16/11/2014 20:10

Effectivenes of cognitive behavioural therapy in treating premenstrual syndrome


Pre menstruation syndrome (PMS) has been known for centuries. In the paper of Ussher (2003) PMS is described as “widely recognised psychological problem” which effects up to 40% of women. It first appeared officially in the 3rd edition of Diagnostic and Statistical Manual of Mental Disorders (DSM). Around that time feminist authors started to rebel against labelling cycle changes as an illness (Rome, 1986). Indeed mood fluctuation and some physical symptoms (like bloating and mild weight gain) are normal during the cycle (Sarris, Bewley & Agnihotri, 2009). According to Nevatte et al (2013) mild to moderate PMS symptoms are so common in female population they should be regarded as normal.

When the symptoms get aggravated it may require medical attention. Various societies classify diagnostic criteria for the diagnosis of PMS however the most used is DSM – IV criteria by American Society of Psychiatrists. The patient has to have at least one of the four core symptoms, which is depressed mood, anxiety, labile mood or anger, plus at least four other symptoms. This has to be assessed over at least two cycles and the fluctuation dependent on the cycle must be proven – symptoms free period after menses and aggravation of symptoms within the last 5 days of the cycle, before the period. This labels PMS as a psychological condition despite the fact that according to results of Dennerstein et al. (2011) and Wittchen et al. (2002) physical symptoms were more prevalent. This will be possibly reflected in the new DSM – V.

The symptomatology of PMS was thought to be due to biological changes throughout cycle. This approach does not however explain why some women suffer from it more than the others. This is further supported by relative inefficiency of hormonal treatment with contraceptive pills and good response to antidepressants (Nevatte et al., 2013).

Importance of other external psychological factors had to be taken into account. Hamilton (1988) investigated a link between relationship disharmony and premenstrual symptoms.  Along with other studies he found higher levels of dissatisfaction in the relationship in women with PMS (Coughlin, 1990; Frank, Dixon & Grosz, 1993). Expanding on that idea of Jones et al (2000) came to a conclusion that couples therapy might be a great help in PMS as those issues seem to aggravate significantly PMS symptoms. Indeed Ussher (2003) documented that women referred to PMS as behaviour in relation to others, never on an individual level. That further evidences PMS as psycho-bio-social construct heavily influenced by interaction with close people.

Connecting the dots between DSM classification of PMS as a psychological disorder and potential benefit of psychotherapeutical interventions it is wise to focus on psychological help. This review will summarise literature on use of CBT in PMS as it is widely accessible via NHS and its effectiveness is well documented.


In 1989 there were two studies published on this topic. Morse (1989) conducted a study in Australia to compare effectiveness of CBT with progesterone. Six women were recruited for a ten-week treatment, continuous assessment with MDQ questionnaire and keeping a diary with symptoms on daily basis. The follow up continued for one year after completing the treatment. Results were confusing. Psychological symptoms improved after 2nd cycle and persisted throughout the follow up. Physical symptoms on the other hand decreased significantly after the 1st period and got back to the previous state, although during the 1- year follow up the physical symptoms were overall improved compared to pre-study baseline. Authors tried to explain it by a possible placebo effect, which would count for improvement of physical symptoms and then bouncing back to baseline, nevertheless it can’t serve as an explanation for general improvement in the 1 – year follow up.

The good aspect of this study is the long follow up and assessment of physical and psychological symptoms separately. Due to the unusual results authors failed to identify possible other causes, such as lifetime events, which might have introduced bias in assessing PMS psychological symptoms. The scientific credit is unfortunately decreased also due to small amount of participants with no preliminary [assessment of their PMS, if it really corresponded to the diagnostic standards in 1989. Authors under the pressure of critiques improved methodology to randomised trial of 42 women and published it in 1991.

The second pioneer study is of Slade (1989) who investigated four British women with pre-established premenstrual syndrome. The criteria for PMS were 30% increase in symptom severity in day 14 – 28 of the cycle. All of them received CBT sessions lasting 30 minutes every week, focused on problem solving, anger control, autogenic training, training in recognising thought attribution patterns. Participants were then followed up for 6 months, meaning 6 consecutive periods. The assessment of improvement of symptoms was done by Moos Menstrual Distress Questionnaire (MMDQ), timed for day 21-28 of the cycle.

Improvement of symptoms was noted however it was not consistent. In all four women the decrease in severity of symptoms was apparent after 2nd cycle post CBT intervention. Nevertheless one woman experienced worsened symptoms between 4th and 5th cycle. It is not the only limitation of the study. Authors originally recruited 17 women, which lead to 25% participant retention rate. Apart of small number and inconsistent results the study loses credit on absence of control group. The good aspects of this study encompass pre assessment of PMS prior to the study and sole focus on CBT (unlike in the study of Morse, 1989). Generally it can be taken as a good pioneer attempt to quantify effect of CBT in PMS with a limited scientific validity.

Both those studies failed to provide control group and were done on a very limited number of participants. As demonstrated they both served as a background for further research with improved design (such as Morse, 1999).

In 1994 Kirby recruited 48 Australian women to assess effectiveness of CBT – based program on PMS. The main component were coping skills strategy alongside with assertiveness and stress reduction techniques. Second group underwent Awareness Through Movement classes to increase their kinaesthetic awareness. This was not known to have any benefit for PMS so it was used as the first control group. Second control group did not receive any treatment. Assessment was done by MMDQ (as in the study of Slade, 1989) and other measures for depression and anxiety. This study did not focus explicitly on physical symptoms. Symptom improvement was described in the treatment group compared to controls after the termination of experiment and in the 9-month follow up.

Unfortunately the diagnostic criteria were set as self – reported severe PMS, which is very inaccurate. Also randomisation did not take place. The CBT – based program for PMS was truly only CBT based as it encompassed wide range of other measurements, such as dietary changes or exercise. It is not entirely clear what made the positive change then as it was shown in the report of Nevatte et al. (2013) that dietary changes alone can improve PMS as well positive effect of exercise on depression is well established (Dimeo, Bauer, Varahram, Proest & Halter, 2001).

With the time new diagnostic and treatment criteria for PMS were employed. Pharmacology treatment developed from hormonal substitution towards psychiatric medication, such as antidepressants, which are though now more effective (Nevatte et al. 2013). In one of the most recent studies of Hunter et al (2002) the effectiveness of CBT was challenged by use of fluoxetine in control group. As in some studies (Faramarzi et al, 2013) fluoxetine was found similarly efficient as CBT, third group was included – CBT and fluoxetine together. Women were assessed prior to the experiment if they fulfil criteria for PMS according to APA standards (1994) with 2-month diary of symptoms (COPE, Calendar of Premenstrual Experiences). CBT was delivered in weekly sessions for the first 3 months with some booster sessions in the following 3 months. Final assessment was done by COPE scores and reassessment of diagnostic criteria for PMS took place.

In the CBT group women meeting PMS diagnostic criteria dropped from 100% to 31.5%, in the fluoxetine group to 22% and in the CBT/fluoxetine group 17% respectively. However after 1 year follow up in the CBT group participants dropped further to 17%, in the fluoxetine group participants increased to 59% and CBT/fluoxetine participants increased to 43%.

These findings are similar to Morsen et al. (1989) that medication does not add any therapeutical value to CBT and might contribute to non – compliance. The side effects and drop – out rate were significantly higher in the fluoxetine or CBT/fluoxetine group, mostly due to “did not like the treatment / side effect”. Also the effect of medication lasts only as long as the treatment continues and once it is stopped symptoms seem to reappear. CBT on the other hand offers long lasting relieve from symptoms, which is more beneficial from the patients’ perspective and also in terms of cost – effectiveness.




Overall those studies prove certain benefit of CBT on PMS. The results were not completely consistent within the studies. The predominant focus of the research is psychological symptoms, which can introduce a significant bias. As shown in the paper of Coughlin (1990) psychological disturbances can be also explained by simple aggravation of pre-existing problems by PMS, not a direct causality of PMS. In the few studies where authors focused on physical symptoms as well CBT proved to be effective.

At the methodology level all studies used various questionnaires to assess the severity of symptoms with a limited qualitative research in this area. Authors mostly used more than one questionnaire to decrease bias. The CBT sessions described in those papers vary in length, quantity and also content. It is understandable that anger management or relaxation techniques were included almost always, however positive self - talk for PMS (Morse, 1989) or kinaesthetic awareness (Kirkby, 1994) were employed only in one study. The lack of randomisation and small number of participants constituted an issue in the early research was corrected for in the latest papers.

In summary CBT as a well established psychotherapeutical intervention can be employed in dealing with PMS and based on this review it has proven benefits in that area. Further research should however be conducted to address further the content of CBT sessions and optimal length of therapy. Although CBT is funded by NHS this information would be beneficial and would improve cost – effectiveness of the therapy.




Chrisler JC, Levy KB (1990). The Media Construct a Menstrual Monster: A content Analysis of PMS Articles in the Popular Press. Women & Health 16 (2): 89-104.

Coughlin PC (1990). Premenstrual syndrome: how marital satisfaction and role choice affect symptom severity. Social Work 35: 351 – 355.

Dimeo F, Bauer M, Varahram I, Proest G, Halter U (2001). Benefits from aerobic exercise in patients with major depression: a pilot study. British Journal of Sports Medicine 35: 114-117.

Faramarzi M, Pasha H, Esmailzadeh S, Kheirkhah F, Heidary S, Afshar Z (2013). The Effect of The Cognitive Behavioral Therapy and Pharmacotherapy on Infertility Stress: A Randomized Controlled Trial. International Journal Fertility Sterility 7(3): 199–206.

Frank B, Dixon DN, Grosz HJ (1993). Conjoint monitoring of symptoms of premenstrual syndrome: impact on marital satisfaction. Journal of Counseling Psychology 40: 109 – 114.

Hamilton TA (1988). Marital adjustment and coping strategies among couples with premenstrual syndrome. Dissertation Abstracts International 49: 463.

Hunter MS, Ussher JM, Cariss M, Browne SJ, Jelley R, Katz M (2002) A randomized comparison of psychological (cognitive behavior therapy), medical (fluoxetine) and combined treatment for women with premenstrual dysphoric disorder. Journal Psychosomatic Obstetrics & Gynecology 23:193–199

Jones A, Theodos V, Canar WJ, Sher TG, Young M (2000). Couples and premenstrual syndrome: partners as moderators of symptoms? The Psychology of Couples and Illness: Theory, Research and Practice (pp 217 – 239). Washington DC: American Psychological Association.

Kirkby RJ (1994) Changes in premenstrual symptoms and irrational thinking following cognitive behavioral coping skills training. J Consult Clin Psychol 61:1026–1032

Morse CA, Bernard ME, Dennerstein L (1989) The effects of rational- emotive therapy and relaxation training on premenstrual syndrome: a preliminary study. J Rational-Emotive & Cog-Behav Ther 7:98–110

Nevatte T, O’Brien PMS, Backstrom C, Dennerstein L, Endicott J, Epperson CN, …Yonkers K (2013). ISPMD consensus on the management of premenstrual disorders. Archives of Women’s Mental Health 16(4): 279 – 291.

Rome E (1986). Premenstrual syndrome (PMS) examined through a feminist lens. Health Care for Women International7(1): 141 – 151

Sarris I, Bewley S, Agnihotri A (2009). Training in Obstetrics and Gynaecology. New York: Oxford University Press.

Slade P (1989) Psychological therapy for premenstrual emotional symptoms. Behav Psychother 17:135–150