Abstract
We investigated the 1-year prevalence, clinical features and mode of treatment of headache in medical students of the University of Lagos, Nigeria, using a self-administered headache questionnaire. Headache prevalence was 46.0% and was significantly higher in women than in men (62.8% vs. 34.1%). Prevalence of tension-type headache was higher than that of migraine (18.1% vs. 6.4%). Although tension-type headache had a similar prevalence in both sexes (male 17.3%, female 19.2%), migraine was three times more common in women (10.9% vs. 3.2%). A family history of headache was present in 22.0%. Only 4.6% sought medical assistance, whereas 68.2% took non-prescription drugs, mainly simple analgesics. Specific drugs for migraine and tension-type headache were rarely used. In conclusion, 1-year headache prevalence is high among medical students at this university. The low consultation rate and the rarity of usage of specific anti-headache drugs probably reflect inadequacies in the management of primary headaches in this population.
Introduction
Headache is a common and important medical condition that, more than being a ‘nuisance’, impacts negatively on the quality of life of affected persons by causing discomfort and impairing the ability to perform routine activities.
Data on the epidemiology of headache in both the general population and specific subgroups are important as a prerequisite for understanding the disease distribution and burden, and planning for equitable distribution of scarce healthcare resources. Research on small and specific populations also has the added advantage of aiding in the identification of factors that influence the frequency and severity of various subtypes of headache (1). Headache research has thus focused on both general populations and specific groups such as adolescents, college undergraduates, people in the workplace and persons with potentially related comorbidities such as epilepsy and psychiatric illnesses.
In our population, relatively little is known about the epidemiology of headache in the past two decades, especially in the light of the more recent case definitions and classifications of the International Headache Society (IHS) (2). Our study objective was thus to explore the epidemiology of headache in our university undergraduates, by determining the 1-year prevalence, clinical characteristics and health-seeking behaviour regarding headaches.
Methods
This was a cross-sectional survey conducted on the population of undergraduate students of the College of Medicine, University of Lagos, Nigeria. The population comprised students in their second to sixth years studying Medicine and Surgery as at May 2007. Approval for the study protocol was obtained from the institutional Health Research and Ethics committee.
We developed a structured questionnaire to obtain information on the 1-year prevalence of headache, the clinical characteristics and the mode of treatment, where relevant. The questionnaire was initially pre-tested in 10 course representatives of the five levels, and modifications made to improve clarity and the ability of participants to self-administer the questionnaire.
To determine its sensitivity and specificity, we distributed the questionnaire to 40 randomly selected students and subsequently conducted face-to-face interviews (as the ‘gold standard’) on all of them to corroborate the diagnosis reached based on the review of the questionnaires. In all, 20/40 had a clinical diagnosis of headaches in the past year based on the questionnaire. Of these, in 19/20 the clinical diagnosis of the type of headache disorder on face-to-face interview tallied with the questionnaire-based diagnosis. In all 20 without headache (based on the questionnaire) the face-to-face interview corroborated the diagnosis. Accordingly, we determined that the questionnaire had a positive predictive value of 95% (true positive 19/20, false positive 1/20) and a negative predictive value of 100% (true negative 20/20, false negative 0/20) and was thus appropriate for the purpose of the study. The single false positive emanated from a questionnaire diagnosis of migraine with aura that on face-to-face interview was actually migraine without aura.
In the survey proper, we distributed the questionnaires through the 10 class representatives. Participation was entirely voluntary, and confidentiality was assured by anonymizing the final questionnaire. Data included in the questionnaire were basic demographic details, questions relating to the occurrence of headache in the preceding year, June 2006 to May 2007, headache characteristics and mode of treatment. The questions were designed to diagnose migraine and tension-type headache according to the IHS criteria (2). The questionnaire considered only the most frequent or burdensome headache in an individual.
Data analyses were performed using
Results
A total of 523 questionnaires were distributed, of which 376 were returned, giving a participation rate of 71.9%. The administrative records for the entire population of second to sixth year medical students showed that there were 576 (369 male, 207 female) students as at May 2007, with a mean age of 23.6 years (
The 376 students who completed the questionnaire comprised 220 men and 156 women. Mean age was 23.4 years (
Headache prevalence
One hundred and seventy-three (46.0%) students, comprising 75 men and 98 women suffered from recurrent headaches.
Table 1 shows the 1-year prevalence of headache in our medical students. Headache was significantly more common in women (62.8%) than in men (34.1%).
One-year prevalence of headache in medical students of the University of Lagos
∗Comparing proportion of men and women in each subgroup.
†Fisher's exact P-value.
Frequency of chronic TTH was 0 (0%).
CI, confidence interval; NS, not statistically significant; RR, relative risk; TTH, tension-type headache.
The overall prevalence of migraine was 6.4% and was significantly higher in women (10.9%) than in men (3.2%). The overall prevalence of tension-type headache was 18.1%, and was similar in both men (17.3%) and women (19.2%).
Thirty-two students (8.5%) had headaches that did not fulfil all criteria for migraine and were classified as probable migraine, whereas 39 students (10.4%) had headache that did not fulfil all criteria for tension-type headache and were classified as probable tension-type headache.
Ten students (four male, six female) had unclassifiable headache.
Headache characteristics
Headache was more commonly bilateral (55.5%), throbbing or pulsating (48.0%), moderate in intensity (49.7%) and frontal in location (41.6%). Aggravation by movement or physical activity was a commonly reported feature (68.8%), closely followed by phonophobia (67.6%). Nausea and/or vomiting occurred in only 9.2% (Tble 2).
Headache characteristics of medical students of the University of Lagos
∗More than one feature may be listed per individual.
Among students with migraine, headache was unilateral in 70.8%, throbbing in 62.5%, associated with nausea and vomiting in 47.8% and with both photophobia and phonophobia in 91.7%. All the students with migraine reported aggravation of headache with routine physical activity, whereas four (16.7%) reported the occurrence of an aura (wavy lines in two students, disturbed speech and blind spots in one student each).
Among students with tension-type headache, headache was bilateral in 77.9%, felt as tightness or pressure in 61.8% and worsened by physical activity in 44.1%. Headache was associated with photophobia in 4.4% and phonophobia in 42.6%. No student with tension-type headache reported nausea or vomiting.
A significantly higher proportion of students with migraine (9/24) than those with tension-type headache (11/68) had a positive family history of headache (P = 0.03).
Headache burden
The mean number of headache episodes reported in the past 1 year was 6.0 (
Headache in most students (71.1%) lasted < 4 h (Table 2). A significantly higher proportion of students with tension-type headache (89.7%) than those with migraine (8.3%) had headache duration < 4 h (P = 0.001)
Headache was described as mild (not interfering with daily activities) or moderate (some interference with daily activities but no restriction to bed) in 90.7% of students (Table 2). Headache was severe (severe restriction in daily activities and/or confinement to bed) in 25% of students with migraine, whereas none with tension-type headache had headache of severe intensity (P = 0.001; Fisher's exact test).
Therapeutic aspects
Table 3 shows the coping strategy of the medical students. Most (68.2%) took non-prescription medication. Only eight (4.6%) sought medical assistance. A similar proportion of students with migraine (18/24) to those with tension-type headache (47/68) self-medicated with non-prescription drugs (P = 0.59). A significantly higher proportion of those with migraine (4/24) than those with tension-type headache (1/68) sought medical attention (P = 0.01).
Coping behaviour of medical students with headache
∗Includes those who used sleep, either alone or in combination with self-medication.
The majority of students (71.7%) used simple analgesics (67.6% took paracetamol, 4.1% took aspirin). No student took a triptan or combination analgesics (Table 4). Similarly, no student took traditional herbal remedies for headache. One student took pizotifen for migraine prophylaxis. No student was on preventive therapy with tricyclic antidepressants, β-blockers, calcium channel blockers or valproate.
Drug option characteristics of medical students with headache
∗More than one drug class may be listed per individual.
Five students with unclassified headache took paracetamol either daily or every other day, but data were insufficient to classify them as analgesic-overuse headache.
Discussion
Headache was found to be common in students of the College of Medicine, University of Lagos, with an overall prevalence of 46%. Several studies that have chosen medical students as the target population have shown similarly high prevalence of headache (1, 3–6). The generally higher prevalence rates for headache among women compared with men reported by most studies was also found in our study. Moreover, women with headache reported more episodes of headache in the past 1 year, on average, than their male counterparts.
Ogunyemi (6), in an earlier study of headache in Nigerian undergraduates, reported a prevalence of migraine of 16%. This is higher than the prevalence of 6.4% in our study. It is known that the reported prevalence of migraine varies with the strictness of the criteria used for diagnosis (7). In our study, diagnosis of migraine was based on the recent operational diagnostic criteria of the IHS. If those students with probable migraine in this study were considered along with those with definite migraine, then the prevalence of migraine would be 14.9%, similar to that reported by Ogunyemi. Epidemiological studies on migraine in Western countries indicate a prevalence ranging from 8 to 14%, with a range of 11–18% in women and 3–8% in men (8). These are mostly population-based studies and so can not be compared with our study. Nevertheless, it appears that there is a racial difference in genetic vulnerability to migraine (9); for example, studies in the Far East (10) and Saudi Arabia (11) have reported much lower migraine prevalence rates of 4.7% and 2.6%, respectively. The female preponderance of migraine in virtually all studies is reflected in our study, with migraine being three times more prevalent in female medical students than in their male counterparts.
Tension-type headache had a higher prevalence than migraine in both men and women in our study. Epidemiological studies show that the prevalence of tension-type headache in a young population varies from one study to another in the range 5.9–34.5% in men and 11.1–40.8% in women (1, 5, 12). The prevalence of 18.1% for tension-type headache in our study is similar to that of 20.4% reported by Kaynak et al. (13) in students of Istanbul University, Turkey, but higher than that of 12.2% found by Deleu et al. (5) in medical students in Oman, and the prevalence of 9.5% in the study by Mitsikostas et al. (1) in medical students in Athens. Apart from possible cultural, regional and genetic differences in the prevalence of tension-type headache, study design and methodology could also affect its reported prevalence. This was a self-reported questionnaire-based study using strict IHS diagnostic criteria. Some authors have pointed out that the IHS criteria for the diagnosis of tension-type headache tended to be more specific than sensitive, leading to high false negatives and an underestimation of the true prevalence of the condition (14). Moreover, our questionnaire considered only the most burdensome headache in an individual and did not take into account those with more than one form of headache such as migraine or tension-type headache.
Most studies (13, 15, 16) report a female preponderance of tension-type headache with a female: male ratio of about 1.5:1 to about 2:1. In our study, the prevalence of tension-type headache was similar in both sexes, with a female: male ratio of 1.1: 1.
Less than one-quarter (22%) of students with headache reported the existence of headache in other family members. This is less than the 58% positive family history of headaches in Oman medical students (5). The positive family history of headache of 37.5% in those with migraine in our study is less than the 79.9% reported by Sanvito et al. (3). Similarly, the 16.2% family history of headache in students with tension-type headache in our study is much lower than the 49.4% found by Kaynak et al. (13). The reason for the low proportion of students with a positive family history of headache in our study is not known. Population-based studies would be necessary to determine if this has any racial or genetic explanation. However, as noted by Sanvito et al. (3), information about family history of headache may not be completely reliable when given by a person who is obviously not qualified to make a correct diagnosis.
Characteristics that were strongly associated with migraine were aggravation by physical activity, presence of photophobia and phonophobia, and moderate to severe intensity of headache. Nausea and/or vomiting occurred in less than half of the students with migraine, whereas one-third had bilateral, non-throbbing headache, characteristics more associated with tension-type headache. The most notable characteristic of tension-type headache was the absence of nausea and vomiting. About a quarter of students with tension-type headache had unilateral headache, whereas over one-third had throbbing headache, characteristics associated with migraine.
Less than 5% of students with headache sought medical assistance during headache episodes. This is despite the fact that the medical school is attached to a teaching hospital that runs a daily students and staff clinic and a weekly neurology clinic. One possible reason for this low recourse to hospital treatment might be the relatively light headache burden. Most students had headache of mild to moderate intensity, lasting < 4 h. As expected, probably because of the greater severity of their headache, a significantly higher proportion of students with migraine than those with tension-type headache consulted a doctor. Most studies (3, 5, 17) have reported that despite the high prevalence of headache in university students, only a small percentage sought medical attention for headache. Most students with headache in our study self-medicated with non-prescription over-the-counter drugs. The range of drugs used was limited to simple analgesics such as paracetamol and aspirin, and the non-steroidal anti-inflammatory drug, ibuprofen. Use of specific antimigraine drugs was rare. Sumatriptan and other ‘triptans’ are virtually unavailable. Surprisingly, readily available and relatively affordable drugs such as propranolol and amitriptyline, which are effective in migraine prophylaxis and in treatment of tension-type headache, were not used. The low consultation rate and the rarity of usage of specific anti-headache drugs probably point to inadequacies in the management of headaches in our population.
Students also reported several non-drug strategies to relieve headache. Sleep, either alone or in combination with analgesics, was the most common non-drug headache-relieving strategy used by many students.
Despite being a widely used therapy for many ailments in Nigeria, traditional herbal remedies were not popular as headache-relieving medications in our student population. The high exposure to Western medicine and culture expected in a university community may account for this, rather than an absence of effective headache-relieving traditional herbal remedies.
We acknowledge that our population is a highly selected one and that our findings may not represent the situation in Nigerian communities. Moreover, our questionnaire may not be as user-friendly in a community-based study.
In conclusion, the results of this study show that the prevalence of headache is high among Nigerian medical students of the University of Lagos. Headache prevalence is higher in women. Tension-type headache occurs more often than migraine. A large majority of students rely on non-prescription simple analgesics for headache relief.