The study was approved by the Ethics Committee of the Chinese PLA General Hospital (2,020,263). The study protocol complied with China’s regulations and Guidelines for Good Clinical Practice. Due to the data collecting nature of the study, oral informed consent was obtained from the patients before inclusion in the study according to the World Medical Association’s Declaration of Helsinki.

This study was conducted in a tertiary headache clinic at the People’s Liberation Army (PLA) General Hospital between January 2015 and January 2020. All females diagnosed with MWA according to the International Classification of Headache Disorders—Third Edition (ICHD-3) diagnostic criteria were inquired for content to take part in the study. Patients who agreed to participate were screened and reassessed by at least two qualified and experienced headache experts to exclude atypical migraineurs. The atypical migraineurs included patients who did not have enough attacks (less than 5 times), did not have enough accompanying symptoms (no nausea/vomiting/photophobia nor phonophobia), did not fulfil the duration of the attack (less than 4 h), and did not fulfil the characteristics criteria (less than 2 out of 4 items). (The detailed procedure is shown in Fig. 2).

Flow chart of recruitment. There were 1444 female patients with migraine without aura participating in the study. The chart shows the screening details

Chronic daily headache (CDH) and medicine overuse headache (MOH) were diagnosed at the same time. Electrical medical records (including general items and migraine related items) were collected for every patient. Male MWA patients were excluded from the study and the reason will be explained in the Discussion.

We will share anonymized data upon reasonable requests from any qualified investigator. The electronic medical data were all collected by software shared through the “cloud”, rendering the data available online.There was no interesting conflict.

Age, gender, diagnosis (MWA, MOH and CDH), group (J1, J2, or J3), general items (BMI, smoking status, consumption of tea/coffee, physical exercise habits, alcohol consumption, and education level), headache characteristics [course, location, side, pulsating pain, duration, frequency, Number Rating Scale (NRS) score, aggravation by routine physical activity, accompanying symptoms (nausea, vomiting, photophobia, or phonophobia), triggers, premonitory symptoms, and aggravation after childbirth (for J1 and J2 groups)] and migraine-related factors (family history, menstrual relationship, sleep state, Patient Health Questionnare-9 (PHQ-9) score and Generalized Anxiety Disorder-7 (GAD-7) score were collected. Premonitory symptoms were asked with a structured questionnaire of 25 items including overactivity, loquacity, mood change, irritability, dysesthesia, drowsiness, fidgeting, concentration changes, photophobia, phonophobia, osmophobia, dysphasia, yawning, stiff neck, food cravings, poor appetite, sensation of coldness, fatigue, diarrhea, constipation, thirst, diuresis, dizziness, edema, and others (unlisted). Triggers were assessed with a questionnaire consisting of 9 items, including sleep disorder, tiredness, foods, nervousness, exercise, sunshine exposure, environmental changes, weather changes and odors.

MOH was diagnosed according to the ICHD-3 criteria. CDH was defined as headache occurring on 15 or more days per month for more than 3 months. Aggravation after childbirth is defined as more than double in frequency, increasing 2–4 points on the pain scale (NRS), or both. Headache location was roughly divided into three parts (front: periorbital, forehead, and temporal areas; back: occipital and neck areas; or other). Among the items listed above, consumption of tea/coffee was analyzed qualitatively and quantitatively (tea: 0 cup, 1–5 cups, 5–10 cups, or more than 10 cups per day; coffee: 0 cup, l-2 cups, or more 2 cups per day), and the education level was divided into three categories according to grades (no education, elementary and advanced).

To evaluate the degree of typicality of migraine, several headache and headache-related characteristics were quantified by weight for analysis, including the unilateral side, pulsating quality, severity of pain (moderate or severe), aggravation by routine physical activity, accompanying factors (nausea, vomiting, photophobia, or phonophobia), menstruation relationship, family history, and presence of premonitory symptoms (PSs), or triggers. These categorical variables were assigned weights, and the cumulative sum was taken as the typical score for qualitative (binary classification of scores as higher or lower than 8/17) and quantitative analyses for each group.

Statistical analyses were performed using SPSS (version 23.0). Numerical data were analyzed using the Kruskal–Wallis test and the Mann–Whitney U test. Categorical data were analyzed using the χ2 tests. Non parametric test was adopted for grade data. A two-sided p value < 0.05 was considered statistically significant. Missing data were excluded from the analyses.

The analysis of PS items was abandoned if the frequency was less than 4.

As we found in clinics, stress and stress-related sleep disorders could not only induce migraine but also promote the aggravation of migraine (migraine new occurrence and migraine aggravation, correspondingly). Thus, to exclude the impact of stress after childbirth on the J1 and J2 groups, the J1 and J2 groups were further subdivided into the J1-1 and J2-1 (aggravation after childbirth) groups and the J1-2 and J2-2 (unchanged after childbirth) groups. In addition, the J1-1 and J2-1 were merged with the J3 group either combined or separately analyzed (migraine new occurrence and/or migraine aggravation after childbirth accordingly). Thus, the statistical analysis was divided into the following four parts: J1vs. J2 vs. J3 (Part 1), J1-2 vs. J2-2 vs. (J1-1 + J2-1 + J3) (Part 2), (J1-1 + J2-1) vs. J3 (Part 3) and (J1-2 + J2-2) vs. (J1-1 + J2-1) (Part 4). Further details on the statistical analysis are shown in Fig. 3.

The detailed statistical project of the study. The patients were divided according to the association of MWA onset with menarche and childbirth. Group J1: onset before menarche, group J2 onset after menarche but before childbirth, group J3 onset after childbirth. Patients who reported aggravation after childbirth in group J1 and group J2 were marked as J1-1 and J2-1, and those who reported no change were marked as J1-2 and J2-2

The following factors were compared among groups in different parts: family history, CDH, MOH, number of headaches per month, headache characteristics (side, location, pulsating quality, NRS score, aggravation by routine physical activity, accompanying symptoms including nausea, vomiting, photophobia, phonophobia, with PSs and sum of PSs, with trigger and sum of triggers), education level, BMI, smoking status, consumption of tea/coffee, alcohol consumption, sleep disorder, depression (PHQ-9), anxiety (GAD-7) and typical score.