The temporal relation between pain and fatigue in individuals receiving treatment for chronic musculoskeletal pain

Table 4 Random intercepts cross-lagged panel path coefficients

Parameter	β	SE	P-value
Within-person (state-like)
Autoregressive paths
MPQ-PRI₀ → MPQ-PRI₁	0.60	0.1	< 0.001
MPQ-PRI₁ → MPQ-PRI₂	0.65	0.1	< 0.001
FM₀ → FM₁	-0.14	0.3	0.63
FM₁ → FM₂	0.23	0.1	0.08
Correlations
MPQ-PRI₀ – FM₀	0.59	0.1	< 0.001
MPQ-PRI₁ – FM₁	0.19	0.1	0.18
MPQ-PRI₂ – FM₂	0.52	0.1	< 0.001
Cross-lagged paths
MPQ-PRI₀ → FM₁	0.55	0.2	0.02
MPQ-PRI₁ → FM₂	0.36	0.1	0.001
FM₀ → MPQ-PRI₁	0.12	0.2	0.37
FM₁ → MPQ-PRI₂	0.12	0.1	0.18

N = 104
β standardized estimate, FM Fatigue Measure, MPQ-PRI McGill Pain Questionnaire-Pain Rating Index, SE standard error. Subscripts: 0 = pre-treatment, 1 = mid-treatment, 2 = post-treatment
Random intercept cross-lagged panel model (RICLPM) was used to estimate path coefficient. Variance and covariance of random-intercepts for the MPQ-PRI were set to zero because random-intercepts of MPQ-PRI were not significant; there was no effect of between-person (trait-like) for the MPQ-PRI. Model fit indices: comparative fit index (CFI) = 1.00, Tucker-Levis index (TLI) = 1.00, root mean square error of approximation (RMSEA) < 0.001 (90% confidence interval: 0.00–0.15), standardized root-mean-square residual (SRMR) = 0.03

ISSN: 1471-2474