Interventions to Promote Follow-up After Trabeculectomy Surgery in Rural Southern China: A Randomized Clinical Trial.

Importance:

Follow-up after trabeculectomy surgery is important to surgical success, but little is known about the effect of interventions on improving follow-up in low-resource areas.

Objective:

To examine whether text message reminders and free eye medications improve follow-up after trabeculectomy in rural southern China.

Design, Setting, and Participants:

This randomized clinical trial studied 222 consecutive patients undergoing trabeculectomy from October 1, 2014, through November 31, 2015, at 4 rural hospitals in Guangdong and Guangxi Provinces, China. Data from the intention-to-treat population were analyzed.

Interventions:

Patients undergoing trabeculectomy were randomized (1:1) to receive text message reminders 3 days before appointments at 1 and 2 weeks and 1 month after surgery and free topical corticosteroid medication (US$5.30) at each visit or to standard follow-up without reminders or free medication.

Main Outcomes and Measure:

Follow-up at 1 month postoperatively.

Results:

Among 222 eligible patients, 13 (5.9%) refused and 209 (94.1%) were enrolled, with 106 (50.7%) randomized to the intervention group (mean [SD] age, 64.4 [12.7] years; 56 women [52.8%]) and 103 (49.3%) to the control group (mean [SD] age, 63.0 [12.7] years; 53 women [51.5%]). A total of 6 patients (2.9%) were unavailable for follow-up. Attendance at 1 month for the intervention group (59 of 102 [57.8%]) was significantly higher than for the control group (34 of 101 [33.7%]) (unadjusted relative risk [RR], 1.72; 95% CI, 1.13-2.63; P = .01). Factors associated with 1-month attendance in multiple regression models included intervention group membership (RR, 1.65; 95% CI, 1.08-2.53; P = .02) and being told to return for suture removal (RR, 1.80; 95% CI, 1.06-3.06; P = .03). One-month attendance among controls not told about suture removal was 3 of 31 (9.7%), whereas it was 44 of 68 (64.7%) among the intervention group with suture removal (unadjusted RR, 6.69; 95% CI, 2.08-21.6; P = .001).

Conclusions and Relevance:

In this setting, low-cost interventions may significantly improve postoperative follow-up after glaucoma surgery, a potential opportunity for interventions known to improve surgical success.

Attitudes and Knowledge Concerning Corneal Donation in a Population-Based Sample of Urban Chinese Adults.

PURPOSE: To better understand knowledge and attitudes concerning corneal donation among Chinese adults.
METHODS: Randomly selected residents in predetermined age strata 20 to 60+ years completed home-based questionnaires in each of 12 randomly chosen communities in Guangzhou, southern China.
RESULTS: Among 1217 selected persons, 430 (35.3%) completed the questionnaires (mean age 40.4 yrs, 57.9% female). Refusers were older (44.8 yrs, P < 0.001), but sex did not differ (52.2% female, P = 0.07). Among participants, 175 (40.7%) were willing to donate their corneas (WTD). Differences between WTD and not WTD included donation knowledge score (range, 1-12) [WTD (SD) 6.91 ± 2.21, not WTD 5.62 ± 2.43, P < 0.001]; having discussed donation (WTD 26.3%, not WTD 8.63%, P < 0.001); viewing donation as unpopular (WTD 88.0%, not WTD 96.5%, P = 0.001); and feeling donation "damages the body" (WTD 15.4%, not WTD 25.7%, P = 0.013). Associated significantly with WTD in multiple regression models were higher knowledge score [odds ratio (OR) = 1.18, 95% confidence interval (CI), 1.04-1.32, P = 0.008]; not feeling donation "damages the body" (OR = 1.91, 95% CI, 1.07-3.43, P = 0.030); and willingness to discuss donation (OR = 10.6, 95% CI, 3.35-33.9, P < 0.001). WTD did not differ by age (>60 yrs: 22/51, 43.1%; ≤60 yrs: 153/379, 40.4%, P = 0.706). Assuming all those refusing the survey would not donate, 14.4% (175/1217) were WTD for themselves, though only 7.1% (86/1217) would do so on behalf of a family member if they did not know the deceased's preference.
CONCLUSIONS: Interventions to increase knowledge and promote discussions about donation, and policies allowing widespread expression of donation preference, are needed in this setting.

Self-refraction, ready-made glasses and quality of life among rural myopic Chinese children: a non-inferiority randomized trial.

PURPOSE: To study, for the first time, the effect of wearing ready-made glasses and glasses with power determined by self-refraction on children's quality of life. METHODS: This is a randomized, double-masked non-inferiority trial. Children in grades 7 and 8 (age 12-15 years) in nine Chinese secondary schools, with presenting visual acuity (VA) ≤6/12 improved with refraction to ≥6/7.5 bilaterally, refractive error ≤-1.0 D and <2.0 D of anisometropia and astigmatism bilaterally, were randomized to receive ready-made spectacles (RM) or identical-appearing spectacles with power determined by: subjective cycloplegic retinoscopy by a university optometrist (U), a rural refractionist (R) or non-cycloplegic self-refraction (SR). Main study outcome was global score on the National Eye Institute Refractive Error Quality of Life-42 (NEI-RQL-42) after 2 months of wearing study glasses, comparing other groups with the U group, adjusting for baseline score. RESULTS: Only one child (0.18%) was excluded for anisometropia or astigmatism. A total of 426 eligible subjects (mean age 14.2 years, 84.5% without glasses at baseline) were allocated to U [103 (24.2%)], RM [113 (26.5%)], R [108 (25.4%)] and SR [102 (23.9%)] groups, respectively. Baseline and endline score data were available for 398 (93.4%) of subjects. In multiple regression models adjusting for baseline score, older age (p = 0.003) and baseline spectacle wear (p = 0.016), but not study group assignment, were significantly associated with lower final score. CONCLUSION: Quality of life wearing ready-mades or glasses based on self-refraction did not differ from that with cycloplegic refraction by an experienced optometrist in this non-inferiority trial.

Joint Analysis of Multiple Algorithms and Performance Measures

There has been an increasing interest in the development of new methods using Pareto optimality to deal with multi-objective criteria (for example, accuracy and time complexity). Once one has developed an approach to a problem of interest, the problem is then how to compare it with the state of art. In machine learning, algorithms are typically evaluated by comparing their performance on different data sets by means of statistical tests. Standard tests used for this purpose are able to consider jointly neither performance measures nor multiple competitors at once. The aim of this paper is to resolve these issues by developing statistical procedures that are able to account for multiple competing measures at the same time and to compare multiple algorithms altogether. In particular, we develop two tests: a frequentist procedure based on the generalized likelihood-ratio test and a Bayesian procedure based on a multinomial-Dirichlet conjugate model. We further extend them by discovering conditional independences among measures to reduce the number of parameters of such models, as usually the number of studied cases is very reduced in such comparisons. Data from a comparison among general purpose classifiers is used to show a practical application of our tests.