Abstract
Objective: Expanding access to high-quality contraceptive care in primary care is key to achieving reproductive health equity. We assessed the impact of an equity-focused quality improvement learning collaborative (QILC) on contraceptive care at community health centers (CHCs) through innovative performance measurement.
Methods: We developed a 9-month QILC including monthly learning sessions on reproductive health equity and person-centered contraceptive care, supporting resources, peer-learning opportunities, and technical assistance. We assessed QILC impact through 3 performance measures collected prepost QILC: the Person-Centered Contraceptive Counseling (PCCC) measure that assesses contraceptive counseling; the Contraceptive Care Screening electronic clinical quality measure (eCQM) (CCS-SINC); and contraceptive use eCQM (CU-SINC). CHCs collected PCCC surveys from patients prepost QILC, and we extracted electronic health record data to calculate eCQMs. To assess intervention impact, we compared prepost PCCC scores and eCQM percentages.
Results: Nine CHCs participated in the QILC. Assessing contraceptive care screening, median increase in CCS-SINC was 14.4% (Interquartile range [IQR]: 7.5%–40.4%) between baseline and endline. CHCs realized an increase in CU-SINC (median relative Δ: 4.9%; IQR [3.7%–22.3%). Compared with baseline, at endline, 5 of 9 CHCs improved their PCCC score (Δ 2.1% to 26.2%) and 3 of those surpassed the 80% benchmark for high-quality care. Greater improvement in performance measure scores was noted among CHCs with leadership buy-in and lower staff turnover.
Conclusions: Participation in an equity-focused and measurement-driven QILC led to improvements in person-centered contraceptive care delivery.
- Community Health Centers
- Contraceptives
- Health Equity
- Quality Improvement
- Quality of Care
- Reproductive Health
Introduction
High-quality contraceptive care is an essential component of health care, as it facilitates individuals to self-determine and achieve their reproductive goals.1 However, in the United States (US), patients face many barriers to quality contraceptive care. Logistic barriers, such as a dearth of clinicians providing contraception, compromise access to services, with this burden disproportionately falling on racial and ethnic minoritized, uninsured, and rural individuals.2–6 Moreover, there are documented gaps in the quality of contraceptive care when it is available, including failure to provide care focused on patient’s own desires, preferences, and values. A common demonstration of this low-quality care is clinician promotion of, and even pressure toward, specific contraceptive methods, notably intrauterine devices and contraceptive implants, regardless of patients’ own method preferences.7 This lack of patient-centered contraceptive care is differentially experienced by patients and communities who have historically been subjected to coercive reproductive health care, including low-income individuals, racial and ethnically minoritized individuals, and those with disabilities.7–9
Primary care is a critical access point for contraceptive care.10–12 Community health centers (CHCs) provide primary care health services across the lifespan for millions of underserved patients in the US,13 including one-third of low-income women of reproductive age14 – the same populations who experience the greatest barriers to high quality, person-centered contraceptive care. However, the services primary care locations, and CHCs in particular, offer can vary, with many not offering the full range of contraceptive methods.6,15,16 Moreover, competing priorities can interfere with delivery of this preventive care service.12,16–18
Improving contraceptive care within CHCs has the potential to enhance reproductive health care access and support people to achieve their reproductive goals. Quality improvement in this area requires attention to both the ongoing reality of nonpatient centered contraceptive care practices and reckoning with the long history of reproductive control and injustice in the US that was perpetuated by the health care system, such as a history of forced sterilization of people of color, low-income people, and people with disabilities, among other groups.19,20,20–23 Moreover, while many patients report a preference for accessing reproductive health services in primary care Federally Qualified Health Centers (FQHCs) and other CHCs,10–12 they also report caution about seeking care in this setting and only find it satisfactory as a means of care if their autonomy and desires about their reproduction are respected.12
Quality improvement learning collaboratives (QILCs) are promising interventions through which to share information and strategies for quality improvement across multiple agencies at once.24–26 QILCs have proven utility in a reproductive health context, however they often focus on practical logistic barriers and do not incorporate a historically informed focus on equity.27–29 We created an innovative QILC, grounded in health equity principles and historic understanding of the history of reproductive health care provision, to support CHCs to assess and improve the quality of their contraceptive care programs. We measured the impact of the QILC in CHC settings on contraceptive care access and quality using novel person-centered contraceptive care performance measures that assessed the multiple components of contraceptive care delivery.
Methods
CHC Partner Recruitment
We partnered with 2 Health Center Controlled Networks (HCCNs) who recruited CHCs from their networks through in-person networking events, e-mail newsletters, and webinars. CHCs were deemed eligible to participate if they provided contraception in any part of their clinic.
Intervention
CHCs participated in a 9-month QILC informed by the Institutes for Health care Improvement (IHI) breakthrough series collaborative model.24 The QILC consisted of the following components: 2 periods of performance measure data collection with baseline data reports used to inform quality improvement and post QILC data used to assess impact (see Measures), 9 monthly virtual learning sessions, a comprehensive online change package, 3 concentrated quality improvement (QI) action periods, and individualized technical assistance sessions (Figure 1).
Inputs and components of project Quality Improvement Learning Collaborative (QILC). Abbreviations: eCQM, electronic clinical quality measure; EHR, Electronic Health Record; PCCC, Person-Centered Contraceptive Counseling; SINC, The Self-Identified Need for Contraception
Learning Sessions
We held eight 90-minute monthly virtual learning sessions and a final harvest session. Content included 4 foundational topics related to racial justice and health equity, measurement, and person-centered care and 4 additional sessions selected by participating agencies from a predefined list (see Figure 1 for list of topics). Each session consisted of a didactic presentation with 1 or more subject matter expert, followed by breakout session discussions and large group peer sharing related to Learning Session topics and QI goals.
Quality Improvement (QI) Activities
QI approaches were introduced through a change package including a self-assessment survey and a toolkit of resources and evidence-based interventions. We organized strategies into 4 key areas of care (Figure 2). The change package, which was hosted on a dedicated website, outlined information on use and interpretation of contraceptive care performance measures and detailed action steps and resources pertaining to each quality improvement strategy.
Organization of quality improvement strategies included in Quality Improvement Learning Collaborative (QILC) change package.
CHCs self-selected QI strategies to focus on in each of three 6-week-long Action Periods based on a self-assessment and review of their baseline performance measure scores. We recommended IHI’s Plan-Do-Study-Act (PDSA) method when establishing the QI structure to test the impact of CHC strategies on a smaller scale, however CHCs were encouraged to any use alternative approaches if already in use in their settings. We supported QI efforts through individualized technical assistance support provided on an ongoing and ad hoc basis as requested by participating agencies.
Contraceptive Care Performance Measures
We utilized 3 performance measures analyzed together to both inform their quality improvement and to assess impact of the QILC intervention. These measures were selected to capture access to contraceptive need screening, a full range of contraceptive methods, and quality of contraceptive counseling to provide a full contraceptive care assessment. Data were collected and reported to participants at baseline and endline of the QILC.
Contraceptive Care Screening Measure (CCS-SINC)
The Self-Identified Need for Contraception (SINC) is a service-bound screening question of contraceptive care designed to be asked of patients at intake that reads: “We ask everyone about their reproductive health needs. Do you want to talk about contraception or pregnancy prevention today?”30 With the aid of their HCCNs, CHCs integrated SINC into their Electronic Health Records (EHRs) within a year before the start of the QILC, optimizing clinical workflow throughout the QILC. An electronic clinical quality measure (eCQM) of access to contraceptive screening is calculated using exported EHR data as the percentage of nonpregnant, women of reproductive ages 15 to 44 years who were asked SINC at least once in the previous calendar year.
Contraceptive Use eCQM (CU-SINC)
We utilized EHR data to measure an eCQM of contraceptive use to evaluate how clinics were meeting contraceptive need31: the percentage of nonpregnant women ages 15 to 44 years interested in contraception who were documented to be using with a most (sterilization, intrauterine device or contraceptive implant) or moderately effective (injectable, contraceptives, oral hormonal pills, patches, and vaginal rings) method of contraception (CU-SINC). Contraceptive “use” included both documented provision of a method or self-report from the patient of a contraceptive method they are currently using. To specify the population of interest to only those who want contraceptive services, patients who answered “No” to SINC were excluded from the denominator of the measure.
Person-Centered Contraceptive Counseling (PCCC) Measure
The PCCC is a 4-item patient reported outcome performance measure (PRO-PM) that measures patient experience of contraceptive counseling.32 Consistent with other patient experience performance measures,33 PCCC is presented as a score compromised of the percent of patients who report “Excellent” across all 4 scale items. A score of 80% is considered a benchmark of high-quality care, based on clustering above this number in validity testing.32
Performance Measure Collection, Calculation, and Reporting
HCCN partners performed 2 sets of EHR extractions: once before the initiation of the QILC (Fall 2022) and a second at the completion of the QILC (Fall 2023). We then utilized extracted data to calculate CCS-SINC and CU-SINC for each CHC.
For PCCC collection, CHCs distributed surveys consisting of the PCCC and demographic questions either electronically or on article At both baseline and endline, we asked CHCs to collect at least 50 surveys, as prior testing of this measure had documented this to be a number that provides adequate reliability.32 To describe the clinical settings and services of the CHCs participating, we also captured organizational characteristics through a survey completed by each CHC at baseline.
Data Analysis and Presentation
We calculated absolute percent change between baseline and endline CCS-SINC and PCCC scores. Absolute percent change was most useful for these measures as the goal of the SINC screening measure is to screen 100% of eligible patients annually, and the PCCC has a clear benchmark of high-quality care (80%).32 Since no target exists for CU-SINC and the goal is to incentivize greater access without compromising autonomy, we calculated relative percent change for this measure. Given our goal of addressing disparities in contraceptive care quality, we also calculate prepost PCCC scores by race/ethnicity of respondents as well as absolute percent change. We report race/ethnicity scores in aggregate due to low sample sizes at the CHC level.
This intervention was deployed while CHCs were managing competing priorities and high staff turnover as a result of the COVID-19 pandemic.34 Given this implementation challenge, we examined the results post hoc through the lens of established factors for successful quality improvement to more accurately assess implementation of the intervention: engaged leadership and sufficient workforce.35 Specifically, we utilized qualitative data collected throughout the project, including notes from technical assistance calls, field notes from learning sessions, and debriefs with staff liaisons, to evaluate whether participants reported leadership that supported the project and contraceptive care optimization and whether they reported challenges with staff turnover or capacity during the course of the project. Notes were reviewed by 2 authors and extracted to a matrix into the a priori identified factors of leadership and staffing by CHC. One author then summarized the findings by dimension, noting facilitating and impeding factors. These summaries were then discussed, and a binary score was assigned to each factor by consensus, generating a two-by-two table of leadership and staffing. We then organized CHCs’ results into 2 tiers: the “high support tier” were CHCs who reported leadership buy-in and ongoing staff capacity and the “low support tier” included CHCs that lacked leadership support and faced challenged with staffing throughout the project.
Results
Characteristics and Participation of CHCs
Nine CHCs participated in the intervention, representing 30 clinical sites across 7 states and 3 geographic regions. CHC characteristics are included in Table 1. All provided some contraceptive care with most providing onsite insertion and removal of intrauterine devices (IUDs) (89%) and implants (78%). Each CHC designated 1 to 3 project liaisons (based on CHC discretion) who attended the Learning Sessions. Of the liaisons who facilitated the project, 7 were Doctors of Medicine or Doctors of Osteopathic Medicine, 3 were nurse practitioners, 1 was a Physician Assistant, 1 was a Health Educator and 5 were administrative staff. Six out of 9 CHCs included both a clinical and nonclinical representative to the project. Notably, 4 of the participating CHCs had turnover in their designated project liaisons, due to either staff leaving the CHC or changing priorities. Seven CHCs attended 90% or 100% of Learning Sessions (remaining 2 CHCs attended 80% and 60% of sessions, respectively). CHCs were also able to view recordings of sessions.
Organizational Characteristics of Community Health Centers (CHCs) that Participated in the Quality Improvement Learning Collaborative (n = 9)
Data on CHC QI strategies are presented in Table 2. CHCs chose diverse strategies to meet their specific agency needs. All focused on some aspect of advancing staff training, attitudes, and beliefs, with most using recordings and training materials provided in the QILC to build foundational knowledge in reproductive health equity and history (67%) and person-centered counseling to train clinicians and standardize contraceptive counseling practices (55%). Most also focused on clinical operations (77%), primarily on optimizing identification of patients’ reproductive health needs through integration of SINC screening (67%). Two agencies used a combination of training and operation strategies to expand contraceptive service provision to additional departments (Pediatrics and Primary Care respectively). One CHC focused on improving care for special populations strategy after noting a disparity in their PCCC scores between Spanish versus English-speaking patients. CHCs on average attended 4 technical assistance calls (range: 2 - 9) to support enactment of these QI strategies.
Quality Improvement (QI) Strategies Utilized by Characteristics of Community Health Centers (CHCs) (n = 9) During QILC Action Periods
Five CHCs were determined to have high support for the intervention. Four of those reported high levels of both leadership buy-in and staffing capacity. One CHC had strong leadership buy-in but did encounter higher staff turnover. After reviewing process data, it was determined that this CHC belonged in the high support tier, as their strong leadership was able to buoy the quality improvement despite staff turnover. Four CHCs had both low leadership buy-in and low staff capacity and were classified into the low support tier.
Performance Measures
Baseline and endline CCS-SINC and CU-SINC, and PCCC scores are presented in Table 3, organized by tier of support.
Change in Performance Measure Scores by Participating Community Health Centers (CHCS) Before (Baseline) and After (Endline) Participating in Nine-Month QILC, Stratified by Support Tier
CCS-SINC
At baseline, most CHCs had only begun to implement SINC in their settings, making the overall median percent of eligible patients screened 0.7%, with very little difference between high and low support (HS, LS) tier CHCs. One CHC in the high support tier had substantial implementation at baseline with 35.9% of eligible patients screened. At endline, the percent of patients screened with SINC ranged from 0.1% to 52.0%. The low support group had a higher median percent change in patients screened at 27.1% increase compared with 16.1% in the high support tier.
CU-SINC
At baseline the percent of eligible patients who received a most or moderately effective contraceptive method ranged from 12.9% to 72.5% (median: 26.8%), with a lower skew of percentages in the high support tier (median: 18.5% v. 41.5%). Relative change among the high support tier was greater than the low support tier (median relative% change: 18.0 v. 4.3%).
PCCC
At baseline, PCCC scores ranged from 37.1% to 93.8%, with the high support tier skewed higher compared with the low support tier (median, 69.7% v. 58.8%). At endline, 5 out of 6 sites in the high support tier exceeded the 80% benchmark for high quality care (median: 82.0%, range: 75.0% −94.0%). Median absolute percent change in scores within this group was 8.3%. Three out of 4 CHCs in the low support tier saw declines in PCCC scores between baseline and end line and only 1 exceeded the 80% benchmark of high-quality care (median: 49.4%; range: 25.0% - 80.8%). Median absolute percent change within the low group was −8.2%.
Prepost PCCC scores by race/ethnicity are included in Table 4. At baseline, scores were comparable between Black (74.7%) and White (75.7%) groups. Scores increased prepost among Black patients (Δ 8.8%) and multi-racial patients (Δ 7.0%). Scores did not change among White patients (Δ 0.3%). The PCCC score among Hispanic/Latina patients was notably lower at baseline (69.4%) and there was no change prepost (Δ −0.5%). While high at baseline (86.2%) the score among Asian patients declined 18.5%.
Person-Centered Contraceptive Counseling (PCCC) Scores (%) by Race/Ethnicity at Baseline and Endline and Pre-Post Change in Score
Discussion
This article demonstrates the implementation of a QILC grounded in health equity and the impact of this intervention on a suite of performance measures capturing multiple components of the contraceptive care process. Novel aspects of the QILC include beginning with a foundation in history of reproductive injustice and equity principles, tailored technical assistance to support individualized quality improvement strategies, and use of innovative person-centered performance measures to assess quality. All participating CHCs demonstrated increases in uptake of SINC and in documented contraceptive use, with PCCC scores increasing in 5 out of 9 CHCs, and 5 CHCs surpassing the 80% benchmark of high-quality care. Improvements in these metrics were linked to CHC capacity/support, with CHCs with high support reporting improvement in all 3 performance measures.
This study is the first time this novel set of contraceptive care performance measures, grounded in person-centered principles, were used in combination for quality improvement to capture multi-dimensional quality across the contraceptive care encounter, assessing: whether patients are being screened for their contraceptive services, whether they receive person-centered counseling, and whether those who want contraception are using a method. Using metrics for quality improvement that are specific to the contraceptive care context and center patient experience enables equitable change in health care delivery.
Prepost data reflect that performance measures were responsive to change across a year span and thus appropriate for quality improvement application. Applied together, health service sites and systems can monitor for increased provision of contraceptive services while also ensuring that delivery was conducted in a manner that centers patients’ needs, values, and preferences. In our cohort, some CHCs saw increases across all measures together, suggesting a move toward higher quality care. However, some CHCs, particularly those in the low support tier, had PCCC scores that started and ended low or even declined, even in cases when CCS-SINC and CU-SINC increased. This suggest that while access improved, contraceptive counseling is still being delivered suboptimally, leaving care overall inadequate. Inversely, some CHCs, particularly in the high support group increased PCCC scores and ended the project with scores above the 80% benchmark, but CCS-SINC and CU-SINC numbers remained low. For these CHCs, focusing on screening is key to ensure that all patients in their settings who want contraception have access to their high-quality counseling.
The range of quality metric scores between CHCs reflects differential quality of care experienced around the country and the need for innovative strategies for quality improvement, and align with recent studies assessing the quality of contraceptive care at a national level.36–38 Having a cohort with differing contraceptive care provision experience and reported quality was likely to the benefit of participating agencies—QILCs are designed for cross-organizational learning and can generate normative pressure and incentive for change.26,39 In addition, establishing a strong foundation in history of health care–driven reproductive control and injustice and the importance of person-centered care as a strategy to protect reproductive autonomy built common understanding for equity at all levels of contraceptive care practice experience. This key learning was further disseminated through the quality improvement strategies chosen by the cohort; all participating agencies focused on staff training, attitudes, and beliefs and most focused on optimizing screening. CHCs with high overall baseline PCCC scores recognized the ability to further optimize care delivery by narrowing in on specific health care quality disparities noted in their baseline data reports and focusing on expanding access through which departments offer contraceptive counseling. A focus on racial/ethnic disparities is reflected in changes to PCCC scores with those among Black and multi-racial patient groups increasing between baseline and endline. However, the comparatively low score and lack of movement in PCCC score among Hispanic/Latina PCCC and the drop in score among Asian patients is important to note as well. Future iterations of this intervention will include more targeted content on providing culturally responsive care.
We also considered the role of external constraints on the success of the intervention and the quality of contraceptive care overall. We saw CHCs with limited leadership support and high staff turnover demonstrate decreases in the PCCC or, in one case, minimal improvement. At the same time, 3 of these sites had robust increases in SINC implementation, with limited change in CU-SINC. This indicates that the structural intervention of screening for contraceptive need may not have resulted in increased counseling and access for those who were interested, and in fact the counseling that did occur was less person-centered. While SINC implementation can be a key tool for access, it must also be paired with person-centered counseling that results in patients receiving a method if desired. Quality improvement requires dedicated staff time and leadership support to be successful.35 This is particularly salient in the model of quality improvement used in this QILC, as it necessitated organizational change in culture, workflows, and practices, which can only be successful with organizational buy-in and the diffusion of information and adoption across the clinical team. All CHCs who did not make progress on quality improvement goals lacked these key foundational factors. Notably, this QILC also occurred closely following the COVID-19 pandemic when CHCs across the US were plagued by competing priorities and high staff turnover,34 which added an additional challenge to dedicated participation, diffusion of ideas and strategies, and staying power of quality improvement interventions. Declines in a range of other performance measures were noted among FQHCs nationally as a result of these challenges.40
Our findings should be interpreted through limitations. Most notably, there was no comparison group. Impact was determined from prepost comparison scores. It is possible that changes in scores may result from external factors, such as temporal trends and differences in survey data collection. In addition, implementation challenges grounded in staff turnover and competing priorities, exacerbated by the overextension of CHCs in the post-COVID era, meant that diffusion of QILC learnings and strategies may have been limited. We tried to address this in our interpretation by dividing results by high/low support tiers. Lastly, our CHC cohort was diverse in size, scope of practice, and experience with contraceptive care. However, we lacked representation from the Southern region of the US and limited representation from the Midwest, where, notably, reproductive health services are increasingly legally and socially restricted.41 Thus, our findings may not be transferable to that setting. We intend to replicate this project in a restricted-setting context to assess and optimize implementation in those complex settings.
Despite limitations, this project demonstrated the application of novel, person-centered performance measures of contraceptive care quality in CHCs and promising results of a QILC grounded in racial justice and health equity. Primary care, and CHCs specifically, can play a key role in meeting patients’ contraceptive service needs and achieving reproductive health equity. Tools such as the suite of measures described here can motivate and ensure equitable access to contraceptive care by capturing multiple dimensions of quality that center patient needs. Interventions to improve contraceptive access should be grounded in historic and social understanding of reproductive health care harms and biases.
Acknowledgments
We are indebted to Fei Dong, Ella Puga, Philip Hastings who facilitated EHR data extractions and calculations of eCQMs. We are also grateful for our guest lecturers during the QILC: Rachel Logan, Diana Carvajal, Elizabeth Jones, Miles Harris, R Claire Roden, and Beth Thielman. Lastly, we want to thank our CHC and HCCN partners who humbly and earnestly dedicated themselves to learning and quality improvement throughout this project.
Notes
This article was externally peer reviewed.
Funding: This project was supported through a grant from an anonymous private donor.
Conflict of interest: The authors have no conflicts to report.
- Received for publication February 21, 2025.
- Revision received May 5, 2025.
- Accepted for publication May 19, 2025.
