Quarshie et al. BMC Res Notes (2020) 13:271 https://doi.org/10.1186/s13104-020-05111-4 BMC Research Notes RESEARCH NOTE Open Access Prevalence of self-harm among lesbian, gay, bisexual, and transgender adolescents: a comparison of personal and social adversity with a heterosexual sample in Ghana Emmanuel N.‑B. Quarshie1* , Mitch G. Waterman2 and Allan O. House3 Abstract Objectives: We sought to estimate the prevalence of self‑reported self‑harm among adolescents identifying as lesbian, gay, bisexual, and transgender (LGBT) in Ghana, and compare self‑reported personal and social adversities related to self‑harm in this group to those in a random sample of heterosexual adolescents from the same locality. Results: A total of 444 adolescents aged 13‑21 years, comprising 74 LGBT adolescents and 370 heterosexual ado‑ lescents, provided data. The lifetime prevalence estimate of self‑harm was higher in the LGBT group (47%) than the heterosexual group (23%). The LGBT group reported a higher rate of self‑harm during the previous 12 months (45%), compared to the heterosexual group (18%). LGBT adolescents reported more alcohol and substance use and more personal social adversities, including various forms of victimisation, than heterosexual adolescents. They were no more likely to report difficulty in making and keeping friends or schoolwork problems than were heterosexual adolescents. Keywords: Accra, Adolescents, Ghana, LGBT, Self‑harm, Sexual minority, Homosexuality, Street‑connected adolescents, Sub‑Saharan Africa, Suicide Introduction prevalence of self-harm in sexual or gender minority Across the world, self-harm is the single strongest risk young people in countries in sub-Saharan Africa. factor for suicide among all age groups [1, 2]. In sub- Based on previous recommendations [12], we included Saharan Africa, suicide remains in the top 12 leading questions on sexual orientation (lesbian, gay or bisexual) causes of death among young persons aged 10–24 years and gender identity (transsexual) in a cross-sectional sur- [3]. Leading researchers and the WHO have reported vey undertaken in the Greater Accra region of Ghana, in that young people identifying as in a sexual or gender 2017; one of our objectives was to estimate the prevalence minority group are at elevated risk of self-harm and even- of self-harm among 2107 adolescents aged 13-21 [13]. Of tual suicide compared to heterosexual young people [1, the 2107 participants, 3.5% (n = 74) self-identified as les- 4–8]. However, neither our own recent systematic lit- bian, gay, bisexual, and transgender (LGBT). The objec- erature review [9], nor those of Aggarwal et al. [10], and tive of this research note is to report on the prevalence of Lim et  al. [11] found studies providing evidence on the self-reported self-harm among those adolescent partici- pants who self-identified as LGBT, and to describe how they compare to a random sample of heterosexual ado- *Correspondence: enquarshie@ug.edu.gh; enquarshie@gmail.com lescents drawn from the same survey, on personal factors 1 Department of Psychology, University of Ghana, Legon, P.O. Box LG 84, and social adversities related to self-harm. Accra, Ghana Full list of author information is available at the end of the article © The Author(s) 2020. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licens es/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/publi cdoma in/ zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Quarshie et al. BMC Res Notes (2020) 13:271 Page 2 of 6 Main text A preliminary logistic regression analysis showed Methods inflated odds ratios and infinite 95% confidence intervals We have described in detail the participants, sampling due to the small sample size and sparse data. Thus, the and data collection procedures followed in the larger sur- present study presents exploratory, descriptive evidence vey [13]. We randomly selected a regionally representa- on the prevalence estimates of self-harm among LGBT tive sample of 2107 adolescents: 1723 adolescents across adolescents in Ghana, and how they compare with heter- 20  second cycle schools, and 384 street-connected ado- osexual adolescents on some personal factors and social lescents from four charity facilities and 10 street census adversities. enumeration areas within the Greater Accra region of Ghana. The survey used a 4-section questionnaire with Results 66 items partly developed by the researchers and partly The mean age of the LGBT sample was 17.1  years adopted from existing key measures (e.g., the Self-Inju- (SD = 1.4, modal age = 17), while the mean age of the rious Thoughts and Behaviors Interview [14], the Suicide heterosexual adolescents was 16.7 years (SD = 1.4, modal Attempt Self-Injury Interview [15], the 2012 WHO– age = 17). Table  1 presents the demographic character- Global School-based Student Health Survey question- istics of participants in this study. More females (n = 45; naire [16]). 60.8%) identified as LGBT than males (n = 29; 39.2%). We asked a composite question about sexual orienta- Nearly half (n = 36; 48.6%) of the LGBT adolescents tion and gender identity, with response options, “hetero- reported as coming from families where the father had sexual”, “lesbian”, “gay”, “bisexual”, “transgender”. Of the more than one wife. 2107 participants, 2030 self-identified as heterosexual, four indicated that they were lesbian, three identified as Prevalence of self‑harm gay, 38 as bisexual, and 29 as transgender; three partici- As shown in Table 2, the lifetime prevalence of self-harm pants did not check any response. We combined all those was higher in the LGBT group (47%) than the heterosex- who replied affirmatively to any of the options into a ual group (23%) [DP 0.24, 95% CI 0.12–0.36]. Similarly, composite LGBT group (n = 74) for the analysis. We used the LGBT group reported a higher rate of self-harm dur- the ‘random sample of cases’ function in SPSS to draw an ing the previous 12 months (44.6%), compared to the het- analytic comparative random sample of 370 adolescents erosexual group (16.2%) [DP 0.27, 95% CI 0.15–0.39]. from the 2030 heterosexual group in a ratio of 5:1, in order to ensure optimal precision of results [17–19]. Personal and social adversity Lifetime self-harm was assessed with the question, The difference in proportions (DP) and associated 95% “Have you actually ever intentionally harmed yourself CIs (Table 2) show that adolescents identifying as LGBT (e.g., cutting, burning, or poisoning yourself, or tried to were more likely than those identifying as heterosexual harm yourself in some other way, for example, hanging, to report personal level factors such as alcohol use, drug jumping from height etc.)?”, while self-harm during the use, and sexual orientation worry during the previous past 12  months was measured with the question, “Did 12 months. Social adversities experienced at higher rates you actually intentionally harm yourself during the past by the LGBT adolescents during the previous 12 months 12 months or 1 year?”. The response options for lifetime included sexual abuse victimisation, trouble with police, self-harm and self-harm during the past 12 months were breakup, conflict between parents, and conflict with par- binary: “No” or “Yes”. Several questions related to per- ents. However, there was no statistically significant dif- sonal factors e.g., sexual orientation worry, alcohol use, ference between the LGBT group and the heterosexual and social adversities e.g., conflict with parents, trouble group in terms of having difficulty making or keeping with police, experienced during the previous 12 months, friends, bullying victimisation, schoolwork problems, and with dichotomous response format (“No” or “Yes”) were attempted suicide in the family. also included. We performed descriptive statistical analysis of the Discussion data using SPSS (version 26.0 for Windows). We used To the best of our knowledge, this is the first study from frequencies and percentages to present the demographic Western sub-Saharan Africa providing evidence on the characteristics of the participants and the prevalence prevalence of self-harm among LGBT adolescents and a estimates of self-harm. We applied 95% confidence inter- comparative group of adolescents identifying as hetero- vals (CI) to assess the uncertainty around the effect esti- sexual. This study has shown that about 4 in 10 adoles- mates of the difference in proportions (DP) between the cents identifying as LGBT, compared to approximately LGBT and heterosexual groups [20–22]. 2 in 10 heterosexual adolescents, report self-harm dur- ing the previous 12 months. LGBT adolescents are likely Q uarshie et al. BMC Res Notes (2020) 13:271 Page 3 of 6 Table 1 Demographics of the two study samples Variable Adolescent group Difference in proportions (95% CI for difference) LGBT (n = 74) Heterosexual (n = 370) n (%) n (%) Adolescent group In‑school 48 (65) 305 (82) − 0.17 (− 0.29, − 0.06) Street‑connected 26 (35) 65 (18) 0.17 (0.06, 0.29) Sex Male 29 (39) 185 (50) − 0.11 (− 0.23, 0.01) Female 45 (61) 185 (50) 0.11 (− 0.01, 0.23) Age 13–15 7 (9) 69 (19) − 0.1 (− 0.17, − 0.01) 16–17 36 (49) 204 (55) − 0.06 (− 0.19, 0.06) 18–21 31 (42) 97 (26) 0.16 (0.04, 0.28) Family structure Father has 1 wife 38 (51) 265 (72) − 0.21 (− 0.32, − 0.08) Father has > 1 wife 36 (49) 105 (28) 0.21 (0.08, 0.32) Sibling size ≤ 4 49 (66) 252 (68) − 0.02 (− 0.14, 0.09) > 4 25 (34) 118 (32) 0.02 (− 0.09, 0.14) Living arrangement One or both parents 38 (51) 264 (71) − 0.2 (0.23, 0.47) Other relation 14 (19) 61 (17) 0.02 (− 0.07, 0.12) Alone/with another person 22 (30) 45 (12) 0.18 (0.07, 0.28) Religious group Christian 57 (80) 319 (87) − 0.07 (− 0.19, 0.01) Muslim 14 (20) 48 (13) 0.07 (− 0.03, 0.15) In romantic relationship No 30 (40) 235 (64) − 0.24 (− 0.35, − 0.11) Yes 44 (60) 135 (36) 0.24 (0.11, 0.35) than heterosexual adolescents to report alcohol and because those individuals  exhibited (sexual) behaviours substance use and to be worried about their sexual ori- that  were considered stereotypically opposite to their entation. LGBT adolescents are likely to report a range sex [32]. This environment of criminality, social hostil- of social adversities than heterosexual adolescents. The ity, stigma, tension and strong heteronormativity, as findings from this study are consistent with what pertains found also in high-income countries [33, 34], gives rise in high-income countries [4–6, 8, 23]. to increased vulnerability of persons identifying as sexual Homosexuality is criminalised in Ghana [24] and cul- and gender  minorities to negative health outcomes and turally tabooed, religiously proscribed, and strongly stig- risky health behaviours including alcohol and substance matised across many countries in Africa [25–27]; about use and abuse, self-harm, and suicide [35–38]. 96% of Ghanaians oppose homosexuality [26]. Anecdo- While sexual orientation is widely discussed in the tal reports from Ghana suggest that persons found to be literature as a risk for mental health problems, gen- engaged in sexual behaviours considered non-heterosex- der identity is much less so. One reason may be that it ual have been arrested by the police for prosecution [28, is under-recognised as a health issue [39]. However, the 29], some have been thrown out or disowned by their World Health Organisation has recognised the need for families [26, 29], while students have been suspended or more focused intervention in this area [1], raising the expelled by school authorities for ‘coming out’ as non- possibility of more active intervention to improve the heterosexual [30, 31]. A recent study on vigilantism in health of LGBT young people in sub-Saharan Africa. Ghana reports that some  individuals have been labelled While homosexual related legislation does not nec- homosexual and assaulted by community members, essarily lead to reductions in self-harm and suicide in Quarshie et al. BMC Res Notes (2020) 13:271 Page 4 of 6 Table 2 Difference in  proportions between  LGBT and  heterosexual adolescents across  personal factors and  social adversities experienced in the previous 12 months Variable Adolescent group Difference in proportions (95% CI for difference) LGBT (n = 74) Heterosexual (n = 370) n (%) n (%) Self‑reported self‑harm Lifetime self‑harm 35 (47) 85 (23) 0.24 (0.12, 0.36) Self‑harm in the past 12 months 33 (45) 65 (18) 0.27 (0.15, 0.39) Suicide in family 11 (15) 10 (3) 0.12 (0.04, 0.20) Attempted suicide in family 14 (19) 37 (10) 0.09 (− 0.01, 0.18) Friend suicide 12 (16) 15 (4) 0.12 (0.03, 0.21) Friend attempted suicide 22 (30) 39 (11) 0.19 (0.08, 0.30) Alcohol and substance use Alcohol use 35 (47) 54 (14) 0.33 (0.21, 0.45) Cigarette smoking 16 (21) 5 (1) 0.20 (0.11, 0.29) Drug use 19 (25) 14 (3) 0.22 (0.12, 0.32) Family conflict Conflict between parents 41 (55) 153 (41) 0.14 (0.02, 0.26) Conflict with parents 28 (37) 94 (25) 0.12 (0.01, 0.24) Other social problems Conflict with friends 47 (63) 183 (49) 0.14 (0.02, 0.26) Breakup 28 (37) 76 (20) 0.17 (0.05, 0.29) Difficulty making/keeping friends 25 (34) 134 (36) − 0.02 (− 0.14, 0.09) Sexual orientation worry 20 (27) 10 (3) 0.24 (0.14, 0.35) Other problems Schoolwork problems 24 (32) 117 (31) 0.1 (− 0.11, 0.12) Bullying victimisation 36 (48) 134 (36) 0.12 (0.00, 0.25) Sexual abuse victimisation 38 (51) 8 (10) 0.41 (0.38, 0.61) Physical abuse victimisation 44 (59) 148 (40) 0.19 (0.07, 0.32) Trouble with police 18 (24) 24 (6) 0.18 (0.08, 0.28) members of sexual or gender minorities [40], it is now Limitations well documented that pro-homosexual legislation, The small sample size of adolescents identifying as together with acceptances by others, supportive family LGBT in our larger cross-sectional survey sample in the and public attitudes and secure environment, have the Greater Accra Region made it impossible to use more potential of leading to positive health outcomes in per- robust multivariate statistical modelling techniques to sons identifying as sexual or gender minority [36, 41–43]. stratify the analysis by sexual orientation categories, The evidence of this study underscores the need for gender, and whether the young people were in school or families, school staff, social and healthcare profession- street-connected. als to show positive attitudes and be supportive of LGBT Also, all the participants were accessed within the adolescents, thereby creating a secure environment to Greater Accra Region of Ghana—which is entirely urban- reduce vulnerabilities and risks to self-harm. Similarly, ised. Therefore, our results may not be necessarily gener- besides developing welcoming attitudes to LGBT young alisable across LGBT adolescents in rural Ghana. people, leaders of religious institutions (e.g., schools) and Social desirability bias is possible—some of the adoles- organisations (e.g., churches, mosques) could consider cents actually  identifying as LGBT might have checked collaborating with the mass media to begin public edu- the heterosexual orientation response option in the larger cation programmes aimed at changing negative attitudes study. However, it is notable that 74 adolescents checked and working towards removing the predominant socio- one of the LGBT options—a sample size that is high cultural factors which give rise to hostilities against sex- compared to a recent study of suicidal behaviours among ual and gender minorities in Ghana. college students in Ghana [44]. Q uarshie et al. BMC Res Notes (2020) 13:271 Page 5 of 6 Clearly, more studies on the health and well-being behalf of street‑connected adolescents aged 13–17 years. We have pseudo‑ of young people identifying as LGBT are needed from anonymised or completely anonymised all potentially identifying information including specific names of schools and charity facilities where participants Ghana, and sub-Saharan Africa. Thus far across sub- were approached. Saharan Africa, only one study from South Africa has been published on the personal experiences of attempted Consent for publication Not applicable. suicide by two students who identify as gay [45]. In par- ticular, future qualitive studies on self-harm exploring Competing interests the first-hand accounts of LGBT young people would be The authors declare that they have no competing interests. potentially informative in mapping out future studies and Author details informing prevention and intervention strategies. 1 Department of Psychology, University of Ghana, Legon, P.O. Box LG 84, Accra, The general homophobic environment and social hos- Ghana. 2 School of Psychology, University of Leeds, Leeds, UK. 3 Leeds Institute of Health Sciences, University of Leeds, Leeds, UK. tilities make sexual and gender minorities a hard-to- reach population for research in sub-Saharan Africa. Received: 3 March 2020 Accepted: 25 May 2020 Future research could consider the use of online research techniques which increase access to this young popula- tion and also ensure anonymity of participants [46]. Potentially, school-based and street-based sexual health References education programmes could benefit from evidence of 1. World Health Organization. Preventing suicide, a global imperative. Geneva: WHO; 2014. such future studies to help young people who have sexual 2. World Health Organization. Practice manual for establishing and main‑ orientation or gender identity concerns. taining surveillance systems for suicide attempts and self‑harm. Geneva: WHO; 2016. 3. Naghavi M. Global burden of disease self‑harm collaborators: global, regional, and national burden of suicide mortality 1990 to 2016: Abbreviations Systematic analysis for the Global Burden of Disease Study 2016. BMJ. CI: Confidence interval; DP: Difference in proportions; LGBT: Lesbian, gay, 2019;364:l94. https: //doi.org/10.1136/bmj.l94. bisexual, and transgender; SPSS: Statistical Package for the Social Sciences; 4. King M, Semlyen J, Tai SS, Killaspy H, Osborn D, Popelyuk D, Nazareth I. A WHO: World health Organisation. systematic review of mental disorder, suicide, and deliberate self harm in lesbian, gay and bisexual people. BMC Psychiatry. 2008;8(1):70. https :// Acknowledgements doi.org/10.1186/1471‑244X‑8‑70. We thank all the key stakeholders who granted permissions for this study: 5. Marshal MP, Dietz LJ, Friedman MS, Stall R, Smith HA, McGinley J, Thoma heads of the participating schools, directors and management of the selected BC, Murray PJ, D’Augelli AR, Brent DA. Suicidality and depression dispari‑ charity organisations, the Greater Accra Regional Office of the Ghana Educa‑ ties between sexual minority and heterosexual youth: a meta‑analytic tion Service, the Department of Social Welfare headquarters, Accra, and review. J Adolesc Health. 2011;49(2):115–23. https: //doi.org/10.1016/j. parents/guardians of the underage participants. Also, we thank all the partici‑ jadohe alth. 2011.02.005. pants who contributed the data about their self‑harm for this study. 6. Plöderl M, Tremblay P. Mental health of sexual minorities. A sys‑ tematic review. Int Rev Psychiatry. 2015;27(5):367–85. https ://doi. Authors’ contributions org/10.3109/09540 261.2015.10839 49. EQ, MW and AH contributed to the study concept and design; EQ performed 7. Liu RT, Sheehan AE, Walsh RF, Sanzari CM, Cheek SM, Hernandez EM. statistical analysis of the data and drafted the manuscript, and MW and AH Prevalence and correlates of non‑suicidal self‑injury among lesbian, gay, critiqued the manuscript for important intellectual content. All the authors bisexual, and transgender individuals: a systematic review and meta‑ contributed to the interpretation of results. All authors read and approved the analysis. Clin Psychol Rev. 2019;74:101783. https ://doi.org/10.1016/j. final manuscript. cpr.2019.10178 3. 8. Miranda‑Mendizábal A, Castellví P, Parés‑Badell O, Almenara J, Alonso Funding I, Blasco M, Cebrià A, Gabilondo A, Gili M, Lagares C. Sexual orientation This study was supported by the Leeds International Research Scholarship and suicidal behaviour in adolescents and young adults: systematic (LIRS) at the University of Leeds, School of Psychology, in the form of a doc‑ review and meta‑analysis. Br J Psychiatry. 2017;211(2):77–87. https: //doi. toral scholarship to the first author (EQ)—[Grant No: CFN/sy/200631403]. The org/10.1192/bjp.bp.116.19634 5. funder of the study had no role in study design, data collection, data analysis, 9. Quarshie ENB, Waterman MG, House AO. Self‑harm with suicidal and data interpretation, or writing of the report. The views expressed in this paper non‑suicidal intent in young people in sub‑Saharan Africa: a systematic are those of the authors and not necessarily those of the LIRS. review. BMC Psychiatry. 2020;20(234):1–26. https ://doi.org/10.1186/s1288 8‑020‑02587‑ z. Availability of data and materials 10. Aggarwal S, Patton G, Reavley N, Sreenivasan SA, Berk M. Youth self‑harm The datasets used and/or analysed during the current study are available from in low‑and middle‑income countries: systematic review of the risk and the corresponding author on reasonable request. protective factors. Int J Soc Psychiatry. 2017;63(4):359–75. https ://doi. org/10.1177/00207 64017 70017 5. Ethics approval and consent to participate 11. Lim K‑S, Wong CH, McIntyre RS, Wang J, Zhang Z, Tran BX, Tan W, Ho CS, This study received ethical approval from two Institutional Review Boards Ho RC. Global lifetime and 12‑month prevalence of suicidal behavior, (the School of Psychology Ethics Committee, University of Leeds, UK, [Ref. №: deliberate self‑harm and non‑suicidal self‑injury in children and adoles‑ 16‑0373] and the Ethics Committee for the Humanities, University of Ghana, cents between 1989 and 2018: a meta‑analysis. Int J Environ Res Public Accra, Ghana [Ref. №: ECH078/16‑17]) and institutional permissions were Health. 2019;16(22):4581. https: //doi.org/10.3390/ijerph 16224 581. also obtained to conduct this study. The participants signed a consent form 12. Dilley JA, Simmons KW, Boysun MJ, Pizacani BA, Stark MJ. Demonstrating prior to taking part in the interview. The consent of the parents/guardians the importance and feasibility of including sexual orientation in public of in‑school adolescents aged 13–17 years was sought, while the underage health surveys: health disparities in the Pacific Northwest. Am J Public adolescents assented to participate. Consent to participate in the study was Health. 2010;100(3):460–7. https ://doi.org/10.2105/AJPH.2007.13033 6. sought from the management of charity facilities and street social workers on Quarshie et al. BMC Res Notes (2020) 13:271 Page 6 of 6 13. Quarshie ENB, Shuweihdi F, Waterman MG, House AO. Self‑harm among 32. Adzimah‑Alade M, Akotia CS, Annor F, Quarshie ENB. Vigilantism in Ghana: in‑school and street‑connected adolescents in Ghana: A cross‑sectional Trends, victim characteristics, and reported reasons. Howard J Crime Just. survey in the Greater Accra Region. 2020. Manuscript submitted for 2020;1:1. https ://doi.org/10.1111/hojo.12364 . publication. 33. Taliaferro LA, McMorris BJ, Rider GN, Eisenberg ME. Risk and protective 14. Nock MK, Holmberg EB, Photos VI, Michel BD. Self‑Injurious Thoughts factors for self‑harm in a population‑based sample of transgender youth. and Behaviors Interview: development, reliability, and validity in an Arch Suicide Res. 2019;23(2):203–21. https ://doi.org/10.1080/13811 adolescent sample. Psychol Assess. 2007;19(3):309–17. https ://doi. 118.2018.14306 39. org/10.1037/1040‑3590.19.3.309. 34. Taylor PJ, Dhingra K, Dickson JM, McDermott E. Psychological correlates of 15. Linehan MM, Comtois KA, Brown MZ, Heard HL, Wagner A. suicide self‑harm within gay, lesbian and bisexual UK university students. Archiv attempt self‑injury interview (SASII): development, reliability, and validity Suicide Res. 2018. https: //doi.org/10.1080/138111 18.2018.143063 9. of a scale to assess suicide attempts and intentional self‑injury. Psychol 35. McDermott E, Roen K. Queer youth, suicide and self‑Harm. Troubled Assess. 2006;18(3):303–12. https ://doi.org/10.1037/1040‑3590.18.3.303. subject, troubling norms. London: Palgrave Macmillan; 2016. 16. Owusu A. Global school‑based student health survey (GSHS): Ghana 36. Rimes KA, Shivakumar S, Ussher G, Baker D, Rahman Q, West E. Psycho‑ report. Murfreesboro: Middle Tennessee State University, Ghana Educa‑ social factors associated with suicide attempts, ideation, and future tion Service, & WHO; 2012. risk in lesbian, gay, and bisexual youth: the Youth Chances study. Crisis. 17. Ury HK. Efficiency of case‑control studies with multiple controls per case: 2019;40(2):83–92. https ://doi.org/10.1027/0227‑5910/a0005 27. continuous or dichotomous data. Biometrics. 1975;31(3):643–9. https :// 37. Oginni OA, Mapayi BM, Afolabi OT, Ebuenyi ID, Akinsulore A, Mosaku KS. doi.org/10.2307/25295 48. Association between risky sexual behavior and a psychosocial syndemic 18. Austin PC. Statistical criteria for selecting the optimal number of among Nigerian men who have sex with men. J Gay Lesbian Mental untreated subjects matched to each treated subject when using Health. 2019;23(2):168–85. https: //doi.org/10.1080/193597 05.2018.15526 many‑to‑one matching on the propensity score. Am J Epidemiol. 40. 2010;172(9):1092–7. https ://doi.org/10.1093/aje/kwq22 4. 38. Oginni OA, Mapayi BM, Afolabi OT, Obiajunwa C, Oloniniyi IO. Internalized 19. Rassen JA, Shelat AA, Myers J, Glynn RJ, Rothman KJ, Schneeweiss S. Homophobia, Coping, and Quality of Life Among Nigerian Gay and Bisex‑ One‑to‑many propensity score matching in cohort studies. Pharmacoepi‑ ual Men. J Homosex. 2019. https ://doi.org/10.1080/009183 69.2019.16008 demiol Drug Saf. 2012;21:69–80. https: //doi.org/10.1002/pds.3263. 99. 20. Sullivan GM, Feinn R. Using effect size ‑ or why the p value is not enough. 39. Jobson GA, Theron LB, Kaggwa JK, Kim H‑J. Transgender in Africa: invis‑ J Grad Med Educ. 2012;4(3):279–82. https ://doi.org/10.4300/jgme‑ ible, inaccessible, or ignored? Sahara J. 2012;9(3):160–3. https ://doi. d‑12‑00156 .1. org/10.1080/172903 76.2012.743829 . 21. Greenland S, Senn SJ, Rothman KJ, Carlin JB, Poole C, Goodman SN, 40. Mathy RM. Homosexual related legislation does not reduce suicidal Altman DG. Statistical tests, P values, confidence intervals, and power: a intent in sexual minority groups. BMJ. 2002;325(7373):1176. https ://doi. guide to misinterpretations. Eur J Epidemiol. 2016;31(4):337–50. https :// org/10.1136/bmj.325.7373.1176. doi.org/10.1007/s1065 4‑016‑0149‑3. 41. Poštuvan V, Podlogar T, Šedivy NZ, De Leo D. Suicidal behaviour among 22. Hespanhol L, Vallio CS, Costa LM, Saragiotto BT. Understanding and inter‑ sexual‑minority youth: a review of the role of acceptance and support. preting confidence and credible intervals around effect estimates. Braz J Lancet Child Adoles Health. 2019;3(3):190–8. https ://doi.org/10.1016/ Phys Ther. 2019;23(4):290–301. https ://doi.org/10.1016/j.bjpt.2018.12.006. S2352 ‑4642(18)30400‑ 0. 23. Plöderl M, Wagenmakers E‑J, Tremblay P, Ramsay R, Kralovec K, Fartacek C, 42. McDermott E, Gabb J, Eastham R, Hanbury A. Family trouble: heteronor‑ Fartacek R. Suicide risk and sexual orientation: a critical review. Arch Sex mativity, emotion work and queer youth mental health. Health. 2019. Behav. 2013;42(5):715–27. https ://doi.org/10.1007/s1050 8‑012‑0056‑y. https ://doi.org/10.1177/13634 59319 860572 . 24. Act 29 of Ghana. The criminal offences act, 1960. Accra, Ghana: Assembly 43. Bagley C, D’Augelli AR. Suicidal behaviour in gay, lesbian, and bisexual Press; 1960. youth: it’s an international problem that is associated with homophobic 25. Gyekye K. African cultural values. An introduction. Accra, Ghana: Sankofa legislation. BMJ. 2000;320:1617–8. https: //doi.org/10.2307/25224 814. publishing company; 2003. 44. Quarshie ENB, Cheataa‑Plange HV, Annor F, Asare‑Doku W, Lartey 26. Kaoma K. Christianity, globalization, and protective homophobia. Cham: JKS. Prevalence of suicidal behaviour among nursing and midwifery Springer; 2018. college students in Ghana. Nurs Open. 2019;6(3):897–906. https ://doi. 27. Thoreson R, Cook S, editors. Nowhere to turn: blackmail and extortion org/10.1002/nop2.271. of LGBT people in Sub‑Saharan Africa. Brooklyn: International Gay and 45. Meissner BL, Bantjes J. Disconnection, reconnection and autonomy: four Lesbian Human Rights Commission; 2011. young South African men’s experience of attempting suicide. J Youth 28. Police arrests two girls over lesbianism in Kumasi. https ://www.newsg Stud. 2017;20(7):781–97. https ://doi.org/10.1080/136762 61.2016.12735 hana.com.gh/polic e‑arres ts‑two‑girls ‑over‑lesbi anism ‑in‑kumas i/. 12. 29. Human Rights Watch. “No Choice but to Deny Who I Am”. Violence and 46. McDermott E, Roen K, Piela A. Hard‑to‑reach youth online: methodologi‑ Discrimination against LGBT People in Ghana. New York: Human Rights cal advances in self‑harm research. Sex Res Social Policy. 2013;10(2):125– Watch; 2018. 34. https: //doi.org/10.1007/s1317 8‑012‑0108‑z. 30. Pope John suspends two “gay” students. https ://www.graph ic.com.gh/ news/gener al‑news/pope‑john‑suspe nds‑two‑gay‑stude nts.html. 31. Two Ghana schools expel 53 students for being gay. https ://www.gayst Publisher’s Note arnews .com/articl e/two‑ghana ‑schoo ls‑expel ‑53‑stude nts‑being ‑gay19 Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 0413/. lished maps and institutional affiliations.