TL; DR Version
Footnotes:
Nature Precedings; JMIR preprints; PeerJ [no longer accepting preprints], etc
Usually for reputable journals; however quality of peer review is variable
Final document is subject to much more political wrangling than most article types
Types of Publications Relevant to COVID-19
Teams are writing papers about COVID-19 at breakneck pace. Below is a table that describes several routes those papers might reach their audiences.
Some COVID-19 papers are not peer reviewed
You may see papers marked with “medRxiv” or “bioRxiv” among the papers about COVID-19. The authors who wrote the paper consider it complete and accurate enough to share with others, and in the interest of rapid information sharing, they have uploaded it to a server.
medRxiv and bioRxiv are called preprint servers, and they exist specifically to allow authors to share their work when the authors deem it ready, without further endorsement by anyone outside the authors. The key thing to understand is that these papers become available without any requirement for vetting or checking of the claims or science. They have not been reviewed by scientists to evaluate their accuracy or their merit, increasing the likelihood that errors could be present. But the benefits of early sharing of information are made obvious by this fast-moving pandemic, and these papers are being heavily inspected and evaluated by the scientific and medical communities once they are available. Those people often share their impressions with one another and the scientific community via various platforms, including Twitter.
A subset of these papers end up favorably peer-reviewed and published in traditional journals after they become available on these preprint servers.
Some types of papers such as those simply reviewing known facts (“review articles”) or expressing opinions (“opinion pieces”, “perspectives”) are not allowed on medRxiv or bioRxiv. They are for new, original research papers.
The key difference between medRxiv and bioRxiv is that medRxiv is more focused on medical papers, and bioRxiv is more focused on more fundamental biology.
To make things a little more interesting, some journals use the term “preprint” in a different context, meaning the paper has been peer reviewed but not published on paper yet. You should assume that this is not the case with a paper marked as a preprint, until you understand which journal it is in. medRxiv and bioRxiv are not journals. Finally, some journals have their own preprint servers competing with medRxiv and bioRvix, allowing people to look at the paper during its period of peer review. Is it unfortunate that the terminology is muddy, but do your due diligence to understand whether what you are reading has been peer reviewed. If you found an important paper on medRxiv or bioRxiv, you can try to locate the same paper in published, peer-reviewed form in a journal somewhere, and sometimes it may be available.
Many COVID-19 papers are being published in traditional medical and scientific journals. In the interest of speedy exchange of useful information, this should be happening after those papers have been shared on a preprint server in the vast majority of cases, unless they are review articles or opinion pieces.
There is more than one type of publication published in medical and scientific journals, and many of each will be appearing on the topic of COVID-19.
Original Research or Original Investigation
This type of publication is how new discoveries are conveyed in the medical and scientific worlds. It describes new results of experiments or systematic observations about some aspect of science or a medical issue.
Review Article
Review articles are not focused on introducing new findings or discoveries, but rather on summarizing the current state of knowledge about a topic. There have already been review articles published on various aspects of COVID-19, and there will be many more in the future.
There are two general types of review articles. Most are so-called narrative reviews, in which the authors decide informally which facts are most important to assemble from the published literature. Fewer review articles--called systematic reviews--include strict, formal criteria for searching the literature and deciding which published papers to include in the review. These are called systematic reviews. It is commonly asserted that systematic reviews are less prone to leaving out important information.
Review articles undergo variable amounts of pre-publication peer review (vetting by other experts before acceptance for publication in a journal). Sometimes it’s very detailed; other times its cursory or may be omitted altogether.
Letter to the Editor
Letters to the editor of a journal can address any topic of interest to the authors and are shorter than the types of publications described above.
They may report new findings, and in that case they are likely to have been peer reviewed. Some letters to the editor are responses to a published article, often critiques of those articles. Authors of that genre of letters to the editor often complain that editors are not interested in publishing such critiques of what has been published in the journal. Some journals may allow publication of letters to the editor critiquing a paper that was published in a different journal.
Correspondence
Many journals use the term correspondence to describe letters to the editor containing critical responses to articles recently published in that journal.
Brief Report
This is a short article reporting the results of original research. This type of article is sometimes easier to publish due to some nuances of how journals are compared to one another.
Perspective or Opinion Pieces
Perspective or opinion pieces offer the authors’ opinion on a topic. Examples might include proposing that we change healthcare policy or medical practice in some way based on the authors’ experiences and perceptions of what would be useful.
Rapid Response
A few journals have a feature that works more like a blog post or tweet, almost. These may be called Rapid Response or something else. There could be a level of vetting by the editor, but no traditional peer review has been applied.
Guidelines
Guideline documents come from a group of authors, often representing a professional medical or scientific organization. Medical guidelines attempt to describe best practices when treating a medical problem or set of medical problems, in hopes of reducing unwarranted variation in how people approach things. The idea is essentially to disseminate wise thoughts of experienced clinicians, who usually are also accomplished researchers. Scientific guidelines also exist and can help standardize how people work in the laboratory for easier comparison across experiments done in different labs. Guidelines are typically peer-reviewed to an extent, but constraints on the degree of review could be imposed by the elite professional status of the people writing the guidelines plus backing by powerful professional societies.
Blog Posts
Blog posts are completely free-form and can be contributed by any person. Accordingly, they vary markedly in quality, from misleading or highly inaccurate to sublime and exceptionally useful. Blogs are not peer reviewed, but the smartest blog authors pay attention to feedback and iterate and improve their posts.