Failure to diagnose retinal detachments

ANNE M MENKE, RN, PhD, OMIC Patient Safety Manager

In a recent study of OMIC diagnostic error (DE) claims, we learned that those involving a retinal condition were the most common, accounting for 38% of the claims closed between 2008 and 2014. By far, the most frequently missed diagnosis in our entire study was retinal detachment (RD). The RD claims represent 79% of the DE retina claims and 48% of the DE retina payments. Remarkably, while there were only six claims for failure to diagnose retinopathy of prematurity, those claims comprised 47% of the DE retina payments.

This issue of the Digest will explore RD DE claims in more detail, first by
presenting data and analysis and then by offering recommendations. The management of RDs once they were diagnosed did not affect the outcome of these claims, so our analysis will focus solely on the diagnostic process.

Forty-two claimants in the study alleged a delay in the diagnosis of RD during the noted 7-year period. They ranged in age from 20 to 79 years of age, with the majority in their 50s and 60s. The delay period ranged from one week (or less) to 32 weeks (Figure 1).

There was no clear relationship between the alleged delay period and indemnity payments. When the diagnosis closely followed the ophthalmologist’s examination, however, claimants had an easier time arguing that the RD was present but missed. Claimants had more difficulty proving negligence when there was a long period between the exam and the eventual diagnosis, although other factors sometimes led to a decision to settle the claim. Visual acuity deteriorated significantly during the alleged delay period, as shown in Figure 2.

The number of claimants with good vision (≥ 20/40) declined from 25 at the initial encounter to 9 at the final encounter, while those with poor vision (<20/200 to LP) doubled; two claimants who had light perception at the initial encounter ended up with no light perception.

Analysis of OMIC defendants and indemnity payments

The number and type of OMIC-insured defendants in the RD DE claims varied. Twenty-five claims were filed against a sole defendant (22 against a single ophthalmologist, 3 against a group); the other 17 claimants named multiple defendants, for a total of 63 claims. Although 40% of the claims involved multiple defendants, OMIC never paid on behalf of more than one defendant. Multiple defendants did increase the costs of defending the care, however.

DE claims are costly. When compared to all OMIC claims during the study period, claims alleging DE of any type resulted in more paid claims, a higher median and mean payment, and the highest payment. DE claims alleging failure to diagnose RD had a slightly lower percentage of paid claims than overall DE claims, as well as lower median, mean, and high payments. Nonetheless, RD claims still had higher median and mean payments than other OMIC claims as a whole (see Figure 3). Comprehensive ophthalmologists (COs) provided the care in 70% of the RD claims, retina specialists (RSs) in the remaining 30%. OMIC made indemnity payments on behalf of a higher percentage of RSs than COs, and the payments tended to be for larger amounts. The small number of claims and payments makes it hard to draw firm conclusions, but experts may have held RSs to a higher standard than COs. The lowest DE payment of $1650 was made on behalf of a CO to reimburse the claimant in an RD claim for out-of-pocket expenses. The highest overall DE claim was for an oncology condition. The highest RD DE payment, for both COs and RDs, was made to a surgeon whose loss of binocular vision affected the ability to perform surgery.

Initial diagnosis and risk for RD

Two of the forty-two RD DE claimants in the study received only telephone care. Ophthalmologists diagnosed 21 of the remaining 40 claimants with a retinal condition other than RD. Retinal conditions diagnosed at the initial encounter in descending order of frequency were posterior vitreous detachment (PVD); vitreous changes such as floaters, degeneration, or syneresis; retinal tear; traumatic injury other than PVD or tear; endophthalmitis following cataract surgery; and non-proliferative diabetic retinopathy with lattice degeneration. The other 19 initial diagnoses, in descending order of frequency, were refractive errors including cataract; neurological conditions; postoperative changes following cataract surgery; glaucoma; and medical conditions (GCA and collagen vascular disease).

The AAO’s Preferred Practice Pattern “Posterior Vitreous Detachment, Retinal Breaks, and Lattice Degeneration” states that risk factors for developing an RD are retinal breaks, myopia, lattice degeneration, cataract surgery, trauma, and a history of RD in the fellow eye or in the patient’s family. As Figure 4 shows, 34 of the 39 claimants (85%) who were given a diagnosis other than RD had one or more of these risk factors at the time of the initial visit. Did the defendant ophthalmologists lack knowledge of the relationship between these conditions and RD? Did they neglect to obtain a history and perform an exam likely to uncover an RD? Did they fail to educate the claimants with risk factors about the condition and the symptoms that should be reported to the ophthalmologist? Did claimants not recognize the symptoms or neglect to notify the ophthalmologist? Experts for both the claimants and defendants sought answers to these questions when they conducted their reviews.

Standard of care (SOC) and causation analysis

As expected, claimant experts concluded that at least some aspect of the care was negligent in the RD DE claims in this study. What did defense experts decide? They reviewed 54 of the 63 claims (9 closed without payment before a review was completed). Additional reviews were usually obtained if the initial one raised concerns. If the reviewers reached different conclusions, we considered this a “mixed” review. According to these experts, the defendant met the standard of care 56% of the time, and failed to do so in 44% of cases (Figure 5). One would expect comprehensive ophthalmologists to have more difficulty diagnosing RDs than retina specialists, and they did: comprehensive ophthalmologists had negative reviews 50% of the time, while defense experts criticized the care of retinal specialists in 29% of the cases.

This condition clearly presents diagnostic challenges to many ophthalmologists. Why is that? To help answer that question, we evaluated claims with negative reviews for the role ophthalmologists, systems (scheduling, telephone care, etc.), and patients played in the delay. All claims had primary causes, while only some had secondary causes. As Figure 6 shows, physician factors account for twice as many delays as systems issues, while patients had the least impact. Specific examples of areas of concern in each category are shown.

Moving from examination to education to diagnosis

As Figure 4 showed, 85% of the patients had risk factors for RD at the time of the initial diagnosis. What obstacles stand in the way of an earlier diagnosis of an RD? The short answer is teamwork. First, the ophthalmologist needs to obtain an adequate history and recognize the risk factors. The Hotline article in this issue provides more detail on what defense experts found lacking in the ophthalmologist’s care. Next, the physician and staff members need to educate the patient about the symptoms and the process for reporting them. Staff members who answer phones need to have written protocols on how to schedule appointments, and have knowledge of symptoms that require urgent or emergent appointments.

The goal of patient education about RDs is to elicit the patient’s agreement to monitor for and report worrisome symptoms. Patient engagement will not happen if the patient does not understand the information, is too distracted by fear of vision loss, or hears only the reassuring news that no RD was found that day. Patients may be confused about the uncertainty and not be able to understand that they may never develop an RD, have one within days, or not for many months. The time gap between the education and the appearance of symptoms can be quite long. The patient can forget the information or lose the written instructions. If the gap is short and the symptoms are the same as the first visit, the patient may decide that there is no need to call, or that she does not have the funds to return so quickly for another visit. Staff who seem brusque or unwelcoming, or lack the knowledge to understand the significance of the reported symptoms, could lead patients to stop advocating for themselves and put off a return visit.

Prompt diagnosis of an RD requires an informed, engaged, and welcoming team. Use an RD risk factor and symptom checklist to train staff on how to help identify patients at risk. Include staff in the development of written protocols for telephone care. Ask them to let you know if they feel the patient doesn’t understand or hasn’t committed to monitoring. Give patients specific language to use to signal a possible RD when they call, such as “Dr. Williams said to bring me in right away if I had this symptom.” In addition, sound like you mean it when you invite patients to call you.