Displaying the Data in a Health Care Quality Report

Graphs and tables remain the most efficient and practical way to convey a large amount of information, especially comparative information and numbers. Visual presentations are powerful tools for concisely making points that are hard to put into words.

However, while some consumers prefer a graphic presentation, others regard charts as a foreign language. Either they do not know how to read them, or they simply find the process too burdensome. Moreover, some people who do understand what they are seeing still respond negatively to charts, although that reaction can change once they get oriented. This suggests that you need to make a special effort to make your charts attractive and easy to interpret.[1]

Consider the following guidelines for making your charts as user-friendly as possible:

Make Graphics Self-Explanatory

Since people are unlikely to read any explanations in the front of a performance report, displays of comparative information have to be as self-explanatory as possible. That is, people should not have to work too hard to interpret them. Anything important for understanding the chart—like a legend or a definition—has to be on the same page, preferably at the top of the page or on the side (i.e., not on the bottom) so that the reader won't miss it. It also needs to be accurate and easy to understand.

The golden rule of chart design is as follows: If people need a lot of explanation to understand your graphic, there's something wrong with the graphic.

To make a chart self-explanatory, you'll need to provide tools that enable the reader to decode the information. Titles, legends, and other explanatory information should tell readers everything they need to know. Here are some suggestions for making these elements of a performance report as useful as possible:

  • Assume that you have to explain the meaning of everything, especially symbols and colors. Don't take anything for granted. Something that seems intuitive to you may not be obvious to your audience. For instance, a bar that shows the average score would seem self-explanatory, but cognitive testing with users indicates it often is not; readers need an explanation of what it is and how to use it.
  • Create eye-catching, user-friendly legends that tell users how to read the chart. Make sure the legend has a prominent place on the page where the reader is sure to see it. Again, don’t assume people can read a chart without the legend. Here’s one example of a good legend.

EXAMPLE: Displaying a User-Friendly Legend

Title: Nursing Home Compare
Sponsor: Centers for Medicare & Medicaid Services (CMS)
URL: http://www.medicare.gov/nursinghomecompare/search.html

The Nursing Home Compare report includes a user-friendly legend in a prominent place to explain the symbols used in the report. Users can also use the ratings in this legend to filter their options.

Source: Centers for Medicare & Medicaid Services (CMS). Nursing Home Compare. Available at http://www.medicare.gov/nursinghomecompare/search.html.

  • Repeat these elements wherever they are needed so that the reader does not have to look somewhere else for an explanation. For instance, if you use the same kind of graph on several pages, include a legend for that graph on each of those pages. Many people do not start at the beginning of a report, whether in print or on the Web, and move forward page by page. This means that key explanations need to appear on each page where they are relevant.
  • Use plain English in the titles and headings. If you have to use abbreviations or industry terms:
    • Explain them in words that readers will understand.
    • Place any definitions on each page where the terms are used, as close to the term as possible. Readers are likely to lose their place if they have to skip over several lines of text.
    • Don’t explain abbreviations or unfamiliar terms in a glossary, which few people will use.

    Learn more about using labels and definitions in Describing Measure in User-friendly Ways.

  • Avoid using too many footnotes. These are often difficult to read, since they are typically in a smaller font. They also make your report appear to be for “academics” rather than ordinary folk.

Make Bar Charts Easy To Interpret

The most common graphic used in comparative quality reports is a bar chart, where the length of the bar is equivalent to the numerical score. Many people are unfamiliar and uncomfortable with graphics of this kind.[2]

There are several ways to make the bars more easily interpretable.

  • Rather than display multiple responses in a segmented bar, offer only the response that is most relevant to your audience. For example, if you are reporting results to a survey item that has multiple responses, don’t segment the bar to show the percentage of each of the responses. Instead, show only one response–typically the most desirable one.
  • Augment the purely visual cue of the length of the bar with the actual number. In some cases, you can do this within the bar itself; in other cases, the number will be more readable if it is just to the right of the bar in a horizontal bar graph or on top of a vertical bar.
  • On the bottom of the chart, provide a scale showing at least zero, 100, and a midpoint to help orient the reader visually to the relationship between the length of the bar and the numerical score.
  • Use an easily readable color for the bars if you can use color. Minimize use of green or red, as a subset of the population cannot easily see these colors. If you are showing one or more comparators, such as a State or National average score, consider using a patterned bar or a lighter tone of the same color, rather than a bar in a different color.
  • Order the bars from the best to the worst performance. Learn more about rank ordering.
  • Write the title of the bar chart very carefully to make sure it describes exactly what the bars represent.

Here’s an example of a horizontal bar chart that is using all these tips:

Title: Model Report on Comparative Hospice Quality
Sponsor: American Hospice Foundation
URL: Not available online.

This chart from a model report illustrates the use of all of the tips provided above.

Source: American Hospice Foundation. Model Report on Comparative Hospice Quality. American Hospice Foundation: Washington, DC; 2008.

Provide Self-Explanatory Symbols

If you are creating a table that uses one or more symbols, it is important to choose the symbols carefully. The ability of consumers to understand and interpret the symbols will play a large role in determining how self-explanatory the table is. Be sure to either test the symbols with your audience or take advantage of symbols that have been tested extensively with positive results. Use this step to confirm that:

  • Your audience can use the symbols to draw accurate conclusions about the relative performance of the entities in your report. A display of arrows pointing up and down, for example, can be difficult for the reader to process.
  • The symbols do not elicit a strong reaction from any subpopulations in your audience. Experienced designers have learned that some symbols are highly charged, i.e., have specific meaning for certain people.

Learn how to test symbols with your audience in The Purpose and Process of Cognitive Testing.

Traditional Symbols for Showing Relative Performance

Traditional symbols include stars, diamonds, checkmarks, and arrows, as well as circles that are either empty, half full, or full. Stars are among the most commonly used symbols in various rating schemes, so it is no surprise that they are also common in quality reports. These symbols are typically used to show performance “relative” to each other or to some other standard. People generally understand that more symbols—for example, more stars—convey better performance. Learn about Choosing a Point of Comparison.

Assigning symbols to each entity rated. The way you assign symbols, and how many you assign, depends upon how you have decided to score your data. Learn about Generating Health Care Quality Scores That Show Differences.

For example, if you use the average score as your basis for comparing providers or plans, you might display one symbol (e.g., a star or diamond) for scores that are lower than average, two for average, and three for better than average. Some reports use five levels of performance, which can provide finer distinctions. In some cases, the five levels are determined by percentiles in a distribution of scores.

Word Icons

Researchers have tested word icons as a way to provide relative information in a chart. Here’s what a word icon chart looks like:

A word icon chart

Source: Adapted from fictional materials used in a CAHPS II research study conducted by American Institutes for Research (led by Kristin Carman, PhD) in collaboration with Jeanne McGee, PhD (McGee & Evers Consulting, Inc.) and Judith Hibbard, DrPH (University of Oregon). For more on this study, go to http://archive.ahrq.gov/cahps/news-and-events/events/UGM10/Day2_b_2_McGee-Carman-Handout.pdf.

This kind of chart uses a symbol that combines three characteristics—color, language and a shape—to distinguish among those rates:

  • Above average performance is indicated by a yellow circle and the word “better.”
  • Below average performance is indicated by a blue downward arrow and the word “below.”
  • Average performance is indicated by the word “average” in a smaller font in a pale, grayed out tone.

The scores that are average are purposely designed to fade into the background while the non-average scores are more visible. Both laboratory studies and cognitive testing of this approach show that it is extremely easy to understand for consumers at various literacy levels, more so than other kinds of symbol-based charts.[3]

Other Purposes of Symbols

There are two other purposes for symbols:

  • To indicate the “best value” or “top performer” in a group. Symbols can be used to reinforce the status of being the highest performer, or in the very highest tier of performance.
  • To highlight recognition awards. Some quality-oriented organizations provide recognition awards to health care providers. For example:
    • Physicians can get recognition from the National Committee for Quality Assurance for meeting standards for overall care for a condition such as diabetes or for having a practice organized to provide patient-centered care.
    • Some hospitals have been awarded “magnet” status to indicate that they do an excellent job of recruiting and retaining their nursing staff.

Using symbols to highlight these awards help to get the attention of readers. The accrediting organization may have created its own symbol for this purpose; if so, use that symbol, which may become a standard that people look for, like the “Good Housekeeping Seal of Approval.”

Limit the Size of Tables

Anyone who has tried to understand a table that shows the benefits offered by a number of health insurance plans knows how hard it is to follow giant matrices. In a table with dozens of data points (whether numerical scores or symbols), there's simply too much information for people to look at, making it almost impossible to make comparisons and find patterns.

In addition to being boring and intimidating, large tables also require too much effort to find specific pieces of information. People can keep only seven, plus or minus two, ideas in their short-term memory at one time,[4] and those who are older or have lower literacy levels are in the “minus” group. For that reason, it is advisable to show no more than seven providers, or no more than seven measures, in a single graph or table.

If you have more data to present, consider breaking the information into smaller chunks, preferably in a way that has meaning for your readers. For example, rather than listing all the hospitals or medical groups in the State in one table:

  • Create multiple tables that show only the providers available in each region or county, or only the providers that the user selects.
  • Show all the clinical quality measures in one table and the measures related to patient experience in another.

One advantage of a Web-based report is that users can decide what they want to see—for example, which providers to compare or which measures to examine. You may want to limit the number of providers or measures they can select to ensure that the amount of information is manageable.

For related guidance, go to Tips on Designing a Quality Report.

Explain How to Read and Interpret Charts

The ability to understand the format and structure of a chart, whether a table or a graph, is something that one learns in school and sometimes, but not always, remembers. Given that, don’t assume that people will intuitively know how to follow the flow of information in a table (i.e., to look up and down the columns and from left to right on the rows). Similarly, they do not necessarily know to look for a legend or to read the axes of a graph. For people who lack the training and experience, a chart can be very intimidating.

In your report, you must explain how to read and interpret any graphic you present. Here are some ways to do that:

  • Refer to the chart within the text accompanying it, making it clear to the reader what to look at and what to pay attention to (e.g., the difference between a plan's score and a National benchmark).
  • When appropriate, summarize the key points illustrated by the graphic (e.g., that all plans performed above the State average).
  • Since people take in information in different ways, repeat and reinforce key points in text and graphics.
  • Explain the purpose of comparators. For example, it may be necessary to explain why a bar showing average performance is included in a horizontal bar graph.

EXAMPLE: Explaining the Meaning of “Average”

Title: Model Report for AHRQ Quality Indicators
Sponsor: Agency for Healthcare Research and Quality
URL: http://qualityindicators.ahrq.gov/Downloads/Modules/QI_Reporting/Model_Report_Composite.pdf

When developing model reports for the AHRQ Quality Indicators, researchers learned through cognitive testing that the following language works to explain the function and meaning of an “average bar.”

Average of hospitals across the state: The average rate of patients who died after having a heart attack, in hospitals across your state. This number is included so you have:

  • a better idea of what is typical for your state.
  • a basis for comparing individual hospitals’ performance.

In the example, the explanation is on the bottom of a bar graph showing hospitals’ performance on a measure of mortality among cardiac patients.

Death rate for coronary artery bypass graft

When you are choosing a hospital, you should look for the hospital that has a lower number of deaths for this operation. A lower number is shown by a shorter bar on the graph below.

1 out of 100 patients who underwent a coronary artery bypass graft died at Hospital D. 2 out of 100 patients who underwent a coronary artery bypass graft died at Hospital B. 3 out of 100 patients who underwent a coronary artery bypass graft died at Hospital A. 11 out of 100 patients who underwent a coronary artery bypass graft died at Hospital C. On average, across the entire state, 3 out of 100 patients who underwent a coronary artery bypass graft died.

Source: Agency for Healthcare Research and Quality. Hospital Quality Model Report: Composites. 2009. Available at http://qualityindicators.ahrq.gov/modules/Default.aspx.

Avoid Abbreviations and Jargon

Sponsors often use abbreviated text in charts, usually in a well-intentioned effort to conserve space. But abbreviations (e.g., PCMH, ACO, NCQA, CABG), like industry jargon, are not familiar to most health care consumers. If your readers cannot understand the language in the chart, they are unlikely to stick around to tackle the data or scores you are trying to present.

While you could explain the abbreviations in text associated with the chart, it is better to make the chart stand on its own by spelling out all the terminology in the chart itself and using words your audiences is likely to understand. One reason for this approach is that people will not read everything you put on the page, so they may miss your explanation. Another reason involves the limited capacity of readers to process multiple pieces of information; researchers have found that the more you make people read, the less likely they are to understand the message you are trying to convey. To learn more about using reader-centered language, go to Tips on Writing a Quality Report.


[1] Peters EM, Dieckmann N, Dixon A, Hibbard JH, Mertz CK. Less is More in Presenting Quality Information to Consumers. Medical Care Research and Review 2007. 64(2):169-190.

[2] To learn more about quantitative literacy, see:

  • Paulos JA. Innumeracy: Mathematical Illiteracy and its Consequences. New York: Hill and Wang; 2001.
  • Peters EM, Hibbard JH, Slovic P, Diechmann N. Numeracy skill and the communication, comprehension, and use of risk and benefit information. Health Affairs 2007 May/June. 26(3):741-748.

[3] Carman K. Improving quality information in a consumer-driven era: Showing differences is crucial to informed consumer choice. Presentation delivered at the 10th National CAHPS User Group Meeting, March 31, 2006. Pages 9-11. Available at http://archive.ahrq.gov/cahps/news-and-events/events/UGM10/DAY2_cd_1_Carman.pdf. Also see: McGee J. Best Practices for Presenting Quality Data. Presentation delivered at the 11th National CAHPS User Group Meeting, CAHPS College, December 3, 2008. Slide 21. Available at http://archive.ahrq.gov/cahps/news-and-events/events/UGM11/McGee.pdf.

[4] Miller GA. The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information. Psychological Review 1956, 63:81-97. Available at http://www.musanim.com/miller1956

Also in "Translate Data Into Information"

Page last reviewed May 2019
Page originally created February 2015
Internet Citation: Displaying the Data in a Health Care Quality Report. Content last reviewed May 2019. Agency for Healthcare Research and Quality, Rockville, MD. https://www.ahrq.gov/talkingquality/translate/display/index.html
Back To Top