Notes on writing a good abstract

Good scientific abstracts (whether they are for research studies, reviews, or perspectives) are clear and pithy. The primary goal of the abstract is to state your claims succinctly and simply. Secondary goals are to highlight the importance of the claim and to provide supporting evidence. If your abstract is a research paper, then a description of the novel methods or findings should be included. If the abstract is for a review or perspective then a brief description of the review strategy should be included. When finished reading the abstract, the reader should be able to restate your point-of-view, the primary result of your study, and, in general, the methods used to come to that conclusion.

Note that these pointers are taken heavily from my scientific mentor, Russ Altman, and filtered through the foggy glass of time and memory.

Here’s a recipe for biomedical focused papers (although, WOLOG, can be adapted for other fields):

  • A title that is declarative. A title that says exactly what you found. If your reader only reads your title (and many will), what do you want them to know. Keep it simple.
  • 1 sentence that introduces an big and important biomedical problem/field. Russ always used to tell me that someone needs to be dead or dying by the end of the first sentence. If not, then why should the reader care about your study. (This is the most specific part to biomedical sciences, in other fields it would be the biggest challenge facing the field, one that pretty much everyone would agree is significant).
  • 1-2 sentences that focus on a significant challenge that is impeding progress. What is the specific roadblock or barrier keeping the field from moving forward that you plan to address in your study. Don’t get into how you address it, just present the more specific challenge.
  • 1-2 sentences on the opportunity. Here is where you introduce a recent advancement that changes how we can address this problem. It could be that new computing power exists that makes certain algorithms possible now, or that a new data resources has been made available that changes how we can look at a problem.
  • 1-2 sentences on what your study is about. What did you do? How did you take advantage of this new opportunity to address the challenge or roadblock you identified.
  • 2-3 sentences of your key results. What did you find? These should be results that support the conclusions (declaration) in your title.
  • 1 sentence on how the world will change as a result of your findings. What will change about how medicine is practiced or research is conducted because of what you found? What are the broader impacts?

Here is that advice in action.

Data-driven prediction of drug effects and interactions.

Adverse drug events remain a leading cause of morbidity and mortality around the world. Many adverse events are not detected during clinical trials before a drug receives approval for use in the clinic. Fortunately, as part of postmarketing surveillance, regulatory agencies and other institutions maintain large collections of adverse event reports, and these databases present an opportunity to study drug effects from patient population data. However, confounding factors such as concomitant medications, patient demographics, patient medical histories, and reasons for prescribing a drug often are uncharacterized in spontaneous reporting systems, and these omissions can limit the use of quantitative signal detection methods used in the analysis of such data. Here, we present an adaptive data-driven approach for correcting these factors in cases for which the covariates are unknown or unmeasured and combine this approach with existing methods to improve analyses of drug effects using three test data sets. We also present a comprehensive database of drug effects (Offsides) and a database of drug-drug interaction side effects (Twosides). To demonstrate the biological use of these new resources, we used them to identify drug targets, predict drug indications, and discover drug class interactions. We then corroborated 47 (P < 0.0001) of the drug class interactions using an independent analysis of electronic medical records. Our analysis suggests that combined treatment with selective serotonin reuptake inhibitors and thiazides is associated with significantly increased incidence of prolonged QT intervals. We conclude that confounding effects from covariates in observational clinical data can be controlled in data analyses and thus improve the detection and prediction of adverse drug effects and interactions.


Special thanks to Russ Altman, my scientific mentor and PhD advisor at Stanford, whose view on scientific writing forms much of the basis of this strategy.