Using a multi-state Learning Community as an implementation strategy for immediate postpartum long-acting reversible contraception

Background Implementation strategies are imperative for the successful adoption and sustainability of complex evidence-based public health practices. Creating a learning collaborative is one strategy that was part of a recently published compilation of implementation strategy terms and definitions. In partnership with the Centers for Disease Control and Prevention and other partner agencies, the Association of State and Territorial Health Officials recently convened a multi-state Learning Community to support cross-state collaboration and provide technical assistance for improving state capacity to increase access to long-acting reversible contraception (LARC) in the immediate postpartum period, an evidence-based practice with the potential for reducing unintended pregnancy and improving maternal and child health outcomes. During 2015–2016, the Learning Community included multi-disciplinary, multi-agency teams of state health officials, payers, clinicians, and health department staff from 13 states. This qualitative study was conducted to better understand the successes, challenges, and strategies that the 13 US states in the Learning Community used for increasing access to immediate postpartum LARC. Methods We conducted telephone interviews with each team in the Learning Community. Interviews were semi-structured and organized by the eight domains of the Learning Community. We coded transcribed interviews for facilitators, barriers, and implementation strategies, using a recent compilation of expert-defined implementation strategies as a foundation for coding the latter. Results Data analysis showed three ways that the activities of the Learning Community helped in policy implementation work: structure and accountability, validity, and preparing for potential challenges and opportunities. Further, the qualitative data demonstrated that the Learning Community integrated six other implementation strategies from the literature: organize clinician implementation team meetings, conduct educational meetings, facilitation, promote network weaving, provide ongoing consultation, and distribute educational materials. Conclusions Convening a multi-state learning collaborative is a promising approach for facilitating the implementation of new reimbursement policies for evidence-based practices complicated by systems challenges. By integrating several implementation strategies, the Learning Community serves as a meta-strategy for supporting implementation.


Background
Long-acting reversible contraception (LARC), which includes subdermal implants and intrauterine devices (IUDs), is an evidence-based method for preventing unintended pregnancies. These methods require no additional user effort once inserted and have failure rates of < 1% [1]. The most common time for receiving LARC after childbirth is at the 4-to 6-week postpartum visit; however, many women miss the postpartum visit or may become pregnant before the visit, putting them at risk for rapid repeat pregnancy and its associated adverse outcomes [2,3]. The current versions of the US and World Health Organization Medical Eligibility Criteria for Contraceptive Use support the placement of immediate postpartum LARC, and the professional organizations of providers from the USA, UK, and Canada have recognized postpartum placement of LARC as safe, effective, and acceptable [4][5][6][7][8].
Although providing LARC immediately postpartum has potential for increasing access to effective contraception, barriers limit widespread adoption by birthing facilities, which include hospitals and birthing centers. In the USA, one substantial barrier is the high costs of LARC devices and insertion procedures; most payers only provide a bundled reimbursement based on the Diagnosis-Related Group (DRG) code for labor and delivery, which is not adequate to cover LARC insertion immediately after birth [9]. To address this barrier, since 2012, several state Medicaid agencies have changed their policies to allow reimbursement to birthing facilities for immediate postpartum LARC devices, and in some states, the insertion procedures, above and beyond the DRG rate for labor and delivery [10,11]. However, experiences of states that were early adopters of a reimbursement policy change demonstrated that a policy change alone was insufficient to increase women's access to immediate postpartum LARC and that implementation strategies were needed to bridge the gap between policy and access to this evidence-based practice [12].
Employing implementation strategies is often necessary for successfully integrating complex, evidence-based public health programs and policies into standard practice [13]. The Expert Recommendations for Implementing Change (ERIC) project recently published compilation of terms and definitions for 73 implementation strategies in order to help implementation science researchers use consistent language in their work [14]. One of the strategies the ERIC project included was the creation of a learning collaborative [14]. Previously, the Institute for Healthcare Improvement in the USA published a white paper on achieving "breakthrough" improvements through the use of a learning collaborative model, and learning collaboratives have been used to support a wide variety of public health projects, including hypertension control and breast cancer disparities [15][16][17].
In 2014, in partnership with the US Centers for Disease Control and Prevention (CDC), other federal agencies, and maternal and child health organizations, the Association of State and Territorial Health Officials (ASTHO) convened a multi-state learning collaborative for increasing access to immediate postpartum LARC. A cohort of six states with Medicaid policies to reimburse for immediate postpartum LARC were invited to participate in this collaborative to identify, document, and address technical assistance needs, promising practices, and barriers to immediate postpartum LARC use [18]. This collaborative, the Immediate Postpartum LARC Learning Community, hereafter referred to as simply the Learning Community, was formed based on the idea that a national-level collaborative could facilitate group learning and information sharing for a systems change approach to better apply policies supporting the use of immediate postpartum LARC [19]. In 2015, ASTHO released a call for letters of interest for any state with Medicaid payment policies in place for immediate postpartum LARC to join a second cohort of the Learning Community. Seven applicant states were accepted at that time to participate alongside the six original states. Activities of the Learning Community include a yearly in-person meeting and quarterly virtual learning sessions. The latter are organized around eight domains, which were selected after baseline interviews with the first cohort of Learning Community states to address the barriers that states face in increasing access to immediate postpartum LARC. These domains include provider training, reimbursement and sustainability, informed consent and ethical considerations, stocking and supply, outreach, stakeholder partnerships, service locations, and data, monitoring, and evaluation (Table 1) [19][20][21].
The purpose of this qualitative study was to better understand successes, challenges, and strategies used for increasing access to immediate postpartum LARC among the 13 states in the Learning Community.

Methods
Between November 2015 and March 2016, we conducted semi-structured interviews via telephone with each of the 13 state teams participating in the Learning Community ( Fig. 1). State teams were multi-disciplinary and multiagency, consisting of three to seven people. They included We developed a semi-structured interview guide that was organized around the eight domains of the Learning Community, which are listed in Table 1 [21]. Each domain covers one aspect of increasing access to immediate postpartum LARC. Within each domain, we asked the teams to describe what their state team has done, what has helped them make their work happen, and what barriers they have faced. Each interview was audio-recorded and transcribed by a third-party vendor. For this analysis, the data were coded for facilitators, barriers, and implementation strategies. We used the compilation of implementation strategies from the ERIC project [14] as a foundation for coding the implementation strategies. Two coders (CD and CE) independently coded all 16 interviews in Dedoose version 7.0.23 (Los Angeles, CA). Following this, the two coders met to resolve coding discrepancies. Coded text was carefully reviewed by members of the primary research team (KR, CD, and CE), who then met to discuss themes in the coded data.
This project received an exemption from the Institutional Review Board at the University of Illinois at Chicago.

Benefits of the Learning Community
One of the key themes from our qualitative data analysis was the mechanism by which the Learning Community serves as a strategy for implementing immediate postpartum LARC policies across and within states. State teams highlighted three ways in which being part of this national learning collaborative has helped them in their policy implementation work: 1. Structure and accountability: Being part of a national initiative with regularly scheduled meetings helps state teams get support from internal organizational leadership and prioritize their work around immediate postpartum LARC. This was explained concisely by one state team member, who said, "Knowing there's a meeting coming up helps us stay focused on the work that we need to get done. " 2. Validity: For many state participants, the activities of the Learning Community require additional time and dedicated effort on top of their normal workload. However, the structure and resources that come with the Learning Community validates state team members' work on immediate postpartum LARC. As one state team member explained, "I think just the validity of having [the learning community] as a shell around the effort…That we're working with an ASTHO Learning Community…as a framework is going to lend so much credibility to this work, rather than [being] just a loosely affiliated team. We've already jumped way ahead of where I thought we would be at this point in time. "

Preparing for potential challenges and opportunities:
Implementation of a statewide immediate postpartum LARC policy is not linear. The Learning Community includes states that initiated their reimbursement policies at various time points, and the states are at varying stages of implementation.
Oftentimes, this means that one state is progressing faster than others within one Learning Community domain, but slower than others in another domain.
Learning from each other's successes and failures helps the states prepare for potential future challenges and opportunities. This is exemplified by the experience of one state team member, who said, "I think the learning collaborative…started us thinking about things that maybe we hadn't thought about that you start to realize more and more as you get deeper into making progress. I think that brought to light things that we should be thinking about and ways to think about them and the realization that there are lots of different approaches as well. I think that's been really valuable. "

Implementation strategies mobilized by the Learning Community
Above and beyond the Learning Community serving as an implementation strategy on its own, the qualitative data demonstrated that the structure of this learning collaborative helps support and integrate six other implementation strategies described by the ERIC project [14]. Throughout the rest of this section, we will provide examples from our interviews with the state teams to demonstrate how this is the case for each strategy. The definitions for the implementation strategies described here are found in Table 2.

Organize clinician implementation team meetings
Implementation efforts for clinical activities can be strengthened by providing time to clinicians involved to reflect on their efforts, share lessons learned, and support each other [14].  Organize clinician implementation team meetings Develop and support teams of clinicians who are implementing the innovation and give them protected time to reflect on the implementation effort, share lessons learned, and support one another's learning In-person meetings provide a unique opportunity for the state teams to work on their immediate postpartum LARC efforts, including creating action plans to prioritize their activities.
Conduct educational meetings Hold meetings targeted toward different stakeholder groups to teach them about the clinical innovation Learning Community virtual learning sessions, which feature guest experts, provide a forum for ongoing education.
Facilitation A process of interactive problem solving and support that occurs in a context of a recognized need for improvement and a supportive interpersonal relationship ASTHO facilitates virtual learning sessions and in-person meetings, which incorporate problem solving and foster the sense of a supportive team across the country.
Promote network weaving Identify and build on existing high-quality working relationships and networks within and outside the organization to promote information sharing, collaborative problem-solving, and a shared vision/goal related to implementing the innovation -State teams include representatives from Medicaid and the state health department, and many of these agencies do not traditionally work together.
-State teams have many opportunities, including through the in-person meetings and virtual learning sessions, to network with other state teams.
-In-person meetings and virtual learning sessions include national partner agencies, which states can collaborate with and learn from.
Provide ongoing consultation Provide ongoing consultation with one or more experts in the strategies used to support implementing the innovation Experts participate in the in-person meetings and virtual learning sessions, and state teams are able to follow up for more in-depth assistance.
Distribute educational materials Distribute educational materials in person, by mail, and/or electronically States developed provider bulletins, patient guides, etc., and then shared them with ASTHO, who then shares the materials with the Learning Community.
a The implementation strategies and definitions in this table were published by the Expert Recommendations for Implementing Change (ERIC) project [14] corrective actions that we can take because we can learn from others."

Promote network weaving
The some of what's going on in states and states can pick their brains a little bit, too. It's also helpful to kind of get on the same page so that different organizations aren't going down the different paths."

Provide ongoing consultation
State team members and individuals from national agencies who are experts in areas related to one (or more) of the eight Learning Community domains participate in the in-person meetings and virtual learning sessions to share their expertise with the Learning Community participants. State teams are able to follow up with these experts for more in-depth assistance as the need arises. As one state team member explained, "I took all the stuff from this toolkit [I am developing] from [the toolkit another state Medicaid representative developed]. We're sending the executive director of our collaborative out to [the other state] to get some mentorship from [the Medicaid representative]."

Distribute educational materials
The structure of the Learning Community helps with the distribution of educational materials related to immediate postpartum LARC. Several states have developed provider bulletins and patient informational handouts related to immediate postpartum LARC. These states have the option of sharing their materials with ASTHO, which in turn shares the materials with the other states in the Learning Community. Some of these materials are also made available on the ASTHO website for public viewing [23]. The state teams reported that this is helpful because they can build on the work of others and tailor materials to the needs of their states, instead of each starting from the beginning. One state team member described the way she uses the materials available, "I have information [to share with hospitals] in an organized folder that has a very brief PowerPoint on why immediate postpartum LARCs are encouraged, information on billing Medicaid, information on a sample hospital policy, sample consents, information for pharmacy, the videos for the physicians…A lot of stuff that I got from the ASTHO learning collaborative are the pieces that I use. That is a [benefit of the Learning Community]… references and resources."

Discussion
Qualitative analysis of semi-structured interviews with the 13 states in the Learning Community demonstrated that this learning collaborative is an important strategy for states implementing this Medicaid policy change around the USA. Further, this learning collaborative serves as a meta-strategy for implementation. In other words, the activities of the Learning Community, including the yearly, 2-day in-person meeting and the quarterly virtual learning sessions, help integrate six other implementation strategies described by the ERIC project [14]: organize clinician implementation team meetings, conduct educational meetings, facilitation, promote network weaving, provide ongoing consultation, and distribute educational materials.
For a clinical innovation as complex as immediate postpartum LARC, establishing a Medicaid reimbursement policy alone is insufficient to bring about widespread access to LARC devices during the delivery hospitalization [12]. As previous literature has suggested, successes in dissemination, implementation, and sustainment of evidence-based practices occur through the development and application of deliberate strategies that may be effective across many different clinical innovations and guidelines [24]. Results from this study suggest that implementation strategies, such as creating a learning collaborative, can help states facilitate implementation of policies designed to improve access to the most effective contraceptive methods in the immediate postpartum period as well as address barriers. In previous studies, implementation strategies have been organized into distinct clusters [25] and grouped according to where they fall in the implementation process [26]. Research teams have sought to test the effectiveness of implementation strategies alone or in combination [27]. To our knowledge, we are the first researchers to describe how one strategy might serve as a meta-strategy to bundle together several other strategies. It should be noted that this is different than what the researchers in the ERIC project proposed, as their implementation strategies are "discrete" and "conceptually distinct" [14,25]. However, this analysis revealed that the Learning Community, as implemented by ASTHO and its partner agencies, demonstrates how creating a learning collaborative could be a unifying strategy to support the implementation of evidence-based public health practices.
The key steps necessary for implementing immediate postpartum LARC policies take place at many levels within the US context. At the national level, the Learning Community brings together teams from several states and federal partners to share strategies and lessons learned to address implementation challenges. At the state level, each team has worked to establish the Medicaid reimbursement policy in its state and then supports implementation of this clinical innovation (i.e., the LARC insertion) by employing statewide strategies within the eight domains listed in Table 1. While the state teams play a large role in supporting implementation, immediate postpartum LARC insertion ultimately takes place in birthing facilities. Future studies should evaluate implementation strategies that can be employed at the facility level, especially related to the Learning Community domains of outreach, provider training, and stocking and supply.
This study has several important strengths and limitations. First, most state team members were interviewed together, which was both a strength and limitation. We had the opportunity to observe the teams' various planning and interaction styles, and their responses to our questions were enhanced because they built upon each other. However, we may not have captured the more challenging aspects of their work as a team because individuals may have felt uncomfortable sharing interpersonal challenges affecting the team's implementation work. Similarly, interviews with state teams were conducted over the telephone. This enabled us to conduct interviews with all 13 state teams, but it prevented us from observing non-verbal communication cues that may have been important for this study. Additionally, the purpose of our interviews was not to identify which of the implementation strategies from the taxonomy proposed in the ERIC project were being applied. Therefore, while we found that the Learning Community served as a meta-strategy to help integrate six other implementation strategies, this list may not be exhaustive. Further, we did not measure state teams' progress over time or implementation outcomes. Finally, the Learning Community was not designed as part of an experiment to study the effectiveness of implementation strategies, and we do not have comparable data from state teams that are not participating in the Learning Community. However, this study adds to the implementation science literature by describing the benefits perceived by the state teams participating in the Learning Community as they relate to a commonly used taxonomy of implementation strategies [14]. Conceptualizing the Learning Community as a meta-strategy may inform other implementation efforts that are considering forming a learning collaborative to promote the adoption or sustainment of evidence-based practices. Future implementation research should use Proctor and colleagues' framework to comprehensively specify each of the implementation strategies packaged within learning collaboratives [13], then empirically test the effect of individual implementation strategies, and the effect of the interaction between multiple implementation strategies within a learning collaborative, on implementation outcomes.
The work performed by state teams in the Learning Community has important public health implications. First, the relationships developed through this collaborative have the capacity to extend into other public health initiatives, creating an added benefit for state teams.