To accurately examine the relationships between the amount of smoking versus levels of craving for smoking, depression, and menstrual phase-associated symptoms among the menstrual, follicular, and luteal phases, we improved the methodological issues described in former studies. The first problematic issue is the definition of the menstrual phases. We clearly defined the menstrual phases for each subject, as described in Materials and Methods. Benowitz et al. reported that estrogens accelerated the nicotine metabolism by comparison between women versus men [
18], but the activity of cytochrome P450 2A6, which is primarily responsible for nicotine metabolism, was not affected by menstrual cycle phase [
19]. Since the level of sex hormones fluctuated during the menstrual cycle, the association of ovarian hormones and smoking behavior has been examined in recent psychological studies [
20‐
22]. These studies attempted to directly examine the pharmacological effect of progesterone and/or estrogen (e.g., hormonal contraception) on women’s sensitivity to nicotine, but such effects on smoking behavior are also unclear. As oral contraceptives were reported to accelerate the nicotine metabolism [
18], women using hormonal contraceptives were excluded. Secondly, there is an issue with the method for measuring breath CO and urinary cotinine levels as objective indices of the amount of smoking. In previous studies [
6,
7,
9,
10], these biomarkers were measured once in each menstrual cycle when the subjects visited a hospital or research institution. Since the amount of smoking can vary every day, measurement should be performed much more frequently. In the present study, breath CO levels were measured every day to reduce such bias. Additionally, Craig et al. [
2] mentioned that women smokers tended to drink alcohol before a menstrual period, which could have caused an increase in the amount of smoking in the luteal phase. In this study, the data on days when the subjects drank alcohol or had no chance to smoke were excluded from the statistical analysis. We then concluded that the number of cigarettes smoked by young Japanese women and their breath CO levels significantly increased in the luteal phase compared with the follicular phase.
In this study, the level of craving for smoking in young Japanese women smokers in the menstrual and luteal phases was higher compared to the follicular phase without smoking cessation. Just as a long period of smoking cessation induces withdrawal symptomatology, a strong craving for smoking is sometimes noted even in smokers. Previous studies reported that the level of craving for smoking was higher and the power of concentration was decreased more in the menstrual phase compared with the follicular phase [
6], and that withdrawal symptoms worsened in the luteal phase [
6,
7,
23]. The present study also identified a strong positive correlation between the levels of craving for smoking and the amount of smoking in the luteal phase.
The levels of depressiveness in young Japanese women smokers in the menstrual and luteal phases were also higher compared to the follicular phase. The subjects of the study did not have premenstrual syndrome (including premenstrual dysphoric disorder) or dysmenorrhea. Nevertheless, the mean CES-D scores in the menstrual and luteal phases were 16 or higher - an indication of depression. This suggests that the level of depressiveness in young Japanese women smokers changes according to the menstrual cycle, and that a large number of these smokers develop depressiveness in the luteal phase. When screening the level of depressiveness in women smokers, careful attention should be paid to the relationship between the time of measurement and the menstrual cycle. There was a stronger correlation between the amount of smoking and the level of depressiveness in the luteal phase compared with other phases. The relationship between smoking and depression is well-recognized [
24,
25] and further studies should be conducted to determine whether depression in the luteal phase causes an increase in the amount of smoking or smoking increases the level of depression.
MDQ scores were high in both the menstrual (44.5 ± 18.2) and luteal (37.7 ± 15.9) phases. Our previous study reported that young Japanese female smokers showed levels that could be considered symptoms (MDQ score; 34.9 ± 19.2 in the menstrual phase, 39.5 ± 24.9 in the luteal phase) [
12]. In the menstrual and luteal phases, a moderate positive correlation was noted between MDQ scores and the amount of smoking. However, there was no significant correlation between them in the follicular phase. Although the MDQ subscale includes “negative emotions” similar to “depressiveness”, the correlations with the amount of smoking differed from each other in the follicular phase. This suggests that factors other than depressiveness may also contribute to the correlation with the amount of smoking in the menstrual and luteal periods.
Limitations
In this study, the subjects were 29 young Japanese female smokers aged from 19 to 25 years old. Smoking behavior and the menstrual phase-associated symptoms vary among countries, cultures and ages. The correlations between the amount of smoking and several symptoms shown in the study were possibly limited to these study subjects. Generalization of our results should be made only with caution. In addition, no cause-effect relationship between the amount of smoking and these symptoms has been determined. Further studies will be needed in the future.