Improving Patient Access to Diabetic Retinopathy Screening Through Telemedicine
Bibliographic record
Abstract
Background The LifeBridge Health (LBH) Accountable Care Organization (ACO) serves approximately 20,000 Medicare beneficiaries, many of whom have type I or type II diabetes. Diabetic retinopathy (DR) screening is extremely important in helping to preserve patient’s eyesight and overall functional status. However, like many other organizations, LBH has struggled with low compliance rates for DR screening. As result, LBH searched for a solution to improve DR screening care and improve ACO quality and financial performance. Objective LifeBridge sought a telemedicine diagnostic solution that was easy for our physicians and clinic teams to use that would enable improved management of patients with diabetes. A pilot was initiated at three large primary care practice locations in the last quarter of 2017. Two of the locations received table top cameras, while the other location received a more mobile, hand held unit. Working with a dedicated LBH IRIS team, the practices created and implemented workflows, documented processes, and instilled best practices. Methods We used a pre-post test design to measure whether implementation of this tool enabled providers to better meet the diabetic retinopathy screening measure. We included the final months of 2017 in the preperiod to account for any operational changes required to implement the new workflow. Manual chart abstraction of patients seen in the previous 4/6 weeks who were eligible to determine the proportion of patients who met the measure. This was done quarterly in every primary care practice throughout the organization. One of the three practice sites was changed halfway through 2018 and switched to another; however, both practices were included in the analysis. We also compared the number of diabetes patients in the populations of each of the four practices. A two sample z test with a P value of .05 was used to test for statistical significance. Results As of April 2019, 810 patients were screened for diabetic retinopathy. Of these, 33.1% (282 patients) were diagnosed with pathology. Approximately 15.6% (n=133) were diagnosed with DR. We also identified 87 patients who are considered “IRIS saves” patients who had pathology identified that was serious enough to put them at imminent risk of losing their sight. For all patients requiring follow up, direct referrals were made to our in-network ophthalmologists at Krieger Eye Institute for treatment that these patients would not have otherwise received. Statistical comparison of DR screening performance of practices pre and post implementation showed mean screening rates of 38.5% and 47.2%, respectively, with P=.01. Conclusions IRIS screenings allowed our primary care providers to provide more comprehensive care to patients with diabetes, eliminating the need for additional office visits. Having IRIS available in the practice was able to demonstrably improve performance in the diabetic retinopathy screening measure. As a result, primary care providerss with IRIS helped facilitate access to care, thus making it easier for patients make better choices related to their health outcomes. We hope to further use the data to study HbA1c control, medication adherence, and cost/utilization in those diagnosed with retinopathy compared to those with a negative screening.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".