Background: Primary headaches are widespread and costly public health problems. However, there are insufficient medical resources for their treatment in Japan due to two reasons. First, the numbers of headache specialists and clinics remain insufficient. Second, neurologists and neurosurgeons mainly treat headaches in Japan. However, they mainly work as general stroke neurologists, so they cannot focus on primary headache treatment. To solve these problems, we preliminarily developed a deep learning (DL)-based automated diagnosis model from patients’ Japanese unstructured sentences in the medical questionnaire using a DL framework. We hypothesized that the model would reduce the time and burden on both doctors and patients and improve their quality of life. Methods: We retrospectively investigated our primary headache database and developed a diagnosis model using the DL framework (Prediction One, Sony Network Communications Inc., Japan). We used age, sex, date, and embedding layer made by the medical questionnaire’s natural language processing (NLP). Results: Eight hundred and forty-eight primary headache patients (495 women and 353 men) are included. The median (interquartile range) age was 59 (40–74). Migraine accounted for 46%, tension-type headache for 47%, trigeminal autonomic cephalalgias for 5%, and other primary headache disorders for 2%. The accuracy, mean precision, mean recall, and mean F value of the developed diagnosis model were 0.7759, 0.8537, 0.6086, and 0.6353, which were satisfactory. Conclusion: The DL-based diagnosis model for primary headaches using the raw medical questionnaire’s Japanese NLP would be useful in performing efficient medical practice after ruling out the secondary headaches.

Background: The prevalence of headache disorders, migraine, chronic daily headache (CDH), and medication-overuse headache (MOH) among the elderly in Japan has not been sufficiently investigated. We performed a questionnaire-based survey and revealed 3-month headache prevalence and headaches’ characteristics. Methods: The population aged over 64 was investigated in Itoigawa during their third coronavirus disease 2019 vaccination. Migraine, MOH was defined as The International Classification of Headache Disorders Third edition. CDH was defined as a headache occurring at least 15 days per month. K-means++ were used to perform clustering. Results: Among 2858 valid responses, headache disorders, migraine, CDH, and MOH prevalence was 11.97%, 0.91%, 1.57%, and 0.70%, respectively. Combined-analgesic and non-opioid analgesic were widely used. Only one migraineur used prophylactic medication. We performed k-means++ to group the 332 MOH patients into four clusters. Cluster 1 seemed to have tension-type headache-like headache characteristics, cluster 2 seemed to have MOH-like headache characteristics, cluster 3 seemed to have severe headaches with comorbidities such as dyslipidemia, stroke, and depression, and cluster 4 seemed to have migraine-like headache characteristics with photophobia and phonophobia. Conclusions: This is the largest prevalence survey in the Japanese elderly. Headache disorders are still the elderly’s burden. Clustering suggested that severe headaches associated with some comorbidities may be unique to the elderly.

Headache disorders are a common neurological disease with socioeconomic burdens among individuals of reproductive age, yet little is known about its impact on fertility intentions. This study examined the association between headache characteristics and fertility intention among parents of school-aged children with headaches in Japan. We prospectively conducted a school-based online survey for students’ parents in Tsubame City, Japan, in 2024. We asked about their age, sex, headache characteristics, the use of acute and prophylactic medications; monthly headache days (MHD); monthly acute medication intake days (AMD); Headache Impact Test-6 (HIT-6); Migraine Interictal Burden Scale-4 (MIBS-4); and the number of children. We also examined the impact of headaches on pregnancy plans by asking, “Are you avoiding or have you avoided pregnancy due to headaches?” and those who answered “yes” to this question were defined as the “avoid pregnancy group.” Of the 5,227 households, we received 1,127 (21.6%) responses, and 599 responses from parents with headaches were analyzed. The median (first quartile-third quartile) age was 43 (40–48) years, and 562 (93.8%) were female. They reported median MHD: 3 (1–4) days, AMD: 3 (1–6) days, HIT-6: 60 (58–68), and MIBS-4: 4 (2–8). Fifty (8.3%) used prophylactic medications, and 492 (82.1%) used acute medications for headache attacks. The median number of children was 2 (2–2). Twenty-two of the 562 female respondents (3.9%) answered that they were avoiding or had avoided pregnancy due to headaches. Both the HIT-6 score (median 58 [53–64] vs. 63 [59–66], p = 0.033) and the MIBS-4 score (4 [2–7] vs. 6 [4–7], p = 0.012) were significantly higher in the avoid pregnancy group. Multivariable analysis showed that avoid pregnancy group was significantly associated with: older age (odds ratio [OR] 1.16, 95% confidence interval [CI]: 1.05–1.29, p = 0.004), shorter headache duration (OR 0.91: 95%CI 0.85–0.98, p = 0.016), and a greater number of MHD (1.08, 95%CI: 1.01–1.16, p = 0.031), the presence of nausea or vomiting (OR 6.11, 95%CI: 1.46–25.60, p = 0.013), and phonophobia (OR 6.40, 95%CI: 1.71–23.99, p = 0.006). The avoid pregnancy group was more likely to express concerns about disability during pregnancy, parenting, and potential harm from medications. Some of the respondents were avoiding or had avoided pregnancy because of their headaches. Those in the avoid pregnancy group had both ictal and interictal severe headache burden and felt that headache disorders negatively affected fertility intention.

The misdiagnosis of headache disorders is a serious issue, and AI-based headache model diagnoses with external validation are scarce. We previously developed an artificial intelligence (AI)-based headache diagnosis model using a database of 4000 patients’ questionnaires in a headache-specializing clinic and herein performed external validation prospectively. The validation cohort of 59 headache patients was prospectively collected from August 2023 to February 2024 at our or collaborating multicenter institutions. The ground truth was specialists’ diagnoses based on the initial questionnaire and at least a one-month headache diary after the initial consultation. The diagnostic performance of the AI model was evaluated. The mean age was 42.55 ± 12.74 years, and 51/59 (86.67%) of the patients were female. No missing values were reported. Of the 59 patients, 56 (89.83%) had migraines or medication-overuse headaches, and 3 (5.08%) had tension-type headaches. No one had trigeminal autonomic cephalalgias or other headaches. The models’ overall accuracy and kappa for the ground truth were 94.92% and 0.65 (95%CI 0.21–1.00), respectively. The sensitivity, specificity, precision, and F values for migraines were 98.21%, 66.67%, 98.21%, and 98.21%, respectively. There was disagreement between the AI diagnosis and the ground truth by headache specialists in two patients. This is the first external validation of the AI headache diagnosis model. Further data collection and external validation are required to strengthen and improve its performance in real-world settings.

Objective: To investigate the prescription patterns for patients aged 6–17 years with headaches in the REZULT database. Methods: We cross-sectionally investigated (Study 1) the pattern of prescription and the proportion of triptan overprescription (≥30 tablets/90 d of triptans) among patients diagnosed with headaches in 2020. Next, we longitudinally studied patients (Study 2) for more than two years from the initial headache diagnosis (July 2010 to April 2022). The number of prescribed tablets was counted every 90 days. Results: In Study 1, headache diagnoses were assigned to 62,568 of 543,628 (11.51%) patients, and 1524 of 62,568 (2.44%) patients received acute medication. Single nonsteroidal anti-inflammatory drugs and triptans were prescribed to 620/624 (99.36%) and 5/624 (0.80%) of patients aged 6–11 years, respectively, and 827/900 (91.89%) and 91/900 (10.11%) of patients aged 12–17 years, respectively. Triptan overprescription was observed in 11/96 (11.46%) patients, and 5/11 (45.45%) of those patients received prophylactic medication. In Study 2, 80,756/845,470 (9.55%) patients aged 6–17 years were diagnosed with headaches that persisted for at least two years. Over two years, 44/80,756 (0.05%) patients were overprescribed triptans, and 3408/80,756 (4.22%) patients were prescribed prophylaxis on at least one occasion. Conclusions: Based on real-world data, the appropriate use of prophylactic treatment is still problematic. Overprescription of triptans was observed, although the number of patients was small.

Introduction Rapid onset and sustained efficacy are important for acute migraine treatment. Global phase 3 trials have demonstrated the early onset and sustained efficacy of the 5-HT_1F receptor agonist lasmiditan. In this prespecified analysis of the MONONOFU study, we assessed the onset and sustained efficacy of lasmiditan in Japanese patients with migraine. Methods MONONOFU was a multicenter, randomized, placebo-controlled, phase 2 study conducted in Japan (May 2019–June 2020). Eligible adults with migraine (N = 846; modified intent-to-treat population, N = 682) were randomized 7:3:7:6 to placebo, lasmiditan 50 mg, 100 mg, or 200 mg, taken orally within 4 h of moderate-to-severe migraine onset. Patients recorded headache severity and symptoms predose and 0.5–48 h postdose. Sustained and modified sustained pain freedom were defined as patients who were headache pain-free 2 h postdose and had no pain (sustained pain freedom) or had mild or no pain (modified sustained pain freedom) at 24 or 48 h without rescue/recurrence medications. Efficacy outcomes were analyzed by logistic regression. Patients also recorded the actual time of pain-free and of meaningful pain relief (Kaplan–Meier analysis). Results Compared with placebo, significantly more lasmiditan-treated (100 or 200 mg) patients were headache pain-free, had pain relief, were free of their most bothersome symptom, or had total migraine freedom (no headache or migraine-associated symptoms) within 30–60 min. Median time to pain-free was 9.26, 6.88, 2.75, and 2.30 h in placebo, 50-mg, 100-mg, and 200-mg lasmiditan groups, respectively. Significantly greater proportions of patients treated with 100 (19.7–29.5%) or 200 mg (21.1–35.7%) lasmiditan had sustained or modified sustained pain freedom at 24 or 48 h compared with placebo (10.4–15.8%). Conclusion This prespecified analysis of data from MONONOFU has confirmed that the efficacy of lasmiditan is rapid in onset and sustained in patients with moderate-to-severe migraine in Japan. Trial Registration ClinicalTrials.gov (NCT03962738).