Distress in cancer patients: Who are the main groups at risk?

Abstract

Objectives:
Psychosocial distress is common in cancer patients and survivors and encompasses a broad range of concerns and psychological symptoms. The aim of the current study was to identify subgroups of respondents who experience a specific constellation of distress symptoms.

Methods:
This study uses data from a large data base (n = 21 680) of cancer patients from diverse settings who provided data in the Questionnaire on Distress in Cancer Patients - Short Form (QSC-R10). Cluster analysis was applied to identify subgroups with a distinct constellation of distress symptoms.

Results:
The results showed five distinct clusters: minimally distressed patients (46.6% of the sample), highly distressed patients (12.7%), mainly physically distressed patients (15.2%), mainly psychologically distressed patients (15.6%) and mainly socially distressed patients (9.9%). These groups differed with regard to age, sex, cancer site, treatment setting and disease progression.

Conclusion:
The results revealed large heterogeneity in the experience of distress. Distress clusters were associated with socio-demographic, and clinical variables. These associations might aid a clinician to tailor interventions and to address specific types of distress. This article is protected by copyright. All rights reserved.

Full text

Distress in cancer patients: who are the main groups at risk?
Peter Herschbach1,3, Ingrid Britzelmeir2, Andreas Dinkel3, Jürgen M. Giesler4, Kathleen Herkommer5,
Alexandra Nest6, Theresia Pichler1, Ralf Reichelt7, Sylvia Tanzer-Küntzer1, Joachim Weis8, Birgitt
Marten-Mittag3
1 Comprehensive Cancer Center, Munich, Germany.
2 Institute for Psychooncology, Klinikum Lippe, Lemgo, Germany
3 Technical University of Munich, School of Medicine, Klinikum rechts der Isar, Department of
Psychosomatic Medicine and Psychotherapy, Munich, Germany
4 Section of Health Care Research and Rehabilitation Research, Faculty of Medicine and Medical
Center University of Freiburg, Germany
5 Technical University of Munich, School of Medicine, Klinikum rechts der Isar, Department of
Urology, Munich, Germany
6 Department of Pediatric Hematology, Oncology, Hemostaseology and Stem Cell Transplantation,
Dr. von Hauner University Children's Hospital, Ludwig Maximilian University Munich, Germany.
7 Onkotrakt AG, Hamburg, Germany
8 Comprehensive Cancer Center, Department of Self-Help Research, Faculty of Medicine and
Medical Center University of Freiburg, Germany
Corresponding author:
Dr. Birgitt Marten-Mittag
Technical University of Munich
Klinikum rechts der Isar
Department of Psychosomatic Medicine and Psychotherapy
Langerstrasse 3
D-81675 Muenchen
Tel.: +49 89 4140-4317
E-Mail: b.marten-mittag@TUM.de
This article is protected by copyright. All rights reserved.
Accepted Article
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/pon.5321

Abstract
Objectives: Psychosocial distress is common in cancer patients and survivors and encompasses a
broad range of concerns and psychological symptoms. The aim of the current study was to identify
subgroups of respondents who experience a specific constellation of distress symptoms.
Methods: This study uses data from a large data base (n = 21,680) of cancer patients from diverse
settings who provided data in the Questionnaire on Distress in Cancer Patients - Short Form (QSC-
R10). Cluster analysis was applied to identify subgroups with a distinct constellation of distress
symptoms.
Results: The results showed five distinct clusters: minimally distressed patients (46.6 % of the
sample), highly distressed patients (12.7 %), mainly physically distressed patients (15.2 %), mainly
psychologically distressed patients (15.6 %) and mainly socially distressed patients (9.9 %). These
groups differed with regard to age, sex, cancer site, treatment setting and disease progression.
Conclusion: The results revealed large heterogeneity in the experience of distress. Distress clusters
were associated with socio-demographic, and clinical variables. These associations might aid a
clinician to tailor interventions and to address specific types of distress.
Key words: cancer, cluster analysis, distress screening, oncology, psycho-oncology, psychosocial
distress
This article is protected by copyright. All rights reserved.
Accepted Article

1 Background
Investigating the distress of cancer patients is one of the main research topics in psycho-oncology.
The importance of this research has increased dramatically since distress screenings were
recommended in guidelines and became certification-criteria for cancer centers1-4. There have been
many studies about the prevalence of elevated levels of distress since the work of van’t Spijker et al.
(1997)5, Pascoe et al. (2000)6 or Zabora et al. (2001)7. Researchers found prevalence rates between
23%8 and 60%9. Herschbach et al. (2004) reported 36%10, Zabora et al. (2001) 35%7, Mitchell et al.
(2011) 39%11, and Mehnert et al. (2018) 52%12. The range of the sample sizes of these studies is
between n = 3039 and n = 6,41411.
The comparability of these distress rates is difficult for the following reasons. Firstly—since distress is
a very broad construct—definition and measurement vary among these studies. Today the definition
of distress by the National Comprehensive Cancer Network (NCCN) is widely accepted: ―a
multifactorial unpleasant experience of a psychological (i.e., cognitive, behavioral, emotional), social,
spiritual and/or physical nature that may interfere with the ability to cope effectively with cancer, its
physical symptoms and its treatment…‖3. Secondly, distress is measured by different instruments.
One of the most commonly used questionnaires to assess distress is the Distress Thermometer
(DT)14. However, even prevalence rates based on the DT vary between 39%11 and 60%9. Moreover, it
is essential to distinguish distress from psychiatric disorders like major depressive disorder or
generalized anxiety disorders, which are more aptly assessed using standardized interviews like the
CIDI13 as opposed to distress questionnaires. Thirdly, in most studies, different subsamples of cancer
patients have been assessed in terms of diagnosis, treatment setting, stage of disease, sex, and age.
Most studies have been conducted with breast cancer patients.
Against this background we present results from our large German database (n = 21,680). It consists
of data on psychosocial distress collected using our Questionnaire on Distress in Cancer Patients -
Short Form (QSC-R10) — a ten-items-questionnaire quick and easy to use, with the ability to
measure different distress areas of cancer patients15. This sample size allows us to compare various
cancer sites using the same questionnaire, in terms of their constellation of distress symptoms.
2 Methods
2.1 Patient recruitment and sample
The database comprises n = 21,680 cancer patients from 16 different subsamples. The data were
provided by cooperating clinics and medical institutions in Germany which used the QSC-R1015.
Patients were recruited from 2007 to 2018 from five treatment settings: aftercare of patients who were
contacted at home (three subsamples), acute care hospitals (six subsamples), specialized medical
practices (one subsample), outpatient departments of university clinics (five subsamples), and
rehabilitation clinics (one subsample). Additionally, the following variables were considered: age, sex,
cancer diagnosis and disease progression, which is defined as either presence of metastases, tumor
recurrence or a second primary tumor.
This article is protected by copyright. All rights reserved.
Accepted Article

As the current analysis was based on anonymized data, no formal ethical approval was sought.
2.2 Measures
The QSC-R10 is a ten-item self-assessment questionnaire for psychosocial distress in cancer
patients15. It comprises the following items: fear of disease progression, feeling tired and weak,
reduced recreation activities, feeling tense and/or nervous, disturbed sleep, feeling physically
imperfect, pain, missing partner's empathy, few opportunities to speak with psycho-oncological
professionals, and not feeling well informed about disease/treatment (see Table 2).
Patients indicate whether each item applies to them, and if so, how severely. Thus, each item is rated
on a six-point scale between 0 (the problem does not apply to me) and 5 (the problem applies to me
and is a very serious problem). A total distress score is calculated by adding the single item ratings.
This questionnaire shows good psychometric properties; in the present sample Cronbach’s Alpha for
the total score is 0.85. A validated cutoff score > 14 was used to identify patients with heightened
distress15.
2.3 Statistical analyses
SPSS/PC23 (SPSS, Chicago, IL, USA) was used for analyses. To identify patients with similar
distress patterns a cluster analysis was conducted. Clustering variables were the QSC-R10 items.
Since clustering requires valid values in all variables, patients showing any missing values in QSC-
R10 items were eliminated. A final sample of n = 19,743 was considered for the clustering procedure.
Due to the large sample size we used quick cluster procedure, which is a k-means method using
Euclidean distances between observations to estimate clusters. Initial cluster centers were computed
by the quick cluster procedure. As this procedure of the statistical software SPSS does not reveal an
optimal number of clusters, we performed quick cluster analyses with two to six clusters: the two-
cluster solution distinguished only between low and high distress; the 3-cluster result revealed low,
medium and high distressed cases; the 4-cluster solution discriminated between low, high, primarily
physical and primarily emotional distress, while the 5-cluster solution revealed an additional group
with mainly social distress. The 6-cluster solution was hard to interpret. Using the criteria of
theoretically meaningful groups, ease of interpretation, and an appropriate number of cases within
each cluster, we chose the 5-cluster solution. To validate the five cluster solution we conducted two
additional cluster analyses (k-means with five clusters), one in a 50% random sample (n = 9,876) and
another in the remaining cases (n = 9,877). Both cluster solutions revealed nearly identical means of
the ten items. Cluster labels are based on the difference between the total sample means and cluster
members’ single item means (see blue and gray bars in Figure 1).
Patient characteristics are reported as absolute and relative frequencies or as means and standard
deviations. Chi-square tests are used to compare the ratios of categorical variables across the five
clusters; an F-test was used to compare the means of continuous variables across the clusters, p-
This article is protected by copyright. All rights reserved.
Accepted Article

values for overall significances are shown. Psychosocial distress in socio-demographic and clinical
subgroups is reported by means of percentage above the cutoff (>14) for the QSC-R10 total score.
Multivariate logistic regression models are used to estimate the probability of the age, sex, treatment
setting, diagnostic and disease progression subgroups to belong to the five clusters. We built five
regression models, one for each cluster. In each model the dependent variable was cluster affiliation
vs. rest of the sample. Odds ratios (OR) and 95% confidence intervals (CI) are reported. In predictor
variables with a high number of missing values, the missing values were treated as a separate
category as available-case analysis could yield to a reduced power and biased estimates.
All statistical tests were two-tailed. P values <0.05 were regarded as statistically significant.
3 Results
3.1 Sample description
Missing values analyses of the excluded n = 1,937 patients were compared to the remaining n =
19,743 and revealed no significant differences regarding gender. Excluded patients were older (M =
66.0 (SD = 12.4) than the final sample, more often had an unknown disease progression status
(47.6% vs. 32.3 %) and showed more distress above the cutoff (42.1% vs. 32.6%). All p values were
< 0.001.
The sample analyzed (n = 19,743) comprised of 54.0% women; mean age was 63.2 years (SD =
12.7). The most frequent cancer diagnoses were female breast cancer (n = 6,863; 34.8%) followed by
prostate cancer (n = 4,603; 23.3%). Nearly one quarter reported disease progression. Most of the
patients were under aftercare treatment and were contacted at home (n = 8,941), followed by patients
in acute care hospitals. Overall 32.6% of the patients showed heightened distress levels. The highest
amount of distressed patients found were in the outpatient departments of university hospitals
(39.0%) (Table 1).
Table 2 shows the mean distress scores of the QSC-R10 items for the total sample (range 0 - 5). The
most distressing problem for the patients is the fear of progression.
3.2 Description of the distress clusters
Cluster analysis revealed five subgroups of patients according to their distress patterns. The labeling
of the five clusters is based on the content of the item, which differed in terms of patient stress level
(blue bars) compared to the total sample (gray bars, see Figure 1). Distributions of sex, cancer sites,
treatment settings and disease progression as well as mean age differ significantly between distress
clusters (Table 1). The five clusters are characterized as follows:
Cluster 1 ―minimally distressed‖: n = 9,198; 46.6% of the total sample; mean QSC-R10 total score: 3.6
(SD = 2.84). These patients have very low levels of all distress items (Figure 1). This cluster consists
of relatively more men, older patients, more prostate cancer and fewer breast cancer patients
This article is protected by copyright. All rights reserved.
Accepted Article

compared to the other clusters (Table 1). Multivariate analysis confirmed significant effects of these
predictors for Cluster 1 (Table 3).
Cluster 2 ―highly distressed‖: n = 2,516; 12.7% of the total sample; mean QSC-R10 total score: 29.8
(SD = 5.55). The patients of this cluster show very high levels in all QSC-R10 items (Figure 1). In this
cluster patients with disease progression, from acute care hospitals or outpatients departments of
university clinics are overrepresented (Table 1). Patients with the following cancer types are also
overrepresented: gynecological, respiratory, upper gastrointestinal, urinary cancer, neuro-oncologic
tumors and ear/neck/throat cancer, hematologic and testicular cancer (Table 3).
Cluster 3 ―mainly physically distressed‖: n = 3,004; 15.2% of the total sample; mean QSC-R10 total
score: 14.8 (SD = 4.37). These patients show a medium average distress level. The highest level of
distress relates to physical problems (Figure 1). This cluster is primarily characterized by patients with
pancreatic, soft tissue, hematologic as well as multiple primary cancers (Table 1); these factors along
with the diagnoses gynecological and neuro-oncological cancers are significant independent
predictors for the classification of a patient into Cluster 3 (Table 3).
Cluster 4 ―mainly psychologically distressed‖: n = 3,072; 15.6% of the total sample; mean QSC-R10
score: 12.9 (SD = 4.01). This group also has a medium average distress level, with peaks in items
relating to psychological problems (e.g. fear of progression, feeling tense or nervous) (Figure 1), and
comprises the highest proportion of women (71 %) and female breast cancer patients (48.6%).
Patients from specialized medical practices are also overrepresented (Table 1). Multivariate logistic
regression confirmed female sex as a strong significant predictor for Cluster 4 independent of the
cancer site (Table 3).
Cluster 5 ―mainly socially distressed‖: n = 1,953; 9.9% of the total sample; mean QSC-R10 score:
17.9 (SD = 4.81). Patients belonging to Cluster 5 have relatively high levels of distress in items
relating to social problems (e.g. missing partner’s empathy, Figure 1); their average distress level is in
the medium range. Patients with endocrine and prostate cancers and patients from rehabilitation
clinics (Table 1) are overrepresented in this cluster. Multivariate analysis confirmed these results
(Table 3).
4 Discussion
In a large sample of 19,743 cancer patients we found that the fear of disease progression is the most
distressing problem. Other important problems are feeling tired and weak, pain and disturbed sleep.
This corresponds with the literature12,16.
The cluster analysis revealed five subgroups showing different distress profiles. Clustering was based
on the ten items of the QSC-R10 distress screening questionnaire. The five distress groups differ not
only in the prevalence of distress but also in distress domains.
This article is protected by copyright. All rights reserved.
Accepted Article

The largest group (46.6%) is not distressed and seems to cope well with their disease. The cited data
within literature varies between 40 and 77%. Bobevsky et al. (2018)17 reported a higher proportion of
54.3% for a subgroup characterized by the absence of distress in a German cancer patient sample.
Factors most strongly associated with affiliation to the minimally distressed cluster when compared to
the rest of the sample, were older age, and male sex and prostate cancer — also reported by
Mehnert et al. (2018)12.
In contrast, 12.7% of our sample patients belong to the highly distressed group. Meggiolaro et al.
(2018)9 reported 9.8% to 16.9% severe caseness on the Distress Thermometer (DT) for a European
cancer patient sample. Our proportion is similar to those of Grassi et al. (2013)18 who found 15% with
severe distress based on DT scores ≥ 8 in an Italian cancer patient sample. In this cluster patients
with disease progress, from acute care hospitals or outpatients departments of university clinics are
overrepresented as are patients with gynecologic, respiratory, upper GI and urinary cancer, neuro-
oncologic tumors and ear/neck/throat cancers. These are patients in a current and invasive medical
treatment situation.
Cormio et al. (2018)19 identified three distinct distress-risk classes and found diagnosis to be the best
differentiating variable. The highest risk of distress was reported for breast, lung, genitourinary or
hematologic cancer patients. This is partially similar to our results. Our data did not confirm Bubis’ et
al. (2018)20 results who found a significantly higher probability of reporting elevated scores in
symptom burden within females and younger patients, in a population-based cohort of newly
diagnosed patients, regardless of cancer sites.
The other three clusters (40.7% of the total sample) show medium average distress levels and differ
in terms of their distress profile, with peaks in the psychological, social or physical field. This
corresponds with George L. Engel`s biopsychosocial model of psychosomatic medicine21,22. The
group characterized by emotional distress symptoms is dominated by female patients with breast
cancer. We believe that this is based on an unspecific correlation between female sex and emotional
distress. The physically distressed patients can be characterized best by their diagnoses: pancreatic,
soft tissue, hematologic and multiple primary cancers. These diagnoses are accompanied by invasive
therapies and / or a poor prognosis. Patients with social distress peaks are especially screened in
rehabilitation clinics.
Our results will help to raise awareness on the existence of three (instead of two) subgroups of
cancer patients. First, the patients who are not distressed and have no need for psychosocial support.
Second, highly distressed patients, who are in urgent need of psychosocial support; and third,
medium distressed patients who need either psychological, social or physical support. We suggest
these results be considered in conjunction with further development of distress screening procedures
e.g. in the form of adaptive testing.
This article is protected by copyright. All rights reserved.
Accepted Article

5 Conclusions
5.1 Study limitations
Our study is a database study which includes 16 subsamples. Since these subsamples have not been
recruited with the purpose of a systematic comparison, the sample sizes and characteristics of the
subsamples vary widely. Refusal to participate in the single studies was not documented. Other
sociodemographic and medical variables — besides sex, age, cancer site, treatment setting and
disease progression — which might have had an impact on the patients’ distress were not
considered. A further limitation of the study is the fact that patients excluded from cluster procedure
show higher QSC-R10 mean scores than the final sample.
5.2 Clinical implications
These results will help raise awareness on three different subgroups of cancer patients who need
psychosocial support. Clinicians should tailor the interventions proposed for distressed patients to
address the specific type of distress identified.
6 Acknowledgements
We thank all the healthcare professionals who contributed to data collection and all the cancer
patients who participated in the screening procedure.
7 Conflict of Interest
The authors have declared no conflicts of interest.
8 Data Availability Statement
The data that support the findings of this study are available on request from the corresponding
author.
This article is protected by copyright. All rights reserved.
Accepted Article

References
1. Carlson LE, Waller A, Mitchell AJ. Screening for distress and unmet needs in patients with
cancer: Review and recommendations. J Clin Oncol 2012; 30:1160-1177.
2. Bultz BD, Carlson LE. Emotional distress: the sixth vital sign-future directions in cancer care.
Psycho-Oncology 2006; 15(2):93-95.
3. NCCN (National Comprehensive Cancer Network) Clinical Practice Guidelines in Oncology.
Distress Management. NCCN guidelines, 2017. [accessed 02.05.2019]; Available from:
https://www.facs.org/~/media/files/quality%20programs/cancer/coc/2016%20coc%20standard
s%20manual_interactive%20pdf.ashx
4. AWMF (Arbeitsgemeinschaft wissenschaftlich medizinischer Fachgesellschaften).
Leitlinienprogramm Onkologie (Deutsche Krebsgesellschaft, Deutsche Krebshilfe, AWMF): S3
Leitlinie für psychoonkologische Diagnostik, Beratung und Behandlung von erwachsenen
Krebspatienten, AWMF-Registernummer: 032/051OL, 2014.
http://www.awmf.org/uploads/tx_szleitlinien/032051OLl_S3_Psychoonkologische_Beratung_B
ehandlung_2014-01.pdf
5. Van't Spijker A, Trijsburg RW, Duivenvoorden HJ. Psychological sequelae of cancer
diagnosis: a meta-analytical review of 58 studies after 1980. Psychosom Med. 1997;
59(3):280-93.
6. Pascoe S, Edelman S, Kidman A. Prevalence of psychological distress and use of support
services by cancer patients at Sydney hospitals. Aust N Z J Psychiatry. 2000; 34(5):785-791.
7. Zabora J, BrintzenhofeSzoc K, Curbow B, Hooker C, Piantadosi S. The prevalence of
psychological distress by cancer site. Psycho-Oncology 2001; 10(1):19-28.
8. Herschbach P, Keller M, Knight L, Brandl T, Huber B, Henrich G, Marten-Mittag B.
Psychological problems of cancer patients: a cancer distress screening with a cancer-specific
questionnaire. Br J Cancer 2004; 91(3):504-511.
9. Meggiolaro E, Berardi MA, Andritsch E, Nanni MG, Sirgo A, Samorì E, Farkas C, Ruffilli F,
Caruso R, Bellé M, Juan Linares E, de Padova S, Grassi L. Cancer patients' emotional
distress, coping styles and perception of doctor‐patient interaction in European cancer
settings. Palliat Support Care 2016; 14(3): 204-11.
10. Marten-Mittag B, Book B, Buchhold B, et al.: The Basic Documentation for Psycho-Oncology
Short Form (PO-Bado SF) - an expert rating scale for distress screening: development and
psychometric properties. Psycho-Oncology 2015; 24: 653–660.
11. Mitchell AJ, Chan M, Bhatti H, et al. Prevalence of depression, anxiety, and adjustment
disorder in oncological, haematological, and palliativecare settings: a meta‐analysis of 94
interview‐based studies. Lancet Oncol. 2011;12(2):160‐174.
12. Mehnert A, Hartung TJ, Friedrich M, Vehling S, Brähler E, Härter M, Keller M, Schulz H,
Wegscheider K, Weis J, Koch U, Faller H. One in two cancer patients is significantly
distressed: Prevalence and indicators of distress. Psycho-Oncology 2018; 27(1):75-82.
This article is protected by copyright. All rights reserved.
Accepted Article

13. World Health Organization. Composite International Diagnostic Interview. Geneva,
Switzerland: World Health Organization, 1990.
14. Roth AJ, Kornblith, AB, Batel-Copel L, Peabody E, Scher HI, Holland JC (1998). Rapid
screening for psychologic distress in men with prostate carcinoma: a pilot study. Cancer 1998;
82(10):1904-8.
15. Book K, Marten-Mittag B, Henrich G, Dinkel A, Scheddel P, Sehlen S, Haimerl W, Schulte T,
Britzelmeir I, Herschbach P. Distress screening in oncology-evaluation of the Questionnaire
on Distress in Cancer Patients-short form (QSC-R10) in a German sample. Psycho-Oncology
2011; 20(3):287-93.
16. Bultz BD, Berman, NJ. Branding distress – as the 6th vital sign: policy implications. In: Holland
J, Breitbard W, Jacobson, PB, Lederberg MS, Loscalzo MJ, McCorkle R, eds. Psycho-
Oncology, Oxford: Oxford University press; 201:663-665.
17. Bovesvski I, Kissane DW, Vehling S, McKenzie DP, Glaesmer H, Mehnert A. Latent class
analysis differentiation of adjustment disorder and demoralization, more severe depressive
and anxiety disorders, and somatic symptoms in patients with cancer. Psycho-Oncology 2018;
27(11):2623-2630.
18. Grassi L, Johansen C, Annunziata MA, Capovilla E, Costantini A, Gritti P, Torta R, Bellani M;
Italian Society of Psycho-Oncology Distress Thermometer Study Group. Screening for distress
in cancer patients: a multicenter, nationwide study in Italy. Cancer 2013; 119(9): 1714-21.
19. Cormio C, Caporale F, Spatuzzi R, Lagattolla F, Lisi A, Graziano G. Psychosocial distress in
oncology: using the distress thermometer for assessing risk classes. Support Care Cancer
2019, Feb 20. doi: 10.1007/s00520-019-04694-4. [Epub ahead of print].
20. Bubis LD, Davis L, Mahar A, Barbera L, Li Q, Moody L, Karanicolas P, Sutradhar R, Coburn
NG. Symptom Burden in the First Year After Cancer Diagnosis: An Analysis of Patient-
Reported Outcomes. J Clin Oncol. 2018; 36(11):1103-11.
21. Engel GL: The need for a new medical model: a challenge for biomedicine. Science 1977;
196: 129–136.
22. Clark K, Bergerot CD, Philip EJ, Buga S, Obenchain R, Loscalzo MJ. Biopsychosocial
problem-related distress in cancer: examining the role of sex and age. Psycho-Oncology
2017; 26(10):1562-1568.
This article is protected by copyright. All rights reserved.
Accepted Article

Table 1: Characteristics total sample and stratified by clusters
Total Sample
Distress Cluster
Frequencies
Distressed
(QSC-R10)
above
Cutoff
Cluster 1:
Minimally
distressed
n = 9,198
Cluster 2:
Highly
distressed
n = 2,516
Cluster 3:
Mainly
physically
distressed
n = 3,004
Cluster 4:
Mainly
psychologic.
distressed
n = 3,072
Cluster 5:
Mainly
socially
distressed
n = 1,953
p †
n
%
%
n
%
n
%
n
%
n
%
n
%
Total sample
19,743
100.0
9,198
46.6
2,516
12.7
3,004
15.2
3,072
15.6
1,953
9.9
Distressed (QSC-R10) n/%
6,431
32.6
0
0.0
2,516
100.0
1,487
49.5
999
32.5
1,429
73.2
<0.001
Sex
Women
10,652
54.0
38.4
4,066
44.2
1,622
64.5
1,675
55.8
2,172
70.7
1,117
54.0
<0.001
Men
9,091
46.0
25.7
5,132
55.8
894
35.5
1,329
44.2
900
29.3
836
46.0
Age (years) M/SD
63.2
12.7
65.5
12.0
60.1
12.5
62.4
13.4
61.1
12.9
60.9
12.3
<0.001‡
Range (years)
11-103§
11-103
15-99
14-95
17-98
18-93
Cancer site
Breast, female
6,863
34.8
37.6
2,660
28.9
1,000
39.7
980
32.6
1,493
48.6
730
37.4
<0.001
Gynecologic
645
3.3
46.0
206
2.2
113
4.5
130
4.3
107
3.5
89
4.6
Respiratory
838
4.2
43.7
293
3.2
175
7.0
195
6.5
97
3.2
78
4.0
Upper GI
476
2.4
37.8
185
2.0
82
3.3
90
3.0
71
2.3
48
2.5
Lower GI
2,052
10.4
28.7
1,023
11.1
202
8.0
352
11.7
298
9.7
177
9.1
Prostate
4,603
23.3
18.1
3,079
33.5
260
10.3
387
12.9
420
13.7
457
23.4
Testicular
40
0.2
52.5
14
0.2
9
0.4
10
0.3
3
0.1
4
0.2
Urinary tract
119
0.6
47.1
43
0.5
29
1.2
20
0.7
11
0.4
16
0.8
Skin
170
0.9
21.2
93
1.0
17
0.7
17
0.6
28
0.9
15
0.8
Hematologic
1,552
7.9
33.1
713
7.8
199
7.9
305
10.2
200
6.5
135
6.9
Soft tissue
537
2.7
37.1
212
2.3
95
3.8
130
4.3
68
2.2
32
1.6
This article is protected by copyright. All rights reserved.
Accepted Article

Endocrine
48
0.2
33.3
28
0.3
3
0.1
3
0.1
6
0.2
8
0.4
Ear/neck/throat
438
2.2
39.7
175
1.9
84
3.3
69
2.3
75
2.4
35
1.8
Breast, male
92
0.5
29.3
45
0.5
11
0.4
18
0.6
10
0.3
8
0.4
Pancreas
433
2.2
39.5
157
1.7
69
2.7
110
3.7
57
1.9
40
2.0
Neuro-oncological
356
1.8
48.9
107
1.2
83
3.3
81
2.7
49
1.6
36
1.8
Multiple primary
402
2.0
39.6
146
1.6
67
2.7
89
3.0
64
2.1
36
1.8
Others
79
0.4
53.2
19
0.2
18
0.7
18
0.6
15
0.5
9
0.5
Setting
Specialized medical
practices
3,194
16.2
31.8
1,499
16.3
386
15.3
388
12.9
580
18.9
341
17.5
<0.001
Outpatient departm.
of university clinics
1,821
9.2
39.0
697
7.6
315
12.5
340
11.3
301
9.8
168
8.6
Acute care hospitals
5,316
26.9
36.7
2,160
23.5
843
33.5
837
27.9
1,005
32.7
471
24.1
Rehabilitation clinics
471
2.4
28.5
261
2.8
39
1.6
41
1.4
58
1.9
72
3.7
After care treatment
8,941
45.3
29.3
4,581
49.8
933
37.1
1,398
46.5
1,128
36.7
901
46.1
Disease progression
Yes
4,567
23.1
39.2
1,868
20.3
737
29.3
835
27.8
628
20.4
499
25.6
<0.001
No
8,798
44.6
29.2
4,392
47.7
975
38.8
1,105
36.8
1,447
47.1
879
45.0
Unknown, missing
6,378
32.3
32.5
2,938
31.9
804
32.0
1,064
35.4
997
32.5
575
29.4
† p values for significance of overall differences between the 5 clusters
if not otherwise stated p values are based on Chi² tests
‡ p value based on one-way ANOVA
§ n = 16 (0.08 %) are under 18 years old
This article is protected by copyright. All rights reserved.
Accepted Article

Table 2: Item distress scores (M = means; SD = standard deviation) of the QSC-R10†
for the total sample
QSC-R10 items
n
M
SD
fear of disease progression
18,853
1.85
1.744
feeling tired and weak
18,829
1.46
1.576
reduced recreation activities
18,760
1.44
1.715
feeling tense and/or nervous
18,871
1.38
1.535
disturbed sleep
18,826
1.38
1.600
feeling physically imperfect
18,743
1.13
1.499
pain
18,863
.96
1.378
missing partner's empathy
18,499
.86
1.387
few opportunities to speak with psychooncol. professionals
18,740
.67
1.251
not feeling well informed about disease/treatment
18,826
.53
1.125
† Item scale 0 - 5
This article is protected by copyright. All rights reserved.
Accepted Article

Table 3: Sociodemographic and clinical factors associated with clusters
Cluster 1*
Minimally
distressed
n = 9,198
Cluster 2*
Highly
distressed
n = 2,516
Cluster 3*
Mainly
physically
distressed
n = 3,004
Cluster 4*
Mainly
psychologic.
distressed
n = 3,072
Cluster 5*
Mainly
socially
distressed
n = 1,953
OR
95 % CI
OR
95 % CI
OR
95 % CI
OR
95 % CI
OR
95 % CI
Age
Per 5 years
1.10
1.08-1.11
0.93
0.91-0.95
0.99
0.98-1.01
0.97
0.95-0.99
0.83
0.89-0.92
Sex
Women
0.75
0.68-0.82
1.15
1.01-1.31
0.80
0.71-0.90
1.88
1.65-2.15
1.11
0.95-1.31
Setting
Specialized medical
practices
1.00
Ref
1.00
Ref
1.00
Ref
1.00
Ref
1.00
Ref
Outpatient
departments of
university clinics
0.62
0.53-0.71
1.84
1.51-2.24
1.53
1.26-1.86
0.92
0.76-1.11
0.90
0.71-1.15
Acute care hospitals
0.60
0.54-0.67
1.64
1.42-1.88
1.46
1.27-1.68
1.12
1.06-1.34
0.82
0.70-0.96
Rehabilitation clinics
0.72
0.57-0.90
1.25
0.85-1.84
0.91
0.63-1.32
0.84
0.61-1.15
1.60
1.16-2.01
aftercare treatment
0.89
0.81-0.97
0.95
0.83-1.09
1.47
1.29-1.67
0.91
0.81-1.03
0.96
0.83-1.11
Disease
progression
Yes
0.60
0.55-0.65
1.73
1.54-1.93
1.38
1.24-1.54
1.01
0.90-1.12
1.19
1.05-1.34
Cancer site
Lower GI
1.00
Ref
1.00
Ref
1.00
Ref
1.00
Ref
1.00
Ref
Breast, female
0.82
0.72-0.92
1.29
1.08-1.56
1.02
0.87-1.20
1.05
0.90-1.23
1.10
0.90-1.35
Gynecologic
0.57
0.47-0.70
1.74
1.33-2.27
1.47
1.16-1.87
0.79
0.62-1.02
1.49
1.11-2.00
Respiratory
0.60
0.50-0.70
2.17
1.74-2.72
1.33
1.09-1.63
0.78
0.61-1.00
1.06
0.80-1.41
Upper GI
0.66
0.54-0.82
1.89
1.42-2.51
1.12
0.86-1.45
0.95
0.71-1.26
1.23
0.87-1.72
Prostate
1.65
1.47-1.85
0.67
0.54-0.81
0.41
0.35-0.48
0.79
0.67-0.94
1.35
1.11-1.64
Testicular
0.72
0.37-1.41
2.21
1.02-4.81
1.59
0.76-3.31
0.45
0.14-1.48
0.92
0.32-2.64
Urinary tract
0.57
0.38-0.86
2.94
1.84-4.71
1.15
0.69-1.92
0.55
0.29-1.06
1.51
0.85-2.67
Skin
1.40
1.00-1.95
0.93
0.54-1.59
0.54
0.32-0.92
0.94
0.60-1.47
1.07
0.60-1.92
Hematologic
0.62
0.53-0.72
2.25
1.77-2.87
1.34
1.09-1.64
0.72
0.58-0.90
1.13
0.86-1.49
Soft tissue
0.84
0.68-1.05
1.71
1.27-2.31
1.58
1.21-2.05
0.61
0.45-0.83
0.62
0.41-0.95
Endocrine
1.79
0.98-3.24
0.58
0.18-1.91
0.34
0.11-1.12
0.50
0.21-1.20
2.31
1.04-5.10
Ear/neck/throat
0.72
0.58-0.90
2.19
1.63-2.96
0.92
0.68-1.24
0.91
0.68-1.23
1.10
0.74-1.63
Breast, male
0.83
0.54-1.27
1.35
0.70-2.59
1.15
0.67-1.96
0.94
0.48-1.84
1.05
0.50-2.21
Pancreas
0.59
0.48-0.74
1.76
1.31-2.38
1.58
1.23-2.03
0.80
0.59-1.09
1.14
0.80-1.65
Neuro-oncological
0.51
0.39-0.65
2.63
1.92-3.58
1.47
1.10-1.98
0.67
0.47-0.94
1.26
0.84-1.88
Multiple primary
0.54
0.43-0.68
1.97
1.45-2.69
1.51
1.15-1.97
0.94
0.70-1.27
1.02
0.71-1.50
† Reference category is the rest of the sample
CI confidence interval; OR odds ratio; Ref reference category
GI gastrointestinal
Bold numbers indicate significant odds ratios
Legend for Figure 1
This article is protected by copyright. All rights reserved.
Accepted Article

Figure 1: Distress patterns based on QSC-R10 items per distress cluster
grey bars: mean item scores of the total sample
blue bars: mean item scores of the respective distress cluster members
This article is protected by copyright. All rights reserved.
Accepted Article

1 2 3 4 5 6 7 8 9 10
0.0
1.0
2.0
3.0
4.0
Cluster�4�(n=3,972)�The�mainly�psychologically�distressed
12345678910
0.0
1.0
2.0
3.0
4.0
Cluster�5�(n=1,953)�The�mainly�socially�distressed
1 2 3 4 5 6 7 8 9 10
0.0
1.0
2.0
3.0
4.0
Cluster�3�(n=3,004)�The�mainly�physically�distressed
12345678910
4.0
3.0
2.0
1.0
0.0
Cluster�2�(n=2,516)�The�highly�distressed
Fig.�1
feeling�ti...
pain
feeling�p...
reduced�r...
fear�of�di...
feeling�te...
disturbed...
missing�...
no�psych...
�not�feeli...
4.0
3.0
2.0
1.0
0.0
Cluster�1�(n=9,198)�The�minimally�distressed
http://mc.manuscriptcentral.com/pon
This article is protected by copyright. All rights reserved.
Accepted Article