The Effectiveness of Diabetic Retinopathy Screening by Primary Care
Providers: A Systematic Review of Literature
Jena Shackelford, PA-S; David B. Day, EdS, MPAS, PA-C
Department of Physician Assistant
College of Health Professions, Wichita, Kansas
INTRODUCTION
Diabetic retinopathy (DR),
is the leading cause of
preventable blindness in the United States and much of the
industrialized world. This complication has the potential to
affect all patients with diabetes, regardless of type. Many
patients with diabetes are unaware of any vision loss and
may not receive treatment before its too late. Primary care
providers play an important role in screening for any retinal
changes in between patients’ annual visits with an
ophthalmologist.
Many health care providers feel
inadequate in their ability to accurately screen for DR using
1
the conventional ophthalmoscope, (CO). There has
recently been a new ophthalmoscope, the PanOptic, (PO),
which claims to be just as accurate. There is also an
emerging form of screening by way of telemedicine.
Telemedicine, (TM), occurs when digital images are
obtained and evaluated off site by an ophthalmologist. The
purpose of this paper was to perform a systematic review of
the literature and examine the effectiveness of screening
for DR by primary care providers by comparing the CO and
PO with TM.
METHODOLOGY
This project was conducted using Medline, FirstSearch and
Infotrac Web Databases.
Journal articles were only
selected if they were peer-reviewed and dated from 1999
to the present, with the exception of two foundational
articles. This time frame was selected to cover changes in
technology and equipment in detecting DR. The search
utilized the following keywords: DM, TM, primary health
care, PanOptic, DigiScope, and eye exam.
REFERENCES
1. Roberts E, Morgan R, King D, Clerkin L. Funduscopy: a forgotten art? Postgrad
Med J. 1999; 75: 282-284.
2. McComiskie JE, Greer RM, Gole GA. Panoptic versus
conventionalophthalmoscope. Clinical Experimental Ophthalmology. 2004; 32: 238242.
3. Zeimer R, Shazhou Z, Meeder T, Quinn K, Vitale S. A fundus camera dedicated to
the screening of diabetic retinopathy in the primary-care physician’s office. Invest
Ophth Vis Sci. 2002 May; 43(5): 1581-1587.
Figure 1: Literature Review Flow Sheet
Effectiveness of Diabetic Retinopathy Screening by Primary Care Providers
Search Terms: Diabetic Retinopathy
Telemedicine
Primary Health Care
Eye Exam
PanOptic®
DigiScope®
Total Articles
n = 29
PanOptic®
effectiveness
n=2
Conventional
ophthalmoscope
effectiveness
n=3
Retrospective
n=2
Retrospective
n=3
Telemedicine
effectiveness
n = 12
Background
n = 12
Random
Control
n=3
Case Series
n=1
Retrospective
n=8
Outcome: Telemedicine is effective
for diabetic retinopathy screening by
primary care providers.
RESULTS
Based on the inclusion material, (Figure 1) twenty-nine
articles were selected. Twelve articles consisted of
background information regarding DR, which included:
epidemiology, signs and symptoms, diagnosis, treatment
options, and inadequacies of screening by primary care
providers. Two studies found PCP’s were effective using
the CO after small workshops; one study found CO not
effective. Only two studies discussed the PO. One
found PO not as accurate as CO and gave a less clear
image. The other compared physicians’ referrals using
PO against an ophthalmologist’s referral based on
standardized criteria. Twelve articles found TM to be
effective. Forty-two percent of articles compared TM
against an ophthalmologist’s use of the gold standard.
Half of the articles chose to dilate the patients’ eyes with
mydriatic drops prior to obtaining the images. The
studies varied on the number of images obtained per
patient eye, ranging from 1-15 images.
DISCUSSION
Several professional groups require a minimum sensitivity
of 80% and specificity of 95% to be effective in DR
screening. Two-thirds of the articles discussing the CO
found it is still an effective piece of equipment in
screening for DR, after receiving a training session.
PCP’s using the PO were not very effective. It gave a
duller image and less clear image than the CO.2 It also
had an unacceptable specificity.
Although the studies regarding TM differed on the number
of digital images obtained, they were all successful in
diagnosing DR. The grading scales used in the studies
varied from whether or not DR was present to grading
levels of severity of DR. The majority of the studies
discussing TM had at least 85% of the obtained images
deemed gradable. If the image was not gradable, the
patient was always sent for referral. An example of TM,
the DigiScope, is a camera specifically designed to
function in a primary care setting.3 The DigiScope is a low
cost camera that had the highest sensitivity and specificity
of all TM studies.
CONCLUSIONS
Screening for DR in the primary care setting will not
replace the vital role that eye care specialists play. Early
detection will not only decrease the amount of
unnecessary referrals, but also allow specialists the time
needed to properly diagnose and treat DR. The CO is
still an effective piece of equipment but TM holds much
promise for protecting vision loss in diabetic patients. It
takes the guess work out of the screening process and
allows the images to be assessed by an eye
professional. It will also allow DR screening for rural
areas or patient populations that would not necessarily
have access to an evaluation by an ophthalmologist or
retinal specialist.