Although most clinicians used intraocular pressure (IOP) as a continuous risk factor in glaucoma treatment decisions, the historical cutoff may still be used.
Intraocular pressure (IOP) levels can have an effect on the ways that clinicians address glaucoma in their patients, as reducing IOP is the primary goal of treatment. A new study published in JAMA Ophthalmology found that the historical IOP cutoff of 22 mm Hg could influence the decision-making of clinicians when it comes to prescribing treatment for glaucoma.1
Glaucoma can sometimes be associated with the elevation of IOP, which is the primary modifiable risk factor when it comes to the progression of the condition.2 Lowering IOP is vital to treating glaucoma in those diagnosed, as it has been shown to help slow visual field progression. This reduction should be by at least 25% compared with baseline levels.1 Some studies have shown that those with higher IOP were more likely to have treatment escalation compared with those with lower IOP, but the effect of IOP levels on the decision to initiate or escalate therapy for glaucoma has been less studied. This study aimed to address how IOP levels can affect the clinician’s decision to initiate or escalate IOP-lowering therapy in patients diagnosed with glaucoma.
Clinicians were more likely to use historical cutoffs for IOP to help in their treatment decisions | Image credit: Clement C/peopleimages.com - stock.adobe.com

The Sight Outcomes Research Collaborative (SOURCE) ophthalmology data repository was used to review electronic health record data that was entered between October 2009 and January 2022. All patients with documented glaucoma who had IOP measurements between 12 and 25 mm Hg in at least 1 eye were identified in the data source. The highest IOP was used in cases where multiple IOP measurements were available in the same eye. The main outcome was initiation of treatment to lower IOP after an encounter, identified through a new prescription order, a billing code for laser trabeculoplasty within 4 weeks, or a billing code for glaucoma surgery within 8 weeks.
The study included 1,866,801 clinic encounters from 184,504 eyes in the study. The 94,232 patients had a mean (SD) patient age of 69.5 (10.8) years, and 58.1% of the patients were women. A total of 248,349 encounters led to the initiation of treatment in a patient, of which 93.6% started with medication, 3.7% had laser treatment, and 2.7% had surgery.
The researchers found that the rate of glaucoma treatment initiation increased as the IOP levels increased for those who had IOPs between 12 and 25 mm Hg. A piecewise logistic regression model found that there was a larger increase in logistic regression slope when the IOP was 22 mm Hg compared with IOPs of 19, 20, or 21 mm Hg. Treatment was more likely when IOP was higher, with a higher increase in odds of treatment at IOP 22 mm Hg.
Treatment initiation or escalation odds were between 1.03 and 1.05 for IOP values of 19, 20, and 21 mm Hg. This increased 1.11 (95% CI, 1.08-1.14) when the IOP was 22 mm Hg. This was also found when focusing on treatment initiation alone, as treatment initiation odds were between 1.03 and 1.12 in those with IOPs of 19, 20, and 21 mm Hg, which increased to 1.23 (95% CI, 1.18-1.29) at 22 mm Hg.
There were some limitations to this study. The exact reasons for starting treatment or escalating treatment for glaucoma could not be gathered due to the retrospective and deidentified nature of the data. IOP levels may not be the only factor in making treatment decisions. De-escalation of treatment could not be assessed. Patient circumstances and differing tonometry devices could have affected the IOP measurements. Clinician characteristics were also not included. Independent ophthalmology and optometry practices were not represented in the dataset.
The researchers concluded that there is likely a lingering influence of the historical IOP cutoff of 22 mm Hg when it comes to decision-making in clinicians. “It is essential for clinicians to recognize how our own cognitive limitations might lead to a reliance on decisional shortcuts that bias our decision-making,” the authors wrote.
References