The Future of Children, Princeton - Brookings: Providing research and analysis to promote effective policies and programs for children.

Obstetricians are very dependent on technology in their practice of medicine, and thus it is essential that controlled trials be carried out for new techniques before they are widely distributed in the marketplace. This ideal is rarely accomplished. We must carefully evaluate the scientific evidence on technology to prevent low birth weight and preterm birth in order to make policy decisions about which technologies should be encouraged and evaluated because of their potential to improve outcomes and which technologies should be discouraged because they do not make a difference.

Home uterine activity monitoring is an example of a system that was disseminated before it was shown to be effective. Most published studies have serious methodologic flaws, prompting the American College of Obstetricians and Gynecologists to advise that it remain investigational. This is not to say that HUAM has no role in obstetrics, but rather that further study is necessary to define that role. Perhaps HUAM will play an important and cost-saving role in the management of twins and triplets or other high-risk groups for preterm labor at home and actually save money in the long run. These are the types of issues which should have been addressed before this technology was released for general use without adequate data to define its role.

Tocolytics are an example of a technology which is so widely accepted and ingrained in medical practice that few clinicians question its usefulness. Many would be surprised to find that little benefit on a population basis has been found for these drugs beyond lengthening a pregnancy by about two days. With the combination of corticosteroids and tocolytics, great benefit has been found, yet corticosteroids are not nearly so widely utilized. This is beginning to change, as more clinicians become aware of the unequivocal benefit of steroids on neonatal outcomes. The reluctance on the part of many clinicians to use corticosteroids is somewhat baffling, but may be because obstetricians do not see the immediate benefit of steroids. It is hoped that the dissemination of the results of the recent Consensus Conference sponsored by the National Institutes of Health will stimulate the use of this life-saving technology.³⁰

Bed rest in the hospital to prevent preterm delivery is an example of an intervention which makes intuitive sense and is seemingly simple to carry out but is based upon little or no empirical evidence that it actually makes a difference. Given the exorbitant costs of a hospital stay and the social disruption that it causes, with little effect on preventing preterm birth, there is no reason to prescribe prophylactic bed rest. However, many physicians continue to recommend bed rest for women at risk for preterm birth. Many interventions in medicine seem to make intuitive sense, and we must condition ourselves to continue to be scientific and test interventions before widely prescribing them simply because they seem to make sense.

Cesarean delivery for vertex, low birth weight infants is an example of a technology which was tested and found not to make a difference in outcome. Today, few institutions routinely perform cesarean deliveries on women who have a low birth weight infant in the vertex position. Likewise, episiotomy to protect the fetal head has become outmoded as the evidence mounted against its usefulness. The use of cesarean delivery for breech low birth weight infants is common practice, but it remains a controversial procedure. Randomized clinical trials are needed to test its efficacy before valid conclusions can be drawn.

Multifetal pregnancy reduction is a technology that is relevant for only a very small number of women and, as such, will not have a major impact on mortality. However, for individual parents, it may be vitally important. To a parent who does not believe in abortion, it may be a tremendous relief to know that reduction from triplets to twins will do little if anything to improve their neonates' outcomes. To a parent who cannot fathom the financial and emotional impact of bringing up four possibly impaired children, it may also be a tremendous relief to know that there is an intervention which is safe and effective in improving outcomes. Technology should be used not only to improve outcomes of multifetal pregnancies, but also to facilitate the lives of the parents taking care of the children created in this way. Ideally, in the future, reproductive technology will improve, and there will be a dramatic reduction in multiple births.

Cervical cerclage is an example of a technology that, despite being widely accepted, has never been tested in a truly randomized controlled trial. It is a technology with relatively minimal morbidity; therefore, the physicians of desperate couples who strongly desire children are often unwilling to take the chance that they will be assigned to the control group which does not receive the intervention. While there does appear to be a benefit for those women with true cervical incompetence, it is very difficult to diagnose this condition accurately. Thus, many more women are diagnosed and unnecessarily treated with cerclage than truly have the disease. In fact, 25 women suspected to have cervical incompetence must undergo cerclage suturing to see a benefit in just one of those 25 women. Finding a way to diagnose true cervical incompetence will be the key to the successful use of this technology.

The reasons physicians use unproven technology or ignore proven ones is unclear. At present there is no easy and effective way to modify physician behavior by encouraging the use of effective technologies such as corticosteroids and discouraging the use of ineffective or unproven technologies such as bed rest or home uterine monitoring. Limiting the availability of new technologies before they are proven to be effective is a first step. But it is difficult to remove familiar technologies despite their proven ineffectiveness. In the current climate of cost consciousness, it is imperative that the widespread use of these unproven and ineffective technologies be abandoned and that the use of proven technologies be encouraged. In addition, we must be careful not to use proven technology in those situations where it does not improve outcomes. Dissemination of technology into low-risk populations has the potential to do more harm than good. Finally, physicians need to be educated to be wise consumers of medical technologies. Up-to-date information, critical reviews, and a systematic synthesis of the thousands of trials that have evaluated obstetric technologies have been electronically compiled in the Oxford Database of Perinatal Trials⁷⁸ and published in Effective Care in Pregnancy and Childbirth⁷⁹ and A Guide to Effective Care in Pregnancy and Childbirth.⁸⁰ These important resources provide physicians with unbiased, clear recommendations about which technologies are effective and warn against using those that are not.

The use of obstetrical technologies appears to have had little impact on reducing the occurrence of low birth weight or preterm births. Nationally, rates of low birth weight and preterm birth have not changed appreciably in the past 25 years despite the introduction of many obstetrical technologies, procedures, and drugs. The reason for this apparent contradiction is that many of the currently used technologies are not effective. There is no effective way to stop the progress of labor once it has started, and attempts to devise an early detection system have generally been unsuccessful. One major handicap in developing methods to stop the progression of labor is that the mechanisms that produce preterm labor are not well understood. Much more work is needed to discover just how and why labor is initiated before new preventive technologies can be developed.