A population health approach to menstrual health research
Introduction
Background
Most women spend nearly half of their lives menstruating (https://sci-hub.tw/10.1016/j.ogc.2019.04.004) and their experience of menstrual and pre-menstrual symptoms can have wide-ranging consequences for their physical and mental health and quality of life.
Reported prevalence rates for menstrual dysfunction and pre-menstrual symptoms show substantial variation between studies and populations, but menstrual pain, heavy menstrual bleeding and premenstrual syndrome (PMS) are thought to affect a high proportion of menstruating women, and are estimated to have a large impact on daily living, quality of life and work productivity.
For example, in an online survey of over 19,000 menstruating women in Japan, 74% reported suffering from “bothersome” menstrual symptoms, 50% reported pain and 19% reported heavy bleeding. The severity of symptoms was related to more outpatient visits, lower work productivity and greater inhibition and limitation of with daily life (PMID: 24015668).
In the United States, painful periods are estimated to occur in 50 to 90% of women and are the most common reason adolescents are repeatedy absent from school (https://www.acog.org/clinical/clinical-guidance/committee-opinion/articles/2018/12/dysmenorrhea-and-endometriosis-in-the-adolescent).
More examples of previous evidence are provided on the next tab
In clinical practice, evaluation of the menstrual cycle is recommended to improve the early identification of potential gynecological and other health problems. Menstrual health is considered an essential component of well-woman care for adult women. However, because of stigma and a lack of open discussion and awareness about menstrual health, women often suffer in silence, feeling uncomfortable discussing their symptoms with a clinician or assuming their symptoms are an inevitable part of menstruation (https://sci-hub.tw/10.1016/j.ogc.2019.04.004).
Despite the high prevalence and potential impact of problematic menstrual symptoms, and the fact that evaluation of the menstrual cycle is considered vital in improving the early identification of potential health problems in clinical practice, remarkably little data has been collected on menstruation in the major cohort studies commonly used for epidemiological research.
In this document, I outline several opportunities to research menstrual health using epidemiological/population-based approaches.
More previous evidence
Japanese study showing association between premenstrual symptoms and negative subjective perceptions of health and stress amond college students: https://bpsmedicine.biomedcentral.com/articles/10.1186/s13030-019-0167-y
Korean study showed premenstrual symptoms (from MDQ) symptoms cluster into “turmoil”, “negative affect”, “general discomfort”: https://www.tandfonline.com/doi/full/10.3109/0167482X.2016.1157159?scroll=top&needAccess=true&instName=University+of+Bristol
Cross-sectional survey of 112 nulliparous elite athletes vs 103 women not practising regular sport found that athletes had a higher incidence of irregular periods and heavy menstrual bleeding and a lower incidence of dysmenorrhea (pain) than controls. Within athletes, HMB was associated with lower mental QoL and higher perceived stress. https://www.ncbi.nlm.nih.gov/pubmed/32046446
Small online survey study of 145 women with primary dysmenorrhea found menstrual pain that is less severe and shorter in duration was associated with better physical QoL, being older at age of onset of painful periods was associated with better mental QoL https://www.ncbi.nlm.nih.gov/pubmed/31931396
Community-based cross-sectional descriptive study on random sample of the Swedish general population (1547 women 40-45 years old) - found that 32% of women experienced heavy menstrual bleeding. Menstrual bleeding in general was associated with negative perceptions and limited social and professional activities, but this was particularly affected in women experiencing heavy menstrual bleeding and these women had worse health-related QoL in all domains. https://www.ncbi.nlm.nih.gov/pubmed/24266506/
Questionnaire survey in 2356 women aged 18–50 years living in Beijing - 18.2% prevalence of HMB. Higher odds in women who drink regularly compared to never drinkers. Women with HMB experienced more practical discomforts and limited life activities. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6360654/
305 Spanish female university students - 76% of the sample suffered from dysmenorrhea. This was associated with not exercising regularly and with the total score for perceived QoL https://www.ncbi.nlm.nih.gov/pubmed/30818861
Nationwide (USA), cross-sectional, internet-based survey among 42,879 women aged 15-45 years found dysmenorrhea was the most common symptom (85%), followed by psychological complaints (77%) and tiredness (71%). During their period, 38% of women reported not being able to perform regular daily activities. From the women that had to skip tasks, only 49% told their families that menstrual symptoms were the reason. https://www.ncbi.nlm.nih.gov/pubmed/30885768
Lower QoL (physical, psychological, social, environmental domains) reported by women with dysmenorrhea in a community-based cross-sectional study of 119 women in an urban field practice area in Puducherry, India. https://www.ncbi.nlm.nih.gov/pubmed/30911494
Prevalence of HMB of 38% in 306 women presenting to an outpatient department of a hospital in Turkey. Ferritin levels and physical functions decreased significantly as the duration of menstruation increased. A “positive but very weak” association was found between menstruation duration and subdimentions of the Brief Fatigue Inventory and the general health perception subscale of a QoL scale. https://www.ncbi.nlm.nih.gov/pubmed/31086516
Self-reported HMB in 44 African American women was associated with anaemia (35%), lower hemoglobin, hematocrit and ferritin. (evidence that self-reported HMB is a reasonably reliable measure of substantial blood loss) https://www.ncbi.nlm.nih.gov/pubmed/27524363
Study of 1575 Australian women - 22.5% reported HMB and 5.3% reported very HMB. Women who experienced severe menstrual pain were more likely to report periods as heavy or very heavy. There was an association between HMB and painful periods and being confined to a bed during their period. https://www.ncbi.nlm.nih.gov/pubmed/27623183
Aims and potential impact
The aims of this research are to:
Raise the profile of menstrual health in the public health agenda by describing the prevalence of menstrual symptoms and their impact in different populations
Estimate the prevalence of problematic menstrual symptoms (like heavy, irregular, painful, prolonged, frequent or infrequent periods and premenstrual symptoms like low mood, breast tenderness, fatigue and GI symptoms) in the UK and other populations.
Characterise differences in prevalence and impact in LMICs vs HICs and in association with health and social inequities (e.g. period poverty) within individual countries.
Identify risk factors for problematic menstrual symptoms that could inform ways to indicate who is at risk of experiencing them (and when) and help develop strategies to prevent or improve them
Identify demographic, molecular and health/lifestyle factors associated with experience of menstrual symptoms.
Characterise within- and between- cycle variation in menstrual symptoms and explore risk factors associated with this variation.
Assess the ability of risk factors in predicting menstrual symptoms, in general (on average) and within and between cycles.
Infer causal relationships between risk factors and menstrual symptoms, in general (on average) and within and between cycles.
The next step would be to develop and trial interventions to improve and/or manage menstrual symptoms by enabling better prediction (and then taking steps to self-manage) or modifying causal risk factors (e.g. adjusting health behaviours)
Identify asssociations between experience of menstrual symptoms and outcomes related to health and quality of life, which will highlight the impact and importance of these experiences and could inform ways to predict who is at risk of poor outcomes and help develop preventative strategies
Identify associations between experience of menstrual symptoms and outcomes related to health and quality of life, such as wellbeing/mental health, participation in sports/exercise, time spent off work and life satisfaction.
Assess the ability of experience of menstrual symptoms to predict outcomes related to health and quality of life.
Infer causal relationships, and the direction of causality, between experience of menstrual symptoms and outcomes related to health and quality of life.
The next step would be to develop and trial interventions to improve health and quality of life by either improving menstrual symptoms or minimising their impact (e.g. through enabling women to better predict and then self-manage their symptoms, or implementing workplace policies around working from home or making sanitary products freely available, etc)
Projects using existing data
There are several opportunities using data that already exists:
Systematic reviews
Possible projects
Systematic review to identify reported prevalences of problematic menstrual symptoms in population-based/cohort studies.
Systematic review of the evidence on risk factors for problematic menstrual symptoms.
Systematic review of the evidence on impact of problematic menstrual symptoms on measures related to quality of life.
Systematic review of the evidence around associations between problematic menstrual symptoms and cardiovascular health.
Analysts
Systematic reviews could be conducted by Reproduction and Development MSc students as ‘Part 1’ of their three part research project unit. These projects might also be relevant to students on the MScs in Epi or Public Health.
Projects in ALSPAC
The Avon Longitudinal Study of Parents and Children (ALSPAC) is one of the only cohorts to collect any data on menstrual symptoms.
These data were collected for mothers (ALSPAC G0) at multiple timepoints: questionnaires D (d012), H (h028, h110:h115g), J (j027, j141:j144), K (k1031, k1035, k1290:k1308), L (l3031, l3350:l3401), M (m4201:m4202), N (n1121:n1133), P (p1031, p1260:p1284), Q (q4020:q4030), R (r2080:r2093), S (s1031, s1260:s1284, s4020), T (t4700-4961), U (u1000:u1060), V (v4551:v4955))
And in their daughters (ALSPAC G1) at age 22 (most detailed time point; questionnaire YPA) and at yearly intervals throughout puberty (larger sample size but less detailed).
These data, in combination with detailed biological, health, lifestyle and demographic data collected at multiple time-points for the past three decades, allow us to study factors that might predict or causally affect menstrual symptoms. Data on health and quality of life were also collected at multiple time points. These data enable us to study associations that, if causal, would tell us about the impact of different menstrual experiences on women’s health and quality of life, which could inform strategies to predict which women are at risk of poor outcomes and help develop preventative interventions.
Preliminary work in ALSPAC G1
To illustrate that this work in ALSPAC is worth doing, I have started to characterise the prevalence of menstrual symptoms in ALSPAC G1 (at 22 years old). As shown below, a high proportion of these women report experiencing problematic menstrual symptoms.
| not at all | mildly | moderately | very | |
|---|---|---|---|---|
| heavy | 6.8 | 27.7 | 50.5 | 14.9 |
| irregular | 48.8 | 21.2 | 14.4 | 15.7 |
| painful | 14.2 | 37.4 | 32.7 | 15.6 |
| 0 | 1 | |
|---|---|---|
| frequent | 93.4 | 6.6 |
| infrequent | 88.2 | 11.8 |
| prolonged | 76.0 | 24.0 |
| 0 | 1 | |
|---|---|---|
| irritable | 58.5 | 41.5 |
| depressed | 75.9 | 24.1 |
| anxious | 86.2 | 13.8 |
| very tired | 76.4 | 23.6 |
| 0 | 1 | |
|---|---|---|
| irritable | 68.0 | 32.0 |
| depressed | 81.9 | 18.1 |
| anxious | 90.1 | 9.9 |
| very tired | 75.4 | 24.6 |
Most symptoms were positively correlated. Hierarchical clustering showed that issues to do with heaviness, frequency, regularity and pain tended to cluster separately from emotions/fatigue experienced during or before periods. Within the heaviness etc cluster, irregular periods tended to cluster with infrequent/frequent periods and prolonged bleeding tended to cluster with heaviness and pain. Heaviness and pain were reasonably highly correlated (r=0.5). Within the emotions/fatigue cluster, feeling anxious and depressed tended to cluster separately from feeling irritable or fatigued. For each of these issues, experiencing the issue before and during the period was very highly correlated (r>0.94 in all cases).
In logistic regression analyses, symptoms were more often experienced by women with a lower socioeconomic position at birth (measured by mother’s occupation and education), who were not on hormone contraception, who smoked in the past year, who had ever been pregnant, who had a higher BMI (although this association was seen less often than others) and who had either endometriosis or PCOS.
| exposure | outcome | sample size | odds ratio | lower 95% CI | upper 95% CI | P value | FDR-corrected P | |
|---|---|---|---|---|---|---|---|---|
| irregular_very_22…43 | pcos | irregular_very_22 | 1696 | 7.03 | 4.20 | 11.77 | 1.1e-13 | 8.9e-12 |
| infrequent_22…45 | pcos | infrequent_22 | 1362 | 6.04 | 2.98 | 12.22 | 5.9e-07 | 1.5e-05 |
| painful_very_22…62 | sep_mums_occup | painful_very_22 | 1709 | 0.71 | 0.62 | 0.81 | 6.0e-07 | 1.5e-05 |
| irritable…9 | contraception_22 | irritable | 1930 | 0.60 | 0.49 | 0.73 | 7.3e-07 | 1.5e-05 |
| depressed…8 | contraception_22 | depressed | 1911 | 0.61 | 0.50 | 0.76 | 9.2e-06 | 1.5e-04 |
| anxious…27 | smoking | anxious | 2057 | 1.62 | 1.27 | 2.07 | 1.1e-04 | 1.4e-03 |
| heavy_very_22…71 | sep_mums_edu | heavy_very_22 | 1937 | 0.77 | 0.67 | 0.88 | 1.9e-04 | 2.2e-03 |
| irregular_very_22…23 | smoking | irregular_very_22 | 2078 | 1.61 | 1.25 | 2.07 | 2.6e-04 | 2.6e-03 |
| anxious…7 | contraception_22 | anxious | 1906 | 0.63 | 0.49 | 0.81 | 3.3e-04 | 2.9e-03 |
| irregular_very_22…13 | ever_pregnant_22 | irregular_very_22 | 2015 | 1.62 | 1.19 | 2.21 | 2.2e-03 | 1.8e-02 |
| painful_very_22…2 | contraception_22 | painful_very_22 | 1928 | 0.67 | 0.51 | 0.87 | 2.4e-03 | 1.8e-02 |
| fatigued…10 | contraception_22 | fatigued | 1912 | 0.72 | 0.58 | 0.89 | 2.8e-03 | 1.9e-02 |
| frequent_22…16 | ever_pregnant_22 | frequent_22 | 1497 | 2.08 | 1.26 | 3.44 | 4.1e-03 | 2.5e-02 |
| depressed…28 | smoking | depressed | 2065 | 1.36 | 1.10 | 1.68 | 4.4e-03 | 2.5e-02 |
| heavy_very_22…61 | sep_mums_occup | heavy_very_22 | 1707 | 0.82 | 0.71 | 0.95 | 6.2e-03 | 3.3e-02 |
| prolonged_22…14 | ever_pregnant_22 | prolonged_22 | 1884 | 1.48 | 1.11 | 1.97 | 7.7e-03 | 3.8e-02 |
| irregular_very_22…3 | contraception_22 | irregular_very_22 | 1919 | 1.45 | 1.08 | 1.95 | 1.3e-02 | 6.1e-02 |
| heavy_very_22…1 | contraception_22 | heavy_very_22 | 1927 | 0.71 | 0.54 | 0.93 | 1.4e-02 | 6.1e-02 |
| fatigued…70 | sep_mums_occup | fatigued | 1692 | 0.87 | 0.78 | 0.97 | 1.5e-02 | 6.3e-02 |
| painful_very_22…22 | smoking | painful_very_22 | 2087 | 1.36 | 1.05 | 1.76 | 2.1e-02 | 8.4e-02 |
| irregular_very_22…73 | sep_mums_edu | irregular_very_22 | 1930 | 0.86 | 0.75 | 0.98 | 2.5e-02 | 9.4e-02 |
| heavy_very_22…51 | bmi_17_z | heavy_very_22 | 1226 | 1.18 | 1.02 | 1.36 | 2.7e-02 | 9.7e-02 |
| prolonged_22…4 | contraception_22 | prolonged_22 | 1783 | 0.77 | 0.61 | 0.98 | 3.4e-02 | 1.2e-01 |
| infrequent_22…5 | contraception_22 | infrequent_22 | 1505 | 0.71 | 0.51 | 1.00 | 4.7e-02 | 1.6e-01 |
| painful_very_22…72 | sep_mums_edu | painful_very_22 | 1937 | 0.88 | 0.77 | 1.00 | 5.1e-02 | 1.6e-01 |
| painful_very_22…32 | endometriosis | painful_very_22 | 1716 | 2.93 | 0.99 | 8.64 | 5.2e-02 | 1.6e-01 |
| heavy_very_22…11 | ever_pregnant_22 | heavy_very_22 | 2024 | 1.32 | 0.95 | 1.83 | 9.5e-02 | 2.7e-01 |
| fatigued…30 | smoking | fatigued | 2066 | 1.19 | 0.97 | 1.47 | 9.6e-02 | 2.7e-01 |
| frequent_22…26 | smoking | frequent_22 | 1544 | 1.44 | 0.93 | 2.22 | 1.0e-01 | 2.8e-01 |
| prolonged_22…64 | sep_mums_occup | prolonged_22 | 1590 | 0.90 | 0.79 | 1.02 | 1.1e-01 | 2.8e-01 |
| irregular_very_22…53 | bmi_17_z | irregular_very_22 | 1222 | 1.13 | 0.97 | 1.30 | 1.1e-01 | 2.8e-01 |
| irritable…69 | sep_mums_occup | irritable | 1707 | 0.92 | 0.83 | 1.02 | 1.2e-01 | 3.0e-01 |
| fatigued…40 | endometriosis | fatigued | 1697 | 2.25 | 0.78 | 6.44 | 1.3e-01 | 3.2e-01 |
| depressed…68 | sep_mums_occup | depressed | 1686 | 0.92 | 0.82 | 1.03 | 1.4e-01 | 3.2e-01 |
| prolonged_22…44 | pcos | prolonged_22 | 1591 | 1.50 | 0.86 | 2.64 | 1.6e-01 | 3.5e-01 |
| frequent_22…46 | pcos | frequent_22 | 1287 | 2.41 | 0.70 | 8.30 | 1.6e-01 | 3.6e-01 |
| fatigued…80 | sep_mums_edu | fatigued | 1921 | 0.93 | 0.84 | 1.03 | 1.8e-01 | 3.8e-01 |
| irregular_very_22…33 | endometriosis | irregular_very_22 | 1707 | 2.17 | 0.68 | 6.86 | 1.9e-01 | 4.0e-01 |
| prolonged_22…34 | endometriosis | prolonged_22 | 1599 | 2.00 | 0.66 | 6.00 | 2.2e-01 | 4.5e-01 |
| heavy_very_22…21 | smoking | heavy_very_22 | 2087 | 1.18 | 0.90 | 1.54 | 2.4e-01 | 4.7e-01 |
| depressed…78 | sep_mums_edu | depressed | 1917 | 1.06 | 0.95 | 1.18 | 2.7e-01 | 5.2e-01 |
| frequent_22…6 | contraception_22 | frequent_22 | 1428 | 1.31 | 0.80 | 2.14 | 2.8e-01 | 5.2e-01 |
| infrequent_22…55 | bmi_17_z | infrequent_22 | 991 | 1.11 | 0.92 | 1.32 | 2.8e-01 | 5.2e-01 |
| prolonged_22…54 | bmi_17_z | prolonged_22 | 1151 | 1.07 | 0.94 | 1.23 | 2.9e-01 | 5.2e-01 |
| irritable…29 | smoking | irritable | 2086 | 1.11 | 0.91 | 1.35 | 3.0e-01 | 5.4e-01 |
| infrequent_22…25 | smoking | infrequent_22 | 1633 | 1.18 | 0.84 | 1.67 | 3.3e-01 | 5.7e-01 |
| infrequent_22…65 | sep_mums_occup | infrequent_22 | 1334 | 1.10 | 0.90 | 1.34 | 3.4e-01 | 5.8e-01 |
| depressed…48 | pcos | depressed | 1685 | 1.28 | 0.76 | 2.17 | 3.6e-01 | 6.0e-01 |
| fatigued…60 | bmi_17_z | fatigued | 1211 | 1.05 | 0.93 | 1.18 | 4.5e-01 | 7.2e-01 |
| frequent_22…76 | sep_mums_edu | frequent_22 | 1435 | 0.92 | 0.73 | 1.15 | 4.5e-01 | 7.2e-01 |
| anxious…17 | ever_pregnant_22 | anxious | 1999 | 1.13 | 0.82 | 1.55 | 4.7e-01 | 7.2e-01 |
| infrequent_22…35 | endometriosis | infrequent_22 | 1364 | 1.77 | 0.38 | 8.26 | 4.7e-01 | 7.2e-01 |
| frequent_22…56 | bmi_17_z | frequent_22 | 926 | 1.10 | 0.85 | 1.43 | 4.8e-01 | 7.2e-01 |
| infrequent_22…15 | ever_pregnant_22 | infrequent_22 | 1590 | 1.17 | 0.75 | 1.81 | 4.9e-01 | 7.2e-01 |
| irritable…79 | sep_mums_edu | irritable | 1940 | 1.03 | 0.94 | 1.14 | 4.9e-01 | 7.2e-01 |
| anxious…57 | bmi_17_z | anxious | 1203 | 0.95 | 0.81 | 1.11 | 5.2e-01 | 7.4e-01 |
| prolonged_22…74 | sep_mums_edu | prolonged_22 | 1807 | 0.97 | 0.86 | 1.09 | 5.6e-01 | 7.6e-01 |
| anxious…67 | sep_mums_occup | anxious | 1681 | 0.96 | 0.84 | 1.10 | 5.7e-01 | 7.6e-01 |
| fatigued…50 | pcos | fatigued | 1686 | 1.16 | 0.68 | 1.96 | 5.8e-01 | 7.6e-01 |
| irregular_very_22…63 | sep_mums_occup | irregular_very_22 | 1702 | 0.96 | 0.83 | 1.11 | 5.8e-01 | 7.6e-01 |
| depressed…38 | endometriosis | depressed | 1696 | 1.36 | 0.45 | 4.08 | 5.8e-01 | 7.6e-01 |
| depressed…58 | bmi_17_z | depressed | 1209 | 0.97 | 0.85 | 1.10 | 5.9e-01 | 7.6e-01 |
| depressed…18 | ever_pregnant_22 | depressed | 2008 | 0.93 | 0.71 | 1.23 | 6.3e-01 | 7.8e-01 |
| anxious…37 | endometriosis | anxious | 1690 | 1.37 | 0.38 | 4.94 | 6.3e-01 | 7.8e-01 |
| infrequent_22…75 | sep_mums_edu | infrequent_22 | 1516 | 1.04 | 0.88 | 1.23 | 6.3e-01 | 7.8e-01 |
| frequent_22…66 | sep_mums_occup | frequent_22 | 1267 | 1.06 | 0.82 | 1.36 | 6.7e-01 | 8.1e-01 |
| anxious…77 | sep_mums_edu | anxious | 1911 | 1.03 | 0.90 | 1.17 | 6.9e-01 | 8.3e-01 |
| heavy_very_22…41 | pcos | heavy_very_22 | 1705 | 0.87 | 0.41 | 1.86 | 7.3e-01 | 8.5e-01 |
| painful_very_22…12 | ever_pregnant_22 | painful_very_22 | 2024 | 1.05 | 0.75 | 1.48 | 7.7e-01 | 8.7e-01 |
| fatigued…20 | ever_pregnant_22 | fatigued | 2008 | 1.04 | 0.80 | 1.35 | 7.9e-01 | 8.7e-01 |
| painful_very_22…52 | bmi_17_z | painful_very_22 | 1226 | 1.02 | 0.87 | 1.19 | 8.0e-01 | 8.7e-01 |
| irritable…19 | ever_pregnant_22 | irritable | 2029 | 0.97 | 0.76 | 1.24 | 8.0e-01 | 8.7e-01 |
| irritable…39 | endometriosis | irritable | 1710 | 0.87 | 0.30 | 2.53 | 8.0e-01 | 8.7e-01 |
| irritable…59 | bmi_17_z | irritable | 1222 | 1.01 | 0.91 | 1.13 | 8.0e-01 | 8.7e-01 |
| painful_very_22…42 | pcos | painful_very_22 | 1705 | 0.92 | 0.45 | 1.89 | 8.3e-01 | 8.8e-01 |
| anxious…47 | pcos | anxious | 1679 | 1.07 | 0.55 | 2.08 | 8.4e-01 | 8.8e-01 |
| prolonged_22…24 | smoking | prolonged_22 | 1940 | 1.02 | 0.80 | 1.29 | 9.0e-01 | 9.4e-01 |
| irritable…49 | pcos | irritable | 1699 | 1.02 | 0.62 | 1.68 | 9.4e-01 | 9.7e-01 |
| heavy_very_22…31 | endometriosis | heavy_very_22 | 1716 | 0.96 | 0.22 | 4.30 | 9.6e-01 | 9.7e-01 |
| frequent_22…36 | endometriosis | frequent_22 | 1289 | 0.00 | 0.00 | Inf | 9.8e-01 | 9.8e-01 |
In linear regression analyses, heaviness of periods, painful periods and irregular periods (ordinal variables treated as continuous) were associated with lower quality of life as measured by the amount of time an individual spends doing vigorous exercise, individual items from the SF-36 index of mental and physical health (e.g. how often an individual feels down in the dumps or how often their emotional or physical health has interfered with their participation in social activities recently), validated indices for wellbeing, life satisfaction and subjective happiness, and a question about how satisfied they are with their sex life. All models were adjusted for taking hormonal contraception, ever having been pregnant, smoking in the past year, maternal education at birth, and BMI at age 17.
| exposure | outcome | sample size | coefficient | lower 95% CI | upper 95% CI | P value | FDR-corrected P | |
|---|---|---|---|---|---|---|---|---|
| tired_22…22 | painful_ordinal_22 | tired_22 | 1016 | 0.14 | 0.08 | 0.20 | 7.2e-06 | 1.9e-04 |
| fulllife_22…24 | painful_ordinal_22 | fulllife_22 | 1017 | -0.14 | -0.20 | -0.08 | 1.1e-05 | 1.9e-04 |
| calmpeaceful_22…43 | irregular_ordinal_22 | calmpeaceful_22 | 1015 | -0.11 | -0.16 | -0.06 | 1.2e-05 | 1.9e-04 |
| tired_22…37 | irregular_ordinal_22 | tired_22 | 1013 | 0.11 | 0.06 | 0.16 | 2.4e-05 | 2.6e-04 |
| downhearted_22…27 | painful_ordinal_22 | downhearted_22 | 1015 | 0.15 | 0.08 | 0.22 | 2.9e-05 | 2.6e-04 |
| downdumps_22…41 | irregular_ordinal_22 | downdumps_22 | 1009 | 0.11 | 0.06 | 0.17 | 9.3e-05 | 7.0e-04 |
| wornout_22…23 | painful_ordinal_22 | wornout_22 | 1018 | 0.12 | 0.06 | 0.19 | 2.2e-04 | 1.4e-03 |
| interfere_social_phys_emo_22…21 | painful_ordinal_22 | interfere_social_phys_emo_22 | 1014 | 0.11 | 0.05 | 0.17 | 3.0e-04 | 1.6e-03 |
| wellbeing_z…16 | painful_ordinal_22 | wellbeing_z | 868 | -0.12 | -0.19 | -0.06 | 3.2e-04 | 1.6e-03 |
| downhearted_22…42 | irregular_ordinal_22 | downhearted_22 | 1012 | 0.11 | 0.05 | 0.16 | 3.6e-04 | 1.6e-03 |
| subjectivehappiness_z…17 | painful_ordinal_22 | subjectivehappiness_z | 867 | -0.12 | -0.19 | -0.06 | 4.2e-04 | 1.7e-03 |
| lotsenergy_22…29 | painful_ordinal_22 | lotsenergy_22 | 1016 | -0.10 | -0.16 | -0.05 | 4.5e-04 | 1.7e-03 |
| downdumps_22…26 | painful_ordinal_22 | downdumps_22 | 1012 | 0.12 | 0.05 | 0.18 | 5.5e-04 | 1.8e-03 |
| subjectivehappiness_z…32 | irregular_ordinal_22 | subjectivehappiness_z | 864 | -0.10 | -0.16 | -0.04 | 5.6e-04 | 1.8e-03 |
| wornout_22…38 | irregular_ordinal_22 | wornout_22 | 1015 | 0.09 | 0.04 | 0.15 | 9.1e-04 | 2.7e-03 |
| vigorous_activity_22…20 | painful_ordinal_22 | vigorous_activity_22 | 886 | -0.16 | -0.25 | -0.06 | 9.5e-04 | 2.7e-03 |
| tired_22…7 | heavy_ordinal_22 | tired_22 | 1015 | 0.12 | 0.05 | 0.19 | 1.0e-03 | 2.7e-03 |
| verynervous_22…40 | irregular_ordinal_22 | verynervous_22 | 1013 | 0.09 | 0.03 | 0.14 | 1.1e-03 | 2.7e-03 |
| fulllife_22…39 | irregular_ordinal_22 | fulllife_22 | 1014 | -0.08 | -0.13 | -0.03 | 1.2e-03 | 2.7e-03 |
| calmpeaceful_22…13 | heavy_ordinal_22 | calmpeaceful_22 | 1017 | -0.11 | -0.18 | -0.04 | 1.4e-03 | 3.1e-03 |
| verynervous_22…25 | painful_ordinal_22 | verynervous_22 | 1016 | 0.10 | 0.04 | 0.16 | 1.5e-03 | 3.3e-03 |
| calmpeaceful_22…28 | painful_ordinal_22 | calmpeaceful_22 | 1018 | -0.09 | -0.15 | -0.03 | 2.6e-03 | 5.3e-03 |
| wornout_22…8 | heavy_ordinal_22 | wornout_22 | 1017 | 0.11 | 0.04 | 0.19 | 3.0e-03 | 5.9e-03 |
| lotsenergy_22…44 | irregular_ordinal_22 | lotsenergy_22 | 1013 | -0.07 | -0.11 | -0.02 | 7.3e-03 | 1.4e-02 |
| happy_22…30 | painful_ordinal_22 | happy_22 | 1015 | -0.07 | -0.13 | -0.02 | 8.4e-03 | 1.5e-02 |
| interfere_social_phys_emo_22…36 | irregular_ordinal_22 | interfere_social_phys_emo_22 | 1011 | 0.07 | 0.02 | 0.12 | 9.8e-03 | 1.7e-02 |
| vigorous_activity_22…5 | heavy_ordinal_22 | vigorous_activity_22 | 885 | -0.14 | -0.24 | -0.03 | 1.2e-02 | 1.9e-02 |
| wellbeing_z…31 | irregular_ordinal_22 | wellbeing_z | 865 | -0.07 | -0.13 | -0.02 | 1.2e-02 | 1.9e-02 |
| happy_22…45 | irregular_ordinal_22 | happy_22 | 1012 | -0.04 | -0.09 | 0.00 | 5.0e-02 | 7.8e-02 |
| interfere_social_phys_emo_22…6 | heavy_ordinal_22 | interfere_social_phys_emo_22 | 1013 | 0.07 | 0.00 | 0.14 | 6.3e-02 | 9.4e-02 |
| life_satisfaction_z…18 | painful_ordinal_22 | life_satisfaction_z | 1024 | -0.06 | -0.12 | 0.00 | 6.6e-02 | 9.6e-02 |
| wellbeing_z…1 | heavy_ordinal_22 | wellbeing_z | 867 | -0.07 | -0.15 | 0.01 | 7.0e-02 | 9.8e-02 |
| sex_satisfaction…19 | painful_ordinal_22 | sex_satisfaction | 861 | -0.08 | -0.16 | 0.01 | 7.2e-02 | 9.9e-02 |
| subjectivehappiness_z…2 | heavy_ordinal_22 | subjectivehappiness_z | 866 | -0.07 | -0.15 | 0.01 | 9.8e-02 | 1.3e-01 |
| verynervous_22…10 | heavy_ordinal_22 | verynervous_22 | 1015 | 0.06 | -0.01 | 0.13 | 1.0e-01 | 1.3e-01 |
| fulllife_22…9 | heavy_ordinal_22 | fulllife_22 | 1016 | -0.06 | -0.12 | 0.01 | 1.2e-01 | 1.5e-01 |
| downhearted_22…12 | heavy_ordinal_22 | downhearted_22 | 1014 | 0.06 | -0.02 | 0.14 | 1.2e-01 | 1.5e-01 |
| life_satisfaction_z…3 | heavy_ordinal_22 | life_satisfaction_z | 1023 | -0.05 | -0.12 | 0.02 | 1.9e-01 | 2.2e-01 |
| lotsenergy_22…14 | heavy_ordinal_22 | lotsenergy_22 | 1015 | -0.04 | -0.11 | 0.03 | 2.3e-01 | 2.7e-01 |
| happy_22…15 | heavy_ordinal_22 | happy_22 | 1014 | -0.04 | -0.10 | 0.03 | 2.5e-01 | 2.8e-01 |
| downdumps_22…11 | heavy_ordinal_22 | downdumps_22 | 1012 | 0.04 | -0.03 | 0.12 | 2.9e-01 | 3.1e-01 |
| life_satisfaction_z…33 | irregular_ordinal_22 | life_satisfaction_z | 1021 | -0.03 | -0.08 | 0.02 | 2.9e-01 | 3.1e-01 |
| sex_satisfaction…4 | heavy_ordinal_22 | sex_satisfaction | 860 | -0.04 | -0.13 | 0.06 | 4.7e-01 | 4.9e-01 |
| vigorous_activity_22…35 | irregular_ordinal_22 | vigorous_activity_22 | 883 | -0.03 | -0.10 | 0.05 | 5.2e-01 | 5.3e-01 |
| sex_satisfaction…34 | irregular_ordinal_22 | sex_satisfaction | 858 | -0.02 | -0.08 | 0.05 | 6.7e-01 | 6.7e-01 |
Possible projects
There are multiple possible projects/tasks using ALSPAC data:
Characterise the prevalence of menstrual and premenstrual symptoms in ALSPAC women (G0 and G1) I have made a start in G1 as described above
Explore the intergenerational correlation/concordance of menstrual experiences in ALSPAC mothers (G0) and daughters (G1)
Explore longitudinal trajectories of menstrual symptoms throughout puberty (in ALSPAC G1)
Explore longitudinal trajectories of menstrual symptoms in adult women (in ALSPAC G0)
Identify predictive or causal demographic and health/lifestyle risk factors for experience of menstrual symptoms
Identify predictive or causal molecular risk factors for experience of menstrual symptoms (see below).
Identify associations between menstrual symptoms and mental health/quality of life related measures in ALSPAC and use longitudinal data (e.g. QoL/depressive symptoms before and after menarche, and repeated measures of menstrual symptoms) to try to establish the direction of identified associations between reporting of menstrual symptoms and quality of life/depressive symptoms.
Identify associations between menstrual symptoms and cardiovascular health related measures in ALSPAC and use longitudinal datato try to establish the direction of effect.
With regards to ‘molecular risk factors’, there are several options, some of which are listed below:
ALSPAC G0:
- GWAS data
- haemoglobin levels
- Red blood cell count
- White blood cell count
- Sex Hormone Binding Globulin
- FSH and LH
- AMH
- C-reactive protein
- Thyroid function tests
- Genome-wide blood DNA methylation data during pregnancy and 17 years later, measured using Illumina arrays
- Metabolomic data
ALSPAC G1:
- GWAS data
- haemoglobin levels (multiple timepoints throughout childhood and puberty and at age 24) - “Hb_TF3”,“Hb_TF4”,“Hb_TF2”,“hb_F9”,“hb_F7”,“hb_F11”,
- platelets at age 24 - “Platelets_F24”
- Red blood cell count at age 24 - “RBC_F24”
- White blood cell count at age 24 - “WBC_F24”
- Sex Hormone Binding Globulin at age 15 - “SHBG_TF3”,
- C-reactive protein at age 15 and 24 - “CRP_TF4”,“CRP_F24”,
- Thyroid function tests at age 15 - “FT3_TF3”,“FT4_TF3”,“TSH_TF3”
- Genome-wide blood DNA methylation data at birth, age 7 and age 15/17, and some (not sure how many as this is a new dataset) in early 20s too, measured using Illumina arrays
- Metabolomic data
Analysts
Projects could be conducted by:
Reproduction and Development MSc students as ‘Part 2’ of their three part research project unit (would have to be particularly able students with experience of R)
Epidemiology MSc students
Public Heath MSc students(?)
Genomic medicine iBSc students
MRes students
First year (mini project) PhD students
Summer students/electives/other short term programmes
Some of these projects could also be used to try out a “team science” approach to research, whereby IEU staff/students with particular complimentary skills come together to tackle different aspects of the project.
Projects in other cohorts
I have been unable to find many other cohorts with data on menstrual symptoms, but it appears that MoBa has asked some questions to mothers, so there is potential for a collaborative project on many of the above ALSPAC projects with MoBa.
The EGG consortium or maybe CHARGE might be a good place to identify other cohorts that could contribute to analyses - particularly GWAS.
Projects using new data
There is enough work to be done in ALSPAC to justify applying for funding. In addition, there are several other possibilities I can think of…
Potential ideas
Extra questions to ALSPAC
Extra questions to ALSPAC (G1) participants (perhaps based on the Menstrual Distress Questionnaire (Moos, 1968)) about other symptoms such as breast tenderness, GI symptoms, water retention, etc.
Extra questions to ALSPAC (G1) participants about quality of life specifically in relation to periods, e.g. how often in the past month have you not gone to work/a social event/participated in sport because of your menstrual symptoms?
Add questions about management of periods (tampons, pads, cup, hormone contraception, etc)?
Repeated symptom reporting within and between cycles
Aim is to get data on symptoms reported repeatedly within and over several cycles - to allow us to study the trajectory of symptoms within and between cycles
Daily questions on other factors (health behaviours, diet, stressful events) would allow us to study menstrual symptoms in relation to these factors within and between cycles (to explore whether these factors influence severity of symptoms).
Daily reporting could be facilitated through an app? Could be developed as part of this project, or we could explore linkage to data collected through an existing period tracker app.
Repeated biological sampling (blood spots/saliva) would enable us to explore DNA methylation changes within and between cycles - to explore mechanisms but also using epigenetics as a biomarker of other factors that might influence menstrual symptoms.
Could be nested within ALSPAC: existing ALSPAC data on exposures over the lifecourse would allow us to study the effect of these on menstrual symptoms, and future follow-up of ALSPAC participants will allow us to explore later-life outcomes related to menstrual symptoms.
Alternatively, we could focus on maximising sample sizes by recruiting online and using an app to collect data. This might better enable us to look at prevalence between and within populations/countries.
GWAS of menstrual symptoms
Would need more data - could the menstrual distress questionnaire be sent to UKBB participants? How many are still pre-menopausal?
Approach existing consortia to check availability of data (e.g. EGG, CHARGE?)
Potential funding
Wellbeing of Women (charity)
Wellcome Trust Collaborative Award
MRC project grant
Potential collaborators
Jackie Maybin and Hilary Critchley from the Unversity of Edinburgh https://www.ed.ac.uk/centre-reproductive-health/dr-jackie-maybin https://www.ed.ac.uk/centre-reproductive-health/professor-hilary-od-critchley Together, they have set up HOPE - Healthy Optimal Periods for Everyone - a website about menstrual health and related research for the public. It was developed as part of a multidisciplinary project in collaboration with the University of Edinburgh’s MRC Centre for Reproductive Health, the RSE Young Academy of Scotland and NHS Lothian.
I spoke to Jackie Maybin in May and she was very keen to put together a proposal around heavy menstrual bleeding and CVD
Menstruation Research Network is a UK-based interdisciplinary network aiming to bring together menstruation experts from inside and outside academia. This project aims to establish a UK-wide Menstruation Research Network, bringing together experts from sciences and humanities, NGOs, the arts, activists and campaigners, industry, and the NHS in order to unify knowledge about the many medical, political, economic, psychological and cultural issues related to menstruation. The project was launched as the Scottish government rolled out its ‘End Period Poverty’ scheme, the medical community acknowledges menstruation as a ‘fifth vital sign’, and activists are calling for more environmentally friendly, inexpensive and positive menstrual products and culture. Through free workshops open to all, Menstruation Research Network aims to empower professionals, activists and academics to gain an overview over the state of the field, knowledge and cultural representation, set research agendas together, and plan future interdisciplinary collaborative work. This project was supported by a Wellcome Trust Small Network Grant (March 2019 - February 2020). https://menstruationresearchnetwork.co.uk
Alexandra Alvergne and Gabriella Kountourides at the University of Oxford dept of anthropology have a project that uses data from the Clue (menstruation tracking) app: “What if a negative premenstrual experience is not a hormonal disorder? What if it’s a message from the body about the socio-ecological causes of ill-health? For example, if people keep saying you’ll be moody, then you expect to be moody does this make you moody? Why individuals and populations vary in how severe their PMS is unclear. The causes of premenstrual distress are poorly understood. Therapies considered to be effective include anti-depressant medication and cognitive behavioural therapy (CBT). However this approach focuses on treating the symptoms associated with PMS rather than its root causes. Our research aims to understand how the environment a person is in can affect their experience of PMS. While PMS has recently been described as an inflammatory disease, there’s little actual data on why levels of inflammation vary between people. We’ll be the first to test the hypothesis that variation in inflammatory markers during the premenstrual experience is partially explained by environmental factors. These are both internal (infection) and external (social expectation of the premenstrual experience). Our research is being carried out in partnership with departments within the medical sciences division at the University of Oxford, and the digital period tracker app, Clue. The findings will have implications for rethinking the discussions around PMS and pave the way for new therapeutic perspectives.” https://www.menstrualmessages.com
Sheelagh McGuiness is working with other colleagues at UoB to research period poverty, so might be interested in this work too.
Ariana De Florio Cardiff University - setting up a PMDD genetics cohort (3000 women to be recruited over 5 years)
Hannah Short GP working near Cambridge with a PMDD/PMS clinic