Skip to main content

Table 7 Multiple linear regression with stepwise inclusion of variables (HI superregion serves as reference for other superregions)

From: The global prevalence of headache: an update, with analysis of the influences of methodological factors on prevalence estimates

Model 1: Original data set Headache, N = 181
(5 variables)
Migraine, N = 242
(8 variables)
TTH, N = 109
(4 variables)
H15+, N = 58
(2 variables)
Whole model F = 10.3, p < 0.001, R2 adj: 0.208 F = 13.9, p < 0.001, R2 adj: 0.299 F = 6.9, p < 0.001, R2 adj: 0.181 F = 6.6, p = 0.003, R2 adj: 0.165
Variables with significance β standardised SPCsqr β standardised SPCsqr β standardised SPCsqr β standardised SPCsqr
Year of publication    0.258 0.060     
Number (ln transformed) of participants −0.207 0.040 − 0.146 0.018     
dMig and pMig    0.269 0.064     
dTTH and pTTH        0.321 0.102
Health-care personnel    0.173 0.022     
Sampling method    −0.169 0.022     
Type of screening question −0.307 0.090    −0.285 0.080   
Number of conditions −0.162 0.026       
Mode of engagement        0.281 0.078
SSA superregion −0.148 0.021 −0.173 0.028 −0.232 0.052   
SEEAO superregion −0.171 0.029 −0.136 0.018 −0.294 0.084   
NAME superregion    −0.113 0.011 −0.231 0.051   
Model 2: ln transformed dependent variable and outliers (±1.5 interquartile range) removed Headache, N = 176
(4 variables)
Migraine, N = 238
(7 variables)
TTH, N = 103
(4 variables)
H15+, N = 57
(2 variables)
Whole model F = 11.3 p < 0.001, R2 adj: 0.191 F = 13.6, p < 0.001, R2 adj: 0.271 F = 5.5, p < 0.001, R2adj:0.150 F = 6.5, p = 0.003, R2 adj: 0.164
Variables with significance β standardised SPCsqr β standardised SPCsqr β standardised SPCsqr β standardised SPCsqr
Year of publication    0.259 0.064     
Number (ln transformed) of participants −0.186 0.032 −0.137 0.016    −0.354 0.125
dMig and pMig    0.291 0.078     
dTTH and pTTH 0.166 0.027       
Health-care personnel    0.133 0.013     
Sampling method    −0.169 0.022     
Type of screening question −0.277 0.072    −0.233 0.054   
SEEAO superregion    −0.110 0.012 −0.206 0.058 −0.250 0.062
SSA superregion −0.207 0.042 −0.180 0.031 −0.245 0.042   
NAME superregion      −0.264 0.068   
  1. R2 adj (Adjusted R2): The proportion of variability explained by the whole model (“Goodness of fit”), adjusted for overestimation caused by the number of effects
  2. β-standardised: The direction and strength of its correlation with the dependent prevalence variable
  3. SPCsqr: the proportion of the variability explained by that variable alone