Sexual orientation-related differences in contraceptive use: a brief report based on a cohort of adolescent and young women

Abstract

Objectives:

To examine contraceptive methods used by adolescent/young adult women of diverse sexual orientations.

Study Design:

We collected data from 12,902 females, born 1982–1995, from the longitudinal Growing Up Today Study.

Results:

Compared to heterosexuals, lesbians were half as likely to use contraceptives; other sexual minority subgroups (e.g., bisexuals) were more likely to use any method, particularly long-acting reversible contraceptives.

Conclusions:

Many sexual minority women use contraception throughout adolescence/young adulthood, though use is low among lesbians.

Implications:

With limited contraception use, lesbians miss opportunities for care and need to be brought into the healthcare system in other ways.

1. INTRODUCTION

Compared to heterosexual women, sexual minority women are as likely to have had sexual intercourse with men and are more likely to experience sexually transmitted infections (STI) and pregnancies as teenagers [1–4]. In contrast, sexual minorities are less likely than heterosexuals to receive preventive gynecologic care, in part due to less contraceptive use by certain sexual minority subgroups who thereby miss opportunities for care [5].

Our team recently examined contraceptive methods across sexual orientation groups among adult women [6]. While certain sexual minority subgroups were more likely than heterosexuals to use contraceptive methods, lesbians were less likely to use any method. The current study aims to expand on those findings among adult women by now elucidating contraceptive patterns among adolescent/young adult women. Contraception use early in the lifecourse is critical to addressing STIs and pregnancy during the teen years. Additionally, because contraceptive use during adolescence and young adulthood predicts use into adulthood, this is a critical time to ensure access to care.

2. METHODS

2.1. Study Population: Growing Up Today Study

Participants from the Nurses’ Health Study 2 enrolled their offspring into the Growing Up Today Study (GUTS). Born between 1982–1995, GUTS participants were 9–16 years old at enrollment. We drew on biannual data from enrollment in 1996 through 2013 (at which point the youngest participant was 18 years old); therefore, participants were 9–32 years old throughout follow-up. Of the 15,035 cisgender, female participants enrolled in GUTS, we excluded participants from this analysis who reported their sexual orientation as "not sure" or never reported their sexual orientation (n=2,095); this resulted in a sample size of 12,940 participants.

2.2. Measures

Detailed information about sexual orientation has been collected on every questionnaire since 1999 (Figure 1). The item was adapted from the Minnesota Adolescent Health Survey [7] and we combined data from this item with a question about the sex of sexual partners.

Figure 1.

Key study design features of the Growing Up Today Study (GUTS)

We grouped contraception methods using previously proposed categories most relevant to sexual orientation-related health disparities [6]. These categories include barrier methods, healthcare-based methods (e.g., provider administered injectables/shots and prescription-based oral contraceptives, ring, and patch), LARC methods, and all other methods (i.e., natural family planning, rhythm, foam/jelly/spermicide/sponge, withdrawal, tubal ligation, vasectomy, emergency contraception, other). The primary analysis focused on ever use of each method during the participant’s lifetime. Secondary analyses examined use before age 20 years and current use (i.e., at the final wave of data collection in 2013). In a sensitivity analysis, we grouped contraception methods according to the World Health Organization’s (WHO) tiered effectiveness chart [8].

Potential confounders included age (in years) and race/ethnicity (white, another race/ethnicity). In sensitivity analyses, we further adjusted for sexual behavior by including age at coitarche and number of sexual partners (male and/or female partners).

2.3. Statistical Analysis

We examined the frequency of using each contraceptive method category across sexual orientation groups. Then, we used multivariable log-binomial regression models to calculate adjusted risk ratios (aRR) and 95% confidence intervals (95%CI) of using each contraceptive method category by sexual orientation groups (referent=completely heterosexual with no same-sex partners), adjusting for potential confounders. When models did not converge, we used log-Poisson models [9]. Analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC).

3. RESULTS

During some portion of follow-up, >80% of lesbians did not use any contraceptives (Table 1). Women in all other sexual orientation subgroups (including the reference group of completely heterosexual women with same-sex partners) most commonly used healthcare-based methods compared to other methods. Compared to the reference group of completely heterosexual women with no-same-sex partners, lesbian women were less likely to ever use healthcare-based, barrier, and other contraceptive methods. For instance, lesbians were 60% less likely than these heterosexual peers to ever use barrier methods (aRR [95%CI]: 0.41 [0.30, 0.57]). Other than lesbians, the remaining sexual minority subgroups were more likely than their heterosexual peers to use each contraceptive method. Most notably, sexual minority women in each of those subgroups were more likely than their heterosexual peers to ever use LARCs: completely heterosexual women with same-sex partners (aRR [95%CI]: 2.02 [1.50, 2.71]), mostly heterosexual women (aRR [95%CI]: 2.31 [1.96, 2.73]), and bisexual women (aRR [95%CI]: 2.83 [2.02, 3.98]).

Table 1.

Frequency and probability of using contraceptives among female participants in the U.S.-based Growing Up Today Study (N=12,940).

	Completely Heterosexual with no same-sex partners		Completely Heterosexual with same-sex partners			Mostly Heterosexual			Bisexual			Lesbian
	(n=10,160, 78.5%)		(n=448, 3.5%)			(n=1,841, 14.2%)			(n=285, 2.2%)			(n=206, 1.6%)
	Frequency % (n)1
Age at baseline, Range: 9–16	12.1	(1.9)	12.0	(1.8)		12.2	(1.9)		12.1	(1.8)		12.4	(1.8)
White race/ethnicity	97.2	(9,754)	95.7	(424)		96.4	(1,761)		95.7	(266)		95.6	(195)
Contraceptive methods ever used2
Barrier	35.7	(3,629)	49.8	(223)		45.9	(845)		45.6	(130)		15.1	(31)
Healthcare-based	71.5	(7,267)	92.6	(415)		80.6	(1,483)		76.5	(218)		45.6	(94)
LARC	4.6	(467)	10.9	(49)		11.1	(205)		12.6	(36)		2.4	(5)
Other	13.4	(1,362)	29.2	(131)		23.1	(425)		23.5	(67)		4.4	(9)
None	65.5	(6,657)	65.2	(292)		58.3	(1,073)		58.6	(167)		83.0	(171)
	Risk Ratio3 (95%CI)
Contraceptive methods ever used
Barrier	ref.		1.39	(1.26,	1.53)	1.27	(1.20,	1.34)	1.25	(1.10,	1.42)	0.41	(0.30,	0.57)
Healthcare-based	ref.		1.13	(1.11,	1.15)	1.09	(1.07,	1.12)	1.06	(1.00,	1.12)	0.63	(0.54,	0.73)
LARC	ref.		2.02	(1.50,	2.71)	2.31	(1.96,	2.73)	2.83	(2.02,	3.98)	0.51	(0.21,	1.24)
Other	ref.		2.01	(1.72,	2.34)	1.68	(1.52,	1.85)	1.75	(1.42,	2.17)	0.32	(0.17,	0.60)
None	ref.		1.05	(0.93,	1.18)	0.91	(0.85,	0.97)	0.89	(0.76,	1.04)	1.29	(1.11,	1.50)
Contraceptive methods used <20 years of age
Barrier	ref.		1.97	(1.50,	2.58)	1.41	(1.22,	1.64)	1.70	(1.26,	2.29)	0.68	(0.37,	1.23)
Healthcare-based	ref.		1.64	(1.44,	1.87)	1.24	(1.14,	1.34)	1.25	(1.04,	1.50)	0.94	(0.72,	1.23)
LARC	ref.		N/A			3.79	(1.13,	12.69)	4.15	(0.52,	33.25)	N/A
Other	ref.		2.52	(1.56,	4.08)	1.91	(1.41,	2.60)	2.46	(1.34,	4.54)	0.65	(0.16,	2.61)
None	ref.		0.87	(0.73,	1.05)	0.89	(0.82,	0.97)	0.89	(0.73,	1.08)	1.12	(0.89,	1.40)
Contraceptive methods currently using
Barrier	ref.		1.22	(1.03,	1.45)	1.29	(1.18,	1.41)	1.26	(1.02,	1.54)	0.22	(0.13,	0.39)
Healthcare-based	ref.		1.10	(0.97,	1.08)	1.01	(0.95,	1.08)	0.85	(0.72,	1.01)	0.39	(0.29,	0.52)
LARC	ref.		1.94	(1.41,	2.65)	2.38	(2.01,	2.82)	2.90	(2.05,	4.12)	0.45	(0.17,	1.20)
Other	ref.		1.93	(1.58,	2.36)	1.64	(1.46,	1.85)	1.72	(1.32,	2.24)	0.16	(0.06,	0.44)
None	ref.		0.96	(0.81,	1.14)	0.74	(0.67,	0.81)	0.83	(0.67,	1.02)	2.05	(1.75,	2.41)

¹Growing Up Today Study (GUTS) participants were born 1982–1994 making them 9–16 years at baseline in 1996 and 18–32 years when follow-up for the current analysis ended in 2013; therefore, participants were aged 9–32 years throughout follow-up. Age data presented as mean years (standard deviation) and race/ethnicity as well as contraceptive methods data presented as % (N).

²Barrier methods=male/female condom, diaphragm/cervical cap; Healthcare-based methods=oral contraceptives, shots, ring, patch; long-acting reversible contraceptive (LARC) methods=hormonal/non-hormonal intrauterine device (IUD), implant; Other methods=natural family planning, rhythm, foam/jelly/spermicide/sponge, withdrawal, tubal ligation, vasectomy, emergency contraception, other; None. Frequencies add up to more than 100% due to participants being able to report more than one method.

³Adjusted for age, race/ethnicity, and cohort (i.e., GUTS1 and 2). Multiple imputation was used during analyses for any missing covariate data. Models estimate the likelihood of using each contraceptive method vs all other methods (including no contraceptive use).

Results were similar for contraceptive use before age 20 and current use. Patterns of ever use and current use were consistent but moderately attenuated after adjusting for sexual behavior. However, after adjusting for sexual behavior, risk ratios for contraceptive use before age 20 were almost all completely attenuated (Supplemental Table 1). Results were also consistent when applying WHO effectiveness tiers.

4. DISCUSSION

These data reveal that some sexual minority groups are more likely than heterosexuals to use various contraceptive methods. However, we know from prior research that these sexual minority women still remain at an increased risk of teen pregnancy [2,4] and STIs [3,4], beyond that of their heterosexual peers, meaning there are unmet contraceptive needs. These data also reveal that there is limited contraception use among lesbians which could result in being at risk for teen pregnancy and STIs. This disparity may in part be explained by factors such as childhood maltreatment (e.g., sexual abuse) being more common among sexual minorities [10]. We documented similar findings among adult women [6], but the fact that these patterns originate during adolescence is further evidence of sexual minority women’s unmet needs across the lifecourse.

The racial/ethnic composition of GUTS is primarily white, which limits generalizability. Data were not available on additional covariates, such as health insurance, that may explain (i.e., mediate) differences in contraception use across sexual orientation groups. Additionally, data on sexual behavior were limited to coitarche and number of sexual partners. However, the large sample size allowed us to disaggregate sexual orientation subgroups and explore several contraceptive method categories. Because of the prospective nature of the cohort commencing early in participant’s lives, we were able to quantify contraceptive use in multiple windows of time, including during the teen years.

Healthcare providers must routinely offer contraceptive counseling to all patients, regardless of sexual orientation, and certain sexual minority subgroups (i.e., lesbians) need to be brought into the healthcare system through other avenues than contraceptive care. Additionally, the increased risk of teen pregnancy and STIs among some sexual minorities compared to heterosexuals— despite more contraception use—suggests that increased contraceptive uptake among sexual minority women will not eliminate these disparities. Instead, other potential points for intervention, such as programs that address childhood maltreatment and sex education that is inclusive of all genders and sexual orientations, should be considered in conjunction with increasing access to the full range of sexual and reproductive healthcare.