Many health care policy researchers believe that non-physician clinicians, such as physician assistants, nurse practitioners and midwives can help to reduce cost while maintaining quality. Midwifery has gained in popularity over recent years. Groups such as the American Public Health Association, Public Citizen and the National Organization for Women all support increased access to midwifery care.
The question remains whether or not midwife care during pregnancy is inferior, equivalent, or superior to physician care. Amalia R. Miller attempts to answer this question in her 2006 B.E. Press paper.
Differences between midwives and physicians
- Skills: Physicians receive an MD while midwife training is comparable to a nursing background. Only physicians can preform surgery such as a cesarean delivery.
- Financial Incentives: Physicians have a financial incentive to recommend cesareans since they will reap the financial rewards of preforming the surgery. Since midwives do not receive any compensation from surgery, they may be more likely to look out for the best interest of the patient.
- Attitude toward childbirth: The midwifery model of care views birth as a natural process and gives the mother more input towards shaping the birth experience. The physician’s medial approach “…highlights the risks of childbirth, viewing the event as inherently medical, even pathological, requiring hospital admission and technological intervention.”
One problem with measuring the quality of care received by those who use midwives is that of selection bias. For instance, healthy women may be more likely to use a midwife. Pregnant women with serious complications will be more likely to use physicians. Thus, a researcher may erroneously conclude that pregnant women treated by physicians are more likely to have cesareans, when a more appropriate conclusion would be that physicians treat a less healthy patient base and thus preform more cesareans.
The best study preformed to date is a Chambliss et al. (1992) paper. The authors conducted “[a] randomized blinded clinical trial was conducted in which 492 low-risk patients were assigned to either physician or midwifery management.” Unlike most non-randomized trials, the authors found a small positive relationship between midwifery and cesarean rates.
The Miller (2006) paper looks at how midwifery affects cesarean rates. They use 1989-1999 Natality Detail Files as well as the 1989-1999 March Current Population Surveys (CPS) as her data source. The author considers a simple OLS regression using cesarean rates as the dependent variable and the use of midwifery–along with other covariates–as the independent variables. The problem of selection bias remains.
Thus, the authors use the enactment of Any Willing Provider laws as a proxy for state-wide midwifery usage. These laws “prohibit discrimination against a class of providers.” While Any Willing Provider laws are not directly related to midwives, the authors also use specific state midwifery reimbursement laws to proxy for midwifery usage. The claim made by Ms. Miller is that these laws are exogenously enacted; this creates a natural experiment and allows a difference-in-difference regression.
Ms. Miller finds that midwifery reimbursement laws, unsurprisingly, do increase midwifery usage by pregnant women but more general Any Willing Provider laws do not alter midwifery usage rates. The authors continues to state that “…The main finding of the Chambliss et al. (1992) study is confirmed: there is no evidence that the expansion of midwifery led to a reduction in cesarean section rates. Hence, the results from the small randomized trial appear to generalize to the population at large, while the non-random trials likely suffered from selection bias due to inadequate controls for patient health and preferences. ”
- Miller, Amalia R. (2006) “The Impact of Midwifery-Promoting Public Policies on Medical Interventions and Health Outcomes,” Advances in Economic Analysis & Policy: Vol. 6 : Iss. 1, Article 6.
- Chambliss LR, Daly C, Mederaris AL, Ames M, Kayne M, Paul R (1992) “The role of selection bias in comparing cesarean birth rates between physician and midwifery management” Obstetrics and gynecology, v80(2) pp. 161-165.