Interactive effects of genetic variants and hormonal contraceptive use on depression

SHL Pre-Registration Worksheet Study Name: Interactive effects of genetic variants and hormonal contraceptive use on depression 1) Have any data been collected for this study yet? 0 (a) No data have been collected 0 (b) Some data have been collected but not analyzed X (c) Some data have been collected and analyzed If (b) or (c), please explain briefly: We will perform secondary data analyses on previously collected data from the UK Biobank. Although some of the co-authors have previously worked data from UK Biobank, no prior analyses were conducted that address the core research questions described below. The first author had access to the data and constructed the analytical sample, but did not perform any analyses prior to completing this preregistration. 2) Hypothesis: What’s the main question being asked or hypothesis being tested? The aim of this study is to investigate interactive effects of genetic variants and hormonal contraceptive use on incident depression using a genome-wide approach. 3) Give a brief (1-2 sentence) overview of the methods. Using data of females from the UK Biobank as discovery sample, we will perform three sets of cox regression analyses: 1) with oral contraceptive (OC) use on incident depression; 2) a genome-wide association study (GWAS) on incident depression; and 3) a genome-wide-by-drug-interaction study (GWDIS) with OC use on incident depression. 4) Independent variables: Describe the conditions (for an experimental study) or predictor variables (for a correlational study). During the initial UK Biobank assessment visit, information on OC use (UKB field: 2784) and the age when first initiated (UKB field: 2794) was obtained through a questionnaire. 5) Dependent variables: Describe the key dependent variable(s) specifying how they will be measured. Information on the first diagnosis of broad depression was obtained from either the verbal interview during the initial UK Biobank assessment visit or the International Classification of Disease code F32 recorded in the inpatient hospital or primary care data. During the initial UK Biobank assessment visit, participants were asked whether they had ever been told by a doctor that they had an illness or disability. The interviewer showed a tree structure of diagnoses which was loosely based on the ICD-10 codes. For each illness the participant recorded, the interviewer recorded their age or the year of the first diagnosis. Their age or the year of diagnosis was processed into a mid-point of that age or year, e.g. age 30 at first diagnosis would be 30 years + 6 months. If a participant had both an ICD10 code identified in the registers (i.e., data from either the death register, hospital inpatient admissions, or primary care) and a self-reported medical condition, the earliest date at which a depression diagnosis was identified in any of these sources was used (UKB field: 130894).1–3 6) Mediator variables: Describe any variables you expect to mediate the relationship between your IV’s and DV. Specify how they will be measured. NA 7) Moderator variables: Describe any variables you expect to moderate the relationship between your IV’s and DV. Specify how they will be measured. DNA was extracted from stored blood samples which was collected from participants during their study visit using either the UK BiLEVE array or the UK Biobank Axiom array. Detailed information about array design, genotyping, and quality control procedures has been described previously.4 In brief, the phasing and imputing were performed using the Haplotype Reference Consortium (HRC) and additionally with the merged UK10K and 1000 Genomes phase 3 reference panels. The imputed data was then combined, using HRC imputation when a SNP was present in both panels.4 Whole Genome Sequencing data for 500,000 participants was re-processed using DRAGEN 3.7.8, with individual and joint-called data available for all participants. Quality control will be performed with PLINK v2.0.5 Variants with a minor allele frequency<1%, or an imputation accuracy Info score <0.8 will be removed.6 8) Analyses: Describe what analyses (e.g., t-test, repeated-measures ANOVA) you will use to test your main hypotheses. We will perform a cox regression model with OC use on incident depression, a genome-wide association study (GWAS) on incident depression, and a genome-wide-by-drug-interaction study (GWDIS) with OC use on incident depression. The participants will be followed from birth until the first occurrence of depression, if they reached menopause (UKB field: 3581), underwent a hysterectomy (UKB field: 2824) or a bilateral oophorectomy (UKB field: 3882), or the end of follow-up (age at UKB assessment visit; UKB field: 21022) whichever came first. A previous study using UK Biobank data showed that the effect of OC use on depression is highest after two years of OC use, therefore OC users will be censored two years after OC initiation.3 For the GWAS and the GWDIS, the software REGENIE will be used to conduct cox regression for time to event data.7 All analyses will be adjusted for year of birth (UKB field: 34),8 Townsend deprivation index (proxy for socioeconomic status; UKB field: 22189),9,10 age at menarche (UKB field: 2714),17,18 age at sexual debut (UKB field: 2139),10,19 ovarian dysfunction (UKB field: 130736),11,12 endometriosis (UKB field: 132122),13,14 dysmenorrhea (UKB field: 132146),15,16 the first eight genetic principal components (UKB field: 22009),2 and genotyping array (for the GWAS and GWDIS; UKB field: 22000). The GWDIS will include a SNP x OC use interaction term. 9) More analyses. Are there any secondary analyses you plan to conduct? (e.g., order or gender effects) NA 10) Sample. Where and from whom will data be collected? How will you decide when to stop collecting data (e.g., target sample size based on power analysis, set amount of time)? If you plan to look at the data using sequential analysis, describe that here. NA 11) Exclusion criteria: Who will be excluded (e.g., outliers, participant who fail manipulation check, demographic exclusions)? Will they be replaced by other participants? We will exclude participants who withdrew their consent, who do not have genetic data available and participants with male sex (UKB field: 31). We will only include participants who self-identified as ‘White British’ and have very similar genetic ancestry based on a principal components analysis of the genotypes (UKB field: 22006). Participants will be excluded in the case of sex mismatches (UKB field: 22001), outliers for heterozygosity (UKB field: 22027) or in case of missing data in OC use. Finally, out of all participants with shared relatedness of up to the third degree (UKB field: 22021; kinship coefficients >0.044), only one participant will be included to screen out genetically related individuals.4 12) How do you plan to handle missing data? Missing data in the covariates will be handled using single imputation procedures. Pre-registration written by (initials): C.A.E. Pre-registration reviewed by (initials): F.S.C. References 1. First Occurrence of Health Outcomes Defined by 3-Character ICD10 Code.; 2019. 2. Silveira PP, Pokhvisneva I, Howard DM, Meaney MJ. A sex-specific genome-wide association study of depression phenotypes in UK Biobank. Mol Psychiatry. 2023;28(6):2469-2479. doi:10.1038/s41380-023-01960-0 3. Johansson T, Vinther Larsen S, Bui M, Ek WE, Karlsson T, Johansson Å. Population-based cohort study of oral contraceptive use and risk of depression. Epidemiol Psychiatr Sci. 2023;32:e39. doi:10.1017/S2045796023000525 4. Bycroft C, Freeman C, Petkova D, et al. The UK Biobank resource with deep phenotyping and genomic data. Nature. 2018;562(7726):203-209. doi:10.1038/s41586-018-0579-z 5. Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ. Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience. 2015;4(1). doi:10.1186/s13742-015-0047-8 6. Adams MJ, Streit F, Meng X, et al. Trans-ancestry genome-wide study of depression identifies 697 associations implicating cell types and pharmacotherapies. Cell. 2025;188(3):640-652.e9. doi:10.1016/j.cell.2024.12.002 7. Mbatchou J, Barnard L, Backman J, et al. Computationally efficient whole-genome regression for quantitative and binary traits. Nat Genet. 2021;53(7):1097-1103. doi:10.1038/s41588-021-00870-7 8. Keyes KM, Nicholson R, Kinley J, et al. Age, Period, and Cohort Effects in Psychological Distress in the United States and Canada. Am J Epidemiol. 2014;179(10):1216-1227. doi:10.1093/aje/kwu029 9. Freeman A, Tyrovolas S, Koyanagi A, et al. The role of socio-economic status in depression: results from the COURAGE (aging survey in Europe). BMC Public Health. 2016;16(1):1098. doi:10.1186/s12889-016-3638-0 10. Doornweerd AM, Branje S, Nelemans SA, et al. Stable Anxiety and Depression Trajectories in Late Adolescence for Oral Contraceptive Users. Front Psychiatry. 2022;13. doi:10.3389/fpsyt.2022.799470 11. Xi D, Chen B, Tao H, Xu Y, Chen G. The risk of depressive and anxiety symptoms in women with premature ovarian insufficiency: a systematic review and meta-analysis. Arch Womens Ment Health. 2023;26(1):1-10. doi:10.1007/s00737-022-01289-7 12. Zehravi M, Maqbool M, Ara I. Depression and anxiety in women with polycystic ovarian syndrome: a literature survey. Int J Adolesc Med Health. 2021;33(6):367-373. doi:10.1515/ijamh-2021-0092 13. Moore J, Kennedy S, Prentice A. Modern combined oral contraceptives for pain associated with endometriosis. In: Kennedy S, ed. Cochrane Database of Systematic Reviews. John Wiley & Sons, Ltd; 1997. doi:10.1002/14651858.CD001019 14. Gambadauro P, Carli V, Hadlaczky G. Depressive symptoms among women with endometriosis: a systematic review and meta-analysis. Am J Obstet Gynecol. 2019;220(3):230-241. doi:10.1016/j.ajog.2018.11.123 15. Zhao S, Wu W, Kang R, Wang X. Significant Increase in Depression in Women With P

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.