Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Jan 2;151(4):529–536. doi: 10.1111/acps.13774

Genetic Confounding of the Association Between Age at First Hormonal Contraception and Depression

Jessica Mundy 1,2, Alisha S M Hall 1,2, Esben Agerbo 3,4, Clara Albiñana 3, Jette Steinbach 3, Bjarni J Vilhjálmsson 3,5,6, Søren D Østergaard 1,2, Katherine L Musliner 1,2,✉
PMCID: PMC11884912  PMID: 39746799

ABSTRACT

Background

Previous research has shown that females who use hormonal contraception are at increased risk of developing depression, and that the risk is highest among adolescents. While this finding could reflect age‐specific effects of exogenous hormones on mental health, genetic liability for mental disorders could be confounding the association. Our goal was to test the plausibility of this hypothesis by determining whether polygenic liabilities for major depressive disorder (MDD), bipolar disorder (BD), schizophrenia (SCZ), and attention deficit hyperactivity disorder (ADHD) are associated with younger age at hormonal contraception initiation.

Methods

We conducted a cohort study using data from the Danish iPSYCH2015 sub‐cohort, a representative sample of people born in Denmark between May 1981 and December 2008. Polygenic scores (PGSs) for MDD, BD, SCZ, and ADHD were created using the most recent genome‐wide association study meta‐analyses from the Psychiatric Genomics Consortium. Associations between PGSs and hormonal contraception initiation in the following age categories: 10–14, 15–19, 20–24, and 25+ were examined via Cox regression. We examined any hormonal contraception, oral contraception, and non‐oral contraception.

Results

PGS‐MDD and PGS‐ADHD showed the strongest associations with hormonal contraception initiation at age 10–14 (PGS‐ADHD: HR = 1.21 [95% CI = 1.16–1.27], p = 6.16 x 10−18; PGS‐MDD: 1.21 [1.16–1.27], p = 1.22 x 10−17). The associations then steadily decreased as age at hormonal contraception initiation increased. Both PGS‐MDD and PGS‐ADHD were also associated with initiation at ages 15–19, but not at 20–24 or 25+. PGS‐BD and PGS‐SCZ were also associated, albeit not as strongly, with initiation at age 10–14 only (PGS‐BD: 1.07 [1.02–1.13], p = 6.87 × 10−3; PGS‐SCZ: 1.09 [1.04–1.14], p = 8.61 × 10−4).

Conclusions and Relevance

These results suggest that genetic confounding could explain some of the association between early hormonal contraception use and depression. Where possible, researchers studying this important topic should account for possible confounding by genetic liability for mental disorders.

Keywords: depression, genetic epidemiology, genetics

Summary.

Significant Outcomes

  • Individuals who initiated hormonal contraception in adolescence had on average a higher polygenic risk for psychiatric disorders, particularly depression and ADHD. This suggests that previous findings that taking hormonal contraception at younger ages increases risk of mental illness may be partly confounded by genetic liability.

Limitations

  • Hormonal contraception prescriptions and all diagnoses were based on register data which are not collected for research purposes. The indication for hormonal contraception is unknown as well as lifestyle and environmental factors that may be relevant for understanding the causal pathway between genetic risk for psychiatric illness and initiating hormonal contraception in adolescence.

  • Polygenic scores only capture a small proportion of the variance in psychiatric disorders and their power is limited by the sample size of the discovery dataset.

  • Analyses were restricted to individuals of European ancestries which means our findings may not generalize to individuals of more diverse genetic ancestries.

1. Background

Previous research has shown that hormonal contraception use is associated with increased risk of developing depression [1, 2, 3]. Studies have also shown that the risk is highest among adolescent users and steadily decreases with age [2, 3, 4, 5]. While it is possible that this trend reflects a causal effect of exogenous hormone exposure during puberty and subsequent depression, several non‐causal explanations have also been proposed. First, people who experience side effects after initiating hormonal contraceptives may be less likely to use them in the future, leaving only those less susceptible to depression in the older age groups (i.e., selective discontinuation bias). Therefore, caution is called for when interpreting results from cross‐sectional studies of hormonal contraception and depression, since studies show that side effects, of which mood problems are commonly cited, are key reasons for discontinuation [6, 7, 8, 9]. Second, there may be factors associated with both early use of hormonal contraceptives and later depression risk (e.g., confounding by indication/extraneous aspects of indication) [10]. For example, younger age at sexual debut is associated with both early use of hormonal contraceptives and depression [11, 12, 13, 14] and therefore potentially confounds this association. Confounding has also been proposed as an explanation for the observed large effect of non‐oral long‐lasting hormonal contraceptives (e.g., levonorgestrel intrauterine system [LNG‐IUS], implant, and depot‐medroxyprogesterone acetate [DMPA] injection) [2, 3, 4, 5], since women living with mental disorders and/or in institutions may be more likely to opt for contraception that is less user‐dependent than a daily oral pill [2, 4, 15].

Another potential confounder in the association between age at hormonal contraception use and depression is genetic liability. Genetic liability for depression is associated both with risk for developing depression [16], and with other factors likely related to early hormonal contraceptive use, including age at sexual debut [17]. In addition, genetic liability for other psychiatric disorders such as bipolar disorder (BD), schizophrenia (SCZ), and attention deficit hyperactivity disorder (ADHD) are also potential confounders, as they, too, are associated with depression and potentially with factors related to early initiation of hormonal contraceptives [17, 18]. Thus, it is possible that an individual's underlying genetic liability for these psychiatric disorders accounts for some of the association between age at hormonal contraceptive use and depression.

Depression is a polygenic disorder, meaning the effect of genetics is not attributable to a single gene, but rather to small additive effects from thousands of genetic variants spread across the genome [19]. These small effects can be summarized in a polygenic score (PGS) which represents a weighted sum of individual genetic variants associated with an outcome [20]. While PGS does not capture all genetic effects, they are useful tools in research for quantifying genetic liability for polygenic disorders such as MDD [20].

Our goal was to evaluate the plausibility of the hypothesis that genetic confounding could play a role in the association between age at hormonal contraception initiation and subsequent depression (Figure 1). To accomplish this, we tested whether PGSs for four psychiatric and neurodevelopmental disorders (SCZ, BD, MDD, and ADHD) were associated with age at first hormonal contraceptive use among a representative sample of the Danish population born between 1981 and 2008. We hypothesized that the association between polygenic risk for these disorders and the likelihood of initiating hormonal contraception would be strongest at younger ages and then progressively decrease with increasing age. Because we were also interested in understanding whether genetic confounding could help explain the strong association between long‐lasting non‐oral forms of hormonal contraception and depression [2, 4], we first conducted our analyses on any hormonal contraceptive, and then stratified by route of administration (oral and non‐oral).

FIGURE 1.

FIGURE 1

Open in a new tab

Directed acyclic graph (DAG) illustrating the hypothesized confounding mechanism. This DAG was created using Daggity and illustrates the mechanism of confounding suggested in the present study. The circles represent measured (in color) and unmeasured (in gray) variables. The arrows represent directed causal relationships between the variables. Genetic risk for psychiatric disorders confounds the suggested causal relationship between hormonal contraception use in adolescence (the exposure) and risk of developing depression (the outcome).

2. Materials and Methods

2.1. Data Source

The study sample was drawn from the population‐representative sub‐cohort (N = 50,615) of the Lundbeck Foundation Initiative for Integrative Psychiatric Research (iPSYCH) study, which is described in detail elsewhere [21, 22]. Briefly, iPSYCH is a case‐cohort study of singletons born in Denmark between 1981 and 2008 conducted in two waves (2012 and 2015). iPSYCH includes genetic data as well as phenotype data obtained from Danish national registers. Demographic information was obtained from the Danish Civil Registration System [23]. Information on prescriptions was obtained from the Danish National Prescription Register, which contains information on all prescriptions redeemed at Danish community pharmacies between 1995 and 2020 [24]. Data on mental disorders were ascertained from the Danish Psychiatric Central Research Register, which contains all diagnoses assigned in inpatient psychiatric hospital settings from 1968 to 2018, and from outpatient and emergency visits from 1995 to 2018 [25]. For exclusion and censoring criteria, data on diagnoses of physical illness were ascertained via the Danish National Patient Register, which includes all diagnoses assigned in hospital settings between 1971 and 2018 [26]. Data linkage between registers was enabled by the CPR number; a unique identification number assigned to all individuals living in Denmark [23].

2.2. Genotyping

Sample collection, genotyping, and genetic quality control of the iPSYCH sample have been described in detail elsewhere [21, 22]. Briefly, DNA was extracted from blood samples collected at birth as part of routine testing and archived in the Danish Neonatal Screening Biobank [27]. The first wave of iPSYCH (iPSYCH2012) was genotyped using the Illumina Infinium Psych chip v1.0 array [21] and the second wave (iPSYCH2015i) was genotyped with the Illumina Global Screening array v2.0 [22]. In both samples, genotyping complied with the manufacturer's guidance. Principal components (PCs) were generated using the autoSVD algorithm from the R package bigsnpr following the guidelines [28]. We identified genetically homogeneous individuals as those with a log Mahalanobis distance below 4.8. The Mahalanobis distance was computed from the first 20 PCs using the function dist_ogk from the R package bigutilsr [28]. We also computed KING‐relatedness robust coefficient and excluded at random one of the individuals in the pairs with > 3rd degree relatedness.

2.3. Study Sample

A flowchart detailing the sample selection process is presented in Figure 2. We selected all unrelated individuals with female sex from the iPSYCH sub‐cohort who were of European ancestry, had genetic data that passed quality control, and were at least 10 years old by the administrative censoring date of December 31, 2016. Exclusion criteria were selected based on the criteria used by Skovlund et al., (2016) and included emigration, diagnosis of cancer, polycystic ovarian syndrome (PCOS), endometriosis, venous thrombosis, or hormonal contraceptive prescription prior to or on their 10th birthday. International Classification of Diseases (ICD) codes for exclusion criteria are listed in Tables S1 and S2. The oldest participants were 35 at the time of administrative censoring.

FIGURE 2.

FIGURE 2

Open in a new tab

Sample selection process.

2.4. Outcome—Hormonal Contraception Initiation

Redeemed prescriptions of hormonal contraception were ascertained from the National Prescription Register using Anatomical Therapeutic Classification (ATC) codes (see Table S3). The date at first redeemed prescription was used as the date for hormonal contraception initiation. For analyses stratified by method of administration, we categorized hormonal contraceptives as either oral or non‐oral. Oral hormonal contraceptives included combined‐hormone and progestin‐only products. Non‐oral hormonal contraceptives included the hormonal patch, vaginal ring, LNG‐IUS, implant, and the DMPA injection.

2.5. Explanatory Variables—Polygenic Scores

PGSs for MDD, ADHD, BD, and SCZ were computed with the MetaPRS method [29] which uses both individual‐level genetic data and summary statistics from external genome‐wide association studies (GWASs) to increase the effective training sample size. PGSs were calculated using GWAS summary statistics from Psychiatric Genomics Consortium (PGC) meta‐analyses [18, 30, 31, 32] and standardized according to the distribution within the iPSYCH sub‐cohort.

2.6. Statistical Analysis

Individuals were followed up from their 10th birthday until they redeemed a prescription for hormonal contraception (the outcome of interest), death, emigration, disqualifying physical health diagnosis (diagnosis of cancer, polycystic ovarian syndrome, endometriosis, or venous thrombosis), or December 31, 2016 (end of study period), whichever came first. For each of the four PGSs, we used Cox regression to estimate hazard ratios (HRs) using the survival package in R [33]. We included the following covariates in the regression models: the first five PCs, genotyping array, and birth year (in three‐/four‐year bands: 1981–1984, 1985–1988, 1989–1992, 1993–1996, 1997–2000, 2001–2003, and 2004–2006). Analyses were stratified by age at hormonal contraception initiation in the following four categories: 10–14, 15–19, 20–24, and 25+ years. Shoenfeld residual plots were used to assess the appropriateness of the age bins with regard to the proportional hazards' assumption. For the models considering oral and non‐oral contraceptives separately, we considered the first time the individual redeemed a prescription of that specific type of hormonal contraceptive regardless of whether they had used the other type before. Statistical significance was assessed using the Bonferroni‐adjusted the alpha level of 0.05/4 = 0.0125.

2.7. Sensitivity Analysis

To check whether the observed associations between the PGSs and hormonal contraception initiation were driven by the manifestation of a mental disorder, we repeated the analyses excluding individuals who were either diagnosed with a mental disorder in a hospital setting or redeemed a psychotropic drug prescription before study entry and censored them if they met these criteria during follow‐up. All ICD and ATC codes are presented in Tables S1–S3. We also conducted a sensitivity analysis excluding individuals born before 1985 (N = 2385), as this group did not have complete medication data when they were 10–14 years old.

3. Results

3.1. Sample Characteristics

Sample characteristics are presented in Table 1. The sample included 19,254 females born between 1981 and 2006. A total of 13,580 (71%) redeemed a prescription for hormonal contraception during the study period. The median age at hormonal contraception initiation was 16 years (interquartile range (IQR) = 2) for any type, 16 (IQR = 2) for oral, and 22 (IQR = 8) for non‐oral hormonal contraceptives, respectively. The vast majority (94%) of hormonal contraception users were prescribed the combined oral pill as their first hormonal contraceptive. Follow‐up time ranged from 0 to 20+ years. Further information about the sample is presented in Table S4.

TABLE 1.

Sample characteristics.

Characteristic N = 19,254 1
Year of birth
1981–1983 1675 (8.7%)
1984–1986 2183 (11%)
1987–1989 2247 (12%)
1990–1992 2495 (13%)
1993–1995 2552 (13%)
1996–1998 2436 (13%)
1999–2001 2409 (13%)
2002–2006 3257 (17%)
Any HC use 13,580 (71%)
Age first HC use 16.00 (2.00)
Oral HC use 13,392 (70%)
Age first oral HC use 16.00 (2.00)
Non‐oral HC use 3192 (17%)
Age first non‐oral HC use 22.0 (8.0)
Proportion of HC type at first use
Combined oral pill 12,820 (94%)
Patch 36 (0.3%)
Vaginal ring 91 (0.7%)
Progestin‐only oral pill 290 (2.1%)
DMPA injection 183 (1.3%)
Implant 111 (0.8%)
LNG‐IUS 49 (0.4%)
Open in a new tab

Abbreviations: DMPA = depot‐medroxyprogesterone acetate, HC = hormonal contraception, LNG‐IUS = levonorgestrel intrauterine system.

1

n (%); Median (IQR).

3.2. Associations Between PGSs and Hormonal Contraception

Associations between PGSs and hormonal contraception initiation in different age groups are shown in Figure 3 and Table S5. For any hormonal contraception use, all four PGS variables were associated with hormonal contraception initiation in the youngest age group (10–14 years) but the associations were strongest for PGS‐ADHD (HR = 1.21 [95% CI = 1.16–1.27], p = 6.16 × 10−18) and PGS‐MDD (1.21 [1.16–1.27], p = 1.22 × 10−17). Among older adolescents (ages 15–19 years), only PGS‐ADHD (1.13, [1.11–1.15], p = 2.97 × 10−34) and PGS‐MDD (1.06, [1.04–1.08], p = 2.82 × 10−9) were associated with hormonal contraception initiation. None of the PGS variables were associated with hormonal contraception initiation at ages 20–24 or 25+.

FIGURE 3.

FIGURE 3

Open in a new tab

Associations between polygenic scores (PGSs) and hormonal contraception (HC) use stratified by age at HC initiation. Analyses were performed with (a) any HC, (b) oral HC, and (c) non‐oral HC as the outcome of interest. *p < 0.0125 (Bonferroni‐adjusted alpha). ADHD = attention deficit hyperactivity disorder, BD = bipolar disorder, MDD = major depressive disorder, SCZ = schizophrenia.

When we stratified by route of administration, we saw an identical pattern of results for oral hormonal contraception, which is unsurprising given that 94% of the sample used the combined oral pill as their first contraceptive. For non‐oral hormonal contraceptives, PGS‐ADHD and PGS‐MDD were also associated with hormonal contraception initiation between ages 20–24 years (PGS‐ADHD: 1.15, [1.08–1.22], p = 1.08 × 10−5; PGS‐MDD: 1.09, [1.02–1.16], p = 0.007) in addition to ages 15–19. The p values fell below the Bonferroni‐corrected threshold for hormonal contraception initiation between ages 10–14 years, but the effect sizes remained high, suggesting that this was due to lack of power attributable to small numbers available for this sub‐analysis. There were no significant associations between PGS‐BD or PGS‐SCZ and age at initiation of non‐oral hormonal contraception.

3.3. Sensitivity Analysis

In the sensitivity analysis (N = 18,640), we excluded or censored individuals who received hospital treatment for any mental/behavioral disorder or were prescribed psychotropic drugs. The overall pattern was the same as in the main analyses (Figure S1; Tables S6 and S7), except for PGS‐BD which was no longer significantly associated with hormonal contraceptive initiation for the age group 10–14. There were also some differences in the analyses examining non‐oral hormonal contraceptives. Furthermore, while non‐oral contraception initiation at ages 20–24 was significantly associated with PGS‐MDD in the main analysis, it was not significantly associated with the sensitivity analysis.

In the analyses excluding individuals born before 1985, the results were similar except that PGS‐SCZ was more strongly associated with initiating hormonal contraception at age 25+ (Tables S8 and S9). This could indicate that secular changes in attitudes towards hormonal contraception have changed the underlying distribution of genetic liability in groups initiating hormonal contraception at older ages; however, the sample size in this age group is too small to draw any concrete conclusions.

4. Discussion

Our goal in this study was to assess the plausibility of the hypothesis that genetic liability for psychiatric and/or neurodevelopmental disorders could confound the relationship between age at hormonal contraception initiation and depression by demonstrating an association between PGS for MDD, BD, SCZ, and ADHD and age at first hormonal contraception use. We found dose–response relationships for PGS‐MDD and, in particular, PGS‐ADHD, such that the associations were strongest in the youngest age groups and decreased with age. These results are consistent with our hypothesis and suggest that previously identified associations [2, 4] between age at hormonal contraception use and depression may be partly confounded by genetic liability.

We found statistically significant associations between all four PGSs and hormonal contraceptive initiation at ages 10–14 years. This finding is not surprising, given that girls within this age range are not usually sexually active and therefore hormonal contraception use at this age is outside the norm. Indeed, Skovlund and colleagues purposely excluded this age group from their analyses in order to avoid exactly the sort of confounding our results suggest. However, the presence of positive associations in the 15–19 age group, particularly for PGS‐ADHD, suggests that confounding may be present in older adolescents as well. Further, sensitivity analyses demonstrated that this association was not mediated by manifested mental or behavioral disorders.

The strongest association, and the clearest dose–response effect, was for PGS‐ADHD. This finding is highly intuitive, as ADHD, a neurodevelopmental disorder, has an earlier onset than MDD, SCZ, and BPD, and is a known risk factor for depression in adulthood [34]. PGS‐ADHD is also strongly correlated with female reproductive traits including age at first birth and age at first sexual intercourse [35].

Past studies have found stronger associations between non‐oral hormonal contraceptives and later depression risk [2, 4]. Several of these studies proposed that women living in institutions and those with severe psychiatric illness or intellectual disability may be more likely to opt for non‐oral hormonal contraceptives as they obviate the need to remember a daily pill [2, 4]. Indeed, research shows that mental illness is associated with poor contraceptive compliance, which could make non‐oral hormonal contraceptives an attractive option for women with psychiatric disorders who wish to avoid pregnancy [36, 37]. Our results showed that women who used non‐oral hormonal contraceptives had, on average, a higher polygenic risk for MDD and ADHD even among adult users (ages 20–24). Furthermore, when we excluded individuals treated for mental/behavioral disorders, the effect sizes for PGSs for MDD and ADHD decreased and ceased to be statistically significant in the 10–14 age group and, for MDD, also in the 20–24 age group. This suggests that for non‐oral hormonal contraceptives, some of the association with PGSs may be mediated by mental illness.

Our results should not be interpreted to suggest that women who start to use hormonal contraception at younger ages are destined to become mentally unwell, or that people with higher polygenic risk for psychiatric disorders in all cases engage in a particular set of behaviors. Our findings say very little about the actual mechanisms underlying the observed correlations between polygenic risk and age at hormonal contraception initiation, which are sure to be complex and influenced by many additional risk factors associated with both age at contraception initiation and mental illness, such as age at menarche [38]. Instead, these results should be interpreted as highlighting the methodological consequences for research of the existence of associations between genetic liability and exposures and outcomes of interest. Genetic confounding is often unaddressed in epidemiological studies even though many exposures and outcomes are heritable [39]. The results of the present study suggest that future studies on this topic should be genetically informed and, where genetic data are available, should adjust for PGSs for MDD and ADHD. However, given that PGSs currently only explain a very small amount of the overall heritability of MDD and ADHD in the general population [18, 32, 40, 41], including PGSs in models of hormonal contraceptive use and depression will not eliminate the issue of genetic confounding. Hopefully this will improve as GWASs of psychiatric phenotypes become larger and better powered, and methods to develop PGSs become more sophisticated. In research cohorts where genetic data have not been collected, family history of psychiatric illness could be used, with an awareness that this also captures shared environmental risk. In a Swedish register‐based study, Lundin et al., (2021) found that the association between hormonal contraceptives and depression did not change when adjusting for family history of psychiatric illness [3].

4.1. Strengths and Limitations

A strength of this study is that the sample is representative of the Danish population born between 1981 and 2006. By examining only the first filled prescription of hormonal contraceptives, we could rule out selective discontinuation bias as an explanation for our findings, which is a key methodological issue of previous research on this topic. It is worth noting, however, that as 94% of the sample used the combined oral pill as their first hormonal contraceptive, our analysis of non‐oral forms could still be affected by selective discontinuation bias. The sample size for individuals whose first hormonal contraceptive was a non‐oral form was too small for meaningful analyses. Another limitation of this study is that it was restricted to individuals of European ancestries. Consequently, our findings may not generalize to individuals of more diverse ancestries living in Denmark or other countries. However, having an ancestrally homogeneous sample is also a strength, as it reduces the risk of confounding by population stratification, which is of particular concern in genetics studies [42]. Finally, because PGSs are incomplete (because they do not capture e.g., rare variants, CNVs, or gene‐by‐gene interactions) and imperfect (due to methodological limitations) measures of genetic liability, they cannot be used to quantify the full impact of genetics on age at first hormonal contraception use.

4.2. Conclusions

Our results suggest that the association between age at hormonal contraception initiation and risk for depression could be partly confounded by genetic liability for psychiatric and neurodevelopmental disorders, particularly MDD and ADHD. Future studies in this area should be aware that genetic confounding could inflate effect sizes and, where possible, adjust for it in statistical models.

Ethical Statement

The iPSYCH study was approved by the Scientific Ethics Committee in the Central Denmark Region (case no. 1‐10‐72‐287‐12) and the Danish Data Protection Agency. The Danish Scientific Ethics Committee waived the need for specific informed consent for the iPSYCH study.

Conflicts of Interest

S.D.Ø. owns/has owned units of mutual funds with stock tickers DKIGI, IAIMWC, SPIC25KL, and WEKAFKI, and owns/has owned units of exchange‐traded funds with stock tickers BATE, TRET, QDV5, QDVH, QDVE, SADM, IQQH, USPY, EXH2, 2B76, IS4S, OM3X, and EUNL.

Supporting information

Data S1.

ACPS-151-529-s001.docx (306.7KB, docx)

Acknowledgments

J.M. and K.L.M. received funding from The Lundbeck Foundation (R303‐2018‐3551). The iPSYCH team was supported by grants from the Lundbeck Foundation (R102‐A9118, R155‐2014‐1724, and R248‐2017‐2003), NIH/NIMH (1U01MH109514‐01 and 1R01MH124851‐01) and the Universities and University Hospitals of Aarhus and Copenhagen. The Danish National Biobank resource was supported by the Novo Nordisk Foundation. High‐performance computer capacity for handling and statistical analysis of iPSYCH data on the GenomeDK HPC facility was provided by the Center for Genomics and Personalized Medicine and the Centre for Integrative Sequencing, iSEQ, Aarhus University, Denmark. We gratefully acknowledge The Broad Institute for genotyping and the Major Depression, Bipolar Disorder, Schizophrenia, and Attention Deficit Hyperactivity Disorder working groups of the Psychiatric Genomics Consortium (PGC) and 23andMe for making summary statistics available. S.D.Ø. received the 2020 Lundbeck Foundation Young Investigator Prize. Outside this study, S.D.Ø. is supported by the Novo Nordisk Foundation (grant number: NNF20SA0062874), the Lundbeck Foundation (grant numbers: R358‐2020‐2341 and R344‐ 2020‐1073), the Danish Cancer Society (grant number: R283‐A16461), the Central Denmark Region Fund for Strengthening of Health Science (grant number: 1‐36‐72‐4‐20), The Danish Agency for Digitisation Investment Fund for New Technologies (grant number 2020–6720), and Independent Research Fund Denmark (grant numbers: 7016‐00048B and 2096‐00055A).

Funding: J.M. and K.L.M. received funding from The Lundbeck Foundation (R303‐2018‐3551).

Data Availability Statement

Owing to the sensitive nature of these iPSYCH data, individual‐level data can be accessed only through secure servers where the download of individual‐level information is prohibited. Each scientific project must be approved before initiation, and approval is granted to a specific Danish research institution. International researchers may gain data access through collaboration with a Danish research institution. More information about getting access to the iPSYCH data can be obtained at https://ipsych.dk/en/about‐ipsych.

References

  • 1. Lindberg M., Foldemo A., Josefsson A., and Wiréhn A. B., “Differences in Prescription Rates and Odds Ratios of Antidepressant Drugs in Relation to Individual Hormonal Contraceptives: A Nationwide Population‐Based Study With Age‐Specific Analyses,” European Journal of Contraception and Reproductive Health Care 17, no. 2 (2012): 106–118. [DOI] [PubMed] [Google Scholar]
  • 2. Skovlund C. W., Mørch L. S., Kessing L. V., and Lidegaard Ø., “Association of Hormonal Contraception With Depression,” JAMA Psychiatry 73, no. 11 (2016): 1154–1162. [DOI] [PubMed] [Google Scholar]
  • 3. Lundin C., Wikman A., Lampa E., et al., “There Is No Association Between Combined Oral Hormonal Contraceptives and Depression: A Swedish Register‐Based Cohort Study,” BJOG: An International Journal of Obstetrics and Gynaecology 129, no. 6 (2022): 917–925. [DOI] [PubMed] [Google Scholar]
  • 4. Zettermark S., Perez Vicente R., and Merlo J., “Hormonal Contraception Increases the Risk of Psychotropic Drug Use in Adolescent Girls but Not in Adults: A Pharmacoepidemiological Study on 800 000 Swedish Women,” PLoS One 13, no. 3 (2018): e0194773. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Stenhammar E., Wikman P., Gemzell Danielsson K., Kopp‐Kallner H., and Sundström P. I., “Levonorgestrel Intrauterine Device and Depression: A Swedish Register‐Based Cohort Study,” International Journal of Psychophysiology 193 (2023): 112230. [DOI] [PubMed] [Google Scholar]
  • 6. Sanders S. A., Graham C. A., Bass J. L., and Bancroft J., “A Prospective Study of the Effects of Oral Contraceptives on Sexuality and Well‐Being and Their Relationship to Discontinuation,” Contraception 64, no. 1 (2001): 51–58. [DOI] [PubMed] [Google Scholar]
  • 7. Westhoff C. L., Heartwell S., Edwards S., et al., “Oral Contraceptive Discontinuation: Do Side Effects Matter?,” American Journal of Obstetrics and Gynecology 196, no. 4 (2007): 412‐e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Ott M. A., Shew M. L., Ofner S., Tu W., and Fortenberry J. D., “The Influence of Hormonal Contraception on Mood and Sexual Interest Among Adolescents,” Archives of Sexual Behavior 37, no. 4 (2008): 605–613. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Martell S., Marini C., Kondas C. A., and Deutch A. B., “Psychological Side Effects of Hormonal Contraception: A Disconnect Between Patients and Providers,” Contraception and Reproductive Medicine 8, no. 1 (2023): 9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Joseph K. S., Mehrabadi A., and Lisonkova S., “Confounding by Indication and Related Concepts,” Current Epidemiology Reports 1, no. 1 (2014): 1–8. [Google Scholar]
  • 11. Ramrakha S., Caspi A., Dickson N., Moffitt T. E., and Paul C., “Psychiatric Disorders and Risky Sexual Behaviour in Young Adulthood: Cross Sectional Study in Birth Cohort,” BMJ 321, no. 7256 (2000): 263–266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Meier A. M., “Adolescent First Sex and Subsequent Mental Health,” American Journal of Sociology 112, no. 6 (2007): 1811–1847. [Google Scholar]
  • 13. Zimmer‐Gembeck M. J. and Helfand M., “Ten Years of Longitudinal Research on U.S. Adolescent Sexual Behavior: Developmental Correlates of Sexual Intercourse, and the Importance of Age, Gender and Ethnic Background,” Developmental Review 28, no. 2 (2008): 153–224. [Google Scholar]
  • 14. Jamieson L. K. and Wade T. J., “Early Age of First Sexual Intercourse and Depressive Symptomatology Among Adolescents,” Journal of Sex Research 48, no. 5 (2011): 450–460. [DOI] [PubMed] [Google Scholar]
  • 15. Francis J., Presser L., Malbon K., Braun‐Courville D., and Linares L. O., “An Exploratory Analysis of Contraceptive Method Choice and Symptoms of Depression in Adolescent Females Initiating Prescription Contraception,” Contraception 91, no. 4 (2015): 336–343. [DOI] [PubMed] [Google Scholar]
  • 16. Musliner K. L., Mortensen P. B., McGrath J. J., et al., “Association of Polygenic Liabilities for Major Depression, Bipolar Disorder, and Schizophrenia With Risk for Depression in the Danish Population,” JAMA Psychiatry 76, no. 5 (2019): 516–525. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Mills M. C., Tropf F. C., Brazel D. M., et al., “Identification of 371 Genetic Variants for Age at First Sex and Birth Linked to Externalising Behaviour,” Nature Human Behaviour 5, no. 12 (2021): 1717–1730. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Wray N. R., Ripke S., Mattheisen M., et al., “Genome‐Wide Association Analyses Identify 44 Risk Variants and Refine the Genetic Architecture of Major Depression,” Nature Genetics 50, no. 5 (2018): 668–681. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Murray G. K., Lin T., Austin J., McGrath J. J., Hickie I. B., and Wray N. R., “Could Polygenic Risk Scores Be Useful in Psychiatry?—A Review,” JAMA Psychiatry 78, no. 2 (2021): 210–219. [DOI] [PubMed] [Google Scholar]
  • 20. Wray N. R., Lin T., Austin J., et al., “From Basic Science to Clinical Application of Polygenic Risk Scores: A Primer,” JAMA Psychiatry 78, no. 1 (2021): 101–109. [DOI] [PubMed] [Google Scholar]
  • 21. Pedersen C. B., Bybjerg‐Grauholm J., Pedersen M. G., et al., “The iPSYCH2012 Case‐Cohort Sample: New Directions for Unravelling Genetic and Environmental Architectures of Severe Mental Disorders,” Molecular Psychiatry 23, no. 1 (2018): 6–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Bybjerg‐Grauholm J., Pedersen C. B., Bækvad‐Hansen M., et al., “The iPSYCH2015 Case‐Cohort Sample: Updated Directions for Unravelling Genetic and Environmental Architectures of Severe Mental Disorders,” medRxiv (2020), 10.1101/2020.11.30.20237768. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Pedersen C. B., “The Danish Civil Registration System,” Scandinavian Journal of Public Health 39, no. 7 Suppl (2011): 22–25. [DOI] [PubMed] [Google Scholar]
  • 24. Kildemoes H. W., Sørensen H. T., and Hallas J., “The Danish National Prescription Registry,” Scandinavian Journal of Public Health 39, no. 7 Suppl (2011): 38–41. [DOI] [PubMed] [Google Scholar]
  • 25. Mors O., Perto G. P., and Mortensen P. B., “The Danish Psychiatric Central Research Register,” Scandinavian Journal of Public Health 39, no. 7 Suppl (2011): 54–57. [DOI] [PubMed] [Google Scholar]
  • 26. Lynge E., Sandegaard J. L., and Rebolj M., “The Danish National Patient Register,” Scandinavian Journal of Public Health 39, no. 7 Suppl (2011): 30–33. [DOI] [PubMed] [Google Scholar]
  • 27. Nørgaard‐Pedersen B. and Hougaard D. M., “Storage Policies and Use of the Danish Newborn Screening Biobank,” Journal of Inherited Metabolic Disease 30, no. 4 (2007): 530–536. [DOI] [PubMed] [Google Scholar]
  • 28. Privé F., Luu K., Blum M. G. B., McGrath J. J., and Vilhjálmsson B. J., “Efficient Toolkit Implementing Best Practices for Principal Component Analysis of Population Genetic Data,” Bioinformatics 36, no. 16 (2020): 4449–4457. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Albiñana C., Grove J., McGrath J. J., et al., “Leveraging Both Individual‐Level Genetic Data and GWAS Summary Statistics Increases Polygenic Prediction,” American Journal of Human Genetics 108, no. 6 (2021): 1001–1011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Schizophrenia Working Group of the Psychiatric Genomics Consortium , “Biological Insights From 108 Schizophrenia‐Associated Genetic Loci,” Nature 511, no. 7510 (2014): 421–427. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Mullins N., Forstner A. J., O'Connell K. S., et al., “Genome‐Wide Association Study of More Than 40,000 Bipolar Disorder Cases Provides New Insights Into the Underlying Biology,” Nature Genetics 53, no. 6 (2021): 817–829. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Demontis D., Walters R. K., Martin J., et al., “Discovery of the First Genome‐Wide Significant Risk Loci for Attention Deficit/Hyperactivity Disorder,” Nature Genetics 51, no. 1 (2019): 63–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Therneau T. M., “A Package for Survival Analysis in R,” 2023, https://CRAN.R‐project.org/package=survival.
  • 34. Meinzer M. C. and Chronis‐Tuscano A., “ADHD and the Development of Depression: Commentary on the Prevalence, Proposed Mechanisms, and Promising Interventions,” Current Developmental Disorders Reports 4, no. 1 (2017): 1–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Ni G., Amare A. T., Zhou X., Mills N., Gratten J., and Lee S. H., “The Genetic Relationship Between Female Reproductive Traits and Six Psychiatric Disorders,” Scientific Reports 9, no. 1 (2019): 12041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Walsemann K. M. and Perez A. D., “Anxiety's Relationship to Inconsistent Use of Oral Contraceptives,” Health Education & Behavior 33, no. 2 (2006): 197–214. [DOI] [PubMed] [Google Scholar]
  • 37. Hall K. S., Moreau C., Trussell J., and Barber J., “Role of Young Women's Depression and Stress Symptoms in Their Weekly Use and Nonuse of Contraceptive Methods,” Journal of Adolescent Health 53, no. 2 (2013): 241–248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Hirtz R., Hars C., Naaresh R., et al., “Causal Effect of Age at Menarche on the Risk for Depression: Results From a Two‐Sample Multivariable Mendelian Randomization Study,” Frontiers in Genetics 13 (2022): 918584. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Pingault J. B., Rijsdijk F., Schoeler T., et al., “Genetic Sensitivity Analysis: Adjusting for Genetic Confounding in Epidemiological Associations,” PLoS Genetics 17, no. 6 (2021): e1009590. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Levey D. F., Stein M. B., Wendt F. R., et al., “Bi‐Ancestral Depression GWAS in the Million Veteran Program and Meta‐Analysis in > 1.2 Million Individuals Highlight New Therapeutic Directions,” Nature Neuroscience 24, no. 7 (2021): 954–963. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Als T. D., Kurki M. I., Grove J., et al., “Depression Pathophysiology, Risk Prediction of Recurrence and Comorbid Psychiatric Disorders Using Genome‐Wide Analyses,” Nature Medicine 29, no. 7 (2023): 1832–1844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Price A. L., Patterson N. J., Plenge R. M., Weinblatt M. E., Shadick N. A., and Reich D., “Principal Components Analysis Corrects for Stratification in Genome‐Wide Association Studies,” Nature Genetics 38, no. 8 (2006): 904–909. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1.

ACPS-151-529-s001.docx (306.7KB, docx)

Data Availability Statement

Owing to the sensitive nature of these iPSYCH data, individual‐level data can be accessed only through secure servers where the download of individual‐level information is prohibited. Each scientific project must be approved before initiation, and approval is granted to a specific Danish research institution. International researchers may gain data access through collaboration with a Danish research institution. More information about getting access to the iPSYCH data can be obtained at https://ipsych.dk/en/about‐ipsych.

Articles from Acta Psychiatrica Scandinavica are provided here courtesy of Wiley

RESOURCES