IHP 450 Module 4 Capital Budget Item Selection example

Selecting a Capital Budget Item: Three Ways to Deliver Diabetic Eye Exams at a Composite Community Health Center

[Student Name]

Southern New Hampshire University

IHP 450: Healthcare Management and Finance

Module Four Assignment

[Instructor Name]

[Date]

The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.

Selecting a Capital Budget Item: Three Ways to Deliver Diabetic Eye Exams at a Composite Community Health Center

The Need

Milestone One identified diabetic eye exams as the need where finances and mission meet at Tamarack Community Health Center, a composite health center with three sites. Only 38 percent of its 2,350 adults with diabetes had an eye exam in the past year. The nearest ophthalmology practice that accepts Medicaid is 40 miles away with a four-month wait, and a bonus from the biggest Medicaid managed care contract depends on passing 60 percent. The adult medicine department must choose how to close the gap, and at least two of the three realistic options involve a capital purchase.

Decision Criteria

Before comparing options, the department manager and the chief financial officer agreed on five criteria and their weights. Effect on the exam rate carries 30 percent, because it is the reason for the project. Net financial impact over five years carries 25 percent, reflecting the center's 1.2 percent margin. Staff burden carries 15 percent, since the center cannot add positions. Patient experience and access carries 15 percent, reflecting the mission. Clinical and operational risk carries 15 percent. Setting the weights before scoring keeps the decision honest, because an option cannot win by having its strengths weighted after the fact. Every option then receives a score between 1 and 5 for every criterion.

Option A: Stronger Outside Referral

The first option buys no equipment. A referral coordinator, funded as an operating expense, would schedule eye appointments, arrange transportation and track results. It requires no capital, but it depends on outside capacity the center does not control. The four-month wait and 40-mile trip remain, and patients who lost Medicaid during the unwinding now face out-of-pocket costs at the eye practice. Even well-run referral systems in safety-net settings have produced low completion; in a randomized trial among youth with diabetes, only 22 percent of those given a scripted referral and education completed an eye exam within six months (Wolf et al., 2024). A coordinator would cost about 52,000 dollars a year with no offsetting revenue to the center, because the exam is billed by the outside practice. The option also leaves the center's quality incentive in someone else's hands: the exam rate would rise only as fast as the eye practice can open appointments, and that practice has not added capacity in three years.

Option B: Conventional Cameras Read Remotely

The second option places a nonmydriatic retinal camera at each site. Medical assistants photograph the retina during the primary care visit, and the images go to a contracted eye specialist reading service, which returns a report in two to three business days. This model has a strong safety-net track record. When the Los Angeles County health system added primary care based teleretinal screening, median wait time for screening fell from 158 days to 17 days and the annual screening rate rose from 40.6 percent to 56.9 percent (Daskivich et al., 2017). Because a human reads the images, the service also finds other eye conditions; 11.6 percent of screened patients in that program were referred for problems other than diabetic retinopathy.

The costs are three cameras at a quoted 22,000 dollars each, 66,000 dollars in total, plus about 9,000 dollars for interfaces and training, and a reading fee of 30 dollars per exam. The weaknesses are operational. Results arrive after the patient has left, so staff must contact patients with abnormal findings, and the center's own experience with outside results, described in its quality reports, is that a share are never acted on.

Option C: Autonomous AI Cameras

The third option uses cameras paired with autonomous artificial intelligence that reports whether more than mild diabetic retinopathy is present before the patient leaves the room. In the trial that led to its authorization by the Food and Drug Administration, the first such system correctly identified 87.2 percent of patients who had more than mild disease and 90.7 percent of those who did not, measured against expert grading, and 96.1 percent of patients produced usable images (Abramoff et al., 2018). The strongest evidence for this option concerns completion. In the randomized trial noted above, every patient offered the exam at the point of care completed it, compared with 22 percent in the referral arm, and among patients with abnormal results, 64 percent went on to see an eye care provider compared with 22 percent in the control group (Wolf et al., 2024).

The quoted cost is 28,500 dollars per camera system, 85,500 dollars for three, plus about 17,500 dollars for installation, interfaces with the health record and staff training, for a capital total of 103,000 dollars. A license fee of 25 dollars is charged per exam. The main limitation is scope: the system screens for diabetic retinopathy and macular edema only and does not replace a comprehensive eye exam, so patients with symptoms still need referral.

Weighted Comparison

Table 1 scores the three options against the agreed criteria.

Table 1

Weighted Decision Matrix for Diabetic Eye Exam Options

Note. Scores from 1 (poor) to 5 (excellent), assigned by the department manager and the chief financial officer. Weighted totals multiply each score by its weight and sum the results.

Recommendation

Option C, autonomous AI cameras at all three sites, is the recommended capital item. It scores highest because it removes the step where follow-up most often fails: the patient receives the result, and a referral if needed, during the same visit. It uses medical assistants the center already employs, it generates billable exams and helps earn the quality incentive, and it brings a service to patients who cannot easily travel. Option B was a close second and would be the better choice if detecting other eye conditions were the priority, which is why Option C scored lower on risk. The center will manage that gap by referring any patient with visual symptoms or an ungradable image for a full exam. Milestone Two will set this purchase inside the adult medicine department's budget and build the projected budget for the coming year.

References

Abramoff, M. D., Lavin, P. T., Birch, M., Shah, N., & Folk, J. C. (2018). Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. npj Digital Medicine, 1, Article 39. https://doi.org/10.1038/s41746-018-0040-6

Daskivich, L. P., Vasquez, C., Martinez, C., Jr., Tseng, C.-H., & Mangione, C. M. (2017). Implementation and evaluation of a large-scale teleretinal diabetic retinopathy screening program in the Los Angeles County Department of Health Services. JAMA Internal Medicine, 177(5), 642-649. https://doi.org/10.1001/jamainternmed.2017.0204

Wolf, R. M., Channa, R., Liu, T. Y. A., Zehra, A., Bromberger, L., Patel, D., Ananthakrishnan, A., Brown, E. A., Prichett, L., Lehmann, H. P., & Abramoff, M. D. (2024). Autonomous artificial intelligence increases screening and follow-up for diabetic retinopathy in youth: The ACCESS randomized control trial. Nature Communications, 15, Article 421. https://doi.org/10.1038/s41467-023-44676-z