Deriving a preference-based utility measure for cancer patients from the European Organisation for the Research and Treatment of Cancer’s Quality of Life Questionnaire C30: a confirmatory versus exploratory approach

Figures & data

Table 1 The 30 items of the Quality of Life Questionnaire C30 and the scalesa to which they belong

Item	Item stem wording	Scale
1	Do you have any trouble doing strenuous activities, like carrying a heavy shopping bag or a suitcase?	PF
2	Do you have any trouble taking a long walk?	PF
3	Do you have any trouble taking a short walk outside of the house?	PF
4	Do you need to stay in bed or a chair during the day?	PF
5	Do you need help with eating, dressing, washing yourself, or using the toilet?	PF
6	Were you limited in doing either your work or other daily activities?	RF
7	Were you limited in pursuing your hobbies or other leisure time activities?	RF
8	Were you short of breath?	Dyspnea (S)
9	Have you had pain?	Pain
10	Did you need to rest?	Fatigue
11	Have you had trouble sleeping?	Insomnia (S)
12	Have you felt weak?	Fatigue
13	Have you lacked appetite?	Appetite loss (S)
14	Have you felt nauseated?	Nausea/vomiting
15	Have you vomited?	Nausea/vomiting
16	Have you been constipated?	Constipation (S)
17	Have you had diarrhea?	Diarrhea (S)
18	Were you tired?	Fatigue
19	Did pain interfere with your daily activities?	Pain
20	Have you had difficulty in concentrating on things, like reading a newspaper or watching television?	CF
21	Did you feel tense?	EF
22	Did you worry?	EF
23	Did you feel irritable?	EF
24	Did you feel depressed?	EF
25	Have you had difficulty remembering things?	CF
26	Has your physical condition or medical treatment interfered with your family life?	SF
27	Has your physical condition or medical treatment interfered with your social activities?	SF
28	Has your physical condition or medical treatment caused you financial difficulties?	Financial difficulties (S)
29	How would you rate your overall health during the past week?	Global
30	How would you rate your overall quality of life during the past week?	Global

Notes:

a There are five multi-item functioning scales (PF, RF, CF, EF, and SF); three multi-item symptom scales (fatigue, pain, and nausea/vomiting); a global health/quality of life scale; and six single-item scales (S).

Abbreviations: CF, cognitive functioning; EF, emotional functioning; PF, physical functioning; RF, role functioning; SF, social functioning.

Table 2 Summary of item statistics based on the dimensions established using exploratory factor analysis

Item	Factors and loadings (exploratory factor analysis)a			Rasch
	Location	Item fit	Differential item functioningb	Local dependencyc	Factor 1	Factor 2	Factor 3
1	0.65	0.00	0.01	−0.92	0.88	Sex, sited
2	0.75	−0.07	−0.07	−1.23	−1.34		3
3	0.82	−0.06	−0.15	0.96	0.06		2
4	0.65	0.03	0.14	0.14	0.73		10
5	0.52	−0.02	−0.05	3.40	0.95
6	0.86	−0.08	0.02	−0.81	−3.89	Site	7
7	0.77	0.02	−0.02	−0.51	−1.37		6
8	0.15	0.17	0.08	Not included in Rasch analysis (weak factor loadings)
9	0.49	0.06	0.05	Misfit
10	0.53	0.26	0.25	−0.42	0.59		4
11	−0.02	0.34	0.02	Misfit
12	0.25	0.38	0.34	−0.97	−0.37		18
13	0.01	0.12	0.62	−0.05	0.34
14	−0.11	0.03	0.83	0.63	0.28	Site	15
15	−0.11	−0.01	0.77	1.64	−1.73		14
16	0.14	−0.03	0.18	Not included in Rasch analysis (weak factor loadings)
17	0.02	−0.07	0.29	Not included in Rasch analysis (weak factor loadings)
18	0.24	0.36	0.36	−1.26	−0.05		12
19	0.68	0.11	−0.02	−0.62	1.51
20	0.15	0.41	0.29	Misfit
21	−0.01	0.83	−0.10	0.28	0.29	Sex
22	−0.01	0.91	−0.23	−0.28	−1.31	Sex, site
23	−0.02	0.60	−0.01	Misfit
24	−0.05	0.77	0.04	−0.001	1.49
25	0.08	0.27	0.27	Not included in Rasch analysis (weak factor loadings)
26	0.15	0.35	0.09	Misfit
27	0.42	0.24	0.13	Misfit

Notes:

Rasch statistics are those obtained from the final analyses, ie, those with misfitting items removed.

a Principal axis factoring extraction, direct oblimin rotation

b grouping variables exhibiting differential item functioning for the item are listed in this column

c values in this column represent numbers of items with which the item has a residual correlation following Rasch analysis

d cancer sites included prostate, breast, lung, and other.

Table 3 Summary of item statistics based on the dimensions established using CFA

Item	A priori factors, guided by conceptual model	CFA loadings	Rasch
			Location	Item fit	Differential item functioninga	Local dependencyb
1	Physical functioning	0.78	−1.01	0.79	Sex, sitec
2	Physical functioning	0.80	−1.33	−1.64		3
3	Physical functioning	0.79	0.90	−0.13		2
4	Physical functioning	0.58, 0.31d	0.06	0.71		10
5	Physical functioning	0.76	3.39	0.76
6	Role functioning	0.94	−0.90	−3.62	Site	7
7	Role functioning	0.90	−0.59	−0.48		6
9	Pain	0.72	−0.38	1.42
10	Fatigue	0.66, 0.39d	−0.52	0.97		4
12	Fatigue	0.87	0.22	−0.07	Site
14	Nausea and vomiting	0.94	−1.20	−0.72	Site
15	Nausea and vomiting	0.92	1.20	−0.21	Site
18	Fatigue	0.88	−0.22	0.38
19	Pain	0.97	0.38	0.24
20	Cognitive functioning	0.89	−0.10	0.52
21	Emotional functioning	0.89	0.28	0.29	Sex
22	Emotional functioning	0.88	−0.28	−1.31	Sex, site
23	Emotional functioning	0.67	Misfit
24	Emotional functioning	0.86	−0.001	1.49
25	Cognitive functioning	0.62	0.10	1.15
26	Social functioning	0.63	0.323	1.03
27	Social functioning	0.87	−0.323	0.71	Site

Notes: The results are for the refined model, in which loadings for items 4 and 10 on both physical functioning and the covariance between items 2 and 3 were estimated. Rasch statistics are those obtained from the final analyses, ie, those with misfitting items removed.

a Grouping variables exhibiting differential item functioning for the item are listed in this column

b values in this column represent numbers of items with which the item has a residual correlation following Rasch analysis

c cancer sites included prostate, breast, lung, and other

d estimate of loading on the non–a priori factor, ie, fatigue for item 4, physical functioning for item 10.

Abbreviation: CFA, confirmatory factor analysis.

Table 4 Correlations between factors obtained from the confirmatory factor analysis

	PF	RF	EF	SF	CF	Pain	Fatigue
RF	0.90
EF	0.28	0.32
SF	0.60	0.62	0.57
CF	0.44	0.50	0.59	0.65
Pain	0.73	0.78	0.39	0.58	0.43
Fatigue	0.52	0.57	0.59	0.61	0.79	0.54
NV	0.19	0.30	0.25	0.37	0.48	0.33	0.58

Abbreviations: CF, cognitive functioning; EF, emotional functioning; NV, nausea and vomiting; PF, physical functioning; RF, role functioning; SF, social functioning.

Table 5 Summary of the factor-level statistics based on the dimensions established using exploratory (top panel) and confirmatory (bottom panel) factor analyses

Exploratory factor analysis
Factor 1 (1–7, 9, 10, 19, 27)	Initial:
	Item fit =2.44 (poor)
	Person fit =1.04 (good)
	Final (items 9, 27 removed):
	Item fit =1.71 (poor)a
	Person fit =0.92 (good)
Factor 2 (11, 20–24, 26)	Initial:
	Item fit =2.82 (poor)
	Person fit =1.11 (good)
	Final (items 11, 20, 23, 26 removed):
	Item fit =1.41 (good)
	Person fit =1.34 (good)
Factor 3 (12–15, 18)	Initial:
	Item fit =0.85 (good)
	Person fit =0.79 (good)
Confirmatory factor analysis
Factor 1 (1–7, 10)	Initial:
	Item fit =1.60 (poor)a
	Person fit =0.85 (good)
Factor 2 (21–24)	Initial:
	Item fit =2.28 (poor)
	Person fit =1.13 (good)
	Final (item 23 removed):
	Item fit =1.41 (good)
	Person fit =1.34 (good)
Factor 3 (26, 27)	Initial:
	Item fit =0.23 (good)
	Person fit =0.77 (good)
Factor 4 (20, 25)	Initial:
	Item fit =0.44 (good)
	Person fit =1.08 (good)
Factor 5 (9, 19)	Initial:
	Item fit =0.83 (good)
	Person fit =0.97 (good)
Factor 6 (12, 18)	Initial:
	Item fit =0.32 (good)
	Person fit =0.99 (good)
Factor 7 (14, 15)	Initial:
	Item fit =0.36 (good)
	Person fit =0.85 (good)

Notes: Item fit for both item and person represent the fit residual standard deviation, where a value greater than 1.5 is considered poor.

a Although item fit was poor, no individual item exhibited misfit.