Review: Unpaywall

Unpaywall is a browser extension available for Firefox and Chrome that allows the user to find free, legal access to academic articles that would otherwise require purchase of the article in question.  From the extension description in the Chrome web store:

Find open-access versions of paywalled research papers, instantly. Get free, legal fulltext of research papers as you browse. When you view a paywalled research article, Unpaywall automatically looks for a copy in our index of over 10 million free, legal fulltext PDFs. If we find one, click the green tab to read the article.

Unpaywall takes advantage of institutional repositories, open access publications, and other sources such as PubMed Central and Google Scholar to offer access to articles (Unpaywall FAQ, How do you find all these fulltext articles?

In the settings, Unpaywall offers two customizations: Color-code tab for Green and Gold OA and Hide content from less trusted sources.  The color-coding option allows the user to see a green tab when a fulltext article is available from an institutional repository or preprint server, a gold tab for journal-hosted content with an Open license, and a blue tab for journal-hosted fulltext with no license.  Content that has no free fulltext displays a grey tab.  The hide-content setting hides results from less reputable sources, including ResearchGate and, as well as researcher homepages and some institutional repositories.  Unpaywall estimates that enabling this option decreases fulltext coverage by about 20%.

Upon installing the extension, the user will see a welcome page ( that contains a very simple overview of how to use the extension, including a screenshot of an online journal article with the green Unpaywall tab.  There is also a “try it now” button that links to an article about the TRAPPIST-1 planetary system published in Nature (Nature 542, 456–460 (23 February 2017); doi:10.1038/nature21360).  With the default settings, the unpaywall tab is green; if the color-coding setting is enabled, it is blue, indicating journal-hosted fulltext with no license.

To assess the performance of the Unpaywall extension, I used a convenience sample of the first ten search results for a PubMed search for the term ‘pharmacoepidemiology’ (search strategy: “pharmacoepidemiology”[MeSH Terms] OR “pharmacoepidemiology”[All Fields]).  Of the ten results, four did not have a link to the article’s fulltext.  Two linked out to Oxford Academic; one to JAMA Fulltext; and three to Wiley Online Library.

Both of the articles that linked out to Oxford Academic were available as journal-hosted fulltext with no license (Unpaywall color code blue).  The article from JAMA Fulltext was also available under the Unpaywall color code blue.  Of the three articles available from Wiley Online Library, two had an Unpaywall color code grey (no free fulltext available) and one had an Unpaywall color code gold (journal-hosted content with an Open license).  The details of each of these six articles are included in Appendix I.

With six out of ten (or 60%) of articles available in fulltext through Unpaywall, this admittedly small sample is nevertheless consistent with the extension’s estimated range of fulltext availability—65-85% of articles, depending on their topic and year of publication (Unpaywall FAQ, How often does Unpaywall find full text?

Appendix I. Article availability through Unpaywall

Article Fulltext from Unpaywall availability (tab color)
Conducting Privacy-Preserving Multivariable Propensity Score Analysis When Patient Covariate Information Is Stored in Separate Locations.

(Am J Epidemiol. 2017 Mar 15;185(6):501-510. doi: 10.1093/aje/kww155)

Oxford Academic Blue (journal-hosted fulltext with no license)
How to compare instrumental variable and conventional regression analyses using negative controls and bias plots.

Int J Epidemiol. 2017 Apr 7. doi: 10.1093/ije/dyx014. [Epub ahead of print]

Oxford Academic Blue (journal-hosted fulltext with no license)
FDA Approval of Eteplirsen for Muscular Dystrophy-Reply.

JAMA. 2017 Apr 11;317(14):1481-1482. doi: 10.1001/jama.2017.2605.

JAMA Fulltext Blue (journal-hosted fulltext with no license)
Prevalence, incidence, indication, and choice of antidepressants in patients with and without chronic kidney disease: a matched cohort study in UK Clinical Practice Research Datalink.

Pharmacoepidemiol Drug Saf. 2017 Apr 11. doi: 10.1002/pds.4212. [Epub ahead of print]

Wiley Online Library Gold (journal-hosted content with an Open license)
Effects of expanding the look-back period to all available data in the assessment of covariates.

Pharmacoepidemiol Drug Saf. 2017 Apr 11. doi: 10.1002/pds.4210. [Epub ahead of print]

Wiley Online Library Grey (fulltext unavailable)
Evaluating the effectiveness of risk minimisation measures: the application of a conceptual framework to Danish real-world dabigatran data.

Pharmacoepidemiol Drug Saf. 2017 Apr 11. doi: 10.1002/pds.4203. [Epub ahead of print]

Wiley Online Library Grey (fulltext unavailable)