Ocular Hypertension Treatment Study (OHTS) — 5-year POAG risk (point system)

Baseline data

Enter right (OD) and left (OS) values. The tool averages each pair to match the published mean-of-eyes approach, then applies the OHTS–EGPS point system (Gordon et al., Ophthalmology 2007, Table 6). Use untreated IOP and Humphrey PSD (24-2 or 30-2 full threshold or SITA standard), or Octopus loss variance converted to Humphrey PSD equivalent per trial protocols.

Age (years)

IOP OD (mm Hg)

IOP OS (mm Hg)

CCT OD (μm)

CCT OS (μm)

Vertical C/D OD (contour)

Vertical C/D OS (contour)

PSD OD (dB)

PSD OS (dB)

Disclaimer: This implements the published simplified point system for education and shared decision support. It is not a medical device, does not store patient data, and does not replace clinical judgment or the official OHTS materials.

Ocular Hypertension Treatment Study (OHTS) — Point system explained

Background

The pooled OHTS observation cohort and EGPS placebo group yielded a multivariate model for 5-year risk of primary open-angle glaucoma (POAG) in ocular hypertension using age, mean intraocular pressure (IOP), mean central corneal thickness (CCT), mean vertical cup-to-disc (C/D) ratio by contour, and mean Humphrey pattern standard deviation (PSD). A simplified point system approximates the Cox model (discrimination similar; see primary publication).

Eye-specific variables should be averaged between right and left eyes after multiple measurements per eye, as in the trials (Goldmann IOP, ultrasonic CCT, vertical C/D by contour, two PSD tests per eye on Humphrey 24-2/30-2 or equivalent Octopus metrics converted to Humphrey PSD).

Points by predictor (0–4 each)

Predictor	0	1	2	3	4
Age (years)	<45	45–<55	55–<65	65–<75	≥75
Mean IOP (mm Hg)	<22	22–<24	24–<26	26–<28	≥28
Mean CCT (μm)	≥600	576–600	551–575	526–550	≤525
Mean vertical C/D (contour)	<0.3	0.3–<0.4	0.4–<0.5	0.5–<0.6	≥0.6
Mean PSD (dB)	<1.8	1.8–<2.0	2.0–<2.4	2.4–<2.8	≥2.8

Total points → estimated 5-year POAG risk

Sum of points	Estimated 5-year risk
0–6	≤4%
7–8	10%
9–10	15%
11–12	20%
>12	≥33%

Reference

Gordon MO, Torri V, Miglior S, et al. A validated prediction model for the development of primary open-angle glaucoma in individuals with ocular hypertension. Ophthalmology. 2007;114(1):10–19. Official materials and continuous-method details: ohts.wustl.edu/risk.

Limitations

• Developed in trial-defined ocular hypertensive cohorts; calibration differs between populations
• Point system is an approximation of the multivariate Cox model
• Excluded exfoliation and pigment dispersion in primary analyses; performance may differ if present
• Does not replace clinical judgment, examination frequency decisions, or treatment choices

What the Ocular Hypertension Treatment Study addressed

The Ocular Hypertension Treatment Study (OHTS) was a landmark randomized trial that examined whether lowering intraocular pressure (IOP) in people with elevated IOP but no detectable glaucomatous damage at baseline could delay or prevent the onset of primary open-angle glaucoma (POAG). Alongside treatment effects, OHTS generated high-quality prospective data on which baseline characteristics best predicted who would later develop POAG.

Many individuals with ocular hypertension never develop glaucoma, while others convert over a few years of follow-up. That heterogeneity makes risk stratification useful: it can inform how intensively to monitor visual fields and optic discs, how to discuss the possible benefits and burdens of IOP-lowering therapy, and how to personalize follow-up intervals. The calculator on this site implements the published simplified point system derived from the pooled OHTS observation cohort and the placebo arm of the European Glaucoma Prevention Study (EGPS), as described in the peer-reviewed prediction model paper and mirrored in materials from the OHTS coordinating center.

From trial cohorts to a five-year POAG risk estimate

Investigators first built multivariate Cox proportional hazards models for time to incident POAG using untreated or control-group participants, so that predictions would reflect natural history under observation rather than treatment effects. The same five structural risk factors emerged independently in OHTS and EGPS: older age, higher IOP, thinner central corneal thickness (CCT), larger vertical cup-to-disc (C/D) ratio by contour, and greater visual field pattern standard deviation (PSD) on standard automated perimetry.

Because clinicians and patients often prefer transparent rules over regression output alone, the authors also published a points-based approximation to the multivariate model. Each predictor is divided into ordered categories; each category maps to an integer score from 0 to 4. Summing the five components yields a total between 0 and 20, which is then mapped to a small set of estimated five-year cumulative risks of developing POAG. Discrimination of the point system is slightly lower than the full Cox model, but it remains a practical teaching and bedside tool when complete continuous data are available.

Why each predictor matters clinically

Age

Glaucoma incidence rises with age in population data and in ocular hypertensive cohorts. In the point system, risk increases stepwise across decade-like bands. Age is non-modifiable but helps contextualize how long a patient might live with elevated IOP and how aggressively to pursue prevention.

Intraocular pressure (IOP)

IOP remains the only major modifiable factor in open-angle disease. The model uses mean Goldmann applanation IOP averaged between eyes after multiple measurements, reflecting the trial protocol rather than a single clinic reading. Higher mean IOP shifts the patient into higher point bands. When IOP has already been lowered by medication, the numerical estimate may no longer represent the same untreated risk stratum; separate analyses have explored using treated IOP in similar frameworks, but users should remain aware of what measurement context their entered pressures represent.

Central corneal thickness (CCT)

Thinner corneas associate with higher risk of conversion in OHTS and related studies. Mechanistically, thinner CCT may correlate with true IOP underestimation by applanation, with structural susceptibility of the optic nerve head, or with both. The calculator expects ultrasonic pachymetry values averaged between eyes, again following multicenter standardization in the trials.

Vertical cup-to-disc ratio by contour

The model uses vertical C/D assessed by contour, not by color alone, and averages right and left eyes. Larger ratios may represent constitutional anatomy or early acquired rim loss; at the first visit the clinician often cannot distinguish those possibilities. The variable therefore functions as a cross-sectional structural marker that nonetheless improves prediction in the trial populations.

Pattern standard deviation (PSD)

PSD summarizes localized visual field variability on automated perimetry. Slightly elevated PSD can occur in technically unreliable tests, early focal loss, or nonspecific variability. The OHTS/EGPS protocols specified particular test strategies (for example Humphrey 24-2 or 30-2 full-threshold or SITA standard); Octopus metrics such as corrected loss variance can be converted to a Humphrey PSD–equivalent when using official tools. The point system uses the mean PSD across two tests per eye, then averages eyes—reducing noise from a single field.

How the point system maps to a risk band

After computing the mean of paired-eye measurements for IOP, CCT, vertical C/D, and PSD, each mean is located in its published category table. Points from age, IOP, CCT, C/D, and PSD are added. The total score is then translated to a discrete estimate: low totals correspond to roughly ≤4% five-year risk, intermediate totals to 10% or 15%, higher totals to 20%, and the highest totals to ≥33% in the original publication’s simplified mapping.

These percentages describe group-level cumulative incidence over five years in closely monitored trial-like cohorts. They are not a guarantee for an individual patient; they are an aid to structured discussion. The multivariate Cox-based continuous calculation can yield a slightly different numeric estimate than the point total for the same patient; both approaches were presented in the primary work as complementary.

Practical use of this calculator

Enter right (OD) and left (OS) values separately; the tool averages them to mimic the trial’s mean-of-eyes convention.
Use untreated baseline IOP when the goal is to approximate observation-group natural history; if you enter pressures on therapy, interpret the output as conditional on that treatment state.
Ensure perimetry and imaging meet the same test types referenced in OHTS/EGPS documentation where possible.
Discuss results together with life expectancy, side-effect tolerance, cost, adherence, and patient goals; low modeled risk does not automatically preclude therapy, and high modeled risk does not mandate it without shared decision-making.

Populations and measurements for which the model fits best

The model was developed in adults with study-defined ocular hypertension, standardized Goldmann tonometry, ultrasonic CCT, and specific Humphrey or Octopus field strategies. Performance may differ in eyes with secondary open-angle mechanisms (for example, exfoliation or pigment dispersion, which were excluded or sparse in the primary analyses), after prior incisional surgery, with corneal pathology invalidating applanation, or when fields are unreliable. Ages and measurement ranges emphasized in official calculator documentation (for example, typical adult age bounds and IOP and CCT intervals used in the trials) flag situations where extrapolation should be interpreted cautiously.

Finally, predictive tools can become miscalibrated when applied to populations with very different baseline incidence of POAG or when care pathways diverge from semiannual trial follow-up. The estimate should therefore be read as structured epidemiologic context, integrated with longitudinal trends in IOP, imaging, field progression, and the clinician’s overall judgment.

CalcMD.

Ocular Hypertension Treatment Study (OHTS) Calculator