DETERMINING DECISION THRESHOLDS FOR PHYSICIANS AND PARENTS FOR INSTITUTING A DO NOT ATTEMPT RESUSCITATION ORDER FOR PEDIATRIC INHOSPITAL CARDIAC ARREST PATIENTS. A CROSS-SECTIONAL STUDY OF PARENTS AND PHYSICIANS IN THE UNITED STATES

This cross-sectional study reveals that physicians in the United States require a significantly higher survival probability (5.3%) to recommend withholding resuscitation for pediatric in-hospital cardiac arrest patients compared to parents (1.2%), highlighting a critical gap in decision-making thresholds that has important implications for clinical counseling.

The Gist

Imagine you are standing at a fork in a very dark, foggy road. On one path, there is a tiny, flickering chance of reaching a bright, sunny destination (survival). On the other path, the road ends abruptly. The question is: How much of a chance do you need to see before you decide to keep walking down that dark road?

This study is all about finding that "magic number" for two different groups of people: doctors and parents of critically ill children.

The Experiment: A Game of "What If?"

The researchers didn't ask people about real-life tragedies. Instead, they created a series of "What If?" stories (called vignettes). They told doctors and parents scenarios like this:

"Imagine a child in the hospital whose heart stops. Based on the data, there is a X% chance they will survive and go home. Would you try to save them?"

They kept changing that X% (the odds) to see exactly where each group said, "Okay, that's too low. We shouldn't try anymore."

The Results: Two Different Compasses

Here is where the two groups parted ways:

1. The Doctors (The Pragmatic Navigators)
The doctors acted like cautious captains steering a ship through a storm. They calculated the odds carefully.

• Their Threshold: They generally said, "We should stop trying to resuscitate the child if the chance of them making it home is 5.3% or lower."
• The Reality Check: The study also found that doctors were often too optimistic about the weather. About 58% of them thought the chances of survival were higher than the actual data from national hospitals showed. They were looking at the horizon and seeing a bit more sun than was actually there.

2. The Parents (The Hopeful Guardians)
The parents, however, were willing to sail into much darker fog.

• Their Threshold: They were willing to try resuscitation even when the chance of success was as low as 1.2%.
• The Difference: To put it simply, if a doctor looks at a situation and says, "The odds are 4%, let's not do it," a parent might look at the same situation and say, "But 4% is better than 0%! Let's try."

The Big Takeaway

Think of it like buying a lottery ticket.

• The Doctor looks at the ticket and says, "The odds of winning are 1 in 20. That's not worth the money or the emotional toll. Let's walk away."
• The Parent looks at the same ticket and says, "But if I don't buy it, the chance is 0. I'll buy it, even if the odds are tiny, because I can't bear the thought of not trying."

Why This Matters

The study highlights a painful but important gap. When a doctor tells a parent, "The chances are too low to try," the parent might hear, "You are giving up." But the doctor isn't giving up; they are just using a different calculator.

The lesson for the real world: When doctors and parents talk about these life-or-death decisions, they need to realize they are speaking different languages. Doctors need to understand that parents are often willing to take risks at much lower odds than the medical guidelines suggest. And parents need to know that doctors aren't being cold; they are trying to be realistic based on data, even if that data sometimes gets the odds slightly wrong.

It's a reminder that in the most difficult moments, hope and statistics often walk in different directions, and good communication is the bridge between them.

Technical Summary

Technical Summary: Decision Thresholds for DNAR Orders in Pediatric Cardiac Arrest

1. Problem Statement

The decision to institute a "Do Not Attempt Resuscitation" (DNAR) order for critically ill children facing in-hospital cardiac arrest is a high-stakes, complex ethical dilemma. A critical gap exists in understanding the divergence between the perspectives of physicians and parents regarding the minimum survival probability required to justify attempting resuscitation. Furthermore, there is a lack of empirical data comparing physicians' subjective estimates of survival against objective data from nationwide registries. This study addresses the need to quantify these decision thresholds to improve clinical counseling and shared decision-making processes.

2. Methodology

The researchers conducted a cross-sectional survey-based study within the United States utilizing a mixed-methods approach:

• Study Population: The study involved 167 parents and 43 physicians.
• Data Collection Instrument: Participants were presented with clinical vignettes based on hypothetical scenarios. These vignettes were derived from the Get-With-The-Guidelines-Resuscitation (GWTG-R) registry to ensure clinical realism.
• Experimental Design:
  • Vignettes varied by hypothetical survival probabilities.
  • Participants were asked to indicate their decision to either "treat" (attempt resuscitation) or "withhold" (institute DNAR) at specific probability levels.
  • Physicians were also asked to estimate survival likelihoods to compare their subjective assessments against the registry data.
• Statistical Analysis: Decision thresholds were calculated using mixed-effect logistic regression models. This statistical approach allowed the researchers to determine the specific probability point at which the likelihood of choosing "withhold" crossed 50% for each group.

3. Key Contributions

• Quantification of Divergent Thresholds: The study provides the first quantitative comparison of the specific survival probability thresholds that trigger a decision to withhold resuscitation for parents versus physicians.
• Validation of Physician Estimates: It offers an empirical assessment of physician accuracy regarding survival predictions by comparing their estimates against the GWTG-R registry data.
• Clinical Framework: The findings establish a data-driven framework for understanding the "expectation gap" in pediatric critical care, moving beyond anecdotal evidence to statistical thresholds.

4. Key Results

The study analyzed decisions from 501 vignettes provided by parents and 257 vignettes provided by physicians. The findings revealed significant disparities:

• Decision Thresholds for Survival to Discharge:
  • Physicians: The threshold was 5.3% (95% CI: 3.7% to 7.0%). Physicians were likely to recommend withholding resuscitation if the chance of survival to discharge fell below this level.
  • Parents: The threshold was significantly lower at 1.2% (95% CI: -0.8% to 3.0%). Parents were willing to attempt resuscitation even when the probability of survival was extremely low.
• Decision Thresholds for Survival with Good Neurological Outcome (PCPC 1 or 2):
  • Physicians: Threshold was 3.5% (95% CI: 1.1% to 7.1%).
  • Parents: Threshold was 0.6% (95% CI: -1.2% to 1.8%).
• Physician Estimation Bias: Approximately 58% of physicians overestimated the likelihood of survival compared to the actual data from the GWTG-R registry.

5. Significance and Implications

The study highlights a profound misalignment in risk tolerance between medical providers and families:

• The Expectation Gap: Parents are willing to pursue resuscitation at survival probabilities (e.g., 1.2%) where physicians would typically recommend a DNAR order (threshold >5.3%). This suggests that physicians may prematurely recommend withholding care based on their own risk aversion or different value systems regarding quality of life.
• Clinical Practice: These findings underscore the necessity for enhanced communication strategies. Physicians must recognize that their "rational" thresholds may not align with parental hope and values. Counseling should explicitly address these probability differences to ensure shared decision-making is truly informed.
• Accuracy in Prognostication: The finding that over half of the physicians overestimated survival rates suggests a need for better integration of registry data into clinical prognostication to avoid false hope or, conversely, to ensure that the decision to withhold is based on accurate, rather than inflated, survival estimates.

In conclusion, this study provides critical evidence that the decision to attempt resuscitation is not solely a function of medical probability but is heavily influenced by the stakeholder's perspective, necessitating a more nuanced approach to end-of-life discussions in pediatric intensive care.