«Previous studies found a relationship between the onset of migraine headaches and negative affect (anger, depression, and anxiety). Suppressed anger, ...»
Migraine Headaches and Anger
Scott W. Boyle, Wesley T. Church II, and Edward Byrnes
Previous studies found a relationship between the onset of migraine headaches and negative affect (anger, depression, and anxiety). Suppressed anger, or anger turned inward,
was found to be an aspect of migraine headaches. This study attempted to support previous research on anger-in. The Anger Disorders Scale (ADS-VI-R) was used. This is a
new instrument developed specifically to assess multidimensional aspects of anger, especially in clinical populations. No evidence was found to support significantly different levels for the trait of anger-in, between the control group and the migraine headache groups. Coercion, a form of anger turned outward, was found to be significant between the outpatient general medical migraine headache group and the nonheadache control group. Those in treatment for migraine headaches were found to be no different from control subjects on mean scores on the instrument.
Key words: anger; anger disorder scale; anger-in; coercion; migraine headache; negative affect Introduction Over the years, researchers have attempted to determine the etiological factors for headaches. Early on, psychoanalysts and clinicians reported a number of specific personality traits associated with patients who experienced migraine headaches. Schnarch and Hunter (1979) reviewed the early literature and identified that the predominant personality trait associated with migraine headaches was repressed hostility. In addition to repressed hostility, the authors also reported the following traits: unexpressed anger, rigid superego, intolerance of frustration, Scott W. Boyle, Ph.D., is associate professor in the College of Social Work at the University of Utah.
Wesley T. Church, II, Ph.D., is assistant professor in the School of Social Work at the University of Alabama, and Edward Cahoon Byrnes, Ph.D., is assistant professor in the School of Social Work at Eastern Washington University. Professor Byrnes is also a senior research associate with Glacier Consulting, Inc. in Annapolis, Maryland.
© 2005 Lyceum Books, Inc., Best Practices in Mental Health, Vol. 1, No. 1, Winter 2005 48 Best Practices in Mental Health perfectionism, pregenital fixation, sexual maladjustment, unresolved parental attachment, poor interpersonal relationships, a strong yet overworked drive to achieve, feelings of resentment, rigidity, and inflexibility. Besides repressed hostility and unexpressed anger (often referred to as anger turned inward), a number of the above identified traits (rigid superego, intolerance of frustration, perfectionism, strong achievement drive, feelings of resentment and rigidity/inflexibility) may be regarded as what Ellis (1995) and Dryden (1990) refer to as low frustration tolerance (LFT) which they find leads to feelings of anger.
According to Schnarch and Hunter (1979) the results from the above cited studies had their weaknesses. Their designs had flaws that led to questionable conclusions. More recently, studies have looked closely at both state and trait aspects of anger and hostility using more sound research designs (including control groups, use of psychological instruments, and self-ratings scales), yet each study has its own limitations.
Literature Review Martin and colleagues have studied the relationship between mood and headaches, as well as the triggers (antecedents) for migraine and tension headaches. Martin, Nathan, Milech, and van Keppel (1988), reported on different moods including “anxiety hostility, depression, unsureness, tiredness, and confusion” (p. 353) that correlated with headache intensity on the same day that the headache occurred. However, the intensity was small and no one mood was found to be more significant than others.
Martin, Milech, and Nathan (1993), identified five factors that are antecedents to headaches, based on subjects’ self-reports. Factor one (negative affect) consisted of anxiety, anger, and depression; factor two was visual disturbance (flicker, glare, and eye strain); factor three was somatic disturbance (sexual activity, sneezing, coughing, and pollen); factor four was environmental stress (humidity, high temperature, and the opposite of relaxation); and factor five was consummatory stimuli (alcohol, certain foods, and hunger).
Subjects in the Paulin, Waal-Manning, Simpson, and Knight (1985) study reported that mental stress, along with too much alcohol (for men only), tiredness, and eye strain were causes of headache (non-specified type). Based on test scores, negative emotions (trait anxiety and anger, along with depression) were found to increase linearly with the frequency of headaches.
Martin and Seneviratne (1997) and Martin and Teoh (1999), based on the findings from the Martin, Nathan, Milech, and van Keppel (1988) study, specifically examined the relationship between negative affect and the variables for hunger and visual disturbance as triggers of headaches. In both studies, they concluded that negative affect can precipitate headaches.
However, each of the above studies had its own limitations. Migraine, tension, or a combination of both types of headaches were combined as a group, thus limiting the strength of the findings for migraine-only headaches (Martin, Nathan, Milech, & van Keppel, 1988). While the study by these authors also found that Migraine Headaches and Anger 49 subjects reported negative affect as an antecedent to headaches, the correlation between same-day headache (compared to the day before and the day after) was low and “no particular emotion emerged as being more closely related to headaches than others” (p. 354).
In the study by Martin, Milech, and Nathan (1993), the authors again did not study subjects by type of headache, but instead combined them (migraine, tension, or both) and looked at headaches as being chronic. They also had subjects report on their mood, or negative affect (anxiety, anger, and depression), as it either precipitated or aggravated the headache, thus making it difficult to determine if the negative affect (mood) was a trigger or an aggravator of headaches.
The studies that examined the impact of negative affect, along with either hunger (Martin & Seneviratne, 1997) or visual (Martin & Teoh, 1999) disturbance as antecedent factors, both used a stressor challenge that was similar to and adopted from what had been used by anxiety researchers. This raises the question as to whether or not the stressor actually elicited anger. Finally, Martin and colleagues combined anxiety, anger, and depression together, thus making it unclear as to the specific relationship between any one of these moods as an antecedent of headaches.
Paulin, Waal-Manning, Simpson, and Knight (1985) used a self-report questionnaire to assess for trait anger, along with measures for trait anxiety and depression. The scores on the three instruments were correlated with headache frequency. The researchers reported a significant relationship between headache frequency and scores for trait anger, as well as anxiety and depression in both men and women. The psychometric scores for these traits were particularly higher for those subjects with the most frequent headaches. Subjects were asked what type of headache they thought they had, plus the researchers apparently only asked where the headache was located and if their type of headache had been diagnosed by a physician. Again, it is difficult to determine if the subjects in this study were suffering from migraine, tension, or other type of headache making it difficult to establish a clear relationship between trait anger and migraine headaches.
When comparing severe migraine patients with severe non-migraine headache patients for personality differences, Schnarch and Hunter (1979) identified two significant differences between the groups: fear of expressing anger, and suspicion of other people, both of which were higher for the migraine group. Though the findings are important, the psychological instrument measuring hostility had its weaknesses. Apparently this earlier edition of the instrument “failed to produce a factor structure that supports the subscales of the test” (DiGiuseppe & Tafrate, p.2) and it has since been revised.
Kinder, Curtis, and Kalichman (1992) examined chronic (primarily tension and vascular) headache sufferers for psychopathology and reactive depression, based on profiles from the Minnesota Multiphasic Personality Inventory (MMPI).
Of the 229 subjects studied, four groups were identified. Thirty-eight were found to have reactive depression and 15 with psychopathology. The two remaining groups were comprised of subjects with normal MMPI profiles. All subjects were 50 Best Practices in Mental Health administered instruments measuring anger. Between the normal groups, the reactive depression, and psychopathological groups, scores for the males and females in the latter two groups mirrored each other. Subjects in the psychopathological group appeared to be more distressed with significant higher trait anger, anger expression, and anger-in, plus anxiety and depression. These subjects were described by Kinder et al. as “chronically anxious and angry individuals who particularly have difficulty in the appropriate expression of their angry feelings. They seem to turn their anger inward...” (p. 522). A main weakness in the study is that the types of headaches in the two groups were not differentiated. Results cannot determine if these effects generalize to all types of chronic headaches or for migraine or tension headaches only.
Henryk-Gutt and Rees (1973) examined psychological factors that may contribute etiologically to migraine headaches. Regarding anger, the researchers found the following factors associated with classic migraine headaches when compared with nonheadache controls: significantly higher scores for the factor measuring hostile behavior, which was comprised of assault, indirect hostility, verbal hostility, and irritability. This was found for both men and women. No difference between women with classic and common migraine headaches was found for this factor. Further, the findings from this study suggested that there is “evidence for increased reactivity of the autonomic nervous system in migraine subjects and that this may provide a predisposing factor for the development of migraine attacks” and that “emotional stress can act as a precipitating factor in migraine” headaches (p. 141). As previously cited in the review of the study by Schnarch and Hunter (1979), the instrument used to measure hostility was one that failed to support the subscales of the instrument through factor analysis.
Studies that specifically examine state aspects, a transitory or situationallyrelated response to a stimulus which is likely to vary, as compared to trait aspects of anger, a proneness or disposition to respond a certain way and is relatively unfluctuating (Endler & Okada, 1975), have found significant results. Migraine subjects, when placed in an anger-provoking situation, were observed to have expressed less anger and to react differently physiologically when compared to pain patients and healthy controls (Grothgar & Scholz, 1987). Specifically, the changes noted were not found on the self report of anger using an adjective checklist, an instrument used to identify “changes of actual states” (p. 207), but on a behavioral level measured by raters observing outward behaviors. Further, physiological changes were found with diastolic blood pressure increasing and pulse pressure decreasing. The authors reported that the physiological changes suggested an anger-in pattern.
According to DiGuiseppe and Tafrate (n.d.), “negative affectivity usually includes measures of depression, anxiety and other negative emotions” (p. 1).
Using factor analysis, these authors suggest that anger is a separate construct from that of the more global negative affectivity. Combining anger, anxiety, and depression into one factor, negative affect, does not allow for the specific examination or planning of interventions. Each affect results in a different emotional experience and, therefore, calls for specific assessment and interventions.
Migraine Headaches and Anger 51 Studies have examined the psychological factor of state and trait anger as a contributor to the etiology of migraine headaches. Results have compared migraine headache subjects with: tension headaches, with no headache controls, and with chronic pain. Data was collected and analyzed from: self-reports through daily diaries of headaches, clinical observations of behaviors, self-report of moods, and psychological and physiological measures. However, as noted in the review of literature, the above studies have their limitations and still leave questions about the relationship between migraine headaches and anger. The variable for migraine headache needs to be better controlled as well as the use of a measure of anger that is multidimensional and based on sound theoretical constructs of anger as a psychological trait.
Psychometric instruments measuring anger are relatively few. Until recently, all of them assessed anger as a normally distributed trait. DiGuiseppe and Tafrate developed the sixth revision of the Anger Disorder Scale (ADS-VI-R). The ADS-VIR is a multidimensional instrument that assesses cognitive, emotional, physiological, and contextual components of trait anger utilizing current theoretical constructs. “Based on factor analytic research... the ADS provides a total score and 13 subscale scores” (anger-in, physiological arousal, physical aggression, verbal aggression, rumination, poor self-control, coercion, duration of anger as a problem, episode length, scope of anger, hurt/social rejection, resentment, and suspiciousness) (DiGuiseppe & Tafrate, n.d., p. 6). It is a 74-item questionnaire with responses based on a 1 to 5 Likert scale. A Cronbach Alpha coefficient for the total scale was.95, while the range for each of the 13 subscales was from.73 to.89 (DiGuiseppe & Tafrate, n.d.). Further, the ADS purports to be able to “distinguish anger as a mental health problem” (p.5). This design was specifically conceptualized by the authors in order to provide clinicians with an instrument that had the ability to distinguish between degrees of disturbance in order to better recognize anger problems in need of clinical intervention.