The Burden of Migraine and Its Association with Sleep Quality, Academic Stress, And Quality of Life Among Health Sciences Students: A Cross-Sectional Study

Background: Migraine is one of the leading causes of disability worldwide. However, its impact on health sciences students as a high-risk group has not received much research attention. The present study aimed to evaluate the prevalence of migraine and its relationship with sleep quality, perceived stress, and quality of life in health sciences students.

Methods: In this cross-sectional study, a total of 460 health sciences students (57.8% females; Mage = 20.57 years) were recruited. ID-Migraine™, MIDAS, PSQI, PSS-10, and WHOQOL-BREF were used to collect data. Chi-square test, independent samples t-test, and binary logistic regression were employed.

Results: The overall prevalence of migraine was found to be 32.4% (n = 149), with 69.8% of them having moderate to severe disability as measured by MIDAS Grade III/IV. The migraineurs had significantly low sleep quality (M = 8.77 vs. 3.44, p < .001, Cohen’s d = −2.11), high perceived stress (M = 26.95 vs. 15.83, p < .001, d = −2.44), and low quality of life in all four domains of WHOQOL-BREF (p < .001, d range = 1.75 to 2.08). Female gender (AOR = 6.18, 95% CI [2.28, 16.77]), sleep quality (AOR = 1.45 per PSQI score, 95% CI [1.21, 1.74]), and perceived stress (AOR = 1.63 per PSS-10 score, 95% CI [1.42, 1.86]) were identified as significant predictors of migraine in the multivariable model. The model accounted for 84.0% variance in the outcome (Nagelkerke R²), with correct classification of 93.0% of the sample.

Conclusion: The present study found that one in three health sciences students experience migraine, with significant levels of disability and severe impact on quality of life. Female gender, sleep quality, and perceived stress levels were identified as significant predictors of migraine.

Migraine is a chronic neurobiological disorder that is defined not just by headache but also by a range of disabling symptoms, including photophobia, phonophobia, nausea, and cognitive impairment (Agosti, 2018). It is one of the major contributors to years lived with disability worldwide, placing a considerable burden on patients and healthcare systems (Leonardi & Raggi, 2019). The burden of migraine has been well-established in the general population, but university students, particularly those with health-related courses, are a unique subgroup of individuals who are more prone to migraine. The high level of academic stress, high-stakes examinations, and irregular schedules of health-related courses create a "perfect storm" of environmental triggers that can increase the frequency and severity of migraine (Szabó et al., 2024). Despite the high prevalence of this debilitating condition, the intricate relationship between migraine burden, sleep architecture, and academic-related stress remains an area of investigation.

The relationship between migraine and sleep is a complex one. Sleep problems are considered one of the most common migraine triggers (Tiseo et al., 2020). Similarly, the migraine attack can also interfere with normal sleep architecture (Tiseo et al., 2020). Studies have also shown that poor quality of sleep is significantly more prevalent in migraine patients compared to non-headache patients (Stanyer et al., 2021). The relationship between migraine and sleep is particularly critical in university students. Sleep-wake cycle, which is normally irregular, has been linked with an increase in headache frequency and severity (Song et al., 2018). Furthermore, the relationship between objective and subjective sleep health and migraine has also been linked with the chronification of migraine, indicating that sleep quality may not just be a comorbidity but also a modifiable risk factor for migraine (Duan et al., 2022).

In parallel, the academic environment itself is an important source of stress. For undergrads, academic stress is an important headache trigger, particularly for migraines and tension-type headaches. Research has shown there is a strong relationship between an intense schedule of coursework and headaches (Omogbiya et al., 2020). In the case of health science students, the academic burden of maintaining high GPAs, coupled with the high academic demands, may trigger significant psychological distress. Current studies suggest that migraines are associated with reduced academic performance, increased absenteeism, and the need to employ coping strategies when compared to non-migraine sufferers (AboElela et al., 2025; Szabó et al., 2024). The complicated relationship between migraines, mental health issues such as anxiety or depression, and academic stress can lead to a debilitating cycle that severely handicaps a student’s career (Jan et al., 2024).

In the end, all of these factors come together to impact the health-Related Quality of Life (HRQoL). Migraines impact different aspects of one’s life, including their physical and emotional well-being, as well as their social functioning (Anand & Sharma, 2008). And the impact of the migraine does not stop there; the constant stress of when the next migraine is going to strike, or the anticipatory anxiety of a migraine, also impacts one’s quality of life (Lantéri-Minet et al., 2011). This can be evidenced by the health sciences student who begins to withdraw from social events and experience burnout. There have been many studies done on these different variables individually and in simple combinations (Rafique et al., 2020; Rafi et al., 2022), but a comprehensive study of the complex interplay of migraine burden, sleep quality, academic stress, and quality of life, especially in the highly stressful health sciences student, has not been done. This study aims to quantify migraine burden and examine the relationship of migraine burden to sleep quality, academic stress, and quality of life.

Study Design and Setting

A descriptive cross-sectional design with an analytical focus was used to conduct a study on undergraduates of Health Sciences programs, which include Medicine, Dentistry, Pharmacy, Nursing, and Applied Medical Sciences, from [University Name]. The data collection was conducted during a three-month period [Period]. Which coincides with the middle of the academic term.

Participants and Sampling

Eligibility criteria for the participants included being students, both males and females, who were 18 years or older. Exclusion criteria included the presence of secondary headaches, neurological disorders, or the use of sleep-modifying drugs. The study employed a stratified sampling technique based on colleges and years of study. The study participants were at least 366, calculated using Cochran’s formula for a 95% confidence level and a 5% margin of error based on a migraine prevalence rate of 38.8% (Szabó et al., 2024). To account for the possibility of some participants not completing the questionnaires, 460 questionnaires were administered, all of which were completed (response rate of 100%).

Data Collection Instruments

Data were collected using a structured, self-administered electronic questionnaire comprising validated instruments with high internal consistency (Cronbach's α > 0.75 for all sub-scales).

Sociodemographic and Academic Profile: Age, gender, college, academic year, cumulative GPA, smoking status, and daily caffeine intake.

Migraine Screening (ID-Migraine™): A three-item screener assessing headache-related disability, nausea, and photophobia. Endorsement of ≥2 items indicated a positive migraine screen (Rafi et al., 2022).

Migraine Disability Assessment (MIDAS): Administered only to participants screening positive for migraine. The MIDAS total score summed missed or reduced-activity days over three months across work/school, household, and social domains. Scores were graded as: Grade I (0–5), II (6–10), III (11–20), and IV (≥21) (Agosti, 2018). Non-migraineurs were not assigned a MIDAS grade.

Sleep Quality (PSQI): The Pittsburgh Sleep Quality Index assessed sleep over the past month. Seven component scores yielded a global score (0–21), with >5 indicating poor sleep quality (Duan et al., 2022).

Perceived Stress (PSS-10): The 10-item Perceived Stress Scale measured the degree to which life situations were appraised as stressful. Scores range from 0–40, with >27 indicating high perceived stress (Omogbiya et al., 2020).

Quality of Life (WHOQOL-BREF): The 26-item instrument evaluated four domains: Physical Health, Psychological Health, Social Relationships, and Environment. Domain scores were transformed to a 0–100 scale; higher scores denote better quality of life.

Statistical Analysis

Analyses were performed using IBM SPSS Version 27.0. Descriptive statistics (frequencies, percentages, means, SD) summarized participant characteristics.

Bivariate Analysis: Chi-square tests compared categorical variables (gender, college, year, GPA, smoking, caffeine) between migraineurs and non-migraineurs. Independent samples t-tests compared continuous variables (age, PSQI, PSS-10, QoL domains). Effect sizes were calculated using Cohen's d.

Multivariable Analysis: Binary logistic regression analysis using the "enter" method was employed to identify the independent predictors of the migraine status (dependent variable: Migraine_Positive; 0 = No, 1 = Yes). The independent variables included the demographic factor of age, the PSQI global score, the PSS-10, and the categorical variables of gender, college, academic year, GPA category, smoker or not, and caffeine use or not. The categorical variables were specified in the SPSS program using the /CATEGORICAL subcommand with the indicator or dummy coding method, with the reference group being male, Medicine, 1st year, GPA < 2>

To assess the goodness-of-fit of the binary logistic regression model, we employed the Omnibus test (likelihood ratio χ² test), the Hosmer-Lemeshow test (goodness-of-fit test; p > 0.05 indicates good calibration), Nagelkerke R², and the accuracy of the classification with a cut-off value of 0.5. Statistical significance was established at p < 0.05 (two-tailed).

Participant Characteristics

A total of 460 health sciences students were included in the study, with 194 (42.2%) men and 266 (57.8%) women. The mean age was 20.57 years, with a standard deviation of 2.04 years and a range of 18-26 years. The participants belonged to the fields of Medicine (31.1%), Dentistry (11.5%), Pharmacy (18.9%), Nursing (20.9%), and Applied Medical Sciences (17.6%). The majority of the participants were in their fourth year (22.4%), who were followed by those in their second year (20.9%), fifth year (20.4%), first year (20.0%), and third year (16.3%). The majority of the participants had a GPA of 3.0-4.0 (66.9%), 15.4% were with a GPA of 2.5-3.0, and 13.3% with a GPA above 4.0. The percentage of smokers was 15.9%, and daily caffeine intake varied: 10.9% consumed none, 39.3% consumed 1–2 cups, 30.7% consumed 3–4 cups, and 19.1% consumed 5 or more cups.

Prevalence of Migraine

Using the ID-Migraine™ screener, defined as a positive answer to two or more of three questions, the overall migraine prevalence was found to be 32.4% (n=149). The females with migraine were mor in number (77.9% n=116), while males with migraine were 22.1% (n=33) of the total.

Migraine Disability

Among the migraineurs screened positive, the average total days over the last three months using the MIDAS scale was 21.99 days (SD=19.56; range=0-136). The breakdown of the MIDAS grade among the screened positive population showed that 10.7% (n=16) had minimal or no disability (Grade I), 19.5% (n=29) had mild disability (Grade II), 32.9% (n=49) had moderate disability (Grade III), and 36.9% (n=55) had severe disability (Grade IV). In total, 69.8% had moderate to severe disability among the migraineurs screened positive.

Bivariate Associations with Migraine Status

Categorical Variables

To examine the relationship between various factors and migraine status, chi-square tests were employed (see Table 1). The results revealed that there was a significant relationship between gender and migraine status: χ²(1, N = 460) = 36.24, p < .001, with 43.6% of females screening positive for migraine compared to 17.0% of males. In addition, there was a significant relationship between smoking status and migraine: χ²(1, N = 460) = 6.92, p = .009, with a higher percentage of non-smokers (34.9%) screening positive for migraine compared to smokers (19.2%). However, being in college, academic year, GPA, and daily caffeine consumption were not significant factors (i.e., p-values > .05).

Table 1: Associations Between Categorical Variables and Migraine Status

Note. Data are presented as n (column %). Percentages may not sum to 100 due to rounding. Degrees of freedom for all chi-square tests = 1 with the exception of college (df = 4), Academic Year (df = 4), GPA (df = 4), and Daily Caffeine (df = 3).

Continuous Variables

Independent samples t-tests were conducted to compare continuous variables between migraineurs and non-migraineurs (Table 2). No significant difference in age was observed, t(458) = 0.16, p = .877, Cohen's d = 0.02, 95% CI [-0.18, 0.21]. However, migraineurs reported significantly poorer sleep quality (higher PSQI scores) compared to non-migraineurs, t(458) = −21.15, p < .001, Cohen's d = −2.11, 95% CI [-2.35, -1.87]. Migraineurs also reported significantly higher perceived stress (PSS-10 scores), t(458) = −24.46, p < .001, Cohen's d = −2.44, 95% CI [-2.69, -2.19].

For all four quality of life domains, migraineurs scored significantly lower than non-migraineurs: Physical Health, t(458) = 19.69, p < .001, Cohen's d = 1.96, 95% CI [1.73, 2.19]; Psychological Health, t(458) = 17.56, p < .001, Cohen's d = 1.75, 95% CI [1.52, 1.97]; Social Relationships, t(458) = 20.86, p < .001, Cohen's d = 2.08, 95% CI [1.84, 2.32]; and Environment, t(458) = 18.27, p < .001, Cohen's d = 1.82, 95% CI [1.59, 2.05]. All effect sizes were large according to Cohen's conventions.

Table 2: Comparison of Continuous Variables Between Migraineurs and Non‑migraineurs

Note. PSQI = Pittsburgh Sleep Quality Index (higher scores indicate worse sleep); PSS-10 = Perceived Stress Scale (higher scores indicate more stress); QoL = Quality of Life (WHOQOL-BREF, 0–100 scale, higher scores indicate better QoL). All t-tests are two-tailed. Confidence intervals for Cohen's d are at the 95% level.

Multivariable Analysis: Predictors of Migraine

In order to pinpoint independent predictors of migraine status, a binary logistic regression was performed. The independent predictors included in the model were age, gender, college, academic year, GPA, smoking status, daily number of caffeinated beverages consumed, PSQI global score, and PSS-10 score. The categorical variables included in the model were contrast-coded using the following reference groups: gender – male, college – Medicine, academic year – first year, GPA – less than 2.5, smoking status – non-smoker, caffeinated beverages – no caffeinated beverages.

The overall model was statistically significant, χ²(20, N = 460) = 423.18, p < .001, and correctly classified 93.0% of cases. The model explained a substantial proportion of variance in migraine status (Nagelkerke R² = .840). The Hosmer‑Lemeshow goodness‑of‑fit test indicated adequate model calibration, χ²(8) = 5.79, p = .670.

After adjusting for all other variables in the model, three factors emerged as significant independent predictors of migraine (Table 3):

• Gender: Female participants had approximately six times higher odds of screening positive for migraine compared to males (AOR = 6.18, 95% CI [2.28, 16.77], p < .001).
• Sleep Quality: Each one‑point increase in PSQI global score (indicating poorer sleep quality) was associated with a 45% increase in the odds of migraine (AOR = 1.45, 95% CI [1.21, 1.74], p < .001).
• Perceived Stress: Each one‑point increase in PSS-10 score was associated with a 63% increase in the odds of migraine (AOR = 1.63, 95% CI [1.42, 1.86], p < .001).

No other variables—including age, college, academic year, GPA category, smoking status, or daily caffeine intake—were statistically significant in the multivariable model (all ps > .05).

Table 3: Binary Logistic Regression Analysis Predicting Migraine Status

Note. AOR = adjusted odds ratio; CI = confidence interval. Reference categories: gender (male), smoking status (no). The model also included college, academic year, GPA category, and daily caffeine intake, none of which were statistically significant (all ps > .05). Model fit: χ² (20, N = 460) = 423.18, p < .001; Nagelkerke R² = .840; Hosmer‑Lemeshow χ² (8) = 5.79, p = .670; overall classification accuracy = 93.0%.

This study examines in detail the impact of migraine on the lives of Health Sciences students, whose academic load is heavy, but whose daily routines can also vary greatly. The findings revealed a very high prevalence of migraine among these students, which is significantly higher than the global prevalence of migraine among the general population. What is more worrisome, however, is the positive correlation between migraine, which is severe in these students, their low quality of sleep, high stress levels, and low HRQoL.

Prevalence and Demographic Patterns

The prevalence of migraine was determined to be 32.4% in our study population using the ID-Migraine™ screener. This is in line with more contemporary findings in our region by Szabó et al. (2024), who reported a 36.8% prevalence of migraine among medical students at Alfaisal University. The prevalence also falls within the range of 22.5% to 44.1% reported in similar student populations in Saudi Arabia (AboElela et al., 2025). The relatively high prevalence of migraine in our study population, in comparison with the global prevalence of around 14-15% (Agosti, 2018), supports our hypothesis that the high-stress environment of a health sciences program can provide a potent trigger for those with a genetic predisposition to primary headaches.

In line with the expected epidemiology of migraine, we also found a considerable gender difference in our study population. The prevalence of migraine was considerably higher in females. This is in line with Rafi et al. (2022), who also found hormonal changes, especially withdrawal of estrogen, to be a well-established trigger for migraine attacks in female migraineurs. The female gender was found to be the most potent independent predictor of migraine in our study population, with females being six times more likely to have a positive screener than their male counterparts (AOR = 6.18, 95% CI [2.28, 16.77]). Although the odds ratio was very high, it was in line with findings in similar student populations in which hormonal changes interact with psychosocial factors.

The Bidirectional Relationship Between Migraine and Sleep Disturbance

The first important finding from the study is the association between migraine and sleep disturbances. The PSQI scores were significantly higher in the migraine group than the non-migraine group, indicating poorer sleep quality (mean PSQI scores were 8.77 and 3.44, p < 0 xss=removed>

In the study, it was observed that with every increase in PSQI scores by one point, the odds of experiencing migraines increased by 45%. This indicates that sleep disturbances are an independent risk factor for migraines because the association between the two conditions is dose-response related. Another important finding from the study is that sleep disturbances may not only be an accompanying factor but also a causative factor in the development and maintenance of migraines because the average PSQI scores in the migraine group were 8.77, which is significantly higher than the cutoff point for poor sleep quality, which is >5 (Duan et al., 2022). The study concluded that sleep quality is a modifiable risk factor in the progression from episodic to chronic migraines.

This is a serious problem for health science students because it creates a cycle in which the student gets less sleep because of academic demands, and the lack of sleep may trigger migraines, which may trigger sleep disturbances because of pain, further compromising academic performance. The association between migraines and sleep disturbances is two-sided, and both conditions trigger and maintain each other (Tiseo et al., 2020). The authors concluded that sleep disorders and migraines share neurobiological links with the hypothalamus and dopamine systems.

Perceived Stress as a Potent Independent Predictor

The relationship of higher PSS-10 scores and the incidence of migraines in the data serves to show the impact of psychological stress on the indication of diseases. In our data, individuals who suffered migraines also demonstrated a significantly higher level of stress than those who did not (M = 26.95 compared to 15.83, p < 0 xss=removed xss=removed>

Health sciences students are living in a world full of pressure of constant examinations and a huge coursework. Our results are also consistent with Omogbiya et al. (2020), who identified "days of intense academic activity" as the major headache trigger for undergrads. The migraines were not just simple headaches. They were disabling. Moreover, as the MIDAS scores indicate, more than two-thirds (69.8%) of those with migraines experience moderate to severe disability (Grades III-IV), which means they cannot attend classes or clinical rotations. This, of course, adds a secondary stressor: "anticipatory anxiety" about failing academically. This is a major aspect of the "triple threat" to academic success, as described by Jan et al. (2024).

It should be added that the cumulative GPA did not demonstrate a significant association with migraine status when using simple (bivariate) and complex (multivariable) statistical methods. This contradicts some of the findings of other studies (AboElela et al., 2025; Szabó et al., 2024), and it is possible that health science students have a “bounce-back ability,” using coping strategies to help maintain a good academic standing despite their poor health. Another possibility is that the cumulative GPA is a poor measure of the episodic impact of migraine on daily academic tasks, such as staying attentive during lectures and achieving well on important exams.

Impact on Quality of Life

The most revealing result is the extent to which migraines impact all aspects of quality of life, as measured by the WHOQOL-BREF. Compared to controls, migraine sufferers scored significantly lower in all aspects of physical, psychological, social, and environmental well-being. The impact was large, with a range of 1.75 to 2.08 in Cohen’s d. This corroborates what was described by Leonardi & Raggi (2019) in their review, which used the narrative review method to explain the impact of migraines not just in terms of pain but in terms of “lost life time.” The Physical Health domain of migraine sufferers scored on average 48.85, which was significantly lower than the 71.36 of the non-migraine group.

The impact on social connections is particularly noteworthy, with a loss of 22.65 points on average. This is a genuine concern for young adults who are building their careers. The loss of quality of life is probably exacerbated by the stigma of invisible pain. As Agosti (2018) noted, “migraines are dismissed by friends and instructors as ‘just a headache,’” which could lead to social withdrawal and reduced participation in school life. Other than pain, another area that scored low was Environment, which encompasses financial, healthcare, and physical environments. This implies that migraines don’t just affect the senses; they also affect one’s feelings of safety, opportunity, and support (Anand & Sharma, 2008).

Clinical and Institutional Implications

These findings have implications for universities as well as health care. Firstly, there is a need to ensure that migraine screenings form part of the routine health care of the students. This can be done using the ID-Migraine™ tool. This is important since one in three students experience this. Secondly, there is a need to ensure that universities implement certain interventions to deal with two of the modifiable factors identified. These factors are sleep quality and stress. It is expected that there will be a large benefit if these two factors are intervened with. Even small benefits have the potential to make a big impact, especially considering the dose-response relationship. Thirdly, there is a need to ensure that special considerations are made for those students who have already been diagnosed with migraines. This can include special considerations such as providing alternative dates for examinations, providing recordings of lectures, as well as providing a quiet area in which they can rest during their headache. Lastly, there is a need to ensure that education is provided to the lecturing staff as well as the students. This can assist in removing the stigma that comes with having a migraine. If the condition of having a migraine is recognized as being neurological rather than "just a headache," then the stigma can be eliminated, and the student can feel comfortable in seeking help without being judged.

Limitations and Strengths

This study has several strengths. We used established scales to measure variables: ID-Migraine, MIDAS, PSQI, PSS-10, and WHOQOL-BREF. This has provided us with a means of obtaining accurate data. We have a substantial sample of 460 students, and the absence of any missing data has provided us with a higher statistical power. We used a stratified sampling method to ensure we sampled students from different colleges and different years of study.

Now, let’s look at the weaknesses of the study. One of the weaknesses of the study is the lack of ability to establish a cause-and-effect relationship. We cannot establish whether poor sleep and stress lead to migraines or whether they are a consequence of migraines. Also, the study relies on self-reported data, and some of the data may be subject to recall bias, especially the data on headaches suffered during the past three months. While the PSS-10 scale has been validated for the measurement of stress, it does not specifically measure stress related to academic pressures. This study may not be generalizable to other schools, as it was conducted in one school only. There may be other unmeasured variables, such as genetic predisposition and psychiatric disorders, which may influence the results.

One in three Health Sciences students is affected by migraines, with more than two-thirds of those affected experiencing moderate to severe disability. Being female, poor sleep quality, and high levels of stress were strong independent predictors. The impact of migraines extends beyond pain, causing a dent in quality of life, including physical, mental, social, and environmental dimensions. These findings call for urgent action. It is time to address the problems of sleep, stress, and migraines, which are intricately linked. It is not just a medical necessity but an academic necessity to address these problems for the future health professionals.