According to Multiple Sources

“There’s three sides to every story, babe: there’s yours and there’s mine and the cold hard truth.”

—Don Henley, Long Way Home

If there is an upside to the mushrooming of “fake news,” it is the increased awareness that reliance on a single source of information (even a trustworthy source) is inferior to seeking multiple perspectives. Facts or data points are(generally) more credible when derived from diverse data sets.

In the 1970s, Norman Denzin, a sociology professor, investigated the concept of triangulation—long used for navigation purposes—as a research technique. He and subsequent investigators understood that arriving at the same conclusion via different pathways boosts our confidence that we are accounting for the bias or limitations of any individual route. This is particularly poignant in the study of patient safety, where a single safety study or report draws attention, or an isolated tragedy triggers calls for action before the breadth and depth of the problem is understood.

Of course, health care providers cannot simply ignore a single patient safety data point—or dismiss it as “fake news.” But raising alarm or directing resources to a potential problem that has not been verified or validated or triangulated with other data sources can be counterproductive. Enthusiasm for process improvement or behavior change will wane without confirmation that that single incident or data point is borne out by other information. Optimally, multiple information sources highlight a common, fertile issue or risk that paves the way for the sustained attention of multiple parties.

Fortunately,health care leaders have access to a rapidly expanding array of data sources to tap into. In addition to malpractice data, many organizations (including PSOs) can employ incident reports, patient complaints,root cause analyses, and a whole raft of quality and regulatory metrics. With some forethought, the terms and taxonomies used to populate these databases can be standardized (or mapped) so that apples can be compared to apples.

When feasible, advanced analytic studies pulling from multiple data sets offer extraordinary opportunities to tease out precise risk factors and hidden vulnerabilities. But in many circumstances, you might just want to be able to point to the fact that three or four independent databases highlight the same broad problem area. At a minimum, triangulation puts your patient safety efforts on the right track to the points where you need to dig deeper.

Additional Material

• The association between patient safety indicators and medical malpractice risk
• Examining proneness to malpractice claims
• Investigating selected patient safety indicators using medical records data
• The use of triangulation in qualitative research [abstract]
• What to do with healthcare incident reporting systems
• Patient complaints in healthcare systems: a systematic review and coding taxonomy
• EHR data better for sepsis surveillance than claims data