Evaluating a virtual reality visual fields analyzer in an urban, underserved glaucoma & glaucoma suspect patient populations to identify disparities

Abstract Purpose To evaluate the utility of wearable visual field perimetry in an urban, underserved patient population and identify disparities in its utility as a screening tool. Methods 175 eyes from 105 participants (46 non-glaucomatous eyes from 34 participants and 113 glaucomatous eyes from 74 participants; 16 eyes failed inclusion criteria) presenting at University Hospital in Newark, New Jersey for glaucoma evaluation underwent testing by both the Humphrey Visual Field Analyzer™ (HFA) and PalmScan VF2000 G2™. Glaucoma severity was classified as per the Hoddap criteria. Mean deviation (MD), pattern standard deviation (PSD), visual field index (VFI), mean sensitivity (MS), & area under the receiver operating curve (AUC) on analysis adjusted for inter-eye correlation. Results The VF2000 and HFA significantly differed in VFI as the VF2000 consistently underestimated VFI ( p  = 0.003) but did not significantly differ in MD ( p  = 0.664) or PSD ( p  = 0.584). The VF2000 had significantly fewer false positives ( p  < 0.001) and fixation losses ( p  = 0.001) but was a significantly longer exam ( p  = 0.018). On a multivariate logistic regression model adjusting for both inter-eye correlation and demographic variables, the VF2000 had an AUC of 0.7007, indicating fair agreement when identifying severe glaucoma. Language, age, and sex did not independently impact odds of agreement between the two devices; however, differences based on the interaction of age and language were observed. Conclusion Our analysis of the Humphrey Visual Field against the virtual reality PalmScan VF2000 G2™ in an urban, diverse population found subtle disparities in predictive staging of glaucoma. Future studies may need to account for these disparities by evaluating the combinations of demographic interactions rather than evaluating them as independent, unrelated factors. Key messages What is known Portable perimetry and virtual reality headsets have been used with moderate efficacy in screenings for glaucoma, but gaps exist in the quality of results as compared to the Humphrey Visual Fields Analyzer. What is new The PalmScan VF2000 G2, a portable perimetry headset, may be suitable as a screening device, but it is not advanced enough to differentiate glaucoma stage as effectively as HFA analysis. Disparities along social determinants of health do exist in VF2000 detection of glaucoma, though these manifestations may be subtle and tied to the interaction of many complex factors. Future studies may benefit from examining the interaction between demographic factors as variables predictive of outcome.