JMIR Publications

ABSTRACT

Background:

Chronic pain is a significant health care problem. To quantify the pain severity in patients and the efficacy of new or current treatments, researchers and clinicians apply tools such as the traditional visual analogue scale (VAS), that lead to inaccurate and subjective interpretation related to the main sensory pain.

Objective:

To validate pain measurements of a neuroscience-based 3D body mobile application called GeoPain.

Methods:

Temporomandibular disorder (TMD) patients were assessed using GeoPain measures in comparison to traditional pain and mood scales, respectively VAS and Positive and Negative Affects (PANAS). Principal Component Analysis (PCA), scatter score analysis, Pearson’s methods and effect size were used to determine the correlation among GeoPain and VAS measures.

Results:

The PCA analysis resulted in two main orthogonal components: PC1 and PC2. PC1 comprises a combination score of all GeoPain measures, which had a high internally consistent, and clustered together in TMD pain. PC2 included VAS and both positive and negative effects (PANAS). All loading coefficient for GeoPain measures in PC1 were above .70, with low loadings for VAS and PANAS. Meanwhile, PC2 were dominated by VAS and PANAS >0.4. Repeated measure analysis revealed a strong correlation between the VAS and mood scores from PANAS over time that might be related with the subjectivity of VAS measure, whereas sensory-discriminative GeoPain measures, not VAS, demonstrated an association between chronicity and TMD pain in locations spread away from the most commonly reported area, or pain epicenter (P=0.01). Analysis using VAS did not detect an association at baseline between TMD pain and chronicity. The long-term reliability (lag > 1 day) was consistently high for P.A.I.N.S. with lag autocorrelations averaging between 0.7 and 0.8, and greater than the autocorrelations for VAS averaging between 0.3 to 0.6. The combination of higher reliability for P.A.I.N.S. and its objectivity displayed by the lack of association with PANAS as compared to VAS indicated that P.A.I.N.S. has a better sensitivity and reliability for measuring treatment effect over time for sensory-discriminative pain. The effect sizes for P.A.I.N.S. were larger compared to VAS, consequently requiring smaller sample sizes to assess the analgesic efficacy of treatment if P.A.I.N.S. is used versus VAS. P.A.I.N.S. effect size was 0.51SD for both facial sides and 0.60SD for the right side versus 0.35SD for VAS. Therefore, the sample size required to detect such effect sizes with 80% power would be n=125/group for VAS, but for P.A.I.N.S. as low as o n=44/group, almost 1/3 of the sample size needed by VAS.

Conclusions:

GeoPain demonstrated precision and reliability as a 3D mobile interface for measuring and analyzing sensory-discriminative aspects of (sub)regional pain regarding its severity and response to treatment, without been influenced by mood variations from patients as noticed in the more traditional VAS.