An experiment on the presumption of consent to emergency medical treatment

Download data and study materials from OSF

Principal investigator:

Winston Chiong

University of California, San Francisco

Email: Winston.Chiong@ucsf.edu

Homepage: http://memory.ucsf.edu/ourcenter/staff/wchiong


Sample size: 2154

Field period: 10/11/2012-03/07/2013

Abstract

In life-threatening emergencies involving incapacitated patients, clinicians may intervene without obtaining informed consent, on the presumption that reasonable people would consent to treatment in such circumstances. Whether this rationale applies to treatment of acute ischemic stroke with thrombolysis, which is not life-preserving, is controversial.
2,154 U.S. adults read vignettes in which they had either suffered an acute ischemic stroke and could be treated with thrombolysis, or had suffered a sudden cardiac arrest and could be treated with cardiopulmonary resuscitation. Participants were then asked either (1) whether they would want the intervention, or (2) whether they would want to be given the intervention even if informed consent could not be obtained.
Overall, 76.2% would want thrombolysis for stroke, while 75.9% would want cardiopulmonary resuscitation for cardiac arrest. If informed consent could not be obtained, the desirability of thrombolysis was unaffected (78.1%), while a greater proportion wanted treatment with cardiopulmonary resuscitation (83.6%).
The proportion of older adults who would want emergency thrombolysis for acute ischemic stroke is comparable to the proportion who would want emergency cardiopulmonary resuscitation for sudden cardiac arrest, and is unaffected by the inability to obtain informed consent. These findings provide empirical support for the treatment of acute ischemic stroke with thrombolysis in appropriate emergency circumstances under a presumption of consent.

Hypotheses

Our primary hypothesis was that there will be a significant interaction between scenario (stroke vs. cardiac arrest) and question type (hypothetical choice vs. treatment without explicit consent), in which the acceptability of thrombolysis but not CPR is affected by question type. This reflects greater public familiarity with CPR than thrombolysis, as well as prior findings that patients desire greater involvement for medical decisions involving morbidity than mortality.
Our secondary hypothesis was that thrombolysis (for which there is no established legal precedent for presumed consent) will be no less acceptable than CPR (for which there is legal precedent for presumed consent).

Experimental Manipulations

Clinical scenario: ischemic stroke vs. cardiac arrest.
Question type: hypothetical choice vs. inability to consent.

Outcomes

Preferences for the described emergency treatment, Likert-scaled as: definitely yes / probably yes / probably no / definitely no.

Summary of Results

Overall, 76.2% would want thrombolysis for stroke, while 75.9% would want cardiopulmonary resuscitation for cardiac arrest. If informed consent could not be obtained, the desirability of thrombolysis was unaffected (78.1%), while a greater proportion wanted treatment with cardiopulmonary resuscitation (83.6%).

Additional Information

Survey includes measures of patient-directed decision-making (from Levinson JGIM 2005), attitudes regarding disability and longevity, confidence in the medical system, and attitudes about the use of statistical information in decision-making.

References

Chiong, Winston. 2014. "Testing the Presumption of Consent to Emergency Treatment for Acute Ischemic Stroke." Journal of American Medical Association 311(16): 1689-1691.

Chiong, Winston et al. 2014. "Inability to Consent Does Not Diminish the Desirability of Stroke Thrombolysis." Annals of Neurology.

Huang, Ivy A., John M. Neuhaus, and Winston Chiong. 2016. "Racial and Ethnic Differences in Advance Directive Possession: Role of Demographic Factors, Religious Affiliation, and Personal Health Values in a National Survey of Older Adults." Journal of Palliative Medicine 19(2): 149-156.