If you don't have any sort of topic, try browsing news sites for health topics of current interest.
Note: You may not find much about the specific study or research highlighted in the news article, but the general topic should be good: i.e. an article about a particular diabetes treatment could give rise to a topic about diabetes treatment in general, diabetes drug treatment, or dietary management of diabetes.
If you still don't have any ideas for topics, check out some of the findings from the largest and longest running health studies, the Nurses Health Study (NHS I and II) and the Health Professionals Follow-up Study (HPFS). These studies have produced some of the best environment and diet related research. Their topics, like diet and environment related diseases, make very good topics.
A topic that is too broad will be hard to search because you'll get too much information, or because the information falls naturally into smaller categories.
'Cancer' is too broad because research focuses on different types of cancer. You'll be best off picking a specific type.
'Breast cancer' may be too broad because there has been a lot of research done on it, and lots of factors contributing to it. You'll be better off if you pick a particular type of treatment, aspect of the disease, or population.
Note: Broad topics are good for searching the library catalog, because books tend to tackle broader topics than articles do.
A topic that is too narrow will be hard to search because you won't find very much that is as specific as you want.
'Nutrition and diet factors of breast cancer in African American women between the ages of 30-35 in Hartford, CT' is too narrow. There might be one study done on this specific aspect in this specific population, but there aren't likely to be many more. You'll need to broaden your search: nutrition and diet factors of breast cancer (in general), breast cancer in areas similar to or with populations like Hartford (i.e. in African Americans, in New England, or in urban centers).
Individual articles will be very narrow, but you are better off searching for a somewhat broader topic so that you get research that you can relate or compare to your preferred topic.
Generally, it's easier to start broad and narrow as you go than the other way around.
Are the following topics broad or narrow? What could you add to or remove from these topics to make them more broad or more narrow?