Update on the psychometric properties and minimal important difference (MID) thresholds of the FACT-M questionnaire for use in treatment-naïve and previously treated patients with metastatic Merkel cell carcinoma

Table 2 Internal consistency reliability and construct (convergent/divergent, clinical) validity of the FACT-M in the MCC population (n = 172)

FACT-M subscale, summary and MCC-specific scores	# items	Cronbach’s alpha^[a]	McDonald’s coefficient omega	Scale structure		ECOG PS (mean [SD])
FACT-M subscale, summary and MCC-specific scores	# items	Cronbach’s alpha^[a]	McDonald’s coefficient omega	Range of item-subscale correlations^[b]	Convergent^[c]/ divergent^[d] validity (% of items)	0 (n = 106)	1 (N = 66)	p-value^[e]
FACT-G subscales
Physical well-being	7	0.84	0.85	0.53–0.71	100 / 57	24.0 (4.2)	21.3 (5.6)	<.001
Social/Family well-being^[f]	6	0.83	0.85	0.55–0.70	100 / 100	22.5 (5.4)	21.6 (5.7)	0.308
Emotional well-being	6	0.80	0.80	0.36–0.70	83 / 67	18.1 (4.1)	16.8 (5.0)	0.0499
Functional well-being	7	0.89	0.89	0.52–0.77	100 / 71	18.1 (6.7)	15.8 (6.4)	0.028
Melanoma-specific subscales
Melanoma subscale	16	0.84	0.86	0.19–0.76	50 / 38	54.3 (7.5)	50.6 (9.0)	0.004
Melanoma surgery scale	8	0.82	0.83	0.20–0.77	75 / 75	27.0 (5.3)	24.4 (7.1)	0.007
Summary scales
FACT-M TOI	30	0.94	–	–	–	96.4 (16.7)	88.1 (19.5)	0.003
FACT-G Total score	27	0.93	–	–	–	82.8 (16.2)	75.5 (18.1)	0.007
FACT-M Total score	43	0.95	–	–	–	137.0 (22.3)	126.6 (25.8)	0.006
MCC-Specific scores
Physical function score (PF)	6	0.89	0.89	0.61–0.76	100 / 100	17.2 (5.5)	14.3 (5.9)	0.001
Psychological impact score (PI)	6	0.84	0.85	0.55–0.72	100 / 100	18.4 (4.3)	17.0 (5.2)	0.064
MCC summary score	12	0.90	–	–	–	35.5 (8.7)	31.3 (10.0)	0.004

# = number; ECOG PS = Eastern Cooperative Oncology Group performance score (0 = fully active; 1 = restricted in physically strenuous activity); MCC = Merkel cell carcinoma; SD = Standard deviation, TOI = Trial Outcome Index
[a] Recommended threshold α > 0.7; [b] Pearson correlation coefficients; [c] % of items correlated with its own dimension ≥0.4; [d] % of items correlated with its own dimension higher that the correlation with any other dimension; [e] p-value from t-test of score between the two ECOG performance status groups at baseline; [f] convergent and divergent validity were calculated without the ‘satisfaction with sex life’ item due to high item missingness resulting from response to this item being voluntary

ISSN: 1477-7525