Limit Technical Information and Caveats in Quality Data Displays

It is very easy to say too much about each of your measures. Many health professionals worry that unless they provide the same level of detail in and around a graph as they would in a report for management or providers, or in an academic paper, they could be misleading consumers rather than informing them. But in presenting data to consumers, less is often better.

What Not To Say

This section discusses various issues that are essential to address in and around your data display. Here are some topics that report sponsors should not present as part of data displays, although you can include them elsewhere in your report:

  • Technical details about scoring methodology.
  • Levels of statistical significance.
  • Confidence intervals.
  • Caveats about the validity or reliability of the data.

Why Not?

It is very difficult to present these topics accurately using language understandable to the many Americans who are only at or below “basic” levels of literacy.[1] Including this information:

  • Makes the page more crowded.
  • Makes many people think the report is not for “them” but rather for “experts.”
  • Makes some people think that the data are being manipulated in ways they do not understand and consequently will not trust.

Exception: Rare Events

There is one exception to the rule of not presenting caveats within a data display: when the measure is of a very rare but serious event. It is difficult to interpret such measures, particularly when making comparisons across plans or providers. Nevertheless, many believe that it is important to be transparent about these rare events and to include them in reports, in part to educate the public about the fact that these events can and should be prevented.

The National Quality Forum has recommended presenting such measures as raw counts rather than as rates.[2] Report designers may want to consider adding a caveat to displays of rare or “never events.” For example, a report on a measure of hip fracture mortality in the hospital may say:

“It is very rare for a patient who comes to the hospital to take care of a hip fracture to die during their stay. Such deaths are considered preventable. (The report sponsor) believes it is important to share information about these rare events with the public, both to increase knowledge and to encourage hospitals to do everything possible to prevent them.”

[1] Kutner M, Greenberg E, Jin Y, et al. Literacy in everyday life: results from the 2003 National Assessment of Adult Literacy (NCES 2007–480). 2007. U.S. Department of Education. Washington, DC: National Center for Education Statistics. Available at http://nces.ed.gov/Pubs2007/2007480_1.pdf.
[2] National Quality Forum (NQF). National Voluntary Consensus Standards for Hospital Care 2007—Guidelines for Consumer-Focused Public Reporting: A Consensus Report. Washington, DC:NQF; 2008. Available at http://www.qualityforum.org/Publications/2008/08/National_Voluntary_Consensus_Standards_for_Hospital_Care_2007__Performance_Measures.aspx.

Also in "Describing Measures in User-Friendly Ways"

Page last reviewed November 2018
Page originally created February 2015
Internet Citation: Limit Technical Information and Caveats in Quality Data Displays. Content last reviewed November 2018. Agency for Healthcare Research and Quality, Rockville, MD. https://www.ahrq.gov/talkingquality/translate/labels/limit-info.html
Back To Top