Evaluation of the Delaware Contraceptive Access Now (DelCAN) initiative: A qualitative analysis of site leaders’ implementation recommendations

Abstract

Objectives:

In 2014, Delaware launched a statewide initiative to reduce the rate of unintended pregnancies and increase access to contraception services. Our study objective was to understand the implementation experiences, barriers, and successes across health care practice settings and to provide recommendations for future, similar initiatives.

Study Design:

As part of a larger multi-component process evaluation, we conducted semi-structured interviews with 32 leaders from 26 practice settings implementing the initiative across the state. We analyzed the qualitative data through iterative open, axial, and selective coding using grounded theory methods, employing thematic analysis to identify common themes in implementation experiences.

Results:

Most practices perceived that patient demand for methods of long-acting reversible contraception (LARC) increased. Many practices had to adapt the intervention to fit the needs and constraints of their settings and patient populations. Primary care practices, smaller practices, and practices that served large numbers of adolescents experienced more barriers compared to obstetrics and gynecology or women’s health practices. For current and future iterations of the initiative, leaders emphasized: (1) the need for greater implementation flexibility, (2) the importance of inclusive communication at multiple levels, and (3) attending to logistical challenges, particularly around billing.

Conclusion:

Varied practice settings required significant flexibility and responsiveness to context for implementation of the initiative. Organizations with greater pre-existing capacity were able to offer the full range of contraceptive care, as the initiative intended, in contrast to practices with less pre-existing capacity for providing methods of LARC and other types of contraception.

1. Introduction

Access to contraceptive care remains a challenge for some women in the US, as evidenced by high rates of unintended pregnancy [1] and unmet need for contraception [2]. State- and city-wide initiatives in Colorado [3], Iowa [4], Utah [5], and Missouri [6] have been successful in reducing unintended pregnancy by expanding access to contraceptive care, including methods of long-acting reversible contraception (LARC). However, uptake of methods of LARC has been slow, largely due to barriers such as cost [7], need for multiple appointments [8], lack of accurate patient knowledge [9–11], lack of provider knowledge about updated recommendations [12], and limited numbers of medical providers trained in their insertion and removal [13,14]. The initiatives aimed at decreasing such barriers reported general success in reducing pregnancy, birth, and abortion rates among adults and adolescents, and increased uptake and continued use of methods of LARC [15].

In the early 2010s, Delaware had the highest rate of unintended pregnancy in the nation [16,17]. In 2014, then-Governor Jack Markell launched the Delaware Contraceptive Access Now (DelCAN) statewide initiative to reduce the rate of unintended pregnancies and increase access to contraceptive services for all women of reproductive age. The program focused on: (a) policy change and implementation at the state level; (b) a statewide public awareness campaign; and (c) statewide clinician and staff trainings. The full initiative has been described elsewhere [18].

Central to the DelCAN initiative was the preparation of primary care providers across the state to offer same-day access to all contraceptive methods, including methods of LARC, at negligible or no cost to women (“All Methods Free”). As part of a public-private partnership, Upstream USA (Upstream), a company that provides health centers with training and technical assistance to eliminate barriers to offering the full range of contraception, trained and provided technical assistance to health centers and medical practices that provided general primary care, pediatric care, and women’s health or obstetrics and gynecology (Ob/Gyn) care. Between 2016 and 2019, Upstream trained healthcare providers and support staff on clinical, counseling, and administrative processes necessary to deliver comprehensive contraceptive services, including asking women of reproductive age about their pregnancy intention at every primary care visit (also known as the One Key Question®: “Would you like to become pregnant in the next year?” [19]), and offering contraceptive counseling, insertion, removal, and management of methods of LARC and other contraceptive methods [18].

An independent evaluation of this initiative, assessing effects of the initiative on statewide rates of unintended pregnancy and related outcomes, is currently ongoing. However, in order to fully understand outcomes in Delaware, this component of a larger mixed-methods process evaluation of DelCAN, seeks to understand the initial facilitators and barriers of the initiative’s implementation as it relates to the goal of increasing same-day contraception provision across primary care and other practice settings, and asking patients about their pregnancy intention.

2. Materials and Methods

Using a semi-structured interview guide, we conducted 26 in-depth interviews with leaders implementing the DelCAN initiative in their practices.

2.1. Study sample

Utilizing a purposive sampling strategy, we recruited 32 leaders from 26 health care settings participating in the DelCAN initiative. At each trained practice site, the eligible participant was the site champion, or a person the site identified as a leader who could speak about the implementation of the intervention. All participants were over 18 years old and provided informed consent.

We recruited participants from May through December 2018, with the goal of obtaining a sample that captured variation by practice types, size, location, and population served. We developed a purposive sample to ensure variability across the range of aforementioned practice characteristics, and we achieved saturation of this variability after interviewing 32 leaders working at 26 sites (out of a total of 32 sites participating in the initiative).

2. Interviews

The research team consisted of two doctoral students with public health, reproductive health, and qualitative research training expertise; an experienced family planning and adolescent health researcher; and a qualitative methodologist. Team members developed a semi-structured interview guide based on a review of the literature on contraceptive care and contraceptive access interventions. The two faculty researchers, A.L. and K.R., conducted initial pilot interviews in person with leaders at Delaware sites and supplemented the interview guide with common follow-up questions, following a debriefing session with the whole team.

We conducted all interviews in English, between June and December 2018. In thirteen interviews (50%), two team members interviewed site leaders in person in clinic office space. One or two team members conducted the remaining half of the interviews by phone. In 20 interviews, we interviewed one participant; in six interviews we interviewed two participants, totaling 32 participants from 26 group sites. Interview length ranged from 45 to 90 minutes. We audio recorded the interviews and advised participants not to use individual or practice names during the interview.

2.3. Data Analysis

The team’s two doctoral students transcribed the audio files verbatim, omitting any identifying information, and imported them into the Dedoose qualitative software program for coding and further analyses. The research team had regular team meetings throughout the interview and data analysis process, where we discussed common themes. We were confident we had reached thematic saturation across the full range of variation in clinic settings after 24 interviews since new observations of various practice settings revealed no new themes [20]. We conducted two additional interviews as confirmation.

The qualitative approach for a process evaluation has frequently been used to give insight into context, process, and meaning [21]. We use a grounded theory approach (GTA), which shaped data collection, protocol development, and data analyses. GTA is a rigorous qualitative approach developed within social sciences, emphasizing iterative comparison and contrast, and a tight relationship fitting data and theory [22]. In particular, GTA requires multiple waves of coding, including open coding, axial coding, and selective coding.

Open coding begins with identification of sensitizing concepts [23], or a priori codes, derived from existing literature on contraceptive access initiatives. GTA goes further, to also identify open codes that emerge from initial reading of textual data – and that are unique to each context observed. We chose to integrate emergent codes from four initial interviews to develop a codebook. We applied these codes to each of the 26 interviews in a wave of open coding. All four team members coded interviews; two researchers coded each interview, and subsequently applied rigorous consideration of interpretations and resolution of divergent coder perspectives.

We then conducted a wave of axial coding, by comparing and contrasting the perspectives of leaders, within and across sites, on specific codes related to recommendations for the intervention. For example, we examined the range of perspectives on what we coded as “preceptorship,” which included challenges to finding supervision opportunities for providers to practice insertion of methods of LARC with qualified physicians in order to become authorized to independently perform this clinical skill. For this analysis, we paid specific attention to differences by type of practice. In the final wave of selective coding, we condensed a diverse set of codes to develop a narrative of implementation recommendations.

We utilized a range of techniques to enhance data quality and qualitative integrity of the study [21]. With multiple sources and sites for interviews, we employed triangulation of the process of implementation. We checked back with leaders to clarify comments and to gather member checks. Finally, we used extensive peer debriefing methods, confirming themes with research team members in the larger study as well as with each other. These discussions allowed for consideration of insights and blind spots, relative to our training and background.

The Institutional Review Board of the University of Maryland, College Park approved this study.

3. Results

We conducted the majority (65%, n=17) of interviews with practices located in the northern region of the state, including urban and suburban sites, and another nine interviews (35% of sample) with champions from practices in the southern, more rural communities in the state. Just over one-quarter (27%, n=7) of these practices were general primary care health clinics; another 42% (n=11) were women’s health or Ob/Gyn practices, 23% (n=6) were pediatric or adolescent medicine practices, and 7% (n=2) were sites that focused primarily on behavioral health care. All but one participant were women, and all worked in administrative roles at practices participating in the DelCAN initiative. In addition to their DelCAN leadership role (“champion”), the participants were office managers, nurses, and upper management, some of whom were also physicians.

We synthesized our discussions with leaders about facilitators and barriers to program implementation in order to identify three key themes from the data that captured participants’ recommendations for implementation of this and other contraceptive access initiatives: (1) the need for early tailoring and implementation flexibility, (2) the importance of inclusive communication at multiple levels, and (3) attending to logistical challenges, particularly around billing.

3.1. Early tailoring of the initiative to the site’s capacities and priorities

All leaders, and most of their staff, were enthusiastic about the initiative and felt they benefitted from the Upstream training, which was the “kick off” of the initiative at most practices. However, they also described facing numerous difficulties in implementing elements of the initiative following the training. These difficulties included challenges around preceptorship for providers, billing complications, and changes to practice workflow to accommodate the goal of same-day provision of methods of LARC. Leaders, particularly those in sites that were not exclusively focused on women’s health, experienced frustration with what felt like a one-size-fits-all initiative that did not easily work across a wide range of clinical settings and social contexts. They suggested that identifying individual practice strengths, needs, and capacities prior to the training would have allowed tailoring of the initiative to better fit each site. As one participant in upper management said, “The success of something like this depends on the nimbleness of the actual organization.”

Many leaders, although enthusiastic and supportive of the goals of the initiative, and grateful for Upstream’s support, found some of Upstream’s requirements and milestones to be unattainable. Participants spoke about the different challenges they faced, such as struggling to meet the reporting needs, to maintain methods of LARC in stock, or to adapt their scheduling enough to enable same-day insertions of methods of LARC. A manager from a pediatric practice explained:

[Upstream] gave us milestones like, ‘You have to do these things and meet these milestones,’ and we’re like, ‘Okay, well, we’re not going to meet all three of those. We’re going meet these two because that’s what makes sense, or that’s what’s practical for us.’ [It was a] square peg in a round hole kind of thing. That’s what they had done in Ob/Gyn practices, but it’s a little different, making it work for pediatrics.

For example, one requirement was training for all clinicians to offer both IUDs (intrauterine devices) and implants. However, this was unattainable for all sites, particularly in pediatric settings, where office infrastructure, such as a lack of gynecological tables, made IUD insertions unfeasible.

Other leaders described challenges based on some clinicians’ concerns about inserting methods of LARC on the same day of contraceptive counseling, or their discomfort in engaging in conversations about sexuality with teenage patients. One leader suggested assessing provider attitudes with internal provider agreements prior to training: “[If providers] sign off on it, then they have an understanding of what they’re going to be expected to do[; …] there’s a meeting of the minds.” She suggested that such provider agreements would serve to jumpstart the conversation about the initiative’s expectations of clinicians and give the practice an overview of their provider capacities and comfort levels.

3.2. Communication at multiple levels

3.2.1. Communication between Upstream and site staff

Leaders praised the practice-wide Upstream training as engaging and instructive. However, many felt they did not receive enough information ahead of time about DelCAN, and what the initiative would mean for their daily site operations. An office manager at an Ob/Gyn site explained, “I would say at least 50% of us were stunned. Probably halfway through [the training, we were] like, ‘Wow! A lot of information.’ I think when we started doing the role play, we realized, ‘Oh, we have to do this in the office.’” Participants wanted to receive specific information prior to the training on how DelCAN would affect each role in their practice (e.g., front desk, clinical support staff, medical records, quality, invoicing). They reported that this would have increased buy-in and focus during the training.

Many leaders indicated that DelCAN is only one of many initiatives active at their sites. They had to juggle multiple priorities and they wished Upstream understood the extent of their reporting burden. A leader of a federally qualified health center, for example, suggested that it would have been helpful “if [Upstream] came in the door, they understood what our reporting [requirements] were, right off the bat [… and had] a working experience and knowledge of federally qualified health centers.” The importance of considering the context of each site was a key reason why leaders wanted earlier communication with Upstream.

3.2.2. Communication with staff across all roles

Participants reported that having staff from all management levels and roles within a practice involved from the start of the initiative enhanced implementation. Even though clinicians delivered the contraceptive care, the administrative team adapted the workflow and did the billing and stocking to make it possible. One high-level manager in a general primary care setting emphasized:

We need to have that representation at the table from the very beginning. […] Physicians think differently because they’re thinking clinical care and ‘How am I going to take care of the patient?’ The operations folks think, ‘How are we going to do this?’

In other words, leaders have multiple responsibilities and need to attend to all of the support services that enable optimal patient care. The ongoing involvement and input of staff from all levels is needed for a practice team to work together effectively. One leader suggested that bigger practices with multiple sites might benefit from having multiple DelCAN champions, drawn from different provider and administrative levels.

3.2.3. Cultural humility in communication with patients

Leaders reported a strong commitment to the goal of DelCAN–to provide access to contraceptive care to all women and girls in the state. However, they felt strongly about the need to improve communication, particularly with specific patient populations. The notion of cultural humility emerged frequently in interviews, both implicitly and explicitly. For instance, leaders emphasized the need to have information and consent forms available in different languages, which was not the case at the start of the initiative. They also stressed the importance of avoiding value judgments about which women need methods of LARC. Some of them were very cognizant about stereotypes, including one leader who asserted:

I think the biggest issue was the kind of cultural insensitivity. […] No one wants to see that broad-brush approach of saying that every poor woman shouldn’t have a baby. […] So I think you have to do something to almost address that thought upfront.

Pre-existing biases, and the potential for the intervention to reinforce them, was clearly a difficult topic of conversation. While some leaders emphasized the need to tackle the issue head on, a minority of participants used language that perpetuated stereotypes, particularly about adolescents (“frequent flyers,” “for others that can’t control themselves”). Each site addressed (or not) the issues around cultural humility separately, with variability among and within practices.

Leaders also cited confidentiality and inability to pay as barriers that might have prevented some women from trusting the initiative. Adolescents, in particular, may have worried about confidentiality. Women from economically disadvantaged communities may not have been reached by the universal statewide public awareness campaign, and thus, might not have known that participating sites near them were providing free contraception and transportation. Women’s inability to pay may have prevented them from seeking care and, consequently, from learning that the service was free once at the participating site. An office manager at a women’s health clinic explained that, in hindsight, she would have tailored outreach and created specific appointments for women who are uninsured to facilitate their access to the free service. She said:

They are a very fragile population. And they’re used to going into an office and being charged or being afraid to even go to a walk-in or another doctor because they’ll be charged. So there is a population that you definitely want to make a very strong relationship with. So I wish that we had made these appointments different[ly].

Leaders at sites serving adolescent patients reported concerns about DelCAN’s appropriateness for teenagers. Even providers who felt comfortable discussing sexual health with their adolescent patients, did not feel that the pregnancy intention screening question was appropriate for teenagers. Furthermore, the absence of condoms and sexually transmitted infections from DelCAN training, funding, and mission concerned both providers and parents, according to the leaders.

3.3. User-friendly logistics

Each leader had at least one story of an unexpected logistical issue that caused implementation hiccups for months. The most common issue was lack of clarity on processing payments and billing for services and devices under the purview of DelCAN. This was a concern for office staff and high-level managers, who worried about double billing either patients or insurance companies. Similarly, many participants struggled with the electronic medical record (EMR) system and its challenges in efficiently meeting DelCAN’s data reporting requirements in relation to state law and the additional burden on staff. Some, particularly those from smaller practices, also described challenges with keeping methods of LARC in stock, and losing money they had to lay out for devices when the patient changed her mind or did not come in for her appointment. While staff were generally enthusiastic post-training about providing more comprehensive contraceptive care, the challenges of ramping up the initiative took considerable time to solve, and many leaders described that enthusiasm waned as they worked out these logistical steps.

4. Discussion

Participants across sites reported being able to successfully implement some aspects of the initiative, including the incorporation of the pregnancy intention question and increased contraceptive counseling. However, other elements of the initiative were more challenging to implement, including same-day insertions of methods of LARC, the timely precepting of providers trained on new insertion methods, and having medical assistants provide contraceptive counseling. Although all leaders appreciated the ongoing technical assistance from Upstream that followed the training, many perceived it as support to make their practice fit the requirements of the initiative, rather than making the initiative more flexibly fit the practice.

Based on these findings and leaders’ suggestions, we recommend that future contraceptive access initiatives begin with a comprehensive pre-implementation assessment at every practice site. Such assessments have been widely discussed and utilized as standard practice in the field of implementation science [24]. Assessments should precede any training and should include gathering input from participants across all roles in a medical practice (e.g., health care providers, medical assistants, office staff, billing department). They begin with an open and thorough understanding of each practice’s needs, capacities, patients, and priorities, and lead to a dialogue that combines the expertise of the intervention staff about best practices in increasing contraceptive access, with the expertise of the practice staff about the needs of their patients and the capacities of their practice. This dialogue would result in a more tailored implementation of the initiative that maintains its core components and mission, while being more responsive to the needs of each practice.

Such tailoring of the intervention happened to some extent by default, as practices found aspects of implementation to be unfeasible or unresponsive to their patients. But these adjustments often created tension, frustration, and extra work as site leaders failed at stated targets, made mid-course corrections, and had to figure out their own version of the intervention. A comprehensive and collaborative pre-implementation assessment, and joint adaptations of the intervention, would enable sites to have greater agency and responsiveness in their efforts to provide more contraceptive care to their patients.

In addition to the initial assessment and tailoring of the intervention, a pre-implementation phase should include more of the logistical work around billing, stock, and workflow planning. Putting more of these components in place prior to training and having more initial conversations about the initiative with practice providers and staff across roles would enable a smoother transition from training to implementation, and greater buy-in across the practice.

Site leaders echoed many of the facilitators, barriers, and recommendations described in other implementation studies of contraceptive care [25–27]. In particular, studies of a hospital-based initiative aiming to enable the postpartum insertion of methods of LARC [25] and a program to provide methods of LARC in school-based health centers [27] described the critical importance of clear and consistent communication across roles and with patients, and of early involvement and engagement of all stakeholders.

This study extends earlier research by analyzing implementation data across multiple practice settings and contexts, and by offering specific recommendations that can be used to inform similar initiatives in other states and communities across the country. This study does have some limitations. Notably, the respondents were all champions of the initiative and, in their support for the initiative, may not have represented the full range of difficulties that others in their practices may have raised. However, we found all respondents to be open and willing to share suggestions for improvement because of their support for the mission of increasing comprehensive contraceptive access for all women. This study did not collect data from other stakeholders, including non-administrative providers or patients. Instead, we sought to capture administrative perspectives that were more of an overview across specific roles. Finally, there was considerable variability in the timing of data collection relative to when each site began participating in the initiative. All sites were in the implementation phase, meaning they had completed the training and were still working with Upstream around implementation. However, they ranged from just a few months to more than two years since their initial training. Despite these limitations, this study offers unique and important insights into the barriers and facilitators of implementation of a statewide contraceptive access initiative. From these, we are also able to offer specific and concrete recommendations that may be of use to other communities working to implement similar initiatives that aim to increase access to contraception for all women.

Implications.

To meet the specific but heterogenous needs of various practices, it is crucial for future contraceptive access initiatives to conduct a comprehensive pre-implementation assessment. Preceding any training, this assessment should gather input from participants across all roles in a medical practice (e.g., providers, medical assistants, office staff, billing department).