Contraceptive Method Uptake at Title X Health Centers in Utah

Abstract

Background:

The Title X family planning program previously supported contraception for Utah clients with low incomes, yet its contributions may not have been sufficient to allow clients to select their preferred methods, including long-acting reversible contraceptives (LARCs). In this study, we compare the contraceptive method choices of self-paying clients with low incomes at three participating Title X health centers in Salt Lake County, UT before and after the removal of additional cost barriers.

Methods:

We used retrospective medical record review to assess clients’ contraceptive choices during two six-month periods: a control period with Title X-assisted sliding scale payment schedules (n=2,776) and an intervention period offering no-cost contraceptive care (n=2,065). We used logistic regression to identify the likelihood of selecting a LARC during the intervention period and multinomial regression to identify the selection probability of different types of available LARCs.

Results:

During the control period, 16% of participants chose a LARC compared to 26% in the intervention period (p=<0.001). During the intervention period, participants were 1.8 times more likely to select LARCs (95% CI 1.65–2.13) compared to non-LARC methods, holding covariates constant. In the multinomial regression, participants were 3 times more likely during the intervention period to select an implant than a pill, patch, or ring, holding all other covariates constant (OR:3.08; 95% CI: 2.47–3.83).

Conclusions:

Title X clients offered contraceptive methods without cost more frequently selected a LARC method. Title X funding reductions may impede individuals’ access to their contraceptive methods of choice.

INTRODUCTION

Despite the increasing popularity of long-acting reversible contraceptive (LARC) methods, inequalities in use and access persist (Goodman, Onwumere, Milam, & Peipert, 2017; Jerman, Frohwirth, Kavanaugh, & Blades, 2017; Mestad et al., 2011; Moniz, Chang, Heisler, & Dalton, 2017; Ricciotti, Dodge, Ramirez, Barnes, & Hacker, 2015; Ricketts, Klingler, & Schwalberg, 2014; Tang, Maurer, & Bartz, 2013). Prohibitive costs continue to prove a barrier, particularly for individuals with low incomes (Eisenberg, McNicholas, & Peipert, 2013; O’Neil-Callahan, Peipert, Zhao, Madden, & Secura, 2013; Ricketts et al., 2014; Thompson, Speidel, Saporta, Waxman, & Harper, 2011). For example, although approximately 12% of U.S. women rely on LARC methods for contraception (Kavanaugh & Jerman, 2018), an estimated 25–29% would select a LARC method in the absence of cost barriers (Foster et al., 2015). In addition, research has acknowledged that the capacity to provide a full range of contraceptive options to clients is an ongoing challenge for clinics (Boudreaux, Xie, Choi, Roby, & Rendall, 2020; Ricketts et al., 2014). While the Patient Protection and Affordable Care Act (or ACA) has improved contraceptive coverage and reduced out of pocket costs for birth control, gaps remain (Blumenthal & Collins, 2014; Castañeda & Mulligan, 2018; Pilliod & Caughey, 2018). Studies have found that many foreign-born clients and those with low incomes still lack insurance and that in some states, such as California, increased coverage has not led to an increase in women’s use of family planning services (Early, Dove, Thiel de Bocanegra, & Schwarz, 2018; Kavanaugh, Zolna, & Burke, 2018). Thus, desired methods of contraception may still be out of reach for many women with low incomes.

Additional research finds that clients with low incomes who use insurance obtained in the wake of the Affordable Care Act may still rely on clinics receiving Title X funding for some family planning services (Kavanaugh et al., 2018). Historically, the Title X federal family planning grant program has provided financial support for clinics to serve women with low incomes. However, the Title X budget has not increased since 2015 (Goldin Evans et al., 2019) and recent restrictions have removed Title X funding from health centers offering abortion referral services (Dawson, 2020; U.S. Department of Health, 2019). Due to recent restrictions, the state of Utah is now without an entity executing the Title X program.

In Utah, women with low incomes face unique challenges to accessing contraceptive care. Utah is a politically conservative state and ranks in the bottom third of U.S. states for health insurance coverage (Berchick, Barnett, & Upton, 2019). At the time of our study, Utah was one of seven states that did not have a Medicaid family planning waiver and did not make a limited expansion of traditional Medicaid until 2019. Absent a family planning waiver, thousands of Utah women are not eligible to receive contraceptive care covered by public funds. Previous research—-including the Contraceptive CHOICE Project in Missouri—has found that cost is a barrier to women seeking contraceptives (Birgisson, Zhao, Secura, Madden, & Peipert, 2015), and Utah women with low incomes may not be able to access their methods of choice when those methods are cost-prohibitive. While LARCs can be cost-prohibitive, other common methods (such as the birth control pill) may also be inaccessible to some self-paying clients due to cost. To help address need and to maximize the number of people served by the Utah Title X program, the bulk of the Title X funds received by participating health centers (namely, those of the Planned Parenthood Association of Utah) supported the allowance of a sliding scale for inexpensive short-acting hormonal methods (pills, ring, injectables). While receiving Title X funds, the Planned Parenthood Association of Utah (PPAU) health centers also tried to meet client demand for LARC methods by offering them on a sliding scale. However, high cost, provider availability, and difficulty keeping LARCs fully in stock likely precluded access for some clients.

In light of the potentially damaging impact of ongoing access challenges when it comes to Utah women’s family planning needs, this study compares, over two periods of time, clients with low incomes at three family planning health centers in Salt Lake County, UT that received Title X funding. We compare self-paying clients’ contraceptive method choices during a control period (operating with Title X contributions) to a study period during which clients were provided with additional assistance, including the removal of costs so that clients could select their contraception of choice without financial barriers. By comparing these self-paying clients, we assess whether Title X financial contributions—which are currently absent in the state—adequately address the contraceptive needs of Utah women with low incomes (including access to their desired methods of choice). Our study is distinctive in that it did not specifically promote LARC usage among participants, and instead focused on the effectiveness of Title X contributions in providing self-paying clients with their methods of choice.

Concentrating on clients at three PPAU health clinics receiving Title X funding, this study compares the clients who received contraceptive care under Title X sliding scale payment schedules to clients receiving care after the study research team provided additional funding to clinics in order to remove all out-of-pocket contraceptive costs and increased clinic capacity to provide clients with their methods of choice. The study team also helped clinics with costly method stocking and provider availability so that clinics could offer clients their desired methods on the day of their appointment. We assess changes in contraceptive method uptake in addition to the varieties of methods chosen by self-paying clients. We hypothesize that even with the most recent levels of Title X support, which previously allowed Utah clinics to offer sliding costs for different methods, not all women with low incomes will have access to their methods of choice. We posit that accessing LARC methods—which are notably cost-prohibitive—could prove a particular challenge in Utah without additional interventions.

METHODS

Study Design

In collaboration, PPAU and the study research team introduced a contraceptive initiative project (funded by the research team) aiming to provide all clients the opportunity to select their preferred methods of contraception, the opportunity to choose LARC methods if desired, and the ability to change methods as often as desired. Although this retrospective review focuses on method uptake at initial visit, the project followed participants for three years (2016–2019) and tracked ongoing contraceptive choices, contraceptive switching, and discontinuation.

The two six-month periods reported in this study consist of a baseline standard of care (or control) period followed by an intervention period. During the intervention period, the research team assisted PPAU by helping to provide all available contraceptive methods at no cost to the clients and by increasing clinic capacity to provide IUDs and implants if clients desired provision the day of their appointment. This included continuous stocking of three LARC products: the copper T380A IUD, the low-cost levonorgestrel 52 mg IUD, and the etonorgestrel contraceptive implant. The intervention period also increased the availability of advanced practice clinicians to insert or remove devices. Between the control and intervention periods, there was no increase in clinical staff—rather, providers from the control period extended their hours during the intervention. For self-paying clients, costs of methods for the control period varied due to the sliding scale. The cheapest and most popular method was birth control pills (with an average cost of $5 per pack), while the average cost of LARC methods ranged on a sliding scale from around $50 to over $200. The implant, which was not covered on the scale, cost around $325. Thus, LARCs were still potentially cost-prohibitive even after Title X contributions.

In an effort to maintain impartiality, the study design, follow up, and analysis was conducted by study staff who were not involved in the clinical care offered at the three study clinics included here. Members of the clinical research staff who enrolled participants and provided contraceptive counseling to clients were PPAU employees. These employees did not participate in data collection, follow up, or analysis aside from handing participants an iPad used to complete their enrollment survey. In preparation for the intervention, all clinic staff learned a new patient-centered contraceptive counseling approach (which emphasized patient needs rather than just method effectiveness) that became the standard of care at all study sites.

The baseline (or control) six-month period was from September 28, 2015 to March 26, 2016. During this time, PPAU clients received client-centered contraceptive counseling and selected their method understanding that they would be responsible for PPAU’s standard charges. This counseling is based on the shared-decision making model and qualitative findings of Planned Parenthood Federation of America research (Dehlendorf, Diedrich, Drey, Postone, & Steinauer, 2010; Dehlendorf, Krajewski, & Borrero, 2014; Makoul & Clayman, 2006). Although they provided approval, the present study is not an official Planned Parenthood study since the organization did not have a role in study design, execution, or analysis. The full intervention period was from March 28, 2016 to March 27, 2017. However, for the purposes of this paper, we have limited the study sample to those who received services in the first six months of the intervention period, as the second half of our intervention period included a media campaign and outreach that may have swayed client demand profiles. We assigned participants to the control or intervention group depending on the time of their first qualifying study visit during the study period. During the intervention period, all new start contraceptive clients (not renewing prescriptions) received the standardized counseling and the ability to obtain their contraceptive method of choice at no cost. Clients who had insurance could bill their insurance and the research project initiative would cover any additional out-of-pocket costs. However, for this analysis we focus on a subset of clients who indicated that they did not have any insurance at the time of their contraceptive visit at the participating clinic and were, thus, eligible for Title X benefits. The Institutional Review Board at the University of Utah approved the study and all participants provided written informed consent for their participation in the project.

Participants

Eligibility requirements included participant age between 16–45 during the study period, being a new client to PPAU or an existing client initiating a new contraceptive method, and an absence of self- or partner-permanent contraception. Lastly, clients had to be self-paying—a proxy measure for having no private insurance or Medicaid coverage. We obtained demographic information along with medical histories (including service descriptions and prescription information) from the Planned Parenthood Associations of Utah’s Electronic Patient Management (EPM) and Electronic Health Records (EHR) systems. Participants came from three clinics located in Salt Lake County. Though all three participating clinics are affiliated with Planned Parenthood, one clinic (“Clinic 2”) is located in a major city center while the others are in neighboring cities in the same county. In addition, as it is located in an area with a higher Hispanic population, Clinic 1 serves more Hispanic-identified clients. Of note, although we use the term “women” intermittently throughout this paper, not all clients seeking reproductive healthcare identify this way. Gender non-conforming individuals and trans men were eligible to participate in the study as well.

Data and Analysis

We used new contraceptive visits as the unit of analysis (excluding prescription refills). We excluded clinic encounters from the sample if the participant was not within the age range or if the encounter did not have a qualifying service description code. We also excluded participants who used insurance (n=4,008). PPAU clinical staff flagged all participants in the medical record system during the study periods and we included clinic encounters in the sample if prospective marking from clinical staff stated qualification. In conjunction with the assigned study status for each patient, billing visit codes identified qualifying clinic visits for the analysis as well as written prescriptions. Use of billing prescription codes to verify eligibility did not permit inclusion of condoms in this analysis, as they are not documented in the medical billing system. We excluded participants selecting internal condoms, cervical caps, diaphragms, fertility awareness-based methods, withdrawal, and those who selected no method, as these methods were either not documented in the electronic medical record or fewer than 15 individuals comprised each of these method selection groups. We also excluded emergency contraception (n=240), as we aimed to capture individuals using a method on an ongoing basis. Figure 1 details a flow diagram for the decisional inclusion of qualifying clinic encounters for statistical analysis.

Figure 1.

Clinic Encounter Inclusion/Exclusion Flow Chart

In the case of participants having multiple contraceptive methods listed at one encounter (e.g., pills and an IUD), we defaulted to the most effective method to produce only one primary method for each clinic encounter. During the control period, the standard practice was to provide a single month of the pill, patch or ring and during the intervention period, clients received three months of supplies at each visit. For participants having multiple clinic encounters during the same period, while receiving the same contraceptive method (i.e., refilling a prescription for the contraceptive pill, patch, ring, or receiving the depot medroxyprogesterone acetate [DMPA] injection), we counted the method only once per period.

To assess potential sociodemographic (e.g., age, race, nulliparity) differences between the control and intervention groups, we conducted chi-square and standard t-tests. To identify differences in method selection (LARC vs. other methods) outcomes between the control and intervention periods, we first identified covariates related to method choice and also those possibly specific to our study, including: age, race, pregnancy history, and geographic location (measured here as Salt Lake County residency and clinic location). Our model-building was limited to variables collected within the electronic medical record. Thus, we did not assess some factors that might impact method choice, such as education level or religious affiliation, as PPAU’s patient management system lacks this information.

We performed unadjusted analyses on each of the covariates and the dependent variable prior to estimating the final model. We then conducted multivariable logistic regression controlling for age, residence within Salt Lake County, race/ethnicity, pregnancy history, and clinic location. We employed robust standard errors to account for the fact that individual clients may have made multiple clinic visits. Finally, we conducted multinomial models assessing likelihood of initiating specific LARC methods (i.e., copper IUD, levonorgestrel IUD, contraceptive implant) between the two study periods. To account for multiple testing, we used the Benjamini Hochberg test. We ran all analyses using Stata version 14 (StataCorp, 2017).

RESULTS

The total sample included 6,021 clinic encounters with 4,841 individual clients. The control period accounted for 2,922 clinic encounters (2,776 individual clients) and the intervention period accounted for 3,099 clinic encounters (2,065 individual clients). Overall, 99% of participants had incomes at or below 100% of the Federal Poverty Level (FPL) according to the electronic records. White, Non-Hispanic women comprised 56% of participants, while 33% identified as Hispanic and 11% identified as belonging to another racial or ethnic group. Further, 79% of the sample reported residing in Salt Lake County. Clinic volume proportions were consistent between the two periods. The highest volume clinic accounted for 49% of all clinic encounters during the two periods. Overall, 52% of the sample was between the ages of 18–24 and 60% were nulliparous.

We did not observe any statistically significant differences for county of residence, previous pregnancies, and clinic location between clients seen during the control and intervention periods (Table 1). However, after the intervention period, we observed a statistically significant change in the age distribution, with a decrease in the percentage of participants under the age of 18 (7% to 4%) and an increase in the percentage of participants between the ages of 18–24 (50% to 54%) (p=<0.001). We also observed a statistically significant change in racial/ethnic distribution between the two periods (p=<0.013), with a decrease in the percentage of Hispanic participants (24% to 31%) and an increase in White, non-Hispanic participants (56% to 57%) as well as those participants belonging to another racial category (10% to 12%).

Table 1.

Demographic Characteristics of Title X Clients During Control and Intervention Periods (N=4,841)

Characteristics	Control Period (N=2,776)		Intervention Period (N=2,065)		Total (N=4,841)		χ2	Pr
	N	%	N	%	N	%
Age(years)
<18	181	7%	75	4%	256	5%	(3)=22.8951	0.001
18–24	1,396	50%	1,119	54%	2,515	52%
25–34	901	32%	643	31%	1,544	32%
35+	298	11%	228	11%	526	11%
Salt Lake County Residency
Yes	2,201	79%	1,613	78%	3,814	79%	(1)=0.9787	0.323
Race/Ethnicity
White, Non-Hispanic	1,554	56%	1,167	57%	2,721	56%
Hispanic	948	34%	647	31%	1,595	33%	(2) = 8.6135	0.013
Other	274	10%	251	12%	525	11%
Ever Been Pregnant
No	1,687	61%	1,212	59%	2,899	60%	(2)=2.1293	0.145
Yes	1,089	39%	853	41%	1,942	40%
Clinic Location
Clinic 1	1,412	48%	1,562	50%	2,974	49%	(2)=3.2927	0.193
Clinic 2	1,003	34%	999	32%	2,002	33%
Clinic 3	507	17%	538	17%	1,045	17%

Contraceptive method distribution between the two study periods is shown in Figure 2. During the control period 16% of the sample chose a LARC method, which increased to 26% in the intervention period (p=<0.001). Of all methods, the implant had the largest proportional increase, from 4% in the control period to 11% after the intervention period. Although pills, patches, and rings remained popular choices in both periods, their combined use decreased from 55% in the control period to 48% in the intervention period (p=<0.001).

Figure 2.

Proportion of Method Selection Among Clients Served at Title X Health Centers During The Project by Control and Intervention Periods

There were statistically significant differences in characteristics of women who selected a LARC method versus those who did not. These results are shown in Table 2. Older age (OR: 1.2; 95% CI 1.0–1.3) and pregnancy history (OR: 1.5; 95% CI 1.3–1.8) were associated with selecting a LARC method; 53% of women selecting a LARC method reported at least one prior pregnancy (p=<0.001). When controlling for covariates (i.e., age, race/ethnicity, pregnancy history, county of residence, and clinic location), the odds of a participant selecting a LARC method was nearly twice as likely during the intervention period compared to the control period (OR:1.86; 95% CI 1.63–2.11).

Table 2.

Logistic Regression Model Predicting Odds of Selecting a LARC Method (N=6,021)

Characteristics	Model 1 Unadjusted	95% CI	Model 2 Adjusted	95% CI
Period
Control	1	1	1	1
Intervention	1.875‡	[1.650–2.132]	1.857‡	[1.632–2.112]
Age
<18	0.452‡	[0.301–0.679]	0.550‡	[0.363–0.831]
18–24	1	1	1	1
25–34	1.398‡	[1.219–1.603]	1.191*	[1.021–1.288]
35+	1.479‡	[1.221–1.804]	1.142	[0.915–1.426]
Race/Ethnicity
Non-Hisp. White	1	1	1	1
Hispanic	1.61*	[1.015–1.327]	0.976	[0.844–1.128]
Other	1.101	[0.895–1.355]	0.998	[0.807–1.235]
Living in Salt Lake County
No	1	1	1	1
Yes	0.848*	[0.729–0.985]	0.834†	[0.715–0.974]
Ever Been Pregnant
No	1	1	1	1
Yes	1.725‡	[1.522–1.956]	1.548‡	[1.327–1.806]
Clinics
Clinic 1	1	1	1	1
Clinic 2	0.904	[0.787–1.038]	1.053	[0.906–1.224]
Clinic 3	0.709‡	[0.519–0.852]	0.788*	[0.651–0.953]

Notes:

^*p<0.05

^†p<0.01

^‡p<0.001

The multinomial regression models (Table 3) use those who chose the pill, patch, or ring as the referent group. We found participants were 3 times more likely to select an implant during the intervention period (OR: 3.08; 95% CI 2.47–3.83). Levonorgestrel IUD selection also increased significantly (OR: 1.5; 95% CI 1.2–1,8), but selection of the copper IUD did not (OR: 1.3; 95% CI 0.99–1.6). Table 3 details adjusted models controlling for all covariates. Women identifying as Hispanic or another (non-White) ethnicity were more likely to select the injectable or implant than the pill/patch/ring. Women in clinic 3 had slightly lower implant provision.

Table 3.

Multinomial Logistic Regression Assessing the Likelihood of Initiating LARC Methods (N=6,021)^*

DMPA Injection	Adjusted Results	95% CI	Copper IUD	Adjusted Results	95% CI
Period			Period
Control	1	1	Control	1	1
Intervention	0.983	[0.871–1.109]	Intervention	1.257	[0.997–1.586]
Age			Age
<18	0.995	[0.758–1.307]	<18	0.637	[0.276–1.473]
18–24	1	1	18–24	1	1
25–34	0.658‡	[0.568–0.764]	25–34	1.683‡	[1.269–2.232]
35+	0.944	[0.757–1.178]	35+	1.655**	[1.121–1.2.443]
Race/Ethnicity			Race/Ethnicity
White, Non-Hispanic	1	1	White, Non-Hispanic	1	1
Hispanic	1.529‡	[1.328–1.761]	Hispanic	1.230	[0.952–1.590]
Other	1.334†	[1.092–1.630]	Other	1.219	[0.831–1.786]
Ever Been Pregnant			Ever Been Pregnant
No	1	1	No	1	1
Yes	1.225‡	[1.055–1.422]	Yes	1.937‡	[1.475–2.545]
Salt Lake County Residency			Salt Lake County Residency
Yes	1.118	[0.958–1.305]	Yes	0.874	[0.659–1.160]
Clinic Location			Clinic Location
Clinic 1	1	1	Clinic 1	1	1
Clinic 2	0.755‡	[0.653–0.873]	Clinic 2	1.124	[0.852–1.483]
Clinic 3	1.011	[0.853–1.197]	Clinic 3	0.842	[0.590–1.202]
Hormonal IUD	Adjusted Results	95% CI	Implant	Adjusted Results	95% CI
Period			Period
Control	1	1	Control	1	1
Intervention	1.509‡	[1.235–1.843]	Intervention	3.077‡	[2.470–3.832]
Age			Age
<18	0.618	[0.337–1.131]	<18	0.451**	[0.226–0.901]
18–24	1	1	18–24	1	1
25–34	1.125	[0.881–1.438]	25–34	0.690†	[0.541–0.880]
35+	1.740†	[1.250–2.423]	35+	0.528†	[0.359–0.777]
Race/Ethnicity			Race/Ethnicity
White, Non-Hispanic	1	1	White, Non-Hispanic	1	1
Hispanic	0.877	[0.692–1.112]	Hispanic	1.397†	[1.120–1.743]
Other	0.736	[0.508–1.065]	Other	1.461**	[1.071–1.994]
Ever Been Pregnant			Ever Been Pregnant
No	1	1	No	1	1
Yes	1.392†	[1.088–1.781]	Yes	1.766‡	[1.387–2.249]
Salt Lake County Residency			Salt Lake County Residency
Yes	0.796	[0.631–1.004]	Yes	0.944	[0.737–1.208]
Clinic Location			Clinic Location
Clinic 1	1	1	Clinic 1	1	1
Clinic 2	0.994	[0.786–1.255]	Clinic 2	0.837	[0.665–1.053]
Clinic 3	1.056	[0.799–1.396]	Clinic 3	0.518‡	[0.372–0.723]

Notes:

^*The pill, patch, or ring users served as the referent group.

^**p<0.05

^†p<0.01

^‡p<0.001

We also conducted a sensitivity analysis for a sub-cohort in order to further test the robustness of our results. In this analysis, we removed all clinic encounters one month before the intervention period and one month after the initiation to account for potential overlap during the transition of the two study periods. Confidence intervals overlapped without significant differences. The similar magnitude and effect of this sensitivity analysis demonstrated the robust nature of the analysis, suggesting little, if any, contamination across the control and intervention periods. This analysis is included as an appendix.

DISCUSSION

In this study at Title X clinics in Utah, LARC method uptake almost doubled with the removal of cost barriers. Similar to previous studies in other contexts (Goldin Evans et al., 2019; Mestad et al., 2011; O’Neil-Callahan et al., 2013; Ricketts et al., 2014), we found that removing cost barriers increased the likelihood that self-paying clients at Utah clinics selected long-acting reversible methods. In the current study, the largest increase occurred with implant use, which went from 4% of women in the control period to 11% in the intervention period. Clients’ chosen methods changed significantly with the removal of cost barriers even in health centers that have long been known to assist clients in navigating financial difficulties and in matching clients with their desired contraception methods. Our findings add insight into the importance of expanding—rather than restricting—programs such as Title X in order to adequately address the contraceptive and family planning needs of clients with low incomes.

Our research also supports previous work in that compared to non-Hispanic white women, women identifying as Hispanic or with another racial/ethnic identity more often selected injections or implant than the pill, patch, or ring (Malat, 2000; Singh et al., 2014). Women in one of the clinics (clinic 3) also had slightly lower implant provision, which we theorize could be due to demographic characteristics, such as an older clientele (Kavanaugh, Jerman, & Finer, 2015). We also observed a lower percentage of LARC uptake with free care than observed in the Contraceptive CHOICE Project (Birgisson et al., 2015). We theorize that this difference may be because during the portion of the project reported here, we did not advertise the offer of free services to women in the community and we did not preferentially encourage LARC use. The insignificant difference with the copper IUD may have occurred due to social perceptions about the copper IUD, such as fears around pain or effectiveness (Heinemann, Reed, Moehner, & Do Minh, 2015; Hubacher, Chen, & Park, 2009; Potter, Rubin, & Sherman, 2014).

We found that more women chose LARCs when cost barriers were removed, but the proportion of women selecting LARC methods was still a minority. Though education and awareness may mediate some of the previously-identified concerns about LARCs –such as pain of insertion or bleeding and other side effects (Akers et al., 2018; Hubacher, Spector, Monteith, Chen, & Hart, 2015)—LARCs are not a one-size-fits-all solution to the needs of all women. Simultaneously, since the proportion of women who did choose to initiate LARC usage increased from 16% to 26%, our findings support previous estimates that approximately one-quarter of women in the United States would use LARC methods if cost barriers were removed (Dehlendorf et al., 2010; Foster et al., 2015).

Limitations

There are several limitations to this study. The observational nature includes a control group and an intervention group where the removal of cost was the only quantifiable change in clinical services. We attempted to control for potential differences between the first period and the second (intervention) period of the study using multivariable regression analysis. Despite our lack of randomized design, our relatively large sample is a notable strength. The work could have been strengthened by more data on the proportion of clients who received same-day LARC insertions between the two periods. Similarly, another limitation is that the provision of demographics for the sample from PPAU’s clinical patient management system lacks information on education level and religious affiliation. Thus, we were not able to control for these factors. A third limitation falls within the qualifying criteria for this study. This project limited participation to women initiating a new contraceptive method, and we excluded women continuing their same method of contraception from this sample. Distribution of method selection may look different if we included women continuing use of the same method in the sample. Lastly, the new contraceptive counseling approach staff took during the intervention could explain some of our results, although we did not measure whether this was the case.

Implications for Policy and/or Practice

This study examines how cost removal relates to the contraceptive choices of clients at Title X clinics, and we found that clients were more likely to select LARC methods that can be cost-prohibitive during our no-cost intervention. These findings suggest policymakers should increase resources to remove contraceptive cost and access difficulties for individuals with low incomes to increase the proportion of clients receiving their desired contraceptive method. Cost and access seemingly continue to be problems for groups with potential coverage gaps, including individuals with low incomes and those who are undocumented (although we cannot speak to the latter here, as we do not collect information on participants’ statuses). Historically, Title X has contributed to health clinics’ ability to offset contraceptive cost and access barriers for women with low incomes, although in Utah, there is currently no Title X grantee as of August 2019.

In Utah, Title X contributions improved the state’s ability to provide contraceptive care to women who lack Medicaid and private insurance coverage. However, in no state in the last decade did Title X provide all women in-need access to any method of contraception. Our research shows that cost barriers inhibit full contraceptive choice and access, and that Title X is underfunded to carry out its goals. If anything, Title X should be expanded rather than restricted. Challenges to access can only worsen when Title X funding to clinics that serve women with low incomes are removed. PPAU, the Utah Title X grantee, returned its unspent portion of the funds in August 2019, and there is no other entity or collection of organizations in the state capable of providing over 50,000 Utah citizens low or no-cost contraception. Going forward, women’s access to the contraceptive methods of their choice (including LARCs) will continue to contribute to health outcomes, including rates of unintended pregnancies and abortions. Comprehensive contraceptive access remains an elusive, but worthwhile, public health pursuit both in Utah and in the United States as a whole. There is an ongoing imperative to expand access to identify the full impact of this goal.

Biography

Corinne Sexsmith, MS, is a Clinical Research Coordinator at the University of Utah with a background in the non-profit sector. Corinne received her Master’s in Human Development and Social Policy and is passionate about implementation research and promoting women’s health in the community.

Jessica Sanders, PhD, MSPH, is an Assistant Professor in Obstetrics and Gynecology at the University of Utah. Using a public health lens, Dr. Sanders conducts and advocates for the use of interdisciplinary academic and clinical research to inform evidence-based policy in Utah and beyond.

Rebecca Simmons, PhD, MPH, is a public health scientist with expertise in evaluating implementation interventions around sexual and reproductive health in both domestic and global settings. She conducts research assessing how contraceptive use and needs change over the life course and fertility-awareness based methods.

Cristen Dalessandro, PhD, is a sociologist and postdoctoral fellow in Family Planning at the University of Utah. Her previous publications and research have focused on the sexual health and intimate relationship lives of young adults.

David Turok, MD, MPH, is an Associate Professor who directs the Division of Family Planning at the University of Utah. He is the principal investigator of several NIH, industry, and foundation grants focusing on overcoming barriers to contraception and developing new contraceptive methods.