Epidemiology of premenstrual symptoms and disorders

Prevalence of premenstrual disorders

Up to 90% of women of childbearing age experience at least one premenstrual symptom during their reproductive years. Symptoms begin up to two weeks before menses and end soon after the onset of the menstrual period. ^1–3 Symptoms include physical (e.g. abdominal bloating, breast tenderness) and emotional (e.g. irritability, depression and mood swings) complaints and differ in their severity (mild to severe), duration and frequency. ^1–3

Different definitions exist for the categorization of premenstrual symptoms. The World Health Organization (WHO) has developed a definition of premenstrual tension syndrome (PMS) in the ICD-10 which requires the presence of just one distressing symptom from a range of physical and emotional symptoms. Other, stricter definitions were used in epidemiological studies on PMS. Deuster et al. ⁴ (1999) conducted a study using a random sample of women between 18–44 years of age in Virginia, USA. Eight percent of the women were categorized as PMS patients using retrospective evaluation of symptoms. In the study of Campbell et al., ⁵ women visiting their general practitioners were asked to complete the validated retrospective premenstrual assessment form. A total of 32% of the study participants fulfilled the criteria for PMS.

The American College of Obstetricians and Gynecologists (ACOG) developed a stricter definition of premenstrual syndrome. ² Specifically, one emotional and one physical symptom must be present during the five days before menses in at least three consecutive menstrual cycles and should be verified over two consecutive months of daily prospective charting.

The ACOG criteria are useful in the diagnosis of PMS, but the strict definition that includes daily symptom recording has only been used in a few studies that have assessed the prevalence of PMS. The prevalence of PMS in the study by Steiner et al. ⁶ (2003), which included 508 women from a convenience sample in Canada, was 20.7%. In the Borenstein et al. ⁷ (2003) study, a convenience sample of 436 women completed the validated daily record of severity of problems (DRSP) scale for two consecutive months. A total of 28.7% of the women met the PMS criteria.

At the most severe end of the premenstrual symptom spectrum is PMDD, defined by the Diagnostic and Statistical Manual of Mental Disorders criteria (DSM-IV). ³ An estimated 3–8% of women of reproductive age are affected by this disorder. ^8–13 The DSM-IV criteria require that at least five symptoms (including at least one emotional symptom) out of 11 symptom categories must occur during the week leading up to menses and improve within a few days of the onset of menses. Symptoms must be severe enough to impair daily functioning. Also, the diagnosis of PMDD has to be confirmed with daily prospective symptom ratings for two or more consecutive menstrual cycles.

Several population-based studies have been conducted to assess the prevalence of PMDD verified by daily prospective symptom charting, and have reported estimates between 3 and 8%. Sveindottir et al. ⁸ (2000) conducted a prospective study using a random sample of 250 women in Iceland. Six percent of the women met the PMDD criteria. A similar prevalence (6.4%) was reported by Banerjee et al. ¹⁴ (2000) in a study in Indian women and by Cohen et al. ¹⁵ in older premenopausal women in a population-based study in the USA. Chawla et al. ¹⁶ conducted a study using randomly selected women from a US Health Maintenance Organization. A total of 1194 women provided prospective daily symptom chartings for two menstrual cycles. The prevalence of PMDD was 4.7%. Studies that defined PMDD through the retrospective evaluation of symptoms found prevalence rates of severe premenstrual symptoms in the same order of magnitude (2–6%). Wittchen et al. ⁹ evaluated the prevalence of PMDD in a representative community sample of 1091 women in Germany, and 5.8% met the PMDD criteria. Steiner et al. ⁶ (2003) categorized 5.1% of the participating women in their study as having PMDD using retrospective evaluation.

More recently the International Society for Premenstrual Disorders (ISPMD) defined precise criteria for core premenstrual disorder. Symptoms occur in ovulatory cycles, during the luteal phase, resolve by the end of menstruation and are followed by a symptom free interval. Significant impairment of day-to-day activities at work/school, social activities/hobbies, interpersonal relationships is a key feature. Symptoms are non-specific but there are typical somatic and psychological symptoms. ¹⁷ There have not, as yet, been epidemiological determinations of the prevalence of the core premenstrual disorder although both PMS and PMDD fulfil the ISPMD criteria for core premenstrual disorders.

Burden of premenstrual disorders

Borenstein et al. ¹⁸ evaluated the impact of premenstrual syndrome on direct and indirect costs in 374 women. The degree of premenstrual symptoms was diagnosed by two months of daily charting using the DRSP scale. Women with PMS showed an average annual increase of $59 in direct costs (P < 0.026) and $4333 in indirect costs per patient (P < 0.0001) compared with women without PMS. Based on the same study data, Borenstein et al. ¹⁹ analysed the health-care utilization of women with and without PMS. Women with no or mild premenstrual symptoms had fewer workdays with reduced productivity per month compared with those with moderate to severe symptoms (13.3 and 22.0, P < 0.0001), and missed fewer workdays per month due to health reasons (1.2 and 2.7, P = 0.001). Women with more severe PMS/PMDD showed an 80% higher risk of having >10 visits to their physician compared with women with fewer symptoms (OR = 1.8, 95% CI 1.0–3.2), and the risk of accumulating more than $500 in health-care charges was twice as high (OR = 1.9, 95% CI 1.2–3.0).

Heinemann et al. conducted a prospective observational study with over 4000 women in 19 countries to evaluate the effects of moderate to severe premenstrual symptoms (PMS/PMDD) on work productivity, absenteeism and daily life activities. The degree of premenstrual problems was evaluated by daily charting over two months using the validated DRSP scale. Women with moderate-to-severe premenstrual problems reported higher levels of work absenteeism (up to 5.7 times higher), of work productivity impairment (up to 5.8 times higher) and of impairment of daily life activities (up to 6.4 times higher) compared with women with no or only mild symptoms. ^20–22 These observations were consistent across the different regions, but less pronounced in most of the Asian countries.

Based on the strict criteria of PMDD, the disability-adjusted life years (DALY) equate to approximately 14 million in the USA alone. However, this is a conservative estimate, since women who suffer from severe premenstrual symptoms but who nevertheless do not fulfill the PMDD criteria were not included in this estimate. ¹⁰ Yang et al. ²³ (2008) compared the burden of PMDD on health-related quality of life (HRQoL) to the US general population and, specifically, chronic diseases. The authors found that the HRQoL burden for PMDD was higher than for chronic back pain and comparable to osteoarthritis and rheumatoid arthritis in all SF-12 scales.

Premenstrual symptoms

Of the disorders described only PMDD has been accepted by regulators, such as the Food and Development Authority. This has led to a research focus on the prevalence, aetiology and treatment of PMDD, which is heavily weighted in symptom type to mental or psychological symptoms. ³ Yet both researchers and clinicians recognize that relatively few of the women presenting for treatment with problematic premenstrual experiences meet criteria for PMDD. ^24,25

A series of cross-sectional survey studies were carried out to determine how women in the community perceive symptoms premenstrually and what factors are associated with more problematic premenstrual symptoms. It was decided to include all symptoms that had been mentioned by the major classificatory systems. As different persistence criteria were given for ACOG ² and WHO ¹ defined PMS (3 months and 1 month respectively) and DSM-IV defined PMDD ³ (most cycles in 12 months), women were asked how many months each premenstrual symptom was experienced. Information was also collected on women's knowledge of the terms PMS and PMDD. These cross-sectional surveys were carried out in Europe and Latin America in 2003–4 and later extended to Asian countries in 2008–9. We have separately published the results for Europe/Latin America ^26,27 and some (but not all) of the Asian countries. ^28–31

The sample included 7226 women (400–500 women of reproductive age, were randomly selected for interview in each of the following countries: European: Germany, Spain, Italy, Hungary, France, UK; Latin American: Mexico and Brazil; and Asian: Hong Kong, Pakistan and Thailand. In addition Internet panels were used to select 1000 Korean and Japanese women and 500 Australian women for a web based questionnaire).

Statistical analysis of all the women in the global data has generated the following key findings, which may have significance in understanding women's experiences of premenstrual problems. ³²

The most prevalent symptoms in terms of women's perception of severity and duration (or persistence of symptoms across cycles) were abdominal bloating, cramps or abdominal pain, irritability, mastalgia and joint/muscle/back pains. Severity of symptoms was directly proportional to duration (number of affected cycles) (R = 0.78).

Clearly these findings have clinical implications. As the most prevalent premenstrual symptoms women perceive are physical, then these are likely to be the predominant complaints to their doctors. There are also research implications as the symptoms women report as most severe and frequent should be included and weighted appropriately in rating scales and addressed by treatment interventions. This has not been the case in the measures designed for PMDD, in which focus has been on psychological symptoms. ³

Hierarchical clustering of symptoms found two well separated categories:

Group 1 (Mental Domain): depressed mood, anger, irritability, mood swings, anxiety, not in control, restlessness, hopelessness, tension, confusion, social withdrawal, sleep disturbances, poor concentration, lack of energy.

The study explored the influence on premenstrual symptom prevalence of various sociodemographic factors. Only four significant explanatory factors were found: firstly age, with a linear and quadratic effect producing an inverse u-curve where younger and older women were characterized by a smaller intensity of symptoms, and a maximum symptom intensity was reached around 35 years. Women's menstrual cycling prior to age 36 is characterized by regular ovulatory cycles and they are likely to enter the early phase of the menopausal transition after this, with increased anovulatory cycles (and less associated symptoms). After the age effect, the next two explanatory variables were parity (with reduced premenstrual symptom severity proportional to number of children), and smoking habits, where smokers were associated with a strongly increased intensity of symptoms. After adjusting for age and smoking, a significant heterogeneity was found between countries but this was only a small effect. Women in some countries reported larger values of premenstrual symptom severity and duration. For example the following countries were characterized by the largest values of severity and duration: (by decreasing value) UK, Brazil, Japan, Korea and Australia. Two countries were associated with the smallest severity and duration: Hong Kong and Pakistan. All the other countries are characterized by median values that do not differ from each other. Thus this global data base has demonstrated that premenstrual symptom experience does not clearly differentiate countries and regions. ³²

Activities of daily life (ADLs) were shown to be equally influenced by mental and physical symptoms. Increasing education level increased perception on ADLs, while exercise participation was associated with a lowered perception. There was no significant difference between countries. Thus the woman's perception of effect of premenstrual symptoms on her ADL is not influenced by country or culture. ³²

Those women with more severe and persistent symptoms were more likely to know of PMS terminology. It was not surprising that older women and those with better educational levels would be more aware of these terms. What had not been previously documented is the very strong difference between countries in knowledge of these terms: by decreasing knowledge: (a) western-style countries (Europe and Australia), (b) Latin American countries and East European countries, (c) Japan, (d) other Asian countries. Clearly Asian women have lower knowledge of these syndromes. ³²

Conclusions

Women frequently experience troublesome premenstrual complaints. Recent studies show that four of the five most prevalent premenstrual symptoms are physical. There were a great deal of similarities of women's experiences of these symptoms across countries and regions. More research attention needs to be given to these symptoms. A number of studies have shown considerable burden of illness of premenstrual symptoms.

Competing interests

Lorraine Dennerstein: speaker and consultant honoraria Bayer Schering; Philippe Lehert: Senior Consulting Statistician for the following laboratories – Merck Kgaa, Sanofi-Aventis, Ipsen, Serono, Bayer Schering; Klaas Heinemann: Bayer Schering Pharma employee at the time the studies were carried out.