Many research questions can be answered quickly and efficiently using data or specimens that have already been collected. There are three general approaches to using these existing resources. Secondary data analysis is the use of existing data to investigate research questions other than the main ones for which the data were originally gathered. Ancillary studies add one or more measurements to a study, often in a subset of the participants, to answer a separate research question. Systematic reviews combine the results of multiple previous studies of a given research question, often including calculation of a summary estimate of effect that has greater precision than the individual study estimates. Making creative use of existing data and specimens is a fast and effective way for new investigators with limited resources to begin to answer important research questions, gain valuable experience in a research area, and sometimes have a publishable finding in a short time frame.
■ ADVANTAGES AND DISADVANTAGES
The main advantages of studies using existing data are speed and economy. A research question that might otherwise require much time and money to investigate can sometimes be answered rapidly and inexpensively.
Studies using existing data or specimens also have disadvantages. The selection of the population to study, which data to collect, the quality of data gathered, and how variables were measured and recorded are all predetermined. The existing data may have been collected from a population that is not ideal (e.g., men only rather than men and women), the measurement approach may not be what the investigator would prefer (history of hypertension, a dichotomous historical variable, in place of actual blood pressure), and the quality of the data may be poor (frequent missing or incorrect values). Important confounders and outcomes may not have been measured or recorded.
Research using secondary data takes advantage of the fact that most of the data needed to answer a research question are already available. In an ancillary study, the investigator adds one or several measurements to an existing study to answer a different research question.
Ancillary studies have many of the advantages of secondary data analysis with fewer constraints.
They are both inexpensive and efficient, and the investigator can design a few key ancillary measurements specifically to answer the research question. Ancillary studies can be added to any type of study, including cross-sectional and case–control studies, but large prospective cohort studies and randomized trials are particularly well suited.
Ancillary studies have the problem that the measurements may be most informative when added before the study begins, and it may be difficult for an outsider to identify studies in the planning phase. Even when a variable was not measured at baseline, however, a single measurement during or at the end of a trial can produce useful information.
Systematic reviews identify a set of completed studies that address a particular research question, and evaluate the results of these studies to arrive at conclusions about a body of research. In contrast to other approaches to reviewing the literature, a systematic review uses a well- defined approach to identify all relevant studies, display the characteristics and results of eligible studies, and, when appropriate, calculate a summary estimate of the overall results. The statistical aspects of a systematic review (calculating summary effect estimates and variance, statistical tests of heterogeneity, and statistical estimates of publication bias) are called meta-analysis.
A systematic review can be a great opportunity for a new investigator. Although it takes a surprising amount of time and effort, a systematic review generally does not require substantial financial or other resources.