The art and science of understanding patent documents – 3 tips for inventors

It is very common for patent examiners to cite patent documents as prior art to support novelty and inventive step objections.  However, the information in patent literature is different to the kind of information published in peer-reviewed journal articles. In the latter case, the technical information is written for other scientists/engineers and is presented in a way that is intended to convey the results of that scientific research and to present conclusions that future researchers can build upon in their own research. In contrast, whilst a patent specification also covers technical subject matter, it is primarily a legal document that aims to set out the terms by which the patent owner and others will be bound. Accordingly, even though a patent specification should always be read through the eyes of a relevantly skilled person, it is written with public officials such as patent examiners, judges and business partners in mind. It is also often drafted around a particular commercial objective.


In light of these differences, the contribution and relevance of patent literature must be assessed in a different manner to non-patent literature. The purpose of this article is to cover some useful tips on how to read and understand prior art patent literature more efficiently.

Don’t just read the claims

The general structure of a patent specification is: title, abstract, background, summary of the invention, detailed description of the invention, examples, claims, and figures. It is useful to read the title and abstract to get an indication of the field of the invention. However, inventors often then immediately jump to the claims to ascertain what technical information is disclosed. In practice, we often encounter inventors who solely focus their review of a prior art patent on the claims. Some inventors deduce that because the claims seem to be quite different from the field of their own invention, the patent is not relevant.  Alternatively, they conclude that because a particular feature of their own invention is not disclosed in the claims, their invention must be new in view of that patent. At first blush, these are reasonable assumptions – after all, the claims should define the invention.  However, this kind of analysis is erroneous – the new invention must be assessed against the entire disclosure of the prior art patent specification.

It is useful at this point to understand what can influence the subject matter of the claims of a patent. In some cases, applicants disclose multiple inventions in a single patent application, but only pursue the main (or most commercially relevant) invention in a first application. Accordingly, just reading the claims will not necessarily give a clear picture of other inventions that may be disclosed in the patent specification. Further, for commercial reasons, claims can sometimes be directed to a specific embodiment of the invention, but a broader form of the invention contained in the detailed description may be highly relevant. An additional complication is that different claims can be pursued in different jurisdictions.

It should also be remembered that, until the claims are granted, they are a work-in-progress. In the usual course of events, patent applications are filed and initially published with broad claims. During the patent prosecution process, the claims are examined and typically narrowed in view of prior art raised by the examiner, and the patent is again published at acceptance/grant with those narrower claims. Under some circumstances, claims can also be amended post acceptance/grant, in which case the patent will be published again.

Therefore, for the above reasons, it is certainly useful to read the claims – but it is more important when evaluating the disclosure of a patent specification to read the detailed description in full, including, in particular, the examples.

It is also worth noting that the status of the patent – i.e., whether it was/is granted, in force, lapsed or expired, is not a relevant consideration when assessing the disclosure of the specification.  Patent documents cannot be dismissed on the basis that they did not proceed to grant or are not currently in force.

Scrutinise the examples

In many cases, applicants have limited resources and only generate a limited number of embodiments of the invention for inclusion in a patent application.  In order to maximise the commercial value of that patent application, the attorney handling the case will usually claim the invention broadly around that limited number of embodiments. Of course, the reason for claiming broadly is to cover enabled, routine alternatives and to maximise claim scope up to the limit of what has been previously disclosed.  In practice, the intent is to claim broadly enough to overlap with the prior art and then amend during prosecution up to the limit of the prior art, or to what is not obvious in view of that prior art. Claiming broadly can also reduce the likelihood that an infringer can avoid infringing the claims.

The net result can be, however, that inventors narrow the claims to their own inventions in view of patent prior art that makes broad claims, or worse – they may set aside their invention altogether.

A better starting point for inventors is therefore to review the examples of the prior art patent and compare their new invention against those actual examples to ascertain whether their invention is new. If there are similarities between the actual examples and a potential new invention, it is then prudent to consider whether the similarities are likely to constitute routine alterations or substitutions (an ethyl group for a methyl group in a chemical compound, for instance) or would otherwise be unlikely to have any measurable effect (a slightly different percentage of a non-active excipient, for instance).  Conversely, if there are one or more differences between actual examples and a new invention that appear to give rise to a surprising or unexpected technical effect (superior efficacy or stability, for instance) – then an inventor may be in a position to argue novelty and inventiveness over a patent disclosure with broad claims.

As mentioned above, patent applications are not peer reviewed, meaning the validity of the technical information in a patent (i.e., experiments in the examples) is not subject to the kind of scrutiny that a peer reviewed journal article will receive.  Because the technical information in a patent will only be assessed in patent validity proceedings, the information is usually assumed to be accurate by a patent examiner. Accordingly, it may be worthwhile repeating experiments in a highly relevant patent specification where the accuracy of the results is in question. This may assist in overcoming inventive step rejections over that patent prior art.

Pay attention to detail – but don’t be deterred by a broad specification

In addition to having broad claims, it is very common to encounter patent documents that include long lists of options and alternatives – be these applications (e.g., medical conditions), active ingredients (e.g., substituents and/or chemical compounds), non-active components (e.g., excipients, stabilisers, colours), physical parts (e.g., joints, adhesives, materials), etc.  Such lists are often included during drafting to demonstrate how components in specific examples can be replaced or altered in a routine manner without departing from the spirit of the invention (in patent terminology, to meet support and enablement requirements).  However, in practice, a patent applicant has rarely tested each and every combination.  Therefore, a new invention that combines selected features from across multiple lists and thus appears to be disclosed in a broad patent document is not necessarily anticipated.  If there were certain technical barriers that had to be overcome to make that combination, for example, or the combination gives rise to a surprising or unexpected technical effect – there is scope to argue rejections over broad patent prior art disclosures.

Notwithstanding the above, it is worth remembering that features which are not explicitly disclosed in a patent prior art document can be implicitly/inherently disclosed. Accordingly, it is important to read patent prior art through the eyes of the skilled person.

Finally, although terms in patent documents are usually given their ordinary meaning in the relevant art, it is permissible for the description to give terms special meanings.  That is, the patent might be its own dictionary.  Furthermore, different attorneys may use different terminology to refer to a feature. Because the language of a patent is very important, and misunderstanding or misinterpreting a technical term or even a commonly used word that has a very specific meaning in patent specifications (e.g., “consisting” in a patent specification is generally interpreted in the exhaustive sense) can lead researchers astray, it is important to read the patent literature carefully to ensure that a relevant disclosure is not dismissed due to terminology differences.

Conclusions and summary

Patent research is a tedious task and sometimes a researcher may dismiss a prior art patent on the basis of a title or abstract, or even the claims, when the relevant disclosure is contained in the examples or detailed description. For this reason, we recommend reading patent documents in full, paying particular attention to the examples.  In this way, you can provide your attorney with valuable information on the content of a prior art patent as read through expert eyes, which can streamline the patent prosecution process. More importantly, this may reduce the possibility of dismissing valuable IP in view of, for example, broad patent prior art disclosures.

The content of this article is general in nature and must not be relied on in lieu of advice from a qualified professional in respect of your particular circumstances.