www.thelancet.com Vol 372 October 4, 2008 1205
independence” to states gaining control of other’s
political machinations. Intervention solely to stop a
crime against humanity infringes on neither of these,
so it does not fall within the Charter’s force prohibition.
Furthermore, humanitarian intervention is within the
principles of the UN, because the Charter’s dual purposes
are preserving peace and promoting human rights.12
Additionally, international law requires the existence
of grave violations of human rights, an exhaustion
of non-forcible responses, and the unavailability of
UN-sanctioned action. The response must be pro-
portionate—no more than necessary to achieve human-
itarian ends—and it must not interfere unnecessarily
with a country’s self-determination. Finally, the
interveners must disengage upon securing fundamental
rights and report their actions to the Security Council.13
Nations should be justifi ably cautious about using
or threatening intervention to stop crimes against
humanity. Policy makers must carefully consider risks to
relief workers, civilians, and troops, as well as the danger
of complicating future health-promotion activities.
Forced intervention is a complex policy question, but
blanket rejection may condemn innocent civilians and
prevent deterrence of crimes against humanity. Where
leaders engage in intentional acts of cruelty toward
their populations, wealthy nations should be prepared
to intervene beyond their borders to safeguard health
and human rights.
John D Kraemer, Dhrubajyoti Bhattacharya,
*Lawrence O Gostin
O’Neill Institute for National and Global Health Law,
Georgetown University Law Center, Washington, DC 20001, USA
We declare that we have no confl ict of interest.
1 Ohmar K. Burma aid eff orts press on despite blocks. UPI Asia Online June 26,
eff orts_press_on_despite_blocks/5365 (accessed July 3, 2008).
Rotberg RI. Who will have the courage to save Zimbabwe? Boston Globe
June 25, 2008: A15.
UN. Charter of the United Nations. June 26, 1945. http://www.un.org/
aboutun/charter (accessed July 3, 2008).
WHO. Constitution of the World Health Organization. October, 2006.
July 3, 2008).
UN Economic and Social Council. The right to the highest attainable
standard of health. Aug 11, 2000. http://www.unhchr.ch/tbs/doc.nsf/
(symbol)/E.C.12.2000.4.En (accessed July 3, 2008).
Offi ce of the United Nations High Commisioner for Human Rights.
Convention on the rights of the child, article 24(4). Sept 2, 1990.
http://www2.ohchr.org/english/law/crc.htm (accessed July 8, 2008).
International Criminal Court. Rome Statute of the International Criminal
Court, article 7(1). July 1, 2002. http://untreaty.un.org/cod/icc/statute/
romefra.htm (accessed July 3, 2008).
Prosecutor v. Milomir Stakić, Case No. IT-97-24-T, International Criminal
Tribunal for the Former Yugoslavia. July 31, 2003. http://www.un.org/icty/
stakic/trialc/judgement/index.htm (accessed July 8, 2008).
Saechao TR. Natural disasters and the responsibility to protect: from chaos
to clarity. Brookland J Int Law 2007; 32: 663–707.
10 Hutchinson MR. Restoring hope: UN Security Council resolutions for
Somalia and an expanded doctrine of humanitarian intervention.
Harvard Int Law J 1993; 34: 624–40.
11 UN General Assembly. 2005 World Summit outcome. Oct 24, 2005.
UN/UNPAN021752.pdf (accessed July 8, 2008).
12 Mertus J. The imprint of Kosovo on the law of humanitarian intervention.
ILSA J Int Comparative Law 2000; 6: 527–40.
13 Terry JP. Rethinking humanitarian intervention after Kosovo: legal reality
and political pragmatism. Army Lawyer 2004; 8: 36–46.
How doctors feel: aff ective issues in patients’ safety
Two books have directed attention to the under-
pinnings of doctors’ thinking.1,2 Thinking (cognitive)
failures abound in clinical decision making, especially
in diagnostic formulation, and taxonomies of com-
mon cognitive errors have been developed.3 Diagnostic
failure has been identifi ed as a major threat to patients’
safety4 and, this year, the American Journal of Medicine
published a supplement on the problem5 to coincide
with the fi rst symposium on diagnostic error.6 Despite
the tardiness of this focus on how doctors think, we
welcome the advance in evolution of patients’ care and
safety. The more diffi cult next step is to recognise that
how doctors feel would also be a complementary and
worthy topic for investigation, especially for any eff ects
on clinical decision making and patients’ safety.
Historically, the prevailing view in medicine is that
clinical decisions should be objective and free from
contextual aff ective issues: one could not be objective and
rational if emotion entered the reasoning process. Indeed,
many of us would consider it a professional virtue to be
able to rise above the emotional pull of clinical situations,
to deliver cool, clear, analytical judgments. However,
despite what we might believe, our feelings (aff ect)
intrude into almost every decision that we make. Our
daily interactions with others are infl uenced by conscious
or unconscious social transference pheno mena7 which
are aff ectively polarised in ways that range from subtle
to substantial. Similarly, specifi c clinical situations pro-
voke lesser or greater degrees of aff ective valence. In
fact, our fi rst response to anything is an aff ective one
www.thelancet.com Vol 372 October 4, 2008
that governs the future direction of our relations8—we
tend to trust our fi rst impressions and stick with them.
To then understand the role that aff ective state has in
clinical decision making seems important.
A consensus is emerging that decision making occurs
through one or a combination of two modes: the fi rst
(system 1) is intuitive, fast, automatic, often involves
an aff ective component, and uses few resources, and
the second (system 2) is analytical, slow, deliberate,
aff ect free, and resource intensive.9 Importantly, most
errors of judgment occur in system 1 in which aff ect
predominates. In system 1 heuristics (mental short-
cuts, maxims, rules of thumb) and biases also occur. The
powerful aff ect heuristic10 might substantially infl uence
judgment. For example, some patients will elicit
aff ective responses from their health-care providers.
Sometimes these responses are positive, but they could
also be negative and lead to labelling,11 patients being
referred to as complainers, diffi cult, high maintenance,
or worse. Labelling not only infl uences a clinician’s
thinking but also that of other health-care providers—
eg, in borderline personality disorder, visceral reactions
elicited by the patient in their provider might be the basis
for making the diagnosis.12 Aff ective valence towards
patients, positive or negative, can compromise decision
making. The best evidence might be degraded when it is
unconsciously passed through an aff ective fi lter.
The idea of aff ective infl uence on decision making
will be unfamiliar to many clinicians. Eff orts should be
made to raise awareness of how aff ect infl uences clinical
performance, and to describe its many forms. The panel
provides a preliminary taxonomy that groups aff ective
dispositions to respond towards patients into three main
categories. The fi rst are aff ective states in the caregiver
that might be induced by the immediate environment or
work conditions—eg, irritability induced by high levels of
ambient noise or negative aff ective states associated with
sleep deprivation. The second are aff ective biases that
are specifi c to the context or patient. One of the most
powerful is counter-transference, in which the caregiver
feels positively or negatively towards the patient because
of experience with previous exemplars. Another example
here is fundamental attribution error, in which patients
can be judged on the basis of dispositional qualities
rather than circumstantial ones—eg, blaming patients
for their obesity rather than underlying socioeconomic
factors that might have led to their condition. The third
are endogenous aff ective states within the clinician:
some depend on various temporal factors, others on
mood disorders, or emotional avoidance leading to
mistreatment or neglect of patients.
In summary, increasing evidence exists, mostly from
the fi eld of psychology, that aff ective factors could
infl uence physicians in the diagnostic process, medical
decision making, and interactions with patients. There is a
growing imperative for medical educators to understand
and incorporate this knowledge into clinical training.
*Pat Croskerry, Allan A Abbass, Albert W Wu
Department of Emergency Medicine and Department of Medical
Education, Dalhousie University, Halifax, NS, Canada B3H 2Y9
(PC); Centre for Emotions and Health, Dalhousie University,
Halifax, NS, Canada (AAA); and Health Policy and Management,
Bloomberg School of Public Health, Johns Hopkins University,
Baltimore, MD, USA (AWW)
We declare that we have no confl ict of interest. AWW is Senior Adviser, World
Alliance for Patient Safety, WHO.
1 Montgomery K. How doctors think: clinical judgment and the practice of
medicine. Oxford: Oxford University Press, 2006.
Groopman J. How doctors think. New York: Houghton Miffl in Company,
Croskerry P. The importance of cognitive errors in diagnosis and strategies
to prevent them. Acad Med 2003; 78: 1–6.
Graber M. Diagnostic errors in medicine: a case of neglect.
Jt Comm J Qual Patient Saf 2005; 31: 106–13.
Graber ML, Berner ES, eds. Diagnostic error: is overconfi dence the problem?
Am J Med 2008; 121: vii.
American Medical Informatics Association. Diagnostic error in medicine:
a national conference—2008. May 29–31, 2008; Phoenix, Arizona. http://
www.amia.org/meetings/s08/dem_program.asp (accessed July 13, 2008).
Glassman NS, Andersen SM. Transference in social cognition: persistence
and exacerbation of signifi cant-other-based inferences over time.
Cognit Ther Res 1999; 23: 75–91.
Ittelson WH. Environmental perception and contemporary perceptual
theory. In: Ittelson WH, ed. Environment and cognition. New York: Seminar
Croskerry P. The theory and practice of clinical decision making.
Can J Anesth 2005; 52: R1–8.
10 Slovic P, Finucane M, Peters E, MacGregor DG. The aff ect heuristic. In: Gilovich
T, Griffi n D, Kahneman D, eds. Heuristics and biases: the psychology of
intuitive judgment. Cambridge: Cambridge University Press, 2002: 397–420.
11 Groves J. The hateful patient. N Engl J Med 1978; 298: 883–87.
12 Olson SC, Rund DA. Behavioral disorders: clinical features. In: Tintinalli JE,
Ruiz E, Krome RL, eds. Emergency medicine: a comprehensive study guide.
4th edn. New York: McGraw-Hill, 1996.
Panel: Taxonomy of aff ective dispositions to respond
• Transitory aff ective states: environmental factors, sleep
deprivation, sleep debt, irritability, stress, fatigue
• Clinical situation induced: specifi c aff ective biases
(eg, fundamental attribution error, counter-transference)
• Endogenous disorders: circadian, infradian, or seasonal
mood variation, mood disorders, anxiety disorders,