Validity and intra-rater reliability of an Android phone application to measure cervical range-of-motion

Table 3 Reliability of the 3DMA (n = 16)

	Phone D1 (Mean ± SD)	Phone D2 (Mean ± SD)	ICC (3,3)	95% CI	Sys bias	Width of 95% LoA	% Error^†	Prop bias^€	Fixed bias^€	SEM	MDC	LOA 2SD (mean diff ±2.1*SDdiff)
Flexion	48.9 ± 7.7	51.9 ± 6.9	0.88	0.54-0.96	N	8.7	9	N	N	3.0	8.91	-14.77 to 8.90
Extension^¥	79.1 ± 9.9	81.6 ± 9.2	0.88	0.67-0.96	N	10.0	6	N	N	3.4	10.1	-12.69 to 7.56
Right Lateral Flexion	41.7 ± 6.7	42.9 ± 6.9	0.94	0.82-0.98	N	6.8	8	N	N	2.3	6.83	-8.01 to 5.51
Left Lateral Flexion	46.7 ± 7.6	47.1 ± 6.6	0.92	0.78-0.97	N	8.0	9	N	N	2.8	8.32	-8.57 to 7.89
Right Rotation	68.8 ± 5.1	72.7 ± 5.9	0.64	-0.010-0.88	N	10.4	7	N	N	3.6	10.69	-14.45 to 6.81
Left Rotation	70.2 ± 6.7	73.4 ± 6.7	0.77	0.32-0.92	N	11.2	8	N	N	3.8	11.29	-14.35 to 8.03

SD = standard deviation; ICC = intra-class coefficients; CI = confidence interval; Sys Bias = systematic bias; LoA = limits of agreement; Prop Bias = proportional bias; N = no; Y = yes; SEM = standard error of measurement; MDC = minimal detectable change; diff = difference.
^¥Based on n = 15 as one subject’s thoracic spine could not be effectively stabilized using the experimental technique utilized.
^†% Error = 0.5*Width of 95% LOA/[(Mean_Phone + Mean_Vicon)/2].
^€Proportional and fixed bias were determined from ordinary least products analysis.

ISSN: 1743-0003