RESEARCH ARTICLE
Characteristics of Design and Statistical Analysis of NEI-funded Ophthalmic Clinical Trials
Wilson Zhang1, #, Richard Zhang1, #, Gui-shuang Ying2, *
Article Information
Identifiers and Pagination:
Year: 2023Volume: 17
E-location ID: e18743641266764
Publisher ID: e18743641266764
DOI: 10.2174/0118743641266764231019092255
Article History:
Received Date: 11/06/2023Revision Received Date: 31/08/2023
Acceptance Date: 13/09/2023
Electronic publication date: 23/10/2023
Collection year: 2023
open-access license: This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0), a copy of which is available at: https://creativecommons.org/licenses/by/4.0/legalcode. This license permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Abstract
Objective:
To describe the characteristics of trial design and statistical analysis of the National Eye Institute (NEI)-funded randomized clinical trials (RCTs) conducted after the year 2000.
Design:
Review of 42 NEI-funded ophthalmic RCTs.
Methods:
Eligible trials were identified from ClinicalTrials.Gov and their primary result papers were identified from PubMed. Data on the design characteristics (primary outcome, number of arms, sample size, statistical power, inclusion of one eye or two eyes) and statistical analysis (statistical method for adjustment of inter-eye correlation, correction for multiple comparisons) as reported in the primary result paper were collected independently by two authors, and the differences were adjudicated by the senior author. Descriptive analyses were performed to summarize the characteristics of trial design and statistical analysis.
Main Outcome Measures:
Characteristics of trial design and statistical analysis.
Results:
Forty-two NEI-funded ophthalmic trials conducted after 2000 were included. The majority of trials were for evaluating the efficacy of drugs (57%), medical devices (21%), or procedures (14%) for the treatment of retinal diseases (45%) or pediatric eye diseases (45%). All trials were designed with at least 80% statistical power for comparing continuous (64%), binary (24%), or time-to-event (12%) primary outcome measures. In 11 (26%) trials enrolling both eyes of a participant, two eyes were in the same treatment group in 6 (55%) trials, and two-eye data were properly analyzed with adjustment for the inter-eye correlation when needed for all these trials. However, none of these trial publications explicitly stated that the inter-eye correlation was considered in the sample size and power calculation. In 13 trials with more than two arms, 12 (92%) trials adjusted for multiplicity using Bonferroni correction (42%), Hochberg procedure (42%) or Turkey’s method (17%).
Conclusion:
While the availability of two eyes of a participant may complicate the ophthalmic trial design and statistical analysis, NEI-funded trials followed good practice in the trial design and statistical analysis, with enrollment of two eyes of a participant when appropriate, and adjustment of the inter-eye correlation in the statistical analysis. The sample size and power calculation can be improved by considering the inter-eye correlation and clearly reporting such information for future ophthalmic trials is important.
