Effect of teleradiology on patient waiting time and service satisfaction in public hospitals, Northwest Ethiopia: a quasi-experimental study

Table 3 The effect of web-based teleradiology intervention on patient waiting time and satisfaction based on rank-biserial correlation effect size, Northwest Ethiopia (N = 836)

Variables	Category	Median (IQR)	Mean Rank	Rank-biserial correlation Effect size (95% CI)
Exposure to web-based teleradiology^a	No (pre-intervention)	43.50 (22.88, 71.63)	593.52	1
Exposure to web-based teleradiology^a	Yes (post-intervention)	4.62 (2.52, 10.53)	244.32	0.84 (0.81, 0.86)*
Exposure to web-based teleradiology^b	No (Pre-intervention)	96 (89, 103)	593.52	1
Exposure to web-based teleradiology^b	Yes (post-intervention)	113 (105, 124)	244.32	−0.65 (−0.69, −0.60)*

- The rank-biserial correlation can be computed as follows:
- \(\:Effect\:size\:\left(rB\right)=\frac{2\left(M1-M2\right)}{n1+n2},\) where M₁= mean rank of the pre-intervention group; M₂= mean rank of the post-intervention group; n₁₌₄₁₇: sample size for pre-intervention group; and n₂₌₄₁₉: sample size for the post-intervention group
- The Rank-biserial correction (rB) value is an effect size where < 0.1 is trivial, 0.1–0.3 is a small effect, 0.3–0.5 is a moderate effect and > 0.5 a large effect [47]
- Positive sign of effect size mean that the mean rank of the pre-intervention group tends to be larger than the post-intervention group
- The negative sign indicates the mean rank of the pre-intervention group is smaller than the post-intervention group
^*Represents for p-value < 0.01; 1- represents the comparison group
^arepresents for patient waiting time
^brepresents for patient service satisfaction

ISSN: 1472-6963