Differences in Anger Expression Between Individuals With and Without Headache After Controlling for Depression and Anxiety


Address all correspondence to Dr. Robert A. Nicholson, Dept. of Community and Family Medicine, St. Louis University School of Medicine, 1402 S. Grand Blvd, St. Louis, MO 63104.


Objective.—To evaluate whether anger and anger expression are different between persons with and without headache after controlling for depression and anxiety.

Background.—Persons with headache may experience more problems with anger and its expression when compared with persons without headache. It is important to establish whether differences exist for persons with and without headache on trait anger and anger expression independent of depression and anxiety. To date, however, this issue has received little attention in the empirical literature. The current study measured trait anger, anger-in, hostility, anxiety, and depression among persons with and without headache, and evaluated whether trait anger and anger-in differentiated groups independent of depression and anxiety.

Methods.—Participants were 422 adults recruited from a larger study within a university setting. Of those, 171 suffered from headache (mean age, 21 years; 81% were female; 69% were white; mean years with pain, 7.53). Another 251 sex-matched individuals (mean age, 21 years; 81% female; 62% white) met criteria for the headache-free group. Participants provided information regarding their headache characteristics and were administered affective trait measures (Trait version of the State-Trait Anxiety Inventory, Brief Symptom Inventory-Depression), trait anger measures (Trait Anger Scale, Cook-Medley Hostility Scale), and a measure of the extent to which individuals hold their anger in.

Results.—Multivariate analysis of variance revealed significant differences between the 2 groups (Wilks λ  =  .86, P  <  .001, η2  =  .14). Step-down analysis revealed that even after controlling for all other variables, those in the headache group had higher levels of anger-in (P  <  .001, η2  =  .08; mean, 18.98 versus 15.68). Trait anger and hostility did not differ between groups after controlling for depression and anxiety. Logistic regression revealed that anger-in contributed most to predicting headache status (P  <  .001; partial r  =  .23).

Conclusions.—The current findings indicate that persons with headache hold their anger in more than those without headache even after controlling for levels of trait anger, depression, and anxiety. However, after controlling for depression and anxiety, individuals no longer differed on trait anger. Also, anger-in was the strongest predictor of headache. The current findings suggest that holding anger in is more common among headache sufferers. Given recent findings regarding the negative effect of holding anger in among persons with pain conditions, this may be an important factor to evaluate when considering psychological/emotional factors affecting headache.