4.2. How to improve the IOLEvidence Score

This IOLEvidence Score recommendations applies to IOLEvidence version 26.2 and later; earlier versions compute the score using quality of reporting only and do not include the Risk of Bias component.

4.2.1. Structure of the Score

The Study Quality Score is a single value from 0 to 100, where:

• Higher score indicates better quality of reporting and/or lower risk of bias

• Lower score indicates poorer quality of reporting and/or higher risk of bias

The score is computed by combining two complementary components:

1. Reporting and Appraisal Component
   This component evaluates whether essential methodological and reporting information is present and sufficiently detailed. It is based on 10 appraisal domains, each producing a normalized value (0–1). For each domain, results are classified as:

• Yes → adequately reported

• Unclear → partially reported

• No → not reported

Some elements are weighted more heavily when they are critical for interpretation and evidence synthesis.

2. Risk of Bias Component
   This component adjusts the score using study features associated with higher or lower risk of bias. It focuses on characteristics related to study design and conduct that can systematically influence results. In the final score, this component is aligned so that better design and lower bias risk increase the score, while higher bias risk decreases it.

4.2.2. Appraisal Domains Explained in Detail (Reporting and Appraisal Component)

4.2.2.1. Inclusion Criteria

This domain evaluates whether the study clearly reports inclusion and exclusion criteria for the study. Special attention is given to factors that can strongly influence outcomes in IOL studies, including:

• Presence or exclusion of ocular comorbidities

• Preoperative corneal astigmatism and residual astigmatism

• Planned postoperative refractive target or surgical plan

• Reporting and handling of adverse events

• Inclusion or exclusion of eyes with prior Nd:YAG capsulotomy

In the algorithm, these elements are independently weighted, meaning partial reporting results in an intermediate score.

Why it matters: Without clearly defined inclusion criteria, results cannot be contextualized or compared across studies.

4.2.2.2. Definition of the condition

This domain assesses whether the clinical condition under study is clearly defined and appropriate. In IOLEvidence, the condition is cataract surgery or refractive lens exchange for presbyopia correction.

The algorithm checks whether the study explicitly reports the condition (e.g. cataract diagnosis) in a way that is coherent with baseline visual status.

Why it matters: Ambiguity in the definition of the surgical indication limits interpretability of outcomes.

4.2.2.3. Identification of the condition

This domain evaluates whether valid methods are used to identify the condition for all participants. Specifically, it checks for reporting of:

• Preoperative corrected distance visual acuity (CDVA)

• Variability of CDVA (standard deviation)

Both elements are weighted, reflecting that reporting a mean alone is insufficient.

Why it matters: Baseline CDVA is essential to understand postoperative change and outcome interpretation.

4.2.2.4. Consecutive inclusion of participants

This domain evaluates whether participants were included consecutively over a defined period. The algorithm distinguishes between:

• Explicit confirmation of consecutive inclusion

• Unclear reporting

• Evidence of selective inclusion

Why it matters: Non-consecutive inclusion may introduce selection bias and distort outcome distributions.

4.2.2.5. Complete inclusion of participants

This domain assesses whether all eligible participants were included and whether follow-up was complete. The algorithm rewards:

• Explicit reporting of no recruitment loss or follow-up loss

• Penalizes missing or unclear follow-up information

Why it matters: Loss to follow-up can systematically bias reported outcomes.

4.2.2.6. Demographic reporting

This domain evaluates the completeness of demographic and biometric reporting, with differential weighting based on clinical relevance. These demographic and biometric variables include:

• Age and its variability

• Photopic and mesopic pupil diameter

• Axial length or IOL power

• Sex distribution

• Anterior chamber depth

• Keratometry

• Corneal astigmatism

• Measurement device information

Why it matters: Demographic and biometric variables are key confounders in IOL performance and outcome variability.

4.2.2.7. Clinical and surgical information

This domain assesses reporting of surgical planning and execution details that may influence outcomes, including:

• Planned refractive target and its variability

• Formula and lens constant used

• Number of surgeons

• Surgical technique (phacoemulsification, incision type)

• Capsulotomy size

Higher weight is assigned to elements directly influencing refractive accuracy.

Why it matters: Inadequate reporting of surgical parameters limits reproducibility and interpretation.

4.2.2.8. Outcome and follow-up reporting

This is one of the most heavily weighted and complex domains, and it is endpoint-specific. The algorithm adapts scoring logic depending on whether the outcome is:

• Visual acuity

• Contrast sensitivity

• Defocus curve

• Patient-reported outcomes

• Measurement conditions (monocular vs binocular)

• Follow-up duration

• Use of standardized tests (e.g. ETDRS)

• Luminance conditions

• Reporting of mean values and variability

• Control of memorization bias in defocus curves

Why it matters: Outcomes are only interpretable if measurement conditions and reporting standards are clearly defined.

4.2.2.9. Reporting of study site(s)

This domain assesses whether the clinical site or center(s) where the study was conducted are clearly reported.

Why it matters: Site information provides context on patient population, practice patterns, and external validity.

4.2.2.10. Statistical analysis

This domain evaluates whether the statistical analysis is appropriate and sufficiently reported, including:

• Explicit statement of acceptable statistical methods

• Reporting of sample size (eyes and subjects)

• Inclusion of dispersion metrics (SD or CI)

The algorithm gives substantial weight to statistical adequacy.

Why it matters: Without proper statistical reporting, quantitative results cannot be reliably interpreted or synthesized.

4.2.3. Risk of Bias Component

In addition to reporting completeness, the score incorporates design and conduct signals that can influence risk of bias. This component typically accounts for:

• Temporal design clarity (e.g., prospective vs retrospective vs unclear)

• Standardization of outcome measurement (e.g., ETDRS use and controlled measurement conditions such as luminance)

• Appropriateness and transparency of statistical methods

• Sample size considerations (number of subjects and/or eyes)

• Recruitment method quality (e.g., consecutive inclusion vs selective or unclear inclusion)

• Endpoint reporting level (monocular and/or binocular outcomes when relevant)

• Study design category (e.g., randomized vs case series)

• End-point specific safeguards (e.g., measures addressing memorization effects when applicable for defocus curves)

These elements are incorporated so that studies with stronger design features and lower bias susceptibility receive a higher final score, even when reporting completeness is similar.

4.2.4.Recommendations for Authors

We recommend authors to follow established checklists and reporting standards for measuring and presenting outcomes in studies involving intraocular lenses. Adhering to these standards ensures transparency, completeness, and comparability of reported results, facilitating correct interpretation and evidence synthesis.

In particular, authors are encouraged to follow the methodological and reporting recommendations described in the following publications:

Sentence to include in the manuscript

1. Fernández, J., Rocha-De-Lossada, C. & Rodríguez-Vallejo, M. Recommendations for Writing Clinical Research Manuscripts: From Monofocal to Multifocal Intraocular Lenses. Int. J. Environ. Res. Public Health 19, 17036 (2022).

2. Fernández, J. et al. Standard for collecting and reporting outcomes of IOL–based refractive surgery: update for enhanced monofocal, EDOF, and multifocal IOLs. J Cataract Refract Surg 48, 1235–1241 (2022).