Medicaid expansion and provision of prescription contraception to Medicaid beneficiaries

Abstract

Introduction:

Medicaid expansion increased access to care, but longitudinal patterns of contraception use after the Medicaid expansion have not been described.

Methods:

We evaluated the effects of Medicaid expansion on the amount and type of contraceptive prescriptions using the Medicaid State Utilization Dataset.

Results:

Overall long-acting reversible contraception (LARC) use increased in both expansion and nonexpansion states. In a difference-in-differences analysis, states that expanded Medicaid had no appreciable increase in per-capita prescription rates of LARC (p = 0.26) or short-acting hormonal contraception (p = 0.09) when compared to nonexpansion states.

Discussion:

The Medicaid expansion was not associated with a change in per-capita LARC or short-acting hormonal contraception use.

1. Introduction

With the passage of the Patient Protection and Affordable Care Act (ACA) in 2010, states were permitted to expand Medicaid eligibility to many additional low-income patients [1]. The ACA specified Medicaid provide a benefits package of preventive services, including all FDA-approved contraceptive medications or devices at no cost to the user [2]. However, not all states expanded Medicaid, leading to variability in expansion of contraceptive coverage to low-income women. Importantly, not all forms of contraception are equally efficacious at preventing pregnancy [3]. The most effective includes long-acting reversible contraception (LARC) and last multiple years, while short-acting hormonal contraception and injectable depot medroxyprogesterone acetate (DMPA) are less effective and provide coverage for shorter durations [4]. The effects of Medicaid expansion and subsequent coverage of contraception on both the amount and type of contraceptive prescription are unknown. Accordingly, we utilized national Medicaid prescription data to determine the effect of Medicaid expansion on the number and type of contraceptive methods prescribed nationally among both Medicaid expansion and non-Medicaid expansion states (Table 1).

Table 1.

Average absolute and per-capita quarterly prescription use among United States Medicaid beneficiaries during pre-Medicaid expansion (2013) and post-Medicaid expansion (2014–2018) periods by state expansion status.

	Expansion states (n = 25)			Nonexpansion states (n = 19)
Absolute quarterly prescriptions	2013	2014–2018	%Change	2013	2014–2018	%Change	Difference in %Change
IUDs	26,452	29,741	12.4%	18,188	17,536	−3.6%	16.0%
Implants	10,174	15,098	48.4%	11,524	15,701	36.3%	12.2%
DMPA (90-day fill)	148,450	165,335	11.4%	121,318	141,544	16.7%	−5.3%
OCP (28-day fill)	1,505,826	1,726,682	14.7%	644,942	650,175	0.8%	13.9%
Patch (28-day fill)	97,988	103,614	5.7%	30,539	31,844	4.3%	1.5%
Rings (28-day fill)	156,019	152,277	−2.4%	51,040	39,001	−23.6%	21.2%
	Expansion states (n = 25)			Nonexpansion states (n = 19)
Quarterly prescriptions per 1000 beneficiaries	2013	2014–2018	%Change	2013	2014–2018	%Change	Difference in %Change
IUDs	0.88	0.81	−8.5%	0.88	0.80	−9.1%	0.6%
Implants	0.34	0.41	7.8%	0.56	0.72	28.4%	−20.7%
DMPA (90-day fill)	4.95	4.49	−9.3%	5.88	6.47	10.0%	−19.3%
OCP (28-day fill)	50.26	46.81	−6.9%	31.26	29.71	−5.0%	−1.9%
Patch (28-day fill)	3.27	2.81	−14.0%	1.48	1.45	−1.7%	−12.3%
Rings (28-day fill)	5.21	4.14	−20.4%	2.47	1.78	−28.9%	7.4%

IUD, intrauterine device; DMPA, depot medroxyprogesterone acetate; OCP, oral contraceptive pill.

2. Material and methods

We analyzed the publicly available 2013 to 2018 Medicaid State Utilization Dataset [5]. This dataset contains state-level information of all outpatient drugs reimbursed partially or fully by state Medicaid agencies. The data are aggregated quarterly, are not linked with patient identifiers, do not contain other payor data, and include both clinic supplied and pharmacy-supplied medications. All formulations of LARC (which include subcutaneous implants and intrauterine devices), DMPA, and short-acting hormonal contraception (which include oral contraceptive pills, transdermal patches, and hormonal vaginal rings) were included (Supplemental Table 1). We extracted product name, number of dispensed units, number of prescriptions, state, quarter, and year for all included formulations between 2013 and 2018. For short-acting hormonal contraception and DMPA injections, we converted prescription duration to standardized 28-day and 90-day fills, respectively. We calculated prescription rate per-1000 Medicaid beneficiaries by obtaining annual state-level estimates of total Medicaid beneficiaries using the Current Population Survey database [6].

We used a quasi-experimental difference-in-differences analysis to compare the effect of Medicaid expansion on absolute and per-capita quarterly prescription averages of LARC, DMPA, and short-acting hormonal contraception among states that expanded Medicaid on January 1, 2014 (n = 24 states + Washington DC) and states that did not expand Medicaid by December 31, 2018 (n = 19 states). The 7 states that expanded Medicaid between January 2, 2014 and December 31, 2018 were excluded. We defined the pre-expansion period from quarter 1, 2013 to quarter 4, 2013 and the postexpansion period as quarter 1, 2014 to quarter 4, 2018. Because pre-expansion curves for per-capita DMPA failed to meet the parallel trends assumption, they were excluded from our difference-in-differences analysis. Analyses were conducted using SASv9.4 (SAS Institute, North Carolina) with p < 0.05 considered significant.

3. Results

Between 2013 and 2018, Medicaid enrollment increased from 30.0 to 37.5 million in expansion states and from 20.6 to 22.0 million in nonexpansion states (Supplemental Figure 1).

3.1. Expansion state contraception use

Absolute number of quarterly prescriptions of LARC (+22.4%), DMPA (+11.4%), and short-acting hormonal contraception (+12.7%) increased between pre-expansion and postexpansion periods. Postexpansion per-capita quarterly prescriptions decreased for LARC (−0.4%; 1.22 to 1.22 prescriptions/1000 beneficiaries), DMPA (−9.3%; 4.95 to 4.49 90-day prescriptions/1000 beneficiaries), and short-acting hormonal contraception (−8.5%; 58.74 to 53.76 28-day prescriptions/1000 beneficiaries).

3.2. Nonexpansion state contraception use

Among nonexpansion states, overall use of LARC, DMPA, and short-acting hormonal contraception changed by +11.9%, +16.7%, and −0.8% between periods, respectively. Postexpansion per-capita quarterly prescription changes were variable between categories, with an increase for LARC (+5.4%; 1.44 to 1.52 prescriptions/1000 beneficiaries), an increase for DMPA (+10.0%; 5.88 to 6.47 90-day prescriptions/1000 beneficiaries), and a decrease for short-acting hormonal contraception (−6.4%; 35.2 to 33.0 28-day prescriptions/1000 beneficiaries).

3.3. Difference-in-differences analysis

States with Medicaid expansion were associated with no appreciable absolute change in LARC (+4,688 prescriptions [−28.8 to 9,406], p = 0.051) or DMPA (−3,341 prescriptions [95% CI: −27,672 to 20,989], p = 0.79) utilization, but were associated with faster absolute growth of short-acting hormonal contraception (+228,240 prescriptions [95% CI: 82,902 to 373,579], p < 0.01) (Fig. 1). However, when analyzed per-1000 Medicaid beneficiaries, the Medicaid expansion was not associated with increases in either LARC (−0.08 prescriptions/1000 beneficiaries [95% CI: −0.23 to 0.06], p = 0.26) or short-acting hormonal contraception (−2.71 prescriptions/1000 beneficiaries [95% CI: −5.85 to 0.43], p = 0.09).

Fig. 1. Trends in per-capita and absolute contraception use before and after the 2014 Medicaid expansion.

*Long acting reversible contraception includes intrauterine and implantable devices. Short-term hormonals include oral contraceptive pills, vaginal rings, and transdermal patches. DMPA = depot medroxyprogesterone acetate

**Dotted line represents Medicaid expansion, which occurred quarter 1, 2014. The pre-expansion period refers to 4 quarters before quarter 1, 2014, and the postexpansion period refers to 20 quarters after (and including) quarter 1, 2014.

4. Discussion

We report that 5 years after the Medicaid expansion, expansion states observed increased absolute prescription averages of all contraceptive methods while nonexpansion states had variable changes in contraception prescriptions. When analyzed per-1000 Medicaid beneficiaries, the Medicaid expansion was not associated with relative increases in LARC or short-acting hormonal contraception use.

The association of the Medicaid expansion with improved access to care, preventive services, and medications has been previously reported [7]. Prior evaluations of the impact of Medicaid contraception have been mixed. In a study of contraceptive care in community health centers, Darney et al. [8] describe a significant association between Medicaid expansion and LARC use, while others have suggested no observed effect [9]. This study adds to the literature by comparing absolute and per-capita rates of LARC and short-acting hormonal contraception prescriptions through a difference-in-differences analysis between states that expanded Medicaid and states that did not. For LARC, states with Medicaid expansion were not associated with increases in either absolute or per-capita prescription rates of LARC. However, for short-acting hormonal contraception, the Medicaid expansion was associated with an absolute increase in the number of prescriptions, but this association was no longer observed in our per-capita analysis. These discordant results between absolute and per-capita analysis may be explained by (1) greater increases in Medicaid enrollees than prescriptions and/or (2) lower per-capita short-acting hormonal contraception utilization rates among newly eligible women.

It is encouraging to note that quarterly prescriptions of most forms of contraception increased in the postexpansion period, irrespective of Medicaid expansion status. Furthermore, the greatest increases in quarterly prescriptions occurred among more effective forms of contraception, with relative increase for implants (+36.3 to +48.4%) and DMPA (+11.4 to +16.7%). These findings may be an early signal of a shift toward more effective forms of contraception.

This study is limited as the public dataset does not provide population demographics nor stratify number of contraception prescriptions in newly qualified Medicaid enrollees compared to previous Medicaid enrollees. Our study, however, uses a national database which may increase generalizability and describes the landscape since Medicaid expansion. Additionally, we are unable to determine pre-expansion use patterns of contraception among newly eligible Medicaid enrollees. Most newly eligible enrollees, however, were previously uninsured and likely gained access to contraception following Medicaid expansion.

In summary, the Medicaid expansion was not associated with changes in the per-capita prescription rates of LARC and short-acting hormonal contraception but was associated with increased total prescriptions of short-acting hormonal contraception. Future studies should evaluate the downstream impact of increased contraception access, including changes in rates of unintended pregnancy, abortion, and adoption.