A novel child-centred core palliative care outcome measure for use in clinical practice and research: findings from a multinational validation study

Table 3 Confirmatory factor analyses model fit indices for the models evaluated

Model	n	X (df)	Comparative fit index	Tucker- Lewis index (TLI)	Root mean square error of approximation (RMSEA)	RMSEA -90%CI	Standardized Root Mean Square Residual (SRMR)
Self-report version
Model 1 with all 7 items	221	34.553*(14)	0.933	0.899	0.062	0.048–0.116	0.046
Model 2 minus feeding item (n for items = 6)	221	25.888*(9)	0.941	0.901	0.043	0.052- 0.135	0.044
Proxy-report
Model 1 with 6 items Excludes item on questions answered (n for items = 6)	213	52.733(9)	0.925	0.876	< 0.001	0.113- 0.192	0.056
Model 2 Excludes feeding and questions answered items (n of items = 5)	213	39.950(5)	0.936	0.871	< 0.001	0.132–0.235	0.050
Assessing the self and proxy report versions for invariance by type of respondent
Configural	434	63.218*(10)	0.933	0.866	0.157	0.121- 0.195	0.048
Metric	Model does not converge due to item 4 invariance item
Scalar	434	146.091(33)	0.858	0.914	0.126	0.105- 0.147	0.065
Metric invariance model – allowing item 4 to vary
Metric invariance model – allowing item 4 to vary	434	63.218(10)	0.933	0.866	0.157	0.121- 0.195	0.048
Family version – 5 items	215	15.892(5)	0.862	0.724	0.057	0.048 -0.158	0.044
Family version – 4 items Less item 12	215	12.312(6)	0.918	0.864	0.234	0.000 -0.126	0.045

Df degrees of freedom, CFI comparative fit index, TLI Tucker–Lewis index, RMSEA root mean square error of approximation, 95% CI 95% confidence intervals

ISSN: 1477-7525