History of Diagnostic Testing

1978 Clinicians bad at updating testing probabilities [1], still bad in 2014 [2]
2003 Cognitive bias leads to diagnostic errors [3]
2005? natural frequencies work better than 2x2 tables to apply probability
Visual grids improve understanding diagnostic probability (Hoffrage, date?)
2011 Laboratory testing leads to Anemia in hospitalized patients [4]
GRADE statement on testing
2014 Testing accounts for ~15 (6-27%)% of diagnoses (most from history & physical) https://www.acpjournals.org/doi/abs/10.7326/M14-0461
2015 NAM report identifying dx error
2014 Diagnostic tests often fail to lead to changes in patient outcomes
2017—unnecessary testing is common, 40% of tests aren’t known by inpatient providers (Koch et al AJM)
Clinical cascades [5] are often triggered by testing [6] (AIM piece? JAMA NO)
Overdiagnosis book on harm of testing
14.8 billion tests/year in US
2017 EBM guidelines for repeat laboratory testing
2021 Clinicians poorly estimate chance of disease before & after testing
2021 Clinicians perform more unnecessary testing who are medical maximizers, poor numeracy, deny uncertainty and have NP/PA training

[1] https://www.nejm.org/doi/full/10.1056/nejm197811022991808

[3] http://www.jround.co.uk/error/reading/crosskerry1.pdf

[4] https://pubmed.ncbi.nlm.nih.gov/21824940/

[5] https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2752986

[6] https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2752991