Premenstrual disorders : risk factors, comorbidities and health impact

Abstract

Premenstrual disorders (PMDs) are characterized by affective and physical symptoms that cyclically occur 7-10 days before the menstruation and affect millions of women of reproductive age worldwide. Despite its commonness, PMDs have received little research attention. Taking advantage of national cohorts and registers, combined with felicitous study designs, we studied the following aspects of PMDs:

Paper I Adverse childhood experiences (ACEs) as early life risk factors: With a subsample of 11,973 menstruating women within the Icelandic Stress‐And‐Gene‐Analysis (SAGA) cohort, we conducted a cross‐sectional analysis and calculated prevalence ratios (PRs) of PMDs in relation to varying numbers and types of ACEs using Poisson regression. The number of ACEs was associated with PMDs in a dose-response manner (PR 1.12 per ACE, 1.11–1.13); and such association was also evident among women without psychiatric comorbidities such as post-traumatic stress disorder (PTSD), anxiety, and depression. All types of ACEs (abuse, neglect, family dysfunction, and violence) were, to a different extent, associated with a higher prevalence of PMDs (PRs ranged from 1.11 to 1.51).

Paper II Association with perinatal depression (PND): Among 1,803,309 singleton pregnancies of 1,041,419 women recorded in the Swedish Medical Birth Register (MBR) from 2001 to 2018, we identified 84,949 incident cases of PND. For each case, ten controls were randomly selected, and individually matched on maternal age and calendar year of pregnancy. We first conducted a nested case-control study and calculated the odds ratio (OR) of PMDs diagnosed before pregnancy using conditional logistic regression, which can be interchangeably interpreted as hazard ratio (HR) of PND; and then conducted a matched cohort study and calculated the HR of subsequent PMDs using stratified Cox regression. We found a bidirectional association between PMDs and PND (HR 3.88, 3.67-4.11 in the nested case-control study and 1.56, 1.49-1.62 in the matched cohort study). Such association was observed for both prenatal and postnatal depression and was evident among women without psychiatric comorbidities. The association was attenuated yet remained statistically robust when comparing PND cases with their PND-free full sisters.

Paper III Subsequent risk of suicidal behavior and accidents: With 1,472,379 Swedish women of reproductive age, we conducted a register-based cohort study with follow-up from 2001 to 2012. HRs of suicidal behavior and accidents were estimated using Cox regression. Women with PMDs were at increased risk of suicidal behavior (HR 2.26, 95% CI 1.97–2.59) and accidents (HR 1.32, 95% CI 1.27–1.38) compared to the general female population. Such associations were attenuated yet remained significant when comparing women with PMDs to their full sisters. Moreover, the associations were evident among those without psychiatric comorbidities.

Paper IV PMD treatment with hormonal contraceptives or antidepressants and risk of suicidal behavior and accidents: A cohort of 23 029 women with PMDs as identified through the registers was followed from 2005 to 2012. Using Poisson regression, we calculated the Incidence Rate Ratios (IRRs) of suicidal behavior/accidents by use of contraceptives or antidepressants in between-individual and within-individual analyses, i.e., comparing the risk of these adverse outcomes between periods of on and off medication use among the women with PMD. Compared to no use of contraceptives, use of contraceptives was associated with a lower risk of suicidal behavior in both between-individual (IRR 0.76, 0.43-1.34) and within-individual analyses (IRR 0.65, 0.51-0.83), particularly for combined products (IRR 0.18, 0.07-0.47 and 0.19, 0.08-0.42 in between- and within-individual analyses specifically). Such risks were observed regardless of psychiatric comorbidities. However, the association for accidents was not consistent in between- and within-individual analyses. Use of antidepressants on the other hand was associated with a higher risk of suicidal behavior/accidents in both between- and withinindividual analyses which may be due to confounding by indication.

Conclusion: Our findings suggest that childhood adversities are important risk factors for PMDs and indicate a bidirectional association between PMDs and PND, including both prenatal and postnatal depression. Our findings also show that women with PMDs are at increased risk of suicidal behavior and accidents and that the use of hormonal contraceptives is associated with a lower risk of suicidal behavior among affected women. Studies in this thesis are among the first of their kind and the findings have important implications in the field of PMDs in terms of future research, public prevention, surveillance, and clinical management of PMDs.