CancerGuide: How to Research the Medical Literature

Introduction

This page explains how to use databases and online resources, but not how to access these resources. For that, see my pages on accessing medical databases, and online information sources. Another option is to pay a medical search service to search the databases for you. Rather than go through the details of the user interface to these databases, I present a conceptual view of how to use them that is independent of the particular database and user interface.

Before starting your search, you need to know some basic information information on your cancer particularly its medical name and stage. For more information on these subjects, see my page on getting basic information on your cancer.

You should also spend a little time to think about exactly what kind of questions you want to answer. Perhaps you want to see if there are any new treatment options that might increase your odds or perhaps you would like to find experts in treating a rare cancer. Maybe you want to look into an alternative therapy that you've heard about.

Your doctor will have the answer to many of your questions, and for lots of things, it makes a lot more sense to ask your doctor than it does to start searching the technical medical literature. If you have a question about a specific symptom or side effect you are having, or about the interpretation of test results, your doctor certainly should be your first source of information. This is also true if you want to know the success rate of a proposed treatment, or about the side effects, although in these cases, it might be also reasonable to look into it yourself. Especially if you aren't satisfied with the answers you're getting or just demand more detail than your doctors have time to give you.

The Research Cycle

Researching the literature is not a single path that proceeds straight from initial question to final answer. It is actually more like a cycle. Initial questions lead to references to the medical literature, which lead to papers, which lead to more references, and more questions, and so on. Eventually, you will zero in on the information that is most valuable to you. This process cannot usually be completely conducted on-line. Following the research cycle will probably take you to medical databases, doctors, and medical libraries.

Before continuing, you should understand the differences between references, abstracts, and papers. References are just that, references to published journal articles, and sometimes to books and conference proceedings. A reference contains the title and author of the work as well as the name of the journal, and the year, volume number, and page numbers as appropriate. Given a reference, you can look-up the paper in the medical library. An abstract is a short summary of a paper. An abstract is usually enough to determine whether a paper is interesting enough to actually be worth looking up at the medical library. Almost all papers start with an abstract, and high quality medical databases such as MedLine contain abstracts for many of the references, but I know of no database that has the full text of articles. The paper is the actual report of the work, and contains all of the details on the background, methods, results, and conclusions. Every paper also contains a list of references at the end.

Where to Get References

There are many sources for references, including:

Other papers
As I mentioned above, every paper in the medical literature has a list of references at the end. Often this list is both lengthy and comprehensive.
Your doctor
I have found many doctors quite willing to give me references and even Xeroxes of papers supporting their positions on treatment questions. All I had to do was ask!
Cancer.gov
Cancer.gov is the website of the US National Cancer Institute. The PDQ Health Professional State of the Art Treatment Statement for each type of cancer contains some references, and Cancer.gov also has some pre-done, but non-specific database searches for most kinds of cancer. See CancerGuide's Tour of Online Resources for a more detailed guide to Cancer.gov.
Protocol Documents
If you are considering a particular clinical trial and want to know more about it, you should know that each clinical trial has a protocol document that defines the complete plan for the trial and includes scholarly discussion of the history and prior results of the treatments being tested in the trial. This includes a list of references to relevant papers.

The doctors conducting the trial you are considering should be willing to make you a copy, but sometimes they do refuse. With some drug company sponsored trials, the protocol document is confidential, and the doctor is "not allowed" to show it to you. I think this is an outrage. If I were in this situation, I would contact the drug company, and demand to be given a copy, while offering to sign a non-disclosure agreement.
Medical Textbooks
Oncology texts such as Cancer Principles and Practices by DeVita, Hellman, and Rosenberg have an extensive list of references at the end of each chapter. This probably won't include references to the most esoteric experimental approaches, and cannot include references to discoveries made since the last edition of the book was published.
Medical Databases
Databases such as MedLine are by far the most powerful way to find references, although some effort is required to learn how to use these resources effectively. The huge size of these databases with millions of references, and abstracts from thousands of journals means that if the information you seek has been published, the references are there. The ability to focus your search on words or (better) concepts that interest you allows you to extract just the references that pertain to these words and concepts. The biggest problem with these databases tends to be getting too much information, not too little.

For detailed information on accessing and using Medline, see my article on MedLine Access along with the associated how-to tips.

Types of Papers in the Medical Literature

Typically a list of references, such as that generated by a database search, has references to many papers, but most of them will be irrelevant. You want to develop the ability to scan the list of references and quickly pick out the relevant ones. Usually you can tell whether a reference might be interesting just by the title. It helps to have some knowledge of what kinds of papers are out there.

Here is a list of some of the kinds of articles you will find in the medical literature, arranged roughly from most to least useful.

Review Articles
Review articles are a virtual gold mine of information, and are one of the very best places to begin your detailed search. Review articles are a summary of all of the current work in some area of science or medicine. Review articles are primarily intended to keep professionals up on the latest research. If they can do that for professionals, they can do it for you! Naturally review articles contain a very extensive list of useful references.

Reviews can actually be of two types, reviews of a particular disease or condition, or reviews of a particular type of treatment or technology. Depending on the situation you may want to look for one or the other.

It is extremely important to get recent review articles. Reviews more than a year or two old are reviewing old data, and may not include important recent developments. Even the most recent review has to be somewhat behind the most recent papers since reviews summarize what's published at the time the review is written, and there are significant publication delays. Still, most important developments go through many years of research before becoming "standard treatment," so the chances are good that a recent review will contain some information on most promising new developments.
Clinical Studies
These are reports of clinical trials of new treatments on patients. For most people, decisions on what treatment is best depends primarily on the prior evidence in humans. That evidence will be found in this key class of papers. The word "trial" or the phrase, "clinical trial" appears in the title of many of these papers.
Case Reports
These are reports of an individual case or a limited number of cases of a particular cancer or situation. Often these are about a cancer that shows up in an unusual way, an unusual side effect, or an unexpected recovery. Because these are reports of exceptions they are not applicable or useful to you unless you happen to be a rare exception as well. If you have a very rare cancer, you may have to rely on case reports to some degree, but for most patients, these references will not be useful. The phrase, "a case of," or "case report" appears in the title of many of these studies.
Animal Studies
Animal studies may be the only source of prior results for some very new treatments, but it is difficult to generalize from animal models of cancer to spontaneous cancers in humans. It is far better to rely on the results of human studies than on animal studies. Note that titles with the word "murine" refer to studies with mice which are, by far, the most common experimental animal.
Reports on Diagnostic Techniques
These reports concern new diagnostic techniques, such as new blood tests, or imaging techniques, or may probe the effectiveness of current diagnostic techniques. Once in a while these reports can be useful to a patient. In two cases where a patient was suspected to have hidden metastases that could not be located, I found new imaging techniques that could help find them so they could be removed surgically.
Reports on Prognostic Factors
These reports attempt to determine which factors about a patient portend a better or worse prognosis. Sometimes consideration of these factors can be useful in determining whether a patient could benefit from additional treatment, but anyone reading these reports will certainly be face to face with some blunt statistics about their prognosis.
Epidemiology Reports
These are reports of the association between various factors and the incidence of various cancers and diseases. These reports do not directly prove that factors that are associated with a disease are actually the cause of the disease, but more to the point, knowing possible causes for your cancer is of little help in deciding what treatment will work best.
Cell Culture Studies
These studies test the reaction of cancer cells grown in a test tube to potential new chemotherapy drugs. These tests are usually done even before animal studies, and it is unwise to place much faith in such results in deciding about treatment. The phrase "in vitro," which literally means "in glass," often appears in the title of these papers.
Basic Science
This is a vast catch-all for research into the basic mechanisms and biochemistry underlying the disease or its treatment. While this kind of research is extremely important for those trying to devise new treatments, it is of little help to a patient trying to find the best current treatment. These papers are often the most technical and difficult to read, and very often can be identified by the extreme (even worse than usual) amount of technicalese in the title.

Next Stop, The Medical Library

After I scan a list of references from a database, or any of the other sources I mentioned, the next stop is the best medical library around. There I make copies of the most interesting complete papers to read at home. While I'm there, I often skim some of the papers and look for other interesting references in them. If I find any, I can get the papers right away. Careful reading of these papers at home may lead questions that prompt another database search, or another visit to the library.

[How to Find and Use a Medical Library]

Most people who are searching for a treatment do not have the luxury of spending an indefinite amount of time on this kind of search. If you are in this situation, you should not delay sharing the information you find with your doctor, and talking to doctors who are doing research or giving innovative new treatments.

You may come up with so many options that it becomes very difficult to make a decision. It is quite common for many of the options to be about equally promising. If you find this, then just pick any one of them. Although it is difficult to make a decision, it is absolutely critical that you actually do make a timely decision. If you don't make a decision, you may end up getting no treatment, rather than the best treatment. No decision is always the worst decision. I have actually seen this happen to patients; don't let it happen to you!

Return To Top