|
 |
In DepthStillbirth Assessment in the Era of Managed CareRichard M. Pauli, M.D., Ph.D.
Introduction
Why might data from the Wisconsin Stillbirth Service Program prove advantageous in assessing the utility of stillbirth evaluation? Well, we have previously shown that this population, which currently numbers around 1200 stillbirths, is an unselected and unbiased series, truly representative of all stillbirths in the population. There has been virtually uniform, unselective and comprehensive evaluation of all of the infants within the series. Finally, because it is community based, the resultant information is analogous to what can be anticipated regarding available local expertise within the community. The last point is of some relevance, since I think it allows us to argue that data from this source probably are a minimal estimate of the diagnostic benefits one can derive from stillbirth postmortem assessment. Benefits of Assessment in a Managed Care Setting
A second dimension relates to what potential benefits could or should derive from such postmortem evaluation. That is, what kinds of benefits should accrue for patients, care providers and care organization? That list includes: The rights and benefits of parents to know why a baby is stillborn. Diagnosis may affect subsequent parental reproductive decisions either positively or negatively. Sometimes assessment will point to prenatal diagnosis or modify plans for prenatal management, or for perinatal management. Such studies may decrease inappropriate litigation. And finally, assessment can function as a means of providing a foundation for assessment of the quality of pre- natal and perinatal care. Levels of Current Evidence: Anecdotal Evidence
Anecdotes and, mostly uncontrolled, data abound concerning grieving following intrauterine death. Few would argue with the statement that grieving is an inevitable accompaniment of intrauterine death. Furthermore there is considerable evidence to support the idea that the grieving following stillbirth, its character, its intensity and its duration are entirely analogous to that following the death of a child at any age. Stillbirth sometimes may be even more difficult than the usual death of a child. A stillborn is a ‘nonperson’ with no past to mourn, who has never been ‘experienced’. When our son, Zachary, was stillborn, I had such desperate feelings of grief and longing -- waves of longing that persisted for months -- to have him with us, to be able to hold him. And one of the hardest realities to deal with was that we would never know why it was that Zachary died. That uncertainty can introduce feelings of guilt, blame, anger and remorse into what is already a difficult, sometimes overwhelming, grieving process. Indeed, during the resolution of the grieving process parents have two primary questions: "Why did this happen? Will it happen again?" And parents’ satisfaction with their care following the birth of a stillborn infant is directly correlated with their level of understanding of the answers to these two questions. Such answers will often be forthcoming only if appropriate evaluation occurs following a stillbirth. How might parental reproductive decision making be influenced by postmortem diagnostic assessment? Specific causes may carry with them recurrence risks anywhere from 0 to 100%. Identification of a fetal cause of stillbirth does not usually imply high risk for subsequent recurrence. In fact, far more often such identification yields an estimated risk for recurrence less than one would estimate based upon population data for all stillborns. Obviously, recognizing those instances in which couples have little or no risk for recurrence is just as important as those in which risks are high. Diagnoses can certainly modify and focus the form of appropriate prenatal diagnostic efforts in next pregnancies. One baby whom we helped to evaluate had what appeared to be a bilateral cleft lip. In isolation this would imply only the option of ultrasonographic assessment in a next pregnancy. Further morphologic investigation showed that the baby also had telecanthus, downslanting palpebral fissures, prominent globes, asymmetric fingers and a left clubfoot — that is, that this likely was a multiple congenital anomalies process rather than an isolated birth defect. Xrays demonstrated that the maxillary abnormality was a midline defect — suggesting that the facial features could be reflective of a holoprosencephalic brain malformation. Post mortem studies showed, in addition, the presence of a large ventricular septal defect. Cytogenetically the infant was found to have additional material, of unknown chromosomal origin, on the tip of the short arm of one chromosome 5. This, in turn, precipitated parental chromosomal assessment, showing that the father had a balanced 5p15/6q23 translocation — implying marked risk in next pregnancies. Of course, that also clearly has modified what prenatal diagnostic assessment could be undertaken — chorionic villus sampling or amniocentesis for cytogenetic assessment rather than sequential morphologic ultrasound evaluation. How might prenatal and perinatal management be modified? Here is a telegraphic history of one couple — in August of 1986 an 18 week gestation pregnancy loss was experienced. The fetus was edematous but otherwise apparently normal. No placental examination was completed. In July of 1987 a 10 week gestation loss occurred. No fetal or placental studies were carried out. Again in August of 1988 intrauterine death occurred at 15 weeks gestation. While a placental abruption was suspected, additional postmortem evaluation showed multiple regions of placental infarction as well as intervillous fibrin deposition. Incidentally, history in the mother for the first time revealed Raynaud-like symptoms in her hands. She was found to have persistent presence of lupus anticoagulant activity. The next pregnancy included more intensive, sequential prenatal evaluation and maternal aspirin therapy and resulted in delivery in June of 1990 of a healthy, term male baby. Can appropriate, uniform and com-prehensive postmortem assessment result in a decrease in the likelihood of inappropriate litigation? Certainly our society has become impressively litigious. Certainly stillbirth can engender feelings of anger and blame. Clearly, suits for wrongful death following fetal death are not prohibited: at least 10 States have allowed wrongful death suits for stillbirth. How then might managed care organizations decrease the probability of such suits being brought? Although there is no direct evidence that can be brought to bear concerning stillbirth per se, I think that one can impute the following — First, there is clear evidence in the literature that the probability that a malpractice proceeding will be initiated is not solely or even primarily the consequence of the practice of bad medicine. Rather it is inversely correlated with the family’s perception of the adequacy of physician-patient communication and is inversely correlated with how knowledgeable a patient has been made, how much they understand about the bad outcome that they have suffered. Recall then, that parents have two primary questions - Why and Will it happen again; that their satisfaction with their care is directly correlated with their level of understanding of the answers to these two questions; and, of course, that to find answers will often require postmortem evaluation. If all of those things are so, then there is a clear implication that wrongful death actions will less likely arise if uniform, comprehensive postmortem evaluation is carried out after every stillbirth. From an economic perspective, after all, one suit is enough to justify hundreds of such evaluations. Levels of Current Evidence: Expert Consensus and Evidence-Based Recommendations
Recently published is a statement of the American College of Obstetrics and Gynecology.* It is not a perfect statement. It is a rather typical consensus statement — in particular, it lacks any real evidence basis but rather mostly says — we all think this is good medicine. Nonetheless, we should celebrate its generation since it should make much more difficult the practice of obstetrics and perinatal medicine without assurance that such post-stillbirth assessments are carried out. Far better, in my opinion, but still to come is an evidence-based summary of the justification for such care. I am currently working on the generation of such an evidence-based set of recommendations from the American College of Medical Genetics.** Such an evidence-based piece will, I hope, be difficult for any health care organization to ignore. I have begun a reconstruction of the data available through the Wisconsin Stillbirth Service Program that may provide the objective evidence of benefit that is so badly needed. If we are to argue that postmortem evaluation of stillborns is critical in reproductive decision making, in prenatal diagnosis, in prenatal and in perinatal management, then it would be terribly helpful to have quantitative evidence to that effect. On the basis of generated diagnoses, I reviewed the first 1196 stillbirths with which we have helped with assessment. In this reconstruction, I asked, "given the diagnoses generated, can one expect that there has resulted — a change in estimated risks in next pregnancies; or, a change in optimal prenatal diagnostic management; or a change in recommendations regarding prenatal treat-ment; or a change in what would be recommended for perinatal and neonatal care. Reconstruction from WiSSP Data
This is as close as anyone has currently come to generating evidence-based data on the utility of such postmortem evaluative efforts [see Table I]. In 37% of stillborns, generated diagnoses can be anticipated to change the estimate of re-currence risk in next pregnancies. In nearly 20% we will provide information that will allow for more precise, or appropriate or directed prenatal diagnostic effort. In a smaller proportion, generated diagnoses will have effect on prenatal management — in around 5%, or in perinatal or neonatal care — in around 2%. Even these latter figures are substantial if the cost of generating such diagnoses is not excessive. Levels of Current Evidence: Cost
On the other hand, one certainly can make arguments from the cost side of the ledger. Managed care organizations are not concerned about traditional charges. Rather, since most assessments will occur internally or through contract, the relevant questions relate to real costs. So, how much does it cost to comprehensively assess a stillborn in search of fetal processes that might have caused the intrauterine death? Table II shows a summary of what seem to be the best estimates of average costs in our state. It would probably be of value to generate a similar real cost list within each managed care organization. In sum, though, one can estimate that real costs should likely be in the vicinity of $1,000 per assessment. Is that excessive or justifiable? That cost translates into about $8.00 per pregnancy. That, then, is the cost that would need to be added to the package of prenatal/perinatal/neonatal costs in order to fund comprehensive evaluation of every stillborn baby. Or, another way to calculate this is by yearly cost per covered life. Ten cents per covered life would be the additional expense of having such evaluations completed. Such economic arguments may carry substantial weight with managed care administrators, if there is a concomitant understanding of the benefits that can accrue — as we have already outlined. Summary and Prospects
The case could be stronger still — a true cost-benefit analysis would be enlightening. I would like to quantify the psychological benefits of generation of a specific diagnosis. Prospective evaluation, rather than retrospective reconstruction as I have done, would yield even more certain evidence of the level of benefit of such evaluations. In the meantime, I hope that the approach I have presented and perhaps some of the data I have demonstrated will help in assuring that, even in cost conscious managed care settings, families who have suffered the death of their baby will be assured of complete and accurate information about why their baby died and the probability that this might happen again. Table II:
*A copy of the ACOG statement is available from WiSSP. ** A copy of the working outline for this evidence-based statement is available from WiSSP. WiSSPers is published quarterly by the Wisconsin Stillbirth Service Program. Richard M. Pauli, M.D., Ph.D., Director
This newsletter is not copyrighted. Readers are free to duplicate all or part of the contents. In accordance with accepted publication standards, we request acknowledgement in print of any article reproduced in whole or in part in another publication. Send comments, letters, address changes and all other correspondence to: Wisconsin Stillbirth Service Program
The WiSSP is supported by Maternal and Child Health block grant funds from the State of Wisconsin, Division of Health. The opinions expressed represent those of the authors and do not necessarily reflect those of the WiSSP staff.
WiSSPers newsletter cover page
|