Implementation and Utility of Non-mydriatic Fundus Photography in the Emergency Department Open Access

Abstract

Background
Non-mydriatic ocular fundus photography provides a way to easily and rapidly examine the ocular fundus without pharmacologic pupil dilation. The candidate developed a novel line of research using fundus photography to screen for ocular fundus abnormalities in the emergency department (ED): the Fundus photography vs. Ophthalmoscopy Trial Outcomes in the Emergency Department (FOTO-ED) study.

Objectives
The goal of this dissertation was to assess factors associated with the implementation and utility of fundus photography in the ED. Our objectives were to evaluate (1) whether the reason the patient came to the ED influences (a) if the patient's ocular fundus was examined and (b) if the ED physician considered the photographs helpful, (2) whether patient characteristics influence if a photograph will be misread, and (3) whether an abnormal photograph predicts return ED visits, hospital admission, and mortality.

Methods
Using the 704 patients enrolled in the FOTO-ED study, we employed a variety of epidemiological methods including logistic regression and time-to-event regression (Cox proportional hazards model, including extensions, and recurrent event models).

Results
Men with neurological complaints (OR for men: 0.31; 95%CI:0.14-0.71 vs. women 0.68; 95%CI:0.38-1.23; p=0.08) and blacks with visual complaints were less likely to be examined (OR for blacks: 1.33; 95%CI:9.65-2.71 vs. patients of other/ unknown race: 3.02; 95%CI:1.48-6.13; p=0.08). Lower quality (OR:1.44 per quality unit; 95%CI:1.03-2.08; p=0.03) and older age (OR:1.03; 95%CI:1.00-1.05; p=0.02) were associated with a higher frequency of false positive errors and black race was associated with a higher frequency of false negative errors (OR:7.46; 95%CI:0.73-60.66; p=0.06). Abnormal photographs were associated with a significantly higher rate of hospital admission (HR:1.94; 95%CI:1.27-2.96; p=0.002) and of death from any cause (HR:4.10; 95%CI:1.51-12.42; p<0.01) controlling for age, race, sex, mean arterial pressure, and body mass index, but not with ED revisit.

Conclusions
Disparities in examination frequency were seen in this high-risk population based upon complaint, age, and race. The simplicity of obtaining non-mydriatic fundus photography and its association with a higher rate of hospital admission and death suggests it may represent a "sixth vital sign." Our findings require validation and further exploration in a larger population.

Table of Contents

1 Background & Aims . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.2.1 Examination of the ocular fundus: important, but neglected . . 2
1.2.2 Why focus on the emergency department setting? . . . . . . . . . . . 3
1.2.3 Why use non-mydriatic ocular fundus photography? . . . . . . . . . 4

1.3 General Methods of the FOTO-ED Study . . . . . . . . . . . . . . . . . . . . . . . 7
1.4 Clinical Results of Phase I of the FOTO-ED Study . . . . . . . . . . . . . . . . 9
1.5 Clinical Results of Phase II of the FOTO-ED Study . . . . . . . . . . . . . . 12
1.6 Dissertation Aims . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.7 General Dissertation Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.8 Description of Overall Cohort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.9 Risk of abnormality by patient characteristics . . . . . . . . . . . . . . . . . . . . 16

2 Patient Factors Associated with the Use and Perceived Helpfulness of
Photographs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2 Part 1: Is patient complaint associated with whether ED
physicians examined the ocular fundus? . . . . . . . . . . . . . . . . . . . . . . . . 18

2.2.1 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

2.3 Part 2: Is patient complaint associated with whether ED
physicians considered photographs helpful? . . . . . . . . . . . . . . . . . . . . . 25

2.3.1 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

3 Patient and Photographic Factors Associated with Errors . . . . . . . . . . . . . . . . . 33

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2 Part 1: Is photographic quality associated with erroneous readings? . . . 37
3.2.1 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.2.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.3 Part 2: Are patient characteristics associated with erroneous
readings and is this association mediated by photographic quality? . . . . 40

3.3.1 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.3.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.3.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

4 Fundus Abnormalities as Predictor of ED Revisit, Hospital Admission,
and Death . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

4.3.1 ED Revisits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.3.2 Hospital Admissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.3.3 All-cause mortality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

4.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

5 Conclusion & Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

5.1 Summary of dissertation-related findings . . . . . . . . . . . . . . . . . . . . . . . 61
5.2 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

5.3.1 Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5.3.2 Expansion & New Directions . . . . . . . . . . . . . . . . . . . . . . . . 66

6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
A Abbreviations Used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
B Assessment of confounding for logistic models . . . . . . . . . . . . . . . . . . . . . . . . . . 73
C Knot placement for non-linearity assessments . . . . . . . . . . . . . . . . . . . . . . . . . . 80
D Assessment of confounding, nonlinearity, and proportional hazard
assumptions for Cox models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

About this Dissertation

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Laney Graduate School
Department	Epidemiology
Degree	PhD
Submission	Dissertation
Language	English
Research Field	Health Sciences, Epidemiology
Keyword	neurology non-mydriatic fundus photography ophthalmology racial disparities mortality emergency medicine
Committee Chair / Thesis Advisor	Drews-Botsch, Carolyn D, Emory University
Committee Members	Wright, David W, Emory University McClellan, William, Emory University Klein, Mitchel, Emory University

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	Implementation and Utility of Non-mydriatic Fundus Photography in the Emergency Department ()	2018-08-28 13:44:26 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions