Scaling evidence: national causal evaluation of perinatal quality collaboratives needs better data.

Scaling evidence: national causal evaluation of perinatal quality collaboratives needs better data - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Health Aff Sch . 2026 May 29;4(6):qxag121. doi: 10.1093/haschl/qxag121 Scaling evidence: national causal evaluation of perinatal quality collaboratives needs better data Rachel Nesbit 1, ✉, 2 , Rachel K Landis 2 , Kandice A Kapinos 3 Author information Article notes Copyright and License information 1 Economics, Sociology, & Statistics, RAND Corporation, Pittsburgh, PA 15213, United States 2 Behavioral & Policy Sciences, RAND Corporation, Arlington, VA 22202, United States 3 Economics, Sociology, & Statistics, RAND Corporation, Santa Monica, CA 90401, United States ✉ Corresponding author: Economics, Sociology, & Statistics, RAND Corporation, 4570 Fifth Ave #600, Pittsburgh, PA 15213, United States. Email: rnesbit@rand.org 2 Conflicts of interest: The authors declare no conflicts of interest. Received 2026 Mar 17; Revised 2026 May 4; Accepted 2026 May 14; Collection date 2026 Jun. © The Author(s) 2026. Published by Oxford University Press on behalf of Project HOPE - The People-To-People Health Foundation, Inc. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License ( https://creativecommons.org/licenses/by-nc/4.0/ ), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact reprints@oup.com for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact journals.permissions@oup.com. PMC Copyright notice PMCID: PMC13234944  PMID: 42253882 Abstract Perinatal quality collaboratives (PQCs) have been widely adopted across US states to improve maternal–infant health care delivery and outcomes, yet rigorous evidence of their national impact on maternal mortality and severe maternal morbidity remains limited. This commentary identifies 2 fundamental challenges constraining causal inference in PQC evaluations. First, inconsistent measurement of key outcomes, stemming from staggered state adoption of revised vital statistics certificates and coding changes, confounds attempts to distinguish genuine health improvements from data artifacts. Second, the causal pathway from state-level PQC adoption to health outcomes involves multiple intermediate changes in clinical practices and care delivery that remain unobserved in administrative data. This imprecision in treatment measurement, combined with rare outcomes, severely limits statistical power, especially for analysis of subpopulations at increased risk of adverse maternal–infant health outcomes (e.g., Black women, those in low-income communities). We recommend establishing a national registry to track quality improvement initiatives systematically, investing in harmonization of vital statistics data across states and time periods, and exploring alternative evaluation designs (e.g., synthetic control methods, stepped-wedge trials) to credibly estimate PQC impacts on maternal health outcomes. Keywords: perinatal quality care collaborative, maternal health, quality improvement, national vital statistics system, maternal mortality Key points. Inconsistent measurement of maternal mortality and severe maternal morbidity across states and over time makes it difficult to distinguish improvements in care from changes in data collection, a fundamental obstacle to evaluating perinatal quality collaboratives (PQCs) nationally. PQCs vary considerably in structure, initiatives, and hospital participation, meaning that national evaluations treating PQC adoption as a uniform intervention cannot capture the mechanisms or heterogeneity that matter most for understanding what works. Federal investment in vital statistics harmonization and standardized Maternal Mortality Review Committee capacity is urgently needed, while PQCs, with sufficient funding, can act now by tracking initiative implementation systematically, measuring process and proximal outcomes, and building prospective evaluations into new initiative rollouts. The US maternal health crisis continues and in 2022, the maternal mortality rate was more than 50% higher than the closest peer country. 1 Severe maternal morbidity (SMM)—life-threatening conditions during the labor and delivery hospitalization—is much more common, affecting up to 60, 000 women annually. 2 In response, many federal and state initiatives have been implemented, including systems-level interventions designed to improve the quality and safety of maternity care across hospitals, health systems, and clinical settings. Among the most prominent of these programs are perinatal quality collaboratives (PQCs), which aim to improve maternal–infant healthcare delivery and outcomes, particularly preventable deaths. 3 All states have implemented PQCs, though the timing and degree of involvement vary. PQCs facilitate data-driven quality improvement (QI) models, typically focusing on provider-level processes adapted to local needs. Despite widespread adoption and some evidence of improvements in specific outcomes, 3 little is known about their national , population-level impact on maternal mortality and SMM. Most studies have focused on descriptive pre–post comparisons of a specific initiative or state and suggested improvements. However, because many initiatives target local maternal health challenges, naïve comparisons might reflect elevated baseline risk driven by environmental, sociopolitical, or broader healthcare system factors, and without a strong design to address confounding, may not reflect true impacts. Two challenges complicate expanding the evidence base. First, measurement of key outcomes (e.g., maternal death or SMM metrics) has changed considerably over time and across states. 4 , 5 Second, the causal pathway between initiative implementation and improved outcomes is complex, multilevel, includes unobservable elements, and varies across states. In this commentary, we discuss each of these issues and their interaction, concluding with recommendations for moving this literature forward. Addressing these challenges requires consistently measured maternal-infant data and precise information about the scope and timing of PQC adoption and implementation. As Congress reauthorized the Preventing Maternal Deaths Act with renewed investments in Maternal Mortality Review Committee (MMRC) data 6 —the only national data on preventable maternal deaths—this is a timely opportunity to consider improvements in related data infrastructure. Although the National Center for Health Statistics (NCHS) National Vital Statistics System (NVSS) provides restricted-access data on birth complications and maternal and infant deaths, these data have important limitations. In 2003, NCHS revised both the standard birth and death certificates, adding a pregnancy checkbox to the death certificate and morbidity checkboxes to the birth certificate, but state adoption was slow and uneven. 4 Moreover, changes in cause-of-death coding and variation in certificate completion compound inconsistencies in the measurement of maternal deaths and SMM. As others have documented, attempts to create time-consistent measures either underestimate maternal deaths 5 (making a rare event rarer) or reduce the number of state-years available for analysis. Consequently, observed differences in maternal health may reflect changes in data collection and measurement rather than changes in care. The Pregnancy-Related Mortality Surveillance System and state MMRCs have their own comparability challenges. State MMRCs vary in how they determine pregnancy-related deaths and classify underlying causes of death, though the CDC is working to standardize this process. 6 Because mortality case reviews are complex, MMRC reports are often years out of date at the time of public release, constraining their utility for timely PQC evaluation. 7 Birth-certificate-based SMM measures have also been shown to be poorly correlated with hospital billing data, raising validity concerns. 8 State Inpatient Databases offer promising measures of SMM but do not contain comprehensive information about deaths, and not all states make their data available to researchers. The standard SMM measure has important structural limitations. It excludes mental health conditions, a critical omission as they are the leading cause of pregnancy-related death, and events related to pregnancy outcomes other than live births. 7 Traditional measures of maternal mortality and SMM also ignore the 60% or more of “pregnancy-related deaths” that occur outside the labor and delivery inpatient stay, 7 including those related to overdose and suicide, events often targeted by PQC initiatives. Without better data, PQC effectiveness in these areas cannot be empirically demonstrated, even if initiatives are working. Similarly, PQCs are increasingly expanding their focus from in-hospital birth care to also engage with emergency departments and ambulatory clinics, but without operationalizable quality metrics, any care improvements in those spaces cannot be assessed. Beyond measurement challenges, treating PQC adoption as a single intervention obscures meaningful heterogeneity across and within PQCs. As Figure 1 shows, the causal chain between state-level PQC adoption and maternal mortality and SMM rates has many intermediate points: states adopt PQCs, hospitals and clinicians choose whether to participate, and PQCs choose specific initiatives. All of these choices jointly affect care processes and outcomes, but because they are often unobservable in administrative data, national assessments can only estimate state-level adoption effects, not the mechanisms underpinning implementation or the effects of actual participation. Resulting estimates are therefore best understood as intent-to-treat rather than treatment-on-the-treated effects of collaboration. This distinction matters because PQCs vary in structure, initiatives implemented, hospital participation, and data-reporting capabilities—differences that may themselves explain variation in outcomes across states. Figure 1. Open in a new tab Causal chain from PQCs to maternal health outcomes. These issues compound the problem of low statistical power. Outcomes such as maternal mortality and SMM are relatively rare, and the level of available variation is constrained by reductions in sample size due to changes in vital statistics data, as well as imprecision in the measurement of the timing of treatment and the identification of mothers who were exposed to PQC initiatives. A difference-in-differences study using state-level death certificates would be underpowered to detect an impact as small as 1 death per 100 000 births. For a minimum detectible effect of this size, researchers would need a sample of almost 200 states. In reality, in analyses using NVSS birth data, only 25 states could be analyzed due to the timing of adoption and certificate updates. 9 Even in individual-level analyses with many observations, the observable variation in PQC adoption comes from the state level, constraining statistical power to detect causal effects. These challenges intensify when examining causal effects of PQCs for subpopulations with disproportionately high rates of SMM and maternal mortality, especially Black women, those in lower-income communities, and those with Medicaid insurance. 10 Restricting the sample to these subpopulations further reduces statistical power and undermines our ability to evaluate efforts to reduce maternal health disparities. The path forward requires action at multiple levels. With sufficient funding, PQCs and their hospital partners can act immediately. Establishing a systematic registry of quality improvement initiatives akin to clinical trial registries or to the Department of Education's What Works Clearinghouse would improve both measurement and cross-hospital learning and sharing. While longer-term data infrastructure is being developed, PQCs should prioritize measuring process and proximal outcomes, like protocol adoption rates and intermediate clinical outcomes, that are more directly observable, more proximate to PQC activities, and more statistically tractable. PQCs should also partner with researchers to build prospective evaluations such as step-wedge trials into implementation from the start, provided funders support these efforts. Federal action is equally essential. Congress's reauthorization of the Preventing Maternal Deaths Act is a timely opportunity to fund harmonization of NCHS birth certificate data across states and over time and expand standardized MMRC capacity—investments that would strengthen the evidence base for PQCs and maternal health policy broadly. These steps are essential to support the rigorous national analyses that PQCs merit and mothers deserve. Supplementary Material qxag121_Supplementary_Data qxag121_supplementary_data.docx (57KB, docx) Acknowledgments This paper received funding from the National Institutes of Health under project R01MD016892. The authors are part of the larger collaborative project team. This methodological commentary reflects the specific expertise and perspectives of the named authors and does not necessarily represent the views of all team members. Contributor Information Rachel Nesbit, Economics, Sociology, & Statistics, RAND Corporation, Pittsburgh, PA 15213, United States. Rachel K Landis, Behavioral & Policy Sciences, RAND Corporation, Arlington, VA 22202, United States. Kandice A Kapinos, Economics, Sociology, & Statistics, RAND Corporation, Santa Monica, CA 90401, United States. Supplementary material Supplementary material is available at Health Affairs Scholar online. Notes 1. Gunja  MZ, Gumas  ED, Masitha  R, Zephyrin  LC. Insights into the U.S. maternal mortality crisis: an international comparison [Internet]. Commonwealth Fund. 2024. Accessed March 2, 2026. https://www.commonwealthfund.org/publications/issue-briefs/2024/jun/insights-us-maternal-mortality-crisis-international-comparison 2. Declercq  E, Zephyrin  LC. Severe maternal morbidity in the United States: a primer [Internet]. The Commonwealth Fund. 2021 Oct 28. Accessed March 2, 2026. https://www.commonwealthfund.org/publications/issue-briefs/2021/oct/severe-maternal-morbidity-united-states-primer 3. Lee King  PA, Henderson  ZT, Borders  AEB. Advances in maternal fetal medicine: perinatal quality collaboratives working together to improve maternal outcomes. Clin Perinatol. 2020;47(4):779–797. 10.1016/j.clp.2020.08.009 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. Hoyert  DL, Miniño  AM. Maternal mortality in the United States: changes in coding, publication, and data release, 2018. Natl Vital Stat Rep. 2020;69(2):1–18. [ PubMed ] [ Google Scholar ] 5. Joseph  KS, Lisonkova  S, Boutin  A, et al.  Maternal mortality in the United States: are the high and rising rates due to changes in obstetrical factors, maternal medical conditions, or maternal mortality surveillance?  Am J Obstet Gynecol. 2024;230(4):440.e1–440.e13. 10.1016/j.ajog.2023.12.038 [ DOI ] [ PubMed ] [ Google Scholar ] 6. Herrick  CL, Biviji  R, Tabb  D. We must protect the only national data on preventable maternal deaths. Health Aff Forefront. 2026. 10.1377/forefront.20260218.186675 [ DOI ] [ Google Scholar ] 7. Trost  S, Beauregard  J, Chandra  G, et al.  2022. Pregnancy-related deaths: data from maternal mortality review committees in 36 US States, 2017–2019 . Centers for Disease Control and Prevention, US Department of Health and Human Services. 2022. 8. Gemmill  A, Passarella  M, Phibbs  CS, et al.  Validity of birth certificate data compared with hospital discharge data in reporting maternal morbidity and disparities. Obstet Gynecol. 2024;143(3):459–462. 10.1097/AOG.0000000000005497 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Nesbit  R, Landis  RK, Park  S, et al. 2026. Toward a national assessment of perinatal quality care collaboratives. [Manuscript under review]. 10. Howell  EA, Egorova  N, Balbierz  A, Zeitlin  J, Hebert  PL. Black-white differences in severe maternal morbidity and site of care. Am J Obstet Gynecol. 2016;214(1):122.e1–122.e7. 10.1016/j.ajog.2015.08.019 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials qxag121_Supplementary_Data qxag121_supplementary_data.docx (57KB, docx) Articles from Health Affairs Scholar are provided here courtesy of Oxford University Press ACTIONS View on publisher site PDF (350.6 KB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Early Postnatal Discharge in a French Perinatal Network#879735
• A comparison of the clinical effectiveness and cost of specialised individually delivered parent training for preschool attention-deficit/hyperactivity disorder and a generic, group-based programme: a multi-centre, randomised controlled trial of the New Forest Parenting Programme versus Incredible Years#352292
• Impact of first-trimester preeclampsia screening on perinatal and maternal morbidity: the RANSPRE open multicenter randomized trial.#161956
• Challenges in implementing optimal perinatal regionalization: a scoping review.#645470
• Challenges in implementing optimal perinatal regionalization: a scoping review#637759