June 2015
Volume 56, Issue 7
ARVO Annual Meeting Abstract  |   June 2015
Diagnostic Misclassification: An Extreme Example in "Pink Eye" Clinical Trials
Author Affiliations & Notes
  • Mae O Gordon
    Ophthal & Vis Sciences, Washington Univ Sch of Med, St Louis, MO
  • Julia Beiser Huecker
    Ophthal & Vis Sciences, Washington Univ Sch of Med, St Louis, MO
  • Leonard Haertter
    Ophthal & Vis Sciences, Washington Univ Sch of Med, St Louis, MO
  • Footnotes
    Commercial Relationships Mae Gordon, None; Julia Huecker, None; Leonard Haertter, None
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science June 2015, Vol.56, 6197. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Mae O Gordon, Julia Beiser Huecker, Leonard Haertter, Reduction in Adenoviral Patient Infected Days (RAPID) Study Group; Diagnostic Misclassification: An Extreme Example in "Pink Eye" Clinical Trials. Invest. Ophthalmol. Vis. Sci. 2015;56(7 ):6197.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

To use therapeutic trials of “pink eye”, adenoviral conjunctivitis (Ad-Cs), to demonstrate how inclusion of patients in therapeutic trials who do not have the Ad-Cs (target disease) dilutes the observed treatment success.


We reviewed therapeutic clinical trials of adenoviral conjunctivitis to identify methods of diagnosis. Most of these studies used clinical signs and symptoms to diagnose Ad-Cs which was seldom confirmed by laboratory tests. In about 2006, AdenoPlusTM “point of care” immunoassay received FDA approval for the diagnosis of Ad-Cs. We estimated the proportion of patients with clinical signs and symptoms of Ad-Cs who test positive for Ad-Cs from published sensitivity/specificity studies that compared AdenoPlusTM to qPCR .1,2 We then calculated the observed success rates for hypothetical drugs with true success rates of 50%, 60%, 70%, 80% given patient samples in which 100%, 70%, 50%, 30% patients actually had Ad-Cs. We assumed that patients who did not have Ad-Cs received no benefit from the drug.


Only about 30% of the patients who present with clinical signs and symptoms of “pink eye” test positive for Ad-Cs by AdenoPlus.1,2 Thus, we infer that about 70% of the patients enrolled in therapeutic trials of Ad-Cs diagnosed by clinical signs and symptoms did not have Ad-Cs. Therefore, it is no surprise that none of the drugs in these trials were “effective”. Table 1 illustrates how the inclusion of patients in a study who do not have the target disease dilutes the observed therapeutic success rate.


Inclusion of patients who do not have the target disease substantially reduces the observed success rate of drugs. These results have important implications for therapeutic trials for eye conditions having unclear diagnostic criteria such as dry eye.<br /> <br /> 1. Sambursky R, Tauber S, Schirra F, Kozich K, Davidson R, Cohen EJ. The RPS adeno detector for diagnosing adenoviral conjunctivitis. Ophthalmology 2006. 113(10):1758-1764.<br /> <br /> 2. Sambursky R, Trattler W, Tauber S, Starr C, Friedberg M, Boland T, McDonald M, DellaVecchia M, Luchs J. Sensitivity and Specificity of the AdenoPlus Test for Diagnosing Adenoviral Conjunctivitis. JAMA Ophthalmol 2013. 131(1):17-22  


This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.