ORIGINAL ARTICLE

Pdf version

A COMPARISON OF HEADACHE AND NON-HEADACHE SUFFERERS ON MEASURES OF SOCIAL SUPPORT AND MENTAL HEALTH PROBLEMS

METHOD

Subjects
The subjects of this study consisted of 138 subjects with headaches and 138 subjects without headaches. The data was collected within an urban college, university and working community in Selangor and Kuala Lumpur, Malaysia, over a period of 10 weeks, between April–June 2007.  Inquiries were made of people within the community as to whether they had headaches, and subsequently whether they were willing to take part in the study.  Criteria for inclusion within the headache group was to have experienced a headache twice a month.  Initially, the data for only the headache sufferers was collected.  Subsequently, the data for the non-headache sufferers was collected based on the demographic profile of the headache sufferers.  No distinction was made as to whether the headaches suffered were migraines or tension headaches.  Those who had only one headache per month were not eligible for inclusion within the headache group.  Given the unpredictability sourcing persons with headaches within a community sample, the overall sample was not controlled for age, gender, race or educational qualification. However, each headache sufferer was matched for race and age group with a non-headache sufferer.

Statistical tests
To answer the two hypotheses of this study, two t-tests were used to calculate the results.  The headache status (i.e., headache vs no-headache) formed the independent variable, whereas both mental health status (i.e., General Health Questionnaire) and social support (i.e., Multidimensional Scale of Perceived Social Support) formed the dependent variables.

Mean differences on respondent mental health and social support variables were assessed via Hotelling’s T2.  The main reason for the choice of this test, as opposed to an Analysis of Variance (ANOVA) is that the Hotelling’s T2 test is calibrated for analysis involving two equal groups of continuous data, as opposed to the ANOVA which considers three groups or more of continuous data.

To further understand the specific differences in mental health problems that headache and non-headache subjects were experiencing, an item analysis of each of the 12 questions in the mental health measure was carried out. To do this, a Hotelling’s T2 was carried out (see Table 4).  Higher scores are indicative of more mental health problems. An item-by-item analysis of the mental health results indicates a similar picture to the overall mental health scale result. 

Measurement instruments
The General Health Questionnaire [GHQ]⁹ was designed as a screening device to detect psychiatric disorder in an adult population.  The device encompasses questions on somatic symptoms, anxiety and insomnia, social dysfunction, and depression.  It is a self-administered 12 item scale where respondents rate themselves on a four point scale (i.e., Better than usual, Same as usual, Worse than usual and Much worse than usual).  Higher scores are indicative of greater symptomatology.  A high degree of internal consistency was observed for each of the 12 items with Cronbach’s alpha value of 0.37-0.79.  Test-retest coefficients for the 12 items were highly significant ⁹.  Whilst the author is aware that there are a number of Malaysian studies that have employed various versions of the GHQ in research. However the GHQ-12 has not been specifically validated with Malaysian populations.

Multidimensional Scale of Perceived Social Support [MSPSS]¹⁰ is a 12 item self-administered scale that measures perceived social support from three sources: friends, family and a significant other.  The item responses are of a 7-point categorical scale, where respondents answer whether they agree to whether they do not agree with the statement.  Higher scores are indicative of higher perceived social support.  Internal consistency for the MSPSS was found to be .91 for the total scale, and .90 to .95 for the subscales.  The norms described are those of North American populations.  To date, there are no reliability data for the MSPSS in Malaysian populations.

Procedure
The questionnaires were administered by a team of 12 trainee psychologists.  The questionnaires were given to participants to complete, either on-the-spot, or to take home.  Each participant was also required to complete a consent form.  All questions were administered in English.

RESULTS

71.4% of the respondents were students and 28.6% were working adults. The age of the subjects ranged from between 14 to 63 years.  The mean age of the total sample was 24.6 years, indicating that the age distribution was somewhat skewed towards younger, rather than older adults.  The age range of students who consisted of 71% of the sample, was 14–24 years old.  The age range of the working adults was 15-62 years old.  In any case, the study did not control for age group.  Racially, the sample consisted of 9 (3.3%) Malays, 181 (65.6%) Chinese, 61 (22.1%) Indian, and 25 (9.1%) other races. Education-wise, there were 48 (17.4%) with SPM qualifications, 50 (18.1%) 6th form qualifications, 34 (12.3%) diploma qualifications, 127 (46%) undergraduate degree qualifications, 15 (5.4%) master degree qualifications, and 2 (0.7%) and doctoral degree qualifications.

The mean of GHQ-12 scores for headache sufferers were 24.91 (SD = 6.46), with a range of 37.0 (Maximum = 45.0, Minimum = 8.0). Whereas the mean of GHQ-12 scores for non-headache sufferers were 23.37 (SD = 5.20) with a range of 31.0 (Maximum = 45.0, Minimum = 14.0). This revealed an overall significant effect for headache status.  Persons with headaches generally had slightly more mental health problems [t (274) = -2.176, p < .05] and higher levels of social support from families [t (274) = -4.023, p < .01], as compared with persons with no headaches.  There was no significant difference across groups for measure of social support from friends and significant others (see Table 2).