BRIEF COMMUNICATIONS Predictors of premenstrual syndrome among Nigerian university students Ahizechukwu C. Eke a, ⁎, Josephat C. Akabuike a , Kelechi Maduekwe b a Department of Obstetrics and Gynecology, Nnamdi Azikiwe University Teaching Hospital, Nnewi, Nigeria b Department of Obstetrics and Gynecology, University of Calabar Teaching Hospital, Calabar, Nigeria article info Article history: Received 14 July 2010 Received in revised form 1 September 2010 Accepted 21 September 2010 Keywords: Nigeria Predictors Premenstrual syndrome Women Premenstrual syndrome (PMS) occurs in the luteal phase of the menstrual cycle, manifesting as physical, behavioral, and psychologic problems in the absence of organic or psychiatric disease [1]. Epi- demiologic surveys have shown that approximately 80% of women of reproductive age experience 1 or more of these symptoms [1], and PMS is associated with a significant impact on personal and public health [2]. The aim of the present study was to determine the pre- dictors of PMS in a sample of young Nigerian undergraduates because there is a paucity of studies of these predictors in Nigeria. The participants were female students at Nnamdi Azikiwe University, Awka, Nigeria. A multi-stage technique was adopted for sampling the students through their off-campus halls of residence—which contained 816 rooms, each inhabited by an average of 6 students. All even- numbered rooms (408 rooms) were selected for the study, and 3 students were randomly selected, via balloting, from each of these rooms. Of these 1224 women, 31 refused to participate. The 1193 participants completed a questionnaire—based on a modified version of the PMS-screening tool developed by Steiner et al. [3]—which required them to rate the severity of their symptoms, including breast tenderness, headache, bloating, weight gain, and insomnia. Severity codes were graded from 1 (mild) to 3 (severe). The questionnaires were initially pre-tested on 80 students for effectiveness, reliability, and validity. Statistical analysis was performed with SPSS version 16 (SPSS, Chicago, IL, USA), and group comparison was carried out via χ 2 test, t test, and analysis of variance. Significant variables were entered into logistic regression analysis to determine predicting variables. P b 0.05 at a 95% confidence interval was considered to be statistically significant. Data were incomplete for 14 of the 1193 participants, so the final analysis was based on 1179 students. Mean age was 20.05 ± 3.18 years, and average age at menarche was 12.22 ± 1.24 years. Mean duration of menstrual flow was 5.15 ± 0.22 days, with an aver- age cycle length of 28.24 ± 4.36 days. Mean body mass index (BMI, calculated as weight in kilograms divided by the square of height in meters) was 19.06 ±5.71, and at least 1 premenstrual symptom out of 14 was experienced by 996 (84.5%) women, with the most severe symptoms experienced by 368 (31.2%). Of the 1179 women, 604 (51.2%) had a family history of PMS, 597 (50.6%) experienced physical and mental stress, 493 (41.8%) reported depressive mood episodes, and 677 (57.4%) consumed sweet-tasting foods. Analysis of the associations between PMS and predictors showed that the variables significantly associated with PMS were family history (χ 2 =3.29; df=1; P =0.02); increased physical and mental stress (t = –3.24; df = 407; P = 0.004); low BMI (t = –3.29; df = 407; P = 0.02); young age at menarche (t =–3.33; df = 451; P = 0.03); longer duration of menstrual flow (t =3.63; df=451; P = 0.001); depressive mood (t =3.37; df=1; P =0.04); sweet foods (t =3.87; df=1; P =0.002); and increasing age after menarche (χ 2 =3.33; df=1; P =0.03). Logistic regression analysis showed that family history (β = 0.23; df = 1; P = 0.004); longer duration of menstrual flow (β =0.46; df=1; P =0.001); and young age at menarche (β =–0.21; df=1; P = 0.002) were independent predictors of PMS. The results are similar to findings from another study in Saudi- Arabia [4], but different from those from a study in Iran [2]. Several factors—including cultural disparity, study population characteristics, and dietary differences—might explain the variation among these populations. The present findings are particularly interesting because cultural beliefs may have affected the participants’ perception of PMS symptoms and their severity. Most of the students experiencing PMS who had a low BMI had lost weight in a bid to look more attractive. There were limitations to the present study. It was performed at a single center and was cross-sectional, so caution must be taken when inferring causality. A prospective large-scale and multicenter study should be carried out to validate these findings. The strength of the investigation was that it involved a culture (Igbo) that had not been studied previously. Because PMS was prevalent among young Nigerian undergraduates in the present study, we recommend that university healthcare providers should screen students routinely and offer treatment, as appropriate, for PMS. Because PMS reportedly affects school performance and attendance, students should be provided with more health education on this topic. Conflict of interest The authors have no conflicts of interest. International Journal of Gynecology and Obstetrics 112 (2010) 63–71BRIEF COMMUNICATIONS ⁎ Corresponding author. Tel.: + 234 806 863 4110. E-mail address: ahizeeke2nd@yahoo.ca (A.C. Eke). International Journal of Gynecology and Obstetrics 112 (2011) 63–71 Contents lists available at ScienceDirect International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo