Scientific evidence alone is not sufficient basis for health policy.
- SourceAvailable from: Mark LevertonAddiction 07/2014; 109(7):1209. · 4.60 Impact Factor
- Addiction 07/2014; 109(7):1209-11. · 4.60 Impact Factor
- Drugs: Education Prevention and Policy 05/2013; 20(3). · 0.53 Impact Factor
Scientific evidence alone is not sufficient basis for
Keith Humphreys and Peter Piot argue that basing health policy solely on evidence is inherently
contrary to the essence of policy development and even potentially dangerous
Keith Humphreys professor
1, Peter Piot director
1Veterans Affairs and Stanford University Medical Centers, VAPAHCS (152-MPD), 795 Willow Road, Menlo Park, CA 94025, USA;2London School
of Hygiene and Tropical Medicine, London WC1E 7HT, UK
about David Nutt’s 2009 conflict with the then UK home
secretary, Alan Johnson. After Professor Nutt publicly accused
the government of ignoring science when formulating drug
policy (for example, by overestimating the dangers of ecstasy),
he was sacked as UK drug policy adviser. As with other
and sex education, many scientists think that the lesson of the
Nutt controversy is that we must take the politics out of health
policy decisions and simply “do what the science says.” Based
on experience as researchers and as policy makers at the White
House and United Nations, we argue that although science
How science should inform health policy
Science can and must inform health policy decisions in several
ways, including by identifying emerging problems. The
discovery of new types of influenza virus, for example, or of
smoking as a cause of cancer made policy makers aware of
grave threats to health. The documentation of the first cases of
AIDS and of the emerging epidemic of methamfetamine
addiction in the western United States provided similar
are likely to produce a desired effect. For example, when the
Obama administration was looking for strategies to reduce
the effectiveness of screening and brief intervention services
within general healthcare. These services were ultimately
reform and the national drug control strategy.7 8Similarly, the
evidence that needle exchange can prevent the transmission of
HIV was the basis for UNAIDS’ support of “harm reduction”
programmes9and the Obama administration’s rationale for
lifting the ban on federal funding for this service.10Other
examples of evidence based interventions with potential policy
relevance include vaccines for human papillomavirus,
provides further help for policy makers by allowing them to
determine the cost and cost effectiveness of interventions.
Role of factors other than scientific
Science also has limits in health policy decision making. For
example, effective policy on smoking required more than the
discovery that it is a cause of lung cancer and other diseases.
the research evidence in the public domain and suppressed its
own findings on the dangers of cigarettes.11Substantial policy
action did not occur for several decades and was instigated not
by a new scientific breakthrough but by sustained health
activism and public interest litigation.
The decision to form policy in response to evidence of
particular individuals to lead their countries. For example, the
Obama administration could have chosen not to fund screening
and brief interventions to prevent substance misuse and instead
directed the money at other interventions with evidence of
effectiveness. Although it may frustrate scientists when
politicians are swayed by the possible electoral consequences
of various policy options, few scientists (including us) would
the views of those who have elected them as their
representatives. Voting, free speech, debate, and the push and
choose to do if the concept of democracy is to be meaningful.
We should remember that progress in public health from its
origins in the late 1800s and early 1900s has been as much
by social justice concerns, with the scientific evidence for their
effectiveness coming later. Without constructive politics,
Correspondence to: K Humphreys Keith.Humphreys@stanford.edu
For personal use only: See rights and reprints http://www.bmj.com/permissionsSubscribe: http://www.bmj.com/subscribe
BMJ 2012;344:e1316 doi: 10.1136/bmj.e1316 (Published 27 February 2012) Page 1 of 3
historical progress in public health would have been far more
limited, as the example of AIDS illustrates more recently.12
More generally, societies have values that don’t need to be
proved in randomised controlled trials and are appropriate
over-riding considerations in policy. A rigorous study showing
that people with an infectious disease could be cheaply and
rule out such a policy. Similarly, current research assessing
whether the death penalty is cost effective or has a deterrent
but it can never tell us whether the taking of a helpless
individual’s life by the state is morally acceptable. That
judgment falls on all our heads and cannot be evaded by saying
that the science made us do it or stopped us from doing it.
None of this is to deny that policy decisions made without
scientific advice can be ineffective (such as, restrictions on
people with HIV entering a country). They can also lead to a
spent worldwide on ineffective youth oriented programmes to
as the over 300 000 deaths resulting from President Mbeki’s
AIDS policies15). This is why the unique system of chief
in nearly all UK government departments is so important and
an example for other countries.
Scientists as human beings
Science is conducted by scientists, who are—thank
goodness—human. Their expertise in science does not
necessarily convey any expertise in governance (if you do not
enshrined in democratic societies, where people with technical
expertise get one vote at the ballot box just like other citizens.
There are at least two reasons why this is wise.
Firstly, scientists, like all people, are wrong at least some of the
time. The scientific “fact” that seems a certain basis for policy
by good scientific evidence have later been found to be
ineffective or even dangerous.16The use of thalidomide to treat
nausea in pregnant women, is one good example.
Secondly, again like all people, scientists can confuse their
opinions on policy matters with objective facts.17The US
addiction, and cognitive impairments and that scientists and
genes from the population through selective sterilisation and
immigrant exclusion.18Supporters of the movement failed to
appreciate that the first of these premises is an objective fact
and the second is a subjective viewpoint.
Conflating facts and values allows scientists to use their
authority inappropriately—that is, to cloak their effort to make
society live by their values as a disinterested, objective, and
unassailable stance. This may lead the public to defer to
scientists on the assumption that they know better, but in a
to say that their policy proposals are determined solely by
objective science is to promote a lack of accountability.
Although scientists have a duty to be a critical voice in society
because of the privileged information and position they may
received a Nobel prize for groundbreaking medical research
entitles him or her to more credibility in societal or moral
In drugs and AIDS policy, we both often hear advocates say
that policy should be based on “evidence rather than morals or
politics.” We have a higher opinion of such advocates than this
we respect them as people, which implies that they have values
them on policy substance or not, we see their commitment to
To say that an advocate or policy maker is guided by more than
technocratic considerations is a compliment and not an insult.
Scientific research is an extremely valuable tool for informing
health policy decisions because it can identify emerging
problems, offer tools to tackle those problems, and forecast the
likely effect of various policy choices. This potential to inform
can or should be cast aside, hence there is no such thing as
to value the influence of forces other than science in forming
health policy can have dangerous consequences for the
accountability of politicians and scientists, and for the
justification of policies that violate fundamental principles in a
democratic society. We should refrain from casting so many
our own political commitments and values, and in that sense
we stand on the same level as our fellow citizens rather than
Contributors and sources: This article emerged from conversations
between two scientists who recently returned to academia after working
in health policy roles. KH served as senior policy adviser at the White
House Drug Policy Office from 2009-10 and was a member of President
secretary-general of the United Nations and founding executive director
of the Joint United Nations Programme on HIV/AIDS from 1995 to 2008.
This article does not necessarily reflect the official position of the White
House, the United Nations, the US Veterans Health Administration, or
any other governmental organisation. KH is guarantor.
Competing interests: Both authors have completed the ICJME unified
declaration form at www.icmje.org/coi_disclosure.pdf (available on
request from the corresponding author) and declare: KH had salary
support from a US Departments of Veterans Affairs career research
scientist award while he completed the submitted work; neither author
has had financial relationships with any organisations that might have
an interest in the submitted work in the past three years; they have no
other relations or activities that could appear to have influenced the
Provenance and peer review: Not commissioned; externally peer
1 Caulkins JP, Reuter P, Coulson C. Basing drug scheduling decisions on scientific ranking
of harmfulness: false promise from false premises. Addiction 2011;106:1886-90.
Fischer B, Kendall P. Nutt et al’s harm scales for drugs—room for improvement but better
policy based on science with limitations than no science at all. Addiction 2011;106:1891-2.
Nutt D. Let not the best be the enemy of the good. Addiction 2011;106:1892-3.
Obot IS. Improved ranking of drugs on harmfulness can bring sense and order to a failed
system. Addiction 2011;106:1894-5.
Room R. Scales and schedules, notes and beams—whose view is obstructed on drug
scheduling? Addiction 2011;106:1895-6.
Rossow I. Can harm ratings be useful? Addiction 2011;106:1893-4.
Humphreys K, McLellan AT. Brief intervention, treatment and recovery support services
for Americans who have substance use disorders: an overview of policy in the Obama
administration. Psychol Services 2010;7:275-84.
White House Office of National Drug Control Policy. National drug control strategy.
Executive Office of the President, 2010.
Knight L. UNAIDS. The first 10 years, 1996-2006. UNAIDS, 2008.
For personal use only: See rights and reprints http://www.bmj.com/permissions Subscribe: http://www.bmj.com/subscribe
BMJ 2012;344:e1316 doi: 10.1136/bmj.e1316 (Published 27 February 2012)Page 2 of 3
10 White House Office of National AIDS Policy. National HIV/AIDS strategy for the United
States. White House, 2010.
Proctor RN. Golden holocaust: origins of the cigarette catastrophe and the case for
abolition. University of California Press, 2012.
Piot P, Russell S, Larson H . Good politics, bad politics: the experience of AIDS. Am J
Pub Health 2007;97:1934-6.
Donohue JJ, Wolfers J. Uses and abuses of empirical evidence in the death penalty
debate. Stanford Law Rev 2005;58:791-846.
Strang J, Babor T, Caulkins J, Fischer B, Foxcroft D, Humphreys K. Drug policy and the
public good: evidence for effective interventions. Lancet 2012;378:71-83.
Chigwedere P, Seage GR, Gruskin S, Lee TH, Essex M. Estimating the lost benefits of
antiretroviral drug use in South Africa. J AIDS 2008;49:410-5.
16 Roth-Cline MD. Clinical trials in the wake of Vioxx. Requiring statistically extreme evidence
of benefit to ensure the safety of new drugs. Circulation 2006;113:2253-9.
Kendler HH. The role of value in the world of psychology. Am Psychol 1999;54:828-35.
Allen GE. Eugenics and American social history, 1880-1950. Genome 1989;31:885-9.
Accepted: 17 January 2012
Cite this as: BMJ 2012;344:e1316
© BMJ Publishing Group Ltd 2012
For personal use only: See rights and reprints http://www.bmj.com/permissionsSubscribe: http://www.bmj.com/subscribe
BMJ 2012;344:e1316 doi: 10.1136/bmj.e1316 (Published 27 February 2012)Page 3 of 3