Randomized Controlled Trial (RCT) of...
Vision for Baltimore, providing free vision screening & eyeglasses in elementary and middle school
Reviewed
JAMA Ophthalmology (Sept. 2021) published results of the Vision for Baltimore RCT, evaluating a citywide program of free vision screening and eyeglasses in elementary and middle school. Despite the study abstract’s largely positive portray of results, the study found no discernible impacts on its preregistered primary outcomes – reading and math scores on the state test over 1-2 years.
Importance Uncorrected refractive error in school-aged children may affect learning.
Objective To assess the effect of a school-based vision program on academic achievement among students in grades 3 to 7.
Design, Setting, and Participants This cluster randomized clinical trial was conducted in Baltimore City Public Schools during school years from 2016 to 2019 among 2304 students in grades 3 to 7 who received eye examinations and eyeglasses.
Intervention Participating schools were randomized 1:1:1 to receive eye examinations and eyeglasses during 1 of 3 school years (2016-2017, 2017-2018, and 2018-2019).
Main Outcomes and Measures The primary outcome was 1-year intervention impact, measured by effect size (ES), defined as the difference in score on an academic test (i-Ready or Partnership for Assessment of Readiness for College and Careers tests on reading and mathematics) between intervention and control groups measured in SD units, comparing cohort 1 (intervention) with cohorts 2 and 3 (control) at the end of program year 1 and comparing cohort 2 (intervention) with cohort 3 (control) at the end of program year 2. The secondary outcome was 2-year intervention impact, comparing ES in cohort 1 (intervention) with cohort 3 (control) at the end of program year 2. Hierarchical linear modeling was used to assess the impact of the intervention. Analysis was performed on an intention-to-treat basis.
Results Among the 2304 students included in the study, 1260 (54.7%) were girls, with a mean (SD) age of 9.4 (1.4) years. The analysis included 964 students (41 schools) in cohort 1, 775 students (41 schools) in cohort 2, and 565 students (38 schools) in cohort 3. There were 1789 Black students (77.6%), 388 Latinx students (16.8%), and 406 students in special education (17.6%). There was an overall 1-year positive impact (ES, 0.09; P = .02) as assessed by the i-Ready reading test during school year 2016-2017. Positive impact was also observed among female students (ES, 0.15; P < .001), those in special education (ES, 0.25; P < .001), and students who performed in the lowest quartile at baseline (ES, 0.28; P < .001) on i-Ready reading and among students in elementary grades on i-Ready mathematics (ES, 0.03; P < .001) during school year 2016-2017. The intervention did not show a sustained impact at 2 years or on Partnership for Assessment of Readiness for College and Careers testing.
Conclusions and Relevance Students in grades 3 to 7 who received eyeglasses through a school-based vision program achieved better reading scores. Students had improved academic achievement over 1 year; however, a sustained impact was not observed after 2 years.
Main Outcomes and Measures The
first
primary outcome was 1-year intervention impact, measured by effect size (ES), defined as the difference in score on
the
an academic test (i-Ready or Partnership for Assessment of Readiness for College and Careers tests on reading and mathematics) between intervention and control groups measured in SD units, comparing cohort 1 (intervention) with cohorts 2 and 3 (control) at the end of program year 1 and comparing cohort 2 (intervention) with cohort 3 (control) at the end of program year 2. The secondary
primary
outcome was 2-year intervention impact, comparing ES in cohort 1 (intervention) with cohort 3 (control) at the end of program year 2. Hierarchical linear modeling was used to assess the impact of the intervention. Analysis was performed on an intention-to-treat basis.
Results Among the 2304 students included in the study, 1260 (54.7%) were girls, with a mean (SD) age of 9.4 (1.4) years. The analysis included 964 students (41 schools) in cohort 1, 775 students (41 schools) in cohort 2, and 565 students (38 schools) in cohort 3. There were 1789 Black students (77.6%), 388 Latinx students (16.8%), and 406 students in special education (17.6%)..
There were no impacts on the preregistered primary outcomes – reading and math scores on the PARCC test – after either 1 or 2 years. Exploratory analyses suggested possible gains for some measures and subgroups. These include
There was an overall 1-year positive impact (ES, 0.09; P = .02) as assessed by the i-Ready reading test during school year 2016-2017. Positive impact was also observed among female students (ES, 0.15; P < .001), those in special education (ES, 0.25; P < .001), and students who performed in the lowest quartile at baseline (ES, 0.28; P < .001) on i-Ready reading and among students in elementary grades on i-Ready mathematics (ES, 0.03; P < .001) during school year 2016-2017. The intervention did not show a sustained impact at 2 years or on Partnership for Assessment of Readiness for College and Careers testing.
Conclusions and Relevance Students in grades 3 to 7 who received eyeglasses through a school-based vision program achieved
no
better reading
or math
scores. Students had improved academic achievement over 1 year; however, a sustained impact was not observed after 2 years.
No-Spin’s Study Overview
High-quality RCT of a citywide program providing free vision screening and eyeglasses in elementary and middle schools finds no discernible impacts on reading and math scores on the state test over 1-2 years.
Study Design:
- The study sample comprised 127 elementary and middle schools, randomly assigned to receive the program starting in 2016-2017 school year vs. 2017-2018 vs. 2018-2019. The study compared state test scores in reading and math for students across the 3 groups who were screened as having vision problems.
- Importantly, vision correction offers many benefits (e.g., quality of life, safety). This study sought to determine if such correction, as delivered by the Baltimore program, could also increase academic achievement in a student population that was 80-90% Black and low-income.
- Based on careful review, this was a mostly well-conducted RCT (e.g., low school and student attrition, good baseline balance, preregistered analyses).
Findings:
- The study found no statistically significant impacts on the primary outcomes - reading and math scores on the state test - after either 1 year (based on 3,170 students) or 2 years (based on 1,528 students). The effect sizes were all close to zero.
Comment:
- Disclosure: Arnold Ventures, which funds No-Spin Evidence Review, helped fund this study.
Click or tap a highlight to see No-Spin’s comment