Table 2.
Summary of study results related to score equivalence.
| Outcome and study (year) | Equivalence examination method and results | |||||
|
|
Score correlation | Score differences | Descriptive | |||
|
|
Method | Results | Method | Results |
|
|
| Studies reporting pain score equivalence | ||||||
|
|
Athale et al (2004) [26] | ICCa | Pain intensity ICC=0.941; pain interference ICC=0.959 | —b | — | — |
|
|
Bandarian-Balooch et al (2017) [27] | — | — | ANOVAc | Mean pain intensity, frequency, duration, medication usage, disability P>.05 of all | — |
|
|
Bishop et al (2010) [29] | ICC | Pain interference ICC=0.965 | — | — | Mean low-back pain interference score difference between method 0.03 (SD 1.43; 95% CI −0.19 to 0.25). Authors predefined acceptable 95% CI was ± 0.5. |
|
|
Byrom et al (2018) [31] | ICC | Pain intensity r=0.87-0.98, 95% CI 0.83-0.99) | — | — | — |
|
|
Castarlanas et al (2015) [22] | Weighted kappa | Pain intensity κ=0.813 | — | — | — |
|
|
Chiu et al (2019) [32] | Pearson correlation | Pain intensity r=0.93-0.96 (P<.001) | — | — | Using Bland-Altman method, an agreement between the data capture techniques shown at 95% CI. |
|
|
Christie et al (2014) [33] | — | — | Paired sample t tests or Wilcoxon Signed Rank Test | Mean, SD, and range of pain intensity P>.46 for all | — |
|
|
Cook et al (2004) [34] | Spearman rho | Pain intensity and interference rho=0.67-084 | — | — | — |
|
|
Cunha Miranda et al (2015) [35] | ICC | Pain intensity and interference ICC=>0.781-0.944 | — | — | — |
|
|
Fanciullo et al (2007) [36] | Spearman rho | Pain intensity rho=−0.72 (P<.001) | — | — | — |
|
|
Gaertner et al (2004) [38] | — | — | t test | Mean pain intensity not significantly different (P value not reported) | — |
|
|
Garcia-Palacios et al (2013) [39] | Pearson correlation | Pain intensity r=0.79 (P<.001) | — | — | — |
|
|
Heiberg et al (2007) [40] | — | — | Wilcoxon’s signed rank test | Mean, SD, and range of pain intensity P>.06 | — |
|
|
Hofstedt et al (2019) [41] | ICC | Pain intensity ICC=0.952 | Paired t test | Mean pain intensity not significantly different (P=.29) | Using Bland-Altman method, an agreement between the data capture techniques shown at 95% CI. |
|
|
Jaatun et al (2014) [42] | — | — | — | — | In 71% (65/92) of cases participants marked the same number of areas and the same anatomical locations on both body map versions, in 20 cases, the markings were relatively similar, and in 7 cases, the markings were dissimilar. |
|
|
Jamison et al (2001) [15] | Pearson correlation | Pain intensity r=0.88, P<.001 | — | — | — |
|
|
Jamison et al (2002) [43] | Pearson correlation | Pain intensity r2>0.999 | — | — | — |
|
|
Jamison et al (2006) [44] | Pearson correlation | Pain intensity r=0.99 (95% CI 0.975-0.996) | — | — | — |
|
|
Jonassaint et al (2015) [45] | ICC | Pain intensity ICC=0.97 (95% CI 0.88-0.99) | — | — | — |
|
|
Kvien et al (2005) [50] | Pearson correlation | Pain intensity r=0.79-0.93 | — | — | — |
|
|
MacKenzie et al (2011) [51] | ICC | Pain intensity and interference ICC=0.95-0.97; 95% CI 0.95-0.98) | — | — | — |
|
|
Marceau et al (2007) [52] | — | — | — | — | Participants reported similar using each data capture methods for pain intensity, pain interference, mood, and helpfulness of medications. |
|
|
Matthews et al (2018) [54] | Pearson correlation and ICC | Pain location pixelated area r=0.93 (P<.001) and ICC=0.966 (P<.001) | t test | Mean pain location pixelated area not significantly different (P=.93) | Using Bland-Altman method, an agreement between the data capture techniques shown at 95% CI. |
|
|
Neudecker et al (2006) [55] | Pearson correlation | Pain intensity r=0.902 (P<.001) | — | — | — |
|
|
Palermo et al (2004) [56] | — | — | t test | Mean pain intensity not significantly different (P value not reported) | — |
|
|
Pawar et al (2017) [57] | ICC | Pain interference ICC=0.994 (95% CI 0.989-0.996) | — | — | — |
|
|
Ritter et al (2004) [58] | — | — | t test, Wilcoxon’s signed rank test and ANCOVAd | Mean pain intensity and pain interference P>.30 | — |
|
|
Saleh et al (2002) [60] | — | — | Test not reported | Mean and SD pain intensity and interference not significantly different (P value not reported) | — |
|
|
Sanchez-Rodrıguez et al (2015) [61] | — | — | — | — | Using Bland-Altman method, an agreement between the data capture techniques shown for the FPS-Re, the VASf, and the CASg at 95% CI. Agreement for the NRSh-11 shown in the 80% CI level. |
|
|
Stinson et al (2012) [63] | — | — | t test | Mean pain intensity P>.09 for younger and older children | — |
|
|
Stinson et al (2015) [7] | Pearson correlation | Pain intensity r=0.49-0.63 (P<.001); pain interference r=0.53-0.65 (P<.001) | — | — | — |
|
|
Stone et al (2003) [65] | — | — | Repeated-measures ANOVA | Mean pain intensity P>.16 | — |
|
|
Sun et al (2015) [66] | Pearson correlation | Pain intensity r=0.87-0.93 | — | — | Using Bland-Altman method, agreement between the data capture techniques shown in the 80% CI level. |
|
|
Symonds et al (2015) [68] | Pearson correlation and ICC | Pain intensity r=0.92 and ICC=0.92; pain interference r=0.97 and ICC=0.97 | — | — | — |
|
|
VanDenKerkhof et al (2003) [70] | — | — | Mann-Whitney test | Median pain intensity not significantly different (P value not reported) | — |
| Wood et al (2011) [21] | Weighted kappa and Spearman rho | Pain intensity κ 0.846 (95% CI 0.79-0.896) and rho=0.911 (P<.001) | — | — | — | |
| Studies reporting pain score nonequivalence | ||||||
|
|
Rolfson et al (2011) [59] | — | — | Mann-Whitney U test | Mean pain intensity P=.02 | — |
| Studies reporting discrepant results | ||||||
|
|
Bedson et al (2019) [28] | Spearman rho | Pain intensity and interference baseline paper-based and first 3 days of electronic reports rho=0.60 −0.79 (P<.006); pain intensity and interference last 3 days of electronic reports and follow-up paper-based rho=0.40 (P<.11)-0.92 (P<.001) | — | — | — |
|
|
Junker et al (2008) [46] | — | — | Paired t test | Mean average and present pain intensity P<.01; mean worst pain P=.68 (null hypothesis was nonequivalence) | — |
|
|
Koho et al (2014) [49] | ICC | Pain-related fear ICC=0.77 (95% CI 0.66-0.85) | Test not reported | Significantly higher mean scores for 2 of 17 scale items using the electronic method (P value not reported) | Using Bland-Altman method, an agreement between the data capture techniques shown at 95% CI. |
|
|
Stinson et al (2008 and 2014) [5,24] | Pearson correlation and ICC | Pain intensity r=0.55-0.76 and ICC=0.52-0.75 (P<.01); pain interference r=0.77-0.84 (P<.01) | — | — | — |
|
|
Stomberg et al (2012) [64] | — | — | Mantel’s test | Mean pain intensity significantly higher in electronic data capture group on 2 of 3 assessment days (P value not reported) | — |
|
|
Suso-Ribera et al (2018) [67] | Pearson correlation | Pain intensity and interference r=0.60-0.81 | Paired sample t tests | Averaged weekly pain interference reports from app significantly lower than verbally or paper-based recalled interference verbal over the week P<.001 | — |
|
|
Wæhrens et al (2015) [72] | ICC | Pain intensity and pain interference ICC=0.76-0.98 (95% CI 0.50-0.99) | — | — | — |
aICC: intraclass correlation coefficient.
bN/A: not applicable.
cANOVA: analysis of variance.
dANCOVA: analysis of covariance.
eFPS-R: Faces Pain Scale-Revised
fVAS: Visual Analog Scale.
gCAS: Color Analogue Scale.
hNRS: Numerical Rating Scale.