 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
We thank Dr Stefánsson for his interest in our clinical report and congratulate him and his colleagues on the development of a successful diabetic retinopathy–screening program in Iceland. We appreciate the opportunity to respond to his comments.
The development of successful screening programs requires the analysis of both scientific data and the practical (real-world) implications of screening recommendations. With regard to scientific data, the articles cited by Dr Stefánsson support the recommendation for biannual eye screening for patients without retinopathy. If these were the only studies to support his recommendation, one would apply caution in extrapolating the data. The study described the findings in a small population of children, and the follow-up conditions would be considered close to ideal, with a small and geographically captive population.1 The authors report a laudable reduction in the prevalence of legal blindness in this setting.
Our clinical report did not address the specific issue of biannual examinations, but there are additional data that support this approach. In a recently published study from Australia, a group of >1000 children and adolescents were longitudinally screened by using fundus photography.2 The authors concluded that adolescents who had no retinopathy at baseline examination, who were under reasonable metabolic control, and who had duration of diabetes <10 years could safely be screened biannually.
One of the goals of screening recommendations is that they be cost-effective. Identification and elimination of unnecessary examinations certainly advances this goal. Another example of this concept is the recommendation that screening for retinopathy is not necessary during the first decade of life. During our literature review, we found no study that specifically evaluated children who were diagnosed with type 1 diabetes mellitus (T1DM) at a very early age (<2 years). We questioned whether screening during the first decade might be indicated in their group, because the duration of disease would be longer in these patients. We examined a database of >2000 patients with T1DM and identified 51 patients who were diagnosed with T1DM before 2 years of age and who were monitored for retinopathy for at least 5 years.3 None of these children developed retinopathy, which supports the contention that screening during the first decade of life is not necessary even when the onset of T1DM occurs at a very young age.
The real-world implementation of screening recommendations presents a different set of problems that are less amenable to scientific study. The unfortunate fact is that compliance with screening programs is often dishearteningly poor despite the best efforts of those involved. There are legitimate concerns regarding whether a recommendation for biannual examinations might impact patient behavior. For instance, increasing the recommended screening interval could make the examinations seem less important to patients, and they therefore might delay their examinations even further. It is difficult to accurately assess effects such as these and certainly difficult to apply a cost/benefit analysis to such considerations.
We congratulate Stefánsson and his colleagues on decreasing the visual morbidity of their patients with T1DM. Good data such as theirs will allow us to further refine our screening recommendations, with the goal of improving the health of our patients in a cost-effective manner. As for translating these recommendations into the real world, we reiterate our statement that research on methods to improve implementation of screening guidelines could potentially provide great benefit to our patients.
REFERENCES
Related articles in Pediatrics:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2006 American Academy of Pediatrics. All rights reserved. |
||
|
||
