A migraine attack occurred somewhere that smelled of perfume or fabric softener. When studying such experiences, clues lie not only in which odors are named but in which odors the same person reports together. Ranking product names by frequency alone cannot fully describe the relationship between fragrance and migraine.
In 2023, Noboru Imai and colleagues classified odors reported as associated with attacks using responses from 101 migraine patients attending a Japanese headache clinic. Perfume and fabric softener were frequently mentioned, but the research focused on finding common response patterns and examining how they related to migraine type and other characteristics.
- Asking 101 patients about their experiences of odors and attacks
- What do the figures of 55.4% for perfume and 32.7% for fabric softener mean?
- What emerges when odors are divided into six groups?
- How should the association with chronic migraine be interpreted?
- Which comes first: the odor or changes accompanying an attack?
- Summary
Asking 101 patients about their experiences of odors and attacks
Participants attended the Department of Neurology at Japanese Red Cross Shizuoka Hospital between April and September 2020 and were diagnosed with migraine by a specialist using international diagnostic criteria. Of 102 enrolled patients, 101 were analyzed after excluding one with insufficiently completed information. Ages ranged from 22 to 75, and 16 had chronic migraine.
The questionnaire listed 35 odors and asked about associations with attacks, related locations, and ways of coping with odors. Free-text responses allowed experiences absent from the options to be reported. This was a retrospective survey of past experiences, not an experiment comparing attacks after exposure to equal odor concentrations.
What do the figures of 55.4% for perfume and 32.7% for fabric softener mean?
Seventy-nine of the 101 patients, or 78.2%, reported an association between some odor and attacks. With all 101 as the denominator, odor-specific proportions included 55.4% for perfume, 47.5% for tobacco, 32.7% for fabric softener, and 32.7% for body odor. People could report multiple odors, so these are not mutually exclusive categories.
These figures describe which odors the participating migraine patients linked to their experiences. They do not mean that a person in the general population has a 55.4% probability of an attack after smelling perfume. Exposure frequency and concentration were not standardized, so reporting frequency cannot simply be translated into product risk.
What emerges when odors are divided into six groups?
The researchers used a statistical method called factor analysis to organize combinations of responses: what other odors people mentioned alongside a particular odor. Six groups emerged: (1) fetid odors, (2) cooking-related products, (3) oil derivatives and others, (4) shampoo and conditioner, (5) cleaning products, and (6) perfumes, insecticides, and rose. The fifth group included hair-styling preparations, laundry detergent, and fabric softener.
These categories bring together similar response patterns and were named by the researchers. Chemical analyses did not establish that every product in a group contained identical ingredients or related to attacks through the same mechanism.
For example, shampoo and laundry-detergent groups can both contain floral fragrances, yet formed separate response clusters. The authors suggested that differences in temperature, humidity, and mental state between bathing and everyday environments might contribute. This explanation was not directly tested, but it suggests studying the circumstances of exposure as well as fragrance ingredients.
Using response patterns for the six odor groups to classify patients then produced three patient clusters. Rather than reporting all odors similarly, patients differed in which groups they associated with attacks. This exploratory result offers a clue for studying differences between patients; it does not establish a new diagnostic classification.
How should the association with chronic migraine be interpreted?
Patients with chronic migraine showed a stronger association between attacks and the fifth group, cleaning products, than those with episodic migraine. The authors considered the possibility that sensitivity to some chemicals increases as migraine becomes chronic.
However, the study compared responses from people with different migraine types at one point in time. It did not follow individuals before and after chronification, so it cannot establish whether chronification increased sensitivity or whether other factors overlapped with odor responses. Only 16 participants had chronic migraine, and 12 of them also had medication-overuse headache. The factors underlying the observed association need to be disentangled.
Which comes first: the odor or changes accompanying an attack?
From the paper’s introduction, the authors advocate initially treating presumed triggers as associated factors. Retrospective responses alone make it difficult to distinguish an odor provoking an attack from an odor being interpreted as a trigger because sensitivity had already increased early in an attack.
This distinction is not about deciding whether a person’s distress is real. It asks what occurred, and in what order, while acknowledging the experience of finding an odor distressing and linking it with an attack. Recall bias and differences in attribution are reasons to improve study methods, not to disregard people’s reports.
The survey also took place at a single outpatient clinic serving as a regional headache center, which may attract more severely affected patients. Its results cannot be generalized to all migraine patients or everyone reporting fragrance-related ill health.
The authors also suggested that identifying associated odors might make avoidance useful for preventing attacks. However, the survey did not test avoidance. Questions about masks, air cleaners, and other measures asked about use rather than comparing effectiveness. An important next step is prospective recording of the time of odor exposure, the onset of odor discomfort, and the course of headache, to test whether these associations recur.
Summary
The study showed that migraine patients’ experiences were associated with multiple odors, including perfume and fabric softener, and that the responses could be organized into six groups. The association between the cleaning-products group and chronic migraine also provides a lead for further investigation.
Looking beyond reporting frequency to differences between patients, exposure circumstances, and the sequence of events around attacks can move research from organizing experiences toward investigating their basis. These findings do not assign a single risk to all perfumes; they are a starting point for more specific investigation of links between fragrance and migraine.
References
Authors: Noboru Imai, Ayako Osanai, Asami Moriya, Masahito Katsuki, Eiji Kitamura
Article: Classification of odors associated with migraine attacks: a cross-sectional study
Journal: Scientific Reports 13, 8469 (2023)
DOI: 10.1038/s41598-023-35211-7
Full original article: Open-access full text in Scientific Reports
Rights: The original article is published under Creative Commons Attribution 4.0 International (CC BY 4.0).
