The phenomenology of premenstrual syndrome in female medical students: a cross sectional study.

BACKGROUND
The premenstrual syndrome (PMS) is particularly common in the younger age groups and, therefore represents a significant public health problem in young girls. This study aims to estimate the prevalence, severity, determinants of premenstrual syndrome (PMS) and its impact among the female medical students in Al-Ahsa, Saudi Arabia.


METHODS
This study was performed at the College of Medicine, King Faisal University, Saudi Arabia, from June through December 2009. It included 250 medical students. They filled different questionnaires covering American College of Obstetrics and Gynecology (ACOG) criteria to diagnose PMS, demographic & reproductive factors, physical activity and mental condition. Regression analysis was conducted for all the predictors.


RESULTS
PMS was diagnosed in 35.6% of cases, distributed as 45% mild, 32.6% moderate and 22.4% severe. There were significant trends for older age, rural residence, family income and family history of PMS. The dominant limited activity was concentration in class (48.3%). Limitations of activities were significantly more frequent among severe cases. The prevalence of anxiety and depression was statistically more evident in the PMS group. Regression analysis revealed that, PMS was significantly associated with older age groups, rural residence, lower age at menarche, regularity of menses and family history.


CONCLUSION
PMS is a common problem in young Saudi students in Al Ahsa. Severe PMS was associated with more impairment of daily activities and psychological distress symptoms. Older student age, rural residence, earlier age of menarche, regular cycles and positive family history are possible risk factors for PMS.


Participants
This cross-sectional study was conducted at the College of Medicine, Al Ahsa, King Faisal University (KFU), Saudi Arabia over a six months period.
All the female students were the target population of the study. Approximately 288 letters and consent forms were distributed to the students. Of these, 271 (94.1%) were returned, signed by participants.
The Students with current medical, psychiatric or gynecological problems were excluded from the study including pregnancies, amenorrhea and significant pelvic pain secondary to a proven or presumptive diagnosis of pelvic inflammatory disease or endometriosis (13 students). Further 8 cases data were incomplete. The final sample was composed of 250 (86.8%) students. This sample completed the ACOG questionnaire, demographic, reproductive and DASS forms.

Instruments
All the questionnaires were self-reported and were completed by the participants with the aide and observation of a trained researcher about all aspects of the questionnaires. ACOG PMS diagnostic criteria: A questionnaire was constructed based on ACOG PMS criteria [1,4] including the following six behavioral and four somatic symptoms; depression, angry outbursts, irritability, anxiety, confusion and social withdrawal breast tenderness, abdominal bloating, headache and swelling of extremities.
Depression Anxiety Stress Scales (DASS): DASS is a set of three self-report scales designed to measure the negative emotional states of depression, anxiety and stress. The essential function of the DASS is to assess the severity of the core symptoms of depression, anxiety and stress.
Overall scores are calculated by summing the scores for the relevant items [11]. The Arabic version carried out by Taouk was used [12 ].

Procedures
The participants were given liberal verbal explanations plus description letters about the topic and the aim of the study with attached consent forms. Body Mass Index (BMI) was calculated using self-reported data on height and weight. BMI was categorized using cut off points recommended by the National Institute of Health. While four BMI categories were described, women categorized as obese (BMI ≥ 30) were of particular interest in our study.
Participants were deemed to meet the ACOG for PMS if they rated their experience of at least one of the six behavioral symptoms and one of the four somatic symptoms. These symptoms must be recorded in the absence of any therapeutic intervention resulting in social or physical dysfunction and if there was no history of psychiatric and non psychiatric conditions. Symptoms should start during the five days before the menses and relieved within four days of the onset of the menses without recurrence until at least cycle day 13 and are evident for two consecutive cycles.
To estimate the severity of PMS, each item was rated on a scale of 0 "not at all" to 3"extreme". The highest score of each symptom in the premenstrual period was calculated. Then the total score of PMS was calculated as the sum of the symptom's score divided by the number of 4 symptoms (mean) and converted to percent. Therefore, the score between 0% -33% represented mild form of PMS, 33% -66% as moderate and more than 66% was accounted as a severe form of PMS.
ACOG PMS criteria were strictly followed but without the prospective recording for two further cycles due to the socio cultural barriers that interfere with daily reporting of such sensitive issues as menstrual-related symptoms. Data were collected by trained female interviewers. They underwent 6-hour training on the ACOG criteria in 3 separate sessions.
In recording of DASS score items, subjects are asked to use 4-point severity/frequency scales to rate the extent to which they have experienced each state over the premenstrual week ranging from 0 (did not apply to me at all) to 3 (applied to me very much, or most of the time).

Statistical Analysis
Data were analyzed using the Statistical Package For the Social Sciences, SPSS Inc., Chicago, IL (SPSS version 16). The different sociodemographic, biological and reproductive variables were presented, compared and analyzed using independent t test for continuous and ordinal variables and x 2 and Fisher exact tests for categorical variables. x 2 test was used to compare the scales different severities of PMS was used.
Variables that were significantly affecting prevalence of PMS on this initial analysis by the x 2 test were introduced into the regression analysis model. Different sets of regression analysis were done for each group of variables, then all the significant variables were grouped into final regression analysis to evaluate the role of the different independent variables on the dependant variable; PMS. A p value of < 0.05 was considered significant.

Results
Of the 250 students approached, PMS was diagnosed in 89 (35.6%) of them using the ACOG criteria. The socio-demographic and reproductive characteristics of the study population are given in Table 1. Among PMS group, the mean age of participants was 20 years. Approximately, twothirds of subjects were from rural areas with unsatisfactory income and their mean age of menarche was 11.88 years and the mean body mass index was 30.8. There were significant difference in the socio-demographic data of those with and without PMS such as age, residence, father occupation, family income and family history of PMS. The PMS had a significant trend for older age, rural residence, unstable jobs of the fathers and unsatisfactory family income. Furthermore, PMS group were more likely to report earlier age of menarche, more regular cycles, body mass index greater than 30 and positive family history of PMS. Also, there was a trend for more significant depression and anxiety based on DASS (Table 1).
Using the ACOG criteria, PMS was diagnosed as described before and the total assigned score was categorized into mild, moderate and severe.
The impact of PMS on women's daily activities was detailed in Table 3. The activities reported to be limited were concentration in class (48.3%), college attendance (46%), going out of the home (43.8%), daily home chores (41.6%) and homework tasks (36%). Limitations of these activities were significantly more frequent among severe cases ( Table 3). As depicted in Table 1, although the prevalence of stress across students with and without PMS did not reveal any significant difference, anxiety and depression scores were statistically more evident in the PMS group. The frequency of depression, anxiety and stress in PMS was presented in Table 3. They were significantly more frequently reported among students with severe PMS.
Regression analysis of the variables that were significantly associated with PMS in the initial x 2 testing was done for all sets of predictors by category. All significant predictors were grouped in the final regression analyses presented in Table 4. PMS was significantly associated with older age groups, rural residence, lower age at menarche, regularity of menses and family history.

Discussion
This study sheds new light on the phenomenology of premenstrual tension syndrome by estimating the prevalence, symptomatology, physical and mental health impact and identifying the risk factors of the syndrome in a sample of Medical students in Al Ahsa, Saudi Arabia using a structured interview.
The prevalence of PMS in the present study (35.6%) was in accordance with the work of Serfaty et al [13] and Dean et al [14] who reported prevalence of 35%, and 19-30% respectively. Other Western investigator reported higher prevalence of 85% [15]. In Egypt, El-Defrawi et al [16], reported prevalence of 69.6 % while Rasheed and Al-Sowielem [17] in Saudi Arabia, reported a prevalence of 96.6%. A cross cultural investigation conducted in 14 different cultural groups in 10 countries found a lower prevalence rate (23-34%) in nonwestern cultures, while a higher prevalence rate (71-73%) was reported in the western countries [18]. The justification for such difference depends on varied definitions; methods of data collection, sampling technique and the type of study population.
The frequency distribution of the PMS cases as measured by ACOG was allocated as; 45% mild, 32.6% moderate and 22.4% severe cases. This order of frequency was nearly similar to what was reported by Tabassum et al [19]. It was interesting to notice that the frequency of severe PMS was high in our research in contrary to what had been reported by Abuhashim et al [20] and Nisar et al [21] (5.8% and 4.4% respectively). The difference could be due to the recent increase in empowerment and positive gender attitude of young women in Saudi Arabia as a result of rapid development and modernization of the society thus increasing the perception and awareness.
In the current study, the most frequently reported symptom was abdominal bloating (75.3%), which was also reported in previous studies [22].
However, Derman et al [23] reported that the most common symptom was negative affect group as stress and nervousness. This difference may be due to different cultural and socio-demographic variables. Grant stated that individuals in low social ladder may not cope with the stress of the increasingly more challenging environment that may negatively impact physical and psychological well-being [24].
Moreover, the most frequent symptoms in mild and moderate cases were somatic (abdominal bloating, breast tenderness and headache).
Whereas, the most frequent symptoms in severe cases included both somatic (abdominal bloating, breast tenderness) and psychological symptoms (confusion, irritability, angry outbursts, social withdrawal and depression). These findings were consistent with Antai et al [25] who showed that somatic symptoms predominated the group with mild -moderate symptoms while mood-related symptoms were predominant in severe conditions. It was found that 37% of students with PMS reported greater impairment of daily activities; concentration in class (48.3%), attending college (46%), going out of the home (43.8%), daily home chores (42%) and homework tasks (36%). Others reported similar findings [14,25]. This study denoted also that severe degree of PMS was associated with more physical impairment. Montero et al [26] and Tenkir et al [27] reported that academic absence and low achievement was significantly more frequent among college students with severe PMS. Moreover, Yang et al [28] reported that severe menstrual distress was associated with greater burden on mental and physical health than any chronic disease and even comparable to the effect of depression.
In this study, although the preva¬lence of stress across students with and without PMS did not reveal any significant difference, anxiety and depression scores were statistically more evident in the PMS group. They were significantly more frequently reported among students who reported having severe PMS. The role of stress and major life events has received considerable attention in terms of potential associations with somatic health. Nisar et al [21] reported that the associations between stressful events such as loss of a loved one, recent breakup, work or financial difficulties, and illness and PMS may, therefore, parallel the effect of stressful events on those who are vulnerable to episodes of major depression.
In a nonclinical sample of 91 college students, Portella et al [29] observed that there was a sizable positive correlation between seasonal depressive symptoms and premenstrual symptoms. Perkonigg et al [30] studied 1488 women aged 14-24 years and found that history of traumatic events, history of anxiety disorders and elevated 'daily hassles' scores were powerful predictors of the development of severe PMS. The difference between these studies and our research may be due to the used tool. We used DASS -21 which had the advantage that: it was psychometrically validated and developed in consideration of cross cultural situations and that our results would reflect the state of these disorders among Saudi young women.
Many factors were analyzed in this study as predictors of PMS using the regression analysis for the age groups, residence, age of menarche, regularity of cycles, and family history.
PMS had a significant association with older age groups in our study. While some authors reported that PMS was increasing with age [31], others failed to find such correlation [32]. Clecknedr-Simth et al [33] found that symptoms were more intense in the 16-18 years group compared to the 13-15 years age group. Bakhshani et al [34] found that the 18-20 years old age group had the highest figures.
Rural residence in our sample was associated with increased PMS. Shershah et al [35] studied PMS in Karachi and found that its prevalence was 33% with the highest figures in lower socioeconomic group living in socially deprived areas. On the contrary, it was reported to be more frequent in young, literate, urban women with more intense symptoms. Despite the level of perception may be high in urban cases, yet the presence of other co-factors, social habits and different living conditions may explain the higher prevalence in rural areas [36]. Our findings of positive associations between PMS and a family history of PMS are similar to some studies done for women in USA and Saudi Arabia [17,37]. Shared biological and/or psychological factors which may influence expectations and self-awareness may explain mother-daughter dyads.
In the current study, there was an association between regular cycles, younger ages of menarche and development of premenstrual syndrome.
These finding were consistent with some previous investigators [2,15,20]. On the contrary, others didn't find any association between PMS and age at menarche [16,21]. The findings in our study could be explained with the fact that earlier age of menarche and regular cycles are associated with early establishment of ovarian functions and ovulation with fluctuation of steroid hormones in such a young age with less physical and psychological maturity which may lead to PMS manifestations.

The limitations of this study must also be recognized.
First, our study included a highly selective sample comprising of medical students from one academic institute which will limit the generalizability of the findings. Second, because of the cross-sectional design of the study, we are unable to determine longitudinal relations between any of the studied predictors and outcome and whether they were coexisting or preexisting. Third, Despite the students were included in the study based on absence of medical chronic disorders, yet they were not screened for other possible medical diagnoses when they reported PMS symptoms. Finally, we depend on the retrospective analysis using questionnaires as it was inherently difficult or even impossible to use the prospective approaches.
Despite the questions asked were standardized and have been used in other Arab studies [38], we believe that the questionnaire filling is likely to pose some biases, either in the recall or differential classification during the filling with either over or under reporting.

Conclusions
PMS is a common problem in young students in this part of the world. Severe PMS was associated with more premenstrual symptoms, impairment of daily activities and psychological distress symptoms. Older student age, rural residence, earlier age of menarche, regular cycles and positive family history could be considered as predictors for PMS. Further studies on large sample of population with more preferably prospective approach need to be conducted to confirm these results and to plan out strategies for better detection and management of PMS in young women. The introduction of a reproductive health component into college health education program could help in providing information, education and support to the young students.

Competing interests
The authors declared they have no competing interests. Also there are no sources of funding. Tables   Table 1: Pertinent clinical and sociodemographic characteristics of the study population Table 2: American College of Obstetrics and Gynecology (ACOG) diagnostic criteria of premenstrual syndrome Table 3: Impairment of student physical and mental activities in premenstrual syndrome group Table 4: Final regression analysis of independent variables significantly associated with with premenstrual syndrome    Items in daily activity were assessed during the premenstrual phase compared to the rest of the month; equal, increased or limited using an activity checklist. Concentration in the class (No of times you were absentminded and the tutor directed you), College attendance (Score of attendance), Going out of the home (No of visits to colleagues or relatives), Daily home activities (Usual home arrangement and sharing in work), Homework tasks (completion of tasks on time)