A recent newspaper article described the potential heath advantages of reducing carbon dioxide emissions from power plants. The piece concludes that the cited scientific report was "peer-reviewed" and published in a scientific journal called Climate Change.
Peer review is a useful system that can help maintain the quality of publications, but it is not without its problems. Peer review typically takes place by a small group of individuals who work in the specific scientific field served by the journal. In other words, climatologists review the work of other climatologists. They rank the submitted articles on their quality and perceived importance.
The problem is that the review group typically consists of selected individuals with similar views, which leads to a built-in bias. The author of the submitted publication is aware of the review team's views, and may even reference publications by committee members, inserting additional bias.
That is why many have criticized peer review as being a "good old boy" system.
But it is worth pointing out that the study of climate change is big business. Scientists get and keep their jobs in part by having long publication lists to their credit. Successful publication depends on having done good work, but unfortunately it also depends on fitting in with the rest of the field.
Saying an article has been peer reviewed doesn't automatically mean it is unbiased. Indeed, the opposite may be true.