Perceptions of health, cognition, and pain among middle‐aged and older adults with migraine: A population‐based cross‐sectional study examining findings from the Canadian Longitudinal Study on Aging

Abstract

Objective

The study compared middle‐aged and older Canadians with and without migraine, examining how self‐perceptions of health, cognition, and pain relate to objective health metrics.

Background

Migraine, a debilitating neurological disorder, affects 8.3% of Canadians and 14.0% of the global population. Research has primarily focused on those aged 18–50 years, leaving middle‐aged and older adults understudied. Individuals’ perceptions of their health, cognition, and pain can tangibly impact their well‐being, with negative health self‐perceptions linked to higher hospitalization, illness, and mortality rates.

Methods

This population‐based cross‐sectional study used 2015–2018 data from the Canadian Longitudinal Study on Aging during the first follow‐up, including the Comprehensive cohort (n = 27,765; 14.0% migraine) and Tracking cohort (n = 17,052; 13.3% migraine). Participants in the Comprehensive cohort were interviewed at one of 11 in‐person sites across seven provinces, located 25–50 km from their homes. Those in the Tracking cohort completed a 60‐min telephone interview.

Results

Females were more than twice as likely to report having a migraine diagnosis. Individuals with migraine rated their physical, mental, and oral health lower than those without migraine and had higher rates of anxiety, depression, and mood disorders. Those with migraine perceived their memory as declining and worried about that decline, yet results displayed only trivial differences in objective memory performance between those with a migraine diagnosis and those without. However, more than twice as many individuals with migraine reported that a physician had told them that they have memory problems. Individuals with migraine reported a higher frequency and intensity of pain and discomfort, resulting in a higher occurrence of missed activities, and higher functional impairment scores.

Conclusion

By exploring how individuals with migraine perceive their health, cognition, and pain, this study highlights the gap between self‐reported health perceptions and objective health assessments. For example, individuals with migraine tend to report poorer mental health, a trend that aligns with the higher prevalence of anxiety, depression, and mood disorders observed within this population. Interestingly, despite individuals with migraine rating their memory as lower than those without migraine, objective memory testing reveals either no difference or even slightly improved scores among those with migraine. Finally, our findings support a Canadian lifetime prevalence rate in this older adult cohort of 13.7%, which aligns with the global prevalence rates of 14.0%.

Plain Language Summary

Previous research suggests that the ways in which people perceive their health significantly impacts their well‐being. In this study, we looked at perceptions of mental, general, and oral health among middle‐aged and older Canadians with and without migraine. Results indicated that people with migraine perceived that they had worse physical, mental, and oral health compared to people without migraine, but these perceptions did not always align with objective data; e.g., people with migraine often reported memory problems even though they scored well on memory tests.

Abbreviations

AFT

Animal Fluency Test

CES‐D‐10

10‐item Center for Epidemiological Studies Depression scale

CLSA

Canadian Longitudinal Study on Aging

COWAT

Controlled Oral Word Association Test

MAT

Mental Alternation Test

MMQ‐Ability

abilityportion of the Multifactorial Memory Questionnaire

OARS

Older Americans Resources and Services scale

RAVLT

Rey Auditory Verbal Learning Test

SD

Standard devation

SWLS

Satisfaction With Life Scale

VST

Stroop Neuropsychological Screening Test‐Victoria version

BACKGROUND

Migraine prevalence is highest among individuals aged 18–50 years, a period during which both the neurological disorder and its associated symptoms are most pronounced. Migraine prevalence tends to increase during puberty and continues to rise throughout early to mid‐adulthood, ¹ with peak prevalence occurring around the age of 35 years. ² In contrast, migraine prevalence gradually declines with age during mid‐to‐late adulthood, ³ estimates suggesting a prevalence of ~10.0% among older adults. ⁴ While migraine progression is intertwined with age, there is no definitive endpoint. Evidence indicates migraine symptoms can persist into older age and can extend beyond the headache phase into other neurological manifestations. ⁴ , ⁵ , ⁶ , ⁷ Furthermore, the clinical presentation of migraine may differ in middle‐aged and older adults compared to younger populations. ⁴ , ⁷ , ⁸ For example, middle‐aged and older adults with migraine may be more likely to experience visual disturbances, sensory changes, or confusion during a migraine attack, rather than the headache pain commonly associated with the condition. ⁹ Therefore, investigating the migraine population aged 46–92 years is crucial, as this may provide insights that have not yet been comprehensively explored in the existing literature.

Migraine is a major health concern, affecting 14.0% of the global population, ¹⁰ and 8.3% of Canadians, ¹¹ highlighting the need to investigate general health, mental health, and oral health within this population. A recent study reported poor self‐rated health was significantly greater in individuals with migraine in comparison to those without migraine. ¹² Individuals’ own reflections of their general health is important because one's self‐rated health can be used as a rough indicator of their health status. ¹³ Having a negative health self‐perception is linked to an increase in frequency of hospitalizations, a higher incidence of illnesses, and a heightened risk of mortality. ¹⁴ , ¹⁵ Individuals with migraine often experience elevated pain levels that disrupt their physical, psychosocial, and financial well‐being, ¹⁶ including reduced enjoyment and participation in basic activities of daily living due to headache pain. ¹⁷ , ¹⁸ Therefore, investigating both self‐rated health and pain in individuals with migraine is crucial.

Migraine is associated with various psychological comorbidities, yet the relationship between self‐perceived mental health and objective psychological measures remains predominately unexplored. This study sought to explore whether these self‐reports aligned with established psychological determinants, which have been extensively investigated within migraine literature. Specifically, we focused on depressive symptoms, psychological distress, satisfaction with life, and several diagnosed psychiatric conditions (i.e., clinical depression, anxiety disorder, and mood disorder).

This study explored subjective cognitive decline, meta‐memory, and several cognitive tests assessing executive function and memory in our target population. Subjective cognitive decline refers to self‐reported sense of cognitive decline in the absence of objective cognitive impairment. Evidence indicates 44.2% of individuals with migraine report subjective cognitive decline, often accompanied by more intense and frequent headaches, greater levels of anxiety and depression, and poorer sleep quality. ¹⁹ However, studies on objective cognitive performance in those with migraine are inconsistent. Some studies suggest there is no relationship between migraine and cognitive impairment, ⁸ , ²⁰ with one study finding no significant cognitive differences between middle‐aged and older adults with and without migraine. ²¹ Conversely, a meta‐analysis reported generally lower cognitive function in those with migraine. ²² Given these discrepancies, our study aimed to investigate differences in subjective cognitive decline, meta‐memory, executive function and memory between middle‐aged and older adults with migraine and those without migraine.

Migraine interacts with many aspects of life, with the current investigation focusing on perceptions of health, cognition, and pain, and its relation to objective health measures. This exploration takes place within a large population‐based sample of Canadian middle‐aged and older adults with migraine compared to their age‐matched peers without migraine. These findings will contribute to a more informed understanding of migraine among those who are middle‐aged or older and offer valuable information to clinicians and migraine researchers.

With the limited research examining perceptions of health, cognition, and pain in middle‐aged and older adults with migraine, it is valuable to conduct a study aimed at exploring those factors. Given the review of the literature, we hypothesize individuals with migraine may be more likely to possess a worse self‐perception of their health in comparison to those without migraine. Our secondary hypothesis suggests individuals with migraine may experience more negative mental health outcomes; specifically, a higher incidence of psychiatric disorders (i.e., depression, anxiety, mood disorders), higher psychological distress, and lower levels of satisfaction with life. Another secondary hypothesis postulates individuals with migraine will have increased pain and discomfort, leading to increased impairments in activities of daily living.

METHODS

Study design and participants

This cross‐sectional study utilized data collected during the first follow‐up evaluation from the Canadian Longitudinal Study on Aging (CLSA). The CLSA is a substantial nationwide longitudinal study that provides detailed information on a population‐based sample of individuals who were within the range of 45 and 85 years of age during the time of recruitment. ²³ The CLSA includes 51,338 participants who are followed for 20 years or until death. These individuals underwent an initial baseline assessment and subsequently continued to take part in refined follow‐up evaluations every 3 years. The CLSA offers two different cohorts of participants: Tracking cohort and Comprehensive cohort, with the follow‐up data collection for both cohorts occurring between 2015 and 2018. At the baseline assessment, the Tracking cohort consisted of 21,241 participants who completed a 60‐min telephone interview and who resided in various regions across Canada, whereas the Comprehensive cohort consisted of 30,097 participants who were interviewed in one of the 11 in‐person data collection sites in seven provinces located within a 25–50 km radius of their residence. ²⁴ Approval to examine these data was obtained from the CLSA and ethical approval was received from the University of Saskatchewan's Biomedical Research Ethics Board.

We utilized two different cohorts, each consisting of similarly aged individuals with and without migraine, the Comprehensive cohort (n = 27,773) and the Tracking cohort (n = 17,050). In most cases, the same questions were asked in both cohorts, demonstrating the replication of our findings across cohorts, which enhanced the generalizability of the results. We used a complete case analysis approach and missingness varied depending on the measure. Furthermore, we did not delete participants who had some missing data, and as such the sample size slightly differed for each question.

Our study focused on participants who self‐reported either having received a migraine diagnosis from a physician or never having received one. The analyses were stratified by migraine diagnosis and non‐migraine diagnosis; the Comprehensive cohort (n _migraine = 3736, n _{non‐migraine} = 22,974; 14.0% migraine), and the Tracking cohort (n _migraine = 2270, n _{non‐migraine} = 14,759; 13.3% migraine).

Migraine

The prevalence of migraine was determined through the participants answering, “yes” to the following dichotomous (“yes/no”) question: “Has a doctor ever told you that you have migraine headaches?”

Sociodemographic factors

For Biological Sex, participants were asked “What was your sex at birth?” and selected a dichotomous option of either male or female.

Health factors

For General Health, participants rated their general health on a 5‐item Likert scale (“excellent,” “very good,” “good,” “fair,” or “poor”), with lower numbers representing better perceived health.

For Mental Health, participants rated their mental health on a 5‐item Likert scale (“excellent,” “very good,” “good,” “fair,” or “poor”), with lower numbers representing better perceived mental health.

Depressive Symptoms were measured using the 10‐item Center for Epidemiological Studies Depression (CES‐D‐10) scale, inquiring about participants depressive symptoms over the past week (e.g., “I was bothered by things that usually don't bother me,” “I had trouble keeping my mind on what I was doing,” and “I felt depressed”). ²⁵ Total score based on summed 10‐item questionnaire ranging from 0 to 30, with higher scores representing greater depressive symptoms. ²⁶ Scores ≥10 = positive screening for depressive symptoms (i.e., experiences depressive symptoms). Scores <10 represented a negative screening for depressive symptoms (i.e., does not experience depressive symptoms).

Psychological Distress was measured with the 10‐item Kessler Psychological Distress Scale, inquiring about participants psychological distress (e.g., “How often did you feel tired out for no good reason,” “How often did you feel nervous,” and “How often did you feel hopeless”). ²⁷ , ²⁸ Total score based on summed questionnaire ranging from 10 to 50, with higher scores representing higher levels of psychological distress. Severity categories: “low” (score 0–15), “moderate” (score 16–21), “high” (score 22–30), “very high” (score 31–50). ²⁹

Satisfaction with Life was measured with the Satisfaction With Life Scale (SWLS) assessing participants judgment of their life satisfaction (e.g., “In most ways my life is close to my ideal,” “The conditions of my life are excellent,” and “I am satisfied with my life”) ³⁰ five‐item questionnaire that participants respond to using a 7‐point Likert scale (ranging from “strongly disagree” to “strongly agree”). Total score based on summed questionnaire ranging from 5 to 35, with higher scores reflecting a higher satisfaction with life. The SWLS further categorization: “extremely dissatisfied” (score 5–9), “dissatisfied” (score 10–14), “slightly dissatisfied” (score 15–19), “neutral” (score 20), “slightly satisfied” (score 21–25), “satisfied” (score 26–30), and “extremely satisfied” (score 31–35).

Psychiatric Comorbidities were measured via dichotomous (“yes/no”) questions regarding the presence of several psychiatric comorbidities, including clinical depression, mood disorders, and anxiety disorders.

For Diagnosed Clinical Depression, participants were asked, “Has a doctor ever told you that you suffer from clinical depression?”

For Diagnosed Mood Disorders, participants were asked, “Has a doctor ever told you that you have a mood disorder such as depression (including manic depression), bipolar disorder, mania, or dysthymia?”

For Diagnosed Anxiety Disorders, participants were asked, “Has a doctor ever told you that you have an anxiety disorder such as a phobia, obsessive‐compulsive disorder or a panic disorder?”

For Oral Health, participants rated their overall oral health on a 5‐item Likert scale (“excellent,” “very good,” “good,” “fair,” or “poor”), with lower numbers representing better perceived health.

For Dental Visit Frequency, participants were asked “When did you last visit a Dental Professional (e.g., dentist, dental hygienists, denturist, denturologist)?” With the following response options: “In the last 12 months,” “In the last 5 years,” “In the last 10 years,” “More than 10 years ago,” or “Never visited a dentist”.

Specific Oral Health Concerns were measured via dichotomous (“yes/no”) questions asking participants if they have experienced any of the following oral health problems over the past 12 months: the gums around their natural teeth being sore, bleeding gums around their natural teeth, or their natural teeth becoming loose.

Cognitive factors

Subjective Cognitive Decline was measured through a dichotomous (“yes/no”) question about participants’ self‐perception of their memory becoming worse. In addition, a 5‐item Likert scale question was used to assess participants’ level of agreement or disagreement that they worry about their perceived memory decline.

Meta‐Memory was measured through the ability portion of the Multifactorial Memory Questionnaire (MMQ‐Ability), testing participants’ self‐perception of their everyday memory function. ³¹ Respondents rated the frequency of experiencing 20 common memory mistakes over the past 14 days, which was summed to create a total score ranging between 0 and 80, with higher scores representing greater self‐reported memory function.

Memory Problem was measured via a dichotomous (“yes/no”) question asking participants “Has a doctor ever told you that you have a memory problem?”

For Neuropsychological Performance, all test scores were corrected for language of administration (French vs. English), age, biological sex, and education level using normative data. ³²

Rey Auditory Verbal Learning Test (RAVLT) measured participants learning and memory retention, involving participants attempting to remember a 15‐item word list, immediately recall as many items as they could (Rey I), and again recall as many items as possible after a 5‐min delay (Rey II). ³³ , ³⁴ Scores ranged between 0 and 15, with higher scores reflecting better performance on the RAVLT.

Mental Alternation Test (MAT) measured processing speed and cognitive flexibility by asking participants to alternate between number 1–26 and the letters of the alphabet A–Z in ascending order given a 30 s time limit (e.g., 1 = A, 2 = B, 3 = C, 4 = D, etc.). ³⁵ , ³⁶ Total scores ranged from 0 to 52, with higher scores reflecting better performance.

Animal Fluency Test (AFT2) measured verbal fluency by asking participants to name as many animals as possible within 60 s. ³⁷ We used the lenient AFT2 scoring, awarding points for each distinct animal named, unlike the strict AFT1 scoring, which counted only taxonomically separate species. Total scores ranged from 0 to 52, with higher scores reflecting better performance.

Controlled Oral Word Association Test (COWAT) measured verbal fluency by asking participants to name as many words as possible beginning with the letters, “F”, “A”, and “S” in 60 s. ³⁸ Total scores ranging from 3 to 105, with higher scores reflecting better performance. ³⁹

Stroop Neuropsychological Screening Test‐Victoria version (VST) measured cognitive flexibility, attention, and processing speed by having participants name the color of dots, name the color of neutral words, and name the color of the words that were displayed in incongruent colors. ⁴⁰ , ⁴¹ Total scores ranged from 0.05 to 38.06, with lower scores reflecting better performance. ³⁹

Memory Latent Construct Scores were comprised of scores from Rey I and Rey II.

Executive Latent Construct Scores were comprised of scores from AFT2, MAT, COWAT, and VST.

Overall Cognitive Latent Construct Scores combined a variety of neuropsychological tests that examined both memory and executive function. ⁴²

Pain factors

Pain and Discomfort was explored through three central questions. First, participants were asked a dichotomous (“yes/no”) question, “Are you usually free of pain or discomfort?” Then, participants were asked to describe the intensity of their pain or discomfort (“mild,” “moderate,” or “severe”). Lastly, participants were asked, “How many activities does your pain or discomfort prevent?” The response options included: “None,” “A few,” “Some,” or “Most.”

For Activities of Daily Living the basic and instrumental activities of daily living classification was based on the Older Americans Resources and Services scale (OARS). The OARS scale assesses individuals’ ability to perform common daily activities (e.g., eating and bathing). The classification options include: “No functional impairment,” “Mild impairment,” “Moderate impairment,” “Severe impairment,” or “Total impairment”.

Statistical analysis

All statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) software version 29.0 (IBM Corp., Armonk, NY, USA), and were stratified by migraine diagnosis and non‐migraine diagnosis. The analyses focus on determining if individuals with migraine and without migraine vary across the following factors: health, cognition, and pain. This is a secondary analysis of previously collected data. An alpha value of 0.05 (two‐sided) was used as the cut‐off for statistical significance and was not modified for multiple comparisons as all analyses were planned a priori; however, we prioritized the size of effect to mitigate Type I error inflation, which was possible because of the large samples and multiple comparisons. No statistical power calculation was conducted prior to the study, rather the sample size was based on available data. We visually inspected our continuous data using both histograms and Q–Q plots, revealing that our data did not significantly deviate from a normal distribution. We chose to use chi‐square tests of independence for categorical data, assessing effect size with the phi coefficient and independent samples t‐tests for continuous data, with effect size measured by Cohen's d, to compare participants with and without migraine.

RESULTS

Sociodemographic results

Biological sex

Aligning with epidemiological literature, there is a statistically significant, but small relationship between migraine status and biological sex (Comprehensive cohort: 26.2% males with migraine and 73.8% females with migraine; p < 0.001, φ = 0.18, percentage missingness 3.8%, and Tracking cohort: 26.7% males with migraine and 73.3% females with migraine; p < 0.001, φ = 0.17, percentage missingness 0.1%). As hypothesized, majority of the individuals with migraine within our sample reported being female (Table 1).

TABLE 1.

Sociodemographic factors and migraine; results from the chi‐square test of independence.

Sociodemographic variable	Comprehensive cohort			Tracking cohort
	Migraine count	Non‐migraine count	Effect size/p	Migraine count	Non‐migraine count	Effect size/p
Birth sex, N	3736	22,973	φ = 0.18 p < 0.001	2266	14,741	φ = 0.17 p < 0.001
Male, n (%)	980 (26.2)	12,126 (52.8)		605 (26.7)	7627 (51.7)
Female, n (%)	2756 (73.8)	10,847 (47.2)		1661 (73.3)	7114 (48.3)

Health results

General health

A chi‐square test of independence revealed a statistically significant relationship between migraine status and self‐rated general health (Comprehensive cohort: with participants with migraine rating their general health as “excellent” [14.0%], “very good” [39.0%], “good” [31.1%], “fair” [12.5%], and “poor” [3.4%]; p < 0.001, φ = 0.08, percentage missingness 3.9%, and Tracking cohort: with participants with migraine rating their general health as “excellent” [11.4%], “very good” [37.0%], “good” [33.2%], “fair” [13.8%], and “poor” [4.7%]; p < 0.001, φ = 0.08, percentage missingness 0.4%). Individuals with migraine were more likely to rate their general health as being poorer in comparison to individuals without migraine (Table 2).

TABLE 2.

Health factors and migraine; results from chi‐square test of independence.

Health variable	Comprehensive cohort			Tracking cohort
	Migraine count	Non‐migraine count	Effect size/p	Migraine count	Non‐migraine count	Effect size/p
Self‐rated general health, N	3731	22,947	φ = 0.08 p < 0.001	2267	14,723	φ = 0.08 p < 0.001
Excellent, n (%)	522 (14.0)	4402 (19.2)		258 (11.4)	2389 (16.2)
Very good, n (%)	1455 (39.0)	9638 (42.0)		838 (37.0)	5996 (40.7)
Good, n (%)	1161 (31.1)	6715 (29.3)		752 (33.2)	4427 (30.1)
Fair, n (%)	468 (12.5)	1829 (8.0)		312 (13.8)	1568 (10.7)
Poor, n (%)	125 (3.4)	363 (1.6)		107 (4.7)	343 (2.3)
Self‐rated mental health, N	3725	22,949	φ = 0.08 p < 0.001	2266	14,739	φ = 0.07 p < 0.001
Excellent, n (%)	739 (19.8)	6013 (26.2)		464 (20.5)	3696 (25.1)
Very good, n (%)	1513 (40.6)	9692 (42.2)		885 (39.1)	6226 (42.2)
Good, n (%)	1115 (29.9)	5932 (25.8)		687 (30.3)	3896 (26.4)
Fair, n (%)	303 (8.1)	1146 (5.0)		191 (8.4)	801 (5.4)
Poor, n (%)	55 (1.5)	166 (0.7)		39 (1.7)	120 (0.8)
Clinical depression diagnosis, N	3705	22,839	φ = 0.12 p < 0.001	–	–	–
Yes, n (%)	954 (25.7)	2971 (13.0)		–	–	–
No, n (%)	2751 (74.3)	19,868 (87.0)		–	–	–
Mood disorder diagnosis, N	3723	22,931	φ = 0.14 p < 0.001	2268	14,737	φ = 0.13 p < 0.001
Yes, n (%)	1207 (32.4)	3761 (16.4)		619 (27.3)	1951 (13.2)
No, n (%)	2516 (67.6)	19,170 (83.6)		1649 (72.7)	12,786 (86.8)
Anxiety disorder diagnosis, N	3724	22,934	φ = 0.10 p < 0.001	2266	14,744	φ = 0.11 p < 0.001
Yes, n (%)	623 (16.7)	1928 (8.4)		366 (16.2)	1091 (7.4)
No, n (%)	3101 (83.3)	21,006 (91.6)		1900 (83.8)	13,653 (92.6)
Self‐rated oral health status, N	3705	22,801	φ = 0.02 p = 0.011	2269	14,741	φ = 0.03 p = 0.002
Excellent, n (%)	916 (24.7)	5885 (25.8)		495 (21.8)	3265 (22.1)
Very good, n (%)	1374 (37.1)	8776 (38.5)		851 (37.5)	5662 (38.4)
Good, n (%)	1046 (28.2)	6166 (27.0)		646 (28.5)	4416 (30.0)
Fair, n (%)	280 (7.6)	1562 (6.9)		209 (9.2)	1072 (7.3)
Poor, n (%)	89 (2.4)	412 (1.8)		68 (3.0)	326 (2.2)
Dental visit frequency, N	3709	22,784	φ = 0.01 p = 0.540	2266	14,735	φ = 0.02 p = 0.212
Last 12 months, n (%)	3235 (87.2)	19,865 (87.2)		1890 (83.4)	12,069 (81.9)
Last 5 years, n (%)	372 (10.0)	2196 (9.6)		283 (12.5)	1894 (12.9)
Last 10 years, n (%)	56 (1.5)	396 (1.7)		48 (2.1)	391 (2.7)
10+ years ago, n (%)	46 (1.2)	321 (1.4)		44 (1.9)	375 (2.5)
Never, n (%)	0 (0.0)	6 (0.0)		1 (0.0)	6 (0.0)
Gums sore, N	3712	22,815	φ = 0.04 p < 0.001	2270	14,759	φ = 0.02 p = 0.003
Yes, n (%)	470 (12.7)	2201 (9.6)		230 (10.1)	1216 (8.2)
No, n (%)	3242 (87.3)	20,614 (90.4)		2040 (89.9)	13,543 (91.8)
Gums bleeding, N	3712	22,815	φ = 0.03 p < 0.001	2270	14,759	φ = 0.02 p = 0.003
Yes, n (%)	570 (15.4)	2788 (12.2)		288 (12.7)	1560 (10.6)
No, n (%)	3141 (84.6)	20,027 (87.8)		1982 (87.3)	13,199 (89.4)
Natural loose tooth, N	3712	22,815	φ = 0.009 p = 0.140	2270	14,759	φ = 0.007 p = 0.372
Yes, n (%)	250 (6.7)	1393 (6.1)		155 (6.8)	935 (6.3)
No, n (%)	3462 (93.3)	21,422 (93.9)		2115 (93.2)	13,824 (93.7)

Mental health

Both cohorts demonstrated a statistically significant, small association between migraine status and self‐rated mental health (Comprehensive cohort: with participants with migraine rating their mental health as “excellent” [19.8%], “very good” [40.6%], “good” [29.9%], “fair” [8.1%], and “poor” [1.5%]; p < 0.001, φ = 0.08, percentage missingness 4.0%, and Tracking cohort: with participants with migraine rating their mental health as “excellent” [20.5%], “very good” [39.1%], “good” [30.3%], “fair” [8.4%], and “poor” [1.7%]; p < 0.001, φ = 0.07, percentage missingness 0.3%). Individuals with migraine were more likely to rate their mental health as being poorer in comparison to individuals without migraine (Table 2).

There was also a statistically significant relationship between migraine status and depressive symptoms given the higher mean (standard deviation [SD]) total score of participants with migraine on the CES‐D‐10 scale (Comprehensive cohort: mean [SD] score 6.2 [5.2]; p < 0.001, d = 0.33, percentage missingness 4.2%, and Tracking cohort: mean [SD] score 6.6 [5.5]; p < 0.001, d = 0.25, percentage missingness 2.2%). This relationship displayed a small to moderate effect size. As hypothesized, individuals with migraine were more likely to have a higher total score on the CES‐D‐10 scale, reflecting more depressive symptoms, in comparison to individuals without migraine (Table 3).

TABLE 3.

Mental health factors and migraine; results of independent t‐tests.

Mental health variable	Comprehensive cohort			Tracking cohort
	Migraine	Non‐migraine	Effect size/p	Migraine	Non‐migraine	Effect size/p
Depressive symptoms total score (CES‐D‐10), mean (SD)	6.2 (5.2) n = 3710	4.8 (4.4) n = 22,862	d = 0.33 p < 0.001	6.6 (5.5) n = 2218	5.4 (4.7) n = 14,430	d = 0.25 p < 0.001
Psychological distress total score (K10), mean (SD)	15.1 (5.4) n = 3705	13.6 (4.2) n = 22,739	d = 0.34 p < 0.001	–	–	–
Satisfaction with life total score (SWLS), mean (SD)	27.4 (7.0) n = 3693	28.5 (6.2) n = 22,687	d = 0.18 p < 0.001	27.6 (7.1) n = 2198	28.7 (6.3) n = 14,288	d = 0.18 p < 0.001

Abbreviations: CES‐D‐10, 10‐item Center for Epidemiological Studies Depression scale; K10, 10‐item Kessler Psychological Distress Scale; SD, standard deviation; SWLS, Satisfaction With Life Scale.

Similarly, there was a statistically significant relationship between migraine status and psychological distress given the higher mean score of participants with migraine (Comprehensive cohort: mean [SD] 15.1 [5.4]; p < 0.001, d = 0.34, percentage missingness 1.1%), the magnitude of effect was small to moderate in size. As hypothesized, individuals with migraine were more likely to score higher on the 10‐item Kessler Psychological Distress Scale, reflecting a higher level of psychological distress in comparison to individuals without migraine (Table 3).

Both cohorts produced a statistically significant association between migraine status and scores on the SWLS given the lower mean score of participants with migraine (Comprehensive cohort: mean [SD] score 27.4 [7.0]; p < 0.001, d = 0.18, percentage missingness 1.3%, and Tracking cohort: mean [SD] score 27.6 [7.1]; p < 0.001, d = 0.18, percentage missingness 3.2%), the magnitude of effect was relatively small in size. Therefore, individuals with migraine had a significantly lower total score on the SWLS, reflecting they were less satisfied with their life, in comparison to individuals without migraine (Table 3).

Several psychological diagnoses also produced a statistically significant association with migraine status, including diagnosed anxiety disorder (Comprehensive cohort: 16.7%; p < 0.001, φ = 0.10, percentage missingness 4.0%, and Tracking cohort: 16.2%; p < 0.001, φ = 0.11, percentage missingness 0.2%). Individuals with migraine had a higher likelihood of also having a diagnosed anxiety disorder in comparison to individuals without migraine. There was a significant association between diagnosed mood disorder and migraine status (Comprehensive cohort: 32.4%; p < 0.001, φ = 0.14, percentage missingness 4.0%, and Tracking cohort: 27.3%; p < 0.001, φ = 0.13, percentage missingness 0.3%). Individuals with migraine had a higher likelihood of also having a diagnosed mood disorder in comparison to individuals without migraine. Additionally, there was a significant association between clinical depression and migraine status (Comprehensive cohort: 25.7%; p < 0.001, φ = 0.12, percentage missingness 4.4%). Individuals with migraine had a higher likelihood of also having clinical depression in comparison to individuals without migraine (Table 2).

Oral health

There was a statistically significant relationship between migraine status and self‐rated oral health (Comprehensive cohort: with participants with migraine rating their oral health as “excellent” [24.7%], “very good” [37.1%], “good” [28.2%], “fair” [7.6%], and “poor” [2.4%]; p = 0.011, φ = 0.02, percentage missingness 4.6%, and Tracking cohort: with participants with migraine rating their oral health as “excellent” [21.8%], “very good” [37.5%], “good” [28.5%], “fair” [9.2%], and “poor” [3.0%]; p = 0.002, φ = 0.03, percentage missingness 0.2%). Individuals with migraine were more likely to rate their oral health worse than individuals without migraine. Both cohorts displayed no significant difference between individuals with migraine and individuals without migraine in terms of frequency of seeing a dental professional (Comprehensive cohort: p = 0.540, φ = 0.01, percentage missingness 4.6%, and Tracking cohort: p = 0.212, φ = 0.02, percentage missingness 0.3%). There was a statistically significant association between migraine status and the gums around participants natural teeth being sore (Comprehensive cohort: with 12.7% of participants with migraine experiencing sore gums; p < 0.001, φ = 0.04, percentage missingness 4.5%, and Tracking cohort: with 10.1% of participants with migraine experiencing sore gums; p = 0.003, φ = 0.02, percentage missingness 0.1%). Individuals with migraine were more likely to report that the gums around their natural teeth were sore in comparison to individuals without migraine. There was a statistically significant association between migraine status and the gums around participants natural teeth bleeding (Comprehensive cohort: with 15.4% of participants with migraine experiencing bleeding gums, p < 0.001, φ = 0.03, percentage missingness 4.5%, and Tracking cohort: with 12.7% of participants with migraine experiencing bleeding gums, p = 0.003, φ = 0.02, percentage missingness 0.1%). Individuals with migraine were more likely to report that the gums around their natural teeth were bleeding in comparison to individuals without migraine. There was no statistically significant relationship between migraine status and the loosening of natural teeth (Comprehensive cohort: p = 0.140, φ = 0.009, percentage missingness 4.5%, and Tracking cohort: p = 0.372, φ = 0.007, percentage missingness 0.1%) (Table 2).

Cognitive results

Subjective cognitive decline

Both cohorts produced a statistically significant relationship between migraine status and self‐perception of memory worsening (Comprehensive cohort: with 61.8% of participants with migraine perceiving their memory as worsening; p < 0.001, φ = 0.03, percentage missingness 4.5%, and Tracking cohort: with 56.6% of participants with migraine perceiving their memory as worsening; p < 0.001, φ = 0.03, percentage missingness 1.0%). Individuals with migraine were more likely to perceive their memory as worsening in comparison to individuals without migraine. Also, there was a significant association between migraine status and worrying about their perceived memory worsening (Comprehensive cohort: with participants with migraine indicating whether they worry about their perceived memory worsening: “strongly agree” [16.5%], “agree” [40.2%], “undecided” [7.9%], “disagree” [30.3%], and “strongly disagree” [5.1%]; p < 0.001, φ = 0.09, percentage missingness 45.1%, and Tracking cohort: with participants with migraine indicating whether they worry about their perceived memory worsening: “strongly agree” [12.7%], “agree” [41.1%], “undecided” [9.5%], “disagree” [28.3%], and “strongly disagree” [8.4%]; p = 0.004, φ = 0.04, percentage missingness 4.7%). Individuals with migraine were more likely to worry about their perceived memory worsening in comparison to individuals without migraine, but this difference was very small (Table 4). Both cohorts produced a statistically significant relationship between migraine status and a physician reporting a memory problem (Comprehensive cohort: with 3.9% of participants with migraine being told by their physician that they have a memory problem; p < 0.001, φ = 0.06, percentage missingness 4.0%, and Tracking cohort: with 3.8% of participants with migraine being told by their physician that they have a memory problem; p < 0.001, φ = 0.05, percentage missingness 0.2%). Individuals with migraine were more likely to be told by their physician that they have memory problems in comparison to individuals without migraine (Table 4).

TABLE 4.

Cognitive factors and migraine; results from chi‐square test of independence.

Cognitive variable	Comprehensive cohort			Tracking cohort
	Migraine	Non‐migraine	Effect size/p	Migraine	Non‐migraine	Effect size/p
Self‐perception of memory worsening, N	3714	22,799	φ = 0.03 p < 0.001	2254	14,627	φ = 0.03 p < 0.001
Yes, n (%)	2297 (61.8)	12,982 (56.9)		1275 (56.6)	7732 (52.9)
No, n (%)	1417 (38.2)	9817 (43.1)		979 (43.4)	6895 (47.1)
Worry about perceived memory worsening, N	2289	12,966	φ = 0.09 p < 0.001	1268	7694	φ = 0.04 p = 0.004
Strongly agree, n (%)	378 (16.5)	1229 (9.5)		161 (12.7)	804 (10.4)
Agree, n (%)	920 (40.2)	4844 (37.4)		521 (41.1)	2970 (38.6)
Undecided, n (%)	181 (7.9)	1177 (9.1)		121 (9.5)	706 (9.2)
Disagree, n (%)	694 (30.3)	4752 (36.6)		359 (28.3)	2396 (31.1)
Strongly disagree	116 (5.1)	964 (7.4)		106 (8.4)	818 (10.6)
Memory problems, N	3729	22,945	φ = 0.06 p < 0.001	2266	14,742	φ = 0.05 p < 0.001
Yes, n (%)	147 (3.9)	355 (1.5)		85 (3.8)	251 (1.7)
No, n (%)	3582 (96.1)	22,590 (98.5)		2181 (96.2)	14,491 (98.3)

Meta‐memory

An independent t‐test was conducted to explore differences between individuals with migraine and individuals without migraine in relation to self‐perception of everyday memory abilities using the MMQ‐Ability scale. A statistically significant difference was evident between individuals with migraine and those without given the lower score of participants with migraine (Comprehensive cohort: mean [SD] score 56.2 [9.8]; p = 0.003, d = 0.17, percentage missingness 0.6%, and Tracking cohort: mean [SD] score 56.1 [10.2]; p < 0.001, d = 0.13, percentage missingness 2.6%). In both cohorts, individuals with migraine were more likely to have a significantly lower score on the MMQ‐Ability scale compared to those who do not have a migraine diagnosis. Therefore, individuals with migraine tended to have a worse self‐perception of their everyday memory ability in comparison to individuals without migraine, but the difference was small (Table 5).

TABLE 5.

Cognitive factors and migraine; results of independent t‐tests.

Cognitive variable	Comprehensive cohort			Tracking cohort
	Migraine	Non‐migraine	Effect size/p	Migraine	Non‐migraine	Effect size/p
Self‐perception of everyday memory ability, MMQ‐Ability scale score, mean (SD)	56.2 (9.8) n = 3720	57.8 (9.4) n = 22,874	d = 0.17 p = 0.003	56.1 (10.2) n = 2212	57.4 (9.7) n = 14,380	d = 0.13 p < 0.001
Rey I (immediate recall) score, mean (SD)	6.9 (2.2) n = 3229	6.8 (2.1) n = 19,768	d = 0.04 p = 0.024	6.4 (2.2) n = 2168	6.3 (2.3) n = 14,077	d = 0.05 p = 0.023
Rey II (delayed recall) score, mean (SD)	5.1 (2.4) n = 3197	5.0 (2.3) n = 19,543	d = 0.04 p = 0.050	4.7 (2.4) n = 2141	4.7 (2.5) n = 13,911	d = 0.02 p = 0.297
MAT score, mean (SD)	27.1 (7.7) n = 3117	26.8 (7.6) n = 18,791	d = 0.04 p = 0.071	26.2 (7.9) n = 2026	26.3 (7.9) n = 12,966	d = 0.01 p = 0.656
AFT score, mean (SD)	22.5 (6.2) n = 3207	22.1 (6.1) n = 19,697	d = 0.07 p < 0.001	22.3 (6.6) n = 2180	21.7 (6.2) n = 14,138	d = 0.09 p < 0.001
COWAT score, mean (SD)	40.2 (12.3) n = 3444	39.9 (12.2) n = 20,996	d = 0.03 p = 0.170	–	–	–
VST score, mean (SD)	2.1 (0.7) n = 3530	2.1 (0.7) n = 21,635	d = 0.05 p = 0.013	–	–	–
Memory latent construct score, mean (SD)	108.0 (17.6) n = 3197	107.3 (17.0) n = 19,542	d = 0.04 p = 0.062	102.1 (14.8) n = 2138	101.6 (15.0) n = 13,855	d = 0.04 p = 0.102
Executive functioning latent construct score, mean (SD)	101.94 (14.4) n = 3197	101.55 (14.6) n = 17,048	d = 0.03 p = 0.189	99.9 (15.0) n = 2015	99.4 (14.4) n = 12,878	d = 0.04 p = 0.128
Overall cognition latent construct score, mean (SD)	105.7 (15.7) n = 2835	105.1 (15.6) n = 16,944	d = 0.04 p = 0.040	101.7 (14.8) n = 1978	100.9 (14.8) n = 12,590	d = 0.05 p = 0.036

Abbreviations: AFT, Animal Fluency Test; COWAT, Controlled Oral Word Association Test; MAT, Mental Alternation Test; MMQ‐Ability, ability portion of the Multifactorial Memory Questionnaire; SD, standard deviation; VST, Stroop Neuropsychological Screening Test‐Victoria version.

Neuropsychological tests

There was a statistically significant association between migraine status and scores on the Rey I (immediate recall) given the higher mean scores of participants with migraine (Comprehensive cohort: mean [SD] score 6.9 [2.2]; p = 0.024, d = 0.04, percentage missingness 14.1%, and Tracking cohort: mean [SD] score 6.4 [2.2]; p = 0.024, d = 0.05, percentage missingness 4.6%). In both cohorts, individuals with migraine were more likely to have a higher score (i.e., better performance) on the Rey I (immediate recall) portion of the Rey compared to those who do not have a migraine diagnosis. In contrast, there was no significant difference between individuals with migraine and individuals without migraine in terms of their Rey II (delayed recall) scores (Comprehensive cohort: p = 0.050, d = 0.04, percentage missingness 15.1%, and Tracking cohort: p = 0.297, d = 0.02, percentage missingness 5.7%). Both cohorts used the MAT to measure executive function, and neither cohort produced statistically significant differences between individuals with migraine and those without (Comprehensive cohort: p = 0.071, d = 0.04, percentage missingness 18.2%, and Tracking cohort: p = 0.656, d = 0.01, percentage missingness 12.0%). However, both cohorts found a significant relationship between migraine status and scores on the AFT2 given the higher mean score of participants with migraine (Comprehensive cohort: mean [SD] 22.5 [6.2]; p < 0.001, d = 0.07, percentage missingness 14.4%, and Tracking cohort: mean [SD] 22.3 [6.6]; p < 0.001, d = 0.09, percentage missingness 4.2%). In both cohorts, individuals with migraine tended to have a higher score (i.e., better performance) on the AFT2 in comparison to individuals without migraine, but the effect was small. The remainder of the tests that examined executive function were exclusively used in the Comprehensive cohort. There was not a statistically significant relationship between migraine status and score on the COWAT (Comprehensive cohort: p = 0.170, d = 0.03, percentage missingness 10.5%), thus, there was not a meaningful difference between individuals with migraine and individuals without migraine on this test. Finally, there was a statistically significant relationship between migraine status and the scores on the VST given the higher mean score of participants with migraine (Comprehensive cohort: mean [SD] 2.1 [0.7]; p = 0.013, d = 0.05, percentage missingness 7.8%). Individuals with migraine tended to have a higher score (i.e., worse performance) on the VST in comparison to individuals without migraine, but this difference was very small (Table 5).

Pain results

Pain and discomfort

Both cohorts displayed a statistically significant relationship between migraine status and report of being free from pain or discomfort (Comprehensive cohort: with 53.4% of participants with migraine reporting being free from pain and discomfort; p < 0.001, φ = 0.13, percentage missingness 4.0%, and Tracking cohort: with 47.2% of participants with migraine reporting being free from pain and discomfort; p < 0.001, φ = 0.10, percentage missingness 0.3%). Individuals with migraine were less likely to report being free from pain or discomfort in comparison to individuals without migraine, but the effect was relatively small. Additionally, there was also a statistically significant association between migraine status and intensity of pain or discomfort (Comprehensive cohort: with participants with migraine reporting the intensity of their pain as “mild” [37.3%], “moderate” [48.8%], and “severe” [13.9%]; p < 0.001, φ = 0.09, percentage missingness 68.8%, and Tracking cohort: with participants with migraine reporting the intensity of their pain as “mild” [29.8%], “moderate” [57.4%], and “severe” [12.8%]; p < 0.001, φ = 0.09, percentage missingness 60.0%). Individuals with migraine were more likely to report the intensity of their pain as being moderate or severe, whereas individuals without migraine were more likely to report the intensity of their pain as being mild. Both cohorts also depicted a significant relationship between migraine status and prevention from activities due to pain or discomfort (Comprehensive cohort: with participants with migraine reporting the frequency of missed activities due to pain: “none” [28.4%], “a few” [31.7%], “some” [24.2%], and “most” [15.6%]; p < 0.001, φ = 0.08, percentage missingness 68.7%, and Tracking cohort: with participants with migraine reporting the frequency of missed activities due to pain: “none” [27.4%], “a few” [29.3%], “some” [25.7%], and “most” [17.7%]; p < 0.001, φ = 0.10, percentage missingness 59.9%). Individuals with migraine were more likely to report missing activities due to pain or discomfort in comparison to individuals without migraine (Table 6).

TABLE 6.

Pain factors and migraine; results from chi‐square test of independence.

Pain variable	Comprehensive cohort			Tracking cohort
	Migraine	Non‐migraine	Effect size/p	Migraine	Non‐migraine	Effect size/p
Usually free from pain or discomfort, N	3722	22,930	φ = 0.13 p < 0.001	2263	14,743	φ = 0.10 p < 0.001
Yes, n (%)	1952 (52.4)	15,967 (69.6)		1068 (47.2)	9074 (61.5)
No, n (%)	1770 (47.6)	6963 (30.4)		1195 (52.8)	5669 (38.5)
Intensity of pain or discomfort, N	1758	6909	φ = 0.09 p < 0.001	1190	5633	φ = 0.09 p < 0.001
Mild, n (%)	655 (37.3)	3145 (45.5)		355 (29.8)	2316 (41.1)
Moderate, n (%)	858 (48.8)	3162 (45.8)		683 (57.4)	2749 (48.8)
Severe, n (%)	245 (13.9)	602 (8.7)		152 (12.8)	568 (10.1)
Activities prevented due to pain or discomfort, N	1762	6933	φ = 0.08 p < 0.001	1188	5646	φ = 0.10 p < 0.001
None, n (%)	501 (28.4)	2525 (36.4)		325 (27.4)	2176 (38.5)
A few, n (%)	559 (31.7)	2158 (31.1)		348 (29.3)	1579 (28.0)
Some, n (%)	427 (24.2)	1459 (21.0)		305 (25.7)	1153 (20.4)
Most, n (%)	275 (15.6)	791 (11.4)		210 (17.7)	738 (13.1)
Basic and instrumental activities of daily living classification, N	3549	22,137	φ = 0.06 p < 0.001	2195	14,251	φ = 0.03 p < 0.001
No impairment, n (%)	2913 (82.1)	19,432 (87.8)		1801 (82.1)	12,120 (85.0)
Mild impairment, n (%)	553 (15.6)	2286 (10.3)		316 (14.4)	1728 (12.1)
Moderate impairment, n (%)	57 (1.6)	331 (1.5)		60 (2.7)	292 (2.0)
Severe impairment, n (%)	19 (0.5)	56 (0.3)		8 (0.4)	79 (0.6)
Total impairment, n (%)	7 (0.2)	32 (0.1)		10 (0.5)	32 (0.2)

Basic activities of daily living

A chi‐square test of independence revealed a statistically significant association between migraine status and participants basic activities of daily living classification, according to the OARS scale (Comprehensive cohort: with participants with migraine having no impairment [82.1%], mild impairment [15.6%], moderate impairment [1.6%], severe impairment [0.5%], total impairment (0.2%); p < 0.001, φ = 0.06, percentage missingness 7.5%, and Tracking cohort: with participants with migraine having no impairment [82.1%], mild impairment [14.4%], moderate impairment [2.7%], severe impairment [0.4%], total impairment [0.5%]; p < 0.001, φ = 0.03, percentage missingness 3.5%). Individuals with migraine were more likely to have total, severe, moderate, and mild impairment in comparison to those without. Also, individuals without migraine were more likely to have no functional impairment in comparison to individuals with migraine (Table 6).

DISCUSSION

This population‐based cross‐sectional study is the first to strictly focus on the differences in self‐ perception of health, cognition, and pain in middle‐aged and older Canadians with and without migraine. In this older sample, we found individuals with migraine rated their physical, mental, and oral health lower than individuals without migraine. Occasionally the self‐perceptions that older individuals with migraine held aligned with objective evaluations, whereas other times they were misaligned. For instance, individuals with migraine reported decreased ratings of their mental health, aligning with their increased likelihood of depression and anxiety. On the other hand, the self‐perceptions older individuals with migraine held were misaligned in relation to their memory abilities, as they believed their memory was worsening, yet the memory tests displayed only trivial or even advantageous differences between their memory and those without migraine. Further, more than twice as many individuals with migraine reported that a physician has told them that they had memory problems. Also, individuals with migraine reported a higher frequency and intensity of pain and discomfort, resulting in a higher occurrence of missed activities, with higher functional impairment scores.

Prevalence and biological sex

Our findings support a Canadian lifetime prevalence rate in this older adult cohort of 13.7%, which aligns with the global prevalence rates of 14.0%. ¹⁰ Specifically, the prevalence of migraine within our population‐based sample is reflective of the general population, with the Comprehensive cohort being comprised of 14.0% of individuals with migraine, and the Tracking cohort containing 13.3% of individuals with migraine. According to Stovner et al. ¹⁰ the global prevalence of migraine is 14.0%, with 8.6% being male and 17.0% being female. The ratio of males to females with migraine within our sample also closely resembled the general population proportions, with the Comprehensive cohort being comprised of 7.5% males with migraine and 20.3% females with migraine, and the Tracking cohort containing 7.3% males with migraine and 18.9% females with migraine. As hypothesized, a significantly higher proportion of females reported a migraine diagnosis compared to males, with epidemiological research attributing this difference to several factors, including fluctuations in estrogen levels, ¹ , ⁴³ , ⁴⁴ differences in brain structures, ⁴⁵ , ⁴⁶ and genetic predispositions. ⁴⁷ These biological sex differences are imperative considerations when conducting migraine research.

Self‐perception of health

Our study found our middle‐aged and older Canadians with migraine self‐perceived their general, mental, and oral health as being significantly worse than those without migraine. These findings aligned with the Popit et al. ¹² study, which found poor self‐rated health was significantly greater in individuals with migraine in comparison to those without migraine. As such, an individual's own self‐perception of their health (i.e., self‐rated health) can have a massive impact on the trajectory of their life. Jylhä ¹³ mentioned how one's self‐rated health can be used as a rough indicator of their health status, with Popit et al. ¹² adding that self‐rated health is a valid and reliable measure that can be utilized in individuals who do not have cognitive impairment. Furthermore, negative health self‐perception often leads to frequently requiring hospitalization, higher incidence of illnesses, and an increased risk of death. ¹⁴ , ¹⁵ To our knowledge there has been no study on this topic that has provided a rationale for why individuals with migraine rate their general health as being significantly worse in comparison to those without migraine. Given the possible adverse implications involved in having a negative self‐perception of one's own health, it is imperative future researchers determine why individuals with migraine tend to have worse self‐perceptions of their health and determine helpful strategies that could be utilized to combat these negative views.

Mental health

Migraine comorbidities are well‐known and have been confirmed by numerous researchers within this field. Our Canadian‐based study produced results aligning with a high volume of studies, demonstrating older individuals with a migraine diagnosis are more likely to also have a variety of diagnosed psychiatric conditions, including mood disorders, ⁴⁸ , ⁴⁹ depression, ⁵⁰ , ⁵¹ , ⁵² and anxiety disorders. ⁴⁸ , ⁵⁰ , ⁵¹ , ⁵² Studies confirm that depression is a common comorbidity of migraine, with patients with migraine being two to three times more likely to have a depression diagnosis. ⁵⁰ , ⁵¹ , ⁵² Additionally, our study also found that older individuals with migraine reported higher levels of psychological distress, which Kristoffersen et al. ⁵³ also associated with migraine. This was an anticipated finding because psychological distress is a stage of emotional distress, commonly associated with depression and anxiety, ⁵⁴ both of which are migraine comorbidities.

Similarly, our study confirmed our hypothesis that older individuals with migraine would experience lower satisfaction with life in comparison to those without migraine. These findings are aligned with the results from the Eskin et al. ⁵⁵ study, which found that individuals with migraine often report lower levels of satisfaction with life, a finding that reflects the broader psychological impact of the condition. The relationship between satisfaction with life and mental health is further underscored by research demonstrating that low satisfaction with life is not only associated with an increase in chronic conditions, ⁵⁶ , ⁵⁷ but is also an indicator of depressive symptoms. ⁵⁷ These findings highlight the complex interplay between migraine and various psychological factors, emphasizing the importance of exploring both self‐perceived mental health, as well as objective psychological determinants.

Unexpectedly, Pohl et al. ⁵⁸ found a positive association between increased headache frequency and increased satisfaction with life. Future studies should endeavor to build on the Pohl et al. ⁵⁸ findings and determine why there might be a proportional relationship between migraine headache frequency and satisfaction with life and if it is replicable. Outside of migraine research, a variety of studies have demonstrated there is an interrelatedness between low satisfaction with life and both an increase in chronic conditions, ⁵⁶ , ⁵⁷ and a higher prevalence of depressive symptoms. ⁵⁷ Our study has added to the growing literature on migraine and its association with psychiatric comorbidities, psychological distress, and satisfaction with life. Notably, it is one of the first studies to specifically demonstrate these associations within a mid‐to‐late adult population.

Oral health

Most migraine literature does not address oral health or the associated risk factors for individuals with migraine. Interestingly, we found that individuals with migraine tend to report sore and/or bleeding gums more frequently than those without migraine. According to the Centers for Disease Control and Prevention, sore and bleeding gums are early signs of periodontal disease, ⁵⁹ which in its mildest form (i.e., gingivitis), can be reversed with proper modifications to oral hygiene practices. ⁶⁰ However, if adjustments are not made, the periodontal disease can progress into chronic periodontitis, which is irreversible. ⁶¹ Huang et al. ⁶² and Mohammed et al. ⁶³ both highlighted the increased risk factor for individuals with migraine developing chronic periodontitis. This increased risk factor for chronic periodontitis may be related to the increase in neurogenic inflammation that accompanies migraine. ⁶² The prevalence of periodontal disease also increases with age, with 70.1% of United Staes adults aged ≥65 years affected. ⁶⁴ Periodontitis is an oral infection that inflames and damages the gum tissue (i.e., gingivae) around natural teeth, and leads to ligament loss and damage to the alveolar bone (i.e., the structure holding the tooth in place). ⁶⁵ Without proper dental treatment, periodontitis can result in mobile teeth and even tooth loss. ⁶⁶ Given this increased risk, individuals with migraine, particularly middle‐aged and older adults, should prioritize visiting a dental professional with regularity and practicing proper oral hygiene (e.g., brushing teeth twice per day, and flossing regularly).

Cognition

In relation to meta‐memory (i.e., an individual's beliefs about their own everyday memory abilities), ⁶⁷ we found middle‐aged and older Canadians with migraine believe their everyday memory ability is worse in comparison to individuals without migraine. Additionally, in terms of subjective cognitive decline, our study discovered individuals with migraine were more likely to perceive their memory as declining and worry about that perceived memory decline, yet the RAVLT showed very few differences between the memory abilities of individuals with migraine in comparison to those without migraine. Furthermore, the Rey I (i.e., immediate recall) demonstrated that individuals with migraine were more likely to have a slightly higher score (i.e., better performance) and the Rey II (i.e., delayed recall) displayed no significant difference between those with migraine and those without migraine. Considering these findings, we wondered why the individuals with migraine within our study appeared to be believing that they were experiencing memory problems, yet validated and reliable objective tests revealed that they did not have any meaningful differences in these measured memory abilities in comparison to the general population. It is worth considering that our participants with migraine may have developed these negative perceptions about their memory abilities due to advice or evaluations from physicians and other healthcare professionals. Our study found individuals with migraine were more likely to have a doctor tell them that they have memory problems in comparison to individuals without migraine. These types of professional evaluations and the advice that is provided to patients by their physicians or healthcare professionals can have a negative impact on their health self‐perception, which is associated with an increased risk of mortality. ¹⁴ As a potential future direction, it would also be of interest to investigate the CLSA correlational data on age and cognitive performance to better understand the interrelatedness of those variables.

Pain and discomfort and basic activities of daily living

The results from our study confirmed our hypotheses, middle‐aged and older Canadians with migraine are more likely to self‐report experiencing a higher intensity of pain and discomfort, leading to a higher number of missed activities. This increase in the self‐reported pain in individuals with migraine has previously been attributed to central sensitization, which is the upsurge in the efficiency of the nociceptive neurons within the central nervous system. ⁶⁸ , ⁶⁹ Central sensitization can result in small irritations leading to highly intense levels of pain, ⁶⁸ which has been shown to impact a variety of areas in one's life (e.g., work, family, socialization, and education). ⁷⁰ Our study also found middle‐aged and older adults with migraine were more likely to have total, severe, moderate, and mild impairment on the OARS scale in comparison to those without migraine. This finding is logical because migraine was ranked as the third highest cause of years lived with disability globally (i.e., among all ages and sexes). ⁷¹ Therefore, it was anticipated that individuals with migraine would have more impairments in their basic activities of daily living in comparison to individuals without migraine. Future longitudinal studies should investigate the OARS classification of individuals with migraine in relation to their self‐reported pain level to determine if there is any type of causal relationship observed over time.

Strengths

One major strength of this research is the large middle‐aged and older Canadian population‐based sample (n = 44,823), which provided a sizable proportion of individuals with migraine (n = 6006). The large sample size allowed for higher statistical power, which was beneficial in providing a greater sense of confidence in the accuracy of our results. Additionally, using the two CLSA cohorts, the Comprehensive and Tracking cohorts, provided us with a unique opportunity to replicate our findings across groups, reinforcing the consistency and generalizability of our results.

Limitations

A limitation of our study was the cross‐sectional format, which relied partially on self‐report measures. These measures have a few potential disadvantages that may impact internal validity, including inaccuracies in recollection and susceptibility of producing socially desirable responses. An additional limitation of our study was that the survey did not ask the participants if they still currently experience migraine attacks, whether the headache symptoms have ceased, and if so, at what age the headache symptoms stopped. As noted above, there are several benefits to having a large population‐based sample; however, large sample sizes also tend to transform marginal differences within the sample into statistically significant differences, many of which would be considered clinically insignificant. Furthermore, our study's large sample size and use of multiple comparisons can be problematic due to the possibility of an inflated Type 1 error rate. Therefore, in this study it was important to conceptualize each finding, compare cohorts, and ensure that we were not simply reporting on every statistically significant result.

CONCLUSION

Our study provided a detailed overview of perceptions of health, cognition, and pain, and how those perceptions aligned with objective health metrics within a sample of middle‐aged and older adults with migraine as compared to those without migraine. Of note, our findings support a Canadian lifetime prevalence rate in this older adult cohort of 13.7%, which aligns with the global prevalence rates of 14.0%. ¹⁰ Importantly, while these older individuals with migraine accurately perceived their general health, mental health, and pain as worse than those without migraine, they also perceived their memory as declining more than those without migraine, but the objective testing did not support this. We hope that these findings will encourage future researchers and clinicians to be aware of the potential misalignment between a middle‐aged or older adult with migraine who perceives their memory as declining when it may not be, and to consider that their overall health and quality of life may be lower than someone who never had a migraine diagnosis.

AUTHOR CONTRIBUTIONS

Study concept and design: Makenna K. N. Jensen, Megan E. O'Connell, Marla J. S. Mickleborough. Acquisition of data: Makenna K. N. Jensen, Megan E. O'Connell, Marla J. S. Mickleborough. Analysis and interpretation of data: Makenna K. N. Jensen. Drafting of the manuscript: Makenna K. N. Jensen. Revising it for intellectual content: Makenna K. N. Jensen, Megan E. O'Connell, Marla J. S. Mickleborough. Final approval of the completed manuscript: Makenna K. N. Jensen, Megan E. O'Connell, Marla J. S. Mickleborough.

FUNDING INFORMATION

This research was made possible using the data collected by the Canadian Longitudinal Study on Aging (CLSA). Funding for the CLSA is provided by the Government of Canada through the Canadian Institutes of Health Research (CIHR) under grant reference: LSA 94473 and the Canada Foundation for Innovation, as well as the following provinces, Newfoundland, Nova Scotia, Quebec, Ontario, Manitoba, Alberta, and British Columbia. This research has been conducted using the CLSA dataset Follow‐up 1 Tracking Dataset version 3.1, Follow‐up 1 Comprehensive Dataset version 5.0, under Application Number 2301017. The CLSA is led by Drs. Parminder Raina, Christina Wolfson, and Susan Kirkland.

CONFLICT OF INTEREST STATEMENT

Makenna K.N. Jensen, Megan E. O'Connell, and Marla J.S. Mickleborough declare there is no conflict of interest.

DISCLAIMER

The opinions expressed in this manuscript are the authors’ own and do not reflect the views of the Canadian Longitudinal Study on Aging.

Jensen MKN, O’Connell ME, Mickleborough MJS. Perceptions of health, cognition, and pain among middle‐aged and older adults with migraine: A population‐based cross‐sectional study examining findings from the Canadian Longitudinal Study on Aging. Headache. 2025;65:1511‐1525. doi: 10.1111/head.14953

DATA AVAILABILITY STATEMENT

Data are available from the Canadian Longitudinal Study on Aging (CLSA; www.clsa‐elcv.ca) for researchers who meet the criteria for access to de‐identified CLSA data.