More LEP patients remained on steroid drops 4 weeks after cataract surgery (14.6% vs 10.1%)

Description #

More LEP patients were still using steroid eyedrops 4 weeks after cataract surgery than EP patients (14.6% vs 10.1%, P < .0002; Table 3 reports P = .0002). The authors consider drop non-compliance, misunderstanding of the taper regimen, greater surgical complexity/CDE, and possible racial differences in inflammation as explanations.

"More LEP patients were on steroids 4 weeks postoperatively when compared with EP patients (14.6% vs 10.1%, P < .0002)." (Zafar, 2023, p. 595)

"Still on steroid drop at 4 wk 127 (14.6) 1289 (10.1) .0002" (Zafar, 2023, p. 598, Table 3)

Methods Context #

What? #ⓘ

The observable: usage of steroid eyedrops at 4 weeks postoperatively, a categorical outcome.

"usage of steroid eyedrops at 4 weeks postoperatively" (Zafar, 2023, p. 596)

How? #ⓘ

Retrospective comparison by LEP status using logistic regression with generalized estimating equations to account for correlation between fellow eyes.

"Comparisons by LEP status were analyzed using logistic regression with generalized estimating equations to account for correlation between eyes from the same subject." (Zafar, 2023, p. 596)

Who? #ⓘ

13,590 phacoemulsification eyes at a single academic eye center (2014–2020); 868 (6.4%) LEP vs 12,722 EP.

"A total of 13 590 eyes were included in the analytic database. Of these, 868 (6.4%) were identified from LEP patients." (Zafar, 2023, p. 596)

Caveats #

• LEP identified only by EMR interpreter-need flag, risking misclassification (Zafar cataract) LEP status was operationalized solely from an electronic-medical-record designation of "needing or preferring an interpreter," recorded by call-center or front-desk staff when the appointment was booked. The authors concede this process may not have accurately captured all LEP patients, and note their observed LEP rate (6.4%) was well below the metro Denver average (8%), suggesting under-ascertainment. Misclassification of the exposure could bias any of the LEP-versus-EP comparisons (mature cataract, complex surgery, preoperative acuity, steroid use, YAG capsulotomy, comorbidity).
• Interpretation modality and language-concordant provider or family presence not captured (Zafar cataract) The study did not record what interpretation modality was used, nor whether a language-concordant family member or physician was present during the encounter. It also did not account for limited literacy, health literacy, education, or other social determinants of health. Because the actual communication support each LEP patient received is unmeasured, the mechanism behind the observed disparities (e.g., prolonged steroid use, lower YAG capsulotomy uptake) cannot be attributed to language access specifically versus these unmeasured confounders.