Strabismus
Conditions
Keywords
strabismus surgery, machine learning, nomogram, real-world evidence, natural history
Brief summary
This retrospective study constructs the largest multicenter registry of strabismus care in Japan. The registry evaluates real-world surgical and conservative treatment outcomes, safety, and natural history for various types of strabismus, including common, rare, and complex cases.
Detailed description
This retrospective multicenter cohort study collects clinical data from participating academic and high-volume centers in Japan. To minimize chronological bias and maintain feasibility, the observation period depends on the target disease. For common types of strabismus, data extraction is limited to a recent timeframe (e.g., the past 3 years) to avoid the influence of behavioral changes during the coronavirus disease 2019 (COVID-19) pandemic. For rare and complex strabismus, the extraction period extends up to 10 years (from January 2016 to the date of Institutional Review Board approval) to secure an adequate sample size. Statistical analyses will incorporate mixed-effects models to adjust for institutional clustering and differences in patient backgrounds. Missing data will be addressed via multiple imputation before developing prediction models with machine learning algorithms such as random forests.
Interventions
PROCEDUREStrabismus Surgery
Surgical correction of ocular deviation.
OTHERConservative Treatment or Observation
Non-surgical management, including clinical observation for natural history, prism therapy, or orthoptic treatment.
Sponsors
Kyoto University, Graduate School of Medicine
Study design
Observational model
COHORT
Time perspective
RETROSPECTIVE
Eligibility
Sex/Gender
ALL
Healthy volunteers
No
Inclusion criteria
* Patients diagnosed with strabismus who underwent clinical evaluation or management (including observation for natural history, conservative treatment, or surgery) at participating institutions. * Observation period for common strabismus: A restricted recent timeframe (e.g., the most recent 3 years) to minimize selection bias. * Observation period for rare and complex strabismus: Up to 10 years (January 1, 2016, to the date of Institutional Review Board approval) to secure an adequate sample size.
Exclusion criteria
* Patients who formally opted out of data utilization for this study.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Success rate of ocular deviation correction and residual deviation angle | At standard observation intervals (2 to 4 weeks, 3 to 6 months, 1 year, and 2 years post-treatment or baseline) and through study completion, up to 10 years. | Assess the angle of ocular deviation and determine the success rate of clinical management (including observation for natural history, surgical, or conservative treatments) based on established clinical criteria. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Recovery of binocular visual function | Through study completion, up to 10 years | Evaluate binocularity using standardized clinical tests. |
| Predictive accuracy of the machine learning nomogram | Through study completion, up to 10 years | Compare the predictive performance of the developed machine learning models against actual surgical outcomes. |
Countries
Japan
Contacts
CONTACTManabu Miyata, MD, PhD
Outcome results
None listed