Negative Studies and the Future of Prenatal Counseling at the Margin of Gestational Viability

Despite advances in the care of extremely preterm infants, the risks of mortality and significant morbidity remain high and difficult to predict with certainty.1Bell E.F. Hintz S.R. Hansen N.I. Bann C.M. Wyckoff M.H. DeMauro S.B. et al.Mortality, in-hospital morbidity, care practices, and 2-year outcomes for extremely preterm infants in the US, 2013-2018.JAMA. 2022; 327: 248-263Crossref PubMed Scopus (126) Google Scholar,2Shukla V.V. Rysavy M.A. Das A. Tyson J.E. Bell E.F. Ambalavanan N. et al.Outcome prediction in newborn infants: past, present, and future.Semin Perinatol. 2022; 46: 151641Crossref PubMed Scopus (1) Google Scholar Obstetricians and neonatologists have long partnered to counsel expectant parents facing extremely preterm birth, unified by the foundational principle that, in some cases, outcomes are so uncertain and potentially so bad that parents should have the ultimate say in whether resuscitative efforts are initiated in the delivery room.3Berger T.M. Decisions in the gray zone: evidence-based or culture-based?.J Pediatr. 2010; 156: 7-9Abstract Full Text Full Text PDF PubMed Scopus (15) Google Scholar For decades, we have obsessed over how we ought to define the borders of the mythical gray zone of gestational viability, in a seemingly quixotic effort to define sharp gestational age edges to the zone of parental discretion. This pursuit has led to more questions than answers and shed light on the near impossibility of defining the grey zone at a population level. Along the way, we have learned that defining the conditions under which parents should decide about initiating resuscitation tells us essentially nothing about how parents are meant to make that decision (or how we are meant to help them).4Janvier A. Lorenz J.M. Lantos J.D. Antenatal counselling for parents facing an extremely preterm birth: limitations of the medical evidence.Acta Paediatr. 2012; 101: 800-804Crossref PubMed Scopus (81) Google Scholar Neonatologists (and other clinicians) are asked to provide prenatal consultation to expectant parents facing extremely preterm birth to provide anticipatory guidance and support decision-making about whether resuscitative efforts should be undertaken. This consultative skill was initially self-taught. However, normative and empirical research on prenatal consultation, focused largely on periviability decision-making, is a growing field owing to the important and correct observation that there is much more to learn about how to do it, how to teach it, and how to measure its effectiveness.5Danziger P. Laventhal N. Prenatal consultation: perspectives on training, relevance, and utilization among pediatric subspecialty program directors.J Perinatol. 2018; 38: 989-996Crossref PubMed Scopus (6) Google Scholar, 6Stokes T.A. Watson K.L. Boss R.D. Teaching antenatal counseling skills to neonatal providers.Semin Perinatol. 2014; 38: 47-51Crossref PubMed Scopus (50) Google Scholar, 7Myers P. Andrews B. Meadow W. Opportunities and difficulties for counseling at the margins of viability.Semin Fetal Neonatal Med. 2018; 23: 30-34Abstract Full Text Full Text PDF PubMed Scopus (11) Google Scholar From the beginning of our clinical training, we are taught that information is at the heart of medical decision-making and informed consent. In the see one, do one, teach one era of training for prenatal consultations, we assumed that parents would make decisions about the initiation of resuscitation based on the outcome statistics that we dutifully regurgitated in a 1-way monologue. We now know that the communication of numerical outcomes data is much harder than it sounds and continue to search for best practices for relaying health information.7Myers P. Andrews B. Meadow W. Opportunities and difficulties for counseling at the margins of viability.Semin Fetal Neonatal Med. 2018; 23: 30-34Abstract Full Text Full Text PDF PubMed Scopus (11) Google Scholar, 8Reyna V.F. Nelson W.L. Han P.K. Dieckmann N.F. How numeracy influences risk comprehension and medical decision making.Psychol Bull. 2009; 135: 943-973Crossref PubMed Scopus (857) Google Scholar, 9Fagerlin A. Zikmund-Fisher B.J. Ubel P.A. Helping patients decide: ten steps to better risk communication.J Natl Cancer Inst. 2011; 103: 1436-1443Crossref PubMed Scopus (411) Google Scholar, 10Guillén Ú. Mackley A. Laventhal N. Kukora S. Christ L. Derrick M. et al.Evaluating the use of a decision aid for parents facing extremely premature delivery: a randomized trial.J Pediatr. 2019; 209: 52-60.e1Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar However, the role of probability of estimates of death or morbidity in parental decision-making for delivery room resuscitation has been hotly debated; some argue, based on a large body of evidence, that outcome statistics simply are not the basis of parents' decisions about whether their newborns should be resuscitated, and that the focus should be on relational skill (in pursuit of helping parents articulate what is important to them), not outcomes communication methods.11Haward M.F. Janvier A. Lorenz J.M. Fischhoff B. Counseling parents at risk of delivery of an extremely premature infant: differing strategies.AJOB Empir Bioeth. 2017; 8: 243-252Crossref PubMed Scopus (22) Google Scholar, 12Arzuaga B.H. Cummings C.L. Deliveries at extreme prematurity: outcomes, approaches, institutional variation, and uncertainty.Curr Opin Pediatr. 2019; 31: 182-187Crossref PubMed Scopus (8) Google Scholar, 13Haward M.F. Gaucher N. Payot A. Robson K. Janvier A. Personalized decision making: practical Recommendations for antenatal counseling for fragile neonates.Clin Perinatol. 2017; 44: 429-445Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar In their paper entitled, Effect of Presenting Survival Information as Text or Pictograph During Periviable Birth Counseling: A Randomized, Controlled Trial, McDonnell et al14McDonnell S.M. Basir M.A. Yan K. Liegl M.N. Windschitl P.D. Effect of presenting survival information as text or pictograph during periviable birth counseling: a randomized, controlled trial.J Pediatr. 2023; : 113382Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar bring robust research methods to this important and vexing topic in contemporary clinical neonatology by studying the interaction of probability communication format and values-based decision making. The authors enrolled 1052 18- to 45-year-old US women contacted through an internet-based survey company to complete electronic surveys for hypothetical scenarios in which they were 22 weeks' pregnant and in preterm labor, being counseled by a physician about a resuscitation decision. Participants were randomized to 1 of 6 groups in which either a 30% or 60% survival likelihood was presented (both with two-thirds of survivors having a disability), and 1 of 3 communication formats: text only, static pictograph, or iterative pictograph. Participants were asked to choose a treatment plan (intensive care vs palliative care) and then provide a recall of the survival estimates. The investigators also assessed participants' intuitive beliefs for their hypothetical baby's chance of survival on a 100-point scale from no chance of survival to definitely will survive. Seventy-five percent of participants chose initiation of intensive care for a hypothetical infant born at 22 weeks, which seems considerably higher than published rates of resuscitation at this gestational age, although it is exceptionally difficult to pin down how many infants born at 22 weeks are actively resuscitated.15Feltman D.M. Fritz K.A. Datta A. Carlos C. Hayslett D. Tonismae T. et al.Antenatal periviability counseling and decision making: a retrospective examination by the investigating neonatal decisions for extremely early deliveries study group.Am J Perinatol. 2020; 37: 184-195Crossref PubMed Scopus (14) Google Scholar,16Smith L.K. Morisaki N. Morken N.H. Gissler M. Deb-Rinker P. Rouleau J. et al.An international comparison of death classification at 22 to 25 weeks’ gestational age.Pediatrics. 2018; 142e20173324Crossref Scopus (23) Google Scholar There was no difference in choice for intensive care in the high (60%) vs low (30%) survival estimate groups, no difference in choice by probability communication format, and no interaction between physician survival estimate and communication format. In other words, the chance of survival in this hypothetical scenario did not impact participants' choices for resuscitation and use of evidence-based tools to improve health communication made no difference; it is not that they did not understand the information. One might point to the modest recall of the information and question this—only 44% of participants reported a survival estimate within 10% of the actual estimate provided, but even among those with perfect recall, survival estimates did not impact treatment choice. Study participants' intuitive beliefs of how their theoretical 22-week baby would do were also not impacted by physician survival estimate, presentation format, or the interaction between the two. These seemingly nihilistic study results are illuminating. Should we be spending our time trying to get better at sharing outcome probabilities with expectant parents if the probabilities are not what matter to them? Some of the participant characteristics that were measured did seem to be associated with differences in resuscitation decisions. White race, previous experience as a parent in the neonatal intensive care unit, intrinsic religiosity, and nonorganizational religiosity were associated were associated with a choice to pursue intensive care over palliative care. One wonders a bit about these findings, because they were not the primary study outcomes and did not necessarily group together intuitively; for example, organizational religiosity and prioritization of sanctity vs quality of life were not associated with a difference in resuscitation preference. However, the investigators performed a classification and regression tree analysis to model experimental conditions and participant characteristics to assess their explanatory power on the binary outcome of treatment choice (intensive care vs palliative care) with interesting results. Participants' intuitive prediction of their hypothetical 22-week baby's chance of survival had the most explanatory power of any variable. This process is well-explained visually in Figure 6 and forces us to reckon with neonatologists' entrenched and consequential pessimism regarding outcomes for infants born extremely preterm.17Wilkinson D. The self-fulfilling prophecy in intensive care.Theor Med Bioeth. 2009; 30: 401-410Crossref PubMed Scopus (103) Google Scholar, 18Lawrence C. Laventhal N. Fritz K.A. Carlos C. Famuyide M. Tonismae T. et al.Ethical cultures in perinatal care: do they exist? Correlation of provider attitudes with periviability practices at six centers.Am J Perinatol. 2021; 38: e193-e200Crossref PubMed Scopus (6) Google Scholar, 19Myers P. Laventhal N. Andrews B. Lagatta J. Meadow W. Population-based outcomes data for counseling at the margin of gestational viability.J Pediatr. 2017; 181: 208-212.e4Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar, 20Blanco F. Suresh G. Howard D. Soll R.F. Ensuring accurate knowledge of prematurity outcomes for prenatal counseling.Pediatrics. 2005; 115: e478-e487Crossref PubMed Scopus (66) Google Scholar The authors frame this as optimism bias. Participants who chose intensive care tended to have higher median intuitive predictions of whether their baby would survive than those who chose palliative care, but it is difficult to know what to make of this when the median intuitive survival estimate for parents who chose palliative care was 58%. In another nod to optimism bias, the authors point to finding that, in both survival risk groups, intuitive estimates of survival probability were higher than the survival probability that they were given moments before. The difference is trivial in the 60% survival group (mean, 64.8%), but striking in the 30% survival group (64%). Given the other study findings, this is not an issue of the participants not getting it. The observed optimism might have something to do with the fundamental limitations of the prognostic information we provide. These numbers are derived retrospectively, telling parents about the proportion of babies like their baby who had a given binary outcome; the future outcome for this baby is fundamentally unknowable. Optimism (or hope), rather than incomprehension or denial, can thus explain a parent thinking that their baby's chance of survival is higher than the quoted statistic. This explanation is distinct from a conclusion that parents simply do not need survival probabilities to make decisions about resuscitation at the margin of viability; rather, the optimism bias explanation might be that parents reject the survival probabilities that we provide in favor of their own. In the pursuit of evidence-based periviability counseling, we should be extremely cautious about ever starting a statement with “parents want...,” as though a single approach will meet the needs of all families. McDonnell et al have shown us that best practices for conveying outcomes probabilities do not seem to impact expectant parents' decisions about initiation of intensive care in hypothetical scenarios; rather, these decisions might have more to do with whatever parents bring with them into the encounter. This does not mean that the information has no value—it is possible that some parents still want to know what happens to babies like theirs. Framed the right way, best practices for sharing this information with parents who want it remains an important area of study. For parents guided by other values, beliefs, and considerations, we still have more to learn about how best support their decision-making. Thoughtfully designed and rigorously conducted studies like the one published in this issue of The Journal are the path to solidifying prenatal consultation as an essential, evidence-based component of modern care of extremely preterm infants. Effect of Presenting Survival Information as Text or Pictograph During Periviable Birth Counseling: A Randomized, Controlled TrialThe Journal of PediatricsVol. 257PreviewTo examine whether presenting a 30% or a 60% chance of survival in different survival information formats would influence hypothetical periviable birth treatment choice and whether treatment choice would be associated with participants’ recall or their intuitive beliefs about the chances of survival. Full-Text PDF