A comparison of the Oxford shoulder score and shoulder pain and disability index: factor structure in the context of a large randomized controlled trial

Figures & data

Table 1 Factor extraction results

Method	OSS (n=648)	SPADI (n=628)
Velicer’s MAP test	1	1
Horn’s PA	2	1
PA-MRFA	1	2
Scree test	2	1–2b
Kaiser’s eigenvalue-over-1 rule*	2a	2b

Notes: Suggested numbers of factors to be retained by the OSS and the SPADI, using different methods to assess dimensionality of the data (based on polychoric correlation matrices).

* Tables 2 and 3 for full details of exploratory factor analysis.

a Eigenvalue 1.233 for second factor.

b Scree test appears borderline between 1 and 2 factors. Eigenvalue 1.003 for second factor (see Supplementary materials).

Abbreviations: MAP, minimum average partial; OSS, Oxford shoulder score; PA, parallel analysis; PA-MRFA, PA based on minimum rank factor analysis; SPADI, shoulder pain and disability index.

Table 2 Exploratory factor analysis solution for the OSS

Pattern matrixa
OSS questionnaire item number		Factor
		1	2
6	OssTray1	0.972	−0.213
4	OssKnFk1	0.834	−0.040
9	OssHang1	0.777	0.031
7	OssBrHr1	0.767	−0.024
10	OssWash1	0.721	0.139
5	OssShop1	0.708	0.075
3	OssTran1	0.685	0.062
11	OssWork1	0.493	0.362
2	OssTrDr1	0.456	0.440
1	OssWsPn1	−0.082	0.836
8	OssUsPn1	0.054	0.766
12	OssPnNt1	−0.099	0.706

Notes: Extraction method: Principal axis factoring. Rotation method: Promax with Kaiser normalization. (Items that cross-load on both factors are shown in bold font.)

a Rotation converged in three iterations. Bold text denotes cross-loading.

Abbreviation: OSS, Oxford shoulder score.

Table 3 Exploratory factor analysis solutions for the SPADI

Pattern matrixa
SPADI Questionnaire item number within customary scale	Factor
	1 Pain	2 Disability
PainScale3	0.920	−0.081
PainScale1	0.801	−0.165
PainScale2	0.768	−0.091
DisabilityScale6	0.717	0.144
PainScale5	0.618	0.141
DisabilityScale7	0.531	0.264
PainScale4	0.516	0.322
DisabilityScale2	0.475	0.387
DisabilityScale5	−0.189	0.943
DisabilityScale4	−0.155	0.919
DisabilityScale1	0.255	0.608
DisabilityScale3	0.305	0.553
DisabilityScale8	0.259	0.496

Notes: Extraction method: Principal axis factoring. Rotation method: Promax with Kaiser normalization. Bold text denotes cross-loading.

a Rotation converged in three iterations.

Abbreviation: SPADI, shoulder pain and disability index.

Table 4 Summary of CFA fit measures for one- and two-factor models of the OSS

Model specification	χ^2	df	CFI	SRMR	RMSEA	RMSEA 90% CI
Model 1 All 12 items=single underlying factor	388.16	54	0.973	0.0710	0.080	(0.0706; 0.0892)
Model 2 2 factors: Pain=Items 1, 8, and 12 Function=all other items	277.32	53	0.982	0.0539	0.0726	(0.0633; 0.0823)
Model 3 2 factors: Pain=Items 1, 8,11 and 12 Function=all other items	331.00	53	0.978	0.0589	0.0772	(0.0679; 0.0868)
Model 4 2 factors: Pain=Items 1, 8,12, 11, and 2 Function=all other items	296.44	53	0.981	0.0564	0.0761	(0.0668; 0.0857)

Abbreviations: CFA, confirmatory factor analysis; CFI, comparative fit index; CI, confidence interval; df, degrees of freedom; OSS, Oxford shoulder score; RMSEA, root mean square approximation; SRMR, standard root mean square residual.

Table 5 Summary of CFA fit measures for one- and two-factor models for the SPADI

Model specification	χ^2	df	CFI	SRMR	RMSEA	RMSEA 90% CI
Model 1 All 13 items=single underlying factor	619.02	65	0.970	0.0058	0.0849	(0.0764; 0.0936)
Model 2 2 factors: Pain=Pain items 1–5 Disability=Disability items 1–8 – as originally specified	540.71	64	0.974	0.0500	0.0772	(0.0686; 0.0861)
Model 3 2 factors: Pain=Pain items 1–5 + Disability items 6, 7, and 2 Disability=all other Disability items	533.40	64	0.974	0.0498	0.0841	(0.0755; 0.0928)

Abbreviations: CFA, confirmatory factor analysis; CFI, comparative fit index; CI, confidence interval; df, degrees of freedom; RMSEA, root mean square approximation; SPADI, shoulder pain and disability index; SRMR, standard root mean square residual.

Figure S1 Oxford shoulder score scree test.

Figure S2 Shoulder pain and disability index scree test.