In recent years a number of events have shaken the measure of trust that the public holds for the application of science and the advances it offers society. Although many of these crises in confidence have arisen due to flawed implementation of policy rather than bad science, the scientific community must remain vigilant in order to preserve the trust and value that society affords scientific progress.

One of the key factors underpinning any mistrust is the public's understanding of the scientific method. Most people's perception of science is that it deals with certainties, an understandable misconception given that elementary scientific education deals with well-established theory derived from substantive research. Much less appreciated is the uncertainty inherent in much cutting-edge research—that by definition there is a limit to understanding when dealing with new phenomena at or beyond the boundaries of current knowledge. The perception of science as established fact means that when scientists do confess uncertainties, confidence in what is known is undermined. Scientific literacy is a pressing matter for education policy that will be increasingly difficult to address given the speed of accumulation of knowledge, but a better public appreciation of the scientific method would considerably alleviate some of the problems.

A better understanding of the scientific method would also allow the public to appreciate that often a scientist holds a particular opinion after weighing the available evidence. Such an appreciation would help clarify situations where different groups of researchers engage in support of opposing viewpoints. Moreover, zealots that advocate a particular scientific viewpoint in the face of overwhelming contrary evidence might also be afforded less credibility. Proponents of the view that HIV does not cause AIDS continue to garner support from some politicians for unknowable reasons. In South Africa, such beliefs are undermining efforts to tackle the grave problems posed by the AIDS epidemic. Recent efforts to provide cheaper antiretrovirals will be ineffective if the general population is led to believe that such approaches are irrelevant to the disease state.

The publication and dissemination of imprecise research studies, particularly correlative studies, can also contribute adversely to the public trust of science. Such studies occasionally have value in that they can frame novel scientific questions and propose new avenues of research, but it should always be clearly stated that only a correlation has been demonstrated. Provocative correlative studies are often picked up by the mass media and presented as accepted wisdom, and correlation soon becomes causation in the public consciousness.

A good example is the 1998 report that linked the measles-mumps-rubella vaccine to autism in children, which has led directly to a decrease in the uptake of this vaccine. Though the hypothesis is now widely disparaged after a number of ensuing studies have failed to find convincing evidence of such a link, negative data do not make good news stories and the information does not resonate with the public in the same manner that the original scare story did. It is a reminder that newsworthiness should never determine priority in scientific publishing and that scientific journalists should always exercise due prudence.

There is also a need to clearly distinguish in the public mind that science and the policies derived from it are separate. Science is concerned with the best way to answer a given question, but in essence it is ethically neutral and not a method for either deciding which questions should be asked or determining the acceptable risks over perceived benefits of new technologies; these considerations require public, scientific and political opinion allied to scientific fact. Policy-makers, keen to cultivate an authoritative stance, often co-opt scientific findings to validate policy. More sinisterly, dissenting scientific opinion is sometimes suppressed to divert attention from unwelcome truths. Subsequent to the bovine spongiform encephalopathy crisis in the United Kingdom, it emerged that in a misguided attempt to allay public concerns, vital information regarding the scientific uncertainties of the situation was not fully disclosed. The public recriminations were particularly damaging to the UK government and its scientific institutions, but also harmed public trust in science in general.

Also fundamental to cultivating public confidence in the veracity of scientific findings are conflict of interest issues that need to be addressed explicitly by researchers, journals and public bodies alike. As boundaries between academic and corporate research become increasingly blurred, especially in the field of biomedicine, there needs to be more transparency and accountability with regard to financial interests, corporate affiliations and patents. Conflict of interest policies are intended to prevent or control situations that might lead to inadvertent and unacceptable bias or to suspicions or actual instances of misconduct. Perceptions of a conflict of interest can erode public trust in science and scientists.

At the present time, it is incumbent upon the scientific establishment to allay public fears and win back crucial public trust before the situation becomes irrevocable. The nascent century has been heralded as the new age of biology, but much of the hoped-for progress will depend on the public's explicit assent for the endeavors science will undertake. Public opinion will be the final arbiter.