Sir

It is disturbing to read about the prevalence of misconduct described by Sandra Titus and colleagues (Nature 453, 980—982; 2008). But, as scientists, we are trained to deal with reality, not to avoid or, worse, misrepresent it. The description of a scientist as honest should be a tautology.

An institution can keep its good name only by reacting vigorously against allegations of misconduct. It should publicly denounce practices such as data falsification, plagiarism and duplication of research results in different publications. Official statements should be issued, warning that misconduct allegations will be subject to formal investigation and the results made public.

Unfortunately, here in Brazil this is not common practice. If misconduct allegations are ever filed, official statements are usually vague and investigations can take several years. Whistle-blowers are typically frowned upon by their colleagues and officials at their institutions.

Even though we agree that regulatory offices cannot catch all misconduct events, we suggest that consideration be given to the creation of international 'offices of research integrity' to pursue universal standards of ethical behaviour. After all, the unethical behaviour of a few scientists can damage the credibility of many.

See also: Integrity: Croatia's standards unusual in much of Europe Integrity: juniors see leaders gain from calculated dishonesty Integrity: how to measure breaches effectively Titus et al. reply