Abstract
<title>Abstract</title> <p>Background. Timely follow-up is essential to vision-saving care, but social and structural barriers keep patients from completing recommended visits. Full-length SDOH screeners are impractical in no-cost ophthalmology clinics, where navigation capacity is the limiting resource. We asked whether a minimal item set could flag such patients for scarce navigation resources. Methods. We analyzed intake data from 630 adult patients from a large no-cost clinic. The outcome was self-reported difficulty attending appointments (prevalence 37.5%). Using 5-fold cross-validation, we ranked 13 SDOH items, built screeners, and assessed outreach yield (PPV, sensitivity) at the highest-risk 5%, 10%, and 20%. We related the score to total SDOH burden, excluding the two screener items, and to outcomes independent of the barrier wording. Results. Discrimination plateaued at two items: transportation and medicine/healthcare need (AUC 0.75, 97% of the full 13-item model's 0.77), matching the full model at every capacity level (PPV 77–97% across the top 5–20%). Reported barriers rose steadily with SDOH burden after excluding the two screener items. The score showed little association with non-overlapping outcomes: self-reported delayed-care markers (AUC 0.54–0.59), no-show rate (rho = − 0.04), and index attendance (AUC 0.53). Conclusions. Two intake questions efficiently identify patients reporting access barriers and flag a high-need subgroup for navigation triage. They add about 30 seconds at registration and are immediately deployable. Because the screener tracks reported need rather than observed attendance, its value is triage, not behavioral prediction; whether prioritizing by reported barriers improves follow-up is a question for prospective study.</p>