The magnitude, long-term trend and lifestyle risk factors of dysmenorrhea and premenstrual syndrome

Ju, Hong (2016). The magnitude, long-term trend and lifestyle risk factors of dysmenorrhea and premenstrual syndrome PhD Thesis, School of Public Health, The University of Queensland. doi:10.14264/uql.2016.349

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
s4045831_phd_correctedthesis.pdf Thesis (open access) application/pdf 1.29MB 0
Author Ju, Hong
Thesis Title The magnitude, long-term trend and lifestyle risk factors of dysmenorrhea and premenstrual syndrome
School, Centre or Institute School of Public Health
Institution The University of Queensland
DOI 10.14264/uql.2016.349
Publication date 2016-06-20
Thesis type PhD Thesis
Supervisor Gita Mishra
Mark Jones
Total pages 228
Language eng
Subjects 1117 Public Health and Health Services
1114 Paediatrics and Reproductive Medicine
Formatted abstract
Background and aim: Dysmenorrhea (painful menstruation) and premenstrual syndrome (PMS) are the two most prevalent menstrual disorders affecting women of reproductive age. It is estimated that overall up to 85% of menstruating women suffer from one or more of the conditions. Moderate to severe forms of the conditions can significantly impact on the quality-of-life of women and cause considerable economic lost. Despite their potential public health significance, uncertainty remains in regard to the long-term trend and potential lifestyle risk factors for both conditions. This is mainly due to the cross-sectional design and retrospective data collection of the underlying studies. There is also a lack of studies which have followed the same group of women long enough to examine the variation of symptoms over time and to relate these with changes of potential time-varying risk factors. This thesis examines the prevalence, the long-term trend and the potential modifiable lifestyle risk factors of dysmenorrhea and PMS in Australian women, with the aim of contributing to the evidence base for preventative health policy-making.

This thesis used self-reported data from women born in 1973–78 who participated in the Australian Longitudinal Study on Women’s Health (ALSWH), a prospective cohort study of random samples from the national health insurance database (Medicare). The 1973–78 cohort includes 14,247 women aged 18–23 years at baseline (1996), who were surveyed every three years after Survey 2 (2000), with the last survey conducted in 2012. For the purpose of this research, data from Surveys 2–6 were used. Applying various statistical methods, including t-test, χ2 test, frequency over time, lasagne plot, latent class analysis, regression analysis including multinominal logistic regression, generalised estimating equations, and generalised linear mixed model, this research mainly examined 1) the prevalence, the variability, the trend and the trajectories of dysmenorrhea and PMS; and 2) the associations between a number of potential modifiable lifestyle risk factors and dysmenorrhea or PMS. In addition, exploratory analyses was performed to assess the impact of the conditions on self-reported quality-of-life. Sensitivity analyses were performed when necessary, including multiple imputations for missing data, to investigate the robustness and the reliability of the estimates.

Findings: Over the 13-year follow-up, dysmenorrhea and PMS affected 21% to 26% and 33% to 41% of Australian women born between 1973 and 1978 respectively. Substantial variation in symptom status was observed for both conditions at the individual level over time, with the majority of women reporting experiencing the conditions intermittently. Four population subgroups with unique symptom trajectories were identified for both conditions: normative (38.3%), low (28.0%), recovering (17.2%) and chronic (16.5%) groups for dysmenorrhea after adjusting for oral contraceptive pill use and pregnancy and/or giving birth in the past 12 months; and normative (22.1%), late onset (21.9%), recovering (26.5%) and chronic (29.5%) groups for PMS. Women in the chronic groups were those who had high probability of reporting the respective conditions consistently over time. Among women reporting dysmenorrhea or PMS, self-reported quality-of-life as measured by the 36-item Short Form Health Survey (SF-36) questionnaire was significantly worse than among women not reporting dysmenorrhea or PMS. Compared with women without dysmenorrhea, women with dysmenorrhea reported worse physical and mental health; whereas PMS predominantly impacted negatively on the mental health of affected women. After adjusting for potential confounders, moderate associations were demonstrated between the conditions and a number of health risks:

• Smoking was associated with higher risk of dysmenorrhea using latent class analysis combined with multinominal logistic regression. Compared with never smokers, baseline smokers were at 40% higher odds of being in the chronic trajectory group. The odds was even higher, 60%, among women with an earlier age of smoking initiation (<14 years).

• A dose-response association was observed between illicit drug use and PMS with illicit drug use in the last 12 months and multiple drug use showing a higher risk of reporting PMS. Compared with never drug use, multiple drug use in the last 12 months showed 30% higher odds of reporting PMS in the generalised estimating equations model. A 40% higher odds was revealed for drug users who first used illicit drugs before age 15 years.

• A U-shaped relationship was demonstrated between body mass index (BMI) and dysmenorrhea in the generalised estimating equations model. Compared with women with a normal BMI, underweight and obese women showed a higher risk of reporting dysmenorrhea (34% and 22% respectively). The risk of dysmenorrhea disappeared among obese women when they lost weight and acquired a healthier body weight.

Conclusion: Dysmenorrhea and PMS are common in Australian women with a stable prevalence at the population level but exhibit great variation among individuals over time. Four subgroups with unique trajectories were identified for each condition in this group of Australian women. Repeatedly over time, approximately one in every six women was likely to experience dysmenorrhea, and nearly one in three PMS. Moderate associations were demonstrated between smoking and dysmenorrhea and between illicit drug use and PMS, with a higher risk demonstrated for early age at first use of the substance. Both underweight and obesity also showed moderately increased risk for dysmenorrhea. The results may have some potential public health importance in supporting interventions to prevent substance use, especially early use, and to maintain a healthy body weight. However, this research was unable to establish causality and the nature of these associations needs further clarification in future studies.
Keyword Dysmenorrhea
Menstrual pain
Premenstrual Syndrome
Risk factors
Illicit drug
Body mass index

Document type: Thesis
Collections: UQ Theses (RHD) - Official
UQ Theses (RHD) - Open Access
Version Filter Type
Citation counts: Google Scholar Search Google Scholar
Created: Wed, 08 Jun 2016, 20:57:35 EST by Hong Ju on behalf of Learning and Research Services (UQ Library)