Fig. 3

Forest plot of the regression results of analysing the number of HCP visits during the previous 6 months, and the number of ER visits as well as inpatient days during the previous 12 months. The shapes indicate the regression coefficient estimate and the accompanying line indicates the 95% CI. A filled shape indicates a significant association (p < 0.05). The x-axis is in the natural logarithmic scale meaning that 0.08 increase in the dependent variable (e.g. number of comorbidities for the number of HCP visits) means 1.08 actual increase, i.e. 8% increase. This increase is reflected by one unit increase in the independent variable, meaning e.g. one more comorbidity or one more monthly migraine day. Exact values of coefficients, exponentiated coefficients, CIs, and p-values are presented in Additional file 6