Contraception 93 (2016) 266 272

Original research article

The relationship between long-acting reversible contraception and

insurance coverage: a retrospective analysis
Jane Broecker a,, Joan Jurich b , Robin Fuchs a
a
Department of Obstetrics and Gynecology, Ohio University Heritage College of Osteopathic Medicine, Athens, OH, United States
b
Department of Social and Public Health, Ohio University, Athens, OH, United States
Received 16 January 2015; revised 2 November 2015; accepted 9 November 2015

Abstract

Objective: The objective was to determine if there is a relationship between patients financial responsibility (out-of-pocket expenses) and
placement of long-acting, reversible contraceptive (LARC) methods among girls and women living in Appalachia who expressed interest in
LARC device placement.
Study design: A retrospective chart analysis of patients prescribed an intrauterine device (IUD) or an etonogestrel implant between
December 2011 and July 2013 in an Appalachian private practice was performed. Of the 571 identified patients aged 13 to 50, the majority
were Caucasian (98.7%) and using Medicaid (53.2%). Outcomes measured the patients decision regarding whether to use LARC after being
informed of out-of-pocket expenses.
Results: There was a dramatic increase in the proportion of patients who had LARC methods placed if expense was under $200 (pb.001).
Placement rate for privately insured patients was 86.6% for those who paid less than $200 compared to 27.8% for those who paid $200 or
more. Medicaid patients, for whom the device was free, had a 78.0% placement rate. For every additional $100 patients had to pay out of
pocket, the odds of deciding to use the prescribed LARC method decreased.
Conclusions: LARC methods are utilized significantly more often when out-of-pocket cost is low. Cost appears to be a significant barrier to
device placement for the group of privately insured Appalachian patients with out-of-pocket expenses over $200. Despite the improvements
in coverage for many women provided under the Affordable Care Act, cost may remain a barrier for privately insured women who are
required to pay some or all of the cost of LARC methods.
Implications: Unintended pregnancy rates in the United States remain high, especially in Appalachia. One contributing factor is reliance on
user-dependent methods which have significantly high typical use failure rates. Placement of LARC methods for more patients could
decrease unintended pregnancy, but device costs may be one barrier to utilization, even for those with private insurance.
2016 Elsevier Inc. All rights reserved.

Keywords: Long-acting, reversible contraceptive (LARC) methods; Contraception; Appalachia; Unintended pregnancy; Insurance coverage; Affordable Care Act

1. Introduction compliance with the ACAs requirement that contraceptives

Identification and reduction of barriers to effective contra-
ceptive use are essential to the reproductive health of women in
the United States and other countries. Despite implementation of
the Affordable Care Act (ACA) which requires new insurance
plans to provide FDA-approved contraceptives and services
without cost, out-of-pocket cost for contraceptives may be a
barrier for many privately insured and uninsured women.
Privately insured women may have insurance plans in
Corresponding author.
E-mail address: broecker@ohio.edu (J. Broecker).
http://dx.doi.org/10.1016/j.contraception.2015.11.006
0010-7824/ 2016 Elsevier Inc. All rights reserved.
be covered without cost sharing or may be enrolled in plans
which are exempt from complying, either because the plan is
grandfathered or because it meets another exemption. Unfortu-
nately, some health plans simply violate the ACA requirements
and illegally impose costs or restrictions on patients who are
entitled to no-cost contraceptive care [1]. Although the number
of workers in grandfathered plans has fallen from 56% in 2011
to 26% in 2014, the number of women still enrolled in these
plans is significant [2]. Out-of-pocket costs are of particular
concern for long-acting, reversible contraceptive (LARC)
methods, which are significantly more effective than short-
er-acting contraceptives but are more expensive to initiate due to
the high cost of the devices and insertion fees.
 J. Broecker et al. / Contraception 93 (2016) 266272
User-independent methods, such as LARC methods, are
more effective at preventing pregnancy than shorter-acting
user-dependent methods. Higher failure rates have been
demonstrated with typical use of short-acting methods due
to inconsistent or incorrect use, and this is of particular
concern in adolescent patients [3]. In 2009, the American
Congress of Obstetricians and Gynecologists (ACOG)
stated that LARC methods should be offered as first-line
contraceptive methods and encouraged as options for most
women and recommended barriers including high upfront
costs be addressed [4]. While there is momentum to
increase utilization of LARC methods, cost may be a
significant barrier.
An increase in utilization of LARC methods can be
expected if barriers to use were reduced or eliminated. The
Contraceptive CHOICE Project clearly demonstrated that
removal of both financial and knowledge barriers to use of
LARC methods resulted in the choice of a LARC method
by 75% of adolescents and women enrolled in that study
[5]. A study of women enrolled in the Kaiser Foundation
Health Plan in California demonstrated that use of all
forms of contraception increased when that health plan
changed its benefits to offer 100% universal coverage for
the most effective forms of contraception, with the largest
increase in utilization seen with IUDs [6]. Even those
who do not face financial or knowledge barriers may face
other barriers such as transportation, fears about medical
intervention and consequences of using a LARC device
(e.g., fear of needles, pain, irregular periods), or concern
about confidentiality [7,8].
Patients may be deterred from using LARC methods if
they are responsible for some or all of the high initial
expenses associated with purchase and placement of the
devices. One small study showed that out-of-pocket costs
of more than $50 was a significant deterrent to urban
women with private insurance seeking placement of an
IUD [9]. While Ohio Medicaid covers LARC methods at
100%, private insurance coverage varies from 0% to 100%,
depending on the plans benefits. Based on previous
studies, we hypothesized that women requesting LARC
methods would be less likely to follow through for
placement if they were required to pay some or all of
the cost.
2. Materials and methods
2.1. Participants
We performed this retrospective chart analysis at Athens
Medical Associates Obstetrics and Gynecology, where six
providers saw an average of 27,000 gynecologic visits a year.
Data were extracted from the electronic medical record (EMR)
for all patients prescribed either an implant (etonogestrel
subdermal implant) or an IUD (levonorgesterel-releasing
IUD or copper IUD) between December 2011 and July 2013.
A report was generated of all for whom an implant or IUD was
267
prescribed. The patients were covered by Medicaid or private
insurance, and a small number were uninsured. The Institutional
Review Board at Ohio University Heritage College of
Osteopathic Medicine approved this study.
During an office visit, the provider educated the patient
about her contraceptive options, and if she expressed
interest in a LARC, the provider then prescribed the
device electronically. Patients included those who were
sure they wanted a device as well as some who were
considering it along with other options. A precertification
specialist then investigated coverage and out-of-pocket
expense, recorded this information in the EMR, reported
back to the patient and documented this process in the
EMR. The patient either proceeded or did not proceed with
placement of the device based on cost information or other
factors, such as choosing another method after more
consideration.
Data points included the following demographic variables:
age, educational achievement (less than high school, high
school, college, postcollege), race (African American, Asian,
White, Other), ethnicity (Hispanic, non-Hispanic), relationship
status (no steady partner, steady partner, married), gravidity
and live births. Other variables included contraceptive method
being used at time of prescriptive visit (IUD, implant, pill/
injection, condoms, no method), LARC method prescribed
(IUD, implant), patients decision to use prescribed LARC
method (yes, no), parity (nulliparous, parous), coverage status
(no insurance, private insurance or Medicaid) and total
out-of-pocket cost for the device and insertion (if insured
total out-of-pocket cost equaled the deductible plus co-
payment, if uninsured or did not have Medicaid, then total
out-of-pocket cost equaled the total cost of the device and
insertion to be paid by the patient coded in the following
intervals $0, $1 to $99, $100 to $199, $200 to $299, etc.).
Unfortunately, data on personal income were not available;
therefore, we considered utilization of Medicaid to be the next
best surrogate measure for income status.
2.2. Data analysis
We assessed basic demographic variables using descrip-
tive statistics. We performed a series of bivariate logistic
regression analyses with the decision to use the prescribed
LARC method as the dichotomous outcome variable and
patient age, education, relationship status, birth control
method at prescriptive visit, parity and total out-of-pocket
expenses for LARC as independent variables. Independent
variables with a significant bivariate relationship with the
decision to use the prescribed LARC were then forced into a
trimmed model with the decision to use LARC as the
dichotomous dependent variable. As previous research has
found patient age, education, parity and relationship status to
be related to use of LARC methods [1014], we included
these variables as independent variables in the analysis.
Since patients already using LARC at the prescriptive visit
may be more likely to continue its use, we included this
 268 J. Broecker et al. / Contraception 93 (2016) 266272

variable as an independent variable as well. Notably, Close to half (46.4%) of the patients completed college, and
although some studies have found race/ethnicity [11,15] 32.9% completed high school as their highest level of
and insurance status [14,16] to be related to decisions to use education. The vast majority of patients were white (98.7%)
LARC, we did not incorporate these variables into the final and non-Hispanic (99.4%). Most (68.1%) of the patients were
analysis due to limited racial/ethnic diversity in the sample in a steady relationship or married. Gravidity ranged from 0 to
and the close association between out-of-pocket expenses 12 with a mean of 1.4, and the number of live births participants
and whether the patient had Medicaid (with no out-of-pocket had ranged from 0 to 6 with a mean of 1.0 birth. Nulliparous
expenses) and no insurance coverage (with substantial patients accounted for 39.6% of the sample.
out-of-pocket expenses.) Finally, we used 2 to examine As reported in Table 2, nearly half of the patients (47.8%) T2
whether there is a critical dollar amount at which patients were not using a method of birth control at the time of the
are less likely to choose to use the prescribed method. We prescriptive visit. For patients using contraception, birth
used a critical value of .05 to assess statistical significance. control pills/injections were the most common methods
We conducted all analyses with PSAW statistical software used. During this visit, IUDs were prescribed for 60.4% of
version 18.0 [17]. the patients, and implants were prescribed for 39.6% of the
patients. Out-of-pocket expenses for these LARC devices
ranged from zero to the 1200 to 1299 interval, with over half
3. Results of the patients (68.8%) having zero out-of-pocket cost.
3.1. Sample demographics
3.2. Missing data
Of the 710 patients who were prescribed LARC during
the study period, 139 were missing data on one or more of We performed t tests and 2 analyses to determine
the study variables, leaving 571 patients available for whether the 139 patients with missing data on one or more of
T1 analysis. Table 1 presents the demographic profile of the the variables included in the logistic regression were
571 patients in the study. Although ages of the patients systematically different from the 571 patients without
included in the analyses ranged from 13 to 51 years, with a missing data on any of the demographic and study variables.
mean age of 24.3, this was a relatively young sample with The t tests revealed that patients with missing data were
63.0% of all patients being under 25 years of age (n= 360). slightly older (26.1 versus 24.3 years of age, t=2.0, df=171,
p=.05) and had slightly more children (1.3 children versus
Table 1 1.0 child, t=2.30, df=705, p=.02) than patients without
Demographic profile of patients. missing data. The t tests were not significant for gravidity and
Variable Mean (SD) n (%)
Age 24.3 (7.1) Table 2
Education completed Patient profile.
Less than high school 83/571 (14.5%)
Variable n (%)
High school 188/571 (32.9%)
College 265/571 (46.4%) Birth control at time of prescriptive visit
Post college 35/571 (6.1%) IUD/implant 72/571 (12.7%)
Race a Pill/injection 153/571 (26.8%)
African American 5/374 (1.3%) Condoms 73/571 (12.8%)
Asian 0/374 (0.0%) No method 273/571 (47.8%)
White 369/374 (98.7%) LARC prescribed
Other 0/374 (0.0%) IUD 345/571 (60.4%)
Ethnicity b Implant 226/571 (39.6%)
Non-Hispanic 350/352 (99.4%) Out-of-pocket cost
Hispanic .2/352 (6%) $0 393/571 (68.8%)
Relationship status Medicaid 303/393 (77.1%)
No steady partner 182/571 (31.9%) Private insurance 90/393 (22.9%)
Steady partner 241/571 (42.2%) $1 to $99 57/571 (10.0%)
Married 148/571 (25.9%) $100 to $199 26/571 (4.6%)
Gravidity 1.4 (1.5) $200 to $299 18/571 (3.2%)
Live births 1.0 (1.1) $300 to $399 9/571 (1.6%)
Parity $400 to $499 1/571 (2%)
Has no children (nulliparous) 226/571 (39.6%) $500 to $599 2/571 (4%)
Has children (parous) 345/571 (60.4%) $600 to $699 1/571 (2%)
Insurance/Medicaid coverage $700 to $799 3/571 (5%)
Private insurance 262/571 (45.9%) $800 to $899 6/571 (1.1%)
Medicaid 304/571 (53.2%) $900 to $999 0/571 (0.0%)
No insurance coverage 5/571 (.9%) $1000 to $1099 3/571 (5%)
a $1100 to $1199 37/571 (6.5%)
Race was not available in 197 patient charts.
b $1200 to $1299 15/571 (2.6%)
Ethnicity was not available in 219 patient charts.
 J. Broecker et al. / Contraception 93 (2016) 266272
Table 3
Logistic regression predicting decision to use LARC (n=571). a
Variable
Age (years, continuous)
Education completed (bhigh school, high school, college, postcollege, continuous)
Relationship status
No steady partner (reference)
Steady partner
Married
Parity
Nulliparous (reference)
Parous
Birth control at time of prescriptive visit
LARCIUD/rod/implant (reference)
Pill/injection
Condoms
No method
Out-of-pocket cost ($0, $100 intervals, continuous)
a
Trimmed model with forced entry of only those variables having a significant bivariate relationship with decision to use LARC.
p.05.
p.01.
out-of-pocket expenses. The 2 analyses showed that patients
with missing data were more likely to have dropped out of high
school (27.5% versus 14.5%) and less likely to have completed
college (33.0% versus 46.4%; 2= 14.18, df=3, p=.003). In
addition, they were more likely to be married (43.0% versus
25.9%) and less likely to have a steady partner (28.0% versus
42.2%, 2= 13.23, df=2, p=.001) than patients without missing
data. No significant 2 was found for race, ethnicity, parity,
birth control being used at the prescriptive visit and the type of
LARC method prescribed.
3.3. The role of out-of-pocket cost in
womens LARC decisions
We performed a series of bivariate logistic regression
analyses with decision to use LARC methods as the
dichotomous outcome variable (the decision not to use
LARC was coded as 0, and the decision to use LARC was
coded as 1). Independent variables included patient age,
education, relationship status, birth control method at
prescriptive visit, parity and total out-of-pocket expenses
for LARC. Only three independent variables had a
significant bivariate relationship with decision to use
LARC: parity, birth control method at the time of the
prescriptive visit and total out-of-pocket expenses. Bivariate
odds ratios (ORs) and their 95% confidence interval (CIs) are
T3 reported in Table 3.
We then used logistic regression to test a trimmed model
including the three independent variables with a significant
bivariate relationship with decision to use LARC entered
simultaneously (parity, birth control method at the time of
the prescriptive visit and out-of-pocket expense) and
decision to use LARC as the dichotomous dependent
variable. Compared against a constant-only model, the
269
OR for choosing to use LARC
(95% CI)
Bivariate OR Adjusted OR a
1.01 (0.981.03)
0.89 (0.711.12)
1
1.00 (0.651.53)
1.16 (0.711.89)
1
1.24 (1.031.50) X
1 1
0.27 (0.110.63) 0.29 (0.110.74)
0.14 (0.060.34) 0.19 (0.070.53)
0.28 (0.120.64) 0.19 (0.080.48)
0.78 (0.740.83) 0.78 (0.730.83)
trimmed model was statistically significant ( 2= 113.99, df=
5, p=.01). Nagelkerke R 2 indicated that the model accounts for
26% of the variation in the decision to use a LARC method.
Table 3 displays the bivariate and fully adjusted ORs, 95% CIs
and significance values for independent variables found to be
significant in this trimmed model.
Only birth control method at prescriptive visit and total
out-of-pocket expenses were statistically significant. Com-
pared to patients already using a LARC method at the
prescriptive visit, the odds of switching to a LARC method
decreased for patients using birth control pills/injections, for
patients using condoms and for patients not using any
method. For every additional $100 patients had to pay in
out-of-pocket expenses, the decision to use a LARC method
also decreased.
To explore the point at which total out-of-pocket
expenses for LARC methods become prohibitive for
patients, we performed a 2 (decision to use a LARC method)
by 6 (total out-of-pocket expenses) 2 analysis. When using
2, Field [18] recommends that no more than 20% of the
cells have an expected frequency below five. With the total
out-of-pocket expense variable coded in $100 intervals, 50%
(n= 14) of the 28 2 cells had an expected frequency below
five. Therefore, several intervals were collapsed to meet this
assumption. Recoded categories included $0, $1 to $99,
$100 to $199, $200 to $299, $300 to $999, and $1000 to
$1299. With this recoding, only 8.3% (n= 1) of the 12 cells
have an expected frequency below five.
The 2 was significant ( 2= 120.99, df=5, pb.001). There
is a dramatic drop in the percentage of patients who chose to
use LARC methods when the expense was equal to or above
$200. We reran the 2 analysis, eliminating the Medicaid
patients from the analysis. The findings remained significant
( 2= 102.21, df=5, pb.001), with a dramatic drop in the
 270 J. Broecker et al. / Contraception 93 (2016) 266272

Table 4
Percentage of patients deciding to use LARC by total out-of-pocket expense
categories. a
Total out-of-pocket expense category
$0
Medicaid
Private insurance
$1 to $99
$100 to $199
$200 to $299
$300 to $999
$1000 to $1299
a
Collapsed out-of-pocket expenses to six categories to meet assump-
tions of 2 analysis.
percentage of patients who chose to use LARC methods
when out-of-pocket expense was equal to or above $200 still
T4 in evidence. Table 4 reports the percentage of patients within
each out-of-pocket expense category who followed through
with LARC placement as compared to patients who decided
not to follow through.
For those who had private insurance and out-of-pocket
costs equal to or exceeding $200, only 27.8% (n= 25)
followed through with placement. A much higher propor-
tion, 86.6% (n= 149), of those with private insurance and no
cost or cost under $200 followed through with LARC
placement. Patients with Ohio Medicaid desiring a LARC
method followed through with placement 78.2% (n= 237) of
the time.
3.4. Contraceptive outcomes for patients deciding not
to use LARC
Finally, as a follow-up to these analyses, we examined the
contraceptive outcomes for patients who decided not to use
T5 LARC in the spring of 2015. As seen in Table 5, more than
half of the women decided to use hormonal contraception
such as the contraceptive pill, patch or injection (n= 77,
52.0%). Seven patients (4.8%) chose condoms, 6 (4.1%)
chose sterilization, and 27 (18.5%) decided not to use any
particular method. Two patients (1.4%) became pregnant and
28 (19.2%) never returned to the office after the prescriptive
visit in which they expressed interest in a LARC method.
Table 5
Contraceptive outcome for patients deciding not to use LARC after initial
prescription.
Contraceptive outcome
Hormonal contraception (e.g. pill, patch, injection)
Condom
Sterilization
No alternative contraceptive decision made
Pregnant
Did not return after prescriptive visit
4. Discussion
4.1. Primary finding: cost is a barrier to utilization of LARC
n (%)
This study demonstrates that for women interested in
315/393 (80.2%) LARC methods, out-of-pocket cost is a significant barrier to
237/303 (78.2%)
78/90 (86.7%)
patient utilization of the most effective methods of reversible
48/57 (84.2%) contraception: IUDs and implants. If patients have to pay
23/26 (88.5%) $200 or more out of pocket, they are less likely to follow
8/18 (50.0%) through for placement of a device even if their financial
6/22 (27.3%) responsibility is only a small proportion of the actual cost.
11/55 (20.0%)
Out-of-pocket costs primarily came in the form of unmet
deductibles and patient financial responsibility for a certain
percentage of covered services. Differences in insurance
coverage resulted in a variety of out-of-pocket costs for
patients including no cost at all, a small co-pay, responsi-
bility for a percentage of the total costs, and costs up to the
total cost of device and placement, which at the time of data
collection was $1300.
The records reviewed for this study were of patients who
had already expressed interest in LARC methods, and our
data showed that those patients had a fairly high follow
through for placement when cost was not a barrier. These
findings in an Appalachian population have some overlap
with the findings of the Contraceptive CHOICE Project [5],
which included women in an urban environment. This chart
review did reveal fairly high device placement rates (77.9%)
for patients with Medicaid living in Appalachia. This finding
is significant, as patients with Medicaid are more likely to
live in poverty and often face other barriers to placement
such as inadequate transportation and lower level of
education [1922]. In order to improve accessibility and
patient satisfaction, we must work toward same-day
placement whenever possible and continue to identify and
break down barriers to LARC provision so that all women
who desire effective contraception have the opportunity to
choose a LARC method.
Further, it should be noted that although a significant
proportion of the variation (27%) was accounted for by the
model examined in this research, clearly other variables not
examined are contributing to womens LARC decisions. As
noted earlier, barriers such as lack of transportation, fears
about the insertion procedure and side effects, and concern
about confidentiality may also impact womens decision to
use a LARC method [7,8]. Outreach services to rural
locations and patientphysician discussion of womens
concerns about LARC may provide greater access to
contraceptive services and help women to make more
informed choices.
n (%) 4.2. Secondary findings: differences in LARC placement
77/146 (52.0%) vary by current contraceptive method
7/146 (4.8%)
6/146 (4.1%) Significant differences in LARC placement were
27/146 (18.5%) found when we stratified by contraceptive status. The
2/146 (1.4%%) patients who were most likely to follow thorough for
28/146 (19.2%)
LARC device placement were those already using a LARC
 J. Broecker et al. / Contraception 93 (2016) 266272
device, followed by those using pill/injection, with the
lowest follow through rates for those patients using no
method or condoms.
There are many possible explanations for these findings.
If a patient is already using a LARC method and requests
another one, she is likely satisfied with a LARC method and
would be expected to have a high rate of follow through for
placement of a new device. The follow through for LARC
methods was lowest among those at greatest risk for
unplanned pregnancy (those using condoms or no method)
and the explanations for this association may be more
complex. For instance, these women may not have the
financial resources (e.g., income, insurance) for a LARC
device or think that they should start with pills and only
consider a LARC method if pills are not satisfactory. It may
also be possible that women using condoms or no method are
not as highly motivated to prevent unplanned pregnancy as
their pill- or LARC-using counterparts and may be less likely
to follow through with LARC placement. It was interesting
to note that there was no difference in follow through for
placement between parous and nulliparous patients. Explor-
atory analysis adding IUD versus implants to the logistic
regression also showed no significant differences between
IUDs and implants.
4.3. Study strengths
A significant strength of this study is a fairly large sample
size. This study is also noteworthy because it examined patients
in an Appalachian region, a rural population less frequently
studied than urban populations. Moreover, all providers in the
practice were knowledgeable and skilled at placement of LARC
methods; thus, provider ability (or inability) to place LARC was
not a barrier to prescription of a LARC device.
4.4. Study limitations
Because this study was completed in rural Appalachia,
our patients were mostly white, and our findings may not be
generalizable to other populations. The retrospective nature
of this study resulted in a lack of information on womens
annual income, a factor which may have been significant
when assessing for whom cost was a barrier. Further, there
were very few uninsured women in the study. Although
Athens Medical Associates Obstetrics and Gynecology has a
few uninsured women in the practice, these patients rarely
request a LARC device due to the cost. Therefore, they were
not flagged for inclusion in the study. We also lacked
detailed information regarding reasons why women who
requested a LARC method did not ultimately have one
placed, as there may have been reasons other than cost.
Additionally, some patients and providers may have been
aware of which insurance plans covered LARC methods and
which did not, and knowledge of coverage status may have
decreased the likelihood of prescription of LARCs to
patients who knew they had no LARC coverage.
271
With regard to data collection, it is important to note that
patients with missing data on one or more study variables
were significantly different from those with complete data on
study variables. Although there were only small differences
in age and parity, larger differences were found for several
other variables. Patients having missing data were more
likely to have dropped out of high school, less likely to have
completed college, more likely to be married and less likely
to have a steady partner than patients without missing data.
The time frame of this study looked at LARC placement
rates between December 2011 and July 2013, and the ACAs
contraceptive coverage provision took effect for the new
plans starting in August 2012. Thus, some of the patients
included would have been impacted by the ACAs
contraceptive provision. The study was designed to identify
if cost is a barrier, not measure the impact of the ACA, so our
findings are still relevant because it demonstrates the effect
of cost on the decision to utilize LARC methods. Despite the
ACA, cost is still a factor for women who are without
insurance, enrolled in grandfathered or exempted plans, or
enrolled in plans which are illegally violating the contraceptive
coverage provision.
4.5. Conclusions and impact of this study
Although women in new insurance plans should be
guaranteed no-cost contraception under the ACA, it is
coming to light that a significant number of insurance
companies are simply not compliant with some of the ACAs
requirements and are imposing fees or denying coverage
illegally [1]. Grandfathered plans and plans that meet another
exemption are not required to comply with many of the
ACAs requirements. Thus, women who are enrolled in these
plans face variable out-of-pocket costs for contraceptives
because these plans are not required to cover contraception.
The number of women affected by these exemptions is not
precisely known, but it is estimated that 26% of workers
were enrolled in grandfathered plans in 2014 [2]. In fact, the
2013 Kaiser Womens Health Study showed that of women
1544, only 35% reported full coverage for contraceptive
services, while 13% had no coverage and 41% had only
partial coverage for contraception [23]. Fortunately, the
number of grandfathered plans is expected to fall as these
older plans are slowly phased out by employers [2].
The study clearly demonstrates that out-of-pocket cost is a
barrier to placement of a desired LARC device [2]. It is likely
that employers could have an impact on the unintended
pregnancy rate of their employees by ensuring low- or
no-cost provision of LARC methods by offering new
insurance plans which cover contraceptive services at no
cost to the patient and phasing out older grandfathered plans
which are not required to comply with the requirement of
no-cost contraceptive services.
Recently, the cost of LARC devices has been rising as
some manufacturers are raising the prices for their devices
[24]. For uninsured women and for those who are without
272 J. Broecker et al. / Contraception 93 (2016) 266272
contraceptive coverage on their plans, we can direct women
to family planning providers who receive funding through
Title X and encourage patients to apply for reduced cost
devices through patient assistance programs offered by
some manufacturers. LARC methods may decrease in cost
as new companies offer lower-cost IUDs and when generic
devices enter the market.
The momentum for increasing provision of LARC
methods to adolescent and adult patients is strong. The
voices of ACOG and the American Academy of Pediatrics
combined with the comprehensive coverage guaranteed to
many by the ACA will likely result in an increasing number
of patients choosing LARC methods. This study indicates
that out-of-pocket cost is likely a barrier for patients whose
insurance companies still do not cover the total cost of
LARC methods. In the short term, we must continue to help
patients find affordable LARC placement whether through
Title X programs, patient assistance programs, use of new
lower-cost IUDs or others. Long term, we must continue to
work toward lowering the costs of long-acting contraception
and advocate for no-cost comprehensive coverage for all.
Acknowledgments
Special thanks to Karen Elliott, precertification specialist
at Athens Medical Associates Obstetrics and Gynecology,
whose complete and accurate recordkeeping made this study
possible. Both Kelly L. Nottingham, M.P.H., C.H.E.S., and
Jennifer Joan Silk, B.S., from Ohio University Heritage
College of Osteopathic Medicine worked with the authors to
prepare this manuscript for publication.
References
[1] National Womens Law Center. State of birth control coverage: health
plan violations of the affordable care act. http://www.nwlc.org/sites/
default/files/pdfs/stateofbirthcontrol2015final.pdf2015.
[2] The Kaiser Family Foundations and Health Research & Educational
Trust. Employer health benefits 2014 annual survey. http://files.kff.org/
attachment/2014-employer-health-benefits-survey-full-report. 2014, 7.
[3] Raine TR, Foster-Rosales A, Upadhyay UD, Boyer CB, Brown BA,
Sokoloff A, et al. One-year contraceptive continuation and pregnancy
in adolescent girls and women initiating hormonal contraceptives.
Obstet Gynecol 2011;117:36371, http://dx.doi.org/10.1097/
AOG.0b013e31820563d3.
[4] American College of Obstetricians and Gynecologists Committee on
Gynecologic 2009:14348, http://dx.doi.org/10.1097/
AOG.0b013e3181c6f965.
[5] Peipert JF, Madden T, Allsworth JE, Secura GM. Preventing
unintended pregnancies by providing no-cost contraception. Obstet
Gynecol 2012;120:12917.
[6] Postlethwaite D, Trussell J, Zoolakis A, Shabear R, Petitti D. A
comparison of contraceptive procurement pre- and post-benefit
change. Contraception 2007;76:3605, http://dx.doi.org/10.1016/
j.contraception.2007.07.006.
[7] Douthit N, Kiv S, Dwolatzky T, Biswas S. Exposing some important
barriers to health care assess in rural USA. Public Health
2015;129(6):61120.
[8] Bharadwaj P, Akintomide H, Brima N, Copas A, DSouza R.
Determinants of long-acting reversible contraceptive (LARC) use by
adolescent girls and young women. Eur J Contracept Reprod Health
Care 2012;17:298306.
[9] Gariepy AM, Simon EJ, Patel DA, Creinin MD, Schwarz EB.
The impact of out-of-pocket expense on IUD utilization among
women with private insurance. Contraception 2011;84:e3942, http://
dx.doi.org/10.1016/j.contraception.2011.07.002.
[10] Finer LB, Jerman J, Kavanaugh ML. Changes in use of long-acting
contraceptive methods in the United States, 20072009. Fertil Steril
2012;98:8937, http://dx.doi.org/10.1016/j.fertnstert.2012.06.027.
[11] Dempsey AR, Billingsley CC, Savage AH, Korte JE. Predictors of
long-acting reversible contraception use among unmarried young
adults. Am J Obstet Gynecol 2012;206:526.e1-5, http://dx.doi.org/
10.1016/j.ajog.2012.02.014.
[12] Haimovich S. Profile of long-acting reversible contraception users
in Europe. Eur J Contracept Reprod Health Care 2009;14:18795,
http://dx.doi.org/10.1080/13625180902741436.
[13] Kottke M, Goedken P, Gidvani M, Cwiak C. Factors
associated with choosing a long-acting reversible contraceptive
method amongst postpartum women in an urban teaching hospital.
Contraception 2010;82:1978, http://dx.doi.org/10.1016/
j.contraception.2010.04.079.
[14] Moreau, C., Bohet, A., Hassoun, D., Teboul, M. Bajos, N.,
FECOND Working Group. Trends and determinants of use of long-
acting reversible contraception use among young women
in France: results from three national surveys conducted between
2000 and 2010. Fertil Steril 2013;100:4518, http://dx.doi.org/
10.1016/j.fertnstert.2013.04.002.
[15] Werth SR, Securra GM, Broughton HO, Jones ME, Dickey V, Peipert JF.
Contraceptive continuation in Hispanic women. Am J Obstet Gynecol
2015;212:312.e18, http://dx.doi.org/10.1016/j.ajog.2014.09.003.
[16] Potter JE, Hopkins K, Aiken ARA, Hubert C, Stevenson AJ, White K,
et al. Unmet demand for highly effective postpartum contraception in
Texas. Contraception 2014;90:48895, http://dx.doi.org/10.1016/
j.contraception.2014.06.039.
[17] SPSS Inc. Released 2009. PASW Statistics for Windows, Version
18.0. Chicago: SPSS Inc.
[18] Field F. Discovering statistics using SPSS. Thousand Oaks, CA: Sage
Publications; 2005.
[19] Cheung PR, Wiler JL, Lowe RA, Ginde AA. National study of barriers
to timely primary care and emergency department utilization among
Medicaid beneficiaries. Ann Emerg Med 2012;60(1):2-10.e2, http://
dx.doi.org/10.1016/j.annemergmed.2012.01.035.
[20] The Kaiser Commission on Medicaid and the Uninsured. What is
Medicaid's impact on access to care, health outcomes, and quality of
care? http://kaiserfamilyfoundation.files.wordpress.com/2013/08/
8467-what-is-medicaids-impact-on-access-to-care1.pdf. 2013
[Accessed July 23, 2014].
[21] U.S Census Bureau. 2012 Statistical abstract: table 148. Medicaid-
selected characteristics of persons covered: 2009. http://www.census.
gov/compendia/statab/cats/health_nutrition/medicare_medicaid.
html2012 [Accessed July 23, 2014].
[22] McDoom MM, Koppelman E, Drainoni M. Barriers to accessible
health care for Medicaid eligible people with disabilities: a
comparative analysis. J Disabil Policy Stud 2012:10, http://
dx.doi.org/10.1177/1044207312469829.
[23] Salganicoff A, Ranji U, Beamesderfer A, Kurani N. Women
and health care in the early years of the Affordable Care Act
key findings from the 2013 Kaiser Womens Health Survey. https://
kaiserfamilyfoundation.files.wordpress.com/2014/05/8590-women-and-
health-care-in-the-early-years-of-the-affordable-care-act.pdf. 2014.
[24] Bayer. Bayer womens healthcare support fast facts. https://www.
whcsupport.com/documents/8_Fast%20Fact%20Sheet_270-010-
0007-12d.pdf. 2014.