Overall, 295 cancer survivors completed the questionnaires and were further analysed. Sociodemographic, clinical, behavioural, and psychological characteristics of the participants are shown in Table 1.
Table 1. Sociodemographic, clinical and psychological characteristics of the sample
Variable
|
n
|
%
|
Age (M = 57.3; SD = 12.6, range = 28-87)
|
Gender
Male
Female
|
62
233
|
21
79
|
Marital status (n, %)
Single
Married
Partnered
Divorced
Widowed
|
27
167
47
16
35
|
9.2
57.2
16.1
5.5
12.00
|
Educational status (n, %)
Primary education
Lower/preparatory vocational education
General secondary education
Higher general secondary education
Post-graduate education
|
16
69
84
96
23
|
5.4
3.4
28.5
32.5
7.8
|
Employment status (n, %)
Unemployed
Full-time employed
Half-time employed
Retired
|
14
94
49
122
|
4.8
32.0
16.7
41.5
|
Cancer type
Breast cancer
Colon cancer
Lymphoma
Melanom
Others
|
148
18
19
9
93
|
51.0
6.2
6.6
3.1
32.1
|
Type of primary treatment
Surgery (S)
S + Chemotherapy (CT)
S + Radiotherapy (RT)
S + CT + RT
|
65
45
45
102
|
22.3
15.5
15.5
35.1
|
Time since the end of primary treatment in years (M = 6.7; SD = 12.6; range = 0-33)
|
Psychological distress
Symptoms of anxiety (M = 6.09; SD = 3.8)
Symptoms of depression (M = 6.23; SD = 3.6)
|
Health-related quality of life
EQ-5D-total (M = 7.85; SD = 3.0)
|
Abbreviations. EQ-5D-total – The EuroQol five-dimension questionnaire total score.
The majority of our sample were female (79.0%) with a mean (SD) age of 57.3 years. Over half of the participants were diagnosed with breast cancer (51.0%) and were at the time of the assessment on average 6.7 (SD = 12.6) years after the end of the treatment.
Factor Analyses
Items 13-17 and item 35 were not included in the EFA due to not meeting the distributional assumptions, endorsed by less than 10.0% of the participants or loading less than 0.30, following the original study [12]. However, they were all included in the final version of the questionnaire in the total needs score, as they were recognized as clinically and theoretically relevant.
After EFA, one item (Item 27) in Existential Survivorship, one (Item 11) in Comprehensive Care, and one (Item 23) in Relationships fell into a different category than reported in the original study [12]. The category from the original CaSUN named “Quality of Life domain” is in the CaSUN-SL re-named as “Psychological and Emotional Support”, as it addresses the newly included factors much better. Item 10 (“Reduce stress”) loaded high on both domains, the Comprehensive Care domain (0.509) and the Psychological and Emotional Support domain (0.436), but was decided to be included in the latter as it fits better theoretically. The total variance explained in the study was 68.4%. Results on factor loadings compared to the original study and the final CaSUN-SL model are available in Table 2.
The goodness-of-fit of the obtained model was not firmly confirmed. The model (Chi square to df = 1.92, p < 0.001) showed values of CFI (0.890) and TLI (0.876) that are smaller than the recommended values of SRMR (0.075) and RMSEA (0.082) that meet the criteria for model fit. However, when applying less strict guidelines (i.e. TLI and CFI values close to 0.90 indicate a good model fit) [52], the criteria would be met. Nevertheless, the results should be taken with caution. Standardized path coefficients are shown in Figure 1. A significant correlation among five CaSUN-SL domains exist, ranging from 0.42-0.86.
Table 2. Results of factor analysis of CaSUN-SL and its categorization
CaSUN
|
CaSUN-SL
|
Scale
|
Item no. and description
|
Factor 1: ES
|
Factor 2: CC
|
Factor 3: PES
|
Factor 4: RE
|
Factor 5: IN
|
ES
|
32. Survivor expectations
|
0.877
|
|
|
|
|
ES
|
31. Acknowledging the impact
|
0.850
|
|
|
|
|
ES
|
25. Handle social/work situations
|
0.649
|
|
|
|
|
ES
|
26. Changes to my body
|
0.569
|
|
|
|
|
ES
|
34. Spiritual beliefs
|
0.513
|
|
|
|
|
RE
|
27. Problems with sex life
|
0.466
|
|
|
|
|
ES
|
24. Talk to others
|
0.302
|
|
|
|
|
CC
|
6. Manage health with teams
|
|
0.940
|
|
|
|
CC
|
8. Complaints addressed
|
|
0.741
|
|
|
|
CC
|
7. Doctors talk to each other
|
|
0.698
|
|
|
|
CC
|
5. Local health care services
|
|
0.688
|
|
|
|
CC
|
4. Best medical care
|
|
0.619
|
|
|
|
CC
|
9. Complimentary therapy
|
|
0.588
|
|
|
|
QL
|
11. Manage side effects
|
|
0.301
|
|
|
|
ES
|
29. Move on with my life
|
|
|
0.803
|
|
|
ES
|
30. Changes to beliefs
|
|
|
0.661
|
|
|
ES
|
19. Recurrence concerns
|
|
|
0.616
|
|
|
ES
|
20. Emotional support
|
|
|
0.567
|
|
|
ES
|
33. Decision about my life
|
0.310
|
|
0.442
|
|
|
ES
|
10. Reduce stress
|
|
0.509
|
0.436
|
|
|
QL
|
12. Changes to quality of life
|
|
|
0.340
|
|
|
RE
|
22. Impact on my relationship
|
|
|
|
0.720
|
|
RE
|
21. Support partner/family
|
|
|
|
0.660
|
|
ES
|
23. New relationships
|
|
|
0.412
|
0.354
|
|
IN
|
1. Up to date information
|
|
|
|
|
0.681
|
IN
|
3. Understandable information
|
|
|
|
|
0.641
|
IN
|
2. Information for others
|
|
|
|
|
0.626
|
|
Eigen values
|
12.861
|
3.171
|
1.146
|
1.084
|
0.879
|
|
% of variance
|
45.931
|
11.324
|
4.091
|
3.871
|
3.140
|
|
Cumulative %
|
45.931
|
57.255
|
61.347
|
65.218
|
68.358
|
Notes. Maximum Likelihood Method, Oblimin with Kaiser Normalization in 7 iterations. With 5 factors and factor loading >.30.
Abbreviations. ES = Existential Survivorship. QL = Quality of Life. CC = Comprehensive Cancer Care. RE = Relations. IN = Information. PES = Psychological and Emotional Support.
Internal consistency and test-retest reliability
Internal consistency of the CaSUN-SL total score showed an excellent reliability with a Cronbach’s α of 0.94. Internal consistency of the five domains is shown in Table 3, ranging from 0.71-0.88.
Test-retest reliability was performed in a group of 31 (retention rate at the re-test assessment: 68.8%) participants with an average of 7-10 days after the first administration of the questionnaires. Correlations between time 1 and time 2 (Pearson's correlation test) between every item (34 items) showed moderate-high stability over time (r = 0.54-0.97; p < 0.01). ICC values ranged from 0.71 (p < 0.001) to 0.98 (p < 0.001). Test-retest reliability for the 34 items showed good to excellent values, for met (r = 0.651, p < 0.001; ICC = 0.79, p < 0.001), unmet (r = 0.91, p < 0.001; ICC = 0.95, p < 0.001), total (r = 0.91, p < 0.001; ICC = 0.95, p < 0.001), and strength of the need (r = 0.88, p < 0.001; ICC = 0.93, p < 0.001).
Table 3. Internal consistency of the CaSUN-SL scale and its subscales
Scale/subscale
|
Cronbach's alpha value
|
CaSUN-SL (34 items)
|
0.940
|
Existential survivorship
|
0.862
|
Comprehensive Care
|
0.821
|
Psychological and emotional support
|
0.877
|
Relationship
|
0.748
|
Information
|
0.705
|
Abbreviations. CaSUN-SL, The Slovenian version of the Cancer Survivors’ Unmet Needs
Construct validity
Table 4 shows the correlations between met, unmet and total needs of the CaSUN-SL and selected variables. Construct validity showed that age was significantly correlated with met (r = 0.13, p < 0.05), unmet (r = 0.29, p < 0.001) and total needs (0.37, p < 0.001), indicating that younger patients have more met and unmet needs. Significant correlations were found between unmet needs and gender (r = 0.14, p < 0.05), indicating that female patients have more unmet needs; type of cancer (r = -0.19, p < 0.001), indicating that breast and lymphoma cancer survivors have the highest unmet needs; time since treatment (r = -0.20, p < 0.001), indicating that with the less time that has passed since the end of the primary treatment, the more unmet needs are observed; and stage of cancer (r = 0.20, p < 0.001), indicating that patients who were diagnosed with more advanced cancer reported more unmet needs.
The subscales HADS anxiety and HADS depression, and the EQ-5D-total were all positively correlated with the unmet and total needs (see Table 4), indicating that cancer patients with higher unmet and total needs have more symptoms of anxiety, depression, and more problems on the quality-related components (i.e., mobility, self-care, usual activities, pain/discomfort, and anxiety/depression). On the contrary, EQ-5D subscale EQ-VAS and RS-14 were found to be negatively correlated to the unmet and total needs (Table 4), indicating that cancer patients with higher unmet and total needs experience lower self-perceived health and resilience.
Table 4. Construct validity between CaSUN-SL and selected variables
Variable
|
Met needs
|
Unmet needs
|
Total needs
|
Age
|
-0.13*
|
-0.29**
|
-0.37**
|
Gender
|
0.10
|
0.14*
|
-0.04
|
Marital status
|
0.03
|
-0.91
|
-0.06
|
Employment
|
0.03
|
-0.26**
|
-0.22**
|
Type of cancer
|
0.07
|
-0.19**
|
-0.13*
|
Stage of cancer
|
0.08
|
0.20**
|
0.25**
|
Time since treatment
|
-0.13
|
-0.20**
|
-0.27**
|
Anxiety (HADS-A)
|
0.07
|
0.49**
|
0.49**
|
Depression (HADS-D)
|
0.05
|
0.44**
|
0.43**
|
Quality of life (EQ-5D-total)
|
0.05
|
0.36**
|
0.36**
|
Quality of life (EQ-VAS)
|
-0.03
|
-0.36**
|
-0.25**
|
Resilience (RS-14)
|
0.02
|
-0.47**
|
-0.41**
|
Notes. The CaSUN-SL included 34 items that were retained after the cultural adaptation of the translated Slovenian version.
* p ≤ 0.05; **p ≤ 0.001