Tuesday, September 5, 2017

Using clinical risk scores wisely

- Jennifer Middleton, MD, MPH

Physicians have several clinical calculator apps to choose from, but guidance about choosing the right score and interpreting its results isn't always as readily available. Busy family physicians looking to enhance their use of clinical risk scores will find several discussed among the articles in the current issue of AFP; understanding the nuances of each may help physicians choose the best ones to "favorite" in their calculator app of choice.

A practice guideline on "Newly Detected Atrial Fibrillation" and an editorial on the "Differences Between the AAFP Atrial Fibrillation Guideline and the AHA/ACC/HRS Guideline" both include a discussion on risk scores to predict stroke and bleeding risk in these patients. Using the CHA2DS2-VASc score increases the number of persons recommended to receive anticoagulation compared to the CHADS2 score, but the authors of both articles argue that these risk scores' ability to predict stroke risk is identical. Interestingly, neither of the clinical calculator apps that I have on my smartphone include the CHA2DS2-VASc score. The practice guideline does describe the HAS-BLED score's ability to predict bleeding risk as "slightly better" than other bleeding risk scores for patients on anticoagulation.

"Pleuritic Chest Pain: Sorting Through the Differential Diagnosis" discusses the importance of ruling out pulmonary embolism (PE), the most common life-threatening cause of pleuritic chest pain. The authors advocate for using a validated risk score in patients presenting with pleuritic chest pain to guide decisions about testing for PE; one of the reference articles describes several available validated risk scores but lists the Wells rule as "widely validated and commonly used;" regardless of the score used, a negative D-dimer test in a patient with a low pre-test probability score usually negates the need for further testing.

Similarly, "Exercise Stress Testing: Indications and Common Questions" discusses the use of the Diamond and Forrester score to calculate the pre-test probability of coronary artery disease (CAD) in patients with chest pain. Exercise stress testing provides the highest diagnostic utility in patients with an intermediate pre-test probability for CAD; low risk patients with negative cardiac enzymes typically require no further testing, and high risk patients should receive prompt intervention.

The AFP By Topic on Point-of-Care Guides provides not only numerous risk scores to use with patients but also an evidence-based summary of how to use them each in practice. You can bookmark this department collection and also save your most-used clinical calculator websites under your AAFP "Favorites" tab for easy future reference.

1 comment:

  1. Clinical risk scores can certainly be helpful in guiding clinicians regarding the likelihood of risks and optimal clinical decision-making. That said, “data out is only as good as data in” — and over dependence on a numerical score without carefully considering context and validity of the “data in” predisposes to erroneous conclusions. This is especially true for the 3 conditions cited here regarding anticoagulation of patients with atrial fibrillation; Ruling out pulmonary embolism; and Use of exercise stress testing. The “art” of clinical decision-making for these conditions (and for other conditions for which “risk scores” are used) extends well beyond a numerical “risk score”.

    ReplyDelete