The tables that make up this section of Understanding Breast Cancer present the evidence that drives many of the recommendations and standards of practice related to breast cancer.
Research tables are useful for presenting data, but they can be hard to understand if you don’t work with them every day. They present a lot of information in a very dense format.
Here, we present some basic concepts of research tables that can help you read the tables in this section of Understanding Breast Cancer.
The table below provides examples for this discussion. It has many of the features that you will see in all of the tables.
To learn about a certain table characteristic, review its number.
Sample Table 5: Dietary fat consumption and breast cancer risk
Selection Criteria
When reviewing the evidence on a specific topic, it is important to recognize “good” studies. Good studies are those that have been well thought out. Not all scientific studies are created equal in their ability to help answer scientific questions. Because of these differences, most scientific reviews—such as these tables—set certain standards for the studies they include. These standards are called “selection criteria” and are listed for each table in this section of Understanding Breast Cancer.
For example, you will see that many of the tables list as one of their selection criteria a minimum number of cases of breast cancer or participants in a study. When all else is equal, a larger number of study participants means that a study is better able to examine scientific questions. Although it is possible to have a large, poorly-designed study, large studies are usually better than small studies. Larger studies have more “statistical power”. This simply means that the results from large studies are less likely to be due to chance than those from small studies. (For more information on statistical power, click here.)
The Studies
This first column, located on the left hand side, describes the studies listed in the tables. The type of study design (for example, randomized controlled trial, prospective cohort, case-control, patient series) is listed first, in bold. This is followed by either the name of the study or the name of the first author of the published article. (For more information, visit Types of Research Studies.)
A reference list is provided after each table so that you can find the original study report. Sometimes, a table will report the results of only one analysis. This can occur for two reasons. Either there is only one study that meets the selection criteria, or, as in the case of Table 32, there is one report that combines data from a number of studies into a single large analysis.
Study Population
The second column describes the sample population of each study. The study population of a randomized controlled trial is the total number of people at the start of the study who were randomized to either the treatment or control condition. For prospective cohort studies present, the study population is the number of people at the start of the study (baseline cohort). And for case-control studies, it is the number of cases and the number of controls.
In some tables, more details of the study participants are included. For example, Table 11, on BRCA1 and BRCA2 gene mutations, has two columns describing the study populations. One column describes some characteristics of the people in the study, while the next column shows the number of families or individuals in the studies.
Length of Follow-up
Randomized controlled trials and prospective cohort studies follow people forward in time to see who will have the outcome of interest (such as breast cancer, recurrence of breast cancer or breast cancer survival). For these studies, the length of time of follow-up (that is, how many months or years people were followed in the study) is provided in the tables.
Because case-control studies do not involve following participants forward in time, there are no columns for length of follow-up for these studies.
Tables that focus on cumulative risk (discussed below) can also show the length of follow-up. These tables give the length of time, or age range, used in the cumulative risk calculation. Table 11, for example, shows the cumulative risk of cancer up to age 70 for people with a BRCA mutation.
Other Characteristics
Many tables have columns that give other information on the study population or the risk factor being studied. For example, Table 29b has a column giving age ranges for the study populations; Table 4 has a column that gives definitions of physical activity used in the studies; and Table 28 has information on the dose and type of risk-lowering drugs used in the studies. These types of information show you not only the specifics of each study, but also how the studies in a table are similar to, as well as different from, each other.
Different studies on the same topic can differ in important ways, such as length of follow-up and definitions of a risk factor. Studies may look at outcomes among women of different ages or menopausal status. They may use different levels such as “high” or “low” to define an exposure. It is important to keep in mind these differences across studies when you review the findings shown in a table. Given differences in study design, study populations and risk factor definitions, differences in findings among studies on the same topic are not surprising.
Quantitative Findings-Understanding the Numbers
All the information in the tables is important, but the main purpose of the tables is to present quantitative findings—the numbers that represent the risk linked to the topic being studied.
These numbers are shown in the remaining columns of the tables. Before looking at the numbers in these columns, it is important to know what they represent. What is the outcome of interest—is it breast cancer? Is it five-year survival? Is it recurrence?
Are groups being compared to each other? If so, what groups are being compared?
The headings of the columns with the quantitative information help to answer these questions. For example, in Table 29b, there is only one column with quantitative findings, the last column. These numbers show the relative risk of dying of breast cancer—comparing those who had screening mammography to those who did not. (For more information on risk, see the Risk Factors and Prevention section.)
Most medical studies report risk measures, such as relative risks, odds ratios and averages, with 95 percent confidence intervals (95% CI). A 95% CI around a risk measure means that there is a 95 percent chance that the "true" measure falls within the interval. Because there is random error in studies, and studies are only samples of a much larger population, a single study does not give the “one” correct answer. There is always a range of likely answers. A single study gives a “best estimate” along with a 95 % CI of a likely range.
For relative risks and odds ratios, a 95% CI that includes the number 1.0 means there may be no link between an exposure (like a risk factor or a treatment) and an outcome (like breast cancer or breast cancer survival). When this happens, the results are said to be “not statistically significant”—meaning that the finding could simply be due to chance. On the other hand, if a 95% CI does not include 1.0, this means the results are statistically significant and it is unlikely that the results are due to chance. Statistical significance is a key concept in health research and it can be measured and presented in many ways. In all cases, however, a statistically significant result means there is very likely a true link between an exposure and an outcome.
Summary Relative Risks from Meta-analyses
Some tables have a summary relative risk reported at the bottom. In Table 29b, the summary relative risk comes from a meta-analysis of many of the studies listed in the table. A meta-analysis takes relative risks reported in different studies and “averages” them to come up with a single, summary measure.
Summary Relative Risks from Pooled Analyses
One of the summary relative risks at the bottom of Table 5 came from a pooled analysis. A pooled analysis is different than a meta-analysis. In a meta-analysis, the quantitative findings from different studies are “averaged” together. In a pooled analysis, all the people in the studies are combined into one large set of data and analyses are done as if it were one big study.
A pooled analysis is almost always better than a meta-analysis. In a meta-analysis, researchers analyze results already published by others, and this has limitations. On the other hand, in a pooled analysis, researchers combine all the data from the various studies and analyze them as they think best. Because researchers have this freedom in a pooled analysis, pooled analyses usually have more statistical power than meta-analyses. More statistical power means there is a greater probability that the results are not simply due to chance.
Cumulative Risk
In some tables, the quantitative findings are presented as a cumulative risk, often in the form of a percentage. Table 11, for example, shows quantitative findings on the cumulative risk of different types of cancer up to age 70 years for women who either have a BRCA1/BRCA2 mutation or have a family member with a BRCA1/BRCA2 mutation.
In the first study listed in this table (the Breast Cancer Linkage Consortium), 87 percent of women who came from families with a BRCA1 mutation had breast cancer by age 70 and 63 percent of these women had ovarian cancer by age 70.
Sensitivity
Some tables show quantitative findings on the sensitivity of a procedure (also in the form of a percentage). The definition of sensitivity is the proportion (or percentage) of people who truly have the condition of interest who “test positive” for that condition. For example, Table 33 gives findings on the sensitivity of sentinel node biopsy in predicting axillary lymph node status. In this case, the percentages listed in the final column show the percentage of women who truly had positive axillary nodes who also had a positive sentinel node biopsy.
The higher the sensitivity (closer to 100 percent), the better a test (e.g., biopsy) is at identifying people with the condition. For example, a sensitivity of 92 percent means that 92 percent of women identified as having positive axillary nodes by a sentinel node biopsy, actually had positive axillary nodes.
Finding Individual Studies
In some cases, you may want to see more detail about an individual study than is given in a summary table. Therefore, the references for all the studies in a table are listed below the table. This list includes the authors of the study article, the title of the article, the year the article was published and the title and specific issue of the medical journal in which the article appeared.
If you live near a university with a medical or public health school, you may be able to go to the school's medical library to get a copy of an article. Local public libraries may not carry medical journals, but they may be able to locate a copy of an article from another source. Many medical journals also have websites and offer full text articles on the Internet—some for free, some for a fee and some only for subscribers. PubMed, the National Library of Medicine's medical literature search engine, is also a good source for finding summaries of journal articles, called abstracts. For some abstracts, PubMed also has links to the full text articles.
For More Information
If you would like to read more about health research, an introductory epidemiology textbook may be a good place to start. One example is Epidemiology in Medicine by C. H. Hennekens and J. E. Buring (Lippincott Williams & Wilkins, 1987). There are many epidemiology textbooks and a medical librarian can help you find one that goes as in-depth as you would like.
Other textbooks:
Rothman KJ. Epidemiology: An Introduction (Oxford University Press, 2002)
Aschengrau A and Seage GR, III. Essentials of Epidemiology in Public Health, 2nd edition (Jones & Bartlett Publishers, 2008).
Updated 09/12/09
|
