by Christine H. Morton, PhD and Monica Basile, PhD, CPM, CD(DONA), CCE (BWI)
Last month there were great discussions after a study was published by the University of Minnesota, examining the potential cost savings to Medicaid if doulas worked with Medicaid clients, helping to reduce interventions and cesareans. Today and next Tuesday, regular contributor, Christine Morton and her colleague Monica Basile, take a look at that study and another from Oregon, and share thoughtful insight about topics that might still need to be addressed if costs savings were to be effectively realized in a two part blog post. – Sharon Muza, Community Manager, Science & Sensibility
How can doula supported births help reduce the cesarean rate and realize cost savings within Medicaid-funded births? Two studies published last month offer the opportunity to address this complex question.
We support the goal of increasing access to doula supported care to childbearing people of diverse racial/ethnic and class backgrounds, and we are pleased that discussions are taking place about how doulas may be able to help reduce racial disparities in maternal and infant health. We recognize that work toward these goals requires policy advocacy, which depends heavily on economic arguments for the benefits of doula care.
However, by limiting the discussion of benefits to the economic impacts of reduced cesareans, advocacy for Medicaid funding of doula supported births—without specifying the doula model of care and without according true value to the doula’s impact—may have unintended consequences for individual doulas, and the organizations that represent them. One such consequence may be that the resulting system will continue to perpetuate a model of economic marginality and potential exploitation for the doulas who serve a low income population of childbearing people.
The AJPH study by Katy Kozhimannil and colleagues in Minnesota received a lot of media attention when it appeared last month, even live coverage in the Huffington Post. This study compared 1,079 selected Medicaid doula patients in Minnesota to Medicaid patients nationwide for their total cesarean rates. They found that doula clients of a community program in Minnesota had a rate of 22.3% while national Medicaid had 31.5%. The authors reported three scenarios, all assuming that if states reduced cesarean rates, by offering doula services, there would be varying levels of cost savings, depending on the cesarean rate achieved, and by reimbursing doulas between $100-300 per birth.
In our view, the Minnesota study design raises several methodological questions, which are applicable to this study and to future research on doula-attended births. We outline those questions here, as well as raise several more substantive concerns about the implications of the study’s stated conclusions.
- Why did the researchers not compare Minnesota Medicaid doula clients to Minnesota Medicaid women who gave birth? Minnesota has a much lower rate of total cesarean that the US as a whole (27.4% during this time period), and this would have been a better matched comparison. A better comparison would be doula attended births vs. non-doula attended births at the same facility. It is not clear from the study whether the doula program whose data was utilized served women at one or multiple hospitals in Minneapolis.
- Why did the researchers not limit their investigation to primary cesareans? Doulas typically support women in labor rather than women undergoing repeat cesareans. The total cesarean rate includes repeat cesarean so it will be much higher than the primary cesarean rate, which is more applicable to doula clients. Including total cesarean rates means that the researchers are comparing a limited universe (doula support of women in labor) to all births (thus including repeat and primary cesarean). The data source for this study, (Nationwide Inpatient Sample), however, does not have this information.
- Cesarean rates are very dependent on the parity distribution of the birthing population, so first time mothers need to be compared to first time mothers and multiparous women to multiparous women. This information is not available in the data source used by the researchers, but in future studies of this type, it is critical to verify that the proportion of each is the same in the intervention and control populations.
- States are implementing a number of payment reform models to reduce cesareans among women covered by Medicaid, with limited success. In part, that is because cesareans are influenced by a number of factors, with payment incentives only one. (Many of these issues are covered in the CMQCC white paper on improvement opportunities to reduce cesareans, which argues that a multi-pronged strategy is necessary).
- Because hospital rates of cesarean have been shown to have high geographic variation in a number of studies (Baicker 2006; Main et al 2011; Caceres 2013; Kozhimannil 2013), it may be more feasible to have comparison groups of hospitals with similar primary cesarean rates. Until we understand what accounts for variation in cesarean rates between institutions (unit culture; facility policies and protocols), it may be premature to assess the independent effect of labor support by a trained doula.
While doula support is associated with fewer cesareans across the board (Hodnett 2012), the methodological issues described above are likely to over estimate the benefits of doula-attended births in terms of reducing the cesarean rate for Medicaid covered births. This, in turn, raises questions about the purported cost savings. In the Minnesota study, the cost breakpoint is no more than $300 dollars for the doula per birth. In most cities, doulas charge well above this amount for fee-for service care.
A cost-benefit analysis by Oregon Health & Science University researchers for the Oregon State Legislature was presented at the Society for Maternal Fetal Medicine in February 2013, which found that doula care in labor provides a cost benefit to payers only when doula costs are below $159.73 per case. In that study, data sources are not entirely clear, but do seem to come from the OHSU facility where a hospital-based doula program is in place. In that program, doulas are on call on weekends only and come to assist in a labor when requested by the woman during her prenatal care or when she arrives at the hospital. A case-control study claiming the benefits of this doula model at OHSU was published as an abstract, and although it claims “women receiving doula care were statistically less likely to have an epidural during labor (p = 0.03), have an episiotomy (p = .03), or cesarean delivery (p = .006) and on average, doula attended women had a shorter hospital stay compared to the control group (p = .002),” nowhere does it show what the actual rates were. This is important, because, they are likely to be relatively low overall, given that OSHU is a teaching hospital, with midwives and family practice physicians providing maternity care.
There are several types of doula models; not all have the same components. The community-based doula model, as exemplified by the HealthConnectOne approach has a solid evidence base. This model employs doulas who are trusted community members, and provides extensive prenatal and postpartum support in addition to continuous labor support. Doulas work collaboratively with community organizations, have extensive training in experiential learning and cultural sensitivity, and are paid a wage commensurate with their value and expertise, serving an important workforce development and grassroots empowerment function. Some so-called community doula programs do not incorporate all these components.
Hospital-based programs usually assign or utilize an on-call doula, who has not met the mother in advance and is not likely to follow up postpartum. Some advocates of Medicaid doula programs utilize the community health worker (CHW) model, which seems to mirror the community-based doula (CBD) model but with important differences. The American Public Health Association has defined CHWs as “frontline public health workers who are trusted members of and/or have an unusually close understanding of the community they serve.” Yet, despite their widespread utilization in public health over the past several years, the conditions of their training, job opportunities, and even job description are idiosyncratic, and highly varied, and this “lack of CHW identity and standards of practice has led employers to contribute to the confusion about who CHWs are and what they do.” While the CHW and CBD models offer important job opportunities to members of under-resourced communities, their wages are often on the low side, with full time work paying $35,000 to $42,000 annually. According to a health careers website, “CHWs often are hired to support a specific health initiative, which may depend on short-term funding sources. As a result, CHWs may have to move from job to job to obtain steady income. This short-term categorical funding of health services is a challenge to the stability and sustainability of the CHW practice.”
In cost-benefit or cost effectiveness studies, it is critical to clearly specify the doula model of care on which the economic model is based. It seems the doula model in the Minnesota study incorporates extensive pre and post partum contact and that there is an attempt to match doulas and clients in terms of race/ethnicity and language, but this is not always possible. The study does not indicate what the doulas in the Minnesota program were paid, however, and that information was unavailable on their website.
Before we move to the topic of reimbursement, we want to note that the type of doula model is critical for assessing the benefits of doula-attended births. The research clearly shows different outcomes for doulas who are affiliated with hospitals compared to those who work independently (Hodnett, 2012). If a cost benefit model shows little gain in terms of outcomes, or yields a price point in the low hundreds of dollars, it may be that findings are affected by the assumptions embedded in the calculations.
More fundamentally, however, we argue that doula benefits cannot be captured solely through an economic model. Neither should doulas be promoted as a primary means to reduce cesarean rates. Both strategies (economic benefits and cesarean reduction) for promoting doulas have significant barrier. In part two of this topic, running on Tuesday, April 2nd, we discuss our concerns about reimbursement and program sustainability alongside a caution against relying too heavily on arguments that position the doula as primarily a money saver and cesarean reducer.
Baicker, K, Kasey S. Buckles, and Amitabh Chandra. Geographic Variation In The Appropriate Use Of Cesarean Delivery: Do higher usage rates reflect medically inappropriate use of thisprocedure? Health Affairs 25 (2006): w355–w367; doi: 10.1377/hlthaff.25.w355
Caceres, Isabel A., Mariana Arcaya, et al., Hospital Differences in Cesarean Deliveries in Massachusetts (US) 2004–2006: The Case against Case-Mix Artifact, PLoS ONE 8(3): e57817. doi:10.1371/journal.pone.0057817
Hodnett ED, Gates S, Hofmeyr GJ, Sakala C. Continuous support for women during childbirth. Cochrane Database of Systematic Reviews, 2012, Issue 10. Art. No.: CD003766. DOI: 10.1002/14651858.CD003766.pub4.
Kozhimannil, Katy Backes, Michael R. Law, and Beth A. Virnig. Cesarean Delivery Rates Vary Tenfold Among US Hospitals; Reducing Variation May Address Quality And Cost Issues, Health Affairs 32, NO. 3 (2013): 527–535; doi: 10.1377/hlthaff.2012.1030
Main EK, Morton CH, Hopkins D, Giuliani G, Melsop K and Gould JB. 2011. Cesarean Deliveries, Outcomes, and Opportunities for Change in California: Toward a Public Agenda for Maternity Care Safety and Quality. Palo Alto, CA: CMQCC. (Available at http://www.cmqcc.org/white_paper)
Pilliod, Rachel; Leslie, Jennie; Tilden, Ellen; et al. Doula care in active labor: a cost benefit analysis. Abstract presented at 33rd Annual Meeting/Pregnancy Meeting of the Society-for-Maternal-Fetal-Medicine (SMFM), San Francisco, CA, February 11-16, 2013, American Journal of Obstetrics and Gynecology, Volume: 208 (1); S348-S349.
About the authors
Monica Basile has been an active birth doula, childbirth educator, and midwifery advocate for 17 years, and holds a PhD in Gender, Women’s and Sexuality Studies. Her 2012 doctoral dissertation, Reproductive Justice and Childbirth Reform: Doulas as Agents of Social Change, is an examination of emerging trends in doula care through the lens of intersectional feminist theory and the reproductive justice movement.
Regular contributor Christine H. Morton, PhD, is a sociologist whose research on doulas is the topic of her forthcoming book, with Elayne Clift, Birth Ambassadors: Doulas and the Re-emergence of Woman-Supported Birth, which will be published by Praeclarus Press in Fall 2013. For more on Christine, please see Science & Sensibility’s Contributor page.