September 2019

Understanding Sensitivity and Specificity of Screening Tools for Early Detection of DFUs

By Lynn Soban, PhD, MPH, RN

In his June anniversary issue commentary, Proper Diabetes Care Can Reduce Rising Lower Extremity Amputation Rates (LER July 2019), Mark Hinkes, DPM, responded to the disheartening news that lower extremity amputations are increasing in the United States with a call to action: “I believe that the time has come for the standard of care of diabetic foot pathology to be PROACTIVE and that change in philosophy embraces PREVENTION.

Today’s standard practices for diabetic foot screening involve physician and patient assessments that are high touch and low tech…and time-consuming. At a time where technology has profoundly changed our daily lives, it is hard to believe these effects have not spilled over into the prevention and early detection of diabetic foot ulcers (DFUs). But reading the new study by Lavery et al., “Unilateral remote temperature monitoring to predict future ulceration for the diabetic foot in remission,”1 makes me optimistic that the tide is about to turn. If this happens, the next question will be: “how do clinicians judge the effectiveness of such new technologies?”

In the spirit of optimism, I provide a very brief answer to this question.

The accuracy of a screening test is measured by its sensitivity and specificity.

Sensitivity refers to the proportion of persons with a condition who correctly test positive when screened. A test with poor sensitivity will yield many false-negatives.

Specificity refers to the proportion of persons without a condition who correctly test negative when screened. A test with poor specificity will yield many false-positives. #1069567532

There is a tradeoff between sensitivity and specificity: an increase in sensitivity (the ability to correctly identify people with the condition) results in a decrease in specificity (the ability to correctly identify those who do not have the condition) and vice versa. Airport metal detectors are a prime example of this: to ensure that true threats are detected, airport metal detectors are set to levels of very high sensitivity which result in lower levels of specificity. As a result, travelers frequently experience false positive screening results related to coins, belts, and keys. The public safety decision was made that it would be safer for everyone to have a few people—the false positives—be re-screened, rather than lower the sensitivity and miss a risk.

In medical screening, false positive results can produce a series of adverse consequences including psychological stress and risks associated with further tests and procedures. The results of sensitivity and specificity analyses need to be weighed within the context of the disease in terms of the relative importance of false positivity versus false negativity.

New Tools Are Here…NOW

A brief description of the findings from Lavery et al appears on page 13. These findings show that use of a smart mat to measure foot temperature has promise for predicting development of DFUs among patients with a prior amputation an average of 41 days before clinical presentation.1 Statistical models indicated that the mat had a sensitivity of 91% and a specificity of 46%.

Similar to the airport metal detectors, the smart mat is characterized by high sensitivity and lower specificity. In other words: the mat correctly predicted 91% of impending non-acute plantar foot ulcers (a 9% false negative rate); and correctly predicted 46% of patients who did not have an impending non-acute plantar foot ulcer (a 54% false positive rate). What makes me optimistic is that the DFU is only impending—still 41 days away! Such early warning means there’s still time for prevention. #513237084

A 54% false positive rate appears to be a reasonable tradeoff given the high the stakes that accompany development of a DFU:

  • 85% of all lower extremity amputations in the diabetic population are proceeded by a foot ulcer;
  • 38% of all diabetic patients who undergo lower extremity amputations will lose the other leg in 3 years;
  • and 75% of diabetic patients who undergo lower extremity amputations will not survive 5 years.

Early results also indicate that this monitoring offers an easy and acceptable way for patients to monitor their feet for signs of impending DFUs.

Technology is not the only answer and, given the likelihood of false positive results, current standards of care will remain an important part of DFU prevention, including following up positive results to determine accuracy. The good news today is that technology appears to be moving us closer to the goal of proactive, preventive care for DFUs.

Lynn Soban, PhD, MPH, RN, is a health services researcher and implementation scientist with expertise in quality improvement related to pressure ulcer prevention.  She lives in Santa Monica, California.

  1. Lavery LA, Petersen BJ, Linders DR, et al. Unilateral remote temperature monitoring to predict future ulceration for the diabetic foot in remission. BMJ Open Diab Res Care. 2019; 7: e000696.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.