Communicating Prognostic Uncertainty
When numerous uncertainties prevent reliable prognostication, the most appropriate approach is to explicitly acknowledge the "indeterminate outcome" and recommend prolonged observation with serial reassessment, while avoiding premature definitive statements about prognosis.
Framework for Addressing Prognostic Uncertainty
Acknowledge Uncertainty Directly
The evidence strongly supports being honest about prognostic limitations rather than forcing premature conclusions:
State explicitly that the outcome is "indeterminate" at the current time point, as recommended in the European Resuscitation Council guidelines for situations where robust predictors are absent 1.
Use clear language such as: "Given the numerous uncertainties in your loved one's condition, I cannot provide a reliable prognosis at this time" 2.
Avoid vague or ambiguous phrasing that may confuse families—be direct about what is known and unknown 3, 4.
Recommend "Observe and Re-evaluate" Strategy
Multiple guidelines emphasize that uncertain outcomes require a time-based approach:
Explicitly recommend prolonged observation as the appropriate next step when prognostic certainty cannot be achieved 1.
Explain that clinical improvement (or lack thereof) over time provides crucial prognostic information that is currently unavailable 1.
Specify a timeframe for reassessment when possible (e.g., 24-72 hours for post-cardiac arrest patients, or several days for intracerebral hemorrhage) 1.
Address Confounding Factors
Postpone prognostication when confounders prevent reliable assessment 1:
- Residual sedation or paralysis
- Inadequate time for physiological stabilization
- Ongoing acute interventions
- Drug interference with clinical examination
State clearly: "We need to wait until [specific confounders] resolve before we can make a more accurate assessment" 1.
Critical Pitfalls to Avoid
The Self-Fulfilling Prophecy
Great caution must be exercised to avoid premature pessimistic prognostication, particularly when considering withdrawal of support 1:
Current prognostic models are biased by failure to account for early care limitations and withdrawal of support 1.
Early DNR orders or withdrawal decisions can create self-fulfilling prophecies of poor outcome in patients who might otherwise have favorable outcomes 1.
For intracerebral hemorrhage specifically, aggressive full care and postponement of new DNR orders until at least the second full day is recommended (Class IIa evidence) 1.
Premature Certainty
No single predictor provides absolute certainty, even the most robust ones 1:
Multimodal prognostication is recommended whenever possible, even when one robust predictor is present 1.
Most survivors recover consciousness within one week, but awakening as late as 25 days has been documented with good neurological outcomes 1.
Prognostic schemes often have inadequate precision for individual patient decision-making, with confidence intervals that may suggest survival in 1 in 20 patients even when predicted mortality approaches 100% 1.
Communicating with Families
Balance Honesty with Hope
Research demonstrates that communication strategies must address both types of uncertainty 2, 3:
Aleatory uncertainty (unpredictability of future events): "Even with the best information, we cannot predict exactly what will happen for your loved one"
Epistemic uncertainty (limitations in our knowledge): "Our prognostic tools have limitations and may not be accurate for individual patients"
Surrogates are often frustrated by both too much ambiguity and failure to acknowledge uncertainty exists 2.
Provide a Clear Plan
Rather than leaving families in limbo:
- Specify what clinical signs or timeframes will guide reassessment 1
- Explain that absence of clinical improvement over time suggests worse outcome, but improvement remains possible 1
- Commit to scheduled follow-up discussions as more information becomes available 1
Special Considerations
When Prolonged Uncertainty Persists
If uncertainty extends beyond typical timeframes 1:
- Acknowledge that newer therapies and more aggressive management may mean current outcomes are better than historical data suggest
- Consider that maximal outcome may take longer to achieve than traditional 6-month assessments capture
- Recognize that severely disabled patients may still express satisfaction with quality of life
Documentation
When documenting prognostic uncertainty:
- Use terminology like "indeterminate outcome" rather than vague descriptors 1
- Document specific confounders preventing prognostication 1
- Record the plan for serial reassessment with timeframes 1
The fundamental principle is that prognostic uncertainty should be explicitly named, not obscured, and should trigger a strategy of continued aggressive care with planned reassessment rather than premature limitation of therapy 1, 5.