Skip to main content

Pediatric Bioethics

July, 2018

Different Thresholds for Overriding Parental Refusals of Life-sustaining Treatment

Author: Hannah Gerdes, Intern, Children's Mercy Bioethics Center John D. Lantos, MD | Director of Pediatric Bioethics | Professor of Pediatrics

When should doctors seek protective custody in order to override a parent’s refusal of potentially lifesaving treatment for their child? 

The answers seems to differ in different clinical situations, for different pediatric subspecialties, at different hospitals, and even between individual doctors. Most neonatologists would defer to parental requests to forego resuscitation for babies born at 23 weeks of gestation. Most would not go along with such requests at 24 weeks.1  At 23 weeks, survival rates are about 50 percent. By 24 weeks, they rise to 75 percent.2    Thus, the threshold for mandating treatment of premature babies seems to be a survival rate of 25-50 percent. This is not the case in pediatric cardiology. We recently surveyed cardiologists and cardiovascular surgeons at Children’s Mercy regarding their willingness to support a parent’s refusal of treatment for their children with congenital heart defects. The vast majority of cardiologists would not pursue legal action when parents refuse unless the anticipated survival rate after surgery is 95 percent. No comparable systematic study on compelling treatment exists in pediatric oncology, but there are case reports of physicians requesting and obtaining protective custody for cancer treatment when the reported mortality rates are 40-50 percent.3    

How can these differences be explained?

One explanation is that survival is not, and should not be, the only factor that is considered. Treatment in all three areas is associated with increased risk of long-term morbidities and with treatments that are painful, prolonged and burdensome. In cardiology, surgeries for correcting congenital heart defects are highly successful, but surgeons may feel responsible for poor outcomes that follow major operations. Those feelings make them less likely to override parental refusals if there is a moderate chance of death or other outcomes.4   For premature babies, another factor may be relevant. The range of possible outcomes for a surviving premature baby is quite wide. Some babies survive with no neurological or physical sequelae. Others are neurologically devastated and have chronic lung disease, blindness, seizures, behavioral problems or other chronic health concerns. Perhaps this uncertainty is a factor in physicians’ unwillingness to override parental refusals of treatment for babies born at 23 weeks. By 24 weeks, though, the higher rates of survival mean that, overall, the likelihood of a good outcome crosses some perceived ethical (and legal) threshold. In oncology, it is harder to generalize. Many people have negative perceptions of cancer and, particularly, of cancer chemotherapy. So, even though advances in medical technology render most cancers far more “curable” than they ever have been, there may be more conflict between physicians and parents on the correct course of action for a child with cancer, reflecting these conflicting perceptions. 

When parents refuse lifesaving treatment, and doctors accept that refusal, the doctors then usually recommend palliative care. Sometimes, palliative care is directed toward symptom relief in a child with a complex chronic condition. At other times, when a child is clearly dying, palliative care means “comfort care only” or “hospice care.” Discussions get complicated when parents refuse life-sustaining treatment, but also do not want palliative care. In some cases, parents request complementary and alternative medicines, instead of standard medical treatments. For doctors, the decisions about whether to accommodate such parental requests can test the limits of our professionalism. Further studies are needed to better understand how doctors negotiate the complex process of shared decision-making in these ethical gray zones. 


  1. Survival of Infants Born at Periviable Gestational Ages. Patel RMRysavy MABell EFTyson JE.  Clin Perinatol. 2017;44:287-303.

  2. Between-hospital Variation in Treatment and Outcomes in Extremely Preterm Infants. Rysavy MA, Li L, Bell EF, Das A, et al.  N Engl J Med. 2015;372:1801-11. 

  3. Refusal of Treatment of Childhood Cancer: A Systematic Review. Caruso Brown AESlutzky AR.Pediatrics. 2017 Dec;140(6).

  4. Risk Taking and Tolerance of Uncertainty: Implications for Surgeons.  Tubbs EPElrod JAFlum DRJ Surg Res. 2006;131:1-6.