September 2016
Volume 57, Issue 12
Open Access
ARVO Annual Meeting Abstract  |   September 2016
Identification of traits of interest in genetic studies of retinopathy of prematurity
Author Affiliations & Notes
  • Ryan Swan
    Department of Medical Informatics and Clinical Epidemiology, Oregon Health & Science University, Portland, Oregon, United States
  • Susan Ostmo
    Casey Eye Institute, Oregon Health & Science University, Portland, Oregon, United States
  • Kemal Sonmez
    Department of Medical Informatics and Clinical Epidemiology, Oregon Health & Science University, Portland, Oregon, United States
  • Kimberly A Drenser
    William Beaumont Hospital, Royal Oak, Michigan, United States
  • Thomas C Lee
    Children's Hospital Los Angeles, Los Angeles, California, United States
  • Audina M Berrocal
    Bascom Palmer Eye Institute, Miami, Florida, United States
  • Charles Simmons
    Cedars-Sinai Medical Center, Los Angeles, California, United States
  • Jerome I Rotter
    Pediatrics, Harbor-UCLA Medical Center, Los Angeles, California, United States
  • Yii-Der Ida Chen
    Pediatrics, Harbor-UCLA Medical Center, Los Angeles, California, United States
  • Robison Vernon Paul Chan
    University of Illinois College of Medicine at Chicago, Chicago, Illinois, United States
  • Michael F Chiang
    Casey Eye Institute, Oregon Health & Science University, Portland, Oregon, United States
  • Footnotes
    Commercial Relationships   Ryan Swan, None; Susan Ostmo, None; Kemal Sonmez, None; Kimberly Drenser, None; Thomas Lee, None; Audina Berrocal, None; Charles Simmons, None; Jerome Rotter, None; Yii-Der Chen, None; Robison Chan, None; Michael Chiang, Clarity Medical Systems (S)
  • Footnotes
    Support  Supported by NIH EY19474, the OSLER TL1 Grant (NIH TL1TR000129-10), the Casey NIH Core grant (P30 EY010572) and an unrestricted grant from Research to Prevent Blindness.
Investigative Ophthalmology & Visual Science September 2016, Vol.57, 6303. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Ryan Swan, Susan Ostmo, Kemal Sonmez, Kimberly A Drenser, Thomas C Lee, Audina M Berrocal, Charles Simmons, Jerome I Rotter, Yii-Der Ida Chen, Robison Vernon Paul Chan, Michael F Chiang; Identification of traits of interest in genetic studies of retinopathy of prematurity. Invest. Ophthalmol. Vis. Sci. 2016;57(12):6303.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Selection of individuals with traits of interest that fall to opposite extremes of a spectrum are important in order to maximize the power of small studies. In cases of rare diseases where multiple phenotypes contribute to the disease definition and where there may be few individuals at these extremes, it is useful to take into account multiple outlier categories. However, incorporation of additional categories can introduce heterogeneity into the study population. We consider the case of retinopathy of prematurity (ROP) and selection of traits from a set including birth weight, gestational age, disease burden, and twin status.

Methods : A multicenter database was generated prospectively from infants screened for ROP at one of 6 major ROP centers in the United States. The database was reviewed to identify infants treated for ROP. Database information included level of ROP progression, treatment required, physical factors (birth weight, postmenstrual age), demographic data, and presence of additional disease factors (sepsis, necrotizing enterocollitis, chronic lung disease, death during hospital stay). Percentile weight for gestational age was calculated using gestational age and birth weight. Expert clinicians were consulted to determine weights reflecting the importance of each trait in the determination of interest as a genetic study participant.

Results : 406 Caucasian infants representing 1603 exams were identified from the full database. A weighting matrix was applied to the data and a ranked list of infants according to genetic interest was generated from the weighted data. Permutations of the ranked data were visualized and reviewed by a panel of expert clinicians and genetic specialists to determine a combination of traits leading to ideal separation in sample and collinearity of existing traits. The ideal factors indicating outlier status were determined to be birth weight, level of treatment, and degree of ROP.

Conclusions : Complex models provide context about the sample population, but incorporation of competing additional traits limits the ability to sample population extremes. Therefore in studies of limited sample size where maximum power is required to detect an effect, it is often necessary to limit the number of factors contributing to phenotypic interest. In retinopathy of prematurity, the highest priority information for outlier status is captured using the three factors listed above.

This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.

×
×

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.

×