Response to “Statistical Concerns About Acupressure on Pain in Cancer Patients With Bone Metastasis Trial”

Santillan-Garcia and Moran¹ had some criticisms as to the statistics of our article on the effect of acupressure on pain in cancer patients with bone metastasis.² Therefore, the statistical analysis of the article was repeated with expert opinion, and the following explanations are needed.

Because none of the sociodemographic and clinical characteristics showed significant differences between groups, we did not use any further statistical method to control possible confounding factors for Table 3. Additionally, we did not aim to show causality; we just stressed that the correlation between pain scores and some conditions can be affected by level of pain.

The normality of distribution of continuous variables was tested by Shapiro-Wilk test. Student’s t test (normal data) and Mann-Whitney U test (nonnormal data) were used to again compare 2 independent groups. Analysis showed no error. However, it was noticed that P values were misplaced. We revised the original Table 4 based on the correct P values.

Table 4.

(Revised) Comparison of the Relationship Between the VAS Pain Mean Scores of Patients Before and After Acupressure.

Features	Intervention Group, X ± SD	Control Group, X ± SD	P
Pre-acupressure pain	7.6 ± 1.9	8.2 ± 1.7	.216
Post-acupressure pain	6.8 ± 1.9	7.7 ± 2.1	.101
Difference between pre-post	0.7 ± 1.1	0.4 ± 1.1	.556
P	.001	.041

These show no significant difference between pain scores of groups in terms of before and after measurements. So no further statistical analysis is necessary to correct the effect of initial pain scores. According to the results of our study, pain decreased in both groups.