Evaluating Publicly Funded Doula Programs: Utilization, Outcomes, and Cost

By Birth & Baby University · Last updated: October 2026 · 7 min read

As states put public money into doula care, the questions from legislators, plans and communities get sharper: how many families are actually using it, are outcomes better, and what does it cost? Answering those questions fairly is harder than it looks. A benefit can have hundreds of enrolled doulas and few billed visits. Outcome comparisons can be skewed by who chooses a doula. Cost figures can look very different depending on which outcomes are counted. This article walks through the three questions, what the available studies show, where they fall short, and what a good evaluation would include.

Key takeaways

Quick answer: the published evidence suggests public doula programs can reach people and may be associated with better outcomes, but utilization has often been low, outcome studies are mostly observational, and cost estimates depend on assumptions. A credible evaluation tracks reach, outcomes and cost together, and compares groups carefully.

• In Virginia, 130 state-certified doulas were enrolled as Medicaid providers, fewer than half billed Medicaid, and use among beneficiaries was below 1 percent from 2022 to 2024.
• A national claims analysis of 722 matched pairs found 47 percent lower risk of cesarean delivery, 29 percent lower risk of preterm birth and 46 percent greater likelihood of a postpartum checkup among Medicaid-affiliated females with doulas, but it was observational.
• Cost studies in Medicaid populations estimate possible savings, but one Texas analysis found costs exceeded counted benefits at $869 per client or more.
• A six-state study now underway is designed to estimate effects with matching and time-series methods, stratified by race, and shows what stronger evidence looks like.

Question 1: Is the benefit being used?

Utilization is the first thing to measure, and it has several layers. Mensah and Cuellar (2026) analyzed Virginia Medicaid claims from January 2022 to December 2024 along with the state registry and interviews with nine doula providers. They found 130 state-certified doula Medicaid providers, but fewer than half billed Medicaid for services, most commonly for postpartum care. Beneficiary utilization was low (less than 1 percent), and availability was limited in some high-need areas. Their policy lessons include revising reimbursement rates and structure, recruiting doulas in underserved areas, providing technical assistance for enrollment and billing, expanding outreach to beneficiaries and engaging a range of doulas in policy discussions.

A 2024 interview study of policy informants in Oregon and Massachusetts (Safon and colleagues, 2024) found that complex billing processes created friction and impeded reimbursement, and that doula services were perceived as not valued by some health care providers. The study had 20 informants and reports perceptions, not counts.

A useful utilization dashboard therefore separates enrolled doulas, doulas who have billed, eligible beneficiaries, beneficiaries who received at least one visit, and visits by type and timing, broken down by region, race and ethnicity and language.

Question 2: Are outcomes better?

Falconi and colleagues (2024) used claims data from a national insurer for 2014 to 2023 and matched 722 pairs of Medicaid-affiliated females with and without a doula. Those with doulas had a 47 percent lower risk of cesarean delivery, a 29 percent lower risk of preterm birth and were 46 percent more likely to attend a postpartum checkup. The design is retrospective and observational, with matching on available variables, so unmeasured differences between those who chose or were offered doula care and those who were not could explain part of the result.

The wider evidence is mixed in quality. Viswanathan and colleagues (2026) found moderate-certainty evidence that doula support is likely associated with lower rates of preterm birth and cesarean delivery, improved birth experiences and increased postpartum visit attendance, with low-certainty evidence for other outcomes. They rated most included studies at moderate or high risk of bias and found insufficient information on the effects of varying intensity or types of support. A program-level evaluation should therefore be modest about attribution.

Stronger designs exist. The protocol for a six-state study (Jarlenski and colleagues, 2024) plans matching techniques to estimate the impact of using doula care on postpartum outcomes and time-series analyses to estimate the effect of doula programs on population outcomes, with analyses stratified by race. Its primary outcomes include severe maternal morbidity, person-reported measures of respectful obstetric care and receipt of evidence-based care for conditions that are leading causes of postpartum mortality. It is a protocol, so it describes a design, not results.

Question 3: What does it cost?

Cost studies tend to model a few outcomes. Kozhimannil and colleagues (2016) estimated potential savings with reimbursement averaging $986, based on associations with preterm and cesarean birth. Nehme and colleagues (2024) estimated a benefit-to-cost ratio of 1.15 in central Texas, cost-beneficial in 65.7 percent of sensitivity analyses, with costs exceeding counted benefits at $869 per client or more. Our article on cost-effectiveness research explains the assumptions. For a public program, the practical cost questions are what the program pays per member, what administrative support it requires, and what outcomes the evaluation will count. When utilization is low, total spending may be small, but so is the number of people reached.

What a good evaluation includes

• A clear question and a comparison. Decide whether you are describing delivery, estimating effects, or both, and choose a comparison group or design that fits.
• Equity from the start. Break every measure down by race, ethnicity, language, geography and insurance, and watch for groups the program is not reaching.
• Workforce data. Count doulas who enroll, bill, stay and leave, and ask why.
• Family experience. Include person-reported measures, with attention to the quality of the tools.
• Honest limits. State what the design can and cannot show, and avoid calling associations proof.
• Doulas at the table. Interviews and policy studies repeatedly point to partnership with doulas as a factor in sustaining programs.

What this means for doulas

• Your records may be part of the evaluation. Keep accurate visit dates and types, and ask what data your plan or state collects about your clients and how it is protected.
• Know the scope of claims. Evaluation findings should not be used to promise outcomes to clients. You provide support, not clinical results.
• Share barriers. Denied claims, delays and enrollment problems are exactly the information evaluators need. Share them through your hub, coalition or state advisory process.
• Read the actual report. Several states have published or are preparing implementation reports. Reading yours helps you advocate accurately.

Putting it into practice

1. Find your state Medicaid agency’s doula page and any evaluation or implementation report, and note the date and data period.
2. Compare the reported numbers for enrolled doulas, billing doulas and beneficiaries served.
3. Check whether outcome comparisons adjust for differences between groups, and whether they report results by race and ethnicity.
4. If you are asked to contribute to an evaluation, ask how data will be protected and how findings will be shared with doulas.
5. When you cite findings to clients or partners, use language such as “associated with” and name the study.

Limits and uncertainty

This article draws on a handful of studies from specific states and insurers. The outcome study is observational and uses an insurer’s claims, the cost studies are models, and the interview studies report perceptions. The six-state study is a protocol with no results yet. Program details and numbers change, and a state’s own evaluation reports may contain figures not covered here. The evidence does not allow a general statement about how public programs perform everywhere.

The bottom line

Evaluating a public doula program means asking whether people are using it, whether outcomes are better, and what it costs, and being honest that the available evidence answers each question only partly. Low utilization and billing friction have been documented, observational studies point toward benefits, and cost estimates depend on assumptions. Better designs are being built, and doulas can help by documenting barriers and reading reports closely.

Birth & Baby University’s For Doulas page outlines its online, self-paced birth and postpartum doula training, recertification and mentorship, for doulas working within public and private programs.

This article is for professional education. It is not medical, legal, financial or tax advice. Check the standards of your certifying organization and the rules where you work.

Sources and further reading

• Mensah DL, Cuellar AE. Doula services for Medicaid beneficiaries in Virginia: access, utilization, and policy lessons. Health Affairs Scholar (2026). Read the source
• Safon CB, McCloskey L, Estela MG, Gordon SH, Cole MB, Clark J. Access to perinatal doula services in Medicaid: a case analysis of 2 states. Health Affairs Scholar (2024). Read the source
• Falconi AM, Ramirez L, Cobb R, Levin C, Nguyen M, Inglis T. Role of doulas in improving maternal health and health equity among Medicaid enrollees, 2014-2023. American Journal of Public Health (2024). Read the source
• Viswanathan M, Middleton JC, Sathe NA, et al. Doula support during pregnancy, childbirth, and postpartum: a systematic review and meta-analysis. Obstetrics & Gynecology (2026). Read the source
• Jarlenski M, Kennedy S, Johnson A, et al. Study protocol: a mixed-methods study of the implementation of doula care to address racial health equity in six state Medicaid programs. Health Research Policy and Systems (2024). Read the source
• Kozhimannil KB, Hardeman RR, Alarid-Escudero F, et al. Modeling the cost-effectiveness of doula care associated with reductions in preterm birth and cesarean delivery. Birth (2016). Read the source
• Nehme EK, Wilson KJ, McGowan R, et al. Providing doula support to publicly insured women in central Texas: a financial cost-benefit analysis. Birth (2024). Read the source
• Chen A. Best Practices for Medicaid Coverage of Doula Care. National Health Law Program (last updated September 2026). Read the source

More for professionals: the Professionals Blog.