ISSN 0017-8748 doi: 10.1111/head.12940 Published by Wiley Periodicals, Inc.

Headache C 2016 American Headache Society V

Research Submissions The Headache Triggers Sensitivity and Avoidance Questionnaire: Establishing the Psychometric Properties of the Questionnaire Sonja U. Kubik, BPsych(Hons); Paul R. Martin, PhD

Objective.—The purpose of this study was to evaluate the psychometric properties of the newly developed Headache Triggers Sensitivity and Avoidance Questionnaire (HTSAQ) designed to measure variables relating to the Trigger Avoidance Model of Headaches. Background.—The Trigger Avoidance Model of Headaches proposes that one pathway to developing a primary headache disorder is through the avoidance of headache triggers, resulting in an increase in trigger potency through sensitization. Conversely, prolonged exposure to certain triggers may reduce a potential trigger’s ability to precipitate a headache. This model has led to an alternative approach to trigger management being proposed called Learning to Cope with Triggers, which advocates the desensitization of certain headache triggers through controlled exposure, while supporting avoidance of those triggers that are detrimental to health and wellbeing. To be able to evaluate predictions based on the Trigger Avoidance Model of Headaches and to investigate the effectiveness of Learning to Cope with Triggers, psychometrically sound measures of trigger sensitivity and trigger avoidance are needed. Method.—A convenience sample of 376 participants (87.5% female; 64.1% European descent) was recruited for the online study which consisted of a battery of measures including the HTSAQ, Brief Headache Screen, Pain Anxiety Symptoms Scale, Pain Sensitivity Questionnaire, Chronic Pain Acceptance Questionnaire, and demographic items. With an interval of 2-6 weeks, a subsample of participants (n 5 201) completed the HTSAQ for a second time. The mean age of the sample was 31.4 years (SD 5 12.8). The most common headache diagnoses (based on the Brief Headache Screen) were migraine (chronic 5 62; episodic 5 160), followed by medication-overuse headache (n 5 87). The remaining participants (n 5 67) were diagnosed as having episodic less severe headache (most likely tension-type headache). Results.—Reliability was assessed through internal consistency and test-retest reliability over a period of 2-6 weeks, and both were excellent (alpha > .80). Strong construct validity was demonstrated by the measure’s scale scores being significantly correlated in theoretically consistent ways with the Pain Sensitivity Questionnaire, Pain Anxiety Symptoms Scale, and Chronic Pain Acceptance Questionnaire. As the Trigger Avoidance Model of Headaches would predict, correlations between the HTSAQ Sensitivity scales and the Avoidance scale were strong (P < .001), and participants with chronic migraine had significantly higher HTSAQ scores than those with episodic headaches. Conclusion.—The findings support the use of the HTSAQ as a valid and reliable tool for assessing sensitivity to headache triggers and avoidance of headache triggers. Further research examining the factor structure of the HTSAQ is warranted. Key words: migraine, tension-type headache, triggers, avoidance, sensitivity, assessment Abbreviations: HTSAQ headache triggers sensitivity and avoidance questionnaire, TTH tension-type headache, TAMH trigger avoidance model of headaches, LCT learning to cope with triggers, S(O) sensitivity to triggers From the School of Applied Psychology, Griffith University, Brisbane, Australia (S. U. Kubik); School of Applied Psychology and Menzies Health Institute Queensland, Griffith University, Brisbane, Australia (P. R. Martin). Address all correspondence to P. R. Martin OAM, School of Applied Psychology & Menzies Health Institute Queensland, Mt Gravatt Campus, Griffith University, 176 Messines Ridge Road, Mt Gravatt, Queensland 4122, Australia, email: sonja.kubik@ griffithuni.edu.au Accepted for publication August 9, 2016.

1

2

Month 2016

compared with others, S(T) sensitivity to triggers compared with time of least sensitivity, BHS brief headache screen, PASS-20 pain anxiety symptom scale, brief version, PSQ pain sensitivity questionnaire, CPAQ chronic pain acceptance questionnaire, CM chronic migraine, EM episodic migraine, EOH episodic other headache, ANOVA analysis of variance (Headache 2016;00:00-00)

Research into the factors believed to precipitate headaches has revealed that the most commonly reported triggers are: stress and negative emotions; sensory triggers (flicker, glare, eyestrain, noise, odors); hunger; too much or not enough sleep; certain food and drinks; alcohol; hormonal factors relating to menstruation; and weather (cold, heat, high humidity).1-7 For many decades, the standard advice for the management of headaches has been to identify and avoid any potential triggers that precipitate a sufferer’s headaches.8-10 It is not always possible, however, to avoid certain triggers (eg, hormonal fluctuations or the weather). The process of attempting to avoid all triggers can in itself also cause a person stress,11 which has been identified as the most common trigger for migraine and tension-type headache (TTH), for both genders.12 There is also a consensus within a variety of psychological literatures which views avoidant coping strategies to be maladaptive, especially in relation to stress, chronic pain, and anxiety. Research into individual differences regarding coping with stress typically indicates that an individual’s coping mechanisms generally take one of two routes approach or avoidance.13 When a person rigidly relies on avoidant coping without regard to situational appropriateness, it can lead to higher levels of general psychopathology and a lower quality of life.14,15 A similar trend also exists in the chronic pain literature which purports that an individual’s fear of pain (or pain anxiety) leads to individual differences regarding pain response on a continuum spanning the extremes of confrontation and avoidance.16-19 It has been argued that individuals who confront their pain are more likely to resume an increasing range of Conflict of Interest: None.

physical and social activities as the organic basis for the pain resolves, resulting in minimal psychological overlay. By contrast, those who are high on avoidance behaviors find themselves in a selfperpetuating vicious cycle of continued fear of pain and avoidance behavior that maintains and exacerbates pain perception, leading to chronic pain and related disability.17 Along with fear of pain, anxiety sensitivity (fear of benign arousalrelated bodily sensations, driven by beliefs about their harmful effects) is now also becoming established as a critical factor in pain-related functioning, as it is believed to be a vulnerability factor for the engagement in fear and avoidance behaviors.19 Finally, the maintenance of fears and phobias is demonstrated in the anxiety literature to often result from short exposure to anxietyeliciting stimuli, which produces a subsequent increased anxiety response to those stimuli. In contrast, prolonged exposure results in decreased subsequent anxiety responses and extinction of the fear.20 In line with these findings and with great success, exposure-based approaches have been implemented for the treatment of a wide range of anxiety disorders.21 Based on this cognate literature, an increasing amount of empirical research has begun to emerge supporting a Trigger Avoidance Model of Headaches (TAMH), whereby the avoidance of a headache trigger may actually increase a trigger’s potency through sensitization, while in contrast, prolonged exposure may weaken a trigger’s ability to precipitate a headache.22-27 Findings from laboratory studies investigating the effects of exposure to noise,25,28 visual disturbance,23,24 and stress26 as triggers for headaches, have all been congruent with the postulates of the TAMH.

Headache One pathway to developing a chronic headache disorder can therefore be conceptualized as a person’s sensitization to headache triggers such that they precipitate headaches more readily.26,27 On the basis of this pathway, an alternative approach to trigger management has been developed called Learning to Cope with Triggers (LCT). This approach advocates the desensitization of certain headache triggers through controlled exposure, while proposing continued avoidance of those triggers that are detrimental to health and well-being.22,27,29 The philosophy is based on the belief that no single strategy can be used for the management of headache triggers and advocates clinical judgment when deciding if avoidance or an exposure strategy is the most adaptive approach for each individual case and corresponding triggers. Since its inception, LCT has been gaining increased support in the field of migraine and headache research. Three recent reviews developing the argument22,27,29 have been cited 162 times, with a number of authors now advocating “coping” with triggers as opposed to the avoidance of all triggers.30-32 Recently published results from a randomized controlled trial demonstrated that LCT was associated with approximately 3 times as large a decrease in headaches as the traditional advice to avoid triggers.33

THE CURRENT STUDY While there is evidence for the efficacy of LCT, there is no evidence relating to how it works. For example, does it result in some triggers being avoided and some being approached as per the management plan, and does approach/exposure result in reduced trigger sensitivity as predicted by the TAMH? For these questions to be addressed a questionnaire is needed that assesses avoidance of triggers and sensitivity to triggers. We had previously developed a questionnaire entitled the Headache Triggers Avoidance Questionnaire (HTAQ), which included items on 24 triggers, and for which respondents had to indicate whether a potential trigger could precipitate headaches for them, and whether they tried to avoid the trigger. Triggers

3 were selected for inclusion in the HTAQ on the basis of 5 studies of the triggers of migraine and tension-type headache.1-5 A table showing the percentage of patients reporting triggers in these 5 studies is presented in Martin and MacLeod.22 We carried out a study to assess the psychometric properties of the HTAQ and found that it had good internal consistency (Cronbach alpha of .81), and good test-retest reliability over 3 to 4 weeks (r 5 .90). There was also support for the construct validity of the HTAQ shown via significant correlations with related scales such as the Anxiety Sensitivity Index and Pain Anxiety Symptoms Scale. The HTAQ and these findings have not been published, however. Testing LCT and the TAMH also requires assessment of sensitivity to triggers. Two variations of sensitivity measures are needed to do this. Firstly, how sensitive are respondents to the triggers compared with other people, and secondly, has their sensitivity changed across time? Hence, the HTAQ was expanded into the Headache Triggers Sensitivity and Avoidance Questionnaire (HTSAQ) by using the same triggers and same avoidance questions, and adding questions about sensitivity to triggers. The HTSAQ lists 24 commonly identified headache triggers, with an option to add 2 not listed triggers. Using a 5-point Likert-scale, respondents are asked to rate for each trigger: (a) is the listed trigger a trigger for the respondent’s headaches; (b) how sensitive is the respondent to the trigger compared with other people; (c) how sensitive is the respondent to the trigger compared with their time of least sensitivity; and (d) does the respondent try to avoid the trigger? Four prescribed scale scores can subsequently be generated corresponding to the above 4 questions: a total Triggers score, a Sensitivity to triggers compared with Others S(O) score, a Sensitivity to triggers compared with Time of least sensitivity S(T) score; and an Avoidance score. Each subscale of the HTSAQ was designed to assess a component of the TAMH. Two sensitivity responses are included because research in this field requires a measurement of comparative trigger sensitivity compared with others (eg, does sensitivity reduce with the use

4

Month 2016

Table 1.—Demographic Characteristics of the Sample Including Follow-Up Survey Completers and Non-Completers and Number of Respondents Endorsing Excluded Items

Follow-Up Survey

TAMH Sample (n 5 289)

Characteristic

Gender Female Male Ethnic Background Aboriginal and TSI African Asian or Pacific Islander European Middle Eastern Other Marital Status Single/Never Married Married Separated/Widowed/Divorced Education Level Completed Less than Year 12 12th Grade College/Vocational training Bachelor or higher Employment Status Full-Time Part-Time Unemployed Headache Type Migraine Chronic Episodic Episodic Less Severe Headache Medication-Overuse Endorsed Menstruation as a trigger Endorsed Smoking as a trigger Excluded from analyses Medication-Overuse

Completers (n 5 201)

Non-Completers (n 5 175)

N

%

N

%

N

%

249 40

86.2 13.8

173 28

86.1 13.9

56 19

89.1 10.9

9 4 21 185 11 59

3.1 1.4 7.2 64.0 3.8 20.4

3 3 15 137 2 41

1.5 1.5 7.5 68.2 1.0 20.3

7 2 9 104 10 43

4.0 1.1 5.1 59.4 5.7 24.6

157 118 14

54.3 40.8 4.9

89 94 18

44.2 46.8 9.0

95 66 14

54.3 37.7 8.0

15 112 54 108

5.2 38.8 18.7 37.3

10 61 42 88

5.0 30.3 20.9 44.9

11 87 28 49

6.1 49.7 16.0 28.2

81 115 93

28.0 39.8 32.8

54 75 72

26.9 37.3 35.8

50 68 57

28.6 38.9 32.6

222 62 160 67 – 129 1

59.0 16.6 42.6 17.8 – 51.8 0.3

114 40 74 36 51 91 1

56.7 19.9 36.8 17.9 25.5 52.6 0.5

108 22 86 31 36 85 2

61.7 12.6 49.1 17.7 20.6 54.5 1.1

87

23.1

–

–

–

–

TAMH, trigger avoidance model of headaches; TSI, Torres Straight Islander.

of exposure strategies), as well as a measurement of change in trigger sensitivity over time, with the latter being important for testing the TAMH (eg, is it the case that avoidance of triggers is associated with increased sensitivity, as predicted?). The study aimed to evaluate the internal consistency, testretest reliability, and construct validity of the HTSAQ.

METHOD Participants and Procedure.—Ethical clearance to conduct this web-based cross-sectional study was granted by the Human Research Ethics Committee of Griffith University. Data collection was completed between 5 June and 15 September, 2015. A convenience sample of 395 participants was recruited for the study. Current and previous Griffith

Headache University alumni were recruited via the Volunteer for Important Research Projects email and through the Psychology first year subject pool. The remaining participants were recruited via web-based advertisements on Australian health-related organization websites. The sample size was determined by the maximum number of participants able to be recruited during the fixed time period. Inclusion criteria required participants to be aged 18-75 years and suffer from at least one headache day per year. Based on these criteria, 19 participants’ responses were excluded, leaving 376 participants for analysis (female, 329; male, 47). The Brief Headache Screen (BHS) was used as a diagnostic tool for classifying participants’ headaches. Any participant classified as suffering from medication-overuse headache (n 5 87), using the BHS, was not considered appropriate to be included in analyses related to the TAMH, which pertains only to primary headaches. It was, however, deemed appropriate to include these participants in the analyses of test-retest reliability. The mean age of the sample including medication-overuse sufferers was 31.4 years (SD 5 12.81). Characteristics of the sample for those who completed the follow-up survey and those who did not are presented in Table 1. The study required the completion of an online survey consisting of a battery of measures. Participants provided consent by “opting-in” via a web link after reading the informed consent procedure. Participants were also given the option to express consent to be contacted for a follow-up survey. A second web link was then sent to their nominated email address 14 days later and remained active for 30 days. Totaling 3 emails, follow-up survey participants were sent a reminder email every 7 days to encourage participation. The follow-up survey was comprised of the Brief Headache Screen (BHS) and HTSAQ, and item presentation matched the original survey. As an incentive for participation, participants were advised that upon completion of both surveys, they would have the opportunity to enter a prize draw for the chance to win a $50 shopping voucher. Griffith University first year student pool participants (n 5 112) received course credit for their participation.

5

MEASURES Headache Triggers Sensitivity and Avoidance Questionnaire (HTSAQ).—The HTSAQ has 26 items and 4 scales. Each item requires a response on a 5-point Likert-scale for each scale via an A, B, C, and D response. For the Triggers scale (A questions) 1 5 Never and 5 5 Always; for the S(O) scale (B questions) 1 5 Not at all sensitive; and 5 5 Very highly sensitive; for the S(T) scale (C questions) 1 5 Same and 5 5 Very much more sensitive; and finally, on the Avoidance scale (D questions) 1 5 Do not try at all to avoid and 5 5 Try to avoid at all costs. Each of the 4 scales can yield a generalized trigger score relating to all 26 possible triggers, as well as a specific trigger score relating to those triggers respondents have identified as a trigger for their headaches. Brief Headache Screen (BHS).—The BHS consists of 6 items and using a scale of 0 (almost never) to 4 (daily or near daily), the measure determines the frequency of severe (disabling) headache, other (mild) headache, and use of symptomatic medication, to generate a diagnosis.34 Any response other than 0 for items pertaining to severe (migraine) and less severe headache was used in this study to reflect a positive diagnosis. A diagnosis of chronic migraine was given to participants with either severe or mild headache frequency more than 3 days per week if question 1 indicated migraine.34 Medication-overuse is defined as symptomatic medication used 3-4 days per week or daily, and modified any primary headache diagnosis indicated by an earlier response.34 Validation studies have shown the measure to be both a sensitive and specific self-report screening tool for migraine, daily headache syndrome, and medication-overuse headache.34,35 Pain Anxiety Symptoms Scale, Brief Version (PASS-20).—The PASS-2036 is a 20-item inventory developed to measure anxiety about pain. It is a shorter version of the Pain Anxiety Symptom Scale.37 The measure is composed of 4 subscales: cognitive anxiety, escape/avoidance, fear, and physiological anxiety, with higher summated total and subscale scores being indicative of higher painrelated anxiety.18 The PASS-20 has good reliability

6 and validity as well as a strong 4-factor structure corresponding to the 4 subscales.36 In the current study, the overall score Cronbach’s alpha coefficient was .93 and was in the .70s and .80s for the subscales. Previous research on chronic pain has shown that high levels of pain-related anxiety—through pain avoidance behavior—can lead to an increased risk for chronic pain and disability.18 Based on the theoretical relationship between pain-related anxiety (fear of pain) and avoidance behavior, to test the convergent validity of the HTSAQ Avoidance scale, a significant direct correlation was expected between the PASS-20 and this scale. As trigger sensitivity is also related to fear of triggers, to test the convergent validity of both the HTSAQ sensitivity scales, a significant direct correlation was also predicted between PASS-20 total score and both of these scales. Pain Sensitivity Questionnaire (PSQ).—The PSQ is a reliable and valid measure of pain sensitivity.38 The measure was chosen to test the convergent validity of the HTSAQ S(O) scale. This Likertstyle measure consists of 17 items each describing a daily life situation. Participants are asked to rate how painful each situation would be for them on a scale ranging from 0 (not painful at all) to 10 (worst pain imaginable).38 In line with previous research which has shown chronic headache sufferers often experience higher than average pain sensitivity,39 it was predicted that a significant positive correlation would exist between the HTSAQ and the PSQ. The PSQ has shown good internal consistency (Cronbach’s alpha of .90) and good test-retest reliability (r 5 .79).38,40 For the present sample, Cronbach’s alpha coefficient was .93. The Chronic Pain Acceptance Questionnaire (CPAQ).—The CPAQ is a 20-item self-report measure that assesses acceptance and experiential avoidance and is specific to pain experiences.41 In line with the notion that acceptance of pain involves both engaging in important activities despite pain and being willing to endure pain, the measure consists of 2 subscale scores: activity engagement and pain willingness. Factor analytic studies have confirmed a 2-factor structure supporting these subscales.41,42 Higher scores for each of

Month 2016 the scales is indicative of greater acceptance of chronic pain. Cronbach’s alpha coefficients for the present study were .86 for the overall score and .84 for each of the subscales. This measure was included in the study as the final test of convergent validity. Based on the theoretical differences between pain acceptance and avoidance behaviors, a significant inverse relationship was expected between the CPAQ and the HTSAQ. Statistical Analysis.—The authors had full access to all study data and take responsibility for both its integrity and the accuracy of the data analysis. Data Screening.—All statistical analyses were conducted using IBM SPSS version 22 for Windows. Data screening procedures were conducted according to the protocol recommended by Tabachnick and Fidell.43 Descriptive statistics were obtained in order to describe the characteristics of the sample and for the purpose of testing assumptions underlying statistical techniques. Frequencies were observed prior to assumption checking for evidence of data entry errors, however none were found. All scales were computed electronically. The final 4 items of the HTSAQ consisted of 2 items that were not applicable to all respondents (menstruation and smoking) and 2 self-nominated triggers for adding triggers that were not listed. As the subsequent 15% missing data on these 4 items was attributed to participants who did not meet the criteria required to respond to these items, the data were deemed missing at random. To avoid a female only and greatly diminished sample size due to the list-wise deletion of affected cases, the HTSAQ scales were instead computed twice. Firstly to include both male and female respondents but excluding the above 4 mentioned items, and secondly to include only female respondents but including the items pertaining to the trigger menstruation. A second HTSAQ Avoidance scale was not created as responses to Item D (How hard do you try to avoid the menstrual cycle?) were not obtained. The numbers of participants endorsing menstruation or smoking as a potential trigger for their headaches have also been provided in Table 1. Readability Analyses.—The Readability Test Tool44 was used to test the readability of the HTSAQ. This online software checks the semantic

Headache

7

Table 2.—Flesch Kincaid Reading Ease Score and US Grade Level Indicators for the HTSAQ

Readability Formula

Flesch Kincaid Reading Ease Flesch Kincaid Grade Level Gunning Fog Score SMOG Index Coleman Liau Index Automated Readability Index

Readability Indices

62.7 8.8 11.9 8.7 9.9 7.8

The higher the Flesch Kincaid Reading Ease score, the easier the text is to read. A value between 60 and 80 should be easy for a 12- to 15-year-old to understand. The Coleman Liau and Automated Readability Index rely on counting characters per word. The other indices consider number of syllables, and sentence lengths.

difficulty and syntactic complexity of written dialect using the following formulas: Flesch Reading Ease,45 Flesch-Kincaid Grade Level,46 Gunning Fog Index,47 Simplified Measure of Gobbledygoop,48 Coleman-Liau Index,49 and Automated Readability Index.50 These formulas are considered to be reliable and consistent tests of readability.51,52 Reliability Analyses.—Cronbach’s alpha was used to measure the internal consistency of each of the 4 HTSAQ scales for both the mixed gender and female only TAMH sample, which excluded medication-overuse headache sufferers. Scale testretest reliability coefficients were calculated using Pearson product-moment correlations with an interval of 2-6 weeks. Reliability was determined using the conventional thresholds of Cronbach’s a and test-retest coefficients of 0.7.53,54 Validity Analyses.—Validity analyses were all assessed using the baseline data from the TAMH sample. Correlation coefficients were used to measure the convergent validity of the 4 HTSAQ scales (both including and excluding menstruation items) with the predicted corresponding scale or subscales of the PASS-20, PSQ, and CPAQ. Construct validity of the HTSAQ was further evaluated using known-groups validity based on predictions from the TAMH. Predictions on the basis of the TAMH were for significant differences between chronic migraine (CM) and the episodic headache groups,

but no significant difference in HTSAQ scores between the episodic headache groups—episodic migraine (EM) and episodic other headache (EOH). ANOVAs were used to compare the mean HTSAQ scale scores (excluding menstruation items) between these groups.

RESULTS Readability Analysis.—The average readabilty test score for the HTSAQ was 9. This score suggests that the wording of the questionnaire should easily be understood by individuals greater than 14 years of age. The exact readabilty indices for each formula utilized by The Readability Test Tool44 are displayed in Table 2. Reliability Analysis.—Internal Consistency.—All scales indicated a high level of internal consistency, as determined by Cronbach’s alpha values above .8.55 Item-total statistics revealed 10 potential items which perhaps were not correlating appropriately with their scale total score, as shown by corrected item-total correlations below .3.55 Removal of each item from its corresponding scale however did not change the Cronbach’s alpha. Table 3 displays a list of these items including their corrected item-total correlation and Cronbach alpha coefficient if deleted. Descriptive statistics including means, standard deviations and Cronbach alpha coefficients are displayed in Table 4. Test-Retest Reliability.—Two-hundred and one participants completed the prospective component of the study (53%). Although a high amount of attrition was evident between the first and follow-up survey, any implications of the missing not at random (MNAR) data were deemed mitigated by the similarities between the follow-up survey completers and non-completers. Test-retest analyses on the HTSAQ scales (excluding items pertaining to menstruation) revealed strong positive correlations for the Triggers scale, S(O) scale, S(T) scale, and Avoidance scale, of .85, .86, .85, and .81, respectively. Strong positive test-retest correlations were also found for the Triggers scale (.88), S(O) scale (.87), and the S(T) scale (.86) for the female only subsample (n 5 173), which included items pertaining to the trigger menstruation.

8

Month 2016 Table 3.—Items With a Low Corrected Item-Total Correlation and Cronbach’s Alpha Coefficient if Deleted

Mixed Gender Sample (n 5 289)

Scale

Female Only Sample (n 5 249)

Corrected item-total r

Cronbach’s a if item deleted

Corrected item-total r

Cronbach’s a if item deleted

How often do you experience headaches because of odors/smell/fragrances? How often do you experience headaches because of drinking alcholol? How often do you experience headaches because of coughing/sneezing?

.251

.85

.225

.85

.077

.85

.098

.85

.267

.84

.256

.85

How sensitive are you to odors/smell/fragrances as a trigger of headaches currently compared with other people? How sensitive are you to drinking alcohol as a trigger of headache currently compared with other people? How sensitive are you to coughing/sneezing compared with other people?

.293

.84

.269

.84

.213

.84

.201

.84

.252

.84

.255

.84

How sensitive are you to drinking alcohol compared with how sensitive you were at the time in your life when you were least sensitive to drinking alcohol? How sensitive are you to low temperature compared with how sensitive you were at the time in your life when you were least sensitive to low temperature?

.218

.86

.216

.86

.215

.86

.173

.86

How hard do you try to avoid drinking alcohol? How hard do you try to avoid coughing/ sneezing?

.287

.82

–

–

.256

.83

–

–

Item

Triggers

S(O)

S(T)

Avoidance

S(O), sensitivity to triggers compared with others; S(T), sensitivity to triggers compared with time of least sensitivity. Mix gender sample excluded items pertaining to menstruation. Female only sample included Items A, B, and C for the trigger menstruation.

Validity Analysis.—Convergent Validity.—Prior to calculating correlation coefficients, assumptions required for correlation analyses were assessed. Potential outliers were indentified via visual inspection of both box plots and a matrix of scatterplots. Although none were extreme cases, the removal of 5 identified outliers corrected violations of normality for both the HTSAQ S(T) (excluding menstruation) and the PASS-20 physiological scale. As correlation analysis is considered sensitive to outliers,43 coefficients were calculated twice both with and without a further 5 repeatedly identified cases. As the strength of the

relationship between variables was being inflated by these outliers, they were also removed leaving 279 remaining cases for the analyses involving the computed HTSAQ scales, which excluded items pertaining to menstruation. Although checked, no outliers were removed for the analyses involving the 3 computed HTSAQ scales, which included items pertaining to menstruation (n 5 249). As the PASS-20 fear scale violated the assumption of normality, due to additional heteroscedasticity, Spearman’s rho was used as a nonparametric alternative to Pearson’s correlation coefficient to investigate the relationship between the

Headache

9

Table 4.—Summary of Means, Standard Deviations, and Cronbach’s Alpha Coefficients for the 4 HTSAQ Scales

Mixed Gender Sample (n 5 289)

Female Only Sample (n 5 249)

HTSAQ Scale

Mean

SD

Cronbach’s a

Mean

SD

Cronbach’s a

1. 2. 3. 4.

9.06 44.92 37.65 51.93

4.20 10.17 11.20 11.53

.85 .84 .86 .83

9.87 47.77 40.05 –

4.31 10.76 11.92 –

.85 .84 .86 –

Triggers S(O) S(T) Avoidance

Both samples excluded medication overuse sufferers (n 5 87). S(O), sensitivity to triggers compared with others; S(T), sensitivity to triggers compared with time of least sensitivity. Maximum trigger score is 22 for mixed gender sample as not all participants answered questions 23–26. Maximum trigger score is 23 for the female only sample as it included menstruation as a trigger. For the remaining scales higher scores are indicative of more extreme responding for that construct, maximum score5 110 (mixed) and 115 (female only).

PASS-20 fear scale and the HTSAQ Avoidance scale. A moderate positive correlation between these variable was found, rho 5 .32, n 5 279, P < .001. Pearson product-moment correlations for all remaining variables are displayed in Tables 5 and 6. Known-Groups Validity.—Comparison of HTSAQ mean scale scores between CM, EM, and EOH are reported in Table 7. Significant differences were present between CM and both the episodic headache groups for all scales (P < .01). There were no significant differences between episodic headache sufferers (EM and EOH) scores on the Avoidance or S(T) scales, but significant differences were found for the Triggers scale (Mdiff 5 1.57, P 5 .016, Hedges’s g 5.40 95% CI [0.22, 2.91]) and the S(O) scale (Mdiff 5 3.82, P 5.010, Hedges’ g 5 .43, 95% CI [0.72, 6.92]). Although the mean differences between the episodic headache groups for these scales reached statistical significance, the actual difference in mean scores was quite small. The effect sizes for the Trigger and S(O) scales for all 3 groups, calculated using g2, were .13 and .17, respectively, which are medium effects.

DISCUSSION Findings from this study suggest that the HTSAQ is a psychometrically sound measure.

Readability analysis revealed the HTSAQ should be easily understood by adults. The internal consistency and test-retest reliability of the 4 prescribed scales were all well above the satisfactory cutoff of .7.55 The convergent validity of each of the HTSAQ scales was supported by significant correlations with measures of pain anxiety, pain acceptance and pain sensitivity. As predicted, the 2 sensitivity scales of the HTSAQ both showed moderate correlations with pain-related anxiety (r 5 0.33 and 0.43 for scales excluding menstruation, and r 5 0.38 and 0.34 for scales including menstruation). Significant correlations were also found between the HTSAQ Avoidance scale and the PASS-20 total and subscale scores. Most of the correlations were moderate in strength, with the strongest association being between the Avoidance scale and pain anxiety (r 5 0.36). These results are in line with previous research showing an association between painrelated anxiety and avoidance behavior,36 and support the construct validity of the HTSAQ. Consistent with predictions, a significant moderate inverse relationship was also observed between the HTSAQ Avoidance and Sensitivity scales and pain willingness (as measured by the CPAQ). Significant inverse correlations were also observed between the HTSAQ and the active process of pain acceptance (as measured by the CPAQ

– .84** .64** .46** .34** .18** .29** .35** 2.15* 2.26** – .17** 8.87 4.06

1

– .73** .55** .43** .28** .34** .40** 2.20** 2.30** – .24** 44.44 9.55

2

– .47** .33** .19** .26** .32** 2.22** 2.30** – .16** 36.97 10.34

3

– .36** .26** .26** .32** 2.21** 2.30** – .24** 51.35 10.83

4

– .79** .84** .77** 2.53** 2.60** 2.26** .31** 31.92 13.98

5

– .62** .43** 2.44** 2.43** 2.26** .15* 9.86 4.52

6

– .49** 2.40** 2.51** 2.15* .29** 11.77 5.06

7

– 2.41** 2.48** 2.18** .25** 5.80 3.85

8

– .72** .82** 2.17** 74.14 15.73

9

– .18** 2.16** 34.15 9.14

10

– 2.11* 39.99 11.19

11

– 3.48 1.31

12

Only significant correlations have been reported. N 5 279 and includes male and female respondents. S(O), sensitivity to triggers compared with others; S(T), sensitivity to triggers compared with time of least sensitivity. *P < .05; **P < .01.

1. HTSAQ Triggers 2. HTSAQ S(O) 3. HTSAQ S(T) 4. HTSAQ Avoidance 5. Pain Anxiety (PASS-20 Total) 6. Escape/Avoidant Behavior (PASS-20) 7. Cognitive Anxiety (PASS-20) 8. Physiological Anxiety (PASS-20) 9. Pain Acceptance (CPAQ Total) 10. Pain Willingness (CPAQ) 11. Activity Engagement (CPAQ) 12. Pain Sensitivity (PSQ Total) M SD

Scale

Table 5.—Peasons Product–Moment Correlations Between HTSAQ Scales (Excluding Menstruation Items) and Measures of Pain Avoidance, Sensitivity, and Acceptance

10 Month 2016

Headache

11

Table 6.—Peasons Product–Moment Correlations Between HTSAQ Scales (Including Menstruation Items) and Measures of Pain Avoidance, Sensitivity, and Acceptance

Scale

1

HTSAQ Triggers HTSAQ S(O) HTSAQ S(T) HTSAQ Avoidance Pain Anxiety (PASS-20 Total) Escape/Avoidant Behavior (PASS-20) 7. Cognitive Anxiety (PASS-20) 8. Physiological Anxiety (PASS-20) 9. Pain Acceptance (CPAQ Total) 10. Pain Willingness (CPAQ) 11. Activity Engagement (CPAQ) 12. Pain Sensitivity (PSQ Total) M SD 1. 2. 3. 4. 5. 6.

2

3

– .83** .64** .47** .33** .21**

– .77** .53** .38** .46**

– .49** .34** .23**

.29** .33** 2.20** 2.27** – .21** 9.87 4.28

.28** .34** 2.21** 2.30** – .25** 47.77 10.76

.28** .31** 2.26** 2.31** – .18** 40.05 11.92

Only significant correlations have been reported. N 5 249 and includes only female repsondents. S(O), sensitivity to triggers compared with others; S(T), sensitivity to triggers compared with time of least sensitivity. Correlations for a second Avoidance Scale are not included in the table as respondants were not asked item D “How hard do you try to avoid the menstrual cycle?.”*P < .05; **P < .01.

total score). There were no significant correlations between the HTSAQ and the second behavior domain of acceptance of chronic pain, activity

engagement. As only 16.6% of the sample was classified as suffering from chronic headache, and the items on CPAQ refer to chronic pain, the nonsignificant correlation may be explained by this characteristic of the sample. Additional correlation analyses with only the chronic headache participants was not conducted as G*Power calculations for an a error probability of .05 and a power of .80 revealed the number of chronic headache participants in this sample (n 5 62) was not large enough to detect a medium effect.56 Nevertheless, the data seem to indicate that sensitivity to headache triggers is more related to willingness to endure pain than preparedness to engage in important activities despite pain. The present results demonstrated the hypothesized correlation between the HTSAQ and general pain sensitivity (as measured by the PSQ). The magnitude of the relationships, however, indicated that the constructs, although related, are quite distinct from one another. Although previous research has shown that general pain sensitivity is often elevated in various localized chronic pain disorders, including TTH headache,39 as mentioned previously, the present sample consisted mainly of episodic headache sufferers as opposed to chronic headache sufferers. This characteristic of the sample may at least

Table 7.—Comparison of Mean HTSAQ Scales Scores for Chronic Migraine, Episodic Migraine, and Episodic Other Headaches

Headache Diagnoses

CM (n 5 57)

Scale

Triggers S(O) S(T) Avoid

M (SD)

11.47 51.47 42.89 55.81

(3.34) (8.27) (10.39) (10.33)

EM (n 5 156)

[95% CI]

[10.59, [49.28, [40.14, [53.07,

M (SD)

[95% CI]

P value

EOH (n 5 66)

M (SD)

12.36] 8.67 (3.86) [8.06, 9.28] 7.11 (4.02) 53.67] 43.78 (8.26) [42.31, 45.24] 39.95 (7.89) 45.65] 36.25 (10.25) [34.63, 37.87] 33.58 (8.34) 58.55] 50.74 (10.52) [49.07, 52.40] 48.95 (11.04)

[95% CI]

[6.12, 8.10] [38.02, 41,48] [31.53, 35.62] [46.24, 51.67]

Comparison of CM with Comparison of

EM EOH

.000 .000 .000 .007

.000 .000 .000 .001

EM with EOH

.016 .010 .198 .760

CM, chronic migraine; EM, episodic migraine; EOH, episodic other headache; CI, confidence interval; S(O), sensitivity to triggers compared with others; S(T), sensitivity to triggers compared with time of least sensitivity; Avoid, avoidance; Higher scores are indicative of more extreme responding for that construct, maximum score 5 110. Maximum trigger score 5 22.

12 partially explain why a stronger relationship between the HTSAQ and pain sensitivity was not found. Further supporting the construct validity of the HTSAQ, the observed correlations between the HTSAQ scales themselves were all moderate to large (r 5 0.46-0.84). The TAMH suggests that the scores on the scales would be related but they are measuring different constructs, which is consistent with these correlations. Construct validity was further confirmed within the framework of knowngroups validity, with results showing that HTSAQ scale scores differed significantly between those sufferering from chronic migraine and episodic headaches. In line with the premise of the TAMH,27 chronic migraine sufferers had a higher trigger score and showed increased trigger sensitivity and trigger avoidance compared with those with episodic headache.

LIMITATIONS While the research produced promising results, some limitations should be taken into account in the interpretation of the study’s findings. First, the development of the HTSAQ is based on the Trigger Avoidance Model of Headaches and the approach to trigger management derived from this model Learning to Cope with Triggers. This line of research has been driven by a particular research group (Martin and colleagues in Australia), and although it is gaining some acceptance in the literature, there is a need for additional studies from other groups of researchers that either confirm or otherwise challenge the findings of the originating group. Second, the main limitation of the study is the lack of a factor analysis. Although high internal consistency for each scale was found (a > .80)— suggesting that the items which make up the scales are measuring a similar underlying construct55—Cronbach’s alpha coefficients cannot determine whether the sample of items being analyzed represent a single dimension or multiple dimensions.57 Procedurally, factor analysis would have given greater confidence regarding the prescribed 4 factor structure of the HTSAQ. It would have also helped identify whether the HTSAQ

Month 2016 has a meaningful total score. Factor analysis could not be used in the current study as the ratio of participants to items was not large enough.43,53 A follow-up study with a larger sample and data analysis via factor analysis is highly recommended. Furthermore, although previous research supports the factors included in the HTSAQ are headache and migraine triggers as reviewed earlier, several items pertaining to the triggers odors/ smell/fragrances, drinking alcohol, coughing/ sneezing, and low temperature yielded lower than anticipated item-total correlations. Although the reasoning for these low correlations is unclear, they do further highlight the importance of undertaking factor analysis in a future study. Another important limitation to acknowledge is that the questionnaire relies on self-report data and hence is subject to all the issues associated with such data. For example, do respondents really know how sensitive they are to triggers compared with others, or how sensitive they are now compared with in the past? Going further, can headache sufferers even accurately identify what are the triggers of their headaches, which is a far from trivial question as there are recent data suggesting that they are poor at doing so.58 Additionally, although online data collection allowed the incorporation of filters and response formulas—preventing both missing data and typing or data entry errors—it may have introduced participation bias because study participants were restricted to those who had access to the internet. Finally, given the way the sample was recruited, it is not clear whether the same results would have been obtained with more severe headache sufferers (eg, sample drawn from a specialist headache clinic or participants in a treatment trial).

IMPLICATIONS Having established the reliability and validity of the HTSAQ, it can now be used by researchers to investigate the factors that correlate with the effectiveness of LCT and its components (exposurebased strategies vs avoidance strategies). The HTSAQ will be particularly useful in randomized control trials that aim to investigate the degree to

Headache which participants continue to avoid certain triggers (Avoidance scale), and to what degree desensitization to triggers has occurred through exposurebased strategies (S[O] scale). In particular, the measure will allow researchers to investigate the degree to which exposure to triggers has succeeded in reducing the number of factors triggering headaches (Trigger scale), as well as the participants’ sensitivity to triggers (S[O] scale and S[T] scale). It will also allow researchers to investigate the degree to which management strategies, based on counseling trigger avoidance, succeed to the degree that participants do avoid triggers (Avoidance scale). The HTSAQ will also be useful in clinical practice, as it will allow practitioners to monitor the progress of their clients when LCT strategies are being utilized in the management of recurrent headaches. At a general level, the development of the HTSAQ should encourage research on exposure versus avoidance of triggers in the context of the TAMH, and such research is much needed to support or refute the Model. In conclusion, the initial psychometric analysis of the HTSAQ supports that it is a reliable and valid measure to be used in adult samples to assess the variables relating to the TAMH. Although further validation research investigating the factor structure of the measure is recommended, the current results warrant the use of the HTSAQ in research on the TAMH and LCT.

STATEMENT OF AUTHORSHIP Category 1 (a) Conception and Design Paul Martin and Sonja Kubik (b) Acquisition of Data Sonja Kubik (c) Analysis and Interpretation of Data Sonja Kubik Category 2 (a) Drafting the Manuscript Paul Martin and Sonja Kubik (b) Revising It for Intellectual Content Paul Martin and Sonja Kubik

13 Category 3 (a) Final Approval of the Completed Manuscript Paul Martin and Sonja Kubik

REFERENCES 1. Kelman L. The triggers or precipitants of the acute migraine attack. Cephalalgia. 2007;27:394402. 2. Karli N, Zarifoglu M, Calisir N, et al. Comparison of pre-headache phases and trigger factors of migraine and episodic tension-type headache: Do they share similar clinical pathophysiology? Cephalalgia. 2005;25:444-451. 3. Deniz O, Aygul R, Kocak N, et al. Precipitating factors of migraine attacks in patients with migraine with and without aura. Pain Clin. 2004; 16:451-456. 4. Ierusalimschy R, Moreira Filho PF. Precipitating factors of migraine attacks in patients with migraine without aura. Arch Neuropsychiat. 2002; 60:609-613. 5. Spierings LH, Ranke AH, Honkoop PC. Precipitating and aggravating factors of migraine versus tension-type headache. Headache. 2001;41:554-558. 6. Leone M, Vila C, McGowan C. Influence of trigger factors on the efficacy of almotriptan as early intervention for the treatment of acute migraine in a primary care setting: The START study. Expert Rev Neurother. 2010;10:1399-1408. 7. Andress-Rothrock D, King W, Rothrock J. An analysis of migraine triggers in a clinic-based population. Headache. 2010;50:1366-1370. 8. Skaer T. Clinical presentation and treatment of migraine. Clin Ther. 1996;18:229-245. 9. Friedman DI, De Ver Dye T. Migraine and the environment. Headache. 2009;49:941-952. 10. World Health Organization. Neurological Disorders: Public Health Challenges. Switzerland: WHO; 2006. 11. Marcus D. Chronic headache: The importance of trigger identification. Headache Pain. 2003;14:139-144. 12. Martin PR. Stress and primary headache: Review of the research and clinical management. Curr Pain Headache Rep. 2016;20:1-8. 13. Snyder CR, Pulvers KM. Dr Seuss, the coping machine. In: Snyder CR, ed. Coping with Stress: Effective People and Processes. New York, NY: Oxford University Press; 2001:3-29.

14 14. Hayes S, Strosahl K, Wilson K, et al. Measuring experiential avoidance. Psychol Rec. 2004;54:553-579. 15. Schmalz J, Murrell A. Measuring experiential avoidance in adults: The avoidance and fusion questionnaire. Int J Behav Consult Ther. 2010;6: 198-213. 16. Asmundson G, Norton P, Vlaeyen J. Fear-avoidance models of chronic pain: An overview. In: Asmundson G, Vlaeyen J, Crombez G, eds. Understanding and Treating Fear of Pain. Oxford, UK: Oxford University Press; 2004:3-24. 17. Lethem J, Slade PD, Troup DG, et al. Outline of fear- avoiding model of exaggerated pain perception. Behav Res Ther. 1983;21:401-408. 18. Murray P, Abrams R, Nicholas C, et al. An exploration of the psychometric properties of the PASS-20 with a nonclinical sample. J Pain. 2007;8:879-886. 19. Ocanez K, McHugh K, Otto M. A meta-analytic review of the association between anxiety sensitivity and pain. Depress Anxiety. 2010;27:760-767. 20. Eysenck H. The conditioning model of neurosis. Behav Brain Sci. 1979;55-199. 21. Moses E, Barlow D. A new unified treatment approach for emotional disorders based on emotion science. Curr Direc Psychol Sci. 2006;15:146150. 22. Martin P, MacLeod C. Behavioral management of headache triggers: Avoidance of triggers is an inadequate strategy. Clin Psychol Rev. 2009;29: 483-495. 23. Martin P. Headache triggers: To avoid or not to avoid, that is the question. Psychol Health. 2000; 15:801-809. 24. Martin P. How do trigger factors acquire the capacity to precipitate headaches? Behav Res Ther. 2001;39:545-554. 25. Martin P, Reece J, Forsyth M. Noise as a trigger for headaches: Relationship between exposure and sensitivity. Headache. 2006;46:962-972. 26. Martin P, Lae L, Reece J. Stress as a trigger for headache: Relationship between exposure and sensitivity. Anxiety Stress Coping. 2007;20:393-407. 27. Martin P. Behavioral management of migraine headache triggers: Learning to cope with triggers. Curr Pain Headache Rep. 2010;14:221-227. 28. Philips H, Jahanshahi M. Chronic pain: An experimental analysis of the effects of exposure. Behav Res Ther. 1985;23:281-290.

Month 2016 29. Martin P. Managing headache triggers: Think ‘coping’ not avoidance. Cephalalgia. 2010;30:634-637. 30. Panconesi A. Alcohol and migraine: Trigger factor, consumption, mechanisms. A review. J Headache Pain. 2008;9:19-27. 31. Bendtsen L, Evers S, Linde M, et al. EFNS guideline on the treatment of tension-type headache– Report of an EFNS task force. Eur J Neurol. 2010; 17:1318. 32. Gaul C, Visscher CM, Bhola R, et al. Team players against headache: Multidisciplinary treatment of primary headaches and medication overuse headache. J Headache Pain. 2011;12:511-519. 33. Martin PR, Reece J, Callen M, et al. Behavioral management of the triggers of recurrent headache: A randomized controlled trial. Behav Res Ther. 2014;61:1-11. 34. Maizels M, Burchette R. Rapid and sensitive paradigm for screening patients with headache in primary care settings. Headache. 2003;43:441-450. 35. Maizels M, Houle T. Results of screening with the brief headache screen compared with a modified IDMigraineTM. Headache. 2008;48:385-394. 36. McCracken L, Dhingra L. A short version of the pain anxiety symptoms scale: Preliminary development and validity. Pain Res Manage. 2002;7:45-50. 37. McCracken L, Zayfert C, Gross R. The pain anxiety symptoms scale: Development and validation of a scale to measure fear of pain. Pain. 1992;50: 67-73. 38. Ruscheweyh R, Marziniak M, Stumpenhorst F, et al. Pain sensitivity can be assessed by self-rating: Development and validation of the pain sensitivity questionnaire. Pain. 2009;146:65-74. 39. Schoenen J, Bottin D, Hardy F, et al. Cephalic and extracephalic pressure pain thresholds in chronic tension-type headache. Pain. 1991;47:145-149. 40. Ruschewyh R, Verneuer B, Dany K, et al. Validation of the pain sensitivity questionnaire in chronic pain patients. Pain. 2012;153:1210-1218. 41. McCracken L, Vowles K, Eccleston C. Acceptance of chronic pain: Component analysis and a revised assessment method. Pain. 2004;107:159-166. 42. Wicksell R, Olsson G, Melin L. The chronic pain acceptance questionnaire (CPAQ)–Further validation including a confirmatory factor analysis and a comparison with the Tampa Scale of kinesiophobia. Eur J Pain. 2009;13:760-768.

Headache 43. Tabachnick BG, Fidell LS. Using Multivariate Statistics, 6th ed. California: Pearson; 2013. 44. The Readability Test Tool [Computer software]. Retrieved from http://www.webpagefx.com/tools/ read-able/check.php; 200922016. 45. Flesch R. A new readability yardstick. J Appl Psychol. 1948;32:221-233. 46. Kincaid JP, Fishburn RP, Rogers RL, et al. Derivation of new readability formulas (Automated Readability Index, Fog Count and Flesh Reading Ease Formula) for Navy enlisted personnel). (Research Branch Report 8-75, Millington, TN: Naval Technical Training, U.S. Naval Air Station, Memphis, TN); 1975. 47. Gunning R. The Technique of Clear Writing. New York: McGraw-Hill; 1952. 48. McLaughlin GH. Clearing the SMOG. J Reading. 1969;13:210-211. 49. Coleman M, Lia TL. A computer readability formula designed for machine scoring. J Appl Psychol. 1975;60:283-284. 50. Smith EA, Senter RJ. Automated Readability Index. Wright-Patterson Air Force Base. 1967;iii:p. AMRL-TR-6620. 51. Mailloux SL, Johnson ME, Fisher DG, et al. How reliable is computerized assessment of readability? Comput Nurs. 1995;5:221-225. 52. Meade CD, Smith CF. Readability formulas: Cautions and criteria. Patient Educ Counsel. 1991;17: 153-158. 53. Nunnally JC. Psychometric Theory. New York: McGraw-Hill; 1978:1994. 54. Lohr K, Aaronson NK, Alonso J, et al. Evaluating quality-of-life and health status instruments: Development of scientific review criteria. Clin Ther. 1996;18:979-992. 55. Pallant J. SPSS Survival Manual, 5th ed. Sydney, Australia: Allen & Unwin; 2013. 56. Faul F, Erdfelder E, Lang A, et al. GPower 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Meth. 2007;39:175-191. 57. Tavakol M, Dennick R. Making sense of Cronbach’s alpha. Int J Med Educ. 2011;2:53-55. 58. Peris F, Donoghue S, Torres F, et al. Towards improved migraine management: Determining potential trigger factors in individual patients. Cephalalgia. 2016; doi:10.1177/0333102416649761.

15

APPENDIX HEADACHE TRIGGERS SENSITIVITY AND AVOIDANCE QUESTIONNAIRE (HTSAQ) Headache/migraine triggers are defined as factors that alone or in combination can precipitate or aggravate headaches in susceptible individuals. Triggers can be thought of as varying in terms of “potency” or “dosage,” which may be determined by variables such as intensity of the trigger (how extreme) and duration of exposure to the trigger. For example, noise can trigger headaches, but the impact of noise is likely to be determined by how loud it is, and the length of exposure to the noise. Chocolate can trigger headaches but does it require consumption of a row of 5-6 squares (30 g), or a family block (250 g)? Lack of sleep can trigger headaches but does this mean 30-60 minutes less than usual, or the level of sleep deprivation that can arise from international plane travel (most of a night without sleep)? Headaches can be triggered by stress, but stress varies from “daily hassles” such as being caught in a traffic jam or burning the toast, to “major life events” such as being made redundant or losing a family member. There are individual differences in “sensitivity” or “susceptibility” to triggers, that is, the level of potency/dosage of a trigger that is required to precipitate a headache/migraine. For example, for some individuals a very loud noise may be needed to precipitate a headache whilst for others a lower level of noise could result in a headache. Also, sensitivity to triggers can vary across time (ie, a lower or higher trigger potency/ dosage may be required to precipitate a headache during different stages of life). Some individuals try to avoid the triggers of headaches because they do not want to experience a headache, and this tendency may be reinforced by advice from doctors or on the internet. Listed below are the most commonly reported triggers. There are 22 triggers that potentially apply to anyone, one trigger that applies to smokers, and one trigger that applies to women only. There is space for adding up to

16

Month 2016

2 triggers that precipitate your headaches but are not on the list. For each of the triggers listed, we would like you to answer 4 questions using the 5-point scales below the questions. The questions are about whether this potential trigger is a trigger of your headaches (question A), your sensitivity to this trigger (questions B and C), and whether you try to avoid this trigger (question D). A. How often do you experience headaches because of this trigger? 1. Never 2. Rarely 3. Sometimes 4. Usually 5. Always B. How sensitive are you to this trigger currently compared with other people? 1. Not at all sensitive (ie, exposure even at high intensities for prolonged periods would not precipitate a headache) 2. Slightly sensitive 3. Moderately sensitive 4. Highly sensitive 5. Very highly sensitive (ie, exposure even at very low intensities for short periods would precipitate a headache) C. How sensitive are you to this trigger currently compared with how sensitive you were at the time in your life when you were least sensitive to the trigger? 1. Same 2. Slightly more sensitive 3. Moderately more sensitive 4. Much more sensitive 5. Very much more sensitive D. How hard do you try to avoid this trigger? 1. Do not try at all 2. Make a small effort to avoid 3. Make a moderate effort to avoid 4. Make a large effort to avoid 5. Try to avoid at all costs Please respond to each question by circling one number. Please remember that the 4 questions for each trigger that is, A, B, C, and D have different scales to use in responding (refer back).

1 A How often do you experience headaches because of stress? 1 B How sensitive are you to stress as a trigger of headaches currently compared with other people? 1 C How sensitive are you to stress as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to stress? 1 D How hard do you try to avoid stress?

1 2 3 4 5

2 A How often do you experience headaches because of anxiety? 2 B How sensitive are you to anxiety as a trigger of headaches currently compared with other people? 2 C How sensitive are you to anxiety as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to anxiety? 2 D How hard do you try to avoid anxiety?

1 2 3 4 5

3 A How often do you experience headaches because of anger? 3 B How sensitive are you to anger as a trigger of headaches currently compared with other people? 3 C How sensitive are you to anger as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to anger? 3 D How hard do you try to avoid anger?

1 2 3 4 5

4 A How often do you experience headaches because of depression? 4 B How sensitive are you to depression as a trigger of headaches currently compared with other people? 4 C How sensitive are you to depression as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to depression? 4 D How hard do you try to avoid depression?

1 2 3 4 5

5 A How often do you experience headaches because of glare? 5 B How sensitive are you to glare as a trigger of headaches currently compared with other people? 5 C How sensitive are you to glare as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to glare? 5 D How hard do you try to avoid glare?

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

6 A How often do you experience headaches 1 2 3 4 5 because of flicker? 6 B How sensitive are you to flicker as a trigger 1 2 3 4 5 of headaches currently compared with other people? (Continued)

Headache

17

6 C How sensitive are you to flicker as a trigger 1 2 3 4 5 of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to flicker? 6 D How hard do you try to avoid flicker? 1 2 3 4 5

compared with how sensitive you were at the time in your life when you were least sensitive to dehydration/lack of water? 11 D How hard do you try to avoid 1 2 3 4 5 dehydration/lack of water?

7 A How often do you experience headaches because of eyestrain? 7 B How sensitive are you to eyestrain as a trigger of headaches currently compared with other people? 7 C How sensitive are you to eyestrain as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to eyestrain? 7 D How hard do you try to avoid eyestrain?

1 2 3 4 5

8 A How often do you experience headaches because of noise? 8 B How sensitive are you to noise as a trigger of headaches currently compared with other people? 8 C How sensitive are you to noise as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to noise? 8 D How hard do you try to avoid noise?

1 2 3 4 5

12 A How often do you experience headaches because of eating “headache foods”? 12 B How sensitive are you to eating “headache foods” as a trigger of headaches currently compared with other people? 12 C How sensitive are you to eating “headache foods” as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to eating “headache foods”? 12 D How hard do you try to avoid eating “headache foods”?

9 A How often do you experience headaches because of odors/smells/fragrances? 9 B How sensitive are you to odors/smells/fragrances as a trigger of headaches currently compared with other people? 9 C How sensitive are you to odors/smells/fragrances as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to odors/smells/fragrances? 9 D How hard do you try to avoid odors/ smells/fragrances? 10 A How often do you experience headaches because of hunger/not eating? 10 B How sensitive are you to hunger/not eating as a trigger of headaches currently compared with other people? 10 C How sensitive are you to hunger/not eating as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to hunger/not eating? 10 D How hard do you try to avoid hunger/not eating?

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5 1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

11 A How often do you experience headaches 1 2 3 4 5 because of dehydration/lack of water? 11 B How sensitive are you to dehydration/lack 1 2 3 4 5 of water as a trigger of headaches currently compared with other people? 11 C How sensitive are you to dehydration/lack 1 2 3 4 5 of water as a trigger of headaches currently (Continued)

13 A How often do you experience headaches because of drinking alcohol? 13 B How sensitive are you to drinking alcohol as a trigger of headaches currently compared with other people? 13 C How sensitive are you to drinking alcohol as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to drinking alcohol? 13 D How hard do you try to avoid drinking alcohol?

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5 1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

14 A How often do you experience headaches because of high humidity? 14 B How sensitive are you to high humidity as a trigger of headaches currently compared with other people? 14 C How sensitive are you to high humidity as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to high humidity? 14 D How hard do you try to avoid high humidity?

1 2 3 4 5

15 A How often do you experience headaches because of high temperature? 15 B How sensitive are you to high temperature as a trigger of headaches currently compared with other people? 15 C How sensitive are you to high temperature as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to high temperature? 15 D How hard do you try to avoid high temperature?

1 2 3 4 5

16 A How often do you experience headaches because of low temperature?

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5 (Continued)

18

Month 2016

16 B How sensitive are you to low temperature as a trigger of headaches currently compared with other people? 16 C How sensitive are you to low temperature 1 2 3 4 5 as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to low temperature? 16 D How hard do you try to avoid low 1 2 3 4 5 temperature? 17 A How often do you experience headaches because of a lack of sleep? 17 B How sensitive are you to a lack of sleep as a trigger of headaches currently compared with other people? 17 C How sensitive are you to a lack of sleep as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to a lack of sleep? 17 D How hard do you try to avoid a lack of sleep?

1 2 3 4 5

18 A How often do you experience headaches because of an excess of sleep? 18 B How sensitive are you to an excess of sleep as a trigger of headaches currently compared with other people? 18 C How sensitive are you to an excess of sleep as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to an excess of sleep? 18 D How hard do you try to avoid an excess of sleep?

1 2 3 4 5

19 A How often do you experience headaches because of fatigue/tiredness? 19 B How sensitive are you to fatigue/tiredness as a trigger of headaches currently compared with other people? 19 C How sensitive are you to fatigue/tiredness as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to fatigue/tiredness? 19 D How hard do you try to avoid fatigue/ tiredness?

1 2 3 4 5

20 A How often do you experience headaches because of head and neck movements? 20 B How sensitive are you to head and neck movements as a trigger of headaches currently compared with other people? 20 C How sensitive are you to head and neck movements as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to head and neck movements? 20 D How hard do you try to avoid head and neck movements?

21 A How often do you experience headaches because of coughing/sneezing? 21 B How sensitive are you to coughing/sneezing as a trigger of headaches currently compared with other people? 21 C How sensitive are you to coughing/sneezing as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to coughing/sneezing? 21 D How hard do you try to avoid coughing/ sneezing?

1 2 3 4 5

22 A How often do you experience headaches because of travel/trips/driving? 22 B How sensitive are you to travel/trips/driving as a trigger of headaches currently compared with other people? 22 C How sensitive are you to travel/trips/driving as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to travel/trips/driving? 22 D How hard do you try to avoid travel/trips/ driving?

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5 1 2 3 4 5

1 2 3 4 5

1 2 3 4 5 (Continued)

If you smoke cigarettes, please respond to the questions for trigger 23, otherwise jump to the questions for trigger 24. 23 A How often do you experience headaches because of smoking cigarettes? 23 B How sensitive are you to smoking cigarettes as a trigger of headaches currently compared with other people? 23 C How sensitive are you to smoking cigarettes as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to smoking cigarettes? 23 D How hard do you try to avoid smoking cigarettes?

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

If you are female, please respond to the questions for trigger 24, otherwise jump to the questions for triggers 25 and 26. Note, question D is not listed for trigger 24 as it is not possible to avoid. 24 A How often do you experience headaches because of the menstrual cycle? 24 B How sensitive are you to the menstrual cycle as a trigger of headaches

1

2

3

4

5

1

2

3

4

5

(Continued)

Headache currently compared with other women? 24 C How sensitive are you to the menstrual cycle as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to the menstrual cycle?

19

1

2

3

4

5

If there is a trigger for your headaches that does not appear in the list of 24 potential triggers above, please respond to the questions for trigger 25, and to the questions for trigger 26 if there are 2 triggers that are not listed. 25 A How often do you experience headaches because of a factor not listed here - please specify what it is ___________ 25 B How sensitive are you to the factor that you have listed as a trigger of headaches currently compared with other people?

1

2

3

4

5

1

2

3

4

5

(Continued)

25 C How sensitive are you to the factor that you have listed as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to this factor? 25 D How hard do you try to avoid the factor that you have listed? 26 A How often do you experience headaches because of a second factor not listed here - please specify what it is ___________ 26 B How sensitive are you to the second factor that you have listed as a trigger of headaches currently compared with other people? 26 C How sensitive are you to the second factor that you have listed as a trigger of headaches currently compared with how sensitive you were at the time in your life when you were least sensitive to this factor? 26 D How hard do you try to avoid the second factor that you have listed?

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5