Headache. 2026;00:1–13. �  | 1wileyonlinelibrary.com/journal/head Received: 17 July 2025  | Accepted: 4 February 2026 DOI: 10.1111/head.70068 R E S E A R C H S U B M I S S I O N Conventional cigarettes, novel heated tobacco products, and self-reported frequent headache association in Japanese individuals: Insights from the JASTIS study, a Japanese cross-sectional analysis Masahito Katsuki MD, PhD1,2,3,4  | Kieran Moran BSc, PhD2,5,6 | Muneto Tatsumoto MD, PhD7,8  | Keisuke Suzuki MD, PhD8  | Tomokazu Shimazu MD, PhD9  | Yasuhiko Matsumori MD, PhD10 | Yutaro Fuse MD, PhD11  | Miguel Á. Huerta PharmD12,13,14  | Daijiro Sugiyama PharmD, PhD15 | Jackson T. S. Cheung MBBS, BSc16 | Siobhán O'Connor MSc, PhD, CAT, FFSEM (Hon.)2 | Tomás Ward BE, MEngSc, PhD3 | Takahiro Tabuchi MD, PhD17 This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. © 2026 The Author(s). Headache: The Journal of Head and Face Pain published by Wiley Periodicals LLC on behalf of American Headache Society. For affiliations refer to page 11. Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; CO, carbon monoxide; HTPs, heated tobacco product; ICHD-3, International Classification of Headache Disorders, 3rd edition; JASTIS, Japan Society and New Tobacco Internet Survey; JPY, Japanese yen; TAC, trigeminal autonomic cephalalgia; TTH, tension-type headache. Correspondence Siobhán O'Connor, School of Health and Human Performance, Dublin City University, Collins Avenue Ext, Whitehall, Dublin 9, D09 V209, Ireland. Email: siobhan.oconnor@dcu.ie Funding information Japan Society for the Promotion of Science (JSPS) KAKENHI Grant, Grant/Award Number: JP20K19633, JP21H04856, JP23K21579, JP24H00663 and JP25H01079; National Institute for Environmental Studies; Taighde Éireann – Research Ireland, Grant/Award Number: 12/RC/2289_P2; Health Labor Sciences Research Grants, Grant/Award Number: 22FA1002 and 23FA1004; The European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant, Grant/Award Number: 101034252 Abstract Objectives/Background: Heated tobacco products (HTPs) are novel alternatives to conventional cigarettes that heat tobacco leaves to generate nicotine-containing aerosols for inhalation. This study aimed to investigate the association between both conventional cigarette use and HTP use with self-reported frequent headache. Although the association between cigarette smoking and headache has been re- ported, evidence regarding HTPs remains limited. Methods: We utilized data from the Japan Society and New Tobacco Internet Survey, an ongoing internet-based cohort cross-sectional study launched in 2015 to monitor conventional cigarette and HTP use. Participants were recruited from a nationally repre- sentative online panel, comprising over 2 million individuals. Among 28,000 respondents in the 2025 wave, we included 23,228 individuals with valid responses. Self-reported frequent headache was defined as responding “occasionally” or “frequently” to the question, “Have you experienced headaches in the past year?” (other options: “never” and “rarely”). We conducted a multivariable analysis adjusting for demographic, socio- economic, lifestyle, and clinical variables. We then compared the adjusted odds ratios (aORs) of conventional cigarettes and HTP use for the prevalence of self-reported fre- quent headache. www.wileyonlinelibrary.com/journal/head https://orcid.org/0000-0002-0192-5430 https://orcid.org/0000-0002-1950-0756 https://orcid.org/0000-0003-0320-8993 https://orcid.org/0000-0001-9146-7770 https://orcid.org/0000-0001-9402-8079 https://orcid.org/0000-0003-2842-0085 mailto: http://creativecommons.org/licenses/by-nc/4.0/ mailto:siobhan.oconnor@dcu.ie 2  |    HEADACHE INTRODUC TION Primary headache disorders, such as migraine, tension-type head- ache (TTH), and trigeminal autonomic cephalalgias (TACs), as classi- fied by the International Classification of Headache Disorders, 3rd edition (ICHD-3),1 are prevalent neurological conditions that cause a significant global socioeconomic burden.2 Effective management requires the appropriate use of both acute and prophylactic medi- cations,3 as well as an accurate diagnosis under the ICHD-3 crite- ria.1 In addition to pharmacological treatments, lifestyle assessment using a headache diary is essential for identifying and avoiding the risk factors associated with headaches. Some studies identified that smoking conventional cigarettes may be a modifiable risk factor for primary headache disorders, including migraine,4 TTH,5,6 and TACs.7 Therefore, avoiding smoking conventional cigarettes may be one of the ways to reduce the risk of having headache disorders. Smoking conventional cigarettes, including exposure to second- hand or passive smoke, is widely recognized as harmful to human health, contributing to a broad spectrum of diseases and increased mortality.8 In response to growing public awareness and regulatory pressures, tobacco companies have introduced heated tobacco products (HTPs) as a novel alternative. Unlike conventional ciga- rettes, which burn tobacco leaves, HTPs just heat tobacco leaves without burning, producing an aerosol that is inhaled by the user.9 These products are marketed as emitting fewer harmful substances compared to conventional cigarettes10 and have rapidly gained pop- ularity in Japan after their introduction in 2013.11 The prevalence of HTP use rose markedly from 0.3% in 2015 to 3.6% in early 2017.11 In 2023, the prevalence of current use was 5.0% (only HTPs), 11.5% (only cigarettes), and 7.4% (dual use).12 As HTPs are relatively new globally, only one study investigated the relationship between HTPs and headache, reporting that 17.7% Results: Of the 23,228 participants, 25.5% (5923/23,228) reported headaches. Regarding tobacco use, 12.2% (2823/23,228) were current cigarette users, 31.8% (7393/23,228) were former cigarette users, 56.0% (13,012/23,228) were never ciga- rette users; 11.6% (2688/23,228) were current HTP users, 10.0% (2326/23,228) were former HTP users, and 78.4% (18,214/23,228) were never HTP users. Multivariable analysis with multiple imputation revealed that current (aOR = 1.71 [95% confidence interval [CI] = 1.44–2.03]) and former (aOR = 1.54 [95% CI = 1.37–1.73]) cigarette use, and current (aOR = 1.15 [95% CI = 1.01–1.32]) HTP use were significantly associated with self-reported frequent headache. Former HTP use was not significantly associ- ated with headache (aOR = 1.09 [95% CI = 0.96–1.24]). Postestimation comparisons indicated a weaker association for HTPs than for cigarettes (current smokers vs. cur- rent HTP users: ratio of aORs = 1.48 [95% CI = 1.19–1.84]). Conclusion: Both current and former cigarette use were significantly associated with a higher prevalence of self-reported frequent headache, and current HTP use also showed a modest but significant association. Although HTPs are often perceived as less harmful, they still emit nicotine and other constituents, which may be associated with self-reported frequent headache. Plain Language Summary Headache is a common health problem, and tobacco products, such as conventional cigarettes and heated tobacco products (HTPs), may be associated with it. Using data from over 23,000 adults in the Japan Society and New Tobacco Internet Survey, we examined the relationships between cigarette and HTP use and self-reported frequent headache. We found that current use of either cigarettes or HTPs was associated with more frequent headaches compared to those who had never smoked; however, the magnitude of this difference was greater among cigarette smokers, suggesting that although the risk of frequent headaches may be less among HTP smokers compared to cigarette smokers, HTP use still has harmful effects on health. K E Y W O R D S cluster headache, electronic cigarettes, heated tobacco products, migraine, tension-type headache, trigeminal autonomic cephalalgias     | 3HEADACHE of individuals experienced headaches after passive smoking of HTPs.13 However, no prior studies have investigated whether HTPs and conventional cigarettes differ in their impact on the prevalence of headache disorders. To address these gaps, this study aimed to examine the association between both conventional cigarette use and HTP use with self-reported frequent headache in a large na- tionwide Japanese cohort. Utilizing data from the 2025 wave of the Japan Society and New Tobacco Internet Survey (JASTIS),9 we in- vestigated the impact of conventional cigarettes and HTP use on the prevalence of self-reported frequent headache, adjusting for socio- demographic, clinical, and behavioral factors. We hypothesized that individuals who are current or former users of either conventional cigarettes or HTPs would have significantly higher odds of reporting frequent headaches compared to individuals who have never used them. METHODS Study sample and data source This study used data from the JASTIS, a nationwide longitudinal internet-based cohort study.9 Participants were recruited through Rakuten Insight, a major internet research company in Japan, which maintains a large-scale panel of approximately 2.3 million individu- als representing a diverse cross-section of the population consistent with the Japanese national census demographics (e.g., education, housing tenure, marital status). For each survey, quota sampling stratified by biological sex, 5-year age groups, and geographic re- gion was employed to ensure demographic balance consistent with the national census distribution. For this analysis, we used the 2025 data set, which was collected from February 23, 2025 to March 31, 2025. The target number of respondents was set at n = 28,000, and the survey was terminated once this predefined target was reached. From this data set, cross-sectional data from a total of 28,000 re- spondents were analyzed. Exclusion criteria To ensure data quality and exclude potentially careless or implausible responses, participants were excluded if they met any of the follow- ing criteria14: (1) failure to correctly answer an attention check item that required selecting the second-to-last choice from a list of five options; (2) reporting an implausibly large household size of 15 or more family members; (3) indicating the presence of all nine queried medical conditions, covering metabolic and cardiovascular diseases (hypertension, diabetes, angina pectoris, or myocardial infarction), respiratory and allergic diseases (asthma, atopic dermatitis, allergic rhinitis), ocular disease (cataracts), immune disorders, and cancer or malignant tumors; or (4) claiming use of all five categories of illicit drugs listed, including prescription-free narcotics (e.g., morphine), organic solvents (e.g., thinner, toluene), designer drugs (e.g., syn- thetic herbs, hallucinogenic mushrooms), cannabis, cocaine, and heroin. We also excluded participants who did not report age or biological sex. Headache assessment The JASTIS study, including the 2025 wave, did not assess headache characteristics following the ICHD-3 criteria.1 The survey also did not include questions regarding whether participants had received a physician's diagnosis of headache or were using headache medi- cations. Therefore, frequent headache was defined based on self- report; participants who responded “occasionally” or “frequently” to the question “Have you experienced headaches in the past year?” were classified as a self-reported frequent headache group (the other options were “never” and “rarely”). Smoking assessment Regarding both conventional cigarettes and HTPs, the survey col- lected detailed information on use history. Based on this infor- mation, individuals who had used these products within the past 30 days were classified as current users, whereas those who had used them previously but not within the past 30 days were classified as former users.15 Individuals who had never used these products were classified as never users.14 Electronic cigarettes are devices that produce vapor for inhalation and do not contain tobacco leaves. Because electronic cigarettes containing nicotine are not legally ap- proved in Japan, their use was not assessed in this study. Other variables as confounding factors Appropriate variables were collected as confounding factors to adjust for the multivariable analysis. Demographic variables included age (years), biological sex (male or female), and body mass index (kg/m2).16 Body mass index was categorized as follows: underweight (<18.5), normal weight (≥18.5–<25.0), overweight (≥25.0–<30.0), and obesity (≥30.0). Socioeconomic indicators included: marital status (married or not married, including never married, widowed, and divorced), annual household income level (seven-point ordinal scale; <2 million Japanese yen [JPY], 2–4 million JPY, 4–6 million JPY, 6–8 million JPY, 8–10 mil- lion JPY, 10–12 million JPY, 12 < million JPY), and educational attain- ment (six-point ordinal scale; junior high school, high school, technical college, two-year university, university, graduate school).17 The pres- ence of comorbid conditions, such as hypertension,18 diabetes mel- litus,19 dyslipidemia,20 depression,21 and sleep disorders,22 were also recorded. Regarding substance use history, daily consumption of cof- fee, caffeinated soft drinks,23 and alcohol consumption more than twice a week17 were assessed. 4  |    HEADACHE Statistical analysis Descriptive statistics and univariable analyses were performed to examine the association between self-reported frequent headache and tobacco product use. To identify the association between the presence of self-reported frequent headache and both conventional cigarette use and HTP use, we performed a logistic regression model for the multivariable analysis, adjusting the confounding factors. Adjusted odds ratios (aORs) with 95% confidence intervals (CIs) were calculated. Missing data were primarily handled by multiple imputa- tions, and a sensitivity analysis using list-wise deletion was conducted to assess the robustness of the results. A two-tailed p < 0.05 was considered statistically significant. We chose not to directly statisti- cally compare the prevalence of headaches between conventional cigarette and HTP users, because some participants used both prod- ucts or switched between them, resulting in up to nine possible com- binations of use (Table S1 in Supporting Information File 2). Instead, we statistically compared their associations by evaluating the aORs in multivariable analyses. Statistical analyses were conducted using SPSS Statistics version 30.0.0 (IBM, Armonk, NY) and Python 3.9.0. Multiple imputation was performed with the IterativeImputer from scikit-learn 1.6.1, and logistic regression models were fitted using statsmodels 0.14.4. Postestimation comparison of regression co- efficients between cigarette and HTP users was performed using SciPy 1.16.2. These comparisons are intended to be descriptive and hypothesis-generating, rather than to imply causal differences between products. Other detailed methods are described in the Supporting Information File 1. Ethical aspects and reporting guidelines The study was reviewed and approved by the Research Ethics Committee of the Osaka International Cancer Institute (no. 1412175183) and the National Institute of Public Health (NIPH- IBRA#12112). All participants provided written informed consent and accessed the survey via a secure website. The research was conducted in accordance with the principles of the Declaration of Helsinki and adhered to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines for reporting observational studies and the Checklist for Reporting Results of Internet E-Surveys (CHERRIES) guidelines for internet surveys. No identifiable personal data were used in the analysis to ensure par- ticipant privacy. RESULTS General characteristics Of the 28,000 panelists who participated in the JASTIS 2025 sur- vey, 17.0% (4772/28,000) were excluded because they met the ex- clusion criteria. Subsequently, 23,228 participants were included in the analysis. Household annual income (missing = 5462/23,228, 23.5%) and educational background (missing = 349/23,228, 1.5%) had missing values. The median (first quartile to third quartile) age was 49 (35–63) years, and 49.3% (11,448/23,228) were female (Table  1). Of the 23,228 individuals, 25.5% (5923/23,228) reported experiencing headaches “occasionally” or “frequently,” which met our study's definition for self-reported frequent headache. In detail, concerning headache frequency over the past year, 50.8% reported “never” expe- riencing headaches, 23.7% reported “rarely,” 20.1% reported “occa- sionally,” and 5.4% reported “frequently.” Regarding smoking, 12.2% (2823/23,228) were current cigarette users, 31.8% (7393/23,228) were former cigarette users, 56.0% (13,012/23,228) were never cig- arette users, 11.6% (2688/23,228) were current HTP users, 10.0% (2326/23,228) former HTP users, and 78.4% (18,214/23,228) never HTP users. Some respondents used both conventional cigarettes and HTPs concurrently or at different times, suggesting switching behav- ior (Table S1 in Supporting Information File 2). Variables associated with frequent headache in univariable analysis In the univariable analysis comparing the headache and nonhead- ache groups, participants with headaches were younger (median age 42 vs. 52 years, p < 0.001) and more likely to be female (63.2% vs. 44.5%, p < 0.001). The results of univariable analysis for other potential confounding factors are shown in Table 1. The proportion of current cigarette use was not different between the headache and nonheadache groups (12.1% vs. 12.2%), but former cigarette use was less frequent (29.5% vs. 32.6%, p < 0.001), and never cigarette use was more frequent in the headache group (58.4% vs. 55.2%, p < 0.001). In contrast, current HTP use was more common in those with headaches (14.5% vs. 10.6%, p < 0.001), as was former HTP use (11.2% vs. 9.6%, p < 0.001), whereas never HTP use was less com- mon (74.4% vs. 79.8%, p < 0.001) (Table 2). Variables associated with frequent headache in multivariable analysis All multivariable model assumptions were met, with no evidence of nonlinearity or multicollinearity. In the multivariable analysis with multiple imputation in Table  3, in comparing the headache group to the nonheadache group, the headache group was significantly associated with greater cigarette use, both current (aOR = 1.71 [95% CI = 1.44–2.03]; p < 0.001) and former use (aOR = 1.54 [95% CI = 1.37–1.73]; p < 0.001), compared to having never smoked. Similarly, current (aOR = 1.15 [95% CI = 1.01–1.32]; p = 0.038) use of HTPs was associated with headache, but former (aOR = 1.09 [95% CI = 0.96–1.24]; p = 0.203) use was not. Additionally, younger age (aOR = 0.97 [95% CI = 0.97–0.98]; p < 0.001), female sex (aOR = 2.30 [95% CI = 2.14–2.47]; p < 0.001), dyslipidemia (aOR = 1.20 [95%     | 5HEADACHE TA B LE 1   G en er al c ha ra ct er is tic s. Va ria bl e To ta l ( n =  2 3, 22 8) Se lf- re po rt ed fr eq ue nt h ea da ch e (2 5. 5% , n  =  5 92 3) N on -s el f- re po rt ed fr eq ue nt he ad ac he (7 4. 5% , n  =  17 ,3 05 ) p- va lu e A ge (y ea rs ), m ed ia n (IQ R) 49 (3 5– 63 ) 42 ↓ (3 0– 53 ) 52 ↑ (3 7– 65 ) < 0. 00 1* (r : − 0. 21 ) Bi ol og ic al s ex ; f em al e, n (% ) 11 ,4 48 (4 9. 30 ) 37 44 ↑ (6 3. 20 ) 77 04 ↓ (4 4. 50 ) < 0. 00 1* (V : 0 .1 6) Bo dy m as s in de x (k g/ m 2 ), m ea n (S D ) 22 .2 (5 .4 0) 21 .9 ↓ (5 .8 0) 22 .3 ↑ (5 .3 0) < 0. 00 1* (d : 0 .0 7) O be si ty : 3 0. 0 ≤, n (% ) 80 1 (3 .4 0) 23 7▲ (4 .0 0) 56 4▽ (3 .3 0) < 0. 00 1* ,a (V : 0 .0 6) O ve rw ei gh t: 25 .0 ≤ – < 30 .0 , n (% ) 34 12 (1 4. 70 ) 75 3▽ (1 2. 70 ) 26 59 ▲ (1 5. 30 ) a U nd er w ei gh t: < 18 .5 , n (% ) 32 18 (1 3. 90 ) 10 16 ▲ (1 7. 20 ) 22 02 ▽ (1 2. 70 ) a N or m al : 1 8. 5 ≤ – < 25 .0 , n (% ) 15 ,7 97 (6 8. 00 ) 39 17 ▽ (6 6. 10 ) 11 ,8 80 ▲ (6 8. 70 ) a So ci oe co no m ic v ar ia bl es M ar rie d, n (% ) 10 ,4 91 (4 5. 20 ) 30 87 ↑ (5 2. 10 ) 74 04 ↓ (4 2. 80 ) < 0. 00 1* (V : 0 .0 8) H ou se ho ld a nn ua l i nc om e (s ev en -p oi nt o rd in al s ca le ) 2– 4 m ill io n (< 2 m ill io n to 4 –6 m ill io n) 2– 4 m ill io n↑ (< 2 m ill io n to 4 –6 m ill io n) 2– 4 m ill io n↓ (< 2 m ill io n to 4 –6 m ill io n) 0. 03 1* (r : 0 .0 2) < 2, n (% ) 81 66 (3 5. 20 ) 19 19 (3 2. 40 ) 62 47 (3 6. 10 ) 2– 4 m ill io n, n (% ) 34 25 (1 4. 70 ) 90 5 (1 5. 30 ) 25 20 (1 4. 60 ) 4– 6 m ill io n, n (% ) 25 39 (1 0. 90 ) 68 6 (1 1. 60 ) 18 53 (1 0. 70 ) 6– 8 m ill io n, n (% ) 28 55 (1 2. 30 ) 69 8 (1 1. 80 ) 21 57 (1 2. 50 ) 8– 10 m ill io n, n (% ) 59 1 (2 .5 0) 13 2 (2 .2 0) 45 9 (2 .7 0) 10 –1 2 m ill io n, n (% ) 79 (0 .3 0) 22 (0 .4 0) 57 (0 .3 0) 12  <  m ill io n, n (% ) 11 1 (0 .5 0) 28 (0 .5 0) 83 (0 .5 0) M is si ng b , n (% ) 54 62 (2 3. 50 ) 15 33 (2 5. 90 ) 39 29 (2 2. 70 ) Ed uc at io na l b ac kg ro un d (s ix -p oi nt o rd in al s ca le ) U ni ve rs ity (h ig h sc ho ol –u ni ve rs ity ) U ni ve rs ity (h ig h sc ho ol –u ni ve rs ity ) U ni ve rs ity (h ig h sc ho ol –u ni ve rs ity ) 0. 37 9 (r: 0 .0 1) Ju ni or h ig h sc ho ol , n (% ) 39 9 (1 .7 0) 11 9 (2 .0 0) 28 0 (1 .6 0) H ig h sc ho ol , n (% ) 57 12 (2 4. 60 ) 13 47 (2 2. 70 ) 43 65 (2 5. 20 ) Te ch ni ca l c ol le ge , n (% ) 25 37 (1 0. 90 ) 73 2 (1 2. 40 ) 18 05 (1 0. 40 ) Tw o- ye ar u ni ve rs ity , n (% ) 23 33 (1 0. 00 ) 65 7 (1 1. 10 ) 16 76 (9 .7 0) U ni ve rs ity , n (% ) 10 ,6 44 (4 5. 80 ) 26 57 (4 4. 90 ) 79 87 (4 6. 20 ) G ra du at e sc ho ol , n (% ) 12 54 (5 .4 0) 29 6 (5 .0 0) 95 8 (5 .5 0) M is si ng b , n (% ) 34 9 (1 .5 0) 11 5 (1 .9 0) 23 4 (1 .4 0) C om or bi di ty H yp er te ns io n, n (% ) 47 23 (2 0. 30 ) 91 3↓ (1 5. 40 ) 38 10 ↑ (2 2. 00 ) < 0. 00 1* (V : 0 .0 7) D ia be te s m el lit us , n (% ) 16 44 (7 .1 0) 37 5↓ (6 .3 0) 12 69 ↑ (7 .3 0) 0. 00 9* (V : 0 .0 2) D ys lip id em ia , n (% ) 34 05 (1 4. 70 ) 84 4 (1 4. 20 ) 25 61 (1 4. 80 ) 0. 30 7 (V : 0 .0 1) D ep re ss io n, n (% ) 11 45 (4 .9 0) 57 8↑ (9 .8 0) 56 7↓ (3 .3 0) < 0. 00 1* (V : 0 .1 3) Sl ee p di so rd er , n (% ) 74 12 (3 1. 90 ) 30 77 ↑ (5 2. 00 ) 43 35 ↓ (2 5. 10 ) < 0. 00 1* (V : 0 .2 5) (C on tin ue s) 6  |    HEADACHE Va ria bl e To ta l ( n =  2 3, 22 8) Se lf- re po rt ed fr eq ue nt h ea da ch e (2 5. 5% , n  =  5 92 3) N on -s el f- re po rt ed fr eq ue nt he ad ac he (7 4. 5% , n  =  17 ,3 05 ) p- va lu e Su bs ta nc e us e hi st or y D ai ly c on su m pt io n of c of fe e, n (% ) 13 ,2 10 (5 6. 90 ) 29 89 ↓ (5 0. 50 ) 10 ,2 21 ↑ (5 9. 10 ) < 0. 00 1* (V : 0 .0 8) D ai ly c on su m pt io n of c af fe in at ed s of t d rin ks , n (% ) 10 05 (4 .3 0) 32 8↑ (5 .5 0) 67 7↓ (3 .9 0) < 0. 00 1* (V : 0 .0 4) A lc oh ol c on su m pt io n m or e th an tw ic e a w ee k, n (% ) 73 73 (3 1. 70 ) 13 88 ↓ (2 3. 40 ) 59 85 ↑ (3 4. 60 ) < 0. 00 1* (V : 0 .1 0) N ot e: C on tin uo us v ar ia bl es w er e co m pa re d us in g th e M an n– W hi tn ey U te st o r i nd ep en de nt t- te st , a s ap pr op ria te . C at eg or ic al v ar ia bl es w er e co m pa re d us in g th e χ2 te st w ith re si du al a na ly si s w he n ap pl ic ab le . E ff ec t s iz es a re re po rt ed a s C ra m er V , C oh en d , o r r . A bb re vi at io ns : ↑ , h ig he r v al ue o r p ro po rt io n; ↓ , l ow er v al ue o r p ro po rt io n; d , e ff ec t s iz e fo r t -t es t; IQ R, in te rq ua rt ile ra ng e; Q 1, fi rs t q ua rt ile ; Q 3, th ird q ua rt ile ; r , e ff ec t s iz e fo r M an n– W hi tn ey U te st . a χ2 te st s (2  ×  4 c on tin ge nc y ta bl e) w er e co nd uc te d to c om pa re th e di st rib ut io n be tw ee n th e tw o gr ou ps . T he ▲ s ym bo l i nd ic at es a c el l w ith a s ig ni fic an tly h ig he r t ha n ex pe ct ed fr eq ue nc y, a nd th e ▽ sy m bo l i nd ic at es a c el l w ith a s ig ni fic an tly lo w er th an e xp ec te d fr eq ue nc y, b as ed o n po st h oc re si du al a na ly si s w ith B on fe rr on i c or re ct io n fo llo w in g th e χ2 te st . b Th es e m is si ng v ar ia bl es (M is si ng ) w er e ex cl ud ed in th e M an n– W hi tn ey U te st fo r o rd in al v ar ia bl e co m pa ris on . T he y w er e in cl ud ed in th e m ul tiv ar ia bl e an al ys is fo llo w in g m ul tip le im pu ta tio n. *p  <  0 .0 5. TA B LE 1   (C on tin ue d) TA B LE 2   D is tr ib ut io n of c on ve nt io na l c ig ar et te a nd h ea te d to ba cc o pr od uc t u se a m on g pa rt ic ip an ts w ith a nd w ith ou t s el f- re po rt ed fr eq ue nt h ea da ch e. Va ria bl e To ta l ( n =  2 3, 22 8) Se lf- re po rt ed fr eq ue nt h ea da ch e (2 5. 5% , n  =  5 92 3) N on -s el f- re po rt ed fr eq ue nt h ea da ch e (7 4. 5% , n  =  17 ,3 05 ) p- va lu e C ig ar et te u se h is to ry < 0. 00 1* (V : 0. 03 ) C ur re nt c ig ar et te u se r, n (% ) 28 23 (1 2. 20 ) 71 7 (1 2. 10 ) 21 06 (1 2. 20 ) Fo rm er c ig ar et te u se r, n (% ) 73 93 (3 1. 80 ) 17 45 ▽ (2 9. 50 ) 56 48 ▲ (3 2. 60 ) N ev er c ig ar et te u se r, n (% ) 13 ,0 12 (5 6. 00 ) 34 61 ▲ (5 8. 40 ) 95 51 ▽ (5 5. 20 ) H ea te d to ba cc o pr od uc t u se h is to ry < 0. 00 1* (V : 0. 06 ) C ur re nt h ea te d to ba cc o pr od uc t u se r, n (% ) 26 88 (1 1. 60 ) 85 6▲ (1 4. 50 ) 18 32 ▽ (1 0. 60 ) Fo rm er h ea te d to ba cc o pr od uc t u se r, n (% ) 23 26 (1 0. 00 ) 66 3▲ (1 1. 20 ) 16 63 ▽ (9 .6 0) N ev er h ea te d to ba cc o pr od uc t u se r, n (% ) 18 ,2 14 (7 8. 40 ) 44 04 ▽ (7 4. 40 ) 13 ,8 10 ▲ (7 9. 80 ) N ot e: G ro up d iff er en ce s w er e te st ed u si ng th e χ2 te st , a nd e ff ec t s iz es a re p re se nt ed a s C ra m er V . T he χ 2 (2  ×  3 c on tin ge nc y ta bl e) w as c on du ct ed to c om pa re th e di st rib ut io n be tw ee n th e tw o gr ou ps . T he ▲ s ym bo l i nd ic at es a c el l w ith a s ig ni fic an tly h ig he r t ha n ex pe ct ed fr eq ue nc y, a nd th e ▽ s ym bo l i nd ic at es a c el l w ith a s ig ni fic an tly lo w er th an e xp ec te d fr eq ue nc y, b as ed o n po st h oc re si du al a na ly si s w ith Bo nf er ro ni c or re ct io n fo llo w in g th e χ2 te st . *p  <  0 .0 5.     | 7HEADACHE TA B L E 3  Multivariable analysis of factors associated with the presence of self-reported frequent headache using a logistic regression model with multiple imputations. Variable Crude odds ratio [95% confidence interval] p-value in univariable analysis Adjusted odds ratio [95% confidence interval] p-value in multivariable analysis Age (years) 0.97 [0.97–0.97] <0.001* 0.97 [0.97–0.98] <0.001* Biological sex; female 2.14 [2.02–2.28] <0.001* 2.30 [2.14–2.47] <0.001* Body mass index (kg/m2) Obesity: 30.0 ≤ 1.16 [1.06–1.27] <0.001* 1.15 [0.99–1.28] 0.106 Overweight: 25.0 ≤ – <30.0 0.78 [0.67–0.92] 0.002* 1.11 [0.99–1.25] 0.058 Underweight: <18.5 0.71 [0.65–0.78] <0.001* 1.01 [0.91–1.09] 0.988 Normal: 18.5 ≤ – <25.0 Reference – Reference – Socioeconomic variables Married 1.46 [1.37–1.54] <0.001* 0.94 [0.88–1.01] 0.093 Higher household annual income (seven-point ordinal scale)a 1.012 [1.004–1.019] 0.071 1.012 [1.004–1.020] 0.125 Higher educational background (six-point ordinal scale)a 0.99 [0.98–1.01] 0.438 1.01 [0.99–1.02] 0.383 Comorbidity Hypertension 0.65 [0.60–0.70] <0.001* 0.95 [0.86–1.05] 0.301 Diabetes mellitus 0.85 [0.76–0.96] 0.010* 0.92 [0.79–1.07] 0.262 Dyslipidemia 0.96 [0.88–1.04] 0.302 1.20 [1.08–1.33] <0.001* Depression 3.19 [2.83–3.60] <0.001* 1.92 [1.68–2.20] <0.001* Sleep disorder 3.23 [3.04–3.44] <0.001* 2.78 [2.61–2.97] <0.001* Substance use history Daily consumption of coffee 0.71 [0.67–0.75] <0.001* 1.071 [1.002–1.153] 0.042* Daily consumption of caffeinated soft drinks 1.44 [1.26–1.65] <0.001* 1.25 [1.08–1.45] 0.003* Alcohol consumption more than twice a week 0.58 [0.54–0.62] <0.001* 0.76 [0.70–0.82] <0.001* Cigarette use history Current cigarette user 1.17 [1.10–1.25] <0.001* 1.71 [1.44–2.03] <0.001* Former cigarette user 1.06 [0.97–1.16] 0.190 1.54 [1.37–1.73] <0.001* No history of cigarette use Reference – Reference – Heated tobacco product use history Current heated tobacco product user 0.80 [0.73–0.88] <0.001* 1.15 [1.01–1.32] 0.038* Former heated tobacco product user 0.68 [0.63–0.75] <0.001* 1.09 [0.96–1.24] 0.203 No history of heated tobacco product use Reference – Reference – Note: This table presents the results of a multivariable analysis with multiple imputation examining which variables were associated with the self-reported frequent headache group (as 1) or the non-self-reported frequent headache group (as 0). Ten-fold multiple imputation for missing values (a) was performed, and the results were pooled using Rubin's Rules. All explanatory variables were entered simultaneously using the forced- entry method. This model adjusted for age, sex, body mass index category, marital status, household income, educational level, comorbidities, and substance use variables shown in this table. Adjusted odds ratios >1 indicate a higher likelihood of having self-reported frequent headache. aSensitivity analysis using list-wise deletion for missing values (a) is presented in Table S2. *p-values <0.05 were considered statistically significant. 8  |    HEADACHE CI = 1.08–1.33]; p < 0.001), depression (aOR = 1.92 [95% CI = 1.68– 2.20]; p < 0.001), sleep disorder (aOR = 2.78 [95% CI = 2.61–2.97]; p < 0.001), the daily consumption of coffee (aOR = 1.071 [95% CI = 1.002–1.153]; p = 0.042), and caffeinated soft drinks (aOR = 1.25 [95% CI = 1.08–1.45]; p = 0.003). Alcohol consumption more than twice a week was negatively associated with headache (aOR = 0.76 [95% CI = 0.70–0.82]; p < 0.001). Postestimation comparison of regression coefficients between cigarette and HTP users To further investigate whether the magnitude of association with headache differed between cigarette use and HTP use, postes- timation pairwise comparisons of regression coefficients were performed using Wald tests (Table 4). In the multiple imputation model, the difference in log-odds coefficients between current cigarette and current HTP users was statistically significant (coef- ficient difference = 0.39, ratio of adjusted odds ratios = 1.48 [95% CI = 1.19–1.84]), indicating that the association with headache was weaker for current HTP use compared with current cigarette use. Furthermore, both current and former HTP use were associated with a smaller magnitude of headache compared to current and former cigarette use. Sensitivity analysis results The results of the sensitivity analysis using list-wise deletion for the missing variables on income and education are presented in Table S2 in Supporting Information File 2. This analysis used list- wise deletion due to some missing data on income and education (finally included; n = 17,638/23,228, 75.9%). Similar to the multi- ple imputation model, the headache group was significantly as- sociated with greater cigarette use, both current (aOR = 1.27 [95% CI = 1.11–1.45]; p < 0.001) and former use (aOR = 1.138 [95% CI = 1.002–1.304]; p = 0.043), compared to having never smoked. Both current (aOR = 1.18 [95% CI = 1.02–1.35]; p = 0.024) and for- mer (aOR = 1.14 [95% CI = 1.04–1.26]; p = 0.007) use of HTPs were also associated with headache. Regarding the pairwise comparisons of regression coefficients between the cigarette and HTP use groups (Table S3 in Supporting Information File 2), no significant differences were observed in the coefficients between current and former users of either cigarettes or HTPs in relation to headache. Overall, the sensitivity analysis supports the robustness of the main results, indicating that current and former cigarette use, as well as current HTP use, are associated with self-reported frequent headache. Considering the findings from the sensitivity analysis, the association between former HTP use and self-reported frequent headache, as well as the differences in the effects between ciga- rettes and HTPs, should be interpreted with caution. DISCUSSION In this large population-based study, we found that both current and former use of conventional cigarettes, as well as current use of HTPs, were independently associated with higher aORs of the prevalence of self-reported frequent headache. Furthermore, the impact on the prevalence of headaches was greater for conventional cigarettes than for HTPs. Importantly, the present findings should be interpreted as associations with self-reported headache symp- toms at the population level, rather than with clinically diagnosed TA B L E 4  Pairwise comparisons of regression coefficients between cigarette and HTP use groups with multiple imputations. Cigarette HTP Coefficient difference (cigarette–HTP) Standard error p-value Ratio of adjusted odds ratio [95% confidence interval] None Former −0.09 0.07 0.206 0.92 [0.80–1.05] None Current −0.15 0.07 0.036* 0.86 [0.75–0.99] Former None 0.43 0.06 <0.001* 1.54 [1.37–1.73] Former Former 0.35 0.09 <0.001* 1.41 [1.18–1.69] Former Current 0.28 0.09 0.002* 1.33 [1.11–1.59] Current None 0.54 0.09 <0.001* 1.72 [1.45–2.03] Current Former 0.45 0.11 <0.001* 1.58 [1.27–1.95] Current Current 0.39 0.11 <0.001* 1.48 [1.19–1.84] Note: This table shows the postestimation pairwise comparisons of regression coefficients (coefficient differences) between cigarette and HTP use categories after multiple imputations of missing data. Coefficient differences represent the log-odds difference (cigarette−HTP) for the presence of self-reported frequent headache, based on the same multivariable logistic regression model as in Table 3. A positive coefficient difference (ratio of adjusted odds ratios >1) indicates a higher self-reported frequent headache prevalence among cigarette users than HTP users. Missing values in covariates were handled by multiple imputation (n = 10 data sets), and pooled estimates were obtained using Rubin's rules. Sensitivity analyses with list-wise deletion yielded different results (see Table S3). Abbreviation: HTP, heated tobacco product. *p-values were derived from Wald tests, and p-values <0.05 were considered statistically significant.     | 9HEADACHE primary headache disorders. Accordingly, this study is positioned as a hypothesis-generating epidemiological analysis, not as an investi- gation of disorder-specific pathophysiology. However, to our knowl- edge, this is the first study worldwide to explore the relationship between conventional cigarettes, HTPs, and the prevalence of self- reported frequent headache. Previous epidemiological surveys on smoking and headache There are no consistent epidemiological results on primary headache disorders and smoking, probably due to the rapid washout effect of certain components, such as nicotine, and because of differences in the definition of smokers, the number of cigarettes smoked, the number of years smoked, and study design.24,25 However, some re- search showed associations between primary headache disorders and smoking. For example, current smokers have a higher risk of migraine (relative risk = 1.3 [95% CI = 1.1–1.5]).4 Current smokers (aOR = 1.3 [95% CI = 1.1–1.6]) and former smokers (aOR = 1.2 [95% CI = 1.0–1.4]) have a higher risk of TTH.5,6 Also, elevated TTH risk among adolescent smokers was confirmed in the Dunedin cohort.26 The weighted-pooled prevalence of smoking in cluster headache, one of the TACs, is 65% (95% CI = 55%–76%).25 Considering that TACs are rare disorders with a prevalence of approximately 1 per 1000, it is likely that most self-reported frequent headache in our study primarily consisted of migraine and TTH. Therefore, similar to the previous studies, our findings reasonably suggest that both cur- rent (aOR = 1.71 [95% CI = 1.44–2.03]) and former use (aOR = 1.54 [95% CI = 1.37–1.73]) of conventional cigarettes may increase the as- sociation with self-reported frequent headache. Moreover, the aORs showed similar values, suggesting that Japanese individuals, such as those in other ethnic groups, may also show a higher prevalence of headache symptoms in association with smoking. Furthermore, there are only two studies in Japan on the rela- tionship between smoking and headache disorders. In patients with migraine with a history of smoking, the effect of triptans as acute medication is reduced.27 The smoking proportion among patients with cluster headache is approximately twice that of the Japanese adult population, at 63% for men and 17% for women.28 However, to date, no studies in Japan have comprehensively investigated the association between headache prevalence and smoking. Therefore, our findings are novel in Japan regarding the relationship between conventional cigarette use and the prevalence of self-reported fre- quent headache. Pathophysiological mechanisms between smoking and headache We herein discuss the potential mechanism of smoking and headache symptoms, given that both conventional cigarettes and HTPs re- lease and deliver similar components upon inhalation.29,30 Cigarette smoking has been reported as a potential trigger for migraine at- tacks.24 Nicotine induces vasoconstriction24 and alters cortical ex- citability, which may increase the likelihood of cortical spreading depression, the presumed mechanism underlying migraine aura.31 Carbon monoxide (CO) has also been implicated as a potential mi- graine trigger through its vasodilatory effects, induction of cerebral hypoxia, and modulation of central pain pathways.32 Additionally, tobacco smoke contains numerous irritants that may stimulate the trigeminovascular system, thereby lowering the threshold for mi- graine attacks.33 This may be through the mechanism that the in- flammatory response provoked by smoking may also contribute to central sensitization and increased headache susceptibility.34 Smoking may also indirectly increase the frequency and severity of headaches through its association with stress, anxiety symptoms,24 and lifestyle factors such as poor diet quality and lack of physical activity.34 Also, the smell or odor of tobacco is a potential trigger for migraine.35 Furthermore, smoking is also a risk for migraine chroni- fication,36 likely through the reinforced and irreversible mechanisms described above. Although the direct pathophysiological link between smoking and TTH is less clearly established compared to migraine,25 several mechanisms may be involved. TTH has been associated with dys- function in the central serotonergic system,37 and cigarette smoking is known to alter serotonin levels.38 Such neurochemical changes may hypothetically influence pain processing and headache suscep- tibility. Similar to migraine, smoking may indirectly exacerbate TTH through stress, anxiety, and unhealthy lifestyles. TACs, especially cluster headache, show an epidemiological as- sociation with smoking.39 A disproportionately high prevalence of current and former smokers, approximately 50%–80%,39 has been observed among patients with cluster headache compared to the general population. Tobacco-related substances may influence hy- pothalamic function and circadian rhythms, which are believed to be involved in the pathophysiology of cluster headache.40 Patients with cluster headache more frequently reported environmental triggers such as alcohol and smoking, and some experienced at- tacks following smoking episodes, suggesting that nicotine might activate trigeminal-autonomic reflex pathways in susceptible individuals.25 These pathophysiological mechanisms may help explain how smoking may exert both acute and chronic effects on headache- related neurovascular systems. Importantly, such effects may not be entirely reversible.36 Persistent neurobiological alterations, such as central sensitization, serotonergic dysregulation, and hypothalamic dysfunction, could remain even after cessation of smoking.41 This may partly explain our findings that former cigarette users also ex- hibited elevated adjusted odds of self-reported frequent headache, although causal interpretations cannot be established in the pres- ent cross-sectional design. Further investigation is needed to clarify how various aspects of tobacco exposure, such as current and for- mer smoking status, acute versus chronic exposure, and occasional or habitual smoking behavior, are differentially associated with pri- mary headache disorders. 10  |    HEADACHE Difference between conventional cigarettes and HTPs The prevalence of conventional cigarette smokers worldwide is approximately 15%.42 Recently, a transition from conventional cigarettes to HTPs has been observed. HTP use increased notably between 2015 and 2020 in the Western Pacific region (from 0.1% to 10.6%) and in the European region (from 0% in 2016 to 1.15% in 2020). Usage rates were higher in males (3.5%) than in females (1.8%). The proportion of individuals who had ever used HTPs among adolescents (13–17 years old) was 5.25%, compared to 2.45% among adults.43 Surveys in European countries showed that HTPs were be- lieved by users to be less harmful than smoking cigarettes (39.5%) and that they are used for stopping or reducing smoking (28.2%).44 The same trend was also confirmed in Japan, where the prevalence of HTP use rose markedly.11 Of more than 4700 chemical substances from tobacco,24 several harmful substances have been implicated in the onset of headaches, including nicotine,31 CO,32 volatile organic compounds,45 and tobacco- specific nitrosamines.46 Compared to conventional cigarettes, HTPs deliver similar levels of nicotine (approximately 0.5–1.4 mg per stick),10,29 which may explain their association with headaches. However, HTPs emit significantly lower levels of CO (1%–14% of cig- arette emissions),10,30 volatile organic compounds such as benzene (reduced by 87%–99%),10,29,30 and tobacco-specific nitrosamines (re- duced by 70%–90%).10,29 The promotion of reduced levels of harmful substances in HTPs, compared to conventional cigarettes, may have contributed to their rapid increase in popularity worldwide.11,43 Despite lower toxicant emissions in HTPs, comparable nicotine exposure may still be relevant to headache-related biological mech- anisms.10 Notably, the aORs for self-reported frequent headache were significantly lower in current and former HTP users than in current and former cigarette smokers in our study, though the sen- sitivity analysis revealed nonsignificance. This suggests that nonnic- otine substances may also play a role in headache pathophysiology, and their reduced presence in HTPs may partially explain the weaker association. In other words, both conventional cigarettes and HTPs may be relevant to headache-related mechanisms through nicotine and other constituents. However, the lower levels of certain non- nicotine toxicants in HTPs may attenuate their overall impact, po- tentially leading to the weaker association observed in the aORs. Further research is needed to identify the specific substances re- sponsible and clarify the biological mechanisms linking tobacco ex- posure to headaches. In addition, HTPs are increasingly viewed by the public as a less harmful alternative. However, this perception may downplay their potential neurological and health risks, under- scoring the importance of continued public education and regula- tory oversight. Future plans for awareness raising and treatment Given the observed associations between both conventional ciga- rette and HTP use with frequent headaches in our study, future public health awareness campaigns should be considered. In Japan, the area dedicated to health warnings on cigarette packages remains relatively small.12 As in other countries, plain packaging and graphic warning labels may be beneficial. Expanding access to smoking cessation medical practice through telemedicine and using mobile cessation support applications may also enhance the reach and ef- fectiveness of treatment. Regionally coordinated awareness cam- paigns47 could further promote public understanding. Notably, no awareness campaigns to date have specifically addressed the link between tobacco use and headaches. Thus, developing targeted in- terventions for patients with headache disorders and smokers, high- lighting the risk of tobacco use for headache disorders prevalence, migraine chronification,24,34 and the effect of second-hand smoking for headache attacks13 may become a novel and impactful public health approach. Limitations First, the survey was conducted online. Although the research agency seeks to ensure the representativeness of the Japanese cen- sus, there is a bias against people who do not use the internet and those who did not enter their age or biological sex in this survey. Also, self-reported information online is not necessarily reliable. Because self-reported data rely on participants' memory and sub- jective perception, there can be recall bias. For instance, older adults may be more likely to forget or underreport their symptoms. Second, we asked the respondents about their experience of headaches in the past year and had them answer using a four-point Likert scale. We formally defined and estimated the prevalence of self-reported frequent headache using the answers to this question. Therefore, we did not actually diagnose specific primary headache disorders, nor did we confirm the diagnosis by other physicians. This nonspecific definition merges different headache disorders (e.g., migraine, TTH, cluster headache), which have potentially different etiologies, risk factor profiles, and relationships to smoking. So, we cannot tell whether the observed associations apply to a specific headache disorder or are diluted across multiple, unrelated condi- tions. Accordingly, we should interpret the result cautiously, because the term headache was used broadly rather than identifying specific headache types. Although our broad definition of self-reported fre- quent headache precludes a diagnosis of specific types, our reported prevalence of overall headache (25.5%) aligns reasonably with pre- viously reported prevalence rates for primary headache disorders in Japan (e.g., 8.4% for migraine and 22.4% for TTH48). Moving for- ward, in the 2026 wave of the JASTIS study, we will explore spe- cific types of headache disorders by administering an ICHD-3-based questionnaire, as well as asking participants whether they have been diagnosed by physicians. Third, because this study was based entirely on self-reported questionnaire data, detailed information on medication use (e.g., antidepressants, anxiolytics, psychostimulants) was not obtained. These medications are known to influence subjective responses on     | 11HEADACHE validated scales and may have affected the results. Similarly, the di- agnosis of sleep disorder was self-reported, and it was not possible to confirm whether participants had undergone polysomnography or whether sleep deprivation or shift work was involved. Fourth, the cross-sectional design cannot determine a causal relationship. Although our analysis suggests an association be- tween smoking and self-reported frequent headache, the possibility of reverse causality cannot be ruled out. Individuals experiencing headaches may turn to smoking as a form of self-medication or stress relief, which may further complicate the causal relationship. Moreover, it remains unclear whether individuals with smoking hab- its experience headaches as a chronic condition or whether smoking itself acts as a trigger for headache attacks. In our future research, we plan to investigate this in more detail by asking about both the prevalence of ICHD-3-based headache disorders and whether smoking serves as a specific trigger. Fifth, unmeasured confounding, such as other socioeconomic factors, stress levels, or genetic predispositions, may still influence the observed associations. Finally, data on the number of sticks, frequency, and intensity of tobacco use were not included in the present analysis. The effects of combining or switching between conventional cigarettes and HTPs have also not been studied. Despite these limitations, our study provides important new evidence that both conventional and HTP use may be associated with self-reported frequent headache. These findings underscore the need for further longitudinal research and raise concerns about the potential neurological risks associated with newly emerging HTPs. CONCLUSION Our nationwide cross-sectional analysis demonstrated that both current and former use of conventional cigarettes, as well as current use of HTP, were independently associated with a higher prevalence of self-reported frequent headache. The aORs for self-reported fre- quent headache were lower in current and former HTP users than in current and former cigarette smokers, suggesting a potentially reduced, but still present, association for headache. Although HTPs are marketed as less harmful than cigarettes, their nicotine content and other substances may be relevant to headache-related biologi- cal mechanisms. These findings highlight the potential neurological associations of both conventional and HTPs and emphasize the need for public health policies to consider headache in tobacco harm re- duction strategies. Future longitudinal research is warranted to clar- ify causal relationships and explore potential interventions targeting cigarettes and HTP-related headache associations. We also need fu- ture studies incorporating ICHD-3-based diagnoses by physicians to clarify disorder-specific effects. AUTHOR CONTRIBUTIONS Masahito Katsuki: Conceptualization; investigation; writing – origi- nal draft; methodology; validation; visualization; writing – review and editing; software; formal analysis; project administration; data cura- tion; resources. Kieran Moran: Funding acquisition; investigation; writing – original draft; supervision; formal analysis; methodology; writing – review and editing. Muneto Tatsumoto: Conceptualization; in- vestigation; writing – review and editing. Keisuke Suzuki: Writing – re- view and editing; conceptualization; investigation. Tomokazu Shimazu: Conceptualization; investigation; writing – review and editing. Yasuhiko Matsumori: Conceptualization; investigation; writing – review and editing. Yutaro Fuse: Conceptualization; investigation; writing – re- view and editing. Miguel Ángel Huerta: Writing – review and editing. Daijiro Sugiyama: Writing – review and editing. Jackson T. S. Cheung: Writing – review and editing. Siobhán O'Connor: Writing – review and editing; supervision. Tomás Ward: Supervision; writing – review and editing. Takahiro Tabuchi: Conceptualization; investigation; funding acquisition; writing – original draft; methodology; writing – review and editing; visualization; validation; software; formal analysis; project ad- ministration; data curation; supervision; resources. AFFILIATIONS 1Physical Education and Health Center, Nagaoka University of Technology, Nagaoka, Niigata, Japan 2School of Health and Human Performance, Dublin City University, Dublin, Ireland 3Insight Research Ireland Centre for Data Analytics, Dublin City University, Dublin, Ireland 4Department of Biostatistics, Graduate School of Medicine, Saitama Medical University, Saitama, Japan 5Insight Research Ireland Centre for Data Analytics, Maynooth University, Kildare, Ireland 6Department of Sport Science and Nutrition, Maynooth University, Kildare, Ireland 7Canon Marketing Japan Inc., Tokyo, Japan 8Department of Neurology, Dokkyo Medical University, Tochigi, Japan 9Department of Neurology, Saitama Neuropsychiatric Institute, Saitama, Japan 10Sendai Headache and Neurology Clinic, Sendai, Japan 11Department of Artificial Intelligence Medicine, Graduate School of Medicine, Chiba University, Chiba, Japan 12Department of Pharmacology, University of Granada, Granada, Spain 13Biosanitary Research Institute ibs.GRANADA, Granada, Spain 14Department of Pharmacology, University of Cambridge, Cambridge, UK 15Research Department, Daiichi Sankyo Healthcare Co. Ltd., Tokyo, Japan 16UCL Faculty of Medical Sciences, London, UK 17Division of Epidemiology, School of Public Health, Tohoku University Graduate School of Medicine, Sendai, Japan ACKNOWLEDG MENTS The authors acknowledge the contributions of all participants in the JASTIS 2025 survey. 12  |    HEADACHE FUNDING INFORMATION This publication has emanated from research jointly funded by Taighde Éireann–Research Ireland under grant number 12/RC/2289_ P2, the European Union's Horizon 2020 research and innovation pro- gram under the Marie Skłodowska-Curie grant agreement number 101034252. This study was also supported by Health Labor Sciences Research Grants (22FA1002, 23FA1004), the Japan Society for the Promotion of Science KAKENHI grants (JP20K19633, JP23K21579, JP24H00663, JP21H04856, JP25H01079), and the intramural fund of the National Institute for Environmental Studies. CONFLIC T OF INTERE S T S TATEMENT Takahiro Tabuchi received financial support for research (research fundings, consulting fees or lecture fees) from Daiichi Sankyo Healthcare, Johnson & Johnson, Data Seed, Workout-Plus, EMMA, and Addness (in the last 36 months). Daijiro Sugiyama is an employee of Daiichi Sankyo Healthcare. Masahito Katsuki, Kieran Moran, Muneto Tatsumoto, Keisuke Suzuki, Tomokazu Shimazu, Yasuhiko Matsumori, Yutaro Fuse, Miguel Ángelngel Huerta, Jackson T.S. Cheung, Siobhán O'Connor, and Tomás Ward report no conflicts of interest concerning the materials or methods used in this study or the findings presented in this article. This article and the contents of this study were not published or presented previously. DATA AVAIL ABILIT Y S TATEMENT The data sets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request. ORCID Masahito Katsuki  https://orcid.org/0000-0002-0192-5430 Muneto Tatsumoto  https://orcid.org/0000-0002-1950-0756 Keisuke Suzuki  https://orcid.org/0000-0003-0320-8993 Tomokazu Shimazu  https://orcid.org/0000-0001-9146-7770 Yutaro Fuse  https://orcid.org/0000-0001-9402-8079 Miguel Á. Huerta  https://orcid.org/0000-0003-2842-0085 R E FE R E N C E S 1. Headache Classification Committee of the International Headache Society (IHS). The International Classification of Headache Disorders, 3rd edition. Cephalalgia. 2018;38(1):1-211. 2. Vos T, Abajobir AA, Abate KH, et al. Global, regional, and national incidence, prevalence, and years lived with disability for 328 dis- eases and injuries for 195 countries, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. 2017;390(10100):1211-1259. 3. Headache Clinical Practice Guideline Development Committee. In: Japanese Society of Neurology J, Japanese Headache Society J, Japanese Society of Neurological Therapeutics J, ed. Clinical Practice Guideline for Headache Disorders 2021 (Japanese). Igaku- Shoin; 2021:494. 4. Hagen K, Åsberg AN, Stovner L, et  al. Lifestyle factors and risk of migraine and tension-type headache. Follow-up data from the Nord-Trøndelag Health Surveys 1995-1997 and 2006-2008. Cephalalgia. 2018;38(13):1919-1926. 5. Robberstad L, Dyb G, Hagen K, Stovner LJ, Holmen TL, Zwart JA. An unfavorable lifestyle and recurrent headaches among adoles- cents: the HUNT study. Neurology. 2010;75(8):712-717. 6. Pan LLH, Ling YH, Wang SJ, et al. Hallmarks of primary headache: part 2- tension-type headache. J Headache Pain. 2025;26(1):164. 7. Manzoni GC, Terzano MG, Bono G, Micieli G, Martucci N, Nappi G. Cluster headache—clinical findings in 180 patients. Cephalalgia. 1983;3(1):21-30. 8. Dai X, Gil GF, Reitsma MB, et al. Health effects associated with smok- ing: a burden of proof study. Nat Med. 2022;28(10):2045-2055. 9. Tabuchi T, Shinozaki T, Kunugita N, Nakamura M, Tsuji I. Study pro- file: the Japan ‘Society and New Tobacco’ Internet Survey (JASTIS): a longitudinal internet cohort study of heat-not-burn tobacco prod- ucts, electronic cigarettes, and conventional tobacco products in Japan. J Epidemiol. 2019;29(11):444-450. 10. Simonavicius E, McNeill A, Shahab L, Brose LS. Heat-not-burn tobacco products: a systematic literature review. Tob Control. 2019;28(5):582-594. 11. Tabuchi T, Gallus S, Shinozaki T, Nakaya T, Kunugita N, Colwell B. Heat-not-burn tobacco product use in Japan: its prevalence, predic- tors and perceived symptoms from exposure to secondhand heat- not-burn tobacco aerosol. Tob Control. 2018;27(e1):e25-e33. 12. Odani S, Tabuchi T. Tobacco usage in the home: a cross-sectional analysis of heated tobacco product (HTP) use and combustible to- bacco smoking in Japan, 2023. Environ Health Prev Med. 2024;29:11. 13. Imura Y, Tabuchi T. Exposure to secondhand heated-tobacco- product aerosol may cause similar incidence of asthma attack and chest pain to secondhand cigarette exposure: the JASTIS 2019 Study. Int J Environ Res Public Health. 2021;18(4):1766. 14. Takaoka S, Saito H, Kawate M, et  al. Exploring the presence of long COVID-like symptoms in patients with chronic pain: a large-scale internet-based cross-sectional study in Japan. Pain. 2025;166:2412-2420. 15. Wang TW, Gentzke AS, Creamer MR, et al. Tobacco product use and associated factors among middle and high school students- United States, 2019. MMWR Surveill Summ. 2019;68(12):1-22. 16. Ornello R, Ripa P, Pistoia F, et  al. Migraine and body mass index categories: a systematic review and meta-analysis of observational studies. J Headache Pain. 2015;16:27. 17. Le H, Tfelt-Hansen P, Skytthe A, Kyvik KO, Olesen J. Association between migraine, lifestyle and socioeconomic factors: a population-based cross-sectional study. J Headache Pain. 2011;​ 12(2):​157-172. 18. Mazzacane F, Vaghi G, Cotta Ramusino M, Perini G, Costa A. Arterial hypertension in the chronic evolution of migraine: bystander or risk factor? An overview. J Headache Pain. 2024;25(1):13. 19. Ha WS, Nguyen VK, Chu MK. Epidemiological linkage between migraine and diabetes mellitus: a systematic review and meta- analysis. J Headache Pain. 2024;25(1):158. 20. Saberi A, Hatamian HR, Kazemnejad E, Ghorbannejad N. Hyperlipidemia in migraine: is it more frequent in migraineurs? Iran J Neurol. 2011;10(3–4):46-50. 21. Viudez-Martínez A, Torregrosa AB, Navarrete F, García-Gutiérrez MS. Understanding the biological relationship between migraine and depression. Biomolecules. 2024;14(2):163. 22. Buse DC, Rains JC, Pavlovic JM, et al. Sleep disorders among peo- ple with migraine: results from the Chronic Migraine Epidemiology and Outcomes (CaMEO) Study. Headache. 2019;59(1):32-45. 23. Zhang L, Yin J, Li J, Sun H, Liu Y, Yang J. Association between di- etary caffeine intake and severe headache or migraine in US adults. Sci Rep. 2023;13(1):10220. 24. Taylor FR. Tobacco, nicotine, and headache. Headache. 2015;55(7):​ 1028-1044. 25. Błaszczyk B, Martynowicz H, Przegrałek J, et al. Smoking in primary headaches—a systematic review and meta-analysis. J Headache Pain. 2025;26(1):133. doi:10.1186/s10194-025-02076-2 26. Waldie KE, McGee R, Reeder AI, Poulton R. Associations between frequent headaches, persistent smoking, and attempts to quit. Headache. 2008;48(4):545-552. https://orcid.org/0000-0002-0192-5430 https://orcid.org/0000-0002-0192-5430 https://orcid.org/0000-0002-1950-0756 https://orcid.org/0000-0002-1950-0756 https://orcid.org/0000-0003-0320-8993 https://orcid.org/0000-0003-0320-8993 https://orcid.org/0000-0001-9146-7770 https://orcid.org/0000-0001-9146-7770 https://orcid.org/0000-0001-9402-8079 https://orcid.org/0000-0001-9402-8079 https://orcid.org/0000-0003-2842-0085 https://orcid.org/0000-0003-2842-0085 https://doi.org//10.1186/s10194-025-02076-2     | 13HEADACHE 27. Hirata K, Komori M, Ueda K, et al. Outcomes and factors associ- ated with insufficient effectiveness of acute treatments of migraine in Japan: results of the ObserVational survey of the epidemiology, tReatment, and Care of MigrainE (OVERCOME [Japan]) Study. Drugs Real World Outcomes. 2023;10(3):415-428. 28. Imai N. [Clinical profile of cluster headaches in Japan] (Japanese). Rinsho Shinkeigaku. 2013;53(11):1128-1130. 29. Hoshino S, Noro K, Amagai T. Quantification of flavors, volatile organic compounds, tobacco markers, and tobacco-specific nitro- samines in heated tobacco products and their mainstream aerosol. Chem Res Toxicol. 2025;38(5):915-922. 30. Lu F, Yu M, Chen C, et  al. The emission of VOCs and CO from heated tobacco products, electronic cigarettes, and conventional cigarettes, and their health risk. Toxics. 2021;10(1):8. 31. Takizawa T, Ihara K, Unekawa M, et  al. Effects of passive smok- ing on cortical spreading depolarization in male and female mice. J Headache Pain. 2024;25(1):162. 32. Badaeva A, Maiolino L, Danilov A, et al. Neurogasobiology of mi- graine: carbon monoxide, hydrogen sulfide, and nitric oxide as emerging pathophysiological trinacrium relevant to nociception regulation. Open Med (Wars). 2025;20(1):20251201. 33. Martin VT, Behbehani MM. Toward a rational understanding of mi- graine trigger factors. Med Clin North Am. 2001;85(4):911-941. 34. Weinberger AH, Seng EK. The relationship of tobacco use and mi- graine: a narrative review. Curr Pain Headache Rep. 2023;27:39-47. 35. Imai N, Osanai A, Moriya A, Katsuki M, Kitamura E. Classification of odors associated with migraine attacks: a cross-sectional study. Sci Rep. 2023;13(1):8469. 36. Bigal ME, Lipton RB. Modifiable risk factors for migraine progres- sion. Headache. 2006;46(9):1334-1343. 37. Bendtsen L. Central sensitization in tension-type head- ache—possible pathophysiological mechanisms. Cephalalgia. 2000;20(5):486-508. 38. Campos MW, Serebrisky D, Castaldelli-Maia JM. Smoking and cog- nition. Curr Drug Abuse Rev. 2016;9(2):76-79. 39. Elbadawi ASA, Albalawi AFA, Alghannami AK, et al. Cluster head- ache and associated risk factors: a systemic review and meta- analysis. Cureus. 2021;13(11):e19294. 40. Naber WC, Fronczek R, Haan J, et  al. The biological clock in cluster headache: a review and hypothesis. Cephalalgia. 2019;39(14):1855-1866. 41. Tiwari RK, Sharma V, Pandey RK, Shukla SS. Nicotine addiction: neurobiology and mechanism. J Pharmacopuncture. 2020;23(1):1-7. 42. Reitsma MB, Flor LS, Mullany EC, Gupta V, Hay SI, Gakidou E. Spatial, temporal, and demographic patterns in prevalence of smok- ing tobacco use and initiation among young people in 204 coun- tries and territories, 1990-2019. Lancet Public Health. 2021;6(7):e4 72-e481. 43. Sun T, Anandan A, Lim CCW, et  al. Global prevalence of heated tobacco product use, 2015-22: a systematic review and meta- analysis. Addiction. 2023;118(8):1430-1444. 44. Laverty AA, Vardavas CI, Filippidis FT. Prevalence and reasons for use of heated tobacco products (HTP) in Europe: an analysis of Eurobarometer data in 28 countries. Lancet Regional Health - Europe. 2021;8:100159. 45. Ricklund N, Bryngelsson IL, Hagberg J. Self-reported symptoms in Swedish hairdressers and association with exposure to volatile or- ganic compounds (VOCs), including aldehydes. BMC Public Health. 2023;23(1):1576. 46. Borkum JM. Migraine triggers and oxidative stress: a narrative re- view and synthesis. Headache. 2016;56(1):12-35. 47. Katsuki M, Matsumori Y, Kawahara J, et al. Headache education by leaflet distribution during COVID-19 vaccination and school-based on-demand e-learning: Itoigawa Geopark Headache Awareness Campaign. Headache. 2023;63(3):429-440. doi:10.21203/rs.3.rs-18​ 22180/v1 48. Sakai F, Igarashi H. Prevalence of migraine in Japan: a nationwide survey. Cephalalgia. 1997;17(1):15-22. SUPPORTING INFORMATION Additional supporting information can be found online in the Supporting Information section at the end of this article. How to cite this article: Katsuki M, Moran K, Tatsumoto M, et al. Conventional cigarettes, novel heated tobacco products, and self-reported frequent headache association in Japanese individuals: Insights from the JASTIS study, a Japanese cross-sectional analysis. Headache. 2026;00:1-13. doi:10.1111/head.70068 https://doi.org//10.21203/rs.3.rs-1822180/v1 https://doi.org//10.21203/rs.3.rs-1822180/v1 https://doi.org/10.1111/head.70068 Conventional cigarettes, novel heated tobacco products, and self-reported frequent headache association in Japanese individuals: Insights from the JASTIS study, a Japanese cross-sectional analysis Abstract Plain Language Summary INTRODUCTION METHODS Study sample and data source Exclusion criteria Headache assessment Smoking assessment Other variables as confounding factors Statistical analysis Ethical aspects and reporting guidelines RESULTS General characteristics Variables associated with frequent headache in univariable analysis Variables associated with frequent headache in multivariable analysis Postestimation comparison of regression coefficients between cigarette and HTP users Sensitivity analysis results DISCUSSION Previous epidemiological surveys on smoking and headache Pathophysiological mechanisms between smoking and headache Difference between conventional cigarettes and HTPs Future plans for awareness raising and treatment Limitations CONCLUSION AUTHOR CONTRIBUTIONS ACKNOWLEDGMENTS FUNDING INFORMATION CONFLICT OF INTEREST STATEMENT DATA AVAILABILITY STATEMENT ORCID REFERENCES