Sources of Contraception Among Adolescents and Young Adults

Abstract

This cross-sectional study describes the nationwide pattern of contraception access by sociodemographic characteristics and health care settings among US youth aged 15 to 24 years.

Contraception is an essential component of comprehensive sexual and reproductive health care for adolescents and young adults (AYAs) because it facilitates individual reproductive goals and bodily autonomy.¹ This population faces unique challenges in accessing contraception at the individual, clinician, health system, and policy levels,² which are exacerbated by long-standing structural inequities.³ Characterizing sources of contraception access among AYAs may reveal opportunities to improve delivery across diverse health care settings. To that end, this cross-sectional study aimed to examine sources of contraception among AYAs in the US and identify associations between these sources and sociodemographic factors.

Methods

We analyzed cross-sectional data from the 2017 to 2019 National Survey of Family Growth (NSFG).⁴ The Yale University Institutional Review Board deemed this cross-sectional study exempt from review and waived informed consent because publicly available, deidentified dataset was used. We followed the STROBE reporting guideline.⁵

The sample included NSFG respondents aged 15 to 24 years who reported receiving a contraception method or prescription in the past 12 months. The outcome of interest was the source of contraception categorized as routine outpatient care (physician offices, health maintenance organization facilities, community or public health clinics), reproductive health safety net (confidential reproductive services outside of primary care, including family planning, Planned Parenthood, school or school-based clinics), and acute care (emergency departments, hospitals, urgent care, and in-store health clinics) settings. Independent variables included age, race and ethnicity, household income, insurance type, and urbanicity. Race and ethnicity were examined to understand the role of well-documented racial and ethnic disparities in reproductive health care access.

Using routine outpatient care setting as the reference category, we developed a multivariable regression model to examine factors associated with reproductive health safety net and acute care settings as sources of contraception. We analyzed estimated probabilities of each source type. Analyses were adjusted for all sociodemographic variables, complex survey design, and sampling weights.

Two-sided P = .05 indicated statistical significance. Analyses were performed between March and July 2023 using Stata 17.0 (StataCorp LLC).

Results

The weighted sample represented 7 505 572 AYAs, of whom 79.3% were aged 18 to 24 years, 63.7% were non-Hispanic White individuals, 39.6% had an income at least 250% of the federal poverty level, 64.0% had private insurance, and 36.6% lived in urban areas (Table 1). Most AYAs (83%) accessed contraception in routine outpatient care settings. In the adjusted model, uninsured AYAs were more likely than privately insured AYAs to access contraception via reproductive health safety net clinics (estimated probability, 23.8% vs 7.6%; adjusted odds ratio [AOR], 3.94; 95% CI, 1.38-11.25) (Table 2). Older vs younger AYAs were more likely to access contraception in reproductive health safety net clinics (estimated probability, 11.4% vs 4.7%; AOR, 2.70; 95% CI, 1.06-6.90). Rural- vs urban-dwelling AYAS were more likely to access contraception from acute care facilities (estimated probability, 12.6% vs 3.6%; AOR, 4.09; 95% CI, 1.03-16.21).

Table 1. Sample Characteristics^a.

Characteristic	Sample, No. (%)
	Unweighted	Weighted
Total No.	650	7 505 572
Age group, y
15-17	146 (22.5)	1 556 081 (20.7)
18-24	504 (77.5)	5 949 492 (79.3)
Race and ethnicityb
Hispanic	150 (23.1)	1 174 075 (15.6)
Non-Hispanic Black	141 (21.7)	970.905 (12.9)
Non-Hispanic White	340 (52.3)	4 784 692 (63.7)
Non-Hispanic other or multiracialc	19 (2.9)	575 901 (7.7)
Household income
≥250% FPL	236 (36.3)	2 968 730 (39.6)
100%-249% FPL	221 (34.0)	2 393 355 (32.9)
0%-99% FPL	193 (29.7)	2 143 488 (28.6)
Insurance type
Private	376 (57.9)	4 804 694 (64.0)
Public	221 (34.0)	2 036 041 (27.1)
Other and no insuranced	53 (8.2)	564 837 (7.5)
Urbanicity
Principal city of MSA or urban	242 (37.2)	2 748 534 (36.6)
Other MSA	298 (45.9)	3 518 765 (46.9)
Not MSA or rural	110 (16.9)	1 238 273 (16.5)

Abbreviations: FPL, federal poverty level; MSA, Metropolitan Statistical Area.

^a The complex sampling design of the National Survey of Family Growth (NSFG) included oversampling of Hispanic and non-Hispanic Black respondents aged 15 to 19 years.

^b Race and ethnicity were self-reported and obtained from the NSFG.

^c Other was not specified.

^d In the NSFG, insurance type consisted of (1) private insurance and Medi-Gap; (2) Medicaid, Children's Health Insurance Program, and state-sponsored health insurance; (3) Medicare, military, and other federally funded health insurance; and (4) single service, Indian Health Service, or not covered. The fourth group represents the other category.

Table 2. Adjusted Odds Ratios (AORs) and Estimated Probabilities of Accessing Contraception at Reproductive Health Safety Net Clinics and Acute Care Facilities vs Routine Outpatient Care Settings^a,b.

Characteristic	Reproductive health safety net clinics			Acute care facilities
	AOR (95% CI)	P value	Estimated probability, %	AOR (95% CI)	P value	Estimated probability, %
Age group, y
15-17	1 [Reference]		4.7	1 [Reference]		7.5
18-24	2.70 (1.06-6.90)	.04	11.4	0.83 (0.29-2.33)	.71	6.0
Race and ethnicity
Hispanic	2.01 (0.82-4.95)	.13	16.0	0.13 (0.01-1.18)	.07	0.7
Non-Hispanic Black	0.82 (0.36-1.89)	.64	6.6	3.05 (0.71-13.1)	.13	13.9
Non-Hispanic White	1 [Reference]		8.5	1 [Reference]		5.1
Other or multiracialc	2.28 (0.63-8.32)	.21	15.0	4.34 (0.87-21.77)	.07	16.9
Household income
≥250% FPL	1 [Reference]		10.0	1 [Reference]		8.6
100%-249% FPL	1.16 (0.51-2.65)	.70	11.7	0.62 (0.13-3.07)	.55	5.6
0%-99% FPL	0.80 (0.35-1.84)	.60	8.5	0.51 (0.10- 2.51)	.40	4.8
Insurance type
Private	1 [Reference]		7.6	1 [Reference]		5.8
Public	1.72 (0.74-3.98)	.20	11.9	1.56 (0.39-6.20)	.52	8.2
Other and no insuranced	3.94 (1.38-11.25)	.01c	23.8	0.85 (0.16-4.48)	.85	4.2
Urbanicity
Principal city of MSA or urban	1 [Reference]		10.0	1 [Reference]		3.6
Other MSA	1.12 (0.55-2.32)	.74	10.8	2.01 (0.62-6.56)	.24	6.8
Not MSA or rural	0.85 (0.30-2.47)	.77	8.0	4.09 (1.03-16.21)	.045c	12.6

Abbreviations: FPL, federal poverty level; MSA, Metropolitan Statistical Area.

^a Adjusted for age, race and ethnicity, household income, insurance type, and urbanicity.

^b Routine outpatient care settings: private physician offices, health maintenance organization facilities, and community or public health clinics.

^c Other was not specified.

^d In the National Survey of Family Growth, insurance type consisted of (1) private insurance and Medi-Gap; (2) Medicaid, Children's Health Insurance Program, and state-sponsored health insurance; (3) Medicare, military, and other federally funded health insurance; and (4) single service, Indian Health Service, or not covered. The fourth group represents the other category.

Discussion

Findings of this study underscore the need to expand contraception access at all health care settings by equipping clinicians to provide high-quality, patient-centered, and evidence-based contraceptive health care. Efforts to bolster Title X funding (which supports many reproductive health safety net clinics) and innovative reproductive health care delivery models, such as telemedicine, mail order and pharmacist-prescribed contraception, or over-the-counter contraceptive pill,⁶ may serve as opportunities to increase contraception access among hard-to-reach populations.

Study limitations included a sample comprising only AYAs with access to contraception, most of whom were aged 18 to 24 years; self-reporting bias; and lack of data on clinician specialty and state-level differences in outcomes. Moreover, the 2017-2019 NSFG dataset did not reflect shifts in reproductive health care access due to the COVID-19 pandemic and changes in federal and state health care legislation over the past 5 years. Nevertheless, the findings revealed distinct patterns in contraceptive access among AYAs and can serve as a baseline for future analyses and the development of evidence-informed policies that protect and broaden contraception access.

Supplement.

Data Sharing Statement