chemother-
apy was 0 days (range 0–7 days), with most performed on day 1 of
treatment. Though not mandated in the study design, for 11 patients
the results of the GA and CARG Score were presented to their treating
oncologist prior to the start of planned chemotherapy.
The GA was consistent with oncologists' ‘overall clinical impression’
for most patients (24, 80%). The GA provided oncologists with new in-
formation for 12 patients (40%) as follows: functional status (n = 6),
cognition (n = 3), psychological health (n = 3), polypharmacy (n =
3), comorbidity (n = 2), nutrition (n = 6), and social supports (n =
2). The GA triggered interventions not otherwise considered for seven
patients (some with ≥1 intervention): social work (one patient); dieti-
tian (four patients); psychologist or psychiatric service (one patient);
medication review (one patient), and community services (one
patient).
Oncologists reported that they would have modified their chemo-
therapy recommendation based on the GA for none of the 30 patients,
and based on the CARG Score for 2 (7%) patients; one patient with a
CARG Score of 4 (low risk) would have been changed to a more inten-
sive regimen and a patient with a score of 12 (high risk) would have
been changed to a less intensive regimen. (Fig. 1) Oncologists thought
the GA was useful for most patients (26, 87%) and easy to interpret
(29, 97%), and the CARG Score was useful for most patients (25, 83%)
and easy to interpret (30, 100%). (Fig. 2) Perceived barriers to the imple-
mentation of GA and CARG Score are in Table 3, with recurring themes
being time and uncertainty about contribution to decision-making and
usual practice.
4. Discussion
Key findings of this study were that oncologists found the results of a
GA and CARG Score useful for most patients but were unlikely to use
them to make changes to recommendations about chemotherapy. Pa-
tients' performances on a GA were mostly consistent with the clinical
impression of their treating oncologist, but for some the GA provided
new information that prompted supportive interventions (in 23%).
The GA provided new information and prompted non-oncological
supportive interventions for one in four patients, lower than other stud-
ies. In a recent systematic review by of 19 studies, Hamaker et al. [5]
identified at least one non-oncological intervention occurred for a me-
dian of 72% of patients (range 26 to 100%) undergoing GA in the oncol-
ogy setting. Some studies reported any intervention following a GA, [26]
whereas others required the intervention would not have happened as
part of usual care (as in our study), [27] in part explaining the wide

I found the results of the CARG Score easy to

interpret
67 33

I found the results of the CARG Score useful 10 7 77 7

I found the results of the geriatric assessment

easy to interpret
3 60 37

I found the results of the geriatric assessment

useful
7 7 80 7

% of patients

Strongly disagree Disagree Neutral Agree Strongly Agree

Fig. 2. Oncologist ratings of ease of use of the Cancer and Aging Research Group's (CARG) Score and Geriatric Assessment. For the 30 enrolled patients, their treating oncologist was asked to
complete a questionnaire regarding the GA and CARG Score in that patient. The proportion of responses in each answer category reflect the proportion of patients for whom their treating
oncologist agreed, on a 5-point Likert scale, with the statements presented.
214 E.B. Moth et al. / Journal of Geriatric Oncology 10 (2019) 210–215
Table 3
Comments from oncologists on barriers to use of the GA and CARG Score.
Comments
What do you see as the main barriers “Time limitation in a busy clinic, lack of
to the implementation of a GA in support staff and space to conduct the
clinical practice? assessment”
“Time”
“Ease of access and time”
“Limited clinic time”
“Expertise and time”
“Time, uncertain how formal assessment
will change decision-making”
“Time in a busy practice”
“Uncertainty about its value/benefit; and
time”
What do you see as the main barriers “Time, practical aspects for example, it
to the implementation of the CARG would need to be added to the electronic
Score in clinical practice? medical record for ease of completion
and storage; training on how to complete
it; not convinced it improves on current
practice”
“Time”
“Time, space, and staff”
“Limited clinic time”
“Expertise and time”
“Time, and uncertain if it adds to clinical
assessment”
“Time in a busy practice”
“Uncertainty about its value; and time”
Abbreviations: GA = Geriatric Assessment; CARG = Cancer and Aging Research Group.
range of results. Oncologists in our study reported that the GA was
consistent with their overall clinical impression for most patients. This
implies that, whilst not having formally assessed geriatric domains,
oncologists had formed a clinical impression of these and were not sur-
prised by their patient's performance on formal GA measures, and may
explain the comparatively lower rate of interventions prompted by the
GA in our study.
Oncologists would not use the results of the GA to modify their
chemotherapy recommendation in our study. Possible reasons for this
include choice-supportive bias (a reluctance to report a different deci-
sion might have been present once a decision has already been made),
lack of experience with GA, inconsistent evidence regarding the compo-
nents of the GA that best predict treatment toxicity, [3,28] fear about
under-treatment, and again the reported consistency of the GA with on-
cologists' overall clinical impression. Using similar methodology,
Decoster et al. [29] found cancer treatment plans for older adults
(n = 902) were modified from standard therapy based on clinical
judgement for 44% (43% in our study), with further changes based on
GA in only an additional 6%. The recent review by Hamaker et al. [5]
found higher rates of change in treatment decisions based on GA. Across
11 studies that reported treatment choice before and after GA, a change
in oncological management (not only chemotherapy) occurred for a
median of 28% of patients (range 8–54%), the majority to less intensive
treatment. Methodologic differences between the studies included in
this review, particularly with regard to the baseline treatment plan
used as the comparator, should be noted. For example, if the baseline
treatment plan was nominated prior to a cancer specialist seeing the pa-
tient, any modifications made based on clinical judgement that was sep-
arate to the effect of the GA might be missed, overestimating the impact
of the GA on treatment decisions.
To our knowledge, oncologists have not previously been asked to re-
port on the impact of the CARG Score on their chemotherapy recom-
mendation and prescribing. We have previously shown that expected
rates of chemotherapy toxicity influence chemotherapy recommenda-
tions. [9] As a chemotherapy toxicity risk score, it would be anticipated
that the CARG Score would have a similar influence. Three studies
[11,12,14] have shown a lack of correlation between oncologists' esti-
mates of the likelihood of chemotherapy toxicity and a patient's CARG
Score, raising potential for the score to be providing information differ-
ent to clinical judgement. Oncologists in our study were unlikely to use
the CARG Score to guide chemotherapy treatment decisions. One reason
for this is the observation that for 13 patients (43%) the chemotherapy
treatment plan had already been modified based on clinical assessment.
Other potential reasons proposed by the authors include lack of famil-
iarity with the score, challenges translating the score into modifications
to chemotherapy, and uncertainty about its local applicability. Whilst
our larger prospective study [12] did not demonstrate predictive value
of the CARG score in our population, oncologists in this substudy were
not aware of this when completing the questionnaires.
Nishijima et al. [14] determined the value of the CARG Score in
decision-making about chemotherapy by assessing the agreement be-
tween treatment decision (reduced or standard intensity chemother-
apy) based on clinical impression and based on the CARG Score in 58
older adults. An assumption of this study was that patients with a
CARG Score ≥ 10 (high-risk) should be recommended reduced intensity
chemotherapy. Patients who received standard intensity chemotherapy
(based on oncologist impression) yet had a CARG Score ≥ 10 had higher
rates of severe toxicity (88% v 40%, p = .006), suggesting the addition of
the CARG Score to clinical judgement may improve treatment decision-
making, at least for high-risk patients. Whether oncologists would
modify their treatment recommendations for these high-risk patients,
considering there may be competing benefits of proceeding with stan-
dard intensity treatment in some settings, is a question our study sought
in part to explore. For only one of the five patients in our study with
a CARG Score of ≥10 would their oncologist have changed their
recommendation to a less intensive chemotherapy regimen. This is a
recognised area for further enquiry.
Strengths of this study include providing novel local data on the
value and use of the GA and CARG Score and being the first study to
seek to evaluate oncologist reported impact of the CARG Score on che-
motherapy recommendations. Limitations include the small number
of oncologists (n = 8) and centres (n = 2), thus results are unlikely to
reflect the views of all Australian oncologists practicing in varied geo-
graphic settings and practice types. The oncologists involved did not
routinely use GAs or screening and so their views may be biased against
their use. Choice-supportive bias is possible due to the design of the
study, as oncologists responded to the questionnaire after they had
made a recommendation about chemotherapy. The study focussed
on the GA and CARG Score with respect to treatment decision-
making but did not evaluate other potential benefits or uses of these
measures.
5. Conclusion
Oncologists found the results of a GA and CARG Score useful for their
older patients commencing chemotherapy. The GA was consistent with
oncologists' clinical impressions for most patients and provided new in-
formation that prompted supportive interventions for 1-in-4 patients.
Oncologists were unlikely to modify their chemotherapy recommenda-
tions based on the GA or CARG Score, and as such their potential to
impact decision-making about chemotherapy prescribing in this setting
was low. Barriers to the use of such tools in routine practice, and their
recommended role in guiding chemotherapy treatment decision-
making and prescribing, need to be addressed to facilitate implementa-
tion of these tools into routine clinical practice.
Ethics Approval and Consent to Participate
Ethics approval was granted by the Sydney Local Health District
Human Research Ethics Committee of Concord Repatriation General
Hospital (HREC/15/CRGH/102). Signed informed consent was obtained
from all participants.
 E.B. Moth et al. / Journal of Geriatric Oncology 10 (2019) 210–215
Conflict of Interest
A/Prof Beale receives compensation as a consultant for Merck Sharp
and Dohme (Aust) P/L, AstraZenica, and Roche Products P/L. The re-
maining authors declare no conflict of interest.
Acknowledgement
This project was funded by a Sydney Local Health District Cancer
Services Research Grant (CIA Dr. EM, other investigators: BK, PLB, PG,
VN). Dr. EM was supported in this work by two PhD scholarships: a
University of Sydney Australian Postgraduate Award (APA), and PhD
funding support from Sydney Catalyst: the Translational Cancer
Research Centre of Central Sydney and regional NSW, University of Syd-
ney, NSW, Australia and Cancer Institute NSW.
Authors' Contributions
Conception and design: EM, PB, BK, AM, and VN conceived the work. All
authors were responsible for the subsequent design of the work.
Data collection: EM and NS were responsible for data acquisition,
Analysis and interpretation of data: EM and AM were responsible for
data analysis and initial interpretation of results. All authors were re-
sponsible for final interpretation of results.
Manuscript writing: EM prepared the first draft of the manuscript. All
authors revised the manuscript.
Approval of final manuscript: All authors approved the final version of
the manuscript. All authors agreed to be accountable for all aspects of
the work.
References
[1] Wildiers H, Heeren P, Puts M, Topinkova E, Janssen-Heijnen ML, Extermann M, et al.
International Society of Geriatric Oncology consensus on geriatric assessment in
older patients with cancer. J Clin Oncol 2014;32(24):2595–603.
[2] Mohile SG, Dale W, Somerfield MR, Schonberg MA, Boyd CM, Burhenn PS, et al. Prac-
tical assessment and management of vulnerabilities in older patients receiving che-
motherapy: ASCO guideline for geriatric oncology. J Clin Oncol 2018;36(22):
2326–47 (JCO2018788687).
[3] Versteeg KS, Konings IR, Lagaay AM, van de Loosdrecht AA, Verheul HM. Prediction
of treatment-related toxicity and outcome with geriatric assessment in elderly pa-
tients with solid malignancies treated with chemotherapy: a systematic review.
Ann Oncol 2014;25(10):1914–8.
[4] Hamaker ME, Schiphorst AH, ten Bokkel Huinink D, Schaar C, van Munster BC. The
effect of a geriatric evaluation on treatment decisions for older cancer patients—a
systematic review. Acta Oncol 2014;53(3):289–96.
[5] Hamaker ME, Te Molder M, Thielen N, van Munster BC, Schiphorst AH, van Huis LH.
The effect of a geriatric evaluation on treatment decisions and outcome for older
cancer patients - A systematic review. J Geriatr Oncol 2018;9(5):430–40.
[6] Kalsi T, Babic-Illman G, Ross PJ, Maisey NR, Hughes S, Fields P, et al. The impact of
comprehensive geriatric assessment interventions on tolerance to chemotherapy
in older people. Br J Cancer 2015;112(9):1435–44.
[7] Hurria A, Mohile S, Gajra A, Klepin H, Muss H, Chapman A, et al. Validation of a pre-
diction tool for chemotherapy toxicity in older adults with cancer. J Clin Oncol 2016;
34(20):2366–71.
[8] Hurria A, Togawa K, Mohile SG, Owusu C, Klepin HD, Gross CP, et al. Predicting che-
motherapy toxicity in older adults with cancer: a prospective multicenter study.
J Clin Oncol 2011;29(25):3457–65.
[9] Moth EB, Kiely BE, Naganathan V, Martin A, Blinman P. How do oncologists make de-
cisions about chemotherapy for their older patients with cancer? A survey of
Australian oncologists. Support Care Cancer 2017;26(2):451–60.
215
[10] To THM, Soo WK, Lane H, Khattak A, Steer C, Devitt B, et al. Utilisation of geriatric as-
sessment in oncology - a survey of Australian medical oncologists. J Geriatr Oncol
2018. https://doi.org/10.1016/j.jgo.2018.07.004 [Epub ahead of print].
[11] Alibhai SM, Aziz S, Manokumar T, Timilshina N, Breunis H. A comparison of the
CARG tool, the VES-13, and oncologist judgment in predicting grade 3+ toxicities
in men undergoing chemotherapy for metastatic prostate cancer. J Geriatr Oncol
2017;8(1):31–6.
[12] Moth EB, Kiely BE, Stefanic N, Naganathan V, Martin A, Grimison P, et al. Predicting
chemotherapy toxicity in older adults: Comparing the predictive value of the CARG
Toxicity Score with oncologists' estimates of toxicity based on clinical judgement. J
Geriatr Oncol 2018 Sep 14. https://doi.org/10.1016/j.jgo.2018.08.010 [Epub ahead
of print].
[13] Nie X, Liu D, Li Q, Bai C. Predicting chemotherapy toxicity in older adults with lung
cancer. J Geriatr Oncol 2013;4(4):334–9.
[14] Nishijima TF, Deal AM, Williams GR, Sanoff HK, Nyrop KA, Muss HB. Chemotherapy
toxicity risk score for treatment decisions in older adults with advanced solid tu-
mors. Oncologist 2018;23(5):573–9.
[15] Podsiadlo D, Richardson S. The timed "Up & Go": a test of basic functional mobility
for frail elderly persons. J Am Geriatr Soc 1991;39(2):142–8.
[16] Rose DJ, Jones CJ, Lucchese N. Predicting the probability of falls in community-
residing older adults using the 8-foot up-and-go: a new measure of functional mo-
bility. JAPA 2002;10(4):466–75.
[17] Katz S, Akpom CA. A measure of primary sociobiological functions. Int J Health Serv
1976;6(3):493–508.
[18] Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illness in the aged.
The index of adl: a standardized measure of biological and psychosocial function.
JAMA 1963;185(12):914–9.
[19] Fillenbaum GG, Smyer MA. The development, validity, and reliability of the OARS
multidimensional functional assessment questionnaire. J Gerontol 1981;36(4):
428–34.
[20] Stewart AL, Kamberg CJ. Physical functioning measures. In: Stewart AL, Ware JE, ed-
itors. Measuring functioning and well-being: The Medical Outcomes Study ap-
proach. Durham, North Carolina: Duke University Press; 1992. p. 86–101.
[21] Miller MD, Paradis CF, Houck PR, Mazumdar S, Stack JA, Rifai AH, et al. Rating chronic
medical illness burden in geropsychiatric practice and research: application of the
Cumulative Illness Rating Scale. Psychiatry Res 1992;41(3):237–48.
[22] Katzman R, Brown T, Fuld P, Peck A, Schechter R, Schimmel H. Validation of a short
Orientation-Memory-Concentration Test of cognitive impairment. Am J Psychiatry
1983;140(6):734–9.
[23] Hoyl MT, Alessi CA, Harker JO, Josephson KR, Pietruszka FM, Koelfgen M, et al. Devel-
opment and testing of a five-item version of the Geriatric Depression Scale. J Am
Geriatr Soc 1999;47(7):873–8.
[24] Sherbourne CD, Stewart AL. The MOS social support survey. Soc Sci Med 1991;32(6):
705–14.
[25] Rubenstein LZ, Harker JO, Salvà A, Guigoz Y, Vellas B. Screening for undernutrition in
geriatric practice: developing the short-form mini-nutritional assessment (MNA-
SF). J Gerontol A Biol Sci Med Sci 2001;56(6):M366–72.
[26] Caillet P, Canoui-Poitrine F, Vouriot J, Berle M, Reinald N, Krypciak S, et al. Compre-
hensive geriatric assessment in the decision-making process in elderly patients with
cancer: ELCAPA study. J Clin Oncol 2011;29(27):3636–42.
[27] Kenis C, Bron D, Libert Y, Decoster L, Van Puyvelde K, Scalliet P, et al. Relevance of a
systematic geriatric screening and assessment in older patients with cancer: results
of a prospective multicentric study. Ann Oncol 2013;24(5):1306–12.
[28] Puts MT, Santos B, Hardt J, Monette J, Girre V, Atenafu EG, et al. An update on a sys-
tematic review of the use of geriatric assessment for older adults in oncology. Ann
Oncol 2014;25(2):307–15.
[29] Decoster L, Kenis C, Van Puyvelde K, Flamaing J, Conings G, De Grève J, et al. The in-
fluence of clinical assessment (including age) and geriatric assessment on treatment
decisions in older patients with cancer. J Geriatr Oncol 2013;4(3):235–41.
[30] Rockwood K, Song X, MacKnight C, Bergman H, Hogan DB, McDowell I, et al. A global
clinical measure of fitness and frailty in elderly people. CMAJ 2005;173(5):489–95.
[31] Oken MM, Creech RH, Tormey DC, Horton J, Davis TE, McFadden ET, et al. Toxicity
and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol
1982;5(6):649–55.
[32] Karnofsky DA, Burchenal JH. The clinical evaluation of chemotherapeutic agents in
cancer. Evaluation of Chemotherapeutic Agents. New York: Columbia University
Press; 1949; 191–205.
[33] Soubeyran PL, Bellera CA, Goyard J, Heitz D, Cure H, Rousselot H. Validation of the G8
screening tool in geriatric oncology: the ONCODAGE project. J Clin Oncol 2011;29
(suppl; abstr 9001).