Open Access

The SAKK cancer-specific geriatric assessment (C-SGA): a pilot study of a brieftool for clinical decision-making in older cancer patients

  • Kerri M Clough-Gorr1, 2Email author,
  • Lea Noti3,
  • Peter Brauchli4,
  • Richard Cathomas5,
  • Marius R Fried6, 7,
  • Gillian Roberts5,
  • Andreas E Stuck3,
  • Felicitas Hitz6 and
  • Ulrich Mey5
Contributed equally
BMC Medical Informatics and Decision MakingBMC series – open, inclusive and trusted201313:93

DOI: 10.1186/1472-6947-13-93

Received: 10 April 2013

Accepted: 14 August 2013

Published: 23 August 2013

Abstract

Background

Recommendations from international task forces on geriatric assessmentemphasize the need for research including validation of cancer-specificgeriatric assessment (C-SGA) tools in oncological settings. This study wasto evaluate the feasibility of the SAKK Cancer-Specific Geriatric Assessment(C-SGA) in clinical practice.

Methods

A cross sectional study of cancer patients 65 years old(N = 51) with pathologically confirmed cancer presenting forinitiation of chemotherapy treatment (07/01/2009-03/31/2011) at two oncologydepartments in Swiss canton hospitals: Kantonsspital Graubünden (KSGRN = 25), Kantonsspital St. Gallen (KSSG N = 26).Data was collected using three instruments, the SAKK C-SGA plus physicianand patient evaluation forms. The SAKK C-SGA includes six measures coveringfive geriatric assessment domains (comorbidity, function, psychosocial,nutrition, cognition) using a mix of medical record abstraction (MRA) andpatient interview. Five individual domains and one overall SAKK C-SGA scorewere calculated and dichotomized as below/above literature-based cut-offs.The SAKK C-SGA was evaluated by: patient and physician estimated time tocomplete, ease of completing, and difficult or unanswered questions.

Results

Time to complete the patient questionnaire was considered acceptable byalmost all (≥96%) patients and physicians. Patients reported slightlyshorter times to complete the questionnaire than physicians(17.33 ± 7.34 vs.20.59 ± 6.53 minutes, p = 0.02). Bothgroups rated the patient questionnaire as easy/fairly easy to complete (91%vs. 84% respectively, p = 0.14) with few difficultor unanswered questions. The MRA took on average8.32 ± 4.72 minutes to complete. Physicians (100%)considered time to complete MRA acceptable, 96% rated it as easy/fairly easyto complete. Individual study site populations differed on health-relatedcharacteristics (excellent/good physician-rated general health KSGR 71%vs. KSSG 32%, p = 0.007). The overall mean C-SGAscore was 2.4 ± 1.12. Patients at KSGR had lower C-SGAscores (2.00 ± 1.19 vs.2.81 ± 0.90, p = 0.009) and a smallerproportion (28% vs. 65%, p = 0.008) was above the C-SGAcut-off score compared to KSSG.

Conclusions

These results suggest the SAKK C-SGA is a feasible practical tool for use inclinical practice. It demonstrated discriminative ability based on objectivegeriatric assessment measures, but additional investigations on use forclinical decision-making are warranted. The SAKK C-SGA also providesimportant usable domain information for intervention to optimize outcomes inolder cancer patients.

Keywords

Assessment Cancer-specific geriatric assessment Decision-making Geriatric assessment Older cancer patients Older adults

Background

Cancer is considered an age-related disease of increasing public health concern dueto worldwide aging populations. The World Health Organization (WHO) reports aprojected continued rise in cancer mortality with an estimated 13.1 million cancerdeaths worldwide in 2030 [1]. The majority of cancer patients presenting for cancer treatment areolder (>65 years old). Older cancer patients are often diagnosed at laterstages, undertreated, and rarely included in clinical trials [2, 3]. In fact, evidence on cancer treatment is mainly generated in youngercancer patients. Because aging is an individualized process older cancer patientsrepresent a heterogeneous group requiring specific management–an oncologicaland research challenge [2, 4].

Comprehensive multidimensional geriatric assessment has been shown in generalpopulation studies to be a promising tool of assessment domains that capture a rangeof patient factors resulting in an individualized intervention-plan for optimizingclinical management and health outcomes [57]. Similarly, cancer-specific geriatric assessment (C-SGA) with multipleassessment domains can aid in identifying, managing and potentially correcting(through intervention) problems that might specifically interfere with cancertreatment [8, 9]. For over a decade C-SGA has been a growing field of cancer-relatedresearch. Much of the earlier literature only allowed for analysis of indirectevidence and clinical opinion supporting the use of C-SGA [10]. More recent publications include a range of study designs as well asreviews that have expanded C-SGA specific knowledge. (for example [1115]) Moreover, recommendations from the International Society of GeriatricOncology (SIOG) task force on geriatric assessment and the European Organization forResearch and Treatment of Cancer (EORTC) Elderly Task Force have underscored theneed for additional research including validation of C-SGA tools in oncologicalsettings. [16, 17] Despite growing evidence additional studies to determine C-SGA’sability to predict relevant outcomes such as choice of treatment, treatmenttolerance, treatment completion, survival, quality of life, and comparativeeffectiveness to physician judgment are still needed [13, 16, 1824]. Incorporating C-SGA into clinical trials of older cancer patients couldprovide such evidence, establish an objective measure for inclusion into clinicaltrials, and further advance the knowledgebase accelerating translation of use intoevidence-based practice. Evidence from use of C-SGA in clinical trial settings hasbeen increasing but does not directly inform on feasibility of implementation indaily oncological practice [2426]. In clinical practice the ease and time to perform a C-SGA is a vital towhether or not it will be adopted in regular practice and requires specificinvestigation.

The Swiss Group for Clinical Cancer Research (Schweizerische Arbeitsgemeinschaftfür Klinische Krebsforschung [SAKK]) developed the SAKK C-SGA specifically foruse in clinical trials including older cancer patients as well as daily oncologicalpractice. The SAKK C-SGA includes six standard geriatric assessment measurescovering five geriatric assessment domains (comorbidity, function, psychosocial,nutrition, and cognition) using a mix of medical record abstraction (MRA) andpatient interview (described in Table 1). The individualmeasures are brief, reliable, valid, and predictive of morbidity and mortality ingeriatric patients [2733]. The measures have been well studied in the geriatric-oncology andgeriatric assessment literature [34]. The SAKK C-SGA was developed based on an extensive literature search andexpert clinical advice to be brief and easy to implement in busy clinical settings(i.e. low physician and patient burden appropriate for clinical practice as well asclinical trials). The patient questionnaire alone was previously pilot tested in asmall sample (N = 5) of older cancer patients to judge phrasing andcomprehension in older adults, but the complete SAKK C-SGA has not been assessed forease of administration in daily oncological settings. To address this gap the aim ofthe current study was to evaluate feasibility and practical use of the SAKK C-SGA inSwiss clinical practices caring for older cancer patients.
Table 1

Content and operationalization of the SAKK cancer-specific geriatricassessment (C-SGA)

Assessment domain

Assessment tool

Number of questions

How administered

Estimated time required (min)

Range: cut-off

Individual domains

Comorbidity

Age-adjusted Charlson Comorbidity Index (CCI)[27], [28]

18

Medical Record Abstraction (MRA)

5-10

0-43: ≥4

Function

Vulnerable Elders Survey (VES-13) [29]

12

Self-report or Interviewer Administered

<5

0-10: ≥3

Psychosocial

Geriatric Depression Scale 5-item short form (GDS-5) [30]

5

Self-report or Interviewer

<5

0-5: ≥2

Modified MOS- Social Support Survey (mMOS-SS) [33]

8

Administered

<5

0-8: ≤2.5

Nutrition

Mini Nutritional Assessment (MNA) [31]

3

Interviewer administered and MRA

<4

0-14: ≤11

Cognition

Mini-Cog [32]

3

Interviewer administered

5

1-3 w/clock draw: 0 or 1–2 w/abnormal clock

SAKK C-SGA

     

5 Domains

Six Measures: CCI, VES-13, GDS-5, mMOS-SS, MNA, Mini-Cog

22– Patient 20– MRA

Self-report, Interviewer Administered, MRA

<30

0-5: ≥3

Methods

The protocol for this study was reviewed and approved by the KantonsspitalGraubünden (KSGR) Institutional Review Board (IRB) Ethics Committee and theKantonsspital St. Gallen (KSSG) IRB Ethics Committee. The study was conducted incompliance with all federal regulations governing the protection and privacy ofhuman subjects, the Helsinki Declaration, and with the informed consent of theparticipants.

Study population

We conducted a cross sectional study of older cancer patients (total studypopulation, N = 51) cared for at two oncology departments in cantonhospitals in Switzerland: KSGR (N = 25) and KSSG(N = 26). The study population included a consecutive case series ofolder adults ( 65 years old) with physician permission toparticipate, a pathologically confirmed cancer diagnosis (newly diagnosed orrelapsed), presenting for initiation of a new chemotherapy treatment (first-lineor subsequent treatment) between July 2009 and March 2011.

Data collection

Data for this study was collected using three instruments, the SAKK C-SGA plusphysician and patient evaluation forms. The SAKK C-SGA (English, German, French,Italian) and scoring instructions (English only) are available by authorrequest.

SAKK C-SGA

The SAKK C-SGA was administered in two parts by trained study personnel(hereafter referred to as physician) before the start of chemotherapytreatment. The interviewer-administered C-SGA patient questionnairecontained 22-questions with a variety of responses (e.g. rating scales,yes/no, word recall, clock drawing and spaces for recording the results ofthe MRA see Additional file 1). The C-SGA MRAincluded 20-questions for extracting health-related information (height,weight, comorbidities) from patient medical records.

Patient and physician evaluation forms

The patient evaluation form was filled-out by patients immediately aftercompletion of the SAKK C-SGA patient interview. It had a total of11-questions divided into two parts: (1) five patient information questions(gender, marital status, education, nationality, mother tongue); and (2) sixquestions relating to the patient questionnaire (estimated/acceptable timeto complete, ease of administration, patient reaction, difficult/unansweredquestions).

Personnel administering the SAKK C-SGA completed the physician evaluationimmediately after each patient assessment. The physician questionnaireincluded 18-questions divided into four sections: (1) six patientinformation questions (age, general health, type cancer, treatment approach,life expectancy, performance status); (2) six questions relating to theC-SGA patient questionnaire (estimated/acceptable time to complete, ease ofadministration, patient reaction, difficult/unanswered questions); (3) fourquestions relating to MRA (estimated time to complete, ease of conducting,difficulty of obtaining information, missing information), and (4) twoquestions relating to training (years of experience, type of training).

Analytic variables

Socio-demographic characteristics

We categorized information on age (65–69,70-79, 80+ years); gender(male, female); education (compulsory-up to 10th grade, secondary-highschool, teacher training or vocational diploma; tertiary-undergraduate,graduate, post-graduate degree); marital status (single, married, widowed,divorced); and nationality (Swiss, other).

Health-related characteristics

Physician-rated general health was assessed by a single question with fiveanswer likert scale ranging from “excellent” to“poor”. We also classified information on type of cancer(bladder, breast, colon, leukemia, lung, lymphoma, pancreatic, prostate,uterine, other); type of treatment (curative, palliative), physicianjudgement of life expectancy (<1 year, 1-2 years,3 + years); and WHO performance score (0–4) with higherscores indicating worse health [35].

SAKK C-SGA

Table 1 shows content and operationalization ofthe SAKK C-SGA and individual domain assesment tools. SAKK C-SGA scoring wascalculated as five individual domain scores (function, psychosocial,nutrition, cognition, comorbidity based on the published scoring rules forindividual measures) and one overall C-SGA score. All scores were based onwhole number ranges. The five individual domain scores were dichotomized asdeficit or not based on literature-based pre-determined cut-off scores foreach measure (see Table 1). Individual domainswith more than one measure (e.g. psychosocial) were considered to be adeficit if at least one measure crossed the cut-off. The overall C-SGA scorewas calculated by summing the number of individual domain deficits (range0–5) and dichotomizing deficits as ≤2 (fit for standardtreatment) vs. ≥3 (unfit for standard treatment) [11, 14].

SAKK C-SGA evaluation

The SAKK C-SGA was assessed by patient and physician estimated time tocomplete (mean time, acceptable yes/no), ease of completing (easy/fairlyeasy, just right, hard/very hard), patient reaction (interested,indifferent, rejecting), and difficult or unanswered questions (yes/no).

Analytic methods

We examined the descriptive characteristics in the total population and comparedthe distributions between study sites using Student’s t-test forcontinuous variables and chi-square or Fisher’s exact tests forcategorical variables testing for statistical significance in their differences.Similar methods were applied to results from the patient and physicianevaluations. Associations between C-SGA scores and health-relatedcharacteristics were assessed using Spearman correlations. All analyses wereperformed using SAS (V9.3 SAS Institute, Cary, NC) and all p values were fromtwo-sided tests.

Results

Population characteristics

Table 2 shows the characteristics of the individualstudy site populations as well as total study population. Most of the studypopulation was Swiss (88%) and over half were aged 70–79 years, malewith excellent/good physician-rated general health. Lung cancer (20%) andlymphoma (25%) were the most common types of cancers being treated in the totalstudy population.
Table 2

Characteristics of the individual site and total study population

 

KSGR

KSSG

Total population

 

N = 25

N = 26

N = 51

Characteristic

n (%)

P value*

Socio-demographic

    

Study site

    

 KSGR

25 (100)

25 (49)

 

 KSSG

26 (100)

26 (51)

 

Age

    

 65–69 years

6 (24)

7 (27)

13 (25)

0.57

 70–79 years

15 (60)

12 (46)

27 (53)

 

 80+ years

4 (16)

7 (27)

11 (22)

 

Gender

    

 Male

17 (68)

11 (42)

28 (55)

0.07

 Female

8 (32)

15 (58)

23 (45)

 

Education

    

 Compulsory

7 (28)

5 (21)

12 (24)

0.60

 Secondary

17 (68)

16 (67)

33 (67)

 

 Tertiary

1 (4)

3 (12)

4 (8.2)

 

Marital status

    

 Single

0 (0)

0 (0)

0 (0)

0.92 #

 Married

18 (72)

17 (65)

35 (69)

 

 Widowed

7 (28)

7 (27)

14 (27)

 

 Divorced

0 (0)

2 (7.7)

2 (3.9)

 

Nationality

    

 Swiss

22 (88)

23 (88)

45 (88)

1.00

 Other

3 (12)

3 (12)

6 (12)

 

Health-related

    

Physician-rated general health

    

 Excellent/good

17 (71)

8 (32)

25 (51)

0.007

 Fair

6 (25)

9 (36)

15 (31)

 

 Poor/very Poor

1 (4.1)

8 (32)

9 (18)

 

Type of cancer

    

 Bladder

1 (4.0)

0 (0)

1 (1.9)

0.40 #

 Breast

0 (0)

1 (3.8)

1 (1.9)

 

 Colon

6 (24)

0 (0)

6 (12)

 

 Leukemia

0 (0)

7 (27)

7 (14)

 

 Lung

5 (20)

5 (19)

10 (20)

 

 Lymphoma

5 (20)

8 (31)

13 (25)

 

 Pancreatic

0 (0)

1 (3.8)

1 (1.9)

 

 Prostate

3 (12)

1 (3.8)

4 (7.8)

 

 Uterine

0(0)

1 (3.8)

1 (1.9)

 

 Other

5 (20)

2 (7.7)

7 (14)

 

Type of treatment

    

 Curative

12 (48)

3 (12)

15 (29)

0.006

 Palliative

13 (52)

23 (88)

36 (71)

 

Life expectancy

    

 <1 year

4 (19)

16 (61)

20 (42)

0.01

 1-2 years

8 (38)

6 (23)

14 (30)

 

 3+ years

9 (43)

4 (15)

13 (28)

 

WHO performance score

    

 0

15 (60)

4 (15)

19 (37)

<0.001

 1

9 (36)

9 (35)

18 (35)

 

 2

1 (4.0)

13 (50)

14 (27)

 

KSGR Kantonsspital Graubünden, KSSG Kantonsspital St. Gallen,WHO World Health Organization.

* Test of difference between study sites.

# Significance test excluding marital status single andcancers with zero cells.

The individual study site populations were evenly distributed (KSGR 49%, KSSG51%) but differed on health-related characteristics. KSGR had more patients withexcellent/good (71% vs. 32%, p = 0.007) physician-ratedgeneral health. The majority of KSSG patients were receiving palliativetreatment (88% vs. 52%, p = 0.006); had a life expectancyof <1 year (61% vs. 19%, p = 0.01); and a WHOperformance score of 2 (50% vs. 4%, p = <0.001).

Patient and physician evaluation of the SAKK C-SGA

Table 3 displays the results of patient and physicianevaluation of the SAKK C-SGA. The time to complete the patient questionnaire wasconsidered acceptable by almost all patients and physicians. Patients reportedslightly shorter times to complete the questionnaire than physicians(17.33 ± 7.34 vs.20.59 ± 6.53 minutes, p = 0.02). Both groupsrated the SAKK C-SGA patient questionnaire as easy/fairly easy to complete (91%vs. 84% respectively, p = 0.14) with few difficult orunanswered questions. Physicians reported the MRA took on average8.32 ± 4.72 minutes, 100% considered the time to completeacceptable with 96% rating it as easy/fairly easy to complete. The majority ofpersonnel administering the SAKK C-SGA were nursing staff with over11 years of clinical experience (data not shown).
Table 3

Patient and physician evaluation of SAKK cancer-specific geriatricassessment (C-SGA)

 

Patient N = 51

Physician N = 9

 
 

n (%) or mean ± SD

P value*

Patient questionnaire

   

Estimated time to complete

   

 Average time in minutes

17.33 ±7.34

20.59 ±6.53

0.02

 Rated time as acceptable

44 (96)

51 (100)

0.22

Ease of completing

   

 Easy/fairly easy

42 (91)

43 (84)

0.14 #

 Just right

4 (8.7)

3 (5.9)

 

 Hard/very hard

0 (0)

5 (9.8)

 

Patient reaction

  

§

 Interested

37 (73)

44 (86)

 

 Indifferent

7 (14)

5 (9.8)

 

 Rejecting

0 (0)

2 (3.9)

 

Difficult to answer questions

   

 Yes

5 (11)

8 (16)

0.91

 No

39 (89)

43 (84)

 

Unanswered questions

   

 Yes

6 (14)

11 (22)

0.03

 No

38 (86)

40 (78)

 

Medical record abstraction

   

Estimated time to complete

   

 Average time in minutes

8.32 ±4.72

 

 Rated time as acceptable

50(100)

 

Ease of completing

   

 Easy/fairly easy

47 (96)

 

 Just right

2 (4.1)

 

 Hard/very hard

0 (0)

 

C-SGA Cancer-Specific Geriatric Assessment; SDstandard deviation.

* Test of difference between patients and physicians.

# Significance test based on two categories created bycollapsing the bottom two categories with the smallest cell countsinto one (e.g. Easy/Fairly easy vs. Just right/Hard/Very hard).

§ Significance test not possible due to zerocells.

C-SGA characteristics of the study population

Table 4 describes the C-SGA characteristics of thestudy population. The overall mean C-SGA score was 2.4 ± 1.12.Patients at KSGR had lower C-SGA scores (2.00 ± 1.19vs. 2.81 ± 0.90, p = 0.009) and asmaller proportion (28% vs. 65%, p = 0.008) was above theC-SGA cut-off score compared to KSSG. In the total population the most commondomain deficits were function, nutrition and comorbidity. By site populationKSGR had statistically significantly fewer patients with a nutrition domaindeficit than KSSG. C-SGA score was correlated with all health-relatedcharacteristics but not with age. Health-related characteristics differed byC-SGA cut-off scores. More patients with C-SGA score below the cut-off hadexcellent/good physician-rated general health (67% vs. 32%,p = 0.007); were receiving curative treatment (44% vs. 13%,p = 0.02); and had a WHO performance score of 0 or 1 (85%vs. 58%, p = 0.002).
Table 4

Cancer-specific geriatric assessment (C-SGA) characteristics in theindividual site and total study populations

C-SGA characteristics by study site and totalpopulation

 

KSGR

KSSG

Total population

 

N = 25

N = 26

N = 51

 
 

n (%)

P value*

Mean C-SGA score

2.00 ±1.19

2.81 ±0.90

2.4 ±1.12

0.009

C-SGA over cut-off score

    

 Yes

7 (28)

17 (65)

24 (47)

0.008

 No

18 (72)

9 (35)

27 (53)

 

C-SGA by domain deficits

    

Function

    

 Yes

14 (56)

18 (69)

32 (63)

0.33

 No

11 (44)

8 (31)

19 (37)

 

Psychosocial

    

 Yes

1 (4.0)

7 (27)

8 (16)

0.05

 No

24 (96)

19 (73)

43 (16)

 

Nutrition

    

 Yes

13 (52)

22 (85)

35 (69)

0.02

 No

12 (48)

4 (15)

16 (31)

 

Cognition

    

 Yes

3 (12)

1 (3.9)

4 (7.8)

0.35

 No

22 (88)

25 (96)

47 (92)

 

Comorbidity

    

 Yes

19 (76)

25 (96)

44 (86)

0.05

 No

6 (24)

1 (3.9)

7 (14)

 

C-SGA association with health-relatedcharacteristics

 

Correlation coefficient

P value

Age

0.12

0.43

Physician-rated general health

0.39

0.008

Type of treatment

−0.38

0.01

Life expectancy

−0.32

0.03

WHO performance score

0.63

<0.0001

Health-related characteristics by C-SGA cut-offscore

 

Below C-SGA cut-off score

Above C-SGA cut-off score

 

N = 27

N = 24

 

n (%)

P value #

Age

   

 65–69 years

8 (30)

5 (21)

0.72

 70–79 years

14 (52)

13 (54)

 

 80+ years

5 (18)

6 (25)

 

Physician-rated general health

   

 Excellent/good

18 (67)

7 (32)

0.007

 Fair

8 (30)

7 (32)

 

 Poor/very poor

1 (3.7)

8 (36)

 

Type of treatment

   

 Curative

12 (44)

3 (13)

0.02

 Palliative

15 (56)

21 (87)

 

Life expectancy

   

 <1 year

7 (30)

13 (54)

0.16

 1–2 years

7 (30)

7 (29)

 

 3+ years

9 (39)

4 (17)

 

WHO performance score

   

 0

16 (59)

3 (12)

0.002

 1

7 (26)

11 (46)

 

 2

4 (15)

10 (42)

 

C-SGA Cancer-Specific Geriatric Assessment, KSGRKantonsspital Graubünden KSSG Kantonsspital St.Gallen, SD Standard Deviation, WHO World HealthOrganization.

* Test of difference between study sites.

# Test of difference between groups by C-SGA cut-off.

Discussion

This study demonstrated that the SAKK C-SGA is feasible and easy to implement indaily clinical practice. The overall time to complete was less than 30 minutesand considered acceptable by patient and physician alike. Importantly, mostparticipants rated the SAKK C-SGA (patient questionnaire and MRA) as easy or fairlyeasy to complete. Only a small number of patients or physicians reported questionsthat were either difficult or unanswered. This likely reflects real-worldpatient-specific difficulties encountered in daily oncological practice as opposedto problems with the questions themselves. This is supported by the fact that allmeasures included in the SAKK C-SGA are widely used and previously validated. Plusthere was no pattern to the individual questions that were reported as difficult orunanswered (e.g. only one question was mentioned twice, questions identified werenot domain-specific).

These findings also suggest that the SAKK C-SGA was able to objectively discriminateolder patients’ health. The difference in SAKK C-SGA scores between studysites mirrored differences in the site-specific patient characteristics. KSGR had anoverall healthier population (as assessed by other health-related measures) with ahigher proportion of patients being treated for curative intent than KSSG.Correspondingly, KSGR SAKK C-SGA scores were lower and a higher proportion was belowthe cut-off score (i.e. fit for standard treatment). In both sites more patientsbelow the cut-off score had better physician-rated general health, a longer lifeexpectancy as well as better WHO performance scores. Interestingly, in thispopulation C-SGA scores were related to other health measures but not age,underscoring the potential advantage of C-SGA versus age-based decision-making. Asexpected (i.e. health-related measures were designed to assess unique health states)there was not complete overlap in how individual health-related measures and theSAKK C-SGA identified the patient population. This is similar to findings by otherresearchers who found physicians and C-SGA do not identify the same patientpopulations as fit/unfit for treatment and that C-SGA compares with but is notidentical to assessments based other health-measures [3640].

Although shorter screening tools exist and the SAKK C-SGA had high correlation withother simpler health-related measures (e.g. WHO performance score) it affordsadditional clinical benefit to patient and provider [4, 25, 4044]. The time to complete though longer than a brief screen is suitable forpre-treatment or pre-trial oncological work-ups. The time to complete the SAKK C-SGAin daily oncological practice was similar to that of another C-SGA tool pilot testedin clinical trial settings [25]. It also requires much less time and no referral for a full comprehensivegeriatric assessment. This is particularly important since not all healthcaresystems offer specialized geriatric care. Second, the SAKK C-SGA identifiesindividual domain deficits for intervention acknowledged within the C-SGA literatureas having specific benefits in the care of older cancer patients [2, 16, 19, 45]. The problems identified can be addressed either within oncologicalpractices and/or by referral depending on available resources and expertise. Forexample, engaging social workers, arranging transportation, or providing nutritionalcounseling before start of treatment could be arranged by staff handling cancertreatment, general practitioners, or referral to a geriatrician depending onindividual patient needs and care situations.

The key advantage is that such interventions, regardless of where they are initiated,may mitigate an older patient’s risk for poor cancer treatment outcomes andincrease their quality of life. A recent study in Spain showed that C-SGA detectsmore information than oncological evaluation alone [46]. Another in Canada found that in 70% of their study patients C-SGAidentified previously unidentified medical problems [47]. In this Swiss population, over half of our patients were identified bythe SAKK C-SGA as not a risk for poor outcomes (i.e. below cut-off/fit for standardtreatment). Nevertheless, all but one of these patients had a deficit in at leastone domain that otherwise may not have been identified by oncological evaluationalone or even a briefer C-SGA screen. Thus use of the SAKK C-SGA provides readilyusable information that can improve outcomes for patients above or below the cut-offwithout delay (i.e. no additional assessment necessarily required). However, we didfind that when dementia was present the SAKK C-SGA (like any geriatric assessment)was challenging to administer. In patients with dementia (especially advanced cases)decision-making regarding treating cancer is likely not be aided by objective C-SGAmeasurement. But instead will require a more complex individualized process betweenpatient-physician-family/caregiver.

Other researchers in the field and SIOG have identified the need for shorter C-SGAtools applicable for busy clinical oncology settings [48]. The SAKK C-SGA has several benefits directly addressing this need.First, assessment using the SAKK C-SGA requires much less time than a full geriatricassessment and does not require referral to a specialist or geriatrics training toadminister. Second, using standard geriatric assessment tools in key domains theSAKK C-SGA maximizes information gathering and minimizes patient/physician burden.In fact, since the tool can be administered by any combination of patient (all butMini-Cog), trained staff, or physician it is easily customized to the individualpatient and clinical setting. Lastly, our findings suggest that use of the SAKKC-SGA is feasible in clinical practice and may be well suited to determineeligibility for clinical trials based on patient health instead of chronologicalage.

A major challenge of C-SGA is to find a balance between time to conduct and producingclinically useful information (i.e. identifying targets for intervention). The SAKKC-SGA is a step forward in this balancing act but can be further improved. Thisstudy used an electronic excel-based CCI calculator that made collecting comorbiditydata and calculating CCI information easier, more accurate, and immediatelyavailable [28]. Based on our positive experience with the excel-based CCI we decidedthat an electronic version of the SAKK C-SGA could offer similar advantages inclinical practice and clinical trials. An electronic SAKK C-SGA would make gatheringdata more efficient, produce real time results that can be immediately incorporatedinto treatment planning, and increase the likelihood of more widespread and uniformuse. Development is underway and validation and feasibility studies are planned.

Several strengths and weaknesses of this study should be considered. The SAKK C-SGAwas developed with input from geriatricians and oncologists specifically for use inbusy oncological and clinical trial settings. The tool includes only standardpsychometrically evaluated geriatric assessment measures covering previouslyidentified key domains. The SAKK C-SGA is easy to score and available in multiplelanguages. However, the results of this study are based on a small number ofpatients in Switzerland. Thus generalizability of these findings to other clinicalsettings and other healthcare systems are limited. We did not require exactmeasurement of time to complete and assume that most patients/physicians estimatedtimes. However, both the patient and physician estimates had similar standarddeviations and the perception of time to complete (i.e. estimated time wasacceptable to nearly all) is an important factor in willingness to adopt the tool.The SAKK C-SGA should be further tested in larger patient samples, a variety ofsettings, and over longer periods of time to include outcome data.

Conclusions

In conclusion, these results suggest that the SAKK C-SGA is a promising tool for usein clinical practice. It demonstrated discriminative ability based on objectivegeriatric assessment measures. It also provides important clinical information thatcould be used for interventions aimed at optimizing outcomes in older cancerpatients. Future studies of SAKK C-SGA reliability, discriminative ability forclinical decision-making based on treatment outcomes, accuracy predictingcancer-related outcomes, ability to improve outcomes, and generalizability arewarranted.

Abbreviations

CCI: 

Charlson comorbidy index

C-SGA: 

Cancer-specific geriatric assessment

GDS: 

Geriatric depression scale

IRB: 

Institutional review board

KSGR: 

Kantonspitalgraubünden

KSSG: 

Kantonspital St. Gallen

mMOS-SS: 

Modified medical outcomesstudy -social support survey

MNA: 

Mini nutritional assessment

MRA: 

Medical recordabstraction

SAKK: 

Schweizerische arbeitsgemeinschaft für klinischekrebsforschung

SIOG: 

International society of geriatric oncology

VES: 

Vulnerableelders survey

WHO: 

World Health Organization.

Declarations

Acknowledgement

We are grateful to the practitioners and participants involved in this study.This work was supported by funds from the Swiss Group for Clinical CancerResearch [38/08].

Authors’ Affiliations

(1)
Institute of Social and Preventive Medicine (ISPM), University of Bern
(2)
Section of Geriatrics, Boston University School of Medicine, Boston Medical Center
(3)
Division of Geriatrics, Department of General Internal Medicine, Inselspital University Hospital and University of Bern
(4)
Swiss Group for Clinical Cancer Research, SAKK Coordinating Centre
(5)
Kantonsspital Graubünden
(6)
Oncology-Haematology Department
(7)
Department of Internal Medicine, Hematology, Medical Oncology, and Pneumology, University Medical Center of the Johannes Gutenberg University

References

  1. World Health Organization: Cancer. Fact sheet 297. 2013, Geneva, Switzerland: WHO,http://​www.​who.​int/​mediacentre/​factsheets/​fs297/​en/​index.​html,
  2. Balducci L, Colloca G, Cesari M, Gambassi G: Assessment and treatment of elderly patients with cancer. Surg Oncol. 2010, 19 (3): 117-123. 10.1016/j.suronc.2009.11.008.View ArticlePubMed
  3. Macleod U, Mitchell ED, Burgess C: Risk factors for delayed presentation and referral of symptomatic cancer:evidence for common cancers. Br J Cancer. 2009, 101 (Suppl 2): S92-S101.PubMed CentralView ArticlePubMed
  4. Hurria A, Lichtman SM, Gardes J: Identifying vulnerable older adults with cancer: integrating geriatricassessment into oncology practice. J Am Geriatr Soc. 2007, 55: 1604-1608. 10.1111/j.1532-5415.2007.01367.x.View ArticlePubMed
  5. Huss A, Stuck AE, Rubenstein LZ: Multidimensional preventive home visit programs for community-dwelling olderadults: a systematic review and meta-analysis of randomized controlledtrials. J Gerontol A Biol Sci Med Sci. 2008, 63: 298-307. 10.1093/gerona/63.3.298.View ArticlePubMed
  6. Wieland D, Ferrucci L: Multidimensional geriatric assessment: back to the future. J Gerontol A Biol Sci Med Sci. 2008, 63: 272-274. 10.1093/gerona/63.3.272.PubMed CentralView ArticlePubMed
  7. Ellis G, Whitehead MA, Robinson D: Comprehensive geriatric assessment for older adults admitted to hospital:meta-analysis of randomised controlled trials. BMJ. 2011, 343: d6553-10.1136/bmj.d6553.PubMed CentralView ArticlePubMed
  8. Brunello A, Sandri R, Extermann M: Multidimensional geriatric evaluation for older cancer patients as a clinicaland research tool. Cancer Treat Rev. 2009, 35: 487-492. 10.1016/j.ctrv.2009.04.005.View ArticlePubMed
  9. Balducci L: New paradigms for treating elderly patients with cancer: the comprehensivegeriatric assessment and guidelines for supportive care. J Support Oncol. 2003, 1: 30-37.PubMed
  10. Chen CC, Kenefick AL, Tang ST, McCorkle R: Utilization of comprehensive geriatric assessment in cancer patients. Crit Rev Oncol Hematol. 2004, 49: 53-67. 10.1016/S1040-8428(03)00098-2.View ArticlePubMed
  11. Clough-Gorr KM, Stuck AE, Thwin SS, Silliman RA: Older breast cancer survivors: geriatric assessment domains are associatedwith poor tolerance of treatment adverse effects and predict mortality over7 years of follow-up. J Clin Oncol. 2010, 28: 380-386. 10.1200/JCO.2009.23.5440.PubMed CentralView ArticlePubMed
  12. Caillet P, Canoui-Poitrine F, Vouriot J: Comprehensive geriatric assessment in the decision-making process in elderlypatients with cancer: ELCAPA study. J Clin Oncol. 2011, 29: 3636-3642. 10.1200/JCO.2010.31.0664.View ArticlePubMed
  13. Liu JJ, Extermann M: Comprehensive geriatric assessment and its clinical impact in oncology. Clin Geriatr Med. 2012, 28: 19-31. 10.1016/j.cger.2011.10.001.View ArticlePubMed
  14. Clough-Gorr KM, Thwin SS, Stuck AE, Silliman RA: Examining five- and ten-year survival in older women with breast cancer usingcancer-specific geriatric assessment. Eur J Cancer. 2012, 48: 805-812. 10.1016/j.ejca.2011.06.016.PubMed CentralView ArticlePubMed
  15. Puts MT, Hardt J, Monette J: Use of geriatric assessment for older adults in the oncology setting: asystematic review. J Natl Cancer Inst. 2012, 104: 1133-1163.View ArticlePubMed
  16. Extermann M, Aapro M, Bernabei R: Use of comprehensive geriatric assessment in older cancer patients:recommendations from the task force on CGA of the International Society ofGeriatric Oncology (SIOG). Crit Rev Oncol Hematol. 2005, 55: 241-252. 10.1016/j.critrevonc.2005.06.003.View ArticlePubMed
  17. Pallis AG, Fortpied C, Wedding U: EORTC elderly task force position paper: approach to the older cancerpatient. Eur J Cancer. 2010, 46: 1502-1513. 10.1016/j.ejca.2010.02.022.View ArticlePubMed
  18. Cohen HJ: The cancer aging interface: a research agenda. J Clin Oncol. 2007, 25: 1945-1948. 10.1200/JCO.2007.10.6807.View ArticlePubMed
  19. Maas HA, Janssen-Heijnen ML, Olde Rikkert MG, Machteld Wymenga AN: Comprehensive geriatric assessment and its clinical impact in oncology. Eur J Cancer. 2007, 43: 2161-2169. 10.1016/j.ejca.2007.08.002.View ArticlePubMed
  20. Balducci L, Extermann M: A practical approach to the older patient with cancer. Curr Probl Cancer. 2001, 25: 6-76.View ArticlePubMed
  21. Roche RJ, Forman WB, Rhyne RL: Formal geriatric assessment, an imperative for the older person withcancer. Cancer Pract. 1997, 5: 81-86.PubMed
  22. Extermann M: Basic assessment of the older cancer patient. Curr Treat Options Oncol. 2011, 12: 276-285. 10.1007/s11864-011-0161-5.View ArticlePubMed
  23. Soubeyran P, Fonck M, Blanc-Bisson C: Predictors of early death risk in older patients treated with first-linechemotherapy for cancer. J Clin Oncol. 2012, 30: 1829-1834. 10.1200/JCO.2011.35.7442.View ArticlePubMed
  24. Aparicio T, Jouve JL, Teillet L: Geriatric factors predict chemotherapy feasibility: ancillary results of FFCD2001–02 phase III study in first-line chemotherapy for metastaticcolorectal cancer in elderly patients. J Clin Oncol. 2013, 31: 1464-1470. 10.1200/JCO.2012.42.9894.View ArticlePubMed
  25. Hurria A, Cirrincione CT, Muss HB: Implementing a geriatric assessment in cooperative group clinical cancertrials: CALGB 360401. J Clin Oncol. 2011, 29: 1290-1296. 10.1200/JCO.2010.30.6985.PubMed CentralView ArticlePubMed
  26. Falandry C, Brain E, Bonnefoy M, Mefti F, Jovenin N, Rigal O, Guillem O, El Kouri C, Uwer L, Abadie-Lacourtoisie S: Impact of geriatric risk factors on pegylated liposomal doxorubicin toleranceand efficacy in elderly metastatic breast cancer patients: Final results ofthe DOGMES multicentre GINECO trial. Eur J Cancer. 2013, 49 (13): 2806-2814. 10.1016/j.ejca.2013.04.027.View ArticlePubMed
  27. Charlson ME, Pompei P, Ales KL, MacKenzie CR: A new method of classifying prognostic comorbidity in longitudinal studies:development and validation. J Chronic Dis. 1987, 40: 373-383. 10.1016/0021-9681(87)90171-8.View ArticlePubMed
  28. Hall WH, Ramachandran R, Narayan S: An electronic application for rapidly calculating Charlson comorbidityscore. BMC Cancer. 2004, 4: 94-10.1186/1471-2407-4-94.PubMed CentralView ArticlePubMed
  29. Saliba D, Elliott M, Rubenstein LZ: The Vulnerable Elders Survey: a tool for identifying vulnerable older peoplein the community. J Am Geriatr Soc. 2001, 49: 1691-1699. 10.1046/j.1532-5415.2001.49281.x.View ArticlePubMed
  30. Rinaldi P, Mecocci P, Benedetti C: Validation of the five-item geriatric depression scale in elderly subjects inthree different settings. J Am Geriatr Soc. 2003, 51: 694-698. 10.1034/j.1600-0579.2003.00216.x.View ArticlePubMed
  31. Vellas B, Villars H, Abellan G: Overview of the MNA–Its history and challenges. J Nutr Health Aging. 2006, 10: 456-463. discussion 463–455PubMed
  32. Borson S, Scanlan JM, Chen P, Ganguli M: The Mini-Cog as a screen for dementia: validation in a population-basedsample. J Am Geriatr Soc. 2003, 51: 1451-1454. 10.1046/j.1532-5415.2003.51465.x.View ArticlePubMed
  33. Moser A, Stuck AE, Silliman RA: The 8-item modified medical outcomes study social support survey (mMOS-SS):psychometric evaluation shows excellent performance. J Clin Epidemiol. 2012, 10.1016/j.jclinepi.2012.04.007. in press,
  34. McDowell I: Measuring health: A guide to rating scales and questionnaires. 2006, New York, NY: Oxford University PressView Article
  35. Oken MM, Creech RH, Tormey DC: Toxicity and response criteria of the eastern cooperative oncology group. Am J Clin Oncol. 1982, 5: 649-655. 10.1097/00000421-198212000-00014.View ArticlePubMed
  36. Wedding U, Kodding D, Pientka L: Physicians’ judgement and comprehensive geriatric assessment (CGA)select different patients as fit for chemotherapy. Crit Rev Oncol Hematol. 2007, 64: 1-9. 10.1016/j.critrevonc.2007.05.001.View ArticlePubMed
  37. Tucci A, Ferrari S, Bottelli C: A comprehensive geriatric assessment is more effective than clinical judgmentto identify elderly diffuse large cell lymphoma patients who benefit fromaggressive therapy. Cancer. 2009, 115: 4547-4553. 10.1002/cncr.24490.View ArticlePubMed
  38. Luciani A, Ascione G, Bertuzzi C: Detecting disabilities in older patients with cancer: comparison betweencomprehensive geriatric assessment and vulnerable elders survey-13. J Clin Oncol. 2010, 28: 2046-2050. 10.1200/JCO.2009.25.9978.View ArticlePubMed
  39. Mohile SG, Bylow K, Dale W: A pilot study of the vulnerable elders survey-13 compared with thecomprehensive geriatric assessment for identifying disability in olderpatients with prostate cancer who receive androgen ablation. Cancer. 2007, 109: 802-810. 10.1002/cncr.22495.View ArticlePubMed
  40. Owusu C, Koroukian SM, Schluchter M: Screening older cancer patients for a comprehensive geriatric assessment: acomparison of three instruments. J Geriatr Oncol. 2011, 2: 121-129. 10.1016/j.jgo.2010.12.002.PubMed CentralView ArticlePubMed
  41. Overcash JA, Beckstead J, Moody L: The abbreviated comprehensive geriatric assessment (aCGA) for use in theolder cancer patient as a prescreen: scoring and interpretation. Crit Rev Oncol Hematol. 2006, 59: 205-210. 10.1016/j.critrevonc.2006.04.003.View ArticlePubMed
  42. Molina-Garrido MJ, Guillen-Ponce C: Development of a cancer-specific comprehensive geriatric assessment in aUniversity Hospital in Spain. Crit Rev Oncol Hematol. 2011, 77: 148-161. 10.1016/j.critrevonc.2010.02.006.View ArticlePubMed
  43. Kellen E, Bulens P, Deckx L: Identifying an accurate pre-screening tool in geriatric oncology. Crit Rev Oncol Hematol. 2010, 75: 243-248. 10.1016/j.critrevonc.2009.12.002.View ArticlePubMed
  44. Bellera CA, Rainfray M, Mathoulin-Pelissier S: Screening older cancer patients: first evaluation of the G-8 geriatricscreening tool. Ann Oncol. 2012, 23: 2166-2172. 10.1093/annonc/mdr587.View ArticlePubMed
  45. Hurria A: Geriatric assessment in oncology practice. J Am Geriatr Soc. 2009, 57 (Suppl 2): S246-S249.View ArticlePubMed
  46. Girones R, Torregrosa D, Maestu I: Comprehensive geriatric assessment (CGA) of elderly lung cancer patients: asingle-center experience. J Geriatr Oncol. 2012, 3: 98-103. 10.1016/j.jgo.2011.12.005.View Article
  47. Horgan AM, Leighl NB, Coate L: Impact and feasibility of a comprehensive geriatric assessment in theoncology setting: a pilot study. Am J Clin Oncol. 2012, 35: 322-328. 10.1097/COC.0b013e318210f9ce.View ArticlePubMed
  48. Balducci L, Extermann M: Cancer and aging. An evolving panorama. Hematol Oncol Clin North Am. 2000, 14: 1-16. 10.1016/S0889-8588(05)70274-4.View ArticlePubMed
  49. Pre-publication history

    1. The pre-publication history for this paper can be accessed here:http://​www.​biomedcentral.​com/​1472-6947/​13/​93/​prepub

Copyright

© Clough-Gorr et al.; licensee BioMed Central Ltd. 2013

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://​creativecommons.​org/​licenses/​by/​2.​0), whichpermits unrestricted use, distribution, and reproduction in any medium, provided theoriginal work is properly cited.

Advertisement